- pip install numpy
- pip install opencv-python
- pip install keras
- pip3 install --upgrade tensorflow
- pip install pillow
- from below link and put in data folder under your project directory
- https://www.kaggle.com/msambare/fer2013
- with all face expression images in the FER2013 Dataset
- command --> python TrainEmotionDetector.py
It will take several hours depends on your processor. (On i7 processor with 20 GB RAM it took me around 2 hours) after Training , you will find the trained model structure and weights are stored in your project directory. emotion_model.json emotion_model.h5
copy these two files create model folder in your project directory and paste it.
python TestEmotionDetector.py