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

https://github.com/app-generator/ai-processor

AI Processor - SQLite / DB information Extractor using Human Language | AppSeed
https://github.com/app-generator/ai-processor

ai-for-databases ai-ml-starter ai-starter ai-template ml-starter ml-template openai-python openai-sqlite python-ai-starter python-ml-starter sqlite-to-ai

Last synced: about 1 month ago
JSON representation

AI Processor - SQLite / DB information Extractor using Human Language | AppSeed

Awesome Lists containing this project

README

        

# `DataBase` [AI Processor](https://github.com/app-generator/ai-processor)

Simple tool that extracts information from an SQLite source using human language queries. The stack uses a `NextJs` frontend and a `Django` (API) for users management and OpenAI interface.


> Download Sources: private repository, for access contact [AppSeed](https://appseed.us/)

```bash
$ git clone https://github.com/app-generator/priv-ai-processor.git
$ cd priv-ai-processor
```


> **Django** Backend

Edit `backend/.env` and add you own `OpenAI API KEY`.

```bash
$ cd backend # change DIR to the backend code
$ virtualenv env # create a new virtual environment
$ source env/bin/activate # activate the VENV
$ pip install -r requirements.txt # install modules
$ python manage.py makemigrations # migrate DB
$ python manage.py migrate # apply DB changes
$ python manage.py runserver # Start the development Server
```

The backend starts on Django's default address: `http://localhost:8000`


> **NextJS** UI

```bash
$ cd frontend # change DIR to the frontend code
$ npm install -g next # Install NextJs globally
$ npm i # install dependencies
$ npm run dev # Start the development Next Server
```


## How to use the tool

> Create a new user or authenticate using the default one:

- **user** : `test`
- **email** : `[email protected]`
- **password** : `pass`

> Add your own OPEN API Key

Access the settings page and save your OpenAI API key

> Upload a new SQLite file

Navigate to the SQLite Uploads file and add a new file. Once uploaded, we can query start quering the database.

> Query the information using OpenAI console

Here are some query samples:

- `List all tables registered in the database`
- `List all products starting with the cheapest`

Once another SQLite file is uploaded, we can query other specific questions.


## Tools

Outside UI, we can query different sources like PDF files or distant APIs:

```bash
$ cd tools
$ virtualenv env # create a new virtual environment
$ source env/bin/activate # activate the VENV
$ pip install -r requirements.txt # install modules
$ vi .env # Save OpenAI KEY
$ python ai-over-api-meteo.py # Extract METEO information using distant API
$ python ai-over-pdf.py # Extract information from a local PDF file

```


## License

[@EULA](https://github.com/app-generator/license-eula)


---
`DataBase` [AI Processor](https://github.com/app-generator/ai-processor) - AI/ML Starter provideed **[AppSeed](https://appseed.us)**.