MCPcopy Create free account

hub / github.com/Ropedia/SpatialBench / functions

Functions5,295 in github.com/Ropedia/SpatialBench

⨍ Functions 5,295◇ Types & classes 758↳ Endpoints 27
↓ 4 callersFunctionNoGradParamDict
(x)
benchmark/models/dust3r_root/dust3r/cloud_opt/commons.py:28
↓ 4 callersFunctionNoGradParamDict
(x)
benchmark/models/mast3r_root/dust3r/cloud_opt/commons.py:28
↓ 4 callersMethod__init__
(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, bias=True, drop=0.)
benchmark/models/worldmirror_root/src/models/layers/patch_embed.py:134
↓ 4 callersMethod__init__
( self, dim: int, num_heads: int, mlp_ratio: float = 4.0, qkv_bias: bo
benchmark/models/loger/models/layers/block.py:40
↓ 4 callersMethod__init__
( self, in_features, hidden_features=None, out_features=None, act_laye
benchmark/models/vgg_ttt/vggttt/nets/vggt/heads/track_modules/modules.py:122
↓ 4 callersMethod__init__
( self, dim_in: int, patch_size: int = 14, output_dim: int = 4, activa
benchmark/models/lingbot_map/heads/dpt_head.py:43
↓ 4 callersMethod__init__
(self, criterion, norm_mode='avg_dis', gt_scale=False)
benchmark/models/dust3r_root/dust3r/losses.py:153
↓ 4 callersMethod__init__
( self, dim: int, num_heads: int, mlp_ratio: float = 4.0, qkv_bias: bo
benchmark/models/pi3/models/layers/block.py:40
↓ 4 callersMethod__init__
(self, criterion, norm_mode='avg_dis', gt_scale=False)
benchmark/models/mast3r_root/dust3r/losses.py:153
↓ 4 callersMethod_add_pos_embed
Simple UV position encoding directly added to feature map.
benchmark/models/depth_anything_3/model/dpt.py:330
↓ 4 callersMethod_add_pos_embed
Simple UV position encoding directly added to feature map.
benchmark/models/r3/depth_anything_3/model/dpt.py:330
↓ 4 callersMethod_apply_activation_single
Apply activation to single channel output, maintaining semantic consistency with value branch in multi-channel case. Supports: exp /
benchmark/models/depth_anything_3/model/dpt.py:286
↓ 4 callersMethod_apply_activation_single
Apply activation to single channel output, maintaining semantic consistency with value branch in multi-channel case. Supports: exp /
benchmark/models/depth_anything_3/model/dualdpt.py:341
↓ 4 callersMethod_apply_activation_single
Apply activation to single channel output, maintaining semantic consistency with value branch in multi-channel case. Supports: exp /
benchmark/models/r3/depth_anything_3/model/dpt.py:286
↓ 4 callersMethod_apply_activation_single
Apply activation to single channel output, maintaining semantic consistency with value branch in multi-channel case. Supports: exp /
benchmark/models/r3/depth_anything_3/model/dualdpt.py:402
↓ 4 callersFunction_bilinear_intepolate
(x, stride, H, W)
benchmark/models/page4d/dependency/track_modules/blocks.py:151
↓ 4 callersFunction_cell_size
(cell2)
benchmark/models/mast3r_root/mast3r/utils/coarse_to_fine.py:43
↓ 4 callersFunction_coerce_int_value
Convert config values that may be ranges/strings/lists into a single int.
benchmark/models/loger/pi3_adapter.py:52
↓ 4 callersMethod_compute_metric_scale_factor
(self, image, pred_depth, pred_conf)
benchmark/models/r3/R3/models/r3_wrapper/online_inference.py:758
↓ 4 callersMethod_encode_image
image has B x 3 x img_size x img_size do_mask: whether to perform masking or not return_all_blocks: if True, return the feat
benchmark/models/dust3r_root/croco/models/croco.py:129
↓ 4 callersMethod_encode_image
image has B x 3 x img_size x img_size do_mask: whether to perform masking or not return_all_blocks: if True, return the feat
benchmark/models/mast3r_root/croco/models/croco.py:129
↓ 4 callersMethod_find_weight_in_dir
(self, directory)
benchmark/evaluation/model_adapters/r3_adapter.py:222
↓ 4 callersMethod_get_msk_indices
(self, msk)
benchmark/models/dust3r_root/dust3r/cloud_opt/optimizer.py:103
↓ 4 callersMethod_get_msk_indices
(self, msk)
benchmark/models/mast3r_root/dust3r/cloud_opt/optimizer.py:103
↓ 4 callersFunction_invert_SE3
Invert a 4x4 SE(3) matrix.
