[Bug]: Evaluation metrics are showing as NaN (Not a Number) values after running an experiment, even though the experiment completes successfully.

[AkhilaTestAutomation]

FloTorch Version

V3.0.1

Bug Description

After completing an experiment in the application, the evaluation metrics section displays NaN (Not a Number) instead of actual values for faithfull context precision and maliciousness. This creates confusion, as users expect meaningful results after running an experiment. It may affect trust in the platform's evaluation capabilities and usability of results.

Steps to Reproduce

Steps to Reproduce

1. Login to FloTorch.ai
2. Click on Get Started button
3. click on Create Project
4. Enter the Project Name, select the region
5. Navigate to the Knowledge Bases selection section.
6. Select an AWS Bedrock Knowledge Base for embedding.Click on Next
7. Select the N shot prompt and attach the N shot prompt file
8. Select the KNN value
9. Select the any of the inferencing models
10. Select the inferencing model temperature
11. Click on Next button
12. Select the Ragas evaluation, embedding model and Inferencing model
13. Click on Submit

Browser and Version

135.0.7049.115

Runtime Version

python 3.13

Error Messages

Additional Context

No response