How to Build a Classification Model with Promptrepo?

Classification models categorize input data into predefined labels. In this tutorial, we’ll build a food classification model to categorize food items using PromptRepo.

Step 1: Preparing Training Data

For classification, our dataset consists of:

  1. Web page title: The name of the web page
  2. Web page content: Content of the page
  3. Nutrition category: The correct category label (e.g., Red Meat, Beverages, Cheese, etc.)

Step 2: Uploading Data to Promptrepo

  1. Open Google Sheets and add or import the training dataset
  2. Click Extensions > Promptrepo > Train AI
  3. Promptrepo sidebar widget will be displayed
  4. Enter the AI model name and click Next
  5. Select Web page title & Web page content as input columns and click Next
  6. Select Nutrition category as the output column, select Enum for the Treat as option and click Next

Note: In the output column, choosing Enum for the Treat as option for nutrition category makes it a classification model.

Step 3: Training & Testing the Model

  1. Click Publish and let Promptrepo fine-tune the classification model.
  2. Once trained, test the model by inputting a food name and its description.
  3. The model should predict the correct category.

Classification models are foundational for structuring data. Now, let’s explore how to build an extraction model for retrieving structured data from unstructured text.