Return [min_id, max_id] for an event type.

This matches the implementation in Denoising Diffusion Probabilistic Models: Create sinusoidal timestep embeddings. :param timesteps: a 1

Initialize a Batchnorm layer.

Interpolate data in time domain. This is used to compensate the resolution reduction in downsampling of a CNN. Args: x: (batch_size

Pad framewise_output to the same length as input frames. The pad value is the same as the value of the last frame. Args: framewise_

Updates paths inside resnets to the new naming scheme (local renaming)

r""" Parameters: `processor (`dict` of `AttentionProcessor` or `AttentionProcessor`): The instantiated processo

Set the verbosity level for the 🤗 Diffusers' root logger. Args: verbosity (`int`): Logging level, e.g., one of:

Create sinusoidal timestep embeddings. :param timesteps: a 1-D Tensor of N indices, one per batch element. These may

r"""Fills the input Tensor with values drawn from a truncated normal distribution. The values are effectively drawn from the normal distribu

Initialize an empty AttentionStore :param step_index: used to visualize only a specific step in the diffusion process

Aggregates the attention for each token and computes the max activation value for each token to alter.

Helper method to cast floating-point values of given parameter `PyTree` to given `dtype`.

r""" Combines a latent image img_vae of shape (B, C, H, W), a CLIP-embedded image img_clip of shape (B, L_img, clip_img_dim), and a

Computes the attend-and-excite loss using the maximum attention value for each token.

r""" Splits a flattened embedding x of shape (B, C * H * W + clip_img_dim) into two tensors of shape (B, C, H, W) and (B, 1, clip_im

r""" Splits a flattened embedding x of shape (B, C * H * W + clip_img_dim + text_seq_len * text_dim] into (img_vae, img_clip, text)

Perform backward process given list of time steps Args: latents: Latents at time timesteps[0]. timestep

DPM-Solver-1 (equivalent to DDIM) from time `s` to time `t`. Args: x: A pytorch tensor. The initial value at time `s

Reverse step process: recover noisy image from generated image. Args: images (`List[PIL Image]`): list of images to encode

Converts all `BatchNorm2d` and `SyncBatchNorm` layers of provided module into `FrozenBatchNorm2d`. If `module` is itself an instance of eit

Calculate and add audio features to sample. Sample: a dict containing all the data of current sample. audio_data: a tensor of shape (T

Return a logger with the specified name. This function is not supposed to be directly accessed unless you are writing a custom diffusers

Reads a checkpoint file, returning properly formatted errors if they arise.

Compute alpha_t of a given continuous-time label t in [0, T].

r"""Parse yaml file. Args: config_yaml (str): config yaml path Returns: yaml_dict (Dict): parsed yaml file

Preprocess the image input, accepted formats are PIL images, numpy arrays or pytorch tensors"

Updates paths inside attentions to the new naming scheme (local renaming)

Singlestep solver DPM-Solver-2 from time `s` to time `t`. Args: x: A pytorch tensor. The initial value at time `s`.

Function invoked when calling the pipeline for generation. Args: prompt (`str` or `List[str]`): The