Functions — github.com/Xiaojiu-z/EasyControl

hub / github.com/Xiaojiu-z/EasyControl / functions

Functions139 in github.com/Xiaojiu-z/EasyControl

r""" Enable sliced VAE decoding. When this option is enabled, the VAE will split the input tensor in slices to compute decoding in sev

train/src/pipeline.py:479

Methodenable_vae_tiling

r""" Enable tiled VAE decoding. When this option is enabled, the VAE will split the input tensor into tiles to compute decoding and en

src/pipeline.py:410

Methodenable_vae_tiling

r""" Enable tiled VAE decoding. When this option is enabled, the VAE will split the input tensor into tiles to compute decoding and en

train/src/pipeline.py:493

Functionencode_prompt

( text_encoders, tokenizers, prompt: str, max_sequence_length, device=

train/src/prompt_helper.py:137

Methodfn_recursive_add_processors

(name: str, module: torch.nn.Module, processors: Dict[str, AttentionProcessor])

src/transformer_flux.py:306

Methodfn_recursive_add_processors

(name: str, module: torch.nn.Module, processors: Dict[str, AttentionProcessor])

train/src/transformer_flux.py:306

Methodfn_recursive_attn_processor

(name: str, module: torch.nn.Module, processor)

src/transformer_flux.py:342

Methodfn_recursive_attn_processor

(name: str, module: torch.nn.Module, processor)

train/src/transformer_flux.py:342

Methodforward

( self, hidden_states: torch.Tensor, cond_hidden_states: torch.Tensor, temb: t

src/transformer_flux.py:57

Methodforward

( self, hidden_states: torch.Tensor, cond_hidden_states: torch.Tensor, encoder

src/transformer_flux.py:147

Methodforward

( self, hidden_states: torch.Tensor, cond_hidden_states: torch.Tensor = None,

src/transformer_flux.py:399

Methodforward

(self, hidden_states: torch.Tensor)

src/layers_cache.py:43

Methodforward

( self, hidden_states: torch.Tensor, cond_hidden_states: torch.Tensor, temb: t

train/src/transformer_flux.py:57

Methodforward

( self, hidden_states: torch.Tensor, cond_hidden_states: torch.Tensor, encoder

train/src/transformer_flux.py:147

Methodforward

( self, hidden_states: torch.Tensor, cond_hidden_states: torch.Tensor = None,

train/src/transformer_flux.py:399

Methodforward

(self, hidden_states: torch.Tensor)

train/src/layers.py:43

Methodfuse_qkv_projections

Enables fused QKV projections. For self-attention modules, all projection matrices (i.e., query, key, value) are fused. For cross-att

src/transformer_flux.py:356

Methodfuse_qkv_projections

Enables fused QKV projections. For self-attention modules, all projection matrices (i.e., query, key, value) are fused. For cross-att

train/src/transformer_flux.py:356

train/src/pipeline.py:576

train/src/pipeline.py:588

Methodjoint_attention_kwargs

(self)

src/pipeline.py:497

Methodjoint_attention_kwargs

(self)

train/src/pipeline.py:580

Functionmulti_condition_generate_image

(prompt, subject_img, spatial_img, height, width, seed)

train/src/pipeline.py:584

Functionpreprocess_train

(examples)

train/src/jsonl_datasets.py:141

Functionset_multi_lora

(transformer, local_paths, lora_weights=[[]], cond_size=512)

train/src/lora_helper.py:180

Functionset_single_lora

(transformer, local_path, lora_weights=[], cond_size=512)

train/src/lora_helper.py:176

Functionsingle_condition_generate_image

(prompt, subject_img, spatial_img, height, width, seed, control_type, style_lora=None)

app.py:53

Functiontokenize_prompt

(tokenizer, prompt, max_sequence_length)

train/src/prompt_helper.py:14

Functiontokenize_prompt_clip

(tokenizer, prompt)

train/src/prompt_helper.py:28

Functiontokenize_prompt_t5

(tokenizer, prompt)

train/src/prompt_helper.py:42

Methodunfuse_qkv_projections

Disables the fused QKV projection if enabled. <Tip warning={true}> This API is 🧪 experimental. </Tip>

src/transformer_flux.py:382

Methodunfuse_qkv_projections

Disables the fused QKV projection if enabled. <Tip warning={true}> This API is 🧪 experimental. </Tip>

train/src/transformer_flux.py:382

Functionunset_lora

(transformer)

src/lora_helper.py:184

Functionunset_lora

(transformer)

train/src/lora_helper.py:184

Functionunwrap_model

(model)

train/train.py:687

← previous101–139 of 139, ranked by callers