FilesExpand file tree

master

/

plotting.py

Copy path

Blame

More file actions

Blame

More file actions

Latest commit

 

History

History

History

150 lines (123 loc) · 7.59 KB

master

/

plotting.py

Copy path

Top

File metadata and controls

• Code
• Blame

150 lines (123 loc) · 7.59 KB

Raw

Copy raw file

Download raw file

Open symbols panel

Edit and raw actions

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

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

# Statistics output for experiments

import os

import argparse

import pandas as pd

import matplotlib.pyplot as plt

import seaborn as sns

# Read provided statistics file

bag_location = os.getenv('BAG_LOCATION')

directory = bag_location + '/CoralRadarEval/'

parser = argparse.ArgumentParser()

parser.add_argument("--filename", "-f", help="csv or text file containing statistics", default="")

args = parser.parse_args()

# Import statistics file

stats_file = directory + args.filename

data = pd.read_csv(stats_file)

# Directory to save figures

filename = args.filename

save_dir = directory + filename.replace(".txt","")

if not os.path.exists(save_dir) :

os.mkdir(save_dir)

# -- Plot 1 (accuracy by method) -- #

# Plot style (seaborn)

sns.set_theme(style="ticks", color_codes=True)

sns.set(style="ticks")

colors = ["#E59866", "#D35400", "#873600", "#85C1E9", "#3498DB", "#2874A6"] #Daniel

color_figures = ["#E59866", "#D35400", "#873600", "#85C1E9", "#3498DB", "#2874A6","#FF90FA", "#FF00F3", "#8F0088", "#94FF76","#39FF00","#26AB00"]

marker_figures = ['o','o','o','o','o','o','o','o','o','o','o','o']

#colors = ["#E59866","#2874A6", "#D35400", "#85C1E9", "#3498DB", "#873600"] #Mine

data = data.rename(columns={'range error': 'Range error (m)'})

data = data.rename(columns={'scan spacing distance': 'Scan spacing distance (m)'})

data = data.rename(columns={'accuracy': 'Accuracy'})

data = data.rename(columns={'method': 'Method'})

data = data.rename(columns={'auc': 'AuC'})

pal = sns.color_palette(colors)

mar = ['o','o','o','s','s','s']

lin = ['-','--',':','-','--',':']

# Plot data

#data['method'] = data.apply(lambda row: row['method'].replace('method', ''), axis=1)

#Lineplot (commented)

#sns_plot = sns.pointplot(x = 'Range error (m)', y = 'Accuracy', data = data, hue = 'Method', dodge = False, palette=pal)

#sns_plot.locator_params(axis='x', nbins=5)

# ----- Uncertainty plot (range error)

#sns_plot = sns.catplot(x='Range error (m)', y='Accuracy', hue='Method', kind="box", dodge=True, data=data, legend_out=True) # Old

method_scan = 'Method and scan spacing (m)'

data[method_scan] = data.apply(lambda row: row['Method'] + ', ' + str(row['Scan spacing distance (m)']), axis=1)

data_scan = data[ (data['Scan spacing distance (m)']==0) | (data['Scan spacing distance (m)']==5) | (data['Scan spacing distance (m)']==10) ]

data_scan_1 = data_scan[ data_scan['Method']=='P2P' ] ; data_scan_1 = data_scan_1.sort_values('Scan spacing distance (m)')

data_scan_2 = data_scan[ data_scan['Method']=='P2D' ] ; data_scan_2 = data_scan_2.sort_values('Scan spacing distance (m)')

data_scan_3 = data_scan[ data_scan['Method']=='P2L' ] ; data_scan_3 = data_scan_3.sort_values('Scan spacing distance (m)')

data_scan_4 = data_scan[ data_scan['Method']=='Coral'] ; data_scan_4 = data_scan_4.sort_values('Scan spacing distance (m)')

sns_plot = sns.lineplot(x="Range error (m)", y="Accuracy", hue=method_scan, style=method_scan, data=data_scan_1, markers=['.','o','s'], dashes=False, palette=["#E59866", "#D35400", "#873600"]) # Current

sns_plot = sns.lineplot(x="Range error (m)", y="Accuracy", hue=method_scan, style=method_scan, data=data_scan_2, markers=['.','o','s'], dashes=False, palette=["#85C1E9", "#3498DB", "#2874A6"])

sns_plot = sns.lineplot(x="Range error (m)", y="Accuracy", hue=method_scan, style=method_scan, data=data_scan_3, markers=['.','o','s'], dashes=False, palette=["#FF90FA", "#FF00F3", "#8F0088"])

sns_plot = sns.lineplot(x="Range error (m)", y="Accuracy", hue=method_scan, style=method_scan, data=data_scan_4, markers=['.','o','s'], dashes=False, palette=["#94FF76","#39FF00","#26AB00"])

#plt.ylim(0.45,1)

#plt.xlim(0.05,0.95)

plt.grid()

plt.show()

#fig = sns_plot.get_figure()

#fig.savefig(save_dir + '/accuracyByMethod.pdf', format='pdf')

