Skip to content

10. Compute historical feature values

To prepare training and testing data (also called "Feature Table" in FeatureByte) for your Machine Learning models, follow these steps:

Step 1: Select a Feature List

Navigate to the Feature List catalog from the Experiment section of the menu.

Name

Select a desired Feature List and click on the calculator icon.

Name

Step 2: Select the Observation Table

Pick the Observation Table that specifies your entity values and historical points-in-time:

  1. If you created the "CUSTOMER_x_PRODUCTGROUP_Sum_of_TotalCost_next_2_weeks" target with the SDK, use the "In_Store_Customer_x_ProductGroup_Spending_next_2_weeks_2022_10K" table.
  2. If you didn't create the target using the SDK, opt for the "In_Store_Customer_x_ProductGroup_2022_10K" table.

Name

Step 3: Name Your Feature Table

Assign a name to your feature table and, if you wish, provide a description. Here, we called the feature table: "In_Store_Customer_x_ProductGroup_Spending_next_2_weeks_2022_10K wtih 15 features".

Name

Step 4: Locate the Completed Table

After the table is computed, find your table in the 'Feature Table Catalog' under the 'Experiment' section.

Name

Name

Step 5: Preview or Download

Select the "In_Store_Customer_x_ProductGroup_Spending_next_2_weeks_2022_10K wtih 15 features"

Go to the 'Preview' tab to preview the table.

Name

Go to the 'About' tab to download it as a parquet file.

Name

For SDK Users

More details are available in our SDK reference for Historical Feature Table.

With the SDK, you can directly compute training data as a Pandas DataFrame or convert a stored Feature Table into a Pandas DataFrame.