# aliquote

## < a quantity that can be divided into another a whole number of time />

Here is the latest bag of tweets*, which covers March 2019.

(*) These are interesting news that I found on Twitter and that I archive periodically. Note that I have slightly updated reporting and fetching methods so that you will get between 80 and 120 of my most recent favorites from my Twitter account (@even4void).

• 2019-03-01 @tallphil: A fun mini-project: SRA-Explorer allows you to collect SRA datasets and get a quick bash download script for either SRA files or now FastQ files! (courtesy of the ENA API) đ https://t.co/UdohG4nHnN https://t.co/YIL5WmoSHj
• 2019-03-01 @rlmcelreath: Done! All 20 Lectures of Statistical Rethinking 2019 â new material includes: Causal inference, instrumental variables, regression splines, better HMC notes, ordered categorical predictors, phylogenetic regression, more https://t.co/2G9FdWge54 https://t.co/kZNz53aAIS
• 2019-03-01 @freakonometrics: “Fundamentals of Linear Algebra and Optimization” https://t.co/kWwgOpn12m (the one thousand page bible)
• 2019-03-01 @stat110: New edition of my probability book with Jessica Hwang is out! Read it FREE online at https://t.co/2iwKDNe0ag https://t.co/5z1zOkJ3Y5
• 2019-03-01 @wrathematics: @vsbuffalo This may interest you https://t.co/GjEJwvHkPC
• 2019-03-01 @driven_by_data: And today I forked Adams chart to use Loess regression instead of the linear model, because I thought temperature isn’t increasing linearly (but maybe exponentially?) https://t.co/GukCajAPKe https://t.co/0tKBe8Gtus
• 2019-03-01 @mattdesl: Testing some unusual ways of displaying a book. (Text from “Pride and Prejudice” by Jane Austen) https://t.co/hTf2jMKIYI
• 2019-03-01 @ttscoff: Lately on https://t.co/K01VXNmbZ0: Your security and privacy: next steps https://t.co/NllBNnPvSz
• 2019-03-02 @hmemcpy: A new version of Category Theory for Programmers PDF by @BartoszMilewski is now available! What’s new? _ Syntax highlighting! _ Scala edition (huge thanks to @impurepics!) * Many more fixes (see https://t.co/vlGU6ssEHV for details) Download here: https://t.co/HIQ0Vzwt2Q https://t.co/5qzzqaPW8p
• 2019-03-02 @AdamChainz: #python tip: taking the first item of a list comprehension is better done as next() on its generator version, to avoid creating the whole list In [1]: [x for x in range(100) if x > 10][0] # okay Out[1]: 11 In [2]: next(x for x in range(100) if x > 10) # better Out[2]: 11
• 2019-03-03 @newsycombinator: Functional Programming in OCaml (2019) https://t.co/nN9OXXHBWz
• 2019-03-03 @dvaughan32: Have you ever wanted to create an R package that calls C code, but didn’t know where to start? Look no further! Now You C Me is a post dedicated to teaching you the bare minimum to create an #rstats package with C that passes CRAN’s checks. Enjoy! https://t.co/JuuosiETq0
• 2019-03-03 @freakonometrics: “A problem in theory” https://t.co/CWTcARfGxC (ht @cathleenogrady’s https://t.co/lFefVLj66I) https://t.co/IvSax3d3SG
• 2019-03-03 @hrbrmstr: #rstats CRAN Mirror âSecurityâ https://t.co/fiNU5Zsm0Q https://t.co/UFOB0VWNNI
• 2019-03-03 @newsycombinator: Python Data Science Handbook: Full Text in Jupyter Notebooks https://t.co/HiNmBLJGrv
• 2019-03-04 @kaz_yos: Tutorials on Bayesian Nonparametrics This page collects references and tutorials on Bayesian nonparametrics: https://t.co/nV55l4pusy
• 2019-03-04 @pascalbugnion: Excited to announce the release of jupyterlab-sql, a SQL GUI for @ProjectJupyter Lab. Installation instructions and documentation on: https://t.co/wPqyf6u5aY https://t.co/wQ6ZN5YpW7
• 2019-03-04 @IKosmidis_: Fresh paper on mean and median bias reduction in GLMs is out (open access)! https://t.co/oYvT7ajRPq Alternatives to ML, IWLS, parameterization invariance, mixed bias reduction, improved inference + more! See my {brglm2} #rstats package for implementation https://t.co/hGsEL229wr https://t.co/IyMCmuFNcz
• 2019-03-04 @ConcejeroPedro: R for Experimental Design an update of the classical exp. design manual by R. E. Kirk. Fancy names as split-plot, latin square, confounded and fractional factorial designs. Free download! https://t.co/0q6oy0gs9B #rstats
• 2019-03-04 @andreymokhov: Is there any intermediate abstraction between applicative functors and monads? And if yes, what is it? In a new paper with @geo2A, @simonmar and @dimenix we explore “selective functors”, which are essentially applicative functors with branching: https://t.co/LHljA0iiUQ https://t.co/tOgdlb1sGG
• 2019-03-04 @newsycombinator: Google reveals âhigh severityâ flaw in MacOS kernel https://t.co/SXQAr3vAlm
• 2019-03-04 @maartenzam: The slides of my talk “Pitfalls in data visualisation, and how to overcome them” at #openbelgium19 https://t.co/hnr03C7f27 (contains 18 of them!) https://t.co/oIAjSE6M6k
• 2019-03-04 @theotheredgar: đŠNew versions of dbplot & modeldb were published to CRAN this weekend. These versions simplify the #rstudioconf demo significantly. The updated demo can be found in the same link from the original tweet. https://t.co/gh1uIjTV45
• 2019-03-04 @garyweissman: I’ve been fortunate to review a number of papers on clinical prediction models this past year and I’ve noticed a number of frequent errors. This thread is mostly for the clinician-investigator considering building and publishing a predictive model of some kind. 3 major points. /1
• 2019-03-04 @garyweissman: A key review of performance metrics for prediction models, what they mean, how they compare: https://t.co/tRNfU0umks Personally, I find the scaled Brier Score to be the most immediately informative in a single value for most cases. /13
• 2019-03-04 @FunctorFact: ‘Literature about Lisp rarely resists that narcissistic pleasure of describing Lisp in Lisp.’ – Christian Queinnec, Lisp in Small Pieces
• 2019-03-04 @newsycombinator: What Every Programmer Should Know About Memory (2007) [pdf] https://t.co/iB5IiBjGVU
• 2019-03-04 @kaz_yos: LaTeX fontawesome package ftp://ftp.dante.de/tex-archive/fonts/fontawesome/doc/fontawesome.pdf https://t.co/p3088rOQ4X
• 2019-03-04 @bascule: 20 years of progress: 1999: IRC, ICQ, AIM, YIM Early 2000s: gaim (nee Pidgin, and other libpurple clients like Adium) provide unified UX across IM protocols 2019: We all run 15 chat apps again. Slack has nothing to do with J.R. Bob Dobbs, or Discord with Eris.
