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 233 additions and 234 deletions
src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2.py
View file @ be4afa0b
......@@ -123,15 +123,15 @@ class KandinskyV22Pipeline(DiffusionPipeline):
@replace_example_docstring(EXAMPLE_DOC_STRING)
def __call__(
self,
image_embeds: Union[torch.FloatTensor, List[torch.FloatTensor]],
negative_image_embeds: Union[torch.FloatTensor, List[torch.FloatTensor]],
image_embeds: Union[torch.Tensor, List[torch.Tensor]],
negative_image_embeds: Union[torch.Tensor, List[torch.Tensor]],
height: int = 512,
width: int = 512,
num_inference_steps: int = 100,
guidance_scale: float = 4.0,
num_images_per_prompt: int = 1,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil",
return_dict: bool = True,
callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
......@@ -142,9 +142,9 @@ class KandinskyV22Pipeline(DiffusionPipeline):
Function invoked when calling the pipeline for generation.
Args:
image_embeds (`torch.FloatTensor` or `List[torch.FloatTensor]`):
image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
The clip image embeddings for text prompt, that will be used to condition the image generation.
negative_image_embeds (`torch.FloatTensor` or `List[torch.FloatTensor]`):
negative_image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
The clip image embeddings for negative text prompt, will be used to condition the image generation.
height (`int`, *optional*, defaults to 512):
The height in pixels of the generated image.
......@@ -164,7 +164,7 @@ class KandinskyV22Pipeline(DiffusionPipeline):
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)
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
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`.
......
src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_combined.py
View file @ be4afa0b
......@@ -213,9 +213,9 @@ class KandinskyV22CombinedPipeline(DiffusionPipeline):
prior_guidance_scale: float = 4.0,
prior_num_inference_steps: int = 25,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil",
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None,
callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
callback_steps: int = 1,
return_dict: bool = True,
prior_callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
......@@ -259,7 +259,7 @@ class KandinskyV22CombinedPipeline(DiffusionPipeline):
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)
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
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`.
......@@ -442,7 +442,7 @@ class KandinskyV22Img2ImgCombinedPipeline(DiffusionPipeline):
def __call__(
self,
prompt: Union[str, List[str]],
image: Union[torch.FloatTensor, PIL.Image.Image, List[torch.FloatTensor], List[PIL.Image.Image]],
image: Union[torch.Tensor, PIL.Image.Image, List[torch.Tensor], List[PIL.Image.Image]],
negative_prompt: Optional[Union[str, List[str]]] = None,
num_inference_steps: int = 100,
guidance_scale: float = 4.0,
......@@ -453,9 +453,9 @@ class KandinskyV22Img2ImgCombinedPipeline(DiffusionPipeline):
prior_guidance_scale: float = 4.0,
prior_num_inference_steps: int = 25,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil",
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None,
callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
callback_steps: int = 1,
return_dict: bool = True,
prior_callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
......@@ -469,7 +469,7 @@ class KandinskyV22Img2ImgCombinedPipeline(DiffusionPipeline):
Args:
prompt (`str` or `List[str]`):
The prompt or prompts to guide the image generation.
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, that will be used as the starting point for the
process. Can also accept image latents as `image`, if passing latents directly, it will not be encoded
again.
......@@ -509,7 +509,7 @@ class KandinskyV22Img2ImgCombinedPipeline(DiffusionPipeline):
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)
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
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`.
......@@ -518,7 +518,7 @@ class KandinskyV22Img2ImgCombinedPipeline(DiffusionPipeline):
(`np.array`) or `"pt"` (`torch.Tensor`).
callback (`Callable`, *optional*):
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):
The frequency at which the `callback` function is called. If not specified, the callback is called at
every step.
......@@ -681,8 +681,8 @@ class KandinskyV22InpaintCombinedPipeline(DiffusionPipeline):
def __call__(
self,
prompt: Union[str, List[str]],
image: Union[torch.FloatTensor, PIL.Image.Image, List[torch.FloatTensor], List[PIL.Image.Image]],
mask_image: Union[torch.FloatTensor, PIL.Image.Image, List[torch.FloatTensor], List[PIL.Image.Image]],
image: Union[torch.Tensor, PIL.Image.Image, List[torch.Tensor], List[PIL.Image.Image]],
mask_image: Union[torch.Tensor, PIL.Image.Image, List[torch.Tensor], List[PIL.Image.Image]],
negative_prompt: Optional[Union[str, List[str]]] = None,
num_inference_steps: int = 100,
guidance_scale: float = 4.0,
......@@ -692,7 +692,7 @@ class KandinskyV22InpaintCombinedPipeline(DiffusionPipeline):
prior_guidance_scale: float = 4.0,
prior_num_inference_steps: int = 25,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil",
return_dict: bool = True,
prior_callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
......@@ -707,7 +707,7 @@ class KandinskyV22InpaintCombinedPipeline(DiffusionPipeline):
Args:
prompt (`str` or `List[str]`):
The prompt or prompts to guide the image generation.
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, that will be used as the starting point for the
process. Can also accept image latents as `image`, if passing latents directly, it will not be encoded
again.
......@@ -746,7 +746,7 @@ class KandinskyV22InpaintCombinedPipeline(DiffusionPipeline):
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)
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
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`.
