Fix need_retrain crash when best model is Ensemble after reload#826
Open
TheChyeahhh wants to merge 3 commits into
Open
Fix need_retrain crash when best model is Ensemble after reload#826TheChyeahhh wants to merge 3 commits into
TheChyeahhh wants to merge 3 commits into
Conversation
When PreprocessingMissingValues._fit_na_fill is called on a column containing exclusively null values, x.value_counts() returns an empty Series. The subsequent sorted(...)[0] then raises: IndexError: list index out of range This adds an early return of None for the empty case, allowing the caller to proceed with a safe fallback fill value. Fixes mljar#770
Adds a unified public method to retrieve global feature importance
computed across all trained models, as requested by the maintainer.
Usage:
automl = AutoML()
automl.fit(X_train, y_train)
importance_df = automl.get_feature_importance()
# Returns DataFrame with columns: feature, mean_rank, models_present
The method returns None when importance data is not available and
raises AutoMLException if fit() hasn't been called yet.
Closes mljar#809
_base_predict already handles the case where _best_model is None by auto-loading from results_path. _need_retrain did not, causing a crash when calling need_retrain on a freshly initialized AutoML object whose best model is an Ensemble. Added the same auto-load guard that _base_predict uses at the top of _need_retrain before accessing _best_model.get_metric(). Fixes mljar#799
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Fixes #799
When the best model is an Ensemble and the AutoML object is instantiated from a saved results path, calling
need_retrain()crashes because_best_modelisNone— the model has not been loaded yet.Root Cause
_base_predict(used bypredict,predict_proba,score) has an auto-load guard at line 1476-1477._need_retrainwas missing this guard, so it crashed onself._best_model.get_metric().Fix
Added the same auto-load guard at the top of
_need_retrainbefore accessing_best_model.Changes
supervised/base_automl.py— 3 lines added