MCPcopy Create free account

hub / github.com/OpenGVLab/HumanBench / functions

Functions1,842 in github.com/OpenGVLab/HumanBench

⨍ Functions 1,842◇ Types & classes 382
↓ 235 callersMethodget
Returns the field called `name`.
PATH/core/data/datasets/images/seg_dataset_dev.py:100
↓ 201 callersMethodsize
(self, *args, **kwargs)
PATH/core/data/transforms/pose_transforms.py:161
↓ 145 callersMethodpermute
(self, x, size)
PATH/core/models/ops/hrt_helpers/multihead_isa_attention.py:59
↓ 90 callersMethodcuda
Returns: Instances: all fields are called with a `cuda`, if the field has this method.
PATH/core/data/datasets/images/seg_dataset_dev.py:116
↓ 83 callersFunctionprintlog
(*args, **kwargs)
PATH/core/utils.py:740
↓ 74 callersMethodcat
Concatenates a list of Boxes into a single Boxes Arguments: boxes_list (list[Boxes]) Returns: Boxes
PATH/core/models/ops/boxes.py:277
↓ 58 callersMethodupdate
(self)
PATH/core/solvers/solver.py:450
↓ 44 callersMethodload
(self, args)
PATH/core/testers/tester.py:138
↓ 41 callersMethodinfo
(self, **kwargs)
PATH/core/models/backbones/vit.py:562
↓ 36 callersMethodto
(self, device: torch.device)
PATH/core/models/ops/boxes.py:166
↓ 33 callersMethodresize
(img, target_size, interpolation=None)
PATH/core/data/datasets/images/image_helper.py:164
↓ 27 callersMethodBNReLU
(num_features, bn_type=None, bn_group=None, **kwargs)
PATH/core/models/decoders/seg_decoders/module_helper.py:51
↓ 27 callersMethoddim
(self)
PATH/core/data/transforms/pose_transforms.py:165
↓ 24 callersMethodstack
(self)
PATH/core/data/transforms/pose_transforms.py:149
↓ 23 callersMethodclip
Clip (in place) the boxes by limiting x coordinates to the range [0, width] and y coordinates to the range [0, height]. Args
PATH/core/models/ops/boxes.py:181
↓ 22 callersMethodstate_dict
Returns a dict containing the current state of this :class:`FP16_Optimizer` instance. This dict contains attributes of :class:`FP16_
PATH/core/fp16/opt.py:292
↓ 21 callersMethoddecode
Decode keypoints from heatmaps. Args: img_metas (list(dict)): Information about data augmentation By default this
PATH/core/models/decoders/pose_decodes/pose_decoder.py:817
↓ 19 callersMethoddevice
(self)
PATH/core/models/ops/boxes.py:297
↓ 18 callersMethodBatchNorm2d
(bn_type="torch", ret_cls=False)
PATH/core/models/decoders/seg_decoders/module_helper.py:68
↓ 18 callersMethodempty
(self)
PATH/core/utils.py:48
↓ 15 callersMethodsave
(self)
PATH/core/solvers/solver.py:546
↓ 15 callersFunctionsync_print
(*args, **kwargs)
PATH/core/utils.py:691
↓ 15 callersMethodtransform
In-place transform all attributes of this class. By "in-place", it means after calling this method, accessing an attribute such
PATH/core/data/transforms/seg_transforms_dev.py:663
↓ 13 callersMethodclone
Clone the Boxes. Returns: Boxes
PATH/core/models/ops/boxes.py:157
↓ 13 callersFunctionvgather
(x_list, x)
PATH/core/distributed_utils.py:1129
↓ 13 callersFunctionvreduce
(x, tensor, group=None)
PATH/core/distributed_utils.py:1121
↓ 12 callersMethod__init__
(self, embed_dim, eps=1e-6)
PATH/core/models/necks/simple_fpn.py:21
↓ 11 callersMethod_init
(self, params=None)
PATH/core/data/transforms/sparsercnn_peddet_transforms_helpers/augmentation.py:103
↓ 11 callersMethod_process
(self, img, data_dict, skip_condition, *args, **kwargs)
PATH/core/data/datasets/images/seg_data_tools/cv2_aug_transforms.py:25
↓ 11 callersFunctioncheckpoint_wrapper
A friendlier wrapper for performing activation checkpointing. Compared to the PyTorch version, this version: - wraps an nn.Module, so
PATH/core/models/ckpt.py:275
↓ 10 callersMethod__call__
(self, img, **kwargs)
PATH/core/data/datasets/images/seg_data_tools/cv2_aug_transforms.py:604
↓ 10 callersMethod__init__
Args: shape: (h, w) tuple or a int interp: PIL interpolation method
PATH/core/data/transforms/sparsercnn_peddet_transforms_helpers/augmentation_impl.py:105
↓ 10 callersMethod_set_attributes
Set attributes from the input list of parameters. Args: params (list): list of parameters.
