Mario O. Bourgoin

Hi, When I used the hyperparameter tuning recipe, I found it useful to create a conda virtual environment in which to run the shell scripts and notebook. I used it...

This updates the version pointed to by TrainMLModelHyperDrive to V.2.0.1, wjich fixes a bug where the Papermill and Jupyter nbconvert installed are not compatible with each other.

Include an alternative to notebook 04 that shows how to use Bayesian hyperparameter sampling.

Put all dependencies for Estimators in a single Conda dependencies file to be used in all cases.

Currently, the testing script uses the maximum number of iterations of the trained model to score the data. Add an "early_stopping_rounds" argument to the training script so it records the...

Use low-priority nodes

1

Consider changing the experiment to use low-priority nodes.

Once an AML SDK LightGBM estimator is available, use it to create the best model on the cluster.