Functions — github.com/ChongQingNoSubway/SelfReg-UNet

hub / github.com/ChongQingNoSubway/SelfReg-UNet / functions

Functions3,840 in github.com/ChongQingNoSubway/SelfReg-UNet

⨍ Functions 3,840◇ Types & classes 731

↓ 1 callersFunctiongrid_partition_nchw

(x, grid_size: List[int])

src/train_acdc/lib/maxxvit_4out.py:1322

↓ 1 callersFunctiongrid_partition_nchw

(x, grid_size: List[int])

src/train_acdc/lib/models_timm/maxxvit.py:1322

↓ 1 callersFunctiongrid_reverse_nchw

(windows, grid_size: List[int], img_size: List[int])

src/train_acdc/lib/maxxvit_4out.py:1332

↓ 1 callersFunctiongrid_reverse_nchw

(windows, grid_size: List[int], img_size: List[int])

src/train_acdc/lib/models_timm/maxxvit.py:1332

↓ 1 callersFunctionhard_mish

Hard Mish Experimental, based on notes by Mish author Diganta Misra at https://github.com/digantamisra98/H-Mish/blob/0da20d4bc58e696b6803f2

src/train_acdc/lib/models_timm/layers/activations.py:104

↓ 1 callersFunctionhard_mish_jit

Hard Mish Experimental, based on notes by Mish author Diganta Misra at https://github.com/digantamisra98/H-Mish/blob/0da20d4bc58e696b6803f2

src/train_acdc/lib/models_timm/layers/activations_jit.py:77

↓ 1 callersFunctionhard_mish_jit_bwd

(x, grad_output)

src/train_acdc/lib/models_timm/layers/activations_me.py:184

↓ 1 callersFunctionhard_mish_jit_fwd

(x)

src/train_acdc/lib/models_timm/layers/activations_me.py:179

↓ 1 callersFunctionhard_sigmoid

(x, inplace: bool = False)

src/train_acdc/lib/models_timm/layers/activations.py:88

↓ 1 callersFunctionhard_sigmoid_jit

(x, inplace: bool = False)

src/train_acdc/lib/models_timm/layers/activations_jit.py:49

↓ 1 callersFunctionhard_sigmoid_jit_bwd

(x, grad_output)

src/train_acdc/lib/models_timm/layers/activations_me.py:105

↓ 1 callersFunctionhard_sigmoid_jit_fwd

(x, inplace: bool = False)

src/train_acdc/lib/models_timm/layers/activations_me.py:100

↓ 1 callersFunctionhard_swish

(x, inplace: bool = False)

src/train_acdc/lib/models_timm/layers/activations.py:74

↓ 1 callersFunctionhard_swish_jit

(x, inplace: bool = False)

src/train_acdc/lib/models_timm/layers/activations_jit.py:63

↓ 1 callersFunctionhard_swish_jit_bwd

(x, grad_output)

src/train_acdc/lib/models_timm/layers/activations_me.py:140

↓ 1 callersFunctionhard_swish_jit_fwd

(x)

src/train_acdc/lib/models_timm/layers/activations_me.py:135

↓ 1 callersFunctionhf_split

(hf_id)

src/train_acdc/lib/models_timm/hub.py:65

↓ 1 callersFunctioninference

(args, model, test_save_path=None)

src/train_synase/test.py:70

↓ 1 callersFunctioninference

(args, model, test_save_path=None)

src/train_synase/test-every.py:70

↓ 1 callersFunctioninference

(args, model, test_save_path=None)

src/train_synase/test_unet.py:71

↓ 1 callersMethodinit_weights

(self, zero_init_last=True)

src/train_acdc/lib/models_timm/resnetv2.py:403

↓ 1 callersMethodinit_weights

(self)

src/train_acdc/lib/models_timm/hrnet.py:675

↓ 1 callersMethodinit_weights

(self, zero_init_last: bool = False)

