Skip to content

GitLab

Unverified Commit be4afa0b authored by Mark Van Aken's avatar Mark Van Aken Committed by GitHub
Browse files

#7535 Update FloatTensor type hints to Tensor (#7883) 

* find & replace all FloatTensors to Tensor

* apply formatting

* Update torch.FloatTensor to torch.Tensor in the remaining files

* formatting

* Fix the rest of the places where FloatTensor is used as well as in documentation

* formatting

* Update new file from FloatTensor to Tensor
parent 04f4bd54
Show whitespace changes
Inline Side-by-side
Showing with 365 additions and 365 deletions
examples/community/llm_grounded_diffusion.py
View file @ be4afa0b
...@@ -756,13 +756,13 @@ class LLMGroundedDiffusionPipeline( ...@@ -756,13 +756,13 @@ class LLMGroundedDiffusionPipeline(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
ip_adapter_image: Optional[PipelineImageInput] = None, ip_adapter_image: Optional[PipelineImageInput] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None, callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
callback_steps: int = 1, callback_steps: int = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
clip_skip: Optional[int] = None, clip_skip: Optional[int] = None,
...@@ -807,14 +807,14 @@ class LLMGroundedDiffusionPipeline( ...@@ -807,14 +807,14 @@ class LLMGroundedDiffusionPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
generation deterministic. generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor is generated by sampling using the supplied random `generator`. tensor is generated by sampling using the supplied random `generator`.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
provided, text embeddings are generated from the `prompt` input argument. provided, text embeddings are generated from the `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument. not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
ip_adapter_image: (`PipelineImageInput`, *optional*): Optional image input to work with IP Adapters. ip_adapter_image: (`PipelineImageInput`, *optional*): Optional image input to work with IP Adapters.
...@@ -825,7 +825,7 @@ class LLMGroundedDiffusionPipeline( ...@@ -825,7 +825,7 @@ class LLMGroundedDiffusionPipeline(
plain tuple. plain tuple.
callback (`Callable`, *optional*): callback (`Callable`, *optional*):
A function that calls every `callback_steps` steps during inference. The function is called with the A function that calls every `callback_steps` steps during inference. The function is called with the
following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`. following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
callback_steps (`int`, *optional*, defaults to 1): callback_steps (`int`, *optional*, defaults to 1):
The frequency at which the `callback` function is called. If not specified, the callback is called at The frequency at which the `callback` function is called. If not specified, the callback is called at
every step. every step.
...@@ -1194,8 +1194,8 @@ class LLMGroundedDiffusionPipeline( ...@@ -1194,8 +1194,8 @@ class LLMGroundedDiffusionPipeline(
num_images_per_prompt, num_images_per_prompt,
do_classifier_free_guidance, do_classifier_free_guidance,
negative_prompt=None, negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
**kwargs, **kwargs,
): ):
...@@ -1227,8 +1227,8 @@ class LLMGroundedDiffusionPipeline( ...@@ -1227,8 +1227,8 @@ class LLMGroundedDiffusionPipeline(
num_images_per_prompt, num_images_per_prompt,
do_classifier_free_guidance, do_classifier_free_guidance,
negative_prompt=None, negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None, clip_skip: Optional[int] = None,
): ):
...@@ -1248,10 +1248,10 @@ class LLMGroundedDiffusionPipeline( ...@@ -1248,10 +1248,10 @@ class LLMGroundedDiffusionPipeline(
The prompt or prompts not to guide the image generation. If not defined, one has to pass The prompt or prompts not to guide the image generation. If not defined, one has to pass
`negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
less than `1`). less than `1`).
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
...@@ -1509,7 +1509,7 @@ class LLMGroundedDiffusionPipeline( ...@@ -1509,7 +1509,7 @@ class LLMGroundedDiffusionPipeline(
data type of the generated embeddings data type of the generated embeddings
Returns: Returns:
`torch.FloatTensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)` `torch.Tensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)`
""" """
assert len(w.shape) == 1 assert len(w.shape) == 1
w = w * 1000.0 w = w * 1000.0
......
examples/community/lpw_stable_diffusion.py
View file @ be4afa0b
...@@ -378,7 +378,7 @@ def preprocess_image(image, batch_size): ...@@ -378,7 +378,7 @@ def preprocess_image(image, batch_size):
def preprocess_mask(mask, batch_size, scale_factor=8): def preprocess_mask(mask, batch_size, scale_factor=8):
if not isinstance(mask, torch.FloatTensor): if not isinstance(mask, torch.Tensor):
mask = mask.convert("L") mask = mask.convert("L")
w, h = mask.size w, h = mask.size
w, h = (x - x % 8 for x in (w, h)) # resize to integer multiple of 8 w, h = (x - x % 8 for x in (w, h)) # resize to integer multiple of 8
...@@ -543,8 +543,8 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -543,8 +543,8 @@ class StableDiffusionLongPromptWeightingPipeline(
do_classifier_free_guidance, do_classifier_free_guidance,
negative_prompt=None, negative_prompt=None,
max_embeddings_multiples=3, max_embeddings_multiples=3,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
): ):
r""" r"""
Encodes the prompt into text encoder hidden states. Encodes the prompt into text encoder hidden states.
...@@ -767,8 +767,8 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -767,8 +767,8 @@ class StableDiffusionLongPromptWeightingPipeline(
self, self,
prompt: Union[str, List[str]], prompt: Union[str, List[str]],
negative_prompt: Optional[Union[str, List[str]]] = None, negative_prompt: Optional[Union[str, List[str]]] = None,
image: Union[torch.FloatTensor, PIL.Image.Image] = None, image: Union[torch.Tensor, PIL.Image.Image] = None,
mask_image: Union[torch.FloatTensor, PIL.Image.Image] = None, mask_image: Union[torch.Tensor, PIL.Image.Image] = None,
height: int = 512, height: int = 512,
width: int = 512, width: int = 512,
num_inference_steps: int = 50, num_inference_steps: int = 50,
...@@ -778,13 +778,13 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -778,13 +778,13 @@ class StableDiffusionLongPromptWeightingPipeline(
add_predicted_noise: Optional[bool] = False, add_predicted_noise: Optional[bool] = False,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
max_embeddings_multiples: Optional[int] = 3, max_embeddings_multiples: Optional[int] = 3,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None, callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
is_cancelled_callback: Optional[Callable[[], bool]] = None, is_cancelled_callback: Optional[Callable[[], bool]] = None,
callback_steps: int = 1, callback_steps: int = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
...@@ -798,10 +798,10 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -798,10 +798,10 @@ class StableDiffusionLongPromptWeightingPipeline(
negative_prompt (`str` or `List[str]`, *optional*): negative_prompt (`str` or `List[str]`, *optional*):
The prompt or prompts not to guide the image generation. Ignored when not using guidance (i.e., ignored The prompt or prompts not to guide the image generation. Ignored when not using guidance (i.e., ignored
if `guidance_scale` is less than `1`). if `guidance_scale` is less than `1`).
image (`torch.FloatTensor` or `PIL.Image.Image`): image (`torch.Tensor` or `PIL.Image.Image`):
`Image`, or tensor representing an image batch, that will be used as the starting point for the `Image`, or tensor representing an image batch, that will be used as the starting point for the
process. process.
mask_image (`torch.FloatTensor` or `PIL.Image.Image`): mask_image (`torch.Tensor` or `PIL.Image.Image`):
`Image`, or tensor representing an image batch, to mask `image`. White pixels in the mask will be `Image`, or tensor representing an image batch, to mask `image`. White pixels in the mask will be
replaced by noise and therefore repainted, while black pixels will be preserved. If `mask_image` is a replaced by noise and therefore repainted, while black pixels will be preserved. If `mask_image` is a
PIL image, it will be converted to a single channel (luminance) before use. If it's a tensor, it should PIL image, it will be converted to a single channel (luminance) before use. If it's a tensor, it should
...@@ -836,14 +836,14 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -836,14 +836,14 @@ class StableDiffusionLongPromptWeightingPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html) One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
to make generation deterministic. to make generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will ge generated by sampling using the supplied random `generator`. tensor will ge generated by sampling using the supplied random `generator`.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
...@@ -857,7 +857,7 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -857,7 +857,7 @@ class StableDiffusionLongPromptWeightingPipeline(
plain tuple. plain tuple.
callback (`Callable`, *optional*): callback (`Callable`, *optional*):
A function that will be called every `callback_steps` steps during inference. The function will be A function that will be called every `callback_steps` steps during inference. The function will be
called with the following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`. called with the following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
is_cancelled_callback (`Callable`, *optional*): is_cancelled_callback (`Callable`, *optional*):
A function that will be called every `callback_steps` steps during inference. If the function returns A function that will be called every `callback_steps` steps during inference. If the function returns
`True`, the inference will be cancelled. `True`, the inference will be cancelled.
...@@ -1032,13 +1032,13 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -1032,13 +1032,13 @@ class StableDiffusionLongPromptWeightingPipeline(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
max_embeddings_multiples: Optional[int] = 3, max_embeddings_multiples: Optional[int] = 3,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None, callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
is_cancelled_callback: Optional[Callable[[], bool]] = None, is_cancelled_callback: Optional[Callable[[], bool]] = None,
callback_steps: int = 1, callback_steps: int = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
...@@ -1072,14 +1072,14 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -1072,14 +1072,14 @@ class StableDiffusionLongPromptWeightingPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html) One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
to make generation deterministic. to make generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will ge generated by sampling using the supplied random `generator`. tensor will ge generated by sampling using the supplied random `generator`.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
...@@ -1093,7 +1093,7 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -1093,7 +1093,7 @@ class StableDiffusionLongPromptWeightingPipeline(
plain tuple. plain tuple.
callback (`Callable`, *optional*): callback (`Callable`, *optional*):
A function that will be called every `callback_steps` steps during inference. The function will be A function that will be called every `callback_steps` steps during inference. The function will be
called with the following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`. called with the following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
is_cancelled_callback (`Callable`, *optional*): is_cancelled_callback (`Callable`, *optional*):
A function that will be called every `callback_steps` steps during inference. If the function returns A function that will be called every `callback_steps` steps during inference. If the function returns
`True`, the inference will be cancelled. `True`, the inference will be cancelled.
