MCPcopy Create free account

hub / github.com/VAST-AI-Research/MIDI-3D / functions

Functions380 in github.com/VAST-AI-Research/MIDI-3D

⨍ Functions 380◇ Types & classes 47↳ Endpoints 1
↓ 14 callersMethodget_save_path
(self, filename)
midi/utils/saving.py:51
↓ 7 callersMethoddo_update_step
( self, epoch: int, global_step: int, on_load_weights: bool = False )
midi/utils/base.py:22
↓ 6 callersMethodconvert_data
(self, data)
midi/utils/saving.py:32
↓ 6 callersMethodget_save_dir
(self)
midi/utils/saving.py:27
↓ 6 callersFunctionparse_structured
(fields: Any, cfg: Optional[Union[dict, DictConfig]] = None)
midi/utils/misc.py:35
↓ 6 callersMethodwrite
(self, msg: str)
midi/utils/callbacks.py:136
↓ 5 callersMethoddo_update_step_end
(self, epoch: int, global_step: int)
midi/utils/base.py:38
↓ 5 callersFunctionwarn
(*args, **kwargs)
midi/utils/core.py:26
↓ 4 callersFunction_parse_scene_list
( scene_list_path: Union[str, List[str]], root_data_dir: Union[str, List[str]] )
midi/data/multi_object.py:55
↓ 4 callersFunctioncleanup
()
midi/utils/misc.py:190
↓ 4 callersMethodinit_custom_adapter
(self, *args, **kwargs)
midi/loaders/custom_adapter.py:13
↓ 4 callersMethodpreprocess_data
(self, batch, stage)
midi/systems/base.py:180
↓ 4 callersMethodset_attn_processor
r""" Sets the attention processor to use to compute attention. Parameters: processor (`dict` of `AttentionProcessor` or o
midi/models/transformers/triposg_transformer.py:572
↓ 4 callersFunctionupdate_end_if_possible
(module: Any, epoch: int, global_step: int)
midi/utils/base.py:65
↓ 4 callersFunctionupdate_if_possible
(module: Any, epoch: int, global_step: int)
midi/utils/base.py:60
↓ 3 callersMethoddecode
( self, z: torch.Tensor, sampled_points: torch.Tensor, return_dict: bool = Tru
midi/models/autoencoders/autoencoder_kl_triposg.py:505
↓ 3 callersMethodenable
(self, enabled: bool)
midi/utils/misc.py:250
↓ 3 callersFunctionget_device
()
midi/utils/misc.py:113
↓ 3 callersMethodget_rgb_image_
(self, img, data_format, data_range, rgba=False)
midi/utils/saving.py:65
↓ 3 callersMethodload_image
( self, path, height, width, background_color, rescale: bool =
midi/data/multi_object.py:262
↓ 3 callersMethodset_attn_processor
r""" Sets the attention processor to use to compute attention. Parameters: processor (`dict` of `AttentionProcessor` or o
midi/models/autoencoders/autoencoder_kl_triposg.py:338
↓ 3 callersFunctionset_system_status
(system: BaseSystem, ckpt_path: Optional[str])
launch.py:181
↓ 3 callersMethodset_timesteps
Sets the discrete timesteps used for the diffusion chain (to be run before inference). Args: num_inference_steps (`int`)
midi/schedulers/scheduling_rectified_flow.py:166
↓ 3 callersFunctionto_pil_image
(image: Float[Tensor, "B C H W"])
midi/systems/model_utils.py:45
↓ 2 callersMethod__init__
( self, cfg: Optional[Union[dict, DictConfig]] = None, *args, **kwargs )
midi/utils/base.py:77
↓ 2 callersMethod__init__
(self, save_path)
midi/utils/callbacks.py:124
↓ 2 callersMethod__init__
( self, in_channels: int = 3, dim: int = 512, num_attention_heads: int = 8,
midi/models/autoencoders/autoencoder_kl_triposg.py:27
↓ 2 callersMethod__len__
(self)
midi/data/multi_object.py:204
↓ 2 callersMethod_decode
( self, z: torch.Tensor, sampled_points: torch.Tensor, num_chunks: int = 50000
midi/models/autoencoders/autoencoder_kl_triposg.py:474
↓ 2 callersFunction_distributed_available
()
midi/utils/misc.py:210
↓ 2 callersMethod_encode
( self, x: torch.Tensor, num_tokens: int = 2048, seed: Optional[int] = None )
midi/models/autoencoders/autoencoder_kl_triposg.py:432
↓ 2 callersMethod_getitem
(self, index)
midi/data/multi_object.py:596
↓ 2 callersFunctioncompute_chamfer_distance
Compute Chamfer Distance between predicted and ground truth point clouds. Args: pred (torch.Tensor): Predicted point cloud of shape
midi/utils/metrics.py:8
↓ 2 callersFunctioncompute_fscore
Compute F-Score between predicted and ground truth point clouds. Args: pred (torch.Tensor): Predicted point cloud of shape (B, N, 3)
midi/utils/metrics.py:34
↓ 2 callersFunctioncreate_palette
()
scripts/grounding_sam.py:17
↓ 2 callersFunctiondetect
Use Grounding DINO to detect a set of labels in an image in a zero-shot fashion.
