MCPcopy Create free account

hub / github.com/PKU-YuanGroup/Helios / functions

Functions847 in github.com/PKU-YuanGroup/Helios

⨍ Functions 847◇ Types & classes 124
↓ 1 callersFunctionload_configs_from_json
(config_path)
helios/videoalign/inference.py:17
↓ 1 callersMethodload_model
Load AMT model
eval/6_get_drifting_motion_smoothness.py:77
↓ 1 callersMethodload_model
Load AMT model
eval/2_get_motion_smoothness.py:73
↓ 1 callersFunctionload_model_from_checkpoint
(model, checkpoint_dir, checkpoint_step)
helios/videoalign/utils.py:174
↓ 1 callersFunctionload_video_frames
(video_path, start_ratio=0.0, end_ratio=1.0, num_frames=8, height=384, width=640)
eval/utils/utils.py:190
↓ 1 callersFunctionmain
(args)
train_helios.py:111
↓ 1 callersFunctionmain
()
infer_helios.py:186
↓ 1 callersFunctionmain
()
tools/others/benchmark/benchmark_compile_performance.py:118
↓ 1 callersFunctionmain
( rank, world_size, global_rank, stride, batch_size, dataloader_num_workers, json_
tools/offload_data/get_short-latents.py:27
↓ 1 callersFunctionmain
()
tools/offload_data/get_ode-pairs.py:92
↓ 1 callersFunctionmain
()
tools/offload_data/get_text-embedding.py:97
↓ 1 callersFunctionmain
( rank, world_size, global_rank, stride, batch_size, dataloader_num_workers, json_
tools/offload_data/get_long-latents.py:27
↓ 1 callersFunctionmain
(args)
eval/8_get_drifting_naturalness.py:191
↓ 1 callersFunctionmain
(args)
eval/3_get_semantic.py:61
↓ 1 callersFunctionmain
(args)
eval/9_merge_all_scores.py:220
↓ 1 callersFunctionmain
(args)
eval/6_get_drifting_motion_smoothness.py:195
↓ 1 callersFunctionmain
(args)
eval/10_merge_all_results.py:102
↓ 1 callersFunctionmain
(args)
eval/4_get_naturalness.py:142
↓ 1 callersFunctionmain
(args)
eval/1_get_motion_amplitude.py:74
↓ 1 callersFunctionmain
(args)
eval/0_get_aesthetic.py:60
↓ 1 callersFunctionmain
(args)
eval/2_get_motion_smoothness.py:168
↓ 1 callersFunctionmain
(args)
eval/5_get_drifting_aesthetic.py:88
↓ 1 callersFunctionmain
(args)
eval/7_get_drifting_semantic.py:115
↓ 1 callersFunctionmain
()
helios/videoalign/inference.py:257
↓ 1 callersFunctionmain_worker
(rank, config)
eval/utils/third_party/amt/train.py:25
↓ 1 callersFunctionmake_colorwheel
Generates a color wheel for optical flow visualization as presented in: Baker et al. "A Database and Evaluation Methodology for Optical F
eval/utils/third_party/amt/utils/flow_utils.py:21
↓ 1 callersMethodmask_tokens
(self, inputs, masking_prob=0.0)
eval/utils/third_party/ViCLIP/viclip_vision.py:157
↓ 1 callersFunctionmeasure_inference_memory
测量推理显存
tools/others/benchmark/benchmark_triton_performance.py:114
↓ 1 callersFunctionmeasure_inference_memory
测量推理显存
tools/others/benchmark/benchmark_patchification_performance.py:110
↓ 1 callersFunctionmeasure_inference_speed
测量推理速度
tools/others/benchmark/benchmark_triton_performance.py:66
↓ 1 callersFunctionmeasure_inference_speed
测量推理速度(单步)
tools/others/benchmark/benchmark_patchification_performance.py:63
↓ 1 callersFunctionmeasure_training_memory
测量训练显存(forward + backward)
tools/others/benchmark/benchmark_triton_performance.py:205
↓ 1 callersFunctionmeasure_training_memory
测量训练显存(包含backward)
tools/others/benchmark/benchmark_patchification_performance.py:147
↓ 1 callersFunctionmeasure_training_speed
测量训练速度(forward + backward)
tools/others/benchmark/benchmark_triton_performance.py:152
↓ 1 callersFunctionmerge_results
(input_dir, output_path=None, is_long=False)
eval/9_merge_all_scores.py:168
↓ 1 callersMethodmotion_score
Calculate motion smoothness score for a video
eval/2_get_motion_smoothness.py:103
↓ 1 callersMethodmultistep_uni_c_bh_update
One step for the UniC (B(h) version). Args: this_model_output (`torch.Tensor`): The model outputs at `x_
helios/scheduler/scheduling_helios.py:585
↓ 1 callersMethodmultistep_uni_c_bh_update
One step for the UniC (B(h) version). Args: this_model_output (`torch.Tensor`): The model outputs at `x_
helios/diffusers_version/scheduling_helios_diffusers.py:585
↓ 1 callersMethodmultistep_uni_p_bh_update
One step for the UniP (B(h) version). Alternatively, `self.solver_p` is used if is specified. Args: model_output (`torch