......
src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_controlnet.py
View file @ be4afa0b
......@@ -151,18 +151,18 @@ class KandinskyV22ControlnetPipeline(DiffusionPipeline):
@torch.no_grad()
def __call__(
self,
image_embeds: Union[torch.FloatTensor, List[torch.FloatTensor]],
negative_image_embeds: Union[torch.FloatTensor, List[torch.FloatTensor]],
hint: torch.FloatTensor,
image_embeds: Union[torch.Tensor, List[torch.Tensor]],
negative_image_embeds: Union[torch.Tensor, List[torch.Tensor]],
hint: torch.Tensor,
height: int = 512,
width: int = 512,
num_inference_steps: int = 100,
guidance_scale: float = 4.0,
num_images_per_prompt: int = 1,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil",
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None,
callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
callback_steps: int = 1,
return_dict: bool = True,
):
......@@ -172,11 +172,11 @@ class KandinskyV22ControlnetPipeline(DiffusionPipeline):
Args:
prompt (`str` or `List[str]`):
The prompt or prompts to guide the image generation.
hint (`torch.FloatTensor`):
hint (`torch.Tensor`):
The controlnet condition.
image_embeds (`torch.FloatTensor` or `List[torch.FloatTensor]`):
image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
The clip image embeddings for text prompt, that will be used to condition the image generation.
negative_image_embeds (`torch.FloatTensor` or `List[torch.FloatTensor]`):
negative_image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
The clip image embeddings for negative text prompt, will be used to condition the image generation.
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
......@@ -199,7 +199,7 @@ class KandinskyV22ControlnetPipeline(DiffusionPipeline):
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)
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
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`.
......@@ -208,7 +208,7 @@ class KandinskyV22ControlnetPipeline(DiffusionPipeline):
(`np.array`) or `"pt"` (`torch.Tensor`).
callback (`Callable`, *optional*):
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):
The frequency at which the `callback` function is called. If not specified, the callback is called at
every step.
......
src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_controlnet_img2img.py
View file @ be4afa0b
......@@ -206,10 +206,10 @@ class KandinskyV22ControlnetImg2ImgPipeline(DiffusionPipeline):
@torch.no_grad()
def __call__(
self,
image_embeds: Union[torch.FloatTensor, List[torch.FloatTensor]],
image: Union[torch.FloatTensor, PIL.Image.Image, List[torch.FloatTensor], List[PIL.Image.Image]],
negative_image_embeds: Union[torch.FloatTensor, List[torch.FloatTensor]],
hint: torch.FloatTensor,
image_embeds: Union[torch.Tensor, List[torch.Tensor]],
image: Union[torch.Tensor, PIL.Image.Image, List[torch.Tensor], List[PIL.Image.Image]],
negative_image_embeds: Union[torch.Tensor, List[torch.Tensor]],
hint: torch.Tensor,
height: int = 512,
width: int = 512,
num_inference_steps: int = 100,
......@@ -218,7 +218,7 @@ class KandinskyV22ControlnetImg2ImgPipeline(DiffusionPipeline):
num_images_per_prompt: int = 1,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
output_type: Optional[str] = "pil",
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None,
callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
callback_steps: int = 1,
return_dict: bool = True,
):
......@@ -226,9 +226,9 @@ class KandinskyV22ControlnetImg2ImgPipeline(DiffusionPipeline):
Function invoked when calling the pipeline for generation.
Args:
image_embeds (`torch.FloatTensor` or `List[torch.FloatTensor]`):
image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
The clip image embeddings for text prompt, that will be used to condition the image generation.
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, that will be used as the starting point for the
process. Can also accept image latents as `image`, if passing latents directly, it will not be encoded
again.
......@@ -238,9 +238,9 @@ class KandinskyV22ControlnetImg2ImgPipeline(DiffusionPipeline):
denoising steps depends on the amount of noise initially added. When `strength` is 1, added noise will
be maximum and the denoising process will run for the full number of iterations specified in
`num_inference_steps`. A value of 1, therefore, essentially ignores `image`.
hint (`torch.FloatTensor`):
hint (`torch.Tensor`):
The controlnet condition.
negative_image_embeds (`torch.FloatTensor` or `List[torch.FloatTensor]`):
negative_image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
The clip image embeddings for negative text prompt, will be used to condition the image generation.
height (`int`, *optional*, defaults to 512):
The height in pixels of the generated image.
......@@ -265,7 +265,7 @@ class KandinskyV22ControlnetImg2ImgPipeline(DiffusionPipeline):
(`np.array`) or `"pt"` (`torch.Tensor`).
callback (`Callable`, *optional*):
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):
The frequency at which the `callback` function is called. If not specified, the callback is called at
every step.
......
src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_img2img.py
View file @ be4afa0b
......@@ -190,9 +190,9 @@ class KandinskyV22Img2ImgPipeline(DiffusionPipeline):
@torch.no_grad()
def __call__(
self,
image_embeds: Union[torch.FloatTensor, List[torch.FloatTensor]],
image: Union[torch.FloatTensor, PIL.Image.Image, List[torch.FloatTensor], List[PIL.Image.Image]],
negative_image_embeds: Union[torch.FloatTensor, List[torch.FloatTensor]],
image_embeds: Union[torch.Tensor, List[torch.Tensor]],
image: Union[torch.Tensor, PIL.Image.Image, List[torch.Tensor], List[PIL.Image.Image]],
negative_image_embeds: Union[torch.Tensor, List[torch.Tensor]],
height: int = 512,
width: int = 512,
num_inference_steps: int = 100,
......@@ -210,9 +210,9 @@ class KandinskyV22Img2ImgPipeline(DiffusionPipeline):
Function invoked when calling the pipeline for generation.
Args:
image_embeds (`torch.FloatTensor` or `List[torch.FloatTensor]`):
image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
The clip image embeddings for text prompt, that will be used to condition the image generation.
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, that will be used as the starting point for the
process. Can also accept image latents as `image`, if passing latents directly, it will not be encoded
again.
......@@ -222,7 +222,7 @@ class KandinskyV22Img2ImgPipeline(DiffusionPipeline):
denoising steps depends on the amount of noise initially added. When `strength` is 1, added noise will
be maximum and the denoising process will run for the full number of iterations specified in
`num_inference_steps`. A value of 1, therefore, essentially ignores `image`.
negative_image_embeds (`torch.FloatTensor` or `List[torch.FloatTensor]`):
negative_image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
The clip image embeddings for negative text prompt, will be used to condition the image generation.
height (`int`, *optional*, defaults to 512):
The height in pixels of the generated image.
