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
Hide whitespace changes
Inline Side-by-side
Showing with 324 additions and 324 deletions
src/diffusers/pipelines/stable_diffusion_diffedit/pipeline_stable_diffusion_diffedit.py
View file @ be4afa0b
......@@ -53,7 +53,7 @@ class DiffEditInversionPipelineOutput(BaseOutput):
Output class for Stable Diffusion pipelines.
Args:
latents (`torch.FloatTensor`)
latents (`torch.Tensor`)
inverted latents tensor
images (`List[PIL.Image.Image]` or `np.ndarray`)
List of denoised PIL images of length `num_timesteps * batch_size` or numpy array of shape `(num_timesteps,
......@@ -61,7 +61,7 @@ class DiffEditInversionPipelineOutput(BaseOutput):
diffusion pipeline.
"""
latents: torch.FloatTensor
latents: torch.Tensor
images: Union[List[PIL.Image.Image], np.ndarray]
......@@ -381,8 +381,8 @@ class StableDiffusionDiffEditPipeline(
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,
**kwargs,
):
......@@ -414,8 +414,8 @@ class StableDiffusionDiffEditPipeline(
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,
):
......@@ -435,10 +435,10 @@ class StableDiffusionDiffEditPipeline(
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.
......@@ -831,15 +831,15 @@ class StableDiffusionDiffEditPipeline(
@replace_example_docstring(EXAMPLE_DOC_STRING)
def generate_mask(
self,
image: Union[torch.FloatTensor, PIL.Image.Image] = None,
image: Union[torch.Tensor, PIL.Image.Image] = None,
target_prompt: Optional[Union[str, List[str]]] = None,
target_negative_prompt: Optional[Union[str, List[str]]] = None,
target_prompt_embeds: Optional[torch.FloatTensor] = None,
target_negative_prompt_embeds: Optional[torch.FloatTensor] = None,
target_prompt_embeds: Optional[torch.Tensor] = None,
target_negative_prompt_embeds: Optional[torch.Tensor] = None,
source_prompt: Optional[Union[str, List[str]]] = None,
source_negative_prompt: Optional[Union[str, List[str]]] = None,
source_prompt_embeds: Optional[torch.FloatTensor] = None,
source_negative_prompt_embeds: Optional[torch.FloatTensor] = None,
source_prompt_embeds: Optional[torch.Tensor] = None,
source_negative_prompt_embeds: Optional[torch.Tensor] = None,
num_maps_per_mask: Optional[int] = 10,
mask_encode_strength: Optional[float] = 0.5,
mask_thresholding_ratio: Optional[float] = 3.0,
......@@ -861,10 +861,10 @@ class StableDiffusionDiffEditPipeline(
target_negative_prompt (`str` or `List[str]`, *optional*):
The prompt or prompts to guide what to not include in image generation. If not defined, you need to
pass `negative_prompt_embeds` instead. Ignored when not using guidance (`guidance_scale < 1`).
target_prompt_embeds (`torch.FloatTensor`, *optional*):
target_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.
target_negative_prompt_embeds (`torch.FloatTensor`, *optional*):
target_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.
source_prompt (`str` or `List[str]`, *optional*):
......@@ -873,11 +873,11 @@ class StableDiffusionDiffEditPipeline(
source_negative_prompt (`str` or `List[str]`, *optional*):
The prompt or prompts to guide semantic mask generation away from using DiffEdit. If not defined, you
need to pass `source_negative_prompt_embeds` or `source_image` instead.
source_prompt_embeds (`torch.FloatTensor`, *optional*):
source_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings to guide the semantic mask generation. Can be used to easily tweak text
inputs (prompt weighting). If not provided, text embeddings are generated from `source_prompt` input
argument.
source_negative_prompt_embeds (`torch.FloatTensor`, *optional*):
source_negative_prompt_embeds (`torch.Tensor`, *optional*):
Pre-generated text embeddings to negatively guide the semantic mask generation. Can be used to easily
tweak text inputs (prompt weighting). If not provided, text embeddings are generated from
`source_negative_prompt` input argument.
