Training a Git Model Version - Teradata Vantage

ClearScape Analytics™ ModelOps User Guide

Deployment
VantageCloud
VantageCore
Edition
Enterprise
IntelliFlex
VMware
Product
Teradata Vantage
Release Number
7.1
Published
December 2024
ft:locale
en-US
ft:lastEdition
2024-12-13
dita:mapPath
zdn1704469623418.ditamap
dita:ditavalPath
azq1671041405318.ditaval
dita:id
rgn1654191066978
lifecycle
latest
Product Category
ClearScape
A default personal connection is required to train models. Teradata recommends creating a default personal connection prior to training a model. See Adding a Connection.
  1. Select a Git model in the models list.
  2. Select Train Model.
    If a default personal connection has not been created, you will be prompted to create it.
    1. Complete the fields and save the connection.

      Once created, the connection displays in the drop-down.

    2. Select Continue.
  3. In the Basic tab, set the properties:
    Property Description
    Model Specify the model name in read-only format.
    Dataset template Specify the required dataset template.
    Dataset Specify the dataset to be used to train the model.
    Hyper parameters Aet the training variables manually with a pre-determined value before starting the training job.
  4. In the Advanced tab, set the properties:
    Property Description
    Engine Specify the engine to train the model in read-only format.
    Docker image Specify the docker version to be used to train the model.
    Resource template Identify a predefined set of resources, including CPU and memory, which are the properties of the container created to run the task in. Select S Standard, M Medium, L Large, or Custom from the Template drop-down list.
    Set any of the following properties:
    • Memory: Specifies the allocated RAM memory for the container. Must be an integer followed by the unit (m for megabytes or g for gigabytes).
    • CPU: Specifies the allocated CPU units (cores) for the container, can be an integer or decimal number. Accepts m as the unit for milicores.
    • GPU (visible when Custom is selected): Specifies the allocated GPU units for the container, can be an integer or decimal number. Accepts m as the unit for milicores.
  5. Select Train model.
    The In progress page displays the model version training progress.

    The Logs tab shows the job logs and events for the selected job.

    The Properties tab shows all properties related to the selected job such as job ID, dataset ID, and hyper parameters.

  6. Select to close the sheet when the training progress completes.

    The List of model versions displays the following properties for each version.

    Property Description
    ID Specifies the auto-generated trained model ID.
    Status Specifies the version status as Trained, Evaluated, Approved, Rejected, Deployed, or Retired.
    Dataset Specifies the dataset used to train the version.
    Created by Shows the username who created the version.
    Tags Displays the list of tags associated with the model version.
    Champion Shows whether the version is champion or not.
    Created at Specifies the timestamp when the model version was created.