This example combines the PSI capabilities with the Vertical Regression
protocol to create a regression model across a subset of records found in three
distinct datasets owned by different organizations. The model is then used to
perform an inference on test datasets distributed across the data owners.

The three datasets contain 40, 60, and 20 columns respectively. Each of the datasets
contains a unique identifier column that is used by the PSI protocol to join the
records that match across the datasets into a single logical record. The target
column used for regression is contained in just one of the datasets.

After the PSI protocol securely performs the match of the identifiers across each
dataset and filters the data, a regression model is trained on top of this logical data.

The last step of this example sequence uses the trained model to perform an inference
using test data that is similarly spread across the participants.
