MCPcopy Create free account

hub / github.com/caiyuanhao1998/Open-OmniVCus / functions

Functions1,933 in github.com/caiyuanhao1998/Open-OmniVCus

⨍ Functions 1,933◇ Types & classes 464↳ Endpoints 1
↓ 5 callersMethodparse_extra_inputs
(self, data, extra_inputs, inputs_shared)
DiffSynth-Studio/examples/wanvideo/model_training/train.py:65
↓ 5 callersMethodparse_model_configs
(self, model_paths, model_id_with_origin_paths, fp8_models=None, offload_models=None, device="cpu")
DiffSynth-Studio/diffsynth/diffusion/training_module.py:114
↓ 5 callersFunctionrearrange_out
(out: torch.Tensor, out_pattern="b n s d", required_out_pattern="b n s d", dims=None)
DiffSynth-Studio/diffsynth/core/attention/attention.py:59
↓ 5 callersFunctionrearrange_qkv
(q: torch.Tensor, k: torch.Tensor, v: torch.Tensor, q_pattern="b n s d", k_pattern="b n s d", v_pattern="b n s
DiffSynth-Studio/diffsynth/core/attention/attention.py:48
↓ 5 callersMethodsplit_pipeline_units
(self, model_names: list[str])
DiffSynth-Studio/diffsynth/diffusion/base_pipeline.py:218
↓ 5 callersMethodtransfer_data_to_device
(self, data, device, torch_float_dtype=None)
DiffSynth-Studio/diffsynth/diffusion/training_module.py:67
↓ 4 callersMethod__init__
(self, device="cuda", torch_dtype=torch.bfloat16)
DiffSynth-Studio/diffsynth/pipelines/z_image.py:21
↓ 4 callersMethod__init__
(self, dim, eps, elementwise_affine=True)
DiffSynth-Studio/diffsynth/models/general_modules.py:105
↓ 4 callersMethod__init__
(self, in_dim, mid_dim, out_dim, bias=False)
DiffSynth-Studio/diffsynth/models/qwen_image_image2lora.py:5
↓ 4 callersMethod__init__
(self, dim_in: int, dim_out: int)
Video-Depth-Anything/video_depth_anything/motion_module/attention.py:346
↓ 4 callersMethod__init__
( self, in_channels, num_attention_heads = 8, num_transformer_b
Video-Depth-Anything/video_depth_anything/motion_module/motion_module.py:33
↓ 4 callersMethod_format_tensor_info
(self, x, name="tensor")
DiffSynth-Studio/examples/wanvideo/model_inference/Wan2.1-VACE-1.3B_tensor_parallel.py:98
↓ 4 callersMethod_get_image_feature
Compute the image features on a given frame.
VideoCus-Factory/sam2/sam2_video_predictor.py:803
↓ 4 callersMethod_is_rank0
(self)
DiffSynth-Studio/examples/wanvideo/model_inference/Wan2.1-VACE-1.3B_tensor_parallel.py:90
↓ 4 callersFunction_make_fusion_block
(features, use_bn, size=None)
Video-Depth-Anything/video_depth_anything/dpt.py:21
↓ 4 callersMethod_modulate
(self, x, mod_params, index=None)
DiffSynth-Studio/diffsynth/models/qwen_image_dit.py:470
↓ 4 callersMethod_predict
Predict masks for the given input prompts, using the currently set image. Input prompts are batched torch tensors and are expected to
VideoCus-Factory/sam2/sam2_image_predictor.py:337
↓ 4 callersMethod_process_attn
(self, q, k, v, shape)
DiffSynth-Studio/diffsynth/models/longcat_video_dit.py:172
↓ 4 callersFunctionadd_image_size_config
(parser: argparse.ArgumentParser)
DiffSynth-Studio/diffsynth/diffusion/parsers.py:12
↓ 4 callersFunctionapply_rotary_emb_qwen
( x: torch.Tensor, freqs_cis: Union[torch.Tensor, Tuple[torch.Tensor]] )
DiffSynth-Studio/diffsynth/models/qwen_image_dit.py:51
↓ 4 callersFunctiondist_barrier
()
DiffSynth-Studio/examples/wanvideo/model_inference/Wan2.1-VACE-1.3B_sequence_parallel.py:23
↓ 4 callersMethodencode
(self, x, scale)
DiffSynth-Studio/diffsynth/models/wan_video_vae.py:984
↓ 4 callersMethodfrom_pretrained
Load a pretrained model from the Hugging Face hub. Arguments: model_id (str): The Hugging Face repository ID. **
VideoCus-Factory/sam2/sam2_image_predictor.py:69
↓ 4 callersFunctionget_device_type
Get device type based on current machine, currently only support CPU, CUDA, NPU.
