Functions — github.com/UX-Decoder/Semantic-SAM

FunctionPYBIND11_MODULE

semantic_sam/body/encoder/ops/src/vision.cpp:18

Method__call__

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

datasets/dataset_mappers/lvis_dataset_mapper.py:126

Method__call__

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

datasets/dataset_mappers/sam_baseline_dataset_mapper_json.py:132

Method__call__

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

datasets/dataset_mappers/part_data_filter_whole_new_instance_dataset_mapper.py:165

Method__call__

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

datasets/dataset_mappers/pascal_instance_new_baseline_dataset_mapper.py:144

Method__call__

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

datasets/dataset_mappers/sam_baseline_dataset_mapper.py:187

Method__call__

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

datasets/dataset_mappers/mask_former_semantic_dataset_mapper.py:104

Method__call__

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

datasets/dataset_mappers/imagenet_dataset_mapper.py:75

Method__call__

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

datasets/dataset_mappers/mask_former_panoptic_dataset_mapper.py:61

Method__call__

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

datasets/dataset_mappers/mask_former_interactive_panoptic_dataset_mapper.py:65

Method__call__

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

datasets/dataset_mappers/o365_instance_new_baseline_dataset_mapper.py:127

Method__call__

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

datasets/dataset_mappers/coco_panoptic_new_baseline_dataset_mapper.py:108

Method__call__

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

datasets/dataset_mappers/inference_mapper_with_gt.py:145

Method__call__

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

datasets/dataset_mappers/coco_interactive_panoptic_new_baseline_dataset_mapper.py:129

Method__call__

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

datasets/dataset_mappers/coco_instance_new_baseline_dataset_mapper.py:123

Method__call__

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

datasets/dataset_mappers/mask_former_instance_dataset_mapper.py:112

Method__delitem__

(self, key: str)

utils/sam_utils/amg.py:35

Method__getitem__

(self, key: str)

utils/sam_utils/amg.py:38

Method__init__

(self, cfg)

train_net.py:62

Method__init__

(self, **kwargs)

demo_auto_generation.py:73

Method__init__

(self, **kwargs)

demo.py:73

Method__init__

(self)

utils/misc.py:46

Method__init__

(self, mask_or_polygons, height, width)

utils/visualizer.py:67

Method__init__

(self, panoptic_seg, segments_info, metadata=None)

utils/visualizer.py:160

Method__init__

Args: img (ndarray): an RGB image of shape (H, W, 3) in range [0, 255]. scale (float): scale the input image

utils/visualizer.py:258

Method__init__

Args: img_rgb: a numpy array of shape (H, W, C), where H and W correspond to the height and width of the image re

utils/visualizer.py:357

Method__init__

(self, target_length: int)

utils/sam_utils/transforms.py:23

Method__init__

( self, model: Sam, return_single_mask: bool, use_stability_score: bool = Fals

utils/sam_utils/onnx.py:25

Method__init__

(self, **kwargs)

utils/sam_utils/amg.py:22

Method__init__

thresh: iou thresh to filter low confidence objects text_size: resize the input image short edge for the model to process hol

tasks/interactive_predictor.py:12

Method__init__

Using a SAM model, generates masks for the entire image. Generates a grid of point prompts over the image, then filters low q

tasks/automatic_mask_generator.py:54

Method__init__

(self, opt, module: nn.Module)

semantic_sam/BaseModel.py:13

Method__init__

(self, dim, proj_drop=0., focal_level=2, focal_window=7, focal_factor=2, use_postln=False, use_postln_in_modul

semantic_sam/backbone/focal_dw.py:56

Method__init__

(self, dim, mlp_ratio=4., drop=0., drop_path=0., act_layer=nn.GELU, norm_layer=nn.LayerNorm,

semantic_sam/backbone/focal_dw.py:132

Method__init__

(self, dim, depth, mlp_ratio=4., drop=0.,

semantic_sam/backbone/focal_dw.py:223

Method__init__

(self, patch_size=4, in_chans=3, embed_dim=96, norm_layer=None, use_conv_embed=False, is_stem=False, use_pre_n

semantic_sam/backbone/focal_dw.py:380

Method__init__

(self, pretrain_img_size=1600, patch_size=4, in_chans=3,

semantic_sam/backbone/focal_dw.py:458

Method__init__

(self, cfg, input_shape)

semantic_sam/backbone/focal_dw.py:696

Method__init__

( self, dim, window_size, num_heads, qkv_bias=True, qk_scale=N

semantic_sam/backbone/swin_new.py:87

Method__init__

( self, dim, num_heads, window_size=7, shift_size=0, mlp_ratio

semantic_sam/backbone/swin_new.py:191

Method__init__

(self, dim, norm_layer=nn.LayerNorm)

