Update-DeepFloyd-IF-Pipelines-Docstrings (#5304)

* added TODOs

* Enhanced and reformatted the docstrings of IFPipeline methods.

* Enhanced and fixed the docstrings of IFImg2ImgSuperResolutionPipeline methods.

* Enhanced and fixed the docstrings of IFImg2ImgPipeline methods.

* Enhanced and fixed the docstrings of IFInpaintingSuperResolutionPipeline methods.

* Enhanced and fixed the docstrings of IFInpaintingPipeline  methods.

* Enhanced and fixed the docstrings of IFSuperResolutionPipeline methods.

* Update src/diffusers/pipelines/deepfloyd_if/pipeline_if.py
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update src/diffusers/pipelines/deepfloyd_if/pipeline_if_img2img.py
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update src/diffusers/pipelines/deepfloyd_if/pipeline_if_img2img_superresolution.py
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update src/diffusers/pipelines/deepfloyd_if/pipeline_if_inpainting.py
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update src/diffusers/pipelines/deepfloyd_if/pipeline_if_superresolution.py
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update src/diffusers/pipelines/deepfloyd_if/pipeline_if_inpainting_superresolution.py
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* remove redundant code

* fix code style

* revert the ordering to not break backwards compatibility

---------
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

src/diffusers/pipelines/deepfloyd_if/pipeline_if.py

@@ -161,11 +161,11 @@ class IFPipeline(DiffusionPipeline, LoraLoaderMixin):
     @torch.no_grad()
     def encode_prompt(
         self,
-        prompt,
-        do_classifier_free_guidance=True,
-        num_images_per_prompt=1,
-        device=None,
-        negative_prompt=None,
+        prompt: Union[str, List[str]],
+        do_classifier_free_guidance: bool = True,
+        num_images_per_prompt: int = 1,
+        device: Optional[torch.device] = None,
+        negative_prompt: Optional[Union[str, List[str]]] = None,
         prompt_embeds: Optional[torch.FloatTensor] = None,
         negative_prompt_embeds: Optional[torch.FloatTensor] = None,
         clean_caption: bool = False,
@@ -176,12 +176,12 @@ class IFPipeline(DiffusionPipeline, LoraLoaderMixin):
         Args:
             prompt (`str` or `List[str]`, *optional*):
                 prompt to be encoded
-            device: (`torch.device`, *optional*):
-                torch device to place the resulting embeddings on
-            num_images_per_prompt (`int`, *optional*, defaults to 1):
-                number of images that should be generated per prompt
             do_classifier_free_guidance (`bool`, *optional*, defaults to `True`):
                 whether to use classifier free guidance or not
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                number of images that should be generated per prompt
+            device: (`torch.device`, *optional*):
+                torch device to place the resulting embeddings on
             negative_prompt (`str` or `List[str]`, *optional*):
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds`. instead. If not defined, one has to pass `negative_prompt_embeds`. instead.
@@ -193,6 +193,8 @@ class IFPipeline(DiffusionPipeline, LoraLoaderMixin):
                 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.
+            clean_caption (bool, defaults to `False`):
+                If `True`, the function will preprocess and clean the provided caption before encoding.
         """
         if prompt is not None and negative_prompt is not None:
             if type(prompt) is not type(negative_prompt):
@@ -568,13 +570,13 @@ class IFPipeline(DiffusionPipeline, LoraLoaderMixin):
             prompt (`str` or `List[str]`, *optional*):
                 The prompt or prompts to guide the image generation. If not defined, one has to pass `prompt_embeds`.
                 instead.
-            num_inference_steps (`int`, *optional*, defaults to 50):
+            num_inference_steps (`int`, *optional*, defaults to 100):
                 The number of denoising steps. More denoising steps usually lead to a higher quality image at the
                 expense of slower inference.
             timesteps (`List[int]`, *optional*):
                 Custom timesteps to use for the denoising process. If not defined, equal spaced `num_inference_steps`
                 timesteps are used. Must be in descending order.
-            guidance_scale (`float`, *optional*, defaults to 7.5):
+            guidance_scale (`float`, *optional*, defaults to 7.0):
                 Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
                 `guidance_scale` is defined as `w` of equation 2. of [Imagen
                 Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >

src/diffusers/pipelines/deepfloyd_if/pipeline_if_img2img.py

@@ -184,14 +184,13 @@ class IFImg2ImgPipeline(DiffusionPipeline, LoraLoaderMixin):
         self.final_offload_hook = None
 