src/train_acdc/lib/models_timm/byobnet.py:1087

↓ 1 callersMethodinit_weights

(self, mode='')

src/train_acdc/lib/models_timm/swin_transformer.py:511

↓ 1 callersMethodinit_weights

(self, mode='')

src/train_acdc/lib/models_timm/nest.py:321

↓ 1 callersMethodinit_weights

(self)

src/train_acdc/lib/models_timm/vision_transformer_relpos.py:224

↓ 1 callersMethodinit_weights

(self)

src/train_acdc/lib/models_timm/vision_transformer_relpos.py:323

↓ 1 callersMethodinit_weights

(self, mode='')

src/train_acdc/lib/models_timm/vision_transformer_relpos.py:449

↓ 1 callersMethodinit_weights

(self, zero_init_last=True)

src/train_acdc/lib/models_timm/resnet.py:679

↓ 1 callersMethodinit_weights

(self, mode='')

src/train_acdc/lib/models_timm/vision_transformer.py:474

↓ 1 callersMethodinit_weights

(self, mode='')

src/train_acdc/lib/models_timm/deit.py:130

↓ 1 callersFunctioninit_weights_vit_timm

ViT weight initialization, original timm impl (for reproducibility)

src/train_acdc/lib/models_timm/vision_transformer.py:554

↓ 1 callersFunctioninstance_std

(x, eps: float = 1e-5)

src/train_acdc/lib/models_timm/layers/evo_norm.py:36

↓ 1 callersFunctionis_main_process

()

src/train_acdc/levit/utils.py:208

↓ 1 callersFunctionis_main_process

()

src/train_synase/levit/utils.py:208

↓ 1 callersFunctionis_model

Check if a model name exists

src/train_acdc/lib/models_timm/registry.py:99

↓ 1 callersFunctionis_static_pad

(kernel_size: int, stride: int = 1, dilation: int = 1, **_)

src/train_acdc/lib/models_timm/layers/padding.py:23

↓ 1 callersFunctionload_cfg_from_json

(json_file: Union[str, os.PathLike])

src/train_acdc/lib/models_timm/hub.py:74

↓ 1 callersFunctionload_checkpoint

(model, checkpoint_path, use_ema=True, strict=True, remap=False)

src/train_acdc/lib/models_timm/helpers.py:66

↓ 1 callersFunctionload_custom_pretrained

r"""Loads a custom (read non .pth) weight file Downloads checkpoint file into cache-dir like torch.hub based loaders, but calls a passed in c

src/train_acdc/lib/models_timm/helpers.py:177

↓ 1 callersMethodload_from

(self, config)

src/train_synase/networks/vision_transformer.py:53

↓ 1 callersFunctionload_model_config_from_hf

(model_id: str)

src/train_acdc/lib/models_timm/hub.py:85

↓ 1 callersFunctionload_pretrained

Load pretrained checkpoint Args: model (nn.Module) : PyTorch model module pretrained_cfg (Optional[Dict]): configuration for pre

src/train_acdc/lib/models_timm/helpers.py:240

↓ 1 callersMethodlocal_init

(self)

src/train_acdc/lib/models_timm/convit.py:119

↓ 1 callersFunctionmake_blocks

( block_fn, channels, block_repeats, inplanes, reduce_first=1, output_stride=32, down_kernel_s

src/train_acdc/lib/models_timm/resnet.py:508

↓ 1 callersFunctionmake_safe

(name)

src/train_acdc/lib/models_timm/factory.py:23

↓ 1 callersMethodmedian

(self)

src/train_acdc/levit/utils.py:51

↓ 1 callersFunctionmish

Mish: A Self Regularized Non-Monotonic Neural Activation Function - https://arxiv.org/abs/1908.08681 NOTE: I don't have a working inplace variant

src/train_acdc/lib/models_timm/layers/activations.py:29

↓ 1 callersFunctionmish_jit

Mish: A Self Regularized Non-Monotonic Neural Activation Function - https://arxiv.org/abs/1908.08681

src/train_acdc/lib/models_timm/layers/activations_jit.py:26

↓ 1 callersFunctionmish_jit_bwd

(x, grad_output)

src/train_acdc/lib/models_timm/layers/activations_me.py:66

↓ 1 callersFunctionmish_jit_fwd

(x)

src/train_acdc/lib/models_timm/layers/activations_me.py:61

↓ 1 callersFunctionmnasnet_100

MNASNet B1, depth multiplier of 1.0.

