AudioML-pipe

Features

• Full machine learning pipeline with data preprocessing, and model training and evaluation.
• Model and training loop implemented in PyTorch.
• Examples of configuration files in config/data/ and config/models/.
• Examples of already processed datasets in data/processed/, as well as test audio files data/test/.
• Examples of already trained models in models/.

Requirements

• Python 3.10.8
• virtualenv 20.0.17

Installation

Clone the repository into your working directory

git clone https://github.com/beltran99/AudioML-pipe.git

Navigate to the repository folder

cd AudioML-pipe

Install virtualenv if you don't have it installed already

pip install virtualenv

Create a new virtual environment called env

virtualenv env

Activate the environment

source env/bin/activate # in Linux

Installed the required packages in the virtual environment

pip install -r requirements.txt

Example usage

Create a dataset

python3 -m src.data.make_dataset --data-path data/raw/ --config-path config/data/data_20f_agg.yml --name 20f_agg -v

Train a model

python3 -m src.pipeline.train --data-path data/processed/20f_agg.npz --config-path config/models/config_20f.yml -r -p -v

Test the trained model

python3 -m src.pipeline.inference --model models/mymodel_20_coeff_v010 --data-path data/test/ --config config/data/data_20f_agg.yml -r -v