Functions — github.com/aistudynow/Comfyui-bitdance

hub / github.com/aistudynow/Comfyui-bitdance / functions

Functions163 in github.com/aistudynow/Comfyui-bitdance

⨍ Functions 163◇ Types & classes 35

Method__init__

(self, runtime: BitDanceModelRuntime)

nodes.py:231

Method__init__

(self, runtime: BitDanceTextRuntime)

nodes.py:254

Method__init__

(self, in_dim: int, out_dim: int, hidden_act: str)

local_model.py:27

Method__init__

(self, hidden_size, frequency_embedding_size=256)

bitdance_arch/vision_head/flow_head_parallel_x.py:154

Method__init__

(self, channels)

bitdance_arch/vision_head/flow_head_parallel_x.py:170

Method__init__

(self, channels, out_channels)

bitdance_arch/vision_head/flow_head_parallel_x.py:186

Method__init__

( self, dim, n_head, )

bitdance_arch/vision_head/flow_head_parallel_x.py:199

Method__init__

(self, channels, use_swiglu=False)

bitdance_arch/vision_head/flow_head_parallel_x.py:262

Method__init__

( self, in_channels, model_channels, z_channels, num_res_blocks,

bitdance_arch/vision_head/flow_head_parallel_x.py:295

Method__init__

(self, in_filters, out_filters, use_conv_shortcut = False,

bitdance_arch/vision_encoder/autoencoder.py:14

Method__init__

(self, *, ch, out_ch, in_channels, num_res_blocks, z_channels, ch_mult=(1, 2, 2, 4), resoluti

bitdance_arch/vision_encoder/autoencoder.py:130

Method__init__

( self, dim, dim_out = None )

bitdance_arch/vision_encoder/autoencoder.py:233

Method__init__

(self, z_channel, in_filters, num_groups=32, eps=1e-6)

bitdance_arch/vision_encoder/autoencoder.py:252

Method__init__

(self, *, ch, out_ch, in_channels, num_res_blocks, z_channels, ch_mult=(1, 2, 2, 4), resoluti

bitdance_arch/vision_encoder/autoencoder.py:280

Method__init__

(self, ddconfig, checkpoint=None, gan_decoder = False,

bitdance_arch/vision_encoder/autoencoder.py:355

Function_apply_loader_optimization

( model_runtime: BitDanceModelRuntime, text_runtime: BitDanceTextRuntime, vae_runtime: BitDance

nodes.py:1149

Method_basic_init

(module)

bitdance_arch/vision_head/flow_head_parallel_x.py:342

Method_inner_step_progress

(cur_step: int, total_steps: int)

nodes.py:2164

Function_module_device_dtype

(module: torch.nn.Module)

nodes.py:686

Function_try_resolve_model_root_from_checkpoint

(ckpt_name: str)

nodes.py:345

Methodapply_model

(self, *args, **kwargs)

nodes.py:237

Functionbuild_projector

(model_root: Path, hidden_size: int)

local_model.py:229

Functionbuild_projector_from_config

(config: Dict[str, Any])

local_model.py:236

Functionbuild_text_model_and_tokenizer

( model_root: Path, torch_dtype: torch.dtype, *, attention_mode: str = "auto", rms_norm_fu

local_model.py:242

Functionbuild_text_model_from_config_dict

( config_dict: Dict[str, Any], torch_dtype: torch.dtype, *, attention_mode: str = "auto",

local_model.py:277

Functionbuild_tokenizer_from_source

(tokenizer_source: str)

Functionbuild_vae_from_config

(config: Dict[str, Any])

local_model.py:222

Functionbuild_vision_head

(model_root: Path)

local_model.py:203

Functionbuild_vision_head_from_config

(config: Dict[str, Any])

(latents: torch.Tensor, tile_size: int = 256, overlap: int = 32)

nodes.py:2298

Functiondepth_to_space

Depth-to-Space DCR mode (depth-column-row) core implementation. Args: x (torch.Tensor): input tensor. The channels-first (*CHW)

bitdance_arch/vision_encoder/autoencoder.py:198

Methodencode

( self, text_encoder: BitDanceTextRuntime, positive_prompt: str, negative_

nodes.py:1700

Methodencode_cached

( self, text_encoder: BitDanceTextRuntime, positive_prompt: str, negative_

nodes.py:1760

Methodencode_from_tokens

(self, *args, **kwargs)

(self, x: torch.Tensor)

nodes.py:207

Methodforward

(self, hidden_states: torch.Tensor)

local_model.py:33

Methodforward

(self, x, cond)

bitdance_arch/vision_head/flow_head_parallel_x.py:79

Methodforward

(self, t)

bitdance_arch/vision_head/flow_head_parallel_x.py:163

Methodforward

(self, x, scale, shift, gate)

bitdance_arch/vision_head/flow_head_parallel_x.py:178

Methodforward

(self, x, y)

bitdance_arch/vision_head/flow_head_parallel_x.py:192

Methodforward

( self, x: torch.Tensor, )

bitdance_arch/vision_head/flow_head_parallel_x.py:215

Methodforward

(self, x, scale1, shift1, gate1, scale2, shift2, gate2)

bitdance_arch/vision_head/flow_head_parallel_x.py:281

Methodforward

(self, x, t, c)

bitdance_arch/vision_head/flow_head_parallel_x.py:364

Methodforward

(self, x, **kwargs)

bitdance_arch/vision_encoder/autoencoder.py:41

Methodforward

(self, x)

bitdance_arch/vision_encoder/autoencoder.py:107

Methodforward

(self, z)

bitdance_arch/vision_encoder/autoencoder.py:172

Methodforward

input_image: [B C H W]

bitdance_arch/vision_encoder/autoencoder.py:243

Methodforward

(self, x, quantizer)

bitdance_arch/vision_encoder/autoencoder.py:260

Methodforward

(self, z)

bitdance_arch/vision_encoder/autoencoder.py:325

Methodforward

(self, input)

bitdance_arch/vision_encoder/autoencoder.py:518

Methodinitialize_weights

(self)

bitdance_arch/vision_head/flow_head_parallel_x.py:145

Methodis_long_image

(H, W)

bitdance_arch/vision_encoder/autoencoder.py:441

Functionload_hidden_size

(model_root: Path)

local_model.py:115

Methodmemory_required

(self, input_shape=None)

nodes.py:242

Methodsample

( self, model: BitDanceModelRuntime, vae: BitDanceVAERuntime, positive: Bi

nodes.py:1993

Methodsample

( self, z, cfg, num_sampling_steps, sampler_name="euler_maruyama",

bitdance_arch/vision_head/flow_head_parallel_x.py:118

Methodselect

(self, resolution: str)

nodes.py:1962

Functionset_vision_attention_mode

(mode: str = "auto")

local_model.py:123

Methodvt_forward

(self, image_list, max_bs=32, ps=1)

bitdance_arch/vision_encoder/autoencoder.py:402

Methodvt_forward_maxpad

image_list: list of [1, 3, H, W] return: Tensor [(sum_i Hi*Wi/stride^2), C]

bitdance_arch/vision_encoder/autoencoder.py:426

← previous101–163 of 163, ranked by callers