index.md

layout	mathjax	title
default	true	TFLR

<script type="text/javascript" src="http://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML"> </script>

Talk

• Talk given on 9/20/2018 at Inside Data Konstanz

Notebooks

The easiest to interactively work with the notebooks, is to use google colab (you will need to have a google account for that). In colab the notebook will be run in the cloud. For doing the hands-on part on your own computer you can either install anaconda or use the provided a docker container (you need no google account). See https://tensorchiefs.github.io/dl_course_2018/ for details.

Tasks / Exercises

• Ninja track (very few hints) linear_regression_tf: notebook static or notebook interactive in colab

• Yellow belt (some hints) linear_regression_tf: notebook static or notebook interactive in colab

If there is time left, you may want to have a look at the solution for linear regression with TensorFlow eager (see below)

Solutions

Please try it first on your own

• Linear Regression with TensorFlow (standard graph based): notebook static or notebook interactive in colab

• Linear Regression with TensorFlow eager: notebook static or notebook interactive in colab

Further material on deep learning, can be found at tensorchiefs $F^{\mu,\nu}$