The tutorials and exerises are best used with the Praktikum lectures and for someone took the summer-semester KIT Neuronale Netze course. For more information, please visit https://ilias.studium.kit.edu/ or http://isl.anthropomatik.kit.edu/english/2667_2747.php

This guideline is for the ones who choose the third option (the own environment and lab machines, eventually you all would need this), for the help on Google Colab and/or Kaggle Kernel (at the beginning phase and free GPUs), please visit the ILIAS of the PNN WS18/19.

I highly recommend installing Anaconda (a python scientific package bundle) and use its environment managing feature.

• Anaconda with python 3.5 (or 3.6)
• CUDA drivers & toolkit 8.0 (if your workstation is Nvidia GPU-enabled or you use our lab machines)

If you are not sure what you are doing, please refer to the steps 1-5 from https://towardsdatascience.com/setup-an-environment-for-machine-learning-and-deep-learning-with-anaconda-in-windows-5d7134a3db10

The tutorials and exerises have been published as Jupyter Notebooks. If you haven't used Jupyter Notebooks before, here is a good tutorial for beginners: https://www.dataquest.io/blog/jupyter-notebook-tutorial/

Install inside the environment those python packages which are needed thoughout the class:

• matplotlib
• seaborn
• scikit-learn
• other stuffs that you need..

We use Pytorch 0.4, not 1.0 or other versions.

• Create an Anaconda environment
• Activate it
• Install pytorch in a suitable way (conda, python version, GPU or not)

If you want to run the sample codes written in Tensorflow and Keras, you need to at first install them. Take a look on Step 7 from the link above.

We use Tensorflow 1.11 and Keras with Tensorflow backend.

Many ideas and examples in this course, especially the exercises are inspired from:

• Deep Learning Specialization, Andrew Ng et al., Coursera (https://www.coursera.org/specializations/deep-learning)
• Neural Networks for Machine Learning Course, Geoffrey Hinton et al., Coursera (https://www.coursera.org/learn/neural-networks)
• CS231N Convolutional Neural Networks for Visual Recognition, Fei-Fei Li et al., Stanford University. (http://cs231n.stanford.edu/)
• CSC 321 Intro to Neural Networks and Machine Learning, Roger Grosse et al., University of Toronto. (http://www.cs.toronto.edu/~rgrosse/courses/csc321_2018/)

If you have any question, please visit the ILIAS of the PNN WS18/19 or write an email to [email protected]