This lab targets image segmentation using deep learning.

It is intended to be very practical: instead of implementing sofisticated
algorithms, we will use a framework to do most of the work for us.

By the end of the lab, you should be able to load data, create a neural
network, train it, and use it for prediction.

We will use Google Colab, an online computational platform. It is available for
free. You just need a Google account. You can use your personal one, or the one
tied to your CTU account. Open

  https://colab.research.google.com/

and log in.

If you do not have a Google account, you can use Jupyter on your computer as
usual. However, it might take ages to train the network using a low-end GPU or
even a CPU.

