1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
|
#!/usr/bin/env python
import numpy as np
from scipy.ndimage import label
with open("day9.txt") as data:
depthmap = []
for line in data:
row = [int(num) for num in line.strip()]
depthmap.append(row)
depthmap = np.array(depthmap)
# part 1
# convolve across grid
def conv_filter(position: (int, int)) -> int:
def neighbors():
x, y = position
neighboring = {(x-1, y), (x+1, y), (x, y-1), (x, y+1)}
return {(xPos, yPos) for xPos, yPos in neighboring if 0 <= xPos <= depthmap.shape[0]-1 and 0 <= yPos <= depthmap.shape[1]-1}
if all(depthmap[position] < depthmap[pos] for pos in neighbors()):
return depthmap[position] + 1
else:
return 0
print(sum(conv_filter((x, y)) for x in range(depthmap.shape[0]) for y in range(depthmap.shape[1])))
# part 2
basins, num_basins = label(depthmap != 9)
counts = [np.count_nonzero(basins == num) for num in range(1, num_basins + 1)]
counts.sort()
print(np.prod(counts[-3:]))
|