MCPcopy Create free account

hub / github.com/OpenImagingLab/FlashVSR / functions

Functions2,110 in github.com/OpenImagingLab/FlashVSR

⨍ Functions 2,110◇ Types & classes 521
↓ 4 callersMethod__init__
(self, dim, channel_first=True, images=True, bias=False)
examples/WanVSR/utils/utils.py:15
↓ 4 callersFunction_clean_tag
(tag: str)
diffsynth/extensions/ImageQualityMetric/open_clip/pretrained.py:235
↓ 4 callersFunction_config_to_kwargs
(args)
diffsynth/models/kolors_text_encoder.py:710
↓ 4 callersMethod_encode_image
(self, images, normalize=True)
diffsynth/extensions/ImageQualityMetric/open_clip/coca_model.py:131
↓ 4 callersMethod_flatten_time
(x: torch.Tensor)
diffsynth/pipelines/flashvsr_tiny.py:93
↓ 4 callersMethod_flatten_time
(x: torch.Tensor)
diffsynth/pipelines/flashvsr_full.py:93
↓ 4 callersMethod_flatten_time
(x: torch.Tensor)
diffsynth/pipelines/flashvsr_tiny_long.py:93
↓ 4 callersMethod_make_layer
(self, planes, blocks, stride=1)
diffsynth/extensions/ImageQualityMetric/open_clip/modified_resnet.py:132
↓ 4 callersFunctionapply_gate
(x, gate, tr_gate=None, tr_token=None)
diffsynth/models/hunyuan_video_dit.py:394
↓ 4 callersMethodapply_nnf_to_image
(self, nnf, source)
diffsynth/extensions/FastBlend/patch_match.py:43
↓ 4 callersFunctionbuild_tcdecoder
构建“更宽”的 decoder；深度增强（IdentityConv2d+ReLU）已在 TAEHV 内部完成。 - 不创建 small / 不做移植 - base_ckpt_path 参数保留但不使用（接口兼容） 返回：big （单个模型）
examples/WanVSR/utils/TCDecoder.py:303
↓ 4 callersMethodclamp_bound
(self, nnf)
diffsynth/extensions/FastBlend/patch_match.py:89
↓ 4 callersMethodclean_mem
(self)
examples/WanVSR/utils/TCDecoder.py:272
↓ 4 callersFunctiondownload_pretrained_from_hf
( model_id: str, filename: str = 'open_clip_pytorch_model.bin', revision=None,
diffsynth/extensions/ImageQualityMetric/open_clip/pretrained.py:337
↓ 4 callersMethodencode_prompt
(self, prompt, positive=True, t5_sequence_length=512)
diffsynth/pipelines/flux_image.py:205
↓ 4 callersMethodencode_video
(self, processed_images, tiled=False, tile_size=64, tile_stride=32)
diffsynth/pipelines/sd_video.py:133
↓ 4 callersMethodgelu
(self, gate: torch.Tensor)
diffsynth/models/stepvideo_dit.py:576
↓ 4 callersMethodget_image_features
(self, *args, **kwargs)
diffsynth/extensions/ImageQualityMetric/trainer/models/clip_model.py:113
↓ 4 callersFunctionimage_transform
( image_size: int, is_train: bool, mean: Optional[Tuple[float, ...]] = None, s
diffsynth/extensions/ImageQualityMetric/open_clip/transform.py:77
↓ 4 callersFunctionlets_dance_xl
( unet: SDXLUNet, motion_modules: SDXLMotionModel = None, controlnet: MultiControlNetManager = Non
diffsynth/pipelines/dancer.py:119
↓ 4 callersMethodpad_image
(self, image)
diffsynth/extensions/FastBlend/patch_match.py:37
↓ 4 callersMethodpatch_multiple_resolutions
(self, latents, padding_latent=None, is_input_images:bool=False)
diffsynth/models/omnigen.py:451
↓ 4 callersMethodprepare_image_ids
(self, latents)
diffsynth/models/flux_dit.py:307
↓ 4 callersMethodprocess_mllm_input
(self, mllm_inputs, target_img_size)
diffsynth/prompters/omnigen_prompter.py:259
↓ 4 callersFunctionsafe_str
(x)
diffsynth/prompters/omost.py:95
↓ 4 callersMethodset_full_adapter
(self)
diffsynth/models/sd_ipadapter.py:26
↓ 4 callersMethodshape
(self)
diffsynth/extensions/FastBlend/data.py:129
↓ 4 callersMethodunpatchify
x: (N, T, patch_size**2 * C) imgs: (N, H, W, C)
diffsynth/models/omnigen.py:413
↓ 3 callersMethod__getitem__
(self, item)
diffsynth/data/video.py:120
↓ 3 callersMethod__init__
(self, num_feat, num_grow_ch=32)
diffsynth/extensions/ESRGAN/__init__.py:29
↓ 3 callersMethod__init__
(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.)
