MCPcopy Create free account

hub / github.com/GeWu-Lab/AnyTouch2 / functions

Functions813 in github.com/GeWu-Lab/AnyTouch2

⨍ Functions 813◇ Types & classes 138↳ Endpoints 2
↓ 1 callersFunctionadd_residual
(x, brange, residual, residual_scale_factor, scaling_vector=None)
sparsh/tactile_ssl/model/layers/block.py:149
↓ 1 callersFunctionalexnet
()
sparsh/tactile_ssl/model/pretrained.py:18
↓ 1 callersMethodapply_block_mask
(self, x: torch.Tensor, mask: torch.Tensor)
sparsh/tactile_ssl/model/multimodal_transformer.py:181
↓ 1 callersMethodapply_center_update
(self)
sparsh/tactile_ssl/loss/ibot_patch_loss.py:154
↓ 1 callersMethodapply_center_update
(self)
sparsh/tactile_ssl/loss/dino_loss.py:90
↓ 1 callersMethodapply_mask
( self, x: torch.Tensor, mask: torch.Tensor, mask_type: Literal["block", "tube"] )
sparsh/tactile_ssl/model/multimodal_transformer.py:166
↓ 1 callersMethodapply_tubelet_mask
(self, x: torch.Tensor, mask: torch.Tensor)
sparsh/tactile_ssl/model/multimodal_transformer.py:176
↓ 1 callersFunctionattempt_resume
(cfg: DictConfig)
sparsh/train_task.py:241
↓ 1 callersFunctioncompress_and_save_dataset
(object)
sparsh/scripts/download_datasets_scratch/compress_feeling_success.py:24
↓ 1 callersFunctioncompress_and_save_dataset
(object)
sparsh/scripts/download_datasets_scratch/compress_object_folder.py:22
↓ 1 callersFunctioncompress_and_save_dataset
(object)
sparsh/scripts/download_datasets_scratch/compress_touch_go.py:42
↓ 1 callersMethodcompute_loss
imgs: [N, C, H, W] mask: [N, L], 0 is keep, 1 is remove, From: https://github.com/facebookresearch/mae/tree/main
sparsh/tactile_ssl/algorithm/mae.py:129
↓ 1 callersMethodcompute_loss
(self, inputs, outputs)
sparsh/tactile_ssl/downstream_task/utils_forcefield/ssl_flow_loss.py:108
↓ 1 callersMethodcompute_losses_normal
Compute the reprojection and smoothness losses for a minibatch
sparsh/tactile_ssl/downstream_task/utils_forcefield/ssl_flow_loss.py:21
↓ 1 callersMethodcompute_losses_shear
Compute the reprojection and smoothness optical flow losses for a minibatch
sparsh/tactile_ssl/downstream_task/utils_forcefield/ssl_flow_loss.py:93
↓ 1 callersMethodcompute_reprojection_loss
Computes reprojection loss between a batch of predicted and target images
sparsh/tactile_ssl/downstream_task/utils_forcefield/ssl_flow_loss.py:77
↓ 1 callersMethodcompute_sl_force
(self, outputs, mask)
sparsh/tactile_ssl/downstream_task/forcefield_sl.py:225
↓ 1 callersMethodconfigure_optimizers
( self, num_iterations_per_epoch: int, num_epochs: int )
sparsh/tactile_ssl/algorithm/module.py:21
↓ 1 callersMethodconnect_digit
(self)
sparsh/tactile_ssl/data/vision_based_interactive.py:91
↓ 1 callersMethodconnect_gelsight
(self)
sparsh/tactile_ssl/data/vision_based_interactive.py:82
↓ 1 callersFunctionconvert_clip_checkpoint
Copy/paste/tweak model's weights to transformers design.
