diffusers 0.27.2__py3-none-any.whl → 0.28.1__py3-none-any.whl

diffusers/pipelines/kandinsky3/pipeline_kandinsky3.py

@@ -8,7 +8,6 @@ from ...models import Kandinsky3UNet, VQModel
 from ...schedulers import DDPMScheduler
 from ...utils import (
     deprecate,
-    is_accelerate_available,
     logging,
     replace_example_docstring,
 )
@@ -24,7 +23,9 @@ EXAMPLE_DOC_STRING = """
         >>> from diffusers import AutoPipelineForText2Image
         >>> import torch
 
-        >>> pipe = AutoPipelineForText2Image.from_pretrained("kandinsky-community/kandinsky-3", variant="fp16", torch_dtype=torch.float16)
+        >>> pipe = AutoPipelineForText2Image.from_pretrained(
+        ...     "kandinsky-community/kandinsky-3", variant="fp16", torch_dtype=torch.float16
+        ... )
         >>> pipe.enable_model_cpu_offload()
 
         >>> prompt = "A photograph of the inside of a subway train. There are raccoons sitting on the seats. One of them is reading a newspaper. The window shows the city in the background."
@@ -70,20 +71,6 @@ class Kandinsky3Pipeline(DiffusionPipeline, LoraLoaderMixin):
             tokenizer=tokenizer, text_encoder=text_encoder, unet=unet, scheduler=scheduler, movq=movq
         )
 
-    def remove_all_hooks(self):
-        if is_accelerate_available():
-            from accelerate.hooks import remove_hook_from_module
-        else:
-            raise ImportError("Please install accelerate via `pip install accelerate`")
-
-        for model in [self.text_encoder, self.unet, self.movq]:
-            if model is not None:
-                remove_hook_from_module(model, recurse=True)
-
-        self.unet_offload_hook = None
-        self.text_encoder_offload_hook = None
-        self.final_offload_hook = None
-
     def process_embeds(self, embeddings, attention_mask, cut_context):
         if cut_context:
             embeddings[attention_mask == 0] = torch.zeros_like(embeddings[attention_mask == 0])
@@ -100,11 +87,11 @@ class Kandinsky3Pipeline(DiffusionPipeline, LoraLoaderMixin):
         num_images_per_prompt=1,
         device=None,
         negative_prompt=None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
         _cut_context=False,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        negative_attention_mask: Optional[torch.FloatTensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        negative_attention_mask: Optional[torch.Tensor] = None,
     ):
         r"""
         Encodes the prompt into text encoder hidden states.
@@ -122,16 +109,16 @@ class Kandinsky3Pipeline(DiffusionPipeline, LoraLoaderMixin):
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds`. instead. If not defined, one has to pass `negative_prompt_embeds`. instead.
                 Ignored when not using guidance (i.e., ignored if `guidance_scale` is less than `1`).
-            prompt_embeds (`torch.FloatTensor`, *optional*):
+            prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
                 provided, text embeddings will be generated from `prompt` input argument.
-            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                 weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
                 argument.
-            attention_mask (`torch.FloatTensor`, *optional*):
+            attention_mask (`torch.Tensor`, *optional*):
                 Pre-generated attention mask. Must provide if passing `prompt_embeds` directly.
-            negative_attention_mask (`torch.FloatTensor`, *optional*):
+            negative_attention_mask (`torch.Tensor`, *optional*):
                 Pre-generated negative attention mask. Must provide if passing `negative_prompt_embeds` directly.
         """
         if prompt is not None and negative_prompt is not None:
@@ -347,10 +334,10 @@ class Kandinsky3Pipeline(DiffusionPipeline, LoraLoaderMixin):
         height: Optional[int] = 1024,
         width: Optional[int] = 1024,
         generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        negative_attention_mask: Optional[torch.FloatTensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        negative_attention_mask: Optional[torch.Tensor] = None,
         output_type: Optional[str] = "pil",
         return_dict: bool = True,
         latents=None,
@@ -393,16 +380,16 @@ class Kandinsky3Pipeline(DiffusionPipeline, LoraLoaderMixin):
             generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
                 One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
                 to make generation deterministic.
-            prompt_embeds (`torch.FloatTensor`, *optional*):
+            prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
                 provided, text embeddings will be generated from `prompt` input argument.
-            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                 weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
                 argument.
-            attention_mask (`torch.FloatTensor`, *optional*):
+            attention_mask (`torch.Tensor`, *optional*):
                 Pre-generated attention mask. Must provide if passing `prompt_embeds` directly.
-            negative_attention_mask (`torch.FloatTensor`, *optional*):
+            negative_attention_mask (`torch.Tensor`, *optional*):
                 Pre-generated negative attention mask. Must provide if passing `negative_prompt_embeds` directly.
             output_type (`str`, *optional*, defaults to `"pil"`):
                 The output format of the generate image. Choose between
@@ -411,7 +398,7 @@ class Kandinsky3Pipeline(DiffusionPipeline, LoraLoaderMixin):
                 Whether or not to return a [`~pipelines.stable_diffusion.IFPipelineOutput`] instead of a plain tuple.
             callback (`Callable`, *optional*):
                 A function that will be called every `callback_steps` steps during inference. The function will be
-                called with the following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`.
+                called with the following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
             callback_steps (`int`, *optional*, defaults to 1):
                 The frequency at which the `callback` function will be called. If not specified, the callback will be
                 called at every step.

