https://github.com/hashlag/spotify_tracks_popularity

https://github.com/hashlag/spotify_tracks_popularity

Last synced: 3 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/hashlag/spotify_tracks_popularity
• Owner: hashlag
• Created: 2023-08-18T12:13:45.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-08-20T19:52:40.000Z (about 1 year ago)
• Last Synced: 2023-08-20T20:32:27.336Z (about 1 year ago)
• Language: Python
• Size: 15 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Spotify track popularity prediction with linear regression

This model tries to predict the popularity score of a track on Spotify based on

[Spotify tracks DB dataset](https://www.kaggle.com/datasets/zaheenhamidani/ultimate-spotify-tracks-db)

## Run model

You should have `pandas`, `torch` and  `sklearn` installed.

Then you can navigate to the project directory and run the model with:

```

python linear.py

```

## Testing

Here is a visualization of MAE change during training

![train loss plot](https://raw.githubusercontent.com/hashlag/spotify_tracks_popularity/main/train_loss_plot.png)

To get actual MAE on testing data just run the `linear.py` script.