diffsynth/extensions/ImageQualityMetric/BLIP/vit.py:22
↓ 3 callersMethod__init__
( self, in_channels: int = 3, out_channels: int = 16, eps=1e-6, dropou
diffsynth/models/hunyuan_video_vae_encoder.py:70
↓ 3 callersMethod__init__
(self, disable_guidance_embedder=False, num_joint_blocks=5, num_single_blocks=10, num_mode=0, mode_dict={}, ad
diffsynth/models/flux_controlnet.py:9
↓ 3 callersMethod__init__
(self)
diffsynth/models/cog_dit.py:109
↓ 3 callersMethod__init__
(self)
diffsynth/models/sdxl_ipadapter.py:44
↓ 3 callersFunction_build_text_tower
( embed_dim: int, text_cfg: CLIPTextCfg, quick_gelu: bool = False, cast_dtype:
diffsynth/extensions/ImageQualityMetric/open_clip/model.py:137
↓ 3 callersFunction_build_vision_tower
( embed_dim: int, vision_cfg: CLIPVisionCfg, quick_gelu: bool = False, cast_dt
diffsynth/extensions/ImageQualityMetric/open_clip/model.py:75
↓ 3 callersFunctionapply_gate
AI is creating summary for apply_gate Args: x (torch.Tensor): input tensor. gate (torch.Tensor, optional): gate tensor. Defaults
diffsynth/models/step1x_connector.py:170
↓ 3 callersFunctionapply_rotary_emb
( xq: torch.Tensor, xk: torch.Tensor, freqs_cis, head_first: bool = False, )
diffsynth/models/hunyuan_video_dit.py:354
↓ 3 callersFunctionbase_conv3d
(x, conv_layer, channel_last=False, residual=None, only_return_output=False)
diffsynth/models/stepvideo_vae.py:74
↓ 3 callersFunctionbase_group_norm
(x, norm_layer, act_silu=False, channel_last=False)
diffsynth/models/stepvideo_vae.py:32
↓ 3 callersFunctionbase_group_norm_with_zero_pad
(x, norm_layer, act_silu=True, pad_size=2)
diffsynth/models/stepvideo_vae.py:405
↓ 3 callersFunctionbasic_clean
(text)
diffsynth/prompters/wan_prompter.py:11
↓ 3 callersMethodbuild_1d_mask
(self, length, left_bound, right_bound, border_width)
diffsynth/models/hunyuan_video_vae_decoder.py:408
↓ 3 callersMethodbuild_1d_mask
(self, length, left_bound, right_bound, border_width)
diffsynth/models/hunyuan_video_vae_encoder.py:207
↓ 3 callersMethodbuild_chat_input
(self, query, history=None, role="user")
diffsynth/prompters/kolors_prompter.py:203
↓ 3 callersFunctioncalc_out_
(in_size, padding, dilation, kernel, stride)
diffsynth/models/stepvideo_vae.py:115
↓ 3 callersMethodclear_cache
(self)
diffsynth/models/wan_video_vae.py:616
↓ 3 callersMethoddecode_image
(self, latent, tiled=False, tile_size=64, tile_stride=32)
diffsynth/pipelines/sd_image.py:75
↓ 3 callersMethoddecode_naive
(self, z, is_init=True)
diffsynth/models/stepvideo_vae.py:983
↓ 3 callersMethodencode_image
(self, image, tiled=False, tile_size=64, tile_stride=32)
diffsynth/pipelines/sd_image.py:70
↓ 3 callersMethodencode_prompt
(self, prompt, positive=True, t5_sequence_length=77)
diffsynth/pipelines/sd3_image.py:62
↓ 3 callersMethodencode_prompt
(self, prompt, clip_skip=1, clip_skip_2=1, positive=True)
diffsynth/pipelines/hunyuan_image.py:177
↓ 3 callersMethodencode_prompt
(self, prompt, clip_skip=1, clip_skip_2=2, positive=True)