......
src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_inpainting.py
View file @ be4afa0b
......@@ -294,17 +294,17 @@ class KandinskyV22InpaintPipeline(DiffusionPipeline):
@torch.no_grad()
def __call__(
self,
image_embeds: Union[torch.FloatTensor, List[torch.FloatTensor]],
image: Union[torch.FloatTensor, PIL.Image.Image],
mask_image: Union[torch.FloatTensor, PIL.Image.Image, np.ndarray],
negative_image_embeds: Union[torch.FloatTensor, List[torch.FloatTensor]],
image_embeds: Union[torch.Tensor, List[torch.Tensor]],
image: Union[torch.Tensor, PIL.Image.Image],
mask_image: Union[torch.Tensor, PIL.Image.Image, np.ndarray],
negative_image_embeds: Union[torch.Tensor, List[torch.Tensor]],
height: int = 512,
width: int = 512,
num_inference_steps: int = 100,
guidance_scale: float = 4.0,
num_images_per_prompt: int = 1,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil",
return_dict: bool = True,
callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
......@@ -315,7 +315,7 @@ class KandinskyV22InpaintPipeline(DiffusionPipeline):
Function invoked when calling the pipeline for generation.
Args:
image_embeds (`torch.FloatTensor` or `List[torch.FloatTensor]`):
image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
The clip image embeddings for text prompt, that will be used to condition the image generation.
image (`PIL.Image.Image`):
`Image`, or tensor representing an image batch which will be inpainted, *i.e.* parts of the image will
......@@ -325,7 +325,7 @@ class KandinskyV22InpaintPipeline(DiffusionPipeline):
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) instead of 3,
so the expected shape would be `(B, H, W, 1)`.
negative_image_embeds (`torch.FloatTensor` or `List[torch.FloatTensor]`):
negative_image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
The clip image embeddings for negative text prompt, will be used to condition the image generation.
height (`int`, *optional*, defaults to 512):
The height in pixels of the generated image.
......@@ -345,7 +345,7 @@ class KandinskyV22InpaintPipeline(DiffusionPipeline):
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)
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
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`.
......
src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_prior.py
View file @ be4afa0b
......@@ -132,12 +132,12 @@ class KandinskyV22PriorPipeline(DiffusionPipeline):
@replace_example_docstring(EXAMPLE_INTERPOLATE_DOC_STRING)
def interpolate(
self,
images_and_prompts: List[Union[str, PIL.Image.Image, torch.FloatTensor]],
images_and_prompts: List[Union[str, PIL.Image.Image, torch.Tensor]],
weights: List[float],
num_images_per_prompt: int = 1,
num_inference_steps: int = 25,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
negative_prior_prompt: Optional[str] = None,
negative_prompt: str = "",
guidance_scale: float = 4.0,
......@@ -147,7 +147,7 @@ class KandinskyV22PriorPipeline(DiffusionPipeline):
Function invoked when using the prior pipeline for interpolation.
Args:
images_and_prompts (`List[Union[str, PIL.Image.Image, torch.FloatTensor]]`):
images_and_prompts (`List[Union[str, PIL.Image.Image, torch.Tensor]]`):
list of prompts and images to guide the image generation.
weights: (`List[float]`):
list of weights for each condition in `images_and_prompts`
......@@ -159,7 +159,7 @@ class KandinskyV22PriorPipeline(DiffusionPipeline):
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)
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
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`.
......@@ -376,7 +376,7 @@ class KandinskyV22PriorPipeline(DiffusionPipeline):
num_images_per_prompt: int = 1,
num_inference_steps: int = 25,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
guidance_scale: float = 4.0,
output_type: Optional[str] = "pt", # pt only
return_dict: bool = True,
......@@ -400,7 +400,7 @@ class KandinskyV22PriorPipeline(DiffusionPipeline):
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)
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
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`.
......
src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_prior_emb2emb.py
View file @ be4afa0b
......@@ -156,12 +156,12 @@ class KandinskyV22PriorEmb2EmbPipeline(DiffusionPipeline):
@replace_example_docstring(EXAMPLE_INTERPOLATE_DOC_STRING)
def interpolate(
self,
images_and_prompts: List[Union[str, PIL.Image.Image, torch.FloatTensor]],
images_and_prompts: List[Union[str, PIL.Image.Image, torch.Tensor]],
weights: List[float],
num_images_per_prompt: int = 1,
num_inference_steps: int = 25,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
negative_prior_prompt: Optional[str] = None,
negative_prompt: str = "",
guidance_scale: float = 4.0,
......@@ -171,7 +171,7 @@ class KandinskyV22PriorEmb2EmbPipeline(DiffusionPipeline):
Function invoked when using the prior pipeline for interpolation.
Args:
images_and_prompts (`List[Union[str, PIL.Image.Image, torch.FloatTensor]]`):
images_and_prompts (`List[Union[str, PIL.Image.Image, torch.Tensor]]`):
list of prompts and images to guide the image generation.
weights: (`List[float]`):
list of weights for each condition in `images_and_prompts`
......@@ -183,7 +183,7 @@ class KandinskyV22PriorEmb2EmbPipeline(DiffusionPipeline):
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)
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
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`.
......@@ -418,7 +418,7 @@ class KandinskyV22PriorEmb2EmbPipeline(DiffusionPipeline):
Conceptually, indicates how much to transform the reference `emb`. Must be between 0 and 1. `image`
will be used as a starting point, adding more noise to it the larger the `strength`. The number of
denoising steps depends on the amount of noise initially added.
emb (`torch.FloatTensor`):
emb (`torch.Tensor`):
The image embedding.
