[moving to Discourse]
Sure—they just need to make sure they have the right
tools on the back end to compile models.
I didn’t try it because the first thing they wanted to
do was have me authorize email and gist access from GitHub.
- Bob
Did Discourse eat forwarded or quoted message? Title is interesting but couldn’t find what it is about…
Aki
https://rcloud.social/edit.html
It is Linux, so you can install rstan and run Stan programs. But they don’t give you enough memory to install rstanarm.
This is from the original email. (I still don’t know how to have Discourse include whole emails.)
On Dec 16, 2016, at 4:18 PM, Andrew Gelman gelman@stat.columbia.edu
wrote:Hi all. I have no idea what this is, but the people at ATT Labs are
reputable, so it makes me wonder whether it would make sense for us to
talk with them about including Stan?
AP.S. I know that stan-dev is disappearing so I didn’t post there. I
would post this on stan discourse but I can’t remember how to post
there.On Dec 16, 2016, at 11:48 AM, rcloud-social info@rcloud.social
wrote:RCloud(R + Javascript) =3D Happy Holidays!
https://rcloud.social/view.html?notebook=3D5410b97fa32e5292555cc6ead20b770
3
Please let me introduce you to the open-source (free) data analysis
cloud platform called RCloud (RCloud - The Social Coding Environment for Big Data Analytics & Visualization | RCloud). Students
and Data Scientists and other analysts may easily create, collaborate
and share results of R and Python analyses with both registered and
unregistered users.This means you can access an R or Python terminal from anywhere with
an internet connection and share results of data analyses through a URL
as a notebook or dashboard (RCloud - The Social Coding Environment for Big Data Analytics & Visualization | RCloud). Similar to
Jupyter, but cloud based so with RCloud you can access the notebooks of
every user in the cloud - which means analysts may leverage the code of
others - no more =E2=80=9Crecreating the wheel=E2=80=9D and a valuable
resource to learn from and build on!AT&T Big Data Research Labs (including Simon Urbanek, core R
contributor) developed RCloud to solve issues related to
reproducibility, collaboration, deployment time and parallelizing code.
They now have many instances of RCloud installed on big data platforms,
including Hadoop and SparkR, and are using big data packages like H20
for Deep Learning across various data lakes (since it is free, set up as
many instances as needed) where results are shared instantaneously with
both registered and unregistered users.Check out our introductory videos:
https://www.youtube.com/watch?v=3D5Dw3Hqg1foU
https://www.youtube.com/watch?v=3D7z5sq_9qk4UPlease email me at any time if you are interested in installing
RCloud in your organization so that you have your own private RCloud or
if you have any questions. We are not selling anything, we just want to
create a community who will use and contribute to RCloud!Kind regards,
Jo Frabetti
frabetti@research.att.com
Data Science, Machine Learning, AI & Analytics - KDnuggets