benchmark/models/loger/models/layers/prope.py:429
↓ 4 callersFunction_invert_SE3
Invert a 4x4 SE(3) matrix.
benchmark/models/pi3/models/layers/prope.py:429
↓ 4 callersFunction_invert_se3_batch
Invert (N, 4, 4) SE3 numpy batch.
benchmark/evaluation/metrics.py:335
↓ 4 callersFunction_is_offloaded_payload
(payload)
benchmark/models/scal3r/utils/offload_utils.py:28
↓ 4 callersFunction_is_run_dir
Return True if `path` looks like a dense run dir or staged sparse reference dir.
benchmark/models/r3/view.py:266
↓ 4 callersFunction_make_dense_resize_layer
(channels: int, resize_scale: float)
benchmark/models/vggt_omega/models/heads/dense_head.py:183
↓ 4 callersFunction_make_fusion_block
(features: int, size: int = None, has_residual: bool = True, groups: int = 1)
benchmark/models/worldmirror_root/src/models/heads/dense_head.py:365
↓ 4 callersFunction_make_fusion_block
(features: int, size: int = None, has_residual: bool = True, groups: int = 1)
benchmark/models/streamvggt/heads/dpt_head.py:287
↓ 4 callersFunction_make_fusion_block
(features: int, size: int = None, has_residual: bool = True, groups: int = 1)
benchmark/models/vggt/heads/dpt_head.py:299
↓ 4 callersFunction_make_fusion_block
(features: int, size: int = None, has_residual: bool = True, groups: int = 1)
benchmark/models/vgg_ttt/vggttt/nets/vggt/heads/dpt_head.py:326
↓ 4 callersFunction_make_fusion_block
(features: int, size: int = None, has_residual: bool = True, groups: int = 1)
benchmark/models/stream3r/models/components/heads/dpt_head.py:312
↓ 4 callersFunction_make_fusion_block
(features: int, size: int = None, has_residual: bool = True, groups: int = 1)
benchmark/models/lingbot_map/heads/dpt_head.py:302
↓ 4 callersFunction_make_fusion_block
(features: int, has_residual: bool = True)
benchmark/models/vggt_omega/models/heads/dense_head.py:222
↓ 4 callersFunction_make_fusion_block
(features: int, size: int = None, has_residual: bool = True, groups: int = 1)
benchmark/models/omnivggt/heads/dpt_head.py:312
↓ 4 callersFunction_make_fusion_block
(features: int, size: int = None, has_residual: bool = True, groups: int = 1)
benchmark/models/page4d/heads/dpt_head.py:299
↓ 4 callersFunction_make_fusion_block
( features: int, size: Optional[int] = None, has_residual: bool = True, groups: int = 1, )
benchmark/models/fastvggt/heads/dpt_head.py:351
↓ 4 callersFunction_make_fusion_block
(features: int, size: int = None, has_residual: bool = True, groups: int = 1)
benchmark/models/scal3r/utils/vggt/heads/dpt_head.py:326
↓ 4 callersFunction_make_fusion_block
(features: int, size: int = None, has_residual: bool = True, groups: int = 1, use_inplace: bool = True)
benchmark/models/zipmap/zipmap/heads/dpt_head_vggt_legacy.py:347
↓ 4 callersFunction_make_fusion_block_slam
(features, use_bn, size=None)
benchmark/models/lingbot_map/heads/dpt_head.py:602
↓ 4 callersMethod_make_layer
(self, dim, stride=1)
benchmark/models/page4d/dependency/track_modules/blocks.py:58
↓ 4 callersFunction_refresh_cameras
()
benchmark/models/r3/view.py:608
↓ 4 callersMethod_set_defer_eviction
Set defer-eviction flag on FlashInfer manager and SDPA caches. While True, eviction is suppressed so that rollback can cleanly undo t
benchmark/models/lingbot_map/models/gct_stream_window.py:362
↓ 4 callersMethod_set_skip_append
Set _skip_append flag on all KV caches (aggregator + camera head). When skip=True, attention layers will attend to [cached_kv + current_kv]
benchmark/models/lingbot_map/models/gct_stream_window.py:341
↓ 4 callersMethod_store_keyframe
Persist one keyframe and evict the most redundant entry at capacity.
