Question

import numpy as np
def readMatrix(filename):
    rows = []
    for line in open(filename):
        columns = []
        for number in string.split(line):
            columns.append(float(number))
        rows.append(columns)
    return numpy.array(rows)

def writeMatrix(a, filename):
    f = open(filename, 'w')
    for row in a:
        for number in row:
            f.write(str(number) + ' ')
        f.write('\n')
    f.close()

def TaylorMatrixExp(A):
    I = identity(len(A))
    return (I + A + (1./2.)*(dot(A,A)) + (1./6.)*(dot(dot(A,A),A)) + (1./24.)*(dot(dot(A,A),dot(A,A))))

A = readMatrix('matrix.txt')

l, v = eig(A)

L = identity(len(l))

for i in xrange(len(l)):
    L[i][i] = array(exp(l))[i]

VLV = dot(dot(v,L),inv(v))

writeMatrix(VLV,'expA.txt')

ExponentA = TaylorMatrixExp(A)
writeMatrix(ExponentA,'expA.txt')  

The matrix it reads is:
2 2
16 6

I’ve defined two 3 functions, readMatrix(reads a matrix from a textfile), writeMatrix(writes matrix to a file), and TaylorMatrixExp(takes array and expands it). I initially use readMatrix to read a text file containing the above matrix and place it in array A. I take the eigenvalues of A and place it in array l as well as eigenvectors of A and place it in array v. I eventually place the values of array l across the diagonal of an identity matrix. Then I call the writeMatrix function and write exponent to ‘expA.txt’ and then call writeMatrix function again and write matrix ExponentA to ‘expA.txt’. However, it replaces the original matrix and I don’t want it to do that.

and I want it to write to a file
some# some#
some# some#

some#2 some#2
some#2 some#2

but instead it replaces the first matrix
some#2 some#2
some#2 some#2

Answer 1

f = open(filename, 'a')

Lets you append to the file rather than rewrite it, which is what you’re currently doing with the 'w' argument, which is why that matrix is replaced.