......@@ -1051,18 +1051,18 @@ class StableDiffusionDiffEditPipeline(
def invert(
self,
prompt: Optional[Union[str, List[str]]] = None,
image: Union[torch.FloatTensor, PIL.Image.Image] = None,
image: Union[torch.Tensor, PIL.Image.Image] = None,
num_inference_steps: int = 50,
inpaint_strength: float = 0.8,
guidance_scale: float = 7.5,
negative_prompt: Optional[Union[str, List[str]]] = None,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = 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,
decode_latents: bool = False,
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: Optional[int] = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
lambda_auto_corr: float = 20.0,
......@@ -1095,10 +1095,10 @@ class StableDiffusionDiffEditPipeline(
generator (`torch.Generator`, *optional*):
A [`torch.Generator`](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 (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.
decode_latents (`bool`, *optional*, defaults to `False`):
......@@ -1111,7 +1111,7 @@ class StableDiffusionDiffEditPipeline(
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.
......@@ -1289,8 +1289,8 @@ class StableDiffusionDiffEditPipeline(
def __call__(
self,
prompt: Optional[Union[str, List[str]]] = None,
mask_image: Union[torch.FloatTensor, PIL.Image.Image] = None,
image_latents: Union[torch.FloatTensor, PIL.Image.Image] = None,
mask_image: Union[torch.Tensor, PIL.Image.Image] = None,
image_latents: Union[torch.Tensor, PIL.Image.Image] = None,
inpaint_strength: Optional[float] = 0.8,
num_inference_steps: int = 50,
guidance_scale: float = 7.5,
......@@ -1298,12 +1298,12 @@ class StableDiffusionDiffEditPipeline(
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,
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,
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,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
clip_skip: int = None,
......@@ -1319,7 +1319,7 @@ class StableDiffusionDiffEditPipeline(
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 expected shape would be `(B, 1, H, W)`.
image_latents (`PIL.Image.Image` or `torch.FloatTensor`):
image_latents (`PIL.Image.Image` or `torch.Tensor`):
Partially noised image latents from the inversion process to be used as inputs for image generation.
inpaint_strength (`float`, *optional*, defaults to 0.8):
Indicates extent to inpaint the masked area. Must be between 0 and 1. When `inpaint_strength` is 1, the
......@@ -1343,14 +1343,14 @@ class StableDiffusionDiffEditPipeline(
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`.
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.
output_type (`str`, *optional*, defaults to `"pil"`):
......@@ -1360,7 +1360,7 @@ class StableDiffusionDiffEditPipeline(
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/stable_diffusion_gligen/pipeline_stable_diffusion_gligen.py
View file @ be4afa0b
......@@ -180,8 +180,8 @@ class StableDiffusionGLIGENPipeline(DiffusionPipeline, StableDiffusionMixin):
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,
**kwargs,
):
......@@ -213,8 +213,8 @@ class StableDiffusionGLIGENPipeline(DiffusionPipeline, StableDiffusionMixin):
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,
):
......@@ -234,10 +234,10 @@ class StableDiffusionGLIGENPipeline(DiffusionPipeline, StableDiffusionMixin):
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.
......@@ -541,12 +541,12 @@ class StableDiffusionGLIGENPipeline(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,
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,
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,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
clip_skip: Optional[int] = None,
......@@ -592,14 +592,14 @@ class StableDiffusionGLIGENPipeline(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.
output_type (`str`, *optional*, defaults to `"pil"`):
......@@ -609,7 +609,7 @@ class StableDiffusionGLIGENPipeline(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/stable_diffusion_gligen/pipeline_stable_diffusion_gligen_text_image.py
View file @ be4afa0b
......@@ -238,8 +238,8 @@ class StableDiffusionGLIGENTextImagePipeline(DiffusionPipeline, StableDiffusionM
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,
):
......@@ -259,10 +259,10 @@ class StableDiffusionGLIGENTextImagePipeline(DiffusionPipeline, StableDiffusionM
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.
......@@ -705,12 +705,12 @@ class StableDiffusionGLIGENTextImagePipeline(DiffusionPipeline, StableDiffusionM
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,
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,
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,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
gligen_normalize_constant: float = 28.7,
......@@ -764,14 +764,14 @@ class StableDiffusionGLIGENTextImagePipeline(DiffusionPipeline, StableDiffusionM
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.
output_type (`str`, *optional*, defaults to `"pil"`):
......@@ -781,7 +781,7 @@ class StableDiffusionGLIGENTextImagePipeline(DiffusionPipeline, StableDiffusionM
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/stable_diffusion_k_diffusion/pipeline_stable_diffusion_k_diffusion.py
View file @ be4afa0b
......@@ -154,8 +154,8 @@ class StableDiffusionKDiffusionPipeline(
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,
**kwargs,
):
......@@ -187,8 +187,8 @@ class StableDiffusionKDiffusionPipeline(
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,
):
......@@ -208,10 +208,10 @@ class StableDiffusionKDiffusionPipeline(
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.
