InternLM-XComposer
InternLM-XComposer copied to clipboard
请问要是中断训练,想继续的话需要加什么参数呢
训练中间一个step断掉了,可以从这个step继续训练吗
You may modify the code and set resume_from_checkpoint =True in the TrainingArguments class.
#423 I found resume_from_checkpoint =True has its own issue with LoRA training, the LOSS restarts itself. Not sure whether you guys got a similar issue. @yuhangzang
I find there is a problem saving checkpoints with 2d5-7b, while internlm-xcomposer2-vl-7b can saves checkpoint correctly in different settings. #423 #426