docs: add a titanic forecasting how-to

Author: lpm0073

azure_ai/commands/compute_cluster.py

@@ -14,7 +14,7 @@
 
 if __name__ == "__main__":
 
-    compute_cluster = AzureAIMLStudioComputeCluster(cluster_name="standard-cluster").cluster
+    compute_cluster = AzureAIMLStudioComputeCluster(cluster_name="tiny-cluster").cluster
     retval = compute_cluster.serialize()  # type: ignore[no-untyped-call]
     retval = json.dumps(retval, indent=4)
     print(retval)

azure_ai/tests/test_batch_endpoint.py

@@ -30,7 +30,9 @@ def setUpClass(cls):
         hash_suffix = hashlib.sha256(datetime.now().isoformat().encode()).hexdigest()[:8]
         cls.endpoint_name = "test-batch-endpoint" + "-" + hash_suffix
         cls.model_name = "best-01"  # FIX NOTE: need to create a temporary model that we can destroy after the test
-        cls.compute_target_name = "standard-cluster"  # FIX NOTE: need to create a temporary compute cluster that we can destroy after the test
+        cls.compute_target_name = (
+            "tiny-cluster"  # FIX NOTE: need to create a temporary compute cluster that we can destroy after the test
+        )
         cls.description = "Custom test batch endpoint"
 
     @classmethod

azure_ai/tests/test_cluster.py

@@ -20,12 +20,12 @@ class AzureMLTestCluster(AzureMLTestBase):
     def test_existing_cluster(self):
         """
         Test that we can create and use an Azure ML workspace.
-        This will get a cluster named "standard-cluster" if it exists,
+        This will get a cluster named "tiny-cluster" if it exists,
         or create it if it does not.
         """
         if not self.is_testable:
             logger.warning("Skipping test_existing_cluster() as the environment is not testable.")
             return
 
-        cluster = AzureAIMLStudioComputeCluster(cluster_name="standard-cluster")
+        cluster = AzureAIMLStudioComputeCluster(cluster_name="tiny-cluster")
         self.assertIsInstance(cluster, AzureAIMLStudioComputeCluster)

docs/AZURE_ACCOUNT_SETUP.md

@@ -147,10 +147,10 @@ az provider list --query "[?namespace=='Microsoft.MachineLearningServices' || na
 
 ```console
 az ml compute create \
-  --name automl-cluster \
+  --name tiny-cluster \
   --type amlcompute \
   --size Standard_D4s_v3 \
-  --max-instances 4 \
+  --max-instances 1 \
   --min-instances 0 \
   --idle-time-before-scale-down 300 \
   --resource-group automl-resources \
@@ -161,7 +161,7 @@ az ml compute create \
 
 ```console
 az ml compute show \
-  --name automl-cluster \
+  --name tiny-cluster \
   --resource-group automl-resources \
   --workspace-name automl-workspace \
   --query "{size:size, location:location, maxInstances:scaleSettings.maxNodeCount}"
@@ -231,18 +231,18 @@ Once your quota is approved, update your cluster:
 ```console
 # Update cluster to use new quota limits
 az ml compute update \
-  --name automl-cluster \
+  --name tiny-cluster \
   --resource-group automl-resources \
   --workspace-name automl-workspace \
-  --max-instances 8 \
+  --max-instances 2 \
   --min-instances 0
 ```
 
 ### Verify Cluster Update
 
 ```console
 az ml compute show \
-  --name automl-cluster \
+  --name tiny-cluster \
   --resource-group automl-resources \
   --workspace-name automl-workspace \
   --query "scaleSettings"
@@ -269,7 +269,7 @@ ws = Workspace.from_config()
 print(f"✅ Connected to workspace: {ws.name}")
 
 # Test compute target
-compute_target = ws.compute_targets["automl-cluster"]
+compute_target = ws.compute_targets["tiny-cluster"]
 print(f"✅ Compute cluster ready: {compute_target.provisioning_state}")
 ```
 
@@ -345,7 +345,7 @@ az account set --subscription "your-subscription-name"
 ```console
 # Set up auto-scale down after 5 minutes
 az ml compute update \
-  --name automl-cluster \
+  --name tiny-cluster \
   --idle-time-before-scale-down 300 \
   --resource-group automl-resources \
   --workspace-name automl-workspace