np.where not returning expected indices in a 3D array operation with NumPy 1.24
I'm experimenting with I'm building a feature where I tried several approaches but none seem to work..... I've encountered an scenario while trying to use `np.where` to extract indices from a 3D numpy array based on a condition. I have a 3D array representing a grid of data points, and I want to find the indices where the values exceed a certain threshold. However, the output is not what I expected. Here’s the relevant part of my code: ```python import numpy as np # Creating a 3D array of shape (4, 4, 4) data = np.random.rand(4, 4, 4) threshold = 0.5 # Trying to find indices where values exceed the threshold indices = np.where(data > threshold) print(indices) ``` The output for `print(data)` shows a grid with some values exceeding 0.5, but when I print `indices`, it returns three arrays, one for each dimension, containing the indices of the True values, which I am expecting. However, when I try to use these indices to access the elements in the original array, I am getting unexpected results: ```python # Attempting to retrieve the actual values using the indices selected_values = data[indices] print(selected_values) ``` Instead of getting the values that exceed the threshold, it's returning a flattened array of values that doesn’t match with what I thought it should return. I want to ensure that I am correctly using the indices to extract the respective values from the original 3D array. Is there something I’m missing in the way I'm trying to fetch these values or in how `np.where` behaves with multi-dimensional arrays? Any insights would be greatly appreciated! This is part of a larger service I'm building. I'm open to any suggestions. I'm developing on Windows 10 with Python. Any feedback is welcome!