diffusers/pipelines/kandinsky3/pipeline_kandinsky3_img2img.py

@@ -12,7 +12,6 @@ from ...models import Kandinsky3UNet, VQModel
 from ...schedulers import DDPMScheduler
 from ...utils import (
     deprecate,
-    is_accelerate_available,
     logging,
     replace_example_docstring,
 )
@@ -29,11 +28,15 @@ EXAMPLE_DOC_STRING = """
         >>> from diffusers.utils import load_image
         >>> import torch
 
-        >>> pipe = AutoPipelineForImage2Image.from_pretrained("kandinsky-community/kandinsky-3", variant="fp16", torch_dtype=torch.float16)
+        >>> pipe = AutoPipelineForImage2Image.from_pretrained(
+        ...     "kandinsky-community/kandinsky-3", variant="fp16", torch_dtype=torch.float16
+        ... )
         >>> pipe.enable_model_cpu_offload()
 
         >>> prompt = "A painting of the inside of a subway train with tiny raccoons."
-        >>> image = load_image("https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky3/t2i.png")
+        >>> image = load_image(
+        ...     "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky3/t2i.png"
+        ... )
 
         >>> generator = torch.Generator(device="cpu").manual_seed(0)
         >>> image = pipe(prompt, image=image, strength=0.75, num_inference_steps=25, generator=generator).images[0]
@@ -92,20 +95,6 @@ class Kandinsky3Img2ImgPipeline(DiffusionPipeline, LoraLoaderMixin):
 
