Training and Deploying Your First Model

This tutorial is an introduction to some of the primary features of AnalyticOps. The tutorial covers basic tasks that are related to Git-based XGBoost Python Demo Model training, evaluation, approval and deployment within multiple environments.

In this tutorial, you will complete the following tasks:

• Train a Model version

• Evaluate Model version

• Approve Model version

• Deploy Model version as RESTFUL

• Deploy Model version as Batch

• Deploy Model version in Vantage

• Retire All Deployments

Using the XGBoost Python Demo Model

XGBoost is an implementation of gradient boosted decision trees designed for speed and performance that is dominative competitive machine learning.

(Comment: Some introduction of XGBoost Python Demo Model will go here)

Train a Model Version

To train an XGBoost Python Demo model version:

1. From List of Projects, open the Demo project.

   

   The List of Models displays.

   

2. For the Python Diabetes Prediction model, click the View Model Versions icon.

   The Python Diabetes Prediction model versions page displays.

   

3. Click the Train Model button.

   The Train Model Version dialog displays.

   

4. In the Basic tab, set the properties:

   Property	Value
   Dataset Template	PIMA Diabetes
   Dataset	PIMA Diabetes Train
   Connection	Vantage Demo
   Hyper Parameters
   eta	0.2
   max_depth	6

   

5. In the Advanced tab, set the properties:

   Property	Value
   Engine
   Docker Image
   Resource Template	S Standard

   

6. Click Train.

   The model version training progress displays on the screen as below.

   

   Note: For more information about Training job progress, see Model Versions in AnalyticOps User Guide.

7. Click Done when the training progress completes.

   The trained model version adds to the list of model versions.

   

   The Training Details display on the model version lifecycle page.

   

See Also

• Models in AnalyticOps User Guide

• Model Versions in AnalyticOps User Guide

• Model Lifecycle in AnalyticOps User Guide

Evaluate Model Version

To initiate the evaluation of a trained XGBoost Python Demo model version, two methods can be followed.

First Method

1. From the Python Diabetes Prediction model versions, select the Trained version of the model.

   The Actions icon displays on the top.

   

2. Click the Actions icon to see the menu. Select Evaluate.

   

Second Method

1. From the Python Diabetes Prediction model versions, select the View Events icon for the Trained version of the model.

   The Model version lifecycle page displays.

   

2. Click the Evaluate button.

   The Evaluate Model Version dialog displays.

   

3. In the Basic tab, set the properties:

   Property	Value
   Dataset Template	PIMA Diabetes
   Dataset	PIMA Diabetes Evaluate
   Connection	Vantage Demo

   

4. In the Advanced tab, set the properties:

   Property	Value
   Engine
   Docker Image
   Resource Template	S Standard

   

5. Click Evaluate.

   The model version evaluation progress displays on the screen as below.

   Note: For more information about Evaluation job progress, see Model Lifecycle in AnalyticOps User Guide.

   

6. Click Done when the evaluation progress completes.

   The model version status is changed to Evaluated.

   

   The Evaluation Details display 0n the model version lifecycle page.

   

See Also

• Models in AnalyticOps User Guide

• Model Versions in AnalyticOps User Guide

• Model Lifecycle in AnalyticOps User Guide

Approve Model Version

To initiate the approval of an evaluated XGBoost Python Demo model version, two methods can be followed.

First Method

1. From the Python Diabetes Prediction model versions, select the Evaluated version of the model.

   The Actions icon displays on the top.

   Comment: Screen for Approval flow from Model versions list will display here. The feature is not available.

2. Click the Actions icon to see the menu. Select Approve.

Second Method

1. From the Python Diabetes Prediction model versions, select the View Events icon for the Evaluated version of the model.

   The Model version lifecycle page displays.

   

2. Click the Approve button.

   The Approve Model Version dialog displays.

   

3. Insert Comments and click Approve.

   The model version status is changed to Approved.

   

   The Approval Details display on the model version lifecycle page.

   

Note: For more information about Approval and Rejection of model version, see Model Approval in AnalyticOps User Guide.

See Also

• Models in AnalyticOps User Guide

• Model Versions in AnalyticOps User Guide

• Model Lifecycle in AnalyticOps User Guide

Deploy Model Version as RESTFUL

To deploy an approved XGBoost Python Demo model version, two methods can be followed.

First Method

1. From the Python Diabetes Prediction model versions, select the Approved version of the model.

   The Actions icon displays on the top.

   

2. Click the Actions icon to see the menu. Select Deploy.

   

Second Method

1. From the Python Diabetes Prediction model versions, select the View Events icon for the Approved version of the model.

   The Model version lifecycle page displays.

   

2. Click the Deploy button.

   The Deploy Model Version dialog displays.

   

3. In the Basic tab, set the properties:

   Property	Value
   Engine Type	RESTful
   Replicas	1

   

4. In the Advanced tab, set the properties:

   Property	Value
   Engine
   Docker Image
   Resource Template	S Standard

   

5. Click Deploy.

   The model version deployment progress displays on the screen as below.

