Back to Community
Using Zipline locally vs. Quantopian's platform

Hey everyone. Would anyone mind clarifying some the differences between running Zipline locally as opposed to using Quantopian's online platform? Also, will minute data be available if I run my algorithms on a locally hosted platform? Thank you!

4 responses

The major differences are based the fact that zipline is an open-sourced backtester that is used both independently and in Q's over products.

Differences include:
- the datasets available...lotsa free-for-use-in-Q-products ones,
-the web-based ide(with debugger)
-the research(jupyter-based) web-based gui
-open sourced add-ons(e.g. pyfolio)
-non-open sourced code( e.g. alphalens, optimization api)
-ability in Zipline to enhance it with your own customization(e.g. GPU, other libs) and remove resource constraints involving computing time/space.

Look at the link Zipline group for community action
Here is the open sourced code on GitHub Zipline github


I think it is just a typo in Alan's response above, but just to clarify alphalens is open-source and available here .

Here is an example jupyter notebook that demonstrates how to use zipline to create the two python pandas objects that are used as input to alphalens .


@Richard, yes, good catch...and nice example!
Alphalens is open sourced. Sorry for the mistake.

Just need to look at the Q github repositories to see what is open sourced.
My belief is that the code in the tree is all open sourced.


@Alan @Richard Thank you both for the great replies. Any idea as to whether or not minute data is available as well if I run locally? Thanks!