• 2019-03-05 @kaz_yos: âidea behind the Lebesgue integral is that instead of approximating the total area by dividing it into vertical strips, one approximates the total area by dividing it horizontally. This corresponds to asking “for each y, how many x produce this value?” https://t.co/kaXOti5Xlx
• 2019-03-05 @zevross: Tiny #rstats tip: if text reading functions are guessing column types incorrectly read all as character and use readr::type_convert() | https://t.co/XgDIpn8lux https://t.co/ebNP8bBrzA
• 2019-03-05 @CompSciFact: Base 85 encoding. Used in PDF files and git patches. https://t.co/BZ3qtjsHeU
• 2019-03-05 @skmorgane: Our modern data workflow for regularly updated data paper is now out! We use #rstats, @github, & @travisci to automatically check field data, acquire sensor data, update supplementary tables and archive the data to @ZENODO_ORG https://t.co/TGcJaV2mz8
• 2019-03-05 @grrrck: đ #rstats and knitr tip! You can apply custom CSS to source and output code using the class.source and class.output chunk opts. đ And with knitr 1.22 (on GitHub now), you can apply custom styles to messages, warnings and errors! đ @xieyihui! https://t.co/p1EzVBOCOE
• 2019-03-05 @juliabloggers: How to make package and set up Travis for JuliaÂ 1.0 https://t.co/tWHiXv5fQk https://t.co/wIRGNmjh0d
• 2019-03-05 @strategist922: Collection of probabilistic models and inference algorithms - https://t.co/4peo3HN7iW https://t.co/PVJftQfOqo
• 2019-03-05 @NlightNFotis: Holy smokes, this is amazing https://t.co/NnT8U4UVNa #racket #racketlang A shell that I can import arbitrary racket modules and pass text streams to racket functions. Yes please. Can’t wait to play with this.
• 2019-03-05 @jesse_vig: New article and updated visualization tool for @OpenAI #GPT2! See how the pre-trained language model is able to finish your sentence. Article: https://t.co/pQvjfsU7pG GitHub: https://t.co/PQIoTjnyhx Colab: https://t.co/DhLrrVccLo Play with it and share what you find! https://t.co/gihmIEkbqn
• 2019-03-06 @nickchk: I’m trying to put together a list of packages that make it easy to get standard data sets into R, ideally without even downloading files. Focusing on stuff USA UG econ students might use for projects. Any ideas? So far: tidycensus, gapminder, quantmod/tidyquant, gtrendsR, ipumsr
• 2019-03-06 @MarcoWirthlin: @kaz_yos There is also a LaTeX package just for DAGs and Bayesian nets. The experience so far has been better than with yEd, because LaTeX supports math symbols and the graphs adapt better to de report. https://t.co/ocKy9K01lk
• 2019-03-06 @math3ma: A nice fact I like: Every matrix corresponds to a graph, and so familiar things (e.g. matrix multiplication) have nice pictures! Another nice fact: joint probability distributions also correspond to graphs. They have telling pictures, too. New blog post! https://t.co/EP67IhGAr0 https://t.co/tqdqImhX9P
• 2019-03-06 @axiomsofxyz: And so begins the #rstats #RNGpacpocalypse related to sampling() changes in R 3.6.x brought on by @kellieotto @philipbstark https://t.co/YlVvh4jjWY (Shout out to John Mount for naming) https://t.co/FagHrKd61e
• 2019-03-07 @newsycombinator: Seven Unix Commands Every Data Scientist Should Know https://t.co/aNNvPn89eB
• 2019-03-07 @grrrck: TIL that @rstudio 1.2 uses header comments in your code (# looks like this —-) to give you feedback on the progress of your script when you “Source as Local Job”. Very cool! #rstats https://t.co/cA0N9xlRV4
• 2019-03-07 @bigdata: You created a machine learning application. Now make sure itâs secure âïž itâs been a busy week, l am looking forward to coming to #rsac https://t.co/xbTbrXoRNT
• 2019-03-07 @paytonjjones: Here is a short tutorial on the “networktree” R package for anyone interested in learning more about this new method for network analysis! I will be posting my #icps19paris slides from today’s talk on the topic soon. https://t.co/6FkWsoX1rG https://t.co/L0OcYXAmw5
• 2019-03-07 @NicolasDeux2: CrĂ©er une url raccourcie et un qr-code en 1 clic. Je ne le quitte plus depuis que @ErunReimsOuest me lâa montrĂ©. Merci @framasoft Et dĂ©solĂ© pas de lien direct car tweeter ne semble pas apprĂ©ciĂ© ce site… allez savoir pourquoi đ donc une capture dâĂ©cran https://t.co/pgBxDN2DGz
• 2019-03-07 @timgluz: Found a comment from 1995: Â« static constexpr uint32_t BASEX = 0x80000000; // Internal radix used in calculations, hope to raise // this to 2^32 after solving scaling problems with // overflow detection esp. in mul Â» #lowhangingfruit #firstcommit #oss
