An ensemble of homogeneous models to classify fashion products

In this example, we'll use the Fashion-MNIST dataset. This dataset has 60,000 images of fashion products from ten categories. The target variable can be classified into ten classes:

• T-shirt/top
• Trouser
• Pullover
• Dress
• Coat
• Sandal
• Shirt
• Sneakers
• Bag
• Ankle boot

Each image is a 28 x 28 grayscale image. We will proceed by reading the data to build a few homogeneous models over a few iterations to see whether the ensemble can deliver a higher accuracy.

Getting ready

We'll use Google Colab to train our models. Google Colab comes with TensorFlow installed, so we don't have to install it separately in our system.

We import the required libraries...