acrolinx

Author: Larry Franks

articles/machine-learning/service/how-to-deploy-and-where.md

@@ -126,8 +126,9 @@ The following types are currently supported:
 To use schema generation, include the `inference-schema` package in your conda environment file. The following example uses `[numpy-support]` since the entry script uses a numpy parameter type: 
 
 #### Example dependencies file
-The following is an example of a Conda dependencies file for inference.
-```python
+The following YAML is an example of a Conda dependencies file for inference.
+
+```YAML
 name: project_environment
 dependencies:
   - python=3.6.2
@@ -343,7 +344,7 @@ For more information on creating an AKS cluster outside of the Azure Machine Lea
 * [Create an AKS cluster](https://docs.microsoft.com/cli/azure/aks?toc=%2Fazure%2Faks%2FTOC.json&bc=%2Fazure%2Fbread%2Ftoc.json&view=azure-cli-latest#az-aks-create)
 * [Create an AKS cluster (portal)](https://docs.microsoft.com/azure/aks/kubernetes-walkthrough-portal?view=azure-cli-latest)
 
-For more information on the `cluster_purpose` parameter ,see the [AksCompute.ClusterPurpose](https://docs.microsoft.com/python/api/azureml-core/azureml.core.compute.aks.akscompute.clusterpurpose?view=azure-ml-py) reference.
+For more information on the `cluster_purpose` parameter, see the [AksCompute.ClusterPurpose](https://docs.microsoft.com/python/api/azureml-core/azureml.core.compute.aks.akscompute.clusterpurpose?view=azure-ml-py) reference.
 
 > [!IMPORTANT]
 > For [`provisioning_configuration()`](https://docs.microsoft.com/python/api/azureml-core/azureml.core.compute.akscompute?view=azure-ml-py), if you pick custom values for agent_count and vm_size, then you need to make sure agent_count multiplied by vm_size is greater than or equal to 12 virtual CPUs. For example, if you use a vm_size of "Standard_D3_v2", which has 4 virtual CPUs, then you should pick an agent_count of 3 or greater.