automl-toolkit icon indicating copy to clipboard operation
automl-toolkit copied to clipboard

'FeatureImportance' object has no attribute 'run_feature_importances'

Open zwag20 opened this issue 4 years ago • 2 comments

I am trying to run the python example on databricks. When I get to this line I get the error in the subject. fi_importances = FI.run_feature_importances("XGBoost", "classifier", dataframe,20.0,"count",generic_overrides)

I have attached the wheel file pyAutoML-0.2.0-py3-none-any.whl to my cluster.

zwag20 avatar Dec 16 '20 22:12 zwag20

Just to add to this. It appears every section in the doc https://github.com/databrickslabs/automl-toolkit/blob/master/python/docs/APIDOCs.md fails, so I am guessing I am missing something.

The AutomatedRunner Class fails with the error ('Your model family but be within any of the following supported model types:', ['RandomForest', 'XGBoost', 'LogisticRegresesion', 'Trees', 'GBT', 'LinearRegression', 'MLPC', 'SVM'])

I've tried both XGBoost and RandomForest, both fail with the same message.

The Family Runner fails with the message 'JavaPackage' object is not callable

zwag20 avatar Dec 17 '20 01:12 zwag20

change 'run_feature_importances' to 'run_feature_importance'

yousenwang avatar Mar 20 '21 13:03 yousenwang