benchmark/models/r3/R3/models/online/fallback.py:940
↓ 4 callersFunction_umeyama_sim3_from_paths
(pose_ref, pose_est)
benchmark/models/depth_anything_3/utils/pose_align.py:84
↓ 4 callersFunction_umeyama_sim3_from_paths
(pose_ref, pose_est)
benchmark/models/r3/depth_anything_3/utils/pose_align.py:84
↓ 4 callersMethodadd_cameras
(self, poses, focals=None, images=None, imsizes=None, colors=None, **kw)
benchmark/models/mast3r_root/dust3r/viz.py:202
↓ 4 callersMethodadd_voxel_feat_to_patch_tokens
Add voxel features to patch tokens. Params: - patch_tokens: dict containing patch tokens - voxel_feat: voxel
benchmark/models/amb3r_root/amb3r/frontend.py:47
↓ 4 callersFunctionargmin
(X, dim)
benchmark/models/mast3r_root/mast3r/fast_nn.py:33
↓ 4 callersFunctioncallback
(_)
benchmark/models/loger/utils/viser_utils.py:142
↓ 4 callersMethodclean_kv_cache
Clean KV cache in aggregator. Call this method when starting a new video sequence to clear cached key-value pairs from previ
benchmark/models/lingbot_map/models/gct_stream.py:284
↓ 4 callersMethodclean_kv_cache
Clean KV cache in aggregator. Call this method when starting a new video sequence to clear cached key-value pairs from previ
benchmark/models/lingbot_map/models/gct_stream_window.py:325
↓ 4 callersMethodcompute_global_alignment
(self, init=None, niter_PnP=10, **kw)
benchmark/models/dust3r_root/dust3r/cloud_opt/base_opt.py:276
↓ 4 callersFunctioncustom_interpolate
Safe interpolation implementation to avoid INT_MAX overflow in torch.nn.functional.interpolate.
benchmark/models/depth_anything_3/model/utils/head_utils.py:205
↓ 4 callersFunctioncustom_interpolate
Safe interpolation implementation to avoid INT_MAX overflow in torch.nn.functional.interpolate.
benchmark/models/r3/depth_anything_3/model/utils/head_utils.py:205
↓ 4 callersFunctiondrop_add_residual_stochastic_depth
( x: Tensor, residual_func: Callable[[Tensor], Tensor], sample_drop_ratio: float = 0.0, )
benchmark/models/loger/models/layers/block.py:114
↓ 4 callersFunctiondrop_add_residual_stochastic_depth
( x: Tensor, residual_func: Callable[[Tensor], Tensor], sample_drop_ratio: float = 0.0, )
benchmark/models/pi3/models/layers/block.py:114
↓ 4 callersFunctiondrop_add_residual_stochastic_depth
( x: Tensor, residual_func: Callable[[Tensor], Tensor], sample_drop_ratio: float = 0.0, pos=None )
benchmark/models/zipmap/zipmap/layers/block_ttt.py:124
↓ 4 callersMethodevict_frame
Return a frame's pages to the free list and drop its page_table entry.
benchmark/models/r3/R3/models/online/flashinfer_backend.py:152
↓ 4 callersMethodevict_frames
Evict old window patch pages (recycle to free list). Special pages are NEVER evicted. Scale pages are NEVER evicted.
benchmark/models/lingbot_map/layers/flashinfer_cache.py:229
↓ 4 callersFunctionexists
(val)
benchmark/models/streamvggt/heads/track_modules/modules.py:21
↓ 4 callersFunctionfind_homography_least_squares_weighted_torch
src_pts: (N,2) source points (torch.Tensor, float32/float64) dst_pts: (N,2) target points (torch.Tensor, float32/float64) confident_weigh
benchmark/models/depth_anything_3/utils/ray_utils.py:112
↓ 4 callersFunctionfind_homography_least_squares_weighted_torch
src_pts: (N,2) source points (torch.Tensor, float32/float64) dst_pts: (N,2) target points (torch.Tensor, float32/float64) confident_weigh
benchmark/models/r3/depth_anything_3/utils/ray_utils.py:112
↓ 4 callersFunctiongamma_loss
(gamma, mul=1, offset=None, clip=np.inf)
benchmark/models/mast3r_root/mast3r/cloud_opt/utils/losses.py:19
↓ 4 callersMethodget_conf_log
(self, x)
benchmark/models/mast3r_root/dust3r/losses.py:217
↓ 4 callersFunctionget_depth_loss
depth0: predicted depth map (B, T, H, W, 1) gt_depth: ground truth depth map (B, T, H, W, 1) valid_mask: binary mask for valid pixels (B,
benchmark/models/amb3r_root/amb3r/loss.py:14
↓ 4 callersFunctionget_distributed
()
benchmark/models/scal3r/utils/dist_utils.py:21
