Skip to content
Nappai Documentation

Natural Language to SQL

This component lets you ask questions about your data in plain English, and Nappai will automatically translate them into a SQL query that retrieves the answers from your database. This makes it easy to get the information you need without needing to know SQL.

Relationship with SQL Databases

This component connects to your SQL database (like MySQL, PostgreSQL, etc.) to execute the SQL queries it generates from your natural language questions. You need to specify which database to use.

Inputs

  • Input: This is where you type your question about the data in your database. For example, you might type “What were the total sales for last month?”. This is a required field.
  • Language Model: This selects the AI model Nappai uses to understand your question and translate it into SQL. You need to choose a model from the available options. This is a required field.
  • SQLDatabase: This is where you select the specific SQL database you want to query. Nappai needs to know which database contains the data you’re asking about. This is a required field.
  • Top K: This setting controls how many results are returned if your question asks for a list. The default is 5, but you can change this to get more or fewer results.
  • Prompt (Optional): This allows you to customize how the SQL query is generated. This is advanced and usually not needed. If you use it, make sure it includes {question} where you want your question to be inserted.

Outputs

The component produces a single output:

  • Text: This contains the generated SQL query. You can usually ignore this unless you want to see the exact SQL code Nappai created. The result of the SQL query will be displayed by other components in your Nappai workflow.

Usage Example

Let’s say you want to know the average price of products in your database. You would:

  1. Type “What is the average price of all products?” into the Input field.
  2. Select your preferred Language Model.
  3. Select your SQLDatabase.
  4. (Optional) Leave the Prompt field blank.
  5. The component will generate a SQL query (which you can see in the Text output if you wish) and then another component will display the average price.

Templates

This component is used in many different Nappai workflows. You’ll find it in various templates depending on the specific data analysis task. The exact configuration will depend on the template.

  • SQL Executor: This component takes the SQL query generated by “Natural Language to SQL” and actually runs it against your database, returning the results.
  • Explain SQL results: This component takes the results from the SQL Executor and explains them in a user-friendly way.

Tips and Best Practices

  • Ask clear and concise questions. The more specific your question, the better the results.
  • Use simple language. Avoid technical jargon or complex sentence structures.
  • If you get unexpected results, double-check your database selection and try rephrasing your question.

Security Considerations

Ensure your SQL database credentials are securely managed within Nappai. Do not expose sensitive database information directly in your workflow. Nappai’s security features help protect your data, but best practices are always recommended.