-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathParent_student_training_output.log
206 lines (174 loc) · 5.66 KB
/
Parent_student_training_output.log
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
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
Files already downloaded and verified
Files already downloaded and verified
Training Child Model:
Epoch 0/7
----------
Loss after 12768 examples: 0.621
Loss after 25568 examples: 0.319
Loss after 38368 examples: 0.503
Train Accuracy tensor(0.8406, dtype=torch.float64)
Validation Loss is 0.2870355574965477
Validation Accuracy is 0.9058
One of the best validation accuracy found.
Epoch 1/7
----------
Loss after 51152 examples: 0.168
Loss after 63952 examples: 0.192
Loss after 76752 examples: 0.220
Loss after 89552 examples: 0.288
Train Accuracy tensor(0.9036, dtype=torch.float64)
Validation Loss is 0.2658479652166367
Validation Accuracy is 0.9099
One of the best validation accuracy found.
Epoch 2/7
----------
Loss after 102336 examples: 0.385
Loss after 115136 examples: 0.146
Loss after 127936 examples: 0.177
Loss after 140736 examples: 0.281
Train Accuracy tensor(0.9232, dtype=torch.float64)
Validation Loss is 0.24817108023762702
Validation Accuracy is 0.9175000000000001
One of the best validation accuracy found.
Epoch 3/7
----------
Loss after 153520 examples: 0.133
Loss after 166320 examples: 0.245
Loss after 179120 examples: 0.336
Loss after 191920 examples: 0.152
Train Accuracy tensor(0.9380, dtype=torch.float64)
Validation Loss is 0.2414812684237957
Validation Accuracy is 0.9221
One of the best validation accuracy found.
Epoch 4/7
----------
Loss after 204704 examples: 0.124
Loss after 217504 examples: 0.054
Loss after 230304 examples: 0.184
Loss after 243104 examples: 0.149
Train Accuracy tensor(0.9473, dtype=torch.float64)
Validation Loss is 0.2919906246095896
Validation Accuracy is 0.9116000000000001
Epoch 5/7
----------
Loss after 255888 examples: 0.013
Loss after 268688 examples: 0.231
Loss after 281488 examples: 0.074
Loss after 294288 examples: 0.252
Train Accuracy tensor(0.9542, dtype=torch.float64)
Validation Loss is 0.24370603072941302
Validation Accuracy is 0.9205000000000001
Epoch 6/7
----------
Loss after 307072 examples: 0.055
Loss after 319872 examples: 0.100
Loss after 332672 examples: 0.281
Loss after 345472 examples: 0.035
Train Accuracy tensor(0.9618, dtype=torch.float64)
Validation Loss is 0.26391933185085653
Validation Accuracy is 0.9188000000000001
Epoch 7/7
----------
Loss after 358256 examples: 0.094
Loss after 371056 examples: 0.036
Loss after 383856 examples: 0.037
Loss after 396656 examples: 0.061
Train Accuracy tensor(0.9644, dtype=torch.float64)
Validation Loss is 0.29729179718494414
Validation Accuracy is 0.9142
Training complete in 15m 49s
Test Loss is 0.29729179772920905
Test Accuracy is 0.9142
AUROC:
0.9957011777777778
precision recall f1-score support
0 0.92 0.93 0.92 1000
1 0.97 0.95 0.96 1000
2 0.94 0.84 0.89 1000
3 0.82 0.86 0.84 1000
4 0.92 0.90 0.91 1000
5 0.85 0.89 0.87 1000
6 0.89 0.95 0.92 1000
7 0.98 0.90 0.94 1000
8 0.92 0.97 0.95 1000
9 0.95 0.94 0.95 1000
accuracy 0.91 10000
macro avg 0.92 0.91 0.91 10000
weighted avg 0.92 0.91 0.91 10000
Training Parent Model:
Epoch 0/7
----------
Loss after 12768 examples: 0.157
Loss after 25568 examples: 0.312
Loss after 38368 examples: 0.094
Train Accuracy tensor(0.8899, dtype=torch.float64)
Validation Loss is 0.16581539884209634
Validation Accuracy is 0.9442
One of the best validation accuracy found.
Epoch 1/7
----------
Loss after 51152 examples: 0.269
Loss after 63952 examples: 0.087
Loss after 76752 examples: 0.225
Loss after 89552 examples: 0.065
Train Accuracy tensor(0.9512, dtype=torch.float64)
Validation Loss is 0.1665045472741127
Validation Accuracy is 0.9470000000000001
One of the best validation accuracy found.
Epoch 2/7
----------
Loss after 102336 examples: 0.228
Loss after 115136 examples: 0.023
Loss after 127936 examples: 0.155
Loss after 140736 examples: 0.195
Train Accuracy tensor(0.9654, dtype=torch.float64)
Validation Loss is 0.17571113832835109
Validation Accuracy is 0.9493
One of the best validation accuracy found.
Epoch 3/7
----------
Loss after 153520 examples: 0.032
Loss after 166320 examples: 0.033
Loss after 179120 examples: 0.173
Loss after 191920 examples: 0.222
Train Accuracy tensor(0.9727, dtype=torch.float64)
Validation Loss is 0.16448353448258712
Validation Accuracy is 0.9524
One of the best validation accuracy found.
Epoch 4/7
----------
Loss after 204704 examples: 0.097
Loss after 217504 examples: 0.092
Loss after 230304 examples: 0.116
Loss after 243104 examples: 0.006
Train Accuracy tensor(0.9764, dtype=torch.float64)
Validation Loss is 0.19640523077249528
Validation Accuracy is 0.9456
Epoch 5/7
----------
Loss after 255888 examples: 0.004
Loss after 268688 examples: 0.113
Loss after 281488 examples: 0.002
Loss after 294288 examples: 0.021
Train Accuracy tensor(0.9792, dtype=torch.float64)
Validation Loss is 0.19950170327818487
Validation Accuracy is 0.9446
Epoch 6/7
----------
Loss after 307072 examples: 0.023
Loss after 319872 examples: 0.126
Loss after 332672 examples: 0.228
Loss after 345472 examples: 0.280
Train Accuracy tensor(0.9822, dtype=torch.float64)
Validation Loss is 0.23222624748442322
Validation Accuracy is 0.9395
Epoch 7/7
----------
Loss after 358256 examples: 0.028
Loss after 371056 examples: 0.007
Loss after 383856 examples: 0.100
Loss after 396656 examples: 0.059
Train Accuracy tensor(0.9833, dtype=torch.float64)
Validation Loss is 0.1818445452242624
Validation Accuracy is 0.9496
Training complete in 40m 11s