convolution.py

import numpy as np

# Input image: 5x5 grayscale image 
image = np.array([[10, 20, 30, 40, 50], 
                  [60, 70, 80, 90, 100], 
                  [110, 120, 130, 140, 150], 
                  [160, 170, 180, 190, 200], 
                  [210, 220, 230, 240, 250]])

# Filter or Kernel: 3x3
filter = np.array([[0, 1, 2], 
                   [2, 2, 0], 
                   [0, 1, 2]])

# Dimensions of the output feature map
output_dim = image.shape[0] - filter.shape[0] + 1

print("Output dimensions:")
print(output_dim)

# Initialize the output feature map
output = np.zeros((output_dim, output_dim))

# Convolve the filter over the image
for i in range(output_dim):
    for j in range(output_dim):
        output[i][j] = np.sum(image[i:i+3, j:j+3] * filter)

print("Output feature map:")
print(output)

const step = w => x => y => z => {
  if (z <= y) return w;
  else return x;
};
const heavyside = step(0)(1)(0);
const sign = step(-1)(1)(0);