diffsynth/pipelines/sdxl_image.py:95
↓ 3 callersMethodestimate_nnf
(self, source_guide, target_guide, source_style, nnf)
diffsynth/extensions/FastBlend/patch_match.py:191
↓ 3 callersFunctionexists
(val)
diffsynth/extensions/ImageQualityMetric/trainer/models/cross_modeling.py:8
↓ 3 callersFunctionflash_attention
(q: torch.Tensor, k: torch.Tensor, v: torch.Tensor, num_heads: int, compatibility_mode=False, attention_mask=N
diffsynth/models/wan_video_dit.py:174
↓ 3 callersFunctionfreeze_batch_norm_2d
Converts all `BatchNorm2d` and `SyncBatchNorm` layers of provided module into `FrozenBatchNorm2d`. If `module` is itself an instance of eithe
diffsynth/extensions/ImageQualityMetric/open_clip/utils.py:8
↓ 3 callersFunctionget_activation_layer
get activation layer Args: act_type (str): the activation type Returns: torch.nn.functional: the activation layer
diffsynth/models/step1x_connector.py:264
↓ 3 callersMethodget_logits
(self, image_features, text_features, logit_scale)
diffsynth/extensions/ImageQualityMetric/open_clip/loss.py:103
↓ 3 callersFunctionget_model_config
(model_name)
diffsynth/extensions/ImageQualityMetric/open_clip/factory.py:70
↓ 3 callersFunctionget_tokenizer
(model_name, open_clip_bpe_path=None)
diffsynth/extensions/ImageQualityMetric/open_clip/factory.py:77
↓ 3 callersMethodinfer_architecture
(self, state_dict)
diffsynth/models/sd3_dit.py:390
↓ 3 callersMethodinit_weights
(self)
diffsynth/models/wan_video_image_encoder.py:582
↓ 3 callersFunctionlets_dance
( unet: SDUNet, motion_modules: SDMotionModel = None, controlnet: MultiControlNetManager = None,
diffsynth/pipelines/dancer.py:7
↓ 3 callersFunctionlets_dance_flux
( dit: FluxDiT, controlnet: FluxMultiControlNetManager = None, step1x_connector: Qwen2Connector =
diffsynth/pipelines/flux_image.py:655
↓ 3 callersFunctionload_model_from_single_file
(state_dict, model_names, model_classes, model_resource, torch_dtype, device)
diffsynth/models/model_manager.py:56
↓ 3 callersMethodlock
lock modules Args: unlocked_groups (int): leave last n layer groups unlocked (default: 0)
diffsynth/extensions/ImageQualityMetric/open_clip/timm_model.py:85
↓ 3 callersMethodpartition
(x: torch.Tensor, win: Tuple[int, int, int])
diffsynth/models/wan_video_dit.py:96
↓ 3 callersFunctionprecompute_freqs_cis
(dim: int, end: int = 1024, theta: float = 10000.0)
diffsynth/models/wan_video_dit.py:257
↓ 3 callersMethodprocess_multi_modal_prompt
(self, text, input_images)
diffsynth/prompters/omnigen_prompter.py:75
↓ 3 callersMethodreinit_cross_kv
(self, new_context: torch.Tensor)
diffsynth/models/wan_video_dit.py:574
↓ 3 callersMethodresample_image
(self, images, level)
diffsynth/extensions/FastBlend/patch_match.py:234
↓ 3 callersFunctionreshape_tensor
(x, heads)
diffsynth/models/flux_infiniteyou.py:17
↓ 3 callersMethodreturn_to_timestep
(self, timestep, sample, sample_stablized)
diffsynth/schedulers/ddim.py:81
↓ 3 callersMethodshift_nnf
(self, nnf, d)
