IN4050/in4050-oblig1
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Made some modifications

Browse Source
This commit is contained in:
Cory Balaton 2024-10-05 15:05:23 +02:00
parent b47a1a4955
commit c3c7860b03
Signed by: coryab
GPG Key ID: F7562F0EC4E4A61B
2 changed files with 47 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

46
exhaustive_search.py
View File

@ -1,11 +1,12 @@
import time
from itertools import permutations
from typing import Tuple
from math import factorial
import numpy as np
import numpy.typing as npt
from common import plot_plan, read_data
from common import indexes_to_cities, plot_plan, read_data
def exhaustive_search(distances: npt.NDArray) -> Tuple[float, npt.NDArray]:
@ -34,14 +35,17 @@ def exhaustive_search(distances: npt.NDArray) -> Tuple[float, npt.NDArray]:
),
permutations(range(size)),
),
key=lambda x: x[0] # Make sure that it finds the minimal distance
)
if __name__ == "__main__":
cities, data = read_data("./european_cities.csv")
times = {}
# A loop timing finding the optimal solution for different n
for n in range(6, 11):
for n in range(6,11):
# Time exhaustive search
t0 = time.time_ns()
distance, perm = exhaustive_search(data[:n, :n])
@ -49,32 +53,32 @@ if __name__ == "__main__":
time_elapsed_ms = (t1 - t0) / 1_000_000.0
times[n] = time_elapsed_ms, distance
if n in (6,10):
city_seq = indexes_to_cities(perm, cities)
plot_plan(city_seq)
print(f"Sequence for {n} cities: {city_seq}")
print("")
for n, (time, distance) in times.items():
print(f"Exhaustive search for the {n} first cities:")
print(f"distance : {distance:>12.6f}km")
print(f"time to find solution: {time_elapsed_ms:>12.6f}ms\n")
print(f"{'distance':<25}: {distance:>12.6f}km")
print(f"{'time to find solution':<25}: {time:>12.6f}ms")
print(f"{f'time / {n}!':<25}: {time / factorial(n):>12.6f}\n")
"""Running example
oblig1 on main [?] via 🐍 v3.12.6 took 7s
oblig1 on main [!] via 🐍 v3.12.6 took 14s
❯ python exhaustive_search.py
Exhaustive search for the 6 first cities:
distance : 5018.810000km
time to find solution: 1.105330ms
Exhaustive search for the 7 first cities:
distance : 5487.890000km
time to find solution: 10.089604ms
Exhaustive search for the 8 first cities:
distance : 6667.490000km
time to find solution: 78.810508ms
Exhaustive search for the 9 first cities:
distance : 6678.550000km
time to find solution: 765.676230ms
distance : 5018.810000km
time to find solution : 1.485208ms
time / 6! : 0.002063
Exhaustive search for the 10 first cities:
distance : 7486.310000km
time to find solution: 8281.795515ms
distance : 7486.310000km
time to find solution : 10980.900480ms
time / 10! : 0.003026
"""

23
hill_climbing.py
View File

@ -59,8 +59,29 @@ def hill_climbing(distances: npt.NDArray) -> Tuple[float, npt.NDArray]:
return (current_distance, perm)
def test_hill_climbing(data: npt.NDArray, cities: npt.NDArray, runs: int):
res = [hill_climbing(data) for _ in range(runs)]
res.sort(key=lambda n: n[0])
distances = list(map(lambda n: n[0], res))
best = res[0][0]
worst = res[-1][0]
avg = sum(distances) / runs
standard_deviation = np.sqrt(sum([(i - avg)**2 for i in distances]) / runs)
print(f"Hill climbing for {len(data)} cities.")
print(f"best distance : {best:>12.6f}km")
print(f"worst distance : {worst:>12.6f}km")
print(f"average distance : {avg:>12.6f}km")
print(f"standard deviation: {standard_deviation:>12.6f}km\n")
plot_plan(indexes_to_cities(res[0][1], cities)) # Plot the best one
if __name__ == "__main__":
np.random.seed(1987)
cities, data = read_data("./european_cities.csv")
distance, perm = hill_climbing(data[:10, :10])
plot_plan(indexes_to_cities(perm, cities))
# plot_plan(indexes_to_cities(perm, cities))
test_hill_climbing(data[:10,:10], cities, 20)
test_hill_climbing(data, cities, 20)