SYMBOL INDEX (223 symbols across 18 files)

FILE: dofa_v1.py
  class OFAViT (line 28) | class OFAViT(nn.Module):
    method __init__ (line 31) | def __init__(self, img_size=224, patch_size=16, drop_rate=0.,
    method forward_features (line 58) | def forward_features(self, x, wave_list):
    method forward_head (line 83) | def forward_head(self, x, pre_logits=False):
    method forward (line 87) | def forward(self, x, wave_list):
  function vit_small_patch16 (line 93) | def vit_small_patch16(**kwargs):
  function vit_base_patch16 (line 99) | def vit_base_patch16(**kwargs):
  function vit_large_patch16 (line 106) | def vit_large_patch16(**kwargs):
  function vit_huge_patch14 (line 113) | def vit_huge_patch14(**kwargs):

FILE: hubconf.py
  function vit_base_dofa (line 9) | def vit_base_dofa(pretrained=True, strict=False, **kwargs):

FILE: pretraining/datasets/ofall_dataset.py
  class DataAugmentation (line 123) | class DataAugmentation(torch.nn.Module):
    method __init__ (line 124) | def __init__(self, mean, std):
    method forward (line 133) | def forward(self,x):
  class Sentinel1Dataset (line 139) | class Sentinel1Dataset(Dataset):
    method __init__ (line 140) | def __init__(self, root_dir, split='train', transform=True):
    method __getitem__ (line 151) | def __getitem__(self, index):
    method __len__ (line 172) | def __len__(self):
  class Sentinel2Dataset (line 176) | class Sentinel2Dataset(Dataset):
    method __init__ (line 177) | def __init__(self, root_dir, split='train', transform=True):
    method __getitem__ (line 189) | def __getitem__(self, index):
    method __len__ (line 207) | def __len__(self):
  class NAIPDataset (line 210) | class NAIPDataset(Dataset):
    method __init__ (line 211) | def __init__(self, root_dir, split='train', transform=True):
    method __getitem__ (line 221) | def __getitem__(self, index):
    method __len__ (line 237) | def __len__(self):
  class HyperDataset (line 240) | class HyperDataset(Dataset):
    method __init__ (line 241) | def __init__(self, root_dir, split='train', transform=True):
    method __getitem__ (line 259) | def __getitem__(self, index):
    method __len__ (line 278) | def __len__(self):
  class GaufenDataset (line 281) | class GaufenDataset(Dataset):
    method __init__ (line 282) | def __init__(self, root_dir, split='train', transform=True):
    method __getitem__ (line 293) | def __getitem__(self, index):
    method __len__ (line 306) | def __len__(self):
  function chunk (line 312) | def chunk(indices, size):
  class MyBatchSampler (line 316) | class MyBatchSampler(Sampler):
    method __init__ (line 317) | def __init__(self, dataset, dataset1, dataset2, dataset3, dataset4, da...
    method __iter__ (line 332) | def __iter__(self):
    method __len__ (line 365) | def __len__(self):
  class MyDistributedBatchSampler (line 375) | class MyDistributedBatchSampler(DistributedSampler):
    method __init__ (line 376) | def __init__(self, dataset, dataset1, dataset2, dataset3, dataset4, da...
    method __iter__ (line 415) | def __iter__(self):
    method __len__ (line 452) | def __len__(self) -> int:

FILE: pretraining/engine_pretrain.py
  function train_one_epoch (line 25) | def train_one_epoch(model: torch.nn.Module,

FILE: pretraining/main_pretrain_ofa.py
  function get_args_parser (line 40) | def get_args_parser():
  function main (line 117) | def main(args):