semantic_sam/backbone/swin_new.py:305

Method__init__

( self, dim, depth, num_heads, window_size=7, mlp_ratio=4.0,

semantic_sam/backbone/swin_new.py:358

Method__init__

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

semantic_sam/backbone/swin_new.py:465

Method__init__

( self, pretrain_img_size=224, patch_size=4, in_chans=3, embed_dim=96,

semantic_sam/backbone/swin_new.py:526

Method__init__

(self, cfg, input_shape)

semantic_sam/backbone/swin_new.py:688

Method__init__

(self, dim, proj_drop=0., focal_level=2, focal_window=7, focal_factor=2, use_postln=False, use_postln_in_modul

semantic_sam/backbone/focal.py:56

Method__init__

(self, dim, mlp_ratio=4., drop=0., drop_path=0., act_layer=nn.GELU, norm_layer=nn.LayerNorm,

semantic_sam/backbone/focal.py:132

Method__init__

(self, dim, depth, mlp_ratio=4., drop=0.,

semantic_sam/backbone/focal.py:214

Method__init__

(self, patch_size=4, in_chans=3, embed_dim=96, norm_layer=None, use_conv_embed=False, is_stem=False)

semantic_sam/backbone/focal.py:299

Method__init__

(self, pretrain_img_size=1600, patch_size=4, in_chans=3,

semantic_sam/backbone/focal.py:364

Method__init__

(self, cfg, input_shape)

semantic_sam/backbone/focal.py:599

Method__init__

( self, dim, window_size, num_heads, qkv_bias=True, qk_scale=N

semantic_sam/backbone/swin.py:92

Method__init__

( self, dim, num_heads, window_size=7, shift_size=0, mlp_ratio

semantic_sam/backbone/swin.py:197

Method__init__

(self, dim, norm_layer=nn.LayerNorm)

semantic_sam/backbone/swin.py:316

Method__init__

( self, dim, depth, num_heads, window_size=7, mlp_ratio=4.0,

semantic_sam/backbone/swin.py:369

Method__init__

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

semantic_sam/backbone/swin.py:476

Method__init__

( self, pretrain_img_size=224, patch_size=4, in_chans=3, embed_dim=96,

semantic_sam/backbone/swin.py:537

Method__init__

(self, cfg, pretrain_img_size, patch_size, in_chans, embed_dim, depths, num_heads, window_si

semantic_sam/backbone/swin.py:770

Method__init__

The `__init__` method of any subclass can specify its own set of arguments.

semantic_sam/backbone/backbone.py:16

Method__init__

Create the criterion. Parameters: num_classes: number of object categories, omitting the special no-object category ma

semantic_sam/modules/criterion_interactive_many_to_many.py:136

Method__init__

Create the criterion. Parameters: num_classes: number of object categories, omitting the special no-object category ma

semantic_sam/modules/criterion_interactive_many_to_one.py:131

Method__init__

Creates the matcher Params: cost_class: This is the relative weight of the classification error in the matching cost

semantic_sam/modules/many2many_matcher.py:83

Method__init__

Creates the matcher Params: cost_class: This is the relative weight of the classification error in the matching cost

semantic_sam/modules/matcher.py:85

Method__init__

(self, embed_dim, num_heads, dropout=0., bias=True, add_bias_kv=False, add_zero_attn=False, kdim=None, vdim=No

semantic_sam/modules/attention.py:364

Method__init__

(self, num_pos_feats=64, temperature=10000, normalize=False, scale=None)

semantic_sam/modules/position_encoding.py:18

Method__init__

(self, tensors, mask: Optional[Tensor])

semantic_sam/utils/misc.py:43

Method__init__

( self, tokenizer, tokenizer_type, lang_encoder, lang_projection,

semantic_sam/language/vlpencoder.py:22

Method__init__

( self, tokenizer, tokenizer_type, lang_encoder, lang_projection,

semantic_sam/language/encoder.py:16

Method__init__

(self, d_model: int, n_head: int, attn_mask: torch.Tensor =

semantic_sam/language/LangEncoder/transformer.py:45

Method__init__

(self, context_length: int, vocab_size: int, width: int,

semantic_sam/language/LangEncoder/transformer.py:82

Method__init__

Args: backbone: a backbone module, must follow detectron2's backbone interface sem_seg_head: a module that predicts s

semantic_sam/architectures/interactive_mask_dino.py:33

Method__init__

(self, encoder_layer, num_layers, norm=None)

