AI/diffusers
1
0
Fork 0
mirror of https://github.com/huggingface/diffusers.git synced 2026-01-27 17:22:53 +03:00
Code Activity

remove methods again

Browse Source
This commit is contained in:
sayakpaul
2025-12-04 09:43:17 +08:00
parent 16d67c4ae0
commit 3be4b01cda
2 changed files with 6 additions and 105 deletions
Download Patch File Download Diff File Expand all files Collapse all files

92
src/diffusers/pipelines/stable_diffusion_xl/pipeline_stable_diffusion_xl.py
View File

@@ -309,98 +309,6 @@ class StableDiffusionXLPipeline(
latents = latents * self.scheduler.init_noise_sigma
return latents
def _get_add_time_ids(
self, original_size, crops_coords_top_left, target_size, dtype, text_encoder_projection_dim=None
):
add_time_ids = list(original_size + crops_coords_top_left + target_size)
passed_add_embed_dim = (
self.unet.config.addition_time_embed_dim * len(add_time_ids) + text_encoder_projection_dim
)
expected_add_embed_dim = self.unet.add_embedding.linear_1.in_features
if expected_add_embed_dim != passed_add_embed_dim:
raise ValueError(
f"Model expects an added time embedding vector of length {expected_add_embed_dim}, but a vector of {passed_add_embed_dim} was created. The model has an incorrect config. Please check `unet.config.time_embedding_type` and `text_encoder_2.config.projection_dim`."
)
add_time_ids = torch.tensor([add_time_ids], dtype=dtype)
return add_time_ids
def upcast_vae(self):
deprecate(
"upcast_vae",
"1.0.0",
"`upcast_vae` is deprecated. Please use `pipe.vae.to(torch.float32)`. For more details, please refer to: https://github.com/huggingface/diffusers/pull/12619#issue-3606633695.",
)
self.vae.to(dtype=torch.float32)
# Copied from diffusers.pipelines.latent_consistency_models.pipeline_latent_consistency_text2img.LatentConsistencyModelPipeline.get_guidance_scale_embedding
def get_guidance_scale_embedding(
self, w: torch.Tensor, embedding_dim: int = 512, dtype: torch.dtype = torch.float32
) -> torch.Tensor:
"""
See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298
Args:
w (`torch.Tensor`):
Generate embedding vectors with a specified guidance scale to subsequently enrich timestep embeddings.
embedding_dim (`int`, *optional*, defaults to 512):
Dimension of the embeddings to generate.
dtype (`torch.dtype`, *optional*, defaults to `torch.float32`):
Data type of the generated embeddings.
Returns:
`torch.Tensor`: Embedding vectors with shape `(len(w), embedding_dim)`.
"""
assert len(w.shape) == 1
w = w * 1000.0
half_dim = embedding_dim // 2
emb = torch.log(torch.tensor(10000.0)) / (half_dim - 1)
emb = torch.exp(torch.arange(half_dim, dtype=dtype) * -emb)
emb = w.to(dtype)[:, None] * emb[None, :]
emb = torch.cat([torch.sin(emb), torch.cos(emb)], dim=1)
if embedding_dim % 2 == 1: # zero pad
emb = torch.nn.functional.pad(emb, (0, 1))
assert emb.shape == (w.shape[0], embedding_dim)
return emb
@property
def guidance_scale(self):
return self._guidance_scale
@property
def guidance_rescale(self):
return self._guidance_rescale
@property
def clip_skip(self):
return self._clip_skip
# here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
# of the Imagen paper: https://huggingface.co/papers/2205.11487 . `guidance_scale = 1`
# corresponds to doing no classifier free guidance.
@property
def do_classifier_free_guidance(self):
return self._guidance_scale > 1 and self.unet.config.time_cond_proj_dim is None
@property
def cross_attention_kwargs(self):
return self._cross_attention_kwargs
@property
def denoising_end(self):
return self._denoising_end
@property
def num_timesteps(self):
return self._num_timesteps
@property
def interrupt(self):
return self._interrupt
@torch.no_grad()
@replace_example_docstring(EXAMPLE_DOC_STRING)
def __call__(

19
src/diffusers/pipelines/stable_diffusion_xl/pipeline_stable_diffusion_xl_utils.py
View File

@@ -19,9 +19,8 @@ import torch
from ...loaders import StableDiffusionXLLoraLoaderMixin, TextualInversionLoaderMixin
from ...models import ImageProjection
from ...models.attention_processor import AttnProcessor2_0, FusedAttnProcessor2_0, XFormersAttnProcessor
from ...models.lora import adjust_lora_scale_text_encoder
from ...utils import USE_PEFT_BACKEND, logging, scale_lora_layers, unscale_lora_layers
from ...utils import USE_PEFT_BACKEND, deprecate, logging, scale_lora_layers, unscale_lora_layers
logger = logging.get_logger(__name__) # pylint: disable=invalid-name
@@ -393,18 +392,12 @@ class StableDiffusionXLMixin:
return image_embeds, uncond_image_embeds
def upcast_vae(self):
dtype = self.vae.dtype
self.vae.to(dtype=torch.float32)
use_torch_2_0_or_xformers = isinstance(
self.vae.decoder.mid_block.attentions[0].processor,
(AttnProcessor2_0, XFormersAttnProcessor, FusedAttnProcessor2_0),
deprecate(
"upcast_vae",
"1.0.0",
"`upcast_vae` is deprecated. Please use `pipe.vae.to(torch.float32)`. For more details, please refer to: https://github.com/huggingface/diffusers/pull/12619#issue-3606633695.",
)
# if xformers or torch_2_0 is used attention block does not need
# to be in float32 which can save lots of memory
if use_torch_2_0_or_xformers:
self.vae.post_quant_conv.to(dtype)
self.vae.decoder.conv_in.to(dtype)
self.vae.decoder.mid_block.to(dtype)
self.vae.to(dtype=torch.float32)
# Copied from diffusers.pipelines.latent_consistency_models.pipeline_latent_consistency_text2img.LatentConsistencyModelPipeline.get_guidance_scale_embedding
def get_guidance_scale_embedding(