MCPcopy Create free account

hub / github.com/MotrixLab/ViMoGen / functions

Functions457 in github.com/MotrixLab/ViMoGen

⨍ Functions 457◇ Types & classes 71
↓ 1 callersFunctionrender_video_from_vertices
( vertices: np.ndarray, joints: Optional[np.ndarray], faces: np.ndarray, output_path: Path,
motion_gating/mbench_render.py:103
↓ 1 callersMethodreset
(self)
datasets/sampler.py:303
↓ 1 callersFunctionresolve_repo_path
(meta_path: Path, raw_path: str)
motion_gating/render_mbench_videos.py:84
↓ 1 callersFunctionretarget_transl_to_fixed_camera
Retarget camera translation to a fixed camera intrinsics. We assume the original (per-frame) intrinsics are bbox-derived: (fx_i, fy_i,
motion_rep/smplest_x_scripts/inference.py:42
↓ 1 callersMethodretrieve_inference_timesteps_and_sigma_given_shift_scale
( self, num_inference_steps: int, shift_scale: int = 3, device=None)
trainer/sd3_scheduler.py:114
↓ 1 callersFunctionrun_alignment_checks
( entries: List[Dict[str, Any]], meta_path: Path, video_field: str, api_key: str, num_thre
motion_gating/apply_quality_gate.py:207
↓ 1 callersFunctionrun_smplx
( motion_tensor: torch.Tensor, smplx_model, device: torch.device, noise_std: float = 0.0, )
motion_gating/mbench_render.py:53
↓ 1 callersMethodsave_ckpt
(self, rank, load_save_dict=None, model_for_opt: tuple =
trainer/base_trainer.py:275
↓ 1 callersMethodsave_ckpt_to_dir
(rank, load_save_dict=None, model_for_opt: tuple = None,
trainer/base_trainer.py:253
↓ 1 callersMethodsave_config
(self, config: DictConfig)
trainer/base_trainer.py:40
↓ 1 callersMethodsave_model
(rank, model, out_path: str, dcp=False, lora=False)
trainer/base_trainer.py:163
↓ 1 callersMethodsave_motion_dict
(self, motion_dict, mean, std, device='cuda:0', vis=True, result_folder=None)
trainer/base_trainer.py:290
↓ 1 callersMethodsave_optimizer
(model, opt, out_path: str, dcp=False)
trainer/base_trainer.py:192
↓ 1 callersMethodsave_txt_dict
(self, txt_dict, result_folder=None)
trainer/base_trainer.py:306
↓ 1 callersMethodsetup_save
(self, save_dir: str, log_level: str, rank: int = 0, mode='train')
trainer/base_trainer.py:45
↓ 1 callersMethodsigma_shift
according to eq 23, and the shift scale for 1024 is 4 here, not consistent with paper, but It's ok. Change of base_scale will make s
trainer/sd3_scheduler.py:66
↓ 1 callersFunctionsinusoidal_embedding_1d
(dim, position)
models/transformer/wan/modules/tm2m_model.py:211
↓ 1 callersFunctionsinusoidal_embedding_1d
(dim, position)
models/transformer/wan/modules/t2m_model.py:210
↓ 1 callersFunctionsmooth_motion_rep
Temporal Gaussian smoothing for motion representation.
utils.py:144
↓ 1 callersMethodstate_dict
(self)
trainer/base_trainer.py:328
↓ 1 callersFunctionsummarize_binary
Return mean/std/count summary for binary values.
mbench/motion_condition_consistency.py:16
↓ 1 callersFunctionsummarize_binary
Return mean/std/count summary for binary values.
mbench/motion_generalizability.py:16
↓ 1 callersFunctionto_bool
(value: Any)
motion_gating/vlm_utils.py:21
↓ 1 callersFunctionto_eval_mode
()
train_eval_vimogen.py:234
↓ 1 callersMethodtraining_target
(self, sample, noise, timestep)
trainer/scheduler.py:224
↓ 1 callersMethodtraining_weight
(self, timestep)
trainer/scheduler.py:229
↓ 1 callersFunctionumt5_xxl
(**kwargs)
models/transformer/wan/modules/t5.py:456
↓ 1 callersFunctionupdate_ema
Step the EMA model towards the current model.
