0

I am trying to run azureml-sdk AutoML on databricks to do a feature/model sweep on a classification task, and when I run the experiment I get the following error:

Running an experiment on spark cluster: ComputeAutoML. 

ERROR: Setup iteration failed: {'exception': ValueError('dataprep_json is None')}

1 Answers1

0

If you are using an old SDK so upgrade to the latest SDK and pandas. Tabular dataset need to be passed for X and y while running on ADB. enter image description here enter image description here Please follow the below notebook for running automl on databricks. https://github.com/Azure/MachineLearningNotebooks/blob/master/how-to-use-azureml/azure-databricks/automl/automl-databricks-local-01.ipynb

Ram
  • 2,459
  • 1
  • 7
  • 14