机器学习-恒星星系分类

sterne 发布于

姑且存一下训练结果，后面应该还会改，比如20个epoch应该不太够

1

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

Epoch 1/20
1038/1038 [==============================] - 22s 21ms/step - loss: 0.2031 - sparse_categorical_accuracy: 0.9348 - val_loss: 0.1866 - val_sparse_categorical_accuracy: 0.9469
Epoch 2/20
1038/1038 [==============================] - 25s 24ms/step - loss: 0.1375 - sparse_categorical_accuracy: 0.9558 - val_loss: 0.1296 - val_sparse_categorical_accuracy: 0.9620
Epoch 3/20
1038/1038 [==============================] - 24s 23ms/step - loss: 0.1155 - sparse_categorical_accuracy: 0.9632 - val_loss: 0.1259 - val_sparse_categorical_accuracy: 0.9599
Epoch 4/20
1038/1038 [==============================] - 23s 22ms/step - loss: 0.1001 - sparse_categorical_accuracy: 0.9670 - val_loss: 0.1430 - val_sparse_categorical_accuracy: 0.9612
Epoch 5/20
1038/1038 [==============================] - 23s 22ms/step - loss: 0.0953 - sparse_categorical_accuracy: 0.9686 - val_loss: 0.1109 - val_sparse_categorical_accuracy: 0.9637
Epoch 6/20
1038/1038 [==============================] - 23s 22ms/step - loss: 0.0877 - sparse_categorical_accuracy: 0.9707 - val_loss: 0.1097 - val_sparse_categorical_accuracy: 0.9648
Epoch 7/20
1038/1038 [==============================] - 23s 22ms/step - loss: 0.0796 - sparse_categorical_accuracy: 0.9744 - val_loss: 0.1058 - val_sparse_categorical_accuracy: 0.9680
Epoch 8/20
1038/1038 [==============================] - 23s 22ms/step - loss: 0.0790 - sparse_categorical_accuracy: 0.9748 - val_loss: 0.0977 - val_sparse_categorical_accuracy: 0.9707
Epoch 9/20
1038/1038 [==============================] - 23s 23ms/step - loss: 0.0696 - sparse_categorical_accuracy: 0.9772 - val_loss: 0.1074 - val_sparse_categorical_accuracy: 0.9672
Epoch 10/20
1038/1038 [==============================] - 23s 22ms/step - loss: 0.0654 - sparse_categorical_accuracy: 0.9793 - val_loss: 0.0979 - val_sparse_categorical_accuracy: 0.9707
Epoch 11/20
1038/1038 [==============================] - 23s 22ms/step - loss: 0.0658 - sparse_categorical_accuracy: 0.9783 - val_loss: 0.1075 - val_sparse_categorical_accuracy: 0.9707
Epoch 12/20
1038/1038 [==============================] - 25s 24ms/step - loss: 0.0604 - sparse_categorical_accuracy: 0.9802 - val_loss: 0.1040 - val_sparse_categorical_accuracy: 0.9680
Epoch 13/20
1038/1038 [==============================] - 25s 24ms/step - loss: 0.0595 - sparse_categorical_accuracy: 0.9802 - val_loss: 0.1135 - val_sparse_categorical_accuracy: 0.9664
Epoch 14/20
1038/1038 [==============================] - 26s 25ms/step - loss: 0.0553 - sparse_categorical_accuracy: 0.9819 - val_loss: 0.1049 - val_sparse_categorical_accuracy: 0.9680
Epoch 15/20
1038/1038 [==============================] - 25s 24ms/step - loss: 0.0546 - sparse_categorical_accuracy: 0.9821 - val_loss: 0.1126 - val_sparse_categorical_accuracy: 0.9715
Epoch 16/20
1038/1038 [==============================] - 25s 24ms/step - loss: 0.0522 - sparse_categorical_accuracy: 0.9825 - val_loss: 0.0917 - val_sparse_categorical_accuracy: 0.9713
Epoch 17/20
1038/1038 [==============================] - 25s 24ms/step - loss: 0.0470 - sparse_categorical_accuracy: 0.9842 - val_loss: 0.1050 - val_sparse_categorical_accuracy: 0.9721
Epoch 18/20
1038/1038 [==============================] - 25s 24ms/step - loss: 0.0444 - sparse_categorical_accuracy: 0.9855 - val_loss: 0.1025 - val_sparse_categorical_accuracy: 0.9686
Epoch 19/20
1038/1038 [==============================] - 25s 24ms/step - loss: 0.0503 - sparse_categorical_accuracy: 0.9830 - val_loss: 0.1110 - val_sparse_categorical_accuracy: 0.9737
Epoch 20/20
1038/1038 [==============================] - 25s 24ms/step - loss: 0.0425 - sparse_categorical_accuracy: 0.9854 - val_loss: 0.1055 - val_sparse_categorical_accuracy: 0.9715
Model: "sequential"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
flatten (Flatten)            (None, 12288)             0         
_________________________________________________________________
dense (Dense)                (None, 64)                786496    
_________________________________________________________________
dense_1 (Dense)              (None, 2)                 130       
=================================================================
Total params: 786,626
Trainable params: 786,626
Non-trainable params: 0
_________________________________________________________________

