LLM Evaluation

[sarah114tran]

Test 1. Prompt Used: One Shot, Schema Provided
Notes:

1. Design files are incorrectly labeled as mechanical design files and visa versa
2. There are a handful of instances where a human would need to find the documentation
3. In instances where the there is a link to a hookup guide, should we consider this to be a bill of materials?

True Positive	False Positive	False Negative	True Negative
134	12	15	91

Accuracy	Precision	Recall	F1 Score	True Positive Rate	False Negative Rate
0.89	0.92	0.90	0.91	0.90	0.10