Functions — github.com/ZhengPeng7/BiRefNet

CodeHub MCPcopy Create free account

hub / github.com/ZhengPeng7/BiRefNet / functions

Functions221 in github.com/ZhengPeng7/BiRefNet

⨍ Functions 221◇ Types & classes 66

(self, window_size=11, size_average=True)

( self, data_loaders, model_opt_lrsch, )

(self, path="log.txt")

(self, datasets, image_size, is_train=True)

(self, bb_pretrained=True)

models/birefnet.py:23

( self, in_channels: int, out_channels: int, inter_channels=64 )

models/birefnet.py:309

(self, channels, reduction=16)

models/modules/attentions.py:8

models/modules/attentions.py:67

(self, in_channels=64, out_channels=64, inter_channels=64, mlp_ratio=4.)

models/modules/ing.py:6

models/modules/utils.py:41

(self, in_channels, planes, kernel_size, padding, dilation)

models/modules/aspp.py:12

(self, in_channels, planes, kernel_size, padding)

models/modules/aspp.py:72

(self, in_channels, out_channels=None, parallel_block_sizes=[1, 3, 7])

models/modules/aspp.py:87

(self, in_channels=64, out_channels=64, inter_channels=64)

models/modules/lateral_blocks.py:14

(self, in_channels, out_channels, kernel_size=3,

models/modules/deform_conv.py:7

(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.)

models/modules/mlp.py:12

(self, dim, num_heads=8, qkv_bias=False, qk_scale=None, attn_drop=0., proj_drop=0., sr_ratio=1)

models/modules/mlp.py:31

(self, img_size=224, patch_size=7, stride=4, in_channels=3, embed_dim=768)

models/modules/mlp.py:99

(self, in_channels=64, out_channels=64, inter_channels=64)

models/modules/decoder_blocks.py:12

(self, in_channels=64, out_channels=None, inter_channels=64)

models/modules/decoder_blocks.py:70

(self, dim, window_size, num_heads, qkv_bias=True, qk_scale=None, attn_drop=0., proj_drop=0.)

models/backbones/swin_v1.py:87

(self, dim, num_heads, window_size=7, shift_size=0, mlp_ratio=4., qkv_bias=True, qk_scale=Non

models/backbones/swin_v1.py:182

(self, dim, norm_layer=nn.LayerNorm)

models/backbones/swin_v1.py:273

(self, dim, depth, num_heads, window_size=

models/backbones/swin_v1.py:328

(self, patch_size=4, in_channels=3, embed_dim=96, norm_layer=None)

models/backbones/swin_v1.py:423

(self, pretrain_img_size=224, patch_size=4, in_channels=3,

models/backbones/swin_v1.py:485

(self, dim, num_heads=8, qkv_bias=False, qk_scale=None, attn_drop=0., proj_drop=0., sr_ratio=1)

models/backbones/pvt_v2.py:53

(self, dim, num_heads, mlp_ratio=4., qkv_bias=False, qk_scale=None, drop=0., attn_drop=0., dr

models/backbones/pvt_v2.py:123

(self, img_size=224, patch_size=7, stride=4, in_channels=3, embed_dim=768)

models/backbones/pvt_v2.py:165

(self, img_size=224, patch_size=16, in_channels=3, num_classes=1000, embed_dims=[64, 128, 256, 512],

models/backbones/pvt_v2.py:205

(self, dim=768)

models/backbones/pvt_v2.py:362

(self, **kwargs)

models/backbones/pvt_v2.py:388

(self, **kwargs)

models/backbones/pvt_v2.py:398

(self, in_channels=3, **kwargs)

models/backbones/pvt_v2.py:406

(self, **kwargs)

models/backbones/pvt_v2.py:414

(self, **kwargs)

models/backbones/pvt_v2.py:422

(self, **kwargs)

models/backbones/pvt_v2.py:431

(self, in_channels=3+1)

models/refinement/refiner.py:20

(self, channels)

models/refinement/refiner.py:110

(self, in_channels=3+1)

models/refinement/refiner.py:160

(self, in_channels=3+1, inter_channels=48, out_channels=96,

models/refinement/stem_layer.py:14

(self, beta: float = 0.3)

evaluation/metrics.py:111

evaluation/metrics.py:169

(self, alpha: float = 0.5)

evaluation/metrics.py:188

evaluation/metrics.py:297

(self, beta: float = 1)

evaluation/metrics.py:410

evaluation/metrics.py:477

Function_conv_filter

convert patch embedding weight from manual patchify + linear proj to conv

models/backbones/pvt_v2.py:375

Method_init_weights

models/backbones/pvt_v2.py:27

Method_init_weights

models/backbones/pvt_v2.py:76

Method_init_weights

models/backbones/pvt_v2.py:139

Method_init_weights

models/backbones/pvt_v2.py:180

Method_init_weights

models/backbones/pvt_v2.py:262

Methoddiscriminator_block

(in_filters, out_filters, bn=Config().batch_size > 1)

target, pred: tensor of shape (B, C, H, W), where target[:,:,region_in_contour] == 1, target[:,:,region_out_contour]

(self, pred, target)

(self, pred, target)

(self, pred, target)

(self, pred, gt=None)

(self, preds, gt)

(self, scaled_preds, gt)

(self, img1, img2)

models/birefnet.py:143

(self, features)

models/birefnet.py:212

models/birefnet.py:316

models/modules/attentions.py:16

models/modules/attentions.py:40

models/modules/attentions.py:75

models/modules/ing.py:17

models/modules/utils.py:44

models/modules/utils.py:53

models/modules/aspp.py:19

models/modules/aspp.py:54

models/modules/aspp.py:79

models/modules/aspp.py:108

models/modules/lateral_blocks.py:19

models/modules/deform_conv.py:50

models/modules/mlp.py:21

(self, x, H, W)

models/modules/mlp.py:51

(self, x, H, W)

models/modules/mlp.py:89

models/modules/mlp.py:112

models/modules/decoder_blocks.py:25

models/modules/decoder_blocks.py:57

models/modules/decoder_blocks.py:89

models/backbones/swin_v1.py:32

Forward function. Args: x: input features with shape of (num_windows*B, N, C) mask: (0/-inf) mask with shape of (num

models/backbones/swin_v1.py:122

Forward function. Args: x: Input feature, tensor size (B, H*W, C). H, W: Spatial resolution of the input feature.

models/backbones/swin_v1.py:206

Forward function. Args: x: Input feature, tensor size (B, H*W, C). H, W: Spatial resolution of the input feature.

models/backbones/swin_v1.py:279

Forward function. Args: x: Input feature, tensor size (B, H*W, C). H, W: Spatial resolution of the input feature.

models/backbones/swin_v1.py:370

Forward function.

models/backbones/swin_v1.py:437

Forward function.

models/backbones/swin_v1.py:582

(self, x, H, W)

models/backbones/pvt_v2.py:42

(self, x, H, W)

models/backbones/pvt_v2.py:91

(self, x, H, W)

models/backbones/pvt_v2.py:154

models/backbones/pvt_v2.py:195

← previous next →101–200 of 221, ranked by callers