Tamar Grey
Tamar Grey
Once https://github.com/alteryx/evalml/issues/1389 is finished, the `OrdinalEncoder` will exist but not actually be used within EvalML. We should integrate it. This will require a change to the OneHotEncoder, since when the...
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...
Our current implementation of the WoodworkTableAccessor has several methods that overwrite Pandas methods: `drop`, `pop`, `rename`, `value_counts`, `index`, and the serialization methods are all examples. Some of these methods add...
In using the Ordinal.order property as part of implementing https://github.com/alteryx/evalml/issues/1389, I ended up digging into Ordinal's implementation to remind myself how nans are handled. It could be useful to others...
The `PerColumnImputer` can impute floating point values into integer data with the `mean` or `median` numeric impute strategies. When this happens, we cannot simply reinitialize the original data's woodwork schema...
Change `_schema_is_equal` check to `_schema_is_compatible` and use training schema for predict data
Currently, at `ComponentGraph._transform_features`, when the graph is not already fit, we do a check for whether or not X's woodwork schema is equal to the `ComponentGraph.input_types`. If the types do...
In https://github.com/alteryx/evalml/issues/4103 we will have added a call to `_handle_nullable_types` in `determine_periodicity` to support IntegerNullable for the decomposer. Once the [pandas bug](https://github.com/pandas-dev/pandas/issues/52127) causing this issue is resolved and released, we...
With https://github.com/alteryx/evalml/issues/4095, we made the handling of integer grid values for partial dependence explicit, rounding the fractional values produced in `_grid_from_X` to allow IntegerNullable types to be used for partial...
Currently, the `TargetDistributionDataCheck` does not allow nullable logical types. This doesn't match the behavior of `InvalidTargetDataCheck`, which does allow nullable types. With the new nullable type support across automl search,...
As a user, I wish I could train a pipeline on data that might not have nans and has non nullable types and then predict/`transform_all_but_final` /score data that has nans...