First CNN projects done after learning about them. Open the files in Colab from the link at the top to view the results.
#1 Cifar10 Classification:
A CNN that studies the Cifar10 dataset and classifies the images into its respective 10 categories with around 80% accuracy.
#2 Cifar10 AutoEncoder:
A CNN that encodes the images in the Cifar10 dataset with minimal loss in legibility.
#3 FMNIST:
A CNN that studies the Fashion MNIST dataset and classifies the images into the 10 categories of fashion items available with 90% accuracy.
#4 Psychometric Functions:
A CNN that studies generated horizontal/vertical lines on gaussian noise than classifies lines generated with a slope of -30 to 30 into the categories of horizontal / vertical.
#4 Style Transfer VGG19
Using the pretrained Vgg19 model to display style transfer from a Vincent Van Gogh painting to the picture of Stedelijk Museum in Amsterdam.