adamlin120 / TCEval

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

TCEval v2

Install

cd lm-evaluation-harness_mr-revised
pip3 install -e ".[vllm]"
pip3 install -U vllm
cd ..

Evaluate Local Models (MMLU, TMMLU+, and Penguin_Table)

please reference examples

Evaluate API Models (MMLU, TMMLU+, and Penguin_Table)

please check scripts/cal_likelihood_by_api.py

Evaluate MTBench-tw

please reference here.

About


Languages

Language:Python 91.6%Language:Jupyter Notebook 7.7%Language:C++ 0.7%