...@@ -1137,7 +1137,7 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -1137,7 +1137,7 @@ class StableDiffusionLongPromptWeightingPipeline(
def img2img( def img2img(
self, self,
image: Union[torch.FloatTensor, PIL.Image.Image], image: Union[torch.Tensor, PIL.Image.Image],
prompt: Union[str, List[str]], prompt: Union[str, List[str]],
negative_prompt: Optional[Union[str, List[str]]] = None, negative_prompt: Optional[Union[str, List[str]]] = None,
strength: float = 0.8, strength: float = 0.8,
...@@ -1146,12 +1146,12 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -1146,12 +1146,12 @@ class StableDiffusionLongPromptWeightingPipeline(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: Optional[float] = 0.0, eta: Optional[float] = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
max_embeddings_multiples: Optional[int] = 3, max_embeddings_multiples: Optional[int] = 3,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None, callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
is_cancelled_callback: Optional[Callable[[], bool]] = None, is_cancelled_callback: Optional[Callable[[], bool]] = None,
callback_steps: int = 1, callback_steps: int = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
...@@ -1159,7 +1159,7 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -1159,7 +1159,7 @@ class StableDiffusionLongPromptWeightingPipeline(
r""" r"""
Function for image-to-image generation. Function for image-to-image generation.
Args: Args:
image (`torch.FloatTensor` or `PIL.Image.Image`): image (`torch.Tensor` or `PIL.Image.Image`):
`Image`, or tensor representing an image batch, that will be used as the starting point for the `Image`, or tensor representing an image batch, that will be used as the starting point for the
process. process.
prompt (`str` or `List[str]`): prompt (`str` or `List[str]`):
...@@ -1190,10 +1190,10 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -1190,10 +1190,10 @@ class StableDiffusionLongPromptWeightingPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html) One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
to make generation deterministic. to make generation deterministic.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
...@@ -1207,7 +1207,7 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -1207,7 +1207,7 @@ class StableDiffusionLongPromptWeightingPipeline(
plain tuple. plain tuple.
callback (`Callable`, *optional*): callback (`Callable`, *optional*):
A function that will be called every `callback_steps` steps during inference. The function will be A function that will be called every `callback_steps` steps during inference. The function will be
called with the following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`. called with the following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
is_cancelled_callback (`Callable`, *optional*): is_cancelled_callback (`Callable`, *optional*):
A function that will be called every `callback_steps` steps during inference. If the function returns A function that will be called every `callback_steps` steps during inference. If the function returns
`True`, the inference will be cancelled. `True`, the inference will be cancelled.
...@@ -1249,8 +1249,8 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -1249,8 +1249,8 @@ class StableDiffusionLongPromptWeightingPipeline(
def inpaint( def inpaint(
self, self,
image: Union[torch.FloatTensor, PIL.Image.Image], image: Union[torch.Tensor, PIL.Image.Image],
mask_image: Union[torch.FloatTensor, PIL.Image.Image], mask_image: Union[torch.Tensor, PIL.Image.Image],
prompt: Union[str, List[str]], prompt: Union[str, List[str]],
negative_prompt: Optional[Union[str, List[str]]] = None, negative_prompt: Optional[Union[str, List[str]]] = None,
strength: float = 0.8, strength: float = 0.8,
...@@ -1260,12 +1260,12 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -1260,12 +1260,12 @@ class StableDiffusionLongPromptWeightingPipeline(
add_predicted_noise: Optional[bool] = False, add_predicted_noise: Optional[bool] = False,
eta: Optional[float] = 0.0, eta: Optional[float] = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
max_embeddings_multiples: Optional[int] = 3, max_embeddings_multiples: Optional[int] = 3,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None, callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
is_cancelled_callback: Optional[Callable[[], bool]] = None, is_cancelled_callback: Optional[Callable[[], bool]] = None,
callback_steps: int = 1, callback_steps: int = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
...@@ -1273,10 +1273,10 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -1273,10 +1273,10 @@ class StableDiffusionLongPromptWeightingPipeline(
r""" r"""
Function for inpaint. Function for inpaint.
Args: Args:
image (`torch.FloatTensor` or `PIL.Image.Image`): image (`torch.Tensor` or `PIL.Image.Image`):
`Image`, or tensor representing an image batch, that will be used as the starting point for the `Image`, or tensor representing an image batch, that will be used as the starting point for the
process. This is the image whose masked region will be inpainted. process. This is the image whose masked region will be inpainted.
mask_image (`torch.FloatTensor` or `PIL.Image.Image`): mask_image (`torch.Tensor` or `PIL.Image.Image`):
`Image`, or tensor representing an image batch, to mask `image`. White pixels in the mask will be `Image`, or tensor representing an image batch, to mask `image`. White pixels in the mask will be
replaced by noise and therefore repainted, while black pixels will be preserved. If `mask_image` is a replaced by noise and therefore repainted, while black pixels will be preserved. If `mask_image` is a
PIL image, it will be converted to a single channel (luminance) before use. If it's a tensor, it should PIL image, it will be converted to a single channel (luminance) before use. If it's a tensor, it should
...@@ -1311,10 +1311,10 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -1311,10 +1311,10 @@ class StableDiffusionLongPromptWeightingPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html) One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
to make generation deterministic. to make generation deterministic.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
...@@ -1328,7 +1328,7 @@ class StableDiffusionLongPromptWeightingPipeline( ...@@ -1328,7 +1328,7 @@ class StableDiffusionLongPromptWeightingPipeline(
plain tuple. plain tuple.
callback (`Callable`, *optional*): callback (`Callable`, *optional*):
A function that will be called every `callback_steps` steps during inference. The function will be A function that will be called every `callback_steps` steps during inference. The function will be
called with the following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`. called with the following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
is_cancelled_callback (`Callable`, *optional*): is_cancelled_callback (`Callable`, *optional*):
A function that will be called every `callback_steps` steps during inference. If the function returns A function that will be called every `callback_steps` steps during inference. If the function returns
`True`, the inference will be cancelled. `True`, the inference will be cancelled.
......
examples/community/lpw_stable_diffusion_xl.py
View file @ be4afa0b
...@@ -694,10 +694,10 @@ class SDXLLongPromptWeightingPipeline( ...@@ -694,10 +694,10 @@ class SDXLLongPromptWeightingPipeline(
do_classifier_free_guidance: bool = True, do_classifier_free_guidance: bool = True,
negative_prompt: Optional[str] = None, negative_prompt: Optional[str] = None,
negative_prompt_2: Optional[str] = None, negative_prompt_2: Optional[str] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
): ):
r""" r"""
...@@ -722,17 +722,17 @@ class SDXLLongPromptWeightingPipeline( ...@@ -722,17 +722,17 @@ class SDXLLongPromptWeightingPipeline(
negative_prompt_2 (`str` or `List[str]`, *optional*): negative_prompt_2 (`str` or `List[str]`, *optional*):
The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and
`text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders `text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument. If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument. input argument.
...@@ -1320,7 +1320,7 @@ class SDXLLongPromptWeightingPipeline( ...@@ -1320,7 +1320,7 @@ class SDXLLongPromptWeightingPipeline(
data type of the generated embeddings data type of the generated embeddings
Returns: Returns:
`torch.FloatTensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)` `torch.Tensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)`
""" """
assert len(w.shape) == 1 assert len(w.shape) == 1
w = w * 1000.0 w = w * 1000.0
...@@ -1378,7 +1378,7 @@ class SDXLLongPromptWeightingPipeline( ...@@ -1378,7 +1378,7 @@ class SDXLLongPromptWeightingPipeline(
prompt_2: Optional[str] = None, prompt_2: Optional[str] = None,
image: Optional[PipelineImageInput] = None, image: Optional[PipelineImageInput] = None,
mask_image: Optional[PipelineImageInput] = None, mask_image: Optional[PipelineImageInput] = None,
masked_image_latents: Optional[torch.FloatTensor] = None, masked_image_latents: Optional[torch.Tensor] = None,
height: Optional[int] = None, height: Optional[int] = None,
width: Optional[int] = None, width: Optional[int] = None,
strength: float = 0.8, strength: float = 0.8,
...@@ -1392,12 +1392,12 @@ class SDXLLongPromptWeightingPipeline( ...@@ -1392,12 +1392,12 @@ class SDXLLongPromptWeightingPipeline(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
ip_adapter_image: Optional[PipelineImageInput] = None, ip_adapter_image: Optional[PipelineImageInput] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
...@@ -1481,23 +1481,23 @@ class SDXLLongPromptWeightingPipeline( ...@@ -1481,23 +1481,23 @@ class SDXLLongPromptWeightingPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html) One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
to make generation deterministic. to make generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will ge generated by sampling using the supplied random `generator`. tensor will ge generated by sampling using the supplied random `generator`.
ip_adapter_image: (`PipelineImageInput`, *optional*): ip_adapter_image: (`PipelineImageInput`, *optional*):
Optional image input to work with IP Adapters. Optional image input to work with IP Adapters.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument. If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument. input argument.
...@@ -1926,12 +1926,12 @@ class SDXLLongPromptWeightingPipeline( ...@@ -1926,12 +1926,12 @@ class SDXLLongPromptWeightingPipeline(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
ip_adapter_image: Optional[PipelineImageInput] = None, ip_adapter_image: Optional[PipelineImageInput] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
...@@ -2001,12 +2001,12 @@ class SDXLLongPromptWeightingPipeline( ...@@ -2001,12 +2001,12 @@ class SDXLLongPromptWeightingPipeline(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
ip_adapter_image: Optional[PipelineImageInput] = None, ip_adapter_image: Optional[PipelineImageInput] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
...@@ -2066,7 +2066,7 @@ class SDXLLongPromptWeightingPipeline( ...@@ -2066,7 +2066,7 @@ class SDXLLongPromptWeightingPipeline(
prompt_2: Optional[str] = None, prompt_2: Optional[str] = None,
image: Optional[PipelineImageInput] = None, image: Optional[PipelineImageInput] = None,
mask_image: Optional[PipelineImageInput] = None, mask_image: Optional[PipelineImageInput] = None,
masked_image_latents: Optional[torch.FloatTensor] = None, masked_image_latents: Optional[torch.Tensor] = None,
height: Optional[int] = None, height: Optional[int] = None,
width: Optional[int] = None, width: Optional[int] = None,
strength: float = 0.8, strength: float = 0.8,
...@@ -2080,12 +2080,12 @@ class SDXLLongPromptWeightingPipeline( ...@@ -2080,12 +2080,12 @@ class SDXLLongPromptWeightingPipeline(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
ip_adapter_image: Optional[PipelineImageInput] = None, ip_adapter_image: Optional[PipelineImageInput] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
......
