liangwq / Chatglm_lora_multi-gpu

chatglm多gpu用deepspeed和
404 stars 61 forks source link

可以增加个finetune之后的推断脚本吗 #13

Closed llplay closed 1 year ago

liangwq commented 1 year ago

No description provided.

inference.py和webui文件夹底下就是

llplay commented 1 year ago

有多gpu的版本吗,使用多gpu推断的时候老是报RuntimeError: expected scalar type Half but found Float的错误

llplay commented 1 year ago

@liangwq