There are two ways to get started:
superduperdb is available on PyPi.org.
Before you begin the installation process, please make sure you have the following prerequisites in place:
superduperdb is compatible with several Linux distributions, including:
If you plan to install SuperDuperDB from source, you'll need the following:
pip22.0.4 or later
Your experience with
superduperdb on Linux may vary depending on your system and compute requirements.
superduperdb on your system using
pip, open your terminal and run the following command:
pip install superduperdb
This command will install
superduperdb along with a minimal set of common dependencies required for running the framework. Some larger dependencies, like
pytorch, are not included to keep the image size small. You can install such dependencies using the following syntax:
pip install superduperdb[<category>]
Here are the available categories you can use:
api: Installs clients for third-party services like OpenAI, Cohere, and Anthropic.
torch: Installs PyTorch dependencies.
docs: Installs tools for rendering Markdown files into websites.
quality: Installs tools for aiding in the development of high-quality code.
testing: Installs tools for testing the SuperDuperDB ecosystem.
dev: Installs all the above categories.
You can find more details on these categories in the pyproject.toml file.
Using Pre-built Images
If you prefer using Docker, you can pull a pre-built Docker image from Docker Hub and run it with Docker version 19.03 or later:
docker run -p 8888:8888 superduperdb/superduperdb:latest
This command installs the base
superduperdb image. If you want to run the ready-to-use examples, you'll need to download the required dependencies at runtime.
Building the image yourself
For more control, you can build the Docker images yourself using the following commands: