[
  {
    "path": ".gitignore",
    "content": "logs\n*.pyc\nlfw/*.png\n"
  },
  {
    "path": "README.md",
    "content": "\nThis repository shows how to train ResNet models in PyTorch on publicly available face recognition datasets.\n\n### Setup\n\n* Install [Anaconda](https://conda.io/docs/user-guide/install/linux.html) if not already installed in the system.\n* Create an Anaconda environment: `conda create -n resnet-face python=2.7` and activate it: `source activate resnet-face`.\n* Install PyTorch and TorchVision inside the Anaconda environment. First add a channel to conda: `conda config --add channels soumith`. Then install: `conda install pytorch torchvision cuda80 -c soumith`.\n* Install the dependencies using conda: `conda install scipy Pillow tqdm scikit-learn scikit-image numpy matplotlib ipython pyyaml`.\n* *Notes*:\n    * Multiple GPUs (we used 5 GeForce GTX 1080Ti in parallel) recommended for the training to finish in reasonable time.\n    * Tested on server running CentOS\n    * Using [PyTorch](http://pytorch.org/tutorials/beginner/deep_learning_60min_blitz.html) \n    * Optional - an explanatory [blogpost](https://blog.waya.ai/deep-residual-learning-9610bb62c355) on Deep Residual Networks (ResNets) structure.\n\n### Contents\n- [ResNet-50 on UMD-Faces](https://github.com/AruniRC/resnet-face-pytorch#pytorch-resnet-on-umd-face)\n    - [Dataset preparation](https://github.com/AruniRC/resnet-face-pytorch#dataset-preparation)\n    - [Training](https://github.com/AruniRC/resnet-face-pytorch#training)\n    - [Evaluation demo](https://github.com/AruniRC/resnet-face-pytorch#evaluation)\n- [ResNet-50 on VGGFace2](https://github.com/AruniRC/resnet-face-pytorch#pytorch-resnet-on-vggface2)\n    - [Dataset preparation](https://github.com/AruniRC/resnet-face-pytorch#dataset-preparation-1)\n    - [Training](https://github.com/AruniRC/resnet-face-pytorch#training-1)\n    - [Evaluation LFW](https://github.com/AruniRC/resnet-face-pytorch#evaluation-1)\n\n\n\n## PyTorch ResNet on UMD-Face\n\nDemo to train a ResNet-50 model on the [UMDFaces](http://www.umdfaces.io/) dataset.\n\n### Dataset preparation\n\n* Download the [UMDFaces](http://www.umdfaces.io/) dataset (the 3 batches of _still_ images), which contains 367,888 face annotations for 8,277 subjects, split into 3 batches.\n* The images need to be cropped into 'train' and 'val' folders. Since the UMDFaces dataset does not specify training and validation sets, by default we select two images from every subject for validation. \n* The cropping code is in the Python script [umd-face/run_crop_face.py](./umd-face/run_crop_face.py). It takes the following command-line arguments:\n    * --dataset_path (-d)\n    * --output_path (-o)\n    * --batch (-b) \n* The following shell command does the cropping for each batch in parallel, using default `dataset_path` and `output_path`.\n`for i in {0..2}; python umd-face/run_crop_face -b $i &; done`.\n\n:small_red_triangle: **TODO** - takes very long, convert into shell+ImageMagick script.\n\n\n### Usage\n\n#### Training:  \n* Training script is [umd-face/train_resnet_umdface.py](./umd-face/train_resnet_umdface.py)\n* Multiple GPUs: \n    * Under section 3 (\"Model\") of the training script, we specify which GPUs to use in parallel: `model = torch.nn.DataParallel(model, device_ids=[0, 1, 2, 3, 4]).cuda()`. Change these numbers depending on the number of available GPUs. \n    * Use `watch -d nvidia-smi` to constantly monitor the multi-GPU usage from the terminal. \n    * :small_red_triangle: **TODO** - make this into command-line args.\n* At the terminal, specify where the cropped face images are saved using an environment variable: `DATASET_PATH=local/path/to/cropped/umd/faces`\n* [config.py](./config.py) lists all the training configurations (i.e. model hyper-parameters) in a numbered Python dict.\n    * The training of ResNet-50 was done in 3 stages (*configs 4, 5 and 6*), each of *30 epochs*. For the first stage, we started with the ImageNet-pre-trained model from PyTorch. After the first stage, we started from the saved model of the previous stage (using the `--model_path` or `-m` command-line argument) and divided the learning rate by a factor of 10.\n    * Stage 1 (config-4): train on  the *full UMDFaces dataset for 30 epochs* (42180 iterations with batchsize 250) with a learning rate of 0.001, starting from an ImageNet pre-trained model. These settings are defined in *config-4* of [config.py](./config.py), which is selected using the `-c 4` flag in the command. Example to train a ResNet-50 on UMDFaces dataset using config-4: run `python umd-face/train_resnet_umdface.py -c 4 -d $DATASET_PATH`.\n    * Stage 2 (config-5): use the best model checkpointed from config-4 to initialize the network and train it using config-5 `python umd-face/train_resnet_umdface.py -c 5 -m ./umd-face/logs/MODEL-resnet_umdfaces_CFG-004_TIMESTAMP/model_best.pth.tar -d $DATASET_PATH` and so on for the subsequent stages.\n* *Training logs:* Each time the training script is run, a new output folder with a timestamp is created by default under `./umd-face/logs` , i.e.  `./umd-face/logs/MODEL-CFG-TIMESTAMP/`. Under an experiment's log folder the settings for each experiment can be viewed in `config.yaml`; metrics such as the training and validation losses are updated in `log.csv`. \nMost of the usual settings (data augmentations, learning rates, number of epochs to train, etc.) can be customized by editing `config.py` and `umd-face/train_resnet_umdface.py`.\n* *Plotting CSV logs:* The log-file plotting utility function can be called from the command line as shown in the snippet below. The figure is saved under the log folder in the output location of that experiment.\n    * `LOG_FILE=umd-face/logs/MODEL-resnet_umdfaces_CFG-004_TIMESTAMP/log.csv`\n    * `python -c \"from utils import plot_log_csv; plot_log_csv('$LOG_FILE')\"`\n    * If that gives parsing errors: `python -c \"from utils import plot_log; plot_log('$LOG_FILE')\"`\n\nstage 1 |   stage 2  | stage 3  \n:------:|:----------:|:--------:\n![](samples/stage1_log_plots.png)|  ![](samples/stage2_log_plots.png) | ![](samples/stage3_log_plots.png) \n\n\n#### Pre-trained model: \n\n:red_circle: TODO - release pre-trained ResNet-50 on UMD-Faces :construction:\n\n\n#### Evaluation: \n\n*Verification demo:* We have a short script, [run_resnet_demo.py](./run_resnet_demo.py) to demonstrate the usage of the model on a toy face verification example. The visualized output of the demo is saved in the the root directory of the project. The 3 sample images are taken from the [LFW](http://vis-www.cs.umass.edu/lfw/) dataset.\n\n![](samples/demo_verif.png)\n\n\n---\n\n\n## PyTorch ResNet on VGGFace2\n\nTraining a *ResNet-50* model in PyTorch on the [VGGFace2](https://www.robots.ox.ac.uk/~vgg/data/vgg_face2/) dataset.\n\n\n### Dataset preparation\n\n* Register on the [VGGFace2](https://www.robots.ox.ac.uk/~vgg/data/vgg_face2/) website and download their dataset\n* VGGFace2 provides loosely-cropped images. We use crops from the [Faster R-CNN face detector](https://github.com/playerkk/face-py-faster-rcnn), saved as a CSV in `[filename, subject_id, xmin, ymin, width, height]` format (the CSV with pre-computed face crops is not yet made available).  \n* The `vgg-face-2/crop_face.sh` script is used to crop the face images into a separate output folder. Please look at the settings section in the script to assign correct paths depending on where the VGGFace2 data was downloaded on your local machine. This takes about a day. **TODO** - multi-process.\n    * Training images are under `OUTPUT_PATH/train-crop`\n    * Validation images (2 images per subject) under `OUTPUT_PATH/val-crop`\n\n### Training\n\n* We used 7 GeForce GTX 1080Ti GPUs in parallel (PyTorch DataParallel) to train the network, using Batch Normalization, following the training procedure described in the [VGGFace2](https://www.robots.ox.ac.uk/~vgg/data/vgg_face2/) paper.\n* Training settings are defined under *configs-20, 21, 22* in [config.py](./config.py).\n* Briefly, training is done using SGD optimizer, starting with a learning rate of 0.1, which gets divided by 10 in subsequent stages. A new stage is begun whenever the validation curve flattens.\n* First stage training command, starting with a ResNet-50 from scratch: `python vgg-face-2/train_resnet50_vggface_scratch.py -c 20`\n* Subsequent training stages (with lowered learning rates) would be: \n    - `python vgg-face-2/train_resnet50_vggface_scratch.py -c 21 -m PATH_TO_BEST_MODEL_CFG-20`\n    - `python vgg-face-2/train_resnet50_vggface_scratch.py -c 22 -m PATH_TO_BEST_MODEL_CFG-21`\n\n### Evaluation\n\nInstructions on how to setup and run the LFW evaluation are at [lfw/README.md](./lfw/README.md). \n\nDevTest |  10 fold   \n:------:|:----------:\n![](samples/lfw_roc_devTest.png)|  ![](samples/lfw_roc_10fold.png)\n\n\n\n\n"
  },
  {
    "path": "config.py",
    "content": "\n\n\nconfigurations = {\n    # same configuration as original work\n    # https://github.com/shelhamer/fcn.berkeleyvision.org\n    0: dict(\n        max_iteration=100000,\n        lr=1.0e-10,\n        momentum=0.99,\n        weight_decay=0.0005,\n        interval_validate=4000,\n    ),\n\n    # python train_*.py -g 0 -c 1\n    1: dict(\n        max_iteration=100,\n        lr=0.01,             # changed learning rate\n        lr_decay_epoch=None, # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=10,\n        optim='Adam',\n        batch_size=100,\n    ),\n\n    2: dict(\n        max_iteration=8670,    # num_iter_per_epoch = ceil(num_images/batch_size)\n        lr=0.01,               # high learning rate\n        lr_decay_epoch=None,   # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=10,\n        optim='Adam',\n        batch_size=250,\n    ),\n\n    3: dict(\n        # num_iter_per_epoch = ceil(num_images/batch_size)\n        max_iteration=8670,  # 10 epochs on subset of images\n        lr=0.001,            # lowered learning rate\n        lr_decay_epoch=None, # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=10,\n        optim='Adam',\n        batch_size=250,\n    ),\n    # ---------------------------------------------------------------------------------\n\n    # ResNet-50 on UMDFaces: stage 1\n    4: dict(\n        # num_iter_per_epoch = ceil(num_images/batch_size)\n        max_iteration=42180,  # 30 epochs on full dataset (about 10 hours)\n        lr=0.001,             # learning rate\n        lr_decay_epoch=None,  # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=50,\n        optim='Adam',\n        batch_size=250,       # DataParallel over 5 gpus\n    ),\n\n    # ResNet-50 on UMDFaces: stage 2\n    5: dict(\n        # num_iter_per_epoch = ceil(num_images/batch_size)\n        max_iteration=42180,  # 30 epochs on full dataset\n        lr=0.0001,            # lowered learning rate\n        lr_decay_epoch=None,  # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=50,\n        optim='Adam',\n        batch_size=250,\n    ),\n\n    # ResNet-50 on UMDFaces: stage 3\n    6: dict(\n        # num_iter_per_epoch = ceil(num_images/batch_size)\n        max_iteration=42180,   # 30 epochs on full dataset\n        lr=0.00001,            # lowered learning rate\n        lr_decay_epoch=None,   # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=50,\n        optim='Adam',\n        batch_size=250,\n    ),\n    # ---------------------------------------------------------------------------------\n\n    # ResNet-101 on VGGFace2: stage 1\n    7: dict(\n        # num_iter_per_epoch = ceil(num_images/batch_size)\n        max_iteration=267630,  # 30 epochs on full dataset\n        lr=0.001,              # learning rate\n        lr_decay_epoch=None,   # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=500,\n        optim='Adam',\n        batch_size=350,        # DataParallel over 7 gpus\n    ),\n\n    # python vgg-face-2/train_resnet_vggface.py -c 8 -m PATH-TO-CFG-7-SAVED-MODEL\n    8: dict(\n        # num_iter_per_epoch = ceil(num_images/batch_size)\n        max_iteration=267630,  # 30 epochs on full dataset\n        lr=0.0001,             # lowered learning rate\n        lr_decay_epoch=None,   # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=500,\n        optim='Adam',\n        batch_size=350,        # DataParallel over 7 gpus\n    ),\n\n    #### \n    # ResNet-101 on VGGFace2: stage 1\n    11: dict(\n        # num_iter_per_epoch = ceil(num_images/batch_size)\n        max_iteration=267630,  # 30 epochs on full dataset\n        lr=0.1,              # learning rate\n        lr_decay_epoch=None,   # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=200,\n        optim='Adam',\n        batch_size=350,        # DataParallel over 7 gpus\n    ),\n\n    # ResNet-101 on VGGFace2: stage 2\n    #   python vgg-face-2/train_resnet_vggface_scratch.py -c 12 -m ./vgg-face-2/logs/MODEL..-CFG-11...\n    12: dict(\n        # num_iter_per_epoch = ceil(num_images/batch_size)\n        max_iteration=267630,  # 30 epochs on full dataset\n        lr=0.01,               # reduced learning rate by factor 10\n        lr_decay_epoch=None,   # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=200,\n        optim='Adam',\n        batch_size=350,        # DataParallel over 7 gpus\n    ),\n\n    13: dict(\n        # num_iter_per_epoch = ceil(num_images/batch_size)\n        max_iteration=267630,  # 30 epochs on full dataset\n        lr=0.001,               # reduced learning rate by factor 10\n        lr_decay_epoch=None,   # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=200,\n        optim='Adam',\n        batch_size=350,        # DataParallel over 7 gpus\n    ),\n\n\n    # ResNet from scratch with SGD\n    20: dict(\n        # num_iter_per_epoch = ceil(num_images/batch_size)\n        max_iteration=267630,  # 22 epochs on full dataset\n        lr=0.1,                \n        lr_decay_epoch=None,   # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=200,\n        optim='SGD',\n        batch_size=256,        # DataParallel over 7 gpus\n    ),\n\n    21: dict(\n        # num_iter_per_epoch = ceil(num_images/batch_size)\n        max_iteration=267630,  # 22 epochs on full dataset\n        lr=0.01,                \n        lr_decay_epoch=None,   # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=200,\n        optim='SGD',\n        batch_size=256,        # DataParallel over 7 gpus\n    ),\n\n    22: dict(\n        # num_iter_per_epoch = ceil(num_images/batch_size)\n        max_iteration=267630,  # 22 epochs on full dataset\n        lr=0.001,                \n        lr_decay_epoch=None,   # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=200,\n        optim='SGD',\n        batch_size=256,        # DataParallel over 7 gpus\n    ),\n\n    # Used to fine-tune Resnet101-512d in the second stage\n    # (after the new layers are converged, entire net is fine-tuned)\n    23: dict(\n        # num_iter_per_epoch = ceil(num_images/batch_size)\n        max_iteration=267630,  # 22 epochs on full dataset\n        lr=0.0001,                \n        lr_decay_epoch=None,   # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=200,\n        optim='SGD',\n        batch_size=256,        # DataParallel over 7 gpus\n    ),\n\n    # lower learning rates on fine-tuning bottleneck (Resnet101-512d)\n    24: dict(\n        # num_iter_per_epoch = ceil(num_images/batch_size)\n        max_iteration=267630,  # 22 epochs on full dataset\n        lr=0.00001,            # lowered learning rate                \n        lr_decay_epoch=None,   # disable automatic lr decay\n        momentum=0.9,  \n        weight_decay=0.0005,\n        interval_validate=200,\n        optim='SGD',\n        batch_size=256,        # DataParallel over 7 gpus\n    ),\n\n\n}\n\n\n\n"
  },
  {
    "path": "data_loader.py",
    "content": "import collections\nimport os.path as osp\n# from __future__ import division\n\nimport numpy as np\nimport PIL.Image\nimport scipy.io\nimport skimage\nimport skimage.color as color\nfrom skimage.transform import rescale\nfrom skimage.transform import resize\nimport torch\nfrom torch.utils import data\n\n\n\nDEBUG = False\n\n\n\nclass DemoFaceDataset(data.Dataset):\n    '''\n        Dataset subclass for demonstrating how to load images in PyTorch.\n\n    '''\n\n    # -----------------------------------------------------------------------------\n    def __init__(self, root, split='train', set='tiny', im_size=250):\n    # -----------------------------------------------------------------------------\n        '''\n            Parameters\n            ----------\n            root        -   Path to root of ImageNet dataset\n            split       -   Either 'train' or 'val'\n            set         -   Can be 'full', 'small' or 'tiny' (5 images)\n        ''' \n        self.root = root  # E.g. '.../ImageNet/images' or '.../vgg-face/images'\n        self.split = split\n        self.files = collections.defaultdict(list)\n        self.im_size = im_size # scale image to im_size x im_size\n        self.set = set\n\n        if set == 'small':\n            raise NotImplementedError()\n            \n        elif set == 'tiny':\n            # DEBUG: 5 images\n            files_list = osp.join(root, 'tiny_face_' + self.split + '.txt')\n\n        elif set == 'full':\n            raise NotImplementedError()\n\n        else:\n        \traise ValueError('Valid sets: `full`, `small`, `tiny`.')\n\n        assert osp.exists(files_list), 'File does not exist: %s' % files_list\n\n        imfn = []\n        with open(files_list, 'r') as ftrain:\n            for line in ftrain:\n                imfn.append(osp.join(root, line.strip()))\n        self.files[split] =  imfn\n\n\n    # -----------------------------------------------------------------------------\n    def __len__(self):\n    # -----------------------------------------------------------------------------\n        return len(self.files[self.split])\n\n\n    # -----------------------------------------------------------------------------\n    def __getitem__(self, index):\n    # -----------------------------------------------------------------------------\n        img_file = self.files[self.split][index]\n        img = PIL.Image.open(img_file)\n\n        # HACK: for non-RGB images - 4-channel CMYK or 1-channel grayscale\n        if len(img.getbands()) != 3:\n            while len(img.getbands()) != 3:\n                index -= 1\n                img_file = self.files[self.split][index] # if -1, wrap-around\n                img = PIL.Image.open(img_file)\n\n        if self.im_size > 0:\n        \t# Scales image to a square of default size 250x250\n        \tscaled_dim = (self.im_size.astype(np.int32), \n        \t\t\t\t  self.im_size.astype(np.int32))\n        \timg = img.resize(scaled_dim, PIL.Image.BILINEAR)\n\n        label = 1 # TODO: read in a class label for each image\n\n        img = np.array(img, dtype=np.uint8)\n        im_out = torch.from_numpy(im_out).float()\n        im_out = im_out.permute(2,0,1) # C x H x W\n\n        return im_out, label\n\n\n\nclass LFWDataset(data.Dataset):\n    '''\n        Dataset subclass for loading LFW images in PyTorch.\n        This returns multiple images in a batch.\n    '''\n\n    def __init__(self, path_list, issame_list, transforms, split = 'test'):\n        '''\n            Parameters\n            ----------\n            path_list    -   List of full path-names to LFW images\n        ''' \n        self.files = collections.defaultdict(list)\n        self.split = split\n        self.files[split] =  path_list\n        self.pair_label = issame_list\n        self.transforms = transforms\n\n    def __len__(self):\n        return len(self.files[self.split])\n\n    def __getitem__(self, index):\n        img_file = self.files[self.split][index]\n        img = PIL.Image.open(img_file)\n        if DEBUG:\n            print img_file\n        im_out = self.transforms(img)\n        return im_out\n\n\n\nclass IJBADataset(data.Dataset):\n    '''\n        Dataset subclass for loading IJB-A images in PyTorch.\n        This returns multiple images in a batch.\n        Path_list -- full paths to cropped images saved as <sighting_id>.jpg \n    '''\n    def __init__(self, path_list, transforms, split=1):\n        '''\n            Parameters\n            ----------\n            path_list    -   List of full path-names to IJB-A images of one split  \n        ''' \n        self.files = collections.defaultdict(list)\n        self.split = split\n        self.files[split] =  path_list\n        self.transforms = transforms\n\n    def __len__(self):\n        return len(self.files[self.split])\n\n    def __getitem__(self, index):\n        img_file = self.files[self.split][index]\n        img = PIL.Image.open(img_file)\n        if not img.mode == 'RGB':\n            img = img.convert('RGB')\n        if DEBUG:\n            print img_file\n        im_out = self.transforms(img)\n        return im_out\n\n\n"
  },
  {
    "path": "environment.yml",
    "content": "name: resnet-demo\ndependencies:\n - python=2.7\n - scipy\n - Pillow\n - tqdm\n - scikit-learn\n - scikit-image\n - numpy\n - matplotlib\n - ipython\n - pyyaml\n\n"
  },
  {
    "path": "finetune.py",
    "content": "import datetime\nimport math\nimport os\nimport os.path as osp\nimport shutil\n\nimport numpy as np\nimport PIL.Image\nimport pytz\nimport scipy.misc\nimport torch\nfrom torch.autograd import Variable\nimport torch.nn.functional as F\nimport tqdm\n\nimport utils\nimport gc\n\n\ndef lr_scheduler(optimizer, epoch, init_lr=0.001, lr_decay_epoch=7):\n    \"\"\"Decay learning rate by a factor of 0.1 every lr_decay_epoch epochs.\"\"\"\n    lr = init_lr * (0.1**(epoch // lr_decay_epoch))\n\n    if epoch % lr_decay_epoch == 0:\n        print('LR is set to {}'.format(lr))\n\n    for param_group in optimizer.param_groups:\n        param_group['lr'] = lr\n\n    return optimizer\n\n\n\nclass Trainer(object):\n\n    # -----------------------------------------------------------------------------\n    def __init__(self, cuda, model, criterion, optimizer, init_lr,\n                 train_loader, val_loader, out, max_iter, \n                 lr_decay_epoch=None, interval_validate=None):\n    # -----------------------------------------------------------------------------\n        self.cuda = cuda\n\n        self.model = model\n        self.criterion = criterion\n        self.optim = optimizer\n        self.train_loader = train_loader\n        self.val_loader = val_loader\n        self.best_acc = 0\n        self.init_lr = init_lr\n        self.lr_decay_epoch = lr_decay_epoch\n        self.epoch = 0\n        self.iteration = 0\n        self.max_iter = max_iter\n        self.best_loss = 0\n\n        self.timestamp_start = \\\n            datetime.datetime.now(pytz.timezone('US/Eastern'))\n\n        if interval_validate is None:\n            self.interval_validate = len(self.train_loader)\n        else:\n            self.interval_validate = interval_validate\n\n        self.out = out\n        if not osp.exists(self.out):\n            os.makedirs(self.out)\n\n        self.log_headers = [\n            'epoch',\n            'iteration',\n            'train/loss',\n            'train/acc',\n            'valid/loss',\n            'valid/acc',\n            'elapsed_time',\n        ]\n        if not osp.exists(osp.join(self.out, 'log.csv')):\n            with open(osp.join(self.out, 'log.csv'), 'w') as f:\n                f.write(','.join(self.log_headers) + '\\n')\n\n        \n\n\n    # -----------------------------------------------------------------------------\n    def validate(self, max_num=500):\n    # -----------------------------------------------------------------------------\n        training = self.model.module.fc.training\n        self.model.eval()\n\n        MAX_NUM = max_num # HACK: stop after 500 images\n\n        n_class = len(self.val_loader.dataset.classes)\n        val_loss = 0\n        label_trues, label_preds = [], []\n\n        for batch_idx, (data, (target)) in tqdm.tqdm(\n                enumerate(self.val_loader), total=len(self.val_loader),\n                desc='Val=%d' % self.iteration, ncols=80,\n                leave=False):\n\n            # Computing val losses\n            if self.cuda:\n                data, target = data.cuda(), target.cuda()\n\n            data, target = Variable(data), Variable(target)\n            score = self.model(data)\n            loss = self.criterion(score, target)\n\n            if np.isnan(float(loss.data[0])):\n                raise ValueError('loss is NaN while validating')\n\n            val_loss += float(loss.data[0]) / len(data)\n\n            lbl_pred = score.data.max(1)[1].cpu().numpy()\n            lbl_true = target.data.cpu()\n            lbl_pred = lbl_pred.squeeze()\n            lbl_true = np.squeeze(lbl_true.numpy())\n\n            del target, score\n\n            label_trues.append(lbl_true)\n            label_preds.append(lbl_pred)\n\n            del lbl_true, lbl_pred, data, loss\n\n            if batch_idx > MAX_NUM:\n                break\n\n        # Computing metrics\n        val_loss /= len(self.val_loader)\n        val_acc = self.eval_metric(label_trues, label_preds)\n\n        # Logging\n        with open(osp.join(self.out, 'log.csv'), 'a') as f:\n            elapsed_time = (\n                datetime.datetime.now(pytz.timezone('US/Eastern')) -\n                self.timestamp_start).total_seconds()\n            log = [self.epoch, self.iteration] + [''] * 2 + \\\n                  [val_loss, val_acc] + [elapsed_time]\n            log = map(str, log)\n            f.write(','.join(log) + '\\n')\n\n        del label_trues, label_preds\n\n        # Saving the best performing model\n        is_best = val_acc > self.best_acc\n        if is_best:\n            self.best_acc = val_acc\n\n        torch.save({\n            'epoch': self.epoch,\n            'iteration': self.iteration,\n            'arch': self.model.__class__.__name__,\n            'optim_state_dict': self.optim.state_dict(),\n            'model_state_dict': self.model.state_dict(),\n            'best_acc': self.best_acc,\n        }, osp.join(self.out, 'checkpoint.pth.tar'))\n\n        if is_best:\n            shutil.copy(osp.join(self.out, 'checkpoint.pth.tar'),\n                        osp.join(self.out, 'model_best.pth.tar'))\n\n        if training:\n            self.model.module.fc.train()\n\n\n\n    # -----------------------------------------------------------------------------\n    def train_epoch(self):\n    # -----------------------------------------------------------------------------\n        self.model.module.fc.train() # for dataParallel finetuning (TODO - make general)\n        n_class = len(self.train_loader.dataset.classes)\n\n        for batch_idx, (data, target) in tqdm.tqdm(\n                enumerate(self.train_loader), total=len(self.train_loader),\n                desc='Train epoch=%d' % self.epoch, ncols=80, leave=False):\n\n            iteration = batch_idx + self.epoch * len(self.train_loader)\n\n            if self.iteration != 0 and (iteration - 1) != self.iteration:\n                continue  # for resuming\n            self.iteration = iteration\n\n            if self.iteration % self.interval_validate == 0:\n                self.validate()\n\n            assert self.model.module.fc.training\n\n            # Computing Losses\n            if self.cuda:\n                data, target = data.cuda(), target.cuda()\n            data, target = Variable(data), Variable(target)\n            score = self.model(data)  # batch_size x num_class\n\n            loss = self.criterion(score, target)\n\n            if np.isnan(float(loss.data[0])):\n                raise ValueError('loss is NaN while training')\n            # print list(self.model.parameters())[0].grad\n\n            # Gradient descent\n            self.optim.zero_grad()\n            loss.backward()\n            self.optim.step()\n\n            # Computing metrics\n            lbl_pred = score.data.max(1)[1].cpu().numpy()\n            lbl_pred = lbl_pred.squeeze()\n            lbl_true = target.data.cpu()\n            lbl_true = np.squeeze(lbl_true.numpy())\n            train_accu = self.eval_metric([lbl_pred], [lbl_true])\n\n            # Logging\n            with open(osp.join(self.out, 'log.csv'), 'a') as f:\n                elapsed_time = (\n                    datetime.datetime.now(pytz.timezone('US/Eastern')) -\n                    self.timestamp_start).total_seconds()\n                log = [self.epoch, self.iteration] + [loss.data[0]] + \\\n                      [train_accu] + [''] * 2 + [elapsed_time]\n                log = map(str, log)\n                f.write(','.join(log) + '\\n')\n                # print '\\nEpoch: ' + str(self.epoch) + ' Iter: ' + str(self.iteration) + \\\n                #         ' Loss: ' + str(loss.data[0])\n\n            if self.iteration >= self.max_iter:\n                break\n\n\n    # -----------------------------------------------------------------------------\n    def eval_metric(self, lbl_pred, lbl_true):\n    # -----------------------------------------------------------------------------\n        # Over-all accuracy\n        # TODO: per-class accuracy\n        accu = []\n        for lt, lp in zip(lbl_true, lbl_pred):\n            accu.append(np.mean(lt == lp))\n        return np.mean(accu)\n\n\n    # -----------------------------------------------------------------------------\n    def train(self):\n    # -----------------------------------------------------------------------------\n        max_epoch = int(math.ceil(1. * self.max_iter / len(self.train_loader)))\n        print 'Number of iters in an epoch: %d' % len(self.train_loader)\n        print 'Total epochs: %d' % max_epoch        \n\n        for epoch in tqdm.trange(self.epoch, max_epoch,\n                                 desc='Train epochs', ncols=80, leave=True):\n            self.epoch = epoch\n\n            if self.lr_decay_epoch is None:\n                pass\n            else:\n                assert self.lr_decay_epoch < max_epoch\n                lr_scheduler(self.optim, self.epoch, \n                             init_lr=self.init_lr, \n                             lr_decay_epoch=self.lr_decay_epoch)\n\n            self.train_epoch()\n            if self.iteration >= self.max_iter:\n                break\n\n"
  },
  {
    "path": "ijba/README.md",
    "content": "\n\n\n## Resnet-101, scale-224x224  [best scaling strategy]\nTAR @ FAR=0.0001 : 0.5276\nTAR @ FAR=0.0010 : 0.7609\nTAR @ FAR=0.0100 : 0.9148\nTAR @ FAR=0.1000 : 0.9845\n\n## Resnet-101, scale-256, center-crop-224\nTAR @ FAR=0.0001 : 0.3485\nTAR @ FAR=0.0010 : 0.6033\nTAR @ FAR=0.0100 : 0.8752\nTAR @ FAR=0.1000 : 0.9767\n\n## Resnet-101, scale-224, center-crop-224\nTAR @ FAR=0.0001 : 0.4173\nTAR @ FAR=0.0010 : 0.6968\nTAR @ FAR=0.0100 : 0.9129\nTAR @ FAR=0.1000 : 0.9850\n\n---\n\n## Resnet-101-512d-norm, scale-224x224 [cfg-23]\nTAR @ FAR=0.0001 : 0.5790\nTAR @ FAR=0.0010 : 0.7704\nTAR @ FAR=0.0100 : 0.9240\nTAR @ FAR=0.1000 : 0.9884\n\n(epoch3)\nTAR @ FAR=0.0001 : *0.6100*\nTAR @ FAR=0.0010 : *0.7848*\nTAR @ FAR=0.0100 : 0.9262\nTAR @ FAR=0.1000 : 0.9878\n\n( + sqrt)\nTAR @ FAR=0.0001 : 0.6112\nTAR @ FAR=0.0010 : 0.7984\nTAR @ FAR=0.0100 : 0.9251\nTAR @ FAR=0.1000 : 0.9889\n\n( + cosine)\nTAR @ FAR=0.0001 : 0.6100\nTAR @ FAR=0.0010 : 0.7914\nTAR @ FAR=0.0100 : 0.9262\nTAR @ FAR=0.1000 : 0.9878\n\n( + cosine  + sqrt)\nTAR @ FAR=0.0001 : 0.6067\nTAR @ FAR=0.0010 : 0.7987\nTAR @ FAR=0.0100 : 0.9248\nTAR @ FAR=0.1000 : 0.9885\n\n[cfg-24]\n\n\n---\n\n\n## Resnet-101-512d-norm, scale-224x224 [cfg-22, ft stage-2]\nTAR @ FAR=0.0001 : 0.5814\nTAR @ FAR=0.0010 : 0.7651\nTAR @ FAR=0.0100 : 0.9242\nTAR @ FAR=0.1000 : 0.9867\n\n---\n\n## Resnet-101-512d, scale-224x224 [cfg-23, ft stage-2]\nTAR @ FAR=0.0001 : 0.5926\nTAR @ FAR=0.0010 : **0.7919**\nTAR @ FAR=0.0100 : 0.9262\nTAR @ FAR=0.1000 : 0.9872\n\n## Resnet-101-512d, scale-224x224 [cfg-21, ft stage-1]\nTAR @ FAR=0.0001 : 0.5723\nTAR @ FAR=0.0010 : 0.7834\nTAR @ FAR=0.0100 : 0.9240\nTAR @ FAR=0.1000 : 0.9845\n\n\n\n"
  },
  {
    "path": "ijba/eval_ijba_1_1.py",
    "content": "import argparse\nimport os\nimport os.path as osp\n\nimport torch\nimport torchvision\nfrom torchvision import models\nimport torch.nn as nn\nimport torch.utils.data\nimport torchvision.transforms as transforms\nimport torchvision.datasets as datasets\nfrom torch.autograd import Variable\nimport torch.nn.functional as F\n\nimport yaml\nimport tqdm\nimport numpy as np\nimport sklearn.metrics\nfrom sklearn import metrics\nfrom scipy import interpolate\nimport scipy.io as sio\nimport matplotlib\nmatplotlib.use('Agg')\nimport matplotlib.pyplot as plt\n\nhere = osp.dirname(osp.abspath(__file__)) # output folder is located here\nroot_dir,_ = osp.split(here)\nimport sys\nsys.path.append(root_dir)\n\nimport models\nimport utils\nimport data_loader\n\n\n'''\nEvaluate a network on the IJB-A 1:1 verification task\n=====================================================\nExample usage: TODO *** \n# Resnet 101 on 10 folds of IJB-A 1:1\n'''\n# MODEL_PATH = '/srv/data1/arunirc/Research/resnet-face-pytorch/vgg-face-2/logs/MODEL-resnet101_vggface_scratch_CFG-022_TIME-20180210-201442/model_best.pth.tar'\n\n\n# Resnet101-512d-norm \n# MODEL_PATH = '/home/renyi/arunirc/data1/Research/resnet-face-pytorch/vgg-face-2/logs/MODEL-resnet101_512d_L2norm_ft2_CFG-023_TIME-20180214-020054/model_best.pth.tar'\nMODEL_TYPE = 'resnet101-512d-norm'\n# MODEL_PATH = '/home/renyi/arunirc/data1/Research/resnet-face-pytorch/vgg-face-2/logs/MODEL-resnet101_512d_L2norm_ft2_CFG-022_TIME-20180214-015313/model_best.pth.tar'\nMODEL_PATH = '/home/renyi/arunirc/data1/Research/resnet-face-pytorch/vgg-face-2/logs/MODEL-resnet101_512d_L2norm_ft2_CFG-024_TIME-20180214-160410/model_best.pth.tar'\n\n\n# Resnet101-512d\n# MODEL_PATH = '/home/renyi/arunirc/data1/Research/resnet-face-pytorch/vgg-face-2/logs/MODEL-resnet101_bottleneck_ft2_CFG-023_TIME-20180213-091016/model_best.pth.tar'\n# MODEL_TYPE = 'resnet101-512d'\n# MODEL_PATH = '/home/renyi/arunirc/data1/Research/resnet-face-pytorch/vgg-face-2/logs/MODEL-resnet101_bottleneck_ft1_CFG-021_TIME-20180212-192332/model_best.pth.tar'\n\n\n\ndef main():\n    parser = argparse.ArgumentParser()\n    parser.add_argument('-e', '--exp_name', default='ijba_eval')\n    parser.add_argument('-g', '--gpu', type=int, default=0)\n\n    parser.add_argument('-d', '--data_dir', \n                        default='/home/renyi/arunirc/data1/datasets/CS2')\n    parser.add_argument('-p', '--protocol_dir', \n                        default='/home/renyi/arunirc/data1/datasets/IJB-A/IJB-A_11_sets/')\n    parser.add_argument('--fold', type=int, default=1, choices=[1,10])\n    parser.add_argument('--sqrt', action='store_true', default=False,\n                        help='Add signed sqrt normalization')\n    parser.add_argument('--cosine', action='store_true', default=False,\n                        help='Use cosine similarity instead of L2 distance')\n    parser.add_argument('--batch_size', type=int, default=100)\n    parser.add_argument('-m', '--model_path', \n                        default=MODEL_PATH, \n                        help='Path to pre-trained model')\n    parser.add_argument('--model_type', default=MODEL_TYPE,\n                        choices=['resnet50', 'resnet101', 'resnet101-512d', 'resnet101-512d-norm'])\n    \n    args = parser.parse_args()\n\n\n    # CUDA setup\n    os.environ['CUDA_VISIBLE_DEVICES'] = str(args.gpu)\n    cuda = torch.cuda.is_available()\n    torch.manual_seed(1337)\n    if cuda:\n        torch.cuda.manual_seed(1337)\n        torch.backends.cudnn.enabled = True\n        torch.backends.cudnn.benchmark = True # enable if all images are same size    \n\n\n    # -----------------------------------------------------------------------------\n    # 1. Model\n    # -----------------------------------------------------------------------------\n    num_class = 8631\n    if args.model_type == 'resnet50':\n        model = torchvision.models.resnet50(pretrained=False)\n        model.fc = torch.nn.Linear(2048, num_class)\n    elif args.model_type == 'resnet101':\n        model = torchvision.models.resnet101(pretrained=False)\n        model.fc = torch.nn.Linear(2048, num_class)\n    elif args.model_type == 'resnet101-512d':\n        model = torchvision.models.resnet101(pretrained=False)\n        layers = []\n        layers.append(torch.nn.Linear(2048, 512))\n        layers.append(torch.nn.Linear(512, num_class))\n        model.fc = torch.nn.Sequential(*layers)\n    elif args.model_type == 'resnet101-512d-norm':\n        model = torchvision.models.resnet101(pretrained=False)\n        layers = []\n        layers.append(torch.nn.Linear(2048, 512))\n        layers.append(models.NormFeat(scale_factor=50.0))\n        layers.append(torch.nn.Linear(512, num_class))\n        model.fc = torch.nn.Sequential(*layers)\n    else:\n        raise NotImplementedError\n    \n    checkpoint = torch.load(args.model_path)       \n    if checkpoint['arch'] == 'DataParallel':\n        model = torch.nn.DataParallel(model, device_ids=[0, 1, 2, 3, 4])\n        model.load_state_dict(checkpoint['model_state_dict'])\n        model = model.module # get network module from inside its DataParallel wrapper\n    else:\n        model.load_state_dict(checkpoint['model_state_dict'])\n\n    if cuda:\n        model = model.cuda()\n\n    # Convert the trained network into a \"feature extractor\"\n    feature_map = list(model.children())\n    if args.model_type == 'resnet101-512d' or args.model_type == 'resnet101-512d-norm':\n        model.eval()\n        extractor = model\n        extractor.fc = nn.Sequential(extractor.fc[0])\n    else: \n        feature_map.pop()\n        extractor = nn.Sequential(*feature_map)\n    extractor.eval() # ALWAYS set to evaluation mode (fixes BatchNorm, dropout, etc.)\n\n\n\n    # -----------------------------------------------------------------------------\n    # 2. Dataset\n    # -----------------------------------------------------------------------------\n    fold_id = 1\n    file_ext = '.jpg'\n    RGB_MEAN = [ 0.485, 0.456, 0.406 ]\n    RGB_STD = [ 0.229, 0.224, 0.225 ]\n\n    test_transform = transforms.Compose([\n        # transforms.Scale(224),\n        # transforms.CenterCrop(224),\n        transforms.Scale((224,224)),\n        transforms.ToTensor(),\n        transforms.Normalize(mean = RGB_MEAN,\n                             std = RGB_STD),\n    ])\n\n\n    pairs_path = osp.join(args.protocol_dir, 'split%d' % fold_id, \n                          'verify_comparisons_%d.csv' % fold_id)\n    pairs = utils.read_ijba_pairs(pairs_path)\n    protocol_file = osp.join(args.protocol_dir, 'split%d' % fold_id, \n                          'verify_metadata_%d.csv' % fold_id)\n    metadata = utils.get_ijba_1_1_metadata(protocol_file) # dict\n    assert np.all(np.unique(pairs) == np.unique(metadata['template_id']))  # sanity-check\n    path_list = np.array([osp.join(args.data_dir, str(x)+file_ext) \n                         for x in metadata['sighting_id'] ]) # face crops saved as <sighting_id.jpg>\n\n    # Create data loader\n    test_loader = torch.utils.data.DataLoader(\n                        data_loader.IJBADataset(\n                        path_list, test_transform, split=fold_id), \n                        batch_size=args.batch_size, shuffle=False )\n\n    # testing\n    # for i in range(len(test_loader.dataset)):\n    #     img = test_loader.dataset.__getitem__(i)\n    #     sz = img.shape\n    #     if sz[0] != 3:\n    #         print sz\n\n\n\n\n    # -----------------------------------------------------------------------------\n    # 3. Feature extraction\n    # -----------------------------------------------------------------------------\n    print 'Feature extraction...'\n    cache_dir = osp.join(here, 'cache-' + args.model_type)\n    if not osp.exists(cache_dir):\n        os.makedirs(cache_dir)\n\n    feat_path = osp.join(cache_dir, 'feat-fold-%d.mat' % fold_id)\n\n    if not osp.exists(feat_path):\n        features = []\n        for batch_idx, images in tqdm.tqdm(enumerate(test_loader), \n                                            total=len(test_loader), \n                                            desc='Extracting features'): \n            x = Variable(images, volatile=True) # test-time memory conservation\n            if cuda:\n                x = x.cuda()\n            feat = extractor(x)\n            if cuda:\n                feat = feat.data.cpu() # free up GPU\n            else:\n                feat = feat.data\n            features.append(feat)\n\n        features = torch.cat(features, dim=0) # (n_batch*batch_sz) x 512\n        sio.savemat(feat_path, {'feat': features.cpu().numpy() })\n    else:\n        dat = sio.loadmat(feat_path)\n        features = torch.FloatTensor(dat['feat'])\n        del dat\n        print 'Loaded.'\n\n\n    # -----------------------------------------------------------------------------\n    # 4. Verification\n    # -----------------------------------------------------------------------------\n    scores = []\n    labels = []\n\n    # labels: is_same_subject\n    print 'Computing pair labels . . . '\n    for pair in tqdm.tqdm(pairs): # TODO - check tqdm\n        sel_t0 = np.where(metadata['template_id'] == pair[0])\n        sel_t1 = np.where(metadata['template_id'] == pair[1])\n        subject0 = np.unique(metadata['subject_id'][sel_t0])\n        subject1 = np.unique(metadata['subject_id'][sel_t1])\n        labels.append(int(subject0 == subject1))\n    labels = np.array(labels)\n    print 'done'\n\n    # templates: average pool, then L2-normalize\n    print 'Pooling templates . . . '\n    pooled_features = []\n    template_set = np.unique(metadata['template_id'])\n    for tid in tqdm.tqdm(template_set):\n        sel = np.where(metadata['template_id'] == tid)\n        # pool template: 1 x n x 512 -> 1 x 512\n        feat = features[sel,:].mean(1)\n        if args.sqrt:  # signed-square-root normalization\n            feat = torch.mul(torch.sign(feat),torch.sqrt(torch.abs(feat)+1e-12))\n        pooled_features.append(F.normalize(feat, p=2, dim=1) )    \n    pooled_features = torch.cat(pooled_features, dim=0) # (n_batch*batch_sz) x 512\n    print 'done'\n\n    print 'Computing pair distances . . . '\n    for pair in tqdm.tqdm(pairs):\n        sel_t0 = np.where(template_set == pair[0])\n        sel_t1 = np.where(template_set == pair[1])\n        if args.cosine:\n            feat_dist = torch.dot(torch.squeeze(pooled_features[sel_t0]), \n                                  torch.squeeze(pooled_features[sel_t1]))\n        else:\n            feat_dist = (pooled_features[sel_t0] - pooled_features[sel_t1]).norm(p=2, dim=1)\n            feat_dist = -torch.squeeze(feat_dist)\n            feat_dist = feat_dist.numpy()\n        scores.append(feat_dist) # score: negative of L2-distance\n    scores = np.array(scores)\n\n    # Metrics: TAR (tpr) at FAR (fpr)\n    fpr, tpr, thresholds = sklearn.metrics.roc_curve(labels, scores)\n    fpr_levels = [0.0001, 0.001, 0.01, 0.1]\n    f_interp = interpolate.interp1d(fpr, tpr)\n    tpr_at_fpr = [ f_interp(x) for x in fpr_levels ]\n\n    for (far, tar) in zip(fpr_levels, tpr_at_fpr):\n        print 'TAR @ FAR=%.4f : %.4f' % (far, tar)\n\n    res = {}\n    res['TAR'] = tpr_at_fpr\n    res['FAR'] = fpr_levels\n    with open( osp.join(cache_dir, 'result-1-1-fold-%d.yaml' % fold_id), \n              'w') as f:\n        yaml.dump(res, f, default_flow_style=False)\n\n    sio.savemat(osp.join(cache_dir, 'roc-1-1-fold-%d.mat' % fold_id), \n                {'fpr': fpr, 'tpr': tpr, 'thresholds': thresholds, \n                    'tpr_at_fpr': tpr_at_fpr})\n\n\nif __name__ == '__main__':\n    main()\n\n"
  },
  {
    "path": "lfw/README.md",
    "content": "## LFW setup\n\nThis README shows how to set up the downloaded face images from the LFW dataset for face verification evaluation. \n\nDownload the deep-funneled (roughly-aligned) images from LFW: `http://vis-www.cs.umass.edu/lfw/lfw-deepfunneled.tgz`. Extract these at some local folder, referred to henceforth as `LFW_DIR`.\n\nNOTE: all scripts are to be executed from the project root directory, *not* the sub-folder where this README file is located.\n\n### Validation results\n\nThe LFW database provides DevTrain and DevTest splits as validation sets for developing code without overfitting to the 10 folds used in the final evaluation. The DevTest pairs are saved at `./lfw/data/pairsDevTest.txt`\n\nRun `python ./lfw/eval_lfw.py -d LFW_DIR -m MODEL_PATH --fold 0` to get AUC score for verification on the dev test split. The number should be around 0.9989. Setting the `--fold` flag to 10 will evaluate on the full 10 folds of LFW test set. Note, ROC curves and AUC metrics are reported (not EER).\n\n"
  },
  {
    "path": "lfw/data/pairs.txt",
    "content": "10\t300\nAbel_Pacheco\t1\t4\nAkhmed_Zakayev\t1\t3\nAkhmed_Zakayev\t2\t3\nAmber_Tamblyn\t1\t2\nAnders_Fogh_Rasmussen\t1\t3\nAnders_Fogh_Rasmussen\t1\t4\nAngela_Bassett\t1\t5\nAngela_Bassett\t2\t5\nAngela_Bassett\t3\t4\nAnn_Veneman\t3\t5\nAnn_Veneman\t6\t10\nAnn_Veneman\t10\t11\nAnthony_Fauci\t1\t2\nAntony_Leung\t1\t2\nAntony_Leung\t2\t3\nAnwar_Ibrahim\t1\t2\nAugusto_Pinochet\t1\t2\nBarbara_Brezigar\t1\t2\nBenjamin_Netanyahu\t1\t4\nBenjamin_Netanyahu\t4\t5\nBernard_Law\t2\t3\nBernard_Law\t3\t4\nBertrand_Bonello\t1\t2\nBill_Frist\t2\t9\nBill_Frist\t4\t5\nBob_Graham\t2\t4\nBob_Graham\t3\t6\nBob_Graham\t4\t6\nBob_Graham\t5\t6\nBoris_Becker\t2\t6\nBrad_Johnson\t1\t3\nBrad_Johnson\t2\t3\nBrad_Johnson\t2\t4\nBrian_Griese\t1\t2\nCandice_Bergen\t1\t2\nCandice_Bergen\t2\t3\nCarla_Myers\t1\t2\nCathy_Freeman\t1\t2\nChang_Dae-whan\t1\t2\nCharles_Grassley\t1\t2\nClare_Short\t1\t3\nClare_Short\t1\t4\nCorinne_Coman\t1\t2\nDai_Bachtiar\t1\t2\nDale_Earnhardt_Jr\t1\t3\nDavid_Caruso\t1\t2\nDemi_Moore\t1\t3\nDick_Vermeil\t1\t2\nDoris_Roberts\t1\t2\nDoris_Roberts\t1\t3\nDrew_Barrymore\t1\t2\nEdmund_Stoiber\t4\t7\nEdmund_Stoiber\t5\t7\nEdmund_Stoiber\t7\t12\nEdmund_Stoiber\t9\t11\nElinor_Caplan\t1\t2\nEmmanuelle_Beart\t1\t2\nEmmanuelle_Beart\t1\t3\nFederico_Trillo\t1\t2\nFederico_Trillo\t1\t3\nFederico_Trillo\t2\t3\nFrancis_Ford_Coppola\t1\t2\nFred_Thompson\t1\t2\nFred_Thompson\t1\t3\nFred_Thompson\t2\t3\nGarry_Trudeau\t1\t2\nGary_Williams\t1\t2\nGene_Robinson\t1\t2\nGene_Robinson\t1\t3\nGene_Robinson\t2\t3\nGeorge_Galloway\t1\t3\nGeorge_Galloway\t2\t3\nGeorge_Galloway\t2\t4\nGeraldine_Chaplin\t1\t2\nGeraldine_Chaplin\t1\t4\nGeraldine_Chaplin\t3\t4\nGerardo_Gambala\t1\t2\nGian_Marco\t1\t2\nGian_Marco\t2\t3\nGillian_Anderson\t1\t2\nGordon_Brown\t6\t9\nGordon_Brown\t10\t13\nGrant_Hackett\t1\t2\nGrant_Hackett\t1\t3\nGrant_Hackett\t1\t5\nGray_Davis\t7\t22\nGray_Davis\t21\t26\nGregg_Popovich\t1\t5\nGregg_Popovich\t3\t4\nGuy_Ritchie\t1\t2\nHamzah_Haz\t1\t2\nHarbhajan_Singh\t1\t2\nHootie_Johnson\t1\t2\nHugo_Chavez\t2\t27\nHugo_Chavez\t27\t53\nHugo_Chavez\t36\t51\nHugo_Chavez\t41\t50\nHugo_Chavez\t45\t56\nIsaiah_Washington\t1\t2\nJack_Grubman\t1\t2\nJacques_Rogge\t1\t9\nJacques_Rogge\t3\t7\nJacques_Rogge\t3\t10\nJacques_Rogge\t4\t6\nJacques_Rogge\t4\t9\nJacques_Rogge\t5\t8\nJacques_Rogge\t6\t8\nJames_Caan\t1\t2\nJames_Caan\t1\t3\nJames_Caan\t2\t3\nJames_Jones\t1\t2\nJamling_Norgay\t1\t2\nJanica_Kostelic\t1\t2\nJeanne_Moreau\t1\t2\nJeffrey_Archer\t1\t2\nJennifer_Garner\t1\t3\nJennifer_Garner\t4\t9\nJennifer_Garner\t5\t6\nJennifer_Garner\t5\t9\nJennifer_Garner\t7\t12\nJennifer_Garner\t8\t11\nJeremy_Greenstock\t2\t24\nJeremy_Greenstock\t3\t22\nJeremy_Greenstock\t5\t11\nJeremy_Greenstock\t10\t14\nJeremy_Greenstock\t17\t21\nJessica_Lange\t1\t2\nJimmy_Carter\t1\t7\nJimmy_Carter\t4\t6\nJimmy_Carter\t5\t8\nJimmy_Carter\t6\t7\nJodie_Foster\t1\t2\nJohn_Manley\t1\t6\nJohn_Manley\t2\t5\nJohn_McCain\t2\t3\nJohn_McCain\t3\t7\nJohn_McCain\t6\t7\nJohn_Snow\t3\t15\nJohn_Snow\t4\t16\nJohn_Snow\t9\t11\nJohnson_Panjaitan\t1\t2\nJorge_Castaneda\t1\t2\nJose_Serra\t1\t9\nJose_Serra\t2\t8\nJose_Serra\t3\t5\nJose_Serra\t3\t7\nJose_Serra\t6\t9\nJose_Theodore\t1\t2\nJoseph_Blatter\t1\t2\nJoseph_Ralston\t1\t2\nJuan_Pablo_Montoya\t1\t8\nJuan_Pablo_Montoya\t2\t4\nJuan_Pablo_Montoya\t4\t5\nJulianna_Margulies\t1\t2\nJustin_Guarini\t1\t2\nJustin_Guarini\t2\t3\nJustine_Henin\t2\t3\nKate_Winslet\t1\t2\nKate_Winslet\t1\t4\nKate_Winslet\t2\t3\nKate_Winslet\t2\t4\nKathy_Winters\t1\t2\nKelvin_Sampson\t1\t2\nKevin_Stallings\t1\t2\nKim_Jong-Il\t2\t3\nKristin_Davis\t1\t2\nKristin_Davis\t2\t3\nLance_Armstrong\t1\t17\nLance_Armstrong\t5\t10\nLance_Armstrong\t7\t11\nLance_Armstrong\t10\t18\nLance_Armstrong\t14\t17\nLaurent_Jalabert\t1\t2\nLindsey_Graham\t1\t2\nLisa_Gottsegen\t1\t2\nLleyton_Hewitt\t1\t13\nLleyton_Hewitt\t16\t27\nLleyton_Hewitt\t18\t37\nLleyton_Hewitt\t24\t36\nLuis_Ernesto_Derbez_Bautista\t3\t6\nMario_Dumont\t1\t2\nMark_Wahlberg\t1\t3\nMark_Wahlberg\t2\t3\nMark_Wahlberg\t2\t4\nMarlene_Weingartner\t1\t2\nMartha_Bowen\t1\t2\nMartin_Cauchon\t1\t2\nMartin_Hoellwarth\t1\t2\nMartin_Sheen\t1\t2\nMatthew_Broderick\t1\t4\nMatthew_Broderick\t2\t3\nMike_Krzyzewski\t1\t6\nMikhail_Gorbachev\t1\t2\nMonica_Bellucci\t1\t4\nMonica_Bellucci\t2\t4\nNaomi_Campbell\t1\t2\nNestor_Kirchner\t3\t30\nNestor_Kirchner\t12\t21\nNestor_Kirchner\t17\t25\nNicolas_Lapentti\t1\t2\nNoah_Wyle\t2\t3\nNora_Bendijo\t1\t2\nNursultan_Nazarbayev\t1\t2\nPamela_Anderson\t1\t2\nPamela_Anderson\t1\t5\nPatricia_Heaton\t1\t2\nPatty_Schnyder\t1\t2\nPatty_Schnyder\t2\t3\nPatty_Schnyder\t2\t4\nPatty_Schnyder\t3\t4\nPaul_Bremer\t6\t17\nPaul_Bremer\t7\t10\nPaul_Bremer\t9\t15\nPaul_Bremer\t14\t16\nPaul_Sarbanes\t1\t3\nPaul_Sarbanes\t2\t3\nPaul_William_Hurley\t1\t2\nPervez_Musharraf\t1\t6\nPervez_Musharraf\t2\t16\nPervez_Musharraf\t3\t10\nPervez_Musharraf\t4\t14\nPervez_Musharraf\t5\t7\nPervez_Musharraf\t5\t15\nPhil_Gramm\t1\t2\nPrince_Edward\t1\t2\nRicardo_Lagos\t3\t12\nRicardo_Lagos\t23\t27\nRicardo_Monasterio\t1\t2\nRich_Gannon\t1\t2\nRick_Perry\t1\t4\nRick_Perry\t1\t5\nRick_Perry\t2\t4\nRick_Perry\t3\t4\nRobert_Bonner\t2\t3\nRobert_Evans\t2\t3\nRobert_Fico\t1\t2\nRomano_Prodi\t3\t5\nRoy_Moore\t2\t5\nRoy_Moore\t3\t6\nSachiko_Yamada\t1\t2\nSachiko_Yamada\t1\t4\nSachiko_Yamada\t2\t3\nSachiko_Yamada\t2\t4\nSaeb_Erekat\t1\t2\nSam_Bith\t1\t2\nSam_Bith\t1\t3\nSam_Bith\t2\t3\nSean_Hayes\t1\t2\nSergio_Garcia\t1\t2\nSilvia_Farina_Elia\t1\t2\nSteve_Ballmer\t1\t2\nSteve_Ballmer\t1\t3\nStockard_Channing\t1\t3\nStockard_Channing\t2\t3\nTerry_McAuliffe\t1\t2\nTerry_McAuliffe\t1\t3\nThomas_Rupprath\t1\t3\nThomas_Rupprath\t2\t3\nTom_Ridge\t6\t16\nTom_Ridge\t8\t25\nTom_Ridge\t9\t26\nTom_Ridge\t18\t22\nTommy_Haas\t4\t5\nTommy_Thompson\t1\t2\nTommy_Thompson\t1\t7\nTommy_Thompson\t2\t8\nTommy_Thompson\t4\t8\nTommy_Thompson\t6\t9\nTomoko_Hagiwara\t1\t2\nTrent_Lott\t1\t2\nTrent_Lott\t2\t3\nTrent_Lott\t2\t8\nTrent_Lott\t6\t11\nTrent_Lott\t7\t10\nTrent_Lott\t8\t16\nTsutomu_Takebe\t1\t2\nTyler_Hamilton\t1\t2\nTyra_Banks\t1\t2\nVaclav_Havel\t2\t4\nVaclav_Havel\t2\t5\nVaclav_Havel\t2\t6\nVaclav_Havel\t4\t9\nVaclav_Havel\t5\t7\nValerie_Harper\t1\t2\nVince_Carter\t3\t4\nVincent_Brooks\t2\t3\nVincent_Brooks\t2\t7\nVincent_Brooks\t4\t5\nVincent_Brooks\t4\t7\nVincent_Gallo\t1\t2\nVitali_Klitschko\t1\t2\nVitali_Klitschko\t3\t4\nWilliam_Macy\t1\t4\nWilliam_Macy\t2\t4\nWilliam_Macy\t3\t4\nWoody_Allen\t2\t4\nWoody_Allen\t3\t5\nYukiko_Okudo\t1\t2\nZico\t1\t2\nZico\t2\t3\nAbdel_Madi_Shabneh\t1\tDean_Barker\t1\nAbdel_Madi_Shabneh\t1\tGiancarlo_Fisichella\t1\nAbdel_Madi_Shabneh\t1\tMikhail_Gorbachev\t1\nAbdul_Rahman\t1\tPortia_de_Rossi\t1\nAbel_Pacheco\t1\tJong_Thae_Hwa\t2\nAbel_Pacheco\t2\tJean-Francois_Lemounier\t1\nAfton_Smith\t1\tDwayne_Wade\t1\nAhmad_Jbarah\t1\tJames_Comey\t1\nAkhmed_Zakayev\t2\tDonna_Morrissey\t1\nAlan_Dershowitz\t1\tBertrand_Bonello\t1\nAlanis_Morissette\t1\tMartin_Cauchon\t1\nAlexander_Lukashenko\t1\tHeather_Chinnock\t1\nAlfonso_Cuaron\t1\tJason_Priestley\t1\nAlfonso_Cuaron\t1\tPatty_Schnyder\t2\nAlfonso_Soriano\t1\tBill_Nelson\t2\nAlfonso_Soriano\t1\tJulio_De_Brun\t1\nAlfonso_Soriano\t1\tPatty_Schnyder\t3\nAlonzo_Mourning\t1\tCecilia_Cheung\t1\nAmber_Tamblyn\t2\tBenjamin_Netanyahu\t1\nAmporn_Falise\t1\tJoe_Pantoliano\t1\nAnders_Fogh_Rasmussen\t2\tJohnson_Panjaitan\t2\nAndre_Bucher\t1\tJoseph_Ralston\t1\nAndre_Bucher\t1\tMaria_Garcia\t1\nAndrew_Gilligan\t1\tHenry_Castellanos\t1\nAndrew_Shutley\t1\tEdmund_Stoiber\t5\nAndrew_Shutley\t1\tMitchell_Swartz\t1\nAndrew_Shutley\t1\tSaeb_Erekat\t2\nAndy_Dick\t1\tSimon_Yam\t1\nAndy_Griffith\t1\tOsrat_Iosef\t1\nAndy_Wisecarver\t1\tDimitar_Berbatov\t1\nAngela_Lansbury\t1\tSteven_Van_Zandt\t1\nAngela_Lansbury\t2\tJohn_Coomber\t1\nAnn_Veneman\t6\tSergio_Garcia\t2\nAnn_Veneman\t8\tTed_Williams\t1\nAnnie-Jeanne_Reynaud\t1\tSJ_Twu\t1\nAnthony_Carter\t1\tEliza_Dushku\t1\nAntonio_Cassano\t1\tPaul_Celluci\t1\nAnwar_Ibrahim\t1\tDavid_Alpay\t1\nArmand_Sargen\t1\tDaryl_Sabara\t1\nArmand_Sargen\t1\tKelvin_Sampson\t3\nArmand_Sargen\t1\tLisa_Gottsegen\t2\nAtom_Egoyan\t1\tBill_Stapleton\t1\nAtom_Egoyan\t1\tJanis_Ruth_Coulter\t1\nBarbara_Brezigar\t2\tDoris_Roberts\t2\nBarbara_Felt-Miller\t1\tLeticia_Dolera\t1\nBart_Freundlich\t1\tErnie_Grunfeld\t1\nBart_Freundlich\t1\tKirsten_Gilham\t1\nBenjamin_Martinez\t1\tGarry_McCoy\t1\nBenjamin_Netanyahu\t1\tMaria_Callas\t1\nBenjamin_Netanyahu\t5\tFrank_Beamer\t1\nBernard_Law\t1\tLiu_Xiaoqing\t1\nBernard_Law\t3\tValerie_Harper\t2\nBertrand_Bonello\t1\tJong_Thae_Hwa\t2\nBill_Bradley\t1\tChen_Tsai-chin\t1\nBill_Bradley\t1\tHelen_Alvare\t1\nBill_Elliott\t1\tMary_Anne_Souza\t1\nBill_Frist\t5\tJimmy_Kimmel\t2\nBill_Maher\t1\tBrad_Russ\t1\nBill_Maher\t1\tJuliette_Binoche\t1\nBill_Nelson\t1\tKim_Jong-Il\t3\nBill_Nelson\t2\tGillian_Anderson\t1\nBill_Stapleton\t1\tSedigh_Barmak\t1\nBill_Stapleton\t1\tValerie_Harper\t1\nBilly_Bob_Thornton\t1\tHerb_Dhaliwal\t1\nBilly_Bob_Thornton\t1\tNong_Duc_Manh\t1\nBob_Alper\t1\tKevin_Millwood\t1\nBob_Graham\t2\tDwayne_Wade\t1\nBob_Petrino\t1\tGeraldine_Chaplin\t4\nBob_Petrino\t1\tJorge_Castaneda\t1\nBoris_Becker\t4\tJulianna_Margulies\t2\nBrad_Russ\t1\tHana_Urushima\t1\nBrad_Russ\t1\tRomeo_Gigli\t1\nBrawley_King\t1\tTom_Glavine\t2\nBrian_Griese\t2\tJeffrey_Archer\t2\nBrian_Griese\t2\tLaura_Elena_Harring\t1\nBrian_Griese\t2\tNicolas_Lapentti\t2\nBryan_Adams\t1\tMichael_Kors\t1\nBryan_Adams\t1\tMohamed_Seineldin\t1\nCalbert_Cheaney\t1\tIan_Smith\t1\nCalbert_Cheaney\t1\tRobert_Downey_Jr\t1\nCarl_Reiner\t1\tHamid_Efendi\t1\nCarl_Reiner\t2\tJohn_Engler\t1\nCarl_Reiner\t2\tPrince_Rainier_III\t1\nCarl_Reiner\t2\tTom_Glavine\t2\nCarlo_Azeglio_Ciampi\t1\tFrancis_Ford_Coppola\t1\nCarlos_Arroyo\t1\tShane_Phillips\t1\nCarlos_Paternina\t1\tEmily_Stevens\t1\nCarlos_Paternina\t1\tPaul_Sarbanes\t1\nCasey_Mears\t1\tMike_Davis\t1\nCasey_Mears\t1\tYukiko_Okudo\t1\nCathy_Freeman\t2\tWilliam_Martin\t2\nCecilia_Cheung\t1\tDaryl_Parks\t1\nCecilia_Cheung\t1\tPascal_Affi_Nguessan\t1\nChen_Tsai-chin\t1\tDereck_Whittenburg\t1\nChen_Tsai-chin\t1\tMamdouh_Habib\t1\nCho_Myung-kyun\t1\tDavid_Bell\t1\nCho_Myung-kyun\t1\tFernando_Sanz\t1\nCho_Myung-kyun\t1\tGeorgia_Giddings\t1\nCho_Myung-kyun\t1\tRichard_Fine\t1\nChoi_Yun-yong\t1\tChuck_Eidson\t1\nChris_Dodd\t1\tTaylor_Twellman\t1\nChris_Swecker\t1\tTom_Vilsack\t1\nChristian_Lacroix\t1\tLaura_Elena_Harring\t1\nChristian_Lacroix\t1\tOrnella_Muti\t1\nChuck_Eidson\t1\tSigourney_Weaver\t1\nClare_Short\t4\tDon_Carcieri\t1\nCoco_dEste\t1\tDarvis_Patton\t1\nCoco_dEste\t1\tMelina_Kanakaredes\t1\nCoco_dEste\t1\tTom_Rouen\t1\nColeen_Rowley\t1\tNong_Duc_Manh\t1\nCorinne_Coman\t2\tFrank_Beamer\t1\nDale_Earnhardt_Jr\t1\tNick_Reilly\t1\nDario_Franchitti\t1\tHenry_Castellanos\t1\nDarren_Campel\t1\tHilary_McKay\t1\nDarvis_Patton\t1\tGerard_Tronche\t1\nDarvis_Patton\t1\tWilliam_Macy\t4\nDaryl_Parks\t1\tGuus_Hiddink\t1\nDaryl_Sabara\t1\tNick_Reilly\t1\nDaryl_Sabara\t1\tValentina_Tereshkova\t1\nDave_Johnson\t1\tHoward_Stern\t1\nDave_Tucker\t1\tGary_Gitnick\t1\nDavid_Collenette\t1\tSalman_Khan\t1\nDavid_Westerfield\t1\tStan_Kroenke\t1\nDean_Jacek\t1\tLarry_Wilmore\t1\nDemi_Moore\t2\tFred_Thompson\t1\nDemi_Moore\t2\tLinus_Roache\t1\nDereck_Whittenburg\t1\tLindsey_Graham\t2\nDianne_Reeves\t1\tLarry_Wilmore\t1\nDianne_Reeves\t1\tRomeo_Gigli\t1\nDon_Carcieri\t1\tJanica_Kostelic\t1\nDora_Bakoyianni\t1\tRichard_Sambrook\t2\nDora_Bakoyianni\t1\tSaeb_Erekat\t2\nDoris_Roberts\t1\tNong_Duc_Manh\t1\nDoug_Wilson\t1\tSzu_Yu_Chen\t1\nDouglas_Gansler\t1\tMartin_Brooke\t1\nDouglas_Gansler\t1\tRonald_Kadish\t1\nDwayne_Wade\t1\tMike_Farrar\t1\nEdward_Arsenault\t1\tJim_Hardin\t1\nEinars_Repse\t1\tMinnie_Mendoza\t1\nEinars_Repse\t1\tTim_Blake_Nelson\t1\nElinor_Caplan\t1\tHilary_McKay\t1\nEliza_Dushku\t1\tGeorge_Lucas\t1\nEliza_Dushku\t1\tItzhak_Perlman\t1\nEmily_Stevens\t1\tJanez_Drnovsek\t1\nEmmanuelle_Beart\t2\tPhil_Jackson\t1\nEric_Daze\t1\tSterling_Hitchcock\t1\nErika_Christensen\t2\tMichael_Dell\t1\nErika_Christensen\t2\tWoody_Allen\t2\nEriko_Tamura\t1\tGeorgia_Giddings\t1\nErnie_Grunfeld\t1\tFrank_Coraci\t1\nEugene_Melnyk\t1\tMahima_Chaudhari\t1\nFatma_Kusibeh\t1\tLee_Baca\t1\nFederico_Trillo\t1\tJonathan_Woodgate\t1\nFernando_Alonso\t1\tSam_Brownback\t1\nFernando_Sanz\t1\tMiranda_Otto\t1\nFernando_Sanz\t1\tRoy_Moore\t1\nFlor_Montulo\t2\tJuan_Pablo_Montoya\t1\nFrancisco_Garcia\t1\tMarsha_Sharp\t1\nFrancois_Ozon\t1\tMakiya_Ali_Hassan\t1\nFrank_Coraci\t1\tTomoko_Hagiwara\t2\nFrank_Van_Ecke\t1\tTsutomu_Takebe\t1\nFred_Thompson\t2\tHelen_Alvare\t1\nFred_Thompson\t2\tSterling_Hitchcock\t1\nFred_Thompson\t3\tMagda_Kertasz\t1\nGarry_Trudeau\t1\tPat_Riley\t1\nGarry_Witherall\t1\tHoward_Stern\t1\nGarry_Witherall\t1\tIngrid_Betancourt\t1\nGarry_Witherall\t1\tMartin_Keown\t1\nGary_Gero\t1\tKim_Hong-gul\t1\nGary_Gero\t1\tPhil_Gramm\t1\nGavin_Degraw\t1\tJeffrey_Archer\t1\nGene_Robinson\t3\tMartha_Bowen\t2\nGeorgia_Giddings\t1\tMahima_Chaudhari\t1\nGeovani_Lapentti\t1\tRodney_Rempt\t1\nGeovani_Lapentti\t1\tSam_Brownback\t1\nGerard_de_Cortanze\t1\tMark_Wahlberg\t1\nGian_Marco\t2\tKevin_Stallings\t2\nGiancarlo_Fisichella\t1\tMaria_Callas\t1\nGideon_Yago\t1\tNatalie_Williams\t1\nGideon_Yago\t1\tPaul_William_Hurley\t1\nGlenn_Plummer\t1\tMaria_Garcia\t1\nGrant_Hackett\t1\tTodd_Robbins\t1\nGrant_Hackett\t3\tMilo_Djukanovic\t3\nGray_Davis\t26\tKaren_Lynn_Gorney\t1\nGregg_Popovich\t3\tVernon_Forrest\t1\nGregor_Gysi\t1\tTomoko_Hagiwara\t1\nGuy_Ritchie\t2\tHerb_Dhaliwal\t1\nGuy_Ritchie\t2\tWilliam_Macy\t1\nHamid_Efendi\t1\tJimmy_Carter\t8\nHamzah_Haz\t2\tHilary_McKay\t1\nHarald_Ringstorff\t1\tPat_Riley\t1\nHarald_Ringstorff\t1\tRomano_Prodi\t6\nHeather_Chinnock\t1\tJean-Francois_Lemounier\t1\nHelen_Alvare\t1\tMilo_Djukanovic\t1\nHenry_Castellanos\t1\tPamela_Anderson\t4\nHenry_Castellanos\t1\tTommy_Shane_Steiner\t1\nHerb_Dhaliwal\t1\tHung_Wan-ting\t1\nHilary_McKay\t1\tKevin_Millwood\t1\nHoward_Stern\t1\tMaria_Callas\t1\nHugo_Chavez\t33\tKaren_Lynn_Gorney\t1\nHugo_Chavez\t60\tSteve_Shiver\t1\nImam_Samudra\t1\tIvana_Trump\t1\nImelda_Marcos\t1\tPatty_Schnyder\t4\nJack_Smith\t1\tMary_Jo_Myers\t1\nJames_Caan\t2\tPaul_Sarbanes\t2\nJames_Comey\t1\tJuan_Carlos_Morales\t1\nJames_Comey\t1\tPaul_William_Hurley\t1\nJamling_Norgay\t1\tZico\t2\nJan_Pronk\t1\tKim_Dong-hwa\t1\nJanez_Drnovsek\t1\tSterling_Hitchcock\t1\nJanica_Kostelic\t2\tYasushi_Akashi\t1\nJanice_Abreu\t1\tKevin_Sorbo\t1\nJeffrey_Ashby\t1\tMichael_Dell\t1\nJennifer_Garner\t6\tMike_Duke\t1\nJennifer_Renee_Short\t1\tTaylor_Twellman\t1\nJerry_Seinfeld\t1\tTim_Blake_Nelson\t1\nJerry_Tarkanian\t1\tThomas_Rupprath\t1\nJessica_Lange\t2\tSedigh_Barmak\t1\nJim_Freudenberg\t1\tNigel_Redden\t1\nJim_Freudenberg\t1\tTina_Pisnik\t1\nJim_Haslett\t1\tTsutomu_Takebe\t1\nJim_Otto\t1\tRafiq_Hariri\t1\nJimmy_Gurule\t1\tTerry_McAuliffe\t1\nJodie_Foster\t3\tJoe_Pantoliano\t1\nJohn_Herrington\t1\tLuis_Ernesto_Derbez_Bautista\t2\nJohn_Richardson\t1\tYasushi_Akashi\t1\nJohn_Snow\t17\tSe_Hyuk_Joo\t1\nJonathan_Arden\t1\tJoseph_Ralston\t1\nJorge_Castaneda\t1\tRobert_Fico\t1\nJose_Rosado\t1\tMicky_Arison\t1\nJoseph_Blatter\t1\tRonald_Kadish\t1\nJoseph_Ralston\t2\tJuan_Pablo_Montoya\t5\nJoseph_Ralston\t2\tYoshiyuki_Kamei\t1\nJuliette_Binoche\t1\tMatthew_Broderick\t3\nJulio_De_Brun\t1\tPatty_Schnyder\t1\nJulio_De_Brun\t1\tVernon_Forrest\t1\nJustin_Guarini\t2\tPrince_Edward\t1\nKate_Winslet\t2\tMike_Duke\t1\nKatie_Wagner\t1\tStan_Kroenke\t1\nKeith_Lowen\t1\tRobert_Evans\t2\nKeith_Lowen\t1\tSilvia_Farina_Elia\t2\nKen_Loach\t1\tTaku_Yamasaki\t1\nKevin_Crane\t1\tMike_Krzyzewski\t2\nKevin_Millwood\t1\tMitchell_Crooks\t1\nKim_Clijsters\t4\tMartin_Short\t1\nKim_Hong-gul\t1\tMilo_Djukanovic\t3\nKim_Jong-Il\t3\tRick_Reed\t1\nLance_Armstrong\t9\tMaria_Garcia\t1\nLaurent_Jalabert\t2\tVincent_Gallo\t1\nLeon_LaPorte\t1\tTed_Williams\t1\nLeon_LaPorte\t1\tTommy_Thompson\t5\nLeonardo_Fernandez\t1\tRomano_Prodi\t7\nLeticia_Dolera\t1\tTom_Glavine\t1\nLorraine_Bracco\t1\tMomcilo_Perisic\t1\nLuis_Ernesto_Derbez_Bautista\t1\tTyra_Banks\t1\nMaria_Burks\t1\tTodd_Parrott\t1\nMario_Lemieux\t1\tStan_Kroenke\t1\nMark_Everson\t1\tMartin_Sheen\t2\nMarquier_Montano_Contreras\t1\tSJ_Twu\t1\nMarsha_Sharp\t1\tSteve_Shiver\t1\nMartin_Cauchon\t2\tVitali_Klitschko\t3\nMartin_Hoellwarth\t2\tMary_Katherine_Smart\t1\nMartina_Hingis\t1\tTerry_McAuliffe\t2\nMelina_Kanakaredes\t1\tOrnella_Muti\t1\nMichael_Dell\t1\tMike_Duke\t1\nMichael_Dell\t1\tNigel_Redden\t1\nMichael_Richards\t1\tSilvia_Farina_Elia\t3\nMilan_Kucan\t1\tSalman_Khan\t1\nNancy_Kerrigan\t1\tSam_Brownback\t1\nNaomi_Campbell\t1\tTom_Ridge\t16\nNina_Jacobson\t1\tPortia_de_Rossi\t1\nNoah_Wyle\t3\tRobbie_Coltrane\t1\nNora_Bendijo\t1\tWilliam_Martin\t2\nNursultan_Nazarbayev\t1\tRobert_Bonner\t1\nPascal_Affi_Nguessan\t1\tTom_Moss\t1\nPat_Summitt\t1\tPaul_Celluci\t1\nPatty_Schnyder\t3\tPernilla_Bjorn\t1\nPatty_Schnyder\t3\tPrince_Philippe\t1\nPatty_Schnyder\t4\tRicardo_Lagos\t25\nPervez_Musharraf\t3\tRichard_Rodriguez\t1\nPhil_Gramm\t2\tStefan_Tafrov\t1\nRachel_Kempson\t1\tZorica_Radovic\t1\nRachel_Roy\t1\tSteve_Shiver\t1\nRichard_Fine\t1\tRichard_Rodriguez\t1\nRick_Reed\t1\tRuth_Bader_Ginsburg\t1\nRobbie_Naish\t1\tZhong_Nanshan\t1\nRobert_Bonner\t2\tVincent_Brooks\t2\nRobert_Downey_Jr\t1\tTommy_Shane_Steiner\t1\nRobert_Evans\t1\tTodd_Robbins\t1\nRomeo_Gigli\t1\tTom_Harkin\t4\nSaeb_Erekat\t1\tTom_Coverdale\t2\nSe_Hyuk_Joo\t1\tTom_Rouen\t1\nSergio_Garcia\t2\tThomas_Watjen\t1\nSimon_Yam\t1\tTerry_McAuliffe\t3\nSimon_Yam\t1\tTommy_Haas\t5\nStan_Kroenke\t1\tWilliam_Hyde\t1\nSteve_Ballmer\t1\tTina_Pisnik\t1\nSteve_Ballmer\t2\tVincent_Gallo\t3\nSteve_Shiver\t1\tThomas_Rupprath\t3\nTina_Fey\t1\tTodd_Parrott\t1\nAbdullah_Gul\t1\t6\nAbdullah_Gul\t1\t8\nAbdullah_Gul\t7\t14\nAbdullah_Gul\t9\t12\nAbdullah_Gul\t9\t15\nAdolfo_Rodriguez_Saa\t1\t2\nAdrien_Brody\t2\t3\nAdrien_Brody\t2\t12\nAdrien_Brody\t5\t10\nAdrien_Brody\t7\t8\nAl_Sharpton\t1\t4\nAl_Sharpton\t2\t4\nAl_Sharpton\t2\t7\nAl_Sharpton\t3\t4\nAlexandra_Stevenson\t1\t2\nAlexandra_Stevenson\t1\t3\nAlexandra_Vodjanikova\t1\t2\nAlicia_Silverstone\t1\t2\nAna_Palacio\t3\t7\nAna_Palacio\t6\t8\nAndre_Agassi\t1\t5\nAndre_Agassi\t2\t16\nAndre_Agassi\t4\t33\nAndre_Agassi\t9\t25\nAndre_Agassi\t17\t25\nAnna_Kournikova\t2\t3\nAnna_Kournikova\t2\t12\nAnna_Kournikova\t3\t8\nAnna_Kournikova\t7\t8\nAnna_Kournikova\t7\t11\nAnnette_Lu\t1\t2\nArnold_Palmer\t1\t3\nArnold_Palmer\t2\t3\nAron_Ralston\t1\t2\nArturo_Gatti\t2\t3\nBashar_Assad\t1\t3\nBashar_Assad\t2\t4\nBernardo_Segura\t1\t2\nBill_Gates\t3\t16\nBill_Gates\t7\t15\nBill_Gates\t11\t13\nBill_Gates\t13\t17\nBo_Pelini\t1\t2\nBob_Stoops\t2\t4\nBob_Stoops\t2\t5\nBobby_Robson\t1\t2\nBode_Miller\t1\t2\nCaroline_Kennedy\t1\t3\nCaroline_Kennedy\t2\t3\nCatherine_Zeta-Jones\t1\t11\nCatherine_Zeta-Jones\t4\t9\nCelso_Amorim\t1\t2\nChan_Gailey\t1\t2\nChanda_Rubin\t2\t5\nChanda_Rubin\t4\t5\nCharles_Bronson\t1\t2\nCharles_Kartman\t1\t2\nCharles_Schumer\t1\t2\nChris_Rock\t1\t2\nChristine_Baumgartner\t1\t2\nChristine_Baumgartner\t2\t4\nColin_Farrell\t1\t7\nColin_Farrell\t2\t7\nColin_Farrell\t3\t6\nColin_Farrell\t4\t8\nDalai_Lama\t1\t2\nDaniel_Radcliffe\t1\t4\nDaryl_Hannah\t1\t2\nDavid_Anderson\t1\t2\nDavid_Anderson\t1\t3\nDavid_Beckham\t1\t15\nDavid_Beckham\t9\t16\nDavid_Beckham\t11\t29\nDavid_Beckham\t13\t29\nDavid_Beckham\t15\t24\nDavid_Beckham\t20\t27\nDavid_Beckham\t27\t31\nDenzel_Washington\t2\t4\nDenzel_Washington\t3\t5\nDianne_Feinstein\t1\t2\nDianne_Feinstein\t1\t3\nDianne_Feinstein\t2\t3\nDick_Clark\t1\t2\nDick_Clark\t1\t3\nDonald_Fehr\t1\t3\nDonald_Fehr\t1\t4\nDonald_Fehr\t2\t3\nDonald_Fehr\t3\t4\nDwayne_Johnson\t1\t2\nEd_Rosenthal\t1\t2\nErik_Morales\t1\t2\nErik_Morales\t2\t3\nEvan_Rachel_Wood\t2\t3\nEvander_Holyfield\t1\t2\nEve_Pelletier\t1\t2\nFarouk_al-Sharaa\t1\t2\nFarouk_al-Sharaa\t1\t3\nFarouk_al-Sharaa\t2\t3\nFrank_Cassell\t1\t2\nFrank_Cassell\t1\t3\nFujio_Cho\t1\t4\nFujio_Cho\t2\t5\nFujio_Cho\t3\t5\nFujio_Cho\t4\t6\nFujio_Cho\t5\t6\nGao_Qiang\t1\t2\nGeoff_Hoon\t1\t2\nGeoff_Hoon\t3\t4\nGeoff_Hoon\t4\t5\nGeorge_Brumley\t1\t2\nGeorge_Papandreou\t1\t2\nGeorge_Papandreou\t1\t3\nGeorge_Papandreou\t1\t4\nGeorge_Papandreou\t3\t4\nGloria_Allred\t1\t2\nGreg_Ostertag\t1\t2\nGreg_Owen\t1\t2\nHanan_Ashrawi\t1\t2\nHarry_Kalas\t1\t2\nHayley_Tullett\t1\t2\nHoward_Dean\t1\t3\nHoward_Dean\t1\t7\nHoward_Dean\t2\t8\nHoward_Dean\t4\t5\nHoward_Dean\t5\t12\nHu_Jintao\t4\t14\nHu_Jintao\t5\t12\nHu_Jintao\t5\t15\nHu_Jintao\t11\t13\nIan_McKellen\t1\t3\nIbrahim_Jaafari\t1\t2\nIsabella_Rossellini\t1\t3\nIshaq_Shahryar\t1\t2\nIsmail_Merchant\t1\t2\nJacques_Chirac\t2\t43\nJacques_Chirac\t19\t28\nJames_Cameron\t1\t3\nJames_McGreevey\t1\t3\nJames_McGreevey\t1\t4\nJason_Alexander\t1\t2\nJason_Kidd\t2\t8\nJason_Kidd\t3\t8\nJason_Kidd\t6\t8\nJelena_Dokic\t1\t2\nJelena_Dokic\t2\t4\nJelena_Dokic\t2\t8\nJelena_Dokic\t3\t4\nJelena_Dokic\t4\t5\nJelena_Dokic\t4\t7\nJelena_Dokic\t4\t8\nJelena_Dokic\t5\t6\nJelena_Dokic\t7\t8\nJennifer_Connelly\t1\t3\nJennifer_Connelly\t1\t4\nJennifer_Connelly\t2\t4\nJennifer_Connelly\t3\t4\nJennifer_Keller\t1\t4\nJennifer_Keller\t3\t4\nJennifer_Reilly\t1\t2\nJennifer_Thompson\t1\t2\nJim_Hahn\t1\t3\nJim_Hahn\t1\t4\nJim_Hahn\t3\t4\nJohnny_Carson\t1\t2\nJorge_Batlle\t1\t2\nJorge_Batlle\t1\t3\nJorge_Rodolfo_Canicoba_Corral\t1\t2\nJuan_Ignacio_Chela\t1\t2\nJuan_Ignacio_Chela\t1\t3\nJuan_Ignacio_Chela\t2\t3\nJustin_Gatlin\t1\t2\nKaren_Mok\t1\t2\nKatie_Harman\t1\t2\nKatie_Harman\t1\t3\nKatie_Harman\t2\t3\nKenneth_Branagh\t1\t2\nKevin_Costner\t1\t3\nKevin_Costner\t2\t6\nKevin_Costner\t2\t8\nKevin_Costner\t6\t8\nLarry_Lindsey\t1\t2\nLeonardo_DiCaprio\t3\t7\nLeonardo_DiCaprio\t5\t6\nLeonardo_DiCaprio\t6\t8\nLindsay_Davenport\t2\t10\nLindsay_Davenport\t5\t9\nLisa_Ling\t1\t2\nLucy_Liu\t2\t4\nLucy_Liu\t2\t5\nLucy_Liu\t3\t5\nManfred_Stolpe\t1\t2\nMaria_Luisa_Mendonca\t1\t2\nMario_Kreutzberger\t1\t2\nMelissa_Etheridge\t1\t2\nMichael_Jordan\t1\t2\nMichael_Jordan\t1\t3\nMichael_Jordan\t2\t4\nMikhail_Youzhny\t1\t2\nMitchell_Daniels\t1\t2\nMitchell_Daniels\t1\t4\nMitchell_Daniels\t3\t4\nMohammad_Khatami\t1\t5\nMohammad_Khatami\t3\t9\nMohammad_Khatami\t8\t10\nMonique_Garbrecht-Enfeldt\t1\t3\nNatalie_Maines\t2\t5\nNatalie_Maines\t4\t5\nNelson_Mandela\t1\t3\nNorodom_Sihanouk\t1\t3\nNorodom_Sihanouk\t2\t3\nOsama_bin_Laden\t2\t4\nOsama_bin_Laden\t3\t4\nOxana_Fedorova\t1\t2\nPeter_Bacanovic\t1\t2\nPhilippe_Noiret\t1\t2\nPlacido_Domingo\t1\t2\nPlacido_Domingo\t2\t3\nPrince_William\t1\t2\nPrincess_Aiko\t1\t2\nPriscilla_Presley\t1\t2\nRichard_Haass\t1\t2\nRichard_Shelby\t1\t2\nRick_Barnes\t1\t2\nRick_Barnes\t1\t3\nRick_Dinse\t1\t3\nRick_Dinse\t2\t3\nRio_Ferdinand\t1\t2\nRita_Grande\t1\t2\nRita_Grande\t2\t3\nRobin_McLaurin_Williams\t1\t2\nSaddam_Hussein\t2\t4\nSaddam_Hussein\t2\t10\nSaddam_Hussein\t3\t11\nSaddam_Hussein\t6\t9\nSally_Field\t1\t2\nSalma_Hayek\t1\t10\nSalma_Hayek\t1\t11\nSandra_Bullock\t1\t3\nSandra_Bullock\t1\t4\nSandra_Bullock\t2\t4\nSandra_Bullock\t3\t4\nSarah_Hughes\t1\t6\nSarah_Hughes\t2\t5\nSarah_Hughes\t3\t5\nSarah_Hughes\t4\t6\nSharon_Frey\t1\t2\nSilvio_Berlusconi\t6\t30\nSilvio_Berlusconi\t8\t23\nSilvio_Berlusconi\t13\t21\nSilvio_Berlusconi\t17\t29\nSilvio_Berlusconi\t18\t25\nSilvio_Berlusconi\t23\t28\nSilvio_Berlusconi\t24\t30\nSilvio_Berlusconi\t26\t32\nSpencer_Abraham\t3\t17\nSteffi_Graf\t2\t5\nSteffi_Graf\t3\t4\nSteven_Spielberg\t1\t5\nSteven_Spielberg\t1\t6\nSteven_Spielberg\t3\t6\nSteven_Spielberg\t4\t6\nSteven_Spielberg\t4\t7\nSteven_Spielberg\t6\t7\nTheresa_May\t1\t2\nTheresa_May\t2\t3\nTippi_Hedren\t1\t2\nTodd_Haynes\t1\t2\nTodd_Haynes\t1\t3\nTodd_Haynes\t1\t4\nTodd_Haynes\t2\t4\nTony_Blair\t18\t86\nTony_Blair\t32\t48\nTony_Blair\t32\t110\nTony_Blair\t53\t102\nTony_Blair\t91\t134\nTony_Blair\t92\t133\nTony_Blair\t105\t121\nVicente_Fox\t4\t30\nVicente_Fox\t9\t16\nVicente_Fox\t11\t17\nVicente_Fox\t15\t30\nVojislav_Kostunica\t1\t3\nVojislav_Kostunica\t2\t4\nVojislav_Kostunica\t4\t7\nVojislav_Kostunica\t5\t7\nWalter_Mondale\t1\t6\nWalter_Mondale\t1\t7\nWalter_Mondale\t5\t7\nWalter_Mondale\t6\t8\nWalter_Mondale\t7\t9\nWang_Yingfan\t2\t3\nWilliam_Bratton\t1\t3\nWilliam_Donaldson\t1\t5\nWilliam_Donaldson\t3\t5\nWilliam_Donaldson\t4\t6\nXavier_Malisse\t1\t3\nXavier_Malisse\t3\t4\nXavier_Malisse\t3\t5\nYevgeny_Kafelnikov\t1\t4\nZarai_Toledo\t1\t2\nAdolfo_Rodriguez_Saa\t1\tDave_Odom\t1\nAdolfo_Rodriguez_Saa\t1\tNancy_Powell\t1\nAdrien_Brody\t3\tDamon_Dash\t1\nAhmed_Ahmed\t1\tMike_Smith\t1\nAl_Sharpton\t5\tCole_Chapman\t1\nAlberto_Sordi\t1\tJames_Cameron\t2\nAlejandro_Lerner\t1\tJesper_Parnevik\t1\nAlejandro_Lerner\t1\tTA_McLendon\t1\nAlex_Penelas\t1\tRobin_Johansen\t1\nAlex_Penelas\t2\tGretchen_Mol\t1\nAlexandra_Stevenson\t3\tRonde_Barber\t1\nAli_Adbul_Karim_Madani\t1\tCarey_Lowell\t1\nAli_Mohammed_Maher\t1\tIsabela_Moraes\t1\nAli_Mohammed_Maher\t1\tIsidro_Pastor\t1\nAlicia_Silverstone\t2\tJennifer_Thompson\t1\nAlicia_Silverstone\t2\tJonathan_Tiomkin\t1\nAlicia_Silverstone\t2\tSerge_Tchuruk\t1\nAndre_Agassi\t31\tBoris_Jordan\t1\nAndre_Techine\t1\tLloyd_Richards\t1\nAndy_Benes\t1\tDoug_Moe\t1\nAndy_Benes\t1\tMitar_Rasevic\t1\nAndy_Bryant\t1\tMike_Slive\t1\nAnna_Kournikova\t8\tColin_Farrell\t1\nAnne_ONeil\t1\tSophie\t1\nAnthony_Lee_Johnson\t1\tDave_Williams\t1\nArnold_Palmer\t1\tStephanie_Zimbalist\t1\nArnold_Palmer\t3\tCharles_Kartman\t1\nAron_Ralston\t1\tBill_Fennelly\t1\nBart_Hendricks\t1\tPhilippe_Noiret\t1\nBen_Kingsley\t1\tDaryl_Hannah\t1\nBen_Kingsley\t1\tJean_Nagel\t1\nBen_Kingsley\t1\tPerry_Farrell\t1\nBenjamin_Neulander\t1\tRobin_Tunney\t1\nBernardo_Segura\t1\tDebra_Shank\t1\nBernardo_Segura\t2\tDouglas_Paal\t1\nBernardo_Segura\t2\tJeanette_Stauffer\t1\nBill_Pryor\t1\tRonde_Barber\t1\nBob_Melvin\t1\tRichard_Hamilton\t1\nBob_Stoops\t3\tWayne_Allard\t1\nBobby_Jackson\t1\tBruce_Gebhardt\t1\nBobby_Jackson\t1\tJP_Suarez\t1\nBobby_Jackson\t1\tMadge_Overhouse\t1\nBobby_Jackson\t1\tNicola_Wells\t1\nBobby_Robson\t1\tRollie_Massimino\t1\nBode_Miller\t1\tDavid_Howard\t1\nBode_Miller\t1\tSteffi_Graf\t1\nBoris_Jordan\t1\tKim_Weeks\t1\nBoris_Jordan\t1\tVincent_Sombrotto\t1\nBrad_Alexander_Smith\t1\tJason_Alexander\t2\nBrad_Alexander_Smith\t1\tKate_Lee\t1\nBrad_Alexander_Smith\t1\tTatiana_Gratcheva\t1\nBrandon_Webb\t1\tHelmut_Panke\t1\nBrandon_Webb\t1\tLarry_Hahn\t1\nBrandon_Webb\t1\tRyan_Leaf\t1\nBrennon_Leighton\t1\tLaurel_Clark\t1\nBrett_Hawke\t1\tTeri_Files\t1\nBruce_Gebhardt\t1\tJean-Luc_Bideau\t1\nBruce_Gebhardt\t1\tMasao_Azuma\t1\nBryan_Cooley\t1\tKatie_Harman\t3\nBryan_Murray\t1\tEric_Schacht\t1\nBryan_Murray\t1\tMikhail_Youzhny\t1\nBryan_Murray\t1\tMitchell_Daniels\t2\nCamille_Colvin\t1\tIrina_Yatchenko\t1\nCarey_Lowell\t1\tCharlie_Coles\t1\nCarl_Pope\t1\tLarry_Hahn\t1\nCarolina_Barco\t1\tDaryl_Hannah\t1\nCarolina_Barco\t1\tLindsay_Davenport\t22\nCarolina_Barco\t1\tNorodom_Sihanouk\t1\nCaroline_Kennedy\t3\tHenry_Suazo\t1\nCatherine_Bell\t1\tGuillaume_Cannet\t1\nCatherine_Zeta-Jones\t10\tPier_Ferdinando_Casini\t1\nCatherine_Zeta-Jones\t11\tHayley_Tullett\t2\nCelso_Amorim\t2\tJuljia_Vysotskij\t1\nCelso_Amorim\t2\tKevin_Costner\t1\nCelso_Amorim\t2\tThomas_Stewart\t1\nChanda_Rubin\t3\tRichard_Tubb\t1\nChante_Jawan_Mallard\t1\tStephen_Glassroth\t1\nChante_Jawan_Mallard\t1\tVicente_Fox\t23\nCharles_Bronson\t2\tHu_Jintao\t4\nCharles_Kartman\t1\tDebra_Shank\t1\nCharles_Schumer\t2\tJames_Watt\t1\nCharles_Schumer\t2\tJustin_Gatlin\t2\nCharles_Schumer\t2\tLionel_Chalmers\t1\nChris_Rock\t2\tDalai_Lama\t1\nChristine_Baumgartner\t2\tVincent_Sombrotto\t1\nChristine_Baumgartner\t3\tLars_Burgsmuller\t1\nChyung_Dai-chul\t1\tDan_Reeves\t1\nCindy_Moll\t1\tDaniel_Radcliffe\t3\nCindy_Moll\t1\tJames_McPherson\t1\nCindy_Moll\t1\tRobin_Johansen\t1\nClaire_De_Gryse\t1\tDeb_Santos\t1\nClaire_Tomalin\t1\tSteve_Case\t1\nClay_Campbell\t1\tLeonardo_DiCaprio\t2\nCole_Chapman\t1\tMarisol_Martinez_Sambran\t1\nColin_Phillips\t1\tIan_McKellen\t3\nColin_Phillips\t1\tLloyd_Richards\t1\nColin_Prescot\t1\tLaurie_Hobbs\t1\nConnie_Freydell\t1\tTommy_Maddox\t1\nCraig_OClair\t1\tSteve_Avery\t1\nDamarius_Bilbo\t1\tJuljia_Vysotskij\t1\nDamon_Dash\t1\tHenry_Suazo\t1\nDaniel_Comisso_Urdaneta\t1\tLarry_Ralston\t1\nDaniel_Radcliffe\t3\tDiane_Green\t2\nDaniel_Radcliffe\t3\tRick_Barnes\t2\nDaniel_Radcliffe\t3\tVicente_Fox\t32\nDany_Heatley\t1\tRichard_Parsons\t1\nDany_Heatley\t1\tTerry_Hoeppner\t1\nDave_Odom\t1\tMaritza_Macias_Furano\t1\nDavid_Beckham\t22\tLaura_Morante\t1\nDavid_Duval\t1\tPhilippe_Noiret\t2\nDavid_Ho\t1\tDoug_Moe\t1\nDavid_Ho\t1\tErskine_Bowles\t1\nDavid_Ho\t1\tEvander_Holyfield\t1\nDavid_Howard\t1\tHugo_Conte\t1\nDenzel_Washington\t4\tHenry_Suazo\t1\nDes_Brown\t1\tEliott_Spitzer\t1\nDes_Brown\t1\tSvetislav_Pesic\t1\nDiana_Ross\t1\tRick_Rickert\t1\nDianne_Feinstein\t3\tDwayne_Johnson\t2\nDick_Clark\t3\tManfred_Stolpe\t1\nDino_Risi\t1\tGabriel_Farhi\t1\nDoc_Rivers\t1\tMelissa_Joan_Hart\t1\nDominick_Dunne\t1\tNicoletta_Braschi\t1\nDon_Flanagan\t1\tEd_Sullivan\t1\nDon_Flanagan\t1\tRatna_Sari_Dewi_Sukarno\t1\nDonald_Keyser\t1\tSalma_Hayek\t2\nDouglas_Paal\t1\tEd_Sullivan\t1\nDouglas_Paal\t1\tRichard_Hamilton\t1\nDusty_Baker\t1\tSaddam_Hussein\t2\nEd_Rendell\t1\tJamie_Dimon\t1\nEd_Rendell\t1\tRichard_Cohen\t1\nEddie_Murray\t1\tKevin_Costner\t7\nElaine_Stritch\t1\tRichard_Parsons\t1\nElena_Tihomirova\t1\tMike_Slive\t1\nElena_Tihomirova\t1\tMohammed_Abu_Sharia\t1\nEric_Schacht\t1\tJennifer_Keller\t2\nErik_Morales\t3\tWerner_Schlager\t1\nErin_Brockovich\t1\tHenry_Suazo\t1\nErin_Brockovich\t1\tMike_Carona\t1\nEvan_Rachel_Wood\t1\tSharon_Frey\t1\nEvander_Holyfield\t1\tJulio_Rossi\t1\nEvander_Holyfield\t1\tMichael_Jordan\t3\nEvander_Holyfield\t1\tMike_Carona\t1\nFatmir_Limaj\t1\tTodd_Haynes\t4\nFelicity_Huffman\t1\tNelson_Acosta\t1\nFelipe_De_Borbon\t1\tJose_Bove\t1\nFrank_Cassell\t2\tJesper_Parnevik\t1\nFrank_Cassell\t3\tRonde_Barber\t1\nFred_Durst\t1\tMarisol_Martinez_Sambran\t1\nFred_Durst\t1\tShavon_Earp\t1\nFujio_Cho\t1\tJavier_Vargas\t1\nFujio_Cho\t2\tTatsuya_Fuji\t1\nFujio_Cho\t3\tMichael_Brandon\t1\nFujio_Cho\t6\tOlene_Walker\t1\nGabriel_Farhi\t1\tLars_Burgsmuller\t1\nGao_Qiang\t2\tKenneth_Brill\t1\nGary_Stevens\t1\tWanda_Ilene_Barzee\t1\nGeoff_Hoon\t1\tOsama_bin_Laden\t3\nGeoff_Hoon\t3\tRichard_Tubb\t1\nGeoff_Hoon\t5\tThomas_Stewart\t1\nGeorge_Brumley\t1\tManfred_Stolpe\t1\nGeorge_Brumley\t1\tRyan_Leaf\t1\nGeorge_Brumley\t2\tSergei_Yushenkov\t1\nGeorge_Papandreou\t2\tTatiana_Panova\t1\nGeorge_Papandreou\t4\tLarry_Lindsey\t2\nGeorge_Papandreou\t4\tPaul_LeClerc\t1\nGhassan_Elashi\t1\tRick_Rickert\t1\nGloria_Allred\t2\tRoman_Tam\t1\nGreg_Kinsey\t1\tPriscilla_Presley\t1\nGreg_Ostertag\t2\tKevin_Costner\t6\nGreg_Owen\t1\tMark_Sisk\t1\nGreg_Owen\t2\tMike_Smith\t1\nGretchen_Mol\t1\tIsabella_Rossellini\t2\nGunilla_Backman\t1\tKathleen_Abernathy\t1\nHanan_Ashrawi\t1\tMario_Kreutzberger\t2\nHans_Leistritz\t1\tSaddam_Hussein\t18\nHans_Leistritz\t1\tSue_Grafton\t1\nHans_Peter_Briegel\t1\tTheresa_May\t2\nHarriet_Lessy\t1\tMike_Carona\t1\nHarry_Kalas\t2\tJoe_Strummer\t1\nHarry_Kalas\t2\tLewis_Booth\t1\nHayley_Tullett\t1\tIan_McKellen\t1\nHenry_Suazo\t1\tRatna_Sari_Dewi_Sukarno\t1\nHernan_Crespo\t1\tRonde_Barber\t1\nHiroyuki_Yoshino\t1\tRollie_Massimino\t1\nHu_Jintao\t6\tJerome_Jenkins\t1\nHugh_Jessiman\t1\tSolomon_Passy\t1\nHugh_Jessiman\t1\tTA_McLendon\t1\nHugo_Conte\t1\tLaurent_Woulzy\t1\nIrina_Yatchenko\t1\tKatie_Harman\t2\nIsidro_Pastor\t1\tMark_Butcher\t1\nIsmail_Merchant\t1\tMack_Brown\t2\nIsmail_Merchant\t2\tKathleen_Abernathy\t1\nIvan_Helguera\t1\tWilliam_Donaldson\t4\nIveta_Benesova\t1\tJacques_Chirac\t50\nJacques_Chirac\t20\tTzipora_Obziler\t1\nJacques_Chirac\t35\tRichard_Shelby\t2\nJames_Cameron\t1\tRichard_Shelby\t2\nJames_McGreevey\t1\tKatie_Harman\t2\nJames_McGreevey\t1\tTzipora_Obziler\t1\nJames_Sensenbrenner\t1\tJimmy_Smits\t1\nJames_Sensenbrenner\t1\tMaria_Shkolnikova\t1\nJames_Watt\t1\tPriscilla_Presley\t2\nJamie_Dimon\t1\tRobert_Korzeniowski\t1\nJason_Alexander\t1\tMike_Maroth\t1\nJason_Alexander\t1\tWayne_Allard\t1\nJason_Kidd\t7\tLouisa_Baileche\t1\nJavier_Vargas\t1\tJesse_Helms\t1\nJeanette_Gray\t1\tKeith_Fotta\t1\nJeanette_Stauffer\t1\tMarie-Josee_Croze\t1\nJeffrey_Donaldson\t1\tLeonardo_DiCaprio\t8\nJeffrey_Donaldson\t1\tPeter_Bacanovic\t1\nJennifer_Connelly\t4\tMartin_Bandier\t1\nJesper_Parnevik\t1\tJustin_Gatlin\t2\nJimmy_Smits\t1\tRobin_Johansen\t1\nJimmy_Smits\t1\tShane_Hmiel\t1\nJoe_Crede\t1\tTom_DeLay\t1\nJohn_Burkett\t1\tLeonardo_DiCaprio\t2\nJohn_McKay\t1\tMary_Sue_Coleman\t1\nJohnny_Hallyday\t1\tOlene_Walker\t1\nJonathan_Tiomkin\t1\tRita_Grande\t2\nJorge_Rodolfo_Canicoba_Corral\t1\tWayne_Allard\t1\nJose_Bove\t1\tJuergen_Trittin\t1\nJuljia_Vysotskij\t1\tPatrick_Coleman\t1\nKate_Lee\t1\tToshimitsu_Motegi\t1\nKathie_Louise_Saunders\t1\tRoberto_Lavagna\t1\nKathie_Louise_Saunders\t1\tRosalyn_Carter\t1\nKenneth_Branagh\t1\tPinar_del_Rio\t1\nKevin_Costner\t5\tStephen_Glassroth\t1\nKhader_Rashid_Rahim\t1\tRichard_Cohen\t1\nKhalid_Khannouchi\t1\tWerner_Schlager\t1\nKim_Su_Nam\t1\tTodd_Haynes\t4\nKim_Weeks\t1\tMarie-Josee_Croze\t1\nKimora_Lee\t1\tRobin_Johansen\t1\nKurt_Suzuki\t1\tScott_Hoch\t1\nLarry_Lindsey\t1\tWanda_Ilene_Barzee\t1\nLarry_Ralston\t1\tRichard_Carl\t1\nLarry_Ralston\t1\tWilliam_Donaldson\t6\nLarry_Ralston\t1\tWillie_Wilson\t1\nLaurel_Clark\t1\tLaurent_Woulzy\t1\nLaurel_Clark\t1\tPriscilla_Presley\t1\nLeandrinho_Barbosa\t1\tMelissa_Joan_Hart\t1\nLeandrinho_Barbosa\t1\tOlene_Walker\t1\nLeon_Barmore\t1\tSachin_Tendulkar\t1\nLeon_Barmore\t1\tYusaku_Miyazato\t1\nLeonard_Schrank\t1\tMarie_Haghal\t1\nLeonard_Schrank\t1\tRick_Rickert\t1\nLinda_Franklin\t1\tMelissa_Joan_Hart\t1\nLindsay_Davenport\t1\tMartin_Burnham\t1\nLindsay_Davenport\t17\tOxana_Fedorova\t3\nLionel_Chalmers\t1\tMohammad_Fares\t1\nLisa_Ling\t2\tMike_Maroth\t1\nLucy_Liu\t3\tPrince_William\t1\nMaria_Luisa_Mendonca\t2\tStephane_Delajoux\t1\nMaria_Luisa_Mendonca\t2\tTerunobu_Maeda\t1\nMaria_Shkolnikova\t1\tMartin_Rodriguez\t1\nMariana_Ohata\t1\tXavier_Malisse\t3\nMario_Kreutzberger\t1\tShavon_Earp\t1\nMario_Kreutzberger\t2\tRaul_Cubas\t1\nMaritza_Macias_Furano\t1\tQusai_Hussein\t1\nMark_Butcher\t1\tScott_Hoch\t1\nMark_Sisk\t1\tMehdi_Baala\t1\nMark_Sisk\t1\tStephen_Swindal\t1\nMasao_Azuma\t1\tMikhail_Kalashnikov\t1\nMcGuire_Gibson\t1\tRichard_Haass\t1\nMehdi_Baala\t1\tSteve_Avery\t1\nMelissa_Joan_Hart\t1\tMikhail_Kalashnikov\t1\nMelissa_Joan_Hart\t1\tSue_Grafton\t1\nMichael_Brandon\t1\tToshimitsu_Motegi\t1\nMichael_Shelby\t1\tOlene_Walker\t1\nMichel_Kratochvil\t1\tSheldon_Silver\t1\nMike_Eskew\t1\tZarai_Toledo\t1\nMike_OConnell\t1\tRoberto_Lavagna\t1\nMike_Sherman\t1\tNatalie_Maines\t3\nMike_Sherman\t1\tPaige_Fitzgerald\t1\nMilt_Heflin\t1\tPier_Ferdinando_Casini\t1\nNicholoas_DiMarzio\t1\tRichard_Parsons\t1\nNikki_McKibbin\t1\tStephen_Swindal\t1\nNikki_McKibbin\t1\tSteven_Tyler\t1\nOsama_Al_Baz\t1\tPatricia_Wartusch\t1\nOsama_Al_Baz\t1\tSandra_Day_OConner\t1\nPatricia_Phillips\t1\tThierry_Falise\t2\nPaula_Locke\t1\tTeri_Files\t1\nPeter_Bacanovic\t2\tRobert_Korzeniowski\t1\nPier_Ferdinando_Casini\t1\tRatna_Sari_Dewi_Sukarno\t1\nPriscilla_Presley\t2\tRaul_Rivero\t1\nRaaf_Schefter\t1\tRick_Dinse\t2\nRichard_Sterner\t1\tSteven_Spielberg\t2\nRick_Dinse\t2\tShavon_Earp\t1\nRoberto_Lavagna\t1\tSandra_Bullock\t4\nRod_Stewart\t1\tSteffi_Graf\t2\nRodolfo_Abalos\t1\tThierry_Falise\t3\nRoman_Tam\t1\tZalmay_Khalilzad\t1\nScott_Hoch\t1\tThomas_Stewart\t1\nSeymour_Cassell\t1\tTodd_Haynes\t4\nShaun_Rusling\t1\tVincent_Sombrotto\t1\nSheldon_Silver\t1\tXavier_Malisse\t3\nStephane_Delajoux\t1\tTristan_Gretzky\t1\nTakahiro_Mori\t1\tTony_Blair\t64\nThierry_Falise\t2\tWerner_Schlager\t1\nAaron_Sorkin\t1\t2\nAbdullah\t1\t3\nAbdullah\t2\t3\nAbdullah\t2\t4\nAbdullah\t3\t4\nAbid_Hamid_Mahmud_Al-Tikriti\t1\t2\nAbid_Hamid_Mahmud_Al-Tikriti\t1\t3\nAlan_Ball\t1\t2\nAlbert_Costa\t3\t5\nAlbert_Costa\t5\t6\nAli_Khamenei\t1\t2\nAli_Khamenei\t1\t3\nAmelia_Vega\t1\t7\nAmelia_Vega\t2\t7\nAmelia_Vega\t3\t5\nAntonio_Banderas\t1\t5\nAntonio_Banderas\t2\t5\nAntonio_Banderas\t3\t5\nArye_Mekel\t1\t2\nAzra_Akin\t1\t3\nAzra_Akin\t1\t4\nBernard_Landry\t2\t3\nBernard_Landry\t2\t4\nBernard_Landry\t3\t4\nBiljana_Plavsic\t1\t2\nBiljana_Plavsic\t1\t3\nBiljana_Plavsic\t2\t3\nBob_Hope\t1\t5\nBob_Hope\t3\t6\nBob_Hope\t4\t5\nBob_Hope\t4\t7\nBridget_Fonda\t1\t3\nBridget_Fonda\t2\t3\nCarlos_Ruiz\t1\t3\nCarlos_Ruiz\t2\t3\nCarly_Fiorina\t1\t2\nCarly_Fiorina\t1\t3\nCherie_Blair\t1\t3\nCherie_Blair\t1\t4\nCherie_Blair\t2\t4\nCherie_Blair\t3\t4\nChuck_Yeager\t1\t2\nChung_Mong-joon\t1\t2\nClaire_Hentzen\t1\t2\nClaire_Leger\t1\t2\nDarren_Clarke\t1\t2\nDavid_Caraway\t1\t2\nDavid_Leahy\t1\t2\nDick_Cheney\t2\t14\nDick_Cheney\t3\t11\nDick_Cheney\t3\t12\nDick_Cheney\t8\t10\nDino_de_Laurentis\t1\t2\nDon_Nickles\t1\t2\nDoris_Schroeder\t1\t3\nDoris_Schroeder\t2\t4\nEduardo_Duhalde\t2\t3\nEduardo_Duhalde\t2\t12\nEduardo_Duhalde\t3\t4\nEduardo_Duhalde\t6\t13\nEduardo_Duhalde\t7\t10\nEduardo_Duhalde\t10\t12\nEdwin_Edwards\t1\t3\nEric_Rosser\t1\t2\nErnie_Eves\t1\t2\nErnie_Fletcher\t1\t2\nEva_Dimas\t1\t2\nFernando_Vargas\t1\t2\nFernando_Vargas\t1\t3\nFernando_Vargas\t1\t4\nFernando_Vargas\t2\t3\nFernando_Vargas\t3\t4\nFrank_Lautenberg\t1\t2\nGeorge_HW_Bush\t1\t3\nGeorge_HW_Bush\t2\t13\nGeorge_HW_Bush\t4\t6\nGeorge_HW_Bush\t5\t8\nGeorge_HW_Bush\t9\t10\nGisele_Bundchen\t1\t2\nGlafcos_Clerides\t1\t2\nGlafcos_Clerides\t1\t3\nGlafcos_Clerides\t2\t3\nGlafcos_Clerides\t2\t4\nGreg_Gilbert\t1\t2\nHashim_Thaci\t1\t2\nHassan_Wirajuda\t1\t2\nHector_Babenco\t1\t3\nHector_Babenco\t2\t3\nHideki_Matsui\t1\t2\nHillary_Clinton\t1\t10\nHillary_Clinton\t1\t14\nHillary_Clinton\t4\t5\nHillary_Clinton\t7\t10\nHillary_Clinton\t11\t13\nHisao_Oguchi\t1\t2\nHitomi_Soga\t1\t4\nHitomi_Soga\t1\t5\nHitomi_Soga\t2\t3\nHitomi_Soga\t3\t4\nJJ_Redick\t1\t2\nJean_Brumley\t1\t2\nJean_Carnahan\t1\t2\nJeremy_Shockey\t1\t2\nJerry_Regier\t1\t3\nJerry_Regier\t2\t3\nJerry_Springer\t1\t2\nJerry_Springer\t1\t4\nJessica_Lynch\t1\t2\nJiang_Zemin\t1\t17\nJiang_Zemin\t2\t3\nJiang_Zemin\t4\t10\nJiang_Zemin\t7\t9\nJiang_Zemin\t15\t18\nJim_Harrick\t1\t2\nJiri_Novak\t3\t4\nJiri_Novak\t3\t8\nJiri_Novak\t3\t9\nJiri_Novak\t5\t6\nJiri_Novak\t5\t10\nJiri_Novak\t5\t11\nJiri_Novak\t7\t10\nJoe_Torre\t1\t3\nJoe_Torre\t2\t4\nJohn_Wolf\t1\t2\nJonathan_Edwards\t1\t6\nJonathan_Edwards\t2\t7\nJonathan_Edwards\t3\t6\nJonathan_Edwards\t5\t6\nJonathan_Edwards\t5\t7\nJonathan_Edwards\t6\t8\nJose_Manuel_Durao_Barroso\t1\t2\nJose_Manuel_Durao_Barroso\t1\t6\nJose_Manuel_Durao_Barroso\t2\t5\nJose_Manuel_Durao_Barroso\t3\t5\nJose_Manuel_Durao_Barroso\t4\t5\nJoseph_Biden\t1\t4\nJoseph_Biden\t2\t3\nJoseph_Biden\t3\t4\nJudi_Dench\t1\t2\nJudy_Genshaft\t1\t2\nKeanu_Reeves\t1\t9\nKeanu_Reeves\t4\t7\nKeanu_Reeves\t6\t9\nKeanu_Reeves\t6\t10\nKeanu_Reeves\t7\t12\nKeira_Knightley\t1\t2\nKen_Watanabe\t1\t2\nKieran_Prendergast\t1\t2\nKing_Abdullah_II\t1\t3\nKing_Abdullah_II\t1\t4\nKing_Abdullah_II\t2\t5\nKing_Abdullah_II\t3\t4\nKirk_Ferentz\t1\t2\nKurt_Busch\t1\t2\nLarry_Bowa\t1\t2\nLarry_Thompson\t1\t3\nLarry_Thompson\t1\t4\nLarry_Thompson\t2\t4\nLaura_Bush\t1\t11\nLaura_Bush\t21\t24\nLaura_Bush\t21\t39\nLaura_Bush\t26\t29\nLauren_Hutton\t1\t2\nLeslie_Ann_Woodward\t1\t2\nLeslie_Moonves\t1\t2\nLyle_Vanclief\t1\t2\nMagui_Serna\t1\t2\nMakhdoom_Amin_Fahim\t1\t3\nMakhdoom_Amin_Fahim\t2\t3\nMarc-Andre_Fleury\t1\t2\nMarco_Antonio_Barrera\t1\t5\nMarco_Antonio_Barrera\t5\t6\nMarie-Reine_Le_Gougne\t1\t2\nMarieta_Chrousala\t1\t2\nMarieta_Chrousala\t1\t3\nMarieta_Chrousala\t2\t3\nMarina_Anissina\t1\t2\nMark_Hamister\t1\t2\nMartha_Stewart\t1\t3\nMartha_Stewart\t1\t5\nMartha_Stewart\t2\t3\nMartha_Stewart\t2\t5\nMartha_Stewart\t3\t4\nMartin_Brodeur\t1\t2\nMartin_McCauley\t1\t2\nMartin_Verkerk\t1\t2\nMartin_Verkerk\t1\t3\nMartin_Verkerk\t2\t3\nMartina_McBride\t1\t3\nMartina_McBride\t2\t4\nMatt_Doherty\t1\t2\nMatt_Doherty\t1\t3\nMatt_Doherty\t2\t3\nMatthew_Perry\t1\t6\nMatthew_Perry\t2\t7\nMatthew_Perry\t3\t4\nMegawati_Sukarnoputri\t6\t23\nMegawati_Sukarnoputri\t7\t22\nMegawati_Sukarnoputri\t10\t15\nMegawati_Sukarnoputri\t11\t24\nMegawati_Sukarnoputri\t20\t24\nMegawati_Sukarnoputri\t20\t30\nMeghann_Shaughnessy\t1\t2\nMichael_Capellas\t1\t2\nMichael_Sullivan\t1\t2\nMike_Brey\t1\t2\nNaji_Sabri\t1\t6\nNaji_Sabri\t2\t4\nNaji_Sabri\t2\t7\nNaji_Sabri\t3\t8\nNaji_Sabri\t6\t7\nNanni_Moretti\t1\t2\nNastassia_Kinski\t1\t2\nNatalie_Coughlin\t2\t3\nNatalie_Coughlin\t4\t6\nNatalie_Coughlin\t5\t6\nNorah_Jones\t3\t15\nNorah_Jones\t4\t12\nNorah_Jones\t7\t15\nNorah_Jones\t9\t15\nNorah_Jones\t11\t12\nNorm_Coleman\t5\t7\nOscar_De_La_Hoya\t1\t3\nOscar_De_La_Hoya\t2\t6\nOscar_De_La_Hoya\t2\t7\nPascal_Lamy\t1\t2\nPat_Burns\t1\t2\nPaul_McCartney\t3\t4\nPaul_McCartney\t3\t5\nPaul_Wellstone\t1\t2\nPaul_Wellstone\t1\t3\nPaul_Wellstone\t2\t3\nPenelope_Cruz\t1\t2\nPenelope_Cruz\t1\t3\nPete_Rose\t1\t2\nPrince_Harry\t1\t2\nPrince_Harry\t2\t3\nPrincess_Caroline\t1\t2\nPrincess_Caroline\t2\t5\nRaquel_Welch\t1\t2\nReggie_Miller\t1\t2\nRenee_Zellweger\t2\t13\nRenee_Zellweger\t2\t16\nRenee_Zellweger\t3\t16\nRicardo_Sanchez\t1\t6\nRicardo_Sanchez\t2\t4\nRichard_Butler\t1\t2\nRubens_Barrichello\t2\t3\nRubens_Barrichello\t4\t5\nRubens_Barrichello\t4\t8\nRubens_Barrichello\t6\t11\nRubens_Barrichello\t9\t12\nSamira_Makhmalbaf\t1\t2\nSamuel_Waksal\t1\t2\nSamuel_Waksal\t1\t3\nSamuel_Waksal\t1\t4\nSamuel_Waksal\t2\t4\nScott_McClellan\t1\t3\nScott_McClellan\t2\t4\nScott_McClellan\t2\t5\nScott_Rudin\t1\t2\nShane_Warne\t1\t2\nSheila_Wellstone\t1\t2\nSheryl_Crow\t1\t3\nSilvan_Shalom\t2\t3\nSilvan_Shalom\t2\t4\nSilvan_Shalom\t2\t6\nSilvan_Shalom\t3\t4\nSilvan_Shalom\t3\t6\nSteve_Mariucci\t1\t3\nSteve_Mariucci\t2\t3\nSteven_Seagal\t1\t2\nTaha_Yassin_Ramadan\t3\t12\nTaha_Yassin_Ramadan\t7\t13\nTammy_Lynn_Michaels\t1\t2\nTheodore_Tweed_Roosevelt\t1\t2\nTheodore_Tweed_Roosevelt\t1\t3\nTheodore_Tweed_Roosevelt\t2\t3\nTom_Crean\t1\t2\nTom_Crean\t3\t5\nTom_Crean\t4\t5\nTony_Bennett\t1\t3\nTorri_Edwards\t1\t2\nTung_Chee-hwa\t1\t2\nTung_Chee-hwa\t1\t8\nTung_Chee-hwa\t2\t8\nVidar_Helgesen\t1\t2\nWarren_Buffett\t1\t2\nWarren_Buffett\t1\t3\nWen_Jiabao\t1\t12\nWen_Jiabao\t2\t11\nWen_Jiabao\t4\t7\nWen_Jiabao\t5\t6\nWen_Jiabao\t7\t9\nWen_Jiabao\t7\t12\nWen_Jiabao\t8\t12\nWen_Jiabao\t10\t11\nWen_Jiabao\t11\t12\nWesley_Clark\t1\t2\nYuri_Malenchenko\t1\t2\nAbbas_Kiarostami\t1\tFujio_Mitarai\t1\nAbdullah\t1\tTeresa_Heinz_Kerry\t1\nAbdullah\t3\tSamuel_Waksal\t1\nAbdullah\t4\tJulio_Cesar_Franco\t1\nAbid_Hamid_Mahmud_Al-Tikriti\t1\tAnjum_Hussain\t1\nAbid_Hamid_Mahmud_Al-Tikriti\t2\tDoris_Schroeder\t1\nAdam_Freier\t1\tHillary_Clinton\t3\nAdam_Freier\t1\tPrincess_Caroline\t1\nAdam_Freier\t1\tRegina_Ip\t1\nAlan_Ball\t1\tKristin_Scott_Thomas\t1\nAlan_Ball\t2\tYuri_Malenchenko\t1\nAlbert_Costa\t6\tGary_Barnett\t1\nAlbert_Costa\t6\tJohn_Marburger\t1\nAlbert_Costa\t6\tWen_Jiabao\t7\nAlberta_Lee\t1\tBabe_Ruth\t1\nAlex_Popov\t1\tKent_Rominger\t2\nAlex_Popov\t1\tMatthew_During\t1\nAli_Khamenei\t2\tRoberto_Canessa\t1\nAmelia_Vega\t1\tGina_Lollobrigida\t1\nAmelia_Vega\t5\tJim_Harrick\t1\nAmy_Pascal\t1\tEduardo_Duhalde\t7\nAmy_Pascal\t1\tHank_Azaria\t1\nAmy_Pascal\t1\tJohn_Marburger\t1\nAmy_Redford\t1\tRoman_Coppola\t1\nAmy_Redford\t1\tVictor_Kraatz\t1\nAnderson_Varejao\t1\tDennis_Oswald\t1\nAnderson_Varejao\t1\tGarth_Drabinsky\t1\nAndrei_Nikolishin\t1\tAngelica_Romero\t1\nAndrew_Bernard\t1\tDon_Nickles\t2\nAndrew_Bernard\t1\tHassan_Wirajuda\t1\nAndrew_Bernard\t1\tMark_Broxmeyer\t1\nAndrew_Fastow\t1\tLuca_Cordero_di_Montezemolo\t1\nAndrew_Fastow\t1\tMeghann_Shaughnessy\t1\nAndrew_Fastow\t1\tTomas_Malik\t1\nAndrew_Luster\t1\tEric_Rosser\t1\nAndrew_Luster\t1\tJose_Manuel_Durao_Barroso\t6\nAndrew_Luster\t1\tLisa_Murkowski\t1\nAnita_DeFrantz\t1\tCarla_Gay_Balingit\t1\nAnita_DeFrantz\t1\tPenny_Lancaster\t1\nAnjum_Hussain\t1\tDavid_Caraway\t2\nAnne_Cavers\t1\tJames_Barksdale\t1\nAnne_Cavers\t1\tStephen_Oake\t1\nAnthony_Hazen\t1\tDebra_Yang\t1\nAntonio_Catania\t1\tBarry_Bonds\t1\nAntonio_Catania\t1\tTaylyn_Solomon\t1\nAretha_Franklin\t1\tJean_Brumley\t1\nAretha_Franklin\t1\tTeruaki_Masumoto\t1\nArt_Lopez\t1\tDiane_Lane\t1\nArthur_Johnson\t1\tSelma_Phoenix\t1\nArye_Mekel\t2\tChung_Mong-joon\t1\nArye_Mekel\t2\tGeorgina_Papin\t1\nBabe_Ruth\t1\tJim_Harrick\t1\nBabe_Ruth\t1\tPete_Aldridge\t1\nBarry_Bonds\t1\tJames_Barksdale\t1\nBarry_Bonds\t1\tMickey_Sherman\t1\nBarry_Switzer\t1\tJiang_Zemin\t11\nBarry_Switzer\t1\tKirk_Ferentz\t2\nBernard_Giraudeau\t1\tKeira_Knightley\t1\nBernard_Landry\t3\tStacey_Jones\t1\nBernard_Landry\t4\tSylvia_Plachy\t1\nBiljana_Plavsic\t1\tKieran_Prendergast\t2\nBill_Cartwright\t1\tClaude_Jorda\t1\nBill_Lerach\t1\tMartin_Verkerk\t2\nBob_Hope\t1\tStanley_Ho\t1\nBob_Hope\t5\tMicah_Knorr\t1\nBob_Hope\t5\tMitchell_Potter\t1\nBob_Menendez\t1\tDavid_Caraway\t2\nBob_Menendez\t1\tLauren_Hutton\t1\nBob_Riley\t1\tDanny_Ainge\t1\nBob_Riley\t1\tNastassia_Kinski\t2\nBob_Riley\t1\tSheila_Wellstone\t2\nBrad_Wilk\t1\tPhil_Cline\t1\nBrad_Wilk\t1\tScott_Dalton\t1\nBrian_Grazier\t1\tTomas_Malik\t1\nBrian_StPierre\t1\tSteve_Mariucci\t2\nCari_Davis\t1\tDon_Henley\t1\nCari_Davis\t1\tMark_Broxmeyer\t1\nCarin_Koch\t1\tJohn_Banko\t2\nCarla_Gay_Balingit\t1\tFrank_Lautenberg\t2\nCarla_Gay_Balingit\t1\tPaula_Dobriansky\t1\nCarlos_Ruiz\t1\tPeter_Schultz\t1\nCarlos_Ruiz\t3\tJennifer_Tilly\t1\nCarly_Fiorina\t2\tPeter_Ueberroth\t1\nCatherine_Woodard\t1\tMike_Sweeney\t1\nChan_Ho_Park\t1\tXiang_Xu\t1\nChance_Mock\t1\tHerb_Ritts\t1\nChance_Mock\t1\tShafal_Mosed\t1\nChance_Mock\t1\tSimona_Hradil\t1\nChance_Mock\t1\tTroy_Jenkins\t1\nCharles_Taylor\t7\tTom_Curley\t1\nChea_Sophara\t1\tJohn_Wolf\t1\nChris_Cirino\t1\tJen_Bice\t1\nChris_Cirino\t1\tMary_Lou_Markakis\t1\nChris_Cirino\t1\tPeri_Gilpin\t1\nChristina_Sawaya\t1\tGreg_Gilbert\t1\nChung_Mong-joon\t1\tDavid_Modell\t1\nChung_Mong-joon\t1\tLuca_Cordero_di_Montezemolo\t1\nChung_Mong-joon\t1\tNick_Markakis\t1\nChung_Mong-joon\t2\tJohn_Rusnak\t1\nClaire_Leger\t1\tDavid_Caraway\t2\nClaire_Leger\t1\tJerry_Sexton\t1\nClaire_Leger\t1\tNobuyuki_Idei\t1\nClark_Randt\t1\tKatie_Boone\t1\nClark_Randt\t1\tMarc-Andre_Fleury\t1\nClemente_de_la_Vega\t1\tRon_Gonzales\t1\nColin_Campbell\t1\tFrank_Keating\t1\nColin_Campbell\t1\tJean_Carnahan\t2\nColin_Campbell\t1\tMitchell_Potter\t1\nColleen_Atwood\t1\tPenelope_Cruz\t2\nCraig_Morgan\t1\tMatthew_McConaughey\t1\nDan_Duquette\t1\tPaddy_Torsney\t1\nDaniel_Ortega\t1\tNorah_Jones\t6\nDaniel_Patrick_Moynihan\t1\tEglis_Yaima_Cruz\t1\nDaniel_Patrick_Moynihan\t1\tNewton_Carlton_Slawson\t1\nDaniel_Rouse\t1\tDoris_Schroeder\t1\nDaniel_Rouse\t1\tMike_Johanns\t1\nDaniell_Sunjata\t1\tMarc-Andre_Fleury\t2\nDanny_Ainge\t1\tLee_Ann_Terlaji\t1\nDavid_Leahy\t1\tJerry_Sexton\t1\nDavid_Scott_Morris\t1\tValerie_Thwaites\t1\nDavid_Zeplowitz\t1\tJerry_Springer\t1\nDawn_Staley\t1\tHasan_Wirayuda\t1\nDebra_Rose\t1\tJonathan_Edwards\t2\nDebra_Yang\t1\tHashim_Thaci\t2\nDebra_Yang\t1\tYekaterina_Guseva\t1\nDenis_Fassou-Nguesso\t1\tWilliam_Overlin\t1\nDerrick_Taylor\t1\tGiuseppe_Morchio\t1\nDerrick_Taylor\t1\tWilliam_Overlin\t1\nDick_Cheney\t5\tJames_Murdoch\t1\nDick_Cheney\t8\tNobuyuki_Idei\t1\nDick_Smothers\t1\tYoon_Won-Sik\t1\nDita_Von_Tesse\t1\tMohammed_Al-Douri\t12\nDon_Henley\t1\tErnie_Eves\t1\nDon_Henley\t1\tGil_Cates\t1\nDon_Henley\t1\tStephen_Frears\t1\nDon_Henley\t1\tZiwang_Xu\t1\nDonald_Regan\t1\tJohn_Gruden\t1\nDoris_Schroeder\t2\tJoe_Mantegna\t1\nEduardo_Duhalde\t1\tPaula_Dobriansky\t1\nEduardo_Duhalde\t2\tTab_Turner\t1\nEduardo_Duhalde\t13\tJames_Murdoch\t1\nEdward_Egan\t1\tErnie_Fletcher\t2\nEdward_Johnson\t1\tJessica_Capshaw\t1\nEl_Hadji_Diouf\t1\tHitomi_Soga\t2\nEmilio_Botin\t1\tLuca_Cordero_di_Montezemolo\t1\nEmilio_Botin\t1\tRobert_Hyatt\t1\nEnrique_Iglesias\t1\tGisele_Bundchen\t2\nEric_Bana\t1\tMike_Sweeney\t1\nEric_Bana\t1\tTab_Turner\t1\nEric_Benet\t1\tMohammad_Mustapha_Miro\t1\nEric_Rosser\t1\tMarc-Andre_Fleury\t2\nEric_Rosser\t1\tPeter_Chan\t1\nErnie_Eves\t2\tHana_Makhmalbaf\t1\nFernando_Vargas\t1\tJulio_Cesar_Franco\t1\nFernando_Vargas\t3\tMartin_Verkerk\t3\nFlavia_Pennetta\t1\tLarry_Flynt\t1\nFlavia_Pennetta\t1\tMicah_Knorr\t1\nFlavia_Pennetta\t1\tPaul_Wellstone\t1\nFrank_Keating\t1\tPaul_McCartney\t7\nFrank_Lautenberg\t2\tMichael_Capellas\t1\nFrank_Lautenberg\t2\tSteve_Allee\t1\nFujio_Mitarai\t1\tHarvey_Weinstein\t1\nGabriella_Bo\t1\tJeremy_Shockey\t2\nGavyn_Arthur\t1\tKevin_Hearn\t1\nGavyn_Arthur\t1\tLynne_Slepian\t1\nGavyn_Arthur\t1\tZahir_Shah\t1\nGeorge_HW_Bush\t7\tGeorge_Maxwell_Richards\t1\nGeorge_HW_Bush\t13\tPenny_Lancaster\t1\nGeorge_Maxwell_Richards\t1\tReggie_Miller\t1\nGeorgina_Papin\t1\tMary_Lou_Retton\t1\nGeorgina_Papin\t1\tSvend_Robinson\t1\nGerry_Kelly\t1\tKeira_Knightley\t1\nGiannina_Facio\t1\tPeri_Gilpin\t1\nGideon_Black\t1\tKevin_Hearn\t1\nGideon_Black\t1\tShafal_Mosed\t1\nGisele_Bundchen\t2\tPenny_Lancaster\t1\nGiuseppe_Morchio\t1\tPenny_Lancaster\t1\nGreg_Frers\t1\tMohammed_Al-Douri\t4\nGreg_Gilbert\t1\tMehmet_Ali_Sahin\t1\nGregorio_Honasan\t1\tJim_Schwarz\t1\nGregorio_Honasan\t1\tJonathan_Fine\t1\nGregorio_Honasan\t1\tMatt_Doherty\t1\nHama_Arba_Diallo\t1\tZhang_Yimou\t1\nHana_Makhmalbaf\t1\tRaquel_Welch\t1\nHassan_Wirajuda\t1\tScott_Dalton\t1\nHector_Babenco\t2\tTalisa_Bratt\t1\nHerb_Ritts\t1\tMark_Broxmeyer\t1\nHugh_Campbell\t1\tKen_Watanabe\t1\nHugh_Campbell\t1\tMasamori_Tokuyama\t1\nHugh_Campbell\t1\tTom_Foy\t1\nImran_Khan\t1\tMatt_Morris\t1\nImran_Khan\t1\tNancy_Humbert\t1\nIvan_Lee\t1\tJennifer_Furminger\t1\nJames_Hallock\t1\tRobert_Beck\t1\nJames_Murdoch\t1\tMartina_McBride\t3\nJamie_Kellner\t1\tTomas_Malik\t1\nJamie_King\t1\tLauren_Hutton\t2\nJane_Fonda\t2\tMartha_Stewart\t4\nJanet_Leigh\t1\tMarianne_Stanley\t1\nJaromir_Jagr\t1\tTroy_Jenkins\t1\nJean_Brumley\t2\tJim_Harrick\t1\nJen_Bice\t1\tJudi_Dench\t2\nJennifer_Furminger\t1\tMatthew_During\t1\nJennifer_Tilly\t1\tSerge_Klarsfeld\t1\nJennifer_Tilly\t1\tStella_Tennant\t1\nJeremy_Shockey\t2\tRobert_Beck\t1\nJerry_Regier\t2\tPaulina_Rodriguez_Davila\t1\nJerry_Regier\t3\tLee_Hyung-taik\t1\nJerry_Regier\t3\tPete_Rose\t2\nJessica_Capshaw\t1\tPenelope_Cruz\t2\nJiang_Zemin\t4\tMichelangelo_Antonioni\t1\nJiang_Zemin\t13\tRoman_Coppola\t1\nJim_Sterk\t1\tJonathan_Edwards\t3\nJim_Sterk\t1\tKen_Watanabe\t2\nJoey_Harrington\t1\tNick_Price\t1\nJohn_Lynch\t1\tMichael_Sullivan\t1\nJohn_Robbins\t1\tTom_Foy\t1\nJohn_Thune\t1\tMilan_Milutinovic\t1\nJohn_Velazquez\t1\tMarco_Antonio_Barrera\t3\nJohn_Velazquez\t1\tTammy_Lynn_Michaels\t1\nJohn_Wolf\t2\tPrince_Harry\t1\nJonathan_Byrd\t1\tMike_Matheny\t1\nJonathan_Byrd\t1\tSheikh_Ahmed_Yassin\t1\nJonathan_Horton\t1\tLee_Ann_Terlaji\t1\nJorge_Alberto_Galindo\t1\tRoger_King\t1\nJoseph_Biden\t5\tMahdi_Al_Bassam\t1\nJoseph_Biden\t5\tTung_Chee-hwa\t4\nJudy_Genshaft\t1\tShane_Warne\t1\nJules_Asner\t1\tRobert_Beck\t1\nJulien_Varlet\t1\tScott_Dalton\t1\nKate_Burton\t1\tMartha_Stewart\t2\nKate_Richardson\t1\tRoman_Coppola\t1\nKatie_Boone\t1\tUthai_Pimchaichon\t1\nKatie_Smith\t1\tLeland_Chapman\t1\nKatrin_Susi\t1\tLuca_Cordero_di_Montezemolo\t1\nKay_Bailey_Hutchison\t1\tTheodore_Tweed_Roosevelt\t3\nKieran_Prendergast\t1\tLachlan_Murdoch\t1\nKing_Abdullah_II\t4\tRobert_Hyatt\t1\nKirk_Ferentz\t2\tMilan_Milutinovic\t1\nLachlan_Murdoch\t1\tMickey_Sherman\t1\nLarry_Flynt\t1\tPhil_Cline\t1\nLarry_Greene\t1\tMike_Brey\t2\nLaurence_Tribe\t1\tUthai_Pimchaichon\t1\nLee_Ann_Terlaji\t1\tRoman_Coppola\t1\nLee_Yuan-tseh\t1\tShafal_Mosed\t1\nLee_Yuan-tseh\t1\tYuri_Malenchenko\t2\nLeland_Chapman\t1\tTheodore_Tweed_Roosevelt\t1\nLeslie_Moonves\t2\tRobert_Hyatt\t1\nLisa_Murkowski\t1\tSerge_Klarsfeld\t1\nMagui_Serna\t1\tMaura_Tierney\t1\nMahdi_Al_Bassam\t1\tMarc-Andre_Fleury\t2\nMahdi_Al_Bassam\t1\tMother_Teresa\t1\nMalak_Habbak\t1\tPrince_Harry\t3\nManuel_Pellegrini\t1\tRolf_Eckrodt\t1\nManuela_Montebrun\t1\tMartina_McBride\t2\nManuela_Montebrun\t1\tSteven_Seagal\t2\nMarion_Fahnestock\t1\tWen_Jiabao\t3\nMark_Hamister\t1\tXiang_Xu\t1\nMartha_Martinez_Flores\t1\tMike_Brey\t2\nMartha_Stewart\t5\tVictor_Kraatz\t1\nMartin_Lawrence\t1\tMickey_Sherman\t1\nMartin_Lawrence\t1\tPenny_Lancaster\t1\nMartin_McCauley\t2\tSteven_Seagal\t1\nMartin_Verkerk\t1\tSteven_Seagal\t1\nMartina_McBride\t3\tPeter_Schultz\t1\nMasamori_Tokuyama\t1\tRegina_Ip\t1\nMasamori_Tokuyama\t1\tSerge_Melac\t1\nMatt_Doherty\t2\tSaoud_Al_Faisal\t1\nMeghann_Shaughnessy\t1\tWilliam_Perry\t1\nMehmet_Ali_Sahin\t1\tPeri_Gilpin\t1\nMehmet_Ali_Sahin\t1\tRaquel_Welch\t2\nMichael_Capellas\t1\tVictor_Garber\t1\nMickey_Sherman\t1\tRoman_Coppola\t1\nMike_Brey\t2\tNick_Markakis\t1\nMike_Johanns\t1\tPascal_Lamy\t1\nMike_Matheny\t1\tNorm_Coleman\t5\nMohammad_Mustapha_Miro\t1\tSteve_Mariucci\t2\nMother_Teresa\t1\tYekaterina_Guseva\t1\nNancy_Humbert\t1\tPark_Na-kyong\t1\nNanni_Moretti\t1\tShigeru_Ishiba\t1\nNatalia_Dmitrieva\t1\tWillie_Nelson\t1\nNatalie_Juniardi\t1\tUthai_Pimchaichon\t1\nNick_Price\t1\tTab_Turner\t1\nNorm_Coleman\t4\tPeri_Gilpin\t1\nNorm_Coleman\t7\tScott_McClellan\t5\nPaddy_Torsney\t1\tTung_Chee-hwa\t9\nPark_Na-kyong\t1\tVecdi_Gonul\t1\nPatrick_Rafter\t1\tPeter_Care\t1\nPierre_Van_Hooijdonk\t1\tScott_McClellan\t1\nRegina_Ip\t1\tRohman_al-Ghozi\t1\nRenee_Zellweger\t9\tYuri_Malenchenko\t2\nRolf_Eckrodt\t2\tTeresa_Heinz_Kerry\t1\nRubens_Barrichello\t10\tStella_Tennant\t1\nSaoud_Al_Faisal\t1\tVecdi_Gonul\t1\nScott_Dalton\t1\tZach_Safrin\t1\nScott_McClellan\t2\tTom_Schnackenberg\t1\nTodd_MacCulloch\t1\tWillie_Nelson\t1\nTom_Curley\t1\tWanda_de_la_Jesus\t1\nTroy_Jenkins\t1\tWalid_Al-Awadi\t1\nWilliam_Overlin\t1\tYekaterina_Guseva\t1\nAbdoulaye_Wade\t1\t2\nAbdoulaye_Wade\t1\t3\nAbdoulaye_Wade\t2\t3\nAdam_Sandler\t1\t2\nAdam_Sandler\t1\t4\nAdam_Sandler\t2\t3\nAicha_El_Ouafi\t1\t3\nAicha_El_Ouafi\t2\t3\nAkbar_Hashemi_Rafsanjani\t1\t3\nAkbar_Hashemi_Rafsanjani\t2\t3\nAl_Pacino\t1\t2\nAl_Pacino\t1\t3\nAlex_Barros\t1\t2\nAllyson_Felix\t1\t3\nAllyson_Felix\t1\t4\nAllyson_Felix\t1\t5\nAllyson_Felix\t4\t5\nAnastasia_Myskina\t1\t2\nAndy_Roddick\t8\t12\nAndy_Roddick\t10\t15\nAndy_Roddick\t13\t15\nAnna_Nicole_Smith\t1\t2\nAntonio_Palocci\t1\t8\nAntonio_Palocci\t3\t6\nAntonio_Palocci\t4\t5\nAntonio_Palocci\t5\t7\nAntonio_Palocci\t6\t8\nArnoldo_Aleman\t1\t3\nArnoldo_Aleman\t3\t5\nAshton_Kutcher\t1\t3\nAshton_Kutcher\t2\t3\nAugusto_Roa_Bastos\t1\t2\nAung_San_Suu_Kyi\t1\t2\nBarry_Zito\t1\t2\nBill_Graham\t1\t9\nBill_Graham\t3\t4\nBill_Graham\t4\t6\nBill_Graham\t5\t6\nBob_Dole\t1\t3\nBruce_Weber\t1\t2\nCarlos_Mesa\t1\t2\nCarolyn_Dawn_Johnson\t1\t2\nCarolyn_Dawn_Johnson\t2\t3\nCeline_Dion\t3\t8\nChakib_Khelil\t1\t2\nChen_Shui-bian\t2\t4\nChen_Shui-bian\t3\t5\nChristopher_Walken\t1\t3\nChristopher_Walken\t1\t4\nClaudia_Pechstein\t1\t2\nClaudia_Pechstein\t1\t4\nClaudia_Pechstein\t3\t4\nClaudia_Pechstein\t3\t5\nClaudia_Pechstein\t4\t5\nClay_Aiken\t2\t4\nClay_Aiken\t3\t4\nClay_Aiken\t3\t5\nColin_Powell\t40\t71\nColin_Powell\t49\t234\nColin_Powell\t133\t170\nColin_Powell\t182\t198\nCristina_Fernandez\t1\t2\nDaisy_Fuentes\t2\t3\nDamon_van_Dam\t1\t2\nDan_Wheldon\t1\t2\nDavid_Coulthard\t1\t2\nDavid_Kelley\t1\t2\nDebra_Brown\t1\t2\nDennis_Erickson\t1\t2\nDerek_Lowe\t1\t2\nEddie_Sutton\t1\t2\nEdie_Falco\t1\t2\nElijah_Wood\t2\t3\nElizabeth_Hurley\t1\t4\nElizabeth_Hurley\t2\t5\nEmily_Robison\t1\t2\nEthan_Hawke\t1\t4\nEunice_Barber\t1\t2\nFelix_Mantilla\t1\t2\nFidel_Castro\t1\t18\nFidel_Castro\t3\t7\nFidel_Castro\t5\t8\nFidel_Castro\t8\t12\nFidel_Castro\t11\t13\nFrancisco_Flores\t1\t2\nFrancisco_Flores\t1\t3\nFrank_Dunham_Jr\t1\t2\nFranko_Simatovic\t1\t2\nFred_Eckhard\t1\t2\nFred_Eckhard\t1\t3\nFred_Eckhard\t2\t3\nGL_Peiris\t1\t2\nGL_Peiris\t1\t3\nGL_Peiris\t2\t3\nGL_Peiris\t2\t4\nGarry_Kasparov\t1\t2\nHassan_Nasrallah\t1\t2\nHeidi_Klum\t1\t3\nHeidi_Klum\t1\t4\nHeidi_Klum\t2\t4\nHeidi_Klum\t3\t4\nHeinz_Feldmann\t1\t2\nHeinz_Feldmann\t2\t3\nIban_Mayo\t1\t2\nImad_Moustapha\t1\t2\nInam-ul-Haq\t1\t2\nJames_Gandolfini\t1\t3\nJames_Gandolfini\t2\t3\nJanet_Thorpe\t1\t2\nJean-Pierre_Raffarin\t1\t2\nJean-Pierre_Raffarin\t1\t6\nJean-Pierre_Raffarin\t3\t4\nJean-Pierre_Raffarin\t3\t5\nJean-Pierre_Raffarin\t4\t7\nJean-Pierre_Raffarin\t5\t7\nJean-Pierre_Raffarin\t6\t7\nJeffrey_Scott_Postell\t1\t2\nJennifer_Capriati\t2\t14\nJennifer_Capriati\t7\t32\nJennifer_Capriati\t33\t42\nJob_Cohen\t1\t2\nJohn_McCormack\t1\t2\nJohn_Paul_II\t1\t4\nJohn_Paul_II\t2\t5\nJohn_Paul_II\t2\t8\nJohn_Paul_II\t4\t9\nJohn_Paul_II\t10\t11\nJohn_Ruiz\t1\t2\nJohn_Stallworth\t1\t2\nJohn_Stockton\t2\t4\nJohn_Travolta\t2\t6\nJohn_Travolta\t3\t5\nJohn_Travolta\t5\t7\nJonathan_Mostow\t1\t2\nJorge_Arce\t1\t2\nJoschka_Fischer\t1\t10\nJoschka_Fischer\t6\t11\nJoschka_Fischer\t7\t11\nJoschka_Fischer\t11\t17\nJoschka_Fischer\t15\t16\nJose_Canseco\t1\t3\nJuan_Manuel_Marquez\t1\t2\nJuan_Manuel_Marquez\t1\t3\nJuan_Manuel_Marquez\t2\t3\nJuan_Valencia_Osorio\t1\t2\nJulie_Gerberding\t9\t13\nJulie_Gerberding\t12\t15\nKate_Hudson\t1\t4\nKate_Hudson\t1\t8\nKate_Hudson\t2\t3\nKate_Hudson\t4\t9\nKate_Hudson\t6\t7\nKemal_Dervis\t1\t3\nKemal_Dervis\t2\t3\nKenneth_Evans\t1\t2\nKifah_Ajouri\t1\t2\nLarry_Lucchino\t1\t2\nLatrell_Sprewell\t1\t2\nLech_Walesa\t1\t2\nLee_Tae-sik\t1\t2\nLisa_Marie_Presley\t1\t3\nLiza_Minnelli\t2\t3\nLiza_Minnelli\t3\t4\nLiza_Minnelli\t3\t6\nLiza_Minnelli\t4\t5\nLiza_Minnelli\t5\t6\nLiza_Minnelli\t6\t7\nMadonna\t1\t4\nMadonna\t2\t3\nMadonna\t4\t5\nMariah_Carey\t3\t6\nMary_Tyler_Moore\t1\t2\nMathias_Reichhold\t1\t2\nMatt_Damon\t1\t2\nMatt_Damon\t1\t3\nMatt_Damon\t2\t4\nMatt_Damon\t3\t4\nMaureen_Fanning\t1\t2\nMelanie_Griffith\t1\t2\nMelanie_Griffith\t1\t3\nMelanie_Griffith\t2\t3\nMichael_Ballack\t1\t2\nMichael_Winterbottom\t1\t3\nMichael_Winterbottom\t2\t3\nMichelle_Collins\t1\t2\nMilo_Maestrecampo\t1\t2\nMohamed_Benaissa\t1\t2\nMohamed_ElBaradei\t2\t4\nMohamed_ElBaradei\t3\t8\nMorgan_Freeman\t1\t2\nMuhammad_Ali\t1\t3\nMuhammad_Ali\t1\t7\nMuhammad_Ali\t2\t5\nMuhammad_Ali\t6\t10\nMuhammad_Ali\t7\t9\nMukesh_Ambani\t1\t2\nMukesh_Ambani\t1\t3\nParis_Hilton\t1\t2\nPat_Cox\t1\t2\nPaul_Burrell\t3\t7\nPaul_Burrell\t5\t11\nPaul_Burrell\t8\t10\nPaul_Wolfowitz\t8\t10\nPaula_Radcliffe\t1\t2\nPaula_Radcliffe\t1\t3\nPaula_Radcliffe\t2\t3\nPaula_Radcliffe\t2\t4\nPaula_Radcliffe\t3\t4\nPaula_Radcliffe\t3\t5\nPaula_Radcliffe\t4\t5\nPaulo_Cesar_Pinheiro\t1\t2\nPedro_Solbes\t1\t3\nPedro_Solbes\t1\t4\nPedro_Solbes\t2\t3\nPedro_Solbes\t3\t4\nPete_Carroll\t1\t2\nPete_Carroll\t1\t3\nPete_Carroll\t2\t3\nPete_Sampras\t2\t12\nPete_Sampras\t2\t13\nPete_Sampras\t3\t15\nPete_Sampras\t4\t20\nPete_Sampras\t6\t7\nPete_Sampras\t6\t8\nPete_Sampras\t10\t13\nPete_Sampras\t12\t15\nPeter_Struck\t1\t5\nPeter_Struck\t2\t5\nPhil_Vassar\t1\t2\nPierre_Boulanger\t1\t2\nPrince_Willem-Alexander\t1\t2\nPrince_Willem-Alexander\t1\t3\nPrince_Willem-Alexander\t2\t3\nQueen_Elizabeth_II\t3\t7\nQueen_Elizabeth_II\t9\t12\nQueen_Elizabeth_II\t10\t11\nQueen_Elizabeth_II\t10\t12\nRay_Nagin\t1\t2\nRicardo_Maduro\t1\t2\nRichard_Branson\t1\t2\nRichard_Virenque\t1\t4\nRichard_Virenque\t1\t6\nRichard_Virenque\t1\t7\nRichard_Virenque\t2\t7\nRichard_Virenque\t2\t8\nRick_Carlisle\t2\t4\nRick_Wagoner\t1\t2\nRobbie_Williams\t1\t2\nRobbie_Williams\t1\t3\nRoberto_Carlos\t2\t3\nRoberto_Carlos\t2\t4\nRoseanne_Barr\t1\t2\nRoseanne_Barr\t2\t3\nRuben_Studdard\t1\t2\nSammy_Sosa\t1\t2\nSarah_Jessica_Parker\t1\t3\nSarah_Jessica_Parker\t2\t4\nSarah_Jessica_Parker\t3\t4\nSharon_Davis\t1\t2\nShaul_Mofaz\t1\t2\nShaul_Mofaz\t2\t3\nStan_Heath\t1\t2\nSvetlana_Koroleva\t1\t2\nTerrell_Suggs\t1\t2\nTim_Henman\t2\t12\nTim_Henman\t8\t19\nTom_Daschle\t7\t8\nTom_Daschle\t15\t21\nTom_Daschle\t15\t22\nTony_Curtis\t1\t2\nValentino_Rossi\t1\t2\nValentino_Rossi\t2\t4\nValentino_Rossi\t3\t6\nValentino_Rossi\t4\t5\nValentino_Rossi\t5\t6\nVanessa_Redgrave\t1\t3\nVanessa_Redgrave\t1\t4\nVanessa_Redgrave\t2\t5\nVanessa_Redgrave\t3\t4\nVictoria_Clarke\t1\t5\nVladimiro_Montesinos\t1\t2\nVladimiro_Montesinos\t1\t3\nVladimiro_Montesinos\t2\t3\nWayne_Ferreira\t1\t2\nWayne_Ferreira\t1\t3\nWayne_Ferreira\t1\t5\nWayne_Ferreira\t2\t5\nWayne_Ferreira\t3\t4\nWill_Smith\t1\t2\nYasser_Arafat\t1\t6\nYasser_Arafat\t1\t8\nYasser_Arafat\t2\t3\nYasser_Arafat\t2\t5\nYasser_Arafat\t3\t4\nYasser_Arafat\t3\t8\nYasser_Arafat\t4\t5\nYasser_Arafat\t5\t8\nYuri_Fedotov\t1\t2\nZoran_Djindjic\t1\t3\nZoran_Djindjic\t1\t4\nAaron_Patterson\t1\tFrank_Bell\t1\nAbdoulaye_Wade\t4\tBruce_Weber\t2\nAbner_Martinez\t1\tCarlos_Alberto\t1\nAdam_Sandler\t2\tMatthew_Ouimet\t1\nAdam_Sandler\t3\tSaeed_Anwar\t1\nAdolfo_Aguilar_Zinser\t3\tJaime_Pressly\t1\nAgnelo_Queiroz\t1\tAung_San_Suu_Kyi\t2\nAgnelo_Queiroz\t1\tDave_Barr\t1\nAicha_El_Ouafi\t3\tMichael_Lechner\t1\nAkbar_Hashemi_Rafsanjani\t1\tLarry_Harris\t1\nAl_Pacino\t2\tCharles_Cope\t1\nAlex_Barros\t1\tBrandon_Jones\t1\nAlex_Barros\t2\tWill_Smith\t2\nAlex_Ferguson\t1\tRainer_Gut\t1\nAlex_Wallau\t1\tShireen_Amir_Begum\t1\nAlexandra_Jackson\t1\tLarry_Harris\t1\nAlfonso_Portillo\t1\tBenito_Santiago\t1\nAlfonso_Portillo\t1\tFaye_Alibocus\t1\nAlfonso_Portillo\t1\tFidel_Castro\t17\nAli_Abdullah_Saleh\t1\tKhalid_Qazi\t1\nAllan_Houston\t1\tAndy_Garcia\t1\nAllan_Houston\t1\tHeidi_Klum\t1\nAllan_Houston\t1\tThomas_Mesereau_Jr\t1\nAlly_Sheedy\t1\tHugh_Carey\t1\nAlly_Sheedy\t1\tMyung_Yang\t1\nAmanda_Marsh\t1\tTony_Curtis\t2\nAnastasia_Myskina\t1\tRaul_Gonzalez\t1\nAnastasia_Myskina\t3\tLen_Jenoff\t2\nAndrzej_Tyszkiewicz\t1\tWes_Craven\t1\nAndy_Griggs\t1\tLech_Walesa\t2\nAndy_Rooney\t1\tJessica_Simpson\t1\nAnna_Nicole_Smith\t2\tMarcus_Garrettson\t1\nAntonio_Palocci\t3\tLiza_Minnelli\t1\nAntonio_Palocci\t5\tJC_Chasez\t1\nAntonio_Palocci\t5\tJose_Woldenberg\t1\nAntonio_Palocci\t6\tJohn_Geoghan\t1\nAntonio_Palocci\t8\tHans_Corell\t1\nArif_Mardin\t1\tEduardo_Fischer\t1\nArnaud_Lagardere\t1\tMelanie_Griffith\t3\nAshton_Kutcher\t2\tDaniel_Barenboim\t1\nAsif_Hanif\t1\tRobbie_Williams\t1\nAsmaa_Assad\t1\tBarry_Hinson\t1\nAung_San_Suu_Kyi\t1\tCharla_Moye\t1\nAzmi_Bishara\t1\tSammy_Sosa\t2\nBarry_Hinson\t1\tNino_DAngelo\t1\nBarry_Zito\t2\tChris_Gratton\t1\nBill_Graham\t8\tMichelle_Hofland\t1\nBill_Graham\t9\tJacqueline_Marris\t1\nBill_Readdy\t1\tBrendan_Gaughan\t1\nBill_Readdy\t1\tJaymon_Crabb\t1\nBill_Readdy\t1\tYasser_Arafat\t3\nBilly_Rork\t1\tEva_Mendes\t1\nBilly_Rork\t1\tGerman_Khan\t1\nBilly_Rork\t1\tPeter_Struck\t2\nBison_Dele\t1\tBrian_McIntyre\t1\nBob_Dole\t2\tDai_Chul_Chyung\t1\nBob_Dole\t2\tJohn_Henry\t1\nBob_Dole\t3\tChris_Gratton\t1\nBob_Dole\t3\tHugh_Carey\t1\nBob_Geldof\t1\tZoran_Djindjic\t2\nBob_Geldof\t2\tEd_Wade\t1\nBob_Holden\t1\tFernando_Leon_de_Aranoa\t1\nBob_Iger\t1\tEdie_Falco\t1\nBob_Iger\t1\tJean-Claude_Van_Damme\t1\nBrian_Clemens\t1\tBrian_Meadors\t1\nBrian_Clemens\t1\tMelanie_Griffith\t2\nBrian_Heidik\t2\tDjabir_Said-Guerni\t1\nBrian_McIntyre\t1\tHans_Corell\t1\nBrian_McIntyre\t1\tMohammed_Abulhasan\t1\nBrian_Pavlich\t1\tRuben_Wolkowyski\t1\nBrook_Robinson\t1\tTom_McClintock\t1\nBrooke_Adams\t1\tPaula_Prentiss\t1\nBrooke_Gordon\t1\tJoschka_Fischer\t12\nBruce_Weber\t1\tHal_Sellers\t1\nBryan_Thomas\t1\tJoey_Mantia\t1\nBustam_A_Zedan_Aljanabi\t1\tKajsa_Bergqvist\t1\nCalvin_Joseph_Coleman\t1\tHassan_Nasrallah\t1\nCarla_Sullivan\t1\tEdie_Falco\t1\nCarlos_Barragan\t1\tChen_Shui-bian\t3\nCarlos_Salinas\t1\tNorman_Mailer\t1\nCarlos_Salinas\t1\tSonya_Walger\t1\nCarolyn_Dawn_Johnson\t3\tLydia_Shum\t1\nCarolyn_Kuhl\t1\tPierre_Boulanger\t1\nCeline_Dion\t2\tJohn_Stallworth\t2\nCeline_Dion\t5\tLinda_Baboolal\t1\nCeline_Dion\t5\tTom_Poston\t1\nChakib_Khelil\t2\tChuck_Woolery\t1\nCharla_Moye\t1\tPatti_Balgojevich\t1\nCharles_Cope\t1\tGarry_Alejano\t1\nCharles_Holzner\t1\tEurico_Guterres\t1\nCharles_Holzner\t1\tGreg_Kinnear\t1\nChen_Shui-bian\t3\tGaston_Gaudio\t1\nChris_Gratton\t1\tMario_Vasquez_Rana\t1\nChris_Kolanas\t1\tJoshua_Gracin\t1\nClaudia_Pechstein\t2\tMireille_Jospin-Dandieu\t1\nClay_Aiken\t1\tSvetlana_Koroleva\t1\nColin_Powell\t95\tFrank_Hsieh\t1\nCraig_David\t1\tTom_McClintock\t1\nCraig_Wilson\t1\tKajsa_Bergqvist\t1\nCristina_Fernandez\t2\tStephen_Cooper\t1\nCurtis_Joseph\t1\tTerrell_Suggs\t2\nCynthia_Rowley\t1\tMichael_Friedman\t1\nDamon_van_Dam\t2\tJason_Sorens\t1\nDaniel_Barenboim\t1\tTyler_Grillo\t1\nDaniel_Bruehl\t1\tGus_Frerotte\t1\nDaniel_Bruehl\t1\tMax_Mosley\t1\nDaniel_Bruehl\t1\tRamon_Cardenas\t1\nDaniele_Bergamin\t1\tKenneth_Evans\t1\nDanielle_Spencer\t1\tRachel_Wheatley\t1\nDarcy_Regier\t1\tWilliam_Hurt\t1\nDave_Matthews\t1\tLinda_Dano\t1\nDave_Matthews\t1\tPaul_Burrell\t7\nDave_McNealey\t1\tGerald_Riley\t1\nDavid_Bisbal\t1\tTerri_Clark\t1\nDavid_Chase\t1\tEkaterina_Dmitriev\t1\nDavid_McCullough\t1\tEvo_Morales\t1\nDavid_McKiernan\t1\tFernando_Leon_de_Aranoa\t1\nDavid_McKiernan\t1\tJane_Leeves\t1\nDennis_Erickson\t2\tHisham_Halawi\t1\nDennis_Erickson\t2\tNoer_Muis\t1\nDerek_Lowe\t2\tHisham_Halawi\t1\nDerek_Lowe\t2\tMadonna\t1\nDerek_Lowe\t2\tMohamed_Benaissa\t1\nDiego_Colorado\t1\tJason_Sorens\t1\nDiego_Colorado\t1\tJohn_Gordnick\t1\nDiego_Diego_Lerman\t1\tFrancisco_Flores\t2\nDiego_Diego_Lerman\t1\tLesley_Coppin\t1\nDjabir_Said-Guerni\t1\tPaul_Tracy\t1\nDjabir_Said-Guerni\t1\tPedro_Solbes\t3\nDominik_Hrbaty\t1\tToshi_Izawa\t1\nDonald_Keck\t1\tKyoko_Nakayama\t1\nDonald_Keck\t1\tTom_Poston\t1\nDonna_Walker\t1\tJacqueline_Marris\t1\nDragan_Covic\t1\tTodd_Reid\t1\nDudley_Rogers\t1\tSyed_Ibrahim\t1\nDunn_Lampton\t1\tJessica_Simpson\t1\nDustin_Hoffman\t1\tNicole_Parker\t1\nEd_Wade\t1\tRoger_Staubach\t1\nEd_Wade\t1\tTerri_Clark\t1\nEddie_Lucio\t1\tPatti_Balgojevich\t1\nEddie_Sutton\t1\tJames_Wattana\t1\nEddie_Sutton\t1\tJeanne_Anne_Schroeder\t1\nEddie_Sutton\t2\tRonald_Ito\t1\nEduardo_Fischer\t1\tKimberly_Bruckner\t1\nEdward_Lohn\t1\tLily_Safra\t1\nEdward_Lohn\t1\tNino_DAngelo\t1\nEkaterina_Dmitriev\t1\tMitch_Kupchak\t1\nEladio_Larez\t1\tFrank_Pallone\t1\nEli_Broad\t1\tRavan_AG_Farhadi\t1\nElijah_Wood\t2\tStefano_Gabbana\t1\nElijah_Wood\t3\tMarcus_Garrettson\t1\nElijan_Ingram\t1\tMichelle_Hofland\t1\nElijan_Ingram\t1\tNastia_Liukin\t1\nElvis_Costello\t1\tJaime_Pressly\t1\nEmelie_Loit\t1\tThomas_Mesereau_Jr\t1\nEric_Staal\t1\tJerry_Lewis\t1\nErin_Hershey_Presley\t1\tFrank_Dunham_Jr\t1\nErin_Hershey_Presley\t1\tYukio_Hatoyama\t1\nEva_Mendes\t1\tLarry_Harris\t1\nFaye_Alibocus\t1\tFrank_Bell\t1\nFaye_Alibocus\t1\tTommy_Tubberville\t1\nFelix_Mantilla\t1\tJerry_Lewis\t1\nFernando_Leon_de_Aranoa\t1\tJohn_Scarlett\t1\nFernando_Leon_de_Aranoa\t1\tMike_Leach\t1\nFidel_Castro\t7\tHu_Maoyuan\t1\nFrancisco_Flores\t4\tHisham_Halawi\t1\nFrank_Bell\t1\tKhatol_Mohammad_Zai\t1\nFrank_Dunham_Jr\t1\tTom_McClintock\t1\nFrank_Hsieh\t1\tPaula_Prentiss\t1\nFrank_Pallone\t1\tJim_Wall\t1\nFrank_Pallone\t1\tMary_Tyler_Moore\t2\nFrank_Schmoekel\t1\tRonald_Ito\t1\nFranko_Simatovic\t2\tTyler_Grillo\t1\nFranz_Gsell\t1\tJohn_Scarlett\t1\nFranz_Gsell\t1\tSarah_Jessica_Parker\t2\nFred_Eckhard\t1\tRachel_Wheatley\t1\nGarry_Alejano\t1\tIban_Mayo\t1\nGarry_Alejano\t1\tMichael_Olowokandi\t1\nGarry_Alejano\t1\tMorgan_Freeman\t2\nGarry_Alejano\t1\tPeter_Struck\t5\nGary_Marshall\t1\tRashid_Qureshi\t1\nGerald_Fitch\t1\tRobin_Williams\t1\nGerald_Riley\t1\tMichael_Hagee\t1\nGerman_Khan\t1\tMorgan_Freeman\t2\nGrady_Little\t1\tRobert_Morvillo\t1\nHal_Sellers\t1\tJanet_Thorpe\t1\nHans_Corell\t1\tJohn_Gordnick\t1\nHeidi_Klum\t2\tPaul_Reiser\t1\nHeidi_Klum\t3\tPedro_Solbes\t1\nHermando_Harton\t1\tParis_Hilton\t2\nHolly_Robinson_Peete\t1\tMichael_Hagee\t1\nHoward_Ross\t1\tKajsa_Bergqvist\t1\nHoward_Ross\t1\tPatsy_Hardy\t1\nHugh_Carey\t1\tLawrence_Roberts\t1\nHugh_Carey\t1\tTracee_Treadwell\t1\nIban_Mayo\t2\tJuan_Valencia_Osorio\t1\nIban_Mayo\t2\tRosalie_Perkov\t1\nIbrahim_Al-Marashi\t1\tJohn_Travolta\t3\nIbrahim_Al-Marashi\t1\tJoshua_Gracin\t1\nImad_Moustapha\t2\tKyoko_Nakayama\t1\nIra_Einhorn\t1\tJohn_Ruiz\t2\nJacqueline_Marris\t1\tMatthew_Ouimet\t1\nJaime_Pressly\t1\tTiago_Splitter\t1\nJames_Brown\t1\tMary_Blige\t1\nJames_Brown\t1\tPaul_Reiser\t1\nJames_Mathis\t1\tYuri_Fedotov\t1\nJames_Wattana\t1\tPaul_Reiser\t1\nJames_Wattana\t1\tRon_Kirk\t1\nJames_Young\t1\tJeffrey_Scott_Postell\t2\nJames_Young\t1\tRonald_Harwood\t1\nJane_Leeves\t1\tTom_McClintock\t1\nJanet_Crawford\t1\tThomas_Scavone\t1\nJason_Sorens\t1\tLidija_Djukanovic\t1\nJason_Vale\t1\tMarcus_Garrettson\t1\nJason_Vale\t1\tZoran_Djindjic\t3\nJaymon_Crabb\t1\tTed_Christopher\t1\nJeff_Weaver\t1\tKyoko_Nakayama\t1\nJeff_Weaver\t1\tRosalie_Perkov\t1\nJeffrey_Scott_Postell\t1\tTara_Dawn_Christensen\t1\nJeffrey_Scott_Postell\t2\tLesley_Coppin\t1\nJennifer_Capriati\t26\tRuben_Wolkowyski\t1\nJim_Bunning\t1\tTerence_Newman\t1\nJim_Jeffords\t1\tPeter_Rasch\t1\nJim_Jeffords\t1\tTom_Sizemore\t1\nJob_Cohen\t1\tJohn_Stockton\t2\nJohn_Franco\t1\tPaul_Wolfowitz\t4\nJohn_Gordnick\t1\tLinda_Dano\t1\nJohn_Gordnick\t1\tPaul_Vathis\t1\nJohn_Gordnick\t1\tYasser_Arafat\t3\nJohn_Stockton\t4\tPatsy_Hardy\t1\nJohn_Travolta\t2\tLech_Walesa\t1\nJuan_Valencia_Osorio\t1\tTom_Daschle\t6\nJuergen_Schrempp\t1\tMitt_Romney\t1\nJulien_Boutter\t1\tSaeed_Anwar\t1\nKaisser_Bazan\t1\tLawrence_Vito\t1\nKaisser_Bazan\t1\tPierre_Lacroix\t1\nKajsa_Bergqvist\t1\tTayshaun_Prince\t1\nKeith_Snyder\t1\tTerri_Clark\t1\nKemal_Dervis\t3\tRoger_Staubach\t1\nKifah_Ajouri\t1\tLaurie_Pirtle\t1\nKifah_Ajouri\t1\tRaul_Gonzalez\t1\nKimberly_Bruckner\t1\tPete_Carroll\t1\nKirk_Douglas\t1\tNida_Blanca\t1\nLarry_Harris\t1\tMichael_Lechner\t1\nLarry_Lucchino\t1\tRudy_Tomjanovich\t1\nLatrell_Sprewell\t1\tRosalie_Perkov\t1\nLaura_Schlessinger\t1\tTony_Curtis\t1\nLaurie_Pirtle\t1\tRoberta_Combs\t1\nLawrence_Roberts\t1\tLynne_Thigpen\t1\nLawrence_Roberts\t1\tNick_Cassavetes\t1\nLech_Walesa\t2\tRick_Wagoner\t2\nLen_Jenoff\t2\tTom_Daschle\t7\nLew_Rywin\t1\tTerri_Clark\t1\nLinda_Dano\t1\tTerence_Newman\t1\nLydia_Shum\t1\tMario_Vasquez_Rana\t1\nLynne_Thigpen\t1\tMary_Blige\t1\nMalcolm_Glazer\t1\tNoer_Muis\t1\nMalcolm_Glazer\t1\tTabare_Vazquez\t1\nMarc_Bulger\t1\tPaul_Wolfowitz\t7\nMarcus_Garrettson\t1\tPham_Sy_Chien\t1\nMarcus_Garrettson\t1\tRoberto_Carlos\t3\nMaribel_Dominguez\t1\tMichael_Winterbottom\t3\nMarion_Barry\t1\tSteve_Peace\t1\nMark_Hogan\t1\tQueen_Elizabeth_II\t11\nMark_Hogan\t1\tRoger_Toussaint\t1\nMary_Blige\t1\tRaul_Gonzalez\t1\nMassoud_Barzani\t1\tPierre_Lacroix\t1\nMax_Mosley\t1\tRaul_Gonzalez\t1\nMelvin_Talbert\t1\tRudy_Tomjanovich\t1\nMichael_Hagee\t1\tSonya_Walger\t1\nMichael_Killeen\t1\tSammy_Sosa\t2\nMichael_Olowokandi\t1\tPete_Carroll\t2\nMichael_Winterbottom\t1\tPierre_Boulanger\t1\nMike_Price\t2\tTom_Miller\t1\nMilo_Maestrecampo\t1\tYuri_Fedotov\t1\nMilo_Maestrecampo\t2\tSteve_Peace\t1\nMilo_Maestrecampo\t3\tStacey_Dales-Schuman\t1\nMitt_Romney\t1\tNida_Blanca\t1\nMohamed_ElBaradei\t8\tRachel_Wheatley\t1\nMohammad_Aktar\t1\tRafeeuddin_Ahmed\t1\nMohammad_Aktar\t1\tTom_McClintock\t1\nMohammed_Abulhasan\t1\tStephane_Rousseau\t1\nMohammed_Abulhasan\t1\tVladimiro_Montesinos\t1\nNastia_Liukin\t1\tNicole_Parker\t1\nNastia_Liukin\t1\tSharon_Davis\t2\nNicolas_Latorre\t1\tPaula_Prentiss\t1\nNicolas_Latorre\t1\tVassilis_Xiros\t1\nNicolas_Latorre\t1\tWilliam_Delahunt\t1\nPete_Sampras\t10\tStan_Heath\t1\nPhil_Vassar\t2\tTabare_Vazquez\t1\nPhilip_Zalewski\t1\tStefan_Holm\t1\nRoberto_Carlos\t1\tYasser_Arafat\t5\nRoger_Cook\t1\tWilbert_Foy\t1\nRoger_Staubach\t1\tSeverino_Antinori\t1\nRon_Kirk\t1\tTroy_Hudson\t1\nSammy_Sosa\t2\tStan_Heath\t2\nSandra_Ceccarelli\t1\tStephen_Cooper\t1\nSandra_Ceccarelli\t1\tTom_Coughlin\t1\nShireen_Amir_Begum\t1\tSushma_Swaraj\t1\nSinead_OConnor\t1\tStephane_Rochon\t1\nAitor_Gonzalez\t1\t2\nAlec_Baldwin\t2\t4\nAllison_Janney\t1\t2\nAlvaro_Noboa\t1\t3\nAlvaro_Noboa\t2\t3\nAmanda_Coetzer\t1\t2\nAmer_al-Saadi\t1\t3\nAmer_al-Saadi\t1\t4\nAmer_al-Saadi\t2\t4\nAna_Guevara\t2\t7\nAna_Guevara\t3\t7\nAnneli_Jaatteenmaki\t1\t2\nAri_Fleischer\t6\t9\nAri_Fleischer\t7\t12\nArianna_Huffington\t1\t2\nArianna_Huffington\t1\t4\nArianna_Huffington\t2\t3\nArianna_Huffington\t2\t4\nArnaud_Clement\t1\t2\nArsinee_Khanjian\t1\t2\nArt_Howe\t1\t2\nArt_Howe\t1\t3\nArt_Howe\t1\t4\nArt_Howe\t2\t3\nArt_Howe\t3\t4\nBen_Affleck\t2\t3\nBen_Affleck\t2\t4\nBen_Affleck\t2\t7\nBen_Affleck\t3\t6\nBen_Affleck\t5\t6\nBetsy_Smith\t1\t2\nBill_Callahan\t1\t3\nBlythe_Hartley\t1\t2\nBob_Huggins\t1\t3\nBob_Huggins\t3\t4\nBobby_Goldwater\t1\t2\nBono\t1\t2\nBono\t1\t3\nBono\t2\t3\nBrad_Garrett\t1\t3\nBrad_Garrett\t2\t3\nBrad_Garrett\t2\t4\nBrian_Mulroney\t1\t2\nBud_Selig\t1\t2\nBud_Selig\t2\t3\nBud_Selig\t2\t4\nCarla_Del_Ponte\t1\t2\nCarla_Del_Ponte\t2\t4\nCarla_Del_Ponte\t2\t5\nCarla_Del_Ponte\t3\t5\nCarlos_Ghosn\t1\t2\nCarlos_Manuel_Pruneda\t1\t2\nCarlos_Manuel_Pruneda\t1\t3\nCarlos_Menem\t2\t21\nCarlos_Menem\t5\t18\nCarlos_Menem\t8\t15\nCarlos_Menem\t11\t12\nCarlos_Menem\t14\t21\nCarlos_Menem\t16\t18\nCarmen_Electra\t3\t4\nCarmen_Electra\t4\t6\nCharlotte_Rampling\t1\t2\nChick_Hearn\t1\t2\nChristine_Todd_Whitman\t2\t3\nChristine_Todd_Whitman\t5\t6\nChristopher_Reeve\t1\t3\nChuck_Amato\t1\t2\nCindy_Crawford\t2\t3\nCindy_Margolis\t1\t2\nClaire_Danes\t1\t3\nClaire_Danes\t2\t3\nConan_OBrien\t1\t2\nConan_OBrien\t2\t3\nConan_OBrien\t2\t4\nConchita_Martinez\t1\t2\nConchita_Martinez\t1\t3\nDan_Morales\t1\t2\nDan_Morales\t1\t3\nDavid_Hyde_Pierce\t1\t2\nDavid_Hyde_Pierce\t1\t3\nDavid_Hyde_Pierce\t2\t3\nDavid_Hyde_Pierce\t2\t4\nDavid_Myers\t1\t2\nDavid_Nalbandian\t1\t3\nDavid_Nalbandian\t2\t9\nDavid_Nalbandian\t3\t4\nDavid_Nalbandian\t4\t13\nDavid_Nalbandian\t11\t12\nDavid_Stern\t1\t2\nDavid_Stern\t2\t3\nDerek_Jeter\t2\t3\nDerek_Jeter\t2\t4\nDonatella_Versace\t1\t3\nDonatella_Versace\t2\t3\nDonna_Shalala\t1\t2\nEdmund_Hillary\t1\t2\nEdmund_Hillary\t1\t3\nEdmund_Hillary\t2\t3\nElisabeth_Schumacher\t1\t2\nElizabeth_Smart\t3\t5\nErin_Runnion\t1\t3\nErin_Runnion\t2\t3\nErin_Runnion\t2\t4\nErin_Runnion\t3\t4\nFernando_Henrique_Cardoso\t1\t2\nFernando_Henrique_Cardoso\t1\t3\nFernando_Henrique_Cardoso\t1\t4\nFernando_Henrique_Cardoso\t2\t7\nFernando_Henrique_Cardoso\t5\t7\nFernando_Henrique_Cardoso\t6\t7\nFrancis_Mer\t1\t2\nFranz_Fischler\t1\t2\nGary_Locke\t1\t2\nGerry_Adams\t1\t3\nGerry_Adams\t1\t7\nGerry_Adams\t2\t6\nGerry_Adams\t3\t5\nGerry_Adams\t4\t6\nGerry_Adams\t5\t6\nGianna_Angelopoulos-Daskalaki\t1\t2\nGil_de_Ferran\t1\t5\nGil_de_Ferran\t2\t4\nGil_de_Ferran\t3\t5\nGoh_Kun\t1\t2\nGrady_Irvin_Jr\t1\t2\nGunter_Pleuger\t3\t5\nGunter_Pleuger\t3\t6\nHarry_Belafonte\t1\t2\nHarry_Schmidt\t1\t3\nHarry_Schmidt\t1\t4\nHarry_Schmidt\t2\t3\nHarry_Schmidt\t3\t4\nHelen_Clark\t2\t4\nHermann_Maier\t1\t2\nIan_Thorpe\t1\t7\nIan_Thorpe\t1\t10\nIan_Thorpe\t3\t9\nIan_Thorpe\t5\t7\nIan_Thorpe\t6\t10\nIsabelle_Huppert\t1\t2\nJames_Butts\t1\t2\nJan-Michael_Gambill\t1\t2\nJan-Michael_Gambill\t2\t3\nJean-Francois_Pontal\t1\t3\nJean-Marc_de_La_Sabliere\t1\t2\nJean-Sebastien_Giguere\t1\t2\nJeb_Bush\t1\t6\nJeb_Bush\t2\t6\nJeb_Bush\t5\t7\nJeffrey_Jones\t1\t2\nJim_OBrien\t1\t2\nJim_OBrien\t1\t3\nJoe_Gatti\t1\t2\nJohn_Cusack\t1\t2\nJohn_Edwards\t1\t3\nJohn_Edwards\t1\t7\nJohn_Edwards\t2\t5\nJohn_Edwards\t3\t7\nJohn_Edwards\t4\t6\nJohn_Edwards\t4\t8\nJohn_Edwards\t6\t7\nJohn_Spencer\t1\t2\nJohn_Taylor\t1\t2\nJohn_Walsh\t1\t2\nJolanta_Kwasniewski\t1\t2\nJose_Mourinho\t1\t2\nJuergen_Peters\t1\t2\nKalpana_Chawla\t1\t4\nKalpana_Chawla\t2\t3\nKalpana_Chawla\t2\t5\nKalpana_Chawla\t3\t4\nKelli_White\t1\t2\nKim_Jin-sun\t1\t2\nLana_Clarkson\t1\t2\nLaura_Hernandez\t1\t2\nLaura_Linney\t1\t2\nLaura_Linney\t1\t3\nLaura_Linney\t2\t4\nLaura_Linney\t3\t4\nLenny_Wilkens\t1\t2\nLenny_Wilkens\t1\t3\nLenny_Wilkens\t2\t3\nLloyd_Ward\t1\t2\nMahathir_Mohamad\t1\t7\nMahathir_Mohamad\t1\t12\nMahathir_Mohamad\t6\t10\nMark_Cuban\t1\t2\nMark_Heller\t1\t2\nMark_Schweiker\t1\t2\nMarty_Mornhinweg\t1\t2\nMarty_Mornhinweg\t1\t3\nMarty_Mornhinweg\t2\t3\nMichael_Phelps\t1\t2\nMichael_Phelps\t2\t4\nMick_Jagger\t2\t4\nMiguel_Estrada\t1\t2\nMike_Myers\t1\t6\nMike_Myers\t2\t3\nMike_Myers\t2\t4\nMike_Myers\t2\t5\nNadine_Vinzens\t1\t2\nNathan_Lane\t1\t2\nNicolas_Cage\t1\t2\nNicolas_Cage\t2\t4\nNicolas_Cage\t3\t4\nNicole_Kidman\t3\t11\nNicole_Kidman\t8\t17\nNicole_Kidman\t11\t