examples/community/masked_stable_diffusion_img2img.py
View file @ be4afa0b
...@@ -16,10 +16,10 @@ class MaskedStableDiffusionImg2ImgPipeline(StableDiffusionImg2ImgPipeline): ...@@ -16,10 +16,10 @@ class MaskedStableDiffusionImg2ImgPipeline(StableDiffusionImg2ImgPipeline):
self, self,
prompt: Union[str, List[str]] = None, prompt: Union[str, List[str]] = None,
image: Union[ image: Union[
torch.FloatTensor, torch.Tensor,
PIL.Image.Image, PIL.Image.Image,
np.ndarray, np.ndarray,
List[torch.FloatTensor], List[torch.Tensor],
List[PIL.Image.Image], List[PIL.Image.Image],
List[np.ndarray], List[np.ndarray],
] = None, ] = None,
...@@ -30,18 +30,18 @@ class MaskedStableDiffusionImg2ImgPipeline(StableDiffusionImg2ImgPipeline): ...@@ -30,18 +30,18 @@ class MaskedStableDiffusionImg2ImgPipeline(StableDiffusionImg2ImgPipeline):
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: Optional[float] = 0.0, eta: Optional[float] = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None, callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
callback_steps: int = 1, callback_steps: int = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
mask: Union[ mask: Union[
torch.FloatTensor, torch.Tensor,
PIL.Image.Image, PIL.Image.Image,
np.ndarray, np.ndarray,
List[torch.FloatTensor], List[torch.Tensor],
List[PIL.Image.Image], List[PIL.Image.Image],
List[np.ndarray], List[np.ndarray],
] = None, ] = None,
...@@ -52,7 +52,7 @@ class MaskedStableDiffusionImg2ImgPipeline(StableDiffusionImg2ImgPipeline): ...@@ -52,7 +52,7 @@ class MaskedStableDiffusionImg2ImgPipeline(StableDiffusionImg2ImgPipeline):
Args: Args:
prompt (`str` or `List[str]`, *optional*): prompt (`str` or `List[str]`, *optional*):
The prompt or prompts to guide image generation. If not defined, you need to pass `prompt_embeds`. The prompt or prompts to guide image generation. If not defined, you need to pass `prompt_embeds`.
image (`torch.FloatTensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.FloatTensor]`, `List[PIL.Image.Image]`, or `List[np.ndarray]`): image (`torch.Tensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.Tensor]`, `List[PIL.Image.Image]`, or `List[np.ndarray]`):
`Image` or tensor representing an image batch to be used as the starting point. Can also accept image `Image` or tensor representing an image batch to be used as the starting point. Can also accept image
latents as `image`, but if passing latents directly it is not encoded again. latents as `image`, but if passing latents directly it is not encoded again.
strength (`float`, *optional*, defaults to 0.8): strength (`float`, *optional*, defaults to 0.8):
...@@ -78,10 +78,10 @@ class MaskedStableDiffusionImg2ImgPipeline(StableDiffusionImg2ImgPipeline): ...@@ -78,10 +78,10 @@ class MaskedStableDiffusionImg2ImgPipeline(StableDiffusionImg2ImgPipeline):
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
generation deterministic. generation deterministic.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
provided, text embeddings are generated from the `prompt` input argument. provided, text embeddings are generated from the `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument. not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
output_type (`str`, *optional*, defaults to `"pil"`): output_type (`str`, *optional*, defaults to `"pil"`):
...@@ -91,14 +91,14 @@ class MaskedStableDiffusionImg2ImgPipeline(StableDiffusionImg2ImgPipeline): ...@@ -91,14 +91,14 @@ class MaskedStableDiffusionImg2ImgPipeline(StableDiffusionImg2ImgPipeline):
plain tuple. plain tuple.
callback (`Callable`, *optional*): callback (`Callable`, *optional*):
A function that calls every `callback_steps` steps during inference. The function is called with the A function that calls every `callback_steps` steps during inference. The function is called with the
following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`. following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
callback_steps (`int`, *optional*, defaults to 1): callback_steps (`int`, *optional*, defaults to 1):
The frequency at which the `callback` function is called. If not specified, the callback is called at The frequency at which the `callback` function is called. If not specified, the callback is called at
every step. every step.
cross_attention_kwargs (`dict`, *optional*): cross_attention_kwargs (`dict`, *optional*):
A kwargs dictionary that if specified is passed along to the [`AttentionProcessor`] as defined in A kwargs dictionary that if specified is passed along to the [`AttentionProcessor`] as defined in
[`self.processor`](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py). [`self.processor`](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
mask (`torch.FloatTensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.FloatTensor]`, `List[PIL.Image.Image]`, or `List[np.ndarray]`, *optional*): mask (`torch.Tensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.Tensor]`, `List[PIL.Image.Image]`, or `List[np.ndarray]`, *optional*):
A mask with non-zero elements for the area to be inpainted. If not specified, no mask is applied. A mask with non-zero elements for the area to be inpainted. If not specified, no mask is applied.
Examples: Examples:
......
examples/community/mixture_canvas.py
View file @ be4afa0b
...@@ -154,7 +154,7 @@ class Text2ImageRegion(DiffusionRegion): ...@@ -154,7 +154,7 @@ class Text2ImageRegion(DiffusionRegion):
class Image2ImageRegion(DiffusionRegion): class Image2ImageRegion(DiffusionRegion):
"""Class defining a region where an image guided diffusion process is acting""" """Class defining a region where an image guided diffusion process is acting"""
reference_image: torch.FloatTensor = None reference_image: torch.Tensor = None
strength: float = 0.8 # Strength of the image strength: float = 0.8 # Strength of the image
def __post_init__(self): def __post_init__(self):
......
examples/community/multilingual_stable_diffusion.py
View file @ be4afa0b
...@@ -147,10 +147,10 @@ class MultilingualStableDiffusion(DiffusionPipeline, StableDiffusionMixin): ...@@ -147,10 +147,10 @@ class MultilingualStableDiffusion(DiffusionPipeline, StableDiffusionMixin):
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[torch.Generator] = None, generator: Optional[torch.Generator] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None, callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
callback_steps: int = 1, callback_steps: int = 1,
**kwargs, **kwargs,
): ):
...@@ -184,7 +184,7 @@ class MultilingualStableDiffusion(DiffusionPipeline, StableDiffusionMixin): ...@@ -184,7 +184,7 @@ class MultilingualStableDiffusion(DiffusionPipeline, StableDiffusionMixin):
generator (`torch.Generator`, *optional*): generator (`torch.Generator`, *optional*):
A [torch generator](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make generation A [torch generator](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make generation
deterministic. deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will ge generated by sampling using the supplied random `generator`. tensor will ge generated by sampling using the supplied random `generator`.
...@@ -196,7 +196,7 @@ class MultilingualStableDiffusion(DiffusionPipeline, StableDiffusionMixin): ...@@ -196,7 +196,7 @@ class MultilingualStableDiffusion(DiffusionPipeline, StableDiffusionMixin):
plain tuple. plain tuple.
callback (`Callable`, *optional*): callback (`Callable`, *optional*):
A function that will be called every `callback_steps` steps during inference. The function will be A function that will be called every `callback_steps` steps during inference. The function will be
called with the following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`. called with the following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
callback_steps (`int`, *optional*, defaults to 1): callback_steps (`int`, *optional*, defaults to 1):
The frequency at which the `callback` function will be called. If not specified, the callback will be The frequency at which the `callback` function will be called. If not specified, the callback will be
called at every step. called at every step.
......
examples/community/pipeline_animatediff_controlnet.py
View file @ be4afa0b
...@@ -198,8 +198,8 @@ class AnimateDiffControlNetPipeline( ...@@ -198,8 +198,8 @@ class AnimateDiffControlNetPipeline(
num_images_per_prompt, num_images_per_prompt,
do_classifier_free_guidance, do_classifier_free_guidance,
negative_prompt=None, negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None, clip_skip: Optional[int] = None,
): ):
...@@ -219,10 +219,10 @@ class AnimateDiffControlNetPipeline( ...@@ -219,10 +219,10 @@ class AnimateDiffControlNetPipeline(
The prompt or prompts not to guide the image generation. If not defined, one has to pass The prompt or prompts not to guide the image generation. If not defined, one has to pass
`negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
less than `1`). less than `1`).
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
...@@ -752,9 +752,9 @@ class AnimateDiffControlNetPipeline( ...@@ -752,9 +752,9 @@ class AnimateDiffControlNetPipeline(
num_videos_per_prompt: Optional[int] = 1, num_videos_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
ip_adapter_image: Optional[PipelineImageInput] = None, ip_adapter_image: Optional[PipelineImageInput] = None,
ip_adapter_image_embeds: Optional[PipelineImageInput] = None, ip_adapter_image_embeds: Optional[PipelineImageInput] = None,
conditioning_frames: Optional[List[PipelineImageInput]] = None, conditioning_frames: Optional[List[PipelineImageInput]] = None,
...@@ -798,20 +798,20 @@ class AnimateDiffControlNetPipeline( ...@@ -798,20 +798,20 @@ class AnimateDiffControlNetPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
generation deterministic. generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for video Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for video
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor is generated by sampling using the supplied random `generator`. Latents should be of shape tensor is generated by sampling using the supplied random `generator`. Latents should be of shape
`(batch_size, num_channel, num_frames, height, width)`. `(batch_size, num_channel, num_frames, height, width)`.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
provided, text embeddings are generated from the `prompt` input argument. provided, text embeddings are generated from the `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument. not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
ip_adapter_image (`PipelineImageInput`, *optional*): ip_adapter_image (`PipelineImageInput`, *optional*):
Optional image input to work with IP Adapters. Optional image input to work with IP Adapters.
ip_adapter_image_embeds (`List[torch.FloatTensor]`, *optional*): ip_adapter_image_embeds (`List[torch.Tensor]`, *optional*):
Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of IP-adapters. Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of IP-adapters.
Each element should be a tensor of shape `(batch_size, num_images, emb_dim)`. It should contain the negative image embedding Each element should be a tensor of shape `(batch_size, num_images, emb_dim)`. It should contain the negative image embedding
if `do_classifier_free_guidance` is set to `True`. if `do_classifier_free_guidance` is set to `True`.
