Angela Bassa

@AngeBassa

I make sure we math on data right. Head of Data Science and Machine Learning . Mathematician from . Opinions mine.

Boston, MA
Vrijeme pridruživanja: studeni 2012.

Tweetovi

Blokirali ste korisnika/cu @AngeBassa

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @AngeBassa

  1. Prikvačeni tweet
    28. velj 2019.

    We’re all winging it. I promise you, even the experts. Especially the experts; we’re dealing with cutting edge problems and there’s no playbook. We’re all figuring it out as we go. Be kind to yourself. Try to figure out what it is you should do, not what the right answer is.

    Prikaži ovu nit
    Poništi
  2. proslijedio/la je Tweet
    prije 23 sata

    writing "The Census Bureau was able to correctly re-identify the records of more than 52 million Americans" which sounds a lot scarier. So its almost besides the point that the 17% number leaves a lot out: see John's tweetorial from awhile back: 2/2

    Prikaži ovu nit
    Poništi
  3. 5. velj

    “How hard could it be?!” It turns out this stuff is hard to do right...

    Poništi
  4. 5. velj

    I referenced the West Wing Bruno Gianelli kelp story yesterday and I’m not even sorry* *I might be a little bit sorry

    Poništi
  5. 4. velj

    “I don’t believe any of these biases were intentionally put in these models, I just think that no one bothered to check if the biases were there. Why did no one bother to check? Because these issues do not (directly) affect the group creating the technology.”

    Poništi
  6. proslijedio/la je Tweet
    2. velj

    It's the year 2020, and we're still finding buffer overflows in programs written in C. Fortunately, those programs are only esoteric, rarely installed programs like... <checks notes> sudo.

    Poništi
  7. proslijedio/la je Tweet
    3. velj

    See, people think Anscombe's quartet is all artificial extreme cases and no one would actually make such a stupid mistake with regression, and then you get figures like Fig 8C

    Prikaži ovu nit
    Poništi
  8. proslijedio/la je Tweet
    30. sij

    I'm baffled by how lazy and dumb this fake data strategy is: 1. Make up some numbers for controls 2. Copy them to a new column for the treatments, and add +2 to each one Not to mention the ethical lapses involved.

    Prikaži ovu nit
    Poništi
  9. 31. sij

    Best way of getting really great feedback on Twitter continues to be to pose an ill formed question 😂 Jokes aside... thanks for some interesting reframing, and food for thought.

    Prikaži ovu nit
    Poništi
  10. 30. sij

    I’m curious about your thoughts: is there ever good reason to intentionally introduce noise to data that doesn’t eventually also end up blowing up in worse ways?

    Prikaži ovu nit
    Poništi
  11. proslijedio/la je Tweet
    30. sij
    Poništi
  12. 30. sij

    An ethics governing body for the software engineering profession is long overdue.

    Poništi
  13. proslijedio/la je Tweet
    27. sij

    The daily standup isn’t a status meeting—it’s a *coordination* meeting. If you’re not planning how to work together, but only reporting status, you can cut it. Or start collaborating more.

    Poništi
  14. 28. sij
    Prikaži ovu nit
    Poništi
  15. 28. sij

    Energy data thread, dear nerds! ⤵️ Makes me almost miss the days of predicting what the utility grid was going to do back at EnerNOC... 🤓⚡️

    Prikaži ovu nit
    Poništi
  16. proslijedio/la je Tweet
    26. sij

    Before engaging in an activity that you are worried you might simply enjoy, be sure to check that it can be worked into a career-centric self-improvement narrative.

    Poništi
  17. 25. sij

    Lots of lessons, but all in all not bad 😁

    Prikaži ovu nit
    Poništi
  18. 25. sij

    Letter fold: 3⁶ = 729 Book fold: 4⁵ = 1,024 But I’m still just learning how to fold, and it’s so hard!

    Prikaži ovu nit
    Poništi
  19. 25. sij

    So today’s “mille-feuilles” will actually be 271 sheets shy, but who’s counting 🙃

    Prikaži ovu nit
    Poništi
  20. 24. sij

    I was feeling queasy all day and I thought it was anxiety about this upcoming important meeting but then I let out a big burp and I guess it was just indigestion 🤷‍♀️

    Poništi
  21. 23. sij

    I finally allowed my husband to name a robot and he called it Robochaz.

    Poništi

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.

    Možda bi vam se svidjelo i ovo:

    ·