Mellen Y. Pu
Mellen Y. Pu
Don't worry about that. It's worth mentioning that this is just a warning when you do not successfully export `Nvidia OptiX ` in `~/.bashrc`. Check your env first, please. :-)
So, in a single NVIDIA `3090`, `Fastchat` **doesn't support finetuning with lora** on `vicuna-7B` model, right?
Hi, all. I tried the above but no effects here. Later I deleted the `venv` dir and commanded `bash ./webui.sh`, and it worked well again. Thanks a lot for your...
so, is there any solution?
One more push for the release.
thank you very much, and another question is that will you consider adding the OCR technology to this framework to enhance the text-reading ability?
> We have 500 gb memory in total and uses
> Could you check the deepspeed config? Using deepspeed zero2.config instead of zero3.config may help. The high demand of RAM may be caused by deepspeed ZeRO-Offload strategy, which needs to...
Hi, thank you, but s+till not yet. @SHUMKASHUN ```python Debugging middleware caught exception in streamed response at a point where response headers were already sent. Traceback (most recent call last):...
May I ask, as the repo mentioned that at least 16GB of memory is essential for pretraining, if that corresponds to only one kind of object such as chairs?