Overview¶
Why dsco?
Dsco takes an opinionated view of a typical data science workflow and wraps all the necessary tools into a container that can be run locally, on a remote VM, or on kubernetes.
A typical project starts in a Jupyter notebook with some exploratory analysis.
Dsco runs a jupyter notebook server.
Jupyter notebooks are great, but can be hard to share with other people. There are some great tools to convert the notebooks into a static format, but they always requires a bit of time to figure it out again and get everything set up. Then of course we need to figure out how we send people the output. If we’re running in a container, can’t we just park it on a VM somewhere and send them a link? Yes!
Dsco generates and serves static html representations of your notebooks
Uh oh, did I loose a few of you? Set up a VM? Not me! Don’t worry, I have a solution for you too. Just configure your github repo to use github pages and serve content out of your master branch docs folder (don’t worry, it’s not hard).
Dsco generates static html representations of your notebooks that are compatible with github pages
Sounds pretty good, you say, but what if I need to add different python dependencies? No problem.
Dsco gives you the tools to manage your python dependencies. No limits.
If all this sounds useful, then dsco is for you! Head over to
the Quickstart to get going.
If you’re still not convinced, I have some good news: there’s more!