steveepreston
steveepreston
@BenjaminBossan Thanks for note. I will try and report here.
@benjamin-marie Nice
Hey @hiyouga I tried, it creates the `log.txt` file but the file is empty. The main challenge of this way is that we should download the log file sec by...
@hiyouga No problem! I will try! I hope guys coming and share ideas about this.
@BUAADreamer Thanks for note **Code 1:** ```python !GRADIO_SHARE=1 llamafactory-cli webui >log.txt 2>&1 & ``` -> `OSError: Background processes not supported.` (by ipykernel) **Code 2:** ```python os.system("GRADIO_SHARE=1 llamafactory-cli webui >log.txt 2>&1...
### Proposal: ```python from llamafactory import webui # start the webui service and prints urls of gradio. webui.start() # user now is able to intract with notebook ... # end...
update: the error `RuntimeError: There are currently no available devices found, must be one of 'XPU', 'CUDA', or 'NPU'` dosen't thrown on `transformers==4.38.2` and llama-3 fine-tune successfully done on TPU...
Do you talking about `\needs_auth\OpenaiChat.py`?
I have just found `temporary=False` in `create_authed()` is this what you told? how can we control it without modifying source code? for example, is there a run arg?
I Think it should be in run args for example `--temporary` / `--temporary_chat`