13\nNicole_Kidman\t13\t14\nNicole_Kidman\t15\t19\nOmar_Sharif\t1\t2\nOmar_Sharif\t1\t3\nOmar_Sharif\t1\t4\nOmar_Sharif\t2\t3\nParris_Glendening\t1\t2\nPatrick_Leahy\t1\t2\nPatrick_McEnroe\t1\t2\nPatrick_Stewart\t1\t2\nPaul_Gascoigne\t1\t3\nPeter_Hillary\t1\t2\nPhan_Van_Khai\t1\t2\nPhan_Van_Khai\t1\t3\nPhan_Van_Khai\t2\t3\nPrince_Claus\t1\t4\nPrince_Claus\t2\t3\nPrince_Claus\t2\t4\nRaghad_Saddam_Hussein\t1\t2\nRay_Allen\t1\t2\nRay_Allen\t2\t3\nRichard_Crenna\t1\t2\nRichard_Gere\t1\t10\nRichard_Gere\t6\t9\nRichard_Myers\t1\t2\nRichard_Myers\t1\t4\nRichard_Myers\t4\t8\nRichard_Myers\t15\t16\nRick_Romley\t1\t3\nRob_Lowe\t1\t3\nRob_Lowe\t1\t4\nRob_Lowe\t2\t3\nRob_Lowe\t3\t4\nRobby_Ginepri\t1\t2\nRobert_Witt\t1\t2\nRod_Blagojevich\t1\t2\nRolandas_Paksas\t1\t2\nRussell_Coutts\t1\t2\nRuth_Dreifuss\t1\t2\nSally_Kirkland\t1\t4\nSean_Patrick_OMalley\t1\t3\nSebastian_Saja\t1\t2\nSebastian_Saja\t1\t3\nSila_Calderon\t1\t2\nSila_Calderon\t1\t3\nSila_Calderon\t1\t4\nSila_Calderon\t2\t3\nStanley_Tong\t1\t2\nSteve_Park\t1\t2\nSusie_Castillo\t1\t2\nSylvester_Stallone\t1\t5\nSylvester_Stallone\t2\t4\nSylvester_Stallone\t3\t8\nSylvester_Stallone\t4\t5\nTakashi_Sorimachi\t1\t2\nTang_Jiaxuan\t2\t6\nTang_Jiaxuan\t2\t11\nTang_Jiaxuan\t4\t9\nTang_Jiaxuan\t5\t9\nTang_Jiaxuan\t7\t10\nTerje_Roed-Larsen\t1\t2\nThomas_Birmingham\t1\t2\nTiger_Woods\t2\t6\nTiger_Woods\t3\t21\nTiger_Woods\t5\t11\nTiger_Woods\t5\t12\nTiger_Woods\t6\t18\nTiger_Woods\t6\t21\nTiger_Woods\t9\t12\nTiger_Woods\t9\t23\nTiger_Woods\t16\t18\nTim_Curry\t1\t2\nTom_Brady\t1\t2\nToshihiko_Fukui\t1\t2\nToshihiko_Fukui\t1\t3\nUma_Thurman\t1\t3\nUma_Thurman\t2\t3\nVaclav_Klaus\t1\t2\nVanessa_Incontrada\t1\t2\nVanessa_Incontrada\t1\t3\nVanessa_Incontrada\t1\t4\nVanessa_Incontrada\t2\t3\nVanessa_Incontrada\t2\t4\nVanessa_Incontrada\t3\t4\nVin_Diesel\t1\t2\nVladimir_Spidla\t1\t2\nVladimir_Spidla\t1\t3\nVladimir_Spidla\t2\t3\nWin_Aung\t1\t3\nWin_Aung\t1\t4\nZhang_Ziyi\t1\t3\nZhang_Ziyi\t2\t3\nAdrian_Annus\t1\tJorge_Marquez-Ruarte\t1\nAdrian_Annus\t1\tPatrick_Bourrat\t1\nAdrian_Murrell\t1\tJose_Cevallos\t1\nAdrian_Murrell\t1\tPaul_Brandt\t1\nAhmed_Ibrahim_Bilal\t1\tBeatrice_Dalle\t1\nAhmed_Ibrahim_Bilal\t1\tLee_Chang-dong\t1\nAileen_Riggin_Soule\t1\tNorio_Ohga\t1\nAitor_Gonzalez\t2\tHorace_Donovan_Reid\t1\nAjit_Agarkar\t1\tJesse_James\t1\nAkbar_Al_Baker\t1\tAndrei_Konchalovsky\t1\nAkbar_Al_Baker\t1\tBobby_Goldwater\t1\nAlain_Cervantes\t1\tBob_Huggins\t3\nAlain_Cervantes\t1\tPierre_Png\t1\nAlanna_Ubach\t1\tPaul_Gascoigne\t1\nAlec_Baldwin\t4\tGoh_Kun\t2\nAlex_Holmes\t1\tBeatrice_Dalle\t1\nAlfred_Sant\t1\tRandall_Tobias\t1\nAlfredo_Moreno\t1\tDyab_Abou_Jahjah\t1\nAlfredo_Moreno\t1\tSuzanne_Torrance\t1\nAlfredo_Pena\t1\tEmmanuel_Milingo\t1\nAlvaro_Noboa\t1\tPhan_Van_Khai\t1\nAlvaro_Noboa\t2\tTimothy_Rigas\t1\nAmanda_Coetzer\t1\tBridgette_Wilson-Sampras\t2\nAmanda_Coetzer\t1\tEddie_Jordan\t1\nAmer_al-Saadi\t4\tMiguel_Cotto\t1\nAmy_Brenneman\t1\tRussell_Coutts\t2\nAmy_Brenneman\t1\tTerje_Roed-Larsen\t1\nAndrew_Firestone\t1\tHarry_Schmidt\t4\nAndrew_Firestone\t1\tIan_Campbell\t1\nAndrew_Firestone\t1\tJose_Acasuso\t1\nAndrew_Firestone\t1\tOmar_Sharif\t4\nAndy_Graves\t1\tGil_de_Ferran\t1\nAngela_Alvarado_Rosa\t1\tGerald_Calabrese\t1\nAngela_Alvarado_Rosa\t1\tJeffrey_Jones\t1\nAnneli_Jaatteenmaki\t2\tEric_Robert_Rudolph\t2\nAnneli_Jaatteenmaki\t2\tGerry_Adams\t6\nAnneli_Jaatteenmaki\t2\tJorge_Marquez-Ruarte\t1\nAnthony_Corso\t1\tJohnny_Benson\t1\nAnthony_Corso\t1\tRay_Allen\t3\nAnthony_Pico\t1\tStephanie_Cohen_Aloro\t1\nAri_Fleischer\t1\tRazali_Ismail\t1\nAri_Fleischer\t9\tJean-Marc_de_La_Sabliere\t2\nArianna_Huffington\t4\tPhilippe_Gagnon\t1\nArnaud_Clement\t1\tJeffrey_Jones\t2\nArnaud_Clement\t1\tKing_Gyanendra\t1\nArnaud_Clement\t2\tGrady_Irvin_Jr\t2\nArnold_Scott\t1\tRandy_Dryer\t1\nArt_Howe\t2\tLuciano_Bovicelli\t1\nArt_Howe\t3\tTeresa_Worbis\t1\nArtieas_Shanks\t1\tDerek_King\t1\nAstrid_Betancourt\t1\tFrederick_Madden\t1\nBarbara_De_Brun\t1\tCosmo_Iacavazzi\t1\nBarry_Collier\t1\tNarendra_Modi\t1\nBen_Affleck\t3\tPatricia_Hearst\t1\nBen_Affleck\t4\tDaniel_Scioli\t1\nBen_Affleck\t7\tMarcio_de_Souza\t1\nBen_Broussard\t1\tRudi_Voeller\t1\nBetsy_Smith\t2\tSteven_Kinlock\t1\nBetty_Williams\t1\tMary_Catherine_Correll\t1\nBianca_Jagger\t1\tDario_Camuffo\t1\nBianca_Jagger\t1\tFabricio_Oberto\t1\nBill_Callahan\t1\tGina_Gershon\t1\nBill_Kollar\t1\tGina_Gershon\t1\nBill_Kollar\t1\tJose_Miguel_Aleman\t1\nBill_Kollar\t1\tMae_Jemison\t1\nBill_Kollar\t1\tMohammed_Ashraf_Hafiz\t1\nBill_Kollar\t1\tPatty_Sheehan\t1\nBill_Parsons\t1\tDonna_Barrera\t1\nBill_Stein\t1\tVanessa_Incontrada\t4\nBlythe_Hartley\t1\tJackie_Dennis\t1\nBlythe_Hartley\t2\tGil_de_Ferran\t2\nBob_Curtis\t1\tHelen_Clark\t1\nBob_Huggins\t3\tSylvester_Stallone\t2\nBobby_Goldwater\t1\tJohn_Moxley\t1\nBobby_Goldwater\t1\tUlrich_Kueperkoch\t1\nBono\t3\tIan_Huntley\t1\nBrad_Garrett\t1\tHoda_Asfor\t1\nBrad_Garrett\t4\tWang_Nan\t1\nBrandon_Hammond\t1\tThomas_Kelly\t1\nBrandon_Robinson\t1\tGiovanny_Cordoba\t1\nBrandon_Robinson\t1\tMichael_Linscott\t1\nBrandon_Robinson\t1\tShi_Guangsheng\t1\nBrendan_Stai\t1\tDan_Guerrero\t1\nBrett_Boone\t1\tJean-Marc_de_La_Sabliere\t1\nBrett_Boone\t1\tTeresa_Worbis\t1\nBrian_Billick\t1\tIan_Huntley\t1\nBrian_Billick\t1\tJohn_Wayne\t1\nBrian_Billick\t1\tStephanie_Moore\t1\nBrian_Olson\t1\tRoberto_Tovar\t1\nBud_Selig\t1\tFranz_Fischler\t1\nBud_Selig\t3\tJohn_Duprey\t1\nCarla_Moreno\t1\tSuzanne_Torrance\t1\nCarla_Moreno\t1\tTang_Jiaxuan\t3\nCarlos_Beltran\t1\tUlrich_Kueperkoch\t1\nCarlos_Ghosn\t1\tJim_Paxson\t1\nCarlos_Ghosn\t2\tRay_Allen\t3\nCarlton_Dotson\t1\tJim_Paxson\t1\nCarlton_Dotson\t1\tPatrick_Bourrat\t1\nCarlton_Dotson\t1\tPorter_Goss\t1\nCarlton_Dotson\t1\tVyacheslav_Fetisov\t1\nCass_Ballenger\t1\tNorio_Ohga\t1\nCecile_de_France\t1\tDyab_Abou_Jahjah\t1\nCecile_de_France\t1\tTerrence_Kiel\t1\nCharles_Bell\t1\tTatjana_Gsell\t1\nCharlotte_Rampling\t1\tLana_Clarkson\t2\nChick_Hearn\t1\tMohammed_Salmane\t1\nChristopher_Reeve\t4\tPauline_Landers\t1\nChristopher_Reeve\t4\tScott_OGrady\t1\nCindy_Margolis\t1\tMark_Cuban\t2\nClaire_Danes\t2\tNadine_Vinzens\t1\nClaudio_Lopez\t1\tGabrielle_Rose\t1\nCollis_Temple_III\t1\tEva_Amurri\t1\nCollis_Temple_III\t1\tRob_Niedermayer\t1\nCollis_Temple_III\t1\tSantiago_Botero\t1\nCollis_Temple_III\t1\tSimon_Chalk\t1\nConan_OBrien\t1\tLiliana_Cavani\t1\nCorinna_Harfouch\t1\tIvo_Dubs\t1\nCorinna_Harfouch\t1\tTim_Curry\t1\nCristina_Kirchner\t1\tStefanie_De_Roux\t1\nDan_Boyle\t1\tPaul_Clark\t1\nDan_Boyle\t1\tThabo_Mbeki\t3\nDaniel_Chin\t1\tIan_Huntley\t1\nDaniel_Chin\t1\tJim_Letten\t1\nDaniel_Chin\t1\tJulia_Glass\t1\nDaniel_Scioli\t1\tLena_Katina\t1\nDaniel_Scioli\t1\tLindsay_Lohan\t1\nDario_Camuffo\t1\tEli_Stutsman\t1\nDavid_Brinkley\t1\tJeff_Bridges\t1\nDavid_Brinkley\t1\tStephen_Arigbabu\t1\nDavid_Montoya\t1\tMary_Elizabeth_Mastrantonio\t1\nDavid_Siegel\t1\tFrancis_Mer\t1\nDennis_Johnson\t1\tSatnarine_Sharma\t1\nDenys_Arcand\t1\tNadine_Vinzens\t2\nDerek_Jeter\t1\tTian_Zhuang_Zhuang\t1\nDerek_Jeter\t3\tJolanta_Kwasniewski\t2\nDerek_King\t1\tYasein_Taher\t1\nDerrick_Battie\t1\tIan_Huntley\t1\nDiego_Armando_Maradona\t1\tRobert_Gordon_Card\t1\nDon_King\t1\tThomas_Kelly\t1\nDon_King\t1\tZurab_Tsereteli\t1\nDonna_Barrera\t1\tFrancis_Mer\t2\nDonna_Shalala\t2\tSereyvuth_Kem\t1\nDuncan_Fletcher\t1\tJohn_Cusack\t2\nDuncan_Fletcher\t1\tMike_Alden\t1\nDustin_Brown\t1\tJose_Cevallos\t1\nDyab_Abou_Jahjah\t1\tJohn_Cornyn\t1\nEd_Mekertichian\t1\tPaul_Clark\t1\nEdmund_Hillary\t2\tMiguel_Cotto\t1\nElena_Likhovtseva\t1\tMarty_Mornhinweg\t2\nEli_Stutsman\t1\tKalpana_Chawla\t4\nEli_Stutsman\t1\tMarcio_de_Souza\t1\nEli_Stutsman\t1\tMichael_Phelps\t1\nElizabeth_Regan\t1\tEugene_Teslovic\t1\nEmmanuel_Milingo\t1\tEnrica_Fico\t1\nEnola_Rice\t1\tLana_Clarkson\t2\nErin_Runnion\t3\tRob_Lowe\t2\nEugene_Teslovic\t1\tGoh_Kun\t1\nEugene_Teslovic\t1\tJean-Marc_de_La_Sabliere\t2\nEva_Amurri\t1\tTanya_Holyk\t1\nFabricio_Oberto\t1\tRudi_Voeller\t1\nFelix_Sanchez\t1\tIan_Moran\t1\nFernando_Henrique_Cardoso\t1\tMichael_Phelps\t1\nFernando_Henrique_Cardoso\t8\tSherry_Fisher\t1\nFrancis_Mer\t2\tPatrick_Bourrat\t1\nFranz_Fischler\t2\tRobert_Lange\t1\nFrederick_Madden\t1\tGary_Condit\t1\nGary_Bauer\t1\tGreg_Hodge\t1\nGary_Condit\t1\tNathan_Lane\t2\nGary_Locke\t1\tMarcio_de_Souza\t1\nGen_Meredith\t1\tMatthew_Vaughan\t1\nGerry_Adams\t8\tJamir_Miller\t1\nGilles_Panizzi\t1\tJim_Piper\t1\nGina_Gershon\t1\tPatricia_Garone\t1\nGina_Gershon\t1\tZhang_Ziyi\t2\nGiovanny_Cordoba\t1\tJackie_Sherrill\t1\nGiovanny_Cordoba\t1\tKeith_Olbermann\t1\nGrady_Irvin_Jr\t2\tOracene_Williams\t1\nGraeme_Smith\t1\tJohn_Salazar\t1\nGraham_Bentley\t1\tJason_Clermont\t1\nGuillaume_Depardieu\t1\tManuel_Llorente\t1\nGunter_Pleuger\t1\tNicole_Kidman\t2\nGunter_Pleuger\t2\tRobby_Ginepri\t1\nGunter_Pleuger\t3\tVaclav_Klaus\t1\nHanns_Schumacher\t1\tMartin_Gecht\t1\nHenri_Proglio\t1\tIbrahim_Haddad\t1\nHenri_Proglio\t1\tJeffery_Hendren\t1\nHermann_Maier\t1\tPhan_Van_Khai\t1\nHermann_Maier\t2\tStephen_Arigbabu\t1\nHideki_Sato\t1\tPeter_Mugyeni\t1\nHideki_Sato\t1\tTroy_Aikman\t1\nHoda_Asfor\t1\tJuergen_Peters\t1\nHoda_Asfor\t1\tLindsay_Lohan\t1\nHorace_Donovan_Reid\t1\tRob_Lowe\t1\nHunter_Kemper\t1\tMarsha_Thomason\t1\nIan_Campbell\t1\tMike_Alden\t1\nIan_Huntley\t1\tJalen_Rose\t1\nIan_Huntley\t1\tSebastian_Saja\t2\nIan_Moran\t1\tStephanie_Cohen_Aloro\t1\nIan_Thorpe\t5\tMickey_Gilley\t1\nIgnacio_Antonio_Velasco\t1\tRich_Brooks\t1\nIran_Brown\t1\tMargaret_Caruso\t1\nIsabelle_Huppert\t1\tMarcio_de_Souza\t1\nIsabelle_Huppert\t2\tPhan_Van_Khai\t3\nIvan_Shvedoff\t1\tJosh_Childress\t1\nIvan_Shvedoff\t1\tMiguel_Estrada\t2\nIvan_Shvedoff\t1\tVojislav_Seselj\t1\nIzzat_Ibrahim\t1\tJerry_Rice\t1\nJacques_Villeneuve\t1\tWim_Duisenberg\t1\nJalen_Rose\t1\tSuzanne_Torrance\t1\nJames_Butts\t1\tNicole_Kidman\t2\nJames_Butts\t2\tJerry_Colangelo\t1\nJames_Butts\t2\tSherry_Fisher\t1\nJames_May\t1\tTrevor_McDonald\t1\nJames_May\t1\tVaclav_Klaus\t2\nJamie_Cooke\t1\tKalpana_Chawla\t5\nJamie_Cooke\t1\tMickey_Rooney\t1\nJamir_Miller\t1\tLaura_Hernandez\t1\nJana_Pittman\t1\tLiv_Tyler\t1\nJason_Clermont\t1\tJose_Carlo_Fernandez\t1\nJason_Clermont\t1\tPatty_Duke\t1\nJeff_Bridges\t1\tLandon_Donovan\t1\nJeff_Bridges\t1\tPatty_Duke\t1\nJeffery_Hendren\t1\tJeremy_Wotherspoon\t1\nJennifer_McCoy\t1\tManuel_Llorente\t1\nJennifer_McCoy\t1\tWilliam_Cocksedge\t1\nJerry_Colangelo\t1\tVin_Diesel\t2\nJerry_Rice\t1\tJoe_Gatti\t2\nJessica_Brungo\t1\tLandon_Donovan\t1\nJim_OBrien\t1\tNadine_Vinzens\t1\nJim_Paxson\t1\tRobert_Gordon_Card\t1\nJoe_Gatti\t2\tMike_Samp\t1\nJoel_Todd\t1\tJohn_Fox\t1\nJoel_Todd\t1\tMomir_Nikolic\t1\nJohn_Cornyn\t1\tJose_Mourinho\t2\nJohn_Cornyn\t1\tRobert_Weitzel\t1\nJohn_Cornyn\t1\tToshihiko_Fukui\t2\nJohn_Dallager\t1\tJose_Canseco_Sr\t1\nJohn_Spencer\t2\tStephanie_Moore\t1\nJohn_Walsh\t1\tKalpana_Chawla\t1\nJohn_Wright\t1\tSandra_Milo\t1\nJohn_Wright\t1\tTrevor_McDonald\t1\nJohnny_Benson\t1\tLana_Clarkson\t2\nJohnny_Benson\t1\tTeresa_Worbis\t1\nJolanta_Kwasniewski\t2\tMartin_Howard\t1\nJose_Canseco_Sr\t1\tJulia_Glass\t1\nJose_Mourinho\t1\tJoseph_LePore\t1\nJose_Mourinho\t2\tLeRoy_Millette_Jr\t1\nJose_Vicente_Rangel\t1\tLeuris_Pupo\t1\nJose_Vicente_Rangel\t1\tScott_OGrady\t1\nJoy_Lee_Sadler\t1\tLaurie_Chan\t1\nJoy_Lee_Sadler\t1\tNorio_Ohga\t1\nJoy_Lee_Sadler\t1\tStephanie_Moore\t1\nKaren_Pereiras\t1\tMichael_Phelps\t2\nKeith_Brown\t1\tNicole_Kidman\t17\nKeith_Brown\t1\tWilliam_Harrison\t1\nKeith_Van_Horn\t1\tSherry_Fisher\t1\nKelli_White\t1\tRudi_Voeller\t1\nKing_Gyanendra\t1\tOtto_Reich\t1\nLana_Clarkson\t1\tMike_Samp\t1\nLandon_Donovan\t1\tRobby_Ginepri\t1\nLarry_Tanenbaum\t1\tMike_Samp\t1\nLaura_Ziskin\t1\tReyyan_Uzuner\t1\nLaura_Ziskin\t1\tRobert_Gordon_Card\t1\nLeRoy_Millette_Jr\t1\tLeuris_Pupo\t1\nLee_Chang-dong\t1\tPhil_Bredesen\t1\nLiliana_Cavani\t1\tRichard_Pennington\t1\nLindsay_Lohan\t1\tMireya_Elisa_Moscoso_Rodriguez\t1\nLloyd_Ward\t2\tTina_Andrews\t1\nMark_Heller\t1\tNicolas_Kiefer\t1\nMark_Heller\t1\tParris_Glendening\t1\nMark_Heller\t1\tPeter_Hillary\t2\nMartin_Gecht\t1\tPeter_Hillary\t1\nMary_Elizabeth_Mastrantonio\t1\tVaclav_Klaus\t1\nMichael_Arif\t1\tSean_Patrick_OMalley\t3\nMichael_Linscott\t1\tTom_Brady\t2\nMick_Jagger\t3\tOracene_Williams\t1\nMo_Elleithee\t1\tTanya_Holyk\t1\nMohammed_Ashraf_Hafiz\t1\tRod_Bryden\t1\nMukhtar_Alytnbayev\t1\tOracene_Williams\t1\nPatricia_Garone\t1\tSean_Patrick_OMalley\t2\nPatricia_Hearst\t1\tScott_Dickson\t1\nPatty_Duke\t1\tSimon_Chalk\t1\nPaul_Bettany\t1\tUlrich_Kueperkoch\t1\nPaul_Gascoigne\t3\tTian_Liang\t1\nPaul_Greengrass\t1\tTim_Pawlenty\t1\nPerry_Compton\t1\tWilliam_Shatner\t1\nPeter_Camejo\t1\tRuth_Christofferson\t1\nPorter_Goss\t1\tTara_Kirk\t1\nRand_Miller\t1\tRobert_Nillson\t1\nRich_Brooks\t1\tSharess_Harrell\t1\nRichard_Pennington\t1\tRobby_Ginepri\t2\nRobert_Gordon_Card\t1\tWilliam_Shatner\t1\nRobert_Kipkoech_Cheruiyot\t1\tSally_Kirkland\t3\nRobert_Kipkoech_Cheruiyot\t1\tSteve_Park\t1\nRussell_Coutts\t1\tTian_Liang\t1\nSandra_Milo\t1\tSatnarine_Sharma\t1\nSteve_Nesbitt\t1\tWin_Aung\t4\nSylvester_Stallone\t7\tTJ_Ford\t1\nSylvie_Guillem\t1\tVadim_Devyatovskiy\t1\nTara_Kirk\t1\tWin_Aung\t1\nAaron_Peirsol\t1\t4\nAaron_Peirsol\t3\t4\nAdrian_Nastase\t1\t2\nAhmed_Chalabi\t1\t3\nAhmed_Chalabi\t1\t5\nAlbrecht_Mentz\t1\t2\nAlejandro_Toledo\t20\t36\nAlejandro_Toledo\t21\t24\nAlejandro_Toledo\t21\t30\nAlejandro_Toledo\t23\t27\nAlejandro_Toledo\t26\t29\nAlexander_Losyukov\t1\t3\nAlexander_Losyukov\t2\t3\nAlexander_Losyukov\t2\t4\nAlimzhan_Tokhtakhounov\t1\t2\nAmelie_Mauresmo\t7\t14\nAmelie_Mauresmo\t11\t17\nAmelie_Mauresmo\t14\t17\nAngelo_Reyes\t1\t2\nAngelo_Reyes\t1\t3\nBegum_Khaleda_Zia\t1\t2\nBen_Curtis\t1\t2\nBijan_Namdar_Zangeneh\t1\t2\nBill_Paxton\t1\t2\nBill_Paxton\t1\t3\nBill_Paxton\t2\t4\nBill_Paxton\t3\t4\nBilly_Crystal\t1\t2\nBilly_Crystal\t1\t3\nBilly_Crystal\t1\t5\nBilly_Crystal\t3\t5\nBilly_Graham\t1\t2\nBob_Colvin\t1\t2\nBrian_Cowen\t1\t2\nButch_Davis\t1\t2\nByron_Scott\t1\t2\nCarol_Burnett\t1\t2\nCharles_Mathews\t1\t2\nChristine_Ebersole\t1\t2\nClaudia_Schiffer\t1\t2\nCondoleezza_Rice\t1\t4\nCondoleezza_Rice\t2\t10\nCondoleezza_Rice\t4\t5\nCondoleezza_Rice\t8\t9\nCondoleezza_Rice\t9\t10\nCostas_Simitis\t1\t4\nCostas_Simitis\t1\t5\nCostas_Simitis\t3\t5\nCostas_Simitis\t4\t6\nCristina_Saralegui\t1\t2\nCruz_Bustamante\t1\t4\nCruz_Bustamante\t3\t5\nDavid_Heymann\t3\t5\nDiana_Krall\t1\t6\nDiana_Krall\t3\t4\nDiana_Munz\t1\t3\nDiana_Munz\t2\t3\nDominique_de_Villepin\t2\t13\nDominique_de_Villepin\t3\t8\nDominique_de_Villepin\t3\t10\nDominique_de_Villepin\t7\t11\nDominique_de_Villepin\t10\t14\nDominique_de_Villepin\t11\t12\nDon_Siegelman\t1\t3\nDon_Siegelman\t2\t4\nDonald_Pettit\t1\t2\nDonald_Pettit\t1\t3\nDonald_Pettit\t2\t3\nDoug_Collins\t1\t2\nEdward_James_Olmos\t1\t2\nElin_Nordegren\t1\t2\nElizabeth_Taylor\t1\t2\nEllen_DeGeneres\t1\t2\nElton_John\t1\t4\nElton_John\t2\t3\nElton_John\t3\t6\nElton_John\t5\t6\nElton_John\t5\t7\nEmma_Watson\t2\t3\nEmma_Watson\t3\t4\nEmma_Watson\t3\t5\nFabrice_Santoro\t1\t3\nFabrice_Santoro\t2\t3\nFlavia_Delaroli\t1\t2\nGeorge_Lopez\t1\t4\nGeorge_Lopez\t2\t4\nGeorge_Lopez\t2\t5\nGeorge_Lopez\t4\t5\nGerard_Depardieu\t1\t2\nGerhard_Schroeder\t5\t43\nGerhard_Schroeder\t17\t25\nGerhard_Schroeder\t27\t32\nGerhard_Schroeder\t45\t67\nGerhard_Schroeder\t66\t100\nGilberto_Rodriguez_Orejuela\t1\t3\nGilberto_Rodriguez_Orejuela\t2\t3\nGilberto_Rodriguez_Orejuela\t3\t4\nGiuseppe_Gibilisco\t1\t3\nGiuseppe_Gibilisco\t1\t4\nGiuseppe_Gibilisco\t3\t4\nGuillermo_Canas\t1\t2\nGuillermo_Canas\t1\t3\nGuillermo_Canas\t2\t4\nGuillermo_Canas\t3\t4\nHans_Blix\t7\t8\nHans_Blix\t7\t18\nHans_Blix\t7\t38\nHans_Blix\t8\t24\nHans_Blix\t9\t12\nHans_Blix\t26\t36\nHans_Blix\t27\t37\nHans_Blix\t31\t37\nHeath_Ledger\t1\t2\nHeath_Ledger\t1\t4\nHeath_Ledger\t2\t3\nHeath_Ledger\t2\t4\nHerta_Daeubler-Gmelin\t1\t2\nHilary_Duff\t1\t3\nHilary_Duff\t2\t3\nHosni_Mubarak\t6\t7\nHun_Sen\t1\t4\nHun_Sen\t2\t4\nIain_Duncan_Smith\t1\t3\nIain_Duncan_Smith\t2\t3\nIain_Duncan_Smith\t2\t4\nIain_Duncan_Smith\t3\t4\nInocencio_Arias\t1\t2\nJK_Rowling\t1\t2\nJK_Rowling\t2\t3\nJK_Rowling\t2\t4\nJK_Rowling\t3\t5\nJackie_Chan\t1\t3\nJackie_Chan\t5\t8\nJane_Kaczmarek\t1\t2\nJanet_Napolitano\t1\t2\nJanet_Napolitano\t1\t4\nJay_Leno\t1\t3\nJay_Leno\t2\t3\nJean-David_Levitte\t1\t2\nJean-David_Levitte\t1\t4\nJean-David_Levitte\t2\t5\nJean-David_Levitte\t3\t4\nJean-David_Levitte\t4\t9\nJeff_Van_Gundy\t1\t2\nJeff_Van_Gundy\t1\t3\nJeff_Van_Gundy\t2\t3\nJim_Furyk\t1\t6\nJim_Furyk\t3\t6\nJohn_Garamendi\t1\t2\nJohn_Rigas\t1\t2\nJohn_Warner\t1\t2\nJohn_Warner\t1\t3\nJohn_Warner\t2\t4\nJohn_Warner\t3\t4\nJulia_Tymoshenko\t1\t2\nJulia_Tymoshenko\t1\t3\nJulia_Tymoshenko\t2\t3\nKeith_Bogans\t1\t3\nKeith_Bogans\t2\t3\nKen_Macha\t1\t3\nKen_Macha\t2\t3\nKenneth_Bowersox\t1\t2\nKenneth_Bowersox\t1\t3\nKenneth_Bowersox\t2\t3\nKristanna_Loken\t1\t2\nKristanna_Loken\t3\t4\nKristanna_Loken\t4\t5\nLK_Advani\t1\t2\nLK_Advani\t1\t3\nLauren_Killian\t1\t2\nLennox_Lewis\t1\t2\nLennox_Lewis\t1\t3\nLili_Taylor\t1\t2\nLily_Tomlin\t1\t2\nLon_Kruger\t1\t2\nLynn_Abraham\t1\t2\nMahmoud_Abbas\t7\t11\nMahmoud_Abbas\t15\t20\nMahmoud_Abbas\t22\t27\nMark_Richt\t1\t2\nMark_Richt\t1\t3\nMary-Kate_Olsen\t1\t2\nMary-Kate_Olsen\t2\t3\nMatt_Dillon\t1\t2\nMax_Mayfield\t1\t2\nMegan_Mullally\t1\t2\nMegan_Mullally\t1\t3\nMichael_Caine\t1\t2\nMichael_Caine\t1\t3\nMichael_Caine\t3\t4\nMichael_J_Sheehan\t1\t2\nMichael_Keaton\t1\t2\nMichael_Powell\t1\t2\nMichael_Powell\t1\t3\nMichael_Powell\t1\t5\nMichael_Powell\t2\t3\nMichael_Powell\t2\t4\nMichael_Powell\t2\t5\nMichael_Powell\t4\t5\nMiguel_Contreras\t1\t2\nMike_Weir\t1\t6\nMike_Weir\t3\t11\nMikhail_Kasyanov\t1\t2\nMikhail_Kasyanov\t1\t3\nMikhail_Kasyanov\t3\t4\nMohammed_Baqir_al-Hakim\t1\t3\nMonica_Lewinsky\t1\t2\nMonica_Lewinsky\t1\t3\nMonica_Lewinsky\t2\t3\nNan_Wang\t1\t3\nNan_Wang\t2\t3\nNan_Wang\t3\t4\nNancy_Demme\t1\t2\nNancy_Reagan\t1\t2\nNaoto_Kan\t1\t3\nNaoto_Kan\t2\t3\nNatasha_McElhone\t1\t2\nNatasha_McElhone\t1\t3\nNeri_Marcore\t1\t2\nNicholas_Tse\t1\t2\nNicolas_Escude\t1\t2\nNoelle_Bush\t1\t3\nNoelle_Bush\t1\t4\nNoelle_Bush\t2\t4\nOrrin_Hatch\t1\t2\nPadraig_Harrington\t1\t3\nPadraig_Harrington\t1\t4\nPadraig_Harrington\t2\t4\nPedro_Almodovar\t1\t2\nPedro_Almodovar\t3\t5\nPedro_Almodovar\t5\t6\nPenelope_Ann_Miller\t1\t2\nPeter_Arnett\t1\t3\nPeter_Arnett\t2\t3\nPetria_Thomas\t1\t2\nPetro_Symonenko\t1\t2\nPrince_Charles\t1\t2\nPrince_Charles\t3\t4\nPrince_Charles\t3\t5\nRaoul_Ruiz\t1\t2\nRaoul_Ruiz\t2\t3\nRaoul_Ruiz\t2\t4\nRaoul_Ruiz\t3\t4\nRicky_Ponting\t1\t2\nRobert_Stack\t1\t2\nRobin_Cook\t1\t2\nRoger_Federer\t4\t5\nRoger_Federer\t4\t7\nRoger_Federer\t6\t7\nRonaldo_Luis_Nazario_de_Lima\t1\t4\nRonaldo_Luis_Nazario_de_Lima\t2\t3\nRonaldo_Luis_Nazario_de_Lima\t2\t4\nRonaldo_Luis_Nazario_de_Lima\t3\t4\nRoy_Williams\t1\t3\nRoy_Williams\t2\t4\nSally_Ride\t1\t2\nScott_McNealy\t1\t2\nScott_Peterson\t1\t2\nScott_Peterson\t1\t5\nScott_Peterson\t2\t5\nScott_Peterson\t3\t4\nScott_Ritter\t1\t2\nSean_OKeefe\t3\t4\nSean_Penn\t1\t2\nSean_Penn\t2\t3\nSebastien_Grosjean\t1\t2\nSebastien_Grosjean\t1\t4\nSharon_Stone\t1\t2\nSharon_Stone\t1\t3\nSharon_Stone\t1\t4\nSharon_Stone\t1\t5\nSharon_Stone\t2\t4\nSonia_Gandhi\t1\t3\nSonia_Gandhi\t1\t4\nSteve_Backley\t1\t2\nSteven_Hatfill\t1\t2\nTassos_Papadopoulos\t1\t3\nTaufik_Hidayat\t1\t2\nTaufik_Hidayat\t1\t3\nTaufik_Hidayat\t2\t3\nThomas_Malchow\t1\t2\nTom_Cruise\t3\t5\nTom_Cruise\t4\t10\nTom_Cruise\t6\t10\nToni_Braxton\t1\t3\nWayne_Gretzky\t1\t2\nWayne_Gretzky\t1\t4\nWilliam_Bulger\t1\t3\nWilliam_Bulger\t2\t4\nWinona_Ryder\t12\t13\nWinona_Ryder\t12\t18\nWinona_Ryder\t12\t19\nXanana_Gusmao\t1\t4\nXanana_Gusmao\t2\t3\nXanana_Gusmao\t3\t5\nXanana_Gusmao\t4\t5\nYann_Martel\t1\t2\nYashwant_Sinha\t1\t3\nYashwant_Sinha\t1\t5\nYashwant_Sinha\t5\t7\nAaron_Guiel\t1\tPascal_Rheaume\t1\nAaron_Guiel\t1\tSteve_Zahn\t1\nAdrian_Nastase\t2\tPrincess_Victoria\t1\nAhmed_Chalabi\t1\tRosario_Dawson\t1\nAhmed_Ghazi\t1\tJulia_Tymoshenko\t3\nAhmet_Demir\t1\tWally_Szczerbiak\t1\nAin_Seppik\t1\tGerhard_Schroeder\t55\nAin_Seppik\t1\tMisty_Dawn_Clymer\t1\nAin_Seppik\t1\tRaoul_Ruiz\t1\nAin_Seppik\t1\tSusan_Whelan\t1\nAishwarya_Rai\t1\tChloe_Sevigny\t1\nAishwarya_Rai\t1\tDinah_Turner\t1\nAishwarya_Rai\t1\tRosario_Dawson\t1\nAlan_Zemaitis\t1\tLauren_Killian\t1\nAlex_Cejka\t1\tTaufik_Hidayat\t2\nAlexandre_Daigle\t1\tBrent_Coles\t1\nAli_Bin_Hussein\t1\tBen_Cahoon\t1\nAli_Bin_Hussein\t1\tOswald_Gruebel\t1\nAli_Bin_Hussein\t1\tRaf_Vallone\t1\nAlimzhan_Tokhtakhounov\t1\tJeane_Kirkpatrick\t1\nAlimzhan_Tokhtakhounov\t2\tButch_Davis\t1\nAlisha_Richman\t1\tCamille_Lewis\t1\nAlisha_Richman\t1\tKaterina_Smrzova\t1\nAlisha_Richman\t1\tMiguel_Rosseto\t1\nAmelie_Mauresmo\t3\tLaura_Flessel\t1\nAmelie_Mauresmo\t18\tBob_Colvin\t1\nAna_Claudia_Talancon\t1\tPiers_Sellers\t1\nAndrea_Kiser\t1\tEllen_Barkin\t1\nAndrea_Kiser\t1\tImre_Kertasz\t1\nAndrea_Kiser\t1\tMatt_Dillon\t1\nAndrea_Kiser\t1\tYann_Martel\t2\nAndres_DAlessandro\t1\tLeticia_Van_de_Putte\t1\nAndres_DAlessandro\t1\tRonaldo_Luis_Nazario_de_Lima\t1\nAnette_Hosoi\t1\tRolf_Zimmermann\t1\nAnette_Hosoi\t1\tSheila_Taormina\t1\nAnette_Hosoi\t1\tTora_Takagi\t1\nAngelo_Reyes\t1\tChip_Knight\t1\nAngelo_Reyes\t1\tOliver_Neuville\t1\nAnne_Heche\t1\tSasha_Alexander\t1\nAnne_Heche\t1\tTanya_Lindenmuth\t1\nAntje_Buschschulte\t1\tJerry_Angelo\t1\nAnton_Balasingham\t1\tMatt_Siebrandt\t1\nAnton_Balasingham\t1\tWill_Ferrell\t1\nArmando_Avila_Panchame\t1\tMonica_Gabrielle\t1\nArmando_Avila_Panchame\t1\tRetief_Goosen\t1\nBJ_Habibie\t1\tCraig_MacTavish\t1\nBaz_Luhrmann\t1\tDanny_Avalon\t1\nBegum_Khaleda_Zia\t1\tEric_Idle\t1\nBegum_Khaleda_Zia\t1\tMichael_Andretti\t1\nBegum_Khaleda_Zia\t1\tPicabo_Street\t1\nBen_Cahoon\t1\tNate_Blackwell\t1\nBen_Cahoon\t1\tRoy_Williams\t3\nBen_Curtis\t3\tJim_Calhoun\t1\nBenicio_Del_Toro\t1\tElizabeth_Taylor\t1\nBenicio_Del_Toro\t1\tIsmail_Khan\t1\nBenicio_Del_Toro\t1\tTonya_Payne\t1\nBijan_Namdar_Zangeneh\t1\tPetro_Symonenko\t1\nBilly_Graham\t1\tKarin_Pilsaeter\t1\nBilly_Graham\t1\tPiers_Sellers\t1\nBilly_Graham\t2\tHoward_Wilkinson\t1\nBob_Colvin\t2\tJanet_Napolitano\t1\nBoris_Henry\t1\tJohn_Reid\t2\nBrandon_Boyd\t1\tJames_Brosnahan\t1\nBrandon_Boyd\t1\tNeil_Moritz\t1\nBrent_Coles\t1\tZoe_Ball\t1\nBrett_Hull\t1\tGregorio_Rosal\t1\nBrett_Hull\t1\tKirsten_Clark\t1\nBrett_Hull\t1\tRalph_Nader\t1\nBrett_Perry\t1\tNeri_Marcore\t1\nBruce_Willis\t1\tCarlos_Juarez\t1\nBruce_Willis\t1\tJim_Carrey\t1\nBruce_Willis\t1\tPetria_Thomas\t2\nBryant_Young\t1\tJim_Bollman\t1\nBryant_Young\t1\tMaurice_Cheeks\t1\nBuddy_Ryan\t1\tJames_Brosnahan\t1\nBuddy_Ryan\t1\tNancy_Reagan\t2\nButch_Davis\t2\tHerta_Daeubler-Gmelin\t2\nByron_Scott\t1\tJane_Clayson\t1\nCamille_Lewis\t1\tLincoln_Chafee\t1\nCamille_Lewis\t1\tNathirah_Hussein\t1\nCarol_Burnett\t1\tRaf_Vallone\t1\nCarol_Niedermayer\t1\tKristin_Scott\t1\nCatherine_Ndereba\t1\tJanet_Napolitano\t4\nChandrika_Kumaratunga\t1\tKevin_Satterfield\t1\nCharles_Mathews\t1\tWill_Ferrell\t1\nCharles_Mathews\t2\tKristin_Scott\t1\nCharlie_Garner\t1\tLuo_Linquan\t1\nCharlie_Williams\t1\tJulia_Tymoshenko\t1\nCheryl_Ford\t1\tChris_Cookson\t1\nChip_Knight\t1\tRonald_Post\t1\nChloe_Sevigny\t1\tEmyr_Jones_Parry\t1\nChloe_Sevigny\t1\tPetria_Thomas\t3\nChris_Cornell\t1\tDesmon_Farmer\t1\nChristine_Ebersole\t2\tPaul_Walker\t1\nChristoph_Daum\t1\tScott_Hubbard\t1\nChristopher_Conyers\t1\tMichael_Milton\t1\nChristopher_Conyers\t1\tRaoul_Ruiz\t1\nChuck_Bednarik\t1\tMario_Jardel\t1\nCindy_Zagorski\t1\tRosie_Perez\t1\nClaudia_Schiffer\t1\tLynn_Abraham\t2\nClaudia_Schiffer\t3\tLin_Yung_Hsi\t1\nColin_Cowie\t1\tMstislav_Rostropovich\t1\nCondoleezza_Rice\t2\tMehmet_Okur\t1\nCondoleezza_Rice\t7\tRyan_Nyquist\t1\nCostas_Simitis\t2\tLorraine_Fenton\t1\nCraig_MacTavish\t1\tVince_Dooley\t1\nCristina_Saralegui\t1\tDave_Robertson\t1\nCruz_Bustamante\t2\tDean_Barkley\t3\nCyndi_Thompson\t1\tNicolas_Escude\t2\nCyndi_Thompson\t2\tOwen_Nolan\t1\nDalia_Rabin-Pelosoff\t1\tSean_Combs\t1\nDaniel_Darnell\t1\tMalcolm_Jamal_Warner\t1\nDanny_Avalon\t1\tSean_Penn\t2\nDavid_Blaine\t1\tNicholas_Tse\t2\nDavid_Blaine\t1\tRamon_Delgado\t1\nDavid_Heymann\t3\tNorm_Macdonald\t1\nDavid_Surrett\t1\tMichael_Taylor\t1\nDavid_Tornberg\t1\tRicky_Ponting\t1\nDeece_Eckstein\t1\tDiana_Munz\t1\nDeece_Eckstein\t1\tTommy_Lewis\t1\nDenis_Coderre\t1\tDon_Siegelman\t1\nDenis_Coderre\t1\tJohn_Eder\t1\nDesmon_Farmer\t1\tLin_Yung_Hsi\t1\nDiana_Krall\t4\tRamon_Delgado\t1\nDick_Armey\t1\tMehmet_Okur\t1\nDinah_Turner\t1\tMekhi_Phifer\t1\nDinah_Turner\t1\tTammy_Helm\t1\nDon_Boudria\t1\tRoy_Chaderton\t1\nDon_Boudria\t1\tSananda_Maitreya\t1\nDon_Matthews\t1\tKirsten_Clark\t1\nDon_Matthews\t1\tPa_Kou_Hang\t1\nDon_Matthews\t1\tPeter_Arnett\t2\nDon_Siegelman\t2\tElizabeth_Taylor\t2\nDonald_Pettit\t1\tEdward_James_Olmos\t2\nDoug_Collins\t1\tKristanna_Loken\t5\nDyana_Calub\t1\tRonald_Post\t1\nEarl_Counter\t1\tJoe_Friedberg\t1\nEdward_Albee\t1\tNicola_Bono\t1\nEdward_Belvin\t1\tJacqueline_Gold\t1\nEdward_James_Olmos\t2\tMariano_Zabaleta\t1\nElin_Nordegren\t1\tGeorge_Brumley_III\t1\nElin_Nordegren\t1\tSamantha_Daniels\t1\nElin_Nordegren\t2\tLeszek_Miller\t3\nElisabeth_Welch\t1\tPiers_Sellers\t1\nElizabeth_Taylor\t1\tOlympia_Dukakis\t1\nEllen_DeGeneres\t2\tLorraine_Fenton\t1\nElva_Hsiao\t1\tSamantha_Daniels\t1\nEmma_Watson\t2\tHenry_Hyde\t1\nEmyr_Jones_Parry\t1\tSally_Clark\t1\nEric_Ryan_Donnelly\t1\tIain_Duncan_Smith\t1\nErnesto_Zedillo\t1\tLenny_Kravitz\t1\nErnesto_Zedillo\t1\tMegan_Mullally\t2\nErwin_Mapasseng\t1\tHilary_Duff\t3\nEvelyn_Lauder\t1\tNathan_Powell\t1\nFernando_Hierro\t1\tPrince_Charles\t2\nFlavia_Delaroli\t2\tJohn_Warner\t4\nFlavia_Delaroli\t2\tVanessa_Laine\t1\nFran_Drescher\t1\tWarren_Truss\t1\nFran_Drescher\t2\tJay_Leno\t3\nFrank_Murkowski\t1\tJames_Brosnahan\t1\nFrank_Stallone\t1\tTommy_Lewis\t1\nFrank_Stallone\t2\tToni_Braxton\t2\nGary_Coleman\t1\tKoichi_Tanaka\t1\nGary_Coleman\t1\tUri_Lopolianski\t1\nGary_Sayler\t1\tRalph_Sampson\t1\nGeorge_Brumley_III\t1\tWill_Ferrell\t1\nGeorge_Lopez\t5\tIain_Duncan_Smith\t1\nGeorgina_Bardach\t1\tHenry_Hilow\t1\nGilberto_Rodriguez_Orejuela\t3\tLennox_Lewis\t1\nGilberto_Rodriguez_Orejuela\t3\tNancy_Demme\t1\nGordon_Lightfoot\t1\tLima_Azimi\t1\nGordon_Lightfoot\t1\tRobert_Nardelli\t1\nGracia_Burnham\t1\tJim_Calhoun\t1\nGregorio_Rosal\t1\tRichard_Greenberg\t1\nGuenter_Verheugen\t1\tKaspar_Villiger\t1\nGuillermo_Canas\t1\tTim_Duncan\t2\nGuillermo_Canas\t2\tMichalis_Chrisohoides\t1\nHal_McCoy\t1\tRudolph_Holton\t1\nHank_Aaron\t1\tSteve_Valentine\t1\nHans_Eichel\t1\tJimmy_Iovine\t1\nHashan_Tillakaratne\t1\tJohn_Warner\t3\nHeath_Ledger\t2\tMelissa_Stark\t1\nHenry_Hilow\t1\tMonica_Gabrielle\t1\nHerta_Daeubler-Gmelin\t2\tKoichi_Tanaka\t1\nHilary_Duff\t1\tLuther_Htu\t1\nHun_Sen\t2\tJohn_Rigas\t2\nIdi_Amin\t1\tStephen_Silas\t1\nIsabel_Orellana\t1\tSpike_Helmick\t1\nIsmael_Miranda\t1\tJanela_Jara\t1\nIsmael_Miranda\t1\tPeter_Greenspun\t1\nIsmail_Khan\t1\tMalcolm_Jamal_Warner\t1\nJK_Rowling\t5\tSasha_Alexander\t1\nJackie_Chan\t3\tJennifer_Gratz\t1\nJacqueline_Edwards\t1\tMeg_Wakeman\t1\nJacqueline_Edwards\t1\tNoor_Mohammed\t1\nJames_Baker\t1\tMike_Tice\t1\nJames_Harris\t1\tKeith_Bogans\t2\nJames_Layug\t1\tRalph_Sampson\t1\nJane_Clayson\t1\tRene_Antonio_Leon_Rodriguez\t1\nJane_Kaczmarek\t2\tMorris_Watts\t1\nJane_Russell\t1\tMartin_ONeill\t1\nJanela_Jara\t1\tKirsten_Clark\t1\nJanela_Jara\t1\tPa_Kou_Hang\t1\nJanela_Jara\t1\tRicky_Cottrill\t1\nJanet_Napolitano\t2\tLaurie_Laychak\t1\nJavier_Camara\t1\tNate_Blackwell\t1\nJeane_Kirkpatrick\t1\tJeff_Van_Gundy\t3\nJeane_Kirkpatrick\t1\tRichard_Greenberg\t1\nJeff_Van_Gundy\t1\tRichard_Naughton\t1\nJeffrey_Katzenberg\t1\tSusan_Whelan\t1\nJennette_Bradley\t1\tTammy_Helm\t1\nJim_Furyk\t6\tLinn_Thornton\t1\nJohn_Eastman\t1\tRicky_Ponting\t1\nJohn_F_Kennedy_Jr\t2\tNathan_Powell\t1\nJohn_Garamendi\t1\tPetro_Symonenko\t2\nJohn_Madden\t1\tSara_Silverman\t1\nJose_Luis_Rodriguez_Zapatero\t1\tRene_Antonio_Leon_Rodriguez\t1\nJose_Luis_Rodriguez_Zapatero\t1\tSteve_Lenard\t1\nJose_Santos\t1\tMichael_Boyce\t1\nJoseph_Kabila\t1\tRose_Linkins\t1\nJuan_Francisco_Palencia\t1\tKristanna_Loken\t2\nJuan_Francisco_Palencia\t1\tWarren_Truss\t1\nJuan_Roman_Riquelme\t1\tRalph_Sampson\t1\nJulia_Tymoshenko\t2\tLuis_Sanchez\t1\nKarin_Pilsaeter\t1\tMike_Flanagan\t1\nKarin_Pilsaeter\t1\tPetria_Thomas\t3\nKeith_Bogans\t2\tMartin_ONeill\t1\nKelly_Leigh\t1\tSpike_Helmick\t1\nKen_Macha\t3\tTommy_Lewis\t1\nKenneth_Carlsen\t1\tRobin_Cook\t1\nKevin_Borseth\t1\tMichael_Caine\t4\nKoichi_Tanaka\t1\tRicardo_Lopez_Murphy\t1\nKristin_Scott\t1\tMax_Mayfield\t2\nLaura_Flessel\t1\tOwen_Nolan\t1\nLauren_Killian\t2\tTommy_Lewis\t1\nLaurie_Laychak\t1\tPa_Kou_Hang\t1\nLawrence_Di_Rita\t1\tMekhi_Phifer\t1\nLennox_Lewis\t3\tMonte_Kiffin\t1\nLeon_Lai\t1\tMonica_Lewinsky\t1\nLeon_Lai\t1\tSun_Myung_Moon\t1\nLily_Tomlin\t1\tMarcos_Daniel_Jimenez\t1\nLily_Tomlin\t2\tSim_Yong\t1\nLinn_Thornton\t1\tSherri_Coale\t1\nLinn_Thornton\t1\tTom_Cruise\t2\nLloyd_Mudiwa\t1\tSebastien_Grosjean\t1\nLois_Smart\t1\tTavis_Smiley\t1\nLuis_Sanchez\t1\tPetro_Symonenko\t1\nLuo_Linquan\t1\tMartin_ONeill\t1\nLuther_Htu\t1\tSteve_Karsay\t1\nLynn_Abraham\t1\tMichael_Keaton\t2\nLynn_Abraham\t2\tMariano_Zabaleta\t1\nMark_Richt\t3\tWilton_Gregory\t1\nMartha_Smith\t1\tRay_Sherman\t1\nMartin_ONeill\t1\tMike_Weir\t11\nMary-Kate_Olsen\t1\tSim_Yong\t1\nMatt_Siebrandt\t1\tRodney_Dangerfield\t1\nMaurice_Cheeks\t1\tSteve_Austin\t1\nMehmet_Okur\t1\tRandy_Travis\t1\nMichael_McNeely\t1\tSean_Combs\t1\nMichael_Milton\t1\tWilfredo_Moreno\t1\nMichael_Powell\t1\tScott_Weiland\t1\nMiguel_Rosseto\t1\tPascal_Rheaume\t1\nMiguel_Rosseto\t1\tRicky_Ponting\t2\nMike_Flanagan\t1\tMohammed_Baqir_al-Hakim\t2\nMike_Tice\t1\tPatti_Smith\t1\nMohammed_Baqir_al-Hakim\t2\tTatyana_Tomashova\t1\nMufti_Mohammad_Syed\t1\tRaoul_Ruiz\t4\nNancy_Demme\t1\tRoger_Mahony\t1\nNancy_Reagan\t1\tRafael_Bielsa\t1\nNatasha_McElhone\t2\tRobert_Nardelli\t1\nNathirah_Hussein\t1\tSusan_Whelan\t1\nNeil_Moritz\t1\tRoy_Williams\t2\nNeil_Moritz\t1\tXanana_Gusmao\t5\nNelson_Shanks\t1\tPedro_Martinez\t1\nNelson_Shanks\t1\tRen_Qingjin\t1\nNoor_Mohammed\t1\tScott_Weiland\t1\nOswald_Gruebel\t1\tRichard_Penniman\t1\nParthiv_Patel\t1\tTonya_Payne\t1\nPauley_Perrette\t1\tRoy_Williams\t4\nPeter_Arnett\t2\tRodney_Dangerfield\t1\nPeter_Greenspun\t1\tRod_Thorn\t1\nPeter_Harvey\t1\tWarren_Truss\t1\nRaja_Ramani\t1\tRichard_Greenberg\t1\nRaja_Ramani\t1\tWill_Young\t1\nRalph_Sampson\t1\tSamantha_Daniels\t1\nRaoul_Ruiz\t3\tTirunesh_Dibaba\t1\nRen_Qingjin\t1\tWilton_Gregory\t1\nRicardo_Lopez_Murphy\t2\tRoberto_Robaina\t1\nRicky_Ponting\t1\tRonaldo_Luis_Nazario_de_Lima\t4\nRobin_Cook\t1\tWilton_Gregory\t1\nRoger_Federer\t4\tSteve_Backley\t2\nRonald_Post\t1\tSteve_Karsay\t1\nRyan_Nyquist\t1\tWinona_Ryder\t22\nSally_Clark\t1\tScott_Ritter\t2\nSamantha_Daniels\t1\tTaufik_Hidayat\t3\nSherri_Coale\t1\tTroy_Garity\t1\nSteve_Valentine\t1\tToni_Braxton\t2\nTammy_Helm\t1\tTom_Smothers\t1\nTanya_Lindenmuth\t1\tTora_Takagi\t1\nToni_Braxton\t3\tYann_Martel\t2\nAdam_Scott\t1\t2\nAhmad_Masood\t1\t2\nAlan_Mulally\t1\t2\nAlexander_Rumyantsev\t1\t2\nAli_Abbas\t1\t2\nAmanda_Bynes\t1\t2\nAmanda_Bynes\t1\t3\nAmanda_Bynes\t1\t4\nAmanda_Bynes\t2\t3\nAndrei_Mikhnevich\t1\t2\nAngela_Merkel\t1\t2\nAngela_Merkel\t4\t5\nAngelina_Jolie\t1\t15\nAngelina_Jolie\t2\t11\nAngelina_Jolie\t8\t15\nAngelina_Jolie\t9\t15\nAngelina_Jolie\t11\t20\nAnthony_LaPaglia\t1\t2\nArlen_Specter\t1\t2\nAtal_Bihari_Vajpayee\t4\t6\nBarry_Alvarez\t1\t2\nBill_Belichick\t1\t2\nBill_Sizemore\t1\t2\nBoris_Yeltsin\t1\t2\nBritney_Spears\t4\t11\nBruce_Van_De_Velde\t1\t2\nCamilla_Parker_Bowles\t1\t2\nCarolina_Kluft\t1\t2\nCarson_Palmer\t1\t3\nCarson_Palmer\t2\t3\nCatherine_Deneuve\t1\t4\nCatherine_Deneuve\t2\t5\nCatherine_Deneuve\t3\t5\nCesar_Maia\t1\t2\nCharles_Moose\t10\t12\nChen_Liang_Yu\t1\t2\nChoi_Sung-hong\t1\t5\nChoi_Sung-hong\t2\t4\nChoi_Sung-hong\t3\t4\nChristina_Aguilera\t1\t2\nChristina_Aguilera\t1\t4\nChristina_Aguilera\t2\t4\nClara_Harris\t1\t5\nClara_Harris\t2\t4\nCoretta_Scott_King\t1\t2\nCoretta_Scott_King\t1\t3\nCoretta_Scott_King\t2\t3\nCourtney_Love\t1\t2\nDaniel_Day-Lewis\t1\t3\nDaniel_Day-Lewis\t2\t3\nDaniela_Hantuchova\t1\t2\nDebra_Messing\t1\t2\nDennis_Kucinich\t2\t7\nDick_Latessa\t1\t2\nDonald_Rumsfeld\t39\t110\nDonald_Rumsfeld\t51\t83\nDonald_Rumsfeld\t60\t67\nEduard_Shevardnadze\t1\t5\nEdward_Lu\t4\t6\nEdwina_Currie\t1\t2\nEdwina_Currie\t2\t4\nElizabeth_Shue\t1\t2\nEllen_Engleman\t1\t2\nEmma_Thompson\t1\t2\nEmma_Thompson\t1\t3\nEmma_Thompson\t2\t3\nEmmit_Smith\t1\t2\nFayssal_Mekdad\t1\t3\nFayssal_Mekdad\t1\t4\nFrances_Fisher\t1\t2\nFrank_Griswold\t1\t2\nGary_Forsee\t1\t2\nGloria_Macapagal_Arroyo\t3\t12\nGloria_Macapagal_Arroyo\t3\t25\nGloria_Macapagal_Arroyo\t8\t23\nGregory_Hines\t1\t2\nGus_Van_Sant\t1\t2\nGus_Van_Sant\t1\t3\nGwendal_Peizerat\t1\t2\nGwendal_Peizerat\t1\t3\nHal_Gehman\t2\t4\nHannah_Stockbauer\t1\t2\nHarrison_Ford\t1\t12\nHeizo_Takenaka\t1\t7\nHeizo_Takenaka\t3\t8\nHeizo_Takenaka\t3\t9\nHichiro_Naemura\t1\t2\nHipolito_Mejia\t1\t3\nHipolito_Mejia\t1\t4\nHolly_Hunter\t2\t6\nHolly_Hunter\t3\t7\nIgor_Ivanov\t5\t16\nIntisar_Ajouri\t2\t3\nJack_Nicholson\t1\t3\nJames_Blake\t1\t10\nJames_Blake\t4\t6\nJames_Blake\t5\t9\nJames_Kelly\t4\t5\nJames_Kopp\t2\t4\nJames_Smith\t1\t2\nJean_Chretien\t2\t39\nJean_Chretien\t6\t40\nJean_Chretien\t8\t35\nJean_Chretien\t19\t28\nJean_Chretien\t28\t52\nJean_Chretien\t29\t37\nJean_Chretien\t34\t36\nJeffrey_Immelt\t1\t2\nJennifer_Aniston\t1\t7\nJennifer_Aniston\t1\t16\nJennifer_Aniston\t6\t21\nJerry_Falwell\t1\t2\nJesse_Harris\t1\t3\nJesse_Harris\t2\t3\nJoan_Claybrook\t1\t2\nJoe_Calzaghe\t1\t2\nJoe_Nichols\t1\t3\nJoe_Nichols\t2\t4\nJoe_Nichols\t3\t4\nJoerg_Haider\t1\t2\nJohn_Abizaid\t1\t2\nJohn_Abizaid\t1\t7\nJohn_Abizaid\t2\t3\nJohn_Abizaid\t4\t7\nJohn_Abizaid\t4\t8\nJohn_Abizaid\t5\t6\nJohn_Jumper\t1\t2\nJohn_Mayer\t1\t2\nJohn_Mayer\t1\t3\nJohn_Negroponte\t5\t26\nJohn_Negroponte\t8\t17\nJohn_Negroponte\t28\t30\nJohn_Reilly\t1\t2\nJohn_Rowland\t1\t2\nJohnny_Depp\t1\t2\nJon_Corzine\t1\t2\nJose_Dirceu\t1\t2\nJose_Sarney\t1\t3\nJoseph_Estrada\t1\t3\nJoseph_Estrada\t2\t3\nJuan_Carlos_Ferrero\t9\t17\nJude_Law\t1\t2\nKatherine_Harris\t1\t2\nKatherine_Harris\t1\t4\nKatherine_Harris\t3\t4\nKelly_Clarkson\t1\t2\nKiki_Vandeweghe\t1\t2\nKofi_Annan\t8\t14\nKofi_Annan\t8\t32\nKofi_Annan\t13\t25\nLance_Bass\t1\t4\nLance_Bass\t2\t4\nLance_Bass\t2\t5\nLance_Bass\t4\t5\nLaurent_Gbagbo\t1\t2\nLee_Soo-hyuck\t1\t2\nLeslie_Caldwell\t2\t3\nLi_Zhaoxing\t2\t7\nLiu_Mingkang\t1\t2\nLuciano_Pavarotti\t1\t2\nLuis_Gonzalez_Macchi\t1\t2\nLuis_Gonzalez_Macchi\t2\t4\nLuis_Gonzalez_Macchi\t3\t4\nLuis_Gonzalez_Macchi\t3\t5\nLuis_Horna\t1\t4\nLuis_Horna\t2\t4\nLuiz_Inacio_Lula_da_Silva\t2\t9\nLuiz_Inacio_Lula_da_Silva\t2\t39\nLuiz_Inacio_Lula_da_Silva\t4\t8\nLuiz_Inacio_Lula_da_Silva\t7\t36\nLuiz_Inacio_Lula_da_Silva\t10\t19\nLuiz_Inacio_Lula_da_Silva\t11\t44\nLuiz_Inacio_Lula_da_Silva\t28\t44\nMadeleine_Albright\t1\t2\nMadeleine_Albright\t2\t3\nMarcus_Gronholm\t1\t2\nMarissa_Jaret_Winokur\t1\t2\nMark_Geragos\t1\t2\nMark_Philippoussis\t2\t7\nMark_Philippoussis\t3\t11\nMark_Philippoussis\t4\t6\nMark_Philippoussis\t4\t9\nMark_Philippoussis\t6\t7\nMark_Philippoussis\t6\t9\nMark_Philippoussis\t7\t9\nMarkus_Naslund\t1\t2\nMartha_Lucia_Ramirez\t2\t4\nMartha_Lucia_Ramirez\t3\t4\nMeryl_Streep\t14\t15\nMesut_Yilmaz\t1\t2\nMichael_Schumacher\t2\t13\nMichael_Schumacher\t2\t17\nMichael_Schumacher\t10\t11\nMichael_Schumacher\t13\t18\nMichel_Temer\t1\t2\nMichelle_Yeoh\t1\t2\nMichelle_Yeoh\t2\t5\nMonica_Seles\t1\t3\nMonica_Seles\t2\t6\nMonica_Seles\t4\t5\nMoshe_Katsav\t1\t2\nMoshe_Katsav\t2\t3\nMuhammad_Saeed_al-Sahhaf\t2\t5\nMuhammad_Saeed_al-Sahhaf\t3\t4\nMuhammad_Saeed_al-Sahhaf\t3\t5\nNancy_Sinatra\t1\t2\nNicanor_Duarte_Frutos\t3\t6\nNicanor_Duarte_Frutos\t3\t7\nNicanor_Duarte_Frutos\t3\t11\nPatti_Labelle\t2\t3\nPaul_Pierce\t1\t2\nPaul_Shanley\t2\t3\nPhil_Mickelson\t1\t2\nPrincess_Anne\t1\t2\nPrincess_Elisabeth\t1\t2\nPriscilla_Owen\t1\t2\nQueen_Rania\t2\t3\nRachel_Hunter\t1\t4\nRachel_Hunter\t2\t3\nRachel_Hunter\t3\t4\nRafael_Ramirez\t1\t2\nRafael_Ramirez\t2\t3\nRafael_Ramirez\t2\t4\nRalph_Lauren\t1\t2\nRecep_Tayyip_Erdogan\t2\t16\nRecep_Tayyip_Erdogan\t17\t19\nReese_Witherspoon\t3\t4\nRichard_Armitage\t2\t5\nRichard_Armitage\t6\t7\nRichard_Armitage\t6\t9\nRichard_Armitage\t8\t9\nRichard_Gephardt\t2\t10\nRichard_Gephardt\t3\t11\nRichard_Gephardt\t5\t9\nRichard_Gephardt\t10\t11\nRichard_Norton-Taylor\t1\t2\nRichie_Adubato\t1\t2\nRobert_Bullock\t1\t2\nRobert_Mueller\t4\t5\nRobert_Mugabe\t1\t2\nRobert_Zoellick\t2\t3\nRobert_Zoellick\t3\t7\nRoberto_Benigni\t1\t2\nRoh_Moo-hyun\t28\t32\nRoman_Polanski\t2\t3\nRoman_Polanski\t4\t6\nRon_Howard\t1\t2\nRonald_Reagan\t1\t3\nRonald_Reagan\t2\t3\nRosemarie_Stack\t1\t2\nRoy_Jones_Jr\t1\t2\nRupert_Murdoch\t1\t2\nSaburo_Kawabuchi\t1\t2\nSam_Mendes\t1\t2\nSam_Torrance\t1\t2\nSam_Torrance\t1\t3\nSam_Torrance\t2\t3\nSergei_Ivanov\t1\t3\nSergio_Vieira_De_Mello\t1\t2\nSergio_Vieira_De_Mello\t4\t9\nSergio_Vieira_De_Mello\t7\t8\nSergio_Vieira_De_Mello\t8\t11\nShannon_OBrien\t1\t2\nSheila_Fraser\t1\t2\nShimon_Peres\t1\t3\nShimon_Peres\t3\t4\nShimon_Peres\t4\t7\nSourav_Ganguly\t1\t3\nStanley_McChrystal\t1\t3\nStanley_McChrystal\t2\t3\nSteve_Lavin\t1\t4\nStrom_Thurmond\t1\t3\nSurakait_Sathirathai\t1\t2\nSusan_Collins\t1\t2\nTariq_Aziz\t3\t5\nTerry_Stotts\t1\t2\nThaksin_Shinawatra\t5\t6\nThomas_OBrien\t3\t9\nTim_Chapman\t1\t2\nTom_Hanks\t1\t7\nTom_Hanks\t4\t6\nTom_Hanks\t4\t9\nTom_Hanks\t5\t6\nTom_Reilly\t1\t2\nTom_Reilly\t2\t3\nTommy_Franks\t2\t4\nTommy_Franks\t4\t8\nValery_Giscard_dEstaing\t1\t2\nValery_Giscard_dEstaing\t1\t6\nVanessa_Williams\t1\t2\nVanessa_Williams\t2\t3\nVictoria_Beckham\t1\t2\nVictoria_Beckham\t1\t3\nVictoria_Beckham\t2\t3\nVladimir_Putin\t16\t21\nVladimir_Putin\t22\t31\nWarren_Beatty\t1\t2\nYasar_Yakis\t1\t4\nYoko_Ono\t1\t4\nYoko_Ono\t2\t4\nAJ_Lamas\t1\tChris_Penn\t1\nAhmed_Qureia\t1\tStanley_McChrystal\t3\nAhmet_Necdet_Sezer\t1\tJohn_Rowe\t1\nAlan_Dreher\t1\tEmily_Mason\t1\nAlan_Dreher\t1\tFrancisco_Santos\t1\nAlan_Dreher\t1\tRobert_Bullock\t2\nAlan_Mulally\t2\tTomomi_Morita\t1\nAlan_Stonecipher\t1\tLuis_Gonzalez\t1\nAlberto_Ruiz_Gallardon\t1\tJames_Smith\t2\nAlessandra_Cerna\t1\tJohn_Darby\t1\nAlessandra_Cerna\t1\tJohn_Rowland\t2\nAlessandra_Cerna\t1\tMarina_Canetti\t1\nAlessandra_Cerna\t1\tRandy_Jackson\t1\nAlessandra_Cerna\t1\tRichard_Jewell\t1\nAlex_Cabrera\t1\tFred_Huff\t1\nAlex_Cabrera\t1\tTommy_Franks\t9\nAlex_Corretja\t1\tJacky_Cheung\t1\nAlexandra_Spann\t1\tBeth_Blough\t1\nAli_Hammoud\t1\tRandy_Jackson\t1\nAlmeida_Baptista\t1\tJan_Peter_Balkenende\t1\nAlmeida_Baptista\t1\tJason_Mewes\t1\nAlmeida_Baptista\t1\tTaoufik_Mathlouthi\t1\nAmanda_Bynes\t2\tGrant_Rossenmeyer\t1\nAmy_Cotton\t1\tMatt_Walters\t1\nAmy_Cotton\t1\tRoberto_Guaterroma\t1\nAndrew_Jarecki\t1\tBarbara_Boxer\t1\nAndrew_Jarecki\t1\tTessa_Jowell\t1\nAndy_North\t1\tEric_Hinske\t1\nAndy_North\t1\tMarina_Kuptsova\t1\nAndy_North\t1\tPhil_Bennett\t1\nAngela_Merkel\t2\tHal_Gehman\t1\nAngela_Merkel\t3\tAnna_Jones\t1\nAngela_Merkel\t3\tFrank_Griswold\t1\nAnna_Jones\t1\tRay_Romano\t8\nAnthony_LaPaglia\t1\tRoberto_Guaterroma\t1\nArlen_Specter\t1\tBill_Butler\t1\nArlen_Specter\t2\tBarbara_Boxer\t1\nAtal_Bihari_Vajpayee\t3\tRuben_Sierra\t1\nAtsushi_Sato\t1\tMonica_Serra\t1\nBaburam_Bhattari\t1\tMuhammad_Saeed_al-Sahhaf\t4\nBarry_Alvarez\t1\tJohn_Jones\t1\nBill_Belichick\t1\tJohn_Petty\t1\nBill_Belichick\t1\tPhillipe_Comtois\t1\nBill_Belichick\t2\tRoger_Machado\t1\nBill_Byrne\t1\tBill_Sizemore\t1\nBill_Curry\t1\tEllen_Martin\t1\nBill_Curry\t1\tRichard_Langille\t1\nBill_Curry\t1\tRoberto_Guaterroma\t1\nBilly_Beane\t1\tTom_Welch\t1\nBoris_Trajkovski\t1\tLaurent_Gbagbo\t2\nBrad_Brownell\t1\tHussam_Mohammed_Amin\t1\nBrandon_Fails\t1\tChristian_Lirette\t1\nBrandon_Fails\t1\tEllen_Engleman\t2\nBrandon_Inge\t1\tEric_Lloyd\t1\nBrenda_Magana\t1\tNikolay_Davydenko\t1\nBrian_Jordan\t1\tJoe_Cravens\t1\nBrian_Lara\t1\tJohn_Darby\t1\nBrian_Lara\t1\tStanley_Nelson\t1\nCalvin_Harrison\t1\tLuis_Gonzalez_Macchi\t3\nCalvin_Harrison\t1\tRichard_Gephardt\t9\nCalvin_Harrison\t1\tSuzanne_Fox\t1\nCamilla_Parker_Bowles\t2\tGustavo_Franco\t1\nCarina_Lau_Ka-ling\t1\tLin_Yi-fu\t1\nCarlos_Ortega\t3\tLionel_Hampton\t1\nCarolina_Kluft\t3\tGwen_Stefani\t1\nCarson_Palmer\t3\tRichard_Jewell\t1\nCatherine_Deneuve\t4\tJade_Jagger\t1\nCatriona_Le_May_Doan\t1\tCraig_Burley\t1\nCatriona_Le_May_Doan\t1\tPhil_Mickelson\t1\nCharlie_Deane\t1\tQueen_Silvia\t1\nCharlotte_Church\t1\tKaoru_Hasuike\t1\nCharlotte_Church\t1\tSourav_Ganguly\t1\nCheryl_Hines\t1\tDu_Qinglin\t1\nCheryl_Hines\t1\tLane_Odom\t1\nChin-Hui_Tsao\t1\tRosemarie_Stack\t1\nChoi_Sung-hong\t4\tDebra_Messing\t1\nChoi_Sung-hong\t4\tFrederique_van_der_Wal\t1\nChris_Penn\t1\tPhil_Bennett\t1\nChris_Penn\t1\tTaoufik_Mathlouthi\t1\nChristian_Lirette\t1\tSargis_Sargsian\t1\nChristina_Aguilera\t2\tJoshua_Harapko\t1\nChristina_Aguilera\t4\tJerry_Falwell\t2\nChristopher_Amolsch\t1\tSteve_Pagliuca\t1\nChristopher_Matero\t1\tValery_Giscard_dEstaing\t1\nClaudine_Farrell\t1\tMike_Miller\t2\nClive_Woodward\t1\tDouglas_Faneuil\t1\nClive_Woodward\t1\tJose_Sarney\t2\nCoretta_Scott_King\t1\tRobert_Torricelli\t1\nCori_Enghusen\t1\tErnie_Stewart\t1\nCourtney_Love\t1\tJennifer_Aniston\t3\nCourtney_Love\t1\tRuben_Sierra\t1\nCraig_Burley\t1\tStanley_McChrystal\t3\nDaniel_Coats\t1\tKathryn_Grayson\t1\nDaniel_Day-Lewis\t3\tScott_Yates\t1\nDaniela_Hantuchova\t1\tJan_Peter_Balkenende\t1\nDanny_Green\t1\tRodrigo_Rato\t1\nDarin_Erstad\t1\tSteve_Fehr\t1\nDaryl_Smith\t1\tReese_Witherspoon\t4\nDavid_Hanson\t1\tRichard_Norton-Taylor\t1\nDavid_Hilt\t1\tHipolito_Mejia\t3\nDennis_Franchione\t1\tHugh_Miller\t1\nDennis_Kucinich\t3\tSam_Torrance\t2\nDennis_Kucinich\t4\tRadovan_Karadzic\t1\nDennis_Kucinich\t6\tIain_Anderson\t1\nDerek_Bond\t1\tJohn_Jumper\t1\nDiane_Ladd\t1\tTrevor_Watson\t1\nDick_Latessa\t1\tJohn_Burnett\t1\nDick_Latessa\t2\tMartha_Beatriz_Roque\t1\nDidier_Defago\t1\tJerry_Falwell\t2\nDon_Hewitt\t1\tGuennadi_Chipouline\t1\nDonald_Carty\t1\tErnie_Stewart\t1\nDonald_Carty\t1\tJeffrey_Immelt\t2\nDonald_Carty\t1\tVladimir_Putin\t34\nDu_Qinglin\t1\tMeryl_Streep\t7\nDustan_Mohr\t1\tEdward_Flynn\t1\nDustan_Mohr\t1\tEnrik_Vendt\t1\nDustan_Mohr\t1\tNikolay_Davydenko\t1\nDustan_Mohr\t1\tSourav_Ganguly\t5\nDustan_Mohr\t1\tTheo_Angelopoulos\t1\nE_Clay_Shaw\t1\tJerry_Jones\t1\nEduard_Shevardnadze\t3\tMartha_Lucia_Ramirez\t1\nEdward_Flynn\t1\tMadeleine_Albright\t2\nEdward_Lu\t1\tRobert_Zoellick\t7\nEdwina_Currie\t4\tEric_Hinske\t2\nElena_de_Chavez\t1\tJack_Goodman\t1\nElizabeth_Shue\t1\tOdai_Hussein\t1\nElizabeth_Shue\t2\tPrincess_Hisako\t1\nElizabeth_Shue\t2\tTerry_Stotts\t1\nEllen_Engleman\t1\tGuennadi_Chipouline\t1\nEllen_Saracini\t1\tRay_Romano\t6\nElsa_Zylberstein\t2\tMatt_Braker\t1\nElvis_Stojko\t1\tJerry_Falwell\t1\nElvis_Stojko\t1\tRobert_Mueller\t5\nEmma_Thompson\t3\tNathan_Doudney\t1\nEmmit_Smith\t2\tRafael_Ramirez\t4\nEnrik_Vendt\t1\tLane_Odom\t1\nEnrique_Oliu\t1\tMarkus_Naslund\t1\nEnrique_Oliu\t1\tWilliam_McDonough\t1\nEric_Hinske\t1\tJuan_Antonio_Samaranch\t1\nEric_Lloyd\t1\tJessica_Alba\t2\nEric_Lloyd\t1\tValery_Giscard_dEstaing\t1\nErnie_Harwell\t1\tQueen_Beatrix\t3\nFelipe_Perez_Roque\t2\tMC_Hammer\t1\nFrank_Griswold\t1\tYana_Klochkova\t1\nFrank_Zappa\t1\tMarkus_Naslund\t1\nFrederique_van_der_Wal\t1\tPedro_Mahecha\t1\nFrederique_van_der_Wal\t1\tRaja_Zafar-ul-Haq\t1\nGary_Leon_Ridgway\t1\tRobert_Wagner\t1\nGary_Leon_Ridgway\t1\tScott_Yates\t1\nGary_Leon_Ridgway\t1\tYana_Klochkova\t1\nGene_Keady\t1\tMark_Sacco\t1\nGeorge_Gregan\t1\tJohn_Negroponte\t27\nGlen_DaSilva\t1\tRachel_Hunter\t4\nGlen_DaSilva\t1\tRobert_Mugabe\t2\nGlen_Sather\t1\tLeslie_Caldwell\t3\nGlen_Sather\t1\tTrevor_Watson\t1\nGoran_Zivkovic\t1\tLee_Soo-hyuck\t2\nGoran_Zivkovic\t1\tMartha_Sahagun_de_Fox\t1\nGoran_Zivkovic\t1\tShane_Reynolds\t1\nGrant_Rossenmeyer\t1\tTrevor_Watson\t1\nGregory_Hines\t1\tMarcus_Gronholm\t2\nGuennadi_Chipouline\t1\tTom_Lantos\t1\nGuillermo_Ortiz\t1\tNestor_Santillan\t1\nGuillermo_Ortiz\t2\tStrom_Thurmond\t2\nGus_Van_Sant\t3\tNatalie_Stewart\t1\nGwen_Stefani\t1\tHugh_Miller\t1\nGwendal_Peizerat\t3\tTurner_Stevenson\t1\nHal_Gehman\t1\tScott_Yates\t1\nHamad_Bin_Isa_al-Khalifa\t1\tMarta_Dominguz\t1\nHarrison_Ford\t2\tMarcus_Gronholm\t1\nHarvey_Wachsman\t1\tNarayan_Singh_Pun\t1\nHeather_Locklear\t1\tJohn_Mayer\t3\nHeather_Locklear\t1\tPablo_Khulental\t1\nHichiro_Naemura\t2\tRichard_Jewell\t1\nHipolito_Mejia\t3\tJim_Ahern\t1\nHolly_Hunter\t3\tJoshua_Perper\t1\nHugh_Hefner\t1\tSam_Mendes\t1\nIain_Anderson\t1\tJames_Kopp\t2\nIain_Anderson\t1\tRecep_Tayyip_Erdogan\t12\nIlan_Goldfajn\t1\tJennifer_Aniston\t17\nIlie_Nastase\t1\tMichael_Lopez-Alegria\t1\nIntisar_Ajouri\t3\tLin_Yi-fu\t1\nItamar_Franco\t1\tJessica_Alba\t1\nJack_Goodman\t1\tKirsten_Dunst\t1\nJack_Goodman\t1\tSaburo_Kawabuchi\t2\nJack_Nicholson\t1\tPat_Wharton\t1\nJack_Valenti\t1\tShinzo_Abe\t1\nJake_Plummer\t1\tJose_Dirceu\t2\nJakob_Kellenberger\t1\tLara_Logan\t1\nJakob_Kellenberger\t1\tPedro_Mahecha\t1\nJames_Maguire\t1\tRonald_Reagan\t1\nJames_Smith\t2\tKyra_Sedgwick\t1\nJason_Statham\t1\tKwame_Kilpatrick\t1\nJawad_Boulus\t1\tNarayan_Singh_Pun\t1\nJean-Rene_Fourtou\t1\tRoh_Moo-hyun\t28\nJeannette_Biedermann\t1\tKenny_Brack\t1\nJennifer_Granholm\t1\tTrudi_Lacey\t1\nJerelle_Kraus\t1\tWarren_Beatty\t1\nJesus_Cardenal\t1\tPablo_Khulental\t1\nJim_Hendry\t1\tLarry_Beinfest\t1\nJoan_Claybrook\t2\tPablo_Latras\t1\nJoaquin_Sanchez\t1\tRichard_Armitage\t1\nJoe_Cocker\t1\tMartha_Beatriz_Roque\t1\nJoe_Cocker\t1\tRandy_Jackson\t1\nJoe_Cravens\t1\tSam_Mendes\t1\nJohn_Barnett\t1\tKathryn_Grayson\t1\nJohn_Darby\t1\tMasum_Turker\t3\nJohn_Ferguson\t1\tPaul_Schrader\t1\nJohn_Paul_DeJoria\t1\tPaul_Pierce\t1\nJohn_Reilly\t2\tPrincess_Hisako\t1\nJohn_Rowland\t2\tSteve_Kerr\t1\nJohnny_Depp\t1\tLeo_Mullin\t1\nJohnny_Depp\t1\tTrudi_Lacey\t1\nJohnny_Htu\t1\tRoger_Machado\t1\nJose_Dirceu\t1\tRobert_Torricelli\t3\nJuan_Antonio_Samaranch\t1\tStefan_Koubek\t1\nJuergen_Braehmer\t1\tShawn_Bradley\t1\nJuergen_Braehmer\t1\tTravis_Rudolph\t1\nKathryn_Grayson\t1\tQueen_Beatrix\t4\nKathryn_Grayson\t1\tRuben_Sierra\t1\nKathryn_Grayson\t1\tShannon_OBrien\t2\nKatie_Holmes\t1\tPark_Jung_Sung\t1\nKenny_Brack\t1\tLin_Yi-fu\t1\nKwame_Kilpatrick\t1\tShannon_OBrien\t2\nKyra_Sedgwick\t1\tReese_Witherspoon\t3\nKyra_Sedgwick\t1\tRichard_Hellfant\t1\nKyra_Sedgwick\t1\tWilbert_Elki_Meza_Majino\t1\nLance_Bass\t3\tQueen_Rania\t5\nLara_Logan\t1\tTurner_Stevenson\t1\nLaszlo_Kovacs\t1\tLuke_Ridnour\t1\nLaszlo_Kovacs\t1\tOlesya_Bonabarenko\t2\nLee_Byung-woong\t1\tWilliam_Hochul\t2\nLee_Soo-hyuck\t1\tRobert_Bullock\t1\nLeslie_Caldwell\t1\tShane_Reynolds\t1\nLeslie_Caldwell\t2\tTurner_Stevenson\t1\nLi_Zhaoxing\t4\tSusan_Sarandon\t4\nLinda_Mason\t1\tRupert_Murdoch\t2\nLiu_Mingkang\t2\tPedro_Pauleta\t1\nLiu_Ye\t1\tRuben_Sierra\t1\nLiu_Ye\t1\tSteve_Alford\t1\nLoretta_Lynn_Harper\t1\tMichel_Temer\t2\nLuis_Gonzalez\t1\tWarren_Beatty\t1\nMadeleine_Albright\t2\tMartha_Lucia_Ramirez\t2\nMaha_Habib\t1\tPrincess_Elisabeth\t1\nMarc_Anthony\t1\tMichael_Pfleger\t1\nMarcus_Gronholm\t1\tTeresa_Graves\t1\nMarcus_Gronholm\t1\tTommy_Franks\t8\nMarina_Canetti\t1\tSaied_Hadi_al_Mudarissi\t1\nMario_Alfaro-Lopez\t1\tRichard_Armitage\t6\nMario_Alfaro-Lopez\t1\tTom_Lantos\t1\nMark_Dacey\t1\tTurner_Stevenson\t1\nMark_Dacey\t2\tSteve_Cox\t1\nMark_Kelly\t1\tMark_Lazarus\t1\nMark_Lazarus\t1\tRonnie_Jagday\t1\nMark_Sacco\t1\tTono_Suratman\t1\nMarkus_Naslund\t1\tRobert_Wagner\t1\nMartin_Kristof\t1\tTaia_Balk\t1\nMatt_Anderson\t1\tSeth_Gorney\t1\nMatt_Braker\t1\tSurakait_Sathirathai\t1\nMatt_Roney\t1\tWilbert_Elki_Meza_Majino\t1\nMichael_Bolton\t1\tRobert_Bullock\t2\nMichael_Pfleger\t1\tRobert_Ehrlich\t2\nMichael_Schumacher\t17\tNatalie_Imbruglia\t1\nMichael_Shane_Jolly\t1\tTrudi_Lacey\t1\nMichel_Temer\t1\tPaul_Shanley\t1\nMichelle_Yeoh\t5\tSaburo_Kawabuchi\t1\nMike_Helton\t1\tOctavio_Lara\t1\nMike_Miller\t2\tToby_Keith\t1\nMonica_Seles\t2\tQueen_Beatrix\t4\nNancy_Sinatra\t1\tSargis_Sargsian\t1\nPablo_Khulental\t1\tRosemarie_Stack\t2\nPark_Jung_Sung\t1\tRobert_Bullock\t2\nPaul_Pierce\t2\tRobert_Bullock\t2\nPedro_Pauleta\t1\tRobert_Tyrrell\t1\nPedro_Pauleta\t1\tTommy_Franks\t14\nPhil_Morris\t1\tPriscilla_Owen\t2\nPhil_Morris\t1\tRoberto_Guaterroma\t1\nPhil_Morris\t1\tYuri_Luzhkov\t1\nQueen_Rania\t3\tRonnie_Jagday\t1\nRadovan_Karadzic\t1\tRichard_Langille\t1\nRandy_Jackson\t1\tSteve_Alford\t1\nReese_Witherspoon\t1\tSam_Torrance\t1\nReese_Witherspoon\t2\tToni_Jennings\t1\nRina_Lazo\t1\tRonald_Reagan\t1\nRobert_Ehrlich\t2\tRon_Lantz\t1\nRobert_Mugabe\t1\tTessa_Jowell\t1\nRobert_Torricelli\t3\tTaoufik_Mathlouthi\t1\nRobert_Tyrrell\t1\tYuri_Luzhkov\t1\nRobert_Wagner\t1\tTommy_Franks\t11\nRobin_Wright_Penn\t1\tSam_Torrance\t3\nRobin_Wright_Penn\t1\tYoko_Ono\t6\nRogelio_Ramos\t1\tRyan_Newman\t1\nRoman_Polanski\t4\tToby_Keith\t1\nRonnie_Jagday\t1\tSidney_Kimmel\t1\nScott_Yates\t1\tSteve_Cox\t1\nSean_Patrick_Thomas\t1\tWilliam_Hochul\t1\nSharon_Osbourne\t2\tShimon_Peres\t6\nStefan_Koubek\t1\tSteve_Alford\t1\nTeri_Garr\t1\tYana_Klochkova\t1\nWarren_Beatty\t2\tWilliam_McDonough\t1\nAleksander_Kwasniewski\t1\t2\nAleksander_Kwasniewski\t2\t4\nAleksander_Kwasniewski\t3\t4\nAli_Naimi\t1\t2\nAli_Naimi\t1\t5\nAli_Naimi\t2\t8\nAmanda_Beard\t1\t2\nAndrew_Cuomo\t1\t2\nArnold_Schwarzenegger\t8\t31\nArnold_Schwarzenegger\t9\t39\nArnold_Schwarzenegger\t11\t41\nArnold_Schwarzenegger\t12\t37\nArnold_Schwarzenegger\t13\t39\nArnold_Schwarzenegger\t22\t33\nArnold_Schwarzenegger\t28\t41\nBernard_Lord\t1\t2\nBeth_Jones\t1\t2\nBranko_Crvenkovski\t2\t3\nBrigitte_Boisselier\t1\t2\nBruce_Springsteen\t1\t2\nBruce_Springsteen\t1\t3\nBruce_Springsteen\t2\t3\nBruce_Springsteen\t2\t4\nCalista_Flockhart\t1\t5\nCalista_Flockhart\t4\t6\nCameron_Diaz\t1\t3\nCameron_Diaz\t3\t6\nCarol_Moseley_Braun\t1\t2\nCarrie-Anne_Moss\t1\t3\nCarrie-Anne_Moss\t1\t5\nCarrie-Anne_Moss\t2\t5\nCarrie-Anne_Moss\t3\t5\nCarrie-Anne_Moss\t4\t5\nCharlton_Heston\t1\t5\nCharlton_Heston\t2\t5\nCharlton_Heston\t4\t6\nChris_Bell\t1\t2\nChung_Mong-hun\t1\t2\nCiro_Gomes\t1\t2\nCiro_Gomes\t2\t5\nCiro_Gomes\t3\t5\nColin_Montgomerie\t1\t4\nColin_Montgomerie\t2\t3\nColin_Montgomerie\t2\t5\nColin_Montgomerie\t4\t5\nDavid_Trimble\t1\t4\nDavid_Trimble\t1\t5\nDavid_Trimble\t3\t4\nDavid_Wolf\t1\t2\nDemetrius_Ferraciu\t1\t2\nDennis_Powell\t1\t2\nDesiree_Lemosi\t1\t2\nDolma_Tsering\t1\t2\nDoug_Melvin\t1\t2\nEdward_Norton\t1\t2\nEmanuel_Ginobili\t2\t3\nEric_Clapton\t1\t2\nFabiola_Zuluaga\t1\t2\nFaye_Dunaway\t1\t2\nFerenc_Madl\t1\t2\nFernando_Gonzalez\t1\t5\nFernando_Gonzalez\t1\t8\nFernando_Gonzalez\t2\t6\nFilippo_Inzaghi\t1\t2\nFilippo_Inzaghi\t1\t3\nFilippo_Inzaghi\t2\t3\nFrancis_George\t1\t2\nFranz_Beckenbauer\t1\t2\nFranz_Muentefering\t1\t2\nFranz_Muentefering\t3\t4\nGary_Bergeron\t1\t2\nGeorge_Foreman\t1\t2\nGeorge_Pataki\t1\t3\nGeorge_Pataki\t3\t4\nGeorge_Pataki\t3\t5\nGeorge_Roy_Hill\t1\t2\nGonzalo_Sanchez_de_Lozada\t7\t8\nGonzalo_Sanchez_de_Lozada\t7\t10\nGonzalo_Sanchez_de_Lozada\t8\t11\nGonzalo_Sanchez_de_Lozada\t8\t12\nGregory_Geoffroy\t1\t2\nGuillermo_Coria\t7\t13\nGuillermo_Coria\t19\t22\nHabib_Rizieq\t1\t3\nHabib_Rizieq\t2\t3\nHamid_Karzai\t1\t3\nHamid_Karzai\t4\t13\nHamid_Karzai\t19\t22\nHilmi_Ozkok\t1\t2\nHorst_Koehler\t1\t2\nHorst_Koehler\t1\t3\nHorst_Koehler\t2\t3\nHoward_Schultz\t1\t2\nIlan_Ramon\t1\t2\nJavier_Solana\t4\t8\nJavier_Solana\t6\t9\nJavier_Solana\t6\t10\nJay_Garner\t1\t3\nJean-Claude_Braquet\t1\t2\nJean-Claude_Trichet\t1\t2\nJean_Charest\t1\t10\nJean_Charest\t2\t11\nJean_Charest\t7\t10\nJean_Charest\t8\t13\nJean_Charest\t8\t17\nJefferson_Perez\t1\t2\nJim_Edmonds\t1\t2\nJohn_Kerry\t1\t5\nJohn_Kerry\t3\t4\nJohn_Kerry\t4\t12\nJohn_Kerry\t9\t13\nJohn_Malkovich\t1\t2\nJohn_Malkovich\t1\t3\nJohn_McCallum\t1\t2\nJohn_McEnroe\t1\t2\nJohn_Rosa\t1\t3\nJohnny_Unitas\t1\t2\nJong_Wook_Lee\t3\t4\nJose_Maria_Aznar\t1\t13\nJose_Maria_Aznar\t6\t13\nJose_Maria_Aznar\t6\t20\nJose_Maria_Aznar\t15\t23\nJose_Maria_Aznar\t17\t22\nJulianne_Moore\t4\t9\nJulianne_Moore\t6\t16\nJulianne_Moore\t6\t19\nJulianne_Moore\t7\t18\nJulianne_Moore\t12\t15\nJulio_Iglesias_Jr\t1\t2\nJustin_Timberlake\t1\t2\nJustin_Timberlake\t1\t3\nJustin_Timberlake\t5\t6\nJustin_Timberlake\t6\t8\nJustine_Pasek\t2\t5\nJustine_Pasek\t2\t6\nJustine_Pasek\t6\t7\nKathleen_Glynn\t1\t2\nKim_Dae-jung\t1\t3\nKim_Dae-jung\t1\t5\nKim_Dae-jung\t2\t3\nKim_Dae-jung\t4\t6\nKirk_Johnson\t1\t2\nKirk_Johnson\t1\t3\nKirk_Johnson\t2\t3\nKosuke_Kitajima\t1\t2\nKurt_Russell\t1\t2\nLarry_Coker\t1\t4\nLarry_Coker\t2\t4\nLarry_Ellison\t1\t2\nLarry_Ellison\t1\t3\nLawrence_MacAulay\t1\t2\nLeBron_James\t1\t4\nLeBron_James\t1\t5\nLeBron_James\t3\t4\nLeBron_James\t4\t5\nLea_Fastow\t1\t2\nLee_Hoi-chang\t1\t2\nLee_Hoi-chang\t2\t4\nLi_Peng\t2\t6\nLi_Peng\t4\t7\nLi_Peng\t4\t8\nLi_Peng\t4\t9\nLim_Dong-won\t1\t2\nLindsay_Benko\t1\t2\nLou_Piniella\t1\t3\nLou_Piniella\t2\t3\nLucio_Gutierrez\t1\t2\nLucio_Gutierrez\t3\t10\nLucio_Gutierrez\t6\t7\nLucio_Gutierrez\t6\t8\nLucio_Gutierrez\t8\t13\nLuis_Figo\t2\t3\nLuis_Figo\t2\t4\nLuke_Walton\t1\t2\nMarat_Safin\t1\t2\nMarat_Safin\t1\t3\nMarat_Safin\t2\t3\nMariana_Pollack\t1\t3\nMariana_Pollack\t2\t3\nMartin_McGuinness\t3\t4\nMasahiko_Nagasawa\t1\t2\nMichael_Chiklis\t1\t4\nMichael_Chiklis\t1\t5\nMichael_Chiklis\t2\t3\nMichael_Chiklis\t2\t4\nMichael_Chiklis\t3\t5\nMichael_Chiklis\t4\t5\nMichael_Douglas\t1\t2\nMichael_Douglas\t1\t3\nMichael_Douglas\t2\t4\nMichael_Douglas\t3\t6\nMichael_Douglas\t4\t5\nMichael_Kostelnik\t1\t2\nMichael_Leavitt\t1\t2\nMichelle_Branch\t1\t2\nMichelle_Kwan\t1\t4\nMichelle_Kwan\t4\t5\nMichelle_Kwan\t6\t8\nMike_Montgomery\t1\t2\nMike_Tyson\t1\t3\nMikhail_Wehbe\t1\t3\nMinnie_Driver\t1\t2\nMiyako_Miyazaki\t1\t2\nMuammar_Gaddafi\t1\t2\nMunir_Akram\t1\t2\nNadia_Petrova\t1\t2\nNadia_Petrova\t1\t3\nNadia_Petrova\t1\t4\nNadia_Petrova\t2\t3\nNadia_Petrova\t3\t4\nNadia_Petrova\t4\t5\nNancy_Pelosi\t1\t7\nNancy_Pelosi\t2\t9\nNancy_Pelosi\t4\t15\nNancy_Pelosi\t12\t14\nNia_Vardalos\t1\t3\nNia_Vardalos\t3\t4\nNia_Vardalos\t3\t5\nNicholas_Byron\t1\t2\nNikki_Reed\t1\t2\nOlivia_Newton-John\t1\t2\nParadorn_Srichaphan\t2\t4\nParadorn_Srichaphan\t4\t5\nPascal_Quignard\t1\t3\nPascal_Quignard\t2\t3\nPatrice_Chereau\t1\t2\nPedro_Malan\t1\t5\nPedro_Malan\t3\t5\nPeter_Harrison\t1\t2\nRainer_Schuettler\t1\t4\nRainer_Schuettler\t2\t3\nRainer_Schuettler\t2\t4\nRaymond_Odierno\t1\t2\nRebecca_Romijn-Stamos\t1\t2\nRebecca_Romijn-Stamos\t1\t3\nRebecca_Romijn-Stamos\t1\t4\nRebecca_Romijn-Stamos\t3\t4\nRichard_Krajicek\t1\t3\nRichard_Krajicek\t2\t3\nRick_Santorum\t1\t2\nRick_Santorum\t1\t3\nRick_Santorum\t2\t3\nRick_Stansbury\t2\t3\nRob_Marshall\t1\t3\nRob_Marshall\t1\t6\nRob_Marshall\t2\t4\nRobert_Blackwill\t1\t2\nRobert_Duvall\t2\t8\nRobert_Horan\t1\t2\nRobert_Redford\t3\t4\nRobert_Redford\t7\t8\nRobinson_Stevenin\t1\t2\nRoger_Clemens\t1\t2\nSadie_Frost\t1\t2\nSaparmurat_Niyazov\t1\t2\nSarah_Michelle_Gellar\t1\t2\nSarah_Michelle_Gellar\t1\t3\nSarah_Michelle_Gellar\t2\t3\nSerena_Williams\t1\t41\nSerena_Williams\t3\t32\nSerena_Williams\t14\t40\nSerena_Williams\t41\t47\nSergey_Lavrov\t1\t3\nSergey_Lavrov\t1\t6\nSergey_Lavrov\t2\t4\nSergey_Lavrov\t3\t5\nSergey_Lavrov\t3\t7\nSergey_Lavrov\t4\t8\nShane_Mosley\t1\t2\nSheila_Copps\t2\t3\nSheila_Copps\t3\t4\nShia_LaBeouf\t1\t2\nSteve_Nash\t1\t3\nSteve_Nash\t1\t4\nSteve_Nash\t2\t4\nSteve_Nash\t2\t5\nSteve_Spurrier\t1\t2\nSteve_Waugh\t1\t2\nSusilo_Bambang_Yudhoyono\t1\t2\nSusilo_Bambang_Yudhoyono\t1\t3\nSusilo_Bambang_Yudhoyono\t1\t4\nSusilo_Bambang_Yudhoyono\t3\t4\nSuzanne_Gaudet\t1\t2\nThomas_Bjorn\t1\t2\nTim_Conway\t1\t2\nTim_Conway\t1\t3\nTim_Robbins\t1\t5\nTim_Robbins\t2\t4\nTim_Robbins\t3\t5\nTom_Craddick\t1\t3\nTom_Craddick\t2\t4\nTom_Craddick\t3\t4\nTracee_Ellis_Ross\t1\t2\nVenus_Williams\t2\t9\nVivica_Fox\t1\t2\nWilliam_Ford_Jr\t1\t2\nWilliam_Ford_Jr\t2\t6\nWilliam_Ford_Jr\t5\t6\nWilliam_Rehnquist\t1\t2\nYossi_Beilin\t1\t2\nAaron_Eckhart\t1\tAkiko_Morigami\t1\nAaron_Eckhart\t1\tAnFernce_Negron\t1\nAaron_Eckhart\t1\tSadie_Frost\t1\nAbdel_Aziz_Al-Hakim\t1\tJoe_Darrell\t1\nAbdullah_al-Attiyah\t1\tRachel_Wadsworth\t1\nAbdullah_al-Attiyah\t2\tJohn_Lisowski\t1\nAbraham_Foxman\t1\tDoug_Melvin\t2\nAbraham_Foxman\t1\tNikki_Reed\t1\nAdam_Rich\t1\tJean-Claude_Trichet\t1\nAdam_Rich\t1\tJohn_Goold\t1\nAdam_Rich\t1\tLisa_Girman\t1\nAdam_Rich\t1\tRoger_Clemens\t1\nAdam_Rich\t1\tSuzanne_Somers\t1\nAdrian_Fernandez\t1\tNicolas_Massu\t1\nAiysha_Smith\t1\tYossi_Beilin\t1\nAkiko_Morigami\t1\tShane_Mosley\t1\nAlastair_Johnston\t1\tAleksander_Kwasniewski\t4\nAlastair_Johnston\t1\tBill_Duffey\t1\nAlbert_Brooks\t1\tRobinson_Stevenin\t1\nAlbert_Brooks\t1\tSheila_Copps\t3\nAleksander_Kwasniewski\t1\tGuangdong_Ou_Guangyuan\t1\nAleksander_Kwasniewski\t2\tGeorge_Plimpton\t1\nAleksander_Kwasniewski\t2\tJuan_Jose_Lucas\t1\nAleksander_Kwasniewski\t3\tBilly_Crawford\t1\nAleksander_Voloshin\t1\tKristen_Rivera\t1\nAlessandro_Nesta\t1\tPaul_Kelleher\t1\nAlfredo_di_Stefano\t1\tChris_Moore\t1\nAlfredo_di_Stefano\t1\tMaryn_McKenna\t1\nAlfredo_di_Stefano\t1\tShamai_Leibowitz\t1\nAli_Naimi\t5\tMorris_Dees\t1\nAline_Chretien\t1\tGuillermo_Coria\t17\nAmy_Yasbeck\t1\tChawki_Armali\t1\nAnFernce_Negron\t1\tKristen_Rivera\t1\nAndrew_Cuomo\t2\tTodd_Petit\t1\nAnil_Ramsook\t1\tTakeo_Hiranuma\t1\nAnnie_Chaplin\t1\tCharles_Tannok\t1\nAntanas_Valionis\t1\tWilliam_Rehnquist\t1\nAvril_Lavigne\t1\tShia_LaBeouf\t2\nBarbara_Becker\t1\tChris_Noth\t1\nBarbara_Becker\t1\tFranz_Beckenbauer\t2\nBarbora_Strycova\t1\tChristiane_Wulff\t1\nBarry_Diller\t1\tMeles_Zenawi\t1\nBartosz_Kizierowski\t1\tFaye_Wong\t1\nBen_Wallace\t1\tNia_Vardalos\t5\nBernard_Lord\t1\tJoxel_Garcia\t1\nBernard_Lord\t2\tBing_Crosby\t1\nBeyonce_Knowles\t1\tMaurice_Papon\t1\nBill_Duffey\t1\tJong_Wook_Lee\t1\nBill_Guerin\t1\tJulie_Goodenough\t1\nBill_Herrion\t1\tFernando_Valenzuela\t1\nBill_Herrion\t1\tJohnny_Unitas\t2\nBill_Richardson\t1\tGeorge_McCloud\t1\nBilly_Andrade\t1\tKellie_Coffey\t1\nBilly_Crawford\t1\tJim_Edmonds\t2\nBilly_Edelin\t1\tHimmler_Rebu\t1\nBing_Crosby\t1\tStephen_Webster\t1\nBixente_LIzarazu\t1\tChris_Bell\t2\nBixente_LIzarazu\t1\tTodd_Wike\t1\nBlythe_Danner\t2\tHank_McKinnell\t1\nBob_Eskridge\t1\tMarco_Pantani\t1\nBob_Hartley\t1\tLou_Piniella\t3\nBob_Krueger\t1\tGordana_Grubin\t1\nBob_Krueger\t1\tMariana_Pollack\t1\nBob_Sulkin\t1\tBranko_Crvenkovski\t3\nBobby_Kielty\t1\tRobert_Horan\t2\nBrajesh_Mishra\t1\tMark_Podlesny\t1\nBrandon_Knight\t1\tClaudia_Cardinale\t1\nBrandon_Knight\t1\tPhil_Donahue\t1\nBrandon_Lloyd\t1\tCha_Yung-gu\t1\nBrandon_Lloyd\t1\tJames_Coburn\t1\nBrian_Schneider\t1\tMichael_Rolinee\t1\nBruce_Springsteen\t4\tIon_Tiriac\t1\nCameron_Diaz\t2\tShia_LaBeouf\t1\nCameron_Diaz\t5\tNadia_Petrova\t2\nCarla_Gugino\t1\tKelly_Osbourne\t1\nCarla_Gugino\t1\tKenny_Chesney\t1\nCarla_Gugino\t1\tMichael_Douglas\t3\nCarla_Gugino\t1\tMiguel_Aldana_Ibarra\t1\nCarlos_Fasciolo\t1\tDebbie_Allen\t1\nCarol_Williams\t1\tGorden_Tallis\t1\nCarrie-Anne_Moss\t2\tHoracio_Julio_Pina\t1\nCarrie-Anne_Moss\t4\tMichael_Goldrich\t1\nCasey_Crowder\t1\tKeiko_Sofia_Fujimori\t1\nCasey_Crowder\t1\tPhoenix_Chang\t1\nCha_Yung-gu\t1\tTakeo_Hiranuma\t1\nCharlene_Barshefsky\t1\tJames_Coburn\t1\nCharles_Tannok\t1\tLarry_Coker\t4\nCharlie_Hunnam\t1\tHestrie_Cloette\t1\nCharlie_Hunnam\t1\tVeronica_Lake\t1\nCharlton_Heston\t4\tDesiree_Lemosi\t1\nChawki_Armali\t1\tShoshannah_Stern\t1\nChris_Bell\t2\tPharrell_Williams\t1\nChris_Forsyth\t1\tNikki_Reed\t2\nChris_Neil\t1\tJohn_Kerry\t14\nChris_Noth\t1\tTakeshi_Kitano\t1\nChristine_Arron\t1\tRobert_Duvall\t5\nChristine_Arron\t1\tSteve_McManaman\t1\nChuck_Hagel\t1\tJeff_Roehm\t1\nChuck_Hagel\t1\tMargaret_Hasley\t1\nChuck_Hagel\t1\tSebastian_Porto\t1\nCliff_Ellis\t1\tJohn_Lisowski\t1\nClifford_Etienne\t1\tPharrell_Williams\t1\nClifford_Etienne\t1\tRay_Lewis\t1\nClive_Lloyd\t1\tHarland_Braun\t1\nColin_Montgomerie\t1\tHoracio_Julio_Pina\t1\nDaniel_Montenegro\t1\tGustavo_Cisneros\t1\nDaniel_Montenegro\t1\tNadia_Petrova\t4\nDaniel_Montenegro\t1\tOmar_Khan_Sharif\t1\nDaniel_Montenegro\t1\tRick_Santorum\t1\nDanny_Glover\t1\tMorris_Dees\t1\nDanny_Glover\t1\tPatrice_Chereau\t2\nDavid_Trimble\t4\tJohn_Elway\t1\nDavid_Wolf\t2\tEmanuel_Ginobili\t4\nDavid_Wolf\t2\tKathleen_Glynn\t2\nDavid_Wolf\t2\tVenus_Williams\t6\nDebbie_Allen\t1\tSebastian_Cuattrin\t1\nDemetrius_Ferraciu\t2\tEdward_Seaga\t1\nDenise_Johnson\t1\tNikki_Cascone\t1\nDenise_Johnson\t2\tEmanuel_Ginobili\t3\nDenise_Johnson\t2\tFilippo_Inzaghi\t2\nDennis_Powell\t1\tHorst_Koehler\t3\nDennis_Powell\t1\tMichael_Chiklis\t2\nDesiree_Lemosi\t2\tDion_Glover\t1\nDion_Glover\t1\tMichael_Weiss\t1\nDoug_Melvin\t1\tGustavo_Cisneros\t1\nDoug_Melvin\t3\tEmma_Nicholson\t1\nDoug_Racine\t1\tMarisol_Breton\t1\nEd_Book\t1\tNur_Jaafar\t1\nEd_Case\t1\tSadam_Hassan\t1\nEdward_Seaga\t1\tGeorge_Roy_Hill\t1\nEdward_Seaga\t1\tSarah_Price\t1\nEileen_Spina\t1\tJeff_Bzdelik\t1\nEkke_Hard_Forberg\t1\tNikki_Reed\t1\nElena_Bereznaya\t1\tLene_Espersen\t1\nElena_Bereznaya\t1\tMario_Lobo_Zagallo\t1\nElena_Bereznaya\t1\tRick_Stansbury\t2\nElisha_Cuthbert\t1\tMariana_Pollack\t2\nEliza_Manningham-Buller\t1\tLawrence_MacAulay\t2\nEric_Clapton\t2\tNiall_Connolly\t1\nEric_Taino\t1\tMichael_Goldrich\t1\nFarida_Ragoonanan\t1\tJorge_Enrique_Jimenez\t1\nFelix_Trinidad\t1\tJoe_Darrell\t1\nFelix_Trinidad\t1\tJustine_Pasek\t4\nFernando_Valenzuela\t1\tLuis_Figo\t3\nFernando_Valenzuela\t1\tNicolas_Macrozonaris\t1\nFilippo_Inzaghi\t1\tMohammed_Al_Hindi\t1\nFilippo_Inzaghi\t1\tShanna_Zolman\t1\nFrancis_Ricciardone\t1\tSven_Goran_Eriksson\t1\nFranz_Beckenbauer\t1\tGerald_Ford\t1\nFranz_Beckenbauer\t2\tRick_Bragg\t1\nGeorge_Foreman\t2\tRalph_Goodale\t1\nGeorge_Pataki\t1\tJames_Dingemans\t1\nGeorge_Pataki\t2\tMasahiko_Nagasawa\t2\nGeorge_Plimpton\t1\tXimena_Bohorquez\t1\nGerald_Ford\t1\tKurt_Budke\t1\nGerald_Ford\t1\tLi_Peng\t1\nGiselle_Estefania_Tavarelli\t1\tJohn_Sununu\t1\nGiulio_Andreotti\t1\tKurt_Russell\t2\nGiulio_Andreotti\t1\tMichael_Denzel\t1\nGonzalo_Sanchez_de_Lozada\t3\tShannyn_Sossamon\t1\nGraciano_Rocchigiani\t1\tJuan_Jose_Lucas\t1\nGregory_Geoffroy\t1\tSybille_Schmid\t1\nGregory_Geoffroy\t2\tKosuke_Kitajima\t1\nGregory_Geoffroy\t2\tWycliffe_Grousbeck\t1\nGuangdong_Ou_Guangyuan\t1\tLi_Ruihuan\t1\nGuangdong_Ou_Guangyuan\t1\tNicole\t1\nGustavo_Cisneros\t1\tSteve_Blake\t1\nGustavo_Noboa\t1\tLea_Fastow\t2\nHabib_Hisham\t1\tPat_Rochester\t1\nHabib_Hisham\t1\tPaul_Wilson\t1\nHabib_Rizieq\t1\tOlivia_Newton-John\t2\nHabib_Rizieq\t3\tKenny_Chesney\t1\nHal_Sutton\t2\tKellie_Coffey\t1\nHarland_Braun\t1\tKathryn_Tucker\t1\nHassanal_Bolkiah\t1\tMichael_Leavitt\t1\nHestrie_Cloette\t1\tNicholas_Byron\t2\nHitoshi_Tanaka\t1\tRamon_Ponce_de_Leon\t1\nHoracio_Julio_Pina\t1\tZaini_Abdullah\t1\nHuang_Suey-Sheng\t1\tLucrecia_Orozco\t1\nHuang_Suey-Sheng\t1\tPaul_Cerjan\t1\nHuang_Suey-Sheng\t1\tRobinson_Stevenin\t1\nIan_Knop\t1\tMichel_Minard\t1\nIlan_Ramon\t3\tTatiana_Kennedy_Schlossberg\t1\nIon_Tiriac\t1\tLawrence_MacAulay\t1\nIsmail_Abu_Shanab\t1\tMichael_Denzel\t1\nIsmail_Abu_Shanab\t1\tRebecca_Romijn-Stamos\t1\nJack_Welch\t1\tKeizo_Yamada\t1\nJames_Dingemans\t1\tRobert_Duvall\t5\nJames_Dingemans\t1\tWilliam_Morrow\t1\nJamie_Lee_Curtis\t1\tWilliam_Joppy\t1\nJamie_Martin\t1\tPatrick_Ewing\t1\nJan_Bjoerklund\t1\tLi_Ruihuan\t1\nJan_Paul_Miller\t1\tParadorn_Srichaphan\t8\nJay_Garner\t3\tReggie_Sanders\t1\nJeff_Bzdelik\t1\tZach_Parise\t1\nJerry_Oliver\t1\tRudolf_Schuster\t1\nJerry_Oliver\t1\tStefaan_Declerk\t1\nJerry_Sloan\t1\tJulianne_Moore\t9\nJerry_Sloan\t1\tKeiko_Sofia_Fujimori\t1\nJerry_Sloan\t1\tNikki_Cascone\t1\nJim_Anderson\t1\tJose_Luis_Santiago_Vasconcelos\t1\nJim_Anderson\t1\tMikhail_Wehbe\t1\nJoaquim_Levy\t1\tJudy_Vassar\t1\nJoe_Leonard\t1\tKristen_Rivera\t1\nJohn_Baldacci\t1\tNona_Gaye\t1\nJohn_Baldacci\t1\tShia_LaBeouf\t1\nJohn_Elway\t1\tMahmoud_Diyab_al-Ahmed\t1\nJohn_Kerry\t9\tRaja_Ibrahim\t1\nJohn_Kerry\t14\tPat_Rochester\t1\nJohn_Lisowski\t1\tMichelle_Kwan\t1\nJohn_Malkovich\t2\tParadorn_Srichaphan\t7\nJohn_Malkovich\t3\tMona_Locke\t1\nJohn_Malkovich\t3\tRichard_Palmer\t1\nJohn_McCallum\t1\tOliver_Phelps\t1\nJohn_McCallum\t2\tRick_Bragg\t1\nJohn_McEnroe\t2\tShane_Mosley\t1\nJohn_Prescott\t1\tWill_Ofenheusle\t1\nJohnnie_Lynn\t1\tLarry_Coker\t4\nJohnnie_Lynn\t1\tRichard_Palmer\t1\nJohnny_Unitas\t1\tMark_Foley\t1\nJonathan_Karsh\t1\tSamantha_Ledster\t1\nJoxel_Garcia\t1\tLena_Olin\t1\nJoy_Bryant\t1\tRichard_Chamberlain\t1\nJulianne_Moore\t4\tShia_LaBeouf\t1\nJulio_Iglesias_Jr\t1\tRamon_Ponce_de_Leon\t1\nJustin_Timberlake\t1\tSuzanne_Somers\t1\nKathleen_Glynn\t2\tSusilo_Bambang_Yudhoyono\t1\nKathryn_Morris\t1\tMinnie_Driver\t2\nKathryn_Tucker\t1\tStella_McCartney\t1\nKatie_Couric\t1\tSanjay_Gupta\t1\nKeiko_Sofia_Fujimori\t1\tSureyya_Ayhan\t1\nKellie_Coffey\t1\tTodd_Petit\t1\nKenny_Chesney\t1\tSteve_Coogan\t1\nKevin_Tarrant\t1\tMarisol_Breton\t1\nKristen_Rivera\t1\tValdas_Adamkus\t2\nKurt_Russell\t2\tMahmoud_Diyab_al-Ahmed\t1\nKurt_Tanabe\t1\tWilliam_Jackson\t1\nKyle_McLaren\t1\tSofia_Milos\t1\nKyle_McLaren\t1\tWill_Ofenheusle\t1\nKyle_Shewfelt\t1\tLarry_Coker\t2\nLawrence_Foley\t1\tRachel_Wadsworth\t1\nLea_Fastow\t1\tUday_Hussein\t1\nLesia_Burlak\t1\tNur_Jaafar\t1\nLori_Berenson\t1\tRudolf_Schuster\t1\nLucio_Angulo\t1\tSarah_Price\t1\nLucio_Gutierrez\t10\tPedro_Malan\t5\nMargaret_Hasley\t1\tMichael_Denzel\t1\nMark_Redman\t1\tStephen_Funk\t1\nMark_Salter\t1\tShanna_Zolman\t1\nMartin_Luther_King_III\t1\tMaryn_McKenna\t1\nMary_Bono\t1\tTodd_Wike\t1\nMatt_LeBlanc\t1\tRobert_Redford\t2\nMauro_Viza\t1\tWilliam_Jackson\t1\nMax_Baucus\t1\tParadorn_Srichaphan\t5\nMax_Baucus\t1\tYossi_Beilin\t2\nMeles_Zenawi\t1\tNawabzada_Nasrullah_Khan\t1\nMelissa_Mulloy\t1\tPaula_Abdul\t1\nMelissa_Mulloy\t1\tRoger_Lyons\t1\nMichael_Chiklis\t1\tMohammed_Al_Hindi\t1\nMichael_Chiklis\t1\tSteve_Rush\t1\nMichael_Doleac\t1\tNur_Jaafar\t1\nMichael_Goldrich\t1\tSuzanne_Somers\t1\nMichael_Kahn\t1\tRick_Caruso\t1\nMichel_Minard\t1\tSuzanne_Gaudet\t1\nMichelle_Bachelet\t1\tSami_Al-Arian\t1\nMiguel_Angel_Rodriguez\t1\tSasha_Cohen\t1\nMike_Bryan\t1\tShanna_Zolman\t1\nMike_Montgomery\t1\tRay_Lewis\t1\nMilton_Wynants\t1\tStuart_Townsend\t1\nMiyako_Miyazaki\t2\tMunir_Akram\t2\nMorris_Dees\t1\tShamai_Leibowitz\t1\nMorris_Dees\t1\tSuzie_McConnell_Serio\t1\nNathalie_Gagnon\t1\tRichard_Reid\t1\nNicklas_Lidstrom\t1\tNorman_Jewison\t1\nNicklas_Lidstrom\t1\tSadie_Frost\t3\nNicole_Hiltz\t1\tZaini_Abdullah\t1\nNona_Gaye\t1\tPaul_Cerjan\t1\nOscar_Bolanos\t1\tPhil_Donahue\t1\nOscar_Bolanos\t1\tTatiana_Kennedy_Schlossberg\t1\nPascal_Quignard\t3\tPatrick_Ewing\t2\nPat_Rochester\t1\tPhoenix_Chang\t1\nPat_Rochester\t1\tWill_Ofenheusle\t1\nPaul_Farley\t1\tPlaton_Lebedev\t1\nPaula_Abdul\t1\tRobert_Vowler\t1\nPharrell_Williams\t1\tTyrone_Medley\t1\nPhoenix_Chang\t1\tPlaton_Lebedev\t1\nRachel_Wadsworth\t1\tRichard_Palmer\t1\nRaymond_Odierno\t1\tRichard_Reid\t1\nReggie_Sanders\t1\tRick_Santorum\t2\nRichard_Chamberlain\t1\tSteve_Patterson\t1\nRichard_Ward\t1\tSteve_Redgrave\t1\nRobert_Vowler\t1\tTab_Baldwin\t1\nRoy_Rogers\t1\tSteven_Feldman\t1\nScott_Rolen\t1\tWilliam_Murabito\t1\nSofia_Milos\t1\tSteve_Nash\t3\nSofia_Milos\t1\tWill_Ofenheusle\t1\nSonja_Kesselschlager\t1\tTim_Robbins\t3\nTakeo_Hiranuma\t1\tTy_Votaw\t1\nTed_Washington\t1\tXimena_Bohorquez\t1\nTy_Votaw\t1\tWilliam_Webster\t1\nAdrian_McPherson\t1\t2\nAl_Davis\t1\t2\nAl_Gore\t2\t6\nAl_Gore\t4\t6\nAlan_Greenspan\t1\t2\nAlan_Greenspan\t1\t5\nAlan_Greenspan\t3\t4\nAlastair_Campbell\t1\t5\nAlastair_Campbell\t2\t4\nAlexander_Downer\t1\t2\nAlexander_Downer\t1\t3\nAlexander_Downer\t1\t4\nAlexander_Downer\t3\t4\nAlice_Fisher\t1\t2\nAlison_Lohman\t1\t2\nAlvaro_Silva_Calderon\t1\t2\nAlvaro_Silva_Calderon\t1\t3\nAlvaro_Uribe\t7\t20\nAlvaro_Uribe\t8\t25\nAlvaro_Uribe\t12\t13\nAlvaro_Uribe\t20\t28\nAnders_Ebbeson\t2\t3\nAndrew_Bunner\t1\t2\nAnibal_Ibarra\t1\t3\nAntonio_Trillanes\t1\t3\nAntonio_Trillanes\t2\t3\nAsa_Hutchinson\t1\t2\nBarbara_Walters\t1\t3\nBarbara_Walters\t1\t4\nBarbara_Walters\t3\t4\nBen_Howland\t1\t2\nBen_Howland\t1\t3\nBen_Howland\t3\t4\nBenazir_Bhutto\t1\t4\nBenazir_Bhutto\t2\t3\nBenazir_Bhutto\t2\t4\nBill_Simon\t5\t9\nBill_Simon\t11\t15\nBilly_Sollie\t1\t2\nBoris_Berezovsky\t1\t2\nBrooke_Shields\t1\t2\nBulent_Ecevit\t1\t4\nBulent_Ecevit\t1\t5\nBulent_Ecevit\t2\t6\nBulent_Ecevit\t4\t6\nCandie_Kung\t1\t2\nCandie_Kung\t1\t4\nCarlo_Ancelotti\t1\t2\nCarlo_Ancelotti\t2\t3\nCarlos_Moya\t8\t17\nCarlos_Moya\t10\t18\nCarlos_Vives\t1\t4\nCarlos_Vives\t2\t4\nCarson_Daly\t1\t2\nCate_Blanchett\t1\t2\nCate_Blanchett\t1\t3\nCate_Blanchett\t1\t4\nCate_Blanchett\t2\t3\nCate_Blanchett\t2\t4\nChok_Tong_Goh\t1\t2\nChris_Byrd\t1\t2\nChris_Cooper\t1\t2\nChris_Tucker\t1\t2\nChristine_Gregoire\t1\t4\nChristine_Gregoire\t2\t3\nChristine_Gregoire\t2\t4\nChristopher_Patten\t1\t2\nClint_Eastwood\t1\t4\nClint_Eastwood\t1\t6\nConstance_Marie\t1\t2\nConstance_Marie\t1\t3\nDennis_Hastert\t2\t3\nDennis_Hastert\t3\t6\nDennis_Hastert\t4\t6\nDennis_Hastert\t5\t6\nDolly_Parton\t1\t2\nDoug_Duncan\t1\t2\nEdward_Kennedy\t1\t2\nEdward_Kennedy\t2\t3\nEdward_Said\t1\t2\nElena_Bovina\t1\t2\nElena_Bovina\t1\t3\nElena_Bovina\t2\t3\nEliane_Karp\t1\t3\nEliane_Karp\t2\t3\nEliane_Karp\t2\t4\nEliane_Karp\t3\t4\nElvis_Presley\t1\t2\nErika_Harold\t1\t3\nErika_Harold\t2\t3\nErika_Harold\t3\t4\nErika_Harold\t4\t5\nErnie_Els\t1\t3\nErnie_Els\t1\t4\nErnie_Els\t2\t3\nFranco_Dragone\t1\t2\nFrank_Solich\t1\t4\nFrank_Solich\t2\t5\nFrank_Solich\t3\t4\nGabriel_Batistuta\t1\t2\nGary_Carter\t1\t2\nGary_Carter\t1\t3\nGary_Doer\t1\t2\nGary_Doer\t2\t3\nGeorge_Tenet\t1\t2\nGeorge_Voinovich\t1\t3\nGeorge_Voinovich\t2\t3\nGeorgi_Parvanov\t1\t2\nGoldie_Hawn\t1\t7\nGoldie_Hawn\t2\t3\nGoldie_Hawn\t3\t7\nGoldie_Hawn\t6\t7\nGoran_Persson\t1\t2\nGro_Harlem_Brundtland\t1\t2\nGuillaume_Soro\t1\t2\nGwyneth_Paltrow\t1\t5\nGwyneth_Paltrow\t1\t6\nGwyneth_Paltrow\t2\t6\nGwyneth_Paltrow\t3\t6\nHee-Won_Han\t1\t2\nHerb_Sendek\t1\t3\nHerb_Sendek\t3\t4\nHoward_Smith\t1\t2\nHugh_Grant\t5\t8\nHugh_Grant\t6\t9\nJack_Straw\t25\t28\nJames_Franco\t1\t2\nJames_Ivory\t1\t2\nJames_Schultz\t1\t2\nJames_Traficant\t1\t2\nJames_Traficant\t2\t3\nJan_Ullrich\t1\t4\nJan_Ullrich\t2\t6\nJan_Ullrich\t3\t6\nJane_Pauley\t1\t2\nJason_Jennings\t1\t2\nJavier_Weber\t1\t2\nJennifer_Rodriguez\t1\t2\nJeong_Se-hyun\t2\t6\nJeong_Se-hyun\t3\t7\nJeong_Se-hyun\t6\t8\nJeong_Se-hyun\t7\t9\nJeong_Se-hyun\t8\t9\nJo_Dee_Messina\t1\t2\nJoe_Lieberman\t8\t11\nJoe_Lieberman\t9\t10\nJoe_Mantello\t1\t2\nJohn_Allen_Muhammad\t2\t9\nJohn_Allen_Muhammad\t5\t7\nJohn_Allen_Muhammad\t6\t10\nJohn_Blaney\t1\t2\nJohn_Brady\t1\t2\nJohn_Howard\t5\t15\nJohn_Howard\t12\t17\nJohnny_Tapia\t1\t2\nJohnny_Tapia\t2\t3\nJorge_Valdano\t1\t2\nJoseph_Deiss\t1\t3\nJunichiro_Koizumi\t1\t53\nJunichiro_Koizumi\t26\t55\nJunichiro_Koizumi\t29\t45\nKate_Capshaw\t1\t2\nKathryn_Bigelow\t1\t2\nKevin_Spacey\t1\t2\nKevin_Spacey\t1\t3\nKim_Ryong-sung\t6\t8\nKim_Ryong-sung\t8\t11\nKlaus_Zwickel\t1\t2\nKristen_Breitweiser\t2\t3\nLaila_Ali\t2\t3\nLarry_Brown\t1\t3\nLarry_Brown\t2\t4\nLarry_Brown\t2\t7\nLarry_Brown\t4\t7\nLarry_Brown\t6\t7\nLarry_Johnson\t1\t2\nLars_Von_Trier\t1\t2\nLars_Von_Trier\t2\t3\nLeander_Paes\t1\t2\nLiam_Neeson\t1\t3\nLiam_Neeson\t2\t3\nLino_Oviedo\t1\t3\nLino_Oviedo\t2\t3\nLudivine_Sagnier\t1\t3\nLudivine_Sagnier\t3\t4\nLynne_Cheney\t1\t2\nLynne_Cheney\t2\t3\nMaggie_Smith\t1\t2\nMarcelo_Salas\t1\t2\nMariangel_Ruiz_Torrealba\t1\t2\nMarisa_Tomei\t1\t2\nMary_Steenburgen\t1\t2\nMel_Brooks\t1\t2\nMel_Gibson\t1\t2\nMian_Khursheed_Mehmood_Kasuri\t2\t3\nMian_Khursheed_Mehmood_Kasuri\t3\t4\nMichael_Moore\t1\t2\nMichael_Moore\t2\t3\nMichel_Duclos\t1\t2\nMichelle_Pfeiffer\t1\t2\nMireya_Moscoso\t2\t3\nMireya_Moscoso\t2\t5\nMireya_Moscoso\t4\t5\nNabil_Shaath\t1\t3\nNabil_Shaath\t2\t3\nNaomi_Watts\t1\t18\nNaomi_Watts\t6\t9\nNaomi_Watts\t13\t18\nNatalie_Cole\t1\t3\nNatalie_Cole\t2\t3\nNathalie_Baye\t1\t2\nNathalie_Baye\t1\t4\nNathalie_Baye\t2\t4\nNathalie_Baye\t3\t4\nOleksandr_Moroz\t1\t2\nOscar_Elias_Biscet\t1\t2\nOswaldo_Paya\t1\t3\nOswaldo_Paya\t3\t4\nPatricia_Clarkson\t1\t2\nPatricia_Clarkson\t1\t3\nPatricia_Clarkson\t2\t3\nPatricia_Clarkson\t2\t4\nPatrick_Roy\t1\t2\nPaul-Henri_Mathieu\t1\t2\nPaul-Henri_Mathieu\t1\t3\nPaul_Byrd\t1\t2\nPaul_Kagame\t1\t2\nPeter_Costello\t1\t2\nPeter_Greenaway\t1\t2\nPrince_Naruhito\t1\t2\nPrince_Naruhito\t1\t3\nPrince_Naruhito\t2\t3\nPrincess_Masako\t1\t2\nPupi_Avati\t2\t3\nQueen_Latifah\t1\t3\nQueen_Latifah\t2\t4\nQueen_Latifah\t3\t4\nRachel_Griffiths\t2\t3\nRalph_Firman\t1\t2\nRalph_Klein\t1\t2\nRanil_Wickremasinghe\t2\t3\nRick_Pitino\t1\t3\nRick_Pitino\t2\t4\nRick_Pitino\t3\t4\nRicky_Martin\t1\t2\nRita_Moreno\t1\t2\nRobert_De_Niro\t1\t4\nRobert_De_Niro\t3\t6\nRobert_Kocharian\t4\t5\nRoberto_Marinho\t2\t3\nRoger_Moore\t3\t5\nRon_Dittemore\t1\t2\nRon_Dittemore\t1\t3\nRon_Dittemore\t4\t6\nRudolph_Giuliani\t1\t20\nRudolph_Giuliani\t2\t5\nRudolph_Giuliani\t3\t20\nRudolph_Giuliani\t4\t17\nRudolph_Giuliani\t15\t20\nRussell_Simmons\t1\t2\nRussell_Simmons\t1\t4\nRussell_Simmons\t2\t4\nSean_Astin\t1\t3\nSean_Astin\t2\t3\nShaukat_Aziz\t1\t2\nSilvio_Fernandez\t1\t2\nSophia_Loren\t1\t2\nSophia_Loren\t1\t3\nSophia_Loren\t1\t7\nSophia_Loren\t6\t7\nStellan_Skarsgard\t1\t2\nTamara_Brooks\t1\t2\nThomas_Fargo\t1\t2\nThomas_Fargo\t1\t3\nThomas_Fargo\t2\t3\nThomas_Fargo\t2\t4\nThomas_Fargo\t3\t4\nTim_Allen\t2\t3\nTim_Allen\t3\t4\nTony_Shalhoub\t1\t2\nTony_Shalhoub\t1\t3\nTracy_McGrady\t1\t2\nVicente_Fernandez\t2\t3\nVicente_Fernandez\t4\t5\nVince_Gill\t1\t2\nWolfgang_Schuessel\t1\t4\nWolfgang_Schuessel\t3\t4\nWu_Yi\t1\t2\nWu_Yi\t1\t3\nWu_Yi\t2\t3\nYao_Ming\t2\t4\nYao_Ming\t5\t6\nYao_Ming\t5\t8\nYao_Ming\t6\t7\nYoriko_Kawaguchi\t3\t5\nYu_Shyi-kun\t1\t3\nYu_Shyi-kun\t1\t4\nYu_Shyi-kun\t2\t3\nZhang_Wenkang\t1\t2\nZinedine_Zidane\t4\t6\nAbdullah_Nasseef\t1\tBruce_Paltrow\t1\nAbdullah_Nasseef\t1\tHoward_Smith\t2\nAbdullah_Nasseef\t1\tJan_De_Bont\t1\nAbdullah_Nasseef\t1\tJim_Nochols\t1\nAbdullah_Nasseef\t1\tOleg_Romantsev\t1\nAdam_Kennedy\t1\tCharlie_Sheen\t1\nAdrian_McPherson\t2\tEduardo_Chillida\t1\nAl_Davis\t1\tJulian_Fantino\t1\nAlan_Greenspan\t4\tKent_McCord\t1\nAlastair_Campbell\t4\tLi_Ka-shing\t1\nAlexa_Vega\t1\tTony_Shalhoub\t1\nAlexander_Downer\t2\tZeljko_Rebraca\t1\nAlexander_Downer\t3\tLuis_Berrondo\t1\nAlexandra_Pelosi\t1\tKen_Kutaragi\t1\nAlexandre_Herchcovitch\t1\tKaren_Clarkson\t1\nAlice_Fisher\t1\tGene_Sauers\t1\nAlicia_Witt\t1\tJorge_Moreno\t1\nAlicia_Witt\t1\tLino_Oviedo\t1\nAllan_Wagner\t1\tLarry_Campbell\t1\nAllan_Wagner\t1\tRita_Moreno\t1\nAlvaro_Silva_Calderon\t2\tStepan_Demirchian\t1\nAlvaro_Uribe\t14\tPatricia_Clarkson\t1\nAna_Sebastiao\t1\tMarcos_Cafu\t1\nAna_Sebastiao\t1\tRaza_Rabbani\t1\nAndrea_De_Cruz\t1\tDionne_Warwick\t1\nAndrea_De_Cruz\t1\tIsmail_Cem\t1\nAndrea_De_Cruz\t1\tNatalya_Sazanovich\t1\nAndrea_De_Cruz\t1\tYoon_Young-kwan\t1\nAndres_Manuel_Lopez_Obrador\t1\tJim_Parque\t1\nAndres_Manuel_Lopez_Obrador\t1\tNorman_Mineta\t1\nAndrew_Bunner\t2\tOscar_Elias_Biscet\t2\nAndrew_Weissmann\t1\tErnie_Els\t2\nAndy_Warhol\t1\tEdith_Masai\t1\nAndy_Warhol\t1\tPaul-Henri_Mathieu\t2\nAnibal_Ibarra\t2\tDavid_Przybyszewski\t1\nAnthony_Ervin\t1\tJuan_Roman_Carrasco\t1\nAntonio_Elias_Saca\t1\tPeter_Holmberg\t1\nAntonio_Elias_Saca\t1\tThomas_Fargo\t4\nAntonio_Trillanes\t1\tKatrin_Cartlidge\t1\nAntonio_Trillanes\t3\tTerry_Gilliam\t1\nAsa_Hutchinson\t2\tGary_Sinise\t1\nAscencion_Barajas\t1\tChris_Hernandez\t1\nAshley_Judd\t1\tJoan_Jett\t1\nAshley_Postell\t1\tCassandra_Heise\t1\nAshley_Postell\t1\tTony_Clement\t1\nAshraf_Ghani\t1\tBoris_Berezovsky\t1\nAshraf_Ghani\t1\tQian_Qichen\t1\nAsif_Ali_Zardari\t1\tSteny_Hoyer\t1\nAssad_Ahmadi\t1\tPercy_Gibson\t1\nBarbara_Walters\t1\tJewel_Howard-Taylor\t1\nBarbara_Walters\t1\tPeter_Greenaway\t1\nBarry_Williams\t1\tWang_Fei\t1\nBenazir_Bhutto\t4\tJane_Pauley\t2\nBenazir_Bhutto\t5\tKaren_Clarkson\t1\nBijan_Darvish\t1\tFranco_Dragone\t1\nBijan_Darvish\t2\tFrank_Taylor\t1\nBilal_Erdogan\t1\tLubomir_Zaoralek\t1\nBill_Self\t1\tYang_Jianli\t1\nBill_Simon\t1\tEdward_Said\t2\nBilly_Gilman\t1\tHoward_Smith\t2\nBilly_Sollie\t1\tMark_Swartz\t1\nBilly_Sollie\t2\tHussein_Malik\t1\nBlas_Ople\t1\tEvan_Marriott\t1\nBlas_Ople\t1\tFelipe_Fernandez\t1\nBlas_Ople\t1\tPeter_Costello\t1\nBob_Cantrell\t1\tGerard_Kleisterlee\t1\nBob_Cantrell\t1\tMary_Landrieu\t3\nBob_Goldman\t1\tLeigh_Winchell\t1\nBob_Goldman\t1\tMaggie_Smith\t1\nBoris_Berezovsky\t2\tLeander_Paes\t1\nBrad_Gushue\t1\tJohannes_Rau\t1\nBrad_Miller\t1\tIrina_Lobacheva\t1\nBrad_Miller\t1\tPercy_Gibson\t1\nBrenda_van_Dam\t1\tNorbert_van_Heyst\t1\nBrian_Cashman\t1\tDelphine_Chuillot\t1\nBrian_Cashman\t1\tRussell_Simmons\t3\nBrian_Cook\t1\tMatt_Welsh\t1\nBruno_Junquiera\t1\tOscar_DLeon\t1\nBulent_Ecevit\t6\tEduardo_Romero\t1\nBulent_Ecevit\t6\tGilberto_Simoni\t1\nCandie_Kung\t1\tSteve_Allan\t1\nCarlo_Ancelotti\t1\tJohn_Tyson\t1\nCarlos_Vives\t2\tRandy_Johnson\t1\nCaroline_Dhavernas\t1\tEvgeni_Plushenko\t1\nCaroline_Dhavernas\t1\tJeremy_Fogel\t1\nCarson_Daly\t1\tJohn_Moe\t1\nCarson_Daly\t1\tTracy_Wyle\t1\nCassandra_Heise\t1\tShaukat_Aziz\t2\nCate_Blanchett\t3\tGina_Centrello\t1\nCate_Blanchett\t3\tRyan_Goodman\t1\nCatherine_Donkers\t1\tEminem\t1\nCedric_Benson\t1\tMarcus_Allen\t1\nCedric_Benson\t1\tRobert_Lee_Yates_Jr\t1\nCedric_Benson\t1\tStacy_Nelson\t1\nChistian_Stahl\t1\tEliane_Karp\t1\nChris_Cooper\t2\tIrfan_Ahmed\t1\nChris_Cooper\t2\tRick_Bland\t1\nChris_Hernandez\t1\tMauricio_Pochetino\t1\nChris_Hernandez\t1\tTonga\t1\nChris_Hernandez\t1\tWolfgang_Schuessel\t1\nChris_Tucker\t2\tJennifer_Pena\t1\nChris_Tucker\t2\tTakuma_Sato\t1\nChris_Whitney\t1\tTom_Scully\t1\nChristine_Gregoire\t4\tRobert_Lee_Yates_Jr\t1\nChristine_Gregoire\t4\tWang_Hailan\t1\nChristine_Rau\t1\tJerry_Hall\t1\nChristine_Rau\t1\tMel_Brooks\t1\nChristopher_Russell\t1\tColleen_OClair\t1\nChristopher_Whittle\t1\tFelipe_Fernandez\t1\nColleen_OClair\t1\tTom_Hanusik\t1\nCurtis_Rodriguez\t1\tJohn_Blaney\t2\nCurtis_Rodriguez\t1\tKathleen_Kennedy_Townsend\t3\nDale_Bosworth\t1\tFrank_Taylor\t1\nDale_Bosworth\t1\tVladimir_Golovlyov\t1\nDan_Quayle\t1\tJim_Ryan\t1\nDan_Quayle\t1\tNick_Turner\t1\nDan_Quayle\t1\tWilliam_Umbach\t1\nDanny_Morgan\t1\tJeff_George\t1\nDanny_Morgan\t1\tJohn_Allen_Muhammad\t10\nDarla_Moore\t1\tPaul_Johnson\t1\nDarrell_Royal\t1\tLouis_Van_Gaal\t1\nDave_McGinnis\t1\tMichel_Therrien\t1\nDavid_Przybyszewski\t1\tElena_Bovina\t2\nDelphine_Chuillot\t1\tJim_Parque\t1\nDemetrin_Veal\t1\tRobert_Hanssen\t1\nDennis_Hastert\t5\tLubomir_Zaoralek\t1\nDewayne_White\t1\tJoe_DeLamielleure\t1\nDick_Posthumus\t1\tJeff_George\t1\nDick_Posthumus\t1\tJim_Wong\t1\nDin_Samsudin\t1\tPupi_Avati\t2\nDionne_Warwick\t1\tJanine_Pietsch\t1\nDominique_Perben\t1\tOsmond_Smith\t1\nDoug_Duncan\t1\tKevin_Spacey\t2\nDoug_Duncan\t2\tJohn_Allen_Muhammad\t2\nEddy_Hartenstein\t1\tLyudmila_Putin\t1\nEddy_Hartenstein\t1\tMatthias_Sammer\t1\nEddy_Hartenstein\t1\tSabah_Al-Ahmad_Al-Jaber_Al-Sabah\t1\nEduardo_Chillida\t1\tStephen_Crampton\t1\nEduardo_Romero\t1\tRobert_Kocharian\t5\nEdward_Greenspan\t1\tRoy_Halladay\t1\nEdward_Greenspan\t1\tSidney_Poitier\t1\nEdward_Kennedy\t1\tMathilda_Karel_Spak\t1\nEdward_Kennedy\t2\tJennifer_Pena\t1\nEdward_Said\t1\tFrank_Abagnale_Jr\t1\nEminem\t1\tWolfgang_Clement\t1\nEnrique_Bolanos\t4\tJanine_Pietsch\t1\nEnrique_Bolanos\t4\tLiam_Neeson\t2\nErnie_Els\t2\tMarcus_Allen\t1\nEvgeni_Plushenko\t1\tQueen_Sofia\t1\nEvgeni_Plushenko\t1\tRiek_Blanjaar\t1\nFranco_Dragone\t2\tLouis_Van_Gaal\t1\nFranco_Frattini\t1\tIrina_Lobacheva\t1\nFrank_Abagnale_Jr\t1\tNate_Hybl\t1\nFrank_Shea\t1\tRob_Moore\t1\nFrank_Taylor\t1\tJason_Jennings\t1\nGabriel_Batistuta\t1\tPeter_Costello\t1\nGabriel_Jorge_Ferreia\t1\tZulfiqar_Ahmed\t1\nGary_Doer\t3\tWang_Hailan\t1\nGary_Sinise\t1\tRoy_Halladay\t1\nGene_Hackman\t1\tQueen_Noor\t1\nGene_Sauers\t1\tGina_Centrello\t1\nGeoff_Dixon\t1\tJean_Todt\t1\nGeorge_Allen\t1\tRoy_Halladay\t1\nGeorge_Allen\t1\tWan_Yanhai\t1\nGeorge_Harrison\t1\tRobert_Lee_Yates_Jr\t1\nGeorgi_Parvanov\t1\tLuis_Fonsi\t1\nGianni_Agnelli\t1\tMarco_Irizarry\t1\nGilberto_Simoni\t1\tJulio_Cesar_Chavez\t1\nGong_Ruina\t1\tHong_Myung\t1\nGong_Ruina\t1\tQian_Qichen\t1\nGro_Harlem_Brundtland\t2\tTurner_Gill\t1\nHector_Grullon\t1\tJeong_Se-hyun\t9\nHenk_Bekedam\t1\tKoichi_Haraguchi\t1\nHenk_Bekedam\t1\tManuel_Gehring\t1\nHenk_Bekedam\t1\tMichael_Kirby\t1\nHiroki_Gomi\t1\tKenneth_Cooper\t1\nHiroki_Gomi\t1\tSophia_Loren\t2\nIrina_Lobacheva\t1\tSteve_Allan\t1\nIsmail_Cem\t1\tLinda_Amicangioli\t1\nJack_Osbourne\t1\tKent_McCord\t1\nJack_Straw\t1\tManuel_Gehring\t1\nJack_Straw\t3\tNorman_Mineta\t1\nJames_Ballenger\t1\tRaza_Rabbani\t1\nJames_Coviello\t1\tJudith_Nathan\t1\nJames_Coviello\t1\tRobert_F_Kennedy_Jr\t1\nJames_Morris\t2\tSilvio_Fernandez\t2\nJames_Schultz\t1\tTim_Curley\t1\nJames_W_Kennedy\t1\tJohn_Connolly\t1\nJames_W_Kennedy\t1\tToutai_Kefu\t1\nJames_Williams\t1\tPhillip_Seymor_Hoffmann\t1\nJames_Williams\t1\tShaun_Pollock\t1\nJan_Ullrich\t3\tJavier_Weber\t1\nJane_Menelaus\t1\tJunichiro_Koizumi\t42\nJane_Menelaus\t1\tSophia_Loren\t1\nJason_Petty\t1\tMayumi_Moriyama\t1\nJean_Todt\t1\tRoger_Etchegaray\t1\nJennifer_Pena\t1\tSonia_Lopez\t1\nJennifer_Pena\t1\tTony_LaRussa\t1\nJennifer_Rodriguez\t1\tTerry_Gilliam\t1\nJennifer_Rodriguez\t2\tJoanna_Poitier\t1\nJennifer_Rodriguez\t2\tTony_Fernandes\t1\nJerome_Golmard\t1\tKristin_Chenoweth\t1\nJewel_Howard-Taylor\t1\tKen_Kutaragi\t1\nJim_Parque\t1\tMichael_Adams\t1\nJim_Parque\t1\tRachel_Leigh_Cook\t1\nJim_Ryan\t1\tLuis_Berrondo\t1\nJimmy_Gobble\t1\tRobert_De_Niro\t5\nJoan_Collins\t1\tZydrunas_Ilgauskas\t1\nJoan_Jett\t1\tKevin_James\t1\nJoanna_Poitier\t1\tMichael_Moore\t1\nJoe_Lieberman\t10\tJoe_Mantello\t1\nJohannes_Rau\t1\tOleksandr_Moroz\t1\nJohn_Connolly\t1\tMel_Gibson\t1\nJohn_Connolly\t1\tMichael_Moore\t3\nJohn_Connolly\t1\tNorman_Mineta\t1\nJohn_Connolly\t1\tUzi_Even\t1\nJohn_Cruz\t1\tSean_Astin\t3\nJohn_Cruz\t1\tSteven_Briggs\t1\nJohn_Hartson\t1\tOscar_Elias_Biscet\t2\nJoseph_Deiss\t1\tTony_Fernandes\t1\nJoseph_Lopez\t1\tRick_Pitino\t1\nJulian_Fantino\t1\tQuin_Snyder\t1\nJulio_Cesar_Chavez\t1\tTom_Brennan\t1\nJunichi_Inamoto\t1\tOleg_Romantsev\t1\nJunichiro_Koizumi\t20\tTony_Fernandes\t1\nKaren_Clarkson\t1\tKristen_Breitweiser\t3\nKarol_Kucera\t1\tQian_Qichen\t1\nKarol_Kucera\t1\tTurner_Gill\t1\nKatrin_Cartlidge\t1\tTony_LaRussa\t1\nKeith_Osik\t1\tPupi_Avati\t1\nKenneth_Cooper\t1\tNorman_Mineta\t1\nKent_McCord\t1\tNatasha_Henstridge\t1\nKevin_James\t1\tRoy_Halladay\t1\nKevin_James\t1\tSuzanne_Haik_Terrell\t1\nKevin_James\t1\tYang_Jianli\t1\nKevin_Spacey\t2\tRay_Evernham\t1\nKoji_Uehara\t1\tRob_Moore\t1\nKrishna_Bhadur_Mahara\t1\tPeter_Holmberg\t1\nKrishna_Bhadur_Mahara\t1\tSilvio_Fernandez\t2\nKristen_Breitweiser\t1\tTatiana_Shchegoleva\t1\nLaila_Ali\t3\tTony_Fernandes\t1\nLarry_Johnson\t1\tTerry_Gilliam\t1\nLarry_Johnson\t2\tMarc_Racicot\t1\nLarry_Johnson\t2\tTom_Hanusik\t1\nLars_Von_Trier\t1\tZhang_Wenkang\t2\nLeander_Paes\t1\tRalph_Firman\t1\nLeo_Ramirez\t1\tSabah_Al-Ahmad_Al-Jaber_Al-Sabah\t1\nLeonardo_Del_Vecchio\t1\tMaggie_Smith\t1\nLeonardo_Del_Vecchio\t1\tTatiana_Shchegoleva\t1\nLiam_Neeson\t1\tPaul_Kagame\t2\nLokendra_Bahadur_Chand\t1\tSteffeny_Holtz\t1\nLouis_Van_Gaal\t1\tNatasha_Henstridge\t1\nLubomir_Zaoralek\t1\tLuis_Berrondo\t1\nLubomir_Zaoralek\t1\tWilliam_Umbach\t1\nLuis_Berrondo\t1\tRita_Moreno\t2\nLuis_Fonsi\t1\tOscar_DLeon\t1\nLuis_Fonsi\t1\tSteven_Briggs\t1\nLuis_Rosario_Huertas\t1\tMary_Steenburgen\t2\nMakiko_Tanaka\t1\tPeter_Greenaway\t1\nManijeh_Hekmat\t1\tRoger_Winter\t1\nManijeh_Hekmat\t1\tT_Boone_Pickens\t1\nMarcelo_Salas\t2\tTara_VanDerveer\t1\nMarco_Irizarry\t1\tYang_Jianli\t1\nMariangel_Ruiz_Torrealba\t3\tPupi_Avati\t1\nMarisa_Tomei\t2\tPeter_Costello\t2\nMark_Polansky\t1\tMiranda_Gaddis\t1\nMary_Landrieu\t3\tPercy_Gibson\t1\nMathilda_Karel_Spak\t1\tMauricio_Macri\t1\nMathilda_Karel_Spak\t1\tTamara_Brooks\t2\nMatt_Welsh\t1\tRandy_Johnson\t1\nMel_Brooks\t2\tShoshana_Johnson\t1\nMichael_Kirby\t1\tRanil_Wickremasinghe\t2\nMichael_Sheehan\t1\tRobert_Hanssen\t1\nMichel_Duclos\t1\tQuin_Snyder\t1\nMichel_Duclos\t1\tSuh_Young-hoon\t1\nMike_Gable\t1\tMohamed_Hammam\t1\nMikhail_Khodorkovsky\t1\tRita_Moreno\t2\nMireya_Moscoso\t4\tSan_Lan\t1\nMohamed_Hammam\t1\tRoger_Moore\t4\nNabil_Shaath\t3\tSilvie_Cabero\t1\nNaomi_Hayashi\t1\tPrakash_Hinduja\t1\nNatalya_Sazanovich\t1\tPupi_Avati\t4\nNikki_Teasley\t1\tRyan_Goodman\t1\nOleksandr_Moroz\t1\tRobert_De_Niro\t1\nOscar_DLeon\t1\tSteven_Curtis_Chapman\t1\nPaul_Byrd\t1\tWolfgang_Schwarz\t1\nPaul_Wollnough\t1\tPhilip_Cummings\t1\nPaul_Wollnough\t1\tYuvraj_Singh\t1\nPenelope_Taylor\t1\tWang_Fei\t1\nPeter_Caruana\t1\tPhilip_Cummings\t1\nPhillip_Seymor_Hoffmann\t1\tSilvio_Fernandez\t1\nPhillips_Idowu\t1\tPrince_Naruhito\t2\nPrincess_Masako\t2\tThomas_Wilkens\t1\nQuin_Snyder\t1\tVagit_Alekperov\t1\nRanil_Wickremasinghe\t3\tStacy_Nelson\t1\nReina_Hayes\t1\tSteffeny_Holtz\t1\nRick_Pitino\t3\tTony_Clement\t1\nRicky_Martin\t2\tRoger_Etchegaray\t1\nRobert_F_Kennedy_Jr\t1\tRon_Dittemore\t1\nSidney_Poitier\t1\tSvend_Aage_Jensby\t1\nAbdel_Nasser_Assidi\t1\t2\nAi_Sugiyama\t1\t2\nAi_Sugiyama\t1\t4\nAi_Sugiyama\t4\t5\nAldo_Paredes\t1\t2\nAlejandro_Avila\t1\t2\nAlejandro_Avila\t1\t3\nAlex_Sink\t2\t3\nAllen_Iverson\t1\t2\nAmram_Mitzna\t1\t2\nAndrew_Niccol\t1\t2\nAndy_Hebb\t1\t2\nAnne_Krueger\t1\t2\nAnne_McLellan\t1\t3\nAnne_McLellan\t2\t3\nAnnette_Bening\t1\t2\nAnthony_Hopkins\t1\t2\nAriel_Sharon\t16\t45\nArminio_Fraga\t1\t6\nArminio_Fraga\t2\t4\nArminio_Fraga\t3\t6\nArminio_Fraga\t4\t6\nArt_Hoffmann\t1\t2\nAshanti\t1\t3\nAshanti\t1\t4\nAshanti\t2\t5\nAshanti\t3\t5\nAshanti\t4\t5\nAugustin_Calleri\t1\t2\nAugustin_Calleri\t2\t4\nAugustin_Calleri\t3\t4\nBertie_Ahern\t1\t2\nBertie_Ahern\t1\t4\nBertie_Ahern\t1\t5\nBertie_Ahern\t2\t3\nBertie_Ahern\t3\t5\nBill_Clinton\t2\t4\nBill_Clinton\t6\t8\nBill_Clinton\t6\t10\nBill_Clinton\t8\t10\nBill_Clinton\t9\t12\nBill_Clinton\t10\t29\nBill_Clinton\t18\t28\nBill_McBride\t4\t7\nBill_McBride\t4\t10\nBill_Parcells\t1\t2\nBinyamin_Ben-Eliezer\t3\t5\nBinyamin_Ben-Eliezer\t5\t6\nBinyamin_Ben-Eliezer\t6\t7\nBrendan_Hansen\t1\t2\nBrian_Wells\t1\t2\nCarlos_Quintanilla_Schmidt\t1\t2\nCarolina_Moraes\t1\t2\nCecilia_Bolocco\t1\t2\nCecilia_Bolocco\t1\t3\nCecilia_Bolocco\t2\t3\nCesar_Gaviria\t2\t4\nCesar_Gaviria\t2\t5\nCesar_Gaviria\t3\t6\nCesar_Gaviria\t3\t7\nCesar_Gaviria\t4\t6\nCharlie_Zaa\t1\t2\nChita_Rivera\t1\t2\nChristian_Longo\t1\t2\nChristian_Longo\t1\t3\nChristian_Wulff\t1\t2\nColin_Jackson\t1\t2\nDarrell_Issa\t1\t2\nDarrell_Porter\t1\t2\nDave_Campo\t1\t3\nDavid_Dodge\t1\t2\nDavid_Heyman\t1\t2\nDavid_Spade\t1\t2\nDavid_Wells\t1\t5\nDavid_Wells\t1\t6\nDavid_Wells\t1\t7\nDavid_Wells\t2\t6\nDavid_Wells\t5\t6\nDebbie_Reynolds\t1\t3\nDebbie_Reynolds\t2\t4\nDexter_Jackson\t1\t2\nDiana_Taurasi\t1\t2\nDonald_Evans\t1\t2\nDuane_Lee_Chapman\t1\t2\nEd_Smart\t1\t3\nEileen_Coparropa\t1\t2\nEileen_Coparropa\t1\t3\nEileen_Coparropa\t2\t3\nElizabeth_Dole\t1\t3\nElizabeth_Dole\t2\t4\nFred_Funk\t1\t2\nGabriel_Valdes\t1\t2\nGary_Winnick\t1\t2\nGeno_Auriemma\t1\t2\nGeorge_Karl\t1\t2\nGeorge_Robertson\t2\t3\nGeorge_Robertson\t3\t19\nGeorge_Robertson\t7\t19\nGeorge_Robertson\t11\t12\nGeorge_Ryan\t1\t4\nGeorge_Ryan\t2\t3\nGeorge_W_Bush\t7\t65\nGeorge_W_Bush\t16\t482\nGeorge_W_Bush\t145\t150\nGeorge_W_Bush\t145\t238\nGeorge_W_Bush\t203\t247\nGloria_Trevi\t1\t2\nGloria_Trevi\t2\t3\nGloria_Trevi\t2\t4\nGordon_Campbell\t1\t2\nGreg_Rusedski\t1\t2\nGreg_Rusedski\t2\t3\nGreg_Rusedski\t2\t4\nGreg_Rusedski\t3\t4\nGustavo_Kuerten\t1\t2\nGustavo_Kuerten\t2\t3\nGuy_Hemmings\t1\t2\nHalle_Berry\t6\t7\nHalle_Berry\t6\t9\nHalle_Berry\t7\t9\nHeather_Mills\t2\t3\nHeather_Mills\t2\t4\nHeidi_Fleiss\t1\t2\nHeidi_Fleiss\t1\t3\nHenrique_Meirelles\t1\t2\nIva_Majoli\t1\t2\nJake_Gyllenhaal\t3\t5\nJake_Gyllenhaal\t4\t5\nJames_Cunningham\t1\t2\nJames_Cunningham\t1\t3\nJames_Cunningham\t2\t3\nJames_Parker\t1\t2\nJames_Wolfensohn\t2\t5\nJason_Lezak\t1\t2\nJay_Rasulo\t1\t2\nJean-Claude_Juncker\t1\t2\nJennifer_Lopez\t1\t3\nJennifer_Lopez\t2\t16\nJennifer_Lopez\t3\t10\nJennifer_Lopez\t3\t14\nJesse_Jackson\t1\t3\nJesse_Jackson\t5\t8\nJesse_Jackson\t7\t8\nJesse_James_Leija\t1\t2\nJesse_Ventura\t2\t3\nJia_Qinglin\t1\t2\nJim_Tressel\t1\t2\nJim_Tressel\t1\t4\nJim_Tressel\t2\t4\nJoan_Laporta\t1\t6\nJoan_Laporta\t6\t7\nJoan_Laporta\t6\t8\nJoe_Dumars\t1\t2\nJohn_Ashcroft\t3\t5\nJohn_Ashcroft\t13\t34\nJohn_Ashcroft\t42\t43\nJohn_Bolton\t6\t7\nJohn_Bolton\t11\t16\nJohn_Swofford\t1\t2\nJohn_Swofford\t1\t3\nJohn_Timoney\t1\t2\nJon_Gruden\t1\t2\nJon_Gruden\t2\t5\nJon_Gruden\t3\t6\nJon_Gruden\t3\t7\nJulie_Taymor\t1\t2\nKamal_Kharrazi\t1\t5\nKamal_Kharrazi\t2\t3\nKamal_Kharrazi\t2\t6\nKarin_Stoiber\t1\t2\nKim_Yong-il\t1\t2\nKim_Yong-il\t1\t3\nKimi_Raikkonen\t1\t3\nKimi_Raikkonen\t2\t3\nKjell_Magne_Bondevik\t1\t3\nKobe_Bryant\t1\t2\nKurt_Warner\t1\t2\nKurt_Warner\t1\t4\nKurt_Warner\t2\t4\nKurt_Warner\t3\t5\nKwon_Yang-sook\t1\t3\nKwon_Yang-sook\t2\t3\nLee_Jun\t1\t2\nLeonid_Kuchma\t1\t2\nLeonid_Kuchma\t1\t3\nLeonid_Kuchma\t2\t3\nLeonid_Kuchma\t2\t5\nLeonid_Kuchma\t4\t5\nLina_Krasnoroutskaya\t1\t2\nLisa_Raymond\t1\t2\nLord_Hutton\t1\t2\nLuiz_Felipe_Scolari\t1\t2\nLynn_Redgrave\t2\t3\nMagdalena_Maleeva\t1\t2\nMagdalena_Maleeva\t1\t3\nMagdalena_Maleeva\t2\t3\nMarcelo_Ebrard\t1\t2\nMarcelo_Ebrard\t1\t3\nMarcelo_Rios\t1\t3\nMarcelo_Rios\t1\t4\nMarcelo_Rios\t2\t3\nMarcelo_Rios\t3\t4\nMarcelo_Rios\t3\t5\nMaria_Shriver\t1\t7\nMaria_Shriver\t1\t8\nMaria_Shriver\t2\t6\nMaria_Shriver\t2\t8\nMaria_Shriver\t3\t8\nMarilyn_Monroe\t1\t2\nMario_Cipollini\t1\t2\nMark_Hurlbert\t1\t5\nMark_Hurlbert\t2\t3\nMark_Hurlbert\t2\t4\nMartin_Scorsese\t2\t5\nMartin_Scorsese\t3\t4\nMarwan_Barghouthi\t1\t2\nMichael_Bloomberg\t1\t12\nMichael_Bloomberg\t3\t5\nMichael_Bloomberg\t7\t11\nMichael_Bloomberg\t11\t13\nMichael_Chang\t1\t3\nMichael_Chang\t5\t6\nMichael_Jackson\t1\t6\nMichael_Jackson\t2\t3\nMichael_Jackson\t2\t11\nMichael_Jackson\t2\t12\nMichael_Jackson\t11\t12\nMichael_Patrick_King\t1\t2\nMichelle_Rodriguez\t1\t2\nMike_Babcock\t1\t2\nMike_Martz\t1\t3\nMike_Scioscia\t1\t2\nMikulas_Dzurinda\t1\t2\nMiroljub\t1\t2\nNasser_al-Kidwa\t1\t2\nNick_Nolte\t1\t2\nNick_Nolte\t1\t3\nNick_Nolte\t1\t5\nNick_Nolte\t2\t4\nNick_Nolte\t3\t4\nOJ_Simpson\t1\t2\nOprah_Winfrey\t1\t2\nOprah_Winfrey\t2\t3\nOrlando_Bloom\t2\t3\nOzzy_Osbourne\t1\t3\nPaul_Coppin\t1\t2\nPaul_Martin\t1\t6\nPaul_Martin\t1\t7\nPaul_McNulty\t1\t2\nPaul_ONeill\t1\t3\nPaul_ONeill\t5\t7\nPaul_Patton\t1\t2\nPaula_Zahn\t1\t2\nPierce_Brosnan\t1\t5\nPierce_Brosnan\t3\t7\nPierce_Brosnan\t6\t9\nPierce_Brosnan\t9\t14\nPierre_Pettigrew\t1\t3\nRalf_Schumacher\t5\t8\nRicky_Barnes\t1\t2\nRita_Wilson\t1\t4\nRob_Schneider\t1\t2\nRobbie_Fowler\t1\t2\nRobert_Blake\t1\t3\nRobert_Blake\t1\t4\nRobert_Blake\t5\t6\nRogerio_Romero\t1\t2\nRupert_Grint\t1\t3\nSalman_Rushdie\t1\t3\nSalman_Rushdie\t2\t3\nScott_Sullivan\t1\t2\nScott_Wolf\t1\t2\nSepp_Blatter\t1\t2\nSepp_Blatter\t1\t3\nSepp_Blatter\t2\t3\nSimon_Cowell\t1\t2\nSlobodan_Milosevic\t1\t3\nSlobodan_Milosevic\t2\t3\nSlobodan_Milosevic\t3\t4\nStacy_Dragila\t1\t2\nStephen_Ambrose\t1\t2\nStephen_Daldry\t1\t2\nStephen_Friedman\t1\t2\nTheo_Epstein\t1\t2\nThomas_Wyman\t1\t2\nTim_Floyd\t1\t2\nTom_Watson\t1\t3\nTom_Watson\t2\t3\nTommy_Robredo\t1\t3\nTommy_Robredo\t2\t3\nTony_Parker\t1\t2\nTony_Stewart\t1\t6\nTony_Stewart\t2\t3\nTony_Stewart\t4\t5\nTony_Stewart\t4\t6\nVladimir_Voltchkov\t1\t2\nWang_Yi\t1\t2\nZafarullah_Khan_Jamali\t1\t2\nZhu_Rongji\t1\t3\nZhu_Rongji\t2\t8\nAaron_Tippin\t1\tEnos_Slaughter\t1\nAaron_Tippin\t1\tJuan_Carlos_Ortega\t1\nAaron_Tippin\t1\tMarlon_Devonish\t1\nAdam_Ant\t1\tJohn_Perrota\t1\nAdam_Ant\t1\tNoel_Forgeard\t1\nAdam_Ant\t1\tRichard_Regenhard\t1\nAdam_Mair\t1\tDaniel_Osorno\t1\nAdoor_Gopalakarishnan\t1\tNathalia_Gillot\t1\nAlain_Ducasse\t1\tPaul_ONeill\t2\nAlan_Greer\t1\tAlan_Trammell\t1\nAlan_Greer\t1\tBob_Hayes\t1\nAlan_Trammell\t1\tHeidi_Fleiss\t2\nAlan_Trammell\t1\tJulie_Taymor\t2\nAldo_Paredes\t1\tSuzanne_Mubarak\t1\nAldo_Paredes\t2\tZafarullah_Khan_Jamali\t1\nAlecos_Markides\t1\tDarryl_Stingley\t1\nAlecos_Markides\t1\tWolfgang_Schneiderhan\t1\nAlejandro_Avila\t3\tBenjamin_Franklin\t1\nAlejandro_Avila\t3\tDarrell_Porter\t1\nAlek_Wek\t1\tBarbara_Bodine\t1\nAlek_Wek\t1\tNasser_al-Kidwa\t1\nAlek_Wek\t1\tRay_Bradbury\t1\nAlina_Kabaeva\t1\tDonald_Trump\t1\nAlina_Kabaeva\t1\tJayson_Williams\t2\nAlyse_Beaupre\t1\tElmar_Brok\t1\nAlyse_Beaupre\t1\tHelo_Pinheiro\t1\nAlyse_Beaupre\t1\tHunter_Bates\t1\nAmbrose_Lee\t1\tBen_Chandler\t1\nAmy_Gale\t1\tHerman_Edwards\t1\nAnastasia_Kelesidou\t1\tBrendan_Hansen\t2\nAnastasia_Kelesidou\t1\tThomas_Wyman\t2\nAndrew_Niccol\t1\tPrince_Felipe\t1\nAndrew_Sabey\t1\tFederico_Castelan_Sayre\t1\nAndrew_Sabey\t1\tGreg_Rusedski\t1\nAndy_Madikians\t1\tJon_Gruden\t1\nAndy_Madikians\t1\tPedro_Alvarez\t1\nAndy_Madikians\t1\tPierce_Brosnan\t1\nAndy_Perez\t1\tElena_Dementieva\t1\nAndy_Perez\t1\tStephen_Joseph\t1\nAngie_Martinez\t1\tRuth_Pearce\t1\nAnn_Godbehere\t1\tTom_Watson\t1\nAnne_McLellan\t1\tTim_Howard\t1\nAnnette_Bening\t1\tRoss_Verba\t1\nAnthony_Hopkins\t2\tOrlando_Bloom\t1\nAntonio_Bernardo\t1\tBilly_Donovan\t1\nAntonio_Bernardo\t1\tDwain_Kyles\t1\nAntonio_Bernardo\t1\tElizabeth_Hill\t1\nAntonio_Bernardo\t1\tGuy_Hemmings\t1\nAntonio_Bernardo\t1\tKareena_Kapoor\t1\nAnzori_Kikalishvili\t1\tCarlos_Quintanilla_Schmidt\t2\nAnzori_Kikalishvili\t1\tEarl_Fritts\t1\nAnzori_Kikalishvili\t1\tSalman_Rushdie\t1\nAram_Adler\t1\tCesar_Gaviria\t3\nAram_Adler\t1\tDeepa_Mehta\t1\nArie_Haan\t1\tTony_Parker\t2\nAriel_Sharon\t30\tDavid_Ballantyne\t1\nArt_Hoffmann\t2\tJuan_Carlos_Ortega\t1\nAtiabet_Ijan_Amabel\t1\tJohn_Perrota\t1\nAugustin_Calleri\t3\tLee_Jun\t2\nBarry_Nakell\t1\tMaria_Simon\t1\nBasdeo_Panday\t1\tFilippo_Volandri\t1\nBen_Betts\t1\tKimi_Raikkonen\t3\nBen_Braun\t1\tCecilia_Chang\t1\nBen_Braun\t1\tHorace_Newcomb\t1\nBen_Chandler\t1\tLarry_Hagman\t1\nBen_Lee\t1\tHorace_Newcomb\t1\nBen_Stein\t1\tDavid_Canary\t1\nBen_Stein\t1\tLionel_Richie\t2\nBernadette_Peters\t1\tEd_Smart\t1\nBertie_Ahern\t4\tJim_Leach\t1\nBill_OReilly\t1\tJim_Wessling\t1\nBilly_Boyd\t1\tSid_Caesar\t1\nBilly_Donovan\t1\tBrandon_Spann\t1\nBilly_Donovan\t1\tCabas\t1\nBilly_Donovan\t1\tWilliam_Nessen\t1\nBinyamin_Ben-Eliezer\t1\tJenny_Romero\t1\nBo_Ryan\t2\tHeidi_Fleiss\t3\nBob_Beauprez\t2\tHedayat_Amin_Arsala\t1\nBob_Beauprez\t2\tJohn_Norquist\t1\nBob_Hayes\t1\tZhu_Rongji\t8\nBrady_Rodgers\t1\tJenna_Elfman\t1\nBrandon_Larson\t1\tChita_Rivera\t1\nBrendan_Hansen\t2\tDorothy_Wilson\t1\nCaio_Blat\t1\tSanja_Papic\t1\nCarlos_Quintanilla_Schmidt\t1\tNova_Esther_Guthrie\t1\nCarlos_Quintanilla_Schmidt\t2\tChris_Thomas\t1\nCarroll_Weimer\t1\tChris_Pronger\t1\nCarroll_Weimer\t1\tDebbie_Reynolds\t2\nCecilia_Bolocco\t1\tJohn_Perrota\t1\nCecilia_Bolocco\t1\tKurt_Schottenheimer\t1\nCelia_Cruz\t1\tRob_Ramsay\t1\nCharlie_Zaa\t2\tTheo_Epstein\t2\nChris_Pronger\t1\tEdgar_Savisaar\t1\nChristian_Longo\t2\tDavid_Carradine\t1\nChristian_Wulff\t2\tKjell_Magne_Bondevik\t3\nCindy_Klassen\t1\tVal_Ackerman\t1\nCindy_Taylor\t1\tFabian_Vargas\t1\nClaudette_Robinson\t1\tEric_Fehr\t1\nClifford_Robinson\t1\tMagdalena_Maleeva\t2\nClifford_Robinson\t1\tShirley_Jones\t1\nClifford_Robinson\t1\tTim_Howard\t1\nColin_Jackson\t1\tMarcelo_Ebrard\t3\nColin_Jackson\t2\tHunter_Bates\t1\nColumba_Bush\t1\tLarry_Anderson\t1\nColumba_Bush\t1\tNeil_Goldman\t1\nCraig_Fitzgibbon\t1\tJoe_Dumars\t1\nCristian_Barros\t1\tSteve_Largent\t1\nCurtis_Strange\t1\tKurt_Schottenheimer\t1\nCurtis_Strange\t1\tRaul_Chacon\t1\nDan_Kellner\t1\tFreda_Black\t1\nDan_LaCoutre\t1\tGustavo_Kuerten\t2\nDan_Monson\t1\tJim_Flaherty\t1\nDan_Monson\t1\tRafael_Vinoly\t1\nDarrell_Porter\t2\tRobert_Towne\t1\nDave_Campo\t2\tJenny_Romero\t1\nDavid_Ballantyne\t1\tDiana_Taurasi\t2\nDavid_Carradine\t1\tKobe_Bryant\t1\nDavid_Dodge\t1\tFrancois_Botha\t1\nDavid_Dodge\t1\tLarry_Anderson\t1\nDavid_Dodge\t2\tThomas_Wyman\t2\nDavid_Heyman\t1\tRod_Paige\t1\nDavid_Oh\t1\tDesiree_McKenzie\t1\nDavid_Oh\t1\tJoe_Garner\t1\nDavid_Spade\t2\tDenise_Locke\t1\nDavid_Suazo\t1\tJane_Krakowski\t1\nDavid_Wells\t1\tMary_Hill\t1\nDavid_Wells\t1\tYannos_Papantoniou\t1\nDawna_LoPiccolo\t1\tJean-Claude_Juncker\t2\nDebbie_Reynolds\t1\tNick_Nolte\t1\nDeniz_Baykal\t1\tKathy_Bates\t1\nDeniz_Baykal\t1\tRobin_Wagner\t1\nDesiree_McKenzie\t1\tJohn_Danforth\t1\nDick_Devine\t1\tMichael_Jackson\t9\nDinora_Rosales\t1\tRobbie_Fowler\t1\nDirk_Kempthorne\t1\tJanusz_Kaminski\t1\nDonald_Evans\t2\tFred_Funk\t2\nDorothy_Wilson\t1\tJim_Thome\t1\nDrew_Gooden\t1\tKimi_Raikkonen\t3\nDwain_Kyles\t1\tGregory_Peck\t1\nEd_Smart\t3\tJason_Gardner\t1\nEdgar_Savisaar\t1\tKirk_Doerger\t1\nEdward_Burns\t1\tPaul_ONeill\t1\nElmar_Brok\t1\tJose_Jose\t1\nEmily_Mortimer\t1\tMaria_Simon\t1\nEnos_Slaughter\t1\tQais_al-Kazali\t1\nEric_Fehr\t1\tLee_Jun\t2\nEric_Fehr\t1\tTamara_Mowry\t1\nFaisal_Saleh_Hayat\t1\tMarwan_Barghouthi\t1\nFederico_Castelan_Sayre\t1\tJimmy_Jimenez\t1\nFiona_Milne\t1\tHartmut_Mehdorn\t1\nFiona_Milne\t1\tJosh_Kronfeld\t1\nFloyd_Keith\t1\tHalle_Berry\t1\nFranklin_Brown\t1\tMichael_Jackson\t10\nFranklin_Brown\t1\tTamara_Mowry\t1\nFranklin_Brown\t1\tYolanda_King\t1\nFreda_Black\t1\tRod_Paige\t1\nFruit_Chan\t1\tKjell_Magne_Bondevik\t3\nGary_Winnick\t1\tMichael_Patrick_King\t2\nGeorge_Robertson\t5\tRobbie_Mc_Ewen\t1\nGloria_Trevi\t4\tWilliam_Ragland\t1\nGrace_Dodd\t1\tHenrique_Meirelles\t1\nGraeme_Lloyd\t1\tHumberto_Coelho\t1\nGraeme_Lloyd\t1\tLaura_Pausini\t1\nGreg_Rusedski\t2\tHorace_Newcomb\t1\nGreg_Rusedski\t3\tJavier_Bardem\t1\nGreg_Rusedski\t3\tMarcos_Milinkovic\t1\nGregory_Peck\t1\tJacqueline_Obradors\t1\nGuillermo_Ruiz_Polanco\t1\tJon_Constance\t1\nGustavo_Kuerten\t1\tRani_Mukherjee\t1\nHalle_Berry\t12\tJanet_Chandler\t1\nHartmut_Mehdorn\t1\tNatanaela_Barnova\t1\nHartmut_Mehdorn\t1\tSanja_Papic\t1\nHartmut_Mehdorn\t1\tSvetlana_Belousova\t1\nHaydar_Aliyev\t1\tNuon_Chea\t1\nHeather_Willson\t1\tLynn_Redgrave\t3\nHector_Mitelman\t1\tJon_Gruden\t2\nHelo_Pinheiro\t1\tRobert_Durst\t1\nHerman_Edwards\t1\tPatrick_Eaves\t1\nHermogenes_Ebdane_Jr\t1\tSanja_Papic\t1\nHorace_Newcomb\t1\tStephan_Eberharter\t1\nHugo_Colace\t1\tKim_Yun-kyu\t1\nHumberto_Coelho\t1\tSue_Wicks\t2\nHunter_Bates\t1\tPedro_Alvarez\t1\nHunter_Bates\t1\tWilliam_Rosenberg\t1\nIgor_Trunov\t1\tPeter_Sejna\t1\nIvan_Stambolic\t1\tMichael_Bloomberg\t18\nJack_LaLanne\t1\tYves_Brodeur\t1\nJacqueline_Obradors\t1\tJulie_Taymor\t1\nJada_Pinkett_Smith\t2\tJenny_Romero\t1\nJames_Cunningham\t3\tStephen_Daldry\t1\nJames_Gibson\t1\tRoger_Corbett\t1\nJames_Parker\t1\tSaeed_Mortazavi\t1\nJane_Krakowski\t1\tShingo_Suetsugu\t1\nJanice_Goldfinger\t1\tPyar_Jung_Thapa\t1\nJanusz_Kaminski\t1\tLisa_Stone\t1\nJason_Lezak\t1\tJohn_Bolton\t4\nJason_Lezak\t1\tSimon_Cowell\t1\nJavier_Vazquez\t1\tKarin_Stoiber\t2\nJay_Rasulo\t1\tNova_Esther_Guthrie\t1\nJean-Marc_Olive\t1\tJohn_Timoney\t1\nJean-Marc_Olive\t1\tTino_Martinez\t1\nJeri_Ryan\t1\tNova_Esther_Guthrie\t1\nJeri_Ryan\t1\tPeter_Goldmark\t1\nJeri_Ryan\t1\tScott_Sullivan\t2\nJeri_Ryan\t1\tSkip_Prosser\t1\nJesse_James_Leija\t2\tJoaquin_Phoenix\t1\nJesse_James_Leija\t2\tNick_Nolte\t4\nJesse_Ventura\t2\tLela_Rochon\t1\nJessica_Biel\t1\tJim_Fassel\t1\nJim_Flaherty\t1\tMark_Andrew\t1\nJim_Tressel\t3\tRalf_Schumacher\t3\nJimmy_Jimenez\t1\tMiles_Stewart\t1\nJimmy_Lee\t1\tNova_Esther_Guthrie\t1\nJoan_Laporta\t3\tRoy_Romanow\t1\nJoaquin_Phoenix\t1\tRainer_Geulen\t1\nJoe_Dumars\t2\tTimothy_McVeigh\t1\nJohn_Bolton\t14\tKim_Yun-kyu\t1\nJohn_Kerr\t1\tLi_Changchun\t1\nJohn_Swofford\t2\tSok_An\t1\nJuan_Fernandez\t1\tPaul_Krueger\t1\nJuan_Fernandez\t1\tSuzanne_Mubarak\t1\nJustin_Wilson\t1\tLarry_Hagman\t1\nJustin_Wilson\t1\tRay_Bradbury\t1\nKara_Lynn_Joyce\t1\tZach_Pillar\t1\nKim_Chinn\t1\tRobert_Flodquist\t1\nKim_Chinn\t1\tRuth_Harlow\t2\nKim_Yong-il\t2\tLinda_Ham\t1\nKwon_Yang-sook\t3\tTommy_Robredo\t1\nLane_Bryant\t1\tYannos_Papantoniou\t1\nLarry_Hagman\t1\tTeddy_Kollek\t1\nLarry_Hagman\t1\tVal_Ackerman\t1\nLaura_Gobai\t1\tRobin_Wagner\t1\nLaura_Pausini\t1\tThad_Matta\t1\nLaura_Romero\t1\tPaula_Zahn\t2\nLee_Jun\t1\tRuth_Pearce\t1\nLee_Jun\t2\tStephen_Friedman\t1\nLela_Rochon\t1\tMichael_Haneke\t1\nLela_Rochon\t1\tYannos_Papantoniou\t1\nLesley_Flood\t1\tPeter_Goldmark\t1\nLi_Changchun\t1\tPieter_Bouw\t1\nLina_Krasnoroutskaya\t1\tTroy_Polamalu\t1\nLinda_Lingle\t1\tVictor_Hanescu\t1\nLisa_Stone\t1\tRod_Paige\t1\nLisa_Stone\t1\tScott_Sullivan\t2\nLou_Lang\t1\tNova_Esther_Guthrie\t1\nLuc_Montagnier\t1\tPaul_Krueger\t1\nLuc_Montagnier\t1\tPedro_Alvarez\t1\nLuc_Montagnier\t1\tPierre_Pettigrew\t2\nLuis_Guzman\t1\tPatsy_Kensit\t1\nMarcelo_Ebrard\t2\tMike_Scioscia\t2\nMarcelo_Rios\t4\tMaria_Shriver\t3\nMarcelo_Rios\t4\tPierre_Pettigrew\t1\nMarcelo_Rios\t4\tSaeed_Mortazavi\t1\nMargaret_Thatcher\t1\tPedro_Alvarez\t1\nMargaret_Thatcher\t1\tSalman_Rushdie\t2\nMaria_Simon\t1\tMona_Rishmawi\t1\nMaria_Simon\t1\tRobert_Durst\t1\nMarlon_Devonish\t1\tPatrick_Clawsen\t1\nMax_von_Sydow\t1\tZhu_Rongji\t1\nMel_Karmazin\t1\tPierre_Pettigrew\t1\nMelana_Scantlin\t1\tWilliam_Ragland\t1\nMichael_Bloomberg\t15\tOzzy_Osbourne\t2\nMichael_J_Fox\t1\tRicky_Barnes\t1\nMichael_J_Fox\t1\tThomas_Daily\t1\nMicky_Ward\t1\tTakenori_Kanzaki\t1\nMike_Babcock\t1\tTony_Parker\t2\nMike_Martz\t5\tWilliam_Burns\t1\nMira_Sorvino\t1\tSimon_Cowell\t2\nMira_Sorvino\t1\tTom_Tunney\t1\nMiroljub\t2\tTheo_Epstein\t1\nMitzi_Gaynor\t1\tRuth_Harlow\t1\nMohammed_Dahlan\t1\tRonald_White\t1\nNatanaela_Barnova\t1\tNuon_Chea\t1\nNoel_Forgeard\t1\tZach_Pillar\t1\nNormand_Legault\t1\tOmar_Vizquel\t1\nNormand_Legault\t1\tRupert_Grint\t3\nNova_Esther_Guthrie\t1\tStephen_Joseph\t1\nOntario_Lett\t1\tWallace_Capel\t1\nOrlando_Bloom\t1\tRay_Liotta\t1\nPatrick_Clawsen\t1\tSandra_Banning\t1\nPaul_Coppin\t2\tRick_Husband\t1\nPaul_Murphy\t1\tQazi_Hussain_Ahmed\t1\nPaul_Newman\t1\tRobert_Blake\t3\nPaula_Zahn\t1\tTamara_Mowry\t1\nPeter_Ahearn\t1\tRomain_Duris\t1\nPeter_Gabriel\t1\tPeter_OToole\t1\nPeter_Lundgren\t1\tWilliam_Rosenberg\t1\nPeter_OToole\t1\tQazi_Afzal\t1\nQais_al-Kazali\t1\tRingo_Starr\t1\nRandy_Brown\t1\tVal_Ackerman\t1\nRani_Mukherjee\t1\tTimothy_McVeigh\t1\nRingo_Starr\t1\tZach_Pillar\t1\nRoger_Corbett\t1\tTocker_Pudwill\t1\nRuth_Harlow\t1\tVirgina_Ruano_Pascal\t1\nSandra_Banning\t1\tWolfgang_Schneiderhan\t1\nScott_Wolf\t2\tTroy_Polamalu\t1\nSergei_Alexandrovitch_Ordzhonikidze\t1\tYolanda_King\t1\nShane_Loux\t1\tVal_Ackerman\t1\nShawn_Marion\t1\tShirley_Jones\t1\nSlobodan_Milosevic\t2\tSok_An\t1\n"
  },
  {
    "path": "lfw/data/pairsDevTest.txt",
    "content": "500\nAbdullah_Gul\t13\t14\nAbdullah_Gul\t13\t16\nAbdullatif_Sener\t1\t2\nAdel_Al-Jubeir\t1\t3\nAl_Pacino\t1\t2\nAlan_Greenspan\t1\t5\nAlbert_Costa\t2\t6\nAlbert_Costa\t4\t6\nAlbert_Costa\t5\t6\nAlejandro_Atchugarry\t1\t2\nAlex_Penelas\t1\t2\nAli_Naimi\t1\t3\nAli_Naimi\t3\t4\nAllison_Janney\t1\t2\nAllyson_Felix\t2\t5\nAlvaro_Uribe\t7\t10\nAlvaro_Uribe\t17\t26\nAlvaro_Uribe\t28\t29\nAmanda_Bynes\t1\t2\nAmanda_Bynes\t1\t3\nAmanda_Bynes\t1\t4\nAmanda_Bynes\t3\t4\nAmelie_Mauresmo\t1\t21\nAmelie_Mauresmo\t4\t11\nAmelie_Mauresmo\t16\t21\nAna_Palacio\t1\t6\nAndrei_Mikhnevich\t1\t2\nAndy_Hebb\t1\t2\nAngelo_Reyes\t1\t2\nAngelo_Reyes\t2\t3\nAnn_Veneman\t1\t2\nAnn_Veneman\t5\t11\nAnn_Veneman\t8\t9\nAnna_Nicole_Smith\t1\t2\nAnne_Krueger\t1\t2\nAnne_Krueger\t2\t3\nAnnette_Lu\t1\t2\nAnnette_Lu\t1\t3\nAnnette_Lu\t2\t3\nAntonio_Palocci\t3\t8\nAntonio_Trillanes\t1\t2\nAntonio_Trillanes\t2\t3\nArlen_Specter\t1\t3\nAugustin_Calleri\t1\t3\nAugustin_Calleri\t1\t4\nBarbara_Brezigar\t1\t2\nBegum_Khaleda_Zia\t1\t2\nBen_Affleck\t1\t3\nBen_Affleck\t2\t6\nBen_Affleck\t3\t6\nBertie_Ahern\t2\t3\nBill_Frist\t1\t3\nBill_Frist\t2\t7\nBill_Frist\t3\t8\nBill_McBride\t2\t6\nBill_McBride\t2\t10\nBill_McBride\t3\t9\nBill_Sizemore\t1\t2\nBono\t1\t2\nBono\t1\t3\nBono\t2\t3\nBrian_Cowen\t1\t2\nBridget_Fonda\t1\t3\nBridgette_Wilson-Sampras\t1\t2\nButch_Davis\t1\t2\nCalista_Flockhart\t2\t6\nCandice_Bergen\t1\t3\nCarla_Del_Ponte\t2\t4\nCarla_Del_Ponte\t4\t5\nCarlos_Manuel_Pruneda\t1\t2\nCarlos_Manuel_Pruneda\t1\t3\nCarlos_Manuel_Pruneda\t2\t3\nCarlos_Moya\t7\t11\nCarlos_Ortega\t1\t3\nCarolyn_Dawn_Johnson\t2\t3\nCarrie-Anne_Moss\t2\t5\nCarson_Daly\t1\t2\nCathy_Freeman\t1\t2\nChanda_Rubin\t2\t4\nChang_Dae-whan\t1\t2\nCharles_Taylor\t3\t9\nChen_Shui-bian\t3\t5\nChita_Rivera\t1\t2\nChok_Tong_Goh\t1\t2\nChris_Bell\t1\t2\nChris_Byrd\t1\t2\nChristian_Longo\t1\t3\nChristian_Longo\t2\t3\nChristian_Wulff\t1\t2\nChristine_Ebersole\t1\t2\nChristine_Todd_Whitman\t5\t6\nChristopher_Reeve\t1\t4\nClaire_Leger\t1\t2\nCondoleezza_Rice\t2\t5\nCondoleezza_Rice\t7\t10\nCristina_Fernandez\t1\t2\nDai_Bachtiar\t1\t2\nDaisy_Fuentes\t3\t4\nDave_Campo\t1\t2\nDave_Campo\t2\t3\nDavid_Stern\t1\t4\nDavid_Stern\t3\t4\nDenise_Johnson\t1\t2\nDennis_Powell\t1\t2\nDenzel_Washington\t1\t3\nDenzel_Washington\t2\t4\nDenzel_Washington\t2\t5\nDerek_Jeter\t2\t3\nDiane_Green\t1\t2\nDick_Clark\t1\t2\nDick_Clark\t1\t3\nDick_Clark\t2\t3\nDick_Vermeil\t1\t2\nDominique_de_Villepin\t3\t5\nDominique_de_Villepin\t7\t11\nDominique_de_Villepin\t14\t15\nDonald_Pettit\t1\t2\nDonald_Pettit\t2\t3\nDonald_Rumsfeld\t22\t36\nDonald_Rumsfeld\t28\t108\nDonald_Rumsfeld\t41\t48\nDoug_Collins\t1\t2\nEdward_Lu\t2\t4\nEdward_Norton\t1\t2\nEdwin_Edwards\t1\t2\nEdwin_Edwards\t2\t3\nEliane_Karp\t1\t4\nEliane_Karp\t2\t3\nElin_Nordegren\t1\t2\nElinor_Caplan\t1\t2\nElisabeth_Schumacher\t1\t2\nElizabeth_Smart\t1\t5\nElizabeth_Smart\t3\t4\nElsa_Zylberstein\t1\t2\nElsa_Zylberstein\t1\t5\nElsa_Zylberstein\t3\t6\nEmma_Watson\t1\t4\nEmma_Watson\t2\t3\nEmma_Watson\t3\t5\nErik_Morales\t1\t3\nErika_Christensen\t1\t2\nErin_Runnion\t3\t4\nErnie_Fletcher\t1\t2\nFabiola_Zuluaga\t1\t2\nFelipe_Perez_Roque\t1\t4\nFelipe_Perez_Roque\t2\t4\nFernando_Vargas\t2\t4\nFlor_Montulo\t1\t2\nFrank_Cassell\t1\t2\nFrank_Dunham_Jr\t1\t2\nGao_Qiang\t1\t2\nGarry_Trudeau\t1\t2\nGary_Doer\t1\t2\nGary_Doer\t1\t3\nGary_Doer\t2\t3\nGary_Winnick\t1\t2\nGene_Robinson\t2\t4\nGeorge_Foreman\t1\t2\nGeorge_Karl\t1\t2\nGeorge_Robertson\t5\t20\nGeorge_Robertson\t12\t19\nGeorge_Robertson\t15\t22\nGeorgi_Parvanov\t1\t2\nGeraldine_Chaplin\t1\t3\nGerhard_Schroeder\t3\t26\nGerry_Adams\t3\t7\nGerry_Adams\t5\t8\nGilberto_Rodriguez_Orejuela\t1\t3\nGilberto_Rodriguez_Orejuela\t2\t4\nGilberto_Rodriguez_Orejuela\t3\t4\nGiuseppe_Gibilisco\t2\t4\nGiuseppe_Gibilisco\t3\t4\nGlafcos_Clerides\t1\t4\nGordon_Brown\t4\t11\nGordon_Campbell\t1\t2\nGray_Davis\t6\t23\nGray_Davis\t13\t23\nGreg_Gilbert\t1\t2\nGreg_Ostertag\t1\t2\nGregory_Hines\t1\t2\nGus_Van_Sant\t1\t2\nGwendal_Peizerat\t1\t2\nGwyneth_Paltrow\t4\t5\nHarry_Kalas\t1\t2\nHassan_Wirajuda\t1\t2\nHeather_Mills\t3\t4\nHeidi_Fleiss\t1\t3\nHeidi_Fleiss\t1\t4\nHermann_Maier\t1\t2\nHorst_Koehler\t1\t3\nIan_Thorpe\t2\t7\nIan_Thorpe\t3\t10\nIan_Thorpe\t6\t8\nIsabella_Rossellini\t1\t2\nIsabelle_Huppert\t1\t2\nIsmail_Merchant\t1\t2\nJack_Grubman\t1\t2\nJack_Nicholson\t2\t3\nJacques_Chirac\t22\t44\nJacques_Chirac\t30\t33\nJacques_Chirac\t31\t52\nJacques_Rogge\t4\t6\nJake_Gyllenhaal\t2\t4\nJake_Gyllenhaal\t3\t4\nJames_Butts\t1\t2\nJames_Franco\t1\t2\nJames_Kelly\t4\t11\nJames_Kelly\t9\t11\nJames_Kopp\t1\t3\nJames_Kopp\t2\t3\nJames_Maguire\t1\t2\nJames_Smith\t1\t2\nJan_Ullrich\t2\t3\nJan_Ullrich\t2\t5\nJanica_Kostelic\t1\t2\nJavier_Weber\t1\t2\nJavier_Weber\t1\t3\nJean-Claude_Juncker\t1\t2\nJean-Claude_Trichet\t1\t2\nJean-Marc_de_La_Sabliere\t1\t2\nJean-Pierre_Raffarin\t2\t6\nJean_Carnahan\t1\t2\nJefferson_Perez\t1\t2\nJennifer_Capriati\t3\t13\nJennifer_Capriati\t24\t29\nJennifer_Connelly\t1\t4\nJennifer_Connelly\t3\t4\nJennifer_Keller\t2\t4\nJennifer_Keller\t3\t4\nJeong_Se-hyun\t4\t5\nJeremy_Shockey\t1\t2\nJesse_Ventura\t1\t3\nJessica_Lange\t1\t2\nJiang_Zemin\t2\t17\nJiang_Zemin\t12\t15\nJim_Harrick\t1\t2\nJimmy_Carter\t1\t9\nJimmy_Carter\t4\t8\nJimmy_Carter\t8\t9\nJoe_Mantello\t1\t2\nJoe_Nichols\t1\t3\nJoe_Nichols\t3\t4\nJohn_Bolton\t3\t10\nJohn_Bolton\t4\t14\nJohn_Bolton\t5\t13\nJohn_Bolton\t11\t14\nJohn_Garamendi\t1\t2\nJohn_McCallum\t1\t2\nJohn_Negroponte\t4\t17\nJohn_Rowland\t1\t2\nJohn_Stallworth\t1\t2\nJohn_Travolta\t4\t5\nJohn_Walsh\t1\t2\nJohnny_Carson\t1\t2\nJorge_Castaneda\t1\t2\nJose_Dirceu\t1\t2\nJoseph_Deiss\t1\t2\nJudi_Dench\t1\t2\nJulie_Gerberding\t7\t12\nJulie_Gerberding\t8\t12\nJustin_Leonard\t1\t3\nJustine_Henin\t1\t3\nJustine_Henin\t2\t3\nKamal_Kharrazi\t3\t5\nKeith_Bogans\t2\t3\nKenneth_Evans\t1\t2\nKim_Yong-il\t1\t2\nKing_Abdullah_II\t1\t2\nKing_Abdullah_II\t1\t3\nKing_Abdullah_II\t1\t4\nKing_Abdullah_II\t2\t3\nKjell_Magne_Bondevik\t1\t2\nKosuke_Kitajima\t1\t2\nKristen_Breitweiser\t1\t2\nKristin_Davis\t1\t2\nKristin_Davis\t2\t3\nKurt_Warner\t1\t5\nLK_Advani\t1\t2\nLK_Advani\t1\t3\nLK_Advani\t2\t3\nLarry_Coker\t2\t4\nLatrell_Sprewell\t1\t2\nLaura_Linney\t1\t3\nLauren_Killian\t1\t2\nLaurent_Jalabert\t1\t2\nLeonardo_DiCaprio\t1\t8\nLeonardo_DiCaprio\t3\t7\nLeonid_Kuchma\t1\t6\nLeonid_Kuchma\t2\t3\nLeonid_Kuchma\t2\t4\nLeonid_Kuchma\t4\t6\nLeslie_Moonves\t1\t2\nLeszek_Miller\t2\t3\nLino_Oviedo\t1\t2\nLisa_Raymond\t1\t2\nLiza_Minnelli\t1\t3\nLloyd_Ward\t1\t2\nLuciano_Pavarotti\t1\t3\nLucy_Liu\t2\t4\nLuis_Figo\t1\t2\nLuis_Horna\t1\t3\nLuis_Horna\t2\t5\nLynn_Abraham\t1\t2\nMack_Brown\t1\t2\nMahmoud_Abbas\t19\t26\nMarcelo_Rios\t1\t4\nMarcelo_Rios\t2\t4\nMarcelo_Rios\t4\t5\nMarieta_Chrousala\t1\t2\nMarieta_Chrousala\t1\t3\nMario_Cipollini\t1\t2\nMark_Dacey\t1\t2\nMark_Richt\t1\t3\nMark_Richt\t2\t3\nMartha_Burk\t2\t4\nMartha_Stewart\t1\t2\nMartha_Stewart\t2\t5\nMartha_Stewart\t3\t5\nMartin_Cauchon\t1\t2\nMartin_Sheen\t1\t2\nMary_Landrieu\t1\t3\nMasum_Turker\t1\t2\nMasum_Turker\t1\t3\nMatt_Doherty\t1\t2\nMatt_Doherty\t1\t3\nMatthew_Broderick\t1\t4\nMatthew_Broderick\t2\t3\nMatthew_Broderick\t3\t4\nMelissa_Etheridge\t1\t2\nMichael_Jackson\t2\t8\nMichael_Phelps\t1\t3\nMichael_Sullivan\t1\t2\nMichel_Duclos\t1\t2\nMichelle_Yeoh\t3\t4\nMiguel_Contreras\t1\t2\nMike_Price\t1\t2\nMikhail_Youzhny\t1\t2\nMikhail_Youzhny\t1\t3\nMikhail_Youzhny\t2\t3\nMikulas_Dzurinda\t1\t2\nMireya_Moscoso\t1\t3\nMireya_Moscoso\t2\t3\nMohamed_ElBaradei\t3\t6\nMohamed_ElBaradei\t4\t5\nMohamed_ElBaradei\t5\t8\nMonica_Lewinsky\t2\t3\nMuhammad_Ali\t4\t7\nMuhammad_Ali\t6\t9\nNancy_Pelosi\t2\t5\nNancy_Pelosi\t6\t13\nNancy_Pelosi\t13\t14\nNaomi_Campbell\t1\t2\nNaoto_Kan\t2\t3\nNasser_al-Kidwa\t1\t2\nNastassia_Kinski\t1\t2\nNatalie_Coughlin\t1\t2\nNatalie_Coughlin\t3\t5\nNatalie_Maines\t2\t4\nNicanor_Duarte_Frutos\t8\t10\nNoah_Wyle\t1\t2\nNorm_Coleman\t2\t7\nOrrin_Hatch\t1\t2\nOsama_bin_Laden\t2\t4\nOwen_Wilson\t1\t2\nOzzy_Osbourne\t1\t2\nOzzy_Osbourne\t2\t3\nPadraig_Harrington\t1\t3\nParis_Hilton\t1\t2\nPatrick_Leahy\t1\t2\nPatti_Labelle\t1\t2\nPatti_Labelle\t2\t3\nPatty_Schnyder\t1\t3\nPatty_Schnyder\t1\t4\nPaul-Henri_Mathieu\t1\t2\nPaul_Burrell\t4\t10\nPaul_Burrell\t5\t11\nPaul_Martin\t1\t8\nPaul_Martin\t2\t4\nPaul_Martin\t3\t6\nPaul_Patton\t1\t2\nPaul_Sarbanes\t2\t3\nPedro_Malan\t1\t2\nPedro_Malan\t1\t4\nPedro_Malan\t2\t3\nPenelope_Cruz\t1\t2\nPeter_Greenaway\t1\t2\nPierce_Brosnan\t4\t12\nPierce_Brosnan\t7\t8\nPierre_Boulanger\t1\t2\nPriscilla_Presley\t1\t2\nRachel_Griffiths\t1\t3\nRachel_Hunter\t2\t3\nRalf_Schumacher\t1\t6\nRebecca_Romijn-Stamos\t2\t4\nRebecca_Romijn-Stamos\t3\t4\nReese_Witherspoon\t1\t2\nReese_Witherspoon\t1\t4\nRicardo_Maduro\t1\t2\nRich_Gannon\t1\t2\nRichard_Gere\t1\t10\nRichard_Gere\t6\t10\nRichard_Norton-Taylor\t1\t2\nRichard_Shelby\t1\t2\nRichie_Adubato\t1\t2\nRick_Dinse\t1\t3\nRick_Perry\t1\t2\nRick_Perry\t2\t6\nRick_Pitino\t1\t2\nRick_Romley\t1\t2\nRick_Wagoner\t1\t2\nRob_Schneider\t1\t2\nRobbie_Fowler\t1\t2\nRobby_Ginepri\t1\t2\nRobert_Horan\t1\t2\nRobert_Mueller\t1\t3\nRobert_Redford\t2\t3\nRobert_Redford\t2\t5\nRobert_Redford\t4\t7\nRobert_Redford\t7\t8\nRod_Blagojevich\t1\t2\nRod_Stewart\t1\t3\nRonaldo_Luis_Nazario_de_Lima\t1\t4\nRoy_Moore\t2\t3\nRupert_Grint\t2\t3\nRussell_Coutts\t1\t2\nRussell_Crowe\t1\t2\nSalma_Hayek\t1\t7\nSalma_Hayek\t8\t10\nSarah_Jessica_Parker\t5\t6\nScott_McNealy\t1\t2\nSean_Astin\t1\t2\nSean_OKeefe\t4\t5\nSean_Patrick_OMalley\t1\t2\nSean_Patrick_OMalley\t1\t3\nSean_Patrick_OMalley\t2\t3\nSean_Penn\t2\t3\nSergey_Lavrov\t1\t10\nShane_Mosley\t1\t2\nShannon_OBrien\t1\t2\nSheila_Wellstone\t1\t2\nSilvio_Berlusconi\t14\t23\nSpencer_Abraham\t12\t13\nStan_Heath\t1\t2\nStefano_Accorsi\t1\t2\nSteve_Lavin\t1\t4\nSteven_Hatfill\t1\t2\nSurakait_Sathirathai\t1\t2\nSusan_Collins\t1\t2\nSusie_Castillo\t1\t2\nSvetlana_Koroleva\t1\t2\nTammy_Lynn_Michaels\t1\t2\nTang_Jiaxuan\t1\t6\nTang_Jiaxuan\t1\t9\nTang_Jiaxuan\t1\t10\nTang_Jiaxuan\t1\t11\nTassos_Papadopoulos\t1\t2\nTerry_McAuliffe\t1\t2\nTerry_McAuliffe\t1\t3\nThabo_Mbeki\t4\t5\nThaksin_Shinawatra\t1\t6\nThaksin_Shinawatra\t3\t6\nTheodore_Tweed_Roosevelt\t2\t3\nThomas_OBrien\t8\t9\nTim_Allen\t1\t2\nTim_Allen\t2\t4\nTim_Curry\t1\t2\nTippi_Hedren\t1\t2\nTodd_Haynes\t1\t3\nTodd_Haynes\t1\t4\nTom_Coverdale\t1\t2\nTom_Cruise\t1\t7\nTom_Cruise\t2\t4\nTom_Cruise\t4\t9\nTommy_Haas\t2\t3\nTomoko_Hagiwara\t1\t2\nTony_Shalhoub\t1\t3\nTony_Shalhoub\t1\t4\nTracee_Ellis_Ross\t1\t2\nTyler_Hamilton\t1\t2\nVaclav_Havel\t1\t6\nVaclav_Havel\t1\t9\nVicente_Fernandez\t2\t5\nVicente_Fernandez\t4\t5\nVictoria_Beckham\t2\t3\nVincent_Brooks\t1\t2\nVincent_Brooks\t2\t6\nWarren_Beatty\t1\t2\nWarren_Buffett\t1\t3\nWesley_Clark\t1\t2\nWill_Smith\t1\t2\nWilliam_Burns\t1\t2\nWilliam_Macy\t1\t4\nWilliam_Macy\t2\t3\nWilliam_Macy\t3\t5\nWilliam_Rehnquist\t1\t2\nWinona_Ryder\t6\t15\nWinona_Ryder\t19\t21\nYevgeny_Kafelnikov\t3\t4\nYoriko_Kawaguchi\t3\t10\nZoran_Djindjic\t3\t4\nAJ_Lamas\t1\tZach_Safrin\t1\nAaron_Guiel\t1\tReese_Witherspoon\t3\nAaron_Tippin\t1\tJose_Luis_Rodriguez_Zapatero\t1\nAbdul_Majeed_Shobokshi\t1\tCharles_Cope\t1\nAbdullah_Gul\t16\tSteve_Cox\t1\nAbid_Hamid_Mahmud_Al-Tikriti\t1\tEli_Broad\t1\nAdam_Kennedy\t1\tAmelie_Mauresmo\t19\nAdel_Al-Jubeir\t1\tElisabeth_Welch\t1\nAdrian_Murrell\t1\tTommy_Franks\t15\nAdriana_Lima\t1\tTerrence_Trammell\t1\nAdriana_Perez_Navarro\t1\tJennifer_Capriati\t8\nAgbani_Darego\t1\tMalik_Mahmud\t1\nAhmed_Ghazi\t1\tHenry_Suazo\t1\nAileen_Riggin_Soule\t1\tDamarius_Bilbo\t1\nAin_Seppik\t1\tDonald_Regan\t1\nAitor_Gonzalez\t1\tLily_Tomlin\t2\nAl_Cardenas\t1\tMary_Landrieu\t3\nAl_Davis\t2\tDennis_Powell\t1\nAl_Pacino\t1\tIzzat_Ibrahim\t1\nAleksander_Voloshin\t1\tAshlea_Talbot\t1\nAlex_Cabrera\t1\tRichard_Carl\t1\nAlex_Corretja\t1\tNewt_Gingrich\t1\nAlex_Corretja\t1\tTippi_Hedren\t2\nAlexandra_Rozovskaya\t1\tDon_King\t1\nAli_Fallahian\t1\tMohammed_Abulhasan\t1\nAli_Naimi\t8\tVanessa_Laine\t1\nAlicia_Keys\t1\tGiannina_Facio\t1\nAlicia_Molik\t1\tLena_Katina\t1\nAline_Chretien\t1\tCarlos_Iturgaitz\t1\nAlisha_Richman\t1\tSpencer_Abraham\t9\nAllison_Searing\t1\tAmy_Gale\t1\nAllison_Searing\t1\tMark_Martin\t1\nAllison_Searing\t1\tPhillip_Fulmer\t1\nAllison_Searing\t1\tWarren_Beatty\t1\nAmanda_Marsh\t1\tHoward_Smith\t2\nAmanda_Marsh\t1\tSvetlana_Koroleva\t2\nAmelie_Mauresmo\t6\tTerri_Clark\t1\nAmporn_Falise\t1\tChristiane_Wulff\t1\nAmporn_Falise\t1\tLiza_Minnelli\t2\nAmporn_Falise\t1\tZhang_Yimou\t1\nAmy_Gale\t1\tFrank_Murkowski\t1\nAna_Palacio\t5\tErika_Christensen\t2\nAna_Palacio\t6\tRobert_Gordon_Card\t1\nAndrew_Caldecott\t1\tJoe_Strummer\t1\nAndrew_Shutley\t1\tCharles_Taylor\t7\nAndrew_Wetzler\t1\tCristina_Torrens_Valero\t1\nAngela_Mascia-Frye\t1\tDerrick_Taylor\t1\nAnn_Godbehere\t1\tPaul_Newman\t1\nAnn_Godbehere\t1\tTatiana_Kennedy_Schlossberg\t1\nAnna_Chicherova\t1\tBob_Iger\t1\nAnna_Nicole_Smith\t1\tTony_Shalhoub\t4\nAnne_Krueger\t3\tLucrecia_Orozco\t1\nAnne_McLellan\t2\tRichard_Penniman\t1\nAnne_McLellan\t2\tRoman_Tam\t1\nAnnette_Lu\t1\tScott_Fawell\t1\nAnnette_Lu\t1\tSvetlana_Belousova\t1\nAnnie_Chaplin\t1\tGloria_Gaynor\t1\nAnnie_Chaplin\t1\tIon_Tiriac\t1\nAnnie_Chaplin\t1\tMichael_Keaton\t1\nAntonio_Palocci\t4\tDonald_Regan\t1\nAretha_Franklin\t1\tMarc_Racicot\t1\nArmando_Avila_Panchame\t1\tHamza_Atiya_Muhsen\t1\nArmando_Avila_Panchame\t1\tHana_Urushima\t1\nArthur_Johnson\t1\tSteve_Pagliuca\t1\nAscencion_Barajas\t1\tMarissa_Jaret_Winokur\t2\nAshlea_Talbot\t1\tSean_Patrick_Thomas\t1\nAshley_Judd\t1\tPat_Burns\t1\nAsif_Ali_Zardari\t1\tCristina_Torrens_Valero\t1\nAssad_Ahmadi\t1\tJose_Cevallos\t1\nAstou_Ndiaye-Diatta\t1\tScott_Yates\t1\nAugustin_Calleri\t4\tBrad_Miller\t1\nAugustin_Calleri\t4\tJohn_Jones\t1\nBabe_Ruth\t1\tJoshua_Perper\t1\nBarbara_Bach\t1\tEli_Rosenbaum\t1\nBarbara_Becker\t1\tGerhard_Boekel\t1\nBarbara_Roberts\t1\tKirsten_Clark\t1\nBarry_Williams\t1\tRichard_Langille\t1\nBarzan_al-Tikriti\t1\tJim_Haslett\t1\nBeecher_Ray_Kirby\t1\tHashan_Tillakaratne\t1\nBegum_Khaleda_Zia\t1\tJesse_Jackson\t5\nBen_Cohen\t1\tRoger_Etchegaray\t1\nBen_Cohen\t1\tZach_Parise\t1\nBenjamin_McKenzie\t1\tLisa_Leslie\t1\nBetsy_Coffin\t1\tGerard_Tronche\t1\nBill_Carmody\t1\tMichael_Sheehan\t1\nBill_Carmody\t1\tTippi_Hedren\t1\nBill_Curry\t1\tCandice_Bergen\t1\nBill_Frist\t2\tMalcolm_Wild\t1\nBill_Frist\t4\tDonald_Pettit\t2\nBill_Kollar\t1\tPhillip_Seymor_Hoffmann\t1\nBill_McBride\t8\tJohn_Rowland\t1\nBill_Sizemore\t1\tCalista_Flockhart\t3\nBilly_Boyd\t1\tLiza_Minnelli\t3\nBilly_Boyd\t1\tRobin_Williams\t1\nBing_Crosby\t1\tEric_Bana\t1\nBing_Crosby\t1\tJohn_Moxley\t1\nBing_Crosby\t1\tLeonid_Kuchma\t5\nBob_Iger\t1\tBrian_Cowen\t1\nBob_Iger\t1\tSteve_Lavin\t6\nBob_Newhart\t1\tMarina_Canetti\t1\nBob_Riley\t1\tJim_Harrick\t1\nBono\t1\tChen_Shui-bian\t2\nBono\t2\tSe_Hyuk_Joo\t1\nBrad_Miller\t1\tChris_Noth\t1\nBrandon_Larson\t1\tJames_Williams\t1\nBrandon_Lloyd\t1\tZach_Parise\t1\nBrandon_Robinson\t1\tMitchell_Garabedian\t1\nBrandon_Webb\t1\tTeddy_Kollek\t1\nBrian_Cowen\t2\tDave_Johnson\t1\nBrian_Griese\t1\tJennie_Garth\t1\nBrock_Berlin\t1\tWendy_Selig\t1\nBrooke_Adams\t1\tGilberto_Rodriguez_Orejuela\t2\nBrooke_Adams\t1\tHussam_Mohammed_Amin\t1\nBruce_Willis\t1\tHeather_Mills\t2\nBuddy_Ryan\t1\tLaurel_Clark\t1\nBuddy_Ryan\t1\tRoy_Romanow\t1\nCalvin_Joseph_Coleman\t1\tEkaterina_Dmitriev\t1\nCalvin_Joseph_Coleman\t1\tVictor_Garber\t1\nCamilla_Parker_Bowles\t2\tMary_Matalin\t1\nCamille_Lewis\t1\tKeith_Lockhart\t1\nCamille_Lewis\t1\tMelissa_Etheridge\t1\nCandice_Bergen\t1\tMasum_Turker\t2\nCandice_Bergen\t2\tJake_Gyllenhaal\t2\nCandice_Bergen\t2\tProspero_Pichay\t1\nCarey_Lowell\t1\tHelio_Castroneves\t1\nCarey_Lowell\t1\tJoshua_Harapko\t1\nCarlo_Ancelotti\t1\tHugh_Miller\t1\nCarlos_Beltran\t1\tShoshannah_Stern\t1\nCarlos_Ghosn\t1\tRenee_Zellweger\t11\nCarlos_Iturgaitz\t1\tFrancisco_Maturana\t1\nCarlos_Manuel_Pruneda\t1\tClaudia_Coslovich\t1\nCarly_Gullickson\t1\tMike_Helton\t2\nCaroline_Dhavernas\t1\tJames_Smith\t1\nCarrie-Anne_Moss\t4\tMike_Richter\t1\nCarson_Daly\t1\tMatthew_During\t1\nCasey_Crowder\t1\tJoe_Nichols\t1\nCasey_Crowder\t1\tLaszlo_Kovacs\t1\nCecilia_Chang\t1\tJeffery_Hendren\t1\nCecilia_Cheung\t1\tKirsten_Clark\t1\nCecilia_Cheung\t1\tSimon_Chalk\t1\nCedric_Benson\t1\tGreg_Hodge\t1\nCesar_Maia\t2\tKarin_Viard\t1\nChang_Jae_On\t1\tOracene_Williams\t1\nChante_Jawan_Mallard\t1\tMuhammad_Ali\t5\nCharla_Moye\t1\tJoe_Mantegna\t1\nCharla_Moye\t1\tRobert_Nillson\t1\nCharles_Cope\t1\tJohn_Franco\t1\nCharley_Armey\t1\tJose_Cevallos\t1\nCharlie_Sheen\t1\tDeece_Eckstein\t1\nCharlie_Sheen\t1\tJanice_Goldfinger\t1\nCharlie_Zaa\t2\tWilliam_Harrison\t1\nCharmaine_Crooks\t1\tEsad_Landzo\t1\nCharmaine_Crooks\t1\tShigeru_Ishiba\t1\nChawki_Armali\t1\tHilmi_Akin_Zorlu\t1\nChea_Sophara\t1\tMuhammad_Ali\t6\nChen_Shui-bian\t2\tSimon_Cowell\t1\nChen_Shui-bian\t2\tWilliam_Nessen\t1\nChen_Shui-bian\t3\tElizabeth_Smart\t1\nChita_Rivera\t2\tJose_Luis_Chilavert\t1\nChris_Andrews\t1\tClaire_De_Gryse\t1\nChris_Andrews\t1\tElinor_Caplan\t2\nChris_Byrd\t1\tDavid_Alpay\t1\nChris_Byrd\t1\tLawrence_Di_Rita\t1\nChris_Noth\t1\tFrank_Sinatra\t1\nChristian_Longo\t2\tDustan_Mohr\t1\nChristiane_Wulff\t1\tPaul_Schrader\t1\nChristine_Ebersole\t2\tVincent_Cianci_Jr\t1\nChristine_Todd_Whitman\t2\tPeter_Rasch\t1\nChristine_Todd_Whitman\t5\tNeil_Goldman\t1\nChristopher_Russell\t1\tRosario_Dawson\t1\nCindy_Taylor\t1\tMelissa_Joan_Hart\t1\nClaire_Danes\t1\tMitchell_Potter\t1\nClaire_De_Gryse\t1\tJim_Parque\t1\nClaire_De_Gryse\t1\tSoenarno\t1\nClaire_Leger\t2\tElin_Nordegren\t2\nColeen_Rowley\t1\tCourtney_Love\t1\nColin_Campbell\t1\tMatt_Walters\t1\nColleen_OClair\t1\tKim_Hong-gul\t1\nColleen_Ryan\t1\tNoah_Wyle\t2\nConchita_Martinez\t1\tMoby\t1\nCora_Cambell\t1\tTodd_Petit\t1\nCourtney_Love\t1\tTsutomu_Takebe\t1\nCristina_Torrens_Valero\t1\tDamarius_Bilbo\t1\nCristina_Torrens_Valero\t1\tEtta_James\t1\nDamarius_Bilbo\t1\tJanis_Ruth_Coulter\t1\nDamarius_Bilbo\t1\tSheila_Taormina\t1\nDan_Ackroyd\t1\tNikki_Teasley\t1\nDan_Bylsma\t1\tJim_Parque\t1\nDan_Bylsma\t1\tScott_Yates\t1\nDaniel_Comisso_Urdaneta\t1\tJean-Marc_de_La_Sabliere\t1\nDaniel_Montgomery\t1\tHassan_Wirajuda\t1\nDaniel_Zelman\t1\tDavid_Surrett\t1\nDaniele_Nardello\t1\tPeter_Hartz\t1\nDariusz_Michalczewski\t1\tJulien_Boutter\t1\nDarvis_Patton\t1\tEstelle_Morris\t1\nDarvis_Patton\t1\tGordon_Lightfoot\t1\nDarvis_Patton\t1\tTammy_Lynn_Michaels\t1\nDave_Johnson\t1\tHank_Aaron\t1\nDavid_Braley\t1\tGerard_Tronche\t1\nDavid_Kelley\t1\tGerald_Calabrese\t1\nDavid_Shayler\t1\tPeter_Mullan\t1\nDavid_Surrett\t1\tOracene_Williams\t1\nDean_Sheremet\t1\tMarc_Racicot\t1\nDenise_van_Outen\t1\tKevin_Satterfield\t1\nDennis_Kozlowski\t2\tJames_Ivory\t1\nDennis_Kozlowski\t2\tRegina_Ip\t1\nDennis_Kozlowski\t2\tWendy_Selig\t1\nDenzel_Washington\t4\tSung_Hong_Choi\t1\nDerrick_Battie\t1\tLisa_Leslie\t1\nDiane_Green\t1\tRuth_Christofferson\t1\nDiane_Green\t1\tTim_Salmon\t1\nDick_Armey\t1\tTom_Glavine\t2\nDirk_Kempthorne\t1\tGeorge_Tenet\t2\nDominique_de_Villepin\t13\tHanns_Schumacher\t1\nDon_Carcieri\t1\tJane_Rooney\t1\nDon_King\t1\tPhilip_Zalewski\t1\nDon_King\t1\tTab_Baldwin\t1\nDon_Lake\t1\tElena_Dementieva\t1\nDonald_Keck\t1\tJulia_Ormond\t1\nDonald_Pettit\t3\tMichel_Kratochvil\t1\nDonald_Rumsfeld\t40\tTamika_Catchings\t1\nDonna_Walker\t1\tPhil_Bredesen\t1\nEddie_Sutton\t1\tGeorge_Maxwell_Richards\t1\nEdith_Masai\t1\tIdi_Amin\t1\nEdward_Belvin\t1\tWilliam_Harrison\t1\nEdward_Egan\t1\tJim_Abbott\t1\nEdward_James_Olmos\t1\tJesse_Harris\t3\nEdward_Seaga\t1\tFatmir_Limaj\t1\nElaine_Chao\t1\tWilbert_Elki_Meza_Majino\t1\nElena_Dementieva\t1\tWilton_Gregory\t1\nElin_Nordegren\t1\tFredric_Seaman\t1\nElin_Nordegren\t2\tGhassan_Elashi\t1\nElinor_Caplan\t1\tHenning_Scherf\t1\nEliott_Spitzer\t1\tJerry_Bruckheimer\t1\nElisabeth_Welch\t1\tFrancisco_Maturana\t1\nElizabeth_Regan\t1\tPierre_Boulanger\t2\nElliott_Mincberg\t1\tGene_Orza\t1\nEnrica_Fico\t1\tErwin_Abdullah\t1\nEric_Dubin\t1\tYang_Pao-yu\t1\nEric_Lindros\t1\tMelvin_Talbert\t1\nEric_Lindros\t1\tSterling_Hitchcock\t1\nEric_Snow\t1\tRobert_McKee\t1\nFabiola_Zuluaga\t1\tRose_Linkins\t1\nFarouk_Kaddoumi\t1\tGina_Gershon\t1\nFarouk_Kaddoumi\t1\tWill_Smith\t2\nFelipe_Perez_Roque\t2\tLloyd_Ward\t2\nFestus_Mogae\t1\tNeil_Moritz\t1\nFilip_De_Winter\t1\tStuart_Townsend\t1\nFilip_De_Winter\t1\tSue_Grafton\t1\nFlavia_Pennetta\t1\tMikulas_Dzurinda\t2\nFlor_Montulo\t1\tSteve_Blankenship\t1\nFloyd_Mayweather\t1\tLou_Lang\t1\nFran_Drescher\t2\tWilliam_Martin\t1\nFrancis_Ricciardone\t1\tGene_Sauers\t1\nFranco_Frattini\t1\tGholamreza_Aghazadeh\t1\nFrank_Sinatra\t1\tVicente_Fox\t23\nFrank_Van_Ecke\t1\tGary_Gero\t1\nFred_Durst\t1\tPatricia_Phillips\t1\nFred_Durst\t1\tRose_Linkins\t1\nFred_Durst\t1\tTara_Reid\t1\nFrederick_Madden\t1\tLuis_Ernesto_Derbez_Bautista\t3\nGary_Leon_Ridgway\t1\tLisa_Raymond\t1\nGary_Leon_Ridgway\t1\tPhillip_Seymor_Hoffmann\t1\nGary_Sayler\t1\tNila_Ferran\t1\nGeorge_Foreman\t1\tGeorge_Lucas\t1\nGeorgia_Giddings\t1\tJohnny_Carson\t1\nGerard_Tronche\t1\tJana_Pittman\t1\nGerhard_Schroeder\t3\tIzzat_Ibrahim\t1\nGhassan_Elashi\t1\tJesse_Ventura\t2\nGhassan_Elashi\t1\tPeter_Rasch\t1\nGideon_Yago\t1\tRobert_Gordon_Card\t1\nGina_Gershon\t1\tStacey_Dales-Schuman\t1\nGlenn_Tilton\t1\tVictor_Hanescu\t1\nGordon_Brown\t12\tMonica_Gabrielle\t1\nGordon_Campbell\t1\tPierre_Boulanger\t2\nGordon_Lightfoot\t1\tJanez_Drnovsek\t1\nGrace_Kelly\t1\tLeslie_Moonves\t1\nGreg_Kinnear\t1\tTim_Salmon\t1\nGregory_Geoffroy\t2\tPaul_Burrell\t11\nGregory_Hines\t1\tMary_Frances_Seiter\t1\nGregory_Peck\t1\tJohn_Lawrence\t1\nGregory_Peck\t1\tJohn_Lynch\t1\nGuillaume_Cannet\t1\tMark_Mulder\t1\nGuillaume_Cannet\t1\tMohammad_Khatami\t5\nGunilla_Backman\t1\tJennifer_Granholm\t1\nGunilla_Backman\t1\tRay_Sherman\t1\nGustavo_Franco\t1\tMelissa_Joan_Hart\t1\nGwendal_Peizerat\t2\tJoe_Nichols\t2\nGwyneth_Paltrow\t5\tMike_Richter\t1\nHalbert_Fillinger\t1\tRachel_Hunter\t1\nHank_Aaron\t1\tRon_Howard\t2\nHarland_Braun\t1\tLisa_Leslie\t1\nHarvey_Fierstein\t1\tMike_Slive\t1\nHelene_Eksterowicz\t1\tMario_Lemieux\t1\nHelio_Castroneves\t1\tJoxel_Garcia\t1\nHermann_Maier\t1\tRobin_Williams\t1\nHoward_Smith\t2\tTim_Pawlenty\t1\nHubie_Brown\t1\tYoon_Jeong_Cho\t1\nHunter_Bates\t1\tJames_Williams\t1\nHussam_Mohammed_Amin\t1\tMike_Smith\t1\nHussam_Mohammed_Amin\t1\tPrincess_Hisako\t1\nHutomo_Mandala_Putra\t1\tThomas_Mesereau_Jr\t1\nIban_Mayo\t2\tYves_Brodeur\t1\nImad_Moustapha\t1\tWill_Ofenheusle\t1\nIon_Tiriac\t1\tLawrence_Roberts\t1\nIrina_Yatchenko\t1\tPedro_Solbes\t4\nIsabelle_Huppert\t1\tPedro_Solbes\t1\nIsaiah_Washington\t2\tLee_Baca\t1\nIslam_Karimov\t1\tShireen_Amir_Begum\t1\nIsmail_Cem\t1\tSeth_Gorney\t1\nIsmail_Merchant\t1\tJoseph_Deiss\t1\nIvan_Lee\t1\tPaul-Henri_Mathieu\t3\nIzzat_Ibrahim\t1\tTina_Sinatra\t1\nJack_Welch\t1\tRick_Dinse\t2\nJack_Welch\t1\tWilliam_Umbach\t1\nJalal_Talabani\t1\tJean-Marc_Olive\t1\nJames_Franco\t2\tLokendra_Bahadur_Chand\t1\nJames_Ivory\t1\tPaul_Martin\t7\nJames_Lockhart\t1\tLisa_Girman\t1\nJan_Paul_Miller\t1\tLynne_Slepian\t1\nJan_Ullrich\t1\tWilliam_Rehnquist\t1\nJanet_Chandler\t1\tZoe_Ball\t1\nJanis_Ruth_Coulter\t1\tLaurent_Woulzy\t1\nJason_Clermont\t1\tVladimir_Ustinov\t1\nJason_Kapono\t1\tPatricia_Medina\t1\nJason_Mewes\t1\tWill_Smith\t1\nJason_White\t1\tWolfgang_Clement\t1\nJean-Pierre_Raffarin\t2\tRobert_Horan\t2\nJean-Pierre_Raffarin\t6\tPaul_Kariya\t1\nJeane_Kirkpatrick\t1\tRichard_Carl\t1\nJeane_Kirkpatrick\t1\tTamika_Catchings\t1\nJeanne_Anne_Schroeder\t1\tYoriko_Kawaguchi\t13\nJeannette_Biedermann\t1\tRoy_Moore\t6\nJeffrey_Pfeffer\t1\tTerry_Lynn_Barton\t1\nJen_Bice\t1\tLily_Tomlin\t2\nJen_Bice\t1\tMartha_Burk\t2\nJenna_Elfman\t1\tMack_Brown\t1\nJenna_Elfman\t1\tSteffi_Graf\t5\nJenna_Elfman\t1\tWillie_Wilson\t1\nJennifer_Keller\t4\tPaula_Abdul\t1\nJennifer_Thompson\t1\tSerge_Melac\t1\nJeremy_Greenstock\t7\tJim_Flaherty\t1\nJerry_Bruckheimer\t1\tJohn_Blaney\t1\nJerry_Jones\t1\tRobert_Woody_Johnson\t1\nJesse_Ventura\t1\tReggie_Lewis\t1\nJesse_Ventura\t3\tRoy_Romanow\t1\nJessica_Biel\t1\tJim_Freudenberg\t1\nJim_Calhoun\t1\tPerry_Farrell\t1\nJim_Cantalupo\t1\tJoe_Pantoliano\t1\nJim_Hahn\t4\tVaclav_Havel\t3\nJim_Hahn\t4\tVicente_Fernandez\t2\nJim_Letten\t1\tVictor_Hanescu\t1\nJoaquin_Phoenix\t1\tKajsa_Bergqvist\t1\nJoe_Pantoliano\t1\tRobin_Tunney\t1\nJohn_Belushi\t1\tMahima_Chaudhari\t1\nJohn_Blaney\t2\tSam_Brownback\t1\nJohn_Franco\t1\tTeri_Files\t1\nJohn_Garamendi\t2\tPeter_Goldmark\t1\nJohn_Kerr\t1\tSusan_Whelan\t1\nJohn_Lynch\t1\tStefano_Gabbana\t1\nJohn_Mayer\t3\tMiguel_Jimenez\t1\nJohn_McCallum\t1\tMark_Andrew\t1\nJohn_Rowland\t2\tRobert_Lee_Yates_Jr\t1\nJohn_Walsh\t2\tPaul_Hogan\t2\nJohnny_Depp\t2\tScott_Dickson\t1\nJorge_Arce\t1\tSamuel_Waksal\t3\nJorge_Quiroga\t1\tRamona_Rispton\t1\nJorge_Quiroga\t1\tSurakait_Sathirathai\t1\nJose_Bove\t1\tMichalis_Chrisohoides\t1\nJose_Bove\t1\tPrincess_Diana\t1\nJose_Luis_Chilavert\t1\tSofyan_Dawood\t1\nJuan_Carlos\t1\tPatricia_Phillips\t1\nJudy_Dean\t1\tKristanna_Loken\t1\nJulia_Ormond\t1\tRandy_Dryer\t1\nJulio_Cesar_Chavez\t1\tMarc_Racicot\t1\nKaoru_Hasuike\t1\tTom_Foy\t1\nKaren_Allen\t1\tLeRoy_Millette_Jr\t1\nKaren_Allen\t1\tRick_Bragg\t1\nKarl-Heinz_Rummenigge\t1\tWayne_Newton\t1\nKatie_Boone\t1\tTerri_Clark\t1\nKeith_Osik\t1\tRobert_Nardelli\t1\nKent_Robinson\t1\tMarquier_Montano_Contreras\t1\nKevin_Gil\t1\tMary_Bono\t1\nKevin_Hearn\t1\tTodd_Haynes\t3\nKhader_Rashid_Rahim\t1\tMartha_Burk\t1\nKhader_Rashid_Rahim\t1\tMichael_Arif\t1\nKim_Yong-il\t2\tTerri_Clark\t1\nKoichi_Tanaka\t1\tSterling_Hitchcock\t1\nKosuke_Kitajima\t2\tQazi_Hussain_Ahmed\t1\nKristin_Davis\t3\tSteven_Van_Zandt\t1\nKurt_Hellstrom\t1\tManuel_Pellegrini\t1\nLaszlo_Kovacs\t1\tMichael_Arif\t1\nLaura_Elena_Harring\t1\tRowan_Williams\t1\nLaurel_Clark\t1\tRahul_Dravid\t1\nLawrence_Di_Rita\t1\tTim_Allen\t2\nLawrence_Roberts\t1\tMalik_Mahmud\t1\nLea_Fastow\t1\tTodd_Parrott\t1\nLee_Baca\t1\tPat_Summerall\t1\nLeisel_Jones\t1\tSteven_Hatfill\t1\nLena_Katina\t1\tQueen_Noor\t1\nLeonard_Glick\t1\tTerrence_Trammell\t1\nLeonard_Schrank\t1\tOzzy_Osbourne\t1\nLeslie_Wiser_Jr\t1\tMireya_Moscoso\t1\nLisa_Murkowski\t1\tSvetlana_Belousova\t1\nLisa_Raymond\t2\tRobert_Redford\t7\nLudwig_Ovalle\t1\tMauricio_Macri\t1\nLuis_Pujols\t1\tRohman_al-Ghozi\t1\nMakhdoom_Amin_Fahim\t1\tTang_Jiaxuan\t5\nMalcolm_Wild\t1\tOwen_Wilson\t2\nMamdouh_Habib\t1\tPolona_Bas\t1\nManijeh_Hekmat\t1\tRichard_Carl\t1\nManijeh_Hekmat\t1\tSe_Hyuk_Joo\t1\nMarc_Anthony\t1\tVince_Vaughan\t1\nMargie_Puente\t1\tRafiq_Hariri\t1\nMaria_Callas\t1\tMelchor_Cob_Castro\t1\nMario_Lemieux\t1\tPedro_Solbes\t4\nMarissa_Jaret_Winokur\t1\tVincent_Sombrotto\t1\nMark_Dacey\t2\tRuss_Ortiz\t1\nMark_Martin\t1\tVladimir_Spidla\t2\nMark_Mishkin\t1\tNick_Reilly\t1\nMark_Mishkin\t1\tShawn_Marion\t1\nMarricia_Tate\t1\tRita_Moreno\t2\nMary_Landrieu\t1\tRobert_Gordon_Card\t1\nMaryn_McKenna\t1\tShawn_Marion\t1\nMaryn_McKenna\t1\tTim_Pawlenty\t1\nMasum_Turker\t1\tNorman_Mineta\t1\nMatt_Dillon\t1\tRegina_Ip\t1\nMatthias_Sammer\t1\tRobbie_Naish\t1\nMelissa_Etheridge\t2\tTeri_Files\t1\nMichael_J_Sheehan\t1\tMichael_Smith_Foster\t1\nMichael_Linscott\t1\tRobin_Tunney\t1\nMichael_Phelps\t5\tPat_Burns\t2\nMichael_Sheehan\t1\tVincent_Cianci_Jr\t1\nMichael_Smith_Foster\t1\tRupert_Murdoch\t1\nMichelle_Yeoh\t5\tShoshannah_Stern\t1\nMiguel_Aldana_Ibarra\t1\tMikhail_Gorbachev\t1\nMiguel_Jimenez\t1\tParis_Hilton\t1\nMike_Helton\t2\tReggie_Lewis\t1\nMike_Leach\t1\tThomas_Haeggstroem\t1\nMike_Maroth\t1\tStefan_Holm\t1\nMike_Slive\t1\tPaul_Sarbanes\t3\nMikhail_Youzhny\t2\tVince_Dooley\t1\nMikulas_Dzurinda\t2\tShirley_Jones\t1\nMiles_Stewart\t1\tYasushi_Akashi\t1\nMirela_Manjani\t1\tRobert_Woody_Johnson\t1\nMitchell_McLaughlin\t1\tTimothy_Goebel\t1\nMohammed_Baqir_al-Hakim\t3\tNeil_Goldman\t1\nMona_Ayoub\t1\tTodd_Parrott\t1\nMufti_Mohammad_Syed\t1\tRalf_Schumacher\t4\nNathalie_Gagnon\t1\tRupert_Grint\t2\nNeil_Goldman\t1\tNiall_Connolly\t1\nNick_Reilly\t1\tTassos_Papadopoulos\t1\nNicolas_Sarkozy\t1\tShanna_Zolman\t1\nNorm_Coleman\t2\tOsmond_Smith\t1\nNorman_Mailer\t1\tSteven_Curtis_Chapman\t1\nOliver_Neuville\t1\tShawn_Marion\t1\nPark_Jie-won\t1\tSteve_Patterson\t1\nPatty_Sheehan\t1\tWilliam_Nessen\t1\nPaul_Hogan\t1\tThomas_Watjen\t1\nPaul_Hogan\t1\tWilliam_Martin\t1\nPaul_Kariya\t1\tShannon_OBrien\t2\nPaul_Martin\t1\tStephanie_Cohen_Aloro\t1\nPaul_Michael_Daniels\t1\tSimon_Chalk\t1\nPaul_Schrader\t1\tTeri_Files\t1\nPaula_Prentiss\t1\tThomas_Mesereau_Jr\t1\nPedro_Martinez\t1\tStephen_Glassroth\t1\nPeter_Fitzgerald\t1\tZelma_Novelo\t1\nPhil_Jackson\t1\tRichard_Penniman\t1\nPhillipe_Comtois\t1\tRichard_Carl\t1\nPierre_Boulanger\t1\tShania_Twain\t1\nPrincess_Diana\t1\tSteffi_Graf\t2\nPriscilla_Presley\t1\tSergey_Lavrov\t5\nPriscilla_Presley\t1\tVictoria_Beckham\t3\nRichard_Butler\t1\tVictoria_Beckham\t1\nRichard_Pennington\t1\tStacy_Nelson\t1\nRick_Caruso\t1\tSteve_Wariner\t1\nRick_Perry\t6\tSvetlana_Koroleva\t1\nRicky_Cottrill\t1\tSananda_Maitreya\t1\nRobert_Flodquist\t1\tWinona_Ryder\t11\nRod_Stewart\t1\tTom_Scully\t1\nRod_Stewart\t3\tSe_Hyuk_Joo\t1\nRod_Thorn\t1\tYves_Brodeur\t1\nRon_Howard\t1\tTim_Pawlenty\t1\nRonaldo_Luis_Nazario_de_Lima\t1\tSamuel_Waksal\t2\nSami_Al-Arian\t1\tTommy_Lasorda\t1\nSarah_Canale\t1\tSteven_Curtis_Chapman\t1\nSasha_Cohen\t1\tValery_Giscard_dEstaing\t5\nScott_Dalton\t1\tTamara_Mowry\t1\nSergio_Castellitto\t1\tSteve_Lavin\t5\nSeth_Gorney\t1\tWilton_Gregory\t1\nShane_Mosley\t1\tStacey_Dales-Schuman\t1\nSheila_Taormina\t1\tStephan_Eberharter\t1\nStefano_Gabbana\t1\tTang_Jiaxuan\t3\nSteve_Wariner\t1\tToshi_Izawa\t1\nSteve_Zahn\t1\tTab_Baldwin\t1\nSusan_Whelan\t1\tWolfgang_Schneiderhan\t1\nTakeo_Fukui\t1\tWill_Ofenheusle\t1\nTamara_Mowry\t1\tZach_Parise\t1\nTatiana_Kennedy_Schlossberg\t1\tThomas_Watjen\t1\nTodd_Petit\t1\tVicente_Fernandez\t3\n"
  },
  {
    "path": "lfw/eval_lfw.py",
    "content": "import argparse\nimport os\nimport os.path as osp\n\nimport torch\nimport torchvision\nfrom torchvision import models\nimport torch.nn as nn\nimport torch.utils.data\nimport torchvision.transforms as transforms\nimport torchvision.datasets as datasets\nfrom torch.autograd import Variable\nimport torch.nn.functional as F\n\nimport tqdm\nimport numpy as np\nimport sklearn.metrics\nfrom sklearn import metrics\nfrom scipy.optimize import brentq\nfrom scipy import interpolate\nimport matplotlib\nmatplotlib.use('Agg')\nimport matplotlib.pyplot as plt\n\nhere = osp.dirname(osp.abspath(__file__)) # output folder is located here\nroot_dir,_ = osp.split(here)\nimport sys\nsys.path.append(root_dir)\n\nimport models\nimport utils\nimport data_loader\n\n\n'''\nEvaluate a network on the LFW verification task\n===============================================\nExample usage:\n# Resnet 101 on 10 folds of LFW\n    python lfw/eval_lfw.py -e lfw_eval_resnet101 --model_type resnet101 --fold 10 -m BEST_MODEL_PATH\n\n'''\n\n\n\ndef main():\n    parser = argparse.ArgumentParser()\n    parser.add_argument('-e', '--exp_name', default='lfw_eval')\n    parser.add_argument('-g', '--gpu', type=int, default=0)\n    parser.add_argument('-d', '--dataset_path', \n                        default='/srv/data1/arunirc/datasets/lfw-deepfunneled')\n    parser.add_argument('--fold', type=int, default=0, choices=[0,10])\n    parser.add_argument('--batch_size', type=int, default=100)\n    parser.add_argument('-m', '--model_path', default=None, required=True,\n                        help='Path to pre-trained model')\n    parser.add_argument('--model_type', default='resnet50',\n                        choices=['resnet50', 'resnet101', 'resnet101-512d'])\n    \n    args = parser.parse_args()\n\n\n    # CUDA setup\n    os.environ['CUDA_VISIBLE_DEVICES'] = str(args.gpu)\n    cuda = torch.cuda.is_available()\n    torch.manual_seed(1337)\n    if cuda:\n        torch.cuda.manual_seed(1337)\n        torch.backends.cudnn.enabled = True\n        torch.backends.cudnn.benchmark = True # enable if all images are same size    \n\n    if args.fold == 0:\n        pairs_path = './lfw/data/pairsDevTest.txt'\n    else:\n        pairs_path = './lfw/data/pairs.txt'\n\n    # -----------------------------------------------------------------------------\n    # 1. Dataset\n    # -----------------------------------------------------------------------------\n    file_ext = 'jpg' # observe, no '.' before jpg\n    num_class = 8631\n\n    pairs = utils.read_pairs(pairs_path)\n    path_list, issame_list = utils.get_paths(args.dataset_path, pairs, file_ext)\n\n    # Define data transforms\n    RGB_MEAN = [ 0.485, 0.456, 0.406 ]\n    RGB_STD = [ 0.229, 0.224, 0.225 ]\n    test_transform = transforms.Compose([\n        transforms.Scale((250,250)),  # make 250x250\n        transforms.CenterCrop(150),   # then take 150x150 center crop\n        transforms.Scale((224,224)),  # resized to the network's required input size\n        transforms.ToTensor(),\n        transforms.Normalize(mean = RGB_MEAN,\n                             std = RGB_STD),\n    ])\n\n    # Create data loader\n    test_loader = torch.utils.data.DataLoader(\n                        data_loader.LFWDataset(\n                        path_list, issame_list, test_transform), \n                        batch_size=args.batch_size, shuffle=False )\n\n\n    # -----------------------------------------------------------------------------\n    # 2. Model\n    # -----------------------------------------------------------------------------\n    if args.model_type == 'resnet50':\n        model = torchvision.models.resnet50(pretrained=False)\n        model.fc = torch.nn.Linear(2048, num_class)\n    elif args.model_type == 'resnet101':\n        model = torchvision.models.resnet101(pretrained=False)\n        model.fc = torch.nn.Linear(2048, num_class)\n    elif args.model_type == 'resnet101-512d':\n        model = torchvision.models.resnet101(pretrained=False)\n        layers = []\n        layers.append(torch.nn.Linear(2048, 512))\n        layers.append(torch.nn.Linear(512, num_class))\n        model.fc = torch.nn.Sequential(*layers)\n    else:\n        raise NotImplementedError\n    \n    checkpoint = torch.load(args.model_path)       \n\n    if checkpoint['arch'] == 'DataParallel':\n        # if we trained and saved our model using DataParallel\n        model = torch.nn.DataParallel(model, device_ids=[0, 1, 2, 3, 4])\n        model.load_state_dict(checkpoint['model_state_dict'])\n        model = model.module # get network module from inside its DataParallel wrapper\n    else:\n        model.load_state_dict(checkpoint['model_state_dict'])\n\n    if cuda:\n        model = model.cuda()\n\n    # Convert the trained network into a \"feature extractor\"\n    feature_map = list(model.children())\n    if args.model_type == 'resnet101-512d':\n        model.eval()\n        extractor = model\n        extractor.fc = nn.Sequential(extractor.fc[0])\n    else: \n        feature_map.pop()\n        extractor = nn.Sequential(*feature_map)\n    \n    extractor.eval() # set to evaluation mode (fixes BatchNorm, dropout, etc.)\n\n\n    # -----------------------------------------------------------------------------\n    # 3. Feature extraction\n    # -----------------------------------------------------------------------------\n    features = []\n\n    for batch_idx, images in tqdm.tqdm(enumerate(test_loader), \n                                        total=len(test_loader), \n                                        desc='Extracting features'): \n        x = Variable(images, volatile=True) # test-time memory conservation\n        if cuda:\n            x = x.cuda()\n        feat = extractor(x)\n        if cuda:\n            feat = feat.data.cpu()\n        else:\n            feat = feat.data\n        features.append(feat)\n\n    features = torch.stack(features)\n    sz = features.size()\n    features = features.view(sz[0]*sz[1], sz[2])\n    features = F.normalize(features, p=2, dim=1) # L2-normalize\n    # TODO - cache features\n\n\n    # -----------------------------------------------------------------------------\n    # 4. Verification\n    # -----------------------------------------------------------------------------\n    num_feat = features.size()[0]\n    feat_pair1 = features[np.arange(0,num_feat,2),:]\n    feat_pair2 = features[np.arange(1,num_feat,2),:]\n    feat_dist = (feat_pair1 - feat_pair2).norm(p=2, dim=1)\n    feat_dist = feat_dist.numpy()\n\n    # Eval metrics\n    scores = -feat_dist\n    gt = np.asarray(issame_list)\n       \n    if args.fold == 0:\n        fig_path = osp.join(here, \n                args.exp_name + '_' + args.model_type + '_lfw_roc_devTest.png')\n        roc_auc = sklearn.metrics.roc_auc_score(gt, scores)\n        fpr, tpr, thresholds = sklearn.metrics.roc_curve(gt, scores)\n        print 'ROC-AUC: %.04f' % roc_auc\n        # Plot and save ROC curve\n        fig = plt.figure()\n        plt.title('ROC - lfw dev-test')\n        plt.plot(fpr, tpr, lw=2, label='ROC (auc = %0.4f)' % roc_auc)\n        plt.xlim([0.0, 1.0])\n        plt.ylim([0.0, 1.05])\n        plt.grid()\n        plt.xlabel('False Positive Rate')\n        plt.ylabel('True Positive Rate')\n        plt.legend(loc='lower right')\n        plt.tight_layout()\n    else:\n        # 10 fold\n        fold_size = 600 # 600 pairs in each fold\n        roc_auc = np.zeros(10)\n        roc_eer = np.zeros(10)\n\n        fig = plt.figure()\n        plt.xlim([0.0, 1.0])\n        plt.ylim([0.0, 1.05])\n        plt.grid()\n        plt.xlabel('False Positive Rate')\n        plt.ylabel('True Positive Rate')\n\n        for i in tqdm.tqdm(range(10)):\n            start = i * fold_size\n            end = (i+1) * fold_size\n            scores_fold = scores[start:end]\n            gt_fold = gt[start:end]\n            roc_auc[i] = sklearn.metrics.roc_auc_score(gt_fold, scores_fold)\n            fpr, tpr, _ = sklearn.metrics.roc_curve(gt_fold, scores_fold)\n            # EER calc: https://yangcha.github.io/EER-ROC/\n            roc_eer[i] = brentq(\n                            lambda x: 1. - x - interpolate.interp1d(fpr, tpr)(x), 0., 1.)\n            plt.plot(fpr, tpr, alpha=0.4, \n                    lw=2, color='darkgreen',\n                    label='ROC(auc=%0.4f, eer=%0.4f)' % (roc_auc[i], roc_eer[i]) )\n\n        plt.title( 'AUC: %0.4f +/- %0.4f, EER: %0.4f +/- %0.4f' % \n                    (np.mean(roc_auc), np.std(roc_auc),\n                     np.mean(roc_eer), np.std(roc_eer)) )\n        plt.tight_layout()\n\n        fig_path = osp.join(here, \n                args.exp_name + '_' + args.model_type + '_lfw_roc_10fold.png')\n        \n\n    plt.savefig(fig_path, bbox_inches='tight')\n    print 'ROC curve saved at: ' + fig_path\n\n\n\n\n\n\nif __name__ == '__main__':\n    main()\n\n"
  },
  {
    "path": "models.py",
    "content": "\n# import fcn\nimport os.path as osp\nimport numpy as np\nimport torch\nimport torch.nn as nn\nimport torch.nn.functional as F\n\n\n\nclass NormFeat(nn.Module):\n    ''' L2 normalization of features '''\n    def __init__(self, scale_factor=1.0):\n        super(NormFeat, self).__init__()\n        self.scale_factor = scale_factor\n\n    def forward(self, input):\n        return self.scale_factor * F.normalize(input, p=2, dim=1)\n\n\nclass ScaleFeat(nn.Module):\n# https://discuss.pytorch.org/t/is-scale-layer-available-in-pytorch/7954/6?u=arunirc\n    def __init__(self, scale_factor=50.0):\n        super().__init__()\n        self.scale = scale_factor\n\n    def forward(self, input):\n        return input * self.scale\n\n\n# https://github.com/shelhamer/fcn.berkeleyvision.org/blob/master/surgery.py\ndef get_upsampling_weight(in_channels, out_channels, kernel_size):\n    \"\"\"Make a 2D bilinear kernel suitable for upsampling\"\"\"\n    factor = (kernel_size + 1) // 2\n    if kernel_size % 2 == 1:\n        center = factor - 1\n    else:\n        center = factor - 0.5\n    og = np.ogrid[:kernel_size, :kernel_size]\n    filt = (1 - abs(og[0] - center) / factor) * \\\n           (1 - abs(og[1] - center) / factor)\n    weight = np.zeros((in_channels, out_channels, kernel_size, kernel_size),\n                      dtype=np.float64)\n    weight[range(in_channels), range(out_channels), :, :] = filt\n    return torch.from_numpy(weight).float()\n\n\n\nclass FCN32sColor(nn.Module):\n\n    def __init__(self, n_class=32, bin_type='one-hot', batch_norm=True):\n        super(FCN32sColor, self).__init__()\n        self.n_class = n_class\n        self.bin_type = bin_type\n        self.batch_norm = batch_norm\n\n        # conv1\n        self.conv1_1 = nn.Conv2d(1, 64, 3, padding=100)\n        self.relu1_1 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv1_1_bn = nn.BatchNorm2d(64)\n        self.conv1_2 = nn.Conv2d(64, 64, 3, padding=1)\n        self.relu1_2 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv1_2_bn = nn.BatchNorm2d(64)\n        self.pool1 = nn.MaxPool2d(2, stride=2, ceil_mode=True)  # 1/2\n\n        # conv2\n        self.conv2_1 = nn.Conv2d(64, 128, 3, padding=1)\n        self.relu2_1 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv2_1_bn = nn.BatchNorm2d(128)\n        self.conv2_2 = nn.Conv2d(128, 128, 3, padding=1)\n        self.relu2_2 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv2_2_bn = nn.BatchNorm2d(128)\n        self.pool2 = nn.MaxPool2d(2, stride=2, ceil_mode=True)  # 1/4\n\n        # conv3\n        self.conv3_1 = nn.Conv2d(128, 256, 3, padding=1)\n        self.relu3_1 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv3_1_bn = nn.BatchNorm2d(256)\n        self.conv3_2 = nn.Conv2d(256, 256, 3, padding=1)\n        self.relu3_2 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv3_2_bn = nn.BatchNorm2d(256)\n        self.conv3_3 = nn.Conv2d(256, 256, 3, padding=1)\n        self.relu3_3 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv3_3_bn = nn.BatchNorm2d(256)\n        self.pool3 = nn.MaxPool2d(2, stride=2, ceil_mode=True)  # 1/8\n\n        # conv4\n        self.conv4_1 = nn.Conv2d(256, 512, 3, padding=1)\n        self.relu4_1 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv4_1_bn = nn.BatchNorm2d(512)\n        self.conv4_2 = nn.Conv2d(512, 512, 3, padding=1)\n        self.relu4_2 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv4_2_bn = nn.BatchNorm2d(512)\n        self.conv4_3 = nn.Conv2d(512, 512, 3, padding=1)\n        self.relu4_3 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv4_3_bn = nn.BatchNorm2d(512)\n        self.pool4 = nn.MaxPool2d(2, stride=2, ceil_mode=True)  # 1/16\n\n        # conv5\n        self.conv5_1 = nn.Conv2d(512, 512, 3, padding=1)\n        self.relu5_1 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv5_1_bn = nn.BatchNorm2d(512)\n        self.conv5_2 = nn.Conv2d(512, 512, 3, padding=1)\n        self.relu5_2 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv5_2_bn = nn.BatchNorm2d(512)\n        self.conv5_3 = nn.Conv2d(512, 512, 3, padding=1)\n        self.relu5_3 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv5_3_bn = nn.BatchNorm2d(512)\n        self.pool5 = nn.MaxPool2d(2, stride=2, ceil_mode=True)  # 1/32\n\n        # fc6\n        self.fc6 = nn.Conv2d(512, 4096, 7)\n        self.relu6 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.fc6_bn = nn.BatchNorm2d(4096)\n        self.drop6 = nn.Dropout2d()\n\n        # fc7\n        self.fc7 = nn.Conv2d(4096, 4096, 1)\n        self.relu7 = nn.ReLU(inplace=True)\n        self.fc7_bn = nn.BatchNorm2d(4096)\n        self.drop7 = nn.Dropout2d()\n\n        if bin_type == 'one-hot':\n            # NOTE: *two* output prediction maps for hue and chroma\n            # TODO - not implemented error should be raised for this!\n            self.score_fr_hue = nn.Conv2d(4096, n_class, 1)\n            self.upscore_hue = nn.ConvTranspose2d(n_class, n_class, 64, stride=32,\n                                              bias=False)\n            self.score_fr_chroma = nn.Conv2d(4096, n_class, 1)\n            self.upscore_chroma = nn.ConvTranspose2d(n_class, n_class, 64, stride=32,\n                                              bias=False)\n            self.upscore_hue.weight.requires_grad = False\n            self.upscore_chroma.weight.requires_grad = False\n        elif bin_type == 'soft':\n            self.score_fr = nn.Conv2d(4096, n_class, 1)\n            self.upscore = nn.ConvTranspose2d(n_class, n_class, 64, stride=32,\n                                              bias=False)            \n            self.upscore.weight.requires_grad = False # fix bilinear upsampler\n\n        self._initialize_weights()\n        # TODO - init from pre-trained network\n\n        \n\n    def _initialize_weights(self):\n        for m in self.modules():\n            if isinstance(m, nn.Conv2d):\n                pass # leave the default PyTorch init\n            if isinstance(m, nn.ConvTranspose2d):\n                assert m.kernel_size[0] == m.kernel_size[1]\n                initial_weight = get_upsampling_weight(\n                    m.in_channels, m.out_channels, m.kernel_size[0])\n                m.weight.data.copy_(initial_weight)\n\n\n    def forward(self, x):\n        h = x\n        h = self.conv1_1(h)\n        if self.batch_norm:\n            h = self.conv1_1_bn(h)\n        h = self.relu1_1(h)\n        h = self.conv1_2(h)\n        if self.batch_norm:\n            h = self.conv1_2_bn(h)\n        h = self.relu1_2(h)\n        h = self.pool1(h)\n\n        if self.batch_norm:\n            h = self.relu2_1(self.conv2_1_bn(self.conv2_1(h)))\n        else:\n            h = self.relu2_1(self.conv2_1(h))\n        if self.batch_norm:\n            h = self.relu2_2(self.conv2_2_bn(self.conv2_2(h)))\n        else:\n            h = self.relu2_2(self.conv2_2_bn(self.conv2_2(h)))\n        h = self.pool2(h)\n\n        if self.batch_norm:\n            h = self.relu3_1(self.conv3_1_bn(self.conv3_1(h)))\n        else:\n            h = self.relu3_1(self.conv3_1(h))\n        if self.batch_norm:\n            h = self.relu3_2(self.conv3_2_bn(self.conv3_2(h)))\n        else:\n            h = self.relu3_2(self.conv3_2(h))\n        if self.batch_norm:\n            h = self.relu3_3(self.conv3_3_bn(self.conv3_3(h)))\n        else:\n            h = self.relu3_3(self.conv3_3(h))\n        h = self.pool3(h)\n\n        if self.batch_norm:\n            h = self.relu4_1(self.conv4_1_bn(self.conv4_1(h)))\n        else:\n            h = self.relu4_1(self.conv4_1(h))\n        if self.batch_norm:\n            h = self.relu4_2(self.conv4_2_bn(self.conv4_2(h)))\n        else:\n            h = self.relu4_2(self.conv4_2(h))\n        if self.batch_norm:\n            h = self.relu4_3(self.conv4_3_bn(self.conv4_3(h)))\n        else:\n            h = self.relu4_3(self.conv4_3(h))\n        h = self.pool4(h)\n\n        if self.batch_norm:\n            h = self.relu5_1(self.conv5_1_bn(self.conv5_1(h)))\n        else:\n            h = self.relu5_1(self.conv5_1(h))\n        if self.batch_norm:\n            h = self.relu5_2(self.conv5_2_bn(self.conv5_2(h)))\n        else:\n            h = self.relu5_2(self.conv5_2(h))\n        if self.batch_norm:\n            h = self.relu5_3(self.conv5_3_bn(self.conv5_3(h)))\n        else:\n            h = self.relu5_3(self.conv5_3(h))\n        h = self.pool5(h)\n\n        if self.batch_norm:\n            h = self.relu6(self.fc6_bn(self.fc6(h)))\n        else:\n            h = self.relu6(self.fc6(h))\n        h = self.drop6(h)\n\n        if self.batch_norm:\n            h = self.relu7(self.fc7_bn(self.fc7(h)))\n        else:\n            h = self.relu7(self.fc7(h))\n        h = self.drop7(h)\n\n        if self.bin_type == 'one-hot':\n            # hue prediction map\n            h_hue = self.score_fr_hue(h)\n            h_hue = self.upscore_hue(h_hue)\n            h_hue = h_hue[:, :, 19:19 + x.size()[2], 19:19 + x.size()[3]].contiguous()\n\n            # chroma prediction map\n            h_chroma = self.score_fr_chroma(h)\n            h_chroma = self.upscore_chroma(h_chroma)\n            h_chroma = h_chroma[:, :, 19:19 + x.size()[2], 19:19 + x.size()[3]].contiguous()\n            h = (h_hue, h_chroma)\n\n        elif self.bin_type == 'soft':\n            h = self.score_fr(h)\n            h = self.upscore(h)\n            h = h[:, :, 19:19 + x.size()[2], 19:19 + x.size()[3]].contiguous()\n\n        return h\n\n\n\n\nclass FCN16sColor(nn.Module):\n\n    def __init__(self, n_class=32, bin_type='one-hot', batch_norm=True):\n        super(FCN16sColor, self).__init__()\n        self.n_class = n_class\n        self.bin_type = bin_type\n        self.batch_norm = batch_norm\n\n        # conv1\n        self.conv1_1 = nn.Conv2d(1, 64, 3, padding=100)\n        self.relu1_1 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv1_1_bn = nn.BatchNorm2d(64)\n        self.conv1_2 = nn.Conv2d(64, 64, 3, padding=1)\n        self.relu1_2 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv1_2_bn = nn.BatchNorm2d(64)\n        self.pool1 = nn.MaxPool2d(2, stride=2, ceil_mode=True)  # 1/2\n\n        # conv2\n        self.conv2_1 = nn.Conv2d(64, 128, 3, padding=1)\n        self.relu2_1 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv2_1_bn = nn.BatchNorm2d(128)\n        self.conv2_2 = nn.Conv2d(128, 128, 3, padding=1)\n        self.relu2_2 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv2_2_bn = nn.BatchNorm2d(128)\n        self.pool2 = nn.MaxPool2d(2, stride=2, ceil_mode=True)  # 1/4\n\n        # conv3\n        self.conv3_1 = nn.Conv2d(128, 256, 3, padding=1)\n        self.relu3_1 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv3_1_bn = nn.BatchNorm2d(256)\n        self.conv3_2 = nn.Conv2d(256, 256, 3, padding=1)\n        self.relu3_2 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv3_2_bn = nn.BatchNorm2d(256)\n        self.conv3_3 = nn.Conv2d(256, 256, 3, padding=1)\n        self.relu3_3 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv3_3_bn = nn.BatchNorm2d(256)\n        self.pool3 = nn.MaxPool2d(2, stride=2, ceil_mode=True)  # 1/8\n\n        # conv4\n        self.conv4_1 = nn.Conv2d(256, 512, 3, padding=1)\n        self.relu4_1 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv4_1_bn = nn.BatchNorm2d(512)\n        self.conv4_2 = nn.Conv2d(512, 512, 3, padding=1)\n        self.relu4_2 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv4_2_bn = nn.BatchNorm2d(512)\n        self.conv4_3 = nn.Conv2d(512, 512, 3, padding=1)\n        self.relu4_3 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv4_3_bn = nn.BatchNorm2d(512)\n        self.pool4 = nn.MaxPool2d(2, stride=2, ceil_mode=True)  # 1/16\n\n        # conv5\n        self.conv5_1 = nn.Conv2d(512, 512, 3, padding=1)\n        self.relu5_1 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv5_1_bn = nn.BatchNorm2d(512)\n        self.conv5_2 = nn.Conv2d(512, 512, 3, padding=1)\n        self.relu5_2 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv5_2_bn = nn.BatchNorm2d(512)\n        self.conv5_3 = nn.Conv2d(512, 512, 3, padding=1)\n        self.relu5_3 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv5_3_bn = nn.BatchNorm2d(512)\n        self.pool5 = nn.MaxPool2d(2, stride=2, ceil_mode=True)  # 1/32\n\n        # fc6\n        self.fc6 = nn.Conv2d(512, 4096, 7)\n        self.relu6 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.fc6_bn = nn.BatchNorm2d(4096)\n        self.drop6 = nn.Dropout2d()\n\n        # fc7\n        self.fc7 = nn.Conv2d(4096, 4096, 1)\n        self.relu7 = nn.ReLU(inplace=True)\n        self.fc7_bn = nn.BatchNorm2d(4096)\n        self.drop7 = nn.Dropout2d()\n\n        if bin_type == 'one-hot':\n            # NOTE: *two* output prediction maps for hue and chroma\n            raise NotImplementedError('TODO - FCN 16s for separate hue-chroma')\n        elif bin_type == 'soft':\n            self.score_fr = nn.Conv2d(4096, n_class, 1)\n            self.score_pool4 = nn.Conv2d(512, n_class, 1)\n\n            self.upscore2 = nn.ConvTranspose2d(n_class, n_class, 4, stride=2,\n                                              bias=False) \n            self.upscore16 = nn.ConvTranspose2d(n_class, n_class, 32, stride=16,\n                                              bias=False)            \n            self.upscore2.weight.requires_grad = False # fix bilinear upsamplers\n            self.upscore16.weight.requires_grad = False\n\n        self._initialize_weights()\n        \n\n    def _initialize_weights(self):\n        for m in self.modules():\n            if isinstance(m, nn.Conv2d):\n                pass # leave the default PyTorch init\n            if isinstance(m, nn.ConvTranspose2d):\n                assert m.kernel_size[0] == m.kernel_size[1]\n                initial_weight = get_upsampling_weight(\n                    m.in_channels, m.out_channels, m.kernel_size[0])\n                m.weight.data.copy_(initial_weight)\n\n\n    def forward(self, x):\n        h = x\n        h = self.conv1_1(h)\n        if self.batch_norm:\n            h = self.conv1_1_bn(h)\n        h = self.relu1_1(h)\n        h = self.conv1_2(h)\n        if self.batch_norm:\n            h = self.conv1_2_bn(h)\n        h = self.relu1_2(h)\n        h = self.pool1(h)\n\n        if self.batch_norm:\n            h = self.relu2_1(self.conv2_1_bn(self.conv2_1(h)))\n        else:\n            h = self.relu2_1(self.conv2_1(h))\n        if self.batch_norm:\n            h = self.relu2_2(self.conv2_2_bn(self.conv2_2(h)))\n        else:\n            h = self.relu2_2(self.conv2_2_bn(self.conv2_2(h)))\n        h = self.pool2(h)\n\n        if self.batch_norm:\n            h = self.relu3_1(self.conv3_1_bn(self.conv3_1(h)))\n        else:\n            h = self.relu3_1(self.conv3_1(h))\n        if self.batch_norm:\n            h = self.relu3_2(self.conv3_2_bn(self.conv3_2(h)))\n        else:\n            h = self.relu3_2(self.conv3_2(h))\n        if self.batch_norm:\n            h = self.relu3_3(self.conv3_3_bn(self.conv3_3(h)))\n        else:\n            h = self.relu3_3(self.conv3_3(h))\n        h = self.pool3(h)\n\n        if self.batch_norm:\n            h = self.relu4_1(self.conv4_1_bn(self.conv4_1(h)))\n        else:\n            h = self.relu4_1(self.conv4_1(h))\n        if self.batch_norm:\n            h = self.relu4_2(self.conv4_2_bn(self.conv4_2(h)))\n        else:\n            h = self.relu4_2(self.conv4_2(h))\n        if self.batch_norm:\n            h = self.relu4_3(self.conv4_3_bn(self.conv4_3(h)))\n        else:\n            h = self.relu4_3(self.conv4_3(h))\n        h = self.pool4(h)\n        pool4 = h  # 1/16\n\n        if self.batch_norm:\n            h = self.relu5_1(self.conv5_1_bn(self.conv5_1(h)))\n        else:\n            h = self.relu5_1(self.conv5_1(h))\n        if self.batch_norm:\n            h = self.relu5_2(self.conv5_2_bn(self.conv5_2(h)))\n        else:\n            h = self.relu5_2(self.conv5_2(h))\n        if self.batch_norm:\n            h = self.relu5_3(self.conv5_3_bn(self.conv5_3(h)))\n        else:\n            h = self.relu5_3(self.conv5_3(h))\n        h = self.pool5(h)\n\n        if self.batch_norm:\n            h = self.relu6(self.fc6_bn(self.fc6(h)))\n        else:\n            h = self.relu6(self.fc6(h))\n        h = self.drop6(h)\n\n        if self.batch_norm:\n            h = self.relu7(self.fc7_bn(self.fc7(h)))\n        else:\n            h = self.relu7(self.fc7(h))\n        h = self.drop7(h)\n\n        if self.bin_type == 'one-hot':\n            raise NotImplementedError('TODO - FCN 16s for separate hue-chroma')\n        elif self.bin_type == 'soft':\n            h = self.score_fr(h)\n            h = self.upscore2(h)\n            upscore2 = h  # 1/16\n\n            h = self.score_pool4(pool4)\n            h = h[:, :, 5:5 + upscore2.size()[2], 5:5 + upscore2.size()[3]]\n            score_pool4c = h # 1/16\n\n            h = upscore2 + score_pool4c\n            \n            h = self.upscore16(h)\n            h = h[:, :, 27:27 + x.size()[2], 27:27 + x.size()[3]].contiguous()\n\n        return h\n\n\n    def copy_params_from_fcn32s(self, fcn32s):\n        for name, l1 in fcn32s.named_children():\n            try:\n                l2 = getattr(self, name)\n                l2.weight  # skip ReLU / Dropout\n            except Exception:\n                continue\n            assert l1.weight.size() == l2.weight.size()\n            assert l1.bias.size() == l2.bias.size()\n            l2.weight.data.copy_(l1.weight.data)\n            l2.bias.data.copy_(l1.bias.data)\n\n\n\nclass FCN8sColor(nn.Module):\n\n    def __init__(self, n_class=32, bin_type='one-hot', batch_norm=True):\n        super(FCN8sColor, self).__init__()\n        self.n_class = n_class\n        self.bin_type = bin_type\n        self.batch_norm = batch_norm\n\n        # conv1\n        self.conv1_1 = nn.Conv2d(1, 64, 3, padding=100)\n        self.relu1_1 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv1_1_bn = nn.BatchNorm2d(64)\n        self.conv1_2 = nn.Conv2d(64, 64, 3, padding=1)\n        self.relu1_2 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv1_2_bn = nn.BatchNorm2d(64)\n        self.pool1 = nn.MaxPool2d(2, stride=2, ceil_mode=True)  # 1/2\n\n        # conv2\n        self.conv2_1 = nn.Conv2d(64, 128, 3, padding=1)\n        self.relu2_1 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv2_1_bn = nn.BatchNorm2d(128)\n        self.conv2_2 = nn.Conv2d(128, 128, 3, padding=1)\n        self.relu2_2 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv2_2_bn = nn.BatchNorm2d(128)\n        self.pool2 = nn.MaxPool2d(2, stride=2, ceil_mode=True)  # 1/4\n\n        # conv3\n        self.conv3_1 = nn.Conv2d(128, 256, 3, padding=1)\n        self.relu3_1 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv3_1_bn = nn.BatchNorm2d(256)\n        self.conv3_2 = nn.Conv2d(256, 256, 3, padding=1)\n        self.relu3_2 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv3_2_bn = nn.BatchNorm2d(256)\n        self.conv3_3 = nn.Conv2d(256, 256, 3, padding=1)\n        self.relu3_3 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv3_3_bn = nn.BatchNorm2d(256)\n        self.pool3 = nn.MaxPool2d(2, stride=2, ceil_mode=True)  # 1/8\n\n        # conv4\n        self.conv4_1 = nn.Conv2d(256, 512, 3, padding=1)\n        self.relu4_1 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv4_1_bn = nn.BatchNorm2d(512)\n        self.conv4_2 = nn.Conv2d(512, 512, 3, padding=1)\n        self.relu4_2 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv4_2_bn = nn.BatchNorm2d(512)\n        self.conv4_3 = nn.Conv2d(512, 512, 3, padding=1)\n        self.relu4_3 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv4_3_bn = nn.BatchNorm2d(512)\n        self.pool4 = nn.MaxPool2d(2, stride=2, ceil_mode=True)  # 1/16\n\n        # conv5\n        self.conv5_1 = nn.Conv2d(512, 512, 3, padding=1)\n        self.relu5_1 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv5_1_bn = nn.BatchNorm2d(512)\n        self.conv5_2 = nn.Conv2d(512, 512, 3, padding=1)\n        self.relu5_2 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv5_2_bn = nn.BatchNorm2d(512)\n        self.conv5_3 = nn.Conv2d(512, 512, 3, padding=1)\n        self.relu5_3 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.conv5_3_bn = nn.BatchNorm2d(512)\n        self.pool5 = nn.MaxPool2d(2, stride=2, ceil_mode=True)  # 1/32\n\n        # fc6\n        self.fc6 = nn.Conv2d(512, 4096, 7)\n        self.relu6 = nn.ReLU(inplace=True)\n        if batch_norm:\n            self.fc6_bn = nn.BatchNorm2d(4096)\n        self.drop6 = nn.Dropout2d()\n\n        # fc7\n        self.fc7 = nn.Conv2d(4096, 4096, 1)\n        self.relu7 = nn.ReLU(inplace=True)\n        self.fc7_bn = nn.BatchNorm2d(4096)\n        self.drop7 = nn.Dropout2d()\n\n        if bin_type == 'one-hot':\n            # NOTE: *two* output prediction maps for hue and chroma\n            raise NotImplementedError('TODO - FCN 16s for separate hue-chroma')\n        elif bin_type == 'soft':\n            self.score_fr = nn.Conv2d(4096, n_class, 1)\n            self.score_pool3 = nn.Conv2d(256, n_class, 1)\n            self.score_pool4 = nn.Conv2d(512, n_class, 1)\n\n            self.upscore2 = nn.ConvTranspose2d(n_class, n_class, 4, stride=2,\n                                              bias=False) \n            self.upscore8 = nn.ConvTranspose2d(n_class, n_class, 16, stride=8,\n                                              bias=False)  \n            self.upscore_pool4 = nn.ConvTranspose2d(n_class, n_class, 4, stride=2,\n                                              bias=False) \n\n            self.upscore2.weight.requires_grad = False # fix bilinear upsamplers\n            self.upscore8.weight.requires_grad = False\n            self.upscore_pool4.weight.requires_grad = False\n\n        self._initialize_weights()\n        \n\n    def _initialize_weights(self):\n        for m in self.modules():\n            if isinstance(m, nn.Conv2d):\n                pass # leave the default PyTorch init\n            if isinstance(m, nn.ConvTranspose2d):\n                assert m.kernel_size[0] == m.kernel_size[1]\n                initial_weight = get_upsampling_weight(\n                    m.in_channels, m.out_channels, m.kernel_size[0])\n                m.weight.data.copy_(initial_weight)\n\n\n    def forward(self, x):\n        h = x\n        h = self.conv1_1(h)\n        if self.batch_norm:\n            h = self.conv1_1_bn(h)\n        h = self.relu1_1(h)\n        h = self.conv1_2(h)\n        if self.batch_norm:\n            h = self.conv1_2_bn(h)\n        h = self.relu1_2(h)\n        h = self.pool1(h)\n\n        if self.batch_norm:\n            h = self.relu2_1(self.conv2_1_bn(self.conv2_1(h)))\n        else:\n            h = self.relu2_1(self.conv2_1(h))\n        if self.batch_norm:\n            h = self.relu2_2(self.conv2_2_bn(self.conv2_2(h)))\n        else:\n            h = self.relu2_2(self.conv2_2_bn(self.conv2_2(h)))\n        h = self.pool2(h)\n\n        if self.batch_norm:\n            h = self.relu3_1(self.conv3_1_bn(self.conv3_1(h)))\n        else:\n            h = self.relu3_1(self.conv3_1(h))\n        if self.batch_norm:\n            h = self.relu3_2(self.conv3_2_bn(self.conv3_2(h)))\n        else:\n            h = self.relu3_2(self.conv3_2(h))\n        if self.batch_norm:\n            h = self.relu3_3(self.conv3_3_bn(self.conv3_3(h)))\n        else:\n            h = self.relu3_3(self.conv3_3(h))\n        h = self.pool3(h)\n        pool3 = h  # 1/8\n\n        if self.batch_norm:\n            h = self.relu4_1(self.conv4_1_bn(self.conv4_1(h)))\n        else:\n            h = self.relu4_1(self.conv4_1(h))\n        if self.batch_norm:\n            h = self.relu4_2(self.conv4_2_bn(self.conv4_2(h)))\n        else:\n            h = self.relu4_2(self.conv4_2(h))\n        if self.batch_norm:\n            h = self.relu4_3(self.conv4_3_bn(self.conv4_3(h)))\n        else:\n            h = self.relu4_3(self.conv4_3(h))\n        h = self.pool4(h)\n        pool4 = h  # 1/16\n\n        if self.batch_norm:\n            h = self.relu5_1(self.conv5_1_bn(self.conv5_1(h)))\n        else:\n            h = self.relu5_1(self.conv5_1(h))\n        if self.batch_norm:\n            h = self.relu5_2(self.conv5_2_bn(self.conv5_2(h)))\n        else:\n            h = self.relu5_2(self.conv5_2(h))\n        if self.batch_norm:\n            h = self.relu5_3(self.conv5_3_bn(self.conv5_3(h)))\n        else:\n            h = self.relu5_3(self.conv5_3(h))\n        h = self.pool5(h)\n\n        if self.batch_norm:\n            h = self.relu6(self.fc6_bn(self.fc6(h)))\n        else:\n            h = self.relu6(self.fc6(h))\n        h = self.drop6(h)\n\n        if self.batch_norm:\n            h = self.relu7(self.fc7_bn(self.fc7(h)))\n        else:\n            h = self.relu7(self.fc7(h))\n        h = self.drop7(h)\n\n        if self.bin_type == 'one-hot':\n            raise NotImplementedError('TODO - FCN 16s for separate hue-chroma')\n        elif self.bin_type == 'soft':\n            h = self.score_fr(h)\n            h = self.upscore2(h)\n            upscore2 = h  # 1/16\n\n            h = self.score_pool4(pool4)\n            h = h[:, :, 5:5 + upscore2.size()[2], 5:5 + upscore2.size()[3]]\n            score_pool4c = h  # 1/16\n\n            h = upscore2 + score_pool4c  # 1/16\n            h = self.upscore_pool4(h)\n            upscore_pool4 = h # 1/8\n            \n            h = self.score_pool3(pool3)\n            h = h[:, :,\n              9:9 + upscore_pool4.size()[2],\n              9:9 + upscore_pool4.size()[3]]\n            score_pool3c = h  # 1/8\n\n            h = upscore_pool4 + score_pool3c # 1/8\n\n            h = self.upscore8(h)\n            h = h[:, :, 31:31 + x.size()[2], 31:31 + x.size()[3]].contiguous()\n\n        return h\n\n\n    def copy_params_from_fcn16s(self, fcn16s):\n        for name, l1 in fcn16s.named_children():\n            try:\n                l2 = getattr(self, name)\n                l2.weight  # skip ReLU / Dropout\n            except Exception:\n                continue\n            assert l1.weight.size() == l2.weight.size()\n            l2.weight.data.copy_(l1.weight.data)\n            if l1.bias is not None:\n                assert l1.bias.size() == l2.bias.size()\n                l2.bias.data.copy_(l1.bias.data)\n\n\n\n"
  },
  {
    "path": "run_resnet_demo.py",
    "content": "\nimport torch\nimport torchvision\nfrom torchvision import models\nimport torch.nn as nn\nimport torch.utils.data\nimport torchvision.transforms as transforms\nfrom torch.autograd import Variable\nimport torch.nn as nn\nimport torch.nn.functional as F\n\nimport os\nimport os.path as osp\nimport yaml\nimport numpy as np\nimport PIL.Image\nimport matplotlib\nmatplotlib.use('Agg')\nimport matplotlib.pyplot as plt\n\nimport utils\n\n\n\n\nhere = osp.dirname(osp.abspath(__file__)) # output folder is located here\n\n    \n# -----------------------------------------------------------------------------\n# 0. User-defined settings\n# -----------------------------------------------------------------------------\ngpu = 0 # use gpu:0 by default\n# specify model path of trained ResNet-50 network:\nmodel_path = './umd-face/logs/MODEL-resnet_umdfaces_CFG-006_TIME-20180114-141943/model_best.pth.tar'  \nnum_class = 8277 # UMD-Faces had this many classes\n\n\n\n# -----------------------------------------------------------------------------\n# 1. GPU setup\n# -----------------------------------------------------------------------------\nos.environ['CUDA_VISIBLE_DEVICES'] = str(gpu)\ncuda = torch.cuda.is_available()\ntorch.manual_seed(1337)\nif cuda:\n    torch.cuda.manual_seed(1337)\n    torch.backends.cudnn.enabled = True\n    torch.backends.cudnn.benchmark = True    \n\n\n\n# -----------------------------------------------------------------------------\n# 2. Data preparation\n# -----------------------------------------------------------------------------\n\n# Samples images taken for demo purpose from LFW:\n#   http://vis-www.cs.umass.edu/lfw/\ndata_root = './samples/verif'\nfile_path = [osp.join(data_root, 'Recep_Tayyip_Erdogan_0012.jpg'), \n             osp.join(data_root, 'Recep_Tayyip_Erdogan_0015.jpg'), \n             osp.join(data_root, 'Quincy_Jones_0001.jpg')]\nimage = [PIL.Image.open(f).convert('RGB') for f in file_path]\n\n# Data transforms\n# http://pytorch.org/docs/master/torchvision/transforms.html\n# NOTE: these should be consistent with the training script val_loader\n# Since LFW images (250x250) are not close-crops, we modify the cropping a bit.\nRGB_MEAN = [ 0.485, 0.456, 0.406 ]\nRGB_STD = [ 0.229, 0.224, 0.225 ]\ntest_transform = transforms.Compose([\n    transforms.CenterCrop(150),   # 150x150 center crop\n    transforms.Scale((224,224)),  # resized to the network's required input size\n    transforms.ToTensor(),\n    transforms.Normalize(mean = RGB_MEAN,\n                         std = RGB_STD),\n])\n\n# apply the transform\ninputs = [test_transform(im) for im in image]\n\n\n\n# -----------------------------------------------------------------------------\n# 3. Model\n# -----------------------------------------------------------------------------\n# PyTorch ResNet model definition: \n#   https://github.com/pytorch/vision/blob/master/torchvision/models/resnet.py\n# ResNet docs:\n#   http://pytorch.org/docs/master/torchvision/models.html#id3\nmodel = torchvision.models.resnet50(pretrained=True)\n\n# Replace last layer (by default, resnet has 1000 output categories)\nmodel.fc = torch.nn.Linear(2048, num_class) # change to current dataset's classes\n\n# Pre-trained PyTorch model loaded from a file\ncheckpoint = torch.load(model_path)        \n\nif checkpoint['arch'] == 'DataParallel':\n    # if we trained and saved our model using DataParallel\n    model = torch.nn.DataParallel(model, device_ids=[0, 1, 2, 3, 4])\n    model.load_state_dict(checkpoint['model_state_dict'])\n    model = model.module # get network module from inside its DataParallel wrapper\nelse:\n    model.load_state_dict(checkpoint['model_state_dict'])\n\nif cuda:\n    model = model.cuda()\n\n\n# Convert the trained network into a \"feature extractor\"\n# From https://github.com/meliketoy/fine-tuning.pytorch/blob/master/extract_features.py#L85\nfeature_map = list(model.children())\nfeature_map.pop()  # remove the final \"class prediction\" layer\nextractor = nn.Sequential(*feature_map) # create feature extractor\n\n# Inspect the structure - it is a nested list of various modules\nprint extractor[-1]      # last layer of the model - avg-pool\nprint extractor[-2][-1]  # second-last layer's last module - output is 2048-dim\n\n\n\n# -----------------------------------------------------------------------------\n# 4. Feature extraction\n# -----------------------------------------------------------------------------\n# - simple, one input sample at a time\nfeatures = []\nfor x in inputs:\n    x = Variable(x, volatile=True)\n    if cuda:\n        x = x.cuda()\n    x = x.view(1, x.size(0), x.size(1), x.size(2)) # add batch_dim=1 in the front\n    feat = extractor(x).view(-1) # extract features of input `x`, reshape to 1-D vector\n    features.append(feat)\nfeatures = torch.stack(features) # N x 2048 for N inputs\n\n# get Tensors on CPU from autograd.Variables on GPU\nif cuda:\n    features = features.data.cpu()\nelse:\n    features = features.data\n\nfeatures = F.normalize(features, p=2, dim=1)  # L2-normalize\n\n\n# -----------------------------------------------------------------------------\n# 5. Face verification \n# -----------------------------------------------------------------------------\n\n# L2-distance between features (Tensors) of same and different pairs\nd1 = (features[0] - features[1]).norm(p=2) # same pair\nd2 = (features[0] - features[2]).norm(p=2) # different pair\n\nprint 'matched pair: %.2f' % d1   \nprint 'mismatched pair: %.2f' % d2\nassert d1 < d2\n\n# visualizations\nfig, ax = plt.subplots(nrows=2, ncols=2)\nplt.subplot(2, 2, 1)\nplt.title('matched pair')\nplt.imshow(image[0])\nplt.tight_layout()\n\nplt.subplot(2, 2, 2)\nplt.imshow(image[1])\nplt.title('d = %.3f' % d1)\nplt.tight_layout()\n\nplt.subplot(2, 2, 3)\nplt.imshow(image[0])\nplt.title('mismatched pair')\nplt.tight_layout()\n\nplt.subplot(2, 2, 4)\nplt.imshow(image[2])\nplt.title('d = %.3f' % d2)\nplt.tight_layout()\n\nplt.savefig(osp.join(here, 'demo_verif.png'), bbox_inches='tight')\n\nprint 'Visualization saved in ' + osp.join(here, 'demo_verif.png')\n\n"
  },
  {
    "path": "tests/test_colorizer.py",
    "content": "import argparse\nimport os\nimport os.path as osp\nimport numpy as np\nimport PIL.Image\nimport skimage.io\nimport skimage.color as color\nfrom skimage import img_as_ubyte\nimport torch\nfrom torch.autograd import Variable\nimport sys\nsys.path.append('/vis/home/arunirc/data1/Research/colorize-fcn/colorizer-fcn')\n# import matplotlib\n# matplotlib.use('agg')\n# import matplotlib.pyplot as plt\n\n# import models\n# import utils\nimport data_loader\n\nroot = '/srv/data1/arunirc/datasets/ImageNet/images/'\ncuda = torch.cuda.is_available()\nGMM_PATH = '/srv/data1/arunirc/Research/colorize-fcn/colorizer-fcn/logs/MODEL-fcn32s_color_CFG-014_VCS-db517d6_TIME-20171230-212406/gmm.pkl'\nMEAN_L_PATH = '/srv/data1/arunirc/Research/colorize-fcn/colorizer-fcn/logs/MODEL-fcn32s_color_CFG-014_VCS-db517d6_TIME-20171230-212406/mean_l.npy'\n\n\ndef test_color_gmm():\n    print 'Entering: test_color'\n    dataset = data_loader.ColorizeImageNet(\n                root, split='val', set='tiny',\n                bins='soft', num_hc_bins=16,\n                gmm_path=GMM_PATH, mean_l_path=MEAN_L_PATH)\n\n    img, labels = dataset.__getitem__(0)\n    gmm = dataset.gmm\n\n    labels = labels.numpy()\n    img = img.squeeze().numpy()\n    labels = labels.astype(gmm.means_.dtype)\n    img = img.astype(gmm.means_.dtype)\n\n    # expectation over GMM centroids\n    hc_means = gmm.means_.astype(labels.dtype)\n    im_hc = np.tensordot(labels, hc_means, (2,0)) \n    im_l = img + dataset.mean_l.astype(img.dtype)\n    im_rgb = dataset.hue_chroma_to_rgb(im_hc, im_l)\n    low, high = np.min(im_rgb), np.max(im_rgb)\n    im_rgb = (im_rgb - low) / (high - low)\n    im_out = img_as_ubyte(im_rgb)\n    skimage.io.imsave(\"tests/output.png\", im_out)\n\n    img_file = dataset.files['val'][0]\n    im_orig = skimage.io.imread(img_file)\n    skimage.io.imsave(\"tests/orig.png\", im_orig)\n\n\n\ndef main():\n    test_color_gmm()\n\n\n\n\n\nif __name__ == '__main__':\n    main()\n"
  },
  {
    "path": "tests/test_data_reader.py",
    "content": "import argparse\nimport os\nimport os.path as osp\n\nimport numpy as np\nimport PIL.Image\nimport skimage.io\nimport skimage.color as color\nimport torch\nfrom torch.autograd import Variable\n\nimport sys\nsys.path.append('/vis/home/arunirc/data1/Research/colorize-fcn/colorizer-fcn')\nimport models\nimport train\nimport utils\nimport data_loader\n\nroot = '/vis/home/arunirc/data1/datasets/ImageNet/images/'\n\n\ndef test_single_read():\n    print 'Entering: test_single_read'\n    dataset = data_loader.ColorizeImageNet(root, split='train', set='small')\n    img, lbl = dataset.__getitem__(0)\n    assert len(lbl)==2 \n    assert np.min(lbl[0].numpy())==0\n    assert np.max(lbl[0].numpy())==30\n    print 'Test passed: test_single_read'\n\n\ndef test_single_read_dimcheck():\n    print 'Entering: test_single_read_dimcheck'\n    dataset = data_loader.ColorizeImageNet(root, split='train', set='small')\n    img, lbl = dataset.__getitem__(0)\n    assert len(lbl)==2\n    im_hue = lbl[0].numpy()\n    im_chroma = lbl[1].numpy()\n    assert im_chroma.shape==im_hue.shape, \\\n            'Labels (Hue and Chroma maps) should have same dimensions.'\n    print 'Test passed: test_single_read_dimcheck'\n\n\ndef test_train_loader():\n    print 'Entering: test_train_loader'\n    train_loader = torch.utils.data.DataLoader(\n        data_loader.ColorizeImageNet(root, split='train', set='small'),\n        batch_size=1, shuffle=False)\n    dataiter = iter(train_loader)\n    img, label = dataiter.next()\n    assert len(label)==2, \\\n        'Network should predict a 2-tuple: hue-map and chroma-map.'\n    im_hue = label[0].numpy()\n    im_chroma = label[1].numpy()\n    assert im_chroma.shape==im_hue.shape, \\\n            'Labels (Hue and Chroma maps) should have same dimensions.'\n    print 'Test passed: test_train_loader'\n\n\ndef test_dataset_read():\n    '''\n        Read through the entire dataset.\n    '''\n    dataset = data_loader.ColorizeImageNet(\\\n                root, split='train', set='small')\n\n    for i in xrange(len(dataset)):\n        # if i > 44890: # HACK: skipping over some stuff\n        img_file = dataset.files['train'][i]\n        img, lbl = dataset.__getitem__(i)\n        assert type(lbl) == torch.FloatTensor\n        assert type(img) == torch.FloatTensor\n        print 'iter: %d,\\t file: %s,\\t imsize: %s' % (i, img_file, img.size())\n\n\ndef test_cmyk_read():\n    '''\n        Handle CMYK images -- skip to previous image.\n    '''\n    print 'Entering: test_cmyk_read'\n    dataset = data_loader.ColorizeImageNet(\\\n                root, split='train', set='small')\n    idx = 44896\n    img_file = dataset.files['train'][idx]\n    im1 = PIL.Image.open(img_file)\n    im1 = np.asarray(im1, dtype=np.uint8)\n    assert im1.shape[2]==4, 'Check that selected image is indeed CMYK.'\n    img, lbl = dataset.__getitem__(idx)    \n    print 'Test passed: test_cmyk_read'\n\n\ndef test_grayscale_read():\n    '''\n        Handle single-channel images -- skip to previous image.\n    '''\n    print 'Entering: test_grayscale_read'\n    dataset = data_loader.ColorizeImageNet(root, split='train', set='small')\n    idx = 4606\n    img_file = dataset.files['train'][idx]\n    im1 = PIL.Image.open(img_file)\n    im1 = np.asarray(im1, dtype=np.uint8)\n    assert len(im1.shape)==2, 'Check that selected image is indeed grayscale.'\n    img, lbl = dataset.__getitem__(idx)    \n    print 'Test passed: test_grayscale_read'\n\n\ndef test_rgb_hsv():\n    # DEFER\n    dataset = data_loader.ColorizeImageNet(\\\n                root, split='train', set='small')\n    img_file = dataset.files['train'][100]\n    img = PIL.Image.open(img_file)\n    img = np.array(img, dtype=np.uint8)\n    assert np.max(img.shape) == 400\n\n\ndef test_soft_bins():\n    dataset = \\\n        data_loader.ColorizeImageNet(root, split='train', set='small', \n                                     bins='soft')\n    img, lbl = dataset.__getitem__(0)    \n    assert type(lbl) == torch.FloatTensor\n    assert type(img) == torch.FloatTensor\n    print 'Test passed: test_soft_bins'\n\n\ndef test_lowpass_image():\n    dataset = \\\n        data_loader.ColorizeImageNet(root, split='train', set='small', \n                                     bins='soft', img_lowpass=8)\n    img, lbl = dataset.__getitem__(0)    \n    assert type(lbl) == torch.FloatTensor\n    assert type(img) == torch.FloatTensor\n    print 'Test passed: test_soft_bins'\n\n\ndef test_init_gmm():\n    # Pass paths to cached GMM and mean Lightness\n    GMM_PATH = '/srv/data1/arunirc/Research/colorize-fcn/colorizer-fcn/logs/MODEL-fcn32s_color_CFG-014_VCS-db517d6_TIME-20171230-212406/gmm.pkl'\n    MEAN_L_PATH = '/srv/data1/arunirc/Research/colorize-fcn/colorizer-fcn/logs/MODEL-fcn32s_color_CFG-014_VCS-db517d6_TIME-20171230-212406/mean_l.npy'\n    dataset = \\\n        data_loader.ColorizeImageNet(\n            root, split='train', set='tiny', bins='soft', \n            gmm_path=GMM_PATH, mean_l_path=MEAN_L_PATH)\n    print 'Test passed: test_init_gmm'\n\n\n\ndef main():\n    test_single_read()\n    test_single_read_dimcheck()\n    test_train_loader()\n    test_cmyk_read()\n    test_grayscale_read()\n    test_soft_bins()\n    test_lowpass_image()\n    test_init_gmm()\n\n    # \n    # dataset.get_color_samples()\n    # test_dataset_read()\n    # TODO - test_labels\n    # TODO - test colorspace conversions\n\n\n\n\n\nif __name__ == '__main__':\n    main()\n"
