JupyterCon 2018 ..
概要を端的に報告していると私が思ったのは.. これ
http://willcrichton.net/notes/lessons-from-jupytercon/
もっとも話題になっているのは... これかな
https://docs.google.com/presentation/d/1n2RlMdmv1p25Xy5thJUhkKGvjtV-dkAIsUXP-AL4ffI/preview?slide=id.g362da58057_0_14
こいつの slide 130 に.. "reproducibility mode" って話があって.. 現在の run_through, freeze に近い idea が提示されていて.. で、"Yeah, something like this would be alleviate a lot of my complaints, BUT I suspect most people who love notebooks would hate this workflow" と、、Jupyter が批判されているのが 興味深いです。
しかし、なんだ shell だってもっと酷い hidden state.. （ひどいひどいステート）あるのになと.. 思ったり
-- 誰か参加しているn人.. 居ます？
#Flipped_Learning
https://gyazo.com/46981bf2d8268932d4903803436e9f03
--
https://conferences.oreilly.com/jupyter/jup-ny/public/schedule/detail/68344
Explorations in reproducible analysis with Notebook - Kevin Zielnicki (Stitch Fix)
2:40pm–3:20pm Friday, August 24, 2018 たぶんこれのネタが以下..
https://jakevdp.github.io/blog/2017/03/03/reproducible-data-analysis-in-jupyter/
#Papermil
https://gyazo.com/18cd44905b62ffa15744bc91da7463e2
Netflix のインフラ.. workflow を ↑ Papermil が
https://gyazo.com/ff78cc78f458817dc7b501ae735d5c05
https://medium.com/netflix-techblog/scheduling-notebooks-348e6c14cfd6
#nbgallery
https://nbgallery.github.io/
https://nbgallery.github.io/minimal_image.html - **Rationale for minimal Jupyter image**
... ”we have developed a minimal Jupyter Docker image (~340MB) based on Alpine Linux. ”
https://nbgallery.github.io/health_paper.html -
#reproducible_publication
Preparing your Jupyter notebook for computationally reproducible publication: A hands-on BYONotebook tutorial for researchers
https://docs.google.com/presentation/d/17h_ylV84m_AY6-0uhYxFcI59nsiYoev4nTESUvoFlrA/edit#slide=id.g3761631756_0_0
https://gyazo.com/0fbdb6f2425ae2b84f27e579a2fed863
https://gyazo.com/7d1e5272302f52d5bc1e18120e2ca7fa
--- 未整理
https://gyazo.com/398076cb61556debacbe14863d4fa77d
共通の感覚か...
https://gyazo.com/346d91426b65059d14847f85df08e8df
CHASM .. technical chasm
https://gyazo.com/4ad3a3912b6674c13adcb5f95b89e36b
そこか..