helios/scheduler/scheduling_helios.py:451
↓ 1 callersMethodmultistep_uni_p_bh_update
One step for the UniP (B(h) version). Alternatively, `self.solver_p` is used if is specified. Args: model_output (`torch
helios/diffusers_version/scheduling_helios_diffusers.py:451
↓ 1 callersFunctionmv_archived_logger
(name)
eval/utils/third_party/amt/trainers/logger.py:10
↓ 1 callersFunctionnormalize_score
(score, metric_key)
eval/9_merge_all_scores.py:151
↓ 1 callersFunctionparse_args
()
infer_helios.py:37
↓ 1 callersFunctionparse_args
()
tools/offload_data/get_ode-pairs.py:313
↓ 1 callersFunctionparse_args
()
tools/offload_data/get_text-embedding.py:229
↓ 1 callersFunctionparse_video_name
Parse video filename to extract step and index
tools/gradio/comparison/gradio_compare_diff-video.py:7
↓ 1 callersFunctionparse_video_name
Parse video filename to extract step and index
tools/gradio/comparison/gradio_compare_diff-ablation.py:7
↓ 1 callersFunctionparse_video_name
Parse video filename to extract step and index
tools/gradio/comparison/gradio_compare_diff-ckpt.py:7
↓ 1 callersMethodprepare_batch
Modified to accept either file paths (str) or Tensors (torch.Tensor) in 'videos'.
helios/videoalign/inference.py:129
↓ 1 callersFunctionprepare_dataset_on_rank0
(txt_file, output_folder, rank)
tools/offload_data/get_ode-pairs.py:42
↓ 1 callersFunctionprepare_dataset_on_rank0
(txt_file, output_folder, rank)
tools/offload_data/get_text-embedding.py:34
↓ 1 callersFunctionprepare_for_saving
(tensor, fps=30, caption=None)
train_helios.py:1947
↓ 1 callersMethodprepare_image_latents
( self, image: torch.Tensor, latents_mean: torch.Tensor, latents_std: torch.Te
helios/diffusers_version/pipeline_helios_diffusers.py:393
↓ 1 callersMethodprepare_image_latents
( self, image: torch.Tensor, latents_mean: torch.Tensor, latents_std: torch.Te
helios/pipelines/pipeline_helios_ode.py:364
↓ 1 callersMethodprepare_image_latents
( self, image: torch.Tensor, latents_mean: torch.Tensor, latents_std: torch.Te
helios/pipelines/pipeline_helios.py:365
↓ 1 callersMethodprepare_latents
( self, batch_size: int, num_channels_latents: int = 16, height: int = 384,
helios/diffusers_version/pipeline_helios_diffusers.py:361
↓ 1 callersMethodprepare_latents
( self, batch_size: int, num_channels_latents: int = 16, height: int = 480,
helios/pipelines/pipeline_helios_ode.py:332
↓ 1 callersMethodprepare_latents
( self, batch_size: int, num_channels_latents: int = 16, height: int = 480,
helios/pipelines/pipeline_helios.py:333
↓ 1 callersMethodprepare_stage1_latent
(self, vae_latent, idx, base_vae_latent=None)
tools/others/get_mean_std_size9.py:139
↓ 1 callersMethodprepare_stage1_latent
(self, vae_latent, idx, base_vae_latent=None)
helios/dataset/dataloader_history_latents_dist.py:133
↓ 1 callersMethodprepare_stage1_latent
(self, vae_latent, idx, base_vae_latent=None, return_secondary=False)
helios/dataset/dataloader_dmd.py:156
↓ 1 callersFunctionprepare_stage1_noise_input
( args, model_input, noise_scheduler, recycle_vars=None, latents_history_short=None, l
helios/utils/utils_helios_base.py:724
↓ 1 callersFunctionprepare_stage2_clean_input
( args, scheduler, latents, # [b c t h w] pyramid_stage_num=3, stage2_sample_ratios=[1, 1
helios/utils/utils_helios_base.py:868
↓ 1 callersFunctionprepare_stage2_noise_input
( args, scheduler, latents, # [b c t h w] pyramid_stage_num=3, stage2_sample_ratios=[1, 1
helios/utils/utils_helios_base.py:1011
↓ 1 callersMethodprepare_video_latents
( self, video: torch.Tensor, latents_mean: torch.Tensor, latents_std: torch.Te
helios/diffusers_version/pipeline_helios_diffusers.py:418
↓ 1 callersMethodprepare_video_latents
( self, video: torch.Tensor, latents_mean: torch.Tensor, latents_std: torch.Te
helios/pipelines/pipeline_helios_ode.py:387
↓ 1 callersMethodprepare_video_latents
( self, video: torch.Tensor, latents_mean: torch.Tensor, latents_std: torch.Te
helios/pipelines/pipeline_helios.py:388
↓ 1 callersMethodprocess_input_hidden_states
( self, latents, indices_hidden_states=None, indices_latents_history_short=Non
