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

↓ 11 callersMethod__init__

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

models/backbones/pvt_v2.py:15

↓ 11 callersMethodinfo

↓ 8 callersMethod__init__

↓ 6 callersMethod__init__

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

models/backbones/swin_v1.py:23

↓ 5 callersFunction_prepare_data

(pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:98

↓ 5 callersMethodget_patches_batch

models/birefnet.py:200

↓ 5 callersMethodstep

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:118

↓ 4 callersFunctionpath_to_image

(path, size=(1024, 1024), color_type=['rgb', 'gray'][0])

↓ 4 callersMethodssim

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:269

↓ 3 callersMethod__init__

(self, in_channels=64, out_channels=None, output_stride=16)

models/modules/aspp.py:27

↓ 3 callersMethod__init__

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

models/modules/mlp.py:75

↓ 3 callersMethod__init__

(self, in_channels=3+1)

models/refinement/refiner.py:64

↓ 3 callersFunctionbuild_backbone

(bb_name, pretrained=True, params_settings='')

models/backbones/build_backbone.py:12

↓ 2 callersMethod__init__

(self, channels)

models/birefnet.py:150

↓ 2 callersMethod__init__

(self, in_channels, S=4, reduction=4)

models/modules/attentions.py:27

↓ 2 callersMethod__init__

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

models/modules/decoder_blocks.py:37

↓ 2 callersMethod_freeze_stages

models/backbones/swin_v1.py:564

↓ 2 callersFunction_get_adaptive_threshold

(matrix: np.ndarray, max_value: float = 1)

evaluation/metrics.py:106

↓ 2 callersMethodapproximate_RDP

(self, boundaries, epsilon=1.0)

evaluation/metrics.py:597

↓ 2 callersFunctionbuild_norm_layer

(dim, norm_layer, in_format='channels_last', ou

models/modules/utils.py:15

↓ 2 callersFunctioncheck_state_dict

(state_dict, unwanted_prefix='_orig_mod.')

↓ 2 callersFunctioncreate_window

(window_size, channel)

↓ 2 callersFunctionevaluate

(pred_dir, method, testset, only_S_MAE=False, epoch=0)

evaluation/evaluate.py:10

↓ 2 callersFunctionevaluator

(gt_paths, pred_paths, metrics=['S', 'MAE', 'E', 'F', 'WF', 'HCE'], verbose=False)

evaluation/metrics.py:15

↓ 2 callersMethodfilter_bdy_cond

(self, bdy_, mask, cond)

evaluation/metrics.py:546

↓ 2 callersMethodgenerate_parts_numel_combinations

(self, fg_fg_numel, fg_bg_numel, pred_fg_numel, pred_bg_numel)

evaluation/metrics.py:381

↓ 2 callersFunctioninference

(model, data_loader_test, pred_root, method, testset, device=0)

inference.py:18

↓ 2 callersFunctionprepare_dataloader

(dataset: torch.utils.data.Dataset, batch_size: int, to_be_distributed=False, is_train=True)

↓ 2 callersMethods_object

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:216

↓ 2 callersMethodtrain

Convert the model into training mode while keep layers freezed.

models/backbones/swin_v1.py:608

↓ 2 callersMethodupdate

(self, val, n=1)

↓ 2 callersFunctionvalid

(model, data_loader_test, pred_dir, method='tmp_val', testset='DIS-VD', only_S_MAE=True, device=0)

evaluation/valid.py:5

↓ 2 callersMethodvalidate_model

↓ 2 callersFunctionwindow_partition

Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C)

models/backbones/swin_v1.py:41

↓ 1 callersFunctionSSIM

↓ 1 callersMethod__init__

models/modules/utils.py:50

↓ 1 callersMethod_load_adv_components

↓ 1 callersFunction_ssim

(img1, img2, window, window_size, channel, size_average=True)

