Live data from Hacker News

Viewing profile — thetrainfold

thetrainfold

HN member
Joined
Sat, Jun 08, 2019, 12:57 PM UTC
HN karma
20
Public activity
10 items

About thetrainfold

No profile information was provided.

Recent public activity

  1. comment
    Comment #20556565

    Question about the algorithm: Is the algorithm run from scratch each time a seam is removed? I.e. energy function computed again on the resized image, all seams recomputed, then on…

  2. comment
    Comment #20552353

    Could it help to force long names, preventing simple, 'elegant' names?

  3. comment
    Comment #20532976

    Just imagine the work that haired people go through.. (sorry)

  4. comment
    Comment #20515946

    I was referring to this specific case. In this housing example, extra infrastructure is being built for the purpose of segregating people. In your examples the segregation emerges …

  5. comment
    Comment #20512910

    (I work in marketing at scale) Thank you. (I don't really)

  6. comment
    Comment #20484939

    Remind them by segregating them from wealthier people?

  7. comment
    Comment #20471286

    Side note, but sharing salaries with colleagues is actually a good idea, due to the unfairness that can arise from salaries being kept secret. Example article: https://www.nytimes.…

  8. comment
    Comment #20381323

    """ A senior dev whom I respect for his technical prowess called me this morning. He called to fret over the meagre 12% raise he got... I gave him the best I could — by listening t…

  9. comment
    Comment #20293414

    "he noted that some combination locks allow for wiggle-room and if this one had a three-digit leeway, Mr Rosenthal put the chances at 1 in 8,000, "which is still a small chance"." …

  10. comment
    Comment #20132172

    It would be unusual not to check the performance of the model at predicting the target variable, which would validate whether or not the derived feature is useful.