• 2019-03-07 @_wilfredh: The thing about memory ownership is that it’s intrinsic (not incidental) complexity on a physical computer. If you don’t want GC, you need ownership. Not all languages provide ownership tooling though (e.g. C). Following Rust’s success I suspect we’ll see more ownership tools. https://t.co/CCmCnX381M
• 2019-03-08 @HNTweets: How does Apple keep secrets so well? (2012): https://t.co/BsTiv9sBLr Comments: https://t.co/GaUu73fWZk
• 2019-03-08 @gappy3000: “Probability and the Real World” by David Aldous (Berkeley). Thoughts about probability and the limits of knowledge. By an excellent probabilist, but non-technical. Love this. https://t.co/0vytEQA6eG
• 2019-03-08 @TeXtip: Math symbols to copy and paste, e.g. into a tweet https://t.co/S2YqeqouRz
• 2019-03-08 @kaz_yos: Two-step epigenetic Mendelian randomization: a strategy for establishing the causal role of epigenetic processes in pathways to disease. https://t.co/qoV01DDoOX https://t.co/hqAzU1wZeJ
• 2019-03-08 @cecilejanssens: The area under the ROC curve (AUC) is so frequently criticized and misunderstood that I often wonder whether I am the metricâs only fan. Let me explain why how I see and value the AUC. (thread) https://t.co/ea69aX2x38
• 2019-03-09 @sanityinc: In Purescript you can elegantly traverse a collection once and calculate several different summaries: https://t.co/CqJyQzsHOJ
• 2019-03-09 @StatsTommy: And then, after another day of odd issues in code and version control, this article by @juliasilge https://t.co/a3voYX0Rj3 was an awesome primer for building continuous integration and automated testing in #rstats
• 2019-03-09 @randomjohn: @abmathewks Part of it is legacy systems, part is the mystique surrounding “validated systems”, and part is the fact that SAS instituted certification a long time ago and enabled people to build careers around this one piece of software.
• 2019-03-09 @kjhealy: BTW if you’re scatterplotting two scaled variables, then you can use ggforce’s geom_circle() to e.g. draw circles to show which observations are within 1 and 2 standard deviations of the mean. Use coord_fixed() to make sure the circles stay circular when drawn. https://t.co/R8mKU2tx7p
• 2019-03-09 @aschinchon: Marine Sponges #rstats #tidyverse #maths #creative https://t.co/PPGzyvO4Xd
• 2019-03-09 @chasewnelson: This book is a veritable thought party â a back-to-basics, crystal clear refresher on what makes R elegant and powerful, and a delightfully easy read. Highly recommended! #rstats #oreilly #Genomics #BioInformatics #datascience https://t.co/uhG3bZQOpT
• 2019-03-09 @dataandme: ICYMI, â° intro to a very cool time-series pkg! “Introduction to the tsfeatures package” by Yangzhuoran Yang & @robjhyndman https://t.co/RCzHa2Dpwq #rstats /_ đŹ couldn’t fit it all on one page _/ https://t.co/MLIaLE96Tw
• 2019-03-09 @george_berry: I wrote a short and hopefully understandable intro to causal inference. I hope it’ll be useful for infosci folks in particular. PDF: https://t.co/ulBrhiyw0E GitHub: https://t.co/BKS9o4ZiF7
• 2019-03-09 @kaz_yos: âThe methods that we describe in this tutorial apply equally to settings in which competing risks are independent of one another and to settings in which competing risks are not independent of one another.â https://t.co/l5W9YdArmw
• 2019-03-09 @stathissideris: Tiny bit of #clojure code to easily avoid reprocessing files that haven’t changed. https://t.co/2iuFBNtIWY
• 2019-03-09 @stathissideris: Lisp-to-Lisp cross-contamination! Using syntax-quoting in #clojure to template some #Emacs Lisp code that will be evaluated in Emacs batch mode. The important point is that the code template is a data structure, not a string! https://t.co/YVWZOsJQNZ
• 2019-03-10 @TechSparx: Common Probability Distributions. Highly recommend this wonderfully easy-to-understand explanation of the most important probability distributions. https://t.co/ChNFUDyaCY https://t.co/Uw03V1HeSQ
• 2019-03-10 @bodil: I’ve been telling you for a while that MIT and BSD are outdated and you should be using Apache 2 instead (or, better yet, a strong copyleft to prevent exploitation, but baby steps etc). Now hear it from a lawyer. https://t.co/8uSstyilaT
• 2019-03-10 @precariobecario: First full draft of my book on INLA “Bayesian inference with INLA and R-INLA” to be published by @CRC_MathStats is available on-line here: https://t.co/0IDJrJpIDz Feedback welcome!!!
• 2019-03-10 @gappy3000: “Clustering of time series is meaningless” https://t.co/XzkbuluTpM
• 2019-03-10 @johnson_davies: An example showing how #Lisp gives you the choice of an imperative or functional programming style: https://t.co/aP4jT3b5iT
• 2019-03-10 @Jose_A_Alonso: Rust for linear algebra and neural networks. ~ J.P. Moresmau. https://t.co/vPzW6dCqOI #RustLang #AI #MachineLearnig #Math
• 2019-03-10 @TechSparx: Generating Uniformly Random points on a d-sphere and d-ball. My blog post gives a comprehensive list of the 20 most common methods to sample from a d-sphere or d-ball. https://t.co/2ScnVOyOrd https://t.co/buDdCcT0nr
• 2019-03-10 @phil_eaton: Adding an LLVM backend to ulisp (small Lisp compiler in JavaScript) https://t.co/VIddKW1r3N
• 2019-03-10 @AdrianRaftery1: Bayesian Model Averaging: A Systematic Review and Conceptual Classification. A very comprehensive review by Tiago Fragoso, Wesley Bertoli and Francisco Louzada. https://t.co/b4dhafCj5L
• 2019-03-10 @fwoaroof: I happened upon a Lisp Metadata Importer on lemonodor, and then I tweaked the source code to work on modern macOS. Now, I get information about definitions in Finder and can use mdfind to search for various kinds of lisp definitions. https://t.co/EjLGrPxMQv https://t.co/gbPGOxL3Kg
• 2019-03-11 @kevin_ushey: I wrote up a small post-mortem on an RStudio bug (https://t.co/aW0tyWoj0i) that took quite a bit of time to fix: https://t.co/gQG8wllvHU
• 2019-03-11 @bigthink: The value of owning more books than you can read https://t.co/SKZFqqbNKe https://t.co/3AS66BNI1Y
• 2019-03-11 @Jose_A_Alonso: Why Jupyter is not my ideal notebook. ~ ClĂ©ment Walter. https://t.co/5jadIQ1W3E #Jupyter
• 2019-03-11 @UnixToolTip: Modern CMake resources https://t.co/yziMP7MGEr
• 2019-03-11 @TechSparx: Quasirandom Sequences. These are less regular than lattices, but more regular than random sampling. In this post, I describe a new quasirandom sequence; explore its remarkable properties; and give some examples of its application in gfx computing . https://t.co/9xmdbyrbmj https://t.co/Zzv86df30f
• 2019-03-11 @math_rachel: OpenML AutoML Benchmarking framework @ledell @open_ml https://t.co/HoBCmcdLiD
• 2019-03-11 @tkremenek: Making Swift more Mathable, now in review: https://t.co/NjkUhv4NPo
• 2019-03-12 @m_deff: Iodide (https://t.co/QZdDuD0lJN), literate programming in the browser by @mozilla. Like @ProjectJupyter, but without the server. Can turn (scientific) Python into WebAssembly! Combine python and the web stack for great visualizations and easy sharing of analyses. đ https://t.co/pYCDwPZO4s
• 2019-03-12 @adelaigue: Si judas avait placĂ© ses trente deniers quand il les a obtenus, combien aurait-il dâargent aujourdâhui?
• 2019-03-12 @kaz_yos: âBayesian approach to estimate the average treatment effect on the treated. builds on developments proposed by Saarela et al in the context of MSMs, using importance sampling weights to adjust for confounding and estimate a causal effectâ https://t.co/aLQEd9DNlC
• 2019-03-12 @askdrstats: Moving beyond P values: Everyday data analysis with estimation plots https://t.co/8gfgXuCoLZ time to fully retire the dynamite plot! @f2harrell @ThisisStats https://t.co/GKX69VjOea
• 2019-03-12 @peterbourgon: You like GitHub, but you wish it had more Lisp and IPFS? Perhaps a CLI-based working UX? And extraordinary, forward-thinking design? Have I got a decentralized treat for you: https://t.co/paDYd9WCm4
• 2019-03-12 @rauschma: Awesome! Chrome 73 lets you install Progressive Web Apps natively on macOS. https://t.co/c6Y4jmz72v
• 2019-03-13 @shravanvasishth: Ever wondered where the 3 standard deviations rule came from for deleting unpleasant data points? https://t.co/ebf4wtf9J3
• 2019-03-13 @hmkyale: Really like this @AnnalsofIM article by great team; proposes visualizations of study designs - simple & easy to understand - should help communicate what was done. ‘help readers understand critical temporal aspects of a longitudinal database study.â https://t.co/YSLuwzr9JS https://t.co/n6qPZkYbDi
• 2019-03-13 @tomroud: Question : comment les gens paient ils les frais dâopen access en France ? https://t.co/7eyp6yLC57
• 2019-03-13 @newsycombinator: Spotify to Apple: Time to Play Fair https://t.co/HaF43pR5BJ
• 2019-03-13 @kaz_yos: MetaâAnalysis and Mendelian randomization: A review - Bowden - - Research Synthesis Methods - Wiley Online Library https://t.co/QwdxIxwIHf
• 2019-03-13 @newsycombinator: Google has added DuckDuckGo as a search engine option for Chrome users https://t.co/CG5zrrQJdk
• 2019-03-13 @StatStas: This is an incredible data set. Raises all sorts of methodological issues, of course: you cannot really fit a survival model based on one NHIS interview and observe the survival time after that… but for stuff like lifetime conditions, it kinda works. https://t.co/yCj72Fj4as
• 2019-03-13 @simplystats: 10 things R can do that might surprise you https://t.co/onxJXz99KK
• 2019-03-13 @bayesean: Seemed like there was a lot of interest on twitter at the mention of the Stan compiler rewrite in OCaml, so I wrote up a little more about why it’s an exciting project. https://t.co/jFF4eXZ7v7
• 2019-03-13 @Teggy: “SQL, Lisp, and Haskell are the only programming languages that I’ve seen where one spends more time thinking than typing.” âPhilip Greenspun
• 2019-03-13 @kaz_yos: The powerâintegrated discriminant improvement: An accurate measure of the incremental predictive value of additional biomarkers - Hayashi - - Statistics in Medicine - Wiley Online Library https://t.co/3idf1UMFsx
• 2019-03-14 @p1cost: The Evolution of Language https://t.co/zNswvdQgN6
• 2019-03-14 @AllenDowney: @Cmrn_DP Really interesting work. And thanks for the link to this tutorial on einsum, which is excellent: https://t.co/IbVQ4oyc5w
• 2019-03-14 @petitroll: A chaque fois que je co-signe un article avec des franĂ§ais, j’ai l’impression d'Ă©crire avec Daenerys Stormborn of the House Targaryen, First of Her Name, the Unburnt, Queen of the Andals and the First Men, Khaleesi of the Great Grass Sea, Breaker of Chains, and Mother of Dragons.
