Music Classifier & Recommendor

A project to classify music genres and generate personalized playlists based on audio features.

Table of Contents

• Prerequisites
• Getting Started
  • Steps to See Results
  • Bonus: Generate Playlists
• Data Download (IMPORTANT)

Prerequisites

Before running the notebooks, ensure you have the following installed:

• Python 3.x
• Required Python libraries:
  • pandas
  • numpy
  • scikit-learn
  • librosa
  • h5py
  • tqdm
  • tensorflow

Getting Started

Steps to See Results

1. Run music_preprocessing.ipynb

   • Description: Processes data from the songs in the Test_Songs directory.

2. Run music_classifier.ipynb

   • Description: Builds the Neural Network model.
   • Note: This step takes approximately 6 minutes, depending on your PC's performance.

3. Run song_Info.ipynb

   • Description: Classifies songs into genres.

Bonus: Generate Playlists

4. Run playlistGenerator.ipynb
   • Description: Creates a playlist based on the subset of music data used.
   • Important: This step requires additional data files (over 1GB) to function correctly.

Data Download

Before Running, Download These Files:

• Music Data Files

Note: Please make sure you have downloaded and that you put these files in the right directory before running the playlistGenerator.ipynb notebook.