Automated Continuous Hugging Face Replication to Google BigQuery

Always-on applications rely on automatic failover capabilities and real-time data access. CData Sync integrates live Hugging Face data into your Google BigQuery instance, allowing you to consolidate all of your data into a single location for archiving, reporting, analytics, machine learning, artificial intelligence and more.

Configure Google BigQuery as a Replication Destination

Using CData Sync, you can replicate Hugging Face data to Google BigQuery. To add a replication destination, navigate to the Connections tab.

1. Click Add Connection. 👁 Add a new connection.
   
2. Select the Destinations tab and locate the Google BigQuery connector. 👁 Configure a Destination connection to Google BigQuery.
   
3. Click the Configure Connection icon at the end of that row to open the New Connection page. If the Configure Connection icon is not available, click the Download Connector icon to install the Google BigQuery connector. For more information about installing new connectors, see Connections in the Help documentation.
4. After the connected is added, enter the necessary connection properties. To connect to Google BigQuery, use OAuth authentication::
   • Connection Name: Enter a connection name of your choice.
   • Auth Scheme: Select OAuth. Sync supports the OAuth, OAuthJWT, GCPInstanceAccount, and AWSWorkloadIdentity authentication methods.
   • ProjectId: Enter the ID of the project you want to connect to.
   • DatasetId: Enter the ID of the dataset you want to connect to.
   • Insert Mode: Set it to Upload. The different Insert Modes available in Sync are Streaming, DML, Upload, and GCSStaging.
👁 Enter the connection details.

5. Click Connect to Google BigQuery. Log in using your Gmail credentials to grant permissions to CData Sync. CData Sync completes the OAuth process and connects successfully to Google BigQuery.
6. Once connected, click Save & Test to save the connection. 👁 Save the connection.
   

You are now connected to Google BigQuery and can use it as both a source and a destination.

NOTE: You can use the Label feature to add a label for a source or a destination.

👁 Add a label.

In this article, we will demonstrate how to load Hugging Face data into Google BigQuery and utilize it as a destination.

Configure the Hugging Face Connection

You can configure a connection to Hugging Face from the Connections tab. To add a connection to your Hugging Face account, navigate to the Connections tab.

1. Click Add Connection.
2. Select a source (Hugging Face).
3. Configure the connection properties.
   
   

   HuggingFace Hub uses token-based authentication to enable access to its API. The API provides access to machine learning models, datasets, spaces, papers, and other resources on the HuggingFace Hub platform.

   Using API Key Authentication

   To authenticate to HuggingFace Hub, you will need to provide an API Key (Access Token). To obtain your access token:

   1. Log in to your HuggingFace account at https://huggingface.co
   2. Navigate to Settings > Access Tokens
   3. Click "New token" to create a new access token
   4. Select the appropriate permissions (read or write)
   5. Copy the token value

   After obtaining your access token, set the following connection properties:

   • AuthScheme: Set this to APIKey.
   • APIKey: Set this to your HuggingFace access token.

   Example connection string

   Profile=C:\profiles\HuggingFace.apip;ProfileSettings='APIKey=hf_xxxxxxxxxxxxxxxxxxxx';
   

   👁 Configuring a Source connection (Salesforce is shown).
   
4. Click Connect to Hugging Face to ensure that the connection is configured properly.
5. Click Save & Test to save the changes.

Configure Replication Queries

CData Sync enables you to control replication with a point-and-click interface and with SQL queries. For each replication you wish to configure, navigate to the Jobs tab and click Add Job. Select the Source and Destination for your replication.

👁 Select Source and Destination connections for the replication.

Replicate Entire Tables

To replicate an entire table, navigate to the Task tab in the Job, click Add Tasks, choose the table(s) from the list of Hugging Face tables you wish to replicate into Google BigQuery, and click Add Tasks again.

👁 Choose entire tables to replicate (Salesforce is shown).

Customize Your Replication

You can use the Columns and Query tabs of a task to customize your replication. The Columns tab allows you to specify which columns to replicate, rename the columns at the destination, and even perform operations on the source data before replicating. The Query tab allows you to add filters, grouping, and sorting to the replication with the help of SQL queries. 👁 Configure a replication Transform.

Schedule Your Replication

Select the Overview tab in the Job, and click Configure under Schedule. You can schedule a job to run automatically by configuring it to run at specified intervals, ranging from once every 10 minutes to once every month.

👁 Schedule your job to run automatically.

Once you have configured the replication job, click Save Changes. You can configure any number of jobs to manage the replication of your Hugging Face data to Google BigQuery.

Run the Replication Job

Once all the required configurations are made for the job, select the Hugging Face table you wish to replicate and click Run. After the replication completes successfully, a notification appears, showing the time taken to run the job and the number of rows replicated.

👁 Run the job.

Free Trial & More Information

Now that you have seen how to replicate Hugging Face data into Google BigQuery, visit our CData Sync page to explore more about CData Sync and download a free 30-day trial. Start consolidating your enterprise data today!

As always, our world-class Support Team is ready to answer any questions you may have.