MCPcopy Create free account

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

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

↓ 1 callersMethoddecode_latents
( self, latents: torch.Tensor, sampled_points: torch.Tensor, decode_progressiv
midi/pipelines/pipeline_midi.py:233
↓ 1 callersMethoddisable_slicing
r""" Disable sliced VAE decoding. If `enable_slicing` was previously enabled, this method will go back to computing decoding in one st
midi/models/autoencoders/autoencoder_kl_triposg.py:388
↓ 1 callersFunctiondot
(x, y)
midi/utils/ops.py:15
↓ 1 callersFunctiondump_config
(path: str, config)
midi/utils/config.py:133
↓ 1 callersMethodenable_vae_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/pipelines/pipeline_utils.py:11
↓ 1 callersMethodencode_image_1
(self, image, device, num_images_per_prompt)
midi/pipelines/pipeline_midi.py:151
↓ 1 callersMethodencode_image_2
( self, image_one, image_two, mask, device, num_images_per_pro
midi/pipelines/pipeline_midi.py:166
↓ 1 callersFunctionfix_hole
(ms, maxholesize=30, verbose=False)
midi/utils/mesh_process.py:107
↓ 1 callersMethodforward
(self, xyz1, xyz2)
midi/utils/chamfer_distance/chamfer_distance.py:68
↓ 1 callersFunctiongenerate_colored_segmentation
(label_image)
scripts/grounding_sam.py:215
↓ 1 callersMethodgenerate_samples
( self, batch, return_dict: bool = True, **kwargs )
midi/systems/system_midi.py:182
↓ 1 callersFunctionget_boxes
(results: DetectionResult)
scripts/grounding_sam.py:185
↓ 1 callersMethodget_dims
(self, input_dim)
midi/models/embeddings.py:70
↓ 1 callersMethodget_file_list
(self)
midi/utils/callbacks.py:66
↓ 1 callersMethodget_image_grid_
(self, imgs, align)
midi/utils/saving.py:239
↓ 1 callersFunctionget_naming_convention
(cfg)
midi/utils/config.py:36
↓ 1 callersMethodget_output_visualizations
(self, batch, outputs)
midi/systems/system_midi.py:152
↓ 1 callersFunctionget_parameters
(model, name)
midi/utils/training_utils.py:21
↓ 1 callersFunctionget_scheduler
(name)
midi/utils/training_utils.py:8
↓ 1 callersFunctiongetattr_recursive
(m, attr)
midi/utils/training_utils.py:15
↓ 1 callersFunctiongrounded_segmentation
( object_detector, processor, segmentator, image: Union[Image.Image, str], labels: Union[s
scripts/grounding_sam.py:313
↓ 1 callersMethodindex_for_timestep
(self, timestep, schedule_timesteps=None)
midi/schedulers/scheduling_rectified_flow.py:212
↓ 1 callersFunctionis_flash3_available
()
midi/utils/import_utils.py:27
↓ 1 callersFunctionis_sdpa_available
()
midi/utils/import_utils.py:31
↓ 1 callersFunctionload_config
( *yamls: str, cli_args: list = [], from_string=False, makedirs=True, **kwargs )
midi/utils/config.py:79
↓ 1 callersMethodload_custom_adapter
( self, pretrained_model_name_or_path_or_dict: Union[str, Dict[str, torch.Tensor]], we
midi/loaders/custom_adapter.py:19
↓ 1 callersFunctionload_image
(image_str: str)
scripts/grounding_sam.py:176
↓ 1 callersMethodload_parts
( self, rgb_path: str, idmap_path: str, indexes: List[int], height: in
midi/data/multi_object.py:298
↓ 1 callersMethodload_weights
( self, weights: str, ignore_modules: Optional[List[str]] = None, mapping: Opt
midi/systems/base.py:64
↓ 1 callersFunctionmask_to_polygon
(mask: np.ndarray)
scripts/grounding_sam.py:136
↓ 1 callersFunctionmerge_close_vertices
(ms, threshold=0.0001, verbose=False)
midi/utils/mesh_process.py:71
↓ 1 callersFunctionmesh_to_meshlab
(vertices, faces)
midi/utils/mesh_process.py:17
↓ 1 callersMethodon_check_train
(self, batch, outputs, **kwargs)
midi/systems/base.py:149
↓ 1 callersFunctionparse_optimizer
(config, model)
midi/utils/training_utils.py:30
↓ 1 callersFunctionparse_scheduler
(config, optimizer)
midi/utils/training_utils.py:70
↓ 1 callersFunctionparse_structured
(fields: Any, cfg: Optional[Union[dict, DictConfig]] = None)
midi/utils/config.py:138
↓ 1 callersFunctionpolygon_to_mask
Convert a polygon to a segmentation mask. Args: - polygon (list): List of (x, y) coordinates representing the vertices of the polygon.
