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

hub / github.com/UX-Decoder/Semantic-SAM / functions

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

⨍ Functions 759◇ Types & classes 112

↓ 2 callersFunctionprompt_engineering

(classnames, topk=1, suffix='.')

utils/prompt_engineering.py:90

↓ 2 callersMethodread_img

(self, row)

datasets/dataset_mappers/sam_baseline_dataset_mapper.py:177

↓ 2 callersFunctionremove_small_regions

Removes small disconnected regions and holes in a mask. Returns the mask and an indicator of if the mask has been modified.

utils/sam_utils/amg.py:267

↓ 2 callersFunctionremove_small_regions

Removes small disconnected regions and holes in a mask. Returns the mask and an indicator of if the mask has been modified.

tasks/interactive_idino_m2m.py:108

↓ 2 callersMethodremove_small_regions

Removes small disconnected regions and holes in a mask. Returns the mask and an indicator of if the mask has been modified.

tasks/interactive_predictor.py:115

↓ 2 callersMethodreset

(self)

utils/misc.py:49

↓ 2 callersFunctionrle_to_mask

Compute a binary mask from an uncompressed RLE.

utils/sam_utils/amg.py:138

↓ 2 callersFunctionsetup

Create configs and perform basic setups.

train_net.py:380

↓ 2 callersFunctionshow_anns

(anns)

tasks/interactive_idino_m2m_auto.py:73

↓ 2 callersFunctionsigmoid_focal_loss

Loss used in RetinaNet for dense detection: https://arxiv.org/abs/1708.02002. Args: inputs: A float tensor of arbitrary shape.

semantic_sam/modules/criterion_interactive_many_to_one.py:23

↓ 2 callersMethodtest

(cls, cfg, model, evaluators=None)

train_net.py:301

↓ 2 callersMethodtrain

Convert the model into training mode while keep layers freezed.

semantic_sam/backbone/focal.py:592

↓ 2 callersFunctionuncrop_boxes_xyxy

(boxes: torch.Tensor, crop_box: List[int])

utils/sam_utils/amg.py:237

↓ 2 callersFunctionwindow_partition

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

semantic_sam/backbone/swin_new.py:44

↓ 2 callersFunctionwindow_partition

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

semantic_sam/backbone/swin.py:49

↓ 2 callersMethodwith_pos_embed

(self, tensor, pos: Optional[Tensor])

semantic_sam/body/transformer_blocks.py:179

↓ 2 callersMethodwith_pos_embed

(self, tensor, pos: Optional[Tensor])

semantic_sam/body/decoder/modules.py:32

↓ 2 callersMethodwith_pos_embed

(tensor, pos)

semantic_sam/body/decoder/utils/dino_decoder.py:232

↓ 1 callersMethod__init__

(self, in_features: int, out_features: int)

semantic_sam/modules/attention.py:327

↓ 1 callersMethod__init__

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

semantic_sam/body/decoder/utils/dino_decoder.py:20

↓ 1 callersMethod_convert_boxes

Convert different format of boxes to an NxB array, where B = 4 or 5 is the box dimension.

utils/visualizer.py:1218

↓ 1 callersFunction_convert_category_id

(segment_info, meta)

datasets/registration/register_scannet_panoptic.py:54

↓ 1 callersFunction_convert_category_id

(segment_info, meta)

datasets/registration/register_coco_panoptic_annos_caption_grounding.py:107

↓ 1 callersFunction_convert_category_id

(segment_info, meta)

datasets/registration/register_coco_panoptic_annos_caption.py:93

↓ 1 callersFunction_convert_category_id

(segment_info, meta)

datasets/registration/register_coco_panoptic_annos_semseg_interactive.py:86

↓ 1 callersFunction_convert_category_id

(segment_info, meta)

datasets/registration/register_coco_panoptic_annos_caption_grounding_interactive.py:107

↓ 1 callersFunction_convert_category_id

(segment_info, meta)

datasets/registration/register_coco_panoptic_annos_semseg_interactive_jointboxpoint.py:86

↓ 1 callersFunction_convert_category_id

(segment_info, meta)

datasets/registration/register_bdd100k_panoseg.py:127

↓ 1 callersFunction_convert_category_id

(segment_info, meta)

datasets/registration/register_ade20k_panoptic.py:228

↓ 1 callersFunction_convert_category_id

(segment_info, meta)

datasets/registration/register_coco_panoptic_annos_semseg.py:86

↓ 1 callersFunction_convert_category_id

(segment_info, meta)

datasets/registration/register_object365_od.py:56

↓ 1 callersFunction_convert_category_id

(segment_info, meta)

datasets/registration/register_coco_panoptic_annos_caption_interactive.py:93

↓ 1 callersMethod_convert_category_id

(self, segment_info)

datasets/evaluation/panoptic_evaluation.py:53

↓ 1 callersMethod_convert_keypoints

(self, keypoints)

utils/visualizer.py:1270

↓ 1 callersMethod_convert_masks

Convert different format of masks or polygons to a tuple of masks and polygons. Returns: list[GenericMask]:

