Program implements a convolutional neural network for classifying images of numbers in the MNIST dataset as either even or odd using GPU framework.

1. Contructing CNN: MNIST dataset is split into training, validation, and testing subsets with 55000, 10000, and 5000 examples, respectively. Digit labels are converted to even or odd using 0 and 1. Also plotting training and validation loss and accuracy as a function of epochs.
2. Hyper-parameters Tuning: Evaluating different variations of basic network by changing:
   • network architecture
   • receptive field using dilation
   • stride
   • optimizer
   • loss function
   • dropout
   • learning rate
   • number of epochs
   • weight initializers
   • batch normalization
   • layer normalization
3. Inference: Taking images with handwritten digits as input and performing image preprocessing like resizing, converting to grayscale and then to binary image using threshold. Classifying input images into odd/even labels.

Program was implemented using Python, TensorFlow, Keras, and OpenCV. Refer the report for further implementation details and instructions to run the code: View Report

1. Contructed CNN for training:

2. After Hyper-parameter Tuning:

3. Inference with handwritten images: