tests.py

import numpy as np
from matplotlib import pyplot as plt
from pathfinder import PointCloud
import time
import logging

POINT_CLOUD_FILE = '../data/construction_site.las'


def perf_time_sample_size():
    logging.disable()
    x = np.arange(0.01, 0.2, 0.01)
    y = np.array([])

    for i in x:
        start_time = time.time()
        pc = PointCloud(filename=POINT_CLOUD_FILE, points_proportion=i)
        e = pc.get_epsilon()
        pc.apply_dbscan(e)
        pc.write_path_output(f'test/perf_time_density_{i*100}.json', 5)
        exec_time = (time.time() - start_time)
        y = np.append(y, exec_time)
        print(f'Computed path with {round(i * 100)}% of the points. Exec. time:\t{exec_time}\tseconds.')

    plt.plot(x, y, ".")
    plt.show()


def perf_clustering_sample_size():
    logging.disable()
    x = np.arange(1, 10, 1)
    y = np.array([])

    for i in x:
        pc = PointCloud(filename=POINT_CLOUD_FILE, points_proportion=i/100)
        e = pc.get_epsilon()
        pc.apply_dbscan(e)
        img = pc.generate_debug_image(1000, 1000, 3)
        img.save(f'test/perf_clustering_sample_size/img_{int(i)}.png')
        print(f"Computed clustering image img_{int(i)}.png")


perf_clustering_sample_size()