Nick Heitzman ๐๐ retweeted
The โData Horror Stories Songโ, inspired by a tweet by @rogierK and commissioned by @LisaDeBruine
Any of these ever happened to you?
#rstats #Statistics #DataScience
Nick Heitzman ๐๐ retweeted
AIgos/AI are UNIQUELY BAD at predicting rare events. They model the world via patterns in data, and by def sudden ruptures aren't marked by such patterns.
While this won't work, it will expand surveillance & likely conflate street protest & insurrection.
washingtonpost.com/technologโฆ
Nick Heitzman ๐๐ retweeted
Am very, very ready for the deep learning hype curve to start dying down. ๐
*So much* business value can be claimed with basic statistics, exploratory data analysis, and traditional machine learningโand those techniques require just a fraction of the time and complexity of DL. โ at Austin, TX
Nick Heitzman ๐๐ retweeted
Holy Shit: CDC now saying that fully vaccinated people no longer need to switch doors in the Monty Hall Problem!
Nick Heitzman ๐๐ retweeted
Top terms searched on Google in 2020. #dataviz
Source: reddit.com/r/dataisbeautifulโฆ
Nick Heitzman ๐๐ retweeted
Learning how and which business problems to solve with machine learning takes practice. I ask myself Qs like these:
- How might I use ML to solve this? What would the deliverable be? Who would use it, when, and how?
And just as important:
- How could I solve this without ML?
Nick Heitzman ๐๐ retweeted
๐คโก scikit-learn tip #37:
New in version 0.23: Create interactive diagrams of Pipelines (and other estimators) in Jupyter!
Click on any element to see more details. You can even export the diagram to a file!
See example ๐
#sklearntips #MachineLearning #Python #DataScience
Nick Heitzman ๐๐ retweeted
Did I spend entirely too long this weekend on this #tidytuesday on #thatswhatshesaid jokes from #TheOffice? Yes, almost certainly, but it's a pandemic- just let me be myself! #rstats
Nick Heitzman ๐๐ retweeted
Tonight it's a Bloomberg chart for the win. Good luck deciphering the starting axis and the gridline increments on this one.
Nick Heitzman ๐๐ retweeted
Some work-in-progress illustrations to gently introduce some concepts for 2-sample t-tests & how to think about p-values. Meant as teaching aids, not as a comprehensive standalone lesson. Caveats & assumptions abound. ๐งต
Nick Heitzman ๐๐ retweeted
OK boomR!!
After discussing base vs tidyverse in the context of generational conflicts with @RStudioJoe, I will from here on out refer to base R as "boom-R". #rstudioconf #rstats @SHaymondSays @srmaster @ibddoctor @DrDanHolmes
Nick Heitzman ๐๐ retweeted
1st blog post: takes a month, attempts to be perfect, read by no-one
.
.
.
100th blog post: takes 2 hours, realize done > perfect, read by lots
You get better by writing a lot, not by trying to be perfect.
Let go and press publish.
Then start your next one.
Nick Heitzman ๐๐ retweeted
[Nickelback voice] Look at this infograph!
This quoted post is unavailable.
Nick Heitzman ๐๐ retweeted
|๏ฟฃ๏ฟฃ๏ฟฃ๏ฟฃ๏ฟฃ๏ฟฃ๏ฟฃ๏ฟฃ|
You've got what it
takes to be a data
scientist if you
want to beโจ
|๏ผฟ๏ผฟ๏ผฟ๏ผฟ๏ผฟ๏ผฟ๏ผฟ๏ผฟ|
ย ย ย ย ย ย ย ย \ (โขโกโข) /
ย ย ย ย ย ย ย ย ย ย \ย ย ย ย ย /
ย ย ย ย ย ย ย ย ย ย ย ---
ย ย ย ย ย ย ย ย ย ย ย |ย ย |
Nick Heitzman ๐๐ retweeted
just learned with horror that deleting any of the "family" emojis in google slides does not remove the emoji, but rather kills off each individual family member one by one, starting with the children
Nick Heitzman ๐๐ retweeted
When something in a piece of software makes you say "what were the developers thinking?"
...they were probably thinking of constraints you're not considering.
Nick Heitzman ๐๐ retweeted
I don't see any good reason for the modifier "domestic" in "domestic terrorism".
Is it lesser than international terrorism, somehow? Is "domestic" meant to soften and make it sound less scary, like "racially charged" in lieu of racist?
Just call it terrorism if it's terrorism.
Nick Heitzman ๐๐ retweeted
The most important thing you can do as an industry data scientist is have regular code reviews/knowledge shares w/ your data warehouse team. They know exactly what/where all the data is and all the little details that others don't. Do this early and prevent a bunch of headaches.