Process finished with exit code 0

试了一圈还是这么存吧。。。

4

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
model = tf.keras.models.Sequential([
    tf.keras.layers.Flatten(),
    Dense(64, activation='relu'),
    Dropout(0.2),
    Dense(32, activation='relu'),
    Dropout(0.2),
    Dense(2, activation='softmax')
])

D:\pyAnaconda\envs\pythonProject\python.exe D:/Code/pythonProject/work/sdss_reco.py
Epoch 1/20
1038/1038 [==============================] - 23s 21ms/step - loss: 0.2327 - sparse_categorical_accuracy: 0.9268 - val_loss: 0.1805 - val_sparse_categorical_accuracy: 0.9447
Epoch 2/20
1038/1038 [==============================] - 23s 22ms/step - loss: 0.1538 - sparse_categorical_accuracy: 0.9497 - val_loss: 0.1279 - val_sparse_categorical_accuracy: 0.9596
Epoch 3/20
1038/1038 [==============================] - 23s 22ms/step - loss: 0.1367 - sparse_categorical_accuracy: 0.9562 - val_loss: 0.1220 - val_sparse_categorical_accuracy: 0.9607
Epoch 4/20
1038/1038 [==============================] - 26s 25ms/step - loss: 0.1210 - sparse_categorical_accuracy: 0.9613 - val_loss: 0.1104 - val_sparse_categorical_accuracy: 0.9607
Epoch 5/20
1038/1038 [==============================] - 24s 23ms/step - loss: 0.1194 - sparse_categorical_accuracy: 0.9620 - val_loss: 0.1141 - val_sparse_categorical_accuracy: 0.9642
Epoch 6/20
1038/1038 [==============================] - 23s 22ms/step - loss: 0.1131 - sparse_categorical_accuracy: 0.9634 - val_loss: 0.1064 - val_sparse_categorical_accuracy: 0.9637
Epoch 7/20
1038/1038 [==============================] - 24s 23ms/step - loss: 0.1059 - sparse_categorical_accuracy: 0.9657 - val_loss: 0.1019 - val_sparse_categorical_accuracy: 0.9691
Epoch 8/20
1038/1038 [==============================] - 24s 23ms/step - loss: 0.1069 - sparse_categorical_accuracy: 0.9670 - val_loss: 0.1052 - val_sparse_categorical_accuracy: 0.9648
Epoch 9/20
1038/1038 [==============================] - 24s 23ms/step - loss: 0.1003 - sparse_categorical_accuracy: 0.9685 - val_loss: 0.0998 - val_sparse_categorical_accuracy: 0.9680
Epoch 10/20
1038/1038 [==============================] - 24s 23ms/step - loss: 0.0980 - sparse_categorical_accuracy: 0.9694 - val_loss: 0.0965 - val_sparse_categorical_accuracy: 0.9696
Epoch 11/20
1038/1038 [==============================] - 25s 24ms/step - loss: 0.1009 - sparse_categorical_accuracy: 0.9677 - val_loss: 0.0952 - val_sparse_categorical_accuracy: 0.9683
Epoch 12/20
1038/1038 [==============================] - 23s 22ms/step - loss: 0.0943 - sparse_categorical_accuracy: 0.9704 - val_loss: 0.0933 - val_sparse_categorical_accuracy: 0.9710