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
......
src/diffusers/pipelines/kandinsky3/pipeline_kandinsky3.py
View file @ be4afa0b
......@@ -87,11 +87,11 @@ class Kandinsky3Pipeline(DiffusionPipeline, LoraLoaderMixin):
num_images_per_prompt=1,
device=None,
negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
_cut_context=False,
attention_mask: Optional[torch.FloatTensor] = None,
negative_attention_mask: Optional[torch.FloatTensor] = None,
attention_mask: Optional[torch.Tensor] = None,
negative_attention_mask: Optional[torch.Tensor] = None,
):
r"""
Encodes the prompt into text encoder hidden states.
......@@ -109,16 +109,16 @@ class Kandinsky3Pipeline(DiffusionPipeline, LoraLoaderMixin):
The prompt or prompts not to guide the image generation. If not defined, one has to pass
`negative_prompt_embeds`. instead. If not defined, one has to pass `negative_prompt_embeds`. instead.
Ignored when not using guidance (i.e., ignored if `guidance_scale` is 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
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
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument.
attention_mask (`torch.FloatTensor`, *optional*):
attention_mask (`torch.Tensor`, *optional*):
Pre-generated attention mask. Must provide if passing `prompt_embeds` directly.
negative_attention_mask (`torch.FloatTensor`, *optional*):
negative_attention_mask (`torch.Tensor`, *optional*):
Pre-generated negative attention mask. Must provide if passing `negative_prompt_embeds` directly.
"""
if prompt is not None and negative_prompt is not None:
......@@ -334,10 +334,10 @@ class Kandinsky3Pipeline(DiffusionPipeline, LoraLoaderMixin):
height: Optional[int] = 1024,
width: Optional[int] = 1024,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
attention_mask: Optional[torch.FloatTensor] = None,
negative_attention_mask: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
attention_mask: Optional[torch.Tensor] = None,
negative_attention_mask: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil",
return_dict: bool = True,
latents=None,
......@@ -380,16 +380,16 @@ class Kandinsky3Pipeline(DiffusionPipeline, LoraLoaderMixin):
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)
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
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
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument.
attention_mask (`torch.FloatTensor`, *optional*):
attention_mask (`torch.Tensor`, *optional*):
Pre-generated attention mask. Must provide if passing `prompt_embeds` directly.
negative_attention_mask (`torch.FloatTensor`, *optional*):
negative_attention_mask (`torch.Tensor`, *optional*):
Pre-generated negative attention mask. Must provide if passing `negative_prompt_embeds` directly.
output_type (`str`, *optional*, defaults to `"pil"`):
The output format of the generate image. Choose between
......@@ -398,7 +398,7 @@ class Kandinsky3Pipeline(DiffusionPipeline, LoraLoaderMixin):
Whether or not to return a [`~pipelines.stable_diffusion.IFPipelineOutput`] instead of a plain tuple.
callback (`Callable`, *optional*):
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):
The frequency at which the `callback` function will be called. If not specified, the callback will be
called at every step.
......
src/diffusers/pipelines/kandinsky3/pipeline_kandinsky3_img2img.py
View file @ be4afa0b
......@@ -112,11 +112,11 @@ class Kandinsky3Img2ImgPipeline(DiffusionPipeline, LoraLoaderMixin):
num_images_per_prompt=1,
device=None,
negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
_cut_context=False,
attention_mask: Optional[torch.FloatTensor] = None,
negative_attention_mask: Optional[torch.FloatTensor] = None,
attention_mask: Optional[torch.Tensor] = None,
negative_attention_mask: Optional[torch.Tensor] = None,
):
r"""
Encodes the prompt into text encoder hidden states.
......@@ -134,16 +134,16 @@ class Kandinsky3Img2ImgPipeline(DiffusionPipeline, LoraLoaderMixin):
The prompt or prompts not to guide the image generation. If not defined, one has to pass
`negative_prompt_embeds`. instead. If not defined, one has to pass `negative_prompt_embeds`. instead.
Ignored when not using guidance (i.e., ignored if `guidance_scale` is 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
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
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument.
attention_mask (`torch.FloatTensor`, *optional*):
attention_mask (`torch.Tensor`, *optional*):
Pre-generated attention mask. Must provide if passing `prompt_embeds` directly.
negative_attention_mask (`torch.FloatTensor`, *optional*):
negative_attention_mask (`torch.Tensor`, *optional*):
Pre-generated negative attention mask. Must provide if passing `negative_prompt_embeds` directly.
"""
if prompt is not None and negative_prompt is not None:
......@@ -403,17 +403,17 @@ class Kandinsky3Img2ImgPipeline(DiffusionPipeline, LoraLoaderMixin):
def __call__(
self,
prompt: Union[str, List[str]] = None,
image: Union[torch.FloatTensor, PIL.Image.Image, List[torch.FloatTensor], List[PIL.Image.Image]] = None,
image: Union[torch.Tensor, PIL.Image.Image, List[torch.Tensor], List[PIL.Image.Image]] = None,
strength: float = 0.3,
num_inference_steps: int = 25,
guidance_scale: float = 3.0,
negative_prompt: Optional[Union[str, List[str]]] = None,
num_images_per_prompt: Optional[int] = 1,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
attention_mask: Optional[torch.FloatTensor] = None,
negative_attention_mask: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
attention_mask: Optional[torch.Tensor] = None,
negative_attention_mask: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil",
return_dict: bool = True,
callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
......@@ -427,7 +427,7 @@ class Kandinsky3Img2ImgPipeline(DiffusionPipeline, LoraLoaderMixin):
prompt (`str` or `List[str]`, *optional*):
The prompt or prompts to guide the image generation. If not defined, one has to pass `prompt_embeds`.
instead.
