PhotonDemocratize model training and inference across the team.
Docs
Credits:$1.00
Jobs/Create

Create Evaluation Job

Run evaluations on your models

Create evaluation jobs to test model performance against datasets

1
Evaluator

Select an Evaluator

Choose an evaluator from your account to run. The evaluator will be used to assess the performance of your model against a dataset.

Evaluators help you understand how well your model performs on specific tasks or criteria. Make sure to select an evaluator that matches your assessment needs.

2
Dataset

Requirements

  • •Evaluator must be created and saved in your account
  • •Evaluator should match your assessment criteria
  • •Python code must be valid and executable

What is an Evaluator?

An evaluator is a Python function that scores your model's outputs against specific criteria. It helps you measure performance, quality, and correctness of your model's responses.