legal_llm

Lawsuit analyzer using LLM, LLM finetuning using legal documents

List of colabs:

(min accuracy is 0, and max is 1)

0. lawsuit_analyze does not measure accuracy (initial approach)
1. finetune_pretrained_llm1 has a low accuracy of 0.25.(see attached low_acc.png)
2. finetune_pretrained_llm2 has a medium accuracy of 0.75.(see attached med_acc.png)
3. finetune_pretrained_llm2_flowchart has a medium accuracy of 0.75.(see attached med2.png and flowchart.png)
4. finetune_pretrained_llm3 has a medium-high accuracy of 0.75-1.00.(see attached med3.png)