helios/modules/transformer_helios.py:1130
↓ 1 callersFunctionprocess_video_tensor
( video: torch.Tensor, nframes: int = 20, resized_height: int = 336, resized_width: int = 588,
helios/videoalign/vision_process.py:379
↓ 1 callersFunctionprompt_clean
(text)
helios/utils/utils_base.py:588
↓ 1 callersFunctionprompt_clean
(text)
helios/diffusers_version/pipeline_helios_diffusers.py:105
↓ 1 callersFunctionprompt_clean
(text)
helios/pipelines/pipeline_helios_ode.py:111
↓ 1 callersFunctionprompt_clean
(text)
helios/pipelines/pipeline_helios.py:112
↓ 1 callersFunctionrandom_crop
(img0, imgt, img1, flow, crop_size=(224, 224))
eval/utils/third_party/amt/datasets/vimeo_datasets.py:21
↓ 1 callersFunctionrandom_horizontal_flip
(img0, imgt, img1, flow, p=0.5)
eval/utils/third_party/amt/datasets/vimeo_datasets.py:51
↓ 1 callersFunctionrandom_resize
(img0, imgt, img1, flow, p=0.1)
eval/utils/third_party/amt/datasets/vimeo_datasets.py:12
↓ 1 callersFunctionrandom_reverse_channel
(img0, imgt, img1, flow, p=0.5)
eval/utils/third_party/amt/datasets/vimeo_datasets.py:33
↓ 1 callersFunctionrandom_reverse_time
(img0, imgt, img1, flow, p=0.5)
eval/utils/third_party/amt/datasets/vimeo_datasets.py:71
↓ 1 callersFunctionrandom_rotate
(img0, imgt, img1, flow, p=0.05)
eval/utils/third_party/amt/datasets/vimeo_datasets.py:61
↓ 1 callersFunctionrandom_vertical_flip
(img0, imgt, img1, flow, p=0.3)
eval/utils/third_party/amt/datasets/vimeo_datasets.py:41
↓ 1 callersFunctionreadFloat
(name)
eval/utils/third_party/amt/utils/utils.py:249
↓ 1 callersFunctionreadFlow
(name)
eval/utils/third_party/amt/utils/utils.py:207
↓ 1 callersFunctionread_cut_crop_and_resize
( video_path, f_prime, h_prime, w_prime, stride=1, start_frame=None, end_frame=None, crop=None )
helios/dataset/dataloader_mp4_dist.py:86
↓ 1 callersFunctionread_frames_decord_by_fps
( video_path, sample_fps=2, sample="rand", fix_start=None, max_num_frames=-1, trimmed3
eval/utils/utils.py:137
↓ 1 callersMethodreset_scheduler_history
(self)
helios/scheduler/scheduling_helios.py:868
↓ 1 callersMethodreset_scheduler_history
(self)
helios/diffusers_version/scheduling_helios_diffusers.py:936
↓ 1 callersFunctionresize_long_side
Resize image keeping aspect ratio
eval/8_get_drifting_naturalness.py:54
↓ 1 callersFunctionresize_long_side
Resize image keeping aspect ratio
eval/4_get_naturalness.py:51
↓ 1 callersMethodrestore
r""" Args: Restore the parameters stored with the `store` method. Useful to validate the model with EMA parameters without: af
helios/utils/create_ema_zero3.py:338
↓ 1 callersMethodresume_training
(self)
eval/utils/third_party/amt/trainers/base_trainer.py:48
↓ 1 callersFunctionrun_benchmark
运行基准测试
tools/others/benchmark/benchmark_compile_performance.py:71
↓ 1 callersFunctionsample_block_noise
(scheduler, batch_size, channel, num_frames, height, width)
helios/utils/utils_helios_post.py:596
↓ 1 callersMethodsample_block_noise
( self, batch_size, channel, num_frames, height, width,
helios/diffusers_version/pipeline_helios_diffusers.py:473
↓ 1 callersMethodsample_block_noise
( self, batch_size, channel, num_frames, height, width,
helios/pipelines/pipeline_helios_ode.py:442
↓ 1 callersMethodsample_block_noise
( self, batch_size, channel, num_frames, height, width,
helios/pipelines/pipeline_helios.py:443
↓ 1 callersFunctionsample_latent_error_from_latent_buffer
Randomly sample an error from the buffer based on timestep grid.
helios/utils/utils_recycle_single.py:263
↓ 1 callersFunctionsample_latent_error_from_latent_buffer
Randomly sample an error from the buffer based on timestep grid.
helios/utils/utils_recycle_batch.py:366
↓ 1 callersFunctionsample_noise_error_from_noise_buffer
Randomly sample an error from the buffer based on timestep grid.
helios/utils/utils_recycle_single.py:242
↓ 1 callersFunctionsample_noise_error_from_noise_buffer
Randomly sample an error from the buffer based on timestep grid.
helios/utils/utils_recycle_batch.py:328
↓ 1 callersFunctionseed_all
(seed)
eval/utils/third_party/amt/utils/utils.py:94
← previousnext →401–500 of 847, ranked by callers