  },
  {
    "path": "tests/test_fcn32s.py",
    "content": "# FIXME: Import order causes error:\n# ImportError: dlopen: cannot load any more object with static TL\n# https://github.com/pytorch/pytorch/issues/2083\nimport torch\n\nimport matplotlib.pyplot as plt\nimport numpy as np\nimport skimage.data\n\nfrom torchfcn.models.fcn32s import get_upsampling_weight\n\n\ndef test_get_upsampling_weight():\n    src = skimage.data.coffee()\n    x = src.transpose(2, 0, 1)\n    x = x[np.newaxis, :, :, :]\n    x = torch.from_numpy(x).float()\n    x = torch.autograd.Variable(x)\n\n    in_channels = 3\n    out_channels = 3\n    kernel_size = 4\n\n    m = torch.nn.ConvTranspose2d(\n        in_channels, out_channels, kernel_size, stride=2, bias=False)\n    m.weight.data = get_upsampling_weight(\n        in_channels, out_channels, kernel_size)\n\n    y = m(x)\n\n    y = y.data.numpy()\n    y = y[0]\n    y = y.transpose(1, 2, 0)\n    dst = y.astype(np.uint8)\n\n    assert abs(src.shape[0] * 2 - dst.shape[0]) <= 2\n    assert abs(src.shape[1] * 2 - dst.shape[1]) <= 2\n\n    return src, dst\n\n\nif __name__ == '__main__':\n    src, dst = test_get_upsampling_weight()\n    plt.subplot(121)\n    plt.imshow(src)\n    plt.title('x1: {}'.format(src.shape))\n    plt.subplot(122)\n    plt.imshow(dst)\n    plt.title('x2: {}'.format(dst.shape))\n    plt.show()\n"
  },
  {
    "path": "tests/vis_prediction.py",
    "content": "import argparse\nimport os\nimport os.path as osp\nimport numpy as np\nimport PIL.Image\nimport skimage.io\nimport skimage.color as color\nimport matplotlib\nmatplotlib.use('Agg')\nimport matplotlib.pyplot as plt\nimport torch\nfrom torch.autograd import Variable\n\n\nimport sys\nsys.path.append('/vis/home/arunirc/data1/Research/colorize-fcn/colorizer-fcn')\nimport utils\nimport data_loader\n\n\nroot = '/vis/home/arunirc/data1/datasets/ImageNet/images/'\nout_path = '/data2/arunirc/Research/colorize-fcn/pytorch-fcn/tests/data_tests/'\nGMM_PATH = '/srv/data1/arunirc/Research/colorize-fcn/colorizer-fcn/logs/MODEL-fcn32s_color_CFG-014_VCS-db517d6_TIME-20171230-212406/gmm.pkl'\nMEAN_L_PATH = '/srv/data1/arunirc/Research/colorize-fcn/colorizer-fcn/logs/MODEL-fcn32s_color_CFG-014_VCS-db517d6_TIME-20171230-212406/mean_l.npy'\ncuda = torch.cuda.is_available()\n\n\n\ndef main():\n    dataset = data_loader.ColorizeImageNet(\n                root, split='val', set='small',\n                bins='soft', num_hc_bins=16,\n                gmm_path=GMM_PATH, mean_l_path=MEAN_L_PATH)\n    img, labels = dataset.__getitem__(0)\n    gmm = dataset.gmm\n    mean_l = dataset.mean_l\n\n    img_file = dataset.files['val'][1]\n    im_orig = skimage.io.imread(img_file)\n\n    # ... predicted labels and input image (mean subtracted)\n    labels = labels.numpy()\n    img = img.squeeze().numpy()\n    im_rgb = utils.colorize_image_hc(labels, img, gmm, mean_l)\n\n    plt.imshow(im_rgb)\n    plt.show()\n\n    # \n    inputs = Variable(img)\n    if cuda:\n      inputs = inputs.cuda()\n    outputs = model(inputs)\n    # TODO: assertions\n    # del inputs, outputs\n\n\n\n\n\n\n\nif __name__ == '__main__':\n    main()\n"
  },
  {
    "path": "train.py",
    "content": "import datetime\nimport math\nimport os\nimport os.path as osp\nimport shutil\n\nimport numpy as np\nimport PIL.Image\nimport pytz\nimport scipy.misc\nimport torch\nfrom torch.autograd import Variable\nimport torch.nn.functional as F\nimport tqdm\n\nimport utils\nimport gc\n\n\ndef lr_scheduler(optimizer, epoch, init_lr=0.001, lr_decay_epoch=7):\n    \"\"\"Decay learning rate by a factor of 0.1 every lr_decay_epoch epochs.\"\"\"\n    lr = init_lr * (0.1**(epoch // lr_decay_epoch))\n\n    if epoch % lr_decay_epoch == 0:\n        print('LR is set to {}'.format(lr))\n\n    for param_group in optimizer.param_groups:\n        param_group['lr'] = lr\n\n    return optimizer\n\n\n\nclass Trainer(object):\n\n    # -----------------------------------------------------------------------------\n    def __init__(self, cuda, model, criterion, optimizer, init_lr,\n                 train_loader, val_loader, out, max_iter, \n                 lr_decay_epoch=None, interval_validate=None):\n    # -----------------------------------------------------------------------------\n        self.cuda = cuda\n\n        self.model = model\n        self.criterion = criterion\n        self.optim = optimizer\n        self.train_loader = train_loader\n        self.val_loader = val_loader\n        self.best_acc = 0\n        self.init_lr = init_lr\n        self.lr_decay_epoch = lr_decay_epoch\n        self.epoch = 0\n        self.iteration = 0\n        self.max_iter = max_iter\n        self.best_loss = 0\n\n        self.timestamp_start = \\\n            datetime.datetime.now(pytz.timezone('US/Eastern'))\n\n        if interval_validate is None:\n            self.interval_validate = len(self.train_loader)\n        else:\n            self.interval_validate = interval_validate\n\n        self.out = out\n        if not osp.exists(self.out):\n            os.makedirs(self.out)\n\n        self.log_headers = [\n            'epoch',\n            'iteration',\n            'train/loss',\n            'train/acc',\n            'valid/loss',\n            'valid/acc',\n            'elapsed_time',\n        ]\n        if not osp.exists(osp.join(self.out, 'log.csv')):\n            with open(osp.join(self.out, 'log.csv'), 'w') as f:\n                f.write(','.join(self.log_headers) + '\\n')\n\n        \n\n\n    # -----------------------------------------------------------------------------\n    def validate(self, max_num=500):\n    # -----------------------------------------------------------------------------\n        training = self.model.training\n        self.model.eval()\n        MAX_NUM = max_num # HACK: stop after 500 images\n\n        n_class = len(self.val_loader.dataset.classes)\n        val_loss = 0\n        label_trues, label_preds = [], []\n\n        for batch_idx, (data, (target)) in tqdm.tqdm(\n                enumerate(self.val_loader), total=len(self.val_loader),\n                desc='Val=%d' % self.iteration, ncols=80,\n                leave=False):\n\n            # Computing val losses\n            if self.cuda:\n                data, target = data.cuda(), target.cuda()\n\n            data, target = Variable(data), Variable(target)\n            score = self.model(data)\n            loss = self.criterion(score, target)\n\n            if np.isnan(float(loss.data[0])):\n                raise ValueError('loss is NaN while validating')\n\n            val_loss += float(loss.data[0]) / len(data)\n\n            lbl_pred = score.data.max(1)[1].cpu().numpy()\n            lbl_true = target.data.cpu()\n            lbl_pred = lbl_pred.squeeze()\n            lbl_true = np.squeeze(lbl_true.numpy())\n\n            del target, score\n\n            label_trues.append(lbl_true)\n            label_preds.append(lbl_pred)\n\n            del lbl_true, lbl_pred, data, loss\n\n            if batch_idx > MAX_NUM:\n                break\n\n        # Computing metrics\n        val_loss /= len(self.val_loader)\n        val_acc = self.eval_metric(label_trues, label_preds)\n\n        # Logging\n        with open(osp.join(self.out, 'log.csv'), 'a') as f:\n            elapsed_time = (\n                datetime.datetime.now(pytz.timezone('US/Eastern')) -\n                self.timestamp_start).total_seconds()\n            log = [self.epoch, self.iteration] + [''] * 2 + \\\n                  [val_loss, val_acc] + [elapsed_time]\n            log = map(str, log)\n            f.write(','.join(log) + '\\n')\n\n        del label_trues, label_preds\n\n        # Saving the best performing model\n        is_best = val_acc > self.best_acc\n        if is_best:\n            self.best_acc = val_acc\n\n        torch.save({\n            'epoch': self.epoch,\n            'iteration': self.iteration,\n            'arch': self.model.__class__.__name__,\n            'optim_state_dict': self.optim.state_dict(),\n            'model_state_dict': self.model.state_dict(),\n            'best_acc': self.best_acc,\n        }, osp.join(self.out, 'checkpoint.pth.tar'))\n\n        if is_best:\n            shutil.copy(osp.join(self.out, 'checkpoint.pth.tar'),\n                        osp.join(self.out, 'model_best.pth.tar'))\n\n        if training:\n            self.model.train()\n\n\n\n    # -----------------------------------------------------------------------------\n    def train_epoch(self):\n    # -----------------------------------------------------------------------------\n        self.model.train()\n        n_class = len(self.train_loader.dataset.classes)\n\n        for batch_idx, (data, target) in tqdm.tqdm(\n                enumerate(self.train_loader), total=len(self.train_loader),\n                desc='Train epoch=%d' % self.epoch, ncols=80, leave=False):\n\n            if batch_idx == len(self.train_loader)-1:\n                break # discard last batch in epoch (unequal batch-sizes mess up BatchNorm)\n\n            iteration = batch_idx + self.epoch * len(self.train_loader)\n            if self.iteration != 0 and (iteration - 1) != self.iteration:\n                continue  # for resuming\n            self.iteration = iteration\n\n            if self.iteration % self.interval_validate == 0:\n                self.validate()\n\n            assert self.model.training\n\n            # Computing Losses\n            if self.cuda:\n                data, target = data.cuda(), target.cuda()\n            data, target = Variable(data), Variable(target)\n            score = self.model(data)  # batch_size x num_class\n\n            loss = self.criterion(score, target)\n\n            if np.isnan(float(loss.data[0])):\n                raise ValueError('loss is NaN while training')\n            # print list(self.model.parameters())[0].grad\n\n            # Gradient descent\n            self.optim.zero_grad()\n            loss.backward()\n            self.optim.step()\n\n            # Computing metrics\n            lbl_pred = score.data.max(1)[1].cpu().numpy()\n            lbl_pred = lbl_pred.squeeze()\n            lbl_true = target.data.cpu()\n            lbl_true = np.squeeze(lbl_true.numpy())\n            train_accu = self.eval_metric([lbl_pred], [lbl_true])\n\n            # Logging\n            with open(osp.join(self.out, 'log.csv'), 'a') as f:\n                elapsed_time = (\n                    datetime.datetime.now(pytz.timezone('US/Eastern')) -\n                    self.timestamp_start).total_seconds()\n                log = [self.epoch, self.iteration] + [loss.data[0]] + \\\n                      [train_accu] + [''] * 2 + [elapsed_time]\n                log = map(str, log)\n                f.write(','.join(log) + '\\n')\n                # print '\\nEpoch: ' + str(self.epoch) + ' Iter: ' + str(self.iteration) + \\\n                #         ' Loss: ' + str(loss.data[0])\n\n            if self.iteration >= self.max_iter:\n                break\n\n\n    # -----------------------------------------------------------------------------\n    def eval_metric(self, lbl_pred, lbl_true):\n    # -----------------------------------------------------------------------------\n        # Over-all accuracy\n        # TODO: per-class accuracy\n        accu = []\n        for lt, lp in zip(lbl_true, lbl_pred):\n            accu.append(np.mean(lt == lp))\n        return np.mean(accu)\n\n\n    # -----------------------------------------------------------------------------\n    def train(self):\n    # -----------------------------------------------------------------------------\n        max_epoch = int(math.ceil(1. * self.max_iter / len(self.train_loader)))\n        print 'Number of iters in an epoch: %d' % len(self.train_loader)\n        print 'Total epochs: %d' % max_epoch        \n\n        for epoch in tqdm.trange(self.epoch, max_epoch,\n                                 desc='Train epochs', ncols=80, leave=True):\n            self.epoch = epoch\n\n            if self.lr_decay_epoch is None:\n                pass\n            else:\n                assert self.lr_decay_epoch < max_epoch\n                lr_scheduler(self.optim, self.epoch, \n                             init_lr=self.init_lr, \n                             lr_decay_epoch=self.lr_decay_epoch)\n\n            self.train_epoch()\n            if self.iteration >= self.max_iter:\n                break\n\n"
  },
  {
    "path": "umd-face/run_crop_face.py",
    "content": "import argparse\nimport os\nimport os.path as osp\n\n# import torch\n# import torchvision\n# import torch.utils.data\n# import torchvision.datasets as datasets\n\nimport yaml\nimport numpy as np\nimport matplotlib\nmatplotlib.use('Agg')\nimport matplotlib.pyplot as plt\nimport PIL\nfrom tqdm import tqdm\n# from multiprocessing.pool import ThreadPool as Pool\n# pool_size = 5  # your \"parallelness\"\n# pool = Pool(pool_size)\n\n'''\n    Crops out the faces from UMDFace images using the annotations in \n    umdfaces_batch*_ultraface.csv. \n    Automatically creates \"train\" and \"val\" folders.\n\n'''\n\ndef main():\n    parser = argparse.ArgumentParser()\n    parser.add_argument('-d', '--dataset_path', \n                        default='/srv/data1/arunirc/datasets/UMDFaces/', \n                        help='Location of the folders containing 3 batches of UMDFaces stills.')\n    parser.add_argument('-o', '--output_path', \n                        default='/srv/data1/arunirc/datasets/UMDFaces/face_crops')\n    parser.add_argument('-n', '--num_val', type=int, default=2)\n    parser.add_argument('-b', '--batch', type=int, default=-1,\n                        help='crop faces of specified UMDFaces batch')\n    args = parser.parse_args()\n\n    # torch.manual_seed(1337)\n\n    if not osp.exists(args.output_path):\n        os.makedirs(args.output_path)   \n\n    if not osp.exists(osp.join(args.output_path, 'train')):\n        os.makedirs(osp.join(args.output_path, 'train')) \n\n    if not osp.exists(osp.join(args.output_path, 'val')):\n        os.makedirs(osp.join(args.output_path, 'val'))\n\n\n    # -----------------------------------------------------------------------------\n    # 1. Dataset\n    # -----------------------------------------------------------------------------\n    dir_batch = (\n                 osp.join(args.dataset_path, 'umdfaces_batch1'),\n                 osp.join(args.dataset_path, 'umdfaces_batch2'),\n                 osp.join(args.dataset_path, 'umdfaces_batch3'))\n\n    # dataset_batch = [datasets.ImageFolder(b) for b in dir_batch]\n\n    annot_files = (\n                   osp.join(dir_batch[0], 'umdfaces_batch1_ultraface.csv'),\n                   osp.join(dir_batch[1], 'umdfaces_batch2_ultraface.csv'),\n                   osp.join(dir_batch[2], 'umdfaces_batch3_ultraface.csv'))\n\n    for fn in annot_files:\n        assert osp.exists(fn)\n\n    if args.batch < 0:\n        # by default loop over batches in order\n        for i in range(len(dir_batch)):\n             crop_batch(dir_batch[i], annot_files[i], \n                        args.output_path, args.num_val)\n    else:\n        i = args.batch \n        crop_batch(dir_batch[i], annot_files[i], args.output_path, args.num_val)\n\n\n\n    # dataset_all = torch.utils.data.ConcatDataset(\n    #                 (dataset_batch1, dataset_batch2, dataset_batch3))\n    # for i in range(100):\n    #     pool.apply_async(f, (item,))\n\n\ndef crop_batch(data_dir, annot_fn, out_dir, nval):\n\n    dat = np.genfromtxt(annot_fn, names=True, delimiter=',', \n                        autostrip=True, dtype=None)\n    im_fn = dat['FILE']\n    (face_x, face_y, face_w, face_h) = (\n                                        dat['FACE_X'],\n                                        dat['FACE_Y'],\n                                        dat['FACE_WIDTH'],\n                                        dat['FACE_HEIGHT'])\n\n    class_ids = dat['SUBJECT_ID']\n\n    for c in tqdm(range(len(class_ids))):\n        sel = (class_ids==class_ids[c])\n\n        class_image_fn = im_fn[sel]\n\n        for i in xrange(len(class_image_fn)):\n            # print class_image_fn[i]\n            im = PIL.Image.open(osp.join(data_dir, class_image_fn[i]))\n            rect = (face_x[sel][i], face_y[sel][i], \n                    face_x[sel][i]+ face_w[sel][i], \n                    face_y[sel][i]+face_h[sel][i])        \n            imc = im.crop(rect)\n            \n            class_name, _ = osp.split(class_image_fn[i])\n\n            if not osp.exists(osp.join(out_dir, 'train', class_name)):\n                os.makedirs(osp.join(out_dir, 'train', class_name))\n\n            if i < len(class_image_fn)-nval:\n                imc.save(osp.join(out_dir, 'train', class_image_fn[i]))\n            else:\n                if not osp.exists(osp.join(out_dir, 'val', class_name)):\n                    os.makedirs(osp.join(out_dir, 'val', class_name))\n                imc.save(osp.join(out_dir, 'val', class_image_fn[i]))\n\n\n\n\nif __name__ == '__main__':\n    main()\n\n"
  },
  {
    "path": "umd-face/train_resnet_umdface.py",
    "content": "import argparse\nimport datetime\nimport os\nimport os.path as osp\nimport pytz\n\nimport torch\nimport torchvision\nfrom torchvision import models\nimport torch.nn as nn\nimport torch.optim\nimport torch.utils.data\nimport torchvision.transforms as transforms\nimport torchvision.datasets as datasets\nfrom torch.autograd import Variable\n\nimport yaml\nimport numpy as np\nimport matplotlib\nmatplotlib.use('Agg')\nimport matplotlib.pyplot as plt\n\nhere = osp.dirname(osp.abspath(__file__)) # output folder is located here\nroot_dir,_ = osp.split(here)\nimport sys\nsys.path.append(root_dir)\n\nimport train\nfrom config import configurations\nimport utils\n\n\n             \n\n\ndef main():\n    parser = argparse.ArgumentParser()\n    parser.add_argument('-e', '--exp_name', default='resnet_umdfaces')\n    parser.add_argument('-c', '--config', type=int, default=1,\n                        choices=configurations.keys())\n    parser.add_argument('-d', '--dataset_path', \n                        default='/srv/data1/arunirc/datasets/UMDFaces/face_crops')\n    parser.add_argument('-m', '--model_path', default=None, \n                        help='Initialize from pre-trained model')\n    parser.add_argument('--resume', help='Checkpoint path')\n    args = parser.parse_args()\n\n    # gpu = args.gpu\n    cfg = configurations[args.config]\n    out = get_log_dir(args.exp_name, args.config, cfg, verbose=False)\n    resume = args.resume\n\n    # os.environ['CUDA_VISIBLE_DEVICES'] = str(gpu)\n    cuda = torch.cuda.is_available()\n\n    torch.manual_seed(1337)\n    if cuda:\n        torch.cuda.manual_seed(1337)\n        torch.backends.cudnn.enabled = True\n        torch.backends.cudnn.benchmark = True # enable if all images are same size    \n\n\n\n    # -----------------------------------------------------------------------------\n    # 1. Dataset\n    # -----------------------------------------------------------------------------\n    #  Images should be arranged like this:\n    #   data_root/\n    #       class_1/....jpg..\n    #       class_2/....jpg.. \n    #       ......./....jpg.. \n    data_root = args.dataset_path\n    kwargs = {'num_workers': 4, 'pin_memory': True} if cuda else {}\n    RGB_MEAN = [ 0.485, 0.456, 0.406 ]\n    RGB_STD = [ 0.229, 0.224, 0.225 ]\n    \n    # Data transforms\n    # http://pytorch.org/docs/master/torchvision/transforms.html\n    train_transform = transforms.Compose([\n        transforms.Scale(256),  # smaller side resized\n        transforms.RandomCrop(224),\n        transforms.RandomHorizontalFlip(),\n        transforms.ToTensor(),\n        transforms.Normalize(mean = RGB_MEAN,\n                             std = RGB_STD),\n    ])\n    val_transform = transforms.Compose([\n        transforms.Scale((224,224)), \n        # transforms.CenterCrop(224),\n        transforms.ToTensor(),\n        transforms.Normalize(mean = RGB_MEAN,\n                             std = RGB_STD),\n    ])\n\n    # Data loaders - using PyTorch built-in objects\n    #   loader = DataLoaderClass(DatasetClass)\n    #   * `DataLoaderClass` is PyTorch provided torch.utils.data.DataLoader\n    #   * `DatasetClass` loads samples from a dataset; can be a standard class \n    #      provided by PyTorch (datasets.ImageFolder) or a custom-made class.\n    #      - More info: http://pytorch.org/docs/master/torchvision/datasets.html#imagefolder\n    #   *  Balanced class sampling: https://discuss.pytorch.org/t/balanced-sampling-between-classes-with-torchvision-dataloader/2703/3\n    traindir = osp.join(data_root, 'train')\n    dataset_train = datasets.ImageFolder(traindir, train_transform)\n    # For unbalanced dataset we create a weighted sampler                       \n    weights = utils.make_weights_for_balanced_classes(\n                dataset_train.imgs, len(dataset_train.classes))                                                                \n    weights = torch.DoubleTensor(weights)\n    sampler = torch.utils.data.sampler.WeightedRandomSampler(weights, len(weights))\n    train_loader = torch.utils.data.DataLoader(\n                    dataset_train, batch_size=cfg['batch_size'], \n                    sampler = sampler, **kwargs)\n\n    valdir = osp.join(data_root, 'val')\n    val_loader = torch.utils.data.DataLoader(\n                    datasets.ImageFolder(valdir, val_transform), \n                    batch_size=cfg['batch_size'], shuffle=False, **kwargs) \n\n    # print 'dataset classes:' + str(train_loader.dataset.classes)\n    num_class = len(train_loader.dataset.classes)\n    print 'Number of classes: %d' % num_class\n\n\n\n    # -----------------------------------------------------------------------------\n    # 2. Model\n    # -----------------------------------------------------------------------------\n    model = torchvision.models.resnet50(pretrained=True) # ImageNet pre-trained for quicker convergence\n\n    # Check if final fc layer sizes match num_class\n    if not model.fc.weight.size()[0] == num_class:\n        # Replace last layer\n        print model.fc\n        model.fc = torch.nn.Linear(2048, num_class)\n        print model.fc\n    else:\n        pass\n\n    # TODO - config options for DataParallel and device_ids\n    model = torch.nn.DataParallel(model, device_ids=[0, 1, 2, 3, 4])\n\n    if cuda:\n        model.cuda()\n\n    if args.model_path:\n        # If existing model is to be loaded from a file\n        checkpoint = torch.load(args.model_path)        \n        model.load_state_dict(checkpoint['model_state_dict'])\n\n    start_epoch = 0\n    start_iteration = 0\n\n    # Loss - cross entropy between predicted scores (unnormalized) and class labels (integers)\n    criterion = nn.CrossEntropyLoss()\n    if cuda:\n        criterion = criterion.cuda()\n\n    if resume:\n        # Resume training from last saved checkpoint\n        checkpoint = torch.load(resume)\n        model.load_state_dict(checkpoint['model_state_dict'])\n        start_epoch = checkpoint['epoch']\n        start_iteration = checkpoint['iteration']\n    else:\n        pass\n\n\n    # -----------------------------------------------------------------------------\n    # 3. Optimizer\n    # -----------------------------------------------------------------------------\n    params = filter(lambda p: p.requires_grad, model.parameters()) \n    # Parameters with p.requires_grad=False are not updated during training.\n    # This can be specified when defining the nn.Modules during model creation\n\n    if 'optim' in cfg.keys():\n    \tif cfg['optim'].lower()=='sgd':\n    \t\toptim = torch.optim.SGD(params,\n\t\t\t\t        lr=cfg['lr'],\n\t\t\t\t        momentum=cfg['momentum'],\n\t\t\t\t        weight_decay=cfg['weight_decay'])\n\n    \telif cfg['optim'].lower()=='adam':\n    \t\toptim = torch.optim.Adam(params,\n\t\t\t\t        lr=cfg['lr'], weight_decay=cfg['weight_decay'])\n\n    \telse:\n    \t\traise NotImplementedError('Optimizers: SGD or Adam')\n    else:\n\t    optim = torch.optim.SGD(params,\n\t\t\t        lr=cfg['lr'],\n\t\t\t        momentum=cfg['momentum'],\n\t\t\t        weight_decay=cfg['weight_decay'])\n\n    if resume:\n        optim.load_state_dict(checkpoint['optim_state_dict'])\n\n\n    # -----------------------------------------------------------------------------\n    # [optional] Sanity-check: forward pass with a single batch\n    # -----------------------------------------------------------------------------\n    DEBUG = False\n    if DEBUG:   \n        # model = model.cpu()\n        dataiter = iter(val_loader)\n        img, label = dataiter.next()\n\n        print 'Labels: ' + str(label.size()) # batchSize x num_class\n        print 'Input: ' + str(img.size())    # batchSize x 3 x 224 x 224\n\n        im = img.squeeze().numpy()\n        im = im[0,:,:,:]    # get first image in the batch\n        im = im.transpose((1,2,0)) # permute to 224x224x3\n        im = im * [ 0.229, 0.224, 0.225 ] # unnormalize\n        im = im + [ 0.485, 0.456, 0.406 ]\n        im[im<0] = 0\n\n        f = plt.figure()\n        plt.imshow(im)\n        plt.savefig('sanity-check-im.jpg')  # save transformed image in current folder\n\n        inputs = Variable(img)\n        if cuda:\n            inputs = inputs.cuda()\n\n        model.eval()\n        outputs = model(inputs)\n        print 'Network output: ' + str(outputs.size())        \n        model.train()\n        import pdb; pdb.set_trace()  # breakpoint c5e7c878 //\n\n\n    else:\n        pass\n\n\n    # -----------------------------------------------------------------------------\n    # 4. Training\n    # -----------------------------------------------------------------------------\n    trainer = train.Trainer(\n        cuda=cuda,\n        model=model,\n        criterion=criterion,\n        optimizer=optim,\n        init_lr=cfg['lr'],\n        lr_decay_epoch = cfg['lr_decay_epoch'],\n        train_loader=train_loader,\n        val_loader=val_loader,\n        out=out,\n        max_iter=cfg['max_iteration'],\n        interval_validate=cfg.get('interval_validate', len(train_loader)),\n    )\n\n    trainer.epoch = start_epoch\n    trainer.iteration = start_iteration\n    trainer.train()\n\n\n\ndef get_log_dir(model_name, config_id, cfg, verbose=True):\n    # Creates an output directory for each experiment, timestamped\n    name = 'MODEL-%s_CFG-%03d' % (model_name, config_id)\n    if verbose:\n        for k, v in cfg.items():\n            v = str(v)\n            if '/' in v:\n                continue\n            name += '_%s-%s' % (k.upper(), v)\n    now = datetime.datetime.now(pytz.timezone('US/Eastern'))\n    name += '_TIME-%s' % now.strftime('%Y%m%d-%H%M%S')\n    log_dir = osp.join(here, 'logs', name)\n    if not osp.exists(log_dir):\n        os.makedirs(log_dir)\n    with open(osp.join(log_dir, 'config.yaml'), 'w') as f:\n        yaml.safe_dump(cfg, f, default_flow_style=False)\n    return log_dir\n    \n\nif __name__ == '__main__':\n    main()\n\n"
  },
  {
    "path": "utils.py",
    "content": "from __future__ import division\n\nimport math\nimport warnings\n\ntry:\n    import cv2\nexcept ImportError:\n    cv2 = None\n\nimport numpy as np\nimport scipy.ndimage\nimport six\nimport skimage\nimport skimage.color\nfrom skimage import img_as_ubyte\nimport os\nimport os.path as osp\nimport matplotlib\nmatplotlib.use('agg')\nimport matplotlib.pyplot as plt\nimport csv\nimport scipy.signal\n\n\n\n\ndef make_weights_for_balanced_classes(images, nclasses):  \n    '''\n        Make a vector of weights for each image in the dataset, based \n        on class frequency. The returned vector of weights can be used \n        to create a WeightedRandomSampler for a DataLoader to have \n        class balancing when sampling for a training batch. \n            images - torchvisionDataset.imgs \n            nclasses - len(torchvisionDataset.classes)\n        https://discuss.pytorch.org/t/balanced-sampling-between-classes-with-torchvision-dataloader/2703/3                      \n    '''\n    count = [0] * nclasses                                                      \n    for item in images:                                                         \n        count[item[1]] += 1  # item is (img-data, label-id)\n    weight_per_class = [0.] * nclasses                                      \n    N = float(sum(count))  # total number of images                  \n    for i in range(nclasses):                                                   \n        weight_per_class[i] = N/float(count[i])                                 \n    weight = [0] * len(images)                                              \n    for idx, val in enumerate(images):                                          \n        weight[idx] = weight_per_class[val[1]]                                  \n    return weight\n\n\ndef get_vgg_class_counts(log_path):\n    '''  Dict of class frequencies from pre-computed text file  '''\n    data_1 = np.genfromtxt(log_path, dtype=None)\n    class_names = [x[0] for x in data_1]\n    class_counts = [x[1] for x in data_1]\n    class_count_dict = dict(zip(class_names, class_counts))\n    return class_count_dict\n\n\n\ndef plot_log_csv(log_path):\n    log_dir, _ = osp.split(log_path)\n    dat = np.genfromtxt(log_path, names=True, \n                        delimiter=',', autostrip=True)\n\n    train_loss =  dat['trainloss']\n    train_loss_sel = ~np.isnan(train_loss)\n    train_loss = train_loss[train_loss_sel]\n    iter_train_loss = dat['iteration'][train_loss_sel]\n\n    train_acc = dat['trainacc']\n    train_acc_sel = ~np.isnan(train_acc)\n    train_acc = train_acc[train_acc_sel]\n    iter_train_acc = dat['iteration'][train_acc_sel]\n\n    val_loss =  dat['validloss']\n    val_loss_sel = ~np.isnan(val_loss)\n    val_loss = val_loss[val_loss_sel]\n    iter_val_loss = dat['iteration'][val_loss_sel]\n\n    mean_iu = dat['validacc']\n    mean_iu_sel = ~np.isnan(mean_iu)\n    mean_iu = mean_iu[mean_iu_sel]\n    iter_mean_iu = dat['iteration'][mean_iu_sel]\n\n    fig, ax = plt.subplots(nrows=2, ncols=2)\n\n    plt.subplot(2, 2, 1)\n    plt.plot(iter_train_acc, train_acc, label='train')\n    plt.ylabel('accuracy')\n    plt.grid()\n    plt.legend()\n    plt.tight_layout()\n\n    plt.subplot(2, 2, 2)\n    plt.plot(iter_mean_iu, mean_iu, label='val')\n    plt.grid()\n    plt.legend()\n    plt.tight_layout()\n\n    plt.subplot(2, 2, 3)\n    plt.plot(iter_train_loss, train_loss, label='train')\n    plt.xlabel('iteration')\n    plt.ylabel('loss')\n    plt.grid()\n    plt.legend()\n    plt.tight_layout()\n\n    plt.subplot(2, 2, 4)\n    plt.plot(iter_val_loss, val_loss, label='val')\n    plt.xlabel('iteration')\n    plt.grid()\n    plt.legend()\n    plt.tight_layout()\n\n    plt.savefig(osp.join(log_dir, 'log_plots.png'), bbox_inches='tight')\n\n\n\n\n\ndef plot_log(log_path):\n    log_dir, _ = osp.split(log_path)\n    epoch = []\n    iteration = []\n    train_loss = []\n    train_acc = []\n    val_loss = []\n    val_acc = []\n    g = lambda x: x if x!='' else float('nan')\n    reader = csv.reader( open(log_path, 'rb'))\n    next(reader)  # Skip header row.\n    for line in reader:\n        line_fields = [g(x) for x in line]\n        epoch.append(float(line_fields[0]))\n        iteration.append(float(line_fields[1]))\n        train_loss.append(float(line_fields[2]))\n        train_acc.append(float(line_fields[3]))\n        val_loss.append(float(line_fields[4]))\n        val_acc.append(float(line_fields[5]))\n\n    epoch = np.array(epoch)\n    iteration = np.array(iteration)\n    train_loss = np.array(train_loss)\n    train_acc = np.array(train_acc)\n    val_loss = np.array(val_loss)\n    val_acc = np.array(val_acc)\n\n    train_loss_sel = ~np.isnan(train_loss)\n    train_loss = train_loss[train_loss_sel]\n    iter_train_loss = iteration[train_loss_sel]\n\n    train_acc_sel = ~np.isnan(train_acc)\n    train_acc = train_acc[train_acc_sel]\n    iter_train_acc = iteration[train_acc_sel]\n\n    val_loss_sel = ~np.isnan(val_loss)\n    val_loss = val_loss[val_loss_sel]\n    iter_val_loss = iteration[val_loss_sel]\n\n    val_acc_sel = ~np.isnan(val_acc)\n    val_acc = val_acc[val_acc_sel]\n    iter_val_acc = iteration[val_acc_sel]\n\n    fig, ax = plt.subplots(nrows=2, ncols=2)\n\n    plt.subplot(2, 2, 1)\n    plt.plot(iter_train_acc, train_acc, label='train', alpha=0.5, color='C0')\n    box_pts = np.rint(np.sqrt(len(train_acc))).astype(np.int)\n    plt.plot(iter_train_acc, savgol_smooth(train_acc, box_pts), color='C0')\n    plt.ylabel('accuracy')\n    plt.grid()\n    plt.legend()\n    plt.title('Training')\n    plt.tight_layout()\n\n    plt.subplot(2, 2, 2)\n    plt.plot(iter_val_acc, val_acc, label='val', alpha=0.5, color='C1')\n    box_pts = np.rint(np.sqrt(len(val_acc))).astype(np.int)\n    plt.plot(iter_val_acc, savgol_smooth(val_acc, box_pts), color='C1')\n    plt.grid()\n    plt.legend()\n    plt.title('Validation')\n    plt.tight_layout()\n\n    plt.subplot(2, 2, 3)\n    plt.plot(iter_train_loss, train_loss, label='train', alpha=0.5, color='C0')\n    box_pts = np.rint(np.sqrt(len(train_loss))).astype(np.int)\n    plt.plot(iter_train_loss, savgol_smooth(train_loss, box_pts), color='C0')\n    plt.xlabel('iteration')\n    plt.ylabel('loss')\n    plt.grid()\n    plt.legend()\n    plt.tight_layout()\n\n    plt.subplot(2, 2, 4)\n    plt.plot(iter_val_loss, val_loss, label='val', alpha=0.5, color='C1')\n    box_pts = np.rint(np.sqrt(len(val_loss))).astype(np.int)\n    plt.plot(iter_val_loss, savgol_smooth(val_loss, box_pts), color='C1')\n    plt.xlabel('iteration')\n    plt.grid()\n    plt.legend()\n    plt.tight_layout()\n\n    plt.savefig(osp.join(log_dir, 'log_plots.png'), bbox_inches='tight')\n\n\ndef savgol_smooth(y, box_pts):\n    # use the Savitzky-Golay filter for 1-D smoothing\n    if box_pts % 2 == 0:\n        box_pts += 1\n    y_smooth = scipy.signal.savgol_filter(y, box_pts, 2)\n    return y_smooth\n\n# -----------------------------------------------------------------------------\n#   LFW helper code from FaceNet: https://github.com/davidsandberg/facenet\n# ----------------------------------------------------------------------------- \n\n# MIT License\n# \n# Copyright (c) 2016 David Sandberg\n# \n# Permission is hereby granted, free of charge, to any person obtaining a copy\n# of this software and associated documentation files (the \"Software\"), to deal\n# in the Software without restriction, including without limitation the rights\n# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell\n# copies of the Software, and to permit persons to whom the Software is\n# furnished to do so, subject to the following conditions:\n# \n# The above copyright notice and this permission notice shall be included in all\n# copies or substantial portions of the Software.\n# \n# THE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\n# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\n# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\n# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\n# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\n# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\n# SOFTWARE.\ndef get_paths(lfw_dir, pairs, file_ext):\n    nrof_skipped_pairs = 0\n    path_list = []\n    issame_list = []\n    for pair in pairs:\n        if len(pair) == 3:\n            path0 = os.path.join(lfw_dir, pair[0], pair[0] + '_' + '%04d' % int(pair[1])+'.'+file_ext)\n            path1 = os.path.join(lfw_dir, pair[0], pair[0] + '_' + '%04d' % int(pair[2])+'.'+file_ext)\n            issame = True\n        elif len(pair) == 4:\n            path0 = os.path.join(lfw_dir, pair[0], pair[0] + '_' + '%04d' % int(pair[1])+'.'+file_ext)\n            path1 = os.path.join(lfw_dir, pair[2], pair[2] + '_' + '%04d' % int(pair[3])+'.'+file_ext)\n            issame = False\n        if os.path.exists(path0) and os.path.exists(path1):    # Only add the pair if both paths exist\n            path_list += (path0,path1)\n            issame_list.append(issame)\n        else:\n            nrof_skipped_pairs += 1\n    if nrof_skipped_pairs>0:\n        print('Skipped %d image pairs' % nrof_skipped_pairs)\n    \n    return path_list, issame_list\n\n\ndef read_pairs(pairs_filename, lfw_flag=True):\n    pairs = []\n    with open(pairs_filename, 'r') as f:\n        if lfw_flag:\n            for line in f.readlines()[1:]:\n                pair = line.strip().split()\n                pairs.append(pair)\n        else:\n            for line in f.readlines():\n                pair = line.strip().split()\n                pairs.append(pair)      \n    return np.array(pairs)\n\n\n# -----------------------------------------------------------------------------\n#   IJB-A helper code\n# ----------------------------------------------------------------------------- \ndef get_ijba_1_1_metadata(protocol_file):\n    metadata  = {}\n    template_id = []\n    subject_id = []\n    img_filename = []\n    media_id = []\n    sighting_id = []\n\n    with open(protocol_file, 'r') as f:\n        for line in f.readlines()[1:]:\n            line_fields = line.strip().split(',')            \n            template_id.append(int(line_fields[0]))\n            subject_id.append(int(line_fields[1]))\n            img_filename.append(line_fields[2])\n            media_id.append(int(line_fields[3]))\n            sighting_id.append(int(line_fields[4]))\n\n    metadata['template_id'] = np.array(template_id)\n    metadata['subject_id'] = np.array(subject_id)\n    metadata['img_filename'] = np.array(img_filename)\n    metadata['media_id'] = np.array(media_id)\n    metadata['sighting_id'] = np.array(sighting_id)\n    return metadata\n\n\ndef read_ijba_pairs(pairs_filename):\n    pairs = []\n    with open(pairs_filename, 'r') as f:\n        for line in f.readlines():\n            pair = line.strip().split(',')\n            pairs.append(pair)      \n    return np.array(pairs).astype(np.int)\n\n"
  },
  {
    "path": "vgg-face-2/calculate_image_mean.py",
    "content": "import argparse\nimport os\nimport os.path as osp\n\nimport torch\nimport torchvision\nimport torch.utils.data\nimport torchvision.transforms as transforms\nimport torchvision.datasets as datasets\n\nimport yaml\nimport tqdm\nimport numpy as np\nimport matplotlib.pyplot as plt\n\nhere = osp.dirname(osp.abspath(__file__)) # output folder is located here\nroot_dir,_ = osp.split(here)\nimport sys\nsys.path.append(root_dir)\n\n\n\n'''\nCalculate mean R, G, B values for VGGFace2 dataset\n--------------------------------------------------\nFollowing implementation: [VGGFace2](https://arxiv.org/pdf/1710.08092.pdf)\n'''\n\ndef main():\n    parser = argparse.ArgumentParser()\n    parser.add_argument('-d', '--dataset_path', \n                        default='/srv/data1/arunirc/datasets/vggface2')\n    args = parser.parse_args()\n\n\n    # os.environ['CUDA_VISIBLE_DEVICES'] = str(gpu)\n    cuda = torch.cuda.is_available()\n\n    torch.manual_seed(1337)\n    if cuda:\n        torch.cuda.manual_seed(1337)\n        torch.backends.cudnn.enabled = True\n        torch.backends.cudnn.benchmark = True # enable if all images are same size\n\n    # -----------------------------------------------------------------------------\n    # 1. Dataset\n    # ----------------------------------------------------------------------------- \n    data_root = args.dataset_path\n    kwargs = {'num_workers': 4, 'pin_memory': True} if cuda else {}\n    \n    # Data transforms\n    train_transform = transforms.Compose([\n        transforms.Scale(256),  # smaller side resized\n        transforms.RandomCrop(224),\n        transforms.RandomHorizontalFlip(),\n        # transforms.RandomGrayscale(p=0.2),\n        transforms.ToTensor()\n    ])\n\n    # Data loaders\n    traindir = osp.join(data_root, 'train')\n    dataset = datasets.ImageFolder(traindir, train_transform)\n    train_loader = torch.utils.data.DataLoader(\n                    dataset, shuffle=True, batch_size=128, **kwargs)\n    \n    rgb_mean = []\n    for batch_idx, (images, lbl) in tqdm.tqdm( enumerate(train_loader), \n                                        total=len(train_loader), \n                                        desc='Sampling images' ): \n        rgb_mean.append( ((images.mean(dim=0)).mean(dim=-1)).mean(dim=-1) )\n        if batch_idx == 100:\n            break\n\n    print len(rgb_mean)\n    rgb_mean = torch.mean( torch.stack(rgb_mean, dim=1), dim=1)\n    print rgb_mean\n\n    res = {}\n    res['R'] = rgb_mean[0]\n    res['G'] = rgb_mean[1]\n    res['B'] = rgb_mean[2]\n\n    with open(osp.join(here, 'mean_rgb.yaml'), 'w') as f:\n        yaml.dump(res, f, default_flow_style=False)\n\n\n\n\nif __name__ == '__main__':\n    main()\n\n"
  },
  {
    "path": "vgg-face-2/class_counts.sh",
    "content": "\n\n# Utility script to get class frequencies of VGGFace2 based on #images in class-folder\n\n# >>>>>> Set paths here <<<<<<\nDATA_PATH=/home/renyi/arunirc/data1/datasets/vggface2/train\nDATA_PATH=/home/renyi/arunirc/data1/datasets/vggface2/train-crop\ntouch vgg-face-2/vggface_class_counts.txt\n\nfor folder in `ls ${DATA_PATH}`; do\n    img_count=`ls ${DATA_PATH}/${folder} | wc -l`\n    echo ${folder}' '${img_count}\n    echo ${folder}' '${img_count} >> vgg-face-2/vggface_class_counts.txt\ndone\n"
  },
  {
    "path": "vgg-face-2/crop_face.sh",
    "content": "\n\n# Utility script to crop large image datasets from the console\n\n# INFO: kill all spawned sub-processes: kill $(ps -s $$ -o pid=)\n\ncrop_image_multi () {\n    DATA_PATH=$1\n    OUT_PATH=$2\n    ANNOT_FILE=$3\n    SUBJECT_FOLDER=$4\n\n    OLDIFS=$IFS\n    IFS=,\n    [ ! -f $ANNOT_FILE ] && { echo \"$ANNOT_FILE file not found\"; exit 99; }\n    \n    COUNT=0\n    while read flname sid xmin ymin width height\n    do\n        IFS='/' read -r -a array <<< \"$flname\"\n        SUBJECT_DIR=\"${array[-2]}\"\n\n        if [ \"${SUBJECT_DIR}\" == \"${SUBJECT_FOLDER}\" ]; then\n            mkdir -p ${OUT_PATH}/${SUBJECT_DIR}\n            IMG_PATH=${DATA_PATH}/${array[-2]}/${array[-1]}\n            IMG_OUT=${OUT_PATH}/${array[-2]}/${array[-1]}\n            convert -crop ${width}x${height}+${xmin}+${ymin} ${IMG_PATH} ${IMG_OUT}\n        fi\n\n    done < $ANNOT_FILE\n    IFS=$OLDIFS\n}\n\n\ncreate_val_dir () {\n# For each subject's folder in train_data, move 2 images into val_folder\n    TRAIN_DATA=$1\n    VAL_DATA=$2\n    mkdir -p ${VAL_DATA}\n    for folder in `ls ${TRAIN_DATA}`; do\n        echo $folder\n        mkdir -p ${VAL_DATA}/${folder}\n        for filename in `ls ${OUT_PATH}/${folder} | tail -n 2`; do\n            mv ${OUT_PATH}/${folder}/${filename} ${VAL_DATA}/${folder}\n        done\n    done\n}\n\n\ncrop_image () {\n    DATA_PATH=$1\n    OUT_PATH=$2\n    ANNOT_FILE=$3\n\n    OLDIFS=$IFS\n    IFS=,\n    [ ! -f $ANNOT_FILE ] && { echo \"$ANNOT_FILE file not found\"; exit 99; }\n    \n    COUNT=0\n    while read flname sid xmin ymin width height\n    do\n        IFS='/' read -r -a array <<< \"$flname\"\n        SUBJECT_DIR=\"${array[-2]}\"\n        mkdir -p ${OUT_PATH}/${SUBJECT_DIR}\n        IMG_PATH=${DATA_PATH}/${array[-2]}/${array[-1]}\n        IMG_OUT=${OUT_PATH}/${array[-2]}/${array[-1]}\n        convert -crop ${width}x${height}+${xmin}+${ymin} ${IMG_PATH} ${IMG_OUT}\n\n    done < $ANNOT_FILE\n    IFS=$OLDIFS\n}\n\n\n\n\n\n# >>>>>> Set paths here <<<<<<\nDATA_PATH=/home/renyi/arunirc/data1/datasets/vggface2/train\nOUT_PATH=/home/renyi/arunirc/data1/datasets/vggface2/train-crop\nANNOT_FILE=/home/renyi/arunirc/data1/datasets/vggface2/vggface2_disk1.csv\n# Annotations format: filename subject_id xmin ymin width height\nVAL_PATH=/home/renyi/arunirc/data1/datasets/vggface2/val-crop\n\n\n# crop faces out of images and save into \"train-crop\" output folder\nmkdir -p ${OUT_PATH}\ndate\n# crop_image ${DATA_PATH} ${OUT_PATH} ${ANNOT_FILE}\ndate\necho \"Done cropping\"\n\n# take 2 face images per subject and save into \"val-crop\" folder\ncreate_val_dir ${OUT_PATH} ${VAL_PATH}\necho \"Done creating validation set\"\n\n\n\n\n# COUNT=0\n# WAIT_COUNT=0\n# MAXPROG=`ls ${DATA_PATH} | wc -l`\n\n# for folder in `ls ${DATA_PATH}`; do\n#     # show progress\n#     ((WAIT_COUNT += 1))\n#     ((COUNT += 1))\n#     echo ${COUNT}\n#     PROGRESS=`echo ${COUNT}*100/${MAXPROG}|bc -l`\n#     echo -n \"${PROGRESS} %     \"\n#     # echo -n \"$((${COUNT}*100/${MAXPROG})) %     \"\n#     echo -n R | tr 'R' '\\r'\n#     # if [ ${WAIT_COUNT} -eq 20 ]; then\n#     #     # don't spawn more than 20 processes at a time\n#     #     echo 'waiting for 20 processes to finish'\n#     #     echo \"Progress: $((${COUNT}*100/${MAXPROG})) %     \"\n#     #     wait\n#     #     ((WAIT_COUNT=0))\n#     # fi\n\n#     # if [ ${COUNT} -ge 20 ]; then\n#     #     break\n#     # fi\n# done\n\n# wait\n# echo \"Done cropping\"\n\n# mkdir -p ${OUT_PATH}\n# date\n# crop_image ${DATA_PATH} ${OUT_PATH} ${ANNOT_FILE}\n# date"
  },
  {
    "path": "vgg-face-2/mean_rgb.yaml",
    "content": "B: 0.36426129937171936\nG: 0.406549334526062\nR: 0.49698397517204285\n"
  },
  {
    "path": "vgg-face-2/train_resnet101_vggface.py",
    "content": "import argparse\nimport datetime\nimport os\nimport os.path as osp\nimport pytz\n\nimport torch\nimport torchvision\nfrom torchvision import models\nimport torch.nn as nn\nimport torch.optim\nimport torch.utils.data\nimport torchvision.transforms as transforms\nimport torchvision.datasets as datasets\nfrom torch.autograd import Variable\n\nimport yaml\nimport numpy as np\nimport matplotlib\nmatplotlib.use('Agg')\nimport matplotlib.pyplot as plt\n\nhere = osp.dirname(osp.abspath(__file__)) # output folder is located here\nroot_dir,_ = osp.split(here)\nimport sys\nsys.path.append(root_dir)\n\nimport train\nimport models\nimport utils\nfrom config import configurations\n\n\n\n\ndef main():\n    parser = argparse.ArgumentParser()\n    parser.add_argument('-e', '--exp_name', default='resnet101_vggface_scratch')\n    parser.add_argument('-c', '--config', type=int, default=1,\n                        choices=configurations.keys())\n    parser.add_argument('-d', '--dataset_path', \n                        default='/srv/data1/arunirc/datasets/vggface2')\n    parser.add_argument('-m', '--model_path', default=None, \n                        help='Initialize from pre-trained model')\n    parser.add_argument('--resume', help='Checkpoint path')\n    parser.add_argument('--bottleneck', action='store_true', default=False,\n                        help='Add a 512-dim bottleneck layer with L2 normalization')\n    args = parser.parse_args()\n\n    # gpu = args.gpu\n    cfg = configurations[args.config]\n    out = get_log_dir(args.exp_name, args.config, cfg, verbose=False)\n    resume = args.resume\n\n    # os.environ['CUDA_VISIBLE_DEVICES'] = str(gpu)\n    cuda = torch.cuda.is_available()\n\n    torch.manual_seed(1337)\n    if cuda:\n        torch.cuda.manual_seed(1337)\n        torch.backends.cudnn.enabled = True\n        torch.backends.cudnn.benchmark = True # enable if all images are same size    \n\n\n\n    # -----------------------------------------------------------------------------\n    # 1. Dataset\n    # -----------------------------------------------------------------------------\n    #  Images should be arranged like this:\n    #   data_root/\n    #       class_1/....jpg..\n    #       class_2/....jpg.. \n    #       ......./....jpg.. \n    data_root = args.dataset_path\n    kwargs = {'num_workers': 4, 'pin_memory': True} if cuda else {}\n    RGB_MEAN = [ 0.485, 0.456, 0.406 ]\n    RGB_STD = [ 0.229, 0.224, 0.225 ]\n    \n    # Data transforms\n    # http://pytorch.org/docs/master/torchvision/transforms.html\n    train_transform = transforms.Compose([\n        transforms.Scale(256),  # smaller side resized\n        transforms.RandomCrop(224),\n        transforms.RandomHorizontalFlip(),\n        transforms.ToTensor(),\n        transforms.Normalize(mean = RGB_MEAN,\n                             std = RGB_STD),\n    ])\n    val_transform = transforms.Compose([\n        transforms.Scale(256), \n        transforms.CenterCrop(224),\n        transforms.ToTensor(),\n        transforms.Normalize(mean = RGB_MEAN,\n                             std = RGB_STD),\n    ])\n\n    # Data loaders - using PyTorch built-in objects\n    #   loader = DataLoaderClass(DatasetClass)\n    #   * `DataLoaderClass` is PyTorch provided torch.utils.data.DataLoader\n    #   * `DatasetClass` loads samples from a dataset; can be a standard class \n    #     provided by PyTorch (datasets.ImageFolder) or a custom-made class.\n    #      - More info: http://pytorch.org/docs/master/torchvision/datasets.html#imagefolder\n    traindir = osp.join(data_root, 'train')\n    dataset_train = datasets.ImageFolder(traindir, train_transform)\n    \n    # For unbalanced dataset we create a weighted sampler\n    #   *  Balanced class sampling: https://discuss.pytorch.org/t/balanced-sampling-between-classes-with-torchvision-dataloader/2703/3                     \n    weights = utils.make_weights_for_balanced_classes(\n                dataset_train.imgs, len(dataset_train.classes))                                                                \n    weights = torch.DoubleTensor(weights)\n    sampler = torch.utils.data.sampler.WeightedRandomSampler(weights, len(weights))\n\n    train_loader = torch.utils.data.DataLoader(\n                    dataset_train, batch_size=cfg['batch_size'], \n                    sampler = sampler, **kwargs)\n\n    valdir = osp.join(data_root, 'val-crop')\n    val_loader = torch.utils.data.DataLoader(\n                    datasets.ImageFolder(valdir, val_transform), \n                    batch_size=cfg['batch_size'], shuffle=False, **kwargs) \n\n    # print 'dataset classes:' + str(train_loader.dataset.classes)\n    num_class = len(train_loader.dataset.classes)\n    print 'Number of classes: %d' % num_class\n\n\n\n    # -----------------------------------------------------------------------------\n    # 2. Model\n    # -----------------------------------------------------------------------------\n    model = torchvision.models.resnet101(pretrained=False)\n\n    if type(model.fc) == torch.nn.modules.linear.Linear:\n        # Check if final fc layer sizes match num_class\n        if not model.fc.weight.size()[0] == num_class:\n            # Replace last layer\n            print model.fc\n            model.fc = torch.nn.Linear(2048, num_class)\n            print model.fc\n        else:\n            pass\n    else:\n        pass    \n\n\n    if args.model_path:\n        # If existing model is to be loaded from a file\n        checkpoint = torch.load(args.model_path) \n\n        if checkpoint['arch'] == 'DataParallel':\n            # if we trained and saved our model using DataParallel\n            model = torch.nn.DataParallel(model, device_ids=[0, 1, 2, 3, 4, 5, 6, 7])\n            model.load_state_dict(checkpoint['model_state_dict'])\n            model = model.module # get network module from inside its DataParallel wrapper\n        else:\n            model.load_state_dict(checkpoint['model_state_dict'])\n    \n    # Optionally add a \"bottleneck + L2-norm\" layer after GAP-layer\n    # TODO -- loading a bottleneck model might be a problem .... do some unit-tests\n    if args.bottleneck:\n        layers = []\n        layers.append(torch.nn.Linear(2048, 512))\n        layers.append(nn.BatchNorm2d(512))\n        layers.append(torch.nn.ReLU(inplace=True))\n        layers.append(models.NormFeat()) # L2-normalization layer\n        layers.append(torch.nn.Linear(512, num_class))\n        model.fc = torch.nn.Sequential(*layers)\n\n    # TODO - config options for DataParallel and device_ids\n    model = torch.nn.DataParallel(model, device_ids=[0, 1, 2, 3, 4, 5, 6, 7])\n\n    if cuda:\n        model.cuda()  \n\n    start_epoch = 0\n    start_iteration = 0\n\n    # Loss - cross entropy between predicted scores (unnormalized) and class labels (integers)\n    criterion = nn.CrossEntropyLoss()\n    if cuda:\n        criterion = criterion.cuda()\n\n    if resume:\n        # Resume training from last saved checkpoint\n        checkpoint = torch.load(resume)\n        model.load_state_dict(checkpoint['model_state_dict'])\n        start_epoch = checkpoint['epoch']\n        start_iteration = checkpoint['iteration']\n    else:\n        pass\n\n\n    # -----------------------------------------------------------------------------\n    # 3. Optimizer\n    # -----------------------------------------------------------------------------\n    params = filter(lambda p: p.requires_grad, model.parameters()) \n    # Parameters with p.requires_grad=False are not updated during training.\n    # This can be specified when defining the nn.Modules during model creation\n\n    if 'optim' in cfg.keys():\n        if cfg['optim'].lower()=='sgd':\n            optim = torch.optim.SGD(params,\n                        lr=cfg['lr'],\n                        momentum=cfg['momentum'],\n                        weight_decay=cfg['weight_decay'])\n\n        elif cfg['optim'].lower()=='adam':\n            optim = torch.optim.Adam(params,\n                        lr=cfg['lr'], weight_decay=cfg['weight_decay'])\n\n        else:\n            raise NotImplementedError('Optimizers: SGD or Adam')\n    else:\n        optim = torch.optim.SGD(params,\n                    lr=cfg['lr'],\n                    momentum=cfg['momentum'],\n                    weight_decay=cfg['weight_decay'])\n\n    if resume:\n        optim.load_state_dict(checkpoint['optim_state_dict'])\n\n\n    # -----------------------------------------------------------------------------\n    # [optional] Sanity-check: forward pass with a single batch\n    # -----------------------------------------------------------------------------\n    DEBUG = False\n    if DEBUG:   \n        # model = model.cpu()\n        dataiter = iter(val_loader)\n        img, label = dataiter.next()\n\n        print 'Labels: ' + str(label.size()) # batchSize x num_class\n        print 'Input: ' + str(img.size())    # batchSize x 3 x 224 x 224\n\n        im = img.squeeze().numpy()\n        im = im[0,:,:,:]    # get first image in the batch\n        im = im.transpose((1,2,0)) # permute to 224x224x3\n        im = im * [ 0.229, 0.224, 0.225 ] # unnormalize\n        im = im + [ 0.485, 0.456, 0.406 ]\n        im[im<0] = 0\n\n        f = plt.figure()\n        plt.imshow(im)\n        plt.savefig('sanity-check-im.jpg')  # save transformed image in current folder\n        inputs = Variable(img)\n        if cuda:\n            inputs = inputs.cuda()\n\n        model.eval()\n        outputs = model(inputs)\n        print 'Network output: ' + str(outputs.size())        \n        model.train()\n\n    else:\n        pass\n\n\n    # -----------------------------------------------------------------------------\n    # 4. Training\n    # -----------------------------------------------------------------------------\n    trainer = train.Trainer(\n        cuda=cuda,\n        model=model,\n        criterion=criterion,\n        optimizer=optim,\n        init_lr=cfg['lr'],\n        lr_decay_epoch = cfg['lr_decay_epoch'],\n        train_loader=train_loader,\n        val_loader=val_loader,\n        out=out,\n        max_iter=cfg['max_iteration'],\n        interval_validate=cfg.get('interval_validate', len(train_loader)),\n    )\n\n    trainer.epoch = start_epoch\n    trainer.iteration = start_iteration\n    trainer.train()\n\n\n\ndef get_log_dir(model_name, config_id, cfg, verbose=True):\n    # Creates an output directory for each experiment, timestamped\n    name = 'MODEL-%s_CFG-%03d' % (model_name, config_id)\n    if verbose:\n        for k, v in cfg.items():\n            v = str(v)\n            if '/' in v:\n                continue\n            name += '_%s-%s' % (k.upper(), v)\n    now = datetime.datetime.now(pytz.timezone('US/Eastern'))\n    name += '_TIME-%s' % now.strftime('%Y%m%d-%H%M%S')\n    log_dir = osp.join(here, 'logs', name)\n    if not osp.exists(log_dir):\n        os.makedirs(log_dir)\n    with open(osp.join(log_dir, 'config.yaml'), 'w') as f:\n        yaml.safe_dump(cfg, f, default_flow_style=False)\n    return log_dir\n\n\n\n\nif __name__ == '__main__':\n    main()\n\n"
  },
  {
    "path": "vgg-face-2/train_resnet50_vggface_scratch.py",