scripts/grounding_sam.py:151
↓ 1 callersMethodpost_configure
executed after weights are loaded
midi/systems/base.py:108
↓ 1 callersMethodprepare_latents
( self, batch_size, num_tokens, num_channels_latents, dtype, d
midi/pipelines/pipeline_midi.py:207
↓ 1 callersFunctionprepare_midi_pipeline
(device, dtype)
scripts/image_to_textured_scene.py:18
↓ 1 callersFunctionprepare_pipeline
( base_model, vae_model, unet_model, lora_model, adapter_path, scheduler, num_view
scripts/mvadapter_ig2mv.py:24
↓ 1 callersFunctionprepare_pipeline
(device, dtype)
scripts/inference_midi.py:18
↓ 1 callersFunctionpreprocess_image
(rgb_image, seg_image)
scripts/inference_midi.py:34
↓ 1 callersFunctionprocess_chunk
(start, end)
midi/utils/smoothing.py:140
↓ 1 callersFunctionprocess_mesh
( vertices, faces, threshold=0.0001, mincomponentRatio=0.02, targetfacenum=50000, maxh
midi/utils/mesh_process.py:167
↓ 1 callersFunctionprocess_raw
(mesh_path, save_path, preprocess=True, device="cpu")
midi/utils/mesh_process.py:255
↓ 1 callersMethodquery_geometry
( self, model_fn: callable, queries: torch.Tensor, sample: torch.Tensor,
midi/models/autoencoders/autoencoder_kl_triposg.py:148
↓ 1 callersFunctionrandom_morphological_transform
Randomly dilate or erode the mask :param mask: [H, W] 0-1 mask :param max_kernel_size: Maximum kernel size; controls the scale of the mor
midi/data/multi_object.py:68
↓ 1 callersFunctionrefine_masks
( masks: torch.BoolTensor, polygon_refinement: bool = False )
scripts/grounding_sam.py:194
↓ 1 callersFunctionremove_bg
(image, net, transform, device)
scripts/mvadapter_ig2mv.py:79
↓ 1 callersFunctionremove_isolated_pieces
(ms, mincomponentsize=25, diameter=None, verbose=False)
midi/utils/mesh_process.py:85
↓ 1 callersFunctionretrieve_timesteps
Calls the scheduler's `set_timesteps` method and retrieves timesteps from the scheduler after the call. Handles custom timesteps. Any kwargs
midi/pipelines/pipeline_midi.py:35
↓ 1 callersFunctionrun_midi
( midi_pipe: MIDIPipeline, rgb_image: Union[str, Image.Image], seg_image: Union[str, Image.Image],
scripts/image_to_textured_scene.py:58
↓ 1 callersFunctionrun_pipeline
( pipe, mesh_path, num_views, text, image, height, width, num_inference_steps,
scripts/mvadapter_ig2mv.py:122
↓ 1 callersMethodsample
(self, generator: Optional[torch.Generator] = None)
midi/models/autoencoders/vae.py:27
↓ 1 callersMethodsave_code_snapshot
(self)
midi/utils/callbacks.py:82
↓ 1 callersMethodsave_config_snapshot
(self)
midi/utils/callbacks.py:106
↓ 1 callersMethodsave_file
(self, filename, src_path, delete=False)
midi/utils/saving.py:435
↓ 1 callersMethodsave_image_grid
( self, filename, imgs, align=DEFAULT_GRID_KWARGS["align"], name: Opti
midi/utils/saving.py:277
↓ 1 callersMethodsave_img_sequence
( self, filename, img_dir, matcher, save_format="mp4", fps=30,
midi/utils/saving.py:372
↓ 1 callersMethodset_resume_status
(self, current_epoch: int, global_step: int)
midi/systems/base.py:80
↓ 1 callersMethodset_save_dir
(self, save_dir: str)
midi/utils/saving.py:24
↓ 1 callersMethodsetup
(self, stage=None)
midi/data/multi_object.py:639
↓ 1 callersFunctionsmooth_constrained
Implementation of the smoothing method from "Surface Extraction from Binary Volumes with Higher-Order Smoothness" Victor Lempitsky, CVPR
midi/utils/smoothing.py:394
↓ 1 callersFunctionsmooth_constrained_gpu
( binary_array: torch.Tensor, band_radius: int = 4, max_iters: int = 250, rel_tol: float = 1e-
midi/utils/smoothing.py:467
↓ 1 callersFunctionsmooth_gaussian
