Skip to content

ask_my_data

ask_my_data(question, fully_qualified_table)

Description

Ask your data any question in natural language.

This function calls generate_sql to convert the question into a SQL query adapted to your data and execute it.
(If the execution fails, the temporary table bigfunction_result gives details).

Param Description
question A question to ask over your data. It may start with a verb
fully_qualified_table A table whose schema will be given to the generative AI model for context.
- fully_qualified_table format must be like PROJECT.DATASET.TABLE.
- fully_qualified_table must exist.
- You must have read permission on the table schema

Usage

Call or Deploy ask_my_data ?
Call ask_my_data directly

The easiest way to use bigfunctions

  • ask_my_data function is deployed in 39 public datasets for all of the 39 BigQuery regions.
  • It can be called by anyone. Just copy / paste examples below in your BigQuery console. It just works!
  • (You need to use the dataset in the same region as your datasets otherwise you may have a function not found error)

Public BigFunctions Datasets

Region Dataset
eu bigfunctions.eu
us bigfunctions.us
europe-west1 bigfunctions.europe_west1
asia-east1 bigfunctions.asia_east1
... ...
Deploy ask_my_data in your project

Why deploy?

  • You may prefer to deploy ask_my_data in your own project to build and manage your own catalog of functions.
  • This is particularly useful if you want to create private functions (for example calling your internal APIs).
  • Get started by reading the framework page

Deployment

ask_my_data function can be deployed with:

pip install bigfunctions
bigfun get ask_my_data
bigfun deploy ask_my_data

Examples

call bigfunctions.eu.ask_my_data("get the 3 products which generated the most revenue in 2023", "bigfunctions.eu.sales");
select * from bigfunction_result;

call bigfunctions.us.ask_my_data("get the 3 products which generated the most revenue in 2023", "bigfunctions.us.sales");
select * from bigfunction_result;

call bigfunctions.europe_west1.ask_my_data("get the 3 products which generated the most revenue in 2023", "bigfunctions.europe_west1.sales");
select * from bigfunction_result;


+------------+---------+
| product_id | revenue |
+------------+---------+
+ 8          + 50286   |
+ 6          + 46172   |
+ 10         + 27972   |
+------------+---------+

Use cases

A business analyst wants to understand sales data for 2023, specifically which products generated the most revenue. They have a BigQuery table named sales in their project (e.g., my-project.sales_data.sales) containing product IDs, revenue figures, and other sales-related information.

Instead of writing a complex SQL query, the analyst uses the ask_my_data function like so:

CALL bigfunctions.us.ask_my_data(
    'Get the 3 products which generated the most revenue in 2023',
    'my-project.sales_data.sales'
);
SELECT * FROM bigfunction_result;

This function call takes two arguments:

  1. 'Get the 3 products which generated the most revenue in 2023': The natural language question the analyst wants to ask the data.
  2. 'my-project.sales_data.sales': The fully qualified name of the table the question pertains to. This allows the function to understand the structure of the data and generate the appropriate SQL query.

The ask_my_data function then internally translates the natural language question into a SQL query (likely involving aggregation, filtering, and ordering), executes it against the specified table, and stores the results in a temporary table named bigfunction_result. The analyst can then retrieve the results by querying bigfunction_result.

This simplified approach allows users without extensive SQL knowledge to extract insights from their data using natural language, making data analysis more accessible and efficient.

Need help or Found a bug?
Get help using ask_my_data

The community can help! Engage the conversation on Slack

We also provide professional suppport.

Report a bug about ask_my_data

If the function does not work as expected, please

  • report a bug so that it can be improved.
  • or open the discussion with the community on Slack.

We also provide professional suppport.

Show your ❤ by adding a ⭐ on