-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathplot.py
159 lines (129 loc) · 4.73 KB
/
plot.py
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
151
152
153
154
155
156
157
158
159
import matplotlib.pyplot as plt
import seaborn
import pandas as pd
import numpy as np
import os
# Load in results
def import_data(path, debug=False):
dfs = []
for i in range(5):
df = pd.read_csv(f"{path}-fold{i}/results.csv")
df['fold'] = [i for j in range(100)]
dfs.append(df)
data = pd.concat(dfs)
data.rename(columns=(lambda x: x.strip()), inplace=True)
if debug:
data.to_excel(path[path.rfind('/') + 1:] + ".xlsx")
return data
ball_fold_results = import_data("runs/detect/ball-tracker-10k")
event_fold_results = import_data("runs/classify/event-tracker")
if not os.path.exists("plots"):
os.makedirs("plots")
# General results for ball model folds
# means: If true, plots an average and standard deviation for all folds instead of each fold independently
def plot_ball_results(means):
extra = dict()
if not means:
extra = {"hue": "fold", "palette": "deep"}
plt.figure(figsize=(15,6))
plt.subplot(2,5,1)
seaborn.lineplot(data=ball_fold_results, x="epoch", y="train/box_loss", **extra)
plt.title("train/box_loss")
plt.ylabel("")
plt.subplot(2,5,2)
seaborn.lineplot(data=ball_fold_results, x="epoch", y="train/cls_loss", **extra)
plt.title("train/cls_loss")
plt.ylabel("")
plt.subplot(2,5,3)
seaborn.lineplot(data=ball_fold_results, x="epoch", y="train/dfl_loss", **extra)
plt.title("train/dfl_loss")
plt.ylabel("")
plt.subplot(2,5,4)
seaborn.lineplot(data=ball_fold_results, x="epoch", y="metrics/precision(B)", **extra)
plt.title("metrics/precision(B)")
plt.ylabel("")
plt.subplot(2,5,5)
seaborn.lineplot(data=ball_fold_results, x="epoch", y="metrics/recall(B)", **extra)
plt.title("metrics/recall(B)")
plt.ylabel("")
plt.subplot(2,5,6)
seaborn.lineplot(data=ball_fold_results, x="epoch", y="val/box_loss", **extra)
plt.title("val/box_loss")
plt.ylabel("")
plt.subplot(2,5,7)
seaborn.lineplot(data=ball_fold_results, x="epoch", y="val/cls_loss", **extra)
plt.title("val/cls_loss")
plt.ylabel("")
plt.subplot(2,5,8)
seaborn.lineplot(data=ball_fold_results, x="epoch", y="val/dfl_loss", **extra)
plt.title("val/dfl_loss")
plt.ylabel("")
plt.subplot(2,5,9)
seaborn.lineplot(data=ball_fold_results, x="epoch", y="metrics/mAP50(B)", **extra)
plt.title("metrics/mAP50(B)")
plt.ylabel("")
plt.subplot(2,5,10)
seaborn.lineplot(data=ball_fold_results, x="epoch", y="metrics/mAP50-95(B)", **extra)
plt.title("metrics/mAP50-95(B)")
plt.ylabel("")
plt.tight_layout()
plt.savefig("plots/ball_results_means.png" if means else "plots/ball_results.png")
plt.clf()
plot_ball_results(True)
plot_ball_results(False)
# General results for event model folds
# means: If true, plots an average and standard deviation for all folds instead of each fold independently
def plot_event_results(means):
extra = dict()
if not means:
extra = {"hue": "fold", "palette": "deep"}
plt.figure(figsize=(9,3))
plt.subplot(1,3,1)
seaborn.lineplot(data=event_fold_results, x="epoch", y="train/loss", **extra)
plt.title("train/loss")
plt.ylabel("")
plt.subplot(1,3,2)
seaborn.lineplot(data=event_fold_results, x="epoch", y="val/loss", **extra)
plt.title("val/loss")
plt.ylabel("")
plt.subplot(1,3,3)
seaborn.lineplot(data=event_fold_results, x="epoch", y="metrics/accuracy_top1", **extra)
plt.title("metrics/accuracy_top1")
plt.ylabel("")
plt.tight_layout()
plt.savefig("plots/event_results_means.png" if means else "plots/event_results.png")
plt.clf()
plot_event_results(True)
plot_event_results(False)
# Fold selection plots
plt.figure(figsize=(6,6))
plt.subplot(2,2,1)
seaborn.lineplot(data=ball_fold_results, x="epoch", y="val/dfl_loss", hue="fold", palette="deep")
plt.title("Ball Detection Validation Loss")
plt.ylabel("Loss")
plt.ylim([0.42,0.52])
plt.xlabel("Epoch")
plt.subplot(2,2,2)
seaborn.lineplot(data=ball_fold_results, x="epoch", y="metrics/recall(B)", hue="fold", palette="deep")
plt.title("Ball Detection Recall")
plt.ylabel("Recall")
plt.ylim([0.94,0.98])
plt.xlabel("Epoch")
plt.legend('',frameon=False)
plt.subplot(2,2,3)
seaborn.lineplot(data=event_fold_results, x="epoch", y="val/loss", hue="fold", palette="deep")
plt.title("Event Classification Validation Loss")
plt.ylabel("Loss")
plt.ylim([0.28, 0.32])
plt.xlabel("Epoch")
plt.legend('',frameon=False)
plt.subplot(2,2,4)
seaborn.lineplot(data=event_fold_results, x="epoch", y="metrics/accuracy_top1", hue="fold", palette="deep")
plt.title("Event Classification Accuracy")
plt.ylabel("Accuracy")
plt.ylim([0.92,1.0])
plt.xlabel("Epoch")
plt.legend('',frameon=False)
plt.tight_layout()
plt.savefig('plots/fold_analysis.png')
plt.clf()