Repository: holyhao/Baidu-Dogs Branch: master Commit: 6c5e56f8b1c3 Files: 16 Total size: 19.9 KB Directory structure: gitextract_0h9pdhwq/ ├── .gitignore ├── README.md ├── data/ │ ├── README.md │ ├── test/ │ │ └── README.md │ ├── train/ │ │ └── README.md │ └── val/ │ └── README.md ├── evaluate/ │ ├── README.md │ ├── predict_bygenerator.py │ └── predict_onebyone.py ├── models/ │ ├── README.md │ ├── dogs.py │ ├── inceptionV3.py │ └── vgg19.py └── preprocessing/ ├── README.md ├── altoFolders.py └── divforValidation.py ================================================ FILE CONTENTS ================================================ ================================================ FILE: .gitignore ================================================ *.h5 *.txt ================================================ FILE: README.md ================================================ # Fine-grained Dog Classification competition - This is a dog classification competition held by Baidu. Further information at http://js.baidu.com/ ## Framework - [Keras](https://keras.io/) - [Tensorflow Backend](https://www.tensorflow.org/) ## Hardware - Geforce GTX TITANX 12G - Intel® Core™ i7-6700 CPU - Memory 16G - Operate system Ubuntu 14.04 ## Data - Download the images from Baidu Cloud - Training Set: http://pan.baidu.com/s/1slLOqBz Key: 5axb - Test set: http://pan.baidu.com/s/1gfaf9rt Key：fl5n - Put the images into diffrent directory by their class labels. Refer to [altoFolders.py](preprocessing/altoFolders.py) for doing this. - Take 20% of the labeled data for validation. Refer to [divforValidation.py](preprocessing/divforValidation.py). ## Base Model - [VGG19](models/vgg19.py) for deep feature extraction, which is provided in keras models. - [InceptionV3](models/InceptionV3) for deep feature extraction, which is provided in keras models. - Softmax for classification. ## Evaluate - Predict the classes for unlabeled data one by one refering to [predict_onebyone](evaluate/predict_onebyone.py) and by generator refering to [predict_bygenerator.by](evaluate/predict_bygenerator.py). ## to be continued > Feel free to contact me if you have any issues or better ideas about anything. > by Holy ================================================ FILE: data/README.md ================================================ #Data Data for train,validation and test. Note that the validation is 20% here from the original data. (Random sampling from different classes.) ================================================ FILE: data/test/README.md ================================================ #Test Data for testing. ================================================ FILE: data/train/README.md ================================================ #TRAIN Data for trainning. ================================================ FILE: data/val/README.md ================================================ #Validation Data for validation. ================================================ FILE: evaluate/README.md ================================================ #evaluate some evaluation methods to be added ================================================ FILE: evaluate/predict_bygenerator.py ================================================ # -*- coding: utf-8 -*- """ Created on Mon Jul 10 21:27:15 2017 @author: Administrator """ import os import shutil import cv2 import h5py import numpy as np import pandas as pd model_path="../models/model_dogs_Xception.h5" test_data_dir="../data/test" val_data_dir="../data/val" model=load_model(model_path) test_datagen = ImageDataGenerator(rescale=1./255) batch_size=64 #for generate class_indices val_generator = test_datagen.flow_from_directory( val_data_dir, target_size=(299, 299), batch_size=batch_size, shuffle=False, class_mode='categorical' ) label_idxs = sorted(valid_generator.class_indices.items(), key=operator.itemgetter(1)) test_generator = test_datagen.flow_from_directory( test_data_dir, target_size=(299, 299), batch_size=batch_size, shuffle=False, class_mode='categorical') y= model.predict_generator(test_generator, test_generator.samples/batch_size + 1) y_max_idx = np.argmax(y1, 1) predict_path = 'submission.txt' with open(predict_path,'a') as fp: for i, idx in enumerate(y_max_idx): fp.write(str(label_idxs[idx][0]) + '\t' + test_generator.filenames[i][6:-4] + '\n') fp.close ================================================ FILE: evaluate/predict_onebyone.py ================================================ # -*- coding: utf-8 -*- """ Created on Mon Jul 10 21:27:15 2017 @author: Administrator """ import os import shutil import cv2 import h5py import numpy as np import pandas as pd model_path=model_dogs_Xception.h5 model=load_model(model_path) test_filenames=os.listdir(test_data_dir) test_img=[] predictions=[] for file_path in test_filenames: img=cv2.imread(test_data_dir+file_path) img=cv2.cvtColor(img,cv2.COLOR_BGR2RGB) img=cv2.resize(img,(299,299),interpolation=cv2.INTER_CUBIC) test_img.append(img) test_img=np.array(test_img) test_img = test_img.astype('float32') test_img/=255 pre=model.predict(test_img,batch_size=1)[0] predictions.append(pre) test_img=[] probs=np.array(predictions) classes=np.argmax(probs,1) #clsses-labels class_indices=np.load('class_indices.txt.npy') #(int,int) class_indices=class_indices.tolist() #(int,str) value_indices={v:k for k,v in class_indices.items()} true_class=[] for i in range(len(classes)): true_class.append(value_indices[classes[i]]) with open('submit.txt','a') as fp: for i in range(len(test_filenames)): fp.write(str(true_class[i])+"\t"+str(test_filenames[i].split(".")[0])+'\n') fp.close ================================================ FILE: models/README.md ================================================ #Models Some models to be added ================================================ FILE: models/dogs.py ================================================ #!/usr/bin/env python2 # -*- coding: utf-8 -*- """ Created on Mon Jul 10 20:49:09 2017 @author: Administrator """ import matplotlib.pyplot as plt import numpy as np import pandas as pd import cv2 import h5py from keras.models import Sequential, Model, load_model from keras import applications from keras import optimizers from keras.layers import Dropout, Flatten, Dense, Input,GlobalAveragePooling2D from keras.utils import vis_utils,plot_model from keras.preprocessing.image import ImageDataGenerator from keras.callbacks import ModelCheckpoint,ReduceLROnPlateau from vgg19 import VGG19 #模型的构建 img_rows, img_cols, img_channel = 400, 400, 3 base_model = VGG19(weights='imagenet', include_top=False,input_shape=(img_rows, img_cols, img_channel)) add_model = Sequential() add_model.add(Flatten(input_shape=base_model.output_shape[1:])) add_model.add(Dense(1024, activation='relu')) add_model.add(Dense(100, activation='softmax')) model = Model(inputs=base_model.input, outputs=add_model(base_model.output)) model.compile(loss='categorical_crossentropy', optimizer=optimizers.SGD(lr=1e-4, momentum=0.9), metrics=['accuracy']) #冻结某些层设置 #for layer in model.layers[:85]: # layer.trainable = False #打印网络结构 #model.summary() for i, layer in enumerate(base_model.layers): print(i, layer.name) #参数设置 batch_size = 32 epochs = 50 train_data_dir="data/train" val_data_dir="data/val" #plot_model(model,to_file='model.png') train_datagen = ImageDataGenerator( rotation_range=30, rescale=1./255, shear_range=0.2, zoom_range=0.2, width_shift_range=0.1, height_shift_range=0.1, horizontal_flip=True) val_datagen=ImageDataGenerator(rescale=1./255) val_generator = val_datagen.flow_from_directory( val_data_dir, target_size=(img_rows, img_cols), batch_size=batch_s, class_mode='categorical') train_generator = train_datagen.flow_from_directory( train_data_dir, target_size=(img_rows, img_cols), batch_size=batch_size, class_mode='categorical') #保存映射表 #with h5py.File("class_indices.h5") as h: #h.create_dataset("class_indices",data=train_generator.class_indices) #np.save('class_indices.txt', train_generator.class_indices) history = model.fit_generator( train_generator, steps_per_epoch=train_generator.samples/batch_size, epochs=epochs, validation_data=val_generator, validation_steps=batch_s #callbacks=[ModelCheckpoint('VGG16-transferlearning.model', monitor='val_acc', save_best_only=True)] ) model.save('model_dogs_VGG19_400*400_full.h5') ================================================ FILE: models/inceptionV3.py ================================================ import os import keras import numpy as np from keras import Input from keras import backend as K from keras.applications import Xception,InceptionV3 from keras.callbacks import EarlyStopping, ModelCheckpoint, ReduceLROnPlateau from keras.layers import Dense, Dropout, Lambda,AveragePooling2D,Flatten from keras.models import Model, load_model from keras.preprocessing.image import ImageDataGenerator from keras.optimizers import SGD from keras.utils import plot_model img_rows,img_cols,img_channel=299,299,3 train_datagen = ImageDataGenerator( rotation_range=30, rescale=1./255, shear_range=0.3, zoom_range=0.2, width_shift_range=0.2, height_shift_range=0.2, horizontal_flip=True) test_datagen = ImageDataGenerator(rescale=1./255) early_stopping = EarlyStopping(monitor='val_loss', patience=3) auto_lr = ReduceLROnPlateau(monitor='val_loss', factor=0.1, patience=3, verbose=0, mode='auto', epsilon=0.001, cooldown=0, min_lr=0) save_model = ModelCheckpoint('InceptionV3{epoch:02d}-{val_acc:.2f}.h5', period=2) # create the base pre-trained model input_tensor = Input(shape=(img_rows, img_cols, img_channel)) base_model = InceptionV3(weights='imagenet', include_top=False,input_shape=(img_rows, img_cols, img_channel)) x=AveragePooling2D(pool_size=(4,4))(base_model.output) x=Dropout(0.5)(x) x=Flatten()(x) x=Dense(100,activation='softmax')(x) #base_model model = Model(inputs=base_model.input, outputs=x) for layer in base_model.layers: layer.trainable = False batch_size = 48 epoch=5 train_generator = train_datagen.flow_from_directory( '../data/train', target_size=(img_rows, img_cols), batch_size=batch_size, class_mode='categorical') validation_generator = test_datagen.flow_from_directory( '../data/val', target_size=(img_rows, img_cols), batch_size=batch_size, class_mode='categorical') model.compile(loss="categorical_crossentropy", optimizer=SGD(lr=1e-3, momentum=0.9), metrics=['accuracy']) # model = make_parallel(model, 3) # train fc model.fit_generator(train_generator, steps_per_epoch=train_generator.samples/batch_size+1, epochs=epoch, validation_data=validation_generator, validation_steps=validation_generator.samples/batch_size+1 #callbacks=[early_stopping, auto_lr, save_model] ) # train all for layer in model.layers: layer.trainable = True model.summary() batch_size=24 epoch=30 train_generator = train_datagen.flow_from_directory( '../data/train', target_size=(img_rows, img_cols), batch_size=batch_size, class_mode='categorical') validation_generator = test_datagen.flow_from_directory( '../data/val', target_size=(img_rows, img_cols), batch_size=batch_size, class_mode='categorical') model.compile(loss="categorical_crossentropy", optimizer=SGD(lr=1e-4, momentum=0.9), metrics=['accuracy']) save_model = ModelCheckpoint('InceptionV3-{epoch:02d}-{val_acc:.2f}.h5', period=2) model.fit_generator(train_generator, steps_per_epoch=train_generator.samples/batch_size+1, epochs=epoch, validation_data=validation_generator, validation_steps=validation_generator.samples/batch_size+1, callbacks=[auto_lr, save_model]) # otherwise the generator would loop indefinitely model.save('inceptionV3.h5') ================================================ FILE: models/vgg19.py ================================================ # -*- coding: utf-8 -*- '''VGG19 model for Keras. # Reference: - [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://arxiv.org/abs/1409.1556) ''' from __future__ import print_function import numpy as np import warnings from keras.models import Model from keras.layers import Flatten, Dense, Input from keras.layers import Conv2D from keras.layers import MaxPooling2D from keras.layers import GlobalMaxPooling2D from keras.layers import GlobalAveragePooling2D from keras.preprocessing import image from keras.utils import layer_utils from keras.utils.data_utils import get_file from keras import backend as K from keras.applications.imagenet_utils import decode_predictions from keras.applications.imagenet_utils import preprocess_input from