Reputation: 1141
I have a specific integer value k
and I want to find matching values in a column of a 2d numpy array (using a "for" loop).
I thought I could use an if
statement and directly compare the single array element with the integer value k
.
Here is my code:
for i in range(mock.shape[0]):
n_cl = int(mock[i,0]/3500.)
zcl = mock[i,5]
pick = [np.random.random_integers(200, size=(n_cl))]
print pick[0]
if(zcl <= 0.05):
for k in range(len(pick)) :
for j in range(z_001.shape[0]):
n = z_001[j,1]
if (int(n) == pick[k]):
binaries[j,7] = mock[i,0]
binaries[j,8] = mock[i,1]
binaries[j,9] = mock[i,2]
binaries[j,10] = mock[i,3]
binaries[j,11] = mock[i,4]
I always get a ValueError about the truth value of an array with more than one element, which is ambiguous. I understand that the problem is in " int(n) == k", but I wonder where am I wrong, and how could I put things better and make this part of my code work.
Upvotes: 2
Views: 6060
Reputation: 1141
I resolved my problem by changing the pick
list definition, from pick = [np.random.random_integers(200, size=(n_cl))]
to pick = np.random.random_integers(200, size=(n_cl))
.
The problem occurred because I had been comparing an array value to an integer value.
Upvotes: 0
Reputation: 353
Well, consider:
So, one possible solution to the problem could be:
import numpy as np
# generating some fake data
x = np.repeat([[1,2,3,4]], 4, axis=0)
x[2,2] = 200
x[3,1] = 200
# retrieving the indices where "x==200" using np.where
indices = np.where(x==200)
print indices
which gives:
(array([2, 3]), array([2, 1]))
so you can just index any other array (with appropriate shape) with indices
and you will get the value of that array in the indices
positions.
Upvotes: 3