A Deep Learning Model which would able to predict your emotions just by looking to your face. The Accuracy it reach upto 55% . It train on 22500 data and on 6 different class. It can also detect multi-face emotion from a image or frame.
I didn't add "X.pkl" file because it is a large file around 2 gb . you get both X and Y_binary files by just running preprocessing.ipynb
'X.pkl' and 'Y_binary.pkl' are file of our data preprocessed. it will very usefull for our training process.
Total 6 class has been used for it. Happy,Sad,Angry,Neutral,Fear and Surprise.
Loss :- binary_crossentropy
Activatiion on last layer :- softmax
