search

StephanRhode / py-algorithms-4-automotive-engineering

This repository contains jupyter notebooks and python code for KIT course: Python Algorithms for Automotive Engineering
http://www.fast.kit.edu/lff/1017_13056.php
MIT License
63 stars 95 forks source link

DeepL notebook #8

Closed StephanRhode closed 4 years ago

StephanRhode commented 4 years ago

Hi @mauricio-fernandez-l ,

great that you are on board. I created the template notebook 04_mini-projects/09_deepl.ipynb on branch deepl. This notebook has some default figures and is tested every time you push in Travis CI. The link to respective Travis is in readme and here for convenience Travis for this repo

You can test all notebooks with pytest and you can create presentation slides with make_slides.py in root folder. All is described in readme. If you need additional python packages, please add them in requirements.txt.

With respect to topic, it would be nice if we could do something with traffic sign recognition, but feel free. Add to this, it would be great if the students could solve some small problems by themselves.

Please let me know if you struggle with the setup of this repo or need any support.

mauricio-fernandez-l commented 4 years ago

Hi Stephan!

Sorry for the delay. I will work on the lecture this and the next week. I checked out the deepl branch and will edit the notebook 04_mini-projects/09_deepl.ipynb. I tested locally the script make_slides.py and everything works fine on my end. I will also add tensorflow to the requirements.txt.

With respect to the examples, I generated today some synthetic image data for traffic sign recognition. I think that will provide a good and relevant example for deep convolutional networks. I will keep you updated.

Best,

Mauricio

StephanRhode commented 4 years ago

Sounds great, just go ahead and me know if I can help somewhere.

mauricio-fernandez-l commented 4 years ago

@StephanRhode I just uploaded the draft for the lecture and created the pull request. Can you take a look into it when you have time?

StephanRhode commented 4 years ago

Is resolved in pull request #12