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, that will be used as the starting point for the
process.
strength (`float`, *optional*, defaults to 0.8):
......@@ -454,16 +454,16 @@ class Kandinsky3Img2ImgPipeline(DiffusionPipeline, LoraLoaderMixin):
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)
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
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
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument.
attention_mask (`torch.FloatTensor`, *optional*):
attention_mask (`torch.Tensor`, *optional*):
Pre-generated attention mask. Must provide if passing `prompt_embeds` directly.
negative_attention_mask (`torch.FloatTensor`, *optional*):
negative_attention_mask (`torch.Tensor`, *optional*):
Pre-generated negative attention mask. Must provide if passing `negative_prompt_embeds` directly.
output_type (`str`, *optional*, defaults to `"pil"`):
The output format of the generate image. Choose between
......
src/diffusers/pipelines/latent_consistency_models/pipeline_latent_consistency_img2img.py
View file @ be4afa0b
......@@ -237,8 +237,8 @@ class LatentConsistencyModelImg2ImgPipeline(
num_images_per_prompt,
do_classifier_free_guidance,
negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None,
):
......@@ -258,10 +258,10 @@ class LatentConsistencyModelImg2ImgPipeline(
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
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
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
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument.
......@@ -565,7 +565,7 @@ class LatentConsistencyModelImg2ImgPipeline(
# Copied from diffusers.pipelines.latent_consistency_models.pipeline_latent_consistency_text2img.LatentConsistencyModelPipeline.get_guidance_scale_embedding
def get_guidance_scale_embedding(
self, w: torch.Tensor, embedding_dim: int = 512, dtype: torch.dtype = torch.float32
) -> torch.FloatTensor:
) -> torch.Tensor:
"""
See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298
......@@ -578,7 +578,7 @@ class LatentConsistencyModelImg2ImgPipeline(
Data type of the generated embeddings.
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
w = w * 1000.0
......@@ -628,7 +628,7 @@ class LatentConsistencyModelImg2ImgPipeline(
prompt: Union[str, List[str]],
strength: float,
callback_steps: int,
prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
ip_adapter_image=None,
ip_adapter_image_embeds=None,
callback_on_step_end_tensor_inputs=None,
......@@ -709,10 +709,10 @@ class LatentConsistencyModelImg2ImgPipeline(
guidance_scale: float = 8.5,
num_images_per_prompt: Optional[int] = 1,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.Tensor] = 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",
return_dict: bool = True,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
......@@ -754,16 +754,16 @@ class LatentConsistencyModelImg2ImgPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) 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
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`.
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
provided, text embeddings are generated from the `prompt` input argument.
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. 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 not
......
src/diffusers/pipelines/latent_consistency_models/pipeline_latent_consistency_text2img.py
View file @ be4afa0b
......@@ -221,8 +221,8 @@ class LatentConsistencyModelPipeline(
num_images_per_prompt,
do_classifier_free_guidance,
negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None,
):
......@@ -242,10 +242,10 @@ class LatentConsistencyModelPipeline(
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
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
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
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument.
......@@ -512,7 +512,7 @@ class LatentConsistencyModelPipeline(
def get_guidance_scale_embedding(
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
......@@ -525,7 +525,7 @@ class LatentConsistencyModelPipeline(
Data type of the generated embeddings.
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
w = w * 1000.0
......@@ -565,7 +565,7 @@ class LatentConsistencyModelPipeline(
height: int,
width: int,
callback_steps: int,
prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
ip_adapter_image=None,
ip_adapter_image_embeds=None,
callback_on_step_end_tensor_inputs=None,
......@@ -646,10 +646,10 @@ class LatentConsistencyModelPipeline(
guidance_scale: float = 8.5,
num_images_per_prompt: Optional[int] = 1,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.Tensor] = 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",
return_dict: bool = True,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
......@@ -691,16 +691,16 @@ class LatentConsistencyModelPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) 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
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`.
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
provided, text embeddings are generated from the `prompt` input argument.
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. 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 not
......
src/diffusers/pipelines/latent_diffusion/pipeline_latent_diffusion.py
View file @ be4afa0b
......@@ -74,7 +74,7 @@ class LDMTextToImagePipeline(DiffusionPipeline):
guidance_scale: Optional[float] = 1.0,
eta: Optional[float] = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil",
return_dict: bool = True,
**kwargs,
......@@ -98,7 +98,7 @@ class LDMTextToImagePipeline(DiffusionPipeline):
generator (`torch.Generator`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) 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
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`.
......@@ -465,17 +465,17 @@ class LDMBertEncoderLayer(nn.Module):
def forward(
self,
hidden_states: torch.FloatTensor,
attention_mask: torch.FloatTensor,
layer_head_mask: torch.FloatTensor,
hidden_states: torch.Tensor,
attention_mask: torch.Tensor,
layer_head_mask: torch.Tensor,
output_attentions: Optional[bool] = False,
) -> Tuple[torch.FloatTensor, Optional[torch.FloatTensor]]:
) -> Tuple[torch.Tensor, Optional[torch.Tensor]]:
"""
Args:
hidden_states (`torch.FloatTensor`): input to the layer of shape `(seq_len, batch, embed_dim)`
attention_mask (`torch.FloatTensor`): attention mask of size
hidden_states (`torch.Tensor`): input to the layer of shape `(seq_len, batch, embed_dim)`
attention_mask (`torch.Tensor`): attention mask of size
`(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size
layer_head_mask (`torch.Tensor`): mask for attention heads in a given layer of size
`(encoder_attention_heads,)`.
output_attentions (`bool`, *optional*):
Whether or not to return the attentions tensors of all attention layers. See `attentions` under
......@@ -587,7 +587,7 @@ class LDMBertEncoder(LDMBertPreTrainedModel):
attention_mask: Optional[torch.Tensor] = None,
position_ids: Optional[torch.LongTensor] = None,
head_mask: Optional[torch.Tensor] = None,
inputs_embeds: Optional[torch.FloatTensor] = None,
inputs_embeds: Optional[torch.Tensor] = None,
output_attentions: Optional[bool] = None,
output_hidden_states: Optional[bool] = None,
return_dict: Optional[bool] = None,
......@@ -615,7 +615,7 @@ class LDMBertEncoder(LDMBertPreTrainedModel):
- 1 indicates the head is **not masked**,
- 0 indicates the head is **masked**.
inputs_embeds (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
inputs_embeds (`torch.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation.
