SciPy-User

Download

scipy-user@python.org

September 2005

53 participants
52 discussions

getting the "best" two eigenvectors for a PCA analysis with a power method
by Syd Diamond Sept. 1, 2005

Sept. 1, 2005

I'll be honest -- I'm mostly interested in the end not the means right now. I need to find the first and second principal components of a 120x520 matrix. Right now, I'm using the "from LinearAlgebra import eigenvectors" method to find eigenvectors of the covariance matrix. This worked fine for my 120x90 matrix, but it's choking on the 120x520 matrix. When I talked to a friend, he said usually shortcuts were taken... like the power method for computing the dominant eigenpair. I've googled around without any luck. Can anyone please give me some guidance here? In return, I can offer some hilariously inefficient code from a guy who still doesn't know what he's doing... enjoy :) Thank you everyone for your continued help. (And if you can offer guidance on the code as well, by all means be my guest :)) from Numeric import array, zeros, shape, Float from LinearAlgebra import eigenvectors from scipy import mean, std, cov def princomp(X): evals, evecs = eigenvectors(cov(X)) eSorter = {} for index in xrange(len(evals)): eval = evals[index] evec = evecs[index] eSorter[eval] = evec evalsList = eSorter.keys().sort() evecNew = zeros(shape(evecs), typecode=Float) counter = 0 for eval in evals: evec = eSorter[eval] evecNew[counter] = evec counter += 1 return evecNew

4 5

python crashes on scipy testing
by rainman Sept. 1, 2005

Sept. 1, 2005

Hello scipy-user, God, that's one thing I hate about Open Source - having to hunt high and low for binaries... or to face tremendous nightmares of building-from-sources. Let's see. I'm running WinXP on A-64. I have Python 2.4. Some time ago I decided to try SciPy. Aw, hell! No prebuilt binaries for Py2.4. And you need a ton of compilers to "make" sources. But I'm stubborn. I had nerve. I took: MSYS-1.0.10.exe MinGW-3.1.0-1.exe atlas3.6.0_WinNT_PIII.zip SciPy_complete-0.3.2.tar.gz Numeric-23.8.win32-py2.4.exe F2PY-2-latest.win32.exe , then I tuned environment vars and started "setup.py build". Oh! how cute! they forgot to put wgnuplot.exe inside! Let's take it from SciPy_complete-0.3.2.win32-py2.3-num23.5.exe. Trying once more... Tada! Compilation finally finished without error message. Let's run: import scipy scipy.test() Yeah, a lot of "no test file" messages, dots, some cryptic mumbling ending with "zswap:n=3", more dots, WHOOPS! python crashes. So, the question is... WHAT ELSE??? -- Best regards, rainman mailto:rainman.rocks@gmail.com

5 4