FILE: pretraining/models_base_ofa_mae.py
  class MaskedAutoencoderViT (line 27) | class MaskedAutoencoderViT(nn.Module):
    method __init__ (line 30) | def __init__(self, img_size=224, patch_size=16, in_chans=[2,9,3,202,4],
    method initialize_weights (line 81) | def initialize_weights(self):
    method _init_weights (line 113) | def _init_weights(self, m):
    method patchify (line 123) | def patchify(self, imgs):
    method unpatchify (line 137) | def unpatchify(self, x):
    method random_select_channels (line 151) | def random_select_channels(self, imgs, wave_list):
    method random_masking (line 164) | def random_masking(self, x, mask_ratio):
    method forward_zs (line 189) | def forward_zs(self, x, wave_list):
    method forward_encoder (line 205) | def forward_encoder(self, x, mask_ratio, wave_list):
    method forward_decoder (line 229) | def forward_decoder(self, x, ids_restore, in_chan):
    method forward_loss (line 255) | def forward_loss(self, imgs, pred, mask):
    method forward_zt (line 273) | def forward_zt(self, imgs, wave_list):
    method forward (line 296) | def forward(self, imgs, in_chans, wave_list, mask_ratio=0.75):
  function mae_vit_small_patch16_dec512d8b (line 306) | def mae_vit_small_patch16_dec512d8b(**kwargs):
  function mae_vit_base_patch16_dec512d8b (line 313) | def mae_vit_base_patch16_dec512d8b(**kwargs):
  function mae_vit_large_patch16_dec512d8b (line 321) | def mae_vit_large_patch16_dec512d8b(**kwargs):
  function mae_vit_huge_patch14_dec512d8b (line 329) | def mae_vit_huge_patch14_dec512d8b(**kwargs):

FILE: pretraining/samplers/distributed.py
  class DistributedRandomGeoSampler (line 11) | class DistributedRandomGeoSampler(RandomGeoSampler):
    method __init__ (line 21) | def __init__(
    method __iter__ (line 58) | def __iter__(self) -> Iterator[BoundingBox]:
    method __len__ (line 83) | def __len__(self) -> int:
    method set_epoch (line 91) | def set_epoch(self, epoch: int) -> None:

FILE: pretraining/util/crop.py
  class RandomResizedCrop (line 15) | class RandomResizedCrop(transforms.RandomResizedCrop):
    method get_params (line 23) | def get_params(img, scale, ratio):

FILE: pretraining/util/datasets.py
  function build_dataset (line 20) | def build_dataset(is_train, args):
  function build_transform (line 31) | def build_transform(is_train, args):

FILE: pretraining/util/lars.py
  class LARS (line 14) | class LARS(torch.optim.Optimizer):
    method __init__ (line 18) | def __init__(self, params, lr=0, weight_decay=0, momentum=0.9, trust_c...
    method step (line 23) | def step(self):

FILE: pretraining/util/lr_decay.py
  function param_groups_lrd (line 15) | def param_groups_lrd(model, weight_decay=0.05, no_weight_decay_list=[], ...
  function get_layer_id_for_vit (line 64) | def get_layer_id_for_vit(name, num_layers):

FILE: pretraining/util/lr_sched.py
  function adjust_learning_rate (line 9) | def adjust_learning_rate(optimizer, epoch, args):

