This project presents a learning based model for identifying the configuration of a given chess board through image classification.
We have about 40,000 training images and 4,000 testing images. There is also a CSV file which consists of the FEN of each image in the dataset. The dataset used can be found here: https://www.aicrowd.com/challenges/ai-blitz-6/problems/chess-configuration
A three layer Convolutional Neural Network is used as the model to predict the piece present in the cell.