I am new to numpy and I am trying to avoid for-loops. My requirement is as below:
Input - decimal value (ex. 3) Output - Binary numpy array ( = 00000 01000) Another example :
Input = 6 Output = 00010 00000 Note: I do not want the binary representation of 3. I only need the index value of array = integer to be set.
Is there any standard library function in numpy? Something analogous to get_dummies function in pandas module.
Try this instead. This doesn't use any for loops and if you add some sanity checks it should work fine.
def oneOfK(label): rows = label.shape[0]; rowsIndex=np.arange(rows,dtype="int") oneKLabel = np.zeros((rows,10)) #oneKLabel = np.zeros((rows,np.max(label)+1)) oneKLabel[rowsIndex,label.astype(int)]=1 return oneKLabel Are you looking for a standard function that does something like:
import numpy as np def foo(d, len=10): a = np.zeros(len) a[len-d-1] = 1 return a print foo(3) # [ 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.] print foo(6) # [ 0. 0. 0. 1. 0. 0. 0. 0. 0. 0.] This is more of a comment with code than an answer. Just trying to be clear about what you're looking for, because I'm not sure this function exists as you specify.
np.zeros(len, dtype='bool'), this will give a boolean array.