         return timesteps, num_inference_steps - t_start
 
-    def remove_all_hooks(self):
-        if is_accelerate_available():
-            from accelerate.hooks import remove_hook_from_module
-        else:
-            raise ImportError("Please install accelerate via `pip install accelerate`")
-
-        for model in [self.text_encoder, self.unet]:
-            if model is not None:
-                remove_hook_from_module(model, recurse=True)
-
-        self.unet_offload_hook = None
-        self.text_encoder_offload_hook = None
-        self.final_offload_hook = None
-
     def _process_embeds(self, embeddings, attention_mask, cut_context):
         # return embeddings, attention_mask
         if cut_context:
@@ -123,11 +112,11 @@ class Kandinsky3Img2ImgPipeline(DiffusionPipeline, LoraLoaderMixin):
         num_images_per_prompt=1,
         device=None,
         negative_prompt=None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
         _cut_context=False,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        negative_attention_mask: Optional[torch.FloatTensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        negative_attention_mask: Optional[torch.Tensor] = None,
     ):
         r"""
         Encodes the prompt into text encoder hidden states.
@@ -145,16 +134,16 @@ class Kandinsky3Img2ImgPipeline(DiffusionPipeline, LoraLoaderMixin):
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds`. instead. If not defined, one has to pass `negative_prompt_embeds`. instead.
                 Ignored when not using guidance (i.e., ignored if `guidance_scale` is less than `1`).
-            prompt_embeds (`torch.FloatTensor`, *optional*):
+            prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
                 provided, text embeddings will be generated from `prompt` input argument.
-            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                 weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
                 argument.
-            attention_mask (`torch.FloatTensor`, *optional*):
+            attention_mask (`torch.Tensor`, *optional*):
                 Pre-generated attention mask. Must provide if passing `prompt_embeds` directly.
-            negative_attention_mask (`torch.FloatTensor`, *optional*):
+            negative_attention_mask (`torch.Tensor`, *optional*):
                 Pre-generated negative attention mask. Must provide if passing `negative_prompt_embeds` directly.
         """
         if prompt is not None and negative_prompt is not None:
@@ -414,17 +403,17 @@ class Kandinsky3Img2ImgPipeline(DiffusionPipeline, LoraLoaderMixin):
     def __call__(
         self,
         prompt: Union[str, List[str]] = None,
-        image: Union[torch.FloatTensor, PIL.Image.Image, List[torch.FloatTensor], List[PIL.Image.Image]] = None,
+        image: Union[torch.Tensor, PIL.Image.Image, List[torch.Tensor], List[PIL.Image.Image]] = None,
         strength: float = 0.3,
         num_inference_steps: int = 25,
         guidance_scale: float = 3.0,
         negative_prompt: Optional[Union[str, List[str]]] = None,
         num_images_per_prompt: Optional[int] = 1,
         generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        negative_attention_mask: Optional[torch.FloatTensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        negative_attention_mask: Optional[torch.Tensor] = None,
         output_type: Optional[str] = "pil",
         return_dict: bool = True,
         callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
@@ -438,7 +427,7 @@ class Kandinsky3Img2ImgPipeline(DiffusionPipeline, LoraLoaderMixin):
             prompt (`str` or `List[str]`, *optional*):
                 The prompt or prompts to guide the image generation. If not defined, one has to pass `prompt_embeds`.
                 instead.
-            image (`torch.FloatTensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.FloatTensor]`, `List[PIL.Image.Image]`, or `List[np.ndarray]`):
+            image (`torch.Tensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.Tensor]`, `List[PIL.Image.Image]`, or `List[np.ndarray]`):
                 `Image`, or tensor representing an image batch, that will be used as the starting point for the
                 process.
             strength (`float`, *optional*, defaults to 0.8):
@@ -465,16 +454,16 @@ class Kandinsky3Img2ImgPipeline(DiffusionPipeline, LoraLoaderMixin):
             generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
                 One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
                 to make generation deterministic.
-            prompt_embeds (`torch.FloatTensor`, *optional*):
+            prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
                 provided, text embeddings will be generated from `prompt` input argument.
-            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                 weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
                 argument.
-            attention_mask (`torch.FloatTensor`, *optional*):
+            attention_mask (`torch.Tensor`, *optional*):
                 Pre-generated attention mask. Must provide if passing `prompt_embeds` directly.
-            negative_attention_mask (`torch.FloatTensor`, *optional*):
+            negative_attention_mask (`torch.Tensor`, *optional*):
                 Pre-generated negative attention mask. Must provide if passing `negative_prompt_embeds` directly.
             output_type (`str`, *optional*, defaults to `"pil"`):
                 The output format of the generate image. Choose between

diffusers/pipelines/latent_consistency_models/pipeline_latent_consistency_img2img.py

@@ -63,6 +63,7 @@ def retrieve_timesteps(
     num_inference_steps: Optional[int] = None,
     device: Optional[Union[str, torch.device]] = None,
     timesteps: Optional[List[int]] = None,
+    sigmas: Optional[List[float]] = None,
     **kwargs,
 ):
     """
@@ -73,19 +74,23 @@ def retrieve_timesteps(
         scheduler (`SchedulerMixin`):
             The scheduler to get timesteps from.
         num_inference_steps (`int`):
-            The number of diffusion steps used when generating samples with a pre-trained model. If used,
-            `timesteps` must be `None`.
+            The number of diffusion steps used when generating samples with a pre-trained model. If used, `timesteps`
+            must be `None`.
         device (`str` or `torch.device`, *optional*):
             The device to which the timesteps should be moved to. If `None`, the timesteps are not moved.
         timesteps (`List[int]`, *optional*):
-                Custom timesteps used to support arbitrary spacing between timesteps. If `None`, then the default
-                timestep spacing strategy of the scheduler is used. If `timesteps` is passed, `num_inference_steps`
-                must be `None`.
+            Custom timesteps used to override the timestep spacing strategy of the scheduler. If `timesteps` is passed,
+            `num_inference_steps` and `sigmas` must be `None`.
+        sigmas (`List[float]`, *optional*):
+            Custom sigmas used to override the timestep spacing strategy of the scheduler. If `sigmas` is passed,
+            `num_inference_steps` and `timesteps` must be `None`.
 
