Functions — github.com/SHI-Labs/Forget-Me-Not

CodeHub MCPcopy Create free account

hub / github.com/SHI-Labs/Forget-Me-Not / functions

Functions31 in github.com/SHI-Labs/Forget-Me-Not

⨍ Functions 31 ◇ Types & classes 4

↓ 2 callersFunctionapply_learned_embed_in_clip

( learned_embeds, text_encoder, tokenizer, token: Optional[Union[str, List[str]]] = None,

src/patch_lora.py:44

↓ 2 callersFunctionloss_step

( batch, unet, vae, text_encoder, scheduler, t_mutliplier=1.0, mixed_precision=Fal

src/train_ti.py:180

↓ 2 callersFunctionparse_safeloras_embeds

Converts a loaded safetensor file that contains Textual Inversion embeds into a dictionary of embed_token: Tensor

src/patch_lora.py:25

↓ 1 callersFunctioncollate_fn

(examples, with_prior_preservation=False)

src/train_attn.py:161

↓ 1 callersFunctionget_models

( pretrained_model_name_or_path, pretrained_vae_name_or_path, revision, placeholder_tokens: Li

src/train_ti.py:47

↓ 1 callersFunctionimport_model_class_from_model_name_or_path

(pretrained_model_name_or_path: str, revision: str)

src/train_attn.py:39

↓ 1 callersMethodloss

src/train_attn.py:291

↓ 1 callersFunctionmain

src/train_ti.py:703

↓ 1 callersFunctionpatch_ti

(pipe, ti_paths)

src/simple_inference.py:6

↓ 1 callersMethodset_concept_positions

(self, concept_positions)

src/train_attn.py:289

↓ 1 callersFunctiontext2img_dataloader

(train_dataset, train_batch_size, tokenizer, vae, text_encoder)

src/train_ti.py:130

↓ 1 callersFunctiontrain_inversion

( unet, vae, text_encoder, dataloader, num_steps: int, scheduler, index_no_updates

src/train_ti.py:262

↓ 1 callersMethodzero_attn_probs

src/train_attn.py:293

(self, attn_prob, m_name)

src/train_attn.py:283

(self, attn: "CrossAttention", hidden_states, encoder_hidden_states=None, attention_mask=None)

src/train_attn.py:302

Method__getitem__

src/dataset.py:114

Method__getitem__

src/train_attn.py:116

( self, instance_data_root, stochastic_attribute, tokenizer, token_map

src/dataset.py:39

( self, tokenizer, size=512, center_crop=False, use_added_token= False

src/train_attn.py:65

src/train_attn.py:280

(self, attn_controller: "AttnController", module_name)

src/train_attn.py:299

src/dataset.py:111

src/train_attn.py:113

Function_randomset

src/dataset.py:20

Function_shuffle

src/dataset.py:28

Functioncollate_fn

src/train_ti.py:131

src/simple_inference.py:23

src/train_attn.py:187

Functionperform_tuning

( unet, vae, text_encoder, dataloader, num_steps, scheduler, optimizer, save_s

src/train_ti.py:419

Functionsafe_save

( tensors: Dict[str, torch.Tensor], filename: str, metadata: Optional[Dict[str, str]]

src/patch_lora.py:12

( instance_data_dir: str, pretrained_model_name_or_path: str, output_dir: str, train_text_enco

src/train_ti.py:503