Epoch 13/20
1038/1038 [==============================] - 24s 23ms/step - loss: 0.0926 - sparse_categorical_accuracy: 0.9707 - val_loss: 0.0978 - val_sparse_categorical_accuracy: 0.9699
Epoch 14/20
1038/1038 [==============================] - 23s 22ms/step - loss: 0.0917 - sparse_categorical_accuracy: 0.9715 - val_loss: 0.0914 - val_sparse_categorical_accuracy: 0.9699
Epoch 15/20
1038/1038 [==============================] - 24s 23ms/step - loss: 0.0861 - sparse_categorical_accuracy: 0.9727 - val_loss: 0.0938 - val_sparse_categorical_accuracy: 0.9713
Epoch 16/20
1038/1038 [==============================] - 24s 23ms/step - loss: 0.0876 - sparse_categorical_accuracy: 0.9725 - val_loss: 0.0984 - val_sparse_categorical_accuracy: 0.9691
Epoch 17/20
1038/1038 [==============================] - 23s 22ms/step - loss: 0.0884 - sparse_categorical_accuracy: 0.9717 - val_loss: 0.0990 - val_sparse_categorical_accuracy: 0.9686
Epoch 18/20
1038/1038 [==============================] - 24s 23ms/step - loss: 0.0832 - sparse_categorical_accuracy: 0.9736 - val_loss: 0.0985 - val_sparse_categorical_accuracy: 0.9686
Epoch 19/20
1038/1038 [==============================] - 23s 23ms/step - loss: 0.0798 - sparse_categorical_accuracy: 0.9748 - val_loss: 0.0892 - val_sparse_categorical_accuracy: 0.9734
Epoch 20/20
1038/1038 [==============================] - 24s 23ms/step - loss: 0.0799 - sparse_categorical_accuracy: 0.9755 - val_loss: 0.0956 - val_sparse_categorical_accuracy: 0.9683
Model: "sequential"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
flatten (Flatten)            (None, 12288)             0         
_________________________________________________________________
dense (Dense)                (None, 64)                786496    
_________________________________________________________________
dropout (Dropout)            (None, 64)                0         
_________________________________________________________________
dense_1 (Dense)              (None, 32)                2080      
_________________________________________________________________
dropout_1 (Dropout)          (None, 32)                0         
_________________________________________________________________
dense_2 (Dense)              (None, 2)                 66        
=================================================================
Total params: 788,642
Trainable params: 788,642
Non-trainable params: 0
_________________________________________________________________

