An open API service indexing awesome lists of open source software.

https://github.com/tina4stack/thoughtdb

ThoughtDB is a mix between traditional relational and vector database for efficient data retrieval
https://github.com/tina4stack/thoughtdb

Last synced: about 1 year ago
JSON representation

ThoughtDB is a mix between traditional relational and vector database for efficient data retrieval

Awesome Lists containing this project

README

          

# ThoughtDB
ThoughtDB is a mix between relational and vector database for efficient data retrieval

## Installation

### Windows
```bash
python -m venv .venv
.\.venv\Scripts\activate
pip install poetry
```

### MacOS / Linux
```bash
python3 -m venv .venv
source ./.venv/bin/activate
pip install poetry
```

```bash
poetry run python app.py 0.0.0.0:8003
```

### Text Embedding Models

Download the following model or model of your choice in GGUF format into the models_db folder
```bash
https://huggingface.co/nomic-ai/nomic-embed-text-v1.5-GGUF/resolve/main/nomic-embed-text-v1.5.Q4_K_M.gguf?download=true
```

# Data Structure

The following is the layout for the ThoughtDB data structure. As we are in Alpha phase this structure may change as we optimize things.

- Organizations
- Collections
- Conversations
- Sessions
- History
- Summaries
- Document Types
- Documents
- Chapters
- Paragraphs
- Sentences
- Words
- Images
- Audio

# Overview

## Installing

Installation should be simple

### Pip
```bash
pip install thoughtdb
```

### Poetry
```bash
poetry add thoughtdb
```

## Creating a new database

Currently, we only support Sqlite3, all the dependencies should be installed by pip or poetry.

```python
from thoughtdb import VectorStore

```

# Building

```bash
python3 -m pip install --upgrade build
python3 -m build
python3 -m pip install --upgrade twine
python3 -m twine upload --repository pipy dist/*
```

# Testing

```bash
pytest tests --verbose
```

# Todo