    "content": "import argparse\nimport datetime\nimport os\nimport os.path as osp\nimport pytz\n\nimport torch\nimport torchvision\nfrom torchvision import models\nimport torch.nn as nn\nimport torch.optim\nimport torch.utils.data\nimport torchvision.transforms as transforms\nimport torchvision.datasets as datasets\nfrom torch.autograd import Variable\n\nimport yaml\nimport numpy as np\nimport matplotlib\nmatplotlib.use('Agg')\nimport matplotlib.pyplot as plt\n\nhere = osp.dirname(osp.abspath(__file__)) # output folder is located here\nroot_dir,_ = osp.split(here)\nimport sys\nsys.path.append(root_dir)\n\nimport train\nimport models\nimport utils\nfrom config import configurations\n\n\n\n\ndef main():\n    parser = argparse.ArgumentParser()\n    parser.add_argument('-e', '--exp_name', default='resnet50_vggface')\n    parser.add_argument('-c', '--config', type=int, default=1,\n                        choices=configurations.keys())\n    parser.add_argument('-d', '--dataset_path', \n                        default='/srv/data1/arunirc/datasets/vggface2')\n    parser.add_argument('-m', '--model_path', default=None, \n                        help='Initialize from pre-trained model')\n    parser.add_argument('--resume', help='Checkpoint path')\n    parser.add_argument('--bottleneck', action='store_true', default=False,\n                        help='Add a 512-dim bottleneck layer with L2 normalization')\n    args = parser.parse_args()\n\n    # gpu = args.gpu\n    cfg = configurations[args.config]\n    out = get_log_dir(args.exp_name, args.config, cfg, verbose=False)\n    resume = args.resume\n\n    # os.environ['CUDA_VISIBLE_DEVICES'] = str(gpu)\n    cuda = torch.cuda.is_available()\n\n    torch.manual_seed(1337)\n    if cuda:\n        torch.cuda.manual_seed(1337)\n        torch.backends.cudnn.enabled = True\n        torch.backends.cudnn.benchmark = True # enable if all images are same size    \n\n\n\n    # -----------------------------------------------------------------------------\n    # 1. Dataset\n    # -----------------------------------------------------------------------------\n    #  Images should be arranged like this:\n    #   data_root/\n    #       class_1/....jpg..\n    #       class_2/....jpg.. \n    #       ......./....jpg.. \n    data_root = args.dataset_path\n    kwargs = {'num_workers': 4, 'pin_memory': True} if cuda else {}\n    RGB_MEAN = [ 0.485, 0.456, 0.406 ]\n    RGB_STD = [ 0.229, 0.224, 0.225 ]\n    \n    # Data transforms\n    # http://pytorch.org/docs/master/torchvision/transforms.html\n    train_transform = transforms.Compose([\n        transforms.Scale(256),  # smaller side resized\n        transforms.RandomCrop(224),\n        transforms.RandomHorizontalFlip(),\n        transforms.ToTensor(),\n        transforms.Normalize(mean = RGB_MEAN,\n                             std = RGB_STD),\n    ])\n    val_transform = transforms.Compose([\n        transforms.Scale(256), \n        transforms.CenterCrop(224),\n        transforms.ToTensor(),\n        transforms.Normalize(mean = RGB_MEAN,\n                             std = RGB_STD),\n    ])\n\n    # Data loaders - using PyTorch built-in objects\n    #   loader = DataLoaderClass(DatasetClass)\n    #   * `DataLoaderClass` is PyTorch provided torch.utils.data.DataLoader\n    #   * `DatasetClass` loads samples from a dataset; can be a standard class \n    #     provided by PyTorch (datasets.ImageFolder) or a custom-made class.\n    #      - More info: http://pytorch.org/docs/master/torchvision/datasets.html#imagefolder\n    traindir = osp.join(data_root, 'train')\n    dataset_train = datasets.ImageFolder(traindir, train_transform)\n    \n    # For unbalanced dataset we create a weighted sampler\n    #   *  Balanced class sampling: https://discuss.pytorch.org/t/balanced-sampling-between-classes-with-torchvision-dataloader/2703/3                     \n    weights = utils.make_weights_for_balanced_classes(\n                dataset_train.imgs, len(dataset_train.classes))                                                                \n    weights = torch.DoubleTensor(weights)\n    sampler = torch.utils.data.sampler.WeightedRandomSampler(weights, len(weights))\n\n    train_loader = torch.utils.data.DataLoader(\n                    dataset_train, batch_size=cfg['batch_size'], \n                    sampler = sampler, **kwargs)\n\n    valdir = osp.join(data_root, 'val-crop')\n    val_loader = torch.utils.data.DataLoader(\n                    datasets.ImageFolder(valdir, val_transform), \n                    batch_size=cfg['batch_size'], shuffle=False, **kwargs) \n\n    # print 'dataset classes:' + str(train_loader.dataset.classes)\n    num_class = len(train_loader.dataset.classes)\n    print 'Number of classes: %d' % num_class\n\n\n\n    # -----------------------------------------------------------------------------\n    # 2. Model\n    # -----------------------------------------------------------------------------\n    model = torchvision.models.resnet50(pretrained=False)\n\n    if type(model.fc) == torch.nn.modules.linear.Linear:\n        # Check if final fc layer sizes match num_class\n        if not model.fc.weight.size()[0] == num_class:\n            # Replace last layer\n            print model.fc\n            model.fc = torch.nn.Linear(2048, num_class)\n            print model.fc\n        else:\n            pass\n    else:\n        pass    \n\n\n    if args.model_path:\n        # If existing model is to be loaded from a file\n        checkpoint = torch.load(args.model_path) \n\n        if checkpoint['arch'] == 'DataParallel':\n            # if we trained and saved our model using DataParallel\n            model = torch.nn.DataParallel(model, device_ids=[0, 1, 2, 3, 4, 5, 6, 7])\n            model.load_state_dict(checkpoint['model_state_dict'])\n            model = model.module # get network module from inside its DataParallel wrapper\n        else:\n            model.load_state_dict(checkpoint['model_state_dict'])\n    \n    # Optionally add a \"bottleneck + L2-norm\" layer after GAP-layer\n    # TODO -- loading a bottleneck model might be a problem .... do some unit-tests\n    if args.bottleneck:\n        layers = []\n        layers.append(torch.nn.Linear(2048, 512))\n        layers.append(nn.BatchNorm2d(512))\n        layers.append(torch.nn.ReLU(inplace=True))\n        layers.append(models.NormFeat()) # L2-normalization layer\n        layers.append(torch.nn.Linear(512, num_class))\n        model.fc = torch.nn.Sequential(*layers)\n\n    # TODO - config options for DataParallel and device_ids\n    model = torch.nn.DataParallel(model, device_ids=[0, 1, 2, 3, 4, 5, 6, 7])\n\n    if cuda:\n        model.cuda()  \n\n    start_epoch = 0\n    start_iteration = 0\n\n    # Loss - cross entropy between predicted scores (unnormalized) and class labels (integers)\n    criterion = nn.CrossEntropyLoss()\n    if cuda:\n        criterion = criterion.cuda()\n\n    if resume:\n        # Resume training from last saved checkpoint\n        checkpoint = torch.load(resume)\n        model.load_state_dict(checkpoint['model_state_dict'])\n        start_epoch = checkpoint['epoch']\n        start_iteration = checkpoint['iteration']\n    else:\n        pass\n\n\n    # -----------------------------------------------------------------------------\n    # 3. Optimizer\n    # -----------------------------------------------------------------------------\n    params = filter(lambda p: p.requires_grad, model.parameters()) \n    # Parameters with p.requires_grad=False are not updated during training.\n    # This can be specified when defining the nn.Modules during model creation\n\n    if 'optim' in cfg.keys():\n        if cfg['optim'].lower()=='sgd':\n            optim = torch.optim.SGD(params,\n                        lr=cfg['lr'],\n                        momentum=cfg['momentum'],\n                        weight_decay=cfg['weight_decay'])\n\n        elif cfg['optim'].lower()=='adam':\n            optim = torch.optim.Adam(params,\n                        lr=cfg['lr'], weight_decay=cfg['weight_decay'])\n\n        else:\n            raise NotImplementedError('Optimizers: SGD or Adam')\n    else:\n        optim = torch.optim.SGD(params,\n                    lr=cfg['lr'],\n                    momentum=cfg['momentum'],\n                    weight_decay=cfg['weight_decay'])\n\n    if resume:\n        optim.load_state_dict(checkpoint['optim_state_dict'])\n\n\n    # -----------------------------------------------------------------------------\n    # [optional] Sanity-check: forward pass with a single batch\n    # -----------------------------------------------------------------------------\n    DEBUG = False\n    if DEBUG:   \n        # model = model.cpu()\n        dataiter = iter(val_loader)\n        img, label = dataiter.next()\n\n        print 'Labels: ' + str(label.size()) # batchSize x num_class\n        print 'Input: ' + str(img.size())    # batchSize x 3 x 224 x 224\n\n        im = img.squeeze().numpy()\n        im = im[0,:,:,:]    # get first image in the batch\n        im = im.transpose((1,2,0)) # permute to 224x224x3\n        im = im * [ 0.229, 0.224, 0.225 ] # unnormalize\n        im = im + [ 0.485, 0.456, 0.406 ]\n        im[im<0] = 0\n\n        f = plt.figure()\n        plt.imshow(im)\n        plt.savefig('sanity-check-im.jpg')  # save transformed image in current folder\n        inputs = Variable(img)\n        if cuda:\n            inputs = inputs.cuda()\n\n        model.eval()\n        outputs = model(inputs)\n        print 'Network output: ' + str(outputs.size())        \n        model.train()\n\n    else:\n        pass\n\n\n    # -----------------------------------------------------------------------------\n    # 4. Training\n    # -----------------------------------------------------------------------------\n    trainer = train.Trainer(\n        cuda=cuda,\n        model=model,\n        criterion=criterion,\n        optimizer=optim,\n        init_lr=cfg['lr'],\n        lr_decay_epoch = cfg['lr_decay_epoch'],\n        train_loader=train_loader,\n        val_loader=val_loader,\n        out=out,\n        max_iter=cfg['max_iteration'],\n        interval_validate=cfg.get('interval_validate', len(train_loader)),\n    )\n\n    trainer.epoch = start_epoch\n    trainer.iteration = start_iteration\n    trainer.train()\n\n\n\ndef get_log_dir(model_name, config_id, cfg, verbose=True):\n    # Creates an output directory for each experiment, timestamped\n    name = 'MODEL-%s_CFG-%03d' % (model_name, config_id)\n    if verbose:\n        for k, v in cfg.items():\n            v = str(v)\n            if '/' in v:\n                continue\n            name += '_%s-%s' % (k.upper(), v)\n    now = datetime.datetime.now(pytz.timezone('US/Eastern'))\n    name += '_TIME-%s' % now.strftime('%Y%m%d-%H%M%S')\n    log_dir = osp.join(here, 'logs', name)\n    if not osp.exists(log_dir):\n        os.makedirs(log_dir)\n    with open(osp.join(log_dir, 'config.yaml'), 'w') as f:\n        yaml.safe_dump(cfg, f, default_flow_style=False)\n    return log_dir\n\n\n\n\nif __name__ == '__main__':\n    main()\n\n"
  },
  {
    "path": "vgg-face-2/vggface_class_counts.txt",
    "content": "n000002 313\nn000003 203\nn000004 385\nn000005 227\nn000006 483\nn000007 257\nn000008 271\nn000010 154\nn000011 378\nn000012 380\nn000013 312\nn000014 286\nn000015 246\nn000016 447\nn000017 317\nn000018 281\nn000019 372\nn000020 448\nn000021 284\nn000022 394\nn000023 396\nn000024 449\nn000025 307\nn000026 369\nn000027 380\nn000028 262\nn000030 265\nn000031 303\nn000032 517\nn000033 355\nn000034 303\nn000035 235\nn000036 372\nn000037 204\nn000038 415\nn000039 374\nn000041 389\nn000042 261\nn000043 491\nn000044 362\nn000045 244\nn000046 409\nn000047 328\nn000048 214\nn000049 437\nn000050 394\nn000051 325\nn000052 210\nn000053 379\nn000054 436\nn000055 448\nn000056 302\nn000057 362\nn000058 442\nn000059 344\nn000060 493\nn000061 374\nn000062 319\nn000063 434\nn000064 376\nn000065 391\nn000066 337\nn000067 509\nn000068 302\nn000069 396\nn000070 420\nn000071 318\nn000072 304\nn000073 317\nn000074 257\nn000075 190\nn000076 302\nn000077 314\nn000079 446\nn000080 592\nn000081 534\nn000083 493\nn000084 445\nn000085 452\nn000086 436\nn000087 432\nn000088 384\nn000089 301\nn000090 467\nn000091 204\nn000092 248\nn000093 174\nn000094 430\nn000095 214\nn000096 423\nn000097 600\nn000098 480\nn000099 344\nn000100 483\nn000101 283\nn000102 370\nn000103 570\nn000104 443\nn000105 513\nn000107 224\nn000108 399\nn000109 476\nn000110 357\nn000111 509\nn000112 280\nn000113 215\nn000114 334\nn000115 299\nn000116 438\nn000117 368\nn000118 390\nn000119 205\nn000120 415\nn000121 347\nn000122 361\nn000123 314\nn000124 308\nn000125 183\nn000126 427\nn000127 269\nn000128 204\nn000130 369\nn000131 356\nn000132 332\nn000133 282\nn000134 402\nn000135 156\nn000136 264\nn000137 222\nn000138 302\nn000139 430\nn000140 444\nn000141 365\nn000142 435\nn000143 256\nn000144 312\nn000145 403\nn000146 310\nn000150 444\nn000151 431\nn000152 390\nn000154 485\nn000155 289\nn000156 355\nn000157 406\nn000158 371\nn000159 492\nn000160 382\nn000161 377\nn000162 290\nn000163 538\nn000164 373\nn000165 287\nn000166 396\nn000167 415\nn000168 287\nn000169 326\nn000170 381\nn000171 457\nn000172 482\nn000173 281\nn000174 278\nn000175 347\nn000176 558\nn000177 264\nn000179 386\nn000180 328\nn000181 397\nn000182 277\nn000183 312\nn000184 414\nn000185 552\nn000186 498\nn000187 374\nn000188 492\nn000189 173\nn000190 344\nn000191 382\nn000192 516\nn000193 269\nn000194 191\nn000195 297\nn000196 186\nn000197 321\nn000198 238\nn000199 410\nn000201 275\nn000202 646\nn000203 524\nn000204 375\nn000205 416\nn000206 366\nn000207 384\nn000208 270\nn000209 357\nn000210 277\nn000211 331\nn000212 307\nn000213 269\nn000214 482\nn000215 438\nn000216 369\nn000217 437\nn000218 334\nn000219 363\nn000220 473\nn000221 369\nn000222 408\nn000223 504\nn000224 294\nn000225 621\nn000226 245\nn000227 562\nn000228 594\nn000229 269\nn000230 270\nn000231 258\nn000232 244\nn000233 309\nn000234 637\nn000235 412\nn000236 479\nn000237 228\nn000238 472\nn000239 328\nn000240 394\nn000241 545\nn000242 435\nn000243 434\nn000244 332\nn000245 297\nn000246 226\nn000247 211\nn000248 515\nn000249 390\nn000250 413\nn000251 306\nn000252 297\nn000253 110\nn000254 327\nn000255 578\nn000256 205\nn000257 518\nn000258 228\nn000260 472\nn000261 326\nn000262 431\nn000263 306\nn000264 436\nn000265 497\nn000266 675\nn000267 347\nn000268 477\nn000269 235\nn000270 718\nn000271 501\nn000272 396\nn000273 244\nn000274 460\nn000275 431\nn000276 316\nn000277 488\nn000278 197\nn000279 204\nn000280 282\nn000281 343\nn000282 586\nn000283 324\nn000285 386\nn000286 440\nn000287 420\nn000288 428\nn000289 325\nn000290 209\nn000291 334\nn000292 386\nn000293 220\nn000294 346\nn000295 396\nn000296 274\nn000297 497\nn000298 240\nn000299 394\nn000300 516\nn000301 345\nn000302 611\nn000303 172\nn000304 270\nn000305 436\nn000306 457\nn000307 318\nn000308 326\nn000309 459\nn000310 317\nn000311 358\nn000312 264\nn000313 223\nn000314 613\nn000315 275\nn000316 495\nn000317 189\nn000318 479\nn000319 353\nn000320 383\nn000321 379\nn000322 325\nn000323 358\nn000324 316\nn000325 252\nn000326 527\nn000327 251\nn000328 238\nn000329 432\nn000330 324\nn000331 508\nn000332 562\nn000333 348\nn000334 260\nn000335 233\nn000336 294\nn000337 348\nn000338 309\nn000339 240\nn000340 266\nn000341 434\nn000342 204\nn000343 344\nn000344 364\nn000345 434\nn000346 411\nn000347 421\nn000348 335\nn000349 206\nn000350 440\nn000351 446\nn000352 473\nn000353 283\nn000354 368\nn000355 382\nn000356 385\nn000357 412\nn000358 517\nn000359 453\nn000360 151\nn000361 382\nn000362 230\nn000364 267\nn000365 375\nn000366 238\nn000367 391\nn000368 380\nn000369 338\nn000370 248\nn000371 280\nn000372 433\nn000373 406\nn000374 390\nn000375 325\nn000376 431\nn000377 471\nn000378 528\nn000379 355\nn000380 450\nn000381 299\nn000382 306\nn000383 398\nn000384 271\nn000385 270\nn000386 327\nn000387 440\nn000388 313\nn000389 253\nn000390 203\nn000391 423\nn000392 351\nn000393 330\nn000395 603\nn000396 305\nn000397 537\nn000398 271\nn000399 536\nn000400 236\nn000401 390\nn000402 349\nn000403 493\nn000405 402\nn000406 432\nn000407 302\nn000408 273\nn000409 345\nn000411 464\nn000412 431\nn000413 435\nn000414 354\nn000415 331\nn000416 297\nn000417 391\nn000418 420\nn000419 687\nn000420 450\nn000421 385\nn000422 413\nn000423 220\nn000424 205\nn000425 452\nn000426 400\nn000427 313\nn000428 371\nn000429 489\nn000430 228\nn000431 253\nn000432 253\nn000433 211\nn000434 468\nn000435 418\nn000436 198\nn000437 173\nn000438 554\nn000439 189\nn000440 431\nn000441 472\nn000442 320\nn000443 377\nn000444 383\nn000445 342\nn000446 395\nn000447 395\nn000448 250\nn000449 364\nn000450 410\nn000451 362\nn000453 347\nn000454 329\nn000455 261\nn000456 479\nn000457 279\nn000458 384\nn000459 439\nn000460 376\nn000461 339\nn000462 453\nn000463 382\nn000464 371\nn000465 434\nn000466 334\nn000467 365\nn000468 214\nn000469 474\nn000470 154\nn000471 369\nn000472 656\nn000473 362\nn000474 250\nn000475 349\nn000476 433\nn000477 300\nn000478 388\nn000479 425\nn000481 346\nn000482 459\nn000483 381\nn000484 461\nn000485 451\nn000486 213\nn000487 338\nn000488 409\nn000489 336\nn000490 336\nn000491 500\nn000492 357\nn000493 100\nn000494 473\nn000495 441\nn000496 291\nn000497 490\nn000498 413\nn000499 278\nn000500 475\nn000501 372\nn000502 333\nn000503 439\nn000504 295\nn000505 254\nn000506 205\nn000507 271\nn000508 515\nn000509 216\nn000510 389\nn000511 348\nn000512 246\nn000513 343\nn000514 285\nn000515 201\nn000516 361\nn000517 348\nn000518 374\nn000519 311\nn000520 495\nn000521 400\nn000522 412\nn000524 385\nn000525 210\nn000526 386\nn000528 444\nn000529 370\nn000530 406\nn000531 221\nn000532 477\nn000533 171\nn000534 289\nn000535 440\nn000536 360\nn000537 489\nn000538 422\nn000539 430\nn000540 389\nn000541 383\nn000542 310\nn000543 361\nn000544 509\nn000545 512\nn000546 237\nn000547 528\nn000548 442\nn000549 418\nn000550 451\nn000551 461\nn000552 433\nn000553 362\nn000554 562\nn000555 263\nn000556 139\nn000557 459\nn000558 546\nn000559 357\nn000560 514\nn000561 404\nn000562 429\nn000563 543\nn000564 426\nn000565 370\nn000566 428\nn000567 363\nn000568 538\nn000569 470\nn000570 280\nn000571 341\nn000573 391\nn000574 429\nn000575 292\nn000576 315\nn000577 629\nn000578 404\nn000579 479\nn000580 246\nn000581 290\nn000582 389\nn000583 260\nn000584 372\nn000585 220\nn000586 329\nn000587 426\nn000588 327\nn000589 391\nn000590 285\nn000591 441\nn000592 255\nn000593 319\nn000594 352\nn000595 394\nn000597 357\nn000598 549\nn000599 385\nn000600 444\nn000601 252\nn000602 341\nn000603 423\nn000604 340\nn000605 208\nn000606 539\nn000607 406\nn000608 358\nn000609 239\nn000610 361\nn000611 384\nn000612 298\nn000613 450\nn000614 487\nn000615 219\nn000616 452\nn000618 401\nn000619 332\nn000620 495\nn000621 309\nn000622 254\nn000623 242\nn000625 391\nn000626 338\nn000627 409\nn000628 620\nn000629 470\nn000630 512\nn000631 335\nn000632 506\nn000633 651\nn000634 246\nn000635 268\nn000636 231\nn000637 547\nn000638 410\nn000639 378\nn000640 593\nn000641 466\nn000642 303\nn000643 451\nn000644 301\nn000645 512\nn000646 499\nn000647 438\nn000648 321\nn000649 207\nn000650 364\nn000651 216\nn000652 334\nn000653 359\nn000655 261\nn000656 297\nn000657 380\nn000660 383\nn000661 443\nn000662 189\nn000663 555\nn000664 250\nn000665 462\nn000666 279\nn000668 512\nn000669 224\nn000670 263\nn000671 406\nn000672 432\nn000673 432\nn000674 443\nn000675 338\nn000676 395\nn000677 394\nn000678 302\nn000679 335\nn000680 393\nn000681 338\nn000682 414\nn000683 426\nn000684 292\nn000685 356\nn000686 347\nn000687 206\nn000688 220\nn000690 390\nn000691 320\nn000692 429\nn000693 351\nn000694 352\nn000695 387\nn000696 383\nn000697 376\nn000698 311\nn000699 363\nn000700 394\nn000701 245\nn000702 380\nn000703 281\nn000704 213\nn000705 516\nn000707 426\nn000708 341\nn000709 449\nn000710 426\nn000711 297\nn000712 493\nn000713 243\nn000714 383\nn000715 474\nn000716 406\nn000717 429\nn000718 336\nn000719 234\nn000720 408\nn000721 401\nn000722 431\nn000723 308\nn000724 488\nn000725 377\nn000726 379\nn000727 578\nn000728 202\nn000729 315\nn000730 194\nn000731 378\nn000732 447\nn000733 456\nn000734 270\nn000735 399\nn000737 473\nn000739 243\nn000741 440\nn000742 345\nn000743 363\nn000744 178\nn000745 457\nn000747 395\nn000748 218\nn000749 376\nn000750 369\nn000751 447\nn000752 394\nn000753 324\nn000754 313\nn000755 530\nn000756 505\nn000757 337\nn000758 506\nn000759 479\nn000760 309\nn000761 211\nn000762 323\nn000763 391\nn000764 346\nn000765 253\nn000766 200\nn000767 258\nn000768 343\nn000769 470\nn000770 464\nn000771 510\nn000772 371\nn000773 455\nn000776 403\nn000777 234\nn000778 344\nn000779 237\nn000780 259\nn000781 356\nn000782 336\nn000783 394\nn000784 460\nn000786 373\nn000787 239\nn000788 388\nn000789 497\nn000790 254\nn000791 303\nn000792 412\nn000793 429\nn000794 305\nn000795 417\nn000796 449\nn000797 328\nn000798 218\nn000799 405\nn000800 245\nn000801 313\nn000802 235\nn000803 399\nn000804 629\nn000805 498\nn000806 345\nn000807 425\nn000808 430\nn000809 287\nn000810 841\nn000811 307\nn000812 649\nn000813 400\nn000814 390\nn000815 226\nn000816 441\nn000817 522\nn000818 475\nn000819 390\nn000820 313\nn000821 245\nn000822 236\nn000823 454\nn000824 350\nn000825 531\nn000826 561\nn000827 497\nn000828 344\nn000829 395\nn000830 329\nn000831 283\nn000832 284\nn000833 535\nn000834 258\nn000835 604\nn000837 282\nn000839 425\nn000840 228\nn000841 337\nn000842 343\nn000843 333\nn000844 430\nn000845 319\nn000846 273\nn000847 464\nn000848 300\nn000849 369\nn000850 230\nn000851 448\nn000852 324\nn000853 251\nn000855 297\nn000856 201\nn000857 555\nn000858 332\nn000859 302\nn000860 324\nn000861 206\nn000862 460\nn000863 463\nn000864 275\nn000865 486\nn000866 465\nn000867 225\nn000868 280\nn000869 494\nn000870 287\nn000871 323\nn000872 386\nn000873 435\nn000874 182\nn000875 424\nn000876 539\nn000878 209\nn000879 508\nn000880 514\nn000881 340\nn000882 259\nn000883 455\nn000884 472\nn000885 595\nn000886 357\nn000887 442\nn000888 375\nn000889 337\nn000890 264\nn000891 346\nn000892 345\nn000893 327\nn000894 253\nn000895 454\nn000896 272\nn000897 230\nn000898 405\nn000899 359\nn000900 346\nn000901 516\nn000902 478\nn000903 427\nn000904 433\nn000905 347\nn000906 215\nn000907 405\nn000908 379\nn000909 323\nn000910 318\nn000911 437\nn000913 315\nn000914 192\nn000915 450\nn000916 381\nn000917 362\nn000918 355\nn000919 424\nn000920 400\nn000921 214\nn000922 354\nn000923 391\nn000924 438\nn000925 278\nn000926 476\nn000927 294\nn000929 126\nn000930 354\nn000931 243\nn000932 310\nn000933 378\nn000934 307\nn000935 312\nn000937 251\nn000938 377\nn000939 453\nn000940 454\nn000941 570\nn000942 547\nn000943 485\nn000944 453\nn000946 189\nn000947 202\nn000948 336\nn000949 241\nn000951 241\nn000952 381\nn000953 199\nn000954 453\nn000955 243\nn000956 588\nn000957 464\nn000959 366\nn000960 372\nn000961 347\nn000962 207\nn000963 419\nn000964 468\nn000965 354\nn000966 329\nn000967 510\nn000968 437\nn000969 434\nn000970 330\nn000971 363\nn000972 220\nn000973 450\nn000974 308\nn000975 346\nn000976 539\nn000977 275\nn000978 366\nn000979 354\nn000980 307\nn000981 477\nn000982 352\nn000983 323\nn000984 413\nn000985 353\nn000986 301\nn000987 393\nn000988 423\nn000989 173\nn000990 427\nn000991 330\nn000992 446\nn000993 274\nn000994 481\nn000995 355\nn000996 347\nn000997 264\nn000999 409\nn001000 450\nn001001 518\nn001002 261\nn001003 361\nn001004 339\nn001005 327\nn001006 588\nn001007 401\nn001008 203\nn001009 462\nn001010 506\nn001011 317\nn001012 476\nn001014 185\nn001015 495\nn001016 425\nn001017 374\nn001018 419\nn001019 465\nn001023 202\nn001024 468\nn001025 533\nn001026 495\nn001027 333\nn001028 585\nn001029 256\nn001030 376\nn001031 329\nn001032 377\nn001033 393\nn001034 254\nn001035 229\nn001036 318\nn001037 384\nn001038 472\nn001039 375\nn001040 306\nn001041 395\nn001042 476\nn001043 102\nn001044 389\nn001045 292\nn001046 146\nn001047 404\nn001048 302\nn001049 247\nn001050 406\nn001051 365\nn001052 507\nn001053 234\nn001054 334\nn001055 231\nn001056 442\nn001057 469\nn001058 301\nn001060 417\nn001061 357\nn001062 334\nn001063 393\nn001064 284\nn001065 347\nn001066 528\nn001067 337\nn001068 471\nn001069 389\nn001070 274\nn001071 394\nn001072 361\nn001073 312\nn001074 247\nn001075 362\nn001076 347\nn001077 357\nn001078 388\nn001079 235\nn001080 368\nn001081 221\nn001082 360\nn001083 265\nn001084 360\nn001085 234\nn001086 344\nn001087 249\nn001088 372\nn001089 389\nn001090 456\nn001091 540\nn001092 348\nn001093 387\nn001094 471\nn001095 411\nn001096 383\nn001097 381\nn001098 294\nn001099 267\nn001100 387\nn001101 344\nn001102 292\nn001103 237\nn001104 337\nn001105 404\nn001106 439\nn001108 264\nn001109 307\nn001110 231\nn001111 391\nn001112 322\nn001113 398\nn001114 234\nn001115 373\nn001116 324\nn001117 369\nn001118 180\nn001119 350\nn001120 398\nn001121 462\nn001122 301\nn001123 352\nn001124 389\nn001126 337\nn001128 215\nn001129 482\nn001130 411\nn001131 492\nn001132 545\nn001133 355\nn001134 479\nn001135 431\nn001136 351\nn001137 473\nn001138 254\nn001139 452\nn001140 348\nn001141 224\nn001142 463\nn001143 367\nn001144 388\nn001145 629\nn001147 403\nn001148 565\nn001149 233\nn001150 452\nn001151 272\nn001152 317\nn001154 408\nn001155 425\nn001157 308\nn001158 312\nn001159 451\nn001160 271\nn001161 460\nn001162 369\nn001163 383\nn001164 332\nn001165 444\nn001166 270\nn001167 237\nn001168 281\nn001169 403\nn001170 398\nn001171 326\nn001172 351\nn001173 292\nn001175 225\nn001176 329\nn001177 166\nn001178 535\nn001179 264\nn001180 315\nn001181 324\nn001182 372\nn001183 221\nn001184 381\nn001185 307\nn001186 362\nn001187 498\nn001188 439\nn001189 387\nn001191 265\nn001192 352\nn001193 444\nn001194 220\nn001195 225\nn001196 280\nn001198 485\nn001200 543\nn001201 334\nn001202 204\nn001203 228\nn001204 401\nn001205 408\nn001206 418\nn001207 254\nn001208 168\nn001209 216\nn001210 382\nn001212 299\nn001213 308\nn001214 234\nn001215 328\nn001216 424\nn001217 206\nn001218 460\nn001219 333\nn001220 365\nn001221 575\nn001222 464\nn001223 490\nn001224 497\nn001225 546\nn001226 417\nn001227 459\nn001228 355\nn001229 420\nn001230 309\nn001231 202\nn001232 327\nn001233 202\nn001234 412\nn001235 607\nn001236 386\nn001237 546\nn001238 383\nn001239 354\nn001240 295\nn001241 528\nn001243 243\nn001244 392\nn001245 359\nn001246 321\nn001247 266\nn001248 303\nn001249 409\nn001250 232\nn001251 307\nn001252 239\nn001253 393\nn001254 346\nn001255 299\nn001257 215\nn001258 324\nn001259 483\nn001260 370\nn001261 278\nn001262 428\nn001263 475\nn001264 526\nn001265 300\nn001266 270\nn001267 272\nn001268 386\nn001269 383\nn001270 217\nn001271 394\nn001272 400\nn001273 229\nn001274 384\nn001275 384\nn001276 322\nn001278 278\nn001279 198\nn001280 207\nn001281 454\nn001282 356\nn001283 203\nn001284 217\nn001285 482\nn001286 305\nn001287 457\nn001288 332\nn001289 314\nn001290 400\nn001292 353\nn001294 427\nn001295 316\nn001297 219\nn001298 387\nn001300 316\nn001301 418\nn001305 362\nn001306 210\nn001307 234\nn001308 333\nn001309 467\nn001310 327\nn001311 277\nn001312 200\nn001313 351\nn001314 370\nn001315 467\nn001316 673\nn001317 342\nn001319 579\nn001320 373\nn001321 273\nn001322 511\nn001323 339\nn001325 398\nn001326 313\nn001327 393\nn001328 406\nn001329 547\nn001330 381\nn001331 345\nn001332 343\nn001333 585\nn001334 280\nn001335 378\nn001336 408\nn001338 281\nn001339 355\nn001340 390\nn001342 265\nn001343 411\nn001344 445\nn001345 356\nn001346 428\nn001347 233\nn001348 414\nn001349 393\nn001350 322\nn001351 342\nn001352 509\nn001353 307\nn001354 385\nn001355 477\nn001356 326\nn001357 263\nn001358 218\nn001359 522\nn001360 486\nn001361 401\nn001362 323\nn001363 464\nn001364 337\nn001365 392\nn001366 204\nn001367 550\nn001369 562\nn001370 490\nn001371 302\nn001372 426\nn001373 409\nn001374 336\nn001375 401\nn001376 294\nn001377 216\nn001378 292\nn001379 419\nn001380 438\nn001381 447\nn001382 365\nn001383 325\nn001384 425\nn001385 311\nn001386 291\nn001387 476\nn001388 343\nn001389 344\nn001390 212\nn001391 647\nn001392 568\nn001393 336\nn001394 233\nn001395 410\nn001396 410\nn001397 559\nn001398 389\nn001399 283\nn001400 392\nn001402 226\nn001403 394\nn001404 497\nn001405 453\nn001406 229\nn001407 439\nn001408 357\nn001409 397\nn001410 469\nn001411 331\nn001412 256\nn001413 441\nn001414 381\nn001415 489\nn001416 232\nn001417 350\nn001419 357\nn001420 325\nn001421 449\nn001422 334\nn001423 378\nn001424 359\nn001425 458\nn001426 217\nn001427 292\nn001428 435\nn001429 259\nn001430 453\nn001431 423\nn001432 360\nn001433 260\nn001434 415\nn001435 202\nn001436 373\nn001437 444\nn001438 239\nn001440 287\nn001441 372\nn001442 491\nn001443 353\nn001444 287\nn001445 228\nn001447 286\nn001448 496\nn001449 308\nn001450 292\nn001451 331\nn001452 416\nn001453 279\nn001454 182\nn001455 312\nn001456 224\nn001457 265\nn001458 274\nn001459 548\nn001460 439\nn001461 434\nn001462 241\nn001463 372\nn001464 321\nn001465 535\nn001466 530\nn001468 415\nn001469 479\nn001470 479\nn001471 347\nn001472 307\nn001473 224\nn001474 398\nn001475 143\nn001476 488\nn001477 305\nn001478 122\nn001479 507\nn001480 434\nn001482 277\nn001483 280\nn001484 463\nn001486 491\nn001487 216\nn001488 332\nn001489 480\nn001490 326\nn001491 404\nn001492 540\nn001493 577\nn001494 450\nn001495 616\nn001496 532\nn001497 468\nn001498 308\nn001499 359\nn001500 343\nn001501 415\nn001502 476\nn001503 184\nn001504 388\nn001505 418\nn001506 410\nn001507 197\nn001508 663\nn001509 506\nn001510 244\nn001511 436\nn001512 257\nn001513 370\nn001514 559\nn001515 353\nn001516 309\nn001517 365\nn001518 483\nn001519 338\nn001520 390\nn001521 287\nn001522 214\nn001523 482\nn001525 250\nn001526 187\nn001528 381\nn001529 342\nn001530 411\nn001531 303\nn001532 348\nn001533 391\nn001534 363\nn001535 585\nn001536 391\nn001537 352\nn001538 468\nn001539 200\nn001540 518\nn001541 470\nn001542 294\nn001543 307\nn001544 310\nn001545 337\nn001546 217\nn001547 603\nn001548 493\nn001549 575\nn001550 411\nn001551 436\nn001552 459\nn001553 435\nn001554 174\nn001555 340\nn001556 204\nn001557 239\nn001558 238\nn001559 537\nn001560 444\nn001561 450\nn001562 321\nn001563 392\nn001564 268\nn001565 167\nn001566 635\nn001567 482\nn001568 427\nn001569 307\nn001570 275\nn001571 313\nn001572 320\nn001573 414\nn001574 258\nn001575 199\nn001577 284\nn001578 411\nn001579 422\nn001580 323\nn001581 227\nn001582 481\nn001583 677\nn001584 601\nn001585 350\nn001586 757\nn001587 562\nn001588 576\nn001589 494\nn001590 321\nn001591 300\nn001592 576\nn001593 248\nn001594 468\nn001595 363\nn001596 249\nn001597 204\nn001598 344\nn001599 324\nn001600 442\nn001601 467\nn001602 397\nn001603 352\nn001604 332\nn001605 251\nn001606 295\nn001607 279\nn001608 531\nn001609 400\nn001610 310\nn001611 279\nn001612 306\nn001613 349\nn001614 437\nn001615 424\nn001616 332\nn001617 537\nn001618 457\nn001619 272\nn001620 428\nn001621 362\nn001622 304\nn001623 411\nn001624 373\nn001625 279\nn001626 262\nn001627 403\nn001628 355\nn001629 453\nn001630 394\nn001631 372\nn001632 403\nn001633 317\nn001634 236\nn001635 371\nn001636 256\nn001637 394\nn001638 212\nn001639 447\nn001640 331\nn001641 400\nn001642 505\nn001643 408\nn001644 473\nn001645 522\nn001646 420\nn001647 493\nn001648 382\nn001649 390\nn001651 325\nn001652 503\nn001653 379\nn001654 375\nn001656 402\nn001657 651\nn001658 582\nn001659 570\nn001660 412\nn001661 242\nn001662 325\nn001663 451\nn001664 354\nn001665 487\nn001666 473\nn001667 214\nn001668 426\nn001670 285\nn001671 359\nn001673 399\nn001674 355\nn001675 448\nn001676 323\nn001677 424\nn001678 173\nn001679 245\nn001680 466\nn001681 455\nn001682 234\nn001684 471\nn001685 137\nn001686 341\nn001688 286\nn001689 274\nn001690 393\nn001691 334\nn001692 399\nn001693 582\nn001694 408\nn001695 465\nn001696 374\nn001697 491\nn001698 355\nn001699 461\nn001700 650\nn001701 386\nn001702 310\nn001703 451\nn001704 420\nn001705 286\nn001706 455\nn001707 344\nn001709 355\nn001711 373\nn001712 432\nn001713 346\nn001714 335\nn001715 422\nn001716 356\nn001717 409\nn001718 325\nn001719 303\nn001720 467\nn001721 369\nn001722 319\nn001723 352\nn001724 421\nn001725 353\nn001726 385\nn001727 623\nn001728 555\nn001729 363\nn001730 301\nn001731 402\nn001732 371\nn001733 417\nn001734 446\nn001735 349\nn001736 218\nn001737 240\nn001738 324\nn001739 254\nn001740 317\nn001741 279\nn001742 393\nn001743 200\nn001744 461\nn001745 313\nn001746 289\nn001747 466\nn001748 488\nn001749 479\nn001750 315\nn001751 311\nn001752 236\nn001753 443\nn001754 424\nn001755 147\nn001756 267\nn001757 401\nn001758 216\nn001759 629\nn001760 359\nn001761 223\nn001762 320\nn001763 547\nn001764 349\nn001765 669\nn001766 348\nn001767 469\nn001768 564\nn001769 370\nn001770 419\nn001771 673\nn001772 407\nn001773 658\nn001774 317\nn001775 649\nn001776 354\nn001777 338\nn001778 395\nn001779 401\nn001780 516\nn001782 303\nn001783 481\nn001784 306\nn001785 372\nn001786 355\nn001787 174\nn001788 542\nn001789 186\nn001790 336\nn001791 215\nn001792 404\nn001793 251\nn001794 417\nn001795 396\nn001796 479\nn001797 438\nn001798 372\nn001799 492\nn001800 292\nn001801 314\nn001802 335\nn001803 531\nn001804 455\nn001805 445\nn001806 206\nn001807 201\nn001808 383\nn001809 353\nn001810 281\nn001812 438\nn001813 415\nn001814 327\nn001815 304\nn001818 158\nn001820 250\nn001821 363\nn001822 342\nn001823 423\nn001824 346\nn001825 320\nn001826 477\nn001827 300\nn001828 443\nn001829 261\nn001831 380\nn001832 335\nn001833 500\nn001834 459\nn001835 392\nn001837 493\nn001839 538\nn001840 432\nn001841 514\nn001842 270\nn001843 558\nn001844 400\nn001845 399\nn001846 397\nn001847 402\nn001848 458\nn001849 241\nn001851 486\nn001852 377\nn001853 352\nn001854 313\nn001855 373\nn001856 345\nn001858 463\nn001859 294\nn001860 492\nn001861 358\nn001862 175\nn001863 364\nn001864 279\nn001865 302\nn001866 454\nn001867 228\nn001868 376\nn001869 279\nn001870 329\nn001871 482\nn001872 272\nn001873 387\nn001874 302\nn001875 470\nn001876 371\nn001877 433\nn001879 365\nn001880 380\nn001881 134\nn001882 320\nn001883 235\nn001884 402\nn001885 344\nn001886 246\nn001887 427\nn001888 397\nn001889 300\nn001890 478\nn001891 298\nn001892 331\nn001893 224\nn001894 397\nn001895 283\nn001896 292\nn001897 248\nn001899 285\nn001900 373\nn001901 307\nn001902 176\nn001903 429\nn001904 384\nn001905 462\nn001906 429\nn001907 530\nn001908 262\nn001909 390\nn001910 280\nn001911 462\nn001912 415\nn001913 231\nn001914 455\nn001915 325\nn001916 219\nn001917 591\nn001918 298\nn001919 344\nn001920 463\nn001921 238\nn001922 407\nn001924 343\nn001925 158\nn001926 404\nn001927 192\nn001928 262\nn001929 158\nn001930 502\nn001931 407\nn001932 461\nn001933 378\nn001936 453\nn001937 319\nn001938 477\nn001939 429\nn001940 390\nn001941 221\nn001942 330\nn001943 250\nn001944 347\nn001945 284\nn001946 388\nn001947 262\nn001948 350\nn001949 497\nn001950 371\nn001951 414\nn001952 215\nn001953 349\nn001954 358\nn001955 468\nn001957 293\nn001958 340\nn001959 475\nn001960 441\nn001961 595\nn001962 222\nn001963 421\nn001964 253\nn001965 404\nn001966 491\nn001967 85\nn001968 318\nn001970 368\nn001971 316\nn001972 384\nn001973 347\nn001974 485\nn001975 303\nn001978 304\nn001979 451\nn001980 474\nn001981 321\nn001982 307\nn001983 301\nn001984 364\nn001985 471\nn001986 357\nn001987 333\nn001988 376\nn001989 278\nn001990 223\nn001991 446\nn001992 395\nn001993 345\nn001994 296\nn001995 368\nn001996 414\nn001998 382\nn001999 305\nn002000 283\nn002001 298\nn002002 382\nn002003 246\nn002004 399\nn002005 224\nn002006 269\nn002007 420\nn002008 442\nn002010 240\nn002011 365\nn002012 388\nn002013 336\nn002014 269\nn002015 345\nn002016 646\nn002017 468\nn002018 410\nn002019 504\nn002020 360\nn002021 387\nn002022 252\nn002023 247\nn002025 500\nn002026 492\nn002027 327\nn002028 481\nn002029 404\nn002031 350\nn002032 476\nn002033 431\nn002034 252\nn002035 506\nn002036 417\nn002037 344\nn002038 539\nn002039 411\nn002040 346\nn002041 404\nn002042 477\nn002043 344\nn002044 211\nn002045 282\nn002046 472\nn002047 539\nn002048 361\nn002049 289\nn002050 284\nn002051 355\nn002052 437\nn002053 341\nn002054 350\nn002055 329\nn002056 482\nn002057 332\nn002058 458\nn002059 292\nn002060 519\nn002061 606\nn002062 406\nn002063 298\nn002064 309\nn002065 251\nn002066 532\nn002067 271\nn002068 306\nn002069 239\nn002070 344\nn002071 353\nn002072 341\nn002073 296\nn002074 445\nn002075 254\nn002076 443\nn002077 441\nn002078 310\nn002079 197\nn002083 363\nn002084 329\nn002085 446\nn002086 424\nn002087 261\nn002088 408\nn002089 304\nn002090 270\nn002091 316\nn002092 360\nn002094 532\nn002095 282\nn002096 450\nn002097 424\nn002098 370\nn002099 289\nn002100 401\nn002101 258\nn002102 424\nn002103 212\nn002104 136\nn002105 414\nn002107 216\nn002108 194\nn002110 345\nn002111 377\nn002112 318\nn002113 372\nn002114 332\nn002115 600\nn002116 388\nn002117 428\nn002118 484\nn002119 360\nn002120 502\nn002121 423\nn002122 403\nn002123 283\nn002124 166\nn002125 356\nn002126 255\nn002127 330\nn002128 291\nn002129 249\nn002130 344\nn002131 306\nn002132 368\nn002133 536\nn002134 347\nn002135 359\nn002136 520\nn002137 472\nn002138 441\nn002139 225\nn002140 298\nn002141 543\nn002142 553\nn002143 374\nn002144 399\nn002145 398\nn002146 361\nn002147 489\nn002148 122\nn002149 372\nn002150 498\nn002151 394\nn002152 341\nn002154 423\nn002155 503\nn002156 428\nn002157 216\nn002159 213\nn002160 537\nn002161 351\nn002162 378\nn002163 446\nn002164 227\nn002165 646\nn002168 304\nn002169 222\nn002170 351\nn002171 307\nn002172 382\nn002173 324\nn002174 436\nn002175 442\nn002176 345\nn002177 307\nn002178 491\nn002179 344\nn002180 461\nn002182 252\nn002183 323\nn002184 261\nn002185 324\nn002186 293\nn002187 518\nn002188 442\nn002189 457\nn002190 351\nn002191 253\nn002192 216\nn002193 300\nn002194 462\nn002195 378\nn002196 462\nn002197 495\nn002198 337\nn002199 393\nn002200 405\nn002201 522\nn002202 487\nn002203 405\nn002204 379\nn002205 247\nn002206 381\nn002207 314\nn002208 275\nn002209 310\nn002210 173\nn002211 369\nn002212 400\nn002213 305\nn002214 270\nn002215 427\nn002217 319\nn002218 262\nn002219 384\nn002220 335\nn002221 561\nn002222 302\nn002224 145\nn002225 518\nn002226 262\nn002227 449\nn002228 171\nn002229 367\nn002230 426\nn002231 395\nn002232 487\nn002233 474\nn002234 317\nn002235 300\nn002236 210\nn002237 365\nn002238 516\nn002239 213\nn002240 445\nn002241 324\nn002242 463\nn002243 397\nn002244 426\nn002246 402\nn002247 349\nn002248 399\nn002249 308\nn002250 311\nn002251 476\nn002252 297\nn002253 325\nn002254 405\nn002255 585\nn002256 307\nn002259 442\nn002260 301\nn002261 429\nn002262 509\nn002265 362\nn002266 312\nn002269 515\nn002270 262\nn002271 425\nn002272 513\nn002273 150\nn002274 273\nn002275 294\nn002276 338\nn002277 522\nn002278 575\nn002279 287\nn002280 416\nn002281 376\nn002283 375\nn002285 407\nn002286 332\nn002287 394\nn002288 374\nn002289 310\nn002290 396\nn002291 266\nn002292 482\nn002293 433\nn002294 364\nn002295 373\nn002296 261\nn002297 562\nn002298 544\nn002299 436\nn002300 310\nn002301 318\nn002302 439\nn002303 285\nn002304 156\nn002305 456\nn002306 241\nn002307 276\nn002308 355\nn002310 454\nn002311 328\nn002312 246\nn002313 374\nn002314 444\nn002315 350\nn002316 445\nn002317 353\nn002318 400\nn002319 454\nn002320 310\nn002321 342\nn002322 324\nn002323 336\nn002324 390\nn002325 371\nn002326 365\nn002327 446\nn002328 398\nn002330 312\nn002331 372\nn002332 406\nn002333 378\nn002334 444\nn002335 331\nn002336 409\nn002337 277\nn002338 428\nn002339 602\nn002340 436\nn002341 238\nn002342 387\nn002343 295\nn002344 270\nn002345 400\nn002346 258\nn002347 367\nn002348 348\nn002349 237\nn002350 585\nn002352 372\nn002353 256\nn002354 312\nn002355 392\nn002356 245\nn002357 335\nn002358 344\nn002359 519\nn002360 521\nn002361 443\nn002362 477\nn002363 295\nn002364 444\nn002365 177\nn002366 558\nn002367 517\nn002368 499\nn002369 104\nn002370 374\nn002371 235\nn002373 389\nn002374 296\nn002375 184\nn002376 274\nn002377 269\nn002378 370\nn002379 391\nn002380 306\nn002382 414\nn002383 528\nn002386 284\nn002387 465\nn002388 419\nn002390 269\nn002391 475\nn002392 302\nn002393 204\nn002394 291\nn002395 231\nn002396 395\nn002397 581\nn002398 289\nn002399 213\nn002400 517\nn002401 308\nn002402 306\nn002403 439\nn002404 354\nn002405 307\nn002406 419\nn002407 392\nn002408 460\nn002409 455\nn002410 264\nn002411 326\nn002412 376\nn002413 322\nn002415 265\nn002416 322\nn002417 198\nn002418 316\nn002419 456\nn002420 397\nn002422 427\nn002423 335\nn002424 182\nn002425 474\nn002426 523\nn002427 426\nn002428 325\nn002430 276\nn002431 434\nn002432 338\nn002433 153\nn002434 227\nn002435 287\nn002436 308\nn002437 349\nn002438 494\nn002439 249\nn002440 411\nn002441 202\nn002442 339\nn002443 360\nn002444 460\nn002445 134\nn002446 290\nn002447 549\nn002448 464\nn002449 402\nn002450 428\nn002451 331\nn002452 497\nn002453 316\nn002454 541\nn002455 227\nn002456 302\nn002457 339\nn002458 318\nn002459 357\nn002460 398\nn002461 353\nn002462 557\nn002463 482\nn002464 423\nn002465 347\nn002466 280\nn002467 413\nn002468 395\nn002469 340\nn002470 223\nn002471 480\nn002472 265\nn002473 363\nn002476 580\nn002477 192\nn002478 501\nn002479 453\nn002480 337\nn002481 326\nn002482 176\nn002483 437\nn002484 478\nn002485 265\nn002486 360\nn002487 388\nn002488 335\nn002489 496\nn002490 298\nn002491 242\nn002492 364\nn002493 246\nn002494 378\nn002495 282\nn002496 213\nn002497 417\nn002498 503\nn002499 398\nn002500 390\nn002501 452\nn002502 593\nn002504 374\nn002505 445\nn002506 390\nn002507 520\nn002508 348\nn002509 347\nn002510 352\nn002512 424\nn002513 191\nn002514 284\nn002515 302\nn002516 461\nn002518 339\nn002519 320\nn002520 496\nn002521 331\nn002522 579\nn002523 258\nn002524 248\nn002525 520\nn002526 424\nn002527 274\nn002528 424\nn002529 221\nn002530 343\nn002531 326\nn002532 437\nn002533 469\nn002534 252\nn002535 246\nn002536 335\nn002537 604\nn002538 597\nn002539 428\nn002541 370\nn002542 285\nn002543 233\nn002544 304\nn002545 550\nn002546 417\nn002547 638\nn002548 357\nn002549 490\nn002550 366\nn002551 199\nn002552 533\nn002553 401\nn002554 529\nn002555 213\nn002557 529\nn002558 343\nn002559 171\nn002560 520\nn002562 327\nn002563 300\nn002564 231\nn002565 310\nn002566 319\nn002567 331\nn002568 497\nn002569 435\nn002570 353\nn002571 446\nn002572 556\nn002573 440\nn002575 364\nn002576 359\nn002577 496\nn002578 443\nn002579 352\nn002580 157\nn002582 357\nn002583 315\nn002584 466\nn002585 369\nn002586 428\nn002588 442\nn002589 258\nn002590 244\nn002591 404\nn002592 450\nn002593 273\nn002594 403\nn002595 468\nn002597 376\nn002598 296\nn002599 362\nn002600 216\nn002601 307\nn002602 316\nn002603 496\nn002605 344\nn002606 388\nn002607 299\nn002608 515\nn002609 323\nn002610 199\nn002611 411\nn002612 464\nn002613 230\nn002614 280\nn002615 250\nn002616 256\nn002617 431\nn002618 521\nn002619 335\nn002620 279\nn002621 518\nn002622 464\nn002623 325\nn002624 381\nn002625 236\nn002626 367\nn002627 410\nn002628 418\nn002629 278\nn002630 471\nn002631 345\nn002632 511\nn002633 410\nn002634 221\nn002635 335\nn002636 351\nn002637 326\nn002638 384\nn002639 387\nn002640 529\nn002641 424\nn002642 251\nn002643 305\nn002644 690\nn002645 324\nn002646 352\nn002648 519\nn002649 361\nn002650 291\nn002651 310\nn002652 293\nn002653 305\nn002654 593\nn002655 289\nn002656 259\nn002657 330\nn002660 278\nn002661 443\nn002662 218\nn002663 311\nn002665 334\nn002666 260\nn002667 417\nn002668 521\nn002670 406\nn002671 437\nn002672 381\nn002673 429\nn002674 250\nn002675 480\nn002676 165\nn002677 379\nn002678 569\nn002679 471\nn002680 249\nn002682 276\nn002683 268\nn002685 402\nn002686 290\nn002687 246\nn002688 272\nn002689 435\nn002690 125\nn002691 336\nn002692 495\nn002693 251\nn002694 359\nn002695 481\nn002696 345\nn002697 292\nn002698 172\nn002699 358\nn002700 336\nn002701 269\nn002702 351\nn002704 327\nn002705 390\nn002706 383\nn002707 263\nn002708 324\nn002709 318\nn002710 400\nn002711 406\nn002712 304\nn002713 309\nn002714 351\nn002716 318\nn002717 309\nn002718 315\nn002719 446\nn002720 440\nn002721 208\nn002722 312\nn002723 503\nn002724 520\nn002725 404\nn002726 233\nn002727 388\nn002728 277\nn002729 371\nn002730 465\nn002731 217\nn002732 530\nn002733 309\nn002734 249\nn002735 244\nn002736 407\nn002737 377\nn002738 457\nn002739 350\nn002740 249\nn002741 564\nn002742 374\nn002744 487\nn002745 385\nn002746 709\nn002747 494\nn002748 491\nn002750 304\nn002751 338\nn002752 335\nn002753 301\nn002754 208\nn002755 213\nn002756 229\nn002757 188\nn002758 404\nn002759 510\nn002760 333\nn002762 201\nn002764 179\nn002765 283\nn002766 418\nn002767 362\nn002769 218\nn002771 403\nn002772 387\nn002774 304\nn002776 259\nn002777 404\nn002778 436\nn002779 479\nn002780 459\nn002781 461\nn002782 512\nn002783 395\nn002784 479\nn002785 464\nn002786 424\nn002787 198\nn002788 277\nn002789 549\nn002790 243\nn002791 185\nn002792 426\nn002793 100\nn002794 324\nn002795 235\nn002796 200\nn002797 205\nn002798 391\nn002799 345\nn002800 343\nn002801 277\nn002802 428\nn002804 384\nn002805 374\nn002806 369\nn002807 299\nn002808 235\nn002809 350\nn002811 306\nn002812 318\nn002813 317\nn002814 385\nn002815 522\nn002816 246\nn002817 272\nn002818 195\nn002819 176\nn002820 260\nn002821 243\nn002822 374\nn002823 310\nn002824 188\nn002825 341\nn002826 208\nn002827 324\nn002828 437\nn002829 354\nn002830 361\nn002831 487\nn002832 193\nn002833 236\nn002834 380\nn002835 278\nn002836 302\nn002837 404\nn002839 326\nn002841 234\nn002842 587\nn002843 312\nn002844 434\nn002845 202\nn002846 179\nn002847 274\nn002848 391\nn002849 289\nn002850 355\nn002851 266\nn002852 360\nn002853 407\nn002854 435\nn002856 282\nn002858 337\nn002859 161\nn002860 280\nn002861 135\nn002862 479\nn002863 261\nn002864 402\nn002865 448\nn002866 177\nn002867 262\nn002868 328\nn002870 350\nn002871 370\nn002872 443\nn002874 371\nn002875 212\nn002876 259\nn002877 253\nn002879 385\nn002881 492\nn002882 442\nn002883 448\nn002885 564\nn002886 594\nn002887 333\nn002888 623\nn002890 265\nn002892 453\nn002893 341\nn002895 334\nn002896 321\nn002897 410\nn002898 310\nn002899 292\nn002900 372\nn002901 285\nn002902 373\nn002903 300\nn002904 139\nn002905 284\nn002906 235\nn002907 249\nn002908 228\nn002909 534\nn002910 479\nn002911 353\nn002912 170\nn002913 244\nn002914 336\nn002915 411\nn002916 277\nn002917 414\nn002918 128\nn002919 456\nn002920 395\nn002921 398\nn002922 315\nn002923 476\nn002924 254\nn002925 366\nn002926 347\nn002927 393\nn002928 333\nn002929 468\nn002930 387\nn002931 139\nn002932 273\nn002933 432\nn002934 229\nn002935 440\nn002936 227\nn002937 293\nn002938 286\nn002939 336\nn002940 386\nn002941 367\nn002942 502\nn002943 307\nn002944 368\nn002945 390\nn002946 365\nn002947 570\nn002948 274\nn002949 305\nn002950 455\nn002951 275\nn002952 350\nn002953 246\nn002954 416\nn002955 436\nn002956 275\nn002957 370\nn002958 265\nn002959 332\nn002960 397\nn002961 198\nn002962 422\nn002963 487\nn002964 238\nn002965 459\nn002966 395\nn002967 378\nn002968 297\nn002969 339\nn002970 469\nn002971 426\nn002972 297\nn002973 350\nn002974 475\nn002975 382\nn002976 320\nn002977 329\nn002978 294\nn002979 259\nn002980 364\nn002981 204\nn002982 478\nn002983 225\nn002984 415\nn002985 390\nn002986 367\nn002987 488\nn002988 462\nn002989 416\nn002990 346\nn002991 255\nn002992 438\nn002993 382\nn002994 457\nn002995 331\nn002997 403\nn002998 461\nn002999 467\nn003000 431\nn003002 366\nn003003 405\nn003004 429\nn003005 345\nn003006 383\nn003007 392\nn003008 318\nn003010 209\nn003011 258\nn003012 269\nn003013 495\nn003014 385\nn003015 447\nn003016 354\nn003017 219\nn003018 235\nn003019 336\nn003020 297\nn003021 565\nn003022 338\nn003023 234\nn003024 265\nn003025 433\nn003026 325\nn003027 474\nn003028 366\nn003029 420\nn003030 683\nn003031 279\nn003032 357\nn003033 366\nn003034 213\nn003035 523\nn003036 320\nn003037 297\nn003038 225\nn003039 339\nn003040 478\nn003041 168\nn003042 403\nn003043 341\nn003044 317\nn003045 315\nn003046 340\nn003047 505\nn003048 518\nn003049 242\nn003050 293\nn003051 276\nn003052 108\nn003053 278\nn003054 360\nn003055 432\nn003056 273\nn003057 318\nn003058 399\nn003059 308\nn003060 310\nn003061 284\nn003062 187\nn003063 474\nn003064 309\nn003065 484\nn003066 410\nn003067 395\nn003068 310\nn003069 334\nn003070 438\nn003071 338\nn003072 451\nn003073 505\nn003074 383\nn003075 221\nn003076 539\nn003077 318\nn003078 330\nn003080 349\nn003081 255\nn003082 312\nn003083 345\nn003084 120\nn003085 364\nn003086 399\nn003087 327\nn003088 322\nn003089 268\nn003090 318\nn003091 205\nn003094 260\nn003095 285\nn003096 376\nn003097 286\nn003098 451\nn003099 267\nn003100 329\nn003101 426\nn003102 467\nn003103 287\nn003105 115\nn003106 313\nn003107 560\nn003108 326\nn003109 457\nn003110 393\nn003111 369\nn003112 259\nn003113 213\nn003114 364\nn003116 247\nn003117 506\nn003118 162\nn003119 295\nn003120 413\nn003121 513\nn003122 277\nn003123 431\nn003124 461\nn003125 198\nn003126 583\nn003127 201\nn003128 378\nn003129 396\nn003130 342\nn003131 316\nn003132 453\nn003133 306\nn003135 401\nn003136 487\nn003137 392\nn003138 442\nn003139 444\nn003142 411\nn003143 270\nn003144 292\nn003145 310\nn003146 488\nn003147 299\nn003148 495\nn003149 283\nn003150 299\nn003151 514\nn003152 295\nn003153 507\nn003154 406\nn003155 432\nn003156 190\nn003157 354\nn003158 291\nn003159 407\nn003160 258\nn003161 410\nn003162 468\nn003163 462\nn003164 240\nn003165 121\nn003166 193\nn003167 536\nn003168 246\nn003169 359\nn003170 512\nn003171 315\nn003172 338\nn003173 389\nn003174 395\nn003175 305\nn003176 391\nn003177 493\nn003178 466\nn003179 472\nn003180 301\nn003181 559\nn003182 550\nn003183 455\nn003184 364\nn003185 301\nn003186 355\nn003187 397\nn003188 321\nn003189 218\nn003190 325\nn003191 305\nn003192 317\nn003193 213\nn003194 267\nn003195 238\nn003196 223\nn003197 521\nn003198 443\nn003199 341\nn003200 341\nn003201 291\nn003202 270\nn003203 366\nn003204 439\nn003206 528\nn003207 251\nn003208 109\nn003209 439\nn003210 550\nn003211 437\nn003212 392\nn003213 376\nn003214 563\nn003216 388\nn003218 321\nn003219 357\nn003220 365\nn003221 290\nn003222 376\nn003223 381\nn003224 374\nn003225 384\nn003226 360\nn003227 225\nn003228 276\nn003229 381\nn003231 333\nn003232 453\nn003233 216\nn003234 202\nn003235 370\nn003236 234\nn003237 386\nn003238 303\nn003239 521\nn003240 261\nn003241 433\nn003242 475\nn003243 333\nn003244 339\nn003245 343\nn003246 283\nn003247 466\nn003248 137\nn003249 253\nn003250 329\nn003251 319\nn003252 412\nn003253 413\nn003254 398\nn003255 321\nn003256 362\nn003257 432\nn003259 416\nn003260 268\nn003261 242\nn003262 440\nn003263 381\nn003264 348\nn003265 494\nn003266 445\nn003267 308\nn003268 209\nn003269 319\nn003270 290\nn003271 399\nn003272 420\nn003273 386\nn003274 406\nn003275 246\nn003276 390\nn003278 377\nn003279 401\nn003280 387\nn003281 427\nn003282 321\nn003283 412\nn003284 487\nn003285 408\nn003286 435\nn003287 499\nn003289 278\nn003290 425\nn003291 416\nn003292 395\nn003293 438\nn003294 232\nn003295 412\nn003296 240\nn003297 330\nn003299 360\nn003300 275\nn003301 243\nn003302 371\nn003303 573\nn003304 306\nn003305 524\nn003306 430\nn003307 373\nn003308 439\nn003310 347\nn003311 401\nn003312 319\nn003313 212\nn003314 252\nn003315 488\nn003316 581\nn003317 243\nn003318 242\nn003319 524\nn003320 250\nn003321 250\nn003322 483\nn003323 455\nn003324 580\nn003325 499\nn003326 212\nn003327 464\nn003328 297\nn003329 204\nn003330 340\nn003331 315\nn003332 164\nn003333 252\nn003334 547\nn003335 509\nn003336 389\nn003337 336\nn003338 430\nn003339 298\nn003340 392\nn003341 332\nn003342 437\nn003343 484\nn003344 531\nn003345 289\nn003346 452\nn003347 327\nn003348 458\nn003349 505\nn003350 510\nn003351 383\nn003352 403\nn003353 418\nn003354 262\nn003355 467\nn003357 270\nn003358 469\nn003359 321\nn003360 475\nn003361 239\nn003362 356\nn003363 440\nn003364 345\nn003365 538\nn003366 456\nn003367 217\nn003368 335\nn003369 300\nn003370 469\nn003371 366\nn003372 276\nn003373 465\nn003374 601\nn003375 213\nn003376 315\nn003377 295\nn003378 518\nn003380 429\nn003381 302\nn003382 336\nn003383 546\nn003384 434\nn003385 449\nn003386 309\nn003387 243\nn003388 298\nn003389 486\nn003390 208\nn003391 334\nn003392 601\nn003393 261\nn003394 294\nn003395 492\nn003396 469\nn003397 279\nn003398 266\nn003399 318\nn003400 302\nn003401 439\nn003402 286\nn003403 323\nn003404 332\nn003405 250\nn003406 285\nn003407 347\nn003408 344\nn003409 324\nn003410 253\nn003411 458\nn003412 290\nn003413 368\nn003414 300\nn003416 416\nn003417 195\nn003418 226\nn003419 339\nn003420 301\nn003421 392\nn003422 288\nn003423 340\nn003424 341\nn003425 446\nn003426 289\nn003427 151\nn003428 366\nn003429 507\nn003431 301\nn003432 246\nn003433 329\nn003434 428\nn003435 311\nn003437 210\nn003438 432\nn003439 420\nn003440 365\nn003441 317\nn003442 212\nn003443 383\nn003444 391\nn003445 374\nn003446 275\nn003447 300\nn003448 386\nn003449 319\nn003450 375\nn003451 236\nn003452 286\nn003453 613\nn003454 259\nn003455 274\nn003456 223\nn003457 197\nn003458 160\nn003459 443\nn003460 413\nn003462 309\nn003463 335\nn003464 234\nn003465 385\nn003466 418\nn003467 159\nn003469 321\nn003470 494\nn003471 227\nn003472 124\nn003473 194\nn003474 413\nn003475 556\nn003476 569\nn003477 219\nn003478 627\nn003479 270\nn003481 424\nn003482 423\nn003483 264\nn003484 523\nn003485 551\nn003486 437\nn003487 434\nn003488 521\nn003489 286\nn003491 198\nn003492 447\nn003493 265\nn003494 349\nn003495 478\nn003496 349\nn003497 336\nn003498 471\nn003499 367\nn003500 480\nn003501 417\nn003502 390\nn003503 261\nn003504 436\nn003505 262\nn003506 437\nn003507 343\nn003508 397\nn003509 349\nn003510 355\nn003511 251\nn003512 324\nn003514 303\nn003515 259\nn003516 453\nn003517 362\nn003518 308\nn003519 248\nn003520 501\nn003521 406\nn003522 239\nn003523 508\nn003524 403\nn003525 453\nn003526 296\nn003527 498\nn003528 469\nn003529 260\nn003530 200\nn003531 443\nn003532 441\nn003533 305\nn003534 212\nn003536 207\nn003537 411\nn003538 282\nn003539 198\nn003541 454\nn003542 444\nn003543 252\nn003544 249\nn003545 301\nn003546 301\nn003547 394\nn003548 239\nn003549 343\nn003550 262\nn003551 336\nn003552 272\nn003553 456\nn003555 356\nn003556 290\nn003557 368\nn003558 202\nn003559 378\nn003560 284\nn003561 242\nn003562 208\nn003563 258\nn003564 467\nn003565 511\nn003566 114\nn003567 345\nn003568 363\nn003569 368\nn003571 286\nn003572 324\nn003573 217\nn003574 266\nn003575 185\nn003576 286\nn003577 300\nn003578 294\nn003579 268\nn003580 427\nn003581 496\nn003582 241\nn003583 533\nn003584 338\nn003585 321\nn003586 491\nn003587 508\nn003588 338\nn003590 225\nn003591 335\nn003593 255\nn003594 403\nn003595 292\nn003596 288\nn003597 315\nn003598 455\nn003599 312\nn003600 212\nn003601 401\nn003602 428\nn003603 338\nn003604 447\nn003605 505\nn003607 140\nn003608 270\nn003609 161\nn003610 350\nn003612 203\nn003613 413\nn003614 379\nn003615 578\nn003616 529\nn003617 518\nn003618 369\nn003619 450\nn003620 470\nn003621 226\nn003622 373\nn003623 417\nn003624 463\nn003625 292\nn003626 415\nn003627 183\nn003628 395\nn003629 268\nn003630 462\nn003631 299\nn003632 459\nn003633 216\nn003634 380\nn003636 341\nn003637 438\nn003638 316\nn003639 360\nn003640 174\nn003641 458\nn003642 183\nn003643 199\nn003644 212\nn003645 336\nn003646 460\nn003647 443\nn003648 294\nn003649 406\nn003650 449\nn003651 541\nn003652 237\nn003654 309\nn003655 216\nn003656 371\nn003657 313\nn003658 591\nn003659 267\nn003660 469\nn003661 216\nn003663 381\nn003664 259\nn003666 201\nn003667 404\nn003668 333\nn003669 357\nn003670 351\nn003671 266\nn003672 453\nn003673 413\nn003674 358\nn003676 389\nn003677 376\nn003678 325\nn003679 337\nn003680 558\nn003681 287\nn003682 380\nn003683 344\nn003684 237\nn003685 479\nn003686 277\nn003687 333\nn003688 340\nn003689 327\nn003690 342\nn003691 394\nn003692 385\nn003693 516\nn003694 478\nn003695 449\nn003696 185\nn003697 206\nn003698 205\nn003699 162\nn003700 386\nn003701 501\nn003702 304\nn003703 341\nn003704 406\nn003705 210\nn003706 495\nn003707 199\nn003708 335\nn003709 442\nn003710 264\nn003711 175\nn003712 660\nn003713 289\nn003714 462\nn003715 482\nn003716 536\nn003717 279\nn003718 257\nn003719 415\nn003720 376\nn003721 213\nn003722 412\nn003723 320\nn003724 514\nn003726 533\nn003727 426\nn003729 443\nn003730 268\nn003731 324\nn003732 206\nn003733 285\nn003734 328\nn003735 543\nn003736 406\nn003737 348\nn003738 409\nn003739 323\nn003740 314\nn003741 361\nn003742 409\nn003743 243\nn003744 331\nn003745 325\nn003746 589\nn003747 416\nn003748 407\nn003749 343\nn003750 380\nn003751 358\nn003752 233\nn003753 350\nn003754 402\nn003755 235\nn003756 374\nn003757 412\nn003758 429\nn003759 262\nn003760 390\nn003761 349\nn003762 247\nn003763 429\nn003764 460\nn003766 353\nn003767 329\nn003768 422\nn003769 525\nn003770 225\nn003771 237\nn003772 422\nn003773 492\nn003774 334\nn003776 535\nn003777 448\nn003778 379\nn003779 368\nn003780 404\nn003781 271\nn003782 214\nn003783 282\nn003784 551\nn003785 450\nn003787 429\nn003788 484\nn003789 466\nn003790 326\nn003791 244\nn003792 383\nn003793 343\nn003794 309\nn003795 404\nn003796 414\nn003797 241\nn003798 205\nn003799 542\nn003800 435\nn003801 250\nn003802 138\nn003803 494\nn003805 280\nn003806 327\nn003807 267\nn003808 433\nn003809 200\nn003810 262\nn003811 323\nn003812 359\nn003813 292\nn003814 358\nn003815 515\nn003816 424\nn003817 581\nn003818 350\nn003819 619\nn003820 394\nn003821 426\nn003822 501\nn003823 298\nn003824 344\nn003825 326\nn003826 330\nn003827 214\nn003828 341\nn003829 404\nn003830 156\nn003831 595\nn003832 488\nn003833 291\nn003834 424\nn003835 333\nn003837 333\nn003838 531\nn003839 324\nn003840 344\nn003841 316\nn003842 276\nn003843 402\nn003844 316\nn003845 180\nn003846 481\nn003847 271\nn003848 324\nn003849 447\nn003850 347\nn003851 201\nn003852 278\nn003853 339\nn003854 416\nn003855 403\nn003856 487\nn003857 389\nn003858 380\nn003859 208\nn003860 491\nn003861 219\nn003862 505\nn003863 470\nn003864 409\nn003865 499\nn003866 447\nn003867 234\nn003868 351\nn003869 251\nn003870 475\nn003871 354\nn003872 355\nn003874 362\nn003875 396\nn003876 371\nn003877 468\nn003878 255\nn003879 398\nn003880 356\nn003882 299\nn003883 421\nn003884 351\nn003885 440\nn003886 562\nn003887 357\nn003888 279\nn003889 270\nn003890 235\nn003891 290\nn003892 139\nn003893 304\nn003895 553\nn003897 539\nn003898 221\nn003899 442\nn003900 329\nn003901 247\nn003902 257\nn003903 330\nn003904 467\nn003905 483\nn003906 402\nn003907 198\nn003908 566\nn003909 264\nn003910 347\nn003911 403\nn003912 551\nn003913 499\nn003914 261\nn003915 224\nn003916 356\nn003918 384\nn003919 524\nn003920 217\nn003921 409\nn003922 378\nn003923 581\nn003924 518\nn003925 270\nn003926 260\nn003927 275\nn003928 355\nn003929 325\nn003930 323\nn003931 440\nn003932 366\nn003933 298\nn003934 242\nn003935 304\nn003936 328\nn003937 455\nn003938 504\nn003939 396\nn003940 394\nn003941 422\nn003942 296\nn003943 374\nn003944 412\nn003945 304\nn003946 458\nn003947 309\nn003948 317\nn003949 334\nn003950 362\nn003951 332\nn003952 327\nn003953 312\nn003954 319\nn003955 359\nn003956 419\nn003957 381\nn003959 411\nn003960 490\nn003961 206\nn003962 504\nn003963 244\nn003964 221\nn003965 519\nn003966 314\nn003967 442\nn003968 542\nn003969 412\nn003970 243\nn003971 155\nn003972 455\nn003973 525\nn003974 327\nn003975 407\nn003976 216\nn003977 613\nn003978 210\nn003979 533\nn003980 471\nn003981 378\nn003982 466\nn003983 443\nn003984 334\nn003985 355\nn003986 474\nn003987 280\nn003988 615\nn003989 351\nn003990 356\nn003991 405\nn003992 242\nn003993 523\nn003994 205\nn003995 570\nn003996 203\nn003997 230\nn003998 502\nn003999 295\nn004000 433\nn004001 510\nn004002 356\nn004003 241\nn004004 334\nn004005 458\nn004006 161\nn004008 326\nn004009 320\nn004011 120\nn004012 305\nn004013 390\nn004014 369\nn004015 214\nn004016 450\nn004017 335\nn004018 284\nn004019 361\nn004020 270\nn004021 390\nn004022 158\nn004023 477\nn004024 188\nn004025 348\nn004026 486\nn004027 358\nn004028 254\nn004029 372\nn004030 370\nn004031 179\nn004032 372\nn004033 323\nn004034 382\nn004035 270\nn004036 386\nn004037 394\nn004038 588\nn004039 220\nn004040 490\nn004041 257\nn004042 654\nn004043 477\nn004044 348\nn004045 580\nn004046 495\nn004047 224\nn004048 488\nn004049 454\nn004051 252\nn004052 523\nn004053 483\nn004054 400\nn004055 185\nn004056 420\nn004057 306\nn004058 365\nn004059 240\nn004060 386\nn004061 484\nn004062 373\nn004063 362\nn004065 333\nn004066 398\nn004067 210\nn004069 390\nn004071 317\nn004072 464\nn004073 435\nn004074 267\nn004075 369\nn004076 374\nn004077 276\nn004079 420\nn004080 319\nn004081 354\nn004082 233\nn004083 352\nn004084 314\nn004087 318\nn004088 357\nn004089 279\nn004090 372\nn004091 206\nn004092 390\nn004093 633\nn004094 326\nn004095 322\nn004096 268\nn004097 642\nn004098 323\nn004099 305\nn004100 345\nn004101 407\nn004102 384\nn004103 254\nn004104 356\nn004105 439\nn004106 413\nn004107 360\nn004108 418\nn004109 395\nn004110 445\nn004111 504\nn004112 135\nn004113 352\nn004114 376\nn004115 373\nn004116 224\nn004117 295\nn004119 480\nn004120 498\nn004121 394\nn004122 474\nn004124 539\nn004125 431\nn004126 378\nn004127 275\nn004128 389\nn004129 422\nn004130 338\nn004131 381\nn004132 373\nn004133 337\nn004134 400\nn004135 328\nn004136 340\nn004137 158\nn004138 285\nn004139 341\nn004140 222\nn004141 447\nn004142 637\nn004143 439\nn004144 555\nn004145 457\nn004146 312\nn004147 332\nn004148 217\nn004149 450\nn004150 507\nn004151 535\nn004152 613\nn004153 409\nn004154 478\nn004155 286\nn004156 375\nn004158 460\nn004159 455\nn004160 364\nn004161 397\nn004162 358\nn004163 363\nn004164 282\nn004165 348\nn004166 329\nn004167 399\nn004168 557\nn004169 358\nn004170 389\nn004171 364\nn004172 360\nn004173 460\nn004174 403\nn004175 293\nn004176 367\nn004177 328\nn004178 565\nn004179 400\nn004181 357\nn004182 454\nn004183 331\nn004184 273\nn004185 421\nn004186 144\nn004187 524\nn004188 374\nn004189 380\nn004190 336\nn004191 481\nn004192 383\nn004193 490\nn004194 412\nn004195 331\nn004196 236\nn004197 363\nn004198 515\nn004201 401\nn004202 322\nn004203 488\nn004204 345\nn004205 308\nn004206 402\nn004209 502\nn004210 437\nn004211 257\nn004212 340\nn004213 303\nn004214 360\nn004215 493\nn004216 594\nn004217 195\nn004218 426\nn004220 458\nn004221 314\nn004222 433\nn004223 309\nn004224 210\nn004225 228\nn004226 390\nn004227 485\nn004228 219\nn004229 332\nn004230 378\nn004231 534\nn004232 296\nn004234 360\nn004235 352\nn004236 349\nn004237 452\nn004238 317\nn004241 241\nn004242 221\nn004244 405\nn004245 223\nn004246 225\nn004247 277\nn004248 330\nn004249 370\nn004250 215\nn004251 506\nn004252 505\nn004253 300\nn004254 320\nn004255 373\nn004256 390\nn004257 558\nn004258 247\nn004259 201\nn004260 254\nn004261 278\nn004262 368\nn004263 444\nn004264 365\nn004265 608\nn004266 334\nn004267 296\nn004268 449\nn004269 415\nn004270 697\nn004271 214\nn004272 359\nn004273 451\nn004274 398\nn004275 557\nn004276 307\nn004277 290\nn004278 426\nn004279 380\nn004280 344\nn004282 281\nn004283 339\nn004284 145\nn004285 373\nn004286 395\nn004287 441\nn004288 454\nn004290 602\nn004291 312\nn004292 343\nn004293 234\nn004294 498\nn004295 447\nn004296 250\nn004297 359\nn004299 453\nn004300 