Sliding windows in numpy with varying window size

Question

I am generating data with a timestamp (counting up). I then want to seperate the array based on the timestamp and calculate the mean of the data in each window. My new array then has a new "timestamp" and the calculated mean data.

My Code is working as it is supposed to, but I do believe there is a more numpy-like way. I believe the while loop can be removed and np.where checking the whole array, as it is already sorted as-well.

Thanks for your help.

# generating test data, first row timestamps, always counting up and random data 
data = np.array([np.cumsum(np.random.randint(100, size=20)), np.random.randint(1, 5, size=20)])
print(data)

window_size = 200
overlap = 100

i, l_lim, u_lim = 0, 0, window_size
timestamps = []
window_mean = []
while u_lim < data[0, -1]:
    window_mean.append(np.mean(data[1, np.where((data[0, :] > l_lim) & (data[0, :] <= u_lim))]))
    timestamps.append(i)
    l_lim = u_lim - overlap
    u_lim = l_lim + window_size
    i += 1

print(np.array([timestamps, window_mean]))

Answer 1

While I may have reduced the number of lines of code, I do not think I have really improved it that much. The main difference is the method of iteration, and its use to define the number selection boundaries, but otherwise, I could not see any way to improve on your code. Here is my attempt for what it is worth:

Code:

import numpy as np

np.random.seed(5)
data = np.array([np.cumsum(np.random.randint(100, size=20)), np.random.randint(1, 5, size=20)])
print("Data:", data)

window_size = 200
overlap = 100

for i in range((max(data[0]) // (window_size-overlap)) + 1):
    result = np.mean(data[1, np.where((data[0] > i*(window_size-overlap)) & (data[0] <= (i*(window_size-overlap)) + window_size))])
    print(f"{i}: {result:.2f}")

Output:

Data: [[  99  177  238  254  327  335  397  424  454  534  541  617  632  685 765  792  836  913  988 1053]
       [   4    3    1    3    2    3    3    2    2    3    2    2    3    2    3    4    1    3    2    3]]

0: 3.50
1: 2.33
2: 2.40
3: 2.40
4: 2.25
5: 2.40
6: 2.80
7: 2.67
8: 2.00
9: 2.67
10: 3.00