LLM/diffusers
1
0
Fork 0
mirror of https://github.com/huggingface/diffusers.git synced 2025-12-06 12:34:13 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix typos in strings and comments (#11476)

Browse Source 
* Fix typos in strings and comments

Signed-off-by: co63oc <co63oc@users.noreply.github.com>

* Update src/diffusers/hooks/hooks.py

Co-authored-by: Aryan <contact.aryanvs@gmail.com>

* Update src/diffusers/hooks/hooks.py

Co-authored-by: Aryan <contact.aryanvs@gmail.com>

* Update layerwise_casting.py

* Apply style fixes

* update

---------

Signed-off-by: co63oc <co63oc@users.noreply.github.com>
Co-authored-by: Aryan <contact.aryanvs@gmail.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
This commit is contained in:
co63oc
2025-05-30 21:19:00 +08:00
committed by GitHub
parent 6508da6f06
commit 8183d0f16e
24 changed files with 34 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/cogvideo/train_cogvideox_image_to_video_lora.py
View File

@@ -555,7 +555,7 @@ class VideoDataset(Dataset):
if any(not path.is_file() for path in instance_videos):
raise ValueError(
"Expected '--video_column' to be a path to a file in `--instance_data_root` containing line-separated paths to video data but found atleast one path that is not a valid file."
"Expected '--video_column' to be a path to a file in `--instance_data_root` containing line-separated paths to video data but found at least one path that is not a valid file."
)
return instance_prompts, instance_videos

2
examples/cogvideo/train_cogvideox_lora.py
View File

@@ -539,7 +539,7 @@ class VideoDataset(Dataset):
if any(not path.is_file() for path in instance_videos):
raise ValueError(
"Expected '--video_column' to be a path to a file in `--instance_data_root` containing line-separated paths to video data but found atleast one path that is not a valid file."
"Expected '--video_column' to be a path to a file in `--instance_data_root` containing line-separated paths to video data but found at least one path that is not a valid file."
)
return instance_prompts, instance_videos

2
examples/research_projects/multi_subject_dreambooth_inpainting/README.md
View File