scripts/grounding_sam.py:255
↓ 2 callersMethodenable_slicing
r""" Enable sliced VAE decoding. When this option is enabled, the VAE will split the input tensor in slices to compute decoding in sev
midi/models/autoencoders/autoencoder_kl_triposg.py:380
↓ 2 callersFunctionfind
(cls_string)
midi/utils/core.py:13
↓ 2 callersMethodfrom_dict
(cls, detection_dict: Dict)
scripts/grounding_sam.py:122
↓ 2 callersFunctiongenerate_dense_grid_points
( bbox_min: np.ndarray, bbox_max: np.ndarray, octree_depth: int, indexing: str = "ij" )
midi/inference_utils.py:9
↓ 2 callersMethodget_bg_color
(self, bg_color)
midi/data/multi_object.py:207
↓ 2 callersMethodget_grayscale_image_
(self, img, data_range, cmap)
midi/utils/saving.py:166
↓ 2 callersFunctionget_obj_from_str
(string, reload=False)
midi/utils/misc.py:70
↓ 2 callersFunctionget_rank
()
midi/utils/misc.py:102
↓ 2 callersMethodget_uv_image_
(self, img, data_format, data_range, cmap)
midi/utils/saving.py:130
↓ 2 callersFunctionicp
Perform ICP algorithm to align src to dst. Args: src (torch.Tensor): Source point cloud of shape (B, N, 3) dst (torch.Tensor
midi/utils/metrics.py:248
↓ 2 callersFunctionload_module_weights
( path, module_name=None, ignore_modules=None, mapping=None, map_location=None )
midi/utils/misc.py:117
↓ 2 callersMethodload_surface
(self, path, num_pc: int = 20480)
midi/data/multi_object.py:227
↓ 2 callersFunctionmain
(args, extras)
launch.py:43
↓ 2 callersFunctionmeshlab_to_mesh
(ms)
midi/utils/mesh_process.py:24
↓ 2 callersFunctionnnsearch
midi/utils/chamfer_distance/chamfer_distance.cpp:59
↓ 2 callersFunctionparse_version
(ver: str)
midi/utils/misc.py:98
↓ 2 callersFunctionplot_segmentation
(image, detections)
scripts/grounding_sam.py:226
↓ 2 callersFunctionprepare_ig2mv_pipeline
(device, dtype)
scripts/image_to_textured_scene.py:22
↓ 2 callersFunctionprepare_model
( device: str = "cuda", detector_id: Optional[str] = None, segmenter_id: Optional[str] = None, )
scripts/grounding_sam.py:236
↓ 2 callersFunctionprepare_texture_pipeline
(device, dtype)
scripts/image_to_textured_scene.py:36
↓ 2 callersFunctionpreprocess_image
(image: Image.Image, height, width)
scripts/mvadapter_ig2mv.py:91
↓ 2 callersFunctionrepair_non_manifold
(ms, verbose=False)
midi/utils/mesh_process.py:121
↓ 2 callersFunctionrun_i2tex
( ig2mv_pipe: Any, texture_pipe: Any, scene: trimesh.Scene, rgb_image: Union[str, Image.Image]
scripts/image_to_textured_scene.py:71
↓ 2 callersFunctionrun_midi
( pipe: Any, rgb_image: Union[str, Image.Image], seg_image: Union[str, Image.Image], seed: int
scripts/inference_midi.py:116
↓ 2 callersMethodsave_image
(self, filename, img)
midi/utils/saving.py:307
↓ 2 callersFunctionsegment
Use Segment Anything (SAM) to generate masks given an image + a set of bounding boxes.
scripts/grounding_sam.py:272
↓ 2 callersMethodset_chunk_feed_forward
(self, chunk_size: Optional[int], dim: int = 0)
midi/models/transformers/triposg_transformer.py:278
↓ 2 callersFunctionset_transformer_attn_processor
( transformer: TripoSGDiTModel, set_self_attn_proc_func: Callable = default_set_attn_proc_func, se
midi/models/transformers/__init__.py:15
↓ 2 callersFunctionsigned_distance_function_gpu
(levelset: torch.Tensor, band_radius: int)
midi/utils/smoothing.py:364
↓ 2 callersFunctiontaubin_smooth
(ms, stepsmoothnum=3, verbose=False)
midi/utils/mesh_process.py:147
↓ 2 callersMethodtest_dataloader
(self)
midi/data/multi_object.py:668
↓ 2 callersFunctionvoxelize
Voxelize the point cloud. Args: points (torch.Tensor): Point cloud of shape (B, N, 3) voxel_size (float): Size of each voxel
midi/utils/metrics.py:94
↓ 1 callersFunctionC
(value: Any, epoch: int, global_step: int)
midi/utils/misc.py:166
↓ 1 callersMethod__init__
(self, cfg: Any, split: str = "train")
midi/data/multi_object.py:161
↓ 1 callersMethod__init__
( self, dim: int, num_attention_heads: int, use_self_attention: bool = True,
midi/models/transformers/triposg_transformer.py:161
↓ 1 callersFunction_build_variable_indices
(band: np.ndarray)
midi/utils/smoothing.py:56
↓ 1 callersFunction_buildq2d
Builds the filterq matrix for the given variables. Version for 2 dimensions.
midi/utils/smoothing.py:216
↓ 1 callersFunction_buildq3d
Builds the filterq matrix for the given variables.
midi/utils/smoothing.py:63
↓ 1 callersMethod_get_next_version
(self)
midi/utils/callbacks.py:36
↓ 1 callersMethod_getitem_mix
(self, index)
midi/data/multi_object.py:536
↓ 1 callersMethod_getitem_scene
(self, index)
midi/data/multi_object.py:384
↓ 1 callersMethod_init_custom_adapter
(self, *args, **kwargs)
midi/loaders/custom_adapter.py:16
↓ 1 callersMethod_init_step_index
(self, timestep)
midi/schedulers/scheduling_rectified_flow.py:226
↓ 1 callersFunction_jacobi
Jacobi method with constraints.
midi/utils/smoothing.py:265
↓ 1 callersMethod_load_custom_adapter
(self, state_dict)
midi/loaders/custom_adapter.py:65
↓ 1 callersFunction_parse_object_list
(object_list_path: Union[str, List[str]])
midi/data/multi_object.py:29
↓ 1 callersFunction_parse_object_list_single
(object_list_path: str)
midi/data/multi_object.py:18
↓ 1 callersFunction_parse_scene_list_single
(scene_list_path: str, root_data_dir: str)
midi/data/multi_object.py:38
↓ 1 callersMethod_sample_features
Sample points from features of the input point cloud. Args: x (torch.Tensor): The input point cloud. shape: (B, N, C)
midi/models/autoencoders/autoencoder_kl_triposg.py:395
↓ 1 callersMethod_save_custom_adapter
(self)
midi/loaders/custom_adapter.py:98
↓ 1 callersMethod_save_grayscale_image
( self, filename, img, data_range, cmap, name: Optional[str] =
midi/utils/saving.py:210
↓ 1 callersMethod_save_rgb_image
( self, filename, img, data_format, data_range, name: Optional
midi/utils/saving.py:101
↓ 1 callersMethod_set_time_proj
( self, time_embedding_type: str, inner_dim: int, flip_sin_to_cos: bool,
midi/models/transformers/triposg_transformer.py:467
↓ 1 callersMethod_t_to_sigma
(self, timestep)
midi/schedulers/scheduling_rectified_flow.py:157
↓ 1 callersMethodcompute_metrics
(self, batch, outputs, use_icp: bool = True)
midi/systems/system_midi.py:231
↓ 1 callersFunctioncompute_nearest_neighbors
Compute the nearest neighbors from src to dst using PyTorch3D's knn_points. Args: src (torch.Tensor): Source point cloud of shape (B
midi/utils/metrics.py:188
↓ 1 callersFunctioncompute_normal
(ms, weightmode="Simple Average", verbose=False)
midi/utils/mesh_process.py:157
↓ 1 callersFunctioncompute_rigid_transform
Compute the rigid transformation (R, t) that aligns A to B. Args: A (torch.Tensor): Source point cloud of shape (B, N, 3) B
midi/utils/metrics.py:205
↓ 1 callersFunctioncompute_volume_iou
Compute Volume IoU between predicted and ground truth point clouds. Args: pred (torch.Tensor): Predicted point cloud of shape (B, N,
midi/utils/metrics.py:126
↓ 1 callersFunctionconfig_to_primitive
(config, resolve: bool = True)
midi/utils/config.py:129
↓ 1 callersMethodconfigure
(self, *args, **kwargs)
midi/utils/base.py:85
↓ 1 callersMethodconfigure
(self, *args, **kwargs)
midi/utils/base.py:115
↓ 1 callersMethodconfigure
(self)
midi/systems/base.py:105
↓ 1 callersMethodcrop_and_pad
(self, rgbs, masks, height, width, padding_ratio=0.1)
midi/data/multi_object.py:342
next →1–100 of 380, ranked by callers