Applying the RAM on a noisy MNIST sample

To understand the working of the RAM in greater detail, let's try to create an MNIST sample incorporating some of the problems as highlighted in the earlier section:

Larger image of noisy and distorted MNIST

The preceding image represents a larger image/collage using an actual and slightly noisy sample of an MNIST image (of number 2), and a lot of other distortions and snippets of other partial samples. Also, the actual digit 2 here is not centered. This example represents all the previously stated problems, yet it is simple enough to understand the working of the RAM.

The RAM uses the concept of a  ...

Get Practical Convolutional Neural Networks now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.