Merge pull request #95564 from trevorbye/master

fixed link and removed old explainability stuff

Author: Jak-MS

articles/machine-learning/service/how-to-auto-train-remote.md

@@ -148,24 +148,6 @@ automl_config = AutoMLConfig(task='classification',
                              )
 ```
 
-### Enable model explanations
-
-Set the optional `model_explainability` parameter in the `AutoMLConfig` constructor. Additionally, a validation dataframe object must be passed as a parameter `X_valid` to use the model explainability feature.
-
-```python
-automl_config = AutoMLConfig(task='classification',
-                             debug_log='automl_errors.log',
-                             path=project_folder,
-                             compute_target=compute_target,
-                             run_configuration=run_config,
-                             X = X,
-                             y = y,
-                             **automl_settings,
-                             model_explainability=True,
-                             X_valid=X_test
-                             )
-```
-
 ## Submit training experiment
 
 Now submit the configuration to automatically select the algorithm, hyper parameters, and train the model.
@@ -234,59 +216,13 @@ remote_run.get_portal_url()
 
 The same information is available in your workspace.  To learn more about these results, see [Understand automated machine learning results](how-to-understand-automated-ml.md).
 
-### View logs
-
-Find logs on the DSVM under `/tmp/azureml_run/{iterationid}/azureml-logs`.
-
-## <a name="explain"></a> Best model explanation
-
-Retrieving model explanation data allows you to see detailed information about the models to increase transparency into what's running on the back-end. In this example, you run model explanations only for the best fit model. If you run for all models in the pipeline, it will result in significant run time. Model explanation information includes:
-
-* shap_values: The explanation information generated by shap lib.
-* expected_values: The expected value of the model applied to set of X_train data.
-* overall_summary: The model level feature importance values sorted in descending order.
-* overall_imp: The feature names sorted in the same order as in overall_summary.
-* per_class_summary: The class level feature importance values sorted in descending order. Only available for the classification case.
-* per_class_imp: The feature names sorted in the same order as in per_class_summary. Only available for the classification case.
-
-Use the following code to select the best pipeline from your iterations. The `get_output` method returns the best run and the fitted model for the last fit invocation.
-
-```python
-best_run, fitted_model = remote_run.get_output()
-```
-
-Import the `retrieve_model_explanation` function and run on the best model.
-
-```python
-from azureml.train.automl.automlexplainer import retrieve_model_explanation
-
-shap_values, expected_values, overall_summary, overall_imp, per_class_summary, per_class_imp = \
-    retrieve_model_explanation(best_run)
-```
-
-Print results for the `best_run` explanation variables you want to view.
-
-```python
-print(overall_summary)
-print(overall_imp)
-print(per_class_summary)
-print(per_class_imp)
-```
-
-Printing the `best_run` explanation summary variables results in the following output.
-
-![Model explainability console output](./media/how-to-auto-train-remote/expl-print.png)
-
-You can also visualize feature importance through the widget UI, or in your workspace in [Azure Machine Learning studio](https://ml.azure.com). 
-
-![Model explainability UI](./media/how-to-auto-train-remote/model-exp.png)
-
 ## Example
 
-The [how-to-use-azureml/automated-machine-learning/remote-amlcompute/auto-ml-remote-amlcompute.ipynb](https://github.com/Azure/MachineLearningNotebooks/blob/master/how-to-use-azureml/automated-machine-learning/remote-amlcompute/auto-ml-remote-amlcompute.ipynb) notebook demonstrates concepts in this article.
+The following [notebook](https://github.com/Azure/MachineLearningNotebooks/blob/master/how-to-use-azureml/automated-machine-learning/regression/auto-ml-regression.ipynb) demonstrates concepts in this article.
 
 [!INCLUDE [aml-clone-in-azure-notebook](../../../includes/aml-clone-for-examples.md)]
 
 ## Next steps
 
-Learn [how to configure settings for automatic training](how-to-configure-auto-train.md).
+* Learn [how to configure settings for automatic training](how-to-configure-auto-train.md).
+* See the [how-to](how-to-machine-learning-interpretability-automl.md) on enabling model interpretability features in automated ML experiments.