Predicting future medical diagnoses with RNNs using Fast AI API from scratch
— Rachel Thomas (@math_rachel) April 22, 2019
(full pytorch implementation of Doctor AI paper using Electronic Health Records) by @SparklePuleri https://t.co/a2wmQEQ7Cg
Predicting future medical diagnoses with RNNs using Fast AI API from scratch
— Rachel Thomas (@math_rachel) April 22, 2019
(full pytorch implementation of Doctor AI paper using Electronic Health Records) by @SparklePuleri https://t.co/a2wmQEQ7Cg
It took several of us more than 24 hours, but @AxiosVisuals indexed the entire #MuellerReport and put together what I hope will be a good and useful interactive for researchers, journalists, and the truly curious. Dive in: https://t.co/mCsyqf9yNO via @axios pic.twitter.com/nx8xW8e2ub
— Harry Stevens (@Harry_Stevens) April 19, 2019
🕵️♂️ Text extraction *and* viz using @Emil_Hvitfeldt's {ggpage}…
— Mara Averick (@dataandme) April 19, 2019
📑 "The Redacted, Text-Extracted Mueller Report" by @grrrckhttps://t.co/UsoT1RPKUA #rstats #dataviz #MuellerReport pic.twitter.com/6fdoa9zVfC
I haven't looked at the report, but I'm fairly confident the people doing the redacting would have been careful to do it in such a way that the redacted info furthermore is not predictable.
— Emily M. Bender (@emilymbender) April 19, 2019
(And, ahem, that the black-out can't just be deleted...) /7
I've seen several different #NLProc folks suggesting today that it would fun/interesting/worthwhile to use BERT or GPT-2 to fill in the redacted bits of the Mueller report. A short thread on why this is a terrible idea /1
— Emily M. Bender (@emilymbender) April 19, 2019
The Mueller Report, frequency of key phrases and names, by volume with highlights. #rstats #tidytext pic.twitter.com/6weiAcgU75
— Ryan Timpe 📊 (@ryantimpe) April 18, 2019
One of my students discovered this using word2vec: good - bad = excellent, bad - good = ['maniacal_killer', 'insuring_repackaged_subprime_mortages'] pic.twitter.com/y0rIHC7hp5
— Andreas Mueller (@amuellerml) April 17, 2019
The app and framework are 100% open-source and based on Markdown + custom elements. I built it for my content, but if you want to use it to publish your own DIY online course, it should be easy to adapt. In theory, it should even work for other languages. https://t.co/br8qeJhWjY
— Ines Montani 〰️ (@_inesmontani) April 17, 2019
Like many of you, I'm incredibly disappointed by DataCamp. I wanted to make a free version of my spaCy course so you don't have to sign up for their service – and ended up building my own interactive app. Powered by the awesome @mybinderteam & @gatsbyjs 💖 https://t.co/2QOuDPoZEX
— Ines Montani 〰️ (@_inesmontani) April 17, 2019
Ding ding ding... what a battle for WMT19 en-de: https://t.co/REboKAOPU2
— Marian NMT (@marian_nmt) April 16, 2019
Including last twists and friendly fire.
PolyAI is releasing a collection of conversational datasets consisting of hundreds of millions of examples https://t.co/mchj2I4VAD
— Peter Skomoroch (@peteskomoroch) April 16, 2019
Gotta side with @yoavgo on this one!
— Stanford NLP Group (@stanfordnlp) April 12, 2019
The genealogy was [early computational semantics roots (FRACAS, Condoravdi et al., Monz & de Rijke) ➔ RTE (Recognizing Textual Entailment, Ido Dagan) ∪ Natural Logic (van Benthem)] ➔ NLI https://t.co/hkrMQw7bHu