Closed williamjamir closed 7 months ago
It looks like the PR checks are not being executed =/
Perhaps it's something that you need to activate or something like that?
It looks like the PR checks are not being executed =/
Perhaps it's something that you need to activate or something like that?
I still need to approve those to run on upstream. Still, they are running in your repo. But there's this weird error 255 which so far I couldn't figure out what it is. When I have more time I'll take a look at it.
It looks like the PR checks are not being executed =/ Perhaps it's something that you need to activate or something like that?
I still need to approve those to run on upstream. Still, they are running in your repo. But there's this weird error 255 which so far I couldn't figure out what it is. When I have more time I'll take a look at it.
I found the issue, some tests were using an import pandas.util.testing
that was deprecated since 1.0, and with version 2.0 it was removed.
Also assert_almost_equal
is not available anymore.
So, it's safe to use that import with the current pandas version pinned on the requirements.txt and the numpy equivalent of assert_almost_equal
I just don't know why pandas 2 was installed just now, and not on others runs 🤷
I'm not sure what this new error is about; locally, the dtype matches only on CI it is changed to int32.
Also, I added check_dype=False
, but the test still fails about dtype
That's some good news. On the bad side it will require a more thorough investigation of what's happening. Hopefully I'll be able to also look into that on this weekend.
Hi @williamjamir ,
I just fixed the windows issue on your branch, could you please confirm if everything is working as expected? As soon as you reply back with a positive I'll merge this patch already.
Thanks for this contribution!
Thanks for the fix!
Confirmed! Just tested locally on Mac M2 with Python 3.11, and your patch works flawlessly.
Ready for the merge 🚀
Close https://github.com/WillianFuks/tfcausalimpact/issues/79
Add support for Python 3.11, TensorFlow has added support for on v2.12