evalml
evalml copied to clipboard
alteryx
EvalML is an AutoML library written in python.
Tolerance can increase as time progresses further curbing search time. Still need to discuss improvement rate (linear? amount?).
Is this the best way to do it? ``` pipeline = clf.best_pipeline pipeline.input_feature_names[pipeline.estimator.name] ``` We might want to think through improving this API or at least documenting this is how...
We should allow a user to specify if they want to impose monotonic constraints on any of the variables that are feeding into their model. Not all modeling approaches support...
Currently users view the `score` of a custom objective as a raw number. Maybe it would make it more clear if units helped define the `score`. For example, in the...
@ParthivNaresh asked in #3305: Do we want to add a name to the feature importance series? _Originally posted by @ParthivNaresh in https://github.com/alteryx/evalml/pull/3305#r805906013_ I don't think it's a bad suggestion! We...
The AutoML user guide page includes a [reference to an outdated method of defining component graphs](https://evalml.alteryx.com/en/stable/user_guide/automl.html#Using-custom-pipelines). We should update this and any other scenarios where this syntax is outdated!
In #3679 `pmdarima` was pinned to `< 2.0.0` due to test failures seen [here](https://github.com/alteryx/evalml/runs/7956162888?check_suite_focus=true). This issue tracks fixing the test failures and upgrading `pmdarima===2.0.0`.
In https://github.com/alteryx/evalml/pull/3657, we added logic to conditionally add the Imputer only when nulls are detected. This led to a marked performance increase in the datasets that had no missing values....
As an EvalML dev, I would like all of my tests that are able to to use `get_test_data_from_configuration()` so that I can reduce total lines of code and confine changes...
I would like the logic in `detect_problem_type` to be updated to include time series problems.
