Comparing Meta-Classifiers for Automatic Music Genre Classification

Source code from proceding Comparing Meta-Classifiers for Automatic Music Genre Classification published at 17th Brazilian Symposium on Computer Music (SBCM).

Getting Started

These instructions will get you a copy of the project up and running on your local machine for experiment reproducibility.

Prerequisites

• Python 3.x
• Libraries listed here

Installing

A step by step series of examples that tell you how to get everything set up to run the experiment.

1. The first step is clone this repository.

$ git clone https://github.com/vitorys/MusicGenreMetaClassifier.git

2. (Optional) Create a virtual enviroment and activate it.

$ virtualenv venv && activate venv/bin/activate

3. Install the requirements.

$ pip install -r requirements.txt

Running experiments

To run the Neural Network experiments:

1. Navigate to NeuralNetwork folder.
2. Execute the python file main.py follow by some dataset. For example:

python main.py data/gtzan-ds_rp-feats_frames

3. The result will be stored at output/ folder.

To run the Hidden Markov experiments:

1. Navigate to HMM folder.
2. Execute the python file main.py follow by --input and some dataset. For example:

python classifier.py --input data/gtzan-ds_rp-feats_frames

4. The result will be stored at output/ folder.

Authors

• Vítor Yudi Shinohara - State University of Campinas
• Juliano Henrique Foleiss - Federal University of Technology - Paraná
• Tiago Fernandes Tavares - State University of Campinas