FILE: pretraining/util/misc.py
  class SmoothedValue (line 27) | class SmoothedValue(object):
    method __init__ (line 32) | def __init__(self, window_size=20, fmt=None):
    method update (line 40) | def update(self, value, n=1):
    method synchronize_between_processes (line 45) | def synchronize_between_processes(self):
    method median (line 59) | def median(self):
    method avg (line 64) | def avg(self):
    method global_avg (line 69) | def global_avg(self):
    method max (line 73) | def max(self):
    method value (line 77) | def value(self):
    method __str__ (line 80) | def __str__(self):
  class MetricLogger (line 89) | class MetricLogger(object):
    method __init__ (line 90) | def __init__(self, delimiter="\t"):
    method update (line 94) | def update(self, **kwargs):
    method __getattr__ (line 103) | def __getattr__(self, attr):
    method __str__ (line 111) | def __str__(self):
    method synchronize_between_processes (line 119) | def synchronize_between_processes(self):
    method add_meter (line 123) | def add_meter(self, name, meter):
    method log_every (line 126) | def log_every(self, iterable, print_freq, header=None):
  function setup_for_distributed (line 173) | def setup_for_distributed(is_master):
  function is_dist_avail_and_initialized (line 190) | def is_dist_avail_and_initialized():
  function get_world_size (line 198) | def get_world_size():
  function get_rank (line 204) | def get_rank():
  function is_main_process (line 210) | def is_main_process():
  function save_on_master (line 214) | def save_on_master(*args, **kwargs):
  function init_distributed_mode (line 219) | def init_distributed_mode(args):
  class NativeScalerWithGradNormCount (line 254) | class NativeScalerWithGradNormCount:
    method __init__ (line 257) | def __init__(self):
    method __call__ (line 260) | def __call__(self, loss, optimizer, clip_grad=None, parameters=None, c...
    method state_dict (line 276) | def state_dict(self):
    method load_state_dict (line 279) | def load_state_dict(self, state_dict):
  function get_grad_norm_ (line 283) | def get_grad_norm_(parameters, norm_type: float = 2.0) -> torch.Tensor:
  function save_model (line 298) | def save_model(args, epoch, model, model_without_ddp, optimizer, loss_sc...
  function load_model (line 318) | def load_model(args, model_without_ddp, optimizer, loss_scaler):
  function all_reduce_mean (line 335) | def all_reduce_mean(x):
  class CLIPLoss (line 347) | class CLIPLoss(nn.Module):
    method __init__ (line 348) | def __init__(self, temperature=1.0):
    method forward (line 352) | def forward(self, image_features, text_features):

FILE: pretraining/util/pos_embed.py
  function get_2d_sincos_pos_embed (line 20) | def get_2d_sincos_pos_embed(embed_dim, grid_size, cls_token=False, promp...
  function get_2d_sincos_pos_embed_with_resolution (line 40) | def get_2d_sincos_pos_embed_with_resolution(
  function get_2d_sincos_pos_embed_from_grid (line 78) | def get_2d_sincos_pos_embed_from_grid(embed_dim, grid):
  function get_2d_sincos_pos_embed_from_grid_torch (line 89) | def get_2d_sincos_pos_embed_from_grid_torch(embed_dim, grid):
  function get_1d_sincos_pos_embed_from_grid_torch (line 104) | def get_1d_sincos_pos_embed_from_grid_torch(embed_dim, pos):
  function get_1d_sincos_pos_embed_from_grid (line 126) | def get_1d_sincos_pos_embed_from_grid(embed_dim, pos):
  function interpolate_pos_embed (line 152) | def interpolate_pos_embed(model, checkpoint_model, num_patches=None):

