[SciPy-user] using linalg.lstsq

[Mitchell Timin]

--- Bill Baxter <wbaxter at gmail.com> wrote:
> Try this:
> 
> #                x   y
> A = array([[ 0., 0.],
>                 [ 0., 1.],
>                 [ 1., 0.],
>                 [ 1., 1.]])
> A = hstack( (A, ones((A.shape[0],1)) ))
> z = array([.5,.6,.4,.55])
> c = scipy.linalg.lstsq(A,z)[0]
<snip>
That seems to work.  I made it into this program:
----------------------------------------------
from scipy import *

A = array([[ 0., 0.],
           [ 0., 1.],
           [ 1., 0.],
           [ 1., 1.]])
A = hstack( (A, ones((A.shape[0],1)) ))
z = array([.5,.6,.4,.55])

print "A"
print A
print
print "z"
print z
print

(X, resids, rank, s) = linalg.lstsq(A,z)   

print "X"
print X
print
print "resids"
print resids
print
print "rank"
print rank
print
print "s"
print s  
print
print "calculate z at original points:"
print X[0]
print X[0] + X[1]
print X[0] + X[2]
print X[0] + X[1] + X[2]
----------------------------------------------
and it produced this result:
----------------------------------------------
A
[[ 0.  0.  1.]
 [ 0.  1.  1.]
 [ 1.  0.  1.]
 [ 1.  1.  1.]]

z
[ 0.5   0.6   0.4   0.55]

X
[-0.075   0.125   0.4875]

resids
0.000625

rank
3

s
[ 2.5243378   1.          0.79228699]

calculate z at original points:
-0.075
0.05
0.4125
0.5375
----------------------------------------------
I can't tell if that answer is correct.  I will create a better test case to 
make it obvious.

Thanks,

Mitchell Timin

I'm proud of http://ANNEvolve.sourceforge.net.  If you want to write software, or articles, or do testing, or research for ANNEvolve, let me know.

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com