Note on the tutorials
The tutorials are presented by Thomas Fischbacher.
Technical
- Bring your laptop with you
- For internet access (use WLAN:
guest-ZHAW) you will need a mobile phone to recieve a text-message with the login code - The easiest way to take part in the tutorials is to use google colab (you need to have a google account for this)
- Alternatively you might install the necessery tools as descriped in https://tensorchiefs.github.io/dl_course_2018/
Notebooks
Lecture notes
- tf.eager() background and history pdf some demo code in colab
- tf.einsum() Einstein summation convention in tf and numpy pdf
The notebook
The use of both tf.eager() and tf.einsum() is demonstrated in the following notebooks.
-
TfEagerDemo to open directly in colab
-
TfEagerDemo Solutions to open directly in colab
Additional resources
- Offical Google page on tf.eager
- Google AI blog on tf.eager
- Alternative notebook to introduce Einstein Summation to open directly in colab provided by tensorchiefs $F^{\mu \nu}$
- Offical Google page on tf.einsum