"Deep Learning in the Eye Tracking World"

the tutorial presented during ETRA 2019

Denver, June 25, 2019

Pawel Kasprowski

Tutorial files:

• Slides in PDF format
• Code examples - feel free to clone or fork

---

The tutorial in brief:

• Very simple examples (max 100 lines of code)
• Not so many formulas (only a few)
• Just the working code (in Python/Keras)
• Simple datasets to show that it works (hopefully...)

Even if you don't have any experience with Machine Learning you will be able to start your own experiments right after the tutorial.

---

Prerequisites to run the examples:

1. Install Python 3.x 64 bit version
• Windows 10: https://www.python.org/ftp/python/3.7.3/python-3.7.3-amd64.exe
• macOs: https://www.python.org/ftp/python/3.7.3/python-3.7.3-macosx10.9.pkg

Remember about adding Python directories to the default path – it is possible in the installer by checking the appropriate check box on the first page of the installer.

2. Open the terminal (cmd for Windows, Terminal.app for macOs) and execute:

pip install -r http://www.kasprowski.pl/tutorial/requirements.txt

[it will take some time to load and install all the required libraries]

Then you may test your environment by typing:

python -c "import tensorflow as tf; print(tf.__version__)"

If you see something like: "1.13.1", you are ready for the tutorial!

Note, that if you are an experienced user and you are already using Python in other projects, you should probably start with creating the virtual environment using the command:

python -m venv DIR

Check: https://docs.python.org/3/tutorial/venv.html for details.

Moreover, if you want to stay with Python for a little bit longer, you should consider installing an IDE tool that will simplify your work, like PyCharm or (my favorite) PyDev.

Any comments would be highly appreciated - pawel{at}kasprowski.pl