search

oarriaga / face_classification

Real-time face detection and emotion/gender classification using fer2013/imdb datasets with a keras CNN model and openCV.
MIT License
5.61k stars 1.59k forks source link

About your model when emotion classification #6

Closed guoxiaolu closed 7 years ago

guoxiaolu commented 7 years ago

impressive job! You have mentioned "fer2013 emotion classification test accuracy: 66%.". To my knowledge, this result is close to the state of art. But I wanna know some information about your training data. Is it fer2013 train or some other data. And the training data total number is ? Thank you very much, it is helpful for me~

oarriaga commented 7 years ago

Hello @guoxiaolu, Thank you, yes I believe the current state-of-the-art is around 70%. I am only using the fer2013 training /test data. I divide the training/test samples in the model.fit function with a 80/20 split. This should divide it accordingly to the original train/test images from the dataset.

guoxiaolu commented 7 years ago

Thank you for your reply~