Sign up to take part
Registered users can ask their own questions, contribute to discussions, and be part of the Community!
Added on October 20, 2023 5:10PM
Likes: 0
Replies: 1
com.dataiku.dip.io.SocketBlockLinkKernelException: Failed to train : <class 'TypeError'> : 'NoneType' object is not iterable at com.dataiku.dip.io.SocketBlockLinkInteraction.throwExceptionFromPython(SocketBlockLinkInteraction.java:302) at com.dataiku.dip.io.SocketBlockLinkInteraction$AsyncResult.checkException(SocketBlockLinkInteraction.java:215) at com.dataiku.dip.io.SocketBlockLinkInteraction$AsyncResult.get(SocketBlockLinkInteraction.java:190) at com.dataiku.dip.io.SingleCommandKernelLink$1.call(SingleCommandKernelLink.java:211) at com.dataiku.dip.analysis.ml.prediction.PredictionTrainAdditionalThread.process(PredictionTrainAdditionalThread.java:76) at com.dataiku.dip.analysis.ml.shared.PRNSTrainThread.run(PRNSTrainThread.java:170)
FIT/PROCESS WITH Step:FeatureSelectionStep /opt/dataiku-dss-12.1.3/python39.packages/sklearn/feature_selection/_univariate_selection.py:112: UserWarning: Features [0 0] are constant. warnings.warn("Features %s are constant." % constant_features_idx, UserWarning) /opt/dataiku-dss-12.1.3/python39.packages/sklearn/feature_selection/_univariate_selection.py:113: RuntimeWarning: invalid value encountered in true_divide f = msb / msw /opt/dataiku-dss-12.1.3/python39.packages/sklearn/feature_selection/_univariate_selection.py:112: UserWarning: Features [0] are constant. warnings.warn("Features %s are constant." % constant_features_idx, UserWarning) [2023-10-20 17:05:30,565] [125197/MainThread] [INFO] [dataiku.doctor.utils.listener] END - Preprocessing train set [2023/10/20-17:05:30.566] [MRT-16096] [INFO] [dku.block.link.interaction] - Check result for nullity exceptionIfNull=true result=null Traceback (most recent call last): File "/opt/dataiku-dss-12.1.3/python/dataiku/doctor/server.py", line 45, in serve ret = api_command(arg) File "/opt/dataiku-dss-12.1.3/python/dataiku/doctor/dkuapi.py", line 45, in aux return api(**kwargs) File "/opt/dataiku-dss-12.1.3/python/dataiku/doctor/commands.py", line 343, in train_prediction_models_nosave transformed_train = pipeline.fit_and_process(train_df) File "/opt/dataiku-dss-12.1.3/python/dataiku/doctor/preprocessing/dataframe_preprocessing.py", line 2537, in fit_and_process new_mf = step.fit_and_process(input_df, cur_mf, result, self.generated_features_mapping) File "/opt/dataiku-dss-12.1.3/python/dataiku/doctor/prediction/feature_selection.py", line 66, in fit_and_process self.selection = get_feature_selector( File "/opt/dataiku-dss-12.1.3/python/dataiku/doctor/prediction/feature_selection.py", line 171, in fit names = self.get_pruned_names(mf, target) File "/opt/dataiku-dss-12.1.3/python/dataiku/doctor/prediction/feature_selection.py", line 205, in get_pruned_names
Hi,
Could you give a little bit of more background e.g. what kind of data, selection of model (is it custom or do you use one from the DSS Lab?), what you are trying to achieve etc.?
From the error you receive, I guess that what you are feeding to the model is None this is why it cannot be trained. Therefore, I recommend trying to check the preprocessing steps for any mistakes (for example check if any of your training columns is empty).
-Konstantina