MCPcopy Create free account

hub / github.com/Tencent-Hunyuan/Hunyuan3D-2 / functions

Functions375 in github.com/Tencent-Hunyuan/Hunyuan3D-2

⨍ Functions 375◇ Types & classes 89↳ Endpoints 2
↓ 70 callersMethodto
(self, device=None, dtype=None)
hy3dgen/shapegen/pipelines.py:300
↓ 26 callersMethodto
(self, device)
hy3dgen/texgen/hunyuanpaint/pipeline.py:153
↓ 24 callersMethodto
(self, device)
hy3dgen/texgen/differentiable_renderer/mesh_render.py:301
↓ 22 callersMethodfrom_pretrained
(cls, model_path, subfolder='hunyuan3d-paint-v2-0-turbo')
hy3dgen/texgen/pipelines.py:55
↓ 11 callersMethod__init__
( self, *, width: int, expand_ratio: int = 4, output_width: int = None,
hy3dgen/shapegen/models/autoencoders/attention_blocks.py:167
↓ 9 callersMethod__init__
(self, approximate='tanh')
hy3dgen/shapegen/models/denoisers/hunyuan3ddit.py:64
↓ 9 callersMethodraster_interpolate
(self, uv, rast_out, uv_idx, rast_db=None, diff_attrs=None)
hy3dgen/texgen/differentiable_renderer/mesh_render.py:179
↓ 8 callersMethod__init__
(self, *, width: int)
hy3dgen/shapegen/models/denoisers/hunyuandit.py:114
↓ 8 callersMethodfrom_pretrained
(pretrained_model_name_or_path, **kwargs)
hy3dgen/texgen/hunyuanpaint/unet/modules.py:423
↓ 7 callersMethoddecode
(self, latents)
hy3dgen/shapegen/models/autoencoders/model.py:295
↓ 7 callersFunctionexport_mesh
(mesh, save_folder, textured=False, type='glb')
gradio_app.py:85
↓ 6 callersFunctionget_neighbor
根据指定轴进行位移并保持维度一致
hy3dgen/shapegen/models/autoencoders/volume_decoders.py:39
↓ 6 callersMethodraster_rasterize
(self, pos, tri, resolution, ranges=None, grad_db=True)
hy3dgen/texgen/differentiable_renderer/mesh_render.py:165
↓ 6 callersFunctionsafe_where
(neighbor)
hy3dgen/shapegen/models/autoencoders/volume_decoders.py:93
↓ 5 callersMethodenable_flashvdm
( self, enabled: bool = True, adaptive_kv_selection=True, topk_mode='mean',
hy3dgen/shapegen/pipelines.py:258
↓ 5 callersFunctionextract_into_tensor
(a, t, x_shape, N_gen)
hy3dgen/texgen/hunyuanpaint/pipeline.py:131
↓ 5 callersFunctiongen_save_folder
(max_size=200)
gradio_app.py:64
↓ 5 callersFunctioninstantiate_from_config
(config, **kwargs)
hy3dgen/shapegen/pipelines.py:120
↓ 5 callersFunctiontransform_pos
(mtx, pos, keepdim=False)
hy3dgen/texgen/differentiable_renderer/camera_utils.py:21
↓ 4 callersFunctionbuild_model_viewer_html
(save_folder, height=660, width=790, textured=False)
gradio_app.py:103
↓ 4 callersFunctionget_attribute_data
(gltf, accessor_index)
hy3dgen/texgen/custom_rasterizer/custom_rasterizer/io_glb.py:35
↓ 4 callersMethodraster_antialias
(self, color, rast, pos, tri, topology_hash=None, pos_gradient_boost=1.0)
hy3dgen/texgen/differentiable_renderer/mesh_render.py:203
↓ 4 callersFunctionscatter_add_nd_with_count
(input, count, indices, values, weights=None)
hy3dgen/texgen/differentiable_renderer/mesh_render.py:39
↓ 4 callersMethodset_timesteps
Sets the discrete timesteps used for the diffusion chain (to be run before inference). Args: num_inference_steps (`int`)
hy3dgen/shapegen/schedulers.py:181
↓ 4 callersMethodunconditional_embedding
(self, batch_size, **kwargs)
hy3dgen/shapegen/models/conditioner.py:106
