r""" Performs Pipeline dtype and/or device conversion. A torch.dtype and torch.device are inferred from the arguments of `self.to(*arg

Sets the discrete timesteps used for the diffusion chain (to be run before inference). Args: num_inference_steps (`int`,

A helper function to create random tensors on the desired `device` with the desired `dtype`. When passing a list of generators, you can seed each

Ensures interchangeability with schedulers that need to scale the denoising model input depending on the current timestep. A

r""" Offloads all models to CPU using accelerate, reducing memory usage with a low impact on performance. Compared to `enable_sequenti

Predict the sample from the previous timestep by reversing the SDE. This function propagates the diffusion process from the learned m

Creates a random float32 tensor

Helper function for reproducible behavior during distributed training. See - https://pytorch.org/docs/stable/notes/randomness.html for pytorc

Disables custom attention processors and sets the default attention implementation.

r""" Function that offloads all components, removes all model hooks that were added when using `enable_model_cpu_offload` and then app

Loads `image` to a PIL Image. Args: image (`str` or `PIL.Image.Image`): The image to convert to the PIL Image format.

Resize image. Args: image (`PIL.Image.Image`, `np.ndarray` or `torch.Tensor`): The image input, can be a

r""" Processes prompts that include a special token corresponding to a multi-vector textual inversion embedding to be replaced with mu

Convert a NumPy image or a batch of images to a PIL image.

r""" Enable sliced attention computation. When this option is enabled, the attention module splits the input tensor in slices to compu

r""" Adds a new adapter to the current model for training. If no adapter name is passed, a default name is assigned to the adapter to

r""" Args: Loads the ExponentialMovingAverage state. This method is used by accelerate during checkpointing to save the ema st

Checks if the LoRA layers are correctly set with peft

r""" Instantiate a [`DiffusionPipeline`] from pretrained pipeline weights saved in the `.ckpt` or `.safetensors` format. The pipeline

Adjust the weightage given to the LoRA layers of the model. Args: model (`torch.nn.Module`): The model to scale.

r""" Reshape the tensor from `[batch_size, seq_len, dim]` to `[batch_size, seq_len, heads, dim // heads]` `heads` is the number of hea

Removes the previously passed weight given to the LoRA layers of the model. Args: model (`torch.nn.Module`): The model t

Loads `image` to a PIL Image. Args: image (`str` or `PIL.Image.Image`): The image to convert to the PIL Image format.

Args: Compares the current PyTorch version to a given reference with an operation. operation (`str`): A string representa

Resize an image to `(size["height"], size["width"])`. Args: image (`np.ndarray`): Image to resize.

Args: Returns a dict of object names and values in a module, while skipping private/internal objects module (ModuleType):

r""" Offloads all models to CPU using 🤗 Accelerate, significantly reducing memory usage. When called, the state dicts of all `torch.nn

`torch.device`: The device on which the module is (assuming that all the module parameters are on the same device).

Preprocess an image or batch of images. Args: images (`ImageInput`): Image to preprocess. Expects a sing

Load LoRA weights specified in `pretrained_model_name_or_path_or_dict` into `self.unet` and `self.text_encoder`. All kwargs

Args: t0: start time has shape [batch_size, *shape, 1] t1: finish time has shape [batch_size, *shape, 1]

Function invoked when using the prior pipeline for interpolation. Args: images_and_prompts (`List[Union[str, PIL.Image.I

Save a configuration object to the directory specified in `save_directory` so that it can be reloaded using the [`~ConfigMixin.from_c

r""" Create a new pipeline from a given pipeline. This method is useful to create a new pipeline from the existing pipeline components

r""" Set the attention processor to use. Args: processor (`AttnProcessor`): The attention processor to us

Preprocess the image input. Args: image (`pipeline_image_input`): The image input, accepted formats are

r""" Returns the state of the ExponentialMovingAverage as a dict. This method is used by accelerate during checkpointing to save the e

Unloads the LoRA parameters. Examples: ```python >>> # Assuming `pipeline` is already loaded with the LoRA paramete

Convert self to a tuple containing all the attributes/keys that are not `None`.

r""" Load a model or scheduler configuration. Parameters: pretrained_model_name_or_path (`str` or `os.PathLike`, *optiona

Helper function to get activation function from string. Args: act_fn (str): Name of activation function. Returns: nn.Module:

Postprocess the image output from tensor to `output_type`. Args: image (`torch.Tensor`): The image input

Prepares a pair (image, mask) to be consumed by the Stable Diffusion pipeline. This means that those inputs will be converted to ``torch.Tens

r""" Prepare the attention mask for the attention computation. Args: attention_mask (`torch.Tensor`): The

r""" Normalize the encoder hidden states. Requires `self.norm_cross` to be specified when constructing the `Attention` class.