↓ 1 callersMethod_train_batch

↓ 1 callersFunctionbuild_act_layer

models/modules/utils.py:4

↓ 1 callersMethodcal_adaptive_em

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:311

↓ 1 callersMethodcal_adaptive_fm

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:129

↓ 1 callersMethodcal_changeable_em

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:316

↓ 1 callersMethodcal_em_with_cumsumhistogram

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:349

↓ 1 callersMethodcal_em_with_threshold

(self, pred: np.ndarray, gt: np.ndarray, threshold: float)

evaluation/metrics.py:320

↓ 1 callersMethodcal_hce

(self, pred: np.ndarray, gt: np.ndarray, gt_ske: np.ndarray, relax=5, epsilon=2.0)

evaluation/metrics.py:491

↓ 1 callersMethodcal_mae

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:178

↓ 1 callersMethodcal_pr

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:141

↓ 1 callersMethodcal_sm

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:198

↓ 1 callersMethodcal_wfm

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:423

↓ 1 callersMethodcentroid

(self, matrix: np.ndarray)

evaluation/metrics.py:235

↓ 1 callersMethodclose

↓ 1 callersFunctioncolor_enhance

↓ 1 callersFunctioncv_random_flip

↓ 1 callersMethoddivide_with_xy

(self, pred: np.ndarray, gt: np.ndarray, x, y)

evaluation/metrics.py:246

↓ 1 callersFunctiondo_eval

eval_existingOnes.py:16

↓ 1 callersMethodforward_enc

models/birefnet.py:66

↓ 1 callersMethodforward_features

models/backbones/pvt_v2.py:314

↓ 1 callersMethodforward_ori

models/birefnet.py:100

↓ 1 callersFunctiongaussian

(window_size, sigma)

↓ 1 callersFunctiongaussianNoisy

(im, mean=mean, sigma=sigma)

↓ 1 callersMethodget_results

evaluation/metrics.py:159

↓ 1 callersMethodget_results

evaluation/metrics.py:182

↓ 1 callersMethodget_results

evaluation/metrics.py:291

↓ 1 callersMethodget_results

evaluation/metrics.py:403

↓ 1 callersMethodget_results

evaluation/metrics.py:471

↓ 1 callersMethodget_results

evaluation/metrics.py:486

↓ 1 callersFunctioninit_data_loaders

(to_be_distributed)

↓ 1 callersFunctioninit_models_optimizers

(epochs, to_be_distributed)

↓ 1 callersFunctionload_weights

(model, model_name)

models/backbones/build_backbone.py:28

↓ 1 callersFunctionmain

↓ 1 callersFunctionmain

waiting4eval.py:75

↓ 1 callersFunctionmain

inference.py:44

↓ 1 callersFunctionmain

evaluation/evaluate.py:50

↓ 1 callersMethodmatlab_style_gauss2D

2D gaussian mask - should give the same result as MATLAB's fspecial('gaussian',[shape],[sigma])

evaluation/metrics.py:457

↓ 1 callersMethodobject

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:209

↓ 1 callersFunctionpreproc

(image, label, preproc_methods=['flip'])

↓ 1 callersMethodprint_task

↓ 1 callersFunctionrandom_crop

↓ 1 callersFunctionrandom_pepper

(img, N=0.0015)

↓ 1 callersFunctionrandom_rotate

(image, label, angle=15)

↓ 1 callersMethodregion

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:222

↓ 1 callersMethodreset

↓ 1 callersFunctionsave_tensor_img

(tenor_im, path)

↓ 1 callersFunctionset_seed

↓ 1 callersMethodstep

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:172

↓ 1 callersMethodstep

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:192

↓ 1 callersMethodstep

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:301

↓ 1 callersMethodstep

(self, pred: np.ndarray, gt: np.ndarray)

evaluation/metrics.py:414

↓ 1 callersMethodstep

(self, pred: np.ndarray, gt: np.ndarray, gt_ske)

evaluation/metrics.py:480

↓ 1 callersMethodtrain_epoch

↓ 1 callersFunctionvalidate_model

(model, test_loaders, epoch)

waiting4eval.py:36

↓ 1 callersFunctionwindow_reverse

Args: windows: (num_windows*B, window_size, window_size, C) window_size (int): Window size H (int): Height of image

models/backbones/swin_v1.py:56

Method__getitem__

(self, channels=1, img_size=256)

next →1–100 of 221, ranked by callers