• 2019-03-14 @BrodieGaslam: A great example of #FOSS at work: #rdatatable team’s contributes their radix sort to #rstats. ~50x faster ordering in this e.g. This type of improvement on such a fundamental operation is invaluable (more on this next week). Thank you @MattDowle @arun_sriniv. https://t.co/RAIu6VhzZQ
• 2019-03-14 @sigfpe: Amazing fact: map Brownian motion with an analytic function and you get (possibly time-warped) Brownian motion again. You can use it to find weird proofs of standard theorems in complex analysis as well as computing expected winding numbers etc. https://t.co/TwZnIqCXhj https://t.co/tPwYQuqiPn
• 2019-03-14 @dsquintana: Googleâs dataset search is a truly remarkable tool https://t.co/UkAkRKjMMp Thereâs more open data out there than we think, we just needed a tool like this to aggregate it all https://t.co/I0weH3hvtH
• 2019-03-15 @eddelbuettel: R^4 #20: Dependencies. Now with badges! Visualize direct and recursive dependencies of your packages and repos With special thanks to @edwindjonge for the badge creation repo! https://t.co/61x8T3wRWj #rstats https://t.co/qRqU2MfVff
• 2019-03-15 @meickenberg: Dear Reviewers, our rebuttal didn’t fit in 5000 characters. Run this: echo H4sIANqgilwAAx3Muw7CMAxG4T1P8Y90CSuXjYmRtaMBq4kIseQ4rfL2mK7nk84sHcpUyoBShaXcImaPnC2xhgNN+FTZsCUyDAdSxltyXa4Q9+e0l5V1wLQ3c4kId/aYmw8Z9LJO5f+uyyW0JGpdSyQ7fm+P0zn8AAUwpEaEAAAA | base64 –decode | gunzip
• 2019-03-15 @JohnSelstad: A (very) short intro to Constraints, which are fundamental to a variety of algorithms ranging from filtering and inverse kinematics to physical simulation https://t.co/Da7oedb46b https://t.co/2G4iF2kXIY
• 2019-03-15 @ChristophMolnar: keras + Interpretable Machine Learning Interpreting neural networks built in R with keras has become a lot easier: The iml package now accepts keras models directly. Try it out by installing the developer version: devtools::install_github(“christophM/iml”) https://t.co/feBTW2thqO
• 2019-03-15 @Kalytis_: HĂ© mais en fait, Framadrop c’est trop bien ? Y’a plus de fonctionnalitĂ©s que WeTransfer, et y’a mĂȘme aucune limite absolue de taille, si on ne laisse le fichier hĂ©bergĂ© que 24h. Utilisez Ă§a si vous voulez envoyer des fichiers/cours facilement. GG @framasoft đȘ https://t.co/0ycGvyyEyi
• 2019-03-15 @ChengSoonOng: Mathematics for machine learning, with @mpd37 and @AnalogAldo , to be published by Cambridge University Press. Final draft submitted. https://t.co/J9HPNTygmd
• 2019-03-15 @hrbrmstr: @carroll_jono @coolbutuseless @groundwalkergmb @BrodieGaslam While I detest browser WASM from a security perspective, this â https://t.co/kIhQWT0fFj â is another reason to at least be able to work with it esp given https://t.co/jXyudF7XK6
• 2019-03-15 @paulbuerkner: #brms 2.8.0 is now on CRAN featuring multinomial and Dirichlet models, data subsetting in multivariate models and parallelization of several post-processing methods. Bayesian multilevel modeling with brms has never been better! @mcmc_stan @rstats
• 2019-03-15 @TensorFlow: Text classification using TensorFlow.js: An example of detecting offensive language in browser. Read the post and get started here â https://t.co/WAdg4dVHrA https://t.co/YkHdXKChYW
• 2019-03-15 @etiennejcb: little experiment https://t.co/0dfb2D7fud
• 2019-03-15 @Md_Harris: The trending of 90âs hacker culture gives my license to relive my days as a warez group founder/courier ca. 1994-97â. The later date is when I rejoined the scene as a budding anthropologist doing my participant observation. Those were the days! #CultoftheDeadCow #warez #2800 https://t.co/VjJlSZ349H
• 2019-03-16 @mjskay: Really excited to announce our (Pierre Dragicevic, Yvonne Jansen, @abhsarma, me, @zirtech) paper on Explorable Multiverse Analyses. We make multiverse analyses (a la Steegen, @StatModeling, et al) interactive through explorable explanations (a la @worrydream). THREAD —> 1/7
• 2019-03-16 @jabawack: 15 data fallacies and biases https://t.co/fpcjoYcZq6
• 2019-03-16 @Atabey_Kaygun: I started listening to the songs I used to listen back in 90’s, and four hours later, I regressed further in the 80’s and ended up listening to motorhead and slayer. Lucky that nobody sees me headbanging in the office alone :)
• 2019-03-16 @newsycombinator: How I Learned to Love Algebraic Geometry https://t.co/uAwfPzpknp
• 2019-03-16 @newsycombinator: Reasons Not to Use Apple https://t.co/gPdPy1pQGK
• 2019-03-16 @zerology: Cool. There is a Julia package that starts to reimplement the Rethinking Statistics package by @rlmcelreath (his is in R) https://t.co/vpyuJp76W1 #StochasticModels #JuliaLang
• 2019-03-16 @rxwei: New documentation about how differentiable programming works in Swift: âą Differentiable Functions and Differentiation APIs: https://t.co/GFhJjaaqb8 âą Differentiable Types: https://t.co/a8DIjFPK6V With language integration, autodiff is just a compiler implementation detail.