↓ 4 callersMethodget_intrinsics
(self)
benchmark/models/dust3r_root/dust3r/cloud_opt/pair_viewer.py:104
↓ 4 callersFunctionget_joint_pointcloud_center_scale
(pts1, pts2, valid_mask1=None, valid_mask2=None, z_only=False, center=True)
benchmark/models/mast3r_root/dust3r/utils/geometry.py:328
↓ 4 callersFunctionget_joint_pointcloud_depth
(z1, z2, valid_mask1, valid_mask2=None, quantile=0.5)
benchmark/models/mast3r_root/dust3r/utils/geometry.py:313
↓ 4 callersFunctionget_point_loss
depth0: predicted depth map (B, T, H, W, 3) gt_depth: ground truth depth map (B, T, H, W, 3) valid_mask: binary mask for valid pixels (B,
benchmark/models/amb3r_root/amb3r/loss.py:49
↓ 4 callersFunctionget_pred_pts3d
(gt, pred, use_pose=False)
benchmark/models/mast3r_root/dust3r/inference.py:81
↓ 4 callersMethodget_principal_points
(self)
benchmark/models/mast3r_root/mast3r/cloud_opt/sparse_ga.py:62
↓ 4 callersMethodget_pts3d
(self, raw=False)
benchmark/models/mast3r_root/dust3r/cloud_opt/base_opt.py:203
↓ 4 callersFunctionget_rank
()
benchmark/models/scal3r/utils/dist_utils.py:33
↓ 4 callersFunctionget_release_root
()
benchmark/models/scal3r/engine/path.py:5
↓ 4 callersMethodget_submap
(self, submap_id)
benchmark/models/scal3r/utils/pgo/optimizer.py:16
↓ 4 callersMethodget_voxel_feat
Process front-end feature with backend to obtain voxel features Params: - res: dict containing front-end feature
benchmark/models/amb3r_root/amb3r/model.py:100
↓ 4 callersFunctionglobal_aligner
(dust3r_output, device, mode=GlobalAlignerMode.PointCloudOptimizer, **optim_kw)
benchmark/models/dust3r_root/dust3r/cloud_opt/__init__.py:20
↓ 4 callersFunctionhash_md5
(s)
benchmark/models/mast3r_root/mast3r/utils/misc.py:16
↓ 4 callersFunctionhmat_to_pose_encoding
Convert homogeneous transform matrices back to pose encoding.
benchmark/models/r3/R3/utils/pose_utils.py:22
↓ 4 callersFunctionhomogenize_points
Convert batched points (xyz) to (xyz1).
benchmark/models/depth_anything_3/utils/geometry.py:241
↓ 4 callersFunctionhomogenize_points
Convert batched points (xyz) to (xyz1).
benchmark/models/r3/depth_anything_3/utils/geometry.py:241
↓ 4 callersMethodinfer
Inference forward pass. Args: images: Input images of shape [#images, 3, H, W] in range [0, 1]. Returns: Dic
benchmark/models/vgg_ttt/vggttt/nets/vggt/models/vggt.py:256
↓ 4 callersFunctioninterleave
(tensor1, tensor2)
benchmark/models/mast3r_root/dust3r/utils/misc.py:48
↓ 4 callersMethodload_weights
Model weights loading & data type preparation
benchmark/models/amb3r_root/amb3r/model.py:57
↓ 4 callersFunctionmake_K
(w, h, fx=1200.0, fy=1100.0)
benchmark/models/depth_anything_3/utils/io/input_processor.py:431
↓ 4 callersFunctionmake_K
(w, h, fx=1200.0, fy=1100.0)
benchmark/models/r3/depth_anything_3/utils/io/input_processor.py:431
↓ 4 callersFunctionmake_fusion_block
(features, use_bn, width_ratio=1)
benchmark/models/stream3r/croco/models/dpt_block.py:253
↓ 4 callersFunctionmake_fusion_block
(features, use_bn, width_ratio=1)
benchmark/models/dust3r_root/croco/models/dpt_block.py:220
↓ 4 callersFunctionmake_fusion_block
(features, use_bn, width_ratio=1)
benchmark/models/mast3r_root/croco/models/dpt_block.py:220
↓ 4 callersFunctionmat_to_quat
Convert rotations given as rotation matrices to quaternions. Args: matrix: Rotation matrices as tensor of shape (..., 3, 3). Re
benchmark/models/lingbot_map/utils/rotation.py:47
↓ 4 callersFunctionnormalize_feature_batch
(feature)
benchmark/models/page4d/models/mask_utils.py:6
↓ 4 callersFunctionnormalize_pointcloud
renorm pointmaps pts1, pts2 with norm_mode
benchmark/models/mast3r_root/dust3r/utils/geometry.py:249
↓ 4 callersFunctionoffload_payload
(payload, path: str, kind: str)
benchmark/models/scal3r/utils/offload_utils.py:61
↓ 4 callersFunctionpost_all2all
(local_seq_2_local_head, seq_world_size)
benchmark/models/vgg_ttt/vggttt/utils/dist_att.py:26
← previousnext →201–300 of 5,295, ranked by callers