Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/TMElyralab/MuseV
/ functions
Functions
388 in github.com/TMElyralab/MuseV
⨍
Functions
388
◇
Types & classes
86
Method
__init__
( self, num_train_timesteps: int = 1000, beta_start: float = 0.0001, beta_end:
musev/schedulers/scheduling_ddpm.py:90
Method
__init__
( self, num_train_timesteps: int = 1000, beta_start: float = 0.00085, beta_end
musev/schedulers/scheduling_lcm.py:43
Method
__init__
( self, num_train_timesteps: int = 1000, beta_start: float = 0.0001, beta_end:
musev/schedulers/scheduling_euler_ancestral_discrete.py:120
Method
__init__
( self, num_train_timesteps: int = 1000, beta_start: float = 0.0001, beta_end:
musev/schedulers/scheduling_euler_discrete.py:21
Method
__init__
( self, num_train_timesteps: int = 1000, beta_start: float = 0.0001, beta_end:
musev/schedulers/scheduling_dpmsolver_multistep.py:151
Method
__init__
( self, num_train_timesteps: int = 1000, beta_start: float = 0.0001, beta_end:
musev/schedulers/scheduling_ddim.py:100
Method
__init__
Args: templates (str, optional): 形象提词模板,若为None,则使用默认的类属性. Defaults to None. portrait prompt template, if None, t
musev/auto_prompt/human.py:19
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/style.py:11
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/render.py:22
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:60
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:68
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:76
Method
__init__
(self, name: str = "singing")
musev/auto_prompt/attributes/human.py:82
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:90
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:98
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:106
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:123
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:131
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:139
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:147
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:155
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:163
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:171
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:179
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:187
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:195
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:203
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:211
Method
__init__
(self, funcs: list = None, use_preset=True, name="portrait")
musev/auto_prompt/attributes/human.py:243
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:354
Method
__init__
(self, name: str = None)
musev/auto_prompt/attributes/human.py:376
Method
__init__
这里类实例初始化设置`name`参数,主要是为了便于一些没有提前实现、通过字符串参数实现的新属性。 Theses class instances are initialized with the `name` parameter to facilitate the imple
musev/auto_prompt/attributes/attributes.py:15
Method
__init__
虽然预置了关注的属性列表和转换类,但也允许定义示例时,进行更新。 注意`self.preset_attributes`的元素只是类名字,以便减少实例化的资源消耗。而funcs是实例化后的属性转换列表。 Although the list of attention a
musev/auto_prompt/attributes/attributes.py:194
Method
__init__
Args: template (str): 提词模板, prompt template. 如果`template`含有{key},则根据key来取值。 if the template field contains {key},
musev/auto_prompt/attributes/attr2template.py:46
Method
__init__
关键词模板属性2文本转化类 1. 获取关键词模板字符串中的关键词属性; 2. 从import * 存储在locals()中变量中获取对应的类; 3. 将集成了多属性转换类的`MultiAttr2Text` Args:
musev/auto_prompt/attributes/attr2template.py:79
Method
__init__
_summary_ Args: sample_size (Optional[int], optional): _description_. Defaults to None. in_channels (int, optional):
musev/models/unet_3d_condition.py:213
Method
__init__
(self, *args, **kwargs)
musev/models/attention_processor.py:157
Method
__init__
( self, attention_op: Optional[Callable] = None, )
musev/models/attention_processor.py:168
Method
__init__
( self, attention_op: Optional[Callable] = None, )
musev/models/attention_processor.py:370
Method
__init__
(self, attention_op: Callable[..., Any] | None = None)
musev/models/attention_processor.py:553
Method
__init__
( self, query_dim: int, cross_attention_dim: int | None = None, heads: int = 8
musev/models/attention_processor.py:573
Method
__init__
( self, num_attention_heads: int = 16, attention_head_dim: int = 88, in_channe
musev/models/temporal_transformer.py:80
Method
__init__
( self, num_attention_heads: int = 16, attention_head_dim: int = 88, in_channe
musev/models/transformer_2d.py:84
Method
__init__
Controlnet 推断函数,用于提取 controlnet backbone的emb,避免训练时重复抽取 Controlnet inference predictor, used to extract the emb of the controlnet backbone
musev/models/controlnet.py:21
Method
__init__
( self, dim: int, num_attention_heads: int, attention_head_dim: int, d
musev/models/attention.py:55
Method
__init__
( self, in_channels: int, temb_channels: int, dropout: float = 0.0, nu
musev/models/unet_2d_blocks.py:546
Method
__init__
( self, in_channels: int, temb_channels: int, dropout: float = 0.0, nu
musev/models/unet_2d_blocks.py:658
Method
__init__
( self, in_channels: int, temb_channels: int, dropout: float = 0.0, nu
musev/models/unet_2d_blocks.py:817
Method
__init__
( self, in_channels: int, out_channels: int, temb_channels: int, dropo
musev/models/unet_2d_blocks.py:953
Method
__init__
( self, in_channels: int, out_channels: int, temb_channels: int, dropo
musev/models/unet_2d_blocks.py:1136
Method
__init__
( self, in_channels: int, out_channels: int, prev_output_channel: int,
musev/models/unet_2d_blocks.py:1236
Method
__init__
( self, sample_size: int | None = None, in_channels: int = 4, out_channels: in
musev/models/referencenet.py:97
Method
__init__
( self, in_channels: int, temb_channels: int, femb_channels: int, drop
musev/models/unet_3d_blocks.py:234
Method
__init__
( self, in_channels: int, out_channels: int, temb_channels: int, femb_
musev/models/unet_3d_blocks.py:439
Method
__init__
( self, in_channels: int, out_channels: int, temb_channels: int, femb_
musev/models/unet_3d_blocks.py:778
Method
__init__
( self, in_channels: int, out_channels: int, prev_output_channel: int,
musev/models/unet_3d_blocks.py:989
Method
__init__
_summary_ Args: unet (nn.Module): _description_ referencenet (nn.Module, optional): _description_. Defaults to None.
musev/models/super_model.py:37
Method
__init__
( self, in_dim, out_dim=None, dropout=0.0, keep_content_condition: boo
musev/models/resnet.py:39
Method
__init__
(self, tokenizer, text_encoder)
musev/models/text_model.py:6
Method
__len__
(self)
musev/schedulers/scheduling_euler_ancestral_discrete.py:355
Method
__len__
(self)
musev/schedulers/scheduling_dpmsolver_multistep.py:814
Method
__setitem__
(self, key, value)
musev/utils/register.py:11
Method
_set_gradient_checkpointing
(self, module, value=False)
musev/models/unet_3d_condition.py:767
Method
_set_gradient_checkpointing
(self, module, value=False)
musev/models/super_model.py:251
Method
add_noise
( self, original_samples: torch.FloatTensor, noise: torch.FloatTensor, timeste
musev/schedulers/scheduling_euler_ancestral_discrete.py:326
Method
attn_processors
r""" Returns: `dict` of attention processors: A dictionary containing all attention processors used in the model with
musev/models/unet_3d_condition.py:614
Function
concat_two_tensor
concat two tensor along dim with given method Args: data1 (torch.Tensor): first in data data2 (torch.Tensor): last in data
musev/data/data_util.py:185
Function
convert_controlnet_checkpoint
( checkpoint, original_config, checkpoint_path, image_size, upcast_attention, extract_ema )
musev/utils/convert_from_ckpt.py:947
Function
convert_ldm_bert_checkpoint
(checkpoint, config)
musev/utils/convert_from_ckpt.py:670
Function
convert_lora
(pipeline, state_dict, LORA_PREFIX_UNET="lora_unet", LORA_PREFIX_TEXT_ENCODER="lora_te", alpha=0.6)
musev/utils/convert_lora_safetensor_to_diffusers.py:52
Function
convert_open_clip_checkpoint
(checkpoint)
musev/utils/convert_from_ckpt.py:826
Function
convert_paint_by_example_checkpoint
(checkpoint)
musev/utils/convert_from_ckpt.py:759
Method
create_custom_forward
(module, return_dict=None)
musev/models/unet_2d_blocks.py:773
Method
create_custom_forward
(module, return_dict=None)
musev/models/unet_2d_blocks.py:1073
Method
create_custom_forward
(module)
musev/models/unet_2d_blocks.py:1204
Method
create_custom_forward
(module, return_dict=None)
musev/models/unet_2d_blocks.py:1374
Method
create_custom_forward
(module)
musev/models/unet_2d_blocks.py:1513
Method
create_custom_forward
(module, return_dict=None)
musev/models/unet_3d_blocks.py:623
Method
create_custom_forward
(module)
musev/models/unet_3d_blocks.py:904
Method
create_custom_forward
(module, return_dict=None)
musev/models/unet_3d_blocks.py:1133
Method
create_custom_forward
(module)
musev/models/unet_3d_blocks.py:1346
Function
create_diffusers_schedular
(original_config)
musev/utils/convert_from_ckpt.py:312
Function
create_ldm_bert_config
(original_config)
musev/utils/convert_from_ckpt.py:322
Function
create_vae_diffusers_config
Creates a config for the diffusers based on the config of the LDM model.
musev/utils/convert_from_ckpt.py:288
Method
custom_forward
(*inputs)
musev/models/unet_2d_blocks.py:774
Method
custom_forward
(*inputs)
musev/models/unet_2d_blocks.py:1074
Method
custom_forward
(*inputs)
musev/models/unet_2d_blocks.py:1205
Method
custom_forward
(*inputs)
musev/models/unet_2d_blocks.py:1375
Method
custom_forward
(*inputs)
musev/models/unet_2d_blocks.py:1514
Method
custom_forward
(*inputs)
musev/models/unet_3d_blocks.py:624
Method
custom_forward
(*inputs)
musev/models/unet_3d_blocks.py:905
Method
custom_forward
(*inputs)
musev/models/unet_3d_blocks.py:1134
Method
custom_forward
(*inputs)
musev/models/unet_3d_blocks.py:1347
Function
ddim_inversion
(pipeline, ddim_scheduler, video_latent, num_inv_steps, prompt="")
musev/utils/util.py:339
Function
decode_unet_latents_with_vae
(vae: nn.Module, latents: torch.tensor)
musev/utils/vae_util.py:7
Function
export_to_video
(videos: torch.Tensor, output_video_path: str, fps=8)
musev/utils/util.py:242
Method
fn_recursive_add_attns
( name: str, module: torch.nn.Module, attns: List[Tuple[str, Attention]],
musev/models/unet_3d_condition.py:1714
Method
fn_recursive_add_attns
( name: str, module: torch.nn.Module, attns: List[Tuple[str, Attention]],
musev/models/referencenet.py:1180
Method
fn_recursive_add_processors
( name: str, module: torch.nn.Module, processors: Dict[str, AttentionProce
musev/models/unet_3d_condition.py:623
Method
fn_recursive_attn_processor
(name: str, module: torch.nn.Module, processor)
musev/models/unet_3d_condition.py:735
← previous
next →
201–300 of 388, ranked by callers