• 2019-03-17 @greghendershott: About a week in. Not missing the MacBook. Have keys, will travel. https://t.co/YXTNyeV0Pp
• 2019-03-17 @kumarshantanu: TIL about two awesome #Clojure learning resources, https://t.co/1Hdm6YBCIw and https://t.co/8JScSZEmDF
• 2019-03-17 @biorxiv_bioinfo: pysradb: A Python package to query next-generation sequencing metadata and data from NCBI Sequence Read Archive https://t.co/q9ihevFva7 #biorxiv_bioinfo
• 2019-03-17 @jkatz05: This is exciting: the ability to query JSON data using the SQL/JSON path syntax was committed to the @PostgreSQL codebase! Another great feature for @PostgreSQL 12: https://t.co/1PWaSP2Ryz Congrats to everyone who worked on it!
• 2019-03-17 @newsycombinator: Thinkpad X210 https://t.co/Ua72ZcLpun
• 2019-03-17 @AndyBugGuy: Genomes of 250 skipper butterflies reveal extensive convergence of wing patterns- resulting in a major overhaul of skipper taxonomy- with 6 new tribes and 9 new genera: https://t.co/53JvzlcmT7 https://t.co/sodU3eayrL
• 2019-03-17 @newsycombinator: A distributed rate counting system with Lua and Redis https://t.co/8dz3aZJw1r
• 2019-03-17 @_wilfredh: Woah, today I learned that Rust has a doubly linked list in its standard library! It’s a common example of something you can’t do in purely safe Rust. Nonetheless, the core Rust devs have done the hard work already! https://t.co/T5ieCzr9E1
• 2019-03-17 @bob_kitt: I’ve been doing it wrong all my life…. https://t.co/gB5J8N76oN
• 2019-03-18 @wviechtb: Of potential interest esp. for people in ecology doing meta-analyses: The metafor package now also allows for spatial correlation structures. An illustrative application: https://t.co/A1WiXrzcOY Note that you need the ‘devel’ version of metafor for this: https://t.co/FNP1FgdkR4
• 2019-03-18 @PierreAblin: Illustration of the Lasso and its path in 2D: for t small enough, the solution is sparse! https://t.co/0JV23e32DF
• 2019-03-18 @hrbrmstr: For any non-“cyber-entrenched” followers who would like some insight into the jargon I use (but likely should not as often as I do) we’re doing a game-show take on “Demystifying Cybersecurity Jargon” tomorrow https://t.co/0Qg6v4CeOb (cc: @kara_woo; finally found the link).
• 2019-03-18 @gigasquid: New Post by Arthur Caillau - MXNet Made Simple - Image Manipulation https://t.co/j5dS29iEA9
• 2019-03-18 @rafalab: A free PDF version of Introduction to Data Science: Data Analysis and Prediction Algorithms with R is now available on https://t.co/vwMW921269 Thanks to all the readers that through GitHub pull requests and issues improved the first gitbook version, specially to @biochemnerd!
• 2019-03-18 @fulhack: Crazy but true: Annoy (everyoneâs favorite approximate nearest neighbor library) just reached 5k stars on Github: https://t.co/7sRLG054tT â­ïžâ­ïžâ­ïžâ­ïž
• 2019-03-18 @NicholasStrayer: Something I embarrassingly never thought about visually until reading @yudapearl ‘s #bookofwhy. People tend to overestimate correlations because people don’t think about regressions using least-squares, they think about them as orthogonal regression. #DataScience #dataviz https://t.co/791MR33CtJ
• 2019-03-18 @newsycombinator: Simple script to analyse your PostgreSQL database config, give tuning advice https://t.co/1ONTIpGdMA
• 2019-03-18 @newsycombinator: Elsevier Left Usersâ Passwords Exposed Online https://t.co/4Lm7J713wa
• 2019-03-18 @carlcarrie: Essential Statistics with Python and R #rstats ebook https://t.co/JPsi1TFC6f https://t.co/Wy5MEHA37Q
• 2019-03-19 @pythontrending: dragonfly - An open source python library for scalable Bayesian optimisation. https://t.co/NKhlQUkk5l
• 2019-03-19 @hrbrmstr: Assumptions Matter More Than Dependencies https://t.co/ZqM4hdIuW7 #rstats
• 2019-03-19 @wviechtb: I am sure the world has been waiting for yet another paper comparing different heterogeneity estimators in meta-analysis, so knock yourselves out: https://t.co/PuM6zg6qaM TLDR: Use REML and use the HartungâKnappâSidikâJonkman method.