...@@ -821,7 +821,7 @@ class AnimateDiffControlNetPipeline( ...@@ -821,7 +821,7 @@ class AnimateDiffControlNetPipeline(
are specified, images must be passed as a list such that each element of the list can be correctly are specified, images must be passed as a list such that each element of the list can be correctly
batched for input to a single ControlNet. batched for input to a single ControlNet.
output_type (`str`, *optional*, defaults to `"pil"`): output_type (`str`, *optional*, defaults to `"pil"`):
The output format of the generated video. Choose between `torch.FloatTensor`, `PIL.Image` or The output format of the generated video. Choose between `torch.Tensor`, `PIL.Image` or
`np.array`. `np.array`.
return_dict (`bool`, *optional*, defaults to `True`): return_dict (`bool`, *optional*, defaults to `True`):
Whether or not to return a [`~pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput`] instead Whether or not to return a [`~pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput`] instead
......
examples/community/pipeline_animatediff_img2video.py
View file @ be4afa0b
...@@ -315,8 +315,8 @@ class AnimateDiffImgToVideoPipeline( ...@@ -315,8 +315,8 @@ class AnimateDiffImgToVideoPipeline(
num_images_per_prompt, num_images_per_prompt,
do_classifier_free_guidance, do_classifier_free_guidance,
negative_prompt=None, negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None, clip_skip: Optional[int] = None,
): ):
...@@ -336,10 +336,10 @@ class AnimateDiffImgToVideoPipeline( ...@@ -336,10 +336,10 @@ class AnimateDiffImgToVideoPipeline(
The prompt or prompts not to guide the image generation. If not defined, one has to pass The prompt or prompts not to guide the image generation. If not defined, one has to pass
`negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
less than `1`). less than `1`).
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
...@@ -746,14 +746,14 @@ class AnimateDiffImgToVideoPipeline( ...@@ -746,14 +746,14 @@ class AnimateDiffImgToVideoPipeline(
num_videos_per_prompt: Optional[int] = 1, num_videos_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
ip_adapter_image: Optional[PipelineImageInput] = None, ip_adapter_image: Optional[PipelineImageInput] = None,
ip_adapter_image_embeds: Optional[PipelineImageInput] = None, ip_adapter_image_embeds: Optional[PipelineImageInput] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None, callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
callback_steps: Optional[int] = 1, callback_steps: Optional[int] = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
clip_skip: Optional[int] = None, clip_skip: Optional[int] = None,
...@@ -791,33 +791,33 @@ class AnimateDiffImgToVideoPipeline( ...@@ -791,33 +791,33 @@ class AnimateDiffImgToVideoPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
generation deterministic. generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for video Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for video
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor is generated by sampling using the supplied random `generator`. Latents should be of shape tensor is generated by sampling using the supplied random `generator`. Latents should be of shape
`(batch_size, num_channel, num_frames, height, width)`. `(batch_size, num_channel, num_frames, height, width)`.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
provided, text embeddings are generated from the `prompt` input argument. provided, text embeddings are generated from the `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument. not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
ip_adapter_image: (`PipelineImageInput`, *optional*): ip_adapter_image: (`PipelineImageInput`, *optional*):
Optional image input to work with IP Adapters. Optional image input to work with IP Adapters.
ip_adapter_image_embeds (`List[torch.FloatTensor]`, *optional*): ip_adapter_image_embeds (`List[torch.Tensor]`, *optional*):
Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of IP-adapters. Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of IP-adapters.
Each element should be a tensor of shape `(batch_size, num_images, emb_dim)`. It should contain the negative image embedding Each element should be a tensor of shape `(batch_size, num_images, emb_dim)`. It should contain the negative image embedding
if `do_classifier_free_guidance` is set to `True`. if `do_classifier_free_guidance` is set to `True`.
If not provided, embeddings are computed from the `ip_adapter_image` input argument. If not provided, embeddings are computed from the `ip_adapter_image` input argument.
output_type (`str`, *optional*, defaults to `"pil"`): output_type (`str`, *optional*, defaults to `"pil"`):
The output format of the generated video. Choose between `torch.FloatTensor`, `PIL.Image` or The output format of the generated video. Choose between `torch.Tensor`, `PIL.Image` or
`np.array`. `np.array`.
return_dict (`bool`, *optional*, defaults to `True`): return_dict (`bool`, *optional*, defaults to `True`):
Whether or not to return a [`AnimateDiffImgToVideoPipelineOutput`] instead Whether or not to return a [`AnimateDiffImgToVideoPipelineOutput`] instead
of a plain tuple. of a plain tuple.
callback (`Callable`, *optional*): callback (`Callable`, *optional*):
A function that calls every `callback_steps` steps during inference. The function is called with the A function that calls every `callback_steps` steps during inference. The function is called with the
following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`. following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
callback_steps (`int`, *optional*, defaults to 1): callback_steps (`int`, *optional*, defaults to 1):
The frequency at which the `callback` function is called. If not specified, the callback is called at The frequency at which the `callback` function is called. If not specified, the callback is called at
every step. every step.
......
examples/community/pipeline_demofusion_sdxl.py
View file @ be4afa0b
...@@ -187,10 +187,10 @@ class DemoFusionSDXLPipeline( ...@@ -187,10 +187,10 @@ class DemoFusionSDXLPipeline(
do_classifier_free_guidance: bool = True, do_classifier_free_guidance: bool = True,
negative_prompt: Optional[str] = None, negative_prompt: Optional[str] = None,
negative_prompt_2: Optional[str] = None, negative_prompt_2: Optional[str] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
): ):
r""" r"""
...@@ -215,17 +215,17 @@ class DemoFusionSDXLPipeline( ...@@ -215,17 +215,17 @@ class DemoFusionSDXLPipeline(
negative_prompt_2 (`str` or `List[str]`, *optional*): negative_prompt_2 (`str` or `List[str]`, *optional*):
The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and
`text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders `text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument. If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument. input argument.
...@@ -642,14 +642,14 @@ class DemoFusionSDXLPipeline( ...@@ -642,14 +642,14 @@ class DemoFusionSDXLPipeline(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = False, return_dict: bool = False,
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None, callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
callback_steps: int = 1, callback_steps: int = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
guidance_rescale: float = 0.0, guidance_rescale: float = 0.0,
...@@ -720,21 +720,21 @@ class DemoFusionSDXLPipeline( ...@@ -720,21 +720,21 @@ class DemoFusionSDXLPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html) One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
to make generation deterministic. to make generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will ge generated by sampling using the supplied random `generator`. tensor will ge generated by sampling using the supplied random `generator`.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument. If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument. input argument.
...@@ -746,7 +746,7 @@ class DemoFusionSDXLPipeline( ...@@ -746,7 +746,7 @@ class DemoFusionSDXLPipeline(
of a plain tuple. of a plain tuple.
callback (`Callable`, *optional*): callback (`Callable`, *optional*):
A function that will be called every `callback_steps` steps during inference. The function will be A function that will be called every `callback_steps` steps during inference. The function will be
called with the following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`. called with the following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
callback_steps (`int`, *optional*, defaults to 1): callback_steps (`int`, *optional*, defaults to 1):
The frequency at which the `callback` function will be called. If not specified, the callback will be The frequency at which the `callback` function will be called. If not specified, the callback will be
called at every step. called at every step.
......
examples/community/pipeline_fabric.py
View file @ be4afa0b
...@@ -190,8 +190,8 @@ class FabricPipeline(DiffusionPipeline): ...@@ -190,8 +190,8 @@ class FabricPipeline(DiffusionPipeline):
num_images_per_prompt, num_images_per_prompt,
do_classifier_free_guidance, do_classifier_free_guidance,
negative_prompt=None, negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
): ):
r""" r"""
...@@ -210,10 +210,10 @@ class FabricPipeline(DiffusionPipeline): ...@@ -210,10 +210,10 @@ class FabricPipeline(DiffusionPipeline):
The prompt or prompts not to guide the image generation. If not defined, one has to pass The prompt or prompts not to guide the image generation. If not defined, one has to pass
`negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
less than `1`). less than `1`).
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
...@@ -512,7 +512,7 @@ class FabricPipeline(DiffusionPipeline): ...@@ -512,7 +512,7 @@ class FabricPipeline(DiffusionPipeline):
neg_scale: float = 0.5, neg_scale: float = 0.5,
pos_bottleneck_scale: float = 1.0, pos_bottleneck_scale: float = 1.0,
neg_bottleneck_scale: float = 1.0, neg_bottleneck_scale: float = 1.0,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
): ):
r""" r"""
The call function to the pipeline for generation. Generate a trajectory of images with binary feedback. The The call function to the pipeline for generation. Generate a trajectory of images with binary feedback. The
......
examples/community/pipeline_prompt2prompt.py
View file @ be4afa0b
...@@ -217,8 +217,8 @@ class Prompt2PromptPipeline( ...@@ -217,8 +217,8 @@ class Prompt2PromptPipeline(
num_images_per_prompt, num_images_per_prompt,
do_classifier_free_guidance, do_classifier_free_guidance,
negative_prompt=None, negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
**kwargs, **kwargs,
): ):
...@@ -250,8 +250,8 @@ class Prompt2PromptPipeline( ...@@ -250,8 +250,8 @@ class Prompt2PromptPipeline(
num_images_per_prompt, num_images_per_prompt,
do_classifier_free_guidance, do_classifier_free_guidance,
negative_prompt=None, negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None, clip_skip: Optional[int] = None,
): ):
...@@ -271,10 +271,10 @@ class Prompt2PromptPipeline( ...@@ -271,10 +271,10 @@ class Prompt2PromptPipeline(
The prompt or prompts not to guide the image generation. If not defined, one has to pass The prompt or prompts not to guide the image generation. If not defined, one has to pass
`negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
less than `1`). less than `1`).
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
...@@ -564,12 +564,12 @@ class Prompt2PromptPipeline( ...@@ -564,12 +564,12 @@ class Prompt2PromptPipeline(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None, callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
callback_steps: Optional[int] = 1, callback_steps: Optional[int] = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
guidance_rescale: float = 0.0, guidance_rescale: float = 0.0,
...@@ -604,7 +604,7 @@ class Prompt2PromptPipeline( ...@@ -604,7 +604,7 @@ class Prompt2PromptPipeline(
generator (`torch.Generator`, *optional*): generator (`torch.Generator`, *optional*):
One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html) One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
to make generation deterministic. to make generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will ge generated by sampling using the supplied random `generator`. tensor will ge generated by sampling using the supplied random `generator`.
...@@ -616,7 +616,7 @@ class Prompt2PromptPipeline( ...@@ -616,7 +616,7 @@ class Prompt2PromptPipeline(
plain tuple. plain tuple.
callback (`Callable`, *optional*): callback (`Callable`, *optional*):
A function that will be called every `callback_steps` steps during inference. The function will be A function that will be called every `callback_steps` steps during inference. The function will be
called with the following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`. called with the following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
callback_steps (`int`, *optional*, defaults to 1): callback_steps (`int`, *optional*, defaults to 1):
The frequency at which the `callback` function will be called. If not specified, the callback will be The frequency at which the `callback` function will be called. If not specified, the callback will be
called at every step. called at every step.
