ML-CIFAR100
The main goal of this project is to perform binary classification of images in two assigned superclass pairs given to our group. 
The robustness of the algorithm is tested against sub-class image data hidden during the training stage.
Milestone 1 : Prediction on randomly selected testing images
Milestone 2 : Prediction on one testing subclass images from each of the two superclasses
Milestone 3 : Prediction on two testing subclass images from each of the two superclasses

About the dataset
The CIFAR-100 dataset has 100 classes containing 600 images each. 
There are 500 training images and 100 testing images per class. 
The 100 classes in the CIFAR-100 are grouped into 20 super classes. 
Each image comes with a "fine" label (the class to which it belongs) 
and a "coarse" label (the superclass to which it belongs) (CIFAR 100 DATASET).

Super Class	Class
Vehicle 1	Bicycle, Bus, Motorcycle, Pickup truck, Train
Vehicle 2	Rocket, Tank, Tractor, Streetcar, Lawn-mower

Results
“lawn_mower” has the most resembling features that could be learned by our CNN model through vehicle 2 subclasses 
and “motorcycle” has the most resembling features that could be learned by our CNN model through vehicle 1. 
The combinations with these two for predictions on two testing subclass images from each of the two superclasses gave consistent accuracies    

“rocket” has the odd one out features in vehicle 2 which our CNN model didn’t learn 
and “train” has the odd one out features in vehicle 1 which our CNN model didn’t learn 
and thus the combinations associated with that pair gave poor accuracies consistently.