mixed-farming / MNIST-digit-classification

Recognizing handwritten digits of the MNIST dataset featuring a deep learning model, providing a comprehensive solution for training, testing, and evaluating digit recognition.

MNIST-digit-classification

Project: Recognizing handwritten digits

• We will discover the MNIST handwritten digit recognition problem and we will develop a deep learning model in Python using the Keras API with Tensorflow backend.
• The MNIST dataset is a collection of handwritten digits from 0-9. It consists of 70,000 labeled 28 x 28 pixel grayscale images of hand-written digits. The dataset is split into 60,000 training images and 10,000 test images. There are 10 classes (one for each of the 10 digits). The task at hand is to train a model using the 60,000 training images and subsequently test its classification accuracy on the 10,000 test images.
  
• Every image is of size 28 x 28 pixels.
  
• The model is 2 layer (hidden layers) sequential fully connected neural network as depicted below.
  

Libraries used:

• Tensorflow
• Keras
• Numpy
• Matplotlib

Evaluation:

The model gets loss a of 0.0497 and an accuracy of 98.55% while evaluating on the test set. If you don't have a GPU powered machine it might take a little longer, you can try reducing the epochs (iterative steps) to reduce computation.

References:

• Coursera
  Course ID: https://www.coursera.org/projects/tensorflow-beginner-basic-image-classification
• https://mxnet.apache.org/versions/1.4.1/tutorials/python/mnist.html