@@ -73,7 +73,7 @@ accelerate launch train_multi_subject_dreambooth_inpaint.py \
## 3. Results
A [![Weights & Biases](https://img.shields.io/badge/Weights%20&%20Biases-Report-blue)](https://wandb.ai/gzguevara/uncategorized/reports/Multi-Subject-Dreambooth-for-Inpainting--Vmlldzo2MzY5NDQ4?accessToken=y0nya2d7baguhbryxaikbfr1203amvn1jsmyl07vk122mrs7tnph037u1nqgse8t) is provided showing the training progress by every 50 steps. Note, the reported weights & baises run was performed on a A100 GPU with the following stetting:
A [![Weights & Biases](https://img.shields.io/badge/Weights%20&%20Biases-Report-blue)](https://wandb.ai/gzguevara/uncategorized/reports/Multi-Subject-Dreambooth-for-Inpainting--Vmlldzo2MzY5NDQ4?accessToken=y0nya2d7baguhbryxaikbfr1203amvn1jsmyl07vk122mrs7tnph037u1nqgse8t) is provided showing the training progress by every 50 steps. Note, the reported weights & biases run was performed on a A100 GPU with the following stetting:
```bash
accelerate launch train_multi_subject_dreambooth_inpaint.py \

2
src/diffusers/hooks/faster_cache.py
View File

@@ -146,7 +146,7 @@ class FasterCacheConfig:
alpha_low_frequency: float = 1.1
alpha_high_frequency: float = 1.1
# n as described in CFG-Cache explanation in the paper - dependant on the model
# n as described in CFG-Cache explanation in the paper - dependent on the model
unconditional_batch_skip_range: int = 5
unconditional_batch_timestep_skip_range: Tuple[int, int] = (-1, 641)

2
src/diffusers/hooks/hooks.py
View File

@@ -45,7 +45,7 @@ class ModelHook:
def deinitalize_hook(self, module: torch.nn.Module) -> torch.nn.Module:
r"""
Hook that is executed when a model is deinitalized.
Hook that is executed when a model is deinitialized.
Args:
module (`torch.nn.Module`):

2
src/diffusers/hooks/layerwise_casting.py
View File

@@ -62,7 +62,7 @@ class LayerwiseCastingHook(ModelHook):
def deinitalize_hook(self, module: torch.nn.Module):
raise NotImplementedError(
"LayerwiseCastingHook does not support deinitalization. A model once enabled with layerwise casting will "
"LayerwiseCastingHook does not support deinitialization. A model once enabled with layerwise casting will "
"have casted its weights to a lower precision dtype for storage. Casting this back to the original dtype "
"will lead to precision loss, which might have an impact on the model's generation quality. The model should "
"be re-initialized and loaded in the original dtype."

2
src/diffusers/loaders/peft.py
View File

@@ -251,7 +251,7 @@ class PeftAdapterMixin:
rank = {}
for key, val in state_dict.items():
# Cannot figure out rank from lora layers that don't have atleast 2 dimensions.
# Cannot figure out rank from lora layers that don't have at least 2 dimensions.
# Bias layers in LoRA only have a single dimension
if "lora_B" in key and val.ndim > 1:
# Check out https://github.com/huggingface/peft/pull/2419 for the `^` symbol.

4
src/diffusers/models/autoencoders/autoencoder_kl.py
View File

@@ -63,8 +63,8 @@ class AutoencoderKL(ModelMixin, ConfigMixin, FromOriginalModelMixin, PeftAdapter
Synthesis with Latent Diffusion Models](https://huggingface.co/papers/2112.10752) paper.
force_upcast (`bool`, *optional*, default to `True`):
If enabled it will force the VAE to run in float32 for high image resolution pipelines, such as SD-XL. VAE
can be fine-tuned / trained to a lower range without loosing too much precision in which case
`force_upcast` can be set to `False` - see: https://huggingface.co/madebyollin/sdxl-vae-fp16-fix
can be fine-tuned / trained to a lower range without losing too much precision in which case `force_upcast`
can be set to `False` - see: https://huggingface.co/madebyollin/sdxl-vae-fp16-fix
mid_block_add_attention (`bool`, *optional*, default to `True`):
If enabled, the mid_block of the Encoder and Decoder will have attention blocks. If set to false, the
mid_block will only have resnet blocks

4
src/diffusers/models/autoencoders/autoencoder_kl_allegro.py
View File

@@ -715,8 +715,8 @@ class AutoencoderKLAllegro(ModelMixin, ConfigMixin):
Synthesis with Latent Diffusion Models](https://huggingface.co/papers/2112.10752) paper.
force_upcast (`bool`, default to `True`):
If enabled it will force the VAE to run in float32 for high image resolution pipelines, such as SD-XL. VAE
can be fine-tuned / trained to a lower range without loosing too much precision in which case
`force_upcast` can be set to `False` - see: https://huggingface.co/madebyollin/sdxl-vae-fp16-fix
can be fine-tuned / trained to a lower range without losing too much precision in which case `force_upcast`
can be set to `False` - see: https://huggingface.co/madebyollin/sdxl-vae-fp16-fix
"""
_supports_gradient_checkpointing = True

4
src/diffusers/models/autoencoders/autoencoder_kl_cogvideox.py
View File

@@ -983,8 +983,8 @@ class AutoencoderKLCogVideoX(ModelMixin, ConfigMixin, FromOriginalModelMixin):
Synthesis with Latent Diffusion Models](https://huggingface.co/papers/2112.10752) paper.