......@@ -469,12 +469,12 @@ class StableDiffusionKDiffusionPipeline(
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,
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,
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,
use_karras_sigmas: Optional[bool] = False,
noise_sampler_seed: Optional[int] = None,
......@@ -512,14 +512,14 @@ class StableDiffusionKDiffusionPipeline(
generator (`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.
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.
......@@ -531,7 +531,7 @@ class StableDiffusionKDiffusionPipeline(
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/stable_diffusion_k_diffusion/pipeline_stable_diffusion_xl_k_diffusion.py
View file @ be4afa0b
......@@ -207,10 +207,10 @@ class StableDiffusionXLKDiffusionPipeline(
do_classifier_free_guidance: bool = True,
negative_prompt: Optional[str] = None,
negative_prompt_2: Optional[str] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None,
):
......@@ -236,17 +236,17 @@ class StableDiffusionXLKDiffusionPipeline(
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
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.
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.
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
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument.
......@@ -584,11 +584,11 @@ class StableDiffusionXLKDiffusionPipeline(
negative_prompt_2: Optional[Union[str, List[str]]] = None,
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,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil",
return_dict: bool = True,
original_size: Optional[Tuple[int, int]] = None,
......@@ -642,21 +642,21 @@ class StableDiffusionXLKDiffusionPipeline(
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.
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.
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.
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
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument.
......
src/diffusers/pipelines/stable_diffusion_ldm3d/pipeline_stable_diffusion_ldm3d.py
View file @ be4afa0b
......@@ -254,8 +254,8 @@ class StableDiffusionLDM3DPipeline(
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,
**kwargs,
):
......@@ -287,8 +287,8 @@ class StableDiffusionLDM3DPipeline(
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,
):
......@@ -308,10 +308,10 @@ class StableDiffusionLDM3DPipeline(
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.
......@@ -666,7 +666,7 @@ class StableDiffusionLDM3DPipeline(
# 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
......@@ -679,7 +679,7 @@ class StableDiffusionLDM3DPipeline(
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
......@@ -740,11 +740,11 @@ class StableDiffusionLDM3DPipeline(
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,
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,
output_type: Optional[str] = "pil",
return_dict: bool = True,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
......@@ -789,19 +789,19 @@ class StableDiffusionLDM3DPipeline(
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.
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/stable_diffusion_panorama/pipeline_stable_diffusion_panorama.py
View file @ be4afa0b
......@@ -226,8 +226,8 @@ class StableDiffusionPanoramaPipeline(
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,
**kwargs,
):
......@@ -259,8 +259,8 @@ class StableDiffusionPanoramaPipeline(
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,
):
......@@ -280,10 +280,10 @@ class StableDiffusionPanoramaPipeline(
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.
......@@ -677,7 +677,7 @@ class StableDiffusionPanoramaPipeline(
# 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
......@@ -690,7 +690,7 @@ class StableDiffusionPanoramaPipeline(
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
......@@ -790,11 +790,11 @@ class StableDiffusionPanoramaPipeline(
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,
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,
output_type: Optional[str] = "pil",
return_dict: bool = True,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
......@@ -839,19 +839,19 @@ class StableDiffusionPanoramaPipeline(
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.
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/stable_diffusion_safe/pipeline_stable_diffusion_safe.py
View file @ be4afa0b
......@@ -516,11 +516,11 @@ class StableDiffusionPipelineSafe(DiffusionPipeline, StableDiffusionMixin, IPAda
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,
ip_adapter_image: Optional[PipelineImageInput] = 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,
sld_guidance_scale: Optional[float] = 1000,
sld_warmup_steps: Optional[int] = 10,
......@@ -555,7 +555,7 @@ class StableDiffusionPipelineSafe(DiffusionPipeline, StableDiffusionMixin, IPAda
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`.
......@@ -568,7 +568,7 @@ class StableDiffusionPipelineSafe(DiffusionPipeline, StableDiffusionMixin, IPAda
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/stable_diffusion_safe/safety_checker.py
View file @ be4afa0b
......@@ -85,7 +85,7 @@ class SafeStableDiffusionSafetyChecker(PreTrainedModel):
return images, has_nsfw_concepts
@torch.no_grad()
def forward_onnx(self, clip_input: torch.FloatTensor, images: torch.FloatTensor):
def forward_onnx(self, clip_input: torch.Tensor, images: torch.Tensor):
pooled_output = self.vision_model(clip_input)[1] # pooled_output
image_embeds = self.visual_projection(pooled_output)
......
src/diffusers/pipelines/stable_diffusion_sag/pipeline_stable_diffusion_sag.py
View file @ be4afa0b
......@@ -169,8 +169,8 @@ class StableDiffusionSAGPipeline(DiffusionPipeline, StableDiffusionMixin, Textua
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,
**kwargs,
):
......@@ -202,8 +202,8 @@ class StableDiffusionSAGPipeline(DiffusionPipeline, StableDiffusionMixin, Textua
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,
):
......@@ -223,10 +223,10 @@ class StableDiffusionSAGPipeline(DiffusionPipeline, StableDiffusionMixin, Textua
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.
......@@ -570,14 +570,14 @@ class StableDiffusionSAGPipeline(DiffusionPipeline, StableDiffusionMixin, Textua
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,
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,
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: Optional[int] = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
clip_skip: Optional[int] = None,
......@@ -611,19 +611,19 @@ class StableDiffusionSAGPipeline(DiffusionPipeline, StableDiffusionMixin, Textua
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.
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 provided, embeddings are computed from the
`ip_adapter_image` input argument.
output_type (`str`, *optional*, defaults to `"pil"`):
......@@ -633,7 +633,7 @@ class StableDiffusionSAGPipeline(DiffusionPipeline, StableDiffusionMixin, Textua
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/stable_diffusion_xl/pipeline_stable_diffusion_xl.py
View file @ be4afa0b
......@@ -282,10 +282,10 @@ class StableDiffusionXLPipeline(
do_classifier_free_guidance: bool = True,
negative_prompt: Optional[str] = None,
negative_prompt_2: Optional[str] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None,
):
......@@ -311,17 +311,17 @@ class StableDiffusionXLPipeline(
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
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.
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.
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
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument.
......@@ -762,7 +762,7 @@ class StableDiffusionXLPipeline(
# 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
......@@ -775,7 +775,7 @@ class StableDiffusionXLPipeline(
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
......@@ -843,13 +843,13 @@ class StableDiffusionXLPipeline(
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,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_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,
......@@ -924,26 +924,26 @@ class StableDiffusionXLPipeline(
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.
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.
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.
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
weighting. If not provided, pooled negative_prompt_embeds will be generated from `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
......
src/diffusers/pipelines/stable_diffusion_xl/pipeline_stable_diffusion_xl_img2img.py
View file @ be4afa0b
......@@ -303,10 +303,10 @@ class StableDiffusionXLImg2ImgPipeline(
do_classifier_free_guidance: bool = True,
negative_prompt: Optional[str] = None,
negative_prompt_2: Optional[str] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None,
):
......@@ -332,17 +332,17 @@ class StableDiffusionXLImg2ImgPipeline(
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
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.
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.
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
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument.
......@@ -902,7 +902,7 @@ class StableDiffusionXLImg2ImgPipeline(
# 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
......@@ -915,7 +915,7 @@ class StableDiffusionXLImg2ImgPipeline(
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
......@@ -988,13 +988,13 @@ class StableDiffusionXLImg2ImgPipeline(
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,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_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,
......@@ -1022,7 +1022,7 @@ class StableDiffusionXLImg2ImgPipeline(
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
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.
strength (`float`, *optional*, defaults to 0.3):
Conceptually, indicates how much to transform the reference `image`. Must be between 0 and 1. `image`
......@@ -1078,26 +1078,26 @@ class StableDiffusionXLImg2ImgPipeline(
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.
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.
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.
