autrainer#
A Modular and Extensible Deep Learning Toolkit for Computer Audition Tasks.
autrainer is built on top of PyTorch and Hydra, offering a modular and extensible way to perform reproducible deep learning experiments for computer audition tasks using YAML configuration files and the command line.
Installation#
To install autrainer, first ensure that PyTorch (along with torchvision and torchaudio) version 2.0 or higher is installed. For installation instructions, refer to the PyTorch website.
It is recommended to install autrainer within a virtual environment. To create a new virtual environment, refer to the Python venv documentation.
Next, install autrainer using pip.
pip install autrainer
The following optional dependencies can be installed to enable additional features:
latexfor LaTeX plotting (requires a LaTeX installation).mlflowfor MLflow logging.tensorboardfor TensorBoard logging.opensmilefor audio feature extraction with openSMILE.albumentationsfor image augmentations with Albumentations.audiomentationsfor audio augmentations with audiomentations.torch-audiomentationsfor audio augmentations with torch-audiomentations.
pip install autrainer[latex]
pip install autrainer[mlflow]
pip install autrainer[tensorboard]
pip install autrainer[opensmile]
pip install autrainer[albumentations]
pip install autrainer[audiomentations]
pip install autrainer[torch-audiomentations]
To install autrainer with all optional dependencies, use the following command:
pip install autrainer[all]
To install autrainer from source, refer to the contribution guide.
Next Steps#
To get started using autrainer, the quickstart guide outlines the creation of a simple training configuration and Tutorials provide examples for implementing custom modules including their configurations.
For a complete list of available CLI commands, refer to the CLI reference or the CLI wrapper.
Overview#
Modules
Development