Skip to content

In this repository, the composer classification problem is described and a new network architecture based on a combination of ConvNets and GRUs is proposed to address the problem. Methods of data augmentation are described and a trend of increasing accuracy with augmentation factor is reported. The best test set accuracy obtained is 72.1%.

Notifications You must be signed in to change notification settings

beginner46/WhoComposedIt

About

In this repository, the composer classification problem is described and a new network architecture based on a combination of ConvNets and GRUs is proposed to address the problem. Methods of data augmentation are described and a trend of increasing accuracy with augmentation factor is reported. The best test set accuracy obtained is 72.1%.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 96.4%
  • Python 3.6%