PATH/core/data/transforms/seg_transforms_dev.py:49
↓ 10 callersFunctionpoint_sample
A wrapper around :function:`torch.nn.functional.grid_sample` to support 3D point_coords tensors. Unlike :function:`torch.nn.functional.grid
PATH/core/models/decoders/losses/test_time.py:6
↓ 10 callersMethodsend
(self, msg, echo=True)
PATH/core/msg_server.py:26
↓ 9 callersMethod__init__
Args: prob (float): probability of flip. horizontal (boolean): whether to apply horizontal flipping verti
PATH/core/data/transforms/seg_aug_dev.py:399
↓ 9 callersMethod__init__
( self, in_channels, key_channels, out_channels, scale=1, drop
PATH/core/models/decoders/seg_decoders/spatial_ocr_block.py:316
↓ 9 callersMethod_crop
(self, x, offset_up, offset_left, target_size)
PATH/core/data/datasets/images/seg_data_tools/cv2_aug_transforms.py:530
↓ 9 callersMethod_pad
(self, x, pad_value, height, width, target_size, offset_left, offset_up)
PATH/core/data/datasets/images/seg_data_tools/cv2_aug_transforms.py:64
↓ 9 callersMethoddecompose
(self)
PATH/core/utils.py:768
↓ 8 callersMethod__init__
(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.)
PATH/core/models/backbones/vit.py:44
↓ 8 callersMethod__init__
(self, configer=None, **kwargs)
PATH/core/models/decoders/losses/seg_losses.py:9
↓ 8 callersMethod_scale_target
(targets_, scaled_size)
PATH/core/models/decoders/losses/seg_losses.py:57
↓ 8 callersFunctioncv2_loader
(img_str)
PATH/core/utils.py:282
↓ 8 callersFunctionget_rank
()
PATH/core/distributed_utils.py:160
↓ 8 callersMethodimg2np
(img)
PATH/core/data/datasets/images/image_helper.py:141
↓ 8 callersMethodload_state_dict
Loads a state_dict created by an earlier call to state_dict(). If ``fp16_optimizer_instance`` was constructed from some ``init_optim
PATH/core/fp16/opt.py:313
↓ 8 callersMethodtest
(self, model, evaluator=None)
PATH/core/solvers/solver_multitask_dev.py:846
↓ 7 callersMethod__init__
(self, ginfo, data_path, data_use_ratio=1,
PATH/core/data/datasets/images/parsing_dataset.py:483
↓ 7 callersMethod__init__
(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.)
PATH/core/models/backbones/vitdet.py:52
↓ 7 callersMethod__init__
(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.)
PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py:52
↓ 7 callersMethod__init__
(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.)
PATH/core/models/backbones/vitdet_for_ladder_attention.py:52
↓ 7 callersMethod__init__
(self, in_features, out_features)
PATH/core/models/decoders/losses/classification_losses.py:244
↓ 7 callersMethod_set_aux_loss
(self, outputs_class=None, outputs_seg_masks=None, outputs_mask=No
PATH/core/models/decoders/peddet_decoders/mask2former_transformer_decoder.py:612
↓ 7 callersFunctionc2_xavier_fill
Initialize `module.weight` using the "XavierFill" implemented in Caffe2. Also initializes `module.bias` to 0. Args: module (torc
PATH/core/models/ops/utils.py:100
↓ 7 callersMethodclose
(self)
PATH/core/msg_server.py:82
↓ 7 callersMethodconvert
(self, img, alpha=1, beta=0)
PATH/core/data/transforms/seg_aug_dev.py:657
↓ 7 callersFunctionget_world_size
()
PATH/core/distributed_utils.py:166
↓ 7 callersMethodremove
Remove the field called `name`.
PATH/core/data/datasets/images/seg_dataset_dev.py:94
↓ 6 callersMethod__init__
(self, optimizer, last_iter=-1)
PATH/core/lr_scheduler/base.py:11
↓ 6 callersMethod__repr__
(self)
PATH/core/models/decoders/losses/classification_losses.py:263
↓ 6 callersFunction_get_max_preds
Get keypoint predictions from score maps. Note: batch_size: N num_keypoints: K heatmap height: H heatmap width: W
PATH/core/models/decoders/pose_decodes/pose_decoder.py:97
↓ 6 callersFunctionconvert_color_factory
(src: str, dst: str)
PATH/core/data/transforms/parsing_transforms.py:374
↓ 6 callersFunctionget_layer_param
(model)
PATH/helper/param_count.py:72
↓ 6 callersMethodget_vis_token_pos_embed
(self, shape=None)
PATH/core/models/decoders/peddet_decoders/mask2former_transformer_decoder.py:481
↓ 6 callersFunctionis_nested
(x)
PATH/core/fp16/utils.py:68
↓ 6 callersFunctionload_state
(path, model, ignore=[], optimizer=None, cuda=False, recover=False, remove_prefix=None, strict=
PATH/core/utils.py:139
↓ 6 callersFunctionloss_entry
(config)
PATH/core/models/decoders/losses/__init__.py:8
↓ 6 callersFunctionpil_loader
(img_str)
PATH/core/utils.py:276
↓ 6 callersFunctionsave_state
(state, path, step)
PATH/core/utils.py:108
↓ 5 callersFunctionaccuracy
Computes the precision@k for the specified values of k
PATH/core/utils.py:87
↓ 5 callersMethodarea
Computes the area of all the boxes. Returns: torch.Tensor: a vector with areas of each box.