src/train_acdc/lib/models_timm/efficientnet.py:1251

↓ 1 callersFunctionmodel_entrypoint

Fetch a model entrypoint for specified model name

src/train_acdc/lib/models_timm/registry.py:105

↓ 1 callersFunctionmodel_factory_v3_NM

(C, D, X, N, drop_path, weights, num_classes, distillation, pretrained, fuse)

src/train_acdc/levit/LeViTUNet128s.py:388

↓ 1 callersFunctionmodel_factory_v3_NM

(C, D, X, N, drop_path, weights, num_classes, distillation, pretrained, fuse)

src/train_synase/levit/LeViTUNet128s.py:388

↓ 1 callersMethodmodel_init

(self)

src/train_acdc/lib/pyramid_vig.py:151

↓ 1 callersMethodmodule_name

feature module name accessor

src/train_acdc/lib/models_timm/features.py:72

↓ 1 callersFunctionnamed_modules_with_params

(module: nn.Module, name='', depth_first=True, include_root=False)

src/train_acdc/lib/models_timm/helpers.py:627

↓ 1 callersFunctionnetParams

computing total network parameters args: model: model return: the number of parameters

src/train_acdc/levit/LeViTUNet128s.py:629

↓ 1 callersFunctionnetParams

computing total network parameters args: model: model return: the number of parameters

src/train_synase/levit/LeViTUNet128s.py:629

↓ 1 callersFunctionnum_groups

(group_size, channels)

src/train_acdc/lib/maxxvit_4out.py:949

↓ 1 callersFunctionnum_groups

(group_size, channels)

src/train_acdc/lib/models_timm/maxxvit.py:949

↓ 1 callersFunctionoutlooker_blocks

generate outlooker layer in stage1 return: outlooker layers

src/train_acdc/lib/models_timm/volo.py:345

↓ 1 callersFunctionpairwise_distance

Compute pairwise distance of a point cloud. Args: x: tensor (batch_size, num_points, num_dims) Returns: pairwise distance

src/train_acdc/lib/gcn_lib/torch_edge.py:9

↓ 1 callersFunctionpart_pairwise_distance

Compute pairwise distance of a point cloud. Args: x: tensor (batch_size, num_points, num_dims) Returns: pairwise distance

src/train_acdc/lib/gcn_lib/torch_edge.py:23

↓ 1 callersFunctionquantize_float

Converts a float to closest non-zero int divisible by q.

src/train_acdc/lib/models_timm/regnet.py:173

↓ 1 callersFunctionrand_bbox

get bounding box as token labeling (https://github.com/zihangJiang/TokenLabeling) return: bounding box

src/train_acdc/lib/models_timm/volo.py:271

↓ 1 callersFunctionrandom_rot_flip

(image, label)

src/train_acdc/utils/dataset_synapse.py:12

↓ 1 callersFunctionrandom_rot_flip

(image, label)

src/train_synase/utils/dataset_synapse.py:12

↓ 1 callersFunctionrandom_rotate

(image, label)

src/train_acdc/utils/dataset_synapse.py:22

↓ 1 callersFunctionrandom_rotate

(image, label)

src/train_synase/utils/dataset_synapse.py:22

↓ 1 callersFunctionrel_pos_indices

(size)

src/train_acdc/lib/models_timm/layers/lambda_layer.py:31

↓ 1 callersFunctionremap_checkpoint

remap checkpoint by iterating over state dicts in order (ignoring original keys). This assumes models (and originating state dict) were created w

