No evaluation script for GSM8K

[duterscmy]

The open-source evaluation scripts can evaluate most benchmarks used in this work, but GSM8K not included. Will authors add it in the future?

[duterscmy]

Hi, I have already found the inference scripts for gsm8k in diffullama/LLaMA-Factory/examples/inference/llama2_full_ddm-gsm-inf.yaml;
however I don't understand why only the evaluation script for this benchmark is built in LLAMA-FACTORY, and other benchmarks' evaluation are built in diffullama/inf_diffullama.py. Is this benchmark different from others?