The speciality of this project is that I have only used Multi-layer Perceptron model for image classification. (Not CNN)
The code is well documented with comments and interpretation of results is given in the report.
This code demonstrates the learning process. We don't have any scientific way to design Neural networks and what we have to do is come up with different settings of hyperparameters. This code depicts what approaches shall we take when building a neural net model.
In simple_image_classifier project I have implemented a simple CNN network for image classification that gives accuracy of 80%.