......
examples/community/pipeline_sdxl_style_aligned.py
View file @ be4afa0b
...@@ -514,10 +514,10 @@ class StyleAlignedSDXLPipeline( ...@@ -514,10 +514,10 @@ class StyleAlignedSDXLPipeline(
do_classifier_free_guidance: bool = True, do_classifier_free_guidance: bool = True,
negative_prompt: Optional[str] = None, negative_prompt: Optional[str] = None,
negative_prompt_2: Optional[str] = None, negative_prompt_2: Optional[str] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None, clip_skip: Optional[int] = None,
): ):
...@@ -543,17 +543,17 @@ class StyleAlignedSDXLPipeline( ...@@ -543,17 +543,17 @@ class StyleAlignedSDXLPipeline(
negative_prompt_2 (`str` or `List[str]`, *optional*): negative_prompt_2 (`str` or `List[str]`, *optional*):
The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and
`text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders `text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument. If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument. input argument.
...@@ -1325,7 +1325,7 @@ class StyleAlignedSDXLPipeline( ...@@ -1325,7 +1325,7 @@ class StyleAlignedSDXLPipeline(
data type of the generated embeddings data type of the generated embeddings
Returns: Returns:
`torch.FloatTensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)` `torch.Tensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)`
""" """
assert len(w.shape) == 1 assert len(w.shape) == 1
w = w * 1000.0 w = w * 1000.0
...@@ -1387,7 +1387,7 @@ class StyleAlignedSDXLPipeline( ...@@ -1387,7 +1387,7 @@ class StyleAlignedSDXLPipeline(
prompt_2: Optional[Union[str, List[str]]] = None, prompt_2: Optional[Union[str, List[str]]] = None,
image: Optional[PipelineImageInput] = None, image: Optional[PipelineImageInput] = None,
mask_image: Optional[PipelineImageInput] = None, mask_image: Optional[PipelineImageInput] = None,
masked_image_latents: Optional[torch.FloatTensor] = None, masked_image_latents: Optional[torch.Tensor] = None,
strength: float = 0.3, strength: float = 0.3,
height: Optional[int] = None, height: Optional[int] = None,
width: Optional[int] = None, width: Optional[int] = None,
...@@ -1401,11 +1401,11 @@ class StyleAlignedSDXLPipeline( ...@@ -1401,11 +1401,11 @@ class StyleAlignedSDXLPipeline(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
ip_adapter_image: Optional[PipelineImageInput] = None, ip_adapter_image: Optional[PipelineImageInput] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
...@@ -1474,21 +1474,21 @@ class StyleAlignedSDXLPipeline( ...@@ -1474,21 +1474,21 @@ class StyleAlignedSDXLPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html) One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
to make generation deterministic. to make generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will ge generated by sampling using the supplied random `generator`. tensor will ge generated by sampling using the supplied random `generator`.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument. If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument. input argument.
......
examples/community/pipeline_stable_diffusion_pag.py
View file @ be4afa0b
...@@ -57,13 +57,13 @@ class PAGIdentitySelfAttnProcessor: ...@@ -57,13 +57,13 @@ class PAGIdentitySelfAttnProcessor:
def __call__( def __call__(
self, self,
attn: Attention, attn: Attention,
hidden_states: torch.FloatTensor, hidden_states: torch.Tensor,
encoder_hidden_states: Optional[torch.FloatTensor] = None, encoder_hidden_states: Optional[torch.Tensor] = None,
attention_mask: Optional[torch.FloatTensor] = None, attention_mask: Optional[torch.Tensor] = None,
temb: Optional[torch.FloatTensor] = None, temb: Optional[torch.Tensor] = None,
*args, *args,
**kwargs, **kwargs,
) -> torch.FloatTensor: ) -> torch.Tensor:
if len(args) > 0 or kwargs.get("scale", None) is not None: if len(args) > 0 or kwargs.get("scale", None) is not None:
deprecation_message = "The `scale` argument is deprecated and will be ignored. Please remove it, as passing it will raise an error in the future. `scale` should directly be passed while calling the underlying pipeline component i.e., via `cross_attention_kwargs`." deprecation_message = "The `scale` argument is deprecated and will be ignored. Please remove it, as passing it will raise an error in the future. `scale` should directly be passed while calling the underlying pipeline component i.e., via `cross_attention_kwargs`."
deprecate("scale", "1.0.0", deprecation_message) deprecate("scale", "1.0.0", deprecation_message)
...@@ -171,13 +171,13 @@ class PAGCFGIdentitySelfAttnProcessor: ...@@ -171,13 +171,13 @@ class PAGCFGIdentitySelfAttnProcessor:
def __call__( def __call__(
self, self,
attn: Attention, attn: Attention,
hidden_states: torch.FloatTensor, hidden_states: torch.Tensor,
encoder_hidden_states: Optional[torch.FloatTensor] = None, encoder_hidden_states: Optional[torch.Tensor] = None,
attention_mask: Optional[torch.FloatTensor] = None, attention_mask: Optional[torch.Tensor] = None,
temb: Optional[torch.FloatTensor] = None, temb: Optional[torch.Tensor] = None,
*args, *args,
**kwargs, **kwargs,
) -> torch.FloatTensor: ) -> torch.Tensor:
if len(args) > 0 or kwargs.get("scale", None) is not None: if len(args) > 0 or kwargs.get("scale", None) is not None:
deprecation_message = "The `scale` argument is deprecated and will be ignored. Please remove it, as passing it will raise an error in the future. `scale` should directly be passed while calling the underlying pipeline component i.e., via `cross_attention_kwargs`." deprecation_message = "The `scale` argument is deprecated and will be ignored. Please remove it, as passing it will raise an error in the future. `scale` should directly be passed while calling the underlying pipeline component i.e., via `cross_attention_kwargs`."
deprecate("scale", "1.0.0", deprecation_message) deprecate("scale", "1.0.0", deprecation_message)
...@@ -493,8 +493,8 @@ class StableDiffusionPAGPipeline( ...@@ -493,8 +493,8 @@ class StableDiffusionPAGPipeline(
num_images_per_prompt, num_images_per_prompt,
do_classifier_free_guidance, do_classifier_free_guidance,
negative_prompt=None, negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
**kwargs, **kwargs,
): ):
...@@ -525,8 +525,8 @@ class StableDiffusionPAGPipeline( ...@@ -525,8 +525,8 @@ class StableDiffusionPAGPipeline(
num_images_per_prompt, num_images_per_prompt,
do_classifier_free_guidance, do_classifier_free_guidance,
negative_prompt=None, negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None, clip_skip: Optional[int] = None,
): ):
...@@ -545,10 +545,10 @@ class StableDiffusionPAGPipeline( ...@@ -545,10 +545,10 @@ class StableDiffusionPAGPipeline(
The prompt or prompts not to guide the image generation. If not defined, one has to pass The prompt or prompts not to guide the image generation. If not defined, one has to pass
`negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
less than `1`). less than `1`).
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
...@@ -966,7 +966,7 @@ class StableDiffusionPAGPipeline( ...@@ -966,7 +966,7 @@ class StableDiffusionPAGPipeline(
dtype: dtype:
data type of the generated embeddings data type of the generated embeddings
Returns: Returns:
`torch.FloatTensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)` `torch.Tensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)`
""" """
assert len(w.shape) == 1 assert len(w.shape) == 1
w = w * 1000.0 w = w * 1000.0
...@@ -1078,11 +1078,11 @@ class StableDiffusionPAGPipeline( ...@@ -1078,11 +1078,11 @@ class StableDiffusionPAGPipeline(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
ip_adapter_image: Optional[PipelineImageInput] = None, ip_adapter_image: Optional[PipelineImageInput] = None,
ip_adapter_image_embeds: Optional[List[torch.FloatTensor]] = None, ip_adapter_image_embeds: Optional[List[torch.Tensor]] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
...@@ -1122,18 +1122,18 @@ class StableDiffusionPAGPipeline( ...@@ -1122,18 +1122,18 @@ class StableDiffusionPAGPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
generation deterministic. generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor is generated by sampling using the supplied random `generator`. tensor is generated by sampling using the supplied random `generator`.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
provided, text embeddings are generated from the `prompt` input argument. provided, text embeddings are generated from the `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument. not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
ip_adapter_image: (`PipelineImageInput`, *optional*): Optional image input to work with IP Adapters. ip_adapter_image: (`PipelineImageInput`, *optional*): Optional image input to work with IP Adapters.
ip_adapter_image_embeds (`List[torch.FloatTensor]`, *optional*): ip_adapter_image_embeds (`List[torch.Tensor]`, *optional*):
Pre-generated image embeddings for IP-Adapter. If not Pre-generated image embeddings for IP-Adapter. If not
provided, embeddings are computed from the `ip_adapter_image` input argument. provided, embeddings are computed from the `ip_adapter_image` input argument.
output_type (`str`, *optional*, defaults to `"pil"`): output_type (`str`, *optional*, defaults to `"pil"`):
......
examples/community/pipeline_stable_diffusion_upscale_ldm3d.py
View file @ be4afa0b
...@@ -164,8 +164,8 @@ class StableDiffusionUpscaleLDM3DPipeline( ...@@ -164,8 +164,8 @@ class StableDiffusionUpscaleLDM3DPipeline(
num_images_per_prompt, num_images_per_prompt,
do_classifier_free_guidance, do_classifier_free_guidance,
negative_prompt=None, negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
**kwargs, **kwargs,
): ):
...@@ -197,8 +197,8 @@ class StableDiffusionUpscaleLDM3DPipeline( ...@@ -197,8 +197,8 @@ class StableDiffusionUpscaleLDM3DPipeline(
num_images_per_prompt, num_images_per_prompt,
do_classifier_free_guidance, do_classifier_free_guidance,
negative_prompt=None, negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None, clip_skip: Optional[int] = None,
): ):
...@@ -218,10 +218,10 @@ class StableDiffusionUpscaleLDM3DPipeline( ...@@ -218,10 +218,10 @@ class StableDiffusionUpscaleLDM3DPipeline(
The prompt or prompts not to guide the image generation. If not defined, one has to pass The prompt or prompts not to guide the image generation. If not defined, one has to pass
`negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
less than `1`). less than `1`).
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
...@@ -535,12 +535,12 @@ class StableDiffusionUpscaleLDM3DPipeline( ...@@ -535,12 +535,12 @@ class StableDiffusionUpscaleLDM3DPipeline(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None, callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
callback_steps: int = 1, callback_steps: int = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
target_res: Optional[List[int]] = [1024, 1024], target_res: Optional[List[int]] = [1024, 1024],
...@@ -551,7 +551,7 @@ class StableDiffusionUpscaleLDM3DPipeline( ...@@ -551,7 +551,7 @@ class StableDiffusionUpscaleLDM3DPipeline(
Args: Args:
prompt (`str` or `List[str]`, *optional*): prompt (`str` or `List[str]`, *optional*):
The prompt or prompts to guide image generation. If not defined, you need to pass `prompt_embeds`. The prompt or prompts to guide image generation. If not defined, you need to pass `prompt_embeds`.
image (`torch.FloatTensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.FloatTensor]`, `List[PIL.Image.Image]`, or `List[np.ndarray]`): image (`torch.Tensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.Tensor]`, `List[PIL.Image.Image]`, or `List[np.ndarray]`):
`Image` or tensor representing an image batch to be upscaled. `Image` or tensor representing an image batch to be upscaled.
num_inference_steps (`int`, *optional*, defaults to 50): num_inference_steps (`int`, *optional*, defaults to 50):
The number of denoising steps. More denoising steps usually lead to a higher quality image at the The number of denoising steps. More denoising steps usually lead to a higher quality image at the
...@@ -570,14 +570,14 @@ class StableDiffusionUpscaleLDM3DPipeline( ...@@ -570,14 +570,14 @@ class StableDiffusionUpscaleLDM3DPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
generation deterministic. generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor is generated by sampling using the supplied random `generator`. tensor is generated by sampling using the supplied random `generator`.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
provided, text embeddings are generated from the `prompt` input argument. provided, text embeddings are generated from the `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument. not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
output_type (`str`, *optional*, defaults to `"pil"`): output_type (`str`, *optional*, defaults to `"pil"`):
...@@ -587,7 +587,7 @@ class StableDiffusionUpscaleLDM3DPipeline( ...@@ -587,7 +587,7 @@ class StableDiffusionUpscaleLDM3DPipeline(
plain tuple. plain tuple.
callback (`Callable`, *optional*): callback (`Callable`, *optional*):
A function that calls every `callback_steps` steps during inference. The function is called with the A function that calls every `callback_steps` steps during inference. The function is called with the
following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`. following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
callback_steps (`int`, *optional*, defaults to 1): callback_steps (`int`, *optional*, defaults to 1):
The frequency at which the `callback` function is called. If not specified, the callback is called at The frequency at which the `callback` function is called. If not specified, the callback is called at
every step. every step.
......
examples/community/pipeline_stable_diffusion_xl_controlnet_adapter.py
View file @ be4afa0b
...@@ -248,10 +248,10 @@ class StableDiffusionXLControlNetAdapterPipeline( ...@@ -248,10 +248,10 @@ class StableDiffusionXLControlNetAdapterPipeline(
do_classifier_free_guidance: bool = True, do_classifier_free_guidance: bool = True,
negative_prompt: Optional[str] = None, negative_prompt: Optional[str] = None,
negative_prompt_2: Optional[str] = None, negative_prompt_2: Optional[str] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None, clip_skip: Optional[int] = None,
): ):
...@@ -277,17 +277,17 @@ class StableDiffusionXLControlNetAdapterPipeline( ...@@ -277,17 +277,17 @@ class StableDiffusionXLControlNetAdapterPipeline(
negative_prompt_2 (`str` or `List[str]`, *optional*): negative_prompt_2 (`str` or `List[str]`, *optional*):
The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and
`text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders `text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument. If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument. input argument.
...@@ -887,14 +887,14 @@ class StableDiffusionXLControlNetAdapterPipeline( ...@@ -887,14 +887,14 @@ class StableDiffusionXLControlNetAdapterPipeline(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None, callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
callback_steps: int = 1, callback_steps: int = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
guidance_rescale: float = 0.0, guidance_rescale: float = 0.0,
...@@ -922,14 +922,14 @@ class StableDiffusionXLControlNetAdapterPipeline( ...@@ -922,14 +922,14 @@ class StableDiffusionXLControlNetAdapterPipeline(
prompt_2 (`str` or `List[str]`, *optional*): prompt_2 (`str` or `List[str]`, *optional*):
The prompt or prompts to be sent to the `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is The prompt or prompts to be sent to the `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is
used in both text-encoders used in both text-encoders
adapter_image (`torch.FloatTensor`, `PIL.Image.Image`, `List[torch.FloatTensor]` or `List[PIL.Image.Image]` or `List[List[PIL.Image.Image]]`): adapter_image (`torch.Tensor`, `PIL.Image.Image`, `List[torch.Tensor]` or `List[PIL.Image.Image]` or `List[List[PIL.Image.Image]]`):
The Adapter input condition. Adapter uses this input condition to generate guidance to Unet. If the The Adapter input condition. Adapter uses this input condition to generate guidance to Unet. If the
type is specified as `Torch.FloatTensor`, it is passed to Adapter as is. PIL.Image.Image` can also be type is specified as `torch.Tensor`, it is passed to Adapter as is. PIL.Image.Image` can also be
accepted as an image. The control image is automatically resized to fit the output image. accepted as an image. The control image is automatically resized to fit the output image.
control_image (`torch.FloatTensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.FloatTensor]`, `List[PIL.Image.Image]`, `List[np.ndarray]`,: control_image (`torch.Tensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.Tensor]`, `List[PIL.Image.Image]`, `List[np.ndarray]`,:
`List[List[torch.FloatTensor]]`, `List[List[np.ndarray]]` or `List[List[PIL.Image.Image]]`): `List[List[torch.Tensor]]`, `List[List[np.ndarray]]` or `List[List[PIL.Image.Image]]`):
The ControlNet input condition to provide guidance to the `unet` for generation. If the type is The ControlNet input condition to provide guidance to the `unet` for generation. If the type is
specified as `torch.FloatTensor`, it is passed to ControlNet as is. `PIL.Image.Image` can also be specified as `torch.Tensor`, it is passed to ControlNet as is. `PIL.Image.Image` can also be
accepted as an image. The dimensions of the output image defaults to `image`'s dimensions. If height accepted as an image. The dimensions of the output image defaults to `image`'s dimensions. If height
and/or width are passed, `image` is resized accordingly. If multiple ControlNets are specified in and/or width are passed, `image` is resized accordingly. If multiple ControlNets are specified in
`init`, images must be passed as a list such that each element of the list can be correctly batched for `init`, images must be passed as a list such that each element of the list can be correctly batched for
...@@ -973,21 +973,21 @@ class StableDiffusionXLControlNetAdapterPipeline( ...@@ -973,21 +973,21 @@ class StableDiffusionXLControlNetAdapterPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html) One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
to make generation deterministic. to make generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will ge generated by sampling using the supplied random `generator`. tensor will ge generated by sampling using the supplied random `generator`.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument. If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument. input argument.
...@@ -999,7 +999,7 @@ class StableDiffusionXLControlNetAdapterPipeline( ...@@ -999,7 +999,7 @@ class StableDiffusionXLControlNetAdapterPipeline(
instead of a plain tuple. instead of a plain tuple.
callback (`Callable`, *optional*): callback (`Callable`, *optional*):
A function that will be called every `callback_steps` steps during inference. The function will be A function that will be called every `callback_steps` steps during inference. The function will be
called with the following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`. called with the following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
callback_steps (`int`, *optional*, defaults to 1): callback_steps (`int`, *optional*, defaults to 1):
The frequency at which the `callback` function will be called. If not specified, the callback will be The frequency at which the `callback` function will be called. If not specified, the callback will be
called at every step. called at every step.
......
examples/community/pipeline_stable_diffusion_xl_controlnet_adapter_inpaint.py
View file @ be4afa0b
...@@ -396,10 +396,10 @@ class StableDiffusionXLControlNetAdapterInpaintPipeline( ...@@ -396,10 +396,10 @@ class StableDiffusionXLControlNetAdapterInpaintPipeline(
do_classifier_free_guidance: bool = True, do_classifier_free_guidance: bool = True,
negative_prompt: Optional[str] = None, negative_prompt: Optional[str] = None,
negative_prompt_2: Optional[str] = None, negative_prompt_2: Optional[str] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None, clip_skip: Optional[int] = None,
): ):
...@@ -425,17 +425,17 @@ class StableDiffusionXLControlNetAdapterInpaintPipeline( ...@@ -425,17 +425,17 @@ class StableDiffusionXLControlNetAdapterInpaintPipeline(
negative_prompt_2 (`str` or `List[str]`, *optional*): negative_prompt_2 (`str` or `List[str]`, *optional*):
The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and
`text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders `text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument. If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument. input argument.
