Classifies music tracks into genres using an ANN trained on audio feature data.
Academic project for the Intelligent Systems course, Electronics Engineering program, Pontificia Universidad Javeriana, Bogotá D.C., Colombia.
Companion to musical-genre-classification-ga — same dataset and task, but using a neural network instead of Genetic Algorithms for comparison.
Challenge_1_Sistemas_Inteligentes.ipynb— ANN implementation, training, and evaluation
The notebook reads from /content/drive/MyDrive/Sistemas Inteligentes/train.csv (Google Drive). The dataset is not included in this repo. The companion repo musical-genre-classification-ga contains train_clean.csv which is the same dataset after cleaning.
Developed in Google Colab with TensorFlow/Keras, AutoKeras, imbalanced-learn, SHAP, and other heavy ML dependencies. Running locally requires installing all dependencies listed in the notebook imports.
pip install jupyter pandas numpy scikit-learn tensorflow
jupyter notebook Challenge_1_Sistemas_Inteligentes.ipynbSee LICENSE.
Sergio Rojas — Electronics Engineering / MSc Artificial Intelligence, Pontificia Universidad Javeriana.
This repository is archived as an academic artifact. It represents the code as delivered for the course and is not under active development. Forks are welcome under the license terms.