Identifying the face expression of a human, given an image of him/her.
Data is taken from Kaggle's Facial Expression Recognition Challenge: https://www.kaggle.com/c/challenges-in-representation-learning-facial-expression-recognition-challenge/data
Evaluation is done based on accuracy and loss between predicted expression and actual expression.
Some information about the data:
- We're dealing with images(unstructured data), so its better we use deep learning / transfer learning.
- Dataset has 3 columns namely emotions, picture, and usage(Training/Testing).
- Dataset has 35887 rows(images).
- There are 28709 training images (with column value as Training).
- There are 3589 testing images (with column value as PublicTest).
- Dataset has 7 classes (emotions).
- 0 -> Angry
- 1 -> Disgust
- 2 -> Fear
- 3 -> Happy
- 4 -> Sad
- 5 -> Surprise
- 6 -> Neutral