Where everything lives
The code for this is stored in gitlab, under
deep-sets-example.
Getting the image you need
I’ve created a less annoying docker image, but we still require some
local directories to be mounted. Clone
the analysisbase-launcher recursively.
The README has instructions. All your data and other work that you
want to use both inside and outside the image should go in the
/work/ directory.
Python training image
For most of what we’re doing I’d recommend using pip and working on your local machine, but if you have issues setting that up you can also use the Atlas Machine Learning Forum’s images. You can get a good starting point by running:
docker pull atlasml/ml-base:py-3.7.2
Getting the files you need
Unfortunately I couldn’t figure out how to do something interesting with the example file we had, so you’ll have to download the one from my public area on EOS.