utils/visualizer.py:1227

↓ 1 callersMethod_derive_coco_results

Additionally plot mAP for 'seen classes' and 'unseen classes'

datasets/evaluation/pascal_part_evaluation.py:37

↓ 1 callersMethod_embed_masks

(self, input_mask: torch.Tensor, has_mask_input: torch.Tensor)

utils/sam_utils/onnx.py:69

↓ 1 callersMethod_embed_points

(self, point_coords: torch.Tensor, point_labels: torch.Tensor)

utils/sam_utils/onnx.py:51

↓ 1 callersMethod_ensure_tsv_opened

(self)

datasets/utils/tsv/tsv_io.py:78

↓ 1 callersMethod_generate_masks

(self, image: np.ndarray)

tasks/automatic_mask_generator.py:215

↓ 1 callersFunction_get_activation_fn

Return an activation function given a string

semantic_sam/body/decoder/utils/utils.py:103

↓ 1 callersFunction_get_ade20k_full_meta

()

datasets/registration/register_ade20k_full.py:926

↓ 1 callersFunction_get_ade_instances_meta

()

datasets/registration/register_ade20k_instance.py:28

↓ 1 callersFunction_get_coco_stuff_meta

()

datasets/registration/register_coco_stuff_10k.py:182

↓ 1 callersFunction_get_paco_metadata

(key)

datasets/registration/register_paco_part_all.py:17

↓ 1 callersFunction_get_partimagenet_metadata

(key)

datasets/registration/register_partimagenet_part_all.py:50

↓ 1 callersFunction_get_partimagenet_metadata

(key)

datasets/registration/register_pascal_part_all.py:206

↓ 1 callersFunction_get_partimagenet_metadata

(key)

datasets/registration/register_pascal_part_all_interactive.py:206

↓ 1 callersFunction_get_sam_instances_meta

()

datasets/registration/register_sam.py:50

↓ 1 callersFunction_get_sam_instances_meta

()

datasets/registration/register_sam_mnode.py:50

↓ 1 callersMethod_get_tgt_permutation_idx

(self, indices)

semantic_sam/modules/criterion_interactive_many_to_many.py:403

↓ 1 callersFunction_is_power_of_2

(n)

semantic_sam/body/encoder/ops/modules/ms_deform_attn.py:28

↓ 1 callersMethod_process_batch

( self, images, points: np.ndarray, im_size: Tuple[int, ...], crop_box

tasks/automatic_mask_generator.py:284

↓ 1 callersMethod_process_crop

( self, image: np.ndarray, crop_box: List[int], crop_layer_idx: int, o

tasks/automatic_mask_generator.py:245

↓ 1 callersMethod_reset_parameters

(self)

semantic_sam/modules/attention.py:403

↓ 1 callersMethod_reset_parameters

(self)

semantic_sam/body/transformer_blocks.py:56

↓ 1 callersMethod_reset_parameters

(self)

semantic_sam/body/encoder/encoder_deform.py:48

↓ 1 callersMethod_reset_parameters

(self)

semantic_sam/body/encoder/transformer_encoder_fpn.py:175

↓ 1 callersMethod_reset_parameters

(self)

semantic_sam/body/encoder/ops/modules/ms_deform_attn.py:66

↓ 1 callersMethod_reset_parameters

(self)

semantic_sam/body/decoder/modules.py:27

↓ 1 callersMethod_reset_parameters

(self)

semantic_sam/body/decoder/modules.py:85

↓ 1 callersMethod_reset_parameters

(self)

semantic_sam/body/decoder/modules.py:149

↓ 1 callersMethod_reset_parameters

(self)

semantic_sam/body/decoder/utils/dino_decoder.py:88

↓ 1 callersMethod_set_aux_loss

(self, outputs_class=None, outputs_seg_masks=None, out_boxes=None, predictions_iou_score=None, predictions_cla

semantic_sam/body/decoder/interactive_mask_dino.py:640

↓ 1 callersMethod_setup_figure

Args: Same as in :meth:`__init__()`. Returns: fig (matplotlib.pyplot.figure): top level container for all th

utils/visualizer.py:269

↓ 1 callersMethod_transform_annotations

(self, dataset_dict, transforms, image_shape)

datasets/dataset_mappers/inference_mapper_with_gt.py:116

↓ 1 callersFunctionalign_and_update_state_dicts

(model_state_dict, ckpt_state_dict)

utils/model.py:29

↓ 1 callersMethodapply_coords

Expects a numpy array of length 2 in the final dimension. Requires the original image size in (H, W) format.

utils/sam_utils/transforms.py:33

↓ 1 callersMethodapply_coords_torch

Expects a torch tensor with length 2 in the last dimension. Requires the original image size in (H, W) format.

utils/sam_utils/transforms.py:67

↓ 1 callersFunctionarea_from_rle

(rle: Dict[str, Any])

utils/sam_utils/amg.py:152

↓ 1 callersMethodattention

(self, x: torch.Tensor, key_padding_mask: torch.Tensor = None)

semantic_sam/language/LangEncoder/transformer.py:63

↓ 1 callersMethodauto_scale_workers

Returns: CfgNode: a new config. Same as original if ``cfg.SOLVER.REFERENCE_WORLD_SIZE==0``.

