Tabular Quality Audit

If you're reading this, it's possible that you were directed here by the Knowledge Hub or maybe you're just exploring the Supportivize Help Center. Either way, welcome!

1 min read

What is Tabular Quality Audit?

The Tabular Quality Audit is a fancy term we use here at Supportivize for the internal audit that takes place before you train your Agent with tabular data. That data might be in the form of a Google Sheet or CSV file.

In simpler words, the Tabular Quality Audit's job is to ensure that each row of your table or sheet doesn't exceed 2048 characters.

Why?

If a row is too long, the AI could potentially fragment the row which might lead to inconsistent answers. It's all about making sure your table or sheet is in the best shape before it's used as for training the Agent.

What can I do?

Ensure that the combined character count of all cells in a row does not exceed 2048 characters. The same rule applies for every row in your table or sheet!

Related articles
Did this answer your question?