.@JennyBryan delivering peak #joygret in this tidyeval talk. I’ve never known to pass the dots until now, and frankly I am thrilled and furious #rstudiconf pic.twitter.com/LRPsqmT8zm
— Brooke Watson (@brookLYNevery1) January 18, 2019
.@JennyBryan delivering peak #joygret in this tidyeval talk. I’ve never known to pass the dots until now, and frankly I am thrilled and furious #rstudiconf pic.twitter.com/LRPsqmT8zm
— Brooke Watson (@brookLYNevery1) January 18, 2019
.@beeonaposy vaccinating the audience against imposter syndrome with a gif game on 💯 #Rstudioconf pic.twitter.com/skhaBa7zUc
— Brooke Watson (@brookLYNevery1) January 18, 2019
Ideally you shouldn't have more than 30 arXiv papers on your desk. Throw away the papers that don't spark joy
— François Chollet (@fchollet) January 18, 2019
“Foster an environment where people can be excellent but have the psychological safety so they don’t always have to be” - @AngeBassa #rstudioconf pic.twitter.com/eVNXh92VbX
— Emily Robinson (@robinson_es) January 18, 2019
it depends: a dialog about dependencies
— Jim Hester (@jimhester_) January 18, 2019
Slides from my #rstudioconf talk at https://t.co/7uUF1H0drK
By getting the data team involved in data collection, you can ensure they get the data they need@hspter's team at Stitch Fix suggested a "Tinder for Clothes" feature, which was perfect for collecting training data and discovering latent structure 👚👕👖📊 #rstudioconf pic.twitter.com/kQabmwnqhh
— David Robinson (@drob) January 18, 2019
Sometimes the solution is not trying matrix factorization again (and again) on sparse data, it’s to create a system to get more data! Stitch Fix created “style shuffle” so people could rate tons of clothes instead of just the 5 they got in each shipment - @hspter #rstudioconf pic.twitter.com/ZW8buZotc3
— Emily Robinson (@robinson_es) January 18, 2019
The main takeaway from @Felienne's talk is "You don't become an expert by doing expert things" #rstudioconf pic.twitter.com/2poojflEjO
— Amelia McNamara (@AmeliaMN) January 18, 2019
A chatbot that learns by chatting.
— Yann LeCun (@ylecun) January 17, 2019
Brought to you by FAIR.
"Learning from Dialogue after Deployment: Feed Yourself, Chatbot!", by Braden Hancock, Antoine... https://t.co/rbnsUWDdJf
#DataScience #10yearschallenge
— Randy Olson (@randal_olson) January 17, 2019
from https://t.co/S5NKPJrHrG pic.twitter.com/tYSD9U6uUH
.@AmeliaMN on the necessity and dangers of working with categorical variables (and the slightly mitigated dangers of doing that in the tidyverse with forcats) #RStudioConf pic.twitter.com/c6LZSYFGDm
— Brooke Watson (@brookLYNevery1) January 17, 2019
.@skyetetra & @heatherklus built a machine learning model in R and deployed it in production, without Python, to 70 million T mobile users (!!!) #RStudioConf pic.twitter.com/0yR7xzrBSG
— Brooke Watson (@brookLYNevery1) January 17, 2019