[FIXED] Replace (fill) values where the previous value is greater

Issue

I have an unsorted array of integers, and I want to fill all decreasing numbers with the previous larger value.

Example:

>>> np.array([10, -1, 2, 5, 19, 5, 5, 4, 10, 2])
[10 10 10 10 19 19 19 19 19 19]

>>> np.array([0, 3, 5, 4, 3, 7, 2]
[0 3 5 5 5 7 7]

I've come up with this solution, but it has to be a more elegant way of doing it.

def func(a):
    a = a.copy()
    for i in range(1, a.shape[0]):
        if a[i] < a[i-1]:
            a[i] = a[i-1]

    return a

Any suggestions?

I have looked at two similar questions, but I am unable to modify the examples to make them work the way I intend.

1. If value is greater than the previous replace with previous in Pandas

2. Most efficient way to forward-fill NaN values in numpy array

Solution

You can use numpy.maximum.accumulate:

np.maximum.accumulate([10, -1, 2, 5, 19, 5, 5, 4, 10, 2])
# array([10, 10, 10, 10, 19, 19, 19, 19, 19, 19])

np.maximum.accumulate([0, 3, 5, 4, 3, 7, 2])
# array([0, 3, 5, 5, 5, 7, 7])

Answered By - mozway