     Returns:
         `Tuple[torch.Tensor, int]`: A tuple where the first element is the timestep schedule from the scheduler and the
         second element is the number of inference steps.
     """
+    if timesteps is not None and sigmas is not None:
+        raise ValueError("Only one of `timesteps` or `sigmas` can be passed. Please choose one to set custom values")
     if timesteps is not None:
         accepts_timesteps = "timesteps" in set(inspect.signature(scheduler.set_timesteps).parameters.keys())
         if not accepts_timesteps:
@@ -96,6 +101,16 @@ def retrieve_timesteps(
         scheduler.set_timesteps(timesteps=timesteps, device=device, **kwargs)
         timesteps = scheduler.timesteps
         num_inference_steps = len(timesteps)
+    elif sigmas is not None:
+        accept_sigmas = "sigmas" in set(inspect.signature(scheduler.set_timesteps).parameters.keys())
+        if not accept_sigmas:
+            raise ValueError(
+                f"The current scheduler class {scheduler.__class__}'s `set_timesteps` does not support custom"
+                f" sigmas schedules. Please check whether you are using the correct scheduler."
+            )
+        scheduler.set_timesteps(sigmas=sigmas, device=device, **kwargs)
+        timesteps = scheduler.timesteps
+        num_inference_steps = len(timesteps)
     else:
         scheduler.set_timesteps(num_inference_steps, device=device, **kwargs)
         timesteps = scheduler.timesteps
@@ -222,8 +237,8 @@ class LatentConsistencyModelImg2ImgPipeline(
         num_images_per_prompt,
         do_classifier_free_guidance,
         negative_prompt=None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
         lora_scale: Optional[float] = None,
         clip_skip: Optional[int] = None,
     ):
@@ -243,10 +258,10 @@ class LatentConsistencyModelImg2ImgPipeline(
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
                 less than `1`).
-            prompt_embeds (`torch.FloatTensor`, *optional*):
+            prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
                 provided, text embeddings will be generated from `prompt` input argument.
-            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                 weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
                 argument.
@@ -548,20 +563,22 @@ class LatentConsistencyModelImg2ImgPipeline(
         return latents
 
     # Copied from diffusers.pipelines.latent_consistency_models.pipeline_latent_consistency_text2img.LatentConsistencyModelPipeline.get_guidance_scale_embedding
-    def get_guidance_scale_embedding(self, w, embedding_dim=512, dtype=torch.float32):
+    def get_guidance_scale_embedding(
+        self, w: torch.Tensor, embedding_dim: int = 512, dtype: torch.dtype = torch.float32
+    ) -> torch.Tensor:
         """
         See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298
 
         Args:
-            timesteps (`torch.Tensor`):
-                generate embedding vectors at these timesteps
+            w (`torch.Tensor`):
+                Generate embedding vectors with a specified guidance scale to subsequently enrich timestep embeddings.
             embedding_dim (`int`, *optional*, defaults to 512):
-                dimension of the embeddings to generate
-            dtype:
-                data type of the generated embeddings
+                Dimension of the embeddings to generate.
+            dtype (`torch.dtype`, *optional*, defaults to `torch.float32`):
+                Data type of the generated embeddings.
 
         Returns:
-            `torch.FloatTensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)`
+            `torch.Tensor`: Embedding vectors with shape `(len(w), embedding_dim)`.
         """
         assert len(w.shape) == 1
         w = w * 1000.0
@@ -611,7 +628,7 @@ class LatentConsistencyModelImg2ImgPipeline(
         prompt: Union[str, List[str]],
         strength: float,
         callback_steps: int,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
         ip_adapter_image=None,
         ip_adapter_image_embeds=None,
         callback_on_step_end_tensor_inputs=None,
@@ -692,10 +709,10 @@ class LatentConsistencyModelImg2ImgPipeline(
         guidance_scale: float = 8.5,
         num_images_per_prompt: Optional[int] = 1,
         generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
-        latents: Optional[torch.FloatTensor] = None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
+        latents: Optional[torch.Tensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
         ip_adapter_image: Optional[PipelineImageInput] = None,
-        ip_adapter_image_embeds: Optional[List[torch.FloatTensor]] = None,
+        ip_adapter_image_embeds: Optional[List[torch.Tensor]] = None,
         output_type: Optional[str] = "pil",
         return_dict: bool = True,
         cross_attention_kwargs: Optional[Dict[str, Any]] = None,
@@ -737,20 +754,20 @@ class LatentConsistencyModelImg2ImgPipeline(
             generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
                 A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
                 generation deterministic.
-            latents (`torch.FloatTensor`, *optional*):
+            latents (`torch.Tensor`, *optional*):
                 Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image
                 generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
                 tensor is generated by sampling using the supplied random `generator`.
-            prompt_embeds (`torch.FloatTensor`, *optional*):
+            prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
                 provided, text embeddings are generated from the `prompt` input argument.
             ip_adapter_image: (`PipelineImageInput`, *optional*):
                 Optional image input to work with IP Adapters.
-            ip_adapter_image_embeds (`List[torch.FloatTensor]`, *optional*):
-                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 provided, embeddings are computed from the `ip_adapter_image` input argument.
+            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
+                provided, embeddings are computed from the `ip_adapter_image` input argument.
             output_type (`str`, *optional*, defaults to `"pil"`):
                 The output format of the generated image. Choose between `PIL.Image` or `np.array`.
             return_dict (`bool`, *optional*, defaults to `True`):
@@ -870,9 +887,10 @@ class LatentConsistencyModelImg2ImgPipeline(
             else self.scheduler.config.original_inference_steps
         )
         latent_timestep = timesteps[:1]
-        latents = self.prepare_latents(
-            image, latent_timestep, batch_size, num_images_per_prompt, prompt_embeds.dtype, device, generator
-        )
+        if latents is None:
+            latents = self.prepare_latents(
+                image, latent_timestep, batch_size, num_images_per_prompt, prompt_embeds.dtype, device, generator
+            )
         bs = batch_size * num_images_per_prompt
 