CLIP-B-16/convert_open_clip_to_hf.py:108
↓ 1 callersFunctioncopy_attn_layer
(hf_attn_layer, pt_attn_layer)
CLIP-B-16/convert_open_clip_to_hf.py:25
↓ 1 callersFunctioncopy_encoder
(hf_encoder, pt_model)
CLIP-B-16/convert_open_clip_to_hf.py:70
↓ 1 callersFunctioncopy_layer
(hf_layer, pt_layer)
CLIP-B-16/convert_open_clip_to_hf.py:53
↓ 1 callersFunctioncopy_mlp
(hf_mlp, pt_mlp)
CLIP-B-16/convert_open_clip_to_hf.py:43
↓ 1 callersFunctioncopy_text_model_and_projection
(hf_model, pt_model)
CLIP-B-16/convert_open_clip_to_hf.py:82
↓ 1 callersFunctioncopy_vison_model_and_projection
(hf_model, pt_model)
CLIP-B-16/convert_open_clip_to_hf.py:90
↓ 1 callersFunctioncreate_ndgrid
Create a n-D grid with coordinates in range [-1, 1] **supports only upto 3D** Args: resolution: Resolution of the grid de
sparsh/tactile_ssl/utils/__init__.py:39
↓ 1 callersFunctiondigit_intrinsics
()
sparsh/tactile_ssl/downstream_task/utils_forcefield/ssl_utils.py:17
↓ 1 callersMethoddiscretize_poses
(self)
sparsh/tactile_ssl/data/vision_based_pose_probes.py:88
↓ 1 callersFunctiondisp_to_depth
Convert network's sigmoid output into depth prediction The formula for this conversion is given in the 'additional considerations' section of
sparsh/tactile_ssl/downstream_task/utils_forcefield/ssl_utils.py:211
↓ 1 callersFunctiondrop_path
(x, drop_prob: float = 0.0, training: bool = False)
model/layers/drop_path.py:14
↓ 1 callersFunctiondrop_path
(x, drop_prob: float = 0.0, training: bool = False)
sparsh/tactile_ssl/model/layers/drop_path.py:14
↓ 1 callersFunctionextract_strokes
(in_contact: np.ndarray)
sparsh/scripts/slip_labelling.py:43
↓ 1 callersMethodfit
The main entrypoint of the trainer, triggering the actual training. Args: module: the LightningModule to train. C
sparsh/tactile_ssl/trainer/trainer.py:157
↓ 1 callersMethodforward
(self, x, produce_targets=False)
model/layers/gumbel_vector_quantizer.py:148
↓ 1 callersMethodforward
(self, x: Tensor)
model/layers/block.py:86
↓ 1 callersMethodforward
(self, x, produce_targets=False)
sparsh/tactile_ssl/model/layers/gumbel_vector_quantizer.py:148
↓ 1 callersMethodforward
(self, x: Tensor)
sparsh/tactile_ssl/model/layers/block.py:89
↓ 1 callersMethodforward
( self, x: torch.Tensor, encoder_masks: List[torch.Tensor] = None, predictor_m
sparsh/tactile_ssl/algorithm/ijepa.py:298
↓ 1 callersMethodforward
( self, x: torch.Tensor, global_masks: List[torch.Tensor], local_masks: List[t
sparsh/tactile_ssl/algorithm/dinov2.py:343
↓ 1 callersMethodforward
( self, x: torch.Tensor, global_masks: List[torch.Tensor], local_masks: List[t
sparsh/tactile_ssl/algorithm/dino.py:329
↓ 1 callersMethodforward
( self, x: torch.Tensor, encoder_masks: List[torch.Tensor] = None, predictor_m
sparsh/tactile_ssl/algorithm/vjepa.py:238
↓ 1 callersMethodforward
(self, x: torch.Tensor)
sparsh/tactile_ssl/algorithm/mae.py:123
↓ 1 callersMethodforward
(self, inputs: dict)
sparsh/tactile_ssl/downstream_task/pose_sl.py:173
↓ 1 callersMethodforward
(self, x: torch.Tensor)
