in reply to Re^4: Data visualisation.
in thread Data visualisation.

LanX:

Yeah, my original triangle check was faulty. I only checked one case, rather than all three. Using your data, I had only a 1/3 chance of detecting that particular triangle as failing the inequality:

47 < 27 + 77 : Passes

27 < 47 + 77 : Passes

77 !< 27 + 47 : Fails...

Since I only checked one case, I'd've missed a bad triangle. But my code didn't check all that many triangles, either, so it missed anyway. So I added an exhaustive check just for the helluvit. I've updated the node above accordingly.

...roboticus

When your only tool is a hammer, all problems look like your thumb.

Replies are listed 'Best First'.
Re^6: Data visualisation.
by BrowserUk (Patriarch) on Jan 05, 2014 at 03:26 UTC
    So I added an exhaustive check just for the helluvit.

    What does your code now do with the "bad triangles"?

    With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
    Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
    "Science is about questioning the status quo. Questioning authority".
    In the absence of evidence, opinion is indistinguishable from prejudice.
[reply]

      BrowserUk:

      Nothing, really. It just mentions them.

      If the dataset were a set of straight-line distances, then it could warn you that the data has an error in it. My code doesn't really care, though. If the triangle is "bad", I'll still calculate the intersection point where the lines *would* meet, were they long enough.

      ...roboticus

      When your only tool is a hammer, all problems look like your thumb.

[reply]
        Nothing, really. It just mentions them.... I'll still calculate the intersection point where the lines *would* meet, were they long enough.

        Oh, okay.

        Though I have to say that it sounds like you are now performing 3 "triangle inequality" checks for every possible triangle (for my 17 node dataset that is 3*15! or nearly 4 Trillion additions and 4 Trillion comparisons) for no real purpose. Ie pure make-work....

        For Dirk80's 24 node dataset, that rises to 1.8 trillion trillion redundant calculations.

        With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
        Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
        "Science is about questioning the status quo. Questioning authority".
        In the absence of evidence, opinion is indistinguishable from prejudice.
[reply]

In Section Seekers of Perl Wisdom