Takes a state dict and a config, and returns a converted checkpoint.

A counter for the `thop` package to count the operations in an attention operation. Meant to be used like: macs, params = tho

A counter for the `thop` package to count the operations in an attention operation. Meant to be used like: macs, params = tho

Creates a dynamic module in the cache directory for modules.

Creates a config for the diffusers based on the config of the LDM model.

Return the data prediction model (with thresholding).

r""" Download and cache a PyTorch diffusion pipeline from pre-trained pipeline weights. Parameters: pretrained_model_

Forward evaluation data and calculate statistics. Args: data_loader: object Returns: statistics: dict,

Generate captions given text embedding features. Returns list[L]. Args: features (`torch.Tensor` of shape `(B, L, D)

embed_dim: output dimension for each position pos: a list of positions to be encoded: size (M,) out: (M, D)

embed_dim: output dimension for each position pos: a list of positions to be encoded: size (M,) out: (M, D)

grid_size: int of the grid height and width return: pos_embed: [grid_size*grid_size, embed_dim] or [1+grid_size*grid_size, embed_dim]

grid_size: int of the grid height and width return: pos_embed: [grid_size*grid_size, embed_dim] or [1+grid_size*grid_size, embed_dim] (w/ o

grid_size: int of the grid height and width return: pos_embed: [grid_size*grid_size, embed_dim] or [1+grid_size*grid_size, embed_dim]

Get the audio embedding from the model Parameters ---------- data: a list of dict the audio input dict list

Convert the continuous-time `t_continuous` (in [epsilon, T]) to the model input time. For discrete-time DPMs, we convert `t_continu

Computes the solver order at each time step. Args: num_inference_steps (`int`): the number of diffu

Return set of symbol pairs in a word. Word is represented as tuple of symbols (symbols being variable-length strings).

Get the list of modules that are relatively imported in a module file. Args: module_file (`str` or `os.PathLike`): The module fi

Get sample rate: Returns: `int`: sample rate of audio

Get tar path from dataset name and type

Get the text embedding from the model Parameters ---------- data: torch.Tensor a tensor of text embedding

Formats a user-agent string with basic info about a request.

A piecewise linear function y = f(x), using xp and yp as keypoints. We implement f(x) in a differentiable way (i.e. applicable for autograd

Checking for safetensors compatibility: - By default, all models are saved with the default pytorch serialization, so we use the list of de

enumerate available model architectures based on config files

r""" Load pretrained attention processor layers into `UNet2DConditionModel`. Attention processor layers have to be defined in

Loads an ONNX Inference session with an ExecutionProvider. Default provider is `CPUExecutionProvider` Arguments: pat

r"""Load trained universal source separation model. Args: configs (Dict) checkpoint_path (str): path of the checkpoint to loa

Returns the log probabilities of the rows from the (cumulative or non-cumulative) transition matrix for each latent pixel in `x_t`.

Multistep DPM-Solver with the order `order` from time `t_prev_list[-1]` to time `t`. Args: x: A pytorch tensor. The

Return the noise prediction model.

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