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

r"""Puts each data field into a tensor with outer dimension batch size

r"""Puts each data field into a tensor with outer dimension batch size

:param h: height :param w: width :return: normalized distance to image border, wtith min distance = 0 at border

:param h: height :param w: width :return: normalized distance to image border, wtith min distance = 0 at border

:param h: height :param w: width :return: normalized distance to image border, wtith min distance = 0 at border

r"""Load list of class labels and return text embeddings

Restore the parameters stored with the `store` method. Useful to validate the model with EMA parameters without affecting the

Save the current parameters for restoring later. Args: parameters: Iterable of `torch.nn.Parameter`; the parameters to b

Initialization. INPUT: - in_features: shape of the input - alpha: trainable parameter alpha is in

r"""Load preprocessed audio and return a audio embeddings

r"""Load preprocessed audio and return a audio embeddings

r"""Load preprocessed text and return text embeddings

r"""Load preprocessed text and return text embeddings

each file may have different caption, we add num to filename to identify each audio-caption pair

Create a 1D, 2D, or 3D average pooling module.

r"""Compute similarity between text and audio embeddings

r"""Load list of audio files and return a audio embeddings

This matches the implementation in Denoising Diffusion Probabilistic Models: From Fairseq. Build sinusoidal embeddings. This matc