Machine Learning: Load Data into Analytics

This page is accessed via Logistics Machine Learning > Machine Learning Project > Actions > Load Data into Analytics.

Load Data into Analytics collects data from multiple shipment-related tables in OTM based on the saved queries defined on the machine learning project. This action loads the data into the Logistics Machine Learning TI subject area. If you change your data, you need to rerun this action.

Note: Only buy shipments with actual arrival and departure times on the first and last stop are loaded into Analytics.

Once the data is in Logistics Machine Learning intelligence, you can do some pre-training exploratory data analysis. You can review the historical data and visual outliers and see what data is actually going to be used for training.

Note: To see complete logging information, turn on the Machine Learning Log IDs.

Loading Data

The Load Data into Analytics Input screen automatically selects and shows the saved queries that were included on the Project Data tab of the machine learning project. You can deselect any saved queries you do not want to load.

1. Select the Saved Query IDs to load into Logistics Machine Learning intelligence.
2. Click Submit to load the data. The page changes to show Load Data into Analytics Results which displays information such as:
   • Project ID
   • Total Number of Records Loaded, Inserted, and Updated for all saved queries
   • Total Number of Records Loaded, Inserted, and Updated for each specific saved query along with the date and time the data was last loaded into Analytics.
3. Click Refresh as needed to see the final results.

The Last Loaded to Analytics field on the Project Data tab of the Project is populated when you run this action.