MCPcopy Create free account

hub / github.com/ThomasWangY/2024-AAAI-HPT / functions

Functions862 in github.com/ThomasWangY/2024-AAAI-HPT

⨍ Functions 862◇ Types & classes 203
↓ 2 callersMethod_check_model_name_is_valid
Validates model name.
dassl/modeling/backbone/efficientnet/model.py:308
↓ 2 callersMethod_compute_alpha
(self, mean_s, var_s, mean_t, var_t)
dassl/modeling/ops/transnorm.py:49
↓ 2 callersMethod_do_register
(self, name, obj, force=False)
dassl/utils/registry.py:36
↓ 2 callersMethod_read_data
(self, data_dir)
dassl/data/datasets/dg/cifar_c.py:87
↓ 2 callersMethod_read_data_all
(self, data_dir)
dassl/data/datasets/ssl/stl10.py:73
↓ 2 callersFunction_transform
(n_px)
clip/clip.py:71
↓ 2 callersFunction_transform
(n_px)
llms/clip/clip.py:71
↓ 2 callersMethod_transform_image
(self, tfm, img0)
dassl/data/data_manager.py:251
↓ 2 callersMethodalign
(self, seq1, seq2)
trainers/hpt_plus.py:375
↓ 2 callersMethodattention
(self, x: torch.Tensor, attn_sp: Optional[Tensor] = None,)
clip/model.py:181
↓ 2 callersFunctionbytes_to_unicode
Returns list of utf-8 byte and a corresponding list of unicode strings. The reversible bpe codes work on unicode strings. This means you
clip/simple_tokenizer.py:16
↓ 2 callersFunctionbytes_to_unicode
Returns list of utf-8 byte and a corresponding list of unicode strings. The reversible bpe codes work on unicode strings. This means you
llms/clip/simple_tokenizer.py:16
↓ 2 callersFunctionconv3x3
3x3 convolution with padding
dassl/modeling/backbone/resnet.py:16
↓ 2 callersMethodcreate_attention_matrix
(self, tokens, relationships)
trainers/hpt.py:361
↓ 2 callersMethodcreate_attention_matrix
(self, tokens, relationships)
trainers/hpt_plus.py:356
↓ 2 callersMethoddiscrepancy
(self, y1, y2)
dassl/engine/da/m3sda.py:186
↓ 2 callersMethoddiscrepancy
(self, y1, y2)
dassl/engine/da/mcd.py:100
↓ 2 callersFunctionefficientnet_params
Map EfficientNet model name to parameter coefficients.
dassl/modeling/backbone/efficientnet/utils.py:266
↓ 2 callersFunctionema_model_update
Exponential moving average of model parameters. Args: model (nn.Module): model being trained. ema_model (nn.Module): ema of the m
dassl/modeling/ops/utils.py:66
↓ 2 callersMethodeuclidean
(self, input1, input2)
dassl/engine/da/m3sda.py:183
↓ 2 callersMethodforward_backward
(self, batch)
dassl/engine/trainer.py:278
↓ 2 callersMethodfrom_name
(cls, model_name, override_params=None)
dassl/modeling/backbone/efficientnet/model.py:281
↓ 2 callersFunctionget_pairs
Return set of symbol pairs in a word. Word is represented as tuple of symbols (symbols being variable-length strings).
clip/simple_tokenizer.py:38
↓ 2 callersFunctionget_pairs
Return set of symbol pairs in a word. Word is represented as tuple of symbols (symbols being variable-length strings).
llms/clip/simple_tokenizer.py:38
↓ 2 callersMethodinit_loc_layer
Initialize the weights/bias with identity transformation.