PATH/core/models/ops/boxes.py:170
↓ 5 callersMethodbox_overlap_opr
(self, dboxes: np.ndarray, gboxes: np.ndarray, if_iou)
PATH/core/solvers/utils/peddet_tester_dev.py:343
↓ 5 callersFunctionclip_grad_norm_
(parameters, max_norm=1000000, norm_type=2, auto_clipper=None, loss_scale=1.0)
PATH/core/clipping.py:17
↓ 5 callersMethodcuda
(self)
PATH/core/utils.py:771
↓ 5 callersMethodintersection
:param boxB: the target bounding box to compute overlap :type boxB: BoxBase :return: the intersection area of current boundin
PATH/core/solvers/utils/detools/box.py:100
↓ 5 callersMethodnp2img
(arr)
PATH/core/data/datasets/images/image_helper.py:132
↓ 5 callersFunctionpad
(image, target, padding)
PATH/core/data/transforms/peddet_transforms_helpers/transforms.py:165
↓ 5 callersMethodset
Set the field named `name` to `value`. The length of `value` must be the number of instances, and must agree with other exist
PATH/core/data/datasets/images/seg_dataset_dev.py:74
↓ 5 callersMethodstep
Performs a single optimization step. Arguments: - closure (:obj:`callable`, optional): A closure that reevaluates the model and r
PATH/core/optimizers/lars.py:50
↓ 5 callersMethodtrain
(self, mode=True)
PATH/core/distributed_utils.py:56
↓ 4 callersMethod__init__
Args: op (Callable): operation to be applied to the image, which takes in an ndarray and returns an ndarray.
PATH/core/data/transforms/sparsercnn_peddet_transforms_helpers/transforms.py:256
↓ 4 callersMethod__init__
(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8, weight_decay=0, amsgrad=False, max_norm
PATH/core/optimizers/adam_clip.py:8
↓ 4 callersMethod__init__
(self, channel=512,reduction=16,kernel_size=7, use_cls_token=False)
PATH/core/models/necks/attention.py:87
↓ 4 callersMethod__init__
(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.)
PATH/core/models/necks/ladder_side_attention_fpn.py:57
↓ 4 callersMethod__init__
(self, input_dim, hidden_dim, output_dim, num_layers)
PATH/core/models/decoders/peddet_decoders/mask2former_transformer_decoder.py:276
↓ 4 callersFunction_gaussian_blur
Modulate heatmap distribution with Gaussian. sigma = 0.3*((kernel_size-1)*0.5-1)+0.8 sigma~=3 if k=17 sigma=2 if k=11; sigma~=1.5
PATH/core/models/decoders/pose_decodes/pose_decoder.py:233
↓ 4 callersFunction_get_activation
Return an activation function given a string
PATH/core/models/necks/simple_fpn.py:74
↓ 4 callersMethod_init
(self, params=None)
PATH/core/data/transforms/seg_transforms_dev.py:441
↓ 4 callersMethod_load_from_state_dict
( self, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs
PATH/core/models/decoders/peddet_decoders/mask2former_transformer_decoder.py:311
↓ 4 callersFunctionadd_task_specific
(m, task_specific)
PATH/core/utils.py:471
↓ 4 callersFunctioncat
Efficient version of torch.cat that avoids a copy if there is only a single element in a list
PATH/core/models/ops/utils.py:132
↓ 4 callersFunctionchange_tensor_cuda
()
PATH/core/utils.py:453
↓ 4 callersFunctioncount_parameters_num
(model)
PATH/core/utils.py:353
↓ 4 callersFunctioncreate_logger
(name, log_file, level=logging.INFO)
PATH/core/utils.py:225
↓ 4 callersFunctioncrop
(image, target, region)
PATH/core/data/transforms/peddet_transforms_helpers/transforms.py:37
↓ 4 callersFunctionget_2d_sincos_pos_embed
grid_size: int of the grid height and width return: pos_embed: [grid_size*grid_size, embed_dim] or [1+grid_size*grid_size, embed_dim] (w/
PATH/core/models/backbones/vitdet.py:814
↓ 4 callersFunctionget_2d_sincos_pos_embed
grid_size: int of the grid height and width return: pos_embed: [grid_size*grid_size, embed_dim] or [1+grid_size*grid_size, embed_dim] (w/
PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py:825
↓ 4 callersFunctionget_2d_sincos_pos_embed
grid_size: int of the grid height and width return: pos_embed: [grid_size*grid_size, embed_dim] or [1+grid_size*grid_size, embed_dim] (w/
PATH/core/models/backbones/vitdet_for_ladder_attention.py:813
↓ 4 callersMethodimshow
(win_name, img, time=0)
PATH/core/data/datasets/images/image_helper.py:124
next →1–100 of 1,842, ranked by callers