This is useful if you want more control over how to convert `input_ids` indices into associated vectors
than the model's internal embedding lookup matrix.
......
src/diffusers/pipelines/ledits_pp/pipeline_leditspp_stable_diffusion.py
View file @ be4afa0b
......@@ -502,8 +502,8 @@ class LEditsPPPipelineStableDiffusion(
enable_edit_guidance,
negative_prompt=None,
editing_prompt=None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
editing_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
editing_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None,
):
......@@ -523,10 +523,10 @@ class LEditsPPPipelineStableDiffusion(
less than `1`).
editing_prompt (`str` or `List[str]`, *optional*):
Editing prompt(s) to be encoded. If not defined, one has to pass `editing_prompt_embeds` instead.
editing_prompt_embeds (`torch.FloatTensor`, *optional*):
editing_prompt_embeds (`torch.Tensor`, *optional*):
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.
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
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument.
......@@ -704,13 +704,13 @@ class LEditsPPPipelineStableDiffusion(
return_dict: bool = True,
editing_prompt: Optional[Union[str, List[str]]] = None,
editing_prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
reverse_editing_direction: Optional[Union[bool, List[bool]]] = False,
edit_guidance_scale: Optional[Union[float, List[float]]] = 5,
edit_warmup_steps: Optional[Union[int, List[int]]] = 0,
edit_cooldown_steps: Optional[Union[int, List[int]]] = None,
edit_threshold: Optional[Union[float, List[float]]] = 0.9,
user_mask: Optional[torch.FloatTensor] = None,
user_mask: Optional[torch.Tensor] = None,
sem_guidance: Optional[List[torch.Tensor]] = None,
use_cross_attn_mask: bool = False,
use_intersect_mask: bool = True,
......@@ -748,7 +748,7 @@ class LEditsPPPipelineStableDiffusion(
editing_prompt_embeds (`torch.Tensor>`, *optional*):
Pre-computed embeddings to use for guiding the image generation. Guidance direction of embedding should
be specified via `reverse_editing_direction`.
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
not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
reverse_editing_direction (`bool` or `List[bool]`, *optional*, defaults to `False`):
......@@ -765,7 +765,7 @@ class LEditsPPPipelineStableDiffusion(
Masking threshold of guidance. Threshold should be proportional to the image region that is modified.
'edit_threshold' is defined as 'λ' of equation 12 of [LEDITS++
Paper](https://arxiv.org/abs/2301.12247).
user_mask (`torch.FloatTensor`, *optional*):
user_mask (`torch.Tensor`, *optional*):
User-provided mask for even better control over the editing process. This is helpful when LEDITS++'s
implicit masks do not meet user preferences.
sem_guidance (`List[torch.Tensor]`, *optional*):
......
src/diffusers/pipelines/ledits_pp/pipeline_leditspp_stable_diffusion_xl.py
View file @ be4afa0b
......@@ -409,14 +409,14 @@ class LEditsPPPipelineStableDiffusionXL(
num_images_per_prompt: int = 1,
negative_prompt: Optional[str] = None,
negative_prompt_2: Optional[str] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None,
enable_edit_guidance: bool = True,
editing_prompt: Optional[str] = None,
editing_prompt_embeds: Optional[torch.FloatTensor] = None,
editing_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
editing_prompt_embeds: Optional[torch.Tensor] = None,
editing_pooled_prompt_embeds: Optional[torch.Tensor] = None,
) -> object:
r"""
Encodes the prompt into text encoder hidden states.
......@@ -432,11 +432,11 @@ class LEditsPPPipelineStableDiffusionXL(
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
`text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders
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
weighting. If not provided, negative_prompt_embeds will be generated from `negative_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
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument.
......@@ -450,11 +450,11 @@ class LEditsPPPipelineStableDiffusionXL(
editing_prompt (`str` or `List[str]`, *optional*):
Editing prompt(s) to be encoded. If not defined and 'enable_edit_guidance' is True, one has to pass
`editing_prompt_embeds` instead.
editing_prompt_embeds (`torch.FloatTensor`, *optional*):
editing_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated edit text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided and 'enable_edit_guidance' is True, editing_prompt_embeds will be generated from
`editing_prompt` input argument.
editing_pooled_prompt_embeds (`torch.FloatTensor`, *optional*):
editing_pooled_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated edit pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled editing_pooled_prompt_embeds will be generated from `editing_prompt`
input argument.
......@@ -713,7 +713,7 @@ class LEditsPPPipelineStableDiffusionXL(
# Copied from diffusers.pipelines.latent_consistency_models.pipeline_latent_consistency_text2img.LatentConsistencyModelPipeline.get_guidance_scale_embedding
def get_guidance_scale_embedding(
self, w: torch.Tensor, embedding_dim: int = 512, dtype: torch.dtype = torch.float32
) -> torch.FloatTensor:
) -> torch.Tensor:
"""
See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298
......@@ -726,7 +726,7 @@ class LEditsPPPipelineStableDiffusionXL(
Data type of the generated embeddings.