dassl/modeling/network/ddaig_fcn.py:236
↓ 2 callersFunctioninit_network_weights
(model, init_type="normal", gain=0.02)
dassl/utils/torchtools.py:317
↓ 2 callersMethodis_input_domain_valid
(self, input_domains)
dassl/data/datasets/base_dataset.py:128
↓ 2 callersFunctionload_clip_to_cpu
(cfg)
trainers/hpt_plus.py:18
↓ 2 callersFunctionmixup
Mixup. Args: x1 (torch.Tensor): data with shape of (b, c, h, w). x2 (torch.Tensor): data with shape of (b, c, h, w). y1 (
dassl/modeling/ops/mixup.py:4
↓ 2 callersFunctionmulti_head_attention_forward
r""" Args: query, key, value: map a query and a set of key-value pairs to an output. See "Attention Is All You Need" for more
clip/attention.py:186
↓ 2 callersFunctionmulti_head_attention_forward
r""" Args: query, key, value: map a query and a set of key-value pairs to an output. See "Attention Is All You Need" for more
llms/clip/attention.py:186
↓ 2 callersMethodpairwise_distance
(self, x, u)
dassl/engine/da/m3sda.py:166
↓ 2 callersMethodparse_batch_train
(self, batch)
trainers/hpt_plus.py:625
↓ 2 callersFunctionpatch_device
(module)
clip/clip.py:136
↓ 2 callersFunctionpatch_device
(module)
llms/clip/clip.py:136
↓ 2 callersFunctionpatch_float
(module)
clip/clip.py:160
↓ 2 callersFunctionpatch_float
(module)
llms/clip/clip.py:160
↓ 2 callersMethodread_data
(self, classnames, split_dir)
datasets/imagenet.py:78
↓ 2 callersMethodread_data
(self, split_file)
datasets/oxford_pets.py:58
↓ 2 callersMethodread_data
(self, cname2lab, text_file)
datasets/sun397.py:61
↓ 2 callersMethodread_data
(self, cname2lab, text_file)
datasets/ucf101.py:64
↓ 2 callersMethodread_data
(self, image_dir, anno_file, meta_file)
datasets/stanford_cars.py:57
↓ 2 callersFunctionread_json
Read json file from a path.
dassl/utils/tools.py:59
↓ 2 callersMethodsave_model
( self, epoch, directory, is_best=False, val_result=None, model_name="" )
dassl/engine/trainer.py:118
↓ 2 callersMethodset_activation_status
(self, status=True)
dassl/modeling/ops/efdmix.py:81
↓ 2 callersMethodset_activation_status
(self, status=True)
dassl/modeling/ops/mixstyle.py:81
↓ 2 callersFunctionsigmoid_rampup
Exponential rampup. Args: current (int): current step. rampup_length (int): maximum step.
dassl/modeling/ops/utils.py:41
↓ 2 callersMethodstep
(self, closure=None)
dassl/optim/radam.py:50
↓ 2 callersMethodtest
A generic testing pipeline.
dassl/engine/trainer.py:451
↓ 2 callersMethodtruncate
(self, array)
trainers/hpt_plus.py:371
↓ 2 callersMethodupdate_mix_method
(self, mix="random")
dassl/modeling/ops/efdmix.py:84
↓ 2 callersMethodupdate_mix_method
(self, mix="random")
dassl/modeling/ops/mixstyle.py:84
↓ 1 callersFunctionCutoutAbs
(img, v)
dassl/data/transforms/randaugment.py:137
↓ 1 callersMethod__init__
( self, dist_metric="cosine", eps=0.01, max_iter=5, bp_to_sinkhorn=Fal
dassl/modeling/ops/optimal_transport.py:38
↓ 1 callersMethod__init__
(self, c_hidden=64)
dassl/modeling/backbone/cnn_digitsdg.py:23
↓ 1 callersMethod__init__
(self, block_args, global_params, image_size=None)
dassl/modeling/backbone/efficientnet/model.py:26
↓ 1 callersMethod__init__
(self, cfg)
dassl/data/datasets/dg/cifar_c.py:49
↓ 1 callersMethod__init__
(self, cfg)
dassl/data/datasets/ssl/cifar.py:22
↓ 1 callersMethod__init__
(self, cfg, model_cfg, num_classes, **kwargs)
dassl/engine/trainer.py:27
↓ 1 callersMethod__init__
(self, cfg)
dassl/engine/dg/daeldg.py:38
↓ 1 callersMethod__init__
(self, cfg)
dassl/engine/da/m3sda.py:33
↓ 1 callersMethod__init__
(self, cfg)
dassl/engine/da/dael.py:36
↓ 1 callersMethod__init__
(self, cfg)
dassl/engine/da/cdac.py:65
↓ 1 callersMethod__init__
(self, cfg)
dassl/engine/da/mme.py:34
↓ 1 callersMethod__init__
(self, cfg)
dassl/evaluation/evaluator.py:13
↓ 1 callersMethod__iter__
(self)
dassl/data/samplers.py:38
↓ 1 callersMethod__iter__
(self)
dassl/data/samplers.py:93
↓ 1 callersMethod__iter__
(self)
dassl/data/samplers.py:149
↓ 1 callersFunction_build_transform_test
(cfg, choices, target_size, normalize)
dassl/data/transforms/transforms.py:326
↓ 1 callersFunction_build_transform_train
(cfg, choices, target_size, normalize)
dassl/data/transforms/transforms.py:206
↓ 1 callersMethod_change_in_channels
(model, in_channels)
dassl/modeling/backbone/efficientnet/model.py:316
↓ 1 callersMethod_check_input
(self, x)
dassl/modeling/ops/transnorm.py:46
↓ 1 callersMethod_check_input
(self, x)
dassl/modeling/backbone/cnn_digitsdg.py:32
↓ 1 callersMethod_check_input
(self, x)
dassl/modeling/backbone/cnn_digit5_m3sda.py:30
↓ 1 callersMethod_check_input
(self, x)
dassl/modeling/backbone/cnn_digitsingle.py:25
↓ 1 callersMethod_decode_block_string
Gets a block through a string notation of arguments.
