Simple operations on every element in numpy arrays

Question

I'm a newbie at numpy, but I was of that notion that you should not iterate over numpy arrays as this defies the purpose of numpy.

I'm trying to perform elementwise operations on a numpy array, but I don't understand the syntax apparantly:

Code:

for line in fidNNC:
    temp = line.strip().split()
    temp.insert(0,0)
    CC.append(temp[0:7])

fidNNC.close()
NNC = np.array(CC)
del(CC)

inds = np.arange(len(NNC))[ NNC[:,4]-1 == NNC[:,1] ]
NNCX = NNC[inds,:]
inds = np.arange(len(NNC))[ NNC[:,5]-1 == NNC[:,2] ]
NNCY = NNC[inds,:]

The file fidNNC contains about a million rows and ten columns of ints.

Error message:

    inds = np.arange(len(NNC))[ NNC[:,4]-1 == NNC[:,1] ]
TypeError: unsupported operand type(s) for -: 'numpy.ndarray' and 'int'

How do I subtract 1 from each and every one element of NNC[:,4] and NNC[:,5] without iterating?

Thanks in advance,

Daniel

Ahh, I should read that question/answer closely, looked similar to my problem. NNC.dtype = |S3 — dansolhe
– dansolhe, Commented Jul 3, 2013 at 13:58
Which I was guessing after looking at your code again :) I removed the comment and added an answer. — Lev Levitsky
– Lev Levitsky, Commented Jul 3, 2013 at 13:59

Lev Levitsky · Accepted Answer · 2013-07-03 13:59:03Z

1

You are not converting to int when reading the file, thus the created array has the string type. Convert to int and it should work:

for line in fidNNC:
    temp = [int(i) for i in line.strip().split()]
    temp.insert(0,0)
    CC.append(temp[0:7])

answered Jul 3, 2013 at 13:59

Lev Levitsky

66.4k23 gold badges155 silver badges184 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

sedavidw · Accepted Answer · 2013-07-03 13:56:55Z

0

You can deconstruct the matrix by column and perform operations on the columns you like and re-create, np.c_ allows this. For example a matrix x we can subtract 1 from the first column like so (and place result in z)

z = np.c_[ (x[:,1] - 1), x[:,1:] ]

This pulls out the first column on x, elementwise subtraction on that column and concatenates the rest of the untouched columns. To do two columns simply break up further

answered Jul 3, 2013 at 13:56

sedavidw

11.8k18 gold badges70 silver badges103 bronze badges

Collectives™ on Stack Overflow

Simple operations on every element in numpy arrays

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related