1. AIML Guide
  2. Step Two: Viewing the Training Data

Step Two: Viewing the Training Data

The process to create one of these models is defined below:

  • Step One: Create a Scheduled Job.
  • Step Two: View the Training Data.
  • Step Three: Define the model with parameters/filters and select it in the AIML Training Template.
  • Step Four: Run the prediction and view results in the AIML Prediction Run UI.
  • Step Five: Deploy, retrain models, and interpret results.

These screens allow you to view training data:

  • AIML Training Data – Allows users to view the prepared data generated by a scheduled job.
  • AIML Training Template – Where users define templates and optionally set a default.
  • AIML Models UI – Displays all models built from training templates.
  • AIML Prediction Run UI – Shows the results from prediction executions.
  • AIML Predictive Dashboard – Displays the latest prediction results (currently available for Order Cycle Time only).

When the job runs, it processes the Username and selected AIML Template, and upon successful completion, generates a model. This job also triggers the Train action on the corresponding AIML Training Template defined in the job parameters, allowing users to generate models on a recurring basis.

The output (model) from each job run is captured in the AIML Models UI, where a new model is logged for every execution.

A new screen called AIML Training Data allows users to view the results of the scheduled job for:

  • Order Cycle Time
  • Order Waiting Time
  • Order Processing Time
  • Intelligent Cycle Count
  • Market Basket Analysis

Users can select one of these metrics from a dropdown. Once selected, a data grid appears showing the relevant training data. This data represents the output of the "Generate Training Data for..." job.

Each metric will have its own unique data grid:

  • Switching the dropdown will load a new grid with fields specific to the selected metric.
  • Each sub-screen supports search functionality.
  • Users can delete records but cannot create or edit them.
  • CSV download is available.
  • Users can also filter by date range using “From Date” and “To Date”.

Two new sub-screens—Intelligent Cycle Count and Order Cycle/Waiting/Processing Time—are now available in the AIML Prediction Run Details UI. This is where the final prediction results are stored and displayed.