244\nn004301 445\nn004302 248\nn004303 357\nn004304 435\nn004305 293\nn004306 240\nn004307 251\nn004308 439\nn004309 380\nn004310 338\nn004311 266\nn004312 372\nn004313 390\nn004314 434\nn004315 241\nn004316 313\nn004317 433\nn004318 332\nn004319 309\nn004320 484\nn004321 437\nn004322 239\nn004323 258\nn004324 446\nn004325 361\nn004326 344\nn004327 419\nn004328 329\nn004329 324\nn004330 262\nn004331 427\nn004332 397\nn004334 293\nn004335 200\nn004336 267\nn004337 539\nn004339 263\nn004340 211\nn004341 303\nn004342 354\nn004343 323\nn004344 387\nn004345 279\nn004347 313\nn004348 462\nn004349 442\nn004350 404\nn004351 442\nn004352 365\nn004354 302\nn004355 313\nn004356 531\nn004358 360\nn004359 424\nn004360 296\nn004361 386\nn004362 293\nn004363 318\nn004364 367\nn004365 501\nn004367 364\nn004368 359\nn004369 398\nn004370 376\nn004371 448\nn004373 663\nn004374 389\nn004375 513\nn004376 284\nn004377 243\nn004378 303\nn004379 309\nn004381 436\nn004382 350\nn004383 360\nn004384 285\nn004385 387\nn004386 456\nn004388 209\nn004389 213\nn004390 375\nn004391 512\nn004392 532\nn004393 328\nn004395 327\nn004396 346\nn004397 383\nn004398 426\nn004399 476\nn004401 442\nn004402 302\nn004403 245\nn004404 198\nn004405 327\nn004406 373\nn004407 617\nn004408 547\nn004409 335\nn004410 454\nn004412 475\nn004413 493\nn004414 491\nn004415 448\nn004416 404\nn004417 259\nn004418 589\nn004419 312\nn004420 365\nn004421 317\nn004422 580\nn004423 104\nn004425 370\nn004426 492\nn004427 282\nn004428 413\nn004429 411\nn004430 269\nn004431 389\nn004432 348\nn004433 339\nn004434 118\nn004435 211\nn004436 277\nn004437 245\nn004438 209\nn004439 598\nn004441 304\nn004442 462\nn004443 575\nn004444 368\nn004445 344\nn004446 465\nn004447 432\nn004448 353\nn004450 323\nn004451 261\nn004452 419\nn004453 146\nn004454 290\nn004455 320\nn004456 393\nn004457 564\nn004458 134\nn004459 357\nn004460 355\nn004462 312\nn004463 122\nn004464 410\nn004465 365\nn004466 274\nn004467 485\nn004468 184\nn004470 446\nn004471 475\nn004472 362\nn004473 478\nn004474 259\nn004475 409\nn004476 447\nn004477 345\nn004478 396\nn004479 401\nn004480 381\nn004481 420\nn004483 410\nn004484 291\nn004485 481\nn004487 496\nn004488 490\nn004489 512\nn004490 551\nn004491 328\nn004492 349\nn004493 512\nn004494 349\nn004495 150\nn004496 235\nn004497 400\nn004498 314\nn004499 333\nn004500 423\nn004501 235\nn004502 377\nn004503 331\nn004504 373\nn004505 462\nn004506 369\nn004507 506\nn004508 607\nn004509 367\nn004510 573\nn004511 478\nn004512 403\nn004513 246\nn004514 278\nn004515 424\nn004516 322\nn004517 431\nn004518 423\nn004519 250\nn004520 227\nn004521 414\nn004522 307\nn004523 392\nn004524 209\nn004525 439\nn004526 298\nn004527 310\nn004528 556\nn004529 319\nn004530 332\nn004531 206\nn004532 202\nn004533 605\nn004534 276\nn004535 530\nn004536 393\nn004537 412\nn004538 458\nn004539 284\nn004540 216\nn004541 388\nn004542 496\nn004543 674\nn004544 325\nn004545 133\nn004546 239\nn004547 361\nn004548 471\nn004549 511\nn004550 526\nn004551 360\nn004552 235\nn004553 436\nn004554 381\nn004556 168\nn004557 224\nn004558 314\nn004559 544\nn004560 431\nn004561 564\nn004562 461\nn004564 342\nn004565 588\nn004566 300\nn004567 361\nn004568 424\nn004569 463\nn004570 234\nn004571 304\nn004572 415\nn004573 197\nn004574 163\nn004575 416\nn004576 467\nn004577 177\nn004578 322\nn004579 216\nn004581 227\nn004582 397\nn004583 251\nn004584 149\nn004585 300\nn004587 376\nn004589 687\nn004590 336\nn004591 562\nn004592 341\nn004593 356\nn004594 375\nn004595 298\nn004596 290\nn004597 379\nn004598 535\nn004599 358\nn004600 293\nn004601 342\nn004602 422\nn004603 435\nn004604 334\nn004605 421\nn004606 220\nn004607 326\nn004608 375\nn004609 191\nn004610 667\nn004611 213\nn004612 366\nn004613 362\nn004614 298\nn004615 696\nn004616 482\nn004617 616\nn004618 340\nn004620 497\nn004621 259\nn004622 326\nn004623 443\nn004624 210\nn004625 216\nn004626 212\nn004627 395\nn004628 349\nn004629 429\nn004630 345\nn004631 169\nn004632 243\nn004633 393\nn004636 332\nn004637 306\nn004638 234\nn004639 403\nn004640 495\nn004641 447\nn004642 202\nn004643 462\nn004644 384\nn004645 252\nn004646 446\nn004647 508\nn004648 424\nn004649 304\nn004650 345\nn004651 539\nn004653 460\nn004654 484\nn004655 552\nn004656 600\nn004657 401\nn004659 471\nn004664 309\nn004665 249\nn004666 530\nn004667 410\nn004668 393\nn004669 234\nn004670 216\nn004671 284\nn004672 457\nn004673 145\nn004674 338\nn004675 295\nn004676 470\nn004677 399\nn004680 251\nn004681 465\nn004682 260\nn004683 429\nn004684 359\nn004685 207\nn004686 529\nn004687 393\nn004688 491\nn004689 383\nn004690 419\nn004691 219\nn004692 152\nn004693 539\nn004694 126\nn004695 561\nn004696 216\nn004697 410\nn004698 202\nn004699 255\nn004700 423\nn004701 329\nn004702 355\nn004703 609\nn004704 260\nn004705 227\nn004706 371\nn004707 195\nn004708 206\nn004710 384\nn004711 394\nn004713 428\nn004714 475\nn004715 253\nn004716 151\nn004717 343\nn004718 564\nn004720 510\nn004721 354\nn004722 302\nn004724 467\nn004725 297\nn004726 251\nn004727 343\nn004728 371\nn004729 468\nn004730 429\nn004731 377\nn004732 348\nn004733 220\nn004734 340\nn004735 442\nn004736 419\nn004737 523\nn004739 315\nn004740 457\nn004741 362\nn004742 455\nn004744 425\nn004745 487\nn004746 299\nn004747 404\nn004748 366\nn004749 267\nn004750 400\nn004751 253\nn004752 170\nn004753 459\nn004754 367\nn004755 375\nn004757 220\nn004758 334\nn004759 364\nn004760 266\nn004761 481\nn004762 411\nn004763 449\nn004764 458\nn004765 339\nn004766 176\nn004767 219\nn004768 236\nn004769 382\nn004770 274\nn004772 186\nn004773 503\nn004774 350\nn004775 245\nn004776 276\nn004777 308\nn004778 271\nn004779 292\nn004780 356\nn004781 462\nn004782 483\nn004783 329\nn004784 361\nn004785 511\nn004786 379\nn004787 523\nn004790 370\nn004791 403\nn004792 397\nn004794 478\nn004796 245\nn004797 403\nn004799 356\nn004800 602\nn004802 419\nn004804 395\nn004805 284\nn004806 812\nn004807 372\nn004808 399\nn004809 403\nn004810 423\nn004814 265\nn004816 303\nn004817 207\nn004818 282\nn004819 363\nn004820 607\nn004821 375\nn004822 385\nn004823 424\nn004824 290\nn004825 483\nn004827 353\nn004828 206\nn004829 410\nn004830 480\nn004831 652\nn004832 302\nn004833 245\nn004834 417\nn004835 453\nn004836 202\nn004837 269\nn004838 358\nn004839 300\nn004840 313\nn004841 327\nn004842 336\nn004843 604\nn004844 652\nn004845 282\nn004846 456\nn004847 373\nn004848 296\nn004849 397\nn004851 451\nn004852 290\nn004853 218\nn004854 350\nn004855 338\nn004856 298\nn004857 353\nn004858 336\nn004859 466\nn004860 257\nn004861 221\nn004862 382\nn004863 322\nn004864 391\nn004865 267\nn004866 351\nn004867 312\nn004868 486\nn004869 339\nn004870 233\nn004871 474\nn004872 366\nn004873 315\nn004874 391\nn004875 320\nn004876 358\nn004877 224\nn004878 479\nn004879 404\nn004880 412\nn004881 490\nn004882 362\nn004883 213\nn004884 440\nn004886 203\nn004887 388\nn004888 519\nn004889 555\nn004890 246\nn004892 135\nn004893 413\nn004894 430\nn004895 347\nn004896 367\nn004897 359\nn004899 287\nn004900 427\nn004901 340\nn004902 473\nn004903 373\nn004904 488\nn004906 358\nn004907 351\nn004908 489\nn004909 318\nn004910 307\nn004912 278\nn004913 626\nn004914 215\nn004916 448\nn004917 353\nn004918 307\nn004919 434\nn004921 186\nn004922 362\nn004923 220\nn004924 416\nn004926 341\nn004927 389\nn004928 387\nn004929 428\nn004930 283\nn004931 208\nn004932 268\nn004933 407\nn004934 303\nn004935 449\nn004936 678\nn004937 320\nn004938 444\nn004939 321\nn004940 525\nn004941 290\nn004942 398\nn004943 343\nn004944 413\nn004946 242\nn004947 377\nn004948 448\nn004949 539\nn004950 331\nn004951 463\nn004952 411\nn004953 421\nn004954 333\nn004955 213\nn004956 289\nn004957 582\nn004958 209\nn004959 449\nn004960 357\nn004961 303\nn004962 317\nn004963 391\nn004964 263\nn004965 371\nn004966 296\nn004967 442\nn004968 399\nn004969 345\nn004970 339\nn004971 352\nn004972 358\nn004973 383\nn004974 228\nn004975 307\nn004976 410\nn004977 416\nn004978 444\nn004979 501\nn004980 397\nn004981 339\nn004982 309\nn004983 233\nn004984 531\nn004985 276\nn004986 361\nn004987 423\nn004988 493\nn004989 302\nn004990 485\nn004991 344\nn004992 502\nn004993 384\nn004994 465\nn004995 277\nn004996 358\nn004997 343\nn004998 475\nn005000 276\nn005001 419\nn005002 222\nn005003 490\nn005004 318\nn005005 439\nn005006 235\nn005007 451\nn005008 215\nn005009 438\nn005010 503\nn005011 485\nn005012 377\nn005013 418\nn005014 503\nn005015 645\nn005016 374\nn005017 512\nn005018 674\nn005019 467\nn005020 346\nn005021 371\nn005022 575\nn005023 262\nn005024 390\nn005025 303\nn005026 437\nn005027 349\nn005028 300\nn005029 352\nn005030 422\nn005031 447\nn005032 297\nn005033 267\nn005034 402\nn005035 302\nn005036 424\nn005037 330\nn005038 308\nn005039 346\nn005040 205\nn005041 187\nn005042 320\nn005043 365\nn005044 195\nn005045 361\nn005046 372\nn005047 368\nn005048 292\nn005050 365\nn005051 225\nn005052 312\nn005053 537\nn005054 406\nn005055 380\nn005056 406\nn005057 480\nn005058 397\nn005059 390\nn005060 390\nn005061 306\nn005062 352\nn005063 246\nn005064 333\nn005065 395\nn005066 368\nn005067 383\nn005069 416\nn005070 324\nn005071 322\nn005072 355\nn005074 363\nn005075 381\nn005076 346\nn005077 297\nn005078 444\nn005079 368\nn005080 576\nn005081 455\nn005082 533\nn005083 300\nn005084 439\nn005085 462\nn005086 528\nn005087 416\nn005089 693\nn005090 374\nn005091 230\nn005092 431\nn005093 298\nn005094 471\nn005095 448\nn005096 333\nn005097 302\nn005098 382\nn005099 304\nn005100 267\nn005102 347\nn005103 486\nn005105 321\nn005106 361\nn005107 279\nn005108 229\nn005109 261\nn005110 413\nn005111 263\nn005113 284\nn005115 515\nn005116 405\nn005117 317\nn005118 243\nn005119 401\nn005121 359\nn005124 522\nn005125 647\nn005126 365\nn005127 322\nn005128 326\nn005129 517\nn005130 393\nn005131 291\nn005132 362\nn005133 350\nn005134 371\nn005138 460\nn005139 373\nn005140 333\nn005141 417\nn005142 517\nn005143 106\nn005144 212\nn005146 271\nn005147 436\nn005149 388\nn005150 287\nn005151 369\nn005152 337\nn005153 473\nn005158 369\nn005160 227\nn005161 283\nn005162 461\nn005163 271\nn005164 578\nn005165 195\nn005166 207\nn005167 534\nn005168 450\nn005169 344\nn005170 364\nn005171 213\nn005172 328\nn005173 331\nn005174 204\nn005175 647\nn005176 618\nn005177 213\nn005178 344\nn005180 587\nn005182 291\nn005183 257\nn005184 248\nn005185 253\nn005186 325\nn005187 318\nn005189 288\nn005190 423\nn005191 604\nn005192 160\nn005193 113\nn005194 323\nn005195 309\nn005196 415\nn005197 397\nn005198 422\nn005199 269\nn005200 406\nn005201 272\nn005202 441\nn005203 340\nn005204 425\nn005205 386\nn005206 514\nn005207 362\nn005208 371\nn005209 216\nn005210 367\nn005211 316\nn005212 393\nn005213 404\nn005214 172\nn005215 303\nn005216 321\nn005217 378\nn005218 341\nn005219 589\nn005220 213\nn005221 391\nn005222 363\nn005223 428\nn005224 285\nn005227 524\nn005228 436\nn005229 330\nn005230 357\nn005231 330\nn005232 284\nn005234 379\nn005235 347\nn005236 340\nn005237 233\nn005238 348\nn005239 383\nn005240 463\nn005241 230\nn005242 372\nn005243 259\nn005244 319\nn005245 327\nn005246 318\nn005247 347\nn005248 316\nn005249 617\nn005250 337\nn005251 422\nn005252 326\nn005253 389\nn005254 450\nn005255 278\nn005256 228\nn005257 291\nn005258 283\nn005259 339\nn005260 303\nn005261 218\nn005262 489\nn005263 403\nn005264 418\nn005265 340\nn005266 513\nn005267 274\nn005268 320\nn005269 527\nn005270 474\nn005271 297\nn005272 301\nn005273 359\nn005274 281\nn005275 384\nn005276 278\nn005277 515\nn005278 304\nn005279 520\nn005281 262\nn005283 532\nn005284 300\nn005285 461\nn005286 237\nn005287 414\nn005288 323\nn005289 368\nn005290 487\nn005291 335\nn005292 591\nn005293 275\nn005295 423\nn005296 317\nn005297 396\nn005298 352\nn005299 397\nn005300 388\nn005302 369\nn005304 272\nn005305 569\nn005307 440\nn005308 144\nn005309 331\nn005310 357\nn005311 422\nn005313 401\nn005314 405\nn005315 233\nn005317 282\nn005318 586\nn005320 417\nn005321 237\nn005322 357\nn005323 490\nn005324 267\nn005325 298\nn005327 298\nn005329 214\nn005330 410\nn005331 339\nn005332 273\nn005333 372\nn005335 501\nn005336 330\nn005337 390\nn005338 583\nn005339 333\nn005341 468\nn005342 482\nn005343 508\nn005344 144\nn005345 473\nn005346 392\nn005348 370\nn005349 395\nn005350 370\nn005351 193\nn005352 226\nn005353 311\nn005354 171\nn005355 315\nn005356 391\nn005357 331\nn005358 339\nn005360 320\nn005361 340\nn005362 448\nn005363 327\nn005364 517\nn005365 247\nn005366 384\nn005367 397\nn005368 279\nn005369 417\nn005370 246\nn005371 412\nn005372 245\nn005374 273\nn005375 386\nn005376 429\nn005378 441\nn005379 292\nn005381 309\nn005382 338\nn005383 445\nn005384 518\nn005385 307\nn005386 350\nn005387 355\nn005388 248\nn005389 282\nn005390 482\nn005391 334\nn005392 415\nn005393 216\nn005394 382\nn005395 500\nn005396 597\nn005397 390\nn005398 347\nn005399 290\nn005400 435\nn005401 444\nn005402 434\nn005403 276\nn005404 227\nn005405 222\nn005406 388\nn005407 226\nn005408 522\nn005409 381\nn005410 431\nn005411 365\nn005412 478\nn005413 603\nn005414 555\nn005415 618\nn005416 534\nn005418 519\nn005419 570\nn005420 548\nn005421 304\nn005422 220\nn005423 483\nn005424 379\nn005426 464\nn005428 269\nn005429 308\nn005430 381\nn005431 412\nn005432 489\nn005433 456\nn005434 165\nn005435 271\nn005436 414\nn005437 441\nn005438 191\nn005439 120\nn005440 362\nn005441 418\nn005442 583\nn005443 466\nn005444 257\nn005445 452\nn005446 413\nn005447 421\nn005449 655\nn005450 241\nn005451 342\nn005452 479\nn005453 404\nn005454 345\nn005455 391\nn005456 489\nn005457 390\nn005458 266\nn005459 375\nn005460 470\nn005461 465\nn005462 282\nn005463 383\nn005464 587\nn005465 287\nn005466 326\nn005467 404\nn005468 391\nn005469 443\nn005470 311\nn005471 396\nn005472 178\nn005475 533\nn005476 468\nn005477 404\nn005478 357\nn005479 437\nn005480 347\nn005481 391\nn005482 334\nn005483 595\nn005484 347\nn005485 445\nn005486 244\nn005487 391\nn005488 255\nn005489 496\nn005491 534\nn005492 215\nn005493 575\nn005494 201\nn005495 478\nn005496 518\nn005497 425\nn005498 264\nn005499 388\nn005500 216\nn005501 475\nn005502 355\nn005503 358\nn005504 310\nn005505 261\nn005506 412\nn005507 425\nn005508 449\nn005509 382\nn005510 262\nn005511 281\nn005512 331\nn005514 448\nn005515 202\nn005516 365\nn005517 317\nn005518 453\nn005519 404\nn005520 288\nn005521 131\nn005522 436\nn005523 365\nn005524 265\nn005525 389\nn005526 368\nn005527 380\nn005528 250\nn005529 208\nn005530 307\nn005531 309\nn005532 448\nn005533 570\nn005534 428\nn005535 270\nn005536 257\nn005537 314\nn005538 469\nn005539 402\nn005540 379\nn005541 497\nn005542 352\nn005543 466\nn005544 295\nn005545 293\nn005546 341\nn005547 272\nn005548 540\nn005549 290\nn005550 272\nn005551 333\nn005553 528\nn005554 353\nn005555 332\nn005556 329\nn005557 374\nn005558 264\nn005559 217\nn005560 467\nn005561 217\nn005562 433\nn005563 310\nn005566 432\nn005567 462\nn005568 460\nn005569 298\nn005570 302\nn005571 433\nn005572 297\nn005573 331\nn005574 366\nn005575 479\nn005576 290\nn005578 403\nn005579 400\nn005580 312\nn005581 299\nn005582 257\nn005583 518\nn005584 384\nn005585 493\nn005586 364\nn005587 318\nn005588 458\nn005589 407\nn005590 348\nn005591 458\nn005592 404\nn005593 451\nn005594 286\nn005595 536\nn005596 348\nn005597 203\nn005598 389\nn005599 426\nn005600 434\nn005601 213\nn005602 351\nn005604 328\nn005605 456\nn005606 306\nn005608 339\nn005609 534\nn005610 451\nn005611 491\nn005613 237\nn005614 314\nn005615 386\nn005616 462\nn005617 396\nn005618 435\nn005620 396\nn005622 418\nn005624 410\nn005625 251\nn005626 241\nn005627 304\nn005628 405\nn005629 290\nn005631 327\nn005632 417\nn005634 397\nn005635 270\nn005637 349\nn005638 457\nn005640 451\nn005641 387\nn005642 333\nn005643 555\nn005644 369\nn005645 420\nn005646 307\nn005647 372\nn005649 359\nn005650 478\nn005651 479\nn005653 590\nn005654 440\nn005655 160\nn005656 305\nn005657 465\nn005658 352\nn005659 296\nn005660 361\nn005661 465\nn005662 486\nn005663 399\nn005665 433\nn005667 354\nn005669 382\nn005671 444\nn005672 379\nn005673 253\nn005674 514\nn005675 303\nn005676 291\nn005677 521\nn005678 262\nn005679 285\nn005681 376\nn005682 432\nn005683 301\nn005684 346\nn005685 253\nn005686 227\nn005687 205\nn005688 387\nn005689 175\nn005690 313\nn005691 284\nn005692 235\nn005694 228\nn005696 334\nn005697 398\nn005698 226\nn005699 265\nn005700 215\nn005701 411\nn005702 316\nn005703 305\nn005704 212\nn005705 339\nn005707 268\nn005708 461\nn005710 289\nn005711 300\nn005712 366\nn005713 513\nn005714 265\nn005715 294\nn005716 367\nn005717 362\nn005718 408\nn005719 364\nn005720 369\nn005721 488\nn005722 267\nn005724 272\nn005725 316\nn005727 197\nn005728 162\nn005729 330\nn005731 339\nn005732 475\nn005733 324\nn005734 267\nn005735 314\nn005736 314\nn005737 458\nn005738 543\nn005739 261\nn005740 363\nn005741 286\nn005742 324\nn005743 451\nn005744 218\nn005745 465\nn005746 280\nn005747 531\nn005749 231\nn005750 372\nn005751 547\nn005752 405\nn005753 436\nn005754 240\nn005756 254\nn005757 314\nn005759 454\nn005760 221\nn005761 500\nn005763 354\nn005765 333\nn005766 327\nn005767 487\nn005768 265\nn005769 358\nn005770 445\nn005771 388\nn005772 133\nn005773 312\nn005774 338\nn005775 390\nn005777 145\nn005778 275\nn005779 587\nn005780 484\nn005782 520\nn005783 455\nn005785 248\nn005787 419\nn005788 302\nn005789 418\nn005790 257\nn005791 332\nn005792 461\nn005793 303\nn005795 275\nn005796 224\nn005797 380\nn005798 125\nn005800 171\nn005801 301\nn005802 247\nn005803 426\nn005804 275\nn005805 280\nn005806 371\nn005807 307\nn005808 227\nn005809 512\nn005810 286\nn005811 232\nn005812 332\nn005813 311\nn005814 566\nn005815 431\nn005816 276\nn005817 265\nn005818 260\nn005819 122\nn005820 437\nn005821 350\nn005822 282\nn005823 204\nn005825 331\nn005826 340\nn005827 352\nn005828 516\nn005829 576\nn005830 420\nn005834 262\nn005835 218\nn005836 353\nn005837 388\nn005838 276\nn005839 401\nn005840 368\nn005841 267\nn005842 350\nn005843 424\nn005844 174\nn005845 205\nn005846 397\nn005847 314\nn005848 364\nn005849 354\nn005850 372\nn005851 137\nn005852 407\nn005853 268\nn005854 362\nn005855 508\nn005857 343\nn005858 245\nn005859 356\nn005860 378\nn005861 297\nn005862 271\nn005863 378\nn005865 288\nn005866 311\nn005867 241\nn005868 397\nn005869 437\nn005870 193\nn005871 271\nn005873 280\nn005874 322\nn005875 316\nn005876 366\nn005877 325\nn005878 314\nn005879 297\nn005880 379\nn005881 292\nn005882 301\nn005883 320\nn005884 422\nn005885 212\nn005886 337\nn005887 346\nn005888 324\nn005889 236\nn005890 361\nn005891 473\nn005892 421\nn005893 207\nn005894 297\nn005895 393\nn005896 205\nn005897 351\nn005898 608\nn005899 211\nn005900 493\nn005901 586\nn005902 482\nn005903 400\nn005904 467\nn005905 337\nn005906 399\nn005907 267\nn005908 360\nn005909 359\nn005910 318\nn005911 207\nn005912 389\nn005913 376\nn005914 487\nn005916 344\nn005917 341\nn005918 384\nn005919 331\nn005920 343\nn005921 367\nn005922 299\nn005923 409\nn005924 236\nn005925 456\nn005926 348\nn005927 330\nn005928 193\nn005929 335\nn005930 340\nn005931 464\nn005932 416\nn005933 418\nn005934 415\nn005935 214\nn005936 288\nn005937 423\nn005938 448\nn005939 246\nn005940 421\nn005941 352\nn005942 300\nn005943 318\nn005944 403\nn005945 354\nn005946 436\nn005947 344\nn005948 412\nn005949 462\nn005950 214\nn005951 435\nn005952 357\nn005953 529\nn005954 281\nn005955 247\nn005957 353\nn005958 335\nn005959 241\nn005960 266\nn005961 416\nn005962 493\nn005965 303\nn005966 477\nn005967 409\nn005968 237\nn005969 254\nn005970 316\nn005971 452\nn005972 371\nn005974 149\nn005975 292\nn005976 414\nn005977 461\nn005978 202\nn005979 250\nn005980 322\nn005981 246\nn005982 578\nn005983 409\nn005984 294\nn005985 534\nn005986 365\nn005987 377\nn005988 419\nn005989 391\nn005990 313\nn005991 275\nn005992 435\nn005993 354\nn005995 231\nn005996 337\nn005997 395\nn005998 296\nn005999 391\nn006000 297\nn006001 546\nn006002 430\nn006003 314\nn006004 370\nn006005 469\nn006006 439\nn006007 316\nn006008 296\nn006009 381\nn006010 340\nn006011 634\nn006012 499\nn006013 418\nn006014 158\nn006015 370\nn006016 480\nn006017 523\nn006018 318\nn006019 510\nn006020 463\nn006021 267\nn006022 436\nn006023 447\nn006024 367\nn006025 386\nn006026 374\nn006027 277\nn006028 355\nn006029 299\nn006030 124\nn006031 293\nn006032 338\nn006033 396\nn006034 429\nn006035 479\nn006036 275\nn006037 448\nn006038 432\nn006039 253\nn006040 350\nn006041 221\nn006042 259\nn006043 232\nn006044 230\nn006045 483\nn006047 386\nn006048 368\nn006049 421\nn006050 447\nn006051 313\nn006052 277\nn006054 257\nn006055 422\nn006056 257\nn006057 458\nn006058 353\nn006059 296\nn006060 347\nn006061 311\nn006062 296\nn006063 283\nn006064 213\nn006065 369\nn006066 261\nn006067 256\nn006068 406\nn006069 471\nn006070 336\nn006071 257\nn006072 454\nn006073 286\nn006074 366\nn006075 335\nn006076 243\nn006077 277\nn006078 347\nn006079 382\nn006080 211\nn006081 250\nn006082 207\nn006083 263\nn006084 343\nn006085 389\nn006086 392\nn006087 460\nn006088 454\nn006089 291\nn006090 189\nn006091 332\nn006092 120\nn006093 485\nn006094 454\nn006095 431\nn006096 236\nn006098 333\nn006099 295\nn006100 207\nn006101 332\nn006102 330\nn006103 399\nn006104 329\nn006105 124\nn006106 237\nn006107 277\nn006108 431\nn006109 190\nn006110 375\nn006111 343\nn006112 340\nn006113 319\nn006114 288\nn006115 378\nn006116 241\nn006117 340\nn006118 408\nn006119 318\nn006120 243\nn006121 452\nn006122 291\nn006124 221\nn006125 297\nn006126 433\nn006127 378\nn006128 141\nn006129 368\nn006130 490\nn006131 319\nn006132 253\nn006133 410\nn006135 391\nn006136 102\nn006137 473\nn006138 499\nn006139 515\nn006140 370\nn006141 494\nn006142 428\nn006143 254\nn006144 385\nn006145 372\nn006146 240\nn006147 357\nn006148 385\nn006149 231\nn006150 276\nn006151 276\nn006152 348\nn006153 525\nn006154 273\nn006155 401\nn006156 472\nn006157 507\nn006158 232\nn006159 261\nn006160 454\nn006161 336\nn006162 435\nn006163 350\nn006164 456\nn006165 424\nn006166 297\nn006167 365\nn006169 333\nn006170 214\nn006171 344\nn006172 428\nn006173 423\nn006174 376\nn006175 282\nn006176 159\nn006177 490\nn006178 311\nn006179 410\nn006180 326\nn006181 332\nn006182 318\nn006183 453\nn006184 385\nn006185 349\nn006186 330\nn006187 395\nn006188 319\nn006189 236\nn006190 307\nn006191 352\nn006192 330\nn006193 371\nn006194 443\nn006195 443\nn006196 198\nn006197 505\nn006198 278\nn006199 199\nn006200 253\nn006201 268\nn006202 416\nn006203 329\nn006204 232\nn006205 411\nn006206 280\nn006207 203\nn006208 243\nn006209 285\nn006210 294\nn006212 263\nn006213 210\nn006214 562\nn006215 150\nn006216 290\nn006217 330\nn006218 266\nn006219 296\nn006220 484\nn006221 544\nn006222 271\nn006223 339\nn006224 225\nn006225 187\nn006226 313\nn006227 400\nn006228 377\nn006229 376\nn006230 299\nn006231 367\nn006232 315\nn006233 366\nn006234 359\nn006235 336\nn006236 361\nn006237 231\nn006238 384\nn006239 547\nn006240 467\nn006241 347\nn006242 478\nn006243 317\nn006244 366\nn006245 358\nn006246 363\nn006248 471\nn006249 253\nn006250 238\nn006251 250\nn006252 258\nn006253 528\nn006254 385\nn006255 262\nn006256 338\nn006257 212\nn006258 361\nn006259 278\nn006260 298\nn006261 434\nn006262 324\nn006263 394\nn006264 378\nn006265 375\nn006266 536\nn006267 446\nn006268 329\nn006269 388\nn006270 319\nn006271 352\nn006272 199\nn006273 394\nn006274 534\nn006275 385\nn006276 351\nn006277 391\nn006278 318\nn006279 295\nn006280 367\nn006281 398\nn006282 363\nn006283 307\nn006284 327\nn006285 336\nn006286 244\nn006287 468\nn006288 412\nn006289 390\nn006290 329\nn006291 337\nn006292 351\nn006293 394\nn006294 404\nn006296 410\nn006297 466\nn006298 486\nn006300 519\nn006302 357\nn006303 432\nn006304 320\nn006305 264\nn006306 330\nn006307 327\nn006308 359\nn006309 213\nn006310 482\nn006311 220\nn006312 234\nn006313 307\nn006314 347\nn006315 303\nn006316 257\nn006317 357\nn006318 330\nn006319 382\nn006320 373\nn006321 268\nn006322 418\nn006323 198\nn006324 237\nn006325 473\nn006326 278\nn006327 285\nn006328 411\nn006329 377\nn006330 402\nn006331 326\nn006332 413\nn006333 457\nn006334 362\nn006335 359\nn006336 234\nn006337 251\nn006338 264\nn006339 387\nn006340 369\nn006341 486\nn006342 372\nn006343 392\nn006344 212\nn006345 481\nn006346 372\nn006348 288\nn006349 526\nn006350 465\nn006351 450\nn006352 545\nn006353 227\nn006354 428\nn006355 321\nn006356 342\nn006357 244\nn006358 262\nn006359 621\nn006360 315\nn006361 352\nn006362 335\nn006363 552\nn006364 270\nn006365 224\nn006366 426\nn006367 465\nn006368 349\nn006369 294\nn006370 265\nn006371 388\nn006372 319\nn006373 244\nn006374 301\nn006375 428\nn006376 387\nn006377 457\nn006378 187\nn006379 531\nn006380 334\nn006381 189\nn006382 447\nn006383 262\nn006384 421\nn006385 217\nn006386 422\nn006387 200\nn006388 260\nn006389 336\nn006390 189\nn006391 521\nn006392 472\nn006393 418\nn006394 552\nn006395 227\nn006396 405\nn006397 211\nn006398 344\nn006399 367\nn006400 607\nn006401 331\nn006402 304\nn006403 401\nn006404 217\nn006405 209\nn006406 412\nn006407 364\nn006408 379\nn006410 583\nn006411 332\nn006412 444\nn006413 380\nn006414 416\nn006415 298\nn006416 406\nn006417 255\nn006418 333\nn006420 517\nn006421 285\nn006422 302\nn006423 334\nn006424 490\nn006425 228\nn006426 513\nn006427 466\nn006428 341\nn006429 280\nn006430 214\nn006431 407\nn006432 372\nn006433 317\nn006434 280\nn006435 221\nn006436 302\nn006437 248\nn006438 226\nn006439 346\nn006440 349\nn006441 255\nn006442 570\nn006443 277\nn006444 310\nn006445 309\nn006446 349\nn006447 213\nn006448 465\nn006449 407\nn006450 440\nn006452 195\nn006453 298\nn006454 211\nn006455 417\nn006456 210\nn006457 334\nn006459 373\nn006460 311\nn006461 457\nn006462 190\nn006463 99\nn006464 419\nn006465 324\nn006466 287\nn006467 302\nn006468 321\nn006469 303\nn006470 354\nn006471 491\nn006473 437\nn006474 308\nn006475 236\nn006476 333\nn006477 315\nn006478 612\nn006479 484\nn006480 299\nn006481 509\nn006482 336\nn006483 365\nn006484 370\nn006485 317\nn006486 341\nn006487 234\nn006488 403\nn006489 304\nn006490 281\nn006491 242\nn006492 327\nn006493 340\nn006494 469\nn006495 288\nn006496 285\nn006498 491\nn006499 481\nn006500 322\nn006501 432\nn006502 323\nn006503 586\nn006504 208\nn006505 308\nn006506 318\nn006507 291\nn006508 315\nn006509 286\nn006510 459\nn006511 339\nn006512 319\nn006513 202\nn006514 318\nn006515 297\nn006516 320\nn006517 283\nn006518 279\nn006519 413\nn006520 239\nn006521 241\nn006522 343\nn006523 485\nn006524 201\nn006525 316\nn006526 387\nn006527 362\nn006528 250\nn006529 360\nn006530 267\nn006533 263\nn006534 323\nn006535 353\nn006536 253\nn006537 336\nn006538 217\nn006539 316\nn006540 348\nn006541 362\nn006542 241\nn006543 283\nn006544 265\nn006545 229\nn006546 327\nn006547 454\nn006548 380\nn006549 410\nn006550 232\nn006551 414\nn006552 238\nn006553 374\nn006554 288\nn006555 176\nn006556 527\nn006557 514\nn006558 454\nn006559 538\nn006560 375\nn006561 261\nn006562 358\nn006563 336\nn006564 476\nn006565 418\nn006566 440\nn006567 332\nn006568 365\nn006569 231\nn006570 482\nn006571 464\nn006573 501\nn006575 258\nn006576 455\nn006577 443\nn006578 387\nn006579 369\nn006580 264\nn006581 324\nn006582 245\nn006583 416\nn006584 245\nn006585 479\nn006587 224\nn006588 429\nn006589 306\nn006590 337\nn006591 210\nn006592 514\nn006593 278\nn006594 362\nn006595 479\nn006596 635\nn006597 318\nn006598 329\nn006599 416\nn006600 356\nn006601 328\nn006602 353\nn006603 288\nn006604 199\nn006605 246\nn006606 292\nn006607 288\nn006608 422\nn006609 153\nn006610 289\nn006611 382\nn006612 342\nn006613 477\nn006614 257\nn006615 386\nn006616 357\nn006617 308\nn006618 478\nn006619 467\nn006620 420\nn006621 286\nn006622 277\nn006623 508\nn006624 407\nn006625 408\nn006627 223\nn006628 246\nn006629 439\nn006630 241\nn006631 187\nn006632 293\nn006633 195\nn006634 270\nn006635 300\nn006636 394\nn006637 324\nn006638 314\nn006639 421\nn006640 404\nn006641 208\nn006642 293\nn006643 209\nn006644 492\nn006645 367\nn006646 351\nn006647 265\nn006648 417\nn006649 295\nn006650 317\nn006651 260\nn006652 317\nn006654 559\nn006655 383\nn006656 281\nn006657 220\nn006658 292\nn006660 543\nn006661 361\nn006662 368\nn006663 347\nn006665 397\nn006666 321\nn006667 306\nn006668 517\nn006669 285\nn006670 183\nn006671 514\nn006672 608\nn006673 357\nn006674 362\nn006675 397\nn006677 133\nn006678 460\nn006679 197\nn006680 159\nn006682 377\nn006683 535\nn006684 332\nn006685 354\nn006686 370\nn006687 273\nn006688 509\nn006689 367\nn006690 414\nn006691 478\nn006692 402\nn006693 366\nn006694 325\nn006695 539\nn006696 387\nn006697 261\nn006698 303\nn006699 314\nn006700 269\nn006701 346\nn006702 384\nn006703 268\nn006704 355\nn006705 498\nn006707 416\nn006708 385\nn006709 312\nn006710 528\nn006711 223\nn006712 398\nn006713 489\nn006714 352\nn006715 201\nn006716 504\nn006717 210\nn006718 321\nn006719 300\nn006720 328\nn006721 452\nn006722 471\nn006723 288\nn006724 402\nn006725 277\nn006726 308\nn006727 354\nn006728 333\nn006729 288\nn006730 203\nn006731 313\nn006732 328\nn006733 337\nn006734 259\nn006735 261\nn006736 398\nn006737 394\nn006738 305\nn006739 299\nn006740 410\nn006741 111\nn006742 239\nn006743 383\nn006744 266\nn006745 283\nn006746 309\nn006747 302\nn006748 474\nn006749 292\nn006750 200\nn006751 325\nn006752 363\nn006753 376\nn006754 246\nn006755 363\nn006756 307\nn006757 353\nn006758 252\nn006759 401\nn006760 248\nn006761 303\nn006762 356\nn006763 412\nn006764 441\nn006765 303\nn006766 435\nn006767 302\nn006768 287\nn006769 352\nn006770 344\nn006771 342\nn006773 326\nn006774 401\nn006775 328\nn006776 648\nn006777 345\nn006778 410\nn006779 486\nn006780 272\nn006781 333\nn006782 349\nn006783 355\nn006784 431\nn006785 309\nn006786 347\nn006787 298\nn006788 244\nn006789 358\nn006790 193\nn006791 381\nn006792 203\nn006793 431\nn006794 335\nn006795 348\nn006796 323\nn006797 287\nn006798 233\nn006799 285\nn006800 480\nn006801 375\nn006802 158\nn006803 257\nn006804 435\nn006805 357\nn006806 472\nn006807 244\nn006809 338\nn006810 377\nn006811 338\nn006812 454\nn006813 238\nn006814 486\nn006815 343\nn006816 369\nn006817 346\nn006818 350\nn006819 345\nn006820 397\nn006821 325\nn006822 377\nn006823 350\nn006824 527\nn006826 393\nn006827 408\nn006828 345\nn006829 366\nn006830 248\nn006831 418\nn006832 498\nn006833 383\nn006834 212\nn006835 347\nn006836 200\nn006837 230\nn006838 235\nn006839 426\nn006840 273\nn006841 266\nn006842 473\nn006843 367\nn006844 329\nn006845 452\nn006846 602\nn006847 355\nn006848 300\nn006849 306\nn006850 430\nn006851 309\nn006852 208\nn006853 300\nn006854 482\nn006855 340\nn006856 295\nn006857 250\nn006859 297\nn006860 606\nn006861 295\nn006862 374\nn006863 361\nn006864 272\nn006865 505\nn006866 222\nn006867 296\nn006868 406\nn006869 282\nn006870 339\nn006872 419\nn006873 334\nn006874 462\nn006875 280\nn006876 345\nn006877 401\nn006878 392\nn006879 299\nn006880 493\nn006881 261\nn006882 355\nn006883 349\nn006884 354\nn006885 344\nn006886 388\nn006887 247\nn006888 549\nn006889 311\nn006890 351\nn006891 350\nn006892 405\nn006893 216\nn006894 323\nn006895 455\nn006896 163\nn006897 430\nn006898 348\nn006899 480\nn006900 256\nn006901 328\nn006902 293\nn006903 255\nn006904 588\nn006905 220\nn006906 328\nn006907 410\nn006908 396\nn006910 414\nn006911 417\nn006912 464\nn006913 479\nn006914 371\nn006915 232\nn006916 237\nn006917 492\nn006918 396\nn006919 363\nn006920 501\nn006921 443\nn006923 223\nn006924 246\nn006925 380\nn006926 253\nn006927 205\nn006928 432\nn006929 292\nn006930 284\nn006931 369\nn006932 330\nn006933 497\nn006934 473\nn006935 315\nn006936 517\nn006937 456\nn006938 329\nn006939 335\nn006940 226\nn006941 421\nn006943 392\nn006944 361\nn006945 531\nn006946 292\nn006947 413\nn006948 365\nn006949 389\nn006950 265\nn006951 385\nn006952 297\nn006953 311\nn006954 209\nn006955 307\nn006956 564\nn006957 531\nn006958 432\nn006959 563\nn006960 435\nn006961 289\nn006962 364\nn006963 298\nn006965 381\nn006966 365\nn006967 405\nn006968 324\nn006969 442\nn006970 449\nn006971 353\nn006972 495\nn006973 197\nn006974 258\nn006975 213\nn006976 355\nn006977 206\nn006978 400\nn006979 507\nn006980 288\nn006981 268\nn006982 405\nn006983 336\nn006984 407\nn006985 289\nn006986 311\nn006988 304\nn006989 261\nn006990 291\nn006991 398\nn006993 297\nn006994 444\nn006995 403\nn006996 324\nn006997 515\nn006998 322\nn006999 236\nn007000 460\nn007001 529\nn007002 529\nn007003 421\nn007004 263\nn007005 188\nn007006 378\nn007007 335\nn007009 563\nn007010 352\nn007011 454\nn007012 369\nn007013 446\nn007015 218\nn007016 450\nn007017 314\nn007018 266\nn007019 298\nn007020 339\nn007022 323\nn007023 453\nn007024 284\nn007025 482\nn007026 322\nn007027 537\nn007028 273\nn007029 375\nn007030 258\nn007031 499\nn007032 229\nn007033 175\nn007034 590\nn007035 380\nn007036 330\nn007037 612\nn007038 282\nn007039 292\nn007040 261\nn007041 272\nn007042 388\nn007043 419\nn007044 511\nn007045 296\nn007046 350\nn007047 385\nn007048 301\nn007049 380\nn007050 478\nn007051 469\nn007052 409\nn007053 295\nn007054 340\nn007055 329\nn007056 295\nn007057 475\nn007058 157\nn007059 435\nn007060 382\nn007061 268\nn007062 367\nn007063 287\nn007064 437\nn007065 423\nn007066 291\nn007067 285\nn007069 412\nn007070 263\nn007071 413\nn007072 393\nn007073 419\nn007074 451\nn007075 451\nn007076 486\nn007077 443\nn007078 294\nn007079 574\nn007080 487\nn007081 495\nn007082 352\nn007083 406\nn007084 398\nn007085 455\nn007086 343\nn007087 323\nn007088 270\nn007089 217\nn007090 427\nn007091 486\nn007092 281\nn007093 469\nn007094 271\nn007095 413\nn007096 230\nn007097 233\nn007098 520\nn007099 158\nn007100 471\nn007101 377\nn007102 419\nn007103 335\nn007105 339\nn007106 536\nn007107 349\nn007108 381\nn007109 441\nn007110 417\nn007111 288\nn007112 346\nn007113 213\nn007114 441\nn007115 479\nn007116 479\nn007117 223\nn007118 419\nn007119 263\nn007120 671\nn007122 461\nn007123 445\nn007124 344\nn007125 279\nn007126 316\nn007127 320\nn007128 289\nn007129 586\nn007130 280\nn007131 204\nn007132 545\nn007134 400\nn007135 379\nn007136 378\nn007137 380\nn007138 528\nn007139 656\nn007140 368\nn007141 126\nn007142 220\nn007143 413\nn007144 557\nn007145 219\nn007147 208\nn007148 325\nn007149 228\nn007150 363\nn007151 389\nn007152 292\nn007153 478\nn007154 260\nn007155 390\nn007156 201\nn007157 256\nn007158 282\nn007160 492\nn007161 396\nn007163 485\nn007164 207\nn007165 485\nn007167 417\nn007168 396\nn007170 451\nn007171 182\nn007172 351\nn007173 559\nn007174 500\nn007175 255\nn007176 371\nn007177 556\nn007178 302\nn007179 290\nn007180 410\nn007181 340\nn007182 232\nn007184 313\nn007185 478\nn007186 367\nn007187 369\nn007188 233\nn007189 296\nn007190 133\nn007191 300\nn007192 261\nn007193 376\nn007194 431\nn007195 473\nn007196 429\nn007197 253\nn007198 397\nn007199 465\nn007200 277\nn007201 298\nn007202 307\nn007203 502\nn007204 179\nn007205 412\nn007206 400\nn007207 439\nn007208 339\nn007209 576\nn007211 304\nn007212 305\nn007213 370\nn007214 431\nn007215 273\nn007216 557\nn007217 388\nn007218 380\nn007219 299\nn007220 288\nn007221 233\nn007222 420\nn007223 404\nn007224 354\nn007225 495\nn007226 473\nn007227 379\nn007228 286\nn007229 604\nn007230 403\nn007231 349\nn007232 437\nn007233 278\nn007234 280\nn007235 607\nn007236 392\nn007237 577\nn007238 227\nn007239 441\nn007242 443\nn007243 340\nn007244 313\nn007245 392\nn007247 361\nn007248 473\nn007249 341\nn007250 494\nn007251 275\nn007252 558\nn007253 376\nn007254 410\nn007255 458\nn007256 259\nn007257 448\nn007258 417\nn007259 282\nn007262 423\nn007263 368\nn007264 425\nn007265 473\nn007266 487\nn007267 510\nn007268 266\nn007269 383\nn007270 446\nn007271 303\nn007272 543\nn007273 320\nn007274 479\nn007275 438\nn007276 426\nn007277 384\nn007278 187\nn007279 404\nn007280 563\nn007281 418\nn007282 463\nn007283 473\nn007284 227\nn007285 402\nn007286 409\nn007287 430\nn007288 230\nn007289 351\nn007290 423\nn007291 440\nn007292 311\nn007293 309\nn007294 335\nn007295 446\nn007297 257\nn007298 504\nn007299 257\nn007300 237\nn007301 350\nn007302 448\nn007303 398\nn007304 473\nn007305 380\nn007306 489\nn007307 449\nn007308 252\nn007309 421\nn007310 296\nn007311 336\nn007312 268\nn007313 505\nn007314 325\nn007315 432\nn007316 378\nn007317 359\nn007318 186\nn007319 376\nn007320 235\nn007321 357\nn007322 430\nn007323 332\nn007324 356\nn007325 453\nn007326 351\nn007327 564\nn007328 326\nn007329 225\nn007330 328\nn007331 599\nn007332 369\nn007333 556\nn007334 267\nn007335 226\nn007336 457\nn007337 540\nn007338 166\nn007339 145\nn007340 267\nn007341 434\nn007342 467\nn007343 391\nn007344 328\nn007345 333\nn007346 292\nn007347 225\nn007348 554\nn007349 314\nn007350 516\nn007351 232\nn007352 363\nn007353 519\nn007354 335\nn007355 340\nn007356 505\nn007357 421\nn007359 409\nn007360 294\nn007361 415\nn007362 452\nn007363 217\nn007364 169\nn007365 280\nn007366 441\nn007367 469\nn007369 214\nn007370 186\nn007371 217\nn007372 275\nn007373 221\nn007374 220\nn007375 387\nn007376 406\nn007377 335\nn007378 291\nn007379 259\nn007380 304\nn007382 418\nn007383 260\nn007384 354\nn007386 317\nn007387 308\nn007388 472\nn007389 514\nn007391 407\nn007392 504\nn007393 480\nn007394 348\nn007395 453\nn007396 340\nn007398 498\nn007399 344\nn007400 243\nn007401 224\nn007402 396\nn007403 260\nn007404 206\nn007405 355\nn007406 254\nn007407 248\nn007408 301\nn007409 292\nn007410 489\nn007411 257\nn007412 298\nn007413 383\nn007414 378\nn007415 326\nn007416 269\nn007417 361\nn007418 335\nn007419 318\nn007420 421\nn007421 441\nn007422 329\nn007423 332\nn007425 432\nn007426 423\nn007427 364\nn007428 492\nn007429 209\nn007430 233\nn007431 268\nn007432 539\nn007433 409\nn007434 226\nn007435 211\nn007436 355\nn007437 433\nn007438 377\nn007440 411\nn007441 219\nn007442 298\nn007443 319\nn007444 316\nn007445 536\nn007446 410\nn007447 365\nn007448 347\nn007449 297\nn007450 384\nn007451 413\nn007452 349\nn007453 288\nn007454 452\nn007455 333\nn007456 369\nn007457 344\nn007458 448\nn007459 212\nn007460 380\nn007461 371\nn007462 274\nn007463 230\nn007464 491\nn007465 528\nn007466 500\nn007467 326\nn007468 386\nn007469 359\nn007470 410\nn007471 292\nn007472 311\nn007473 466\nn007475 312\nn007476 487\nn007477 314\nn007478 222\nn007479 247\nn007480 429\nn007481 487\nn007482 442\nn007483 352\nn007484 480\nn007485 290\nn007486 265\nn007487 224\nn007488 363\nn007489 374\nn007490 322\nn007491 393\nn007492 211\nn007493 551\nn007494 388\nn007495 198\nn007496 251\nn007497 216\nn007498 272\nn007499 319\nn007501 352\nn007502 387\nn007503 206\nn007504 324\nn007505 109\nn007506 425\nn007507 329\nn007508 505\nn007509 302\nn007510 398\nn007511 417\nn007512 371\nn007513 213\nn007514 368\nn007515 244\nn007516 343\nn007517 425\nn007518 269\nn007519 451\nn007520 346\nn007521 275\nn007522 254\nn007523 440\nn007524 308\nn007525 465\nn007526 201\nn007527 354\nn007528 374\nn007529 275\nn007530 414\nn007532 215\nn007533 268\nn007534 362\nn007535 343\nn007536 385\nn007537 436\nn007538 281\nn007539 516\nn007540 572\nn007542 323\nn007543 352\nn007544 482\nn007545 418\nn007546 575\nn007547 269\nn007549 402\nn007550 419\nn007551 399\nn007552 359\nn007553 446\nn007554 272\nn007555 277\nn007557 385\nn007558 383\nn007559 564\nn007560 588\nn007561 284\nn007562 324\nn007563 357\nn007564 597\nn007565 392\nn007566 461\nn007567 488\nn007568 340\nn007569 373\nn007570 253\nn007572 324\nn007573 419\nn007574 255\nn007575 593\nn007576 474\nn007577 450\nn007578 305\nn007579 435\nn007580 234\nn007581 366\nn007582 381\nn007583 296\nn007584 383\nn007585 531\nn007586 347\nn007587 354\nn007588 245\nn007589 276\nn007590 636\nn007591 498\nn007592 525\nn007593 350\nn007595 368\nn007596 375\nn007597 311\nn007598 407\nn007599 334\nn007600 384\nn007601 288\nn007603 208\nn007604 400\nn007605 279\nn007606 313\nn007607 341\nn007608 410\nn007609 319\nn007610 502\nn007611 350\nn007612 491\nn007613 197\nn007614 265\nn007615 401\nn007616 566\nn007617 445\nn007618 366\nn007619 638\nn007620 540\nn007621 303\nn007622 243\nn007623 454\nn007624 308\nn007625 222\nn007626 332\nn007627 464\nn007628 293\nn007629 530\nn007630 522\nn007632 354\nn007633 334\nn007634 344\nn007635 298\nn007636 368\nn007637 207\nn007638 218\nn007639 320\nn007640 422\nn007641 323\nn007642 504\nn007643 307\nn007644 283\nn007645 405\nn007646 295\nn007647 525\nn007648 278\nn007649 318\nn007652 224\nn007653 442\nn007654 199\nn007655 345\nn007656 270\nn007657 370\nn007658 589\nn007659 368\nn007660 371\nn007661 487\nn007662 272\nn007663 407\nn007665 412\nn007666 322\nn007667 199\nn007669 439\nn007670 438\nn007671 167\nn007672 250\nn007674 292\nn007675 371\nn007676 306\nn007677 493\nn007678 148\nn007679 422\nn007680 374\nn007681 325\nn007682 351\nn007683 407\nn007684 482\nn007685 397\nn007686 204\nn007687 281\nn007688 506\nn007689 280\nn007690 271\nn007691 402\nn007692 332\nn007693 262\nn007694 337\nn007695 395\nn007696 336\nn007697 208\nn007698 408\nn007699 382\nn007701 345\nn007702 480\nn007704 356\nn007705 412\nn007706 396\nn007707 311\nn007708 210\nn007710 222\nn007711 404\nn007712 357\nn007713 410\nn007714 338\nn007715 372\nn007716 462\nn007717 320\nn007718 228\nn007719 379\nn007720 379\nn007721 254\nn007722 480\nn007723 419\nn007724 321\nn007725 387\nn007726 386\nn007727 299\nn007728 179\nn007729 280\nn007730 205\nn007731 295\nn007732 349\nn007733 115\nn007734 427\nn007735 426\nn007736 494\nn007737 452\nn007738 420\nn007739 235\nn007740 421\nn007741 358\nn007742 250\nn007743 487\nn007744 379\nn007745 259\nn007746 306\nn007747 365\nn007748 370\nn007749 342\nn007750 344\nn007751 271\nn007752 488\nn007754 434\nn007755 414\nn007756 310\nn007757 430\nn007758 310\nn007759 272\nn007760 408\nn007761 228\nn007762 314\nn007763 497\nn007764 247\nn007765 479\nn007767 419\nn007768 463\nn007769 324\nn007770 395\nn007771 292\nn007772 429\nn007773 313\nn007774 286\nn007775 459\nn007776 336\nn007777 513\nn007778 367\nn007779 441\nn007780 352\nn007781 338\nn007782 124\nn007783 284\nn007784 399\nn007785 605\nn007786 349\nn007787 638\nn007788 331\nn007789 308\nn007790 429\nn007791 543\nn007792 390\nn007793 228\nn007794 352\nn007795 522\nn007796 184\nn007797 390\nn007798 206\nn007799 281\nn007800 300\nn007801 475\nn007802 244\nn007803 459\nn007804 383\nn007805 366\nn007806 353\nn007807 356\nn007808 272\nn007809 315\nn007810 246\nn007811 259\nn007812 233\nn007813 448\nn007814 452\nn007815 427\nn007816 283\nn007817 290\nn007818 346\nn007819 356\nn007821 333\nn007822 315\nn007823 228\nn007824 348\nn007825 304\nn007826 458\nn007827 454\nn007828 441\nn007830 419\nn007831 299\nn007832 369\nn007833 207\nn007834 244\nn007835 195\nn007836 468\nn007837 354\nn007838 209\nn007839 225\nn007840 350\nn007841 366\nn007842 291\nn007843 394\nn007844 348\nn007845 236\nn007846 355\nn007847 371\nn007848 470\nn007849 490\nn007850 368\nn007851 364\nn007852 208\nn007853 384\nn007855 417\nn007856 393\nn007857 297\nn007858 377\nn007859 397\nn007860 414\nn007861 408\nn007862 188\nn007863 431\nn007864 115\nn007866 211\nn007867 314\nn007868 249\nn007869 458\nn007870 445\nn007871 288\nn007873 460\nn007874 374\nn007875 178\nn007876 117\nn007877 458\nn007878 392\nn007879 457\nn007880 342\nn007881 200\nn007882 262\nn007883 504\nn007884 233\nn007885 323\nn007886 258\nn007887 582\nn007888 487\nn007889 291\nn007890 297\nn007891 435\nn007892 321\nn007893 249\nn007894 483\nn007895 262\nn007896 332\nn007897 341\nn007899 281\nn007901 465\nn007902 300\nn007904 184\nn007906 456\nn007907 351\nn007908 297\nn007910 249\nn007911 272\nn007912 148\nn007913 331\nn007915 433\nn007916 344\nn007917 288\nn007918 336\nn007920 389\nn007921 324\nn007922 305\nn007923 407\nn007924 329\nn007925 482\nn007926 483\nn007927 355\nn007928 477\nn007929 421\nn007930 356\nn007931 336\nn007932 475\nn007933 410\nn007934 281\nn007935 286\nn007936 377\nn007937 259\nn007938 564\nn007939 151\nn007940 336\nn007941 287\nn007942 263\nn007943 305\nn007944 324\nn007945 223\nn007946 556\nn007947 306\nn007948 407\nn007950 543\nn007952 420\nn007953 431\nn007954 297\nn007955 434\nn007956 333\nn007957 363\nn007958 275\nn007959 335\nn007960 592\nn007961 426\nn007962 379\nn007963 296\nn007966 286\nn007967 302\nn007968 518\nn007969 564\nn007970 433\nn007971 546\nn007972 534\nn007973 258\nn007974 480\nn007975 572\nn007976 369\nn007977 349\nn007978 417\nn007979 522\nn007980 518\nn007981 598\nn007982 256\nn007983 484\nn007984 334\nn007985 317\nn007986 348\nn007987 367\nn007988 296\nn007989 318\nn007990 465\nn007991 346\nn007992 400\nn007993 599\nn007994 381\nn007995 390\nn007996 308\nn007997 273\nn007998 454\nn007999 503\nn008000 339\nn008001 479\nn008002 289\nn008004 392\nn008005 365\nn008006 414\nn008007 213\nn008008 367\nn008009 238\nn008010 445\nn008011 489\nn008012 357\nn008013 605\nn008014 377\nn008016 285\nn008017 357\nn008018 200\nn008019 327\nn008021 358\nn008022 242\nn008023 502\nn008024 416\nn008025 392\nn008026 373\nn008027 375\nn008030 249\nn008032 301\nn008033 481\nn008034 265\nn008038 444\nn008039 407\nn008040 373\nn008041 479\nn008042 477\nn008043 328\nn008044 249\nn008045 383\nn008046 476\nn008048 529\nn008049 457\nn008050 319\nn008051 378\nn008052 440\nn008053 432\nn008054 340\nn008055 321\nn008057 290\nn008058 293\nn008059 672\nn008060 491\nn008061 307\nn008062 316\nn008063 253\nn008064 580\nn008065 493\nn008066 360\nn008067 497\nn008068 200\nn008069 271\nn008070 316\nn008071 391\nn008072 395\nn008073 534\nn008074 187\nn008075 358\nn008076 237\nn008077 310\nn008078 499\nn008079 502\nn008080 432\nn008081 116\nn008082 427\nn008083 415\nn008084 348\nn008085 380\nn008086 358\nn008087 298\nn008088 298\nn008089 268\nn008090 301\nn008091 419\nn008092 536\nn008093 325\nn008094 339\nn008095 332\nn008096 477\nn008097 461\nn008098 373\nn008099 390\nn008100 370\nn008101 325\nn008102 263\nn008103 427\nn008104 295\nn008107 476\nn008109 528\nn008111 282\nn008112 409\nn008113 139\nn008114 421\nn008115 288\nn008116 332\nn008117 237\nn008118 238\nn008119 339\nn008120 135\nn008121 403\nn008122 426\nn008123 357\nn008124 368\nn008125 390\nn008126 346\nn008127 444\nn008128 388\nn008129 296\nn008130 292\nn008131 245\nn008132 232\nn008133 409\nn008134 334\nn008135 274\nn008136 434\nn008137 343\nn008138 309\nn008139 425\nn008140 228\nn008141 367\nn008142 487\nn008143 201\nn008144 468\nn008145 320\nn008146 308\nn008147 330\nn008148 372\nn008149 151\nn008150 252\nn008151 360\nn008152 328\nn008153 424\nn008154 393\nn008156 496\nn008157 440\nn008158 467\nn008159 528\nn008160 372\nn008161 385\nn008162 252\nn008163 535\nn008165 670\nn008166 494\nn008168 472\nn008169 253\nn008170 502\nn008171 453\nn008172 427\nn008173 399\nn008174 438\nn008175 454\nn008176 304\nn008177 370\nn008178 166\nn008180 439\nn008181 429\nn008182 308\nn008184 302\nn008185 289\nn008186 587\nn008187 477\nn008188 306\nn008189 291\nn008190 379\nn008191 371\nn008192 498\nn008193 192\nn008194 398\nn008196 282\nn008197 407\nn008198 359\nn008199 261\nn008200 296\nn008201 478\nn008202 410\nn008203 347\nn008204 271\nn008205 458\nn008206 297\nn008207 550\nn008208 397\nn008209 318\nn008210 326\nn008211 321\nn008212 374\nn008214 501\nn008215 400\nn008216 430\nn008217 374\nn008218 412\nn008219 528\nn008220 323\nn008221 356\nn008223 637\nn008224 397\nn008225 498\nn008226 341\nn008227 285\nn008228 316\nn008229 252\nn008230 332\nn008231 247\nn008232 196\nn008233 467\nn008234 442\nn008235 307\nn008236 288\nn008237 347\nn008238 439\nn008239 421\nn008240 478\nn008241 287\nn008242 558\nn008243 381\nn008244 315\nn008245 292\nn008246 230\nn008247 396\nn008248 473\nn008249 390\nn008250 317\nn008252 185\nn008253 306\nn008254 204\nn008255 216\nn008256 252\nn008257 402\nn008258 331\nn008259 321\nn008260 299\nn008261 375\nn008262 440\nn008263 536\nn008265 234\nn008266 444\nn008267 359\nn008268 430\nn008270 387\nn008272 386\nn008273 248\nn008274 450\nn008275 383\nn008276 526\nn008277 372\nn008278 204\nn008279 383\nn008280 405\nn008281 397\nn008282 495\nn008283 392\nn008284 409\nn008285 323\nn008286 322\nn008287 449\nn008288 485\nn008289 375\nn008290 453\nn008291 269\nn008292 285\nn008293 440\nn008294 327\nn008295 226\nn008296 508\nn008297 373\nn008298 304\nn008299 286\nn008300 121\nn008301 346\nn008302 340\nn008303 365\nn008304 287\nn008305 344\nn008306 364\nn008307 319\nn008308 422\nn008309 497\nn008310 389\nn008311 446\nn008312 459\nn008313 460\nn008316 523\nn008318 303\nn008319 383\nn008320 319\nn008321 416\nn008322 308\nn008323 348\nn008324 433\nn008326 388\nn008327 321\nn008328 744\nn008329 365\nn008330 356\nn008332 433\nn008333 503\nn008334 428\nn008335 312\nn008336 321\nn008337 246\nn008338 401\nn008339 413\nn008340 338\nn008341 218\nn008342 436\nn008343 261\nn008344 139\nn008345 237\nn008346 299\nn008347 347\nn008348 369\nn008349 356\nn008350 414\nn008351 211\nn008352 377\nn008353 382\nn008354 320\nn008355 274\nn008356 346\nn008357 447\nn008358 450\nn008359 314\nn008360 289\nn008362 268\nn008363 375\nn008364 359\nn008365 381\nn008366 350\nn008367 363\nn008368 234\nn008369 500\nn008370 360\nn008371 334\nn008372 225\nn008373 428\nn008374 448\nn008375 270\nn008376 465\nn008377 251\nn008378 355\nn008379 425\nn008380 359\nn008381 217\nn008382 492\nn008383 348\nn008384 518\nn008386 437\nn008387 505\nn008388 659\nn008389 442\nn008390 239\nn008391 451\nn008392 238\nn008393 222\nn008394 228\nn008395 398\nn008396 244\nn008397 333\nn008398 387\nn008399 357\nn008400 473\nn008401 349\nn008402 398\nn008403 214\nn008404 283\nn008405 485\nn008406 253\nn008407 252\nn008408 385\nn008409 214\nn008410 465\nn008411 307\nn008412 510\nn008413 458\nn008414 351\nn008415 350\nn008416 483\nn008417 373\nn008418 287\nn008419 456\nn008420 471\nn008421 364\nn008422 276\nn008423 511\nn008424 255\nn008425 292\nn008427 423\nn008428 322\nn008429 351\nn008430 306\nn008431 437\nn008432 240\nn008433 286\nn008434 299\nn008437 326\nn008438 251\nn008439 405\nn008440 569\nn008441 463\nn008442 301\nn008443 438\nn008444 206\nn008445 208\nn008446 201\nn008447 310\nn008448 436\nn008449 319\nn008450 249\nn008451 243\nn008452 267\nn008453 235\nn008454 183\nn008455 525\nn008456 516\nn008457 391\nn008458 430\nn008459 485\nn008460 373\nn008461 348\nn008462 436\nn008463 436\nn008464 504\nn008465 327\nn008466 312\nn008467 450\nn008468 254\nn008469 106\nn008470 243\nn008471 344\nn008472 262\nn008473 438\nn008474 214\nn008475 271\nn008476 501\nn008477 314\nn008479 465\nn008480 281\nn008481 378\nn008482 226\nn008483 353\nn008485 232\nn008487 448\nn008489 328\nn008490 404\nn008491 390\nn008492 236\nn008493 314\nn008494 308\nn008495 253\nn008496 340\nn008497 317\nn008498 442\nn008499 372\nn008500 455\nn008501 309\nn008502 641\nn008503 233\nn008504 360\nn008505 370\nn008506 373\nn008507 387\nn008508 454\nn008509 311\nn008510 379\nn008511 503\nn008512 395\nn008513 240\nn008514 260\nn008515 431\nn008516 213\nn008517 525\nn008519 425\nn008520 608\nn008521 220\nn008522 318\nn008523 304\nn008524 504\nn008525 242\nn008526 350\nn008527 295\nn008529 522\nn008531 319\nn008532 378\nn008533 455\nn008534 391\nn008535 485\nn008536 306\nn008537 480\nn008538 304\nn008540 197\nn008541 278\nn008542 321\nn008543 528\nn008544 333\nn008545 559\nn008546 478\nn008547 373\nn008548 432\nn008549 335\nn008550 379\nn008551 373\nn008552 316\nn008553 313\nn008554 189\nn008555 425\nn008556 446\nn008557 384\nn008560 343\nn008561 409\nn008562 462\nn008563 267\nn008564 324\nn008565 365\nn008566 427\nn008568 670\nn008569 259\nn008570 331\nn008571 374\nn008572 490\nn008573 456\nn008574 326\nn008575 304\nn008576 381\nn008577 133\nn008578 355\nn008579 322\nn008580 363\nn008581 366\nn008582 298\nn008583 342\nn008584 517\nn008585 286\nn008586 586\nn008587 254\nn008588 390\nn008589 382\nn008590 265\nn008591 271\nn008592 360\nn008593 260\nn008594 422\nn008596 299\nn008597 373\nn008598 420\nn008599 455\nn008600 522\nn008601 433\nn008602 257\nn008603 551\nn008604 323\nn008605 432\nn008606 357\nn008607 486\nn008608 349\nn008609 276\nn008610 298\nn008611 331\nn008612 435\nn008614 345\nn008616 304\nn008617 521\nn008618 213\nn008619 517\nn008620 302\nn008621 364\nn008622 543\nn008623 525\nn008624 373\nn008625 309\nn008626 296\nn008627 297\nn008628 208\nn008631 399\nn008632 383\nn008633 499\nn008634 514\nn008635 200\nn008636 429\nn008637 249\nn008638 519\nn008639 478\nn008640 557\nn008641 466\nn008642 345\nn008643 278\nn008644 342\nn008645 283\nn008646 458\nn008647 453\nn008648 458\nn008649 262\nn008650 343\nn008651 362\nn008652 405\nn008654 431\nn008656 265\nn008657 463\nn008658 500\nn008659 306\nn008660 234\nn008661 342\nn008663 322\nn008664 309\nn008665 353\nn008666 454\nn008667 360\nn008668 375\nn008669 461\nn008670 344\nn008671 289\nn008672 340\nn008673 471\nn008675 401\nn008676 356\nn008677 356\nn008678 216\nn008679 355\nn008680 199\nn008681 332\nn008683 506\nn008684 188\nn008685 152\nn008686 202\nn008687 339\nn008688 328\nn008689 216\nn008690 293\nn008691 201\nn008692 307\nn008693 365\nn008694 347\nn008695 522\nn008696 259\nn008697 374\nn008698 440\nn008699 552\nn008700 386\nn008701 471\nn008702 486\nn008703 323\nn008704 351\nn008705 311\nn008706 224\nn008707 406\nn008708 224\nn008709 219\nn008711 210\nn008712 243\nn008713 339\nn008714 340\nn008715 551\nn008716 370\nn008718 435\nn008719 278\nn008720 289\nn008721 319\nn008722 431\nn008723 369\nn008724 333\nn008725 274\nn008726 338\nn008727 519\nn008728 149\nn008729 177\nn008730 202\nn008731 455\nn008732 403\nn008733 235\nn008734 362\nn008735 254\nn008736 429\nn008737 236\nn008738 379\nn008739 342\nn008740 327\nn008741 305\nn008742 288\nn008743 404\nn008744 259\nn008745 408\nn008746 299\nn008747 290\nn008748 433\nn008749 327\nn008750 493\nn008751 473\nn008752 407\nn008753 463\nn008754 300\nn008755 306\nn008756 309\nn008757 460\nn008758 379\nn008759 347\nn008760 378\nn008761 552\nn008762 207\nn008763 160\nn008764 265\nn008765 406\nn008766 243\nn008767 480\nn008768 335\nn008769 303\nn008770 385\nn008771 365\nn008772 384\nn008774 472\nn008775 217\nn008776 307\nn008779 267\nn008780 258\nn008781 392\nn008782 292\nn008783 367\nn008784 106\nn008785 321\nn008786 421\nn008787 243\nn008788 488\nn008789 408\nn008790 286\nn008791 419\nn008792 269\nn008793 264\nn008794 355\nn008795 462\nn008796 303\nn008797 312\nn008798 375\nn008799 366\nn008800 419\nn008801 539\nn008802 367\nn008803 315\nn008804 338\nn008805 264\nn008806 327\nn008807 394\nn008808 386\nn008809 281\nn008810 420\nn008811 334\nn008812 307\nn008813 445\nn008814 467\nn008815 299\nn008816 230\nn008817 372\nn008818 320\nn008819 417\nn008820 311\nn008821 297\nn008822 457\nn008823 209\nn008824 289\nn008825 422\nn008826 387\nn008827 210\nn008830 340\nn008831 355\nn008832 437\nn008833 475\nn008834 335\nn008835 225\nn008836 351\nn008837 227\nn008838 418\nn008839 516\nn008840 496\nn008841 333\nn008842 271\nn008843 286\nn008844 491\nn008845 279\nn008846 226\nn008847 203\nn008848 255\nn008849 451\nn008850 341\nn008851 351\nn008852 510\nn008853 201\nn008854 338\nn008855 338\nn008856 280\nn008857 263\nn008859 430\nn008860 299\nn008861 235\nn008862 317\nn008863 154\nn008864 354\nn008865 452\nn008866 287\nn008867 428\nn008868 371\nn008869 400\nn008870 397\nn008871 282\nn008872 497\nn008873 273\nn008874 399\nn008875 321\nn008877 342\nn008878 294\nn008879 504\nn008880 317\nn008881 335\nn008882 477\nn008883 368\nn008884 486\nn008885 481\nn008886 228\nn008887 248\nn008891 369\nn008892 339\nn008893 442\nn008894 308\nn008895 308\nn008896 301\nn008897 219\nn008898 270\nn008899 282\nn008900 411\nn008901 506\nn008902 330\nn008903 479\nn008904 524\nn008905 195\nn008906 310\nn008907 331\nn008908 223\nn008909 437\nn008910 294\nn008911 473\nn008912 257\nn008913 176\nn008914 354\nn008915 508\nn008916 202\nn008917 601\nn008918 330\nn008919 438\nn008920 257\nn008921 246\nn008922 214\nn008923 440\nn008924 371\nn008925 386\nn008926 384\nn008927 583\nn008928 369\nn008929 375\nn008930 511\nn008931 349\nn008933 526\nn008934 285\nn008935 211\nn008936 472\nn008938 447\nn008939 503\nn008940 366\nn008941 397\nn008942 483\nn008943 284\nn008944 441\nn008945 268\nn008946 320\nn008947 305\nn008949 266\nn008950 408\nn008951 389\nn008952 373\nn008953 482\nn008954 290\nn008955 452\nn008956 364\nn008957 344\nn008959 421\nn008960 277\nn008961 369\nn008962 324\nn008963 220\nn008964 316\nn008965 312\nn008966 330\nn008967 371\nn008968 206\nn008969 372\nn008970 337\nn008971 319\nn008972 536\nn008973 156\nn008974 483\nn008975 277\nn008976 350\nn008977 359\nn008978 247\nn008979 298\nn008980 469\nn008981 331\nn008982 281\nn008983 476\nn008984 318\nn008985 245\nn008986 293\nn008987 439\nn008988 247\nn008990 218\nn008991 408\nn008992 361\nn008993 294\nn008994 255\nn008995 231\nn008996 331\nn008997 159\nn008998 497\nn008999 428\nn009000 261\nn009001 378\nn009002 407\nn009003 265\nn009004 213\nn009005 413\nn009006 311\nn009007 293\nn009008 373\nn009009 432\nn009010 401\nn009011 242\nn009012 312\nn009013 211\nn009015 196\nn009016 286\nn009017 369\nn009018 378\nn009019 214\nn009020 266\nn009021 280\nn009022 397\nn009023 193\nn009024 243\nn009025 333\nn009026 361\nn009027 307\nn009029 284\nn009030 477\nn009031 385\nn009032 387\nn009033 469\nn009034 252\nn009035 549\nn009036 551\nn009037 486\nn009038 232\nn009039 442\nn009040 266\nn009041 157\nn009042 272\nn009043 478\nn009044 324\nn009045 310\nn009046 549\nn009047 499\nn009048 498\nn009049 516\nn009050 328\nn009051 377\nn009052 524\nn009053 459\nn009054 263\nn009055 307\nn009056 417\nn009057 382\nn009058 171\nn009059 426\nn009060 321\nn009061 360\nn009062 302\nn009063 402\nn009064 263\nn009065 286\nn009066 363\nn009067 203\nn009068 507\nn009069 427\nn009070 392\nn009071 427\nn009072 220\nn009073 235\nn009074 387\nn009075 529\nn009076 274\nn009077 312\nn009078 403\nn009079 359\nn009080 201\nn009081 320\nn009082 375\nn009083 283\nn009084 467\nn009085 177\nn009086 272\nn009087 234\nn009088 293\nn009089 253\nn009091 411\nn009092 312\nn009093 240\nn009094 382\nn009095 251\nn009096 335\nn009097 388\nn009098 387\nn009099 439\nn009100 395\nn009101 238\nn009102 294\nn009103 536\nn009104 427\nn009105 324\nn009106 368\nn009107 136\nn009108 358\nn009109 445\nn009110 300\nn009111 261\nn009112 244\nn009113 286\nn009115 417\nn009116 249\nn009117 449\nn009118 159\nn009119 233\nn009120 294\nn009121 472\nn009122 450\nn009124 373\nn009125 369\nn009126 338\nn009127 389\nn009129 149\nn009130 336\nn009131 415\nn009132 373\nn009133 372\nn009134 474\nn009135 369\nn009136 558\nn009137 520\nn009138 375\nn009139 407\nn009140 288\nn009141 304\nn009142 236\nn009143 449\nn009144 492\nn009145 487\nn009146 185\nn009147 539\nn009148 117\nn009149 323\nn009150 406\nn009151 365\nn009152 374\nn009153 272\nn009154 328\nn009155 278\nn009156 407\nn009157 346\nn009158 400\nn009159 455\nn009160 362\nn009161 510\nn009162 337\nn009163 328\nn009164 507\nn009165 463\nn009166 496\nn009167 306\nn009168 254\nn009169 350\nn009170 372\nn009171 608\nn009172 283\nn009173 287\nn009174 388\nn009176 152\nn009177 377\nn009179 250\nn009180 383\nn009181 252\nn009182 150\nn009183 375\nn009184 426\nn009186 318\nn009187 281\nn009188 485\nn009189 257\nn009190 233\nn009191 282\nn009192 297\nn009193 363\nn009194 500\nn009195 270\nn009196 354\nn009197 433\nn009198 489\nn009200 364\nn009201 464\nn009202 278\nn009203 323\nn009204 354\nn009205 390\nn009207 306\nn009208 366\nn009209 265\nn009210 267\nn009211 230\nn009212 387\nn009214 292\nn009215 364\nn009216 478\nn009217 364\nn009218 439\nn009219 384\nn009220 487\nn009221 435\nn009222 444\nn009223 438\nn009224 267\nn009226 365\nn009227 509\nn009228 370\nn009229 279\nn009230 489\nn009231 425\nn009233 436\nn009234 357\nn009236 298\nn009237 368\nn009238 376\nn009239 397\nn009240 270\nn009241 292\nn009242 389\nn009243 469\nn009244 206\nn009245 405\nn009246 376\nn009247 402\nn009248 282\nn009249 263\nn009250 331\nn009251 330\nn009252 393\nn009253 272\nn009254 360\nn009255 351\nn009256 264\nn009257 331\nn009258 403\nn009259 417\nn009260 377\nn009261 295\nn009262 263\nn009263 503\nn009264 507\nn009265 361\nn009266 250\nn009267 353\nn009268 507\nn009269 385\nn009270 293\nn009271 478\nn009272 403\nn009273 396\nn009274 194\nn009275 499\nn009276 173\nn009277 406\nn009278 179\nn009279 363\n"
  }
]