math So to prove F cannot be shattered, we must prove that there exists a subset of F, S, which cannot be intersected without intersecting a point in F that is not in In other words, we must prove that there exists a point q in F that intersects ALL balls that intersect that's where this earlier lemma comes it makes our job the one about 3 points on a line in since with that lemma, all we need to prove that there exists a point q in F that is in the smallest ball that intersects all points in ah, when I said "intersects S" earlier I meant "intersects all points in i am following but it feels like we're still very close to the deffinition of shattering, we havent actually touched on anything that has to do with with closed balls and well that's the usefulness of that I was looking for a way to prove something about ALL balls that intersect all points in + KD19 do u see where this will go or are you still just what the definition of a hi, are there any websites that have good I am finding a way to traverse a complete path if given a matrix of `m0 yes there I'm looking to see if anyone has a classpad, classpad manager, or knows how to get os3 for Say I have a matrix of numebrs 5x5 all 0's, and we have a T shape in I would like to count how many 1's are there in that T so basically, count all the adjacent (up|down|left|right) Flood fill 7 adjacent ones, I am trying to do this for my research in the quickest way possible (for some other `m0: Look up flood fill Olathe: I have done an algorithm that But it seems it just to accept 67x67 matrixes.

#apache #archlinux #asterisk #centos #debian #gentoo #haskell #kde #kubuntu #lisp #math #mysql #perl #python #ruby-lang #rubyonrails #suse #ubuntu #vim #wikipedia 0 1 2 3 4 5 6 7 8 9 10 11 12
Top Prev 5203 5204 5205 5206 5207 5208 5209 5210 5211 5212 5213 5214 5215 5216 5217 5218 5219 5220 5221 Next

#math

<KD19> So to prove F cannot be shattered, we must prove that there exists a subset of F, S, which cannot be intersected without intersecting a point in F that is not in S.
<ed> absolutely
<KD19> In other words, we must prove that there exists a point q in F that intersects ALL balls that intersect S.
<ed> yup
<KD19> that's where this earlier lemma comes in
<KD19> it makes our job easier
<ed> the one about 3 points on a line in 2d?
<KD19> since with that lemma, all we need to prove that there exists a point q in F that is in the smallest ball that intersects all points in S.
<KD19> ah, when I said "intersects S" earlier I meant "intersects all points in S"
<ed> i am following you
<ed> but it feels like we're still very close to the deffinition of shattering, we havent actually touched on anything that has to do with with closed balls and dimensions
<KD19> well that's the usefulness of that lemma
<KD19> I was looking for a way to prove something about ALL balls that intersect all points in S
<cyclicFifths> 2^64
<Olathe> + 5
<ed> KD19 do u see where this will go or are you still exploring
<KD19> hm
<KD19> just exploring
<KD19> what is
<KD19> the definition of a line?
<`m0> hi, are there any websites that have good algorithms?
<`m0> I am finding a way to traverse a complete path if given a matrix of numbers
<KD19> `m0 yes there are
<srart> I'm looking to see if anyone has a classpad, classpad manager, or knows how to get os3 for it.
<KD19> hmm
<`m0> Say I have a matrix of numebrs 5x5 all 0's, and we have a T shape in 1's
<`m0> I would like to count how many 1's are there in that T shape
<koro> 9
<`m0> so basically, count all the adjacent (up|down|left|right) paths
<Olathe> Flood fill !
<KD19> hm
<`m0> http://rafb.net/p/voTCs832.html
<`m0> 7 adjacent ones, I am trying to do this for my research in the quickest way possible (for some other module)
<Olathe> `m0: Look up flood fill algoriths.
<`m0> Olathe: I have done an algorithm that recurses
<`m0> But it seems it just to accept 67x67 matrixes. Antyhing greater than that, it crashes
<`m0> so I assume something is bad in my algorithm , so I am gonna redo
<`m0> http://www.codeproject.com/cs/media/floodfillincsharp.asp
<`m0> aha, excellent
<KD19> people, I just wanted to let you know that I AM REALLY HAPPY
<KD19> well I'm not THAT happy
<KD19> just a little
<cyclicFifths> asphyxia: you still there? sorry I dropped the ball on you bud, I just realized the time and the lack of me having my own HW done.
<asphyxia> cyclicFifths: no problem
<ed> KD19 why do you dance
<ed> i hate dancing
<cyclicFifths> asphyxia: you make any progress?
<asphyxia> I cant really demand anything, can I :)
<ed> but i did just figure out how to play comfortably numb on the guitar
<KD19> I found the Tex source files of the book

Previous Page Next Page

#math

Search ircarchive.info