keras.applications.imagenet_utils import _obtain_input_shape from keras.engine.topology import get_source_inputs WEIGHTS_PATH = 'https://github.com/fchollet/deep-learning-models/releases/download/v0.1/vgg19_weights_tf_dim_ordering_tf_kernels.h5' WEIGHTS_PATH_NO_TOP = 'https://github.com/fchollet/deep-learning-models/releases/download/v0.1/vgg19_weights_tf_dim_ordering_tf_kernels_notop.h5' def VGG19(include_top=True, weights='imagenet', input_tensor=None, input_shape=None, pooling=None, classes=1000): """Instantiates the VGG19 architecture. Optionally loads weights pre-trained on ImageNet. Note that when using TensorFlow, for best performance you should set `image_data_format="channels_last"` in your Keras config at ~/.keras/keras.json. The model and the weights are compatible with both TensorFlow and Theano. The data format convention used by the model is the one specified in your Keras config file. # Arguments include_top: whether to include the 3 fully-connected layers at the top of the network. weights: one of `None` (random initialization) or "imagenet" (pre-training on ImageNet). input_tensor: optional Keras tensor (i.e. output of `layers.Input()`) to use as image input for the model. input_shape: optional shape tuple, only to be specified if `include_top` is False (otherwise the input shape has to be `(224, 224, 3)` (with `channels_last` data format) or `(3, 224, 244)` (with `channels_first` data format). It should have exactly 3 inputs channels, and width and height should be no smaller than 48. E.g. `(200, 200, 3)` would be one valid value. pooling: Optional pooling mode for feature extraction when `include_top` is `False`. - `None` means that the output of the model will be the 4D tensor output of the last convolutional layer. - `avg` means that global average pooling will be applied to the output of the last convolutional layer, and thus the output of the model will be a 2D tensor. - `max` means that global max pooling will be applied. classes: optional number of classes to classify images into, only to be specified if `include_top` is True, and if no `weights` argument is specified. # Returns A Keras model instance. # Raises ValueError: in case of invalid argument for `weights`, or invalid input shape. """ if weights not in {'imagenet', None}: raise ValueError('The `weights` argument should be either ' '`None` (random initialization) or `imagenet` ' '(pre-training on ImageNet).') if weights == 'imagenet' and include_top and classes != 1000: raise ValueError('If using `weights` as imagenet with `include_top`' ' as true, `classes` should be 1000') # Determine proper input shape input_shape = _obtain_input_shape(input_shape, default_size=224, min_size=48, data_format=K.image_data_format(), include_top=include_top) if input_tensor is None: img_input = Input(shape=input_shape) else: if not K.is_keras_tensor(input_tensor): img_input = Input(tensor=input_tensor, shape=input_shape) else: img_input = input_tensor # Block 1 x = Conv2D(64, (3, 3), activation='relu', padding='same', name='block1_conv1')(img_input) x = Conv2D(64, (3, 3), activation='relu', padding='same', name='block1_conv2')(x) x = MaxPooling2D((2, 2), strides=(2, 2), name='block1_pool')(x) # Block 2 x = Conv2D(128, (3, 3), activation='relu', padding='same', name='block2_conv1')(x) x = Conv2D(128, (3, 3), activation='relu', padding='same', name='block2_conv2')(x) x = MaxPooling2D((2, 2), strides=(2, 2), name='block2_pool')(x) # Block 3 x = Conv2D(256, (3, 3), activation='relu', padding='same', name='block3_conv1')(x) x = Conv2D(256, (3, 3), activation='relu', padding='same', name='block3_conv2')(x) x = Conv2D(256, (3, 3), activation='relu', padding='same', name='block3_conv3')(x) x = Conv2D(256, (3, 3), activation='relu', padding='same', name='block3_conv4')(x) x = MaxPooling2D((2, 2), strides=(2, 2), name='block3_pool')(x) # Block 4 x = Conv2D(512, (3, 3), activation='relu', padding='same', name='block4_conv1')(x) x = Conv2D(512, (3, 3), activation='relu', padding='same', name='block4_conv2')(x) x = Conv2D(512, (3, 3), activation='relu', padding='same', name='block4_conv3')(x) x = Conv2D(512, (3, 