https://github.com/yyyasin19/discord-emb-search

https://github.com/yyyasin19/discord-emb-search

Last synced: 20 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/yyyasin19/discord-emb-search
• Owner: YYYasin19
• Created: 2024-01-13T13:26:33.000Z (12 months ago)
• Default Branch: main
• Last Pushed: 2024-01-14T22:34:17.000Z (12 months ago)
• Last Synced: 2024-10-16T16:48:23.681Z (2 months ago)
• Language: Python
• Size: 227 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Discord Embedding Search



Being somewhat dissatisfied with Discord's search functionality, I decided to make my own.

This is a simple Discord-Bot implemented in Python that will index messages using a `SentenceTransformer` embedding.

## Usage

### Configuration

Configuratio requires setting up a Discord bot.

After that, set the `BOT_TOKEN` environment variable (e.g. in a local `.env` file).

Finally, invite the bot to your server using the OAuth2 URL generated by the Discord Developer Portal.

You can also set which embeddings to use using the `HF_EMBEDDING_MODEL` environment variable, e.g. setting it to `deutsche-telekom/gbert-large-paraphrase-cosine` will use a German BERT model.

### with Docker

```bash

docker compose up -d --build

```

### locally

```bash

# create environment with (micro)mamba/conda

micromamba env create -f environment.yml

# activate environment

micromamba activate discord-bot

# install package

pip install -e .

# run bot

python -m discord-embed-bot

```

:warning: Note that the first usage might take a few minutes, because the model needs to be downloaded.

The volume mapping in the `docker-compose.yml` file is set up to cache the huggingface models, though.