FILE: pretraining/util/vision_transformer.py
  function _cfg (line 21) | def _cfg(url='', **kwargs):
  class Mlp (line 90) | class Mlp(nn.Module):
    method __init__ (line 91) | def __init__(self, in_features, hidden_features=None, out_features=Non...
    method forward (line 100) | def forward(self, x):
  class Attention (line 109) | class Attention(nn.Module):
    method __init__ (line 110) | def __init__(self, dim, num_heads=8, qkv_bias=False, qk_scale=None, at...
    method forward (line 122) | def forward(self, x):
  class Block (line 137) | class Block(nn.Module):
    method __init__ (line 139) | def __init__(self, dim, num_heads, mlp_ratio=4., qkv_bias=False, qk_sc...
    method forward (line 151) | def forward(self, x):
  class PatchEmbed (line 157) | class PatchEmbed(nn.Module):
    method __init__ (line 160) | def __init__(self, img_size=224, patch_size=16, in_chans=3, embed_dim=...
    method forward (line 171) | def forward(self, x):
  class HybridEmbed (line 180) | class HybridEmbed(nn.Module):
    method __init__ (line 184) | def __init__(self, backbone, img_size=224, feature_size=None, in_chans...
    method forward (line 213) | def forward(self, x):
  class VisionTransformer (line 221) | class VisionTransformer(nn.Module):
    method __init__ (line 224) | def __init__(self, img_size=224, patch_size=16, in_chans=3, num_classe...
    method _init_weights (line 264) | def _init_weights(self, m):
    method no_weight_decay (line 274) | def no_weight_decay(self):
    method get_classifier (line 277) | def get_classifier(self):
    method reset_classifier (line 280) | def reset_classifier(self, num_classes, global_pool=''):
    method forward_features (line 284) | def forward_features(self, x):
    method forward (line 299) | def forward(self, x):
  function resize_pos_embed (line 305) | def resize_pos_embed(posemb, posemb_new):
  function checkpoint_filter_fn (line 325) | def checkpoint_filter_fn(state_dict, model):
  function _create_vision_transformer (line 343) | def _create_vision_transformer(variant, pretrained=False, distilled=Fals...
  function vit_small_patch16_224 (line 371) | def vit_small_patch16_224(pretrained=False, **kwargs):
  function vit_base_patch16_224 (line 384) | def vit_base_patch16_224(pretrained=False, **kwargs):
  function vit_base_patch32_224 (line 394) | def vit_base_patch32_224(pretrained=False, **kwargs):
  function vit_base_patch16_384 (line 403) | def vit_base_patch16_384(pretrained=False, **kwargs):
  function vit_base_patch32_384 (line 413) | def vit_base_patch32_384(pretrained=False, **kwargs):
  function vit_large_patch16_224 (line 423) | def vit_large_patch16_224(pretrained=False, **kwargs):
  function vit_large_patch32_224 (line 433) | def vit_large_patch32_224(pretrained=False, **kwargs):
  function vit_large_patch16_384 (line 442) | def vit_large_patch16_384(pretrained=False, **kwargs):
  function vit_large_patch32_384 (line 452) | def vit_large_patch32_384(pretrained=False, **kwargs):
  function vit_base_patch16_224_in21k (line 462) | def vit_base_patch16_224_in21k(pretrained=False, **kwargs):
  function vit_base_patch32_224_in21k (line 473) | def vit_base_patch32_224_in21k(pretrained=False, **kwargs):
  function vit_large_patch16_224_in21k (line 484) | def vit_large_patch16_224_in21k(pretrained=False, **kwargs):
  function vit_large_patch32_224_in21k (line 495) | def vit_large_patch32_224_in21k(pretrained=False, **kwargs):
  function vit_huge_patch14_224_in21k (line 506) | def vit_huge_patch14_224_in21k(pretrained=False, **kwargs):
  function vit_base_resnet50_224_in21k (line 518) | def vit_base_resnet50_224_in21k(pretrained=False, **kwargs):
  function vit_base_resnet50_384 (line 534) | def vit_base_resnet50_384(pretrained=False, **kwargs):
  function vit_small_resnet26d_224 (line 548) | def vit_small_resnet26d_224(pretrained=False, **kwargs):
  function vit_small_resnet50d_s3_224 (line 558) | def vit_small_resnet50d_s3_224(pretrained=False, **kwargs):
  function vit_base_resnet26d_224 (line 568) | def vit_base_resnet26d_224(pretrained=False, **kwargs):
  function vit_base_resnet50d_224 (line 578) | def vit_base_resnet50d_224(pretrained=False, **kwargs):
  function vit_deit_tiny_patch16_224 (line 588) | def vit_deit_tiny_patch16_224(pretrained=False, **kwargs):
  function deit_small_resnet50_224 (line 597) | def deit_small_resnet50_224(pretrained=False, **kwargs):

FILE: pretraining/util/wandb_log.py
  class WANDB_LOG_IMG_CONFIG (line 19) | class WANDB_LOG_IMG_CONFIG:
  function equalize (line 25) | def equalize(x):
  function wandb_dump_input_output (line 34) | def wandb_dump_input_output(x, ys, epoch=0, texts=""):
  function wandb_dump_images (line 57) | def wandb_dump_images(imgs, name="vis", epoch=0):
  function compare_pos_embedding (line 71) | def compare_pos_embedding(posa, posb, ns=[0]):
  function wandb_log_metadata (line 90) | def wandb_log_metadata(metadata):

