二维numpy数组的块平均值(两个维度)

import numpy as np

# 创建一个二维NumPy数组
arr = np.array([[1, 2, 3, 4],
                [5, 6, 7, 8],
                [9, 10, 11, 12],
                [13, 14, 15, 16]])

# 定义块的大小
block_size = (2, 2)

# 计算块平均值
block_means = np.zeros((arr.shape[0] // block_size[0], arr.shape[1] // block_size[1]))

for i in range(0, arr.shape[0], block_size[0]):
    for j in range(0, arr.shape[1], block_size[1]):
        block = arr[i:i + block_size[0], j:j + block_size[1]]
        block_means[i // block_size[0], j // block_size[1]] = np.mean(block)

print("原始数组：\n", arr)
print("块平均值：\n", block_means)

# 确保块大小能够整除数组维度
block_size = (2, 4)

# 使用分块处理
for i in range(0, arr.shape[0], block_size[0]):
    for j in range(0, arr.shape[1], block_size[1]):
        block = arr[i:i + block_size[0], j:j + block_size[1]]
        # 处理每个块

# 调整块大小
block_size = (4, 4)