diffsynth/extensions/FastBlend/patch_match.py:116
↓ 3 callersMethodtile
(self, model_input, tile_size, tile_stride, tile_device, tile_dtype)
diffsynth/models/tiler.py:20
↓ 3 callersFunctiontokenize_long_prompt
(tokenizer, prompt, max_length=None)
diffsynth/prompters/base_prompter.py:6
↓ 2 callersMethod__getitem__
(self, item)
diffsynth/extensions/FastBlend/data.py:136
↓ 2 callersMethod__init__
(self, num_inference_steps, rel_l1_thresh, model_id)
diffsynth/pipelines/flashvsr_tiny.py:439
↓ 2 callersMethod__init__
(self, num_inference_steps, rel_l1_thresh, model_id)
diffsynth/pipelines/flashvsr_full.py:456
↓ 2 callersMethod__init__
(self, num_inference_steps, rel_l1_thresh, model_id)
diffsynth/pipelines/flashvsr_tiny_long.py:440
↓ 2 callersMethod__init__
(self, model_path: str)
diffsynth/prompters/kolors_prompter.py:13
↓ 2 callersMethod__init__
(self, tokenizer_path=None, model=None, system_prompt="")
diffsynth/prompters/prompt_refiners.py:60
↓ 2 callersMethod__init__
(self, input_size)
diffsynth/extensions/ImageQualityMetric/imagereward.py:25
↓ 2 callersMethod__init__
(self, inplanes, planes, stride=1)
diffsynth/extensions/ImageQualityMetric/open_clip/modified_resnet.py:13
↓ 2 callersMethod__init__
(self, **kwargs)
diffsynth/extensions/RIFE/__init__.py:61
↓ 2 callersMethod__init__
(self)
diffsynth/models/svd_vae_decoder.py:72
↓ 2 callersMethod__init__
(self, cross_attention_dim=768, id_embeddings_dim=512, num_tokens=4)
diffsynth/models/flux_ipadapter.py:8
↓ 2 callersMethod__init__
(self, num_attention_heads, attention_head_dim, in_channels, num_layers=1, norm_num_groups=32, eps=1e-5)
diffsynth/models/sd_motion.py:54
↓ 2 callersMethod__init__
(self, vocab_size=49408)
diffsynth/models/sd3_text_encoder.py:8
↓ 2 callersMethod__len__
(self)
diffsynth/data/video.py:107
↓ 2 callersFunction_adain
(content_feat: torch.Tensor, style_feat: torch.Tensor)
diffsynth/pipelines/flashvsr_tiny.py:34
↓ 2 callersFunction_adain
(content_feat: torch.Tensor, style_feat: torch.Tensor)
diffsynth/pipelines/flashvsr_full.py:34
↓ 2 callersFunction_adain
(content_feat: torch.Tensor, style_feat: torch.Tensor)
diffsynth/pipelines/flashvsr_tiny_long.py:34
↓ 2 callersFunction_calc_mean_std
(feat: torch.Tensor, eps: float = 1e-5)
diffsynth/pipelines/flashvsr_tiny.py:25
↓ 2 callersFunction_calc_mean_std
(feat: torch.Tensor, eps: float = 1e-5)
diffsynth/pipelines/flashvsr_full.py:25
↓ 2 callersFunction_calc_mean_std
(feat: torch.Tensor, eps: float = 1e-5)
diffsynth/pipelines/flashvsr_tiny_long.py:25
↓ 2 callersMethod_calculate_score
Calculate the aesthetic score for a single image. Args: image (torch.Tensor): The processed image tensor. Returns:
diffsynth/extensions/ImageQualityMetric/aesthetic.py:76
↓ 2 callersMethod_calculate_score
Calculate the CLIP score for a single image and prompt. Args: image (torch.Tensor): The processed image tensor. promp
diffsynth/extensions/ImageQualityMetric/clip.py:44
← previousnext →101–200 of 2,110, ranked by callers