Version:

Models + Analytics

A model is a mathematical or programmatical representation of a real-world process. AAW currently supports three types of models: TensorFlow, RAPIDS, and Blackbox.

  • Point to + Add Model then click one of the options to begin the model setup/import process.
  • Type into Filter to filter down the models
  • Click show to display any archived models.
  • Click refresh to refresh the table
  • Click export to export the table's values as JSON or CSV
  • Click an existing model to display additional actions. Actions vary depending on the type of model:
    • NEURAL_NET models:
      • Click View Selection to open the Model Details page
      • Click Train to start training the model
      • Click Terminate Training to stop training the model
      • Click Deploy to deploy the model
      • Click Clone Model to clone the model
      • Click Export Entity to export the model as a JSON object
      • Click Archive to archive the model; it will be removed from the list of models and will be no longer useable
      • Click Description / Configuration to review summary information for the model
    • BLACKBOX models:
      • Click View Selection to open the Model Details page
      • Click Deploy to deploy the model
      • Click Clone Model to clone the model
      • Click Export Entity to export the model as a JSON object
      • Click Archive to archive the model; it will be removed from the list of models and will be no longer useable
      • Click Description / Configuration to review summary information for the model
../_images/aaw_ui_models.png

Details

The Model Details page provides a detailed look at a given model, including configuration information, feature set, training and test datasets, and any deployments. Available actions depend on the type of model being detailed.

  • NEURAL_NET models:
    • Click Back to return to the Models page
    • Click Train to start training the model
    • Click Terminate Training to stop training the model
    • Click Deploy to deploy the model
    • Click Clone Model to clone the model
    • Click Export Entity to export the model as a JSON object
    • Click Archive to archive the model; it will be removed from the list of models and will be no longer useable
  • BLACKBOX models:
    • Click Back to return to the Models page
    • Click Deploy to deploy the model
    • Click Clone Model to clone the model
    • Click Export Entity to export the model as a JSON object
    • Click Archive to archive the model; it will be removed from the list of models and will be no longer useable
    • If the model has been deployed previously, click the deployment's name in the Model Deployments table to open the Deployment Details
../_images/aaw_ui_model_details.png

Model Creation

New TensorFlow Model

To create a new TensorFlow model:

  1. From the Models and Analytics page, click + Add Model ‣ New TensorFlow.
  2. Provide a Model Name.
  3. Optionally, provide a Model Description.
  4. Optionally, for Model Template, click Search to search for and select a template. Depending on the selected template, several Training Parameters will be automatically created and configured.
  5. For Feature Set, click Search to search for and select an existing feature set. In the Search window, click + Add New Feature Set to add a new feature set and use it. See New Feature Set for more details.
  6. For Training Dataset, click Search to search for and select an existing dataset. In the Search window, click + Add New Dataset to add a new dataset and use it. See New Dataset for more details.
  7. For Testing Dataset, click Search to search for and select an existing dataset. In the Search window, click + Add New Dataset to add a new dataset and use it. See New Dataset for more details.
  8. For Training Parameters:
    • Finish configuring the automatically-created parameters (if there were any)
    • Add additional parameters:
      • + Boolean creates a boolean parameter. Provide a Parameter Name and click the slider as necessary. Slide it to the right for true; left for false.
      • + Text creates a text parameter. Provide a Parameter Name and Parameter Value.
      • + Number creates a number parameter. Provide a Parameter Name and Parameter Value.
      • + JSON creates a JSON parameter. Provide a Parameter Name and click Edit JSON to provide a JSON Parameter Value.
      • Click the trashcan icon to remove a parameter.
  9. Click Create.

New RAPIDS Model

To create a new RAPIDS model:

  1. From the Models and Analytics page, click + Add Model ‣ New RAPIDS.
  2. Provide a Model Name.
  3. Optionally, provide a Model Description.
  4. Optionally, for Model Template, click Search to search for and select a template.
  5. For Dataset, click Search to search for and select an existing dataset. In the Search window, click + Add New Dataset to add a new dataset and use it. See New Dataset for more details.
  6. For Dataset Features select one or more features from the selected dataset.
  7. For Dataset Label, select a label from the list of features in the selected dataset.
  8. Set the Training Percentage for the label to the desired amount.
  9. Click Create.

New Blackbox Model

After opting to create a new Blackbox model, two paths are available:

  • Inspect Docker Container -- inspect an existing Docker container (one that is already in a Docker repository), select an available module/function, and automatically import the model

    1. From the Models and Analytics page, click + Add Model ‣ New Blackbox
    2. Under Inspect Docker Container, provide a Docker container URI (e.g., <repo-name>/<image-name>:<tag>)
    3. Click Inspect. AAW will pull the container's functions
    4. Optionally, click View to display details about a given function
    5. Under the desired function, click Import. AAW will automatically configure the Blackbox model
    6. Optionally, adjust the Blackbox model configuration fields as necessary
    7. Click Create
  • Create a Blackbox Model Manually -- start from a blank template and provide various parameters to manually create and import a Blackbox model

    Note

    Models can be created manually from existing containers as well; this option provides more manual control over the model import process

    1. From the Models and Analytics page, click + Add Model ‣ New Blackbox.
    2. Under Create a Blackbox Model Manually, click Create.
    3. Provide a Model Name.
    4. Optionally, provide a Model Description.
    5. Provide a Blackbox Docker Container repository URI or opt to create a Blackbox model container.
      • If a Blackbox model has already been created using the Kinetica Blackbox SDK and published to a Docker repository, provide the URI in the following format: <repo-name>/<image-name>:<tag>, e.g., kinetica/kinetica-blackbox-quickstart:latest
      • If opting to create a Blackbox model container automatically using AAW, click + New Container.
        1. Provide the Docker Repository Name.
        2. Optionally, provide the Docker Repository Description.
        3. Optionally, upload a Requirements File.
        4. Upload the Module File (must be a Python file).
        5. Provide the Module Function name from the Module File.
        6. Select a Docker credential.
        7. Click Create.
    6. Provide a Module and Function from the model
    7. For Input Columns:
      1. Click Add Input Column to create input columns.
      2. Provide a Column name and Type.
    8. For Output Columns:
      1. Click Add Output Column to create output columns.
      2. Provide a Column name and Type.
    9. Click Create.

Model Deployment

After creating a model, a deployment can be created from the Models + Analytics page or the Model Details page.

  1. Select a model from the Models + Analytics page or open the Model Details page.

  2. Click Deploy in the right-hand menu.

  3. Provide a Name for the deployment.

  4. Optionally, provide a Description for the deployment.

  5. Select a compute target:

    • If selecting CPU, set the number of replicas of the model to deploy.

    • If selected GPU, set the number of GPUs used per replica and the number of replicas of the model to deploy. The Resources graphic will update accordingly.

      ../_images/aaw_ui_gpu_compute_target.png
  6. Select a Mode. See Deployments for more information on the different types of modes.

    • If selecting On-Demand, skip to the next step.
    • If selecting Continuous or Batch, select a Source Table from the drop-down menu and provide an Output Table name.
  7. Optionally, provide key and value pairs for any Environment Variables available to the model

  8. Click Deploy.