# ----- Uncertainty plot (scan spacing)

#sns_plot = sns.catplot(x='Range error (m)', y='Accuracy', hue='Method', kind="box", dodge=False, data=data, legend_out=True) # Old

method_range = 'Method and range error (m)'

data[method_range] = data.apply(lambda row: row['Method'] + ', ' + str(row['Range error (m)']), axis=1)

data_spacing = data[ (data['Range error (m)']==0.3) | (data['Range error (m)']==0.6) | (data['Range error (m)']==0.9)]

data_spacing_1 = data_spacing[data_spacing['Method']=='P2P']

data_spacing_2 = data_spacing[data_spacing['Method']=='P2D']

data_spacing_3 = data_spacing[data_spacing['Method']=='P2L']

data_spacing_4 = data_spacing[data_spacing['Method']=='Coral']

sns_plot = sns.lineplot(x="Scan spacing distance (m)", y="Accuracy", hue=method_range, style=method_range, data=data_spacing_1, markers=['.','o','s'], dashes=False, palette=["#E59866", "#D35400", "#873600"]) # Current

sns_plot = sns.lineplot(x="Scan spacing distance (m)", y="Accuracy", hue=method_range, style=method_range, data=data_spacing_2, markers=['.','o','s'], dashes=False, palette=["#85C1E9", "#3498DB", "#2874A6"])

sns_plot = sns.lineplot(x="Scan spacing distance (m)", y="Accuracy", hue=method_range, style=method_range, data=data_spacing_3, markers=['.','o','s'], dashes=False, palette=["#FF90FA", "#FF00F3", "#8F0088"])

sns_plot = sns.lineplot(x="Scan spacing distance (m)", y="Accuracy", hue=method_range, style=method_range, data=data_spacing_4, markers=['.','o','s'], dashes=False, palette=["#94FF76","#39FF00","#26AB00"])

#sns_plot = sns.lineplot(x="scan spacing distance", y="Accuracy", hue="Method", style="Method", data= data[data["Range error (m)"]==0.5], markers=['s','s','s'], dashes=False, palette=["#85C1E9", "#3498DB", "#2874A6"]) # Current

#sns_plot = sns.lineplot(x="scan spacing distance", y="Accuracy", hue="Method", style="Method", data= data[data["Range error (m)"]==0.9], markers=['^','^','^'], dashes=False, palette=["#E59866", "#D35400", "#873600"]) # Current

#plt.ylim(0.45,1)

#plt.xlim(0.05,0.95)

plt.grid()

plt.show()

#fig = sns_plot.get_figure()

#fig.savefig(save_dir + '/accuracyByMethod.pdf', format='pdf')

# Print parameters for 'best' configuration for each method

#data_p2p = data[data['Method'] == 'P2P']

#data_p2d = data[data['Method'] == 'P2D']

#data_p2l = data[data['Method'] == 'P2L']

#print('Best configurations for all methods: ')

#print(data_p2p[data_p2p.Accuracy == data_p2p.Accuracy.max()]) ; print(' ')

#print(data_p2d[data_p2d.Accuracy == data_p2d.Accuracy.max()]) ; print(' ')

#print(data_p2l[data_p2l.Accuracy == data_p2l.Accuracy.max()]) ; print(' ')

# Box plot (accuracies and AuC for all methods)

#sns_plot_2 = sns.boxplot(x=data['Method'], y=data['Accuracy'], width=0.3)

#plt.show()

#fig = sns_plot_2.get_figure()

#fig.savefig(save_dir + '/accuracyBoxplot.pdf', format='pdf')

#sns_plot_3 = sns.boxplot(x=data['Method'], y=data['AuC'], width=0.3)

#plt.show()

#fig = sns_plot_3.get_figure()

#fig.savefig(save_dir + '/aucBoxplot.pdf', format='pdf')

# --- OLD CODE (standalone matplot lib, no seaborn) --- #

# Filter data by method

# data_p2p = data[data['method'] == 'P2P']

# data_p2d = data[data['method'] == 'P2D']

# data_p2l = data[data['method'] == 'P2L']

# # Get data

# x_p2p = data_p2p['range error'] ; y_p2p = data_p2p['accuracy']

# x_p2d = data_p2d['range error'] ; y_p2d = data_p2d['accuracy']

# x_p2l = data_p2l['range error'] ; y_p2l = data_p2l['accuracy']

# # Plot data

# plt.plot(x_p2p,y_p2p,marker='o',label='P2P')

# plt.plot(x_p2d,y_p2d,marker='o',label='P2D')

# plt.plot(x_p2l,y_p2l,marker='o',label='P2L')

# plt.legend(loc="upper left")

# plt.xlabel('Range error (m)')

# plt.ylabel('Classification accuracy')

# plt.savefig(save_dir + '/accuracyByMethod.pdf')

# plt.show()

# # -- Plot 2 (boxplot by method) -- #

# accuracies = [y_p2p, y_p2d, y_p2l]

# plt.boxplot(accuracies)

# plt.xticks([1, 2, 3], ['P2P','P2D','P2L'])

# plt.ylabel('Classification accuracy')

# plt.savefig(save_dir + '/boxplotByMethod.pdf')

# plt.show()