PuzzleLib/Statistics.py at master · EmilPi/PuzzleLib · GitHub

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
import numpy as np


def confusion(labels, predictions, dim=0, log=True):
	if dim <= 0:
		for label in labels:
			if int(label) >= dim:
				dim = int(label)+1

		for label in predictions:
			if int(label) >= dim:
				dim = int(label) + 1

	cm = [[0] * dim for _ in range(dim)]

	for i in range(min(len(labels), len(predictions))):
		cm[int(labels[i])][int(predictions[i])] += 1

	if log:
		print("Confusion Matrix:")

		for mst in cm:
			print(str(mst))

	return cm


def precision(cm, log=True, verbose=True):
	dim = len(cm)
	prs, pr = [], 0

	for i in range(dim):
		summary = 0

		for j in range(dim):
			summary += cm[j][i]

		if summary == 0:
			tpr = 1.0
		else:
			tpr = cm[i][i] / summary

		pr += tpr
		prs.append(tpr)

		if log and verbose:
			print("Precision on class %s: %s" % (i, tpr))

	pr /= dim

	if log:
		print("Precision mean: %s" % pr)

	return pr, prs


def recall(cm, log=True, verbose=True):
	dim = len(cm)
	rcs, rc = [], 0

	for i in range(dim):
		summary = 0

		for j in range(dim):
			summary += cm[i][j]

		if summary == 0:
			trc = 1.0
		else:
			trc = cm[i][i] / summary

		rc += trc
		rcs.append(trc)

		if log and verbose:
			print("Recall on class %d: %f" % (i, trc))

	rc /= dim

	if log:
		print("Recall mean: %s" % rc)

	return rc, rcs


def accuracy(cm, log=True):
	dim = len(cm)
	acc, summary = 0, 0

	for i in range(dim):
		for j in range(dim):
			summary += cm[i][j]

		acc += cm[i][i]

	acc /= summary

	if log:
		print("Accuracy: %s" % acc)

	return acc


def fullstats(labels, predictions, dim=0, printing=True, verbose=True):
	cm = confusion(labels, predictions, dim, printing)
	pr, prs = precision(cm, printing, verbose)
	rc, rcs = recall(cm, printing, verbose)

	return cm, pr, rc, prs, rcs


def unittest():
	cm = np.ones(shape=(1000, 1000), dtype=np.float32)

	assert np.isclose(precision(cm, verbose=False)[0], 0.001)
	assert np.isclose(recall(cm, verbose=False)[0], 0.001)
	assert np.isclose(accuracy(cm), 0.001)

	import random
	labels = [random.randint(0, 5) for _ in range(10000)]
	results = [random.randint(0, 5) for _ in range(10000)]

	cm, pr, rc, prs, rcs = fullstats(labels, results, verbose=False)

	labels = np.array(labels)
	results = np.array(results)

	npcm, nppr, nprc, npprs, nprcs = fullstats(labels, results, verbose=False)

	assert np.allclose(np.array(cm), npcm)
	assert np.allclose(np.array(pr), nppr)
	assert np.allclose(np.array(rc), nprc)
	assert np.allclose(np.array(prs), npprs)
	assert np.allclose(np.array(rcs), nprcs)


if __name__ == "__main__":
	unittest()