I can't install the Discord module into Conda

[SvetZitrka]

Description

I would like to use this script in the "oobabooga" environment to help link language models to the Discord bot. But I can't get it to work in the "oobabooga" environment. I probably can't install what Conda module Discord :(.

Is it possible to include this Discord bot feature in the "oobabooga" environment or reveal how to install the Discord module?

Additional Context

[If applicable, please provide any extra information, external links, or screenshots that could be useful.]

[bmoconno]

Check out this fork that has been adapted to work with text-generation-webui

[xNul]

@SvetZitrka hey, thanks for using my bot! To install the discord module, just open the cmd_windows.bat (or cmd_linux.sh or cmd_macos.sh) and run pip install discord. My instructions were updated recently so maybe they are easier to understand now https://github.com/xNul/chat-llama-discord-bot#setup

[SvetZitrka]

Thanks for your work. In the end I had to use this branch which seems to work better https://github.com/mercm8/chat-llama-discord-bot

Are you planning any updates, for this script bot?

[xNul]

@SvetZitrka strange, they are setup the same way.

Yep! I have a lot of plans actually, but I'm focusing on making the installation process easier so I've been working on the webui's one-click installer. I made the new version that everyone has been using for the past week or two and am ironing out the bugs rn. Afterwards, I need to rearchitect the bot a bit to solve xNul/chat-llama-discord-bot#8 and then I'll be going hard mode on features.

[matatonic]

@SvetZitrka
FYI, I've updated openai extension api to include enough support so that the offical gpt-discord-bot from openai works now. (also re: #1504). The pull request is here if you want to try it out: #1475

[xydreen]

@SvetZitrka strange, they are setup the same way.

Yep! I have a lot of plans actually, but I'm focusing on making the installation process easier so I've been working on the webui's one-click installer. I made the new version that everyone has been using for the past week or two and am ironing out the bugs rn. Afterwards, I need to rearchitect the bot a bit to solve xNul/chat-llama-discord-bot#8 and then I'll be going hard mode on features.

Consider the other method of just sending the reply without the embed/constant editing (set status to typing during processing, send reply on complete). Was actually looking to modify the code to do this but the mercm8 fork did this so it's a little less work.

Either way, thanks for the work on all of this everyone.

[github-actions]

This issue has been closed due to inactivity for 6 weeks. If you believe it is still relevant, please leave a comment below. You can tag a developer in your comment.