Spaces:
Running
on
CPU Upgrade
Gradio ApI stopped working.
Used to be able to call like this
client = Client("https://huggingfaceh4-open-llm-leaderboard.hf.space/")
job = client.submit( fn_index=6)
json_data = job.result(timeout=120)
Now there is a parameter required and no info on what is supposed to be.
Is there a way to pin a stable gradio API endpoint so it will return the full view of the data without filters? I think a lot of people would find that useful.
Hi!
It's more important for us that people are able to access the app in both web and mobile very fast (which is now the case thanks to the hard work of the gradio team) than for people to have the full table on one page, especially since all the results can also be downloaded from the public dataset. We therefore won't change the pinned gradio version we use.
Can you open an issue on the gradio github with the error you encountered with this gradio version? Thank you :)
I'll give it a try.. i may just go for way to read the result sets directly.
I created a space that clones the Leaderboard and provides a Gradio API to download the full leaderboard list (with instructions)
https://huggingface.co/spaces/felixz/open_llm_leaderboard
@clefourrier
May be useful to add to community resources
Added it, thank you for your work!
I love this leaderboard - but it is not working today - runtime error :(
https://huggingface.co/spaces/felixz/open_llm_leaderboard
It says:
Running on local URL: http://0.0.0.0:7860
To create a public link, set share=True in launch().
Hi
@lselector
,
This is a normal problem which can happen from time to time, as indicated in the FAQ :)
No need to create an issue for this, unless the problem lasts for more than a day.
Btw, to make the discussions easier to manage, please create your own issues instead of writing on those of others with unrelated topics. Thank you for your interest in the leaderboard :)
I love this leaderboard - but it is not working today - runtime error :(
https://huggingface.co/spaces/felixz/open_llm_leaderboard
It says:
Running on local URL: http://0.0.0.0:7860
To create a public link, setshare=Trueinlaunch().
I think I need to restart it periodically, i will put some code in for that. It is working as of now.