The problem Fashion Garment Recognizer solves

Developed a custom CNN (Convolutional Neural Network) model for classifying the images of the fashion_mnist dataset containing 60,000 training images and 10,000 testing images.
This dataset consists of 28 x28 grayscale images from ten different classes of fashion clothes.
The different classes are: T-shirt/top, Trouser, Pullover, Dress, Coat, Sandal, Shirt, Sneaker, Bag, Ankle boot.
The images are stored in the form of pixels with values ranging from 0 to 255.
Accuracy achieved : 93.49%
The total number of correctly and incorrectly classified images was also calculated and the figures for the same are: 9349 and 651.

Challenges I ran into

Improving accuracy.