DiffSynth-Studio/diffsynth/core/device/npu_compatible_device.py:19
↓ 4 callersMethodinfer_video_depth
(self, frames, target_fps, input_size=518, device='cuda')
Video-Depth-Anything/video_depth_anything/video_depth.py:67
↓ 4 callersFunctionmove_files
(prefix, target_folder)
DiffSynth-Studio/examples/dev_tools/unit_test.py:74
↓ 4 callersFunctionparse_device_type
(device)
DiffSynth-Studio/diffsynth/core/device/npu_compatible_device.py:85
↓ 4 callersMethodprepare_image_ids
(self, latents)
DiffSynth-Studio/diffsynth/models/flux_dit.py:306
↓ 4 callersMethodprepare_tokens_with_masks
(self, x, masks=None)
Video-Depth-Anything/video_depth_anything/dinov2.py:212
↓ 4 callersMethodtile
(self, model_input, tile_size, tile_stride, tile_device, tile_dtype)
DiffSynth-Studio/diffsynth/models/flux_vae.py:20
↓ 4 callersMethodtransform_coords
Expects a torch tensor with length 2 in the last dimension. The coordinates can be in absolute image or normalized coordinates, If th
VideoCus-Factory/sam2/utils/transforms.py:46
↓ 4 callersMethodunpatchify
(self, hidden_states, height, width)
DiffSynth-Studio/diffsynth/models/flux_dit.py:301
↓ 3 callersMethod__getitem__
(self, item)
DiffSynth-Studio/diffsynth/utils/data/__init__.py:122
↓ 3 callersMethod__init__
( self, offload_dtype: torch.dtype = None, offload_device: Union[str, torch.device] =
DiffSynth-Studio/diffsynth/core/vram/layers.py:10
↓ 3 callersMethod__init__
(self, disable_guidance_embedder=False, num_joint_blocks=5, num_single_blocks=10, num_mode=0, mode_dict={}, ad
DiffSynth-Studio/diffsynth/models/flux_controlnet.py:62
↓ 3 callersMethod__init__
(self, cross_attention_dim=768, id_embeddings_dim=512, num_tokens=4)
DiffSynth-Studio/diffsynth/models/flux_ipadapter.py:24
↓ 3 callersMethod__init__
(self, layer, num_layers, dim=None, input_projection=False)
VideoCus-Factory/sam2/modeling/memory_encoder.py:121
↓ 3 callersMethod__init__
( self, embedding_dim: int, num_heads: int, downsample_rate: int = 1,
VideoCus-Factory/sam2/modeling/sam/transformer.py:196
↓ 3 callersMethod__len__
(self)
DiffSynth-Studio/diffsynth/utils/data/__init__.py:109
↓ 3 callersMethod_apply_non_overlapping_constraints
Apply non-overlapping constraints to the object scores in pred_masks. Here we keep only the highest scoring object at each spatial lo
VideoCus-Factory/sam2/modeling/sam2_base.py:811
↓ 3 callersMethod_consolidate_temp_output_across_obj
Consolidate the per-object temporary outputs in `temp_output_dict_per_obj` on a frame into a single output for all objects, including
VideoCus-Factory/sam2/sam2_video_predictor.py:433
↓ 3 callersMethod_get_obj_num
Get the total number of unique object ids received so far in this session.
