You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
If only pytest-harvest had already existed when we were working on MEMOTE. We found a fairly hackish way by storing information on the test function itself and then collecting that info with a plugin, but the result_bag makes so much more sense! And it even works with xdist.
I didn't know where else to communicate besides issues, so this is just to say, your package looks awesome. I wish we had it back then. I'm now considering re-writing MEMOTE, but I don't really have the time right now.
The text was updated successfully, but these errors were encountered:
This is a late answer but your kind words are very much appreciated ! Thanks a lot for taking the time to write this.
MEMOTE is exactly the kind of usage that I had in mind when I developed pytest-harvest and pytest-cases: data science benchmarks that are reproducible by anyone. I'd be curious to see where you actually store the reference datasets - that is the piece where I do not have an elegant solution, apart from using a blob storage or a data catalog such as OpenDataSoft (maybe using https://smarie.github.io/python-odsclient/ :) )
Anyway, kudos for MEMOTE and thanks again for your message!
If only pytest-harvest had already existed when we were working on MEMOTE. We found a fairly hackish way by storing information on the test function itself and then collecting that info with a plugin, but the
result_bag
makes so much more sense! And it even works with xdist.I didn't know where else to communicate besides issues, so this is just to say, your package looks awesome. I wish we had it back then. I'm now considering re-writing MEMOTE, but I don't really have the time right now.
The text was updated successfully, but these errors were encountered: