Graph Setup

After connecting your tables, the next step is to link them into a graph schema to enable predictions. This step is crucial, as these relationships define how Kumo derives meaningful input signals.

  • To begin, navigate to the Graph and click Create Graph.

  • Select the tables to include. As you add tables, you can see the graph on the right pane.

  • Click Next to proceed.

Table Linkages

Once tables are selected, link them by specifying key (ID) columns that define relationships between tables. These links help Kumo understand the structure of your data for making predictions.

  • Links – Created when linking a primary key column to another table’s foreign key column.

  • Backlinks – Appear when a table is referenced by other tables, indicating relationships originating from that table.

For an e-commerce dataset, you might link:

  • Customer ID between customers and transactions.

  • Transaction ID between transactions and articles.

Table Linkage Considerations

Kumo automatically suggests links matching column names across tables. These linkages appear as dotted lines in the graph visualization.

When creating table links, keep in mind:

  • Timestamps do not need manual linking—Kumo automatically detects them.

  • If two linked columns have different types, both are cast to strings—ensure values still match after casting.

  • Rows with missing key values are ignored when creating linkages.

  • Rows with unmatched key values across tables are also ignored.

To adjust linkages:

  • Click Table Name to modify an forward link.

  • Select tables and column names to add links or Delete icon to remove links.

  • Click Save Graph to finalize your schema.

Fixing Data Issues After Graph Creation

If you discover data quality issues after creating your graph:

  • Fix issues in your source data.

  • Kumo will automatically re-ingest and reconnect tables the next time you train a model.