src/train_acdc/lib/models_timm/helpers.py:81

↓ 1 callersMethodreset

(self)

src/train_acdc/utils/misc.py:87

↓ 1 callersMethodreset

(self)

src/train_acdc/utils/utils.py:60

↓ 1 callersMethodreset

(self)

src/train_synase/utils/misc.py:87

↓ 1 callersMethodreset

(self)

src/train_synase/utils/utils.py:60

↓ 1 callersMethodreset_parameters

(self)

src/train_acdc/lib/models_timm/layers/non_local_attn.py:57

↓ 1 callersMethodreset_parameters

(self)

src/train_acdc/lib/models_timm/layers/bottleneck_attn.py:130

↓ 1 callersMethodreset_parameters

(self)

src/train_acdc/lib/models_timm/layers/inplace_abn.py:74

↓ 1 callersMethodreset_parameters

(self)

src/train_acdc/lib/models_timm/layers/global_context.py:43

↓ 1 callersMethodreset_parameters

(self)

src/train_acdc/lib/models_timm/layers/lambda_layer.py:102

↓ 1 callersMethodreset_parameters

(self)

src/train_acdc/lib/models_timm/layers/evo_norm.py:111

↓ 1 callersMethodreset_parameters

(self)

src/train_acdc/lib/models_timm/layers/evo_norm.py:149

↓ 1 callersMethodreset_parameters

(self)

src/train_acdc/lib/models_timm/layers/evo_norm.py:184

↓ 1 callersMethodreset_parameters

(self)

src/train_acdc/lib/models_timm/layers/evo_norm.py:223

↓ 1 callersMethodreset_parameters

(self)

src/train_acdc/lib/models_timm/layers/evo_norm.py:278

↓ 1 callersMethodreset_parameters

(self)

src/train_acdc/lib/models_timm/layers/evo_norm.py:327

↓ 1 callersMethodreset_parameters

(self)

src/train_acdc/lib/models_timm/layers/cond_conv2d.py:73

↓ 1 callersMethodreset_parameters

(self)

src/train_acdc/lib/models_timm/layers/filter_response_norm.py:30

↓ 1 callersMethodreset_parameters

(self)

src/train_acdc/lib/models_timm/layers/filter_response_norm.py:58

↓ 1 callersMethodreset_parameters

(self)

src/train_acdc/lib/gcn_lib/torch_nn.py:70

↓ 1 callersFunctionresize_pos_embed

Rescale the grid of position embeddings when loading from state_dict Expected shape of position embeddings is (1, T, N, C), and considers onl

src/train_acdc/lib/models_timm/nest.py:395

↓ 1 callersMethodrgb_loader

(self, path)

src/train_acdc/utils/dataloader.py:180

↓ 1 callersMethodrgb_loader

(self, path)

src/train_synase/utils/dataloader.py:180

↓ 1 callersFunctionround_channels

Round number of filters based on depth multiplier.

src/train_acdc/lib/models_timm/efficientnet_builder.py:57

↓ 1 callersFunctionsave_for_hf

(model, save_directory, model_config=None)

src/train_acdc/lib/models_timm/hub.py:102

↓ 1 callersFunctionscale_image

Pulled out of CrossViT.forward_features to bury conditional logic in a leaf node for FX tracing. Args: x (Tensor): input image

src/train_acdc/lib/models_timm/crossvit.py:265

↓ 1 callersFunctionsemnasnet_100

MNASNet A1 (w/ SE), depth multiplier of 1.0.

src/train_acdc/lib/models_timm/efficientnet.py:1285

↓ 1 callersFunctionseresnext26t_32x4d

Constructs a SE-ResNet-26-T model. This is technically a 28 layer ResNet, like a 'D' bag-of-tricks model but with tiered 24, 32, 64 channels i

src/train_acdc/lib/models_timm/resnet.py:1373

← previous next →801–900 of 3,840, ranked by callers