trainer/base_trainer.py:389
↓ 1 callersFunctionupdate_entry_paths
(entry: Dict[str, Any], repo_root: Path, pt_path: Path, mp4_path: Path)
motion_gating/render_mbench_videos.py:115
Method__call__
(self, )
datasets/multi_resolution_sampler.py:24
Method__call__
(self, n_frame: int, height: int, width: int, rnd_state: np.random.RandomState)
datasets/bucket_config.py:261
Method__call__
(self, n_frame: int, height: int, width: int, rnd_state: np.random.RandomState)
datasets/bucket_config.py:375
Method__call__
(self, n_frame: int, height: int, width: int, rnd_state: np.random.RandomState)
datasets/bucket_config.py:420
Method__call__
(self, sequence, **kwargs)
models/transformer/wan/modules/tokenizers.py:49
Method__call__
(self, texts, device)
models/transformer/wan/modules/t5.py:506
Method__enter__
(self)
trainer/base_trainer.py:372
Method__eq__
(self, other)
datasets/bucket_config.py:119
Method__eq__
(self, other)
datasets/bucket_config.py:150
Method__eq__
(self, other)
datasets/bucket_config.py:182
Method__exit__
(self, exc_type, exc_val, exc_tb)
trainer/base_trainer.py:377
Method__getitem__
(self, index)
trainer/__init__.py:29
Method__getitem__
(self, index)
datasets/video_datasets.py:47
Method__getitem__
(self, index)
datasets/video_datasets.py:246
Method__getitem__
(self, data_id)
models/transformer/wan/text_encoding_batch.py:15
Method__gt__
(self, other)
datasets/bucket_config.py:127
Method__gt__
(self, other)
datasets/bucket_config.py:156
Method__gt__
(self, other)
datasets/bucket_config.py:188
Method__hash__
(self, )
datasets/bucket_config.py:159
Method__hash__
(self, )
datasets/bucket_config.py:191
Method__init__
Initialize MBench evaluator Args: device: Device to use for evaluation ('cuda' or 'cpu') output_path
mbench/__init__.py:18
Method__init__
(self, scale, translation, znear=pyrender.camera.DEFAULT_Z_
mbench/render.py:25
Method__init__
(self, save_dir, log_level: str, rank: int = 0, mode='train')
trainer/base_trainer.py:30
Method__init__
(self, model, optimizer=None)
trainer/base_trainer.py:324
Method__init__
( self, name, logger=None, )
trainer/base_trainer.py:358
Method__init__
(self, dp_group: dist.ProcessGroup, num_train_timesteps: int = 1000)
trainer/scheduler.py:72
Method__init__
(self, num_inference_steps=100, num_train_timesteps=1000, shift=5.0, sigma_max=1.0, sigma_min=0.0, inverse_tim
trainer/scheduler.py:155
Method__init__
(self)
trainer/__init__.py:16
Method__init__
( self, logit_mean: float = 0.0, logid_std: float = 1.0, base_height: int = 25
trainer/sd3_scheduler.py:13
Method__init__
(self, bucket_config)
datasets/bucket.py:31
Method__init__
( self, ori_thw_list: List[Tuple], dp_size: int, rnd_state: np.random.RandomSt
datasets/multi_resolution_sampler.py:12
Method__init__
( self, dataset, dp_rank: int, dp_size: int, global_seed: int = 0,
datasets/multi_resolution_sampler.py:94
Method__init__
(self, *size)
datasets/bucket_config.py:115
Method__init__
(self, t, h, w)
datasets/bucket_config.py:137
Method__init__
(self, t, h, w, hw_value_arr: np.ndarray, t_arr: np.ndarray)
datasets/bucket_config.py:172
Method__init__
(self, )
datasets/bucket_config.py:269
Method__init__
(self, )
datasets/bucket_config.py:285
Method__init__
(self, )
datasets/bucket_config.py:302
Method__init__
