Scorer Node

This node assigns scores to an item based on custom criteria. You need to define the criteria in the scoring criteria field.

​

Node Inputs

1. Item for Scoring: The text to be scored
   • Type: Text
2. Scoring Criteria: The criteria which the text will be scored
   • Type: Text
3. Additional Context: The additional information or instructions for the AI when doing the scoring
   • Type: Text

​

Node Parameters

On the face of the node:

1. Include Justification: When this parameter is turned on, the justification output field for why the score was given will be available.

In the gear:

1. Provider: Provider of the AI model you want to use. The default provider is OpenAI.
2. Model: Model name of the AI model you want to use. The default model is gpt-4o.

​

Node Outputs

1. Score: The score of the text based on the criteria
   • Type: Text
   • Example usage: {{scorer_0.score}}
2. Justification: The justification given by the AI for selecting the score (if include justification is switched on)
   • Type: Text
   • Example usage: {{scorer_0.justification}}

​

Considerations

• Define the bounds of the score in the scoring criteria, specifically the upwards and lower bounds. Define what criteria warrants a high vs low score.

​

Example

The below example is a pipeline that grades text based on grammar. The scoring criteria stipulates that a score of 1 means the text has bad grammar and a score of 10 means good grammar.

• Text Node: Text for grading.
• Scorer Node: Provide the text and specify the criteria for scoring.
  • Item for scoring: {{text_0.text}}
• Output Node: Display the score from the Scorer Node.
  • Output: {{scorer_0.score}}