ZXStudio
ZXStudio
我本地电脑上能正常运行cli_demo.py,但部署在腾讯GPU服务器上却有问题,我是通过远程桌面连接到服务器,然后打开VSCdoe跑cli_demo.py,提问之后AI没有回复消息,也没报错,字符串变量response没有任何内容(cli_demo.py第18~21行),我没改动过代码,都是本地测试过在上传到服务器的。 试过换了windows server 2022系统,还是同样的问题。 ---------4.15更新-------- int4量化版可以输出内容,标准版还是有问题,所以bug可能出在模型。 ### Expected Behavior _No response_ ### Steps To Reproduce 我没改动过代码,直接VSCdoe运行cli_demo.py ### Environment ```markdown - 显卡:P40 - OS:windwos server 2019 - Python:Python3.10 - Transformers:4.27.1 -...
### Is your feature request related to a problem? Please describe. 训练出来的模型可以在训练环境下运行,但无法在标准环境运行。 发生异常: TypeError argument of type 'WindowsPath' is not iterable File "D:\GLM\web_demo.py", line 4, in from peft import PeftModel...
### Is your feature request related to a problem? Please describe. 只能自己搞爬虫抓吗? ### Solutions 只能自己搞爬虫抓吗? ### Additional context _No response_
How to switch to GPU mode?
I need a scene where Gaussian Splatting and Mesh are mixed, and I don't know if GSPLAT generates Mesh or not.