!apt-get update && apt-get install -y python3-opencv !pip install opencv-python==4.5.1.48

# setting up notebook all modules %matplotlib inline import matplotlib.pyplot as plt plt.style.use('ggplot') import numpy as np import pandas as pd import cv2 from functools import reduce

df1 = pd.read_csv('n_primitives_tests/tests_pdp_intersection.txt') df2 = pd.read_csv('n_primitives_tests/tests_pdp_union.txt') df3 = pd.read_csv('n_primitives_tests/tests_pdp_difference.txt')

df_normal = pd.concat([df1, df2, df3], axis=1) df_normal.columns = ["Triangle Number", "Intersection Time", "tris2", "Union Time", "tris3", "Difference Time"] df_normal = df_normal.drop(columns=['tris2', 'tris3']) col = df_normal.loc[: , "Intersection Time":"Difference Time"] df_normal['NaiCSG Mean Time'] = col.mean(axis=1)

df_normal

df_normal.plot(x ='Triangle Number', y=['Intersection Time', 'Union Time', 'Difference Time', 'NaiCSG Mean Time'], kind = 'line', lw=1) plt.show() # step size in triangles grow exponentiall; therefore the bar chart shows different results than the scaled one below here.

bsp1 = pd.read_csv('n_primitives_tests/tests_bsp_pdp_intersection.txt') bsp2 = pd.read_csv('n_primitives_tests/tests_bsp_pdp_union.txt') bsp3 = pd.read_csv('n_primitives_tests/tests_bsp_pdp_difference.txt')

df_bsp = pd.concat([bsp1, bsp2, bsp3], axis=1) df_bsp.columns = ["Triangle Number", "Intersection Time", "tris2", "Union Time", "tris3", "Difference Time"] df_bsp = df_bsp.drop(columns=['tris2', 'tris3']) col2 = df_bsp.loc[: , "Intersection Time":"Difference Time"] df_bsp['BinCSG Mean Time'] = col2.mean(axis=1)

df_bsp

df_bsp.plot(x ='Triangle Number', y=['Intersection Time', 'Union Time', 'Difference Time', 'BinCSG Mean Time'], kind = 'line', lw=1) plt.show()

optim1 = pd.read_csv('n_primitives_tests/tests_bsp_optim_intersection.txt') optim2 = pd.read_csv('n_primitives_tests/tests_bsp_optim_union.txt') optim3 = pd.read_csv('n_primitives_tests/tests_bsp_optim_difference.txt')

df_optim = pd.concat([optim1, optim2, optim3], axis=1) df_optim.columns = ["Triangle Number", "Intersection Time", "tris2", "Union Time", "tris3", "Difference Time"] df_optim = df_optim.drop(columns=['tris2', 'tris3']) col3 = df_optim.loc[: , "Intersection Time":"Difference Time"] df_optim['OptiCSG Mean Time'] = col3.mean(axis=1)

df_optim

df_optim.plot(x ='Triangle Number', y=['Intersection Time', 'Union Time', 'Difference Time', 'OptiCSG Mean Time'], kind = 'line', lw=1) plt.show()

df_comp = pd.concat([df_normal["Triangle Number"], df_normal["NaiCSG Mean Time"], df_bsp[0:len(df_normal.index)]["BinCSG Mean Time"], df_optim[0:len(df_normal.index)]["OptiCSG Mean Time"]], axis=1) df_comp

df_comp.plot(x="Triangle Number", y=["NaiCSG Mean Time", "BinCSG Mean Time", "OptiCSG Mean Time"], kind="line", lw=1) plt.title('Performance of all versions') plt.show()

nest_1 = pd.read_csv('n_nests_tests/nesting_base_union.txt', header=None) nest_2 = pd.read_csv('n_nests_tests/nesting_optim_union.txt', header=None) nest_3 = pd.read_csv('n_nests_tests/nesting_bin_union.txt', header=None) nest_viewport = pd.read_csv('n_nests_tests/nest_to_viewport.txt', header=None)

df_nest = pd.concat([nest_1, nest_2, nest_3, nest_viewport], axis=1) df_nest.columns = ["Triangle Number", "Number of Nests", "NaiCSG Time", "tris2", "nests2", "OptiCSG Time", "tris3", "nests3", "BinCSG Time", "nests4", "Viewport Fill"] df_nest = df_nest.drop(columns=['tris2', 'nests2', 'tris3', 'nests3', 'nests4']) mean_viewport_fill = df_nest['Viewport Fill'].mean() print(mean_viewport_fill)

df_nest

df_nest.plot(x ='Number of Nests', y=['NaiCSG Time', 'BinCSG Time', 'OptiCSG Time'], kind = 'line', lw=2) plt.title("Render time in comparasion to an increase in nests") plt.show()

## First get the view port rates n_step = [] n_viewPort = [] for i in range(1, 21, 1): n_step.append(i) img = cv2.imread('view_port_renders/bin_port_union_' + str(i - 1) + '.png', cv2.IMREAD_GRAYSCALE) n_pixels = img.size n_white_pix = np.sum(img == 255) n_rate = n_white_pix/n_pixels * 100 n_viewPort.append(n_rate) n_viewPort.reverse() data_tuples = list(zip(n_step,n_viewPort)) generic_rate = pd.DataFrame(data_tuples, columns=['Step','Viewport Fill Rate'])

viewport_1 = pd.read_csv('viewport_tests/viewport_bin_union.txt', header=None) viewport_1.columns = ["tris1", "step2", "BinCSG Time"] viewport_1 = viewport_1.drop(columns=['tris1', 'step2']) viewport_2 = pd.read_csv('viewport_tests/viewport_optim_union.txt', header=None) viewport_2.columns = ["tris1", "step2", "OptiCSG Time"] viewport_2 = viewport_2.drop(columns=['tris1', 'step2']) viewport_3 = pd.read_csv('viewport_tests/viewport_base_union.txt', header=None) viewport_3.columns = ["tris1", "step2", "NaiCSG Time"] viewport_3 = viewport_3.drop(columns=['tris1', 'step2']) df_viewport = pd.concat([generic_rate, viewport_1, viewport_2, viewport_3], axis=1)

subplot = df_viewport.plot(x ='Viewport Fill Rate', y=['NaiCSG Time', 'BinCSG Time', 'OptiCSG Time'], lw=2) plt.title("Render time with respect to viewport fill percentage") plt.show()

df_viewport