Description
     Problem Statement   Consider the mnist dataset consisting of 50,000 training
images, and 10,000 test images. Each instance is a 28 28 pixel handwritten digit
     zero through nine. Train a (optionally convolutional) neural network for
     classification using the training set that achieves at least 95.5% accuracy on the test
set. Do not explicitly tune hyperparameters based on the test set performance,
    a validation set taken from the training set as discussed in class. Use dropout and
    an L2 penalty for regularization. Note: if you write a sufficiently general program
the next assignment will be very easy.
     Do not use the built in mnist data class from tensorflow.
Extra challenge (optional)Â Â Â In addition to the above, the student with the fewest
number of parameters for a network that gets at least 80% accuracy on the test set
will receive a prize. There will be an extra prize if any one can achieve 80% on the
test set with a single digit number of parameters. For this extra challenge you c
make your network have any crazy kind of topology you’d like, it just needs to be
optimized by a gradient based algorithm.