FILE: pretraining/wave_dynamic_layer.py
  class TransformerWeightGenerator (line 26) | class TransformerWeightGenerator(nn.Module):
    method __init__ (line 27) | def __init__(self, input_dim, output_dim, embed_dim, num_heads=4, num_...
    method forward (line 43) | def forward(self, x):
  class GaussianFourierFeatureTransform (line 55) | class GaussianFourierFeatureTransform(torch.nn.Module):
    method __init__ (line 67) | def __init__(self, num_input_channels, mapping_size=256, scale=10):
    method forward (line 75) | def forward(self, x):
  class Basic1d (line 98) | class Basic1d(nn.Module):
    method __init__ (line 99) | def __init__(self, in_channels, out_channels, bias=True):
    method forward (line 107) | def forward(self, x):
  class FCResLayer (line 111) | class FCResLayer(nn.Module):
    method __init__ (line 112) | def __init__(self, linear_size=128):
    method forward (line 121) | def forward(self, x):
  class Dynamic_MLP_Decoder (line 131) | class Dynamic_MLP_Decoder(nn.Module):
    method __init__ (line 132) | def __init__(self, wv_planes, inter_dim=128, kernel_size=16, decoder_e...
    method _get_weights (line 147) | def _get_weights(self, waves, batch=True):
    method weight_init (line 159) | def weight_init(self, m):
    method _init_weights (line 164) | def _init_weights(self):
    method forward (line 170) | def forward(self, img_feat, waves):
  class Dynamic_MLP_OFA (line 181) | class Dynamic_MLP_OFA(nn.Module):
    method __init__ (line 189) | def __init__(self, wv_planes, inter_dim = 128, kernel_size=3, embed_di...
    method _get_weights (line 206) | def _get_weights(self, waves):
    method weight_init (line 212) | def weight_init(self, m):
    method _init_weights (line 217) | def _init_weights(self):
    method forward (line 225) | def forward(self, img_feat, wvs):

FILE: wave_dynamic_layer.py
  function get_1d_sincos_pos_embed_from_grid_torch (line 11) | def get_1d_sincos_pos_embed_from_grid_torch(embed_dim, pos):
  class TransformerWeightGenerator (line 31) | class TransformerWeightGenerator(nn.Module):
    method __init__ (line 32) | def __init__(self, input_dim, output_dim, embed_dim, num_heads=4, num_...
    method forward (line 58) | def forward(self, x):
  class Basic1d (line 71) | class Basic1d(nn.Module):
    method __init__ (line 72) | def __init__(self, in_channels, out_channels, bias=True):
    method forward (line 82) | def forward(self, x):
  class FCResLayer (line 87) | class FCResLayer(nn.Module):
    method __init__ (line 88) | def __init__(self, linear_size=128):
    method forward (line 96) | def forward(self, x):
  class Dynamic_MLP_Decoder (line 105) | class Dynamic_MLP_Decoder(nn.Module):
    method __init__ (line 106) | def __init__(self, wv_planes, inter_dim=128, kernel_size=16, decoder_e...
    method _get_weights (line 121) | def _get_weights(self, waves, batch=True):
    method weight_init (line 133) | def weight_init(self, m):
    method _init_weights (line 138) | def _init_weights(self):
    method forward (line 144) | def forward(self, img_feat, waves):
  class Dynamic_MLP_OFA (line 157) | class Dynamic_MLP_OFA(nn.Module):
    method __init__ (line 165) | def __init__(self, wv_planes, inter_dim=128, kernel_size=3, embed_dim=...
    method _get_weights (line 185) | def _get_weights(self, waves):
    method weight_init (line 190) | def weight_init(self, m):
    method _init_weights (line 195) | def _init_weights(self):
    method forward (line 202) | def forward(self, img_feat, wvs):