Pin mlflow<3 and fix deployment in azureml-in-a-day tutorial

tutorials/azureml-in-a-day/azureml-in-a-day.ipynb

@@ -193,8 +193,10 @@
                 "  - pip:\n",
                 "    - inference-schema[numpy-support]==1.3.0\n",
                 "    - xlrd==2.0.1\n",
-                "    - mlflow\n",
+                "    - mlflow==2.17.2\n",
                 "    - azureml-mlflow\n",
+                "    - azureml-ai-monitoring\n",
+                "    - azureml-inference-server-http\n",
                 "    - psutil\n",
                 "    - tqdm\n",
                 "    - ipykernel~=6.29\n",
@@ -648,12 +650,20 @@
                 "# picking the model to deploy. Here we use the latest version of our registered model\n",
                 "model = ml_client.models.get(name=registered_model_name, version=latest_model_version)\n",
                 "\n",
+                "# create an inference environment that includes azureml-ai-monitoring\n",
+                "# (required by the scoring script generated by the latest inference base image)\n",
+                "inference_env = Environment(\n",
+                "    name=\"credit-defaults-inference\",\n",
+                "    conda_file=os.path.join(dependencies_dir, \"conda.yaml\"),\n",
+                "    image=\"mcr.microsoft.com/azureml/openmpi4.1.0-ubuntu22.04:latest\",\n",
+                ")\n",
                 "\n",
                 "# create an online deployment.\n",
                 "blue_deployment = ManagedOnlineDeployment(\n",
                 "    name=\"blue\",\n",
                 "    endpoint_name=online_endpoint_name,\n",
                 "    model=model,\n",
+                "    environment=inference_env,\n",
                 "    instance_type=\"Standard_DS3_v2\",\n",
                 "    instance_count=1,\n",
                 ")\n",