In this notebook, image classification is performed on the CIFAR 10 dataset using Convolutional Neural Networks.
CIFAR 10 dataset is used for object recognition in images.It consists of 60000 32x32 colour images in 10 classes, with 6000 images per class.
The validation accuracy obtained with this is 75.4%
In the second notebook, the ResNet architecture is used. Along with that data transformations are also applied such as cropping, horizontal flipping,etc.
After training for 16 epochs with optimisation function as Adam Optimizer and other parameters, the final validation accuracy achieved is 93.13%.