Mail Archive Search: <+from:fchen@stern.nyu.edu>

Results:

References: [ +from:fchen@stern.nyu.edu: 15 ]

Total 15 documents matching your query.

1. [S] loops in likelihood in Gaussian mixtures (score: 1): Author: Fei Chen <fchen@stern.nyu.edu>; Date: Wed, 9 Dec 1998 13:29:14 -0500 (EST); Hi all, I need help with how to avoid the following loop. I'm trying to calculate the likelihood of some data. The likelihood function for each data point is a mixture of K weighted normals, i.e. P(y; /archives/html/s-news/1998-12/msg00068.html (8,317 bytes)
2. Thanks!: Re: [S] loops in likelihood in Gaussian mixtures (score: 1): Author: Fei Chen <fchen@stern.nyu.edu>; Date: Wed, 9 Dec 1998 16:47:47 -0500 (EST); Thanks to Patrick Burns, Rolf Turner and Claude Nadeau for their quick responses. The following from Rolf Turner works well: in terms of unix.time, for N==1000, K=20, with.loops.unix.time: [1] 13.389; /archives/html/s-news/1998-12/msg00071.html (7,550 bytes)
3. [S] growing matrices (score: 1): Author: Fei Chen <fchen@stern.nyu.edu>; Date: Sat, 19 Sep 1998 19:28:44 -0400 (EDT); hi all, i was wondering if there is a way to grow matrices dynamically without using rbind()? which exhausts my memory after some number of loops. i have a function, say, myfun(), that returns a matr; /archives/html/s-news/1998-09/msg00131.html (7,150 bytes)
4. [S] thanks and another question (score: 1): Author: Fei Chen <fchen@stern.nyu.edu>; Date: Tue, 22 Sep 1998 12:03:54 -0400 (EDT); hi all, thanks for all the response to my original question of how to grow a matrix w/o using rbind(). i ended up using the following solution: (myfun() returns a matrix of fixed columns but variable; /archives/html/s-news/1998-09/msg00140.html (8,172 bytes)
5. [S] is there a cumouter() like function? (score: 1): Author: Fei Chen <fchen@stern.nyu.edu>; Date: Sun, 23 Aug 1998 12:51:23 -0400 (EDT); hi all, i'm trying to do the following, is there a way to avoid the loop? a_matrix(rnorm(30),10,3) # some random N x 3 matrix b_matrix(0,3,3) for(i in 2:10) b_b+a[i-1,]%o%a[i,] it seems this sort of; /archives/html/s-news/1998-08/msg00167.html (7,685 bytes)
6. Thanks: [S] is there a cumouter() like function? (score: 1): Author: Fei Chen <fchen@stern.nyu.edu>; Date: Mon, 24 Aug 1998 12:22:20 -0400 (EDT); many thanks to Nick Ellis, Christian Keller, Lutz Prechelt, Patrick Burns, Claude Nadeau and Bill Dunlap for answer my question: crossprod is the solution! -- original question: how to avoid the foll; /archives/html/s-news/1998-08/msg00170.html (8,011 bytes)
7. [S] vector subscripting question (score: 1): Author: Fei Chen <fchen@stern.nyu.edu>; Date: Thu, 25 Jun 1998 11:41:39 -0400 (EDT); hi all, i thought i had understood this, but maybe not. let's say i have two vetors a and b of length 120000, should these two expressions be the same? 1. a[b>0] 2. a[b[b>0]] in my case they turned o; /archives/html/s-news/1998-06/msg00186.html (7,244 bytes)
8. Re: [S] vector subscripting question (score: 1): Author: Fei Chen <fchen@stern.nyu.edu>; Date: Thu, 25 Jun 1998 12:28:41 -0400 (EDT); hi all, thanks for all these quick response. i can't believe how obvious this should have been in hind sight. i guess i was just a little confused over a coincidental example i had. thanks to Z. Todd; /archives/html/s-news/1998-06/msg00187.html (8,240 bytes)
9. [S] out of memory (score: 1): Author: Fei Chen <fchen@stern.nyu.edu>; Date: Thu, 25 Jun 1998 21:09:16 -0400 (EDT); hi all, i encountered a rather strange problem and not sure what to do about it. i'm runing splus3.5 on ultra sparc 2 with 500 ram and 900m swap. i try to load a file of 3.5 meg size using the scan(); /archives/html/s-news/1998-06/msg00199.html (7,047 bytes)
10. msg00212.html (score: 1): Author: Fei Chen <fchen@stern.nyu.edu>; Date: Thu, 21 May 1998 17:08:05 -0400 (EDT); f; /archives/html/s-news/1998-05/msg00212.html (7,588 bytes)
11. Re: [S] Re: your mail (score: 1): Author: Fei Chen <fchen@stern.nyu.edu>; Date: Thu, 21 May 1998 17:57:32 -0400 (EDT); :; /archives/html/s-news/1998-05/msg00216.html (9,501 bytes)
12. [S] ModAppStat compilation error, etc (score: 1): Author: Fei Chen <fchen@stern.nyu.edu>; Date: Fri, 22 May 1998 12:47:59 -0400 (EDT); k; /archives/html/s-news/1998-05/msg00226.html (10,394 bytes)
13. [S] matrix square root (score: 1): Author: Fei Chen <fchen@stern.nyu.edu>; Date: Tue, 21 Apr 1998 17:58:52 -0400 (EDT); hi all, this might be a thing easy to do. but i don't know how: in MATLAB, there is a function sqrtm() which takes the square root of a matrix, as opposed to sqrt() which takes the square root of eve; /archives/html/s-news/1998-04/msg00164.html (7,100 bytes)
14. [S] histogram color (score: 1): Author: Fei Chen <fchen@stern.nyu.edu>; Date: Thu, 5 Feb 1998 21:21:10 -0500 (EST); hi all, i was wondering if anyone can tell me how to set the histogram bar colors to be black? i can't figure it out for the life of me. thanks in advance, fei -- This message was distributed by s-ne; /archives/html/s-news/1998-02/msg00043.html (6,665 bytes)
15. [S] thanks! and EM (score: 1): Author: Fei Chen <fchen@stern.nyu.edu>; Date: Sat, 7 Feb 1998 08:30:22 -0500 (EST); hi all, thanks so much for all the responses for the histogram color problem. i can't believe it was so easy! one more question: does anyone know of a splus function that does clustering with Gaussia; /archives/html/s-news/1998-02/msg00055.html (6,809 bytes)

Current List: 1 - 15
Page: [1]