5

训练集接近99，测试集差不多还那样，有点明显过拟合，感觉效果还算不错

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
model = tf.keras.models.Sequential([
    tf.keras.layers.Flatten(),
    Dense(128, activation='relu'),
    Dense(64, activation='relu'),
    Dense(32, activation='relu'),
    Dense(2, activation='softmax')
])
D:\pyAnaconda\envs\pythonProject\python.exe D:/Code/pythonProject/work/sdss_reco.py
Epoch 1/20
1038/1038 [==============================] - 13s 12ms/step - loss: 0.1989 - sparse_categorical_accuracy: 0.9375 - val_loss: 0.1320 - val_sparse_categorical_accuracy: 0.9620
Epoch 2/20
1038/1038 [==============================] - 15s 15ms/step - loss: 0.1232 - sparse_categorical_accuracy: 0.9612 - val_loss: 0.1179 - val_sparse_categorical_accuracy: 0.9669
Epoch 3/20
1038/1038 [==============================] - 9s 8ms/step - loss: 0.1024 - sparse_categorical_accuracy: 0.9665 - val_loss: 0.1009 - val_sparse_categorical_accuracy: 0.9683
Epoch 4/20
1038/1038 [==============================] - 5s 5ms/step - loss: 0.0898 - sparse_categorical_accuracy: 0.9708 - val_loss: 0.1039 - val_sparse_categorical_accuracy: 0.9677
Epoch 5/20
1038/1038 [==============================] - 5s 5ms/step - loss: 0.0798 - sparse_categorical_accuracy: 0.9733 - val_loss: 0.0972 - val_sparse_categorical_accuracy: 0.9680
Epoch 6/20
1038/1038 [==============================] - 5s 5ms/step - loss: 0.0747 - sparse_categorical_accuracy: 0.9748 - val_loss: 0.0818 - val_sparse_categorical_accuracy: 0.9737
Epoch 7/20
1038/1038 [==============================] - 5s 5ms/step - loss: 0.0667 - sparse_categorical_accuracy: 0.9772 - val_loss: 0.0849 - val_sparse_categorical_accuracy: 0.9740
Epoch 8/20
1038/1038 [==============================] - 5s 5ms/step - loss: 0.0611 - sparse_categorical_accuracy: 0.9788 - val_loss: 0.0816 - val_sparse_categorical_accuracy: 0.9764
Epoch 9/20
1038/1038 [==============================] - 5s 5ms/step - loss: 0.0582 - sparse_categorical_accuracy: 0.9808 - val_loss: 0.0859 - val_sparse_categorical_accuracy: 0.9715
Epoch 10/20
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0558 - sparse_categorical_accuracy: 0.9805 - val_loss: 0.1014 - val_sparse_categorical_accuracy: 0.9694
Epoch 11/20
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0506 - sparse_categorical_accuracy: 0.9828 - val_loss: 0.0815 - val_sparse_categorical_accuracy: 0.9745
Epoch 12/20
1038/1038 [==============================] - 5s 5ms/step - loss: 0.0510 - sparse_categorical_accuracy: 0.9824 - val_loss: 0.1146 - val_sparse_categorical_accuracy: 0.9653
Epoch 13/20
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0476 - sparse_categorical_accuracy: 0.9835 - val_loss: 0.0982 - val_sparse_categorical_accuracy: 0.9724
Epoch 14/20
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0433 - sparse_categorical_accuracy: 0.9856 - val_loss: 0.0820 - val_sparse_categorical_accuracy: 0.9770
Epoch 15/20
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0420 - sparse_categorical_accuracy: 0.9852 - val_loss: 0.0855 - val_sparse_categorical_accuracy: 0.9737
Epoch 16/20
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0376 - sparse_categorical_accuracy: 0.9868 - val_loss: 0.1125 - val_sparse_categorical_accuracy: 0.9691
Epoch 17/20
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0431 - sparse_categorical_accuracy: 0.9853 - val_loss: 0.1019 - val_sparse_categorical_accuracy: 0.9729
Epoch 18/20
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0356 - sparse_categorical_accuracy: 0.9877 - val_loss: 0.0895 - val_sparse_categorical_accuracy: 0.9748
Epoch 19/20
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0313 - sparse_categorical_accuracy: 0.9888 - val_loss: 0.0921 - val_sparse_categorical_accuracy: 0.9759
Epoch 20/20
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0294 - sparse_categorical_accuracy: 0.9897 - val_loss: 0.1218 - val_sparse_categorical_accuracy: 0.9694
Model: "sequential"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
flatten (Flatten)            (None, 12288)             0         
_________________________________________________________________
dense (Dense)                (None, 128)               1572992   
_________________________________________________________________
dense_1 (Dense)              (None, 64)                8256      
_________________________________________________________________
dense_2 (Dense)              (None, 32)                2080      
_________________________________________________________________
dense_3 (Dense)              (None, 2)                 66        
=================================================================
Total params: 1,583,394
Trainable params: 1,583,394
Non-trainable params: 0
_________________________________________________________________