train_net.py:269

↓ 1 callersFunctionbatch_iterator

(batch_size: int, *args)

utils/sam_utils/amg.py:98

↓ 1 callersMethodbbox

(self)

utils/visualizer.py:146

↓ 1 callersFunctionbuild_all_layer_point_grids

Generates point grids for all crop layers.

utils/sam_utils/amg.py:189

↓ 1 callersMethodbuild_attention_mask

(self)

semantic_sam/language/LangEncoder/transformer.py:122

↓ 1 callersFunctionbuild_backbone

(config, **kwargs)

semantic_sam/backbone/build.py:6

↓ 1 callersFunctionbuild_decoder

(config, *args, **kwargs)

semantic_sam/body/decoder/build.py:5

↓ 1 callersFunctionbuild_encoder

(config, *args, **kwargs)

semantic_sam/body/encoder/build.py:7

↓ 1 callersFunctionbuild_eval_dataloader

(cfg, )

datasets/build.py:381

↓ 1 callersMethodbuild_hooks

Build a list of default hooks, including timing, evaluation, checkpointing, lr scheduling, precise BN, writing events. Retur

train_net.py:105

↓ 1 callersFunctionbuild_language_encoder

(config, **kwargs)

semantic_sam/language/build.py:5

↓ 1 callersMethodbuild_lr_scheduler

It now calls :func:`detectron2.solver.build_lr_scheduler`. Overwrite it if you'd like a different scheduler.

train_net.py:171

↓ 1 callersMethodbuild_optimizer

(cls, cfg, model)

train_net.py:179

↓ 1 callersFunctionbuild_point_grid

Generates a 2D grid of points evenly spaced in [0,1]x[0,1].

utils/sam_utils/amg.py:179

↓ 1 callersFunctionbuild_semantic_sam_head

(config, *args, **kwargs)

semantic_sam/body/build.py:6

↓ 1 callersMethodbuild_test_loader

(cls, cfg, dataset_name)

train_net.py:165

↓ 1 callersFunctionbuild_train_dataloader

(cfg, )

datasets/build.py:406

↓ 1 callersMethodbuild_train_loader

(cls, cfg)

train_net.py:161

↓ 1 callersFunctionbuild_transform_gen

Create a list of default :class:`Augmentation` from config. Now it includes resizing and flipping. Returns: list[Augmentation]

datasets/dataset_mappers/lvis_dataset_mapper.py:22

↓ 1 callersFunctionbuild_transform_gen

Create a list of default :class:`Augmentation` from config. Now it includes resizing and flipping. Returns: list[Augmentation]

datasets/dataset_mappers/sam_baseline_dataset_mapper_json.py:41

↓ 1 callersFunctionbuild_transform_gen

Create a list of default :class:`Augmentation` from config. Now it includes resizing and flipping. Returns: list[Augmentation]

datasets/dataset_mappers/part_data_filter_whole_new_instance_dataset_mapper.py:74

↓ 1 callersFunctionbuild_transform_gen

Create a list of default :class:`Augmentation` from config. Now it includes resizing and flipping. Returns: list[Augmentation]

datasets/dataset_mappers/pascal_instance_new_baseline_dataset_mapper.py:59

↓ 1 callersFunctionbuild_transform_gen

Create a list of default :class:`Augmentation` from config. Now it includes resizing and flipping. Returns: list[Augmentation]

datasets/dataset_mappers/sam_baseline_dataset_mapper.py:58

↓ 1 callersFunctionbuild_transform_gen

Create a list of default :class:`Augmentation` from config. Now it includes resizing and flipping. Returns: list[Augmentation]

datasets/dataset_mappers/o365_instance_new_baseline_dataset_mapper.py:42

↓ 1 callersFunctionbuild_transform_gen

Create a list of default :class:`Augmentation` from config. Now it includes resizing and flipping. Returns: list[Augmentation]

datasets/dataset_mappers/coco_panoptic_new_baseline_dataset_mapper.py:21

↓ 1 callersFunctionbuild_transform_gen

Create a list of default :class:`Augmentation` from config. Now it includes resizing and flipping. Returns: list[Augmentation]

datasets/dataset_mappers/coco_interactive_panoptic_new_baseline_dataset_mapper.py:42

↓ 1 callersFunctionbuild_transform_gen

Create a list of default :class:`Augmentation` from config. Now it includes resizing and flipping. Returns: list[Augmentation]

datasets/dataset_mappers/coco_instance_new_baseline_dataset_mapper.py:38

↓ 1 callersFunctioncalculate_uncertainty

We estimate uncerainty as L1 distance between 0.0 and the logit prediction in 'logits' for the foreground class in `classes`. Args:

semantic_sam/modules/criterion_interactive_many_to_many.py:112

↓ 1 callersFunctioncalculate_uncertainty

We estimate uncerainty as L1 distance between 0.0 and the logit prediction in 'logits' for the foreground class in `classes`. Args:

semantic_sam/modules/criterion_interactive_many_to_one.py:107

← previous next →101–200 of 759, ranked by callers