-
### Contribution guidelines
- [X] I've read the [contribution guidelines](https://tvdboom.github.io/ATOM/latest/contributing/).
- [X] The documentation does not mention anything about my problem.
- […
-
Update: currently, the input checkers don't let nullable types pass, an error is thrown.
I think we should move to a detect & coerce solution, with optional config to turn off the input checks/coer…
-
When https://github.com/scikit-learn/scikit-learn/issues/25578 is completed and released, we can upgrade evalml and remove the nullable type handling put in place by https://github.com/alteryx/evalml/…
-
- We can improve our time to run unit tests by running koalas, dask, and pandas separately, and then combining the coverage report (CodeCov automatically [supports](https://docs.codecov.io/docs/mergin…
-
When `interpolate` is the impute method, we use pandas' `series.impute` method, which cannot handle nullable types: https://github.com/pandas-dev/pandas/issues/40252
As part of component-specific n…
-
Per [this](https://github.com/alteryx/evalml/actions/runs/3024888320) Woodwork/EvalML main job run, it looks like we are raising errors related to inconsistent dtypes. Given the timing, it might be re…
-
Not sure if this is reproducible or was just a weird state I got myself into, but after creating a new virtualenv on python 3.8.2 and following contributing.md (running `make installdeps-dev` and then…
-
The following block of code will raise the `ValueError: Unknown label type: 'unknown'` error.
```python
import woodwork as ww
X, y = X_y_binary
y = ww.init_series(y, logical_type="Bo…
-
The conda version is outdated (see https://anaconda.org/conda-forge/evalml). Please update to the newest release.
-
## Overview
It has been over a year since the last release. A number of bugs and deprecations have been implemented on the community-led [SHAP Community Fork](https://github.com/dsgibbons/shap), wh…