songrex

Lightweight utilities for ingesting Last.fm-style scrobble CSVs and producing simple music recommendations.

Usage

• Ensure scrobbles_2022.csv and loved_Bobcat4242_2022.csv live in the repository root (or update paths).
• Run the demo:

python dataLoad.py

Files

• dataLoad.py: CSV ingestion, simple recommender, and an "abandoned songs" detector.
• scrobbles_2022.csv, loved_Bobcat4242_2022.csv: example input files (user export).
• Sandbox.py: scratchpad for experiments.

Dependencies

• Python 3.8+
• pandas

Install dependencies:

pip install pandas

Next steps

• Tweak thresholds in find_abandoned_songs() or add CLI flags.
• Replace popularity baseline with collaborative filtering or embeddings for better personalization.

Web UI

Start a lightweight web UI (Flask) to browse recommendations:

pip install -r requirements.txt
python webapp.py
# then open http://127.0.0.1:5000

Or run the Streamlit app (recommended for quick interactive data exploration):

pip install -r requirements.txt
streamlit run streamlit_app.py
# then open http://localhost:8501

License

Personal project.