force_upcast (`bool`, *optional*, default to `True`):
If enabled it will force the VAE to run in float32 for high image resolution pipelines, such as SD-XL. VAE
can be fine-tuned / trained to a lower range without loosing too much precision in which case
`force_upcast` can be set to `False` - see: https://huggingface.co/madebyollin/sdxl-vae-fp16-fix
can be fine-tuned / trained to a lower range without losing too much precision in which case `force_upcast`
can be set to `False` - see: https://huggingface.co/madebyollin/sdxl-vae-fp16-fix
"""
_supports_gradient_checkpointing = True

4
src/diffusers/models/autoencoders/autoencoder_kl_temporal_decoder.py
View File

@@ -161,8 +161,8 @@ class AutoencoderKLTemporalDecoder(ModelMixin, ConfigMixin):
Synthesis with Latent Diffusion Models](https://huggingface.co/papers/2112.10752) paper.
force_upcast (`bool`, *optional*, default to `True`):
If enabled it will force the VAE to run in float32 for high image resolution pipelines, such as SD-XL. VAE
can be fine-tuned / trained to a lower range without loosing too much precision in which case
`force_upcast` can be set to `False` - see: https://huggingface.co/madebyollin/sdxl-vae-fp16-fix
can be fine-tuned / trained to a lower range without losing too much precision in which case `force_upcast`
can be set to `False` - see: https://huggingface.co/madebyollin/sdxl-vae-fp16-fix
"""
_supports_gradient_checkpointing = True

2
src/diffusers/pipelines/consisid/consisid_utils.py
View File

@@ -166,7 +166,7 @@ def process_face_embeddings(
raise RuntimeError("facexlib align face fail")
align_face = face_helper_1.cropped_faces[0] # (512, 512, 3) # RGB
# incase insightface didn't detect face
# in case insightface didn't detect face
if id_ante_embedding is None:
logger.warning("Failed to detect face using insightface. Extracting embedding with align face")
id_ante_embedding = face_helper_2.get_feat(align_face)

4
tests/lora/utils.py
View File

@@ -1092,7 +1092,7 @@ class PeftLoraLoaderMixinTests:
def test_simple_inference_with_text_denoiser_multi_adapter_block_lora(self):
"""
Tests a simple inference with lora attached to text encoder and unet, attaches
multiple adapters and set differnt weights for different blocks (i.e. block lora)
multiple adapters and set different weights for different blocks (i.e. block lora)
"""
for scheduler_cls in self.scheduler_classes:
components, text_lora_config, denoiser_lora_config = self.get_dummy_components(scheduler_cls)
@@ -1636,7 +1636,7 @@ class PeftLoraLoaderMixinTests:
pipe.fuse_lora(components=self.pipeline_class._lora_loadable_modules, adapter_names=["adapter-1"])
self.assertTrue(pipe.num_fused_loras == 1, f"{pipe.num_fused_loras=}, {pipe.fused_loras=}")
# Fusing should still keep the LoRA layers so outpout should remain the same
# Fusing should still keep the LoRA layers so output should remain the same
outputs_lora_1_fused = pipe(**inputs, generator=torch.manual_seed(0))[0]
self.assertTrue(

2
tests/pipelines/cogvideo/test_cogvideox_image2video.py
View File

@@ -270,7 +270,7 @@ class CogVideoXImageToVideoPipelineFastTests(PipelineTesterMixin, unittest.TestC
generator_device = "cpu"
components = self.get_dummy_components()
# The reason to modify it this way is because I2V Transformer limits the generation to resolutions used during initalization.
# The reason to modify it this way is because I2V Transformer limits the generation to resolutions used during initialization.
# This limitation comes from using learned positional embeddings which cannot be generated on-the-fly like sincos or RoPE embeddings.
# See the if-statement on "self.use_learned_positional_embeddings" in diffusers/models/embeddings.py
components["transformer"] = CogVideoXTransformer3DModel.from_config(

2
tests/pipelines/consisid/test_consisid.py
View File

@@ -280,7 +280,7 @@ class ConsisIDPipelineFastTests(PipelineTesterMixin, unittest.TestCase):
generator_device = "cpu"
components = self.get_dummy_components()
# The reason to modify it this way is because ConsisID Transformer limits the generation to resolutions used during initalization.
# The reason to modify it this way is because ConsisID Transformer limits the generation to resolutions used during initialization.
# This limitation comes from using learned positional embeddings which cannot be generated on-the-fly like sincos or RoPE embeddings.
# See the if-statement on "self.use_learned_positional_embeddings" in diffusers/models/embeddings.py
components["transformer"] = ConsisIDTransformer3DModel.from_config(

2
tests/pipelines/kolors/test_kolors_img2img.py
View File

@@ -155,6 +155,6 @@ class KolorsPipelineImg2ImgFastTests(PipelineTesterMixin, unittest.TestCase):
def test_float16_inference(self):
super().test_float16_inference(expected_max_diff=7e-2)
@unittest.skip("Test not supported because kolors img2img doesn't take pooled embeds as inputs unline kolors t2i.")
