-
Notifications
You must be signed in to change notification settings - Fork 1
/
benchmarkoutput.txt
90 lines (90 loc) · 4 KB
/
benchmarkoutput.txt
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
(60000,)
(10000,)
(60000, 28, 28)
(10000, 28, 28)
(60000, 28, 28)
(10000, 28, 28)
(60000, 28, 28)
(10000, 28, 28)
(60000, 28, 28)
(10000, 28, 28)
(60000, 28, 28)
(10000, 28, 28)
(60000, 28, 28)
(10000, 28, 28)
Train on 60000 samples, validate on 10000 samples
Epoch 1/5
2018-04-19 12:01:30.619801: I T:\src\github\tensorflow\tensorflow\core\platform\cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
60000/60000 [==============================] - 259s 4ms/step - loss: 0.3910 - acc: 0.8773 - val_loss: 0.0928 - val_acc: 0.9696
Epoch 2/5
60000/60000 [==============================] - 259s 4ms/step - loss: 0.1076 - acc: 0.9679 - val_loss: 0.0515 - val_acc: 0.9835
Epoch 3/5
60000/60000 [==============================] - 259s 4ms/step - loss: 0.0786 - acc: 0.9760 - val_loss: 0.0398 - val_acc: 0.9868
Epoch 4/5
60000/60000 [==============================] - 258s 4ms/step - loss: 0.0655 - acc: 0.9799 - val_loss: 0.0333 - val_acc: 0.9891
Epoch 5/5
60000/60000 [==============================] - 259s 4ms/step - loss: 0.0563 - acc: 0.9827 - val_loss: 0.0383 - val_acc: 0.9867
Test loss: 0.03825525404292857
Test accuracy: 0.9867
Train on 60000 samples, validate on 10000 samples
Epoch 1/5
60000/60000 [==============================] - 262s 4ms/step - loss: 0.3231 - acc: 0.9077 - val_loss: 0.6596 - val_acc: 0.7954
Epoch 2/5
60000/60000 [==============================] - 260s 4ms/step - loss: 0.1104 - acc: 0.9675 - val_loss: 1.2605 - val_acc: 0.6648
Epoch 3/5
60000/60000 [==============================] - 261s 4ms/step - loss: 0.0868 - acc: 0.9737 - val_loss: 1.5547 - val_acc: 0.5784
Epoch 4/5
60000/60000 [==============================] - 260s 4ms/step - loss: 0.0728 - acc: 0.9786 - val_loss: 2.3885 - val_acc: 0.4907
Epoch 5/5
60000/60000 [==============================] - 260s 4ms/step - loss: 0.0622 - acc: 0.9814 - val_loss: 2.1992 - val_acc: 0.5064
Original data set
Test loss: 2.199169082069397
Test accuracy: 0.5064
Second data set
Test loss: 0.034937220360540956
Test accuracy: 0.9881
Train on 60000 samples, validate on 10000 samples
Epoch 1/5
60000/60000 [==============================] - 260s 4ms/step - loss: 0.2074 - acc: 0.9400 - val_loss: 0.6088 - val_acc: 0.7797
Epoch 2/5
60000/60000 [==============================] - 259s 4ms/step - loss: 0.0829 - acc: 0.9749 - val_loss: 0.3715 - val_acc: 0.8732
Epoch 3/5
60000/60000 [==============================] - 259s 4ms/step - loss: 0.0640 - acc: 0.9803 - val_loss: 0.6067 - val_acc: 0.7986
Epoch 4/5
60000/60000 [==============================] - 259s 4ms/step - loss: 0.0551 - acc: 0.9836 - val_loss: 0.8112 - val_acc: 0.7491
Epoch 5/5
60000/60000 [==============================] - 263s 4ms/step - loss: 0.0472 - acc: 0.9853 - val_loss: 0.8369 - val_acc: 0.7524
Original data set
Test loss: 0.8368887896299362
Test accuracy: 0.7524
Second data set
Test loss: 3.361742441177368
Test accuracy: 0.3679
Third data set
Test loss: 0.028098958297784703
Test accuracy: 0.9905
Train on 60000 samples, validate on 10000 samples
Epoch 1/5
60000/60000 [==============================] - 260s 4ms/step - loss: 0.3193 - acc: 0.9083 - val_loss: 2.2299 - val_acc: 0.3726
Epoch 2/5
60000/60000 [==============================] - 310s 5ms/step - loss: 0.1178 - acc: 0.9658 - val_loss: 2.6773 - val_acc: 0.3289
Epoch 3/5
60000/60000 [==============================] - 346s 6ms/step - loss: 0.0885 - acc: 0.9740 - val_loss: 3.8655 - val_acc: 0.2255
Epoch 4/5
60000/60000 [==============================] - 341s 6ms/step - loss: 0.0721 - acc: 0.9785 - val_loss: 3.8812 - val_acc: 0.2398
Epoch 5/5
60000/60000 [==============================] - 387s 6ms/step - loss: 0.0640 - acc: 0.9808 - val_loss: 4.6580 - val_acc: 0.2152
-----------------------------
Original data set
Test loss: 4.658011458587646
Test accuracy: 0.2152
Second data set
Test loss: 5.044480073547363
Test accuracy: 0.2149
Third data set
Test loss: 3.389402949142456
Test accuracy: 0.468
Fourth data set
Test loss: 0.03599673119182698
Test accuracy: 0.9884
Train on 60000 samples, validate on 10000 samples