3), activation='relu', padding='same', name='block4_conv4')(x) x = MaxPooling2D((2, 2), strides=(2, 2), name='block4_pool')(x) # Block 5 x = Conv2D(512, (3, 3), activation='relu', padding='same', name='block5_conv1')(x) x = Conv2D(512, (3, 3), activation='relu', padding='same', name='block5_conv2')(x) x = Conv2D(512, (3, 3), activation='relu', padding='same', name='block5_conv3')(x) x = Conv2D(512, (3, 3), activation='relu', padding='same', name='block5_conv4')(x) x = MaxPooling2D((2, 2), strides=(2, 2), name='block5_pool')(x) if include_top: # Classification block x = Flatten(name='flatten')(x) x = Dense(4096, activation='relu', name='fc1')(x) x = Dense(4096, activation='relu', name='fc2')(x) x = Dense(classes, activation='softmax', name='predictions')(x) else: if pooling == 'avg': x = GlobalAveragePooling2D()(x) elif pooling == 'max': x = GlobalMaxPooling2D()(x) # Ensure that the model takes into account # any potential predecessors of `input_tensor`. if input_tensor is not None: inputs = get_source_inputs(input_tensor) else: inputs = img_input # Create model. model = Model(inputs, x, name='vgg19') # load weights if weights == 'imagenet': if include_top: weights_path = get_file('vgg19_weights_tf_dim_ordering_tf_kernels.h5', WEIGHTS_PATH, cache_subdir='models') else: weights_path = get_file('vgg19_weights_tf_dim_ordering_tf_kernels_notop.h5', WEIGHTS_PATH_NO_TOP, cache_subdir='models') model.load_weights(weights_path) if K.backend() == 'theano': layer_utils.convert_all_kernels_in_model(model) if K.image_data_format() == 'channels_first': if include_top: maxpool = model.get_layer(name='block5_pool') shape = maxpool.output_shape[1:] dense = model.get_layer(name='fc1') layer_utils.convert_dense_weights_data_format(dense, shape, 'channels_first') if K.backend() == 'tensorflow': warnings.warn('You are using the TensorFlow backend, yet you ' 'are using the Theano ' 'image data format convention ' '(`image_data_format="channels_first"`). ' 'For best performance, set ' '`image_data_format="channels_last"` in ' 'your Keras config ' 'at ~/.keras/keras.json.') return model if __name__ == '__main__': model = VGG19(include_top=True, weights='imagenet') img_path = 'cat.jpg' img = image.load_img(img_path, target_size=(224, 224)) x = image.img_to_array(img) x = np.expand_dims(x, axis=0) x = preprocess_input(x) print('Input image shape:', x.shape) preds = model.predict(x) print('Predicted:', decode_predictions(preds)) ================================================ FILE: preprocessing/README.md ================================================ #Preprocessing Some preprocessing to be added. ================================================ FILE: preprocessing/altoFolders.py ================================================ # -*- coding: utf-8 -*- """ Created on Mon Jul 10 19:15:14 2017 @author: holy """ import os import shutil import pandas as pd import numpy as np label = pd.read_csv("originaldata.txt",sep='\s+',encoding='utf-8',escapechar='\n') train_filenames=os.listdir('originaldata') def ex_mkdir(dirname): if not os.path.exists(dirname): os.mkdir(dirname) def rmrf_mkdir(dirname): if os.path.exists(dirname): shutil.rmtree(dirname) os.mkdir(dirname) ex_mkdir('../data/train') for iter in label.index: name=label.iloc[iter,0] i=label.iloc[iter,1] ex_mkdir('train2/'+str(i)) shutil.copy('originaldata'+name+'.jpg', '../data/train/'+str(i)+'/'+name+'.jpg') ================================================ FILE: preprocessing/divforValidation.py ================================================ # -*- coding: utf-8 -*- """ Created on Mon Jul 10 19:15:14 2017 @author: holy """ filename ='../data/train' train_dir='../data/train' val_dir='../data/val' ls=os.listdir(filename) def ex_mkdir(dirname): if not os.path.exists(dirname): os.mkdir(dirname) def rmrf_mkdir(dirname): if os.path.exists(dirname): shutil.rmtree(dirname) os.mkdir(dirname) ex_mkdir(val_dir) for i in range(0,len(ls)): ex_mkdir(val_dir+str(ls[i])) data=os.listdir(train_dir+str(ls[i])) for j in range(0,int(0.2*len(data))):#%20 for validation name=data[j] shutil.move(train_dir+str(ls[i])+'/'+name,val_dir+str(ls[i])+'/'+name)