data-science 2020-04-01 | Slack Archive

whilo01:04:30

In case people are interested in probabilistic programming, I think we have put a good show case in our paper about epidemiological simulations https://twitter.com/wh1lo/status/1245142471035637764

🙏 12

whilo01:04:54

Unfortunately that is without Clojure, but that can change in the future 🙂.

grumplet05:04:16

@daslu @chris441 - Thanks for looking! Now isn’t the time to look further, but I feel that there is growing potential for better R/clj or R/cljs integration now we’ve seen what can be done with libpython-clj.

genmeblog10:04:33

I hope you are aware of existence of clojisr project which is Clj and R integration library: https://github.com/scicloj/clojisr

grumplet14:08:11

Thanks for the link. Something to explore later. I ended up using openCPU for that job which worked a treat.

Daniel Slutsky14:08:46

Nice, did you do it through https://github.com/behrica/opencpu-clj ?

grumplet19:08:56

No. I needed an R link to clojurescript. This was to test the next release of https://breast.predict.nhs.uk/ - an upgrade to a revised statistical model developed in R.

👍 2

grumplet19:08:03

The app is entirely client side.

grumplet05:04:42

@whilo. Try this: https://medium.com/wintoncentre/how-much-normal-risk-does-covid-represent-4539118e1196. It offers a different perspective. There is also this accessible article by Julia Gog on modelling https://plus.maths.org/content/how-can-maths-fight-pandemic

➕ 4

alan10:04:58

Updated & upgraded version of the Math for clojurists post, now featuring Neanderthal & tech.datatype https://alanmarazzi.gitlab.io/blog/posts/2020-3-23-math-for-clojurists/

➕ 8

jsa-aerial13:04:10

Very nice! This makes me even more want to turn this into a live Saite doc where users can run the code!

2020-04-01

Channels