sparsh/tactile_ssl/downstream_task/textile_sl.py:104
↓ 1 callersMethodforward
(self, imgs: torch.Tensor, force: torch.Tensor = None)
sparsh/tactile_ssl/downstream_task/slip_sl.py:80
↓ 1 callersMethodforward
(self, x: torch.Tensor)
sparsh/tactile_ssl/downstream_task/grasp_sl.py:93
↓ 1 callersMethodforward
(self, x: torch.Tensor)
sparsh/tactile_ssl/downstream_task/force_sl.py:104
↓ 1 callersMethodforward_decoder
(self, x, ids_restore, sensor_type=None, data_type = None)
sparsh/tactile_ssl/model/point_mae.py:627
↓ 1 callersMethodforward_decoder
(self, z: torch.Tensor, **kwargs)
sparsh/tactile_ssl/probe/online_probe.py:30
↓ 1 callersMethodforward_encoder
(self, x=None, sensor_type=None, use_mask = False, get_cls = False, probe=False)
model/tactile_mae.py:156
↓ 1 callersMethodforward_encoder
(self, x=None, points=None, points_num=None, sensor_type=None, data_type = None, use_mask = True, get_cls = Fa
sparsh/tactile_ssl/model/point_mae.py:1142
↓ 1 callersMethodforward_encoder
( self, x: torch.Tensor, ids_mask_visible: Optional[torch.Tensor] = None )
sparsh/tactile_ssl/algorithm/mae.py:117
↓ 1 callersMethodforward_features
( self, xs: List[torch.Tensor], masks: Optional[List[torch.Tensor]] = None, ma
sparsh/tactile_ssl/model/multimodal_transformer.py:308
↓ 1 callersMethodforward_features_list
(self, x_list, masks_list)
sparsh/tactile_ssl/model/vision_transformer.py:319
↓ 1 callersMethodforward_loss_img
(self, x, pred, mask, data_type = None)
sparsh/tactile_ssl/model/point_mae.py:725
↓ 1 callersMethodforward_loss_pts
(self, points, pred, mask, data_type = None)
sparsh/tactile_ssl/model/point_mae.py:747
↓ 1 callersMethodforward_nested
x_list contains a list of tensors to nest together and run
model/layers/block.py:209
↓ 1 callersMethodforward_nested
x_list contains a list of tensors to nest together and run
sparsh/tactile_ssl/model/layers/block.py:212
↓ 1 callersMethodforward_point_decoder
(self, points, points_num, ids_restore, sensor_type=None, data_type = None)
sparsh/tactile_ssl/model/point_mae.py:676
↓ 1 callersMethodgenerate_images_pred
Generate the warped (reprojected) color images for a minibatch. Generated images are saved into the `outputs` dictionary.
sparsh/tactile_ssl/downstream_task/forcefield_sl.py:242
↓ 1 callersFunctionget_2d_sincos_pos_embed_from_grid
(embed_dim, grid)
model/util/pos_embed.py:38
↓ 1 callersFunctionget_2d_sincos_pos_embed_from_grid
(embed_dim, grid)
sparsh/tactile_ssl/model/util/pos_embed.py:38
↓ 1 callersFunctionget_2d_sincos_pos_embed_from_grid
(embed_dim, grid)
util/pos_embed.py:38
↓ 1 callersMethodget_accuracy
(self, data)
sparsh/tactile_ssl/test/test_t4_grasp.py:65
↓ 1 callersMethodget_accuracy
(self, data)
sparsh/tactile_ssl/test/test_t3_pose.py:84
↓ 1 callersMethodget_accuracy
(self, data)
sparsh/tactile_ssl/test/test_t6_textile.py:67
↓ 1 callersFunctionget_branges_scales
(x, sample_drop_ratio=0.0)
model/layers/block.py:138
↓ 1 callersFunctionget_branges_scales
(x, sample_drop_ratio=0.0)
sparsh/tactile_ssl/model/layers/block.py:141