     @torch.no_grad()
-    # Copied from diffusers.pipelines.deepfloyd_if.pipeline_if.IFPipeline.encode_prompt
     def encode_prompt(
         self,
-        prompt,
-        do_classifier_free_guidance=True,
-        num_images_per_prompt=1,
-        device=None,
-        negative_prompt=None,
+        prompt: Union[str, List[str]],
+        do_classifier_free_guidance: bool = True,
+        num_images_per_prompt: int = 1,
+        device: Optional[torch.device] = None,
+        negative_prompt: Optional[Union[str, List[str]]] = None,
         prompt_embeds: Optional[torch.FloatTensor] = None,
         negative_prompt_embeds: Optional[torch.FloatTensor] = None,
         clean_caption: bool = False,
@@ -202,12 +201,12 @@ class IFImg2ImgPipeline(DiffusionPipeline, LoraLoaderMixin):
         Args:
             prompt (`str` or `List[str]`, *optional*):
                 prompt to be encoded
-            device: (`torch.device`, *optional*):
-                torch device to place the resulting embeddings on
-            num_images_per_prompt (`int`, *optional*, defaults to 1):
-                number of images that should be generated per prompt
             do_classifier_free_guidance (`bool`, *optional*, defaults to `True`):
                 whether to use classifier free guidance or not
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                number of images that should be generated per prompt
+            device: (`torch.device`, *optional*):
+                torch device to place the resulting embeddings on
             negative_prompt (`str` or `List[str]`, *optional*):
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds`. instead. If not defined, one has to pass `negative_prompt_embeds`. instead.
@@ -219,6 +218,8 @@ class IFImg2ImgPipeline(DiffusionPipeline, LoraLoaderMixin):
                 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.
+            clean_caption (bool, defaults to `False`):
+                If `True`, the function will preprocess and clean the provided caption before encoding.
         """
         if prompt is not None and negative_prompt is not None:
             if type(prompt) is not type(negative_prompt):
@@ -686,19 +687,19 @@ class IFImg2ImgPipeline(DiffusionPipeline, LoraLoaderMixin):
             image (`torch.FloatTensor` or `PIL.Image.Image`):
                 `Image`, or tensor representing an image batch, that will be used as the starting point for the
                 process.
-            strength (`float`, *optional*, defaults to 0.8):
+            strength (`float`, *optional*, defaults to 0.7):
                 Conceptually, indicates how much to transform the reference `image`. Must be between 0 and 1. `image`
                 will be used as a starting point, adding more noise to it the larger the `strength`. The number of
                 denoising steps depends on the amount of noise initially added. When `strength` is 1, added noise will
                 be maximum and the denoising process will run for the full number of iterations specified in
                 `num_inference_steps`. A value of 1, therefore, essentially ignores `image`.
-            num_inference_steps (`int`, *optional*, defaults to 50):
+            num_inference_steps (`int`, *optional*, defaults to 80):
                 The number of denoising steps. More denoising steps usually lead to a higher quality image at the
                 expense of slower inference.
             timesteps (`List[int]`, *optional*):
                 Custom timesteps to use for the denoising process. If not defined, equal spaced `num_inference_steps`
                 timesteps are used. Must be in descending order.
-            guidance_scale (`float`, *optional*, defaults to 7.5):
+            guidance_scale (`float`, *optional*, defaults to 10.0):
                 Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
                 `guidance_scale` is defined as `w` of equation 2. of [Imagen
                 Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >

src/diffusers/pipelines/deepfloyd_if/pipeline_if_img2img_superresolution.py

@@ -338,11 +338,11 @@ class IFImg2ImgSuperResolutionPipeline(DiffusionPipeline, LoraLoaderMixin):
     # Copied from diffusers.pipelines.deepfloyd_if.pipeline_if.IFPipeline.encode_prompt
     def encode_prompt(
         self,
-        prompt,
-        do_classifier_free_guidance=True,
-        num_images_per_prompt=1,
-        device=None,
-        negative_prompt=None,
+        prompt: Union[str, List[str]],
+        do_classifier_free_guidance: bool = True,
+        num_images_per_prompt: int = 1,
+        device: Optional[torch.device] = None,
+        negative_prompt: Optional[Union[str, List[str]]] = None,
         prompt_embeds: Optional[torch.FloatTensor] = None,
         negative_prompt_embeds: Optional[torch.FloatTensor] = None,
         clean_caption: bool = False,
@@ -353,12 +353,12 @@ class IFImg2ImgSuperResolutionPipeline(DiffusionPipeline, LoraLoaderMixin):
         Args:
             prompt (`str` or `List[str]`, *optional*):
                 prompt to be encoded
-            device: (`torch.device`, *optional*):
-                torch device to place the resulting embeddings on
-            num_images_per_prompt (`int`, *optional*, defaults to 1):
-                number of images that should be generated per prompt
             do_classifier_free_guidance (`bool`, *optional*, defaults to `True`):
                 whether to use classifier free guidance or not
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                number of images that should be generated per prompt
+            device: (`torch.device`, *optional*):
+                torch device to place the resulting embeddings on
             negative_prompt (`str` or `List[str]`, *optional*):
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds`. instead. If not defined, one has to pass `negative_prompt_embeds`. instead.
@@ -370,6 +370,8 @@ class IFImg2ImgSuperResolutionPipeline(DiffusionPipeline, LoraLoaderMixin):
                 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.
+            clean_caption (bool, defaults to `False`):
+                If `True`, the function will preprocess and clean the provided caption before encoding.
         """
         if prompt is not None and negative_prompt is not None:
             if type(prompt) is not type(negative_prompt):
@@ -784,7 +786,7 @@ class IFImg2ImgSuperResolutionPipeline(DiffusionPipeline, LoraLoaderMixin):
             timesteps (`List[int]`, *optional*):
                 Custom timesteps to use for the denoising process. If not defined, equal spaced `num_inference_steps`
                 timesteps are used. Must be in descending order.
-            guidance_scale (`float`, *optional*, defaults to 7.5):
+            guidance_scale (`float`, *optional*, defaults to 4.0):
                 Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
                 `guidance_scale` is defined as `w` of equation 2. of [Imagen
                 Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >

src/diffusers/pipelines/deepfloyd_if/pipeline_if_inpainting.py

@@ -190,11 +190,11 @@ class IFInpaintingPipeline(DiffusionPipeline, LoraLoaderMixin):
     # Copied from diffusers.pipelines.deepfloyd_if.pipeline_if.IFPipeline.encode_prompt
     def encode_prompt(
         self,
-        prompt,
-        do_classifier_free_guidance=True,
-        num_images_per_prompt=1,
-        device=None,
-        negative_prompt=None,
+        prompt: Union[str, List[str]],
+        do_classifier_free_guidance: bool = True,
+        num_images_per_prompt: int = 1,
+        device: Optional[torch.device] = None,
+        negative_prompt: Optional[Union[str, List[str]]] = None,
         prompt_embeds: Optional[torch.FloatTensor] = None,
         negative_prompt_embeds: Optional[torch.FloatTensor] = None,
         clean_caption: bool = False,
@@ -205,12 +205,12 @@ class IFInpaintingPipeline(DiffusionPipeline, LoraLoaderMixin):
         Args:
             prompt (`str` or `List[str]`, *optional*):
                 prompt to be encoded
-            device: (`torch.device`, *optional*):
-                torch device to place the resulting embeddings on
-            num_images_per_prompt (`int`, *optional*, defaults to 1):
-                number of images that should be generated per prompt
             do_classifier_free_guidance (`bool`, *optional*, defaults to `True`):
                 whether to use classifier free guidance or not
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                number of images that should be generated per prompt
+            device: (`torch.device`, *optional*):
+                torch device to place the resulting embeddings on
             negative_prompt (`str` or `List[str]`, *optional*):
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds`. instead. If not defined, one has to pass `negative_prompt_embeds`. instead.
@@ -222,6 +222,8 @@ class IFInpaintingPipeline(DiffusionPipeline, LoraLoaderMixin):
                 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.
+            clean_caption (bool, defaults to `False`):
+                If `True`, the function will preprocess and clean the provided caption before encoding.
         """
         if prompt is not None and negative_prompt is not None:
             if type(prompt) is not type(negative_prompt):
@@ -786,7 +788,7 @@ class IFInpaintingPipeline(DiffusionPipeline, LoraLoaderMixin):
                 repainted, while black pixels will be preserved. If `mask_image` is a PIL image, it will be converted
                 to a single channel (luminance) before use. If it's a tensor, it should contain one color channel (L)
                 instead of 3, so the expected shape would be `(B, H, W, 1)`.
-            strength (`float`, *optional*, defaults to 0.8):
+            strength (`float`, *optional*, defaults to 1.0):
                 Conceptually, indicates how much to transform the reference `image`. Must be between 0 and 1. `image`
                 will be used as a starting point, adding more noise to it the larger the `strength`. The number of
                 denoising steps depends on the amount of noise initially added. When `strength` is 1, added noise will
@@ -798,7 +800,7 @@ class IFInpaintingPipeline(DiffusionPipeline, LoraLoaderMixin):
             timesteps (`List[int]`, *optional*):
                 Custom timesteps to use for the denoising process. If not defined, equal spaced `num_inference_steps`
                 timesteps are used. Must be in descending order.
-            guidance_scale (`float`, *optional*, defaults to 7.5):
+            guidance_scale (`float`, *optional*, defaults to 7.0):
                 Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
                 `guidance_scale` is defined as `w` of equation 2. of [Imagen
                 Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >

src/diffusers/pipelines/deepfloyd_if/pipeline_if_inpainting_superresolution.py

@@ -340,11 +340,11 @@ class IFInpaintingSuperResolutionPipeline(DiffusionPipeline, LoraLoaderMixin):
     # Copied from diffusers.pipelines.deepfloyd_if.pipeline_if.IFPipeline.encode_prompt
     def encode_prompt(
         self,
-        prompt,
-        do_classifier_free_guidance=True,
-        num_images_per_prompt=1,
-        device=None,
-        negative_prompt=None,
+        prompt: Union[str, List[str]],
+        do_classifier_free_guidance: bool = True,
+        num_images_per_prompt: int = 1,
+        device: Optional[torch.device] = None,
+        negative_prompt: Optional[Union[str, List[str]]] = None,
         prompt_embeds: Optional[torch.FloatTensor] = None,
         negative_prompt_embeds: Optional[torch.FloatTensor] = None,
         clean_caption: bool = False,
@@ -355,12 +355,12 @@ class IFInpaintingSuperResolutionPipeline(DiffusionPipeline, LoraLoaderMixin):
         Args:
             prompt (`str` or `List[str]`, *optional*):
                 prompt to be encoded
-            device: (`torch.device`, *optional*):
-                torch device to place the resulting embeddings on
-            num_images_per_prompt (`int`, *optional*, defaults to 1):
-                number of images that should be generated per prompt
             do_classifier_free_guidance (`bool`, *optional*, defaults to `True`):
                 whether to use classifier free guidance or not
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                number of images that should be generated per prompt
+            device: (`torch.device`, *optional*):
+                torch device to place the resulting embeddings on
             negative_prompt (`str` or `List[str]`, *optional*):
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds`. instead. If not defined, one has to pass `negative_prompt_embeds`. instead.
@@ -372,6 +372,8 @@ class IFInpaintingSuperResolutionPipeline(DiffusionPipeline, LoraLoaderMixin):
                 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.
+            clean_caption (bool, defaults to `False`):
+                If `True`, the function will preprocess and clean the provided caption before encoding.
         """
         if prompt is not None and negative_prompt is not None:
             if type(prompt) is not type(negative_prompt):
@@ -874,13 +876,13 @@ class IFInpaintingSuperResolutionPipeline(DiffusionPipeline, LoraLoaderMixin):
             prompt (`str` or `List[str]`, *optional*):
                 The prompt or prompts to guide the image generation. If not defined, one has to pass `prompt_embeds`.
                 instead.
-            num_inference_steps (`int`, *optional*, defaults to 50):
+            num_inference_steps (`int`, *optional*, defaults to 100):
                 The number of denoising steps. More denoising steps usually lead to a higher quality image at the
                 expense of slower inference.
             timesteps (`List[int]`, *optional*):
                 Custom timesteps to use for the denoising process. If not defined, equal spaced `num_inference_steps`
                 timesteps are used. Must be in descending order.
-            guidance_scale (`float`, *optional*, defaults to 7.5):
+            guidance_scale (`float`, *optional*, defaults to 4.0):
                 Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
                 `guidance_scale` is defined as `w` of equation 2. of [Imagen
                 Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >

src/diffusers/pipelines/deepfloyd_if/pipeline_if_superresolution.py

@@ -296,11 +296,11 @@ class IFSuperResolutionPipeline(DiffusionPipeline, LoraLoaderMixin):
     # Copied from diffusers.pipelines.deepfloyd_if.pipeline_if.IFPipeline.encode_prompt
     def encode_prompt(
         self,
-        prompt,
-        do_classifier_free_guidance=True,
-        num_images_per_prompt=1,
-        device=None,
-        negative_prompt=None,
+        prompt: Union[str, List[str]],
+        do_classifier_free_guidance: bool = True,
+        num_images_per_prompt: int = 1,
+        device: Optional[torch.device] = None,
+        negative_prompt: Optional[Union[str, List[str]]] = None,
         prompt_embeds: Optional[torch.FloatTensor] = None,
         negative_prompt_embeds: Optional[torch.FloatTensor] = None,
         clean_caption: bool = False,
@@ -311,12 +311,12 @@ class IFSuperResolutionPipeline(DiffusionPipeline, LoraLoaderMixin):
         Args:
             prompt (`str` or `List[str]`, *optional*):
                 prompt to be encoded
-            device: (`torch.device`, *optional*):
-                torch device to place the resulting embeddings on
-            num_images_per_prompt (`int`, *optional*, defaults to 1):
-                number of images that should be generated per prompt
             do_classifier_free_guidance (`bool`, *optional*, defaults to `True`):
                 whether to use classifier free guidance or not
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                number of images that should be generated per prompt
+            device: (`torch.device`, *optional*):
+                torch device to place the resulting embeddings on
             negative_prompt (`str` or `List[str]`, *optional*):
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds`. instead. If not defined, one has to pass `negative_prompt_embeds`. instead.
@@ -328,6 +328,8 @@ class IFSuperResolutionPipeline(DiffusionPipeline, LoraLoaderMixin):
                 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.
+            clean_caption (bool, defaults to `False`):
+                If `True`, the function will preprocess and clean the provided caption before encoding.
         """
         if prompt is not None and negative_prompt is not None:
             if type(prompt) is not type(negative_prompt):
@@ -637,19 +639,19 @@ class IFSuperResolutionPipeline(DiffusionPipeline, LoraLoaderMixin):
             prompt (`str` or `List[str]`, *optional*):
                 The prompt or prompts to guide the image generation. If not defined, one has to pass `prompt_embeds`.
                 instead.
-            height (`int`, *optional*, defaults to self.unet.config.sample_size):
+            height (`int`, *optional*, defaults to None):
                 The height in pixels of the generated image.
-            width (`int`, *optional*, defaults to self.unet.config.sample_size):
+            width (`int`, *optional*, defaults to None):
                 The width in pixels of the generated image.
             image (`PIL.Image.Image`, `np.ndarray`, `torch.FloatTensor`):
                 The image to be upscaled.
             num_inference_steps (`int`, *optional*, defaults to 50):
                 The number of denoising steps. More denoising steps usually lead to a higher quality image at the
                 expense of slower inference.
-            timesteps (`List[int]`, *optional*):
+            timesteps (`List[int]`, *optional*, defaults to None):
                 Custom timesteps to use for the denoising process. If not defined, equal spaced `num_inference_steps`
                 timesteps are used. Must be in descending order.
-            guidance_scale (`float`, *optional*, defaults to 7.5):
+            guidance_scale (`float`, *optional*, defaults to 4.0):
                 Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
                 `guidance_scale` is defined as `w` of equation 2. of [Imagen
                 Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >