Hand-written-digit-recognition

A mini project for Data Science assignment.

This is a 2 layers Sequential Convolutional Neural Network for digits recognition trained on MNIST dataset. It uses TensorFlow with keras API (Tensorflow backend) which is very intuitive

Steps we followed

1. Load data from MNIST Dataset

2. Data Exploration

3. Dataset Inspection

4. Data Preprocessing

5. Models Used

   a) Simple Neural Network

   b) Simple Convolutional Neural Networks(CNN)

   c) CNN with Dropouts

   d) CNN with Dropouts and Batch Normalization

6. Precision and recall and Confusion Matrix of each model

7. Results