configuration.ipynb should specify default compute target

[dem108]

Issue summary

Current GitHub sample run: fails when using
https://github.com/Azure/MachineLearningNotebooks/blob/master/configuration.ipynb does not include specifying default compute target when creating a workspace.
https://github.com/Azure/MachineLearningNotebooks/blob/master/how-to-use-azureml/training-with-deep-learning/train-hyperparameter-tune-deploy-with-keras/train-hyperparameter-tune-deploy-with-keras.ipynb tries to use the default compute target, which fails.
These two are not consistent.

How to reproduce:

Just git clone and run above two notebooks.

Proposed Solution:

The configuration.ipynb should be updated: when creating workspace, it should specify default compute target.

import os
from azureml.core.compute import AmlCompute

compute_min_nodes = os.environ.get("AML_COMPUTE_CLUSTER_MIN_NODES", 0)
compute_max_nodes = os.environ.get("AML_COMPUTE_CLUSTER_MAX_NODES", 4)
vm_size = os.environ.get("AML_COMPUTE_CLUSTER_SKU","STANDARD_D2_V2")
provisioning_config = AmlCompute.provisioning_configuration(vm_size = vm_size, min_nodes = compute_min_nodes, max_nodes = compute_max_nodes)

from azureml.core import Workspace
ws = Workspace.create(name='<WORKSPACE_NAME>' ,
                      subscription_id='<SUBSCRIPTION_ID>' ,
                      resource_group='<RESOURCE_GROUP>' ,
                      create_resource_group=True ,
                      default_cpu_compute_target=provisioning_config ,
                      location='<LOCATION>')

Please confirm and fix the issue if this is correct.