↓ 3 callersMethod__init__
( self, version=None, config=None, use_cls_token=True, image_size=224,
hy3dgen/shapegen/models/conditioner.py:58
↓ 3 callersFunctionattention
(q: Tensor, k: Tensor, v: Tensor, **kwargs)
hy3dgen/shapegen/models/denoisers/hunyuan3ddit.py:33
↓ 3 callersFunctionbuild_image_encoder
(config)
hy3dgen/shapegen/models/conditioner.py:203
↓ 3 callersMethodencode_images
(self, images)
hy3dgen/texgen/hunyuanpaint/pipeline.py:231
↓ 3 callersFunctionexport_mesh
(input, output)
hy3dgen/shapegen/postprocessors.py:91
↓ 3 callersFunctionexport_to_trimesh
(mesh_output)
hy3dgen/shapegen/pipelines.py:95
↓ 3 callersFunctiongenerate_dense_grid_points
( bbox_min: np.ndarray, bbox_max: np.ndarray, octree_resolution: int, indexing: str = "ij", )
hy3dgen/shapegen/models/autoencoders/volume_decoders.py:122
↓ 3 callersFunctionget_mv_matrix
(elev, azim, camera_distance, center=None)
hy3dgen/texgen/differentiable_renderer/camera_utils.py:37
↓ 3 callersMethodget_pos_from_mvp
(self, elev, azim, camera_distance, center)
hy3dgen/texgen/differentiable_renderer/mesh_render.py:512
↓ 3 callersFunctionimport_mesh
(mesh: Union[pymeshlab.MeshSet, trimesh.Trimesh, Latent2MeshOutput, str])
hy3dgen/shapegen/postprocessors.py:104
↓ 3 callersFunctionlinear_grid_put_2d
(H, W, coords, values, return_count=False)
hy3dgen/texgen/differentiable_renderer/mesh_render.py:67
↓ 3 callersMethodprepare_latents
(self, batch_size, dtype, device, generator, latents=None)
hy3dgen/shapegen/pipelines.py:473
↓ 3 callersFunctionretrieve_timesteps
Calls the scheduler's `set_timesteps` method and retrieves timesteps from the scheduler after the call. Handles custom timesteps. Any kwargs
hy3dgen/shapegen/pipelines.py:35
↓ 3 callersMethodrun
(self, *args, **kwargs)
hy3dgen/shapegen/models/autoencoders/surface_extractors.py:47
↓ 3 callersMethodsample
(self)
hy3dgen/shapegen/models/autoencoders/model.py:48
↓ 3 callersFunctionsample_pointcloud
(mesh, num=200000)
hy3dgen/shapegen/surface_loaders.py:18
↓ 3 callersMethodstep
Predict the sample from the previous timestep by reversing the SDE. This function propagates the diffusion process from the learned m
hy3dgen/shapegen/schedulers.py:245
↓ 2 callersMethod_export
( self, latents, output_type='trimesh', box_v=1.01, mc_level=0.0,
hy3dgen/shapegen/pipelines.py:648
↓ 2 callersFunction_gen_shape
( caption=None, image=None, mv_image_front=None, mv_image_back=None, mv_image_left=None,
gradio_app.py:135
↓ 2 callersMethod_sigma_to_t
(self, sigma)
hy3dgen/shapegen/schedulers.py:175
↓ 2 callersFunctionarray_to_tensor
(np_array)
hy3dgen/shapegen/preprocessors.py:22
↓ 2 callersMethodback_project
(self, image, elev, azim, camera_distance=None, center=None, method=None)
hy3dgen/texgen/differentiable_renderer/mesh_render.py:653
↓ 2 callersMethodcompile
(self)
hy3dgen/shapegen/pipelines.py:253
↓ 2 callersMethodenable_flashvdm_decoder
( self, enabled: bool = True, adaptive_kv_selection=True, topk_mode='mean',
hy3dgen/shapegen/models/autoencoders/model.py:178
↓ 2 callersMethodencode
(self, surface, sample_posterior=True)
hy3dgen/shapegen/models/autoencoders/model.py:284
↓ 2 callersMethodencode_cond
