Closed MarcSkovMadsen closed 11 months ago
I've marked this ready for review. I've not checked all the boxes. But I compared to what Streamlit and Gradio has done. And then I tried to do the same. Please help me out as a first time contributor and let me know what needs to be done.
Thanks.
at a first glance this looks really interesting, will review!
alright @MarcSkovMadsen, we got this over the line! Thanks for all your effort on this. The reason why test was failing was since we were running .run
in base.py. I changed the warning back to raising a value error -- hope you're cool with that!
I'm VERY cool with that. Thanks for for all the help.
When will this be available on the Hub?
Great! I'll cut a release for sure tomorrow :)
@MarcSkovMadsen thanks for your awesome contribution! just a heads up, planning to feature this tomorrow on our socials.
The LlamaPack page is here: https://llamahub.ai/l/llama_packs-panel_chatbot?from=llama_packs - the asset doesn't seem to be rendering but doesn't prevent us from publishing this
Thanks. Do you know how I can best get the asset working?
A workaround would be to use the github link https://raw.githubusercontent.com/run-llama/llama-hub/main/llama_hub/llama_packs/panel_chatbot/panel_chatbot.png
like below
yeah was just about to suggest that :)
I identified some more issues including the app would not be be served via the commands described.
I hope I can get the fixes in https://github.com/run-llama/llama-hub/pull/759 in before you start promoting it. Otherwise users might get a bad experience. @jerryjliu
I identified some more issues including the app would not be be served via the commands described.
I hope I can get the fixes in #759 in before you start promoting it. Otherwise users might get a bad experience. @jerryjliu
taking a look now, thanks for flagging
Description
[HoloViz Panel]() is a powerful framework for data exploration and data apps. It has ~750.000 downloads per month and a very active community. Panel 1.3.0 added powerful chat components that support
async
, multi-modal, multi-agent interfaces. It can display any Python object in its chat interface including markdown, plots, dataframes, audio, video etc. Check out Panel-Chat-Examples to see some examples.I'm biased as a contributor to Panel and the Panel chat components. But luckily others also find our chat components really valuable.
.
Panel is also used in courses by Andrew Ng on deeplearning.ai. Probably because we also work really well in notebooks.
Thus I would like to add a Panel Chat Pack to LLama Hub to help users get started.
Panel Chat Pack in Action
https://github.com/run-llama/llama-hub/assets/42288570/e6db098d-96f8-4502-8ef1-046360c9f221
How Has This Been Tested?
Suggested Checklist:
make format; make lint
to appease the lint gods