(self, side_list: list, ar_list: list, ar_max: float, ar_min: float, pixel_num_list: list,
datasets/bucket_config.py:325
Method__init__
(self, )
datasets/bucket_config.py:341
Method__init__
(self, pixel_grading_num: int = 100, frame_grading_num: int = 13)
datasets/bucket_config.py:364
Method__init__
(self, )
datasets/bucket_config.py:386
Method__init__
(self, pixel_grading_num: int = 100, frame_grading_num: int = 13)
datasets/bucket_config.py:409
Method__init__
( self, dataset: Dataset, num_replicas: Optional[int] = None, rank: Optional[i
datasets/sampler.py:29
Method__init__
( self, dataset, bucket_config: dict, num_replicas: Optional[int] = None,
datasets/sampler.py:62
Method__init__
(self, train_json_file_list, test_json_file_list, motion_mean_path, motion_std_path, min_motion_length, max_mo
datasets/video_datasets.py:9
Method__init__
(self, *args, **kwargs)
datasets/video_datasets.py:214
Method__init__
(self, logging_dir, **kwargs)
parallel/logging.py:85
Method__init__
(self, json_file, text_key)
models/transformer/wan/text_encoding_batch.py:9
Method__init__
(self, name, seq_len=None, clean=None, **kwargs)
models/transformer/wan/modules/tokenizers.py:39
Method__init__
(self, dim, eps=1e-6)
models/transformer/wan/modules/t5.py:55
Method__init__
(self, dim, dim_attn, num_heads, dropout=0.1)
models/transformer/wan/modules/t5.py:71
Method__init__
(self, dim, dim_ffn, dropout=0.1)
models/transformer/wan/modules/t5.py:125
Method__init__
(self, dim, dim_attn, dim_ffn, num_heads,
models/transformer/wan/modules/t5.py:146
Method__init__
(self, dim, dim_attn, dim_ffn, num_heads,
models/transformer/wan/modules/t5.py:180
Method__init__
(self, num_buckets, num_heads, bidirectional, max_dist=128)
models/transformer/wan/modules/t5.py:223
Method__init__
(self, vocab, dim, dim_attn, dim_ffn,
models/transformer/wan/modules/t5.py:269
Method__init__
(self, vocab, dim, dim_attn, dim_ffn,
models/transformer/wan/modules/t5.py:317
Method__init__
( self, text_len, dtype=torch.bfloat16, device=torch.cuda.current_device(),
models/transformer/wan/modules/t5.py:474
Method__init__
(self, dim, eps=1e-5)
models/transformer/wan/modules/tm2m_model.py:237
Method__init__
(self, dim, eps=1e-6, elementwise_affine=False)
models/transformer/wan/modules/tm2m_model.py:252
Method__init__
(self, dim, num_heads, window_size=(-1, -1),
models/transformer/wan/modules/tm2m_model.py:261
Method__init__
(self, cross_attn_type, dim, ffn_dim, num_
models/transformer/wan/modules/tm2m_model.py:339
Method__init__
(self, dim, ffn_dim, num_heads, window_siz
models/transformer/wan/modules/tm2m_model.py:441
Method__init__
(self, in_dim, out_dim)
models/transformer/wan/modules/tm2m_model.py:523
Method__init__
(self, d_model, max_len=5000)
models/transformer/wan/modules/tm2m_model.py:536
Method__init__
(self, model_type='t2v', patch_size=(1, 2, 2), text_len=512
models/transformer/wan/modules/tm2m_model.py:557
Method__init__
(self)
models/transformer/wan/modules/tm2m_model.py:790
Method__init__
(self, dim, eps=1e-5)
models/transformer/wan/modules/t2m_model.py:236
Method__init__
(self, dim, eps=1e-6, elementwise_affine=False)
models/transformer/wan/modules/t2m_model.py:251
Method__init__
(self, dim, num_heads, window_size=(-1, -1),
models/transformer/wan/modules/t2m_model.py:260
← previousnext →201–300 of 457, ranked by callers