dassl/modeling/backbone/efficientnet/utils.py:288
↓ 1 callersFunction_download
(url: str, root: str = os.path.expanduser("~/.cache/clip"))
clip/clip.py:39
↓ 1 callersFunction_download
(url: str, root: str = os.path.expanduser("~/.cache/clip"))
llms/clip/clip.py:39
↓ 1 callersMethod_encode_block_string
Encodes a block to a string.
dassl/modeling/backbone/efficientnet/utils.py:317
↓ 1 callersMethod_forward_impl
(self, x: Tensor)
dassl/modeling/backbone/resnet_dynamic.py:523
↓ 1 callersMethod_init_params
(self)
dassl/modeling/backbone/resnet.py:171
↓ 1 callersMethod_initialize_weights
(self)
dassl/modeling/backbone/vgg.py:51
↓ 1 callersMethod_make_layer
( self, block, in_planes, out_planes, nb_layers, stride, dropRate )
dassl/modeling/backbone/wide_resnet.py:71
↓ 1 callersMethod_read_data_test
(self, data_dir)
dassl/data/datasets/ssl/cifar.py:79
↓ 1 callersMethod_read_data_train
(self, data_dir, num_labeled, val_percent)
dassl/data/datasets/ssl/cifar.py:43
↓ 1 callersMethod_read_data_train
(self, data_dir, fold, fold_file)
dassl/data/datasets/ssl/stl10.py:52
↓ 1 callersMethod_reset_parameters
(self)
clip/attention.py:94
↓ 1 callersMethod_reset_parameters
(self)
llms/clip/attention.py:94
↓ 1 callersFunction_scaled_dot_product_attention
r""" Computes scaled dot product attention on query, key and value tensors, using an optional attention mask if passed, and applying dropout i
clip/attention.py:447
↓ 1 callersFunction_scaled_dot_product_attention
r""" Computes scaled dot product attention on query, key and value tensors, using an optional attention mask if passed, and applying dropout i
llms/clip/attention.py:447
↓ 1 callersFunction_vgg
(arch, cfg, batch_norm, pretrained)
dassl/modeling/backbone/vgg.py:133
↓ 1 callersMethodafter_epoch
(self)
dassl/engine/trainer.py:263
↓ 1 callersMethodafter_train
(self)
dassl/engine/trainer.py:257
↓ 1 callersMethodassess_y_pred_quality
(self, y_pred, y_true, mask)
dassl/engine/da/cdac.py:139
↓ 1 callersMethodassess_y_pred_quality
(self, y_pred, y_true, mask)
dassl/engine/ssl/fixmatch.py:40
↓ 1 callersMethodattention
(self, x: torch.Tensor)
llms/clip/model_new.py:185
↓ 1 callersMethodattention
(self, x: torch.Tensor, attn_sp: Optional[Tensor] = None,)
llms/clip/model.py:181
↓ 1 callersFunctionavailable_models
Returns the names of available CLIP models
clip/clip.py:81
↓ 1 callersFunctionavailable_models
Returns the names of available CLIP models
llms/clip/clip.py:81
↓ 1 callersFunctionbasic_clean
(text)
clip/simple_tokenizer.py:50
← previousnext →101–200 of 862, ranked by callers