feat: Support Dataframes as Fields in Pydantic Models #148
Conversation
ytausch
requested review from
borchero,
AndreasAlbertQC and
delsner
as code owners
|
|
||
| @classmethod | ||
| @abstractmethod | ||
| def validate( |
There was a problem hiding this comment.
I had to move this here as an abstract class method because the generic argument to dy.Dataframe is only bound to BaseSchema, but I need to be able to call validate on it.
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #148 +/- ##
=========================================
Coverage 100.00% 100.00%
=========================================
Files 50 51 +1
Lines 2851 2906 +55
=========================================
+ Hits 2851 2906 +55 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
pixi.toml
Outdated
| pytest-cov = "*" | ||
| pytest-md = "*" | ||
| scikit-learn = "*" | ||
| pydantic = ">=2.11.9,<3" |
There was a problem hiding this comment.
Just to be sure: Since we also use the Pydantic mypy plugin, does Pydantic have to be added to the docs feature? I think no, but I haven't checked exactly how you generate the docs for dataframely.mypy.
There was a problem hiding this comment.
Thanks for the PR @ytausch!
I have no pre-existing experience with custom types in pydantic, so my review on that front relies on understanding the tests. Overall, I just have pretty minor comments.
|
side note: there are some CI checks not marked as required, is that intended? |
Nope, fixed, thanks for pointing it out :) |
Co-authored-by: Andreas Albert <[email protected]>
There was a problem hiding this comment.
LGTM now, thanks for bearing with all the fine-tuning :) I just checked in with @borchero and he asked to take a look before we merge, so let's please wait for his review.
Closes #143.
If this is released, the feedstock should add a version constraint on
pydantic_core.