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
weighting. If not provided, pooled negative_prompt_embeds will be generated from `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
......
src/diffusers/pipelines/stable_diffusion_xl/pipeline_stable_diffusion_xl_inpaint.py
View file @ be4afa0b
......@@ -531,10 +531,10 @@ class StableDiffusionXLInpaintPipeline(
do_classifier_free_guidance: bool = True,
negative_prompt: Optional[str] = None,
negative_prompt_2: Optional[str] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None,
):
......@@ -560,17 +560,17 @@ class StableDiffusionXLInpaintPipeline(
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
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.
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.
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
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument.
......@@ -1132,7 +1132,7 @@ class StableDiffusionXLInpaintPipeline(
# 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
......@@ -1145,7 +1145,7 @@ class StableDiffusionXLInpaintPipeline(
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
......@@ -1207,7 +1207,7 @@ class StableDiffusionXLInpaintPipeline(
prompt_2: Optional[Union[str, List[str]]] = None,
image: PipelineImageInput = None,
mask_image: PipelineImageInput = None,
masked_image_latents: torch.FloatTensor = None,
masked_image_latents: torch.Tensor = None,
height: Optional[int] = None,
width: Optional[int] = None,
padding_mask_crop: Optional[int] = None,
......@@ -1223,13 +1223,13 @@ class StableDiffusionXLInpaintPipeline(
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,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_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,
......@@ -1329,22 +1329,22 @@ class StableDiffusionXLInpaintPipeline(
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
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.
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.
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
weighting. If not provided, pooled negative_prompt_embeds will be generated from `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
......@@ -1357,7 +1357,7 @@ class StableDiffusionXLInpaintPipeline(
generator (`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/stable_diffusion_xl/pipeline_stable_diffusion_xl_instruct_pix2pix.py
View file @ be4afa0b
......@@ -222,10 +222,10 @@ class StableDiffusionXLInstructPix2PixPipeline(
do_classifier_free_guidance: bool = True,
negative_prompt: Optional[str] = None,
negative_prompt_2: Optional[str] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None,
):
r"""
......@@ -250,17 +250,17 @@ class StableDiffusionXLInstructPix2PixPipeline(
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
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.
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.
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
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument.
......@@ -622,14 +622,14 @@ class StableDiffusionXLInstructPix2PixPipeline(
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,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: 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,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
guidance_rescale: float = 0.0,
......@@ -647,7 +647,7 @@ class StableDiffusionXLInstructPix2PixPipeline(
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
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.
height (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
The height in pixels of the generated image.
......@@ -689,21 +689,21 @@ class StableDiffusionXLInstructPix2PixPipeline(
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.
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.
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.
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
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument.
......@@ -715,7 +715,7 @@ class StableDiffusionXLInstructPix2PixPipeline(
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/stable_diffusion_xl/watermark.py
View file @ be4afa0b
......@@ -21,7 +21,7 @@ class StableDiffusionXLWatermarker:
self.encoder.set_watermark("bits", self.watermark)
def apply_watermark(self, images: torch.FloatTensor):
def apply_watermark(self, images: torch.Tensor):
# can't encode images that are smaller than 256
if images.shape[-1] < 256:
return images
......
src/diffusers/pipelines/stable_video_diffusion/pipeline_stable_video_diffusion.py
View file @ be4afa0b
......@@ -128,12 +128,12 @@ class StableVideoDiffusionPipelineOutput(BaseOutput):
Output class for Stable Video Diffusion pipeline.
Args:
frames (`[List[List[PIL.Image.Image]]`, `np.ndarray`, `torch.FloatTensor`]):
frames (`[List[List[PIL.Image.Image]]`, `np.ndarray`, `torch.Tensor`]):
List of denoised PIL images of length `batch_size` or numpy array or torch tensor of shape `(batch_size,
num_frames, height, width, num_channels)`.
