Fantastic! Great work, Buildpacks team. I'm especially excited that this
will allow data scientists to more easily run their applications on Cloud
Foundry.


On Wed, May 18, 2016 at 1:56 PM, Danny Rosen <drosen(a)pivotal.io> wrote:

Over the last few weeks the Buildpacks team has experimented with
different methods of getting the Python buildpack working with various data
science dependencies (ex: sci-py, num-py, sklearn). We believe we have
found a solution that does not create significant challenges and fulfills
the underlying goal of enabling these dependencies in a native CF
application.

We have included the addition of Miniconda into the newest version [1
<https://github.com/cloudfoundry/python-buildpack/releases/tag/v1.5.6>]
of the python-buildpack and urge you to give it a try. To do so, include an
environment.yml [2] in your app and cf push *or* try out this sample app
[3 <https://github.com/ihuston/pydata_package_test>].

We're very excited about this addition and note that this is our first
step towards providing the functionality. As always, we're interested in
obtaining feedback and pull requests for improvement.

[1] -
https://github.com/cloudfoundry/python-buildpack/releases/tag/v1.5.6
[2] -
http://conda.pydata.org/docs/using/envs.html#use-environment-from-file
[3] - https://github.com/ihuston/pydata_package_test/