The code behind the picture

3 teaching cells · Python standard library · Read-only

1. 1. Prepare the inputs

   The three rows of the watermelon around the edge pixel at row 10, column 0, first two columns, and the lesson's hand-set filter.

   In [ ]:

   import json
   
   rows = [[8, 8], [8, 0], [0, 4]]  # picture rows 9-11, columns 0-1
   kernel = [[-1, 0, 1], [-1, 0, 1], [-1, 0, 1]]

2. 2. Pad the left edge three ways

   Zero supplies 0, replicate copies column 0, reflect copies column 1 (the mirror without repeating the edge). Then take the weighted sum of each 3 × 3 window.

   In [ ]:

   def window(policy):
       left = {"zero": lambda r: 0, "replicate": lambda r: r[0], "reflect": lambda r: r[1]}[policy]
       return [[left(r), r[0], r[1]] for r in rows]
   
   def respond(patch):
       return sum(k * v for kr, pr in zip(kernel, patch) for k, v in zip(kr, pr))
   
   result = {p: respond(window(p)) for p in ["zero", "replicate", "reflect"]}

3. 3. Inspect the result

   Print the result as JSON so both languages use the same output format.

   In [ ]:

   print(json.dumps(result))

{
  "zero": 12,
  "replicate": -4,
  "reflect": 0
}