将Matlab代码C = B(:,ismember(B(1,:),A))转换为Python实现

Got it, let's translate that MATLAB logic to Python using NumPy—since it's the closest equivalent for matrix operations like this. Here's how you can replicate the exact behavior:

Step-by-Step Breakdown #

Your MATLAB code picks columns from matrix B where the first row's element exists in the row vector A. In Python, we'll use NumPy's isin() function to create a boolean mask, then index B with that mask to get the desired columns.

Example Code #

import numpy as np

# Define your input data (A can be a list or NumPy array)
A = np.array([11, 12, 13, 14])  # 1D array works like MATLAB's row vector
B = np.array([
    [10, 11, 12, 15, 13],
    [20, 21, 22, 25, 23],
    [30, 31, 32, 35, 33]
])

# Extract columns where B's first row matches any element in A
C = B[:, np.isin(B[0, :], A)]

print(C)

Output #

[[11 12 13]
 [21 22 23]
 [31 32 33]]

Key Details #

• np.isin(): This function checks each element in B[0, :] (first row of B) against all elements in A, returning a boolean array where True means a match was found.
• Column Indexing: The syntax [:, mask] in NumPy is identical to MATLAB's (:, mask) for selecting columns—super intuitive if you're coming from MATLAB.
• Flexibility: If A is a regular Python list instead of a NumPy array, np.isin() still works perfectly, no conversion needed.
• Duplicates: Just like MATLAB, if B's first row has duplicate elements present in A, all corresponding columns will be included in C.

内容的提问来源于stack exchange，提问作者Python1