(self, image, additional_cond_inputs, do_classifier_free_guidance, dual_guidance)
hy3dgen/shapegen/pipelines.py:424
↓ 2 callersFunctionextract_near_surface_volume_fn
(input_tensor: torch.Tensor, alpha: float)
hy3dgen/shapegen/models/autoencoders/volume_decoders.py:29
↓ 2 callersMethodfast_bake_texture
(self, textures, cos_maps)
hy3dgen/texgen/differentiable_renderer/mesh_render.py:783
↓ 2 callersFunctionfps
( src: torch.Tensor, batch: Optional[Tensor] = None, ratio: Optional[Union[Tensor, float]] = None,
hy3dgen/shapegen/models/autoencoders/attention_blocks.py:497
↓ 2 callersMethodfrom_pretrained
( cls, model_path, device='cuda', dtype=torch.float16, use_safetensors
hy3dgen/shapegen/pipelines.py:201
↓ 2 callersFunctionget_buffer_data
(gltf, buffer_view)
hy3dgen/texgen/custom_rasterizer/custom_rasterizer/io_glb.py:26
↓ 2 callersMethodget_guidance_scale_embedding
See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298 Args: timeste
hy3dgen/shapegen/pipelines.py:512
↓ 2 callersFunctionget_image_data
(gltf, image, folder)
hy3dgen/texgen/custom_rasterizer/custom_rasterizer/io_glb.py:62
↓ 2 callersFunctionget_node_transform
(node)
hy3dgen/texgen/custom_rasterizer/custom_rasterizer/io_glb.py:101
↓ 2 callersFunctionget_obj_from_str
(string, reload=False)
hy3dgen/shapegen/pipelines.py:112
↓ 2 callersMethodindex_for_timestep
(self, timestep, schedule_timesteps=None)
hy3dgen/shapegen/schedulers.py:223
↓ 2 callersMethodinit_attention
(self, unet, use_ma=False, use_ra=False, is_turbo=False)
hy3dgen/texgen/hunyuanpaint/unet/modules.py:458
↓ 2 callersMethodlatents2mesh
(self, latents: torch.FloatTensor, **kwargs)
hy3dgen/shapegen/models/autoencoders/model.py:171
↓ 2 callersMethodload_image
(self, image, border_ratio=0.15, to_tensor=True)
hy3dgen/shapegen/preprocessors.py:90
↓ 2 callersFunctionnormalize_mesh
(mesh, scale=0.9999)
hy3dgen/shapegen/surface_loaders.py:7
↓ 2 callersMethodprepare_extra_step_kwargs
(self, generator, eta)
hy3dgen/shapegen/pipelines.py:456
↓ 2 callersMethodprepare_image
(self, image)
hy3dgen/shapegen/pipelines.py:490
↓ 2 callersMethodraster_texture
(self, tex, uv, uv_da=None, mip_level_bias=None, mip=None, filter_mode='auto', boundary
hy3dgen/texgen/differentiable_renderer/mesh_render.py:193
↓ 2 callersMethodrecenter
recenter an image to leave some empty space at the image border. Args: image (ndarray): input image, float/uint8 [H, W, 3/4]
hy3dgen/shapegen/preprocessors.py:36
↓ 2 callersMethodset_surface_extractor
(self, mc_algo)
hy3dgen/shapegen/pipelines.py:540
↓ 2 callersMethodset_texture
(self, tex)
hy3dgen/texgen/differentiable_renderer/mesh_render.py:263
↓ 2 callersFunctionsmart_load_model
( model_path, subfolder, use_safetensors, variant, )
hy3dgen/shapegen/utils.py:89
↓ 2 callersFunctiontimestep_embedding
Create sinusoidal timestep embeddings. :param t: a 1-D Tensor of N indices, one per batch element. These may be fractio
hy3dgen/shapegen/models/denoisers/hunyuan3ddit.py:39
↓ 2 callersMethodwrite
(self, buf)
api_server.py:109
↓ 1 callersFunctionLoadGlb
(path)
hy3dgen/texgen/custom_rasterizer/custom_rasterizer/io_glb.py:134
↓ 1 callersMethod__getattr__
(self, name: str)
