-
Notifications
You must be signed in to change notification settings - Fork 68
/
regression_test.go
188 lines (170 loc) · 4.5 KB
/
regression_test.go
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
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
package regression
import (
"fmt"
"math"
"testing"
)
func TestRun(t *testing.T) {
r := new(Regression)
r.SetObserved("Murders per annum per 1,000,000 inhabitants")
r.SetVar(0, "Inhabitants")
r.SetVar(1, "Percent with incomes below $5000")
r.SetVar(2, "Percent unemployed")
r.Train(
DataPoint(11.2, []float64{587000, 16.5, 6.2}),
DataPoint(13.4, []float64{643000, 20.5, 6.4}),
DataPoint(40.7, []float64{635000, 26.3, 9.3}),
DataPoint(5.3, []float64{692000, 16.5, 5.3}),
DataPoint(24.8, []float64{1248000, 19.2, 7.3}),
DataPoint(12.7, []float64{643000, 16.5, 5.9}),
DataPoint(20.9, []float64{1964000, 20.2, 6.4}),
DataPoint(35.7, []float64{1531000, 21.3, 7.6}),
DataPoint(8.7, []float64{713000, 17.2, 4.9}),
DataPoint(9.6, []float64{749000, 14.3, 6.4}),
DataPoint(14.5, []float64{7895000, 18.1, 6}),
DataPoint(26.9, []float64{762000, 23.1, 7.4}),
DataPoint(15.7, []float64{2793000, 19.1, 5.8}),
DataPoint(36.2, []float64{741000, 24.7, 8.6}),
DataPoint(18.1, []float64{625000, 18.6, 6.5}),
DataPoint(28.9, []float64{854000, 24.9, 8.3}),
DataPoint(14.9, []float64{716000, 17.9, 6.7}),
DataPoint(25.8, []float64{921000, 22.4, 8.6}),
DataPoint(21.7, []float64{595000, 20.2, 8.4}),
DataPoint(25.7, []float64{3353000, 16.9, 6.7}),
)
r.Run()
fmt.Printf("Regression formula:\n%v\n", r.Formula)
fmt.Printf("Regression:\n%s\n", r)
// All vars are known to positively correlate with the murder rate
for i, c := range r.coeff {
if i == 0 {
// This is the offset and not a coeff
continue
}
if c < 0 {
t.Errorf("Coefficient is negative, but shouldn't be: %.2f", c)
}
}
// We know this set has an R^2 above 80
if r.R2 < 0.8 {
t.Errorf("R^2 was %.2f, but we expected > 80", r.R2)
}
}
func TestCrossApply(t *testing.T) {
r := new(Regression)
r.SetObserved("Input-Squared plus Input")
r.SetVar(0, "Input")
r.Train(
DataPoint(6, []float64{2}),
DataPoint(20, []float64{4}),
DataPoint(30, []float64{5}),
DataPoint(72, []float64{8}),
DataPoint(156, []float64{12}),
)
r.AddCross(PowCross(0, 2))
r.AddCross(PowCross(0, 7))
err := r.Run()
if err != nil {
t.Error(err)
}
fmt.Printf("Regression formula:\n%v\n", r.Formula)
fmt.Printf("Regression:\n%s\n", r)
if r.names.vars[1] != "(Input)^2" {
t.Error("Name incorrect")
}
for i, c := range r.coeff {
if i == 0 {
// This is the offset and not a coeff
continue
}
if c < 0 {
t.Errorf("Coefficient is negative, but shouldn't be: %.2f", c)
}
}
// We know this set has an R^2 above 80
if r.R2 < 0.8 {
t.Errorf("R^2 was %.2f, but we expected > 80", r.R2)
}
// Test that predict uses the cross as well
val, err := r.Predict([]float64{6})
if err != nil {
t.Error(err)
}
if val <= 41.999 && val >= 42.001 {
t.Errorf("Expected 42, got %.2f", val)
}
}
func TestMakeDataPoints(t *testing.T) {
a := [][]float64{
{1, 2, 3, 4},
{2, 2, 3, 4},
{3, 2, 3, 4},
}
correct := []float64{2, 3, 4}
dps := MakeDataPoints(a, 0)
for i, dp := range dps {
for i, v := range dp.Variables {
if correct[i] != v {
t.Errorf("Expected variables to be %v. Got %v instead", correct, dp.Variables)
}
}
if dp.Observed != float64(i+1) {
t.Error("Expected observed to be the same as the index")
}
}
a = [][]float64{
{1, 2, 3, 4},
{1, 2, 3, 4},
{1, 2, 3, 4},
}
correct = []float64{1, 3, 4}
dps = MakeDataPoints(a, 1)
for _, dp := range dps {
for i, v := range dp.Variables {
if correct[i] != v {
t.Errorf("Expected variables to be %v. Got %v instead", correct, dp.Variables)
}
}
if dp.Observed != 2.0 {
t.Error("Expected observed to be the same as the index")
}
}
correct = []float64{1, 2, 3}
dps = MakeDataPoints(a, 3)
for _, dp := range dps {
for i, v := range dp.Variables {
if correct[i] != v {
t.Errorf("Expected variables to be %v. Got %v instead", correct, dp.Variables)
}
}
if dp.Observed != 4.0 {
t.Error("Expected observed to be the same as the index")
}
}
}
func TestGetCoeffs(t *testing.T) {
a := [][]float64{
{651, 1, 23},
{762, 2, 26},
{856, 3, 30},
{1063, 4, 34},
{1190, 5, 43},
{1298, 6, 48},
{1421, 7, 52},
{1440, 8, 57},
{1518, 9, 58},
}
r := new(Regression)
r.Train(MakeDataPoints(a, 0)...)
r.Run()
coeffs := r.GetCoeffs()
if len(coeffs) != 3 {
t.Errorf("Expected 3 coefficients. Got %v instead", len(coeffs))
}
expected := []float64{323.54, 46.60, 13.99}
for i := range expected {
if math.Abs(expected[i]-coeffs[i]) > 0.01 {
t.Errorf("Expected coefficient %v to be %v. Got %v instead", i, expected[i], coeffs[i])
}
}
}