This dataset consists of songs collected by year, with 515345 instances (samples) and 90 attributes (variables). Songs are mostly western, commercial tracks ranging from 1922 to 2011, with a peak in the year 2000s. The task  is to predict the albums release year (i.e., V1) by the audio features transformed under MFCC.  In this competition, you have to train a model based on the train.csv file, and make a prediction about the release year of the songs for those given in test.csv.

## Files

*   **train.csv** - the training set
*   **test.csv** - the test set
*   **sample_submission.csv** - a sample submission file in the correct format

## Columns

*   `V1` - The year of the albums were published
*   `V2` to `V14` -  Timbre average
*   `V15` to `V91` -  Timbre covariance

Features extracted from the 'timbre' features from The Echo Nest API.
We take the average and covariance over all 'segments', each segment being described by a 12-dimensional timbre vector. If you are interested in the feature extraction process, you may search using keywords like MFCC and DCT.