Unexpected behavior when using NumPy's np.where with multiple conditions on 3D arrays
Could someone explain I've searched everywhere and can't find a clear answer. I'm stuck on something that should probably be simple. I'm sure I'm missing something obvious here, but I'm encountering an issue while trying to apply multiple conditions on a 3D NumPy array using `np.where`. I have a 3D array representing a grid of values, and I want to set values based on two conditions: one for values greater than a threshold and another for checking if the corresponding values in a mask array are `True`. However, when I run my code, the result doesn't match my expectations. Here's my setup: ```python import numpy as np grid = np.random.rand(4, 4, 4) # Random values between 0 and 1 mask = np.array([[[True, False, True, True], [True, True, False, True], [False, True, True, False], [True, False, True, True]], [[False, True, False, False], [True, False, True, True], [True, True, False, True], [True, False, False, True]], [[True, True, True, False], [False, True, False, True], [True, True, True, True], [False, True, True, False]], [[True, False, True, True], [False, False, True, True], [True, True, False, True], [True, False, True, False]]]) threshold = 0.5 # Trying to set values to -1 where grid > threshold and mask is True result = np.where((grid > threshold) & mask, -1, grid) ``` I've checked the shapes of both `grid` and `mask`, and they match (both are 4x4x4). However, the values in `result` are not being set to -1 as I anticipated. Instead, the output retains many of the original grid values. When I print `result`, I see that it has many instances of the original values instead of -1. Additionally, I've tried breaking down the condition like this: ```python condition = grid > threshold combined_condition = condition & mask result = np.where(combined_condition, -1, grid) ``` But the outcome remains the same. Is there something I'm missing here regarding the logical operations or the behavior of `np.where` with multi-dimensional arrays? I'm using NumPy version 1.21.0, in case that matters. Any insights would be greatly appreciated! What's the best practice here? This is part of a larger service I'm building. Is there a better approach? My team is using Python for this application. Has anyone dealt with something similar? Am I missing something obvious? Any ideas how to fix this?