[Tutor] Matrix Read-in and Inversion

[Corran Webster]

At 1:15 PM -0400 4/9/01, Robin B. Lake wrote:
>I have a 100 x 150 matrix which I want to invert using the Generalized
>Inverse capability of MathPy.
>
>How might I read this dataset in from either a comma-separated file or
>an Excel spreadsheet.  I'm on a Mac G3, OS 8.6.

Something like the following should work for comma-separated, 
assuming each matrix row is on a line by itself and the numbers are 
floats (warning, untested code):

import Numeric

# open the file
f = open("filename")

# read data into a list of lists
data = []
for line in f.readlines():
      linedata = []
      # split the line at each comma
      for value in line.split(","):
           linedata.append(float(value))
      data.append(linedata)
f.close()

# convert to numpy array
a = Numeric.array(data)

In a python version with list comprehensions (2.0 or better), this 
can be compressed considerably:

import Numeric

f = open("filename")
a = Numeric.array([[float(value) for value in line.split(",")] for 
line in f.readlines()])
f.close()

Regards,
Corran