Skip to content

Latest commit

 

History

History
45 lines (25 loc) · 1.28 KB

README.md

File metadata and controls

45 lines (25 loc) · 1.28 KB

decadance

Core Code School - Machine Learning Bootcamp Final Project

Project

The objective of this project is to make an analysis of the audio features of songs between the 40s and the 21s decades.

Dataset

Dataset used for this project is retrieved from the SpotifyAPI.

🪄 How is the dataset created?

By using the Search API in Spotify, querying by range of years from 1900 to 2021 od release date of tracks, filtered by genre pop

You can generate the csv and download the mp3 files used for the project by running:

$ python data/spotify

This will create the dir csv and mp3 with the datasets and mp3 files of the preview-url of the tracks.

The dataset will be a collection of tracks of pop genre witch have been released between 1900 to 2021, tagged with the decade.

📆 Consideration on decade

According to Wikipedia, a decade is:

Any period of ten years is a "decade"

There are two methods to group the years on decades:

  • 0-to-9 decade
  • 1-to-0 decade

For the purpose of this project, the first one is considered, so a decade will be all the years from 0 to 9.

For example:

  • decade 1900 - from 1900 to 1909
  • decade 2010 - from 2010 to 2019

You can read more about this here