VideoCus-Factory/sam2/sam2_video_predictor.py:175
↓ 3 callersMethod_get_orig_video_res_output
Resize the object scores to the original video resolution (video_res_masks) and apply non-overlapping constraints for final output.
VideoCus-Factory/sam2/sam2_video_predictor.py:411
↓ 3 callersFunction_list_files_with_ext
(d: str, exts: Tuple[str, ...])
DiffSynth-Studio/examples/wanvideo/model_training/validate_full/Wan2.1-VACE-14B_tensor_parallel_all.py:90
↓ 3 callersFunction_list_files_with_ext
(d: str, exts: Tuple[str, ...])
DiffSynth-Studio/examples/wanvideo/model_training/validate_full/Wan2.2-VACE-Fun-A14B_tensor_parallel_all.py:90
↓ 3 callersFunction_list_files_with_ext
(d: str, exts: Tuple[str, ...])
DiffSynth-Studio/examples/wanvideo/model_training/validate_full/Wan2.1-VACE-14B_tensor_parallel_all_reverse.py:90
↓ 3 callersFunction_list_files_with_ext
(d: str, exts: Tuple[str, ...])
DiffSynth-Studio/examples/wanvideo/model_training/validate_full/Wan2.1-VACE-1.3B_tensor_parallel_all.py:90
↓ 3 callersMethod_prepare_backbone_features
Prepare and flatten visual features.
VideoCus-Factory/sam2/modeling/sam2_base.py:477
↓ 3 callersMethod_run_single_frame_inference
Run tracking on a single frame based on current inputs and previous memory.
VideoCus-Factory/sam2/sam2_video_predictor.py:836
↓ 3 callersFunctionapply_gate
AI is creating summary for apply_gate Args: x (torch.Tensor): input tensor. gate (torch.Tensor, optional): gate tensor. Defaults
DiffSynth-Studio/diffsynth/models/step1x_connector.py:170
↓ 3 callersFunctionattention_forward
(q: torch.Tensor, k: torch.Tensor, v: torch.Tensor, q_pattern="b n s d", k_pattern="b n s d", v_pattern="b n s
DiffSynth-Studio/diffsynth/core/attention/attention.py:108
↓ 3 callersFunctionbasic_clean
(text)
DiffSynth-Studio/diffsynth/models/wan_video_text_encoder.py:260
↓ 3 callersMethodbatch_decode
This method forwards all its arguments to Qwen2TokenizerFast's [`~PreTrainedTokenizer.batch_decode`]. Please refer to the docstring o
DiffSynth-Studio/diffsynth/models/nexus_gen_ar_model.py:1083
↓ 3 callersMethodblend_h
(self, a: torch.Tensor, b: torch.Tensor, blend_extent: int)
DiffSynth-Studio/diffsynth/models/flux2_vae.py:2142
↓ 3 callersMethodblend_v
(self, a: torch.Tensor, b: torch.Tensor, blend_extent: int)
DiffSynth-Studio/diffsynth/models/flux2_vae.py:2136
↓ 3 callersFunctionbuild_sam2_video_predictor
( config_file, ckpt_path=None, device="cuda", mode="eval", hydra_overrides_extra=[], a
VideoCus-Factory/sam2/build_sam.py:44
↓ 3 callersMethodcreate_coordinate_grid
(size, start=None, device=None)
DiffSynth-Studio/diffsynth/models/z_image_dit.py:396
↓ 3 callersMethoddecode
(self, z, scale)
DiffSynth-Studio/diffsynth/models/wan_video_vae.py:1011
↓ 3 callersFunctionenable_vram_management
(model: torch.nn.Module, module_map: dict, vram_config: dict, vram_limit=None, disk_map=None, **kwargs)
DiffSynth-Studio/diffsynth/core/vram/layers.py:468
↓ 3 callersMethodextract_masked_hidden
(self, hidden_states: torch.Tensor, mask: torch.Tensor)
DiffSynth-Studio/diffsynth/pipelines/qwen_image.py:330
↓ 3 callersMethodfetch_input_params
(self)
DiffSynth-Studio/diffsynth/diffusion/base_pipeline.py:31
↓ 3 callersMethodfetch_output_params
(self)
DiffSynth-Studio/diffsynth/diffusion/base_pipeline.py:45
↓ 3 callersFunctionfilter_unprocessed_tasks
(script_path)
DiffSynth-Studio/examples/dev_tools/unit_test.py:9
↓ 3 callersMethodforward_image
Get the image feature on the input batch.
VideoCus-Factory/sam2/modeling/sam2_base.py:463
↓ 3 callersFunctionget_activation_layer
get activation layer Args: act_type (str): the activation type Returns: torch.nn.functional: the activation layer
DiffSynth-Studio/diffsynth/models/step1x_connector.py:264
↓ 3 callersFunctionget_connected_components
Get the connected components (8-connectivity) of binary masks of shape (N, 1, H, W). Inputs: - mask: A binary mask tensor of shape (N, 1
VideoCus-Factory/sam2/utils/misc.py:47
↓ 3 callersMethodget_rope_index
Calculate the 3D rope index based on image and video's temporal, height and width in LLM. Explanation: Each embedding se
DiffSynth-Studio/diffsynth/models/nexus_gen_ar_model.py:110
↓ 3 callersFunctionget_torch_device
Get torch attribute based on device type, e.g. torch.cuda or torch.npu
DiffSynth-Studio/diffsynth/core/device/npu_compatible_device.py:31
↓ 3 callersMethodload_from_disk
(self, torch_dtype, device, copy_module=False)
DiffSynth-Studio/diffsynth/core/vram/layers.py:126
↓ 3 callersMethodload_from_disk
(self, torch_dtype, device, assign=True)
DiffSynth-Studio/diffsynth/core/vram/layers.py:359
↓ 3 callersFunctionmerge_lora
(loras: List[Dict[str, torch.Tensor]], alpha=1)
DiffSynth-Studio/diffsynth/utils/lora/merge.py:11
↓ 3 callersFunctionmodulate_fp32
(norm_func, x, shift, scale)
DiffSynth-Studio/diffsynth/models/longcat_video_dit.py:343
↓ 3 callersFunctionprecompute_freqs_cis_3d
(dim: int, end: int = 1024, theta: float = 10000.0)
DiffSynth-Studio/diffsynth/models/wan_video_dit.py:74
↓ 3 callersFunctionread_video_frames
(video_path, process_length, target_fps=-1, max_res=-1)
Video-Depth-Anything/utils/dc_utils.py:19
↓ 3 callersMethodreset_predictor
Resets the image embeddings and other state variables.