(binary_array: np.ndarray, sigma: float = 3)
midi/utils/smoothing.py:551
↓ 1 callersFunctionsmooth_gaussian_gpu
(binary_array: torch.Tensor, sigma: float = 3)
midi/utils/smoothing.py:556
↓ 1 callersFunctionsmooth_gpu
(binary_array: torch.Tensor, method: str = "auto", **kwargs)
midi/utils/smoothing.py:634
↓ 1 callersFunctionsplit_rgb_mask
(rgb_image, seg_image)
scripts/inference_midi.py:81
↓ 1 callersMethodstep
Predict the sample from the previous timestep by reversing the SDE. This function propagates the diffusion process from the learned m
midi/schedulers/scheduling_rectified_flow.py:234
↓ 1 callersMethodtime_shift
(self, t: torch.Tensor)
midi/schedulers/scheduling_rectified_flow.py:163
↓ 1 callersMethodtime_shift_dynamic
(self, mu: float, sigma: float, t: torch.Tensor)
midi/schedulers/scheduling_rectified_flow.py:160
↓ 1 callersFunctiontotal_variation_denoising
(x, weight=0.1, num_iterations=5, eps=1e-8)
midi/utils/smoothing.py:453
↓ 1 callersMethodupdate_step
(self, epoch: int, global_step: int, on_load_weights: bool = False)
midi/utils/base.py:50
↓ 1 callersMethodupdate_step_end
(self, epoch: int, global_step: int)
midi/utils/base.py:56
↓ 1 callersFunctionuv_parameterize_uvatlas
( vertices, faces, size=1024, gutter=2.5, max_stretch=0.1666666716337204, parallel_par
midi/utils/mesh_process.py:223
↓ 1 callersMethodvalidate
(variable)
midi/utils/typing.py:52
↓ 1 callersFunctionwrite_to_text
(file, lines)
launch.py:143
MethodC
(self, value: Any)
midi/systems/base.py:114
FunctionPYBIND11_MODULE
midi/utils/chamfer_distance/chamfer_distance.cpp:180
Method__call__
( self, image: PipelineImageInput, mask: PipelineImageInput, image_scene: Pipe
midi/pipelines/pipeline_midi.py:314
Method__call__
( self, attn: Attention, hidden_states: torch.Tensor, encoder_hidden_states: O
midi/models/attention_processor.py:27
Method__call__
( self, attn: Attention, hidden_states: torch.Tensor, encoder_hidden_states: O
midi/models/attention_processor.py:156
Method__call__
( self, attn: Attention, hidden_states: torch.Tensor, encoder_hidden_states: O
midi/models/attention_processor.py:278
Method__getitem__
(self, index)
midi/data/multi_object.py:606
Method__init__
(self)
launch.py:32
Method__init__
(self)
midi/utils/misc.py:237
Method__init__
(self, cfg: Optional[dict] = None)
midi/utils/base.py:16
Method__init__
( self, cfg: Optional[Union[dict, DictConfig]] = None, *args, **kwargs )
midi/utils/base.py:96
Method__init__
(self, save_root, version=None, use_version=True)
midi/utils/callbacks.py:20
Method__init__
(self, save_root, version=None, use_version=True)
midi/utils/callbacks.py:63
Method__init__
(self, config_path, config, save_root, version=None, use_version=True)
midi/utils/callbacks.py:100
Method__init__
(self, cfg, resumed=False)
midi/systems/base.py:40
Method__init__
(self, cfg: Optional[Union[dict, DictConfig]] = None)
midi/data/multi_object.py:635
Method__init__
( self, vae: TripoSGVAEModel, transformer: TripoSGDiTModel, scheduler: FlowMat
midi/pipelines/pipeline_midi.py:105
Method__init__
( self, num_train_timesteps: int = 1000, shift: float = 1.0, use_dynamic_shift
midi/schedulers/scheduling_rectified_flow.py:100
Method__init__
(self)
midi/models/attention_processor.py:21
Method__init__
(self)
midi/models/attention_processor.py:150
Method__init__
(self, use_mi: bool = True)
midi/models/attention_processor.py:270
Method__init__
The initialization
midi/models/embeddings.py:42
← previousnext →101–200 of 380, ranked by callers