...@@ -1229,14 +1229,14 @@ class StableDiffusionXLControlNetAdapterInpaintPipeline( ...@@ -1229,14 +1229,14 @@ class StableDiffusionXLControlNetAdapterInpaintPipeline(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[Union[torch.FloatTensor]] = None, latents: Optional[Union[torch.Tensor]] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None, callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
callback_steps: int = 1, callback_steps: int = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
guidance_rescale: float = 0.0, guidance_rescale: float = 0.0,
...@@ -1270,14 +1270,14 @@ class StableDiffusionXLControlNetAdapterInpaintPipeline( ...@@ -1270,14 +1270,14 @@ class StableDiffusionXLControlNetAdapterInpaintPipeline(
repainted, while black pixels will be preserved. If `mask_image` is a PIL image, it will be converted repainted, while black pixels will be preserved. If `mask_image` is a PIL image, it will be converted
to a single channel (luminance) before use. If it's a tensor, it should contain one color channel (L) to a single channel (luminance) before use. If it's a tensor, it should contain one color channel (L)
instead of 3, so the expected shape would be `(B, H, W, 1)`. instead of 3, so the expected shape would be `(B, H, W, 1)`.
adapter_image (`torch.FloatTensor`, `PIL.Image.Image`, `List[torch.FloatTensor]` or `List[PIL.Image.Image]` or `List[List[PIL.Image.Image]]`): adapter_image (`torch.Tensor`, `PIL.Image.Image`, `List[torch.Tensor]` or `List[PIL.Image.Image]` or `List[List[PIL.Image.Image]]`):
The Adapter input condition. Adapter uses this input condition to generate guidance to Unet. If the The Adapter input condition. Adapter uses this input condition to generate guidance to Unet. If the
type is specified as `Torch.FloatTensor`, it is passed to Adapter as is. PIL.Image.Image` can also be type is specified as `torch.Tensor`, it is passed to Adapter as is. PIL.Image.Image` can also be
accepted as an image. The control image is automatically resized to fit the output image. accepted as an image. The control image is automatically resized to fit the output image.
control_image (`torch.FloatTensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.FloatTensor]`, `List[PIL.Image.Image]`, `List[np.ndarray]`,: control_image (`torch.Tensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.Tensor]`, `List[PIL.Image.Image]`, `List[np.ndarray]`,:
`List[List[torch.FloatTensor]]`, `List[List[np.ndarray]]` or `List[List[PIL.Image.Image]]`): `List[List[torch.Tensor]]`, `List[List[np.ndarray]]` or `List[List[PIL.Image.Image]]`):
The ControlNet input condition to provide guidance to the `unet` for generation. If the type is The ControlNet input condition to provide guidance to the `unet` for generation. If the type is
specified as `torch.FloatTensor`, it is passed to ControlNet as is. `PIL.Image.Image` can also be specified as `torch.Tensor`, it is passed to ControlNet as is. `PIL.Image.Image` can also be
accepted as an image. The dimensions of the output image defaults to `image`'s dimensions. If height accepted as an image. The dimensions of the output image defaults to `image`'s dimensions. If height
and/or width are passed, `image` is resized accordingly. If multiple ControlNets are specified in and/or width are passed, `image` is resized accordingly. If multiple ControlNets are specified in
`init`, images must be passed as a list such that each element of the list can be correctly batched for `init`, images must be passed as a list such that each element of the list can be correctly batched for
...@@ -1330,21 +1330,21 @@ class StableDiffusionXLControlNetAdapterInpaintPipeline( ...@@ -1330,21 +1330,21 @@ class StableDiffusionXLControlNetAdapterInpaintPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html) One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
to make generation deterministic. to make generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will ge generated by sampling using the supplied random `generator`. tensor will ge generated by sampling using the supplied random `generator`.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument. If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument. input argument.
...@@ -1356,7 +1356,7 @@ class StableDiffusionXLControlNetAdapterInpaintPipeline( ...@@ -1356,7 +1356,7 @@ class StableDiffusionXLControlNetAdapterInpaintPipeline(
instead of a plain tuple. instead of a plain tuple.
callback (`Callable`, *optional*): callback (`Callable`, *optional*):
A function that will be called every `callback_steps` steps during inference. The function will be A function that will be called every `callback_steps` steps during inference. The function will be
called with the following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`. called with the following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
callback_steps (`int`, *optional*, defaults to 1): callback_steps (`int`, *optional*, defaults to 1):
The frequency at which the `callback` function will be called. If not specified, the callback will be The frequency at which the `callback` function will be called. If not specified, the callback will be
called at every step. called at every step.
......
examples/community/pipeline_stable_diffusion_xl_differential_img2img.py
View file @ be4afa0b
...@@ -280,10 +280,10 @@ class StableDiffusionXLDifferentialImg2ImgPipeline( ...@@ -280,10 +280,10 @@ class StableDiffusionXLDifferentialImg2ImgPipeline(
do_classifier_free_guidance: bool = True, do_classifier_free_guidance: bool = True,
negative_prompt: Optional[str] = None, negative_prompt: Optional[str] = None,
negative_prompt_2: Optional[str] = None, negative_prompt_2: Optional[str] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None, clip_skip: Optional[int] = None,
): ):
...@@ -309,17 +309,17 @@ class StableDiffusionXLDifferentialImg2ImgPipeline( ...@@ -309,17 +309,17 @@ class StableDiffusionXLDifferentialImg2ImgPipeline(
negative_prompt_2 (`str` or `List[str]`, *optional*): negative_prompt_2 (`str` or `List[str]`, *optional*):
The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and
`text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders `text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument. If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument. input argument.
...@@ -868,7 +868,7 @@ class StableDiffusionXLDifferentialImg2ImgPipeline( ...@@ -868,7 +868,7 @@ class StableDiffusionXLDifferentialImg2ImgPipeline(
# Copied from diffusers.pipelines.latent_consistency_models.pipeline_latent_consistency_text2img.LatentConsistencyModelPipeline.get_guidance_scale_embedding # Copied from diffusers.pipelines.latent_consistency_models.pipeline_latent_consistency_text2img.LatentConsistencyModelPipeline.get_guidance_scale_embedding
def get_guidance_scale_embedding( def get_guidance_scale_embedding(
self, w: torch.Tensor, embedding_dim: int = 512, dtype: torch.dtype = torch.float32 self, w: torch.Tensor, embedding_dim: int = 512, dtype: torch.dtype = torch.float32
) -> torch.FloatTensor: ) -> torch.Tensor:
""" """
See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298 See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298
...@@ -881,7 +881,7 @@ class StableDiffusionXLDifferentialImg2ImgPipeline( ...@@ -881,7 +881,7 @@ class StableDiffusionXLDifferentialImg2ImgPipeline(
Data type of the generated embeddings. Data type of the generated embeddings.
Returns: Returns:
`torch.FloatTensor`: Embedding vectors with shape `(len(w), embedding_dim)`. `torch.Tensor`: Embedding vectors with shape `(len(w), embedding_dim)`.
""" """
assert len(w.shape) == 1 assert len(w.shape) == 1
w = w * 1000.0 w = w * 1000.0
...@@ -942,10 +942,10 @@ class StableDiffusionXLDifferentialImg2ImgPipeline( ...@@ -942,10 +942,10 @@ class StableDiffusionXLDifferentialImg2ImgPipeline(
prompt: Union[str, List[str]] = None, prompt: Union[str, List[str]] = None,
prompt_2: Optional[Union[str, List[str]]] = None, prompt_2: Optional[Union[str, List[str]]] = None,
image: Union[ image: Union[
torch.FloatTensor, torch.Tensor,
PIL.Image.Image, PIL.Image.Image,
np.ndarray, np.ndarray,
List[torch.FloatTensor], List[torch.Tensor],
List[PIL.Image.Image], List[PIL.Image.Image],
List[np.ndarray], List[np.ndarray],
] = None, ] = None,
...@@ -960,13 +960,13 @@ class StableDiffusionXLDifferentialImg2ImgPipeline( ...@@ -960,13 +960,13 @@ class StableDiffusionXLDifferentialImg2ImgPipeline(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
ip_adapter_image: Optional[PipelineImageInput] = None, ip_adapter_image: Optional[PipelineImageInput] = None,
ip_adapter_image_embeds: Optional[List[torch.FloatTensor]] = None, ip_adapter_image_embeds: Optional[List[torch.Tensor]] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
...@@ -982,12 +982,12 @@ class StableDiffusionXLDifferentialImg2ImgPipeline( ...@@ -982,12 +982,12 @@ class StableDiffusionXLDifferentialImg2ImgPipeline(
clip_skip: Optional[int] = None, clip_skip: Optional[int] = None,
callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None, callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
callback_on_step_end_tensor_inputs: List[str] = ["latents"], callback_on_step_end_tensor_inputs: List[str] = ["latents"],
map: torch.FloatTensor = None, map: torch.Tensor = None,
original_image: Union[ original_image: Union[
torch.FloatTensor, torch.Tensor,
PIL.Image.Image, PIL.Image.Image,
np.ndarray, np.ndarray,
List[torch.FloatTensor], List[torch.Tensor],
List[PIL.Image.Image], List[PIL.Image.Image],
List[np.ndarray], List[np.ndarray],
] = None, ] = None,
...@@ -1003,7 +1003,7 @@ class StableDiffusionXLDifferentialImg2ImgPipeline( ...@@ -1003,7 +1003,7 @@ class StableDiffusionXLDifferentialImg2ImgPipeline(
prompt_2 (`str` or `List[str]`, *optional*): prompt_2 (`str` or `List[str]`, *optional*):
The prompt or prompts to be sent to the `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is The prompt or prompts to be sent to the `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is
used in both text-encoders used in both text-encoders
image (`torch.FloatTensor` or `PIL.Image.Image` or `np.ndarray` or `List[torch.FloatTensor]` or `List[PIL.Image.Image]` or `List[np.ndarray]`): image (`torch.Tensor` or `PIL.Image.Image` or `np.ndarray` or `List[torch.Tensor]` or `List[PIL.Image.Image]` or `List[np.ndarray]`):
The image(s) to modify with the pipeline. The image(s) to modify with the pipeline.
strength (`float`, *optional*, defaults to 0.3): strength (`float`, *optional*, defaults to 0.3):
Conceptually, indicates how much to transform the reference `image`. Must be between 0 and 1. `image` Conceptually, indicates how much to transform the reference `image`. Must be between 0 and 1. `image`
...@@ -1051,26 +1051,26 @@ class StableDiffusionXLDifferentialImg2ImgPipeline( ...@@ -1051,26 +1051,26 @@ class StableDiffusionXLDifferentialImg2ImgPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html) One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
to make generation deterministic. to make generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will ge generated by sampling using the supplied random `generator`. tensor will ge generated by sampling using the supplied random `generator`.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument. If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument. input argument.
ip_adapter_image: (`PipelineImageInput`, *optional*): Optional image input to work with IP Adapters. ip_adapter_image: (`PipelineImageInput`, *optional*): Optional image input to work with IP Adapters.
ip_adapter_image_embeds (`List[torch.FloatTensor]`, *optional*): ip_adapter_image_embeds (`List[torch.Tensor]`, *optional*):
Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of IP-adapters. Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of IP-adapters.
Each element should be a tensor of shape `(batch_size, num_images, emb_dim)`. It should contain the negative image embedding Each element should be a tensor of shape `(batch_size, num_images, emb_dim)`. It should contain the negative image embedding
if `do_classifier_free_guidance` is set to `True`. if `do_classifier_free_guidance` is set to `True`.
...@@ -1083,7 +1083,7 @@ class StableDiffusionXLDifferentialImg2ImgPipeline( ...@@ -1083,7 +1083,7 @@ class StableDiffusionXLDifferentialImg2ImgPipeline(
plain tuple. plain tuple.
callback (`Callable`, *optional*): callback (`Callable`, *optional*):
A function that will be called every `callback_steps` steps during inference. The function will be A function that will be called every `callback_steps` steps during inference. The function will be
called with the following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`. called with the following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
callback_steps (`int`, *optional*, defaults to 1): callback_steps (`int`, *optional*, defaults to 1):
The frequency at which the `callback` function will be called. If not specified, the callback will be The frequency at which the `callback` function will be called. If not specified, the callback will be
called at every step. called at every step.
