训练成功了，但是没有合并的脚本，请问如何合并？

Question

训练成功了，但是没有合并的脚本，请问如何合并？

Opened this issue a year ago · 4 comments

apachemycat commented a year ago

Can't find 'adapter_config.json' at '/wzh/Chinese-Guanaco/models/guanaco-13b-merged-wzhtrain

Answer 1 · 2023-06-15T07:54:48.000Z

Here is the merge code: https://github.com/jianzhnie/Efficient-Tuning-LLMs/blob/main/chatllms/utils/apply_lora.py

Answer 2 · 2023-07-27T07:10:30.000Z

404 - page not found

Answer 3 · 2023-08-04T01:13:13.000Z

https://github.com/jianzhnie/Efficient-Tuning-LLMs/blob/main/chatllms/utils/apply_lora.py

Answer 4 · 2023-08-04T01:14:18.000Z

CUDA_VISIBLE_DEVICES=0 python chatllms/utils/apply_lora.py
--base-model-path ~/checkpoints/baichuan7b/
--lora-model-path ./work_dir/vicuna_merge_vicuna-baichuan-7b-1gpu/checkpoint-15000
--target-model-path ./work_dir/vicuna_merge_vicuna-baichuan-7b/merged_model