Evaluate Assistant
1. For the Assistant you want to evaluate, click on the three dots next to it.
2. Click on Evaluate to open the evaluation tab
3. Here you will be able to see the list of datasets
4. Lets create a dataset by clicking on Create Dataset
5. Type a Name for the dataset
6. Add a description for the dataset
7. From the list of queries, select relevant queries for the dataset
8. Select by clicking the checkboxes next to queries
9. Click on Save Changes
10. The datasets table shows the newly added dataset now
11. Lets create another dataset following the same steps
12. Select queries for dataset
15. Type a Name for the dataset
16. Add the description for the dataset
17. Click on Save Changes
18. Here we have two datasets for running our evaluation test now
19. Select a dataset from the table for the evaluation test
20. you can also select multiple datasets at the same time for a bulk evaluation
21. Lets start the evaluation for a single dataset by clicking on Start Evaluation
22. Click on Confirm
23. Go to the Test Runs Tab
24. The evaluation has started with the Status "Pending"
25. The status will soon update to "Running"
26. Click on the Refresh Table button after a while to check for results
27. Finally, the evaluation scores will be available
