Keras code for Emotion Recognition using Facial Expresions. The code contains a Standard Convolutional Neural Network that detects 5 different types of Emotions in Humans:
[sad, happy, angry, surprise and Neutral]
The dataset used is the Facial Expression Recognition (FER) dataset from Kaggle. The link to the data is given below:
https://www.kaggle.com/deadskull7/fer2013
Open the Emotion recognition jupyter notebook, specify the apprpriate paths for Images and csv file. Run the training. After the training is finished, the model will get saved as model.json and weights will be saved as modelfer.h5
Open the load model jupyter notebook (loadmodel.ipynb) and provide the paths for saved model and weights. Run the notebook. Yipee!