         # 6. Get Guidance Scale Embedding

diffusers/pipelines/latent_consistency_models/pipeline_latent_consistency_text2img.py

@@ -67,6 +67,7 @@ def retrieve_timesteps(
     num_inference_steps: Optional[int] = None,
     device: Optional[Union[str, torch.device]] = None,
     timesteps: Optional[List[int]] = None,
+    sigmas: Optional[List[float]] = None,
     **kwargs,
 ):
     """
@@ -77,19 +78,23 @@ def retrieve_timesteps(
         scheduler (`SchedulerMixin`):
             The scheduler to get timesteps from.
         num_inference_steps (`int`):
-            The number of diffusion steps used when generating samples with a pre-trained model. If used,
-            `timesteps` must be `None`.
+            The number of diffusion steps used when generating samples with a pre-trained model. If used, `timesteps`
+            must be `None`.
         device (`str` or `torch.device`, *optional*):
             The device to which the timesteps should be moved to. If `None`, the timesteps are not moved.
         timesteps (`List[int]`, *optional*):
-                Custom timesteps used to support arbitrary spacing between timesteps. If `None`, then the default
-                timestep spacing strategy of the scheduler is used. If `timesteps` is passed, `num_inference_steps`
-                must be `None`.
+            Custom timesteps used to override the timestep spacing strategy of the scheduler. If `timesteps` is passed,
+            `num_inference_steps` and `sigmas` must be `None`.
+        sigmas (`List[float]`, *optional*):
+            Custom sigmas used to override the timestep spacing strategy of the scheduler. If `sigmas` is passed,
+            `num_inference_steps` and `timesteps` must be `None`.
 
     Returns:
         `Tuple[torch.Tensor, int]`: A tuple where the first element is the timestep schedule from the scheduler and the
         second element is the number of inference steps.
     """
+    if timesteps is not None and sigmas is not None:
+        raise ValueError("Only one of `timesteps` or `sigmas` can be passed. Please choose one to set custom values")
     if timesteps is not None:
         accepts_timesteps = "timesteps" in set(inspect.signature(scheduler.set_timesteps).parameters.keys())
         if not accepts_timesteps:
@@ -100,6 +105,16 @@ def retrieve_timesteps(
         scheduler.set_timesteps(timesteps=timesteps, device=device, **kwargs)
         timesteps = scheduler.timesteps
         num_inference_steps = len(timesteps)
+    elif sigmas is not None:
+        accept_sigmas = "sigmas" in set(inspect.signature(scheduler.set_timesteps).parameters.keys())
+        if not accept_sigmas:
+            raise ValueError(
+                f"The current scheduler class {scheduler.__class__}'s `set_timesteps` does not support custom"
+                f" sigmas schedules. Please check whether you are using the correct scheduler."
+            )
+        scheduler.set_timesteps(sigmas=sigmas, device=device, **kwargs)
+        timesteps = scheduler.timesteps
+        num_inference_steps = len(timesteps)
     else:
         scheduler.set_timesteps(num_inference_steps, device=device, **kwargs)
         timesteps = scheduler.timesteps
@@ -206,8 +221,8 @@ class LatentConsistencyModelPipeline(
         num_images_per_prompt,
         do_classifier_free_guidance,
         negative_prompt=None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
         lora_scale: Optional[float] = None,
         clip_skip: Optional[int] = None,
     ):
@@ -227,10 +242,10 @@ class LatentConsistencyModelPipeline(
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
                 less than `1`).
-            prompt_embeds (`torch.FloatTensor`, *optional*):
+            prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
                 provided, text embeddings will be generated from `prompt` input argument.
-            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                 weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
                 argument.
@@ -474,7 +489,12 @@ class LatentConsistencyModelPipeline(
 
     # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.prepare_latents
     def prepare_latents(self, batch_size, num_channels_latents, height, width, dtype, device, generator, latents=None):
-        shape = (batch_size, num_channels_latents, height // self.vae_scale_factor, width // self.vae_scale_factor)
+        shape = (
+            batch_size,
+            num_channels_latents,
+            int(height) // self.vae_scale_factor,
+            int(width) // self.vae_scale_factor,
+        )
         if isinstance(generator, list) and len(generator) != batch_size:
             raise ValueError(
                 f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
@@ -490,20 +510,22 @@ class LatentConsistencyModelPipeline(
         latents = latents * self.scheduler.init_noise_sigma
         return latents
 
-    def get_guidance_scale_embedding(self, w, embedding_dim=512, dtype=torch.float32):
+    def get_guidance_scale_embedding(
+        self, w: torch.Tensor, embedding_dim: int = 512, dtype: torch.dtype = torch.float32
+    ) -> torch.Tensor:
         """
         See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298
 
         Args:
-            timesteps (`torch.Tensor`):
-                generate embedding vectors at these timesteps
+            w (`torch.Tensor`):
+                Generate embedding vectors with a specified guidance scale to subsequently enrich timestep embeddings.
             embedding_dim (`int`, *optional*, defaults to 512):
-                dimension of the embeddings to generate
-            dtype:
-                data type of the generated embeddings
+                Dimension of the embeddings to generate.
+            dtype (`torch.dtype`, *optional*, defaults to `torch.float32`):
+                Data type of the generated embeddings.
 
         Returns:
-            `torch.FloatTensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)`
+            `torch.Tensor`: Embedding vectors with shape `(len(w), embedding_dim)`.
         """
         assert len(w.shape) == 1
         w = w * 1000.0
@@ -543,7 +565,7 @@ class LatentConsistencyModelPipeline(
         height: int,
         width: int,
         callback_steps: int,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
         ip_adapter_image=None,
         ip_adapter_image_embeds=None,
         callback_on_step_end_tensor_inputs=None,
@@ -624,10 +646,10 @@ class LatentConsistencyModelPipeline(
         guidance_scale: float = 8.5,
         num_images_per_prompt: Optional[int] = 1,
         generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
-        latents: Optional[torch.FloatTensor] = None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
+        latents: Optional[torch.Tensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
         ip_adapter_image: Optional[PipelineImageInput] = None,
-        ip_adapter_image_embeds: Optional[List[torch.FloatTensor]] = None,
+        ip_adapter_image_embeds: Optional[List[torch.Tensor]] = None,
         output_type: Optional[str] = "pil",
         return_dict: bool = True,
         cross_attention_kwargs: Optional[Dict[str, Any]] = None,
@@ -669,20 +691,20 @@ class LatentConsistencyModelPipeline(
             generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
                 A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
                 generation deterministic.
-            latents (`torch.FloatTensor`, *optional*):
+            latents (`torch.Tensor`, *optional*):
                 Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image
                 generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
                 tensor is generated by sampling using the supplied random `generator`.
-            prompt_embeds (`torch.FloatTensor`, *optional*):
+            prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
                 provided, text embeddings are generated from the `prompt` input argument.
             ip_adapter_image: (`PipelineImageInput`, *optional*):
                 Optional image input to work with IP Adapters.
-            ip_adapter_image_embeds (`List[torch.FloatTensor]`, *optional*):
-                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 provided, embeddings are computed from the `ip_adapter_image` input argument.
+            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
+                provided, embeddings are computed from the `ip_adapter_image` input argument.
             output_type (`str`, *optional*, defaults to `"pil"`):
                 The output format of the generated image. Choose between `PIL.Image` or `np.array`.
             return_dict (`bool`, *optional*, defaults to `True`):