Skip to content

LahiruJayasinghe/DCASE2018

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

DCASE 2018

An ensemble learning system based on convolutional neural network and long short memory recurrent neural network. With extracted features such as spectrogram and mel-frequency cepstrum coefficients from different channels, the proposed system can classify different domestic activities effectively. Our model achieved 92.19% F1-score performance, which is a 7.69% performance gain when compared with its baseline system.

Releases

No releases published

Packages

No packages published

Languages

  • HTML 54.2%
  • Jupyter Notebook 45.3%
  • Python 0.5%