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
w = w * 1000.0
......@@ -804,8 +804,8 @@ class LEditsPPPipelineStableDiffusionXL(
denoising_end: Optional[float] = None,
negative_prompt: Optional[Union[str, List[str]]] = None,
negative_prompt_2: Optional[Union[str, List[str]]] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
ip_adapter_image: Optional[PipelineImageInput] = None,
output_type: Optional[str] = "pil",
return_dict: bool = True,
......@@ -824,7 +824,7 @@ class LEditsPPPipelineStableDiffusionXL(
sem_guidance: Optional[List[torch.Tensor]] = None,
use_cross_attn_mask: bool = False,
use_intersect_mask: bool = False,
user_mask: Optional[torch.FloatTensor] = None,
user_mask: Optional[torch.Tensor] = None,
attn_store_steps: Optional[List[int]] = [],
store_averaged_over_steps: bool = True,
clip_skip: Optional[int] = None,
......@@ -851,11 +851,11 @@ class LEditsPPPipelineStableDiffusionXL(
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
`text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders
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
weighting. If not provided, negative_prompt_embeds will be generated from `negative_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
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument.
......@@ -869,7 +869,7 @@ class LEditsPPPipelineStableDiffusionXL(
of a plain tuple.
callback (`Callable`, *optional*):
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):
The frequency at which the `callback` function will be called. If not specified, the callback will be
called at every step.
......
src/diffusers/pipelines/musicldm/pipeline_musicldm.py
View file @ be4afa0b
......@@ -120,8 +120,8 @@ class MusicLDMPipeline(DiffusionPipeline, StableDiffusionMixin):
num_waveforms_per_prompt,
do_classifier_free_guidance,
negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
):
r"""
Encodes the prompt into text encoder hidden states.
......@@ -139,10 +139,10 @@ class MusicLDMPipeline(DiffusionPipeline, StableDiffusionMixin):
The prompt or prompts not to guide the audio generation. If not defined, one has to pass
`negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
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
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
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument.
......@@ -427,11 +427,11 @@ class MusicLDMPipeline(DiffusionPipeline, StableDiffusionMixin):
num_waveforms_per_prompt: Optional[int] = 1,
eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
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,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
output_type: Optional[str] = "np",
......@@ -465,21 +465,21 @@ class MusicLDMPipeline(DiffusionPipeline, StableDiffusionMixin):
generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) 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
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`.
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
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
not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
return_dict (`bool`, *optional*, defaults to `True`):
Whether or not to return a [`~pipelines.AudioPipelineOutput`] instead of a plain tuple.
callback (`Callable`, *optional*):
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):
The frequency at which the `callback` function is called. If not specified, the callback is called at
every step.
......
src/diffusers/pipelines/paint_by_example/pipeline_paint_by_example.py
View file @ be4afa0b
......@@ -266,7 +266,7 @@ class PaintByExamplePipeline(DiffusionPipeline, StableDiffusionMixin):
and not isinstance(image, list)
):
raise ValueError(
"`image` has to be of type `torch.FloatTensor` or `PIL.Image.Image` or `List[PIL.Image.Image]` but is"
"`image` has to be of type `torch.Tensor` or `PIL.Image.Image` or `List[PIL.Image.Image]` but is"
f" {type(image)}"
)
......@@ -393,9 +393,9 @@ class PaintByExamplePipeline(DiffusionPipeline, StableDiffusionMixin):
@torch.no_grad()
def __call__(
self,
example_image: Union[torch.FloatTensor, PIL.Image.Image],
image: Union[torch.FloatTensor, PIL.Image.Image],
mask_image: Union[torch.FloatTensor, PIL.Image.Image],
example_image: Union[torch.Tensor, PIL.Image.Image],
image: Union[torch.Tensor, PIL.Image.Image],
mask_image: Union[torch.Tensor, PIL.Image.Image],
height: Optional[int] = None,
width: Optional[int] = None,
num_inference_steps: int = 50,
......@@ -404,22 +404,22 @@ class PaintByExamplePipeline(DiffusionPipeline, StableDiffusionMixin):
num_images_per_prompt: Optional[int] = 1,
eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil",
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,
):
r"""
The call function to the pipeline for generation.
Args:
example_image (`torch.FloatTensor` or `PIL.Image.Image` or `List[PIL.Image.Image]`):
example_image (`torch.Tensor` or `PIL.Image.Image` or `List[PIL.Image.Image]`):
An example image to guide image generation.
image (`torch.FloatTensor` or `PIL.Image.Image` or `List[PIL.Image.Image]`):
image (`torch.Tensor` or `PIL.Image.Image` or `List[PIL.Image.Image]`):
`Image` or tensor representing an image batch to be inpainted (parts of the image are masked out with
`mask_image` and repainted according to `prompt`).
mask_image (`torch.FloatTensor` or `PIL.Image.Image` or `List[PIL.Image.Image]`):
mask_image (`torch.Tensor` or `PIL.Image.Image` or `List[PIL.Image.Image]`):
`Image` or tensor representing an image batch to mask `image`. White pixels in the mask are repainted,
while black pixels are preserved. If `mask_image` is a PIL image, it is converted to a single channel
(luminance) before use. If it's a tensor, it should contain one color channel (L) instead of 3, so the
......@@ -445,7 +445,7 @@ class PaintByExamplePipeline(DiffusionPipeline, StableDiffusionMixin):
generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) 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
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`.
......@@ -456,7 +456,7 @@ class PaintByExamplePipeline(DiffusionPipeline, StableDiffusionMixin):
plain tuple.
callback (`Callable`, *optional*):
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):
The frequency at which the `callback` function is called. If not specified, the callback is called at
every step.
......
src/diffusers/pipelines/pia/pipeline_pia.py
View file @ be4afa0b
......@@ -207,8 +207,8 @@ class PIAPipeline(
num_images_per_prompt,
do_classifier_free_guidance,
negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None,
):
......@@ -228,10 +228,10 @@ class PIAPipeline(
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
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
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
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument.