   Note: For more information about Deployment job progress, see Model Lifecycle in AnalyticOps User Guide.

   

6. Click Done when the deployment progress completes.

   The model version status changes to Deployed.

   

   The Deployment Details display on the model version lifecycle page.

   

See Also

• Models in AnalyticOps User Guide

• Model Versions in AnalyticOps User Guide

• Model Lifecycle in AnalyticOps User Guide

Deploy Model Version as Batch

To deploy an approved or deployed XGBoost Python Demo model version, two methods can be followed.

First Method

1. From the Python Diabetes Prediction model versions, select the Deployed version of the model.

   The Actions icon displays on the top.

   

2. Click the Actions icon to see the menu. Select Deploy.

   

Second Method

1. From the Python Diabetes Prediction model versions, select the View Events icon for the Deployed version of the model.

   The Model version lifecycle page displays.

   

2. Click the Deploy button.

   The Deploy Model Version dialog displays.

   

3. In the Basic tab, set the properties:

   Property	Value
   Engine Type	Batch
   Dataset Template	PIMA Diabetes
   Connextion	Vantage Demo
   Schedule	Hourly
   Hours	1
   Mins	0

   

4. In the Advanced tab, set the properties:

   Property	Value
   Engine
   Docker Image
   Resource Template	S Standard

   

5. Click Deploy.

   The model version deployment progress displays on the screen as below.

   Note: For more information about Deployment job progress, see Model Lifecycle in AnalyticOps User Guide.

   

6. Click Done when the deployment progress completes.

   The model version status remains as Deployed.

   

   The Deployment Details display on the model version lifecycle page.

   

See Also

• Models in AnalyticOps User Guide

• Model Versions in AnalyticOps User Guide

• Model Lifecycle in AnalyticOps User Guide

Deploy Model Version in Vantage

To deploy an approved or deployed XGBoost Python Demo model version as Vantage, two methods can be followed.

First Method

1. From the Python Diabetes Prediction model versions, select the Deployed version of the model.

   The Actions icon displays on the top.

   

2. Click the Actions icon to see the menu. Select Deploy.

   

Second Method

1. From the Python Diabetes Prediction model versions, select the View Events icon for the Deployed version of the model.

   The Model version lifecycle page displays.

   

2. Click the Deploy button.

   The Deploy Model Version dialog displays.

   

3. In the Basic tab, set the properties:

   Property	Value
   Engine Type	In-Vantage
   Connection	Vantage Demo
   Database	AOA_DEMO

   

4. In the Advanced tab, set the properties:

   Property	Value
   Engine
   Docker Image
   Resource Template	S Standard
   Build Properties
   LANGUAGE	PMML

   

5. Click Deploy.

   The model version deployment progress displays on the screen as below.

   Note: For more information about Deployment job progress, see Model Lifecycle in AnalyticOps User Guide.

   

6. Click Done when the deployment progress completes.

   The model version status remains as Deployed.

   

   The Deployment Details display on the model version lifecycle page.

   

See Also

• Models in AnalyticOps User Guide

• Model Versions in AnalyticOps User Guide

• Model Lifecycle in AnalyticOps User Guide

Retire All Deployments

To retire the active deployments for XGBoost Python Demo model version, two methods can be followed.

First Method

1. From the Python Diabetes Prediction model versions, select the Deployed version of the model.

   The Actions icon displays on the top.

   

2. Click the Actions icon to see the menu. Select Retire.

   

Second Method

1. From the Python Diabetes Prediction model versions, select the View Events icon for the Deployed version of the model.

   The Model version lifecycle page displays.

   

2. Click the Retire button.

   The Retire Model Version dialog displays.

   

Retire RESTful Deployment

1. Select RESTful Deployment ID.

   

2. Click Retire.

   The model version retirement progress displays on the screen as below.

   Note: For more information about Retiring job progress, see Model Lifecycle in AnalyticOps User Guide.

   

3. Click Done when the retirement progress completes.

   The model version status remains as Deployed as the model version has still active deployments.

   

   The Retirement Details display on the model version lifecycle page.

   

Retire Batch Deployment

1. Select the Batch Deployment ID in the Retire Model Version dialog.

   

2. Click Retire.

   The model version retirement progress displays on the screen as below.

   Note: For more information about Retiring job progress, see Model Lifecycle in AnalyticOps User Guide.

   

3. Click Done when the retirement progress completes.

   The model version status is Deployed as the model version still has active deployments.

   

   The Retirement Details display on the model version lifecycle page.

   

Retire Vantage Deployment

1. Select the Vantage Deployment ID in the Retire Model Version dialog.

   

2. Click Retire.

   The model version retirement progress displays on the screen as below.

   Note: For more information about Retiring job progress, see Model Lifecycle in AnalyticOps User Guide.

   

3. Click Done when the retirement progress completes.

   The model version status is changed to Retired since there is no active deployment for the version.

   

   The Retirement Details display on the model version lifecycle page.

   

See Also

• Models in AnalyticOps User Guide

• Model Versions in AnalyticOps User Guide

• Model Lifecycle in AnalyticOps User Guide