β οΈ envd is still under heavy development, and subject to change. it is not feature-complete or production-ready. Please contact us in discord if there is any problem.
envd is a machine learning development environment for data science and AI/ML engineering teams.
π No Docker, only Python - Focus on writing Python code, we will take care of Docker and development environment setup.
π¨οΈ Built-in Jupyter/VSCode - First-class support for Jupyter and VSCode remote extension.
β±οΈ Save time - Better cache management to save your time, keep the focus on the model, instead of dependencies.
βοΈ Local & cloud - envd
integrates seamlessly with Docker so that you can easily share, version, and publish envd
environments with Docker Hub or any other OCI image registries.
π Repeatable builds & reproducible results - You can reproduce the same dev environment on your laptop, public cloud VMs, or Docker containers, without any change in setup.
It is still too difficult to configure development environments and reproduce results in AI/ML applications.
envd
is a machine learning development environment for data science and AI/ML engineering teams. Environments built with envd
provide the following features out-of-the-box:
π Life is short, use Python1
Development environments are full of Dockerfiles, bash scripts, Kubernetes YAML manifests, and many other clunky files that are always breaking. envd
builds are isolated and clean. You can write simple instructions in Python, instead of Bash / Makefile / Dockerfile / ...
β±οΈ Save you plenty of time
envd
adopts a multi-level cache mechanism to accelerate the building process. For example, the PyPI cache is shared across builds and thus the package will be cached if it has been downloaded before. It saves plenty of time, especially when you update the environment by trial and error.
envd |
Docker2 |
$ envd build
=> pip install tensorflow 5s
+ => Using cached tensorflow-...-.whl (511.7 MB) |
$ docker build
=> pip install tensorflow 278s
- => Downloading tensorflow-...-.whl (511.7 MB) |
βοΈ Local & cloud native
envd
integrates seamlessly with Docker, you can share, version, and publish envd
environments with Docker Hub or any other OCI image registries. The envd
environments can be run on Docker or Kubernetes.
π Repeatable builds & reproducible results
You can reproduce the same dev environment, on your laptop, public cloud VMs, or Docker containers, without any change in setup. You can also collaborate with your colleagues without "let me configure the environment in your machine".
π¨οΈ Seamless experience of Jupyter/VSCode
envd
provides first-class support for Jupyter and VSCode remote extension. You benefit without sacrificing any developer experience.
Weβre focused on helping data scientists and teams that develop AI/ML models. And they may suffer from:
- building the development environments with Python, CUDA, Docker, SSH, and so on. Do you have a complicated Dockerfile or build script that sets up all your dev environments, but is always breaking?
- Updating the environment. Do you always need to ask infrastructure engineers how to add a new python package in the Dockerfile?
- Managing environments and machines. Do you always forget which machines are used for the specific project, because you handle multiple projects concurrently?
Before envd | After envd |
---|---|
See envd documentation.
- Docker (20.10.0 or above)
envd
can be installed with pip
. After the installation, please run envd bootstrap
to bootstrap.
pip install envd
envd bootstrap
You can add
--dockerhub-mirror
or-m
flag when runningenvd bootstrap
, to configure the mirror for docker.io registry:envd bootstrap --dockerhub-mirror https://docker.mirrors.sjtug.sjtu.edu.cn
Please clone the envd-quick-start
:
git clone https://github.com/tensorchord/envd-quick-start.git
The build manifest build.envd
looks like:
def build():
base(os="ubuntu20.04", language="python3")
install.python_packages(name = [
"numpy",
])
shell("zsh")
Then please run the command below to set up a new environment:
cd envd-quick-start && envd up
$ cd envd-quick-start && envd up
[+] β parse build.envd and download/cache dependencies 2.8s β
(finished)
=> download oh-my-zsh 2.8s
[+] π build envd environment 18.3s (25/25) β
(finished)
=> create apt source dir 0.0s
=> local://cache-dir 0.1s
=> => transferring cache-dir: 5.12MB 0.1s
...
=> pip install numpy 13.0s
=> copy /oh-my-zsh /home/envd/.oh-my-zsh 0.1s
=> mkfile /home/envd/install.sh 0.0s
=> install oh-my-zsh 0.1s
=> mkfile /home/envd/.zshrc 0.0s
=> install shell 0.0s
=> install PyPI packages 0.0s
=> merging all components into one 0.3s
=> => merging 0.3s
=> mkfile /home/envd/.gitconfig 0.0s
=> exporting to oci image format 2.4s
=> => exporting layers 2.0s
=> => exporting manifest sha256:7dbe9494d2a7a39af16d514b997a5a8f08b637f 0.0s
=> => exporting config sha256:1da06b907d53cf8a7312c138c3221e590dedc2717 0.0s
=> => sending tarball 0.4s
(envd) β demo git:(master) β # You are in the container-based environment!
You can run ssh envd-quick-start.envd
to reconnect if you exit from the environment. Or you can execute git
or python
commands inside.
$ python demo.py
[2 3 4]
$ git fetch
$
Please edit the build.envd
to enable jupyter notebook:
def build():
base(os="ubuntu20.04", language="python3")
install.python_packages(name = [
"numpy",
])
shell("zsh")
config.jupyter(password="", port=8888)
You can get the endpoint of the running Jupyter notebook via envd get envs
.
$ envd up --detach
$ envd get env
NAME JUPYTER SSH TARGET CONTEXT IMAGE GPU CUDA CUDNN STATUS CONTAINER ID
envd-quick-start http://localhost:8888 envd-quick-start.envd /home/gaocegege/code/envd-quick-start envd-quick-start:dev false <none> <none> Up 54 seconds bd3f6a729e94
Please checkout ROADMAP.
We welcome all kinds of contributions from the open-source community, individuals, and partners.
- Join our discord community!
- To build from the source, please read our contributing documentation and development tutorial.
Thanks goes to these wonderful people (emoji key):
This project follows the all-contributors specification. Contributions of any kind welcome!