Aruvansh1997 / Handwritten-Digits-Classification

Handwritten Digits Classification using Neural Networks

1. Implemented a Multilayer Perceptron Neural Network on the MNIST dataset and evaluated its performance in classifying handwritten digits using Python
2. Upon testing the network to classify the hand written digits into the training set, validation set and the test set, the network gave a high accuracy of 95% approx. within 60sec.
3. The regularization of the network is implemented using the regularization coefficient which avoids the over fitting problem.