"""
frames: Union[List[List[PIL.Image.Image]], np.ndarray, torch.FloatTensor]
frames: Union[List[List[PIL.Image.Image]], np.ndarray, torch.Tensor]
class StableVideoDiffusionPipeline(DiffusionPipeline):
......@@ -186,7 +186,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
device: Union[str, torch.device],
num_videos_per_prompt: int,
do_classifier_free_guidance: bool,
) -> torch.FloatTensor:
) -> torch.Tensor:
dtype = next(self.image_encoder.parameters()).dtype
if not isinstance(image, torch.Tensor):
......@@ -279,7 +279,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
return add_time_ids
def decode_latents(self, latents: torch.FloatTensor, num_frames: int, decode_chunk_size: int = 14):
def decode_latents(self, latents: torch.Tensor, num_frames: int, decode_chunk_size: int = 14):
# [batch, frames, channels, height, width] -> [batch*frames, channels, height, width]
latents = latents.flatten(0, 1)
......@@ -315,7 +315,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
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)}"
)
......@@ -332,7 +332,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
dtype: torch.dtype,
device: Union[str, torch.device],
generator: torch.Generator,
latents: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
):
shape = (
batch_size,
......@@ -377,7 +377,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
@replace_example_docstring(EXAMPLE_DOC_STRING)
def __call__(
self,
image: Union[PIL.Image.Image, List[PIL.Image.Image], torch.FloatTensor],
image: Union[PIL.Image.Image, List[PIL.Image.Image], torch.Tensor],
height: int = 576,
width: int = 1024,
num_frames: Optional[int] = None,
......@@ -391,7 +391,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
decode_chunk_size: Optional[int] = None,
num_videos_per_prompt: Optional[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_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
callback_on_step_end_tensor_inputs: List[str] = ["latents"],
......@@ -401,7 +401,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
The call function to the pipeline for generation.
Args:
image (`PIL.Image.Image` or `List[PIL.Image.Image]` or `torch.FloatTensor`):
image (`PIL.Image.Image` or `List[PIL.Image.Image]` or `torch.Tensor`):
Image(s) to guide image generation. If you provide a tensor, the expected value range is between `[0,
1]`.
height (`int`, *optional*, defaults to `self.unet.config.sample_size * self.vae_scale_factor`):
......@@ -440,7 +440,7 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
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`.
......@@ -465,8 +465,8 @@ class StableVideoDiffusionPipeline(DiffusionPipeline):
Returns:
[`~pipelines.stable_diffusion.StableVideoDiffusionPipelineOutput`] or `tuple`:
If `return_dict` is `True`, [`~pipelines.stable_diffusion.StableVideoDiffusionPipelineOutput`] is
returned, otherwise a `tuple` of (`List[List[PIL.Image.Image]]` or `np.ndarray` or `torch.FloatTensor`)
is returned.
returned, otherwise a `tuple` of (`List[List[PIL.Image.Image]]` or `np.ndarray` or `torch.Tensor`) is
returned.
"""
# 0. Default height and width to unet
height = height or self.unet.config.sample_size * self.vae_scale_factor
......
src/diffusers/pipelines/t2i_adapter/pipeline_stable_diffusion_adapter.py
View file @ be4afa0b
......@@ -271,8 +271,8 @@ class StableDiffusionAdapterPipeline(DiffusionPipeline, StableDiffusionMixin):
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,
**kwargs,
):
......@@ -304,8 +304,8 @@ class StableDiffusionAdapterPipeline(DiffusionPipeline, StableDiffusionMixin):
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,
):
......@@ -325,10 +325,10 @@ class StableDiffusionAdapterPipeline(DiffusionPipeline, StableDiffusionMixin):
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.
......@@ -635,7 +635,7 @@ class StableDiffusionAdapterPipeline(DiffusionPipeline, StableDiffusionMixin):
# 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
......@@ -648,7 +648,7 @@ class StableDiffusionAdapterPipeline(DiffusionPipeline, StableDiffusionMixin):
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
......@@ -690,12 +690,12 @@ class StableDiffusionAdapterPipeline(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,
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,
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,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
adapter_conditioning_scale: Union[float, List[float]] = 1.0,
......@@ -708,9 +708,9 @@ class StableDiffusionAdapterPipeline(DiffusionPipeline, StableDiffusionMixin):
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`, `List[torch.FloatTensor]` or `List[PIL.Image.Image]` or `List[List[PIL.Image.Image]]`):
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
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.
height (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
The height in pixels of the generated image.
......@@ -745,14 +745,14 @@ class StableDiffusionAdapterPipeline(DiffusionPipeline, StableDiffusionMixin):
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.
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.
