Skip to content

Kaggle Facial Expression Recognition 2013 Using TensorFlow

Notifications You must be signed in to change notification settings

nhduong/fer2013

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

24 Commits
 
 
 
 

Repository files navigation

Kaggle Facial Expression Recognition 2013

Author: Nguyen Hai Duong

Using CNN with TensorFlow

How to use:

  1. Download all files from this repo
  2. Download the dataset from here (fer2013_training_onehot.csv, and fer2013_publictest_onehot.csv): https://drive.google.com/drive/folders/15EozdYlGeO3d-ny8nbqFM3ojTGsj2hdE?usp=sharing
  3. Run fer2013.ipynb with Jupyter, Python 2.7, and TensorFlow 1.

Note: for testing only, you can download the trained model (fer2013_weights.h5) from the link above

Accuracy

Classes Accuracy
angry 52.46%
disgust 62.50%
fear 46.77%
happy 78.88%
sad 44.87%
surprise 80.00%
neutral 55.52%
overal 60.74%

About

Kaggle Facial Expression Recognition 2013 Using TensorFlow

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published