↓ 1 callersFunctionget_dataloader_digit
(cfg: DictConfig)
sparsh/train_task.py:123
↓ 1 callersFunctionget_dataloader_reskin
(cfg: DictConfig)
sparsh/train_task.py:102
↓ 1 callersFunctionget_dataloaders
(cfg: DictConfig)
sparsh/train_task.py:225
↓ 1 callersFunctionget_digit_intrinsics
(img_sz)
sparsh/tactile_ssl/data/digit/utils.py:69
↓ 1 callersFunctionget_grad_norm_
(parameters, norm_type: float = 2.0)
util/misc.py:280
↓ 1 callersMethodget_latest_checkpoint
Returns the latest checkpoint from the ``checkpoint_dir`` Args: checkpoint_dir: the directory to search for checkpoints
sparsh/tactile_ssl/trainer/trainer.py:626
↓ 1 callersFunctionget_local_rank
()
sparsh/tactile_ssl/utils/__init__.py:12
↓ 1 callersMethodget_map_idx2traj
(self)
sparsh/tactile_ssl/data/vision_based_slip_probes.py:88
↓ 1 callersMethodget_map_idx2traj
(self)
sparsh/tactile_ssl/data/vision_based_forces_slip_probes.py:104
↓ 1 callersFunctionget_path_dataset
(config, dataset_name)
sparsh/tactile_ssl/data/digit/utils.py:37
↓ 1 callersFunctionget_rank
()
utils.py:50
↓ 1 callersFunctionget_rank
()
sparsh/tactile_ssl/model/utils.py:50
↓ 1 callersFunctionget_rank
()
util/misc.py:201
↓ 1 callersMethodget_smooth_loss
Computes the smoothness loss for a disparity image The color image is used for edge-aware smoothness
sparsh/tactile_ssl/downstream_task/utils_forcefield/ssl_flow_loss.py:58
↓ 1 callersFunctionget_translation_matrix
Convert a translation vector into a 4x4 transformation matrix
sparsh/tactile_ssl/downstream_task/utils_forcefield/pose_estimator/utils.py:29
↓ 1 callersMethodinit
(self)
sparsh/tactile_ssl/test/demo_t1_forcefield.py:40
↓ 1 callersMethodinit_pos_embed
(self, pos_embed_fn)
sparsh/tactile_ssl/model/multimodal_transformer.py:140
↓ 1 callersFunctioninit_wandb
(cfg: DictConfig)
sparsh/train_task.py:89
↓ 1 callersFunctionis_main_process
()
util/misc.py:207
↓ 1 callersFunctionlabel_slip
(frictional_coeff: float, forces: np.ndarray)
sparsh/scripts/slip_labelling.py:57
↓ 1 callersFunctionload_data
(args)
scripts/quick_start.py:33
↓ 1 callersFunctionload_dataset_forces_digit_gelsight
(config, dataset_name, sensor)
sparsh/tactile_ssl/data/digit/utils.py:257
↓ 1 callersFunctionload_dataset_poses
(config, dataset_name, finger_type, t_stride)
sparsh/tactile_ssl/data/digit/utils.py:282
↓ 1 callersMethodload_decoder
(self, decoder_partial: partial)
sparsh/tactile_ssl/algorithm/mae.py:74
↓ 1 callersMethodload_encoder
(self, checkpoint_encoder: str)
sparsh/tactile_ssl/downstream_task/sl_module.py:82
↓ 1 callersFunctionload_from_eval_checkpoint
(ckpt, model)
train/main_touchdbench.py:23
↓ 1 callersFunctionload_model_from_multi_clip
(ckpt, model)
scripts/quick_start.py:56
↓ 1 callersFunctionload_model_from_multi_clip
(ckpt, model)
sparsh/train_task.py:54
↓ 1 callersFunctionload_model_from_multi_clip
(ckpt, model)
train/main_touchdbench.py:37
↓ 1 callersFunctionload_model_from_multi_clip
(ckpt, model)
train/main_objbench.py:21
← previousnext →201–300 of 813, ranked by callers