Multiple actions for multiple arrays in list comprehension-CodePudding

I would like to substitute the '1's on an array nparray1 = np.array([1, 1, 0, 0, 1]) by the values on another array nparray2 = np.array([8,7,4]), to produce nparray1 = np.array([8, 7, 0, 0, 4]).

I was hoping to do this as efficiently as possible, hence I thought using a list comprehension would be the best alternative, however, I am unable to find a way to do this. The alternative for loop would look like:

nparray1 = np.array([1, 1, 0, 0, 1])
nparray2 = np.array([8,7,4])

for i in range(len(nparray1)):
    if nparray1[i]==True:
        nparray1[i] = nparray2[0]
        nparray2    = nparray2[1:]

CodePudding user response：

You can use:

nparray1[nparray1 == 1] = nparray2
print(nparray1)

# Output
array([8, 7, 0, 0, 4])

CodePudding user response：

Alternatively, I think you can also keep track of an idx variable which you only increment when you see a 1 in the first numpy array.

nparray1 = np.array([1, 1, 0, 0, 1])
nparray2 = np.array([8,7,4])

idx = 0;
for i in range(len(nparray1)):
    if nparray1[i] == 1: 
        nparray1[i] = nparray1[i] * nparray2[idx]
        idx =1

print(nparray1)

or alternatively, via the use of iter():

nparray1 = numpy.array([1, 1, 0, 0, 1])
nparray2 = iter(numpy.array([8,7,4]))

for i in range(len(nparray1)):
    if nparray1[i]== 1:
        nparray1[i] = next(nparray2)

print(nparray1)