/audio_classifier_example

Example repository for the CV4Ecology 2022 Summer School.

Primary LanguagePythonMIT LicenseMIT

Bird Vocalization Classification

Example repository for the CV4Ecology 2022 Summer School to train and test deep learning models on audio data, using the SSW60 Dataset.

Installation instructions

  1. Install Conda

  2. Create environment and install requirements

conda create -n cv4ecology python=3.8 -y
conda activate cv4ecology
pip install -r requirements.txt
  1. Download dataset
sh scripts/download_dataset.sh

This downloads the SSW60 Dataset to the datasets/ssw60 folder.

Reproduce results

  1. Train
python audio_classifier/train.py --config configs/exp_resnet18.yaml
  1. Test/inference

@CV4Ecology participants: Up to you to figure that one out. :)