Assign value to array (Python, Numpy)

Question

I have initialized an array in the form of

for x in range(0,21000): 
    if np.amax(probability[x]) > 0.2:
        i = i+1

Label = np.zeros(shape=(i,130))

and now with the following for cycle I would like to assign the value to each row of Label

for x in range(0,21000):
    if np.amax(probability[x]) > 0.2:
        Label[i] = [feature[x],y[x],np.amax(probability)[x]]
    i = i+1

Where feature is an array 128*21000. Unfortunately, I keep receving the following error:

Label[i] = [feature[x],y[x],np.amax(probability)[x]]
IndexError: invalid index to scalar variable.

You probably want to provide the axis value for the np.amax call otherwise it returns a scalar: docs.scipy.org/doc/numpy-1.14.0/reference/generated/… — sshashank124
– sshashank124, Commented May 16, 2018 at 13:48
This isn't coming from the assignment, but from np.amax(probability)[x], where you're trying to index a scalar. When in doubt about something try each expression individually and see what you get. — Iguananaut
– Iguananaut, Commented May 16, 2018 at 13:50
Please don't change your question mid-stream. Earlier you were asking about a different error. — Iguananaut
– Iguananaut, Commented May 16, 2018 at 13:53
Okey, sorry I dind't know that. Can I ask another question ? — Tommaso Bendinelli
– Tommaso Bendinelli, Commented May 16, 2018 at 13:54

sshashank124 · Accepted Answer · 2018-05-16 13:51:09Z

1

You don't want a single scalar value for np.amax but rather an array of values. Therefore you need to specify a specific axis for the call as follows:

np.amax(probability, axis=1)

You can see examples of the use of the axis argument from the docs

Or if you really do want to use the scalar value, don't index it using [x] in np.amax(probability)[x]

answered May 16, 2018 at 13:51

sshashank124

32.3k10 gold badges72 silver badges76 bronze badges

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

Collectives™ on Stack Overflow

Assign value to array (Python, Numpy)

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related