semantic_sam/body/transformer_blocks.py:79

Method__init__

(self, decoder_layer, num_layers, norm=None, return_intermediate=False)

semantic_sam/body/transformer_blocks.py:106

Method__init__

( self, d_model, nhead, dim_feedforward=2048, dropout=0.1, act

semantic_sam/body/transformer_blocks.py:155

Method__init__

( self, d_model, nhead, dim_feedforward=2048, dropout=0.1, act

semantic_sam/body/transformer_blocks.py:232

Method__init__

Args: input_shape: shapes (channels and stride) of the input features num_classes: number of classes to predict

semantic_sam/body/general_head.py:21

Method__init__

(self, d_model=256, nhead=8, num_encoder_layers=6, dim_feedforward=1024, dropout=0.1,

semantic_sam/body/encoder/encoder_deform.py:30

Method__init__

(self, d_model=256, d_ffn=1024, dropout=0.1, activation="relu",

semantic_sam/body/encoder/encoder_deform.py:104

Method__init__

(self, encoder_layer, num_layers)

semantic_sam/body/encoder/encoder_deform.py:146

Method__init__

( self, d_model=512, nhead=8, num_encoder_layers=6, dim_feedforward=20

semantic_sam/body/encoder/transformer_encoder_fpn.py:152

Method__init__

NOTE: this interface is experimental. Args: input_shape: shapes (channels and stride) of the input features t

semantic_sam/body/encoder/transformer_encoder_fpn.py:195

Method__init__

Multi-Scale Deformable Attention Module :param d_model hidden dimension :param n_levels number of feature levels

semantic_sam/body/encoder/ops/modules/ms_deform_attn.py:35

Method__init__

NOTE: this interface is experimental. Args: in_channels: channels of the input features mask_classification:

semantic_sam/body/decoder/interactive_mask_dino.py:26

Method__init__

(self, d_model, nhead, dropout=0.0, activation="relu", normalize_before=False)

semantic_sam/body/decoder/modules.py:14

Method__init__

(self, d_model, nhead, dropout=0.0, activation="relu", normalize_before=False)

semantic_sam/body/decoder/modules.py:72

Method__init__

(self, d_model, dim_feedforward=2048, dropout=0.0, activation="relu", normalize_before=False)

semantic_sam/body/decoder/modules.py:134

Method__init__

(self, input_dim, hidden_dim, output_dim, num_layers)

semantic_sam/body/decoder/utils/utils.py:14

Method__init__

(self, d_model=256, d_ffn=1024, dropout=0.1, activation="relu", n_levels=4,

semantic_sam/body/decoder/utils/dino_decoder.py:194

Method__init__

(self, loaders, key_dataset)

datasets/build.py:63

Method__init__

(self, tsv_file, silence=True)

datasets/utils/tsv/tsv_io.py:13

Method__init__

(self, fileobj, msg)

datasets/utils/tsv/io_common.py:21

Method__init__

NOTE: this interface is experimental. Args: is_train: for training or inference augmentations: a list of augm

datasets/dataset_mappers/lvis_dataset_mapper.py:73

Method__init__

( self, is_train=True, *, augmentation, image_format, )

datasets/dataset_mappers/sam_baseline_dataset_mapper_json.py:94

Method__init__

NOTE: this interface is experimental. Args: is_train: for training or inference augmentations: a list of augm

datasets/dataset_mappers/part_data_filter_whole_new_instance_dataset_mapper.py:127

Method__init__

NOTE: this interface is experimental. Args: is_train: for training or inference augmentations: a list of augm

datasets/dataset_mappers/pascal_instance_new_baseline_dataset_mapper.py:109

Method__init__

( self, is_train=True, *, augmentation, image_format, )

datasets/dataset_mappers/sam_baseline_dataset_mapper.py:111

Method__init__

NOTE: this interface is experimental. Args: is_train: for training or inference augmentations: a list of augm

datasets/dataset_mappers/mask_former_semantic_dataset_mapper.py:33

Method__init__

NOTE: this interface is experimental. Args: is_train: for training or inference augmentations: a list of augm

datasets/dataset_mappers/imagenet_dataset_mapper.py:40

Method__init__

NOTE: this interface is experimental. Args: is_train: for training or inference augmentations: a list of augm

datasets/dataset_mappers/mask_former_panoptic_dataset_mapper.py:35

Method__init__

NOTE: this interface is experimental. Args: is_train: for training or inference augmentations: a list of augm

datasets/dataset_mappers/mask_former_interactive_panoptic_dataset_mapper.py:39

Functions759 in github.com/UX-Decoder/Semantic-SAM