......@@ -764,7 +764,7 @@ class StableDiffusionAdapterPipeline(DiffusionPipeline, StableDiffusionMixin):
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/t2i_adapter/pipeline_stable_diffusion_xl_adapter.py
View file @ be4afa0b
......@@ -296,10 +296,10 @@ class StableDiffusionXLAdapterPipeline(
do_classifier_free_guidance: bool = True,
negative_prompt: Optional[str] = None,
negative_prompt_2: Optional[str] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None,
):
......@@ -325,17 +325,17 @@ class StableDiffusionXLAdapterPipeline(
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
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.
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.
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
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument.
......@@ -806,7 +806,7 @@ class StableDiffusionXLAdapterPipeline(
# 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
......@@ -819,7 +819,7 @@ class StableDiffusionXLAdapterPipeline(
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
......@@ -864,16 +864,16 @@ class StableDiffusionXLAdapterPipeline(
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,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
latents: Optional[torch.Tensor] = None,
prompt_embeds: Optional[torch.Tensor] = None,
negative_prompt_embeds: Optional[torch.Tensor] = None,
pooled_prompt_embeds: Optional[torch.Tensor] = None,
negative_pooled_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,
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None,
callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
callback_steps: int = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
guidance_rescale: float = 0.0,
......@@ -897,9 +897,9 @@ class StableDiffusionXLAdapterPipeline(
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
used in both text-encoders
image (`torch.FloatTensor`, `PIL.Image.Image`, `List[torch.FloatTensor]` or `List[PIL.Image.Image]` or `List[List[PIL.Image.Image]]`):
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
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.
height (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
The height in pixels of the generated image. Anything below 512 pixels won't work well for
......@@ -948,26 +948,26 @@ class StableDiffusionXLAdapterPipeline(
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.
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.
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.
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
weighting. If not provided, pooled negative_prompt_embeds will be generated from `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
......@@ -980,7 +980,7 @@ class StableDiffusionXLAdapterPipeline(
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/text_to_video_synthesis/pipeline_text_to_video_synth.py
View file @ be4afa0b
......@@ -114,8 +114,8 @@ class TextToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInve
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,
**kwargs,
):
......@@ -147,8 +147,8 @@ class TextToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInve
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,
):
......@@ -168,10 +168,10 @@ class TextToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInve
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.
......@@ -442,12 +442,12 @@ class TextToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInve
negative_prompt: Optional[Union[str, List[str]]] = 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,
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,
output_type: Optional[str] = "np",
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,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
clip_skip: Optional[int] = None,
......@@ -482,25 +482,25 @@ class TextToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInve
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.
output_type (`str`, *optional*, defaults to `"np"`):
The output format of the generated video. Choose between `torch.FloatTensor` or `np.array`.
The output format of the generated video. Choose between `torch.Tensor` 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.
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/text_to_video_synthesis/pipeline_text_to_video_synth_img2img.py
View file @ be4afa0b
......@@ -149,8 +149,8 @@ class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInv
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,
**kwargs,
):
......@@ -182,8 +182,8 @@ class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInv
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,
):
......@@ -203,10 +203,10 @@ class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInv
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.
......@@ -499,19 +499,19 @@ class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInv
def __call__(
self,
prompt: Union[str, List[str]] = None,
video: Union[List[np.ndarray], torch.FloatTensor] = None,
video: Union[List[np.ndarray], torch.Tensor] = None,
strength: float = 0.6,
num_inference_steps: int = 50,
guidance_scale: float = 15.0,
negative_prompt: Optional[Union[str, List[str]]] = 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,
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,
output_type: Optional[str] = "np",
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,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
clip_skip: Optional[int] = None,
......@@ -522,7 +522,7 @@ class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInv
Args:
prompt (`str` or `List[str]`, *optional*):
The prompt or prompts to guide image generation. If not defined, you need to pass `prompt_embeds`.
video (`List[np.ndarray]` or `torch.FloatTensor`):
video (`List[np.ndarray]` or `torch.Tensor`):
`video` frames or tensor representing a video batch to be used as the starting point for the process.
Can also accept video latents as `image`, if passing latents directly, it will not be encoded again.
strength (`float`, *optional*, defaults to 0.8):
......@@ -546,25 +546,25 @@ class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInv
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.
output_type (`str`, *optional*, defaults to `"np"`):
The output format of the generated video. Choose between `torch.FloatTensor` or `np.array`.
The output format of the generated video. Choose between `torch.Tensor` 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.
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.
......
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