......@@ -680,11 +680,11 @@ class PIAPipeline(
num_videos_per_prompt: Optional[int] = 1,
eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = 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,
motion_scale: int = 0,
output_type: Optional[str] = "pil",
return_dict: bool = True,
......@@ -725,20 +725,20 @@ class PIAPipeline(
generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) 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 video
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
`(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
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
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_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. 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 not
......@@ -749,8 +749,7 @@ class PIAPipeline(
added. Must be between 0 and 8. Set between 0-2 to only increase the amount of motion. Set between 3-5
to create looping motion. Set between 6-8 to perform motion with image style transfer.
output_type (`str`, *optional*, defaults to `"pil"`):
The output format of the generated video. Choose between `torch.FloatTensor`, `PIL.Image` or
`np.array`.
The output format of the generated video. Choose between `torch.Tensor`, `PIL.Image` or `np.array`.
return_dict (`bool`, *optional*, defaults to `True`):
Whether or not to return a [`~pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput`] instead
of a plain tuple.
......
src/diffusers/pipelines/pixart_alpha/pipeline_pixart_alpha.py
View file @ be4afa0b
......@@ -296,10 +296,10 @@ class PixArtAlphaPipeline(DiffusionPipeline):
negative_prompt: str = "",
num_images_per_prompt: int = 1,
device: Optional[torch.device] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_attention_mask: Optional[torch.FloatTensor] = None,
negative_prompt_attention_mask: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
prompt_attention_mask: Optional[torch.Tensor] = None,
negative_prompt_attention_mask: Optional[torch.Tensor] = None,
clean_caption: bool = False,
max_sequence_length: int = 120,
**kwargs,
......@@ -320,10 +320,10 @@ class PixArtAlphaPipeline(DiffusionPipeline):
number of images that should be generated per prompt
device: (`torch.device`, *optional*):
torch device to place the resulting embeddings on
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
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. For PixArt-Alpha, it's should be the embeddings of the ""
string.
clean_caption (`bool`, defaults to `False`):
......@@ -694,14 +694,14 @@ class PixArtAlphaPipeline(DiffusionPipeline):
width: Optional[int] = None,
eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_attention_mask: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_attention_mask: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
prompt_attention_mask: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_attention_mask: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil",
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,
clean_caption: bool = True,
use_resolution_binning: bool = True,
......@@ -748,18 +748,18 @@ class PixArtAlphaPipeline(DiffusionPipeline):
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)
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
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`.
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
provided, text embeddings will be generated from `prompt` input argument.
prompt_attention_mask (`torch.FloatTensor`, *optional*): Pre-generated attention mask for text embeddings.
negative_prompt_embeds (`torch.FloatTensor`, *optional*):
prompt_attention_mask (`torch.Tensor`, *optional*): Pre-generated attention mask for text embeddings.
negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. For PixArt-Alpha this negative prompt should be "". If not
provided, negative_prompt_embeds will be generated from `negative_prompt` input argument.
negative_prompt_attention_mask (`torch.FloatTensor`, *optional*):
negative_prompt_attention_mask (`torch.Tensor`, *optional*):
Pre-generated attention mask for negative text embeddings.
output_type (`str`, *optional*, defaults to `"pil"`):
The output format of the generate image. Choose between
......@@ -768,7 +768,7 @@ class PixArtAlphaPipeline(DiffusionPipeline):
Whether or not to return a [`~pipelines.stable_diffusion.IFPipelineOutput`] instead of a plain tuple.
callback (`Callable`, *optional*):
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):
The frequency at which the `callback` function will be called. If not specified, the callback will be
called at every step.
......
src/diffusers/pipelines/pixart_alpha/pipeline_pixart_sigma.py
View file @ be4afa0b
......@@ -222,10 +222,10 @@ class PixArtSigmaPipeline(DiffusionPipeline):
negative_prompt: str = "",
num_images_per_prompt: int = 1,
device: Optional[torch.device] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_attention_mask: Optional[torch.FloatTensor] = None,
negative_prompt_attention_mask: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
prompt_attention_mask: Optional[torch.Tensor] = None,
negative_prompt_attention_mask: Optional[torch.Tensor] = None,
clean_caption: bool = False,
max_sequence_length: int = 120,
**kwargs,
......@@ -246,10 +246,10 @@ class PixArtSigmaPipeline(DiffusionPipeline):
number of images that should be generated per prompt
device: (`torch.device`, *optional*):
torch device to place the resulting embeddings on
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
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. For PixArt-Alpha, it's should be the embeddings of the ""
string.
clean_caption (`bool`, defaults to `False`):
......@@ -621,14 +621,14 @@ class PixArtSigmaPipeline(DiffusionPipeline):
width: Optional[int] = None,
eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_attention_mask: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_attention_mask: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
prompt_attention_mask: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_attention_mask: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil",
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,
clean_caption: bool = True,
use_resolution_binning: bool = True,
......@@ -675,18 +675,18 @@ class PixArtSigmaPipeline(DiffusionPipeline):
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)
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
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`.
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
provided, text embeddings will be generated from `prompt` input argument.
prompt_attention_mask (`torch.FloatTensor`, *optional*): Pre-generated attention mask for text embeddings.
negative_prompt_embeds (`torch.FloatTensor`, *optional*):
prompt_attention_mask (`torch.Tensor`, *optional*): Pre-generated attention mask for text embeddings.
negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated negative text embeddings. For PixArt-Sigma this negative prompt should be "". If not
provided, negative_prompt_embeds will be generated from `negative_prompt` input argument.
negative_prompt_attention_mask (`torch.FloatTensor`, *optional*):
negative_prompt_attention_mask (`torch.Tensor`, *optional*):
Pre-generated attention mask for negative text embeddings.
output_type (`str`, *optional*, defaults to `"pil"`):
The output format of the generate image. Choose between
......@@ -695,7 +695,7 @@ class PixArtSigmaPipeline(DiffusionPipeline):
Whether or not to return a [`~pipelines.stable_diffusion.IFPipelineOutput`] instead of a plain tuple.
callback (`Callable`, *optional*):
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):
The frequency at which the `callback` function will be called. If not specified, the callback will be
called at every step.
......
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