• 2019-03-19 @f2harrell: Wonder how a #machinelearning algorithm can cross-validate (CV) when the sample size can’t support the number of candidate predictive features? It only seems to, with changing features selected for each CV-cycle. Reseachers wanting stability of features selected: watch out.
• 2019-03-19 @lambda_calculus: Glad to see Elsevier embrace an open-access model. https://t.co/NHs17QiY10
• 2019-03-19 @f2harrell: What may be 2 of the most important predictors of failure to translate animal research to humans? Failure to demonstrate dose-response relationships in animals, and poor #Statistics/experimental design: https://t.co/aLn36yehkX https://t.co/0Qi2GiFFtS @venkmurthy @ProfDFrancis
• 2019-03-19 @strnr: CGAT-core: a python framework for building scalable, reproducible computational biology workflows https://t.co/Z1D98HvWiw https://t.co/Fz3pNLcJuf
• 2019-03-19 @deaneckles: This book contains some approaches that will be quite unfamiliar to most social scientists, even those who sometimes think with causal DAGs. (Not all are necessarily as suited to the social sciences though.) https://t.co/rpEMHFc9gO https://t.co/LODD1OOprT
• 2019-03-19 @strnr: Sequencing-based methods and resources to study antimicrobial resistance https://t.co/VHOalRIWXc https://t.co/zbANFfSUhW
• 2019-03-19 @chowthedog: New post cruising around @Minecraft â using R and the A* algorithm. Notice it quickly gets close to the end, then spends a lot of time optimizingâ. Inspired by @revodavid’s @rstatsnyc talk last year, using miner lib to run a maze! #rstats #minecraft https://t.co/TZyDvUc5Vb https://t.co/iqbVNeVbId
• 2019-03-19 @JLRohmann: A nice summary of the last @bemcolloquium talk on regression shrinkage by one of our Health Data Sciences PhD students! The slides are linked on the blog, too. Thanks again, @maartenvsmeden https://t.co/IbxngK407b
• 2019-03-19 @Richard_D_Riley: Methodology grant submissions are a lottery: Once I submitted a proposal to develop sample size calculations to improve upon the 10 EPV rule … it didnât even make the panel meeting because main reviewer said âEPV is never used for sample size calculations in risk predictionâ đČ
• 2019-03-19 @eddelbuettel: The new ‘options(“conflicts.policy"=….)’ in the upcoming R 3.6.0 will be very useful – and this write-up by @LukeTierney4 is something many of us may come back to a number if times (as the section in help(library) in R-devel is rather brief). Excellent stuff. #rstats https://t.co/rmgulpP1qC
• 2019-03-20 @StatModeling: Maybe itâs time to let the old ways die; or We broke R-hat so now we have to fix it. https://t.co/ckgycU0VZD
• 2019-03-20 @kaz_yos: R and labelled data: Using quasiquotation to add variable and value labels #rstats | R-bloggers https://t.co/3lHAOQUSWL
• 2019-03-20 @GaelVaroquaux: New paper: Population shrinkage of covariance (PoSCE) for better individual brain functional-connectivity estimation https://t.co/H7Yh1PVXCn Better covariance estimation using a reference population, for James-Stein shrinkage on the covariance Riemannian manifold. https://t.co/6WvR1MaB9E
• 2019-03-20 @gappy3000: PSA: The second edition of the Machine Learning textbook by Mori et al. is free to download. The 1st ed. was already excellent. https://t.co/4HPYQ6FQiH
• 2019-03-20 @CodeWisdom: âSometimes, the elegant implementation is just a function. Not a method. Not a class. Not a framework. Just a function.â - John Carmack
• 2019-03-20 @paulbuerkner: In our (@avehtari, @StatModeling, @dan_p_simpson, Bob, and my) new paper, we fix the Rhat convergence diagnostic and provide additional diagnostics that we argue should become the new standard to assess convergence of MCMC algorithms. Preprint available at https://t.co/bhxSjKN3eO
• 2019-03-20 @rafalab: All materials for the 4 hour workshop Data Science for Statisticians are available on GitHub. We covered tidyverse, dataviz, wrangling and machine learning. Includes 6 lectures in R markdown and html 5 labs solutions in R markdown and markdown https://t.co/Zz027ZeGLz
• 2019-03-20 @plotlygraphs: Today we are launching Plotly Express, a new high-level Python visualization library! Check out our announcement at https://t.co/mb6UNT2NeF or just watch this show reel of charts each made in a single line of Python: https://t.co/DXrKwXrrTm
• 2019-03-20 @magnars: This is long overdue. Sorry to anyone negatively affected. #emacs https://t.co/z5NXbdiRPl
• 2019-03-20 @ben_j_lindsay: One of the most underrated #Pandas functions in #Python is .query(). I use it all the time. data = data.query(‘age==42’) looks so much nicer than: data = data[data[‘age’] == 42] And it allows chaining like: data = data.query(‘age >18’).query(‘age < 32’)
• 2019-03-20 @mattdesl: The recent MySpace & Tumblr revelations is a reminder to artists… - Don’t put all your eggs in one basket, instead share on many platforms - Have cloud & physical backups of all your art at raw/source size - Self-host where possible (Below comment is from HN) https://t.co/9dg5LImIey
• 2019-03-20 @strnr: A Systematic Evaluation of Single Cell RNA-Seq Analysis Pipelines https://t.co/sL4PAUi9wb https://t.co/AwAw26W8jR
• 2019-03-20 @Jose_A_Alonso: Map/Reduce operations for scientific computing in Julia. ~ X. Vasseur. https://t.co/hwVPQz1lVr #JuliaLang
• 2019-03-20 @drizopoulos: In two weeks I’ll be teaching again my Repeated Measurements course (https://t.co/NNw2emPTjE) with improved slides, featuring a new version of the shiny app that contains all code, output & extra material: https://t.co/8wT06nkWnR - next year it will also be fully available online
• 2019-03-20 @rasbt: Another optimization algorithm visualization tool. This is a really great one, though. Coincidentally, I am going to cover optim algos in the next lecture and this will be super neat for the conceptual illustrations https://t.co/tggJoZPjwd https://t.co/WstjHKB11B
• 2019-03-21 @newsycombinator: How I’m able to take notes in mathematics lectures using LaTeX and Vim https://t.co/gECwQlxKIs
• 2019-03-21 @kaz_yos: âKnowing what not to do with p-values is indeed necessary, but it does not suffice. It is as though statisticians were asking users of statistics to tear out the beams and struts holding up the edifice of modern scientific research without offering solidâ https://t.co/JAUkSlxwXn
• 2019-03-21 @strnr: Elements and Principles of Data Analysis (PDF) https://t.co/L8oIfIRuCH from @stephaniehicks @rdpeng https://t.co/L0SlgiCOeZ
• 2019-03-21 @MuinJKhoury: Check out the @NEJM highly informative glossary on genetics and genomics. https://t.co/GS1IfMla2e https://t.co/zee7BgMvhl
• 2019-03-21 @AllenDowney: “We must learn to embrace uncertainty.” https://t.co/jrIIizQDx6
• 2019-03-21 @cocoweixu: I condensed my semester-long CS course “Social Media & Text Analysis” into a 45-minute tutorial for high school kids. Going to debut it at High School I/O @hackohio this Saturday! đ https://t.co/MqHbZ9bIoj #NLProc #Python @OSUOutreach https://t.co/6jbJp3bRgI
• 2019-03-21 @TeXtip: Mirroring text in LaTeX https://t.co/Q2dX6qnYAY by @thurnherr
• 2019-03-21 @CompSciFact: Books for CS grad students https://t.co/KikhCzWTz5
• 2019-03-21 @Viral_B_Shah: The #julialang debugger is here. The command line UI and Juno integration is perfect. Enough said. https://t.co/p90vI0pXfO https://t.co/8Lx2MEdjTA
• 2019-03-21 @nitot: 20.000 employĂ©s de Facebook ont eu accĂšs Ă  vos mots de passe non chiffrĂ©s. Des centaines de millions dâutilisateurs sont concernĂ©s. AprĂšs la data faĂ§on far west, la sĂ©cu faĂ§on YOLO. Dans le pire, Facebook nâest jamais dĂ©cevant ! https://t.co/nrGyK0OODe
• 2019-03-21 @the_repl: Listen to @carinmeier talk about Clojure MXNet and calling Scala from Clojure (!): https://t.co/8pUcYdSjQL
• 2019-03-21 @tmcw: never, ever use the @disqus embed snippet. it hits google, facebook, and viglink (an ad network) and binds ‘click’ events on every link on the surrounding page.
• 2019-03-21 @dotMudge: Just noticed that OâReilly has my chapter from Beautiful Security up on their site Infosec examples of: Learned Helplessness Confirmation Traps Analystâs Dilemma Functional Fixation https://t.co/3nR4WQnr6z
• 2019-03-21 @marco_cova: The “Hack for hire” paper on the underground market of Gmail hacking services is pretty neat: https://t.co/RtLBTo4dGw
• 2019-03-22 @janezhgw: I visited #DesignTO in January by @DesignTOFest and really enjoyed the experience. But I had difficulties planning out my itinerary. I saw a #dataviz opportunity here and decided to make one as a project. Check it out here https://t.co/otLfmXrzSL. Process belowđ (1/9) https://t.co/IiI7VCVF3h
• 2019-03-22 @newsycombinator: Functional programming explained for the pragmatic programmer https://t.co/Ja8Hd7g9NY
• 2019-03-22 @gigasquid: Kicking off a monthly update for the Clojure MXNet project https://t.co/0zECwNNDgO
• 2019-03-23 @TechSparx: “Algorithms” by Sedgewick & Wayne. One of the most readable and respected books on this topic! Over 900 pages of amazing content all for free! PDF: https://t.co/U7s1UbkkiF Code: https://t.co/ZTuUPAbDss Solutions: https://t.co/4kZsMD7tjz Course: https://t.co/QaUk4w2PSW https://t.co/MsFxQl91Az
• 2019-03-24 @ken_bellows: I wrote a thing about semantic HTML on @ThePracticalDev: “Stop using so many divs! An intro to semantic HTML” https://t.co/h1sv6ZvZ4L #DEVcommunity #html #webdev
• 2019-03-24 @Jose_A_Alonso: Permutations by sorting. ~ Donnacha OisĂ­n Kidney (@oisdk). https://t.co/kRT32U4SwC #Haskell #FunctionalProgramming
• 2019-03-24 @Atabey_Kaygun: loom: a clojure graph library https://t.co/2p11sKXXQF