Feature: mean of the array np.mean => simsimd.mean

[ternaus]

Describe what you are looking for

If I have image with shape (height, width, num_channel) and I want to compute mean:

numpy

mean = np.mean(x)

If num_channels = 3, I can use OpenCV:

mean = np.mean(cv2.mean(x)[:3])

OpenCV computes mean per channel for RGBA images returning for RGB images array (mean_channel_0, mean_channel_1, mean_2, 0) and we use np.mean() to take average of that.

First works for any shape but slower
Second works for images with 3 channels, but faster

Request:
mean operation that

• works on any shape
• is faster than OpenCV or numpy version

Can you contribute to the implementation?

• I can contribute

Is your feature request specific to a certain interface?

It applies to everything

Contact Details

No response

Is there an existing issue for this?

• I have searched the existing issues

Code of Conduct

• I agree to follow this project's Code of Conduct