Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Avoid dfs transformer checking in feat importance fast mode check if pipline is stacked ensemble #4083

Merged
merged 2 commits into from
Mar 17, 2023

Conversation

tamargrey
Copy link
Contributor

fixes #4082

if has_dfs:
# If the pipeline is a stacked ensemble, the DFS Transformer name will be different
# and we don't need to check for the pre_existing features anyway
if has_dfs and not has_more_than_one_estimator:
Copy link
Contributor Author

@tamargrey tamargrey Mar 17, 2023

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The other option for fixing was to change self.component_graph[DFSTransformer.name] to use whatever name is actually being used in the component graph for the dfs transformer(s), but I don't know of any other way the name could be anything other than "DFS Transformer" other than the stacked ensemble case, so this seemed simpler.

@codecov
Copy link

codecov bot commented Mar 17, 2023

Codecov Report

Merging #4083 (1ec4228) into main (345c05c) will not change coverage.
The diff coverage is 100.0%.

@@          Coverage Diff          @@
##            main   #4083   +/-   ##
=====================================
  Coverage   99.7%   99.7%           
=====================================
  Files        349     349           
  Lines      37548   37548           
=====================================
  Hits       37429   37429           
  Misses       119     119           
Impacted Files Coverage Δ
...understanding_tests/test_permutation_importance.py 100.0% <ø> (ø)
evalml/pipelines/pipeline_base.py 98.5% <100.0%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

@tamargrey tamargrey marked this pull request as ready for review March 17, 2023 19:28
@chukarsten chukarsten merged commit a84fe26 into main Mar 17, 2023
@chukarsten chukarsten deleted the fix-dfs-in-ensembling-feat-importance branch March 17, 2023 19:54
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Feature importance for stacked ensemble pipelines will error if DFS Transformer is present
5 participants