Conversation
|
Thanks for this PR. When updating the files, please submit notebooks without output and (if you can) run ruff. |
|
@dbrakenhoff could it be that these tests are failing because the testing environment uses a newer version of numpy? |
dbrakenhoff
left a comment
There was a problem hiding this comment.
Nice work! No comments on the implementation.
My only question is why timflow isn't able to reproduce some of the results obtained with Aqtesolve/MLU? I guess we sometimes calibrate to slightly different data, and if the RMSE is similar, I guess we've found another (local) minimum? So maybe not so surprising, but if that is the case we could mention that in the comment on the final results (which I think is already done for some, but I think not all)?
|
Ready to go. Thanks for your PR. |
No description provided.