......
examples/community/pipeline_stable_diffusion_xl_instandid_img2img.py
View file @ be4afa0b
...@@ -561,12 +561,12 @@ class StableDiffusionXLInstantIDImg2ImgPipeline(StableDiffusionXLControlNetImg2I ...@@ -561,12 +561,12 @@ class StableDiffusionXLInstantIDImg2ImgPipeline(StableDiffusionXLControlNetImg2I
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
image_embeds: Optional[torch.FloatTensor] = None, image_embeds: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
...@@ -596,10 +596,10 @@ class StableDiffusionXLInstantIDImg2ImgPipeline(StableDiffusionXLControlNetImg2I ...@@ -596,10 +596,10 @@ class StableDiffusionXLInstantIDImg2ImgPipeline(StableDiffusionXLControlNetImg2I
prompt_2 (`str` or `List[str]`, *optional*): prompt_2 (`str` or `List[str]`, *optional*):
The prompt or prompts to be sent to `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is The prompt or prompts to be sent to `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is
used in both text-encoders. used in both text-encoders.
image (`torch.FloatTensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.FloatTensor]`, `List[PIL.Image.Image]`, `List[np.ndarray]`,: image (`torch.Tensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.Tensor]`, `List[PIL.Image.Image]`, `List[np.ndarray]`,:
`List[List[torch.FloatTensor]]`, `List[List[np.ndarray]]` or `List[List[PIL.Image.Image]]`): `List[List[torch.Tensor]]`, `List[List[np.ndarray]]` or `List[List[PIL.Image.Image]]`):
The ControlNet input condition to provide guidance to the `unet` for generation. If the type is The ControlNet input condition to provide guidance to the `unet` for generation. If the type is
specified as `torch.FloatTensor`, it is passed to ControlNet as is. `PIL.Image.Image` can also be specified as `torch.Tensor`, it is passed to ControlNet as is. `PIL.Image.Image` can also be
accepted as an image. The dimensions of the output image defaults to `image`'s dimensions. If height accepted as an image. The dimensions of the output image defaults to `image`'s dimensions. If height
and/or width are passed, `image` is resized accordingly. If multiple ControlNets are specified in and/or width are passed, `image` is resized accordingly. If multiple ControlNets are specified in
`init`, images must be passed as a list such that each element of the list can be correctly batched for `init`, images must be passed as a list such that each element of the list can be correctly batched for
...@@ -632,24 +632,24 @@ class StableDiffusionXLInstantIDImg2ImgPipeline(StableDiffusionXLControlNetImg2I ...@@ -632,24 +632,24 @@ class StableDiffusionXLInstantIDImg2ImgPipeline(StableDiffusionXLControlNetImg2I
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
generation deterministic. generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor is generated by sampling using the supplied random `generator`. tensor is generated by sampling using the supplied random `generator`.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
provided, text embeddings are generated from the `prompt` input argument. provided, text embeddings are generated from the `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument. not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs (prompt weighting). If Pre-generated pooled text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
not provided, pooled text embeddings are generated from `prompt` input argument. not provided, pooled text embeddings are generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs (prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs (prompt
weighting). If not provided, pooled `negative_prompt_embeds` are generated from `negative_prompt` input weighting). If not provided, pooled `negative_prompt_embeds` are generated from `negative_prompt` input
argument. argument.
image_embeds (`torch.FloatTensor`, *optional*): image_embeds (`torch.Tensor`, *optional*):
Pre-generated image embeddings. Pre-generated image embeddings.
output_type (`str`, *optional*, defaults to `"pil"`): output_type (`str`, *optional*, defaults to `"pil"`):
The output format of the generated image. Choose between `PIL.Image` or `np.array`. The output format of the generated image. Choose between `PIL.Image` or `np.array`.
......
examples/community/pipeline_stable_diffusion_xl_instantid.py
View file @ be4afa0b
...@@ -559,12 +559,12 @@ class StableDiffusionXLInstantIDPipeline(StableDiffusionXLControlNetPipeline): ...@@ -559,12 +559,12 @@ class StableDiffusionXLInstantIDPipeline(StableDiffusionXLControlNetPipeline):
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
image_embeds: Optional[torch.FloatTensor] = None, image_embeds: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
cross_attention_kwargs: Optional[Dict[str, Any]] = None, cross_attention_kwargs: Optional[Dict[str, Any]] = None,
...@@ -592,10 +592,10 @@ class StableDiffusionXLInstantIDPipeline(StableDiffusionXLControlNetPipeline): ...@@ -592,10 +592,10 @@ class StableDiffusionXLInstantIDPipeline(StableDiffusionXLControlNetPipeline):
prompt_2 (`str` or `List[str]`, *optional*): prompt_2 (`str` or `List[str]`, *optional*):
The prompt or prompts to be sent to `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is The prompt or prompts to be sent to `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is
used in both text-encoders. used in both text-encoders.
image (`torch.FloatTensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.FloatTensor]`, `List[PIL.Image.Image]`, `List[np.ndarray]`,: image (`torch.Tensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.Tensor]`, `List[PIL.Image.Image]`, `List[np.ndarray]`,:
`List[List[torch.FloatTensor]]`, `List[List[np.ndarray]]` or `List[List[PIL.Image.Image]]`): `List[List[torch.Tensor]]`, `List[List[np.ndarray]]` or `List[List[PIL.Image.Image]]`):
The ControlNet input condition to provide guidance to the `unet` for generation. If the type is The ControlNet input condition to provide guidance to the `unet` for generation. If the type is
specified as `torch.FloatTensor`, it is passed to ControlNet as is. `PIL.Image.Image` can also be specified as `torch.Tensor`, it is passed to ControlNet as is. `PIL.Image.Image` can also be
accepted as an image. The dimensions of the output image defaults to `image`'s dimensions. If height accepted as an image. The dimensions of the output image defaults to `image`'s dimensions. If height
and/or width are passed, `image` is resized accordingly. If multiple ControlNets are specified in and/or width are passed, `image` is resized accordingly. If multiple ControlNets are specified in
`init`, images must be passed as a list such that each element of the list can be correctly batched for `init`, images must be passed as a list such that each element of the list can be correctly batched for
...@@ -628,24 +628,24 @@ class StableDiffusionXLInstantIDPipeline(StableDiffusionXLControlNetPipeline): ...@@ -628,24 +628,24 @@ class StableDiffusionXLInstantIDPipeline(StableDiffusionXLControlNetPipeline):
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
generation deterministic. generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor is generated by sampling using the supplied random `generator`. tensor is generated by sampling using the supplied random `generator`.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
provided, text embeddings are generated from the `prompt` input argument. provided, text embeddings are generated from the `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument. not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs (prompt weighting). If Pre-generated pooled text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
not provided, pooled text embeddings are generated from `prompt` input argument. not provided, pooled text embeddings are generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs (prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs (prompt
weighting). If not provided, pooled `negative_prompt_embeds` are generated from `negative_prompt` input weighting). If not provided, pooled `negative_prompt_embeds` are generated from `negative_prompt` input
argument. argument.
image_embeds (`torch.FloatTensor`, *optional*): image_embeds (`torch.Tensor`, *optional*):
Pre-generated image embeddings. Pre-generated image embeddings.
output_type (`str`, *optional*, defaults to `"pil"`): output_type (`str`, *optional*, defaults to `"pil"`):
The output format of the generated image. Choose between `PIL.Image` or `np.array`. The output format of the generated image. Choose between `PIL.Image` or `np.array`.
......
examples/community/pipeline_stable_diffusion_xl_ipex.py
View file @ be4afa0b
...@@ -276,10 +276,10 @@ class StableDiffusionXLPipelineIpex( ...@@ -276,10 +276,10 @@ class StableDiffusionXLPipelineIpex(
do_classifier_free_guidance: bool = True, do_classifier_free_guidance: bool = True,
negative_prompt: Optional[str] = None, negative_prompt: Optional[str] = None,
negative_prompt_2: Optional[str] = None, negative_prompt_2: Optional[str] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None, lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None, clip_skip: Optional[int] = None,
): ):
...@@ -305,17 +305,17 @@ class StableDiffusionXLPipelineIpex( ...@@ -305,17 +305,17 @@ class StableDiffusionXLPipelineIpex(
negative_prompt_2 (`str` or `List[str]`, *optional*): negative_prompt_2 (`str` or `List[str]`, *optional*):
The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and
`text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders `text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument. If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument. input argument.
...@@ -687,7 +687,7 @@ class StableDiffusionXLPipelineIpex( ...@@ -687,7 +687,7 @@ class StableDiffusionXLPipelineIpex(
data type of the generated embeddings data type of the generated embeddings
Returns: Returns:
`torch.FloatTensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)` `torch.Tensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)`
""" """
assert len(w.shape) == 1 assert len(w.shape) == 1
w = w * 1000.0 w = w * 1000.0
...@@ -750,11 +750,11 @@ class StableDiffusionXLPipelineIpex( ...@@ -750,11 +750,11 @@ class StableDiffusionXLPipelineIpex(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
ip_adapter_image: Optional[PipelineImageInput] = None, ip_adapter_image: Optional[PipelineImageInput] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
...@@ -826,21 +826,21 @@ class StableDiffusionXLPipelineIpex( ...@@ -826,21 +826,21 @@ class StableDiffusionXLPipelineIpex(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html) One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
to make generation deterministic. to make generation deterministic.
latents (`torch.FloatTensor`, *optional*): latents (`torch.Tensor`, *optional*):
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will ge generated by sampling using the supplied random `generator`. tensor will ge generated by sampling using the supplied random `generator`.
prompt_embeds (`torch.FloatTensor`, *optional*): prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument. provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*): negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument. argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*): pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument. If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument. input argument.
...@@ -1190,11 +1190,11 @@ class StableDiffusionXLPipelineIpex( ...@@ -1190,11 +1190,11 @@ class StableDiffusionXLPipelineIpex(
num_images_per_prompt: Optional[int] = 1, num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0, eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None, latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None, pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
ip_adapter_image: Optional[PipelineImageInput] = None, ip_adapter_image: Optional[PipelineImageInput] = None,
output_type: Optional[str] = "pil", output_type: Optional[str] = "pil",
return_dict: bool = True, return_dict: bool = True,
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment