Args: root (string): Root directory of dataset where ``MNIST/processed/training.pt`` and ``MNIST/processed/test.pt`` exist.

Either pads or truncates an audio into a fixed length.

Loads an audio into a numpy array.

Args: root (string): Root directory of dataset where ``MNIST/processed/training.pt`` and ``MNIST/processed/test.pt`` exist.

Args: root (string): Root directory of dataset where ``MNIST/processed/training.pt`` and ``MNIST/processed/test.pt`` exist.

Converts into a tensor.

Args: root (string): Root directory of dataset where ``MNIST/processed/training.pt`` and ``MNIST/processed/test.pt`` exist.

Adds a random background noise on the frequency domain.

Dataset for silence / background noise.

Changes amplitude of an audio randomly.

Change the speed of an audio. This transform also changes the pitch of the audio.

Pytorch doesn't like complex numbers, use this transform to remove STFT after computing the mel spectrogram.

Either pads or truncates in the time axis on the frequency domain, applied after stretching, time shifting etc.

Dataset partitioning helper

Stretches an audio on the frequency domain.

A simple timeshift on the frequency domain without multiplying with exp.

Creates the mel spectrogram from an audio. The result is a 32x32 matrix.

Creates the mel spectrogram from the short time fourier transform of a file. The result is a 32x32 matrix.

Applies on an audio the short time fourier transform.

Oort's testing selector We provide two kinds of selector: select_by_deviation: testing participant selection that preserves data representati

Oort's training selector

Adds a random background noise.

Inverse short time fourier transform.

Overrides the ElasticDistributedSampler to ensure we reset the start index when an epoch is finished. This is essential since we support savi

Implementation of a Random Sampler for sampling the dataset. Added to ensure we reset the start index when an epoch is finished. This is

Basic decoder class from which all other decoders inherit. Implements several helper functions. Subclasses should implement the decode() meth

Support Vector Machine

CNN Network architecture.

Container module with an encoder, a recurrent module, and a decoder.

Stretches an audio randomly.

Shifts an audio randomly.