Skip to main content

Google BigQuery

Heimdall offers a direct connection to your BigQuery databases, allowing you to import your tables for use in Heimdall ML and Heimdall Forecast. While selecting the data to use, you will be able to select BigQuery as a data source. You will then be prompted to enter your Redshift connection parameters.

Connection Parameters​

  • Project ID: your bigquery's project id
  • Dataset ID: your project's dataset id
  • JSON Credentials File: upload your service account credentials in JSON file format
note

To create your JSON Credentials File, you have to create a service account

  1. Click on the Google Cloud sidebar menu on the top left corner
  2. Hover over the IAM & Admin and select "Service Accounts" in the sub menu
  3. On the service accounts page, click on the "+ CREATE SERVICE ACCOUNT" button
  4. Give a name, account ID, and description for the account and click the "CREATE AND CONTINUE" button
  5. Grant the account with a role that has enough permissions such as "BigQuery User" or "BigQuery Admin"
  6. On the next page click on the "DONE" button
  7. After being redirected back to the service accounts home page, click on the service account you just created
  8. On the top menu, click on the keys section
  9. Click on the "ADD KEY" button and the "Create new key" option
  10. For the key type, select the recommended JSON file format for the private key and click "CREATE"
  11. You've now created your key to use for credential authentication!
note

Heimdall does not save the connection parameters for your BigQuery database.

Importing and Using your Data​

Once you've provided the connection parameters for your BigQuery database, you will be provided a list of tables to import data from.

tip

Only tables within schemas accessible to the user will be shown. If you cannot find a specific table, make sure the user associated with the username and password provided has permission to access the table.

Select the table you'd like to import data from and click Next. Heimdall will now select all rows from the table and continue the modeling or forecasting process.

Once the process is complete, you'll be provided some high level metrics for your model or forecaster, and can save it with a unique name. It will appear in your inventory, and you'll be able to select it to view more details.