I had a bit of a breakthrough in terms of my thinking of data science thanks to all the interesting discussions at #rstudioconf2020 --
-
-
Like the fun fact that every time you incorporate new data fields into a model/system, all the historical data that fed into that old system has an immediate decrease in utility.
-
omg tell me about it
Kraj razgovora
Novi razgovor -
-
-
Could this go against blameless post mortem if something ends up wrong with the dataset?
-
That's a good point. Ideally in a blameless culture you have people not blaming individuals when something goes wrong, so you don't have to keep things anonymous to enforce that outcome. But reality is always much more messy and imperfect
Kraj razgovora
Novi razgovor -
-
-
I’m so alien to the stitchfix world - is there a simple example that illustrates how data gets created (or was it the dataset that was created after say some processing - joining tables for e.g.)
-
the data I'm referencing is outfits, so like -- what items of clothing in our inventory go together? This is all hand-created by fashion experts, vs. more behavioral click data that is usually the fodder of data science. We have that too, of course.
- Još 1 odgovor
Novi razgovor -
-
-
I always try to name check the people whose stuff I use <3
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
I love this!!! It makes data real and reminds you that it matters where it comes from and how you use it
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
small but such a good idea!
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
TELL ME ABOUT IT! People seem to assume that anything from the SQL bank is good quality. I literally had a similar conversation that we just couldn't do anything with such lousy dataset. That's why I had to create my own dataset. I want clean water and better quality data!
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.