hy3dgen/texgen/hunyuanpaint/unet/modules.py:492
↓ 1 callersMethod__init__
(self, alpha_cumprods, timesteps=1000, ddim_timesteps=50)
hy3dgen/texgen/hunyuanpaint/pipeline.py:140
↓ 1 callersMethod__init__
(self, unet: UNet2DConditionModel)
hy3dgen/texgen/hunyuanpaint/unet/modules.py:405
↓ 1 callersMethod__init__
(self, size=512, border_ratio=None)
hy3dgen/shapegen/preprocessors.py:31
↓ 1 callersMethod__init__
(self, embed_dim, num_experts=16, num_experts_per_tok=2, aux_loss_alpha=0.01)
hy3dgen/shapegen/models/denoisers/moe_layers.py:31
↓ 1 callersMethod__init__
( self, *, num_latents: int, embed_dim: int, width: int, heads
hy3dgen/shapegen/models/autoencoders/model.py:199
↓ 1 callersFunction_chunked_feed_forward
(ff: nn.Module, hidden_states: torch.Tensor, chunk_dim: int, chunk_size: int)
hy3dgen/texgen/hunyuanpaint/unet/modules.py:29
↓ 1 callersMethod_compute_box_stat
(self, bounds: Union[Tuple[float], List[float], float], octree_resolution: int)
hy3dgen/shapegen/models/autoencoders/surface_extractors.py:38
↓ 1 callersMethod_init_step_index
(self, timestep)
hy3dgen/shapegen/schedulers.py:237
↓ 1 callersMethod_init_step_index
(self, timestep)
hy3dgen/shapegen/schedulers.py:428
↓ 1 callersMethod_initialize_attn_weights
(self)
hy3dgen/texgen/hunyuanpaint/unet/modules.py:83
↓ 1 callersMethod_render
( self, glctx, mvp, pos, pos_idx, uv, uv_idx,
hy3dgen/texgen/differentiable_renderer/mesh_render.py:338
↓ 1 callersMethodbake_from_multiview
(self, views, camera_elevs, camera_azims, view_weights, method='graphcut')
hy3dgen/texgen/pipelines.py:128
↓ 1 callersFunctionbuild_app
()
gradio_app.py:353
↓ 1 callersFunctionbuild_logger
(logger_name, logger_filename)
api_server.py:52
↓ 1 callersFunctioncenter_vertices
Translate the vertices so that bounding box is centered at zero.
hy3dgen/shapegen/models/autoencoders/surface_extractors.py:29
↓ 1 callersMethodcompile
(self)
hy3dgen/text2image.py:49
↓ 1 callersFunctioncompute_discrete_voxel_indice
(position, grid_resolution=8, voxel_resolution=128)
hy3dgen/texgen/hunyuanpaint/unet/modules.py:359
↓ 1 callersFunctioncompute_multi_resolution_discrete_voxel_indice
( position_maps, grid_resolutions=[64, 32, 16, 8], voxel_resolutions=[512, 256, 128, 64] )
hy3dgen/texgen/hunyuanpaint/unet/modules.py:391
↓ 1 callersFunctioncompute_multi_resolution_mask
(position_maps, grid_resolutions=[32, 16, 8])
hy3dgen/texgen/hunyuanpaint/unet/modules.py:349
↓ 1 callersFunctioncompute_voxel_grid_mask
(position, grid_resolution=8)
hy3dgen/texgen/hunyuanpaint/unet/modules.py:302
↓ 1 callersFunctionconvert_triangle_fan_to_triangles
(indices)
hy3dgen/texgen/custom_rasterizer/custom_rasterizer/io_glb.py:93
↓ 1 callersFunctionconvert_triangle_strip_to_triangles
(indices)
hy3dgen/texgen/custom_rasterizer/custom_rasterizer/io_glb.py:82
↓ 1 callersMethoddenoise
r""" The call function to the pipeline for generation. Args: prompt (`str` or `List[str]`, *optional*): T
hy3dgen/texgen/hunyuanpaint/pipeline.py:390
↓ 1 callersMethodenable_model_cpu_offload
(self, gpu_id: Optional[int] = None, device: Union[torch.device, str] = "cuda")
hy3dgen/texgen/pipelines.py:106
next →1–100 of 375, ranked by callers