VideoCus-Factory/sam2/sam2_image_predictor.py:459
↓ 3 callersMethodreshape_heads_to_4d
(self, tensor)
Video-Depth-Anything/video_depth_anything/motion_module/attention.py:100
↓ 3 callersFunctionreshape_tensor
(x, heads)
DiffSynth-Studio/diffsynth/models/flux_infiniteyou.py:17
↓ 3 callersFunctionrope_precompute
(x, grid_sizes, freqs, start=None)
DiffSynth-Studio/diffsynth/models/wan_video_dit_s2v.py:26
↓ 3 callersFunctionrun_train_multi_GPU
(script_path)
DiffSynth-Studio/examples/dev_tools/unit_test.py:52
↓ 3 callersFunctionskip_model_initialization
(device=torch.device("meta"))
DiffSynth-Studio/diffsynth/core/vram/initialization.py:6
↓ 3 callersMethodto
(self, *args, **kwargs)
DiffSynth-Studio/diffsynth/diffusion/training_module.py:12
↓ 3 callersMethodvae_output_to_video
(self, vae_output, pattern="B C T H W", min_value=-1, max_value=1)
DiffSynth-Studio/diffsynth/diffusion/base_pipeline.py:138
↓ 2 callersMethod__getitem__
(self, index)
VideoCus-Factory/sam2/utils/misc.py:147
↓ 2 callersMethod__init__
( self, mot_layers=(0, 4, 8, 12, 16, 20, 24, 28, 32, 36), patch_size=(1, 2, 2),
DiffSynth-Studio/diffsynth/models/wan_video_mot.py:95
↓ 2 callersMethod__init__
(self, q_dim, num_heads, head_dim, kv_dim=None, bias_q=False, bias_kv=False, bias_out=False)
DiffSynth-Studio/diffsynth/models/sd_text_encoder.py:18
↓ 2 callersMethod__init__
(self, dim)
DiffSynth-Studio/diffsynth/models/flux_lora_patcher.py:251
↓ 2 callersMethod__init__
(self, entry)
DiffSynth-Studio/diffsynth/models/wan_video_camera_controller.py:80
↓ 2 callersMethod__init__
(self, q_dim, num_heads, head_dim, kv_dim=None, bias_q=False, bias_kv=False, bias_out=False)
DiffSynth-Studio/diffsynth/models/flux_text_encoder_clip.py:6
↓ 2 callersMethod__init__
( self, input_dim: int, hidden_dim: int, output_dim: int, num_layers:
VideoCus-Factory/sam2/modeling/sam2_utils.py:109
↓ 2 callersMethod__init__
( self, embed_dim: int = 96, # initial embed dim num_heads: int = 1, # initial numbe
VideoCus-Factory/sam2/modeling/backbones/hieradet.py:174
↓ 2 callersMethod_add_output_per_object
Split a multi-object output into per-object output slices and add them into `output_dict_per_obj`. The resulting slices share the sam
VideoCus-Factory/sam2/sam2_video_predictor.py:743
↓ 2 callersFunction_apply_tp_to_transformer
对一个 transformer/dit 模块做 TP。
DiffSynth-Studio/examples/wanvideo/model_inference/Wan2.1-VACE-14B_tensor_parallel.py:98
↓ 2 callersMethod_attention
(self, query, key, value, attention_mask=None)
Video-Depth-Anything/video_depth_anything/motion_module/attention.py:182
↓ 2 callersMethod_clear_non_cond_mem_around_input
Remove the non-conditioning memory around the input frame. When users provide correction clicks, the surrounding frames' non-conditio
VideoCus-Factory/sam2/sam2_video_predictor.py:956
↓ 2 callersMethod_decode
(self, z: torch.Tensor, return_dict: bool = True)
DiffSynth-Studio/diffsynth/models/flux2_vae.py:2088
↓ 2 callersMethod_encode
(self, x: torch.Tensor)
DiffSynth-Studio/diffsynth/models/flux2_vae.py:2045
↓ 2 callersMethod_encode_new_memory
Encode the current image and its prediction into a memory feature.
VideoCus-Factory/sam2/modeling/sam2_base.py:664
↓ 2 callersMethod_encode_xy
(self, x, y)
VideoCus-Factory/sam2/modeling/position_encoding.py:42
↓ 2 callersMethod_expand_pos_freqs_if_needed
(self, video_fhw, txt_seq_lens)
DiffSynth-Studio/diffsynth/models/qwen_image_dit.py:94
↓ 2 callersMethod_forward_sam_heads
Forward SAM prompt encoders and mask heads. Inputs: - backbone_features: image features of [B, C, H, W] shape - poin
VideoCus-Factory/sam2/modeling/sam2_base.py:251
← previousnext →101–200 of 1,933, ranked by callers