Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Simple chat app for beginners #919

Closed
rickyloynd-microsoft opened this issue Dec 8, 2023 · 11 comments
Closed

Simple chat app for beginners #919

rickyloynd-microsoft opened this issue Dec 8, 2023 · 11 comments
Assignees
@danyrouh
Labels
setup installation and setup issues

Comments

@rickyloynd-microsoft
Copy link
Contributor

As a first step, many new users struggle to run the NVDA-TESLA example or the notebooks, since they all involve various kinds of complexity. Our discord users often ask for simpler sample code that just replicates a basic, chatgpt-like conversational experience. Such an applet would also be convenient for anyone to use when testing new model configuration settings, or a new LLM.
@rickyloynd-microsoft rickyloynd-microsoft added the setup installation and setup issues label Dec 8, 2023
@rickyloynd-microsoft rickyloynd-microsoft self-assigned this Dec 8, 2023
@danyrouh danyrouh mentioned this issue Dec 8, 2023
Merged
3 tasks
@janaka
Copy link

janaka commented Dec 10, 2023 •
edited
Loading

Happy to see this. A few thoughts though:

I feel a big contributor to the issue the design of ConversableAgent. The current implementation has console I/O hardcoded rather than using an abstraction.

Re the simple web app sample:

  • pls keep it all Python
  • less than 5 commands after git clone to run it

🙏🏾

@rickyloynd-microsoft
Copy link
Contributor Author

Happy to see this. A few thoughts though:

I feel a big contributor to the issue the design of ConversableAgent. The current implementation has console I/O hardcoded rather than using an abstraction.

Re the simple web app sample:

  • pls keep it all Python
  • less than 5 commands after git clone to run it

🙏🏾

@janaka Does #921 follow your suggestions?

@janaka
Copy link

janaka commented Dec 10, 2023

Happy to see this. A few thoughts though:

I feel a big contributor to the issue the design of ConversableAgent. The current implementation has console I/O hardcoded rather than using an abstraction.

Re the simple web app sample:

  • pls keep it all Python
  • less than 5 commands after git clone to run it

🙏🏾

@janaka Does #921 follow your suggestions?

So that looks like a CLI app - basically what's in the quick start doc.

I would like to see the same but with a web UI. If you use a framework like Streamlit or Reflex it will avoid a web API layer.

@rickyloynd-microsoft
Copy link
Contributor Author

Yes, #921 is a CLI. We also have a new web UI.

@janaka
Copy link

janaka commented Dec 10, 2023

Yes, #921 is a CLI. We also have a new web UI.

That web UI sample is way to complicated with the flow* abstractions, web api, and gatsby.

@janaka
Copy link

janaka commented Dec 10, 2023

#740 is asking the same as me.

@rickyloynd-microsoft
Copy link
Contributor Author

If somebody wants to write a simple UI to demo basic autogen functionality, I'm sure we would add a link to it in the repo and docs.

@janaka
Copy link

janaka commented Dec 11, 2023

fwiw @victordibia 's early version here https://github.com/victordibia/autogen-ui seems like the right level. Even though the same layers exist in that as in the latest version included in this repo, those layers are far simpler to follow.

It needs a little work updating to work with autogen v0.2.*

@sonichi
Copy link
Contributor

sonichi commented Dec 11, 2023

There are more UI examples using Streamlit, Chainlit, etc., which haven't been added to https://microsoft.github.io/autogen/docs/Gallery.

@janaka
Copy link

janaka commented Dec 11, 2023

PR for autogen v0.2.*. It was pretty simple to fix. /cc @victordibia

There's value in this particular example as it's a web app version of the example in getting started with the same prompt.

I don't know you alls criteria to include in this repo but linking in the getting started next to the CLI example is probably a useful.

victordibia/autogen-ui#16

🙂

whiskyboy pushed a commit to whiskyboy/autogen that referenced this issue Apr 17, 2024
@coffepowered @andreaw-ag @qingyun-wu
Improve annotations in automl and ml modules (microsoft#919)
8263310 
* begin annotation in automl.py and ml.py

* EstimatorSubclass + annotate metric

* review: fixes + setting fit_kwargs as proper Optional

* import from flaml.automl.model (import from flaml.model is deprecated)

* comment n_jobs in train_estimator as well

* better annotation in _compute_with_config_base

Co-authored-by: Qingyun Wu <qingyun.wu@psu.edu>

---------

Co-authored-by: Andrea W <a.ruggerini@ammagamma.com>
Co-authored-by: Qingyun Wu <qingyun.wu@psu.edu>
@thinkall
Copy link
Collaborator

Closing this issue due to inactivity. If you have further questions, please open a new issue or join the discussion in AutoGen Discord server: https://discord.com/invite/Yb5gwGVkE5

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@danyrouh danyrouh

Labels
setup installation and setup issues
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
5 participants
@janaka @thinkall @sonichi @rickyloynd-microsoft @danyrouh