|
1
|
|
|
# Author: Simon Blanke |
|
2
|
|
|
# Email: [email protected] |
|
3
|
|
|
# License: MIT License |
|
4
|
|
|
|
|
5
|
|
|
|
|
6
|
|
|
import numpy as np |
|
7
|
|
|
import pandas as pd |
|
8
|
|
|
import matplotlib.pyplot as plt |
|
9
|
|
|
from matplotlib.ticker import MultipleLocator |
|
10
|
|
|
|
|
11
|
|
|
# import matplotlib as mpl |
|
12
|
|
|
import seaborn as sns |
|
13
|
|
|
|
|
14
|
|
|
ml = MultipleLocator(5) |
|
15
|
|
|
sns.set(style="whitegrid") |
|
16
|
|
|
|
|
17
|
|
|
|
|
18
|
|
|
def plot_optimizer_time(model_name, y_min, y_max, step_major, title): |
|
19
|
|
|
file_name = "optimizer_calc_time_" + model_name |
|
20
|
|
|
|
|
21
|
|
|
data = pd.read_csv("./data/" + file_name, header=0) |
|
22
|
|
|
|
|
23
|
|
|
columns = data.columns |
|
24
|
|
|
values = data.values |
|
25
|
|
|
|
|
26
|
|
|
no_opt = values[:, 0] |
|
27
|
|
|
|
|
28
|
|
|
values_norm = values / no_opt[:, None] |
|
29
|
|
|
|
|
30
|
|
|
fig, ax = plt.subplots() |
|
31
|
|
|
plt.figure(figsize=(10, 8)) |
|
32
|
|
|
# plt.grid(True) |
|
33
|
|
|
# plt.ylabel(r"$\dfrac{T_{norm}}{iteration}$", rotation=0) |
|
34
|
|
|
plt.ylabel(r"$T_{norm}$", rotation=0) |
|
35
|
|
|
|
|
36
|
|
|
data = pd.DataFrame(values_norm, columns=columns) |
|
37
|
|
|
ax = sns.barplot(data=data, alpha=0.75, capsize=0.1) |
|
38
|
|
|
# ax = sns.barplot(x=columns, y=values_norm, alpha=0.75, ci="sd") |
|
39
|
|
|
ax.set_xticklabels(ax.get_xticklabels(), rotation=75) |
|
40
|
|
|
ax.set_title(title) |
|
41
|
|
|
|
|
42
|
|
|
# ax.tick_params(axis="x", which="minor", bottom=False) |
|
43
|
|
|
# ax.yaxis.set_minor_locator(ml) |
|
44
|
|
|
|
|
45
|
|
|
# ax.set_yscale("log") |
|
46
|
|
|
y_min = y_min |
|
47
|
|
|
y_max = y_max |
|
48
|
|
|
step_major = step_major |
|
49
|
|
|
# step_minor = 0.025 |
|
50
|
|
|
|
|
51
|
|
|
ax.set_ylim(y_min, y_max) |
|
52
|
|
|
ticks_major = list(np.arange(y_min, y_max, step_major)) |
|
53
|
|
|
# ticks_minor = list(np.arange(y_min, y_max, step_minor)) |
|
54
|
|
|
|
|
55
|
|
|
ax.yaxis.set_ticks(ticks_major) |
|
56
|
|
|
# ax.yaxis.set_ticks(ticks_minor, minor=True) |
|
57
|
|
|
|
|
58
|
|
|
ax.xaxis.set_label_coords(-0.05, 0.5) |
|
59
|
|
|
ax.get_yaxis().set_label_coords(-0.05, 1.1) |
|
60
|
|
|
|
|
61
|
|
|
# ax.yaxis.grid(which="minor", color="r", linestyle="-", linewidth=2) |
|
62
|
|
|
fig = ax.get_figure() |
|
63
|
|
|
fig.tight_layout() |
|
64
|
|
|
fig.savefig("optimizer_time_" + model_name + ".png") |
|
65
|
|
|
|
|
66
|
|
|
|
|
67
|
|
|
plot_optimizer_time( |
|
68
|
|
|
"sklearn.neighbors.KNeighborsClassifier", |
|
69
|
|
|
y_min=0.50, |
|
70
|
|
|
y_max=4.50001, |
|
71
|
|
|
step_major=0.50, |
|
72
|
|
|
title="KNeighborsClassifier - iris_data", |
|
73
|
|
|
) |
|
74
|
|
|
|
|
75
|
|
|
plot_optimizer_time( |
|
76
|
|
|
"sklearn.ensemble.GradientBoostingClassifier", |
|
77
|
|
|
y_min=0.50, |
|
78
|
|
|
y_max=1.50001, |
|
79
|
|
|
step_major=0.25, |
|
80
|
|
|
title="GradientBoostingClassifier - cancer_data", |
|
81
|
|
|
) |
|
82
|
|
|
|
|
83
|
|
|
plot_optimizer_time( |
|
84
|
|
|
"sklearn.tree.DecisionTreeClassifier", |
|
85
|
|
|
y_min=0.50, |
|
86
|
|
|
y_max=3.000001, |
|
87
|
|
|
step_major=0.50, |
|
88
|
|
|
title="DecisionTreeClassifier - iris_data", |
|
89
|
|
|
) |
|
90
|
|
|
|
|
91
|
|
|
plot_optimizer_time( |
|
92
|
|
|
"lightgbm.LGBMClassifier", |
|
93
|
|
|
y_min=0.50, |
|
94
|
|
|
y_max=1.500001, |
|
95
|
|
|
step_major=0.25, |
|
96
|
|
|
title="LGBMClassifier - cancer_data", |
|
97
|
|
|
) |
|
98
|
|
|
|
SimonBlanke /
Hyperactive