6

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
model = tf.keras.models.Sequential([
    tf.keras.layers.Flatten(),
    Dense(128, activation='relu'),
    Dropout(0.2),
    Dense(64, activation='relu'),
    Dense(32, activation='relu'),
    Dense(2, activation='softmax')
])
D:\pyAnaconda\envs\pythonProject\python.exe D:/Code/pythonProject/work/sdss_reco.py
Epoch 1/20
1038/1038 [==============================] - 9s 8ms/step - loss: 0.2111 - sparse_categorical_accuracy: 0.9320 - val_loss: 0.1328 - val_sparse_categorical_accuracy: 0.9599
Epoch 2/20
1038/1038 [==============================] - 8s 8ms/step - loss: 0.1339 - sparse_categorical_accuracy: 0.9560 - val_loss: 0.1196 - val_sparse_categorical_accuracy: 0.9626
Epoch 3/20
1038/1038 [==============================] - 8s 8ms/step - loss: 0.1209 - sparse_categorical_accuracy: 0.9606 - val_loss: 0.1059 - val_sparse_categorical_accuracy: 0.9658
Epoch 4/20
1038/1038 [==============================] - 8s 8ms/step - loss: 0.1100 - sparse_categorical_accuracy: 0.9645 - val_loss: 0.1252 - val_sparse_categorical_accuracy: 0.9599
Epoch 5/20
1038/1038 [==============================] - 7s 7ms/step - loss: 0.1062 - sparse_categorical_accuracy: 0.9652 - val_loss: 0.0966 - val_sparse_categorical_accuracy: 0.9686
Epoch 6/20
1038/1038 [==============================] - 7s 7ms/step - loss: 0.1027 - sparse_categorical_accuracy: 0.9663 - val_loss: 0.1018 - val_sparse_categorical_accuracy: 0.9661
Epoch 7/20
1038/1038 [==============================] - 7s 7ms/step - loss: 0.0965 - sparse_categorical_accuracy: 0.9683 - val_loss: 0.0893 - val_sparse_categorical_accuracy: 0.9696
Epoch 8/20
1038/1038 [==============================] - 7s 7ms/step - loss: 0.0938 - sparse_categorical_accuracy: 0.9699 - val_loss: 0.0873 - val_sparse_categorical_accuracy: 0.9696
Epoch 9/20
1038/1038 [==============================] - 7s 7ms/step - loss: 0.0873 - sparse_categorical_accuracy: 0.9716 - val_loss: 0.0977 - val_sparse_categorical_accuracy: 0.9658
Epoch 10/20
1038/1038 [==============================] - 7s 6ms/step - loss: 0.0869 - sparse_categorical_accuracy: 0.9717 - val_loss: 0.0934 - val_sparse_categorical_accuracy: 0.9680
Epoch 11/20
1038/1038 [==============================] - 8s 7ms/step - loss: 0.0816 - sparse_categorical_accuracy: 0.9737 - val_loss: 0.0905 - val_sparse_categorical_accuracy: 0.9705
Epoch 12/20
1038/1038 [==============================] - 7s 6ms/step - loss: 0.0797 - sparse_categorical_accuracy: 0.9745 - val_loss: 0.0858 - val_sparse_categorical_accuracy: 0.9729
Epoch 13/20
1038/1038 [==============================] - 7s 6ms/step - loss: 0.0783 - sparse_categorical_accuracy: 0.9742 - val_loss: 0.0815 - val_sparse_categorical_accuracy: 0.9729
Epoch 14/20
1038/1038 [==============================] - 7s 6ms/step - loss: 0.0766 - sparse_categorical_accuracy: 0.9748 - val_loss: 0.0921 - val_sparse_categorical_accuracy: 0.9710
Epoch 15/20
1038/1038 [==============================] - 7s 6ms/step - loss: 0.0731 - sparse_categorical_accuracy: 0.9763 - val_loss: 0.0839 - val_sparse_categorical_accuracy: 0.9729
Epoch 16/20
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0732 - sparse_categorical_accuracy: 0.9761 - val_loss: 0.0871 - val_sparse_categorical_accuracy: 0.9713
Epoch 17/20
1038/1038 [==============================] - 7s 6ms/step - loss: 0.0697 - sparse_categorical_accuracy: 0.9775 - val_loss: 0.0992 - val_sparse_categorical_accuracy: 0.9694
Epoch 18/20