@unittest.skip("Test not supported because kolors img2img doesn't take pooled embeds as inputs unlike kolors t2i.")
def test_encode_prompt_works_in_isolation(self):
pass

2
tests/pipelines/pag/test_pag_pixart_sigma.py
View File

@@ -254,7 +254,7 @@ class PixArtSigmaPAGPipelineFastTests(PipelineTesterMixin, unittest.TestCase):
assert_mean_pixel_difference(to_np(output_with_slicing1[0]), to_np(output_without_slicing[0]))
assert_mean_pixel_difference(to_np(output_with_slicing2[0]), to_np(output_without_slicing[0]))
# Because we have `pag_applied_layers` we cannot direcly apply
# Because we have `pag_applied_layers` we cannot directly apply
# `set_default_attn_processor`
def test_dict_tuple_outputs_equivalent(self, expected_slice=None, expected_max_difference=1e-4):
components = self.get_dummy_components()

2
tests/pipelines/stable_unclip/test_stable_unclip.py
View File

@@ -227,7 +227,7 @@ class StableUnCLIPPipelineIntegrationTests(unittest.TestCase):
pipe.enable_sequential_cpu_offload()
generator = torch.Generator(device="cpu").manual_seed(0)
output = pipe("anime turle", generator=generator, output_type="np")
output = pipe("anime turtle", generator=generator, output_type="np")
image = output.images[0]

4
tests/pipelines/stable_unclip/test_stable_unclip_img2img.py
View File

@@ -250,7 +250,7 @@ class StableUnCLIPImg2ImgPipelineIntegrationTests(unittest.TestCase):
pipe.enable_sequential_cpu_offload()
generator = torch.Generator(device="cpu").manual_seed(0)
output = pipe(input_image, "anime turle", generator=generator, output_type="np")
output = pipe(input_image, "anime turtle", generator=generator, output_type="np")
image = output.images[0]
@@ -277,7 +277,7 @@ class StableUnCLIPImg2ImgPipelineIntegrationTests(unittest.TestCase):
pipe.enable_sequential_cpu_offload()
generator = torch.Generator(device="cpu").manual_seed(0)
output = pipe(input_image, "anime turle", generator=generator, output_type="np")
output = pipe(input_image, "anime turtle", generator=generator, output_type="np")
image = output.images[0]

10
tests/pipelines/test_pipelines_common.py
View File

@@ -2096,11 +2096,11 @@ class PipelineTesterMixin:
with torch.no_grad():
encoded_prompt_outputs = pipe_with_just_text_encoder.encode_prompt(**encode_prompt_inputs)
# Programatically determine the reutrn names of `encode_prompt.`
ast_vistor = ReturnNameVisitor()
encode_prompt_tree = ast_vistor.get_ast_tree(cls=self.pipeline_class)
ast_vistor.visit(encode_prompt_tree)
prompt_embed_kwargs = ast_vistor.return_names
# Programmatically determine the return names of `encode_prompt.`
ast_visitor = ReturnNameVisitor()
encode_prompt_tree = ast_visitor.get_ast_tree(cls=self.pipeline_class)
ast_visitor.visit(encode_prompt_tree)
prompt_embed_kwargs = ast_visitor.return_names
prompt_embeds_kwargs = dict(zip(prompt_embed_kwargs, encoded_prompt_outputs))
# Pack the outputs of `encode_prompt`.

2
tests/quantization/bnb/test_4bit.py
View File

@@ -205,7 +205,7 @@ class BnB4BitBasicTests(Base4bitTests):
def test_original_dtype(self):
r"""
A simple test to check if the model succesfully stores the original dtype
A simple test to check if the model successfully stores the original dtype
"""
self.assertTrue("_pre_quantization_dtype" in self.model_4bit.config)
self.assertFalse("_pre_quantization_dtype" in self.model_fp16.config)

2
tests/quantization/bnb/test_mixed_int8.py
View File

@@ -195,7 +195,7 @@ class BnB8bitBasicTests(Base8bitTests):
def test_original_dtype(self):
r"""
A simple test to check if the model succesfully stores the original dtype
A simple test to check if the model successfully stores the original dtype
"""
self.assertTrue("_pre_quantization_dtype" in self.model_8bit.config)
self.assertFalse("_pre_quantization_dtype" in self.model_fp16.config)

2
tests/single_file/test_model_autoencoder_dc_single_file.py
View File

@@ -95,7 +95,7 @@ class AutoencoderDCSingleFileTests(unittest.TestCase):
# `in` variant checkpoints require passing in a `config` parameter
# in order to set the scaling factor correctly.
# `in` and `mix` variants have the same keys and we cannot automatically infer a scaling factor.
# We default to using teh `mix` config
# We default to using the `mix` config
repo_id = "mit-han-lab/dc-ae-f128c512-in-1.0-diffusers"
ckpt_path = "https://huggingface.co/mit-han-lab/dc-ae-f128c512-in-1.0/blob/main/model.safetensors"

2
utils/custom_init_isort.py
View File

@@ -252,7 +252,7 @@ def sort_imports(file: str, check_only: bool = True):
code, start_prompt="_import_structure = {", end_prompt="if TYPE_CHECKING:"
)
# We ignore block 0 (everything untils start_prompt) and the last block (everything after end_prompt).
# We ignore block 0 (everything until start_prompt) and the last block (everything after end_prompt).
for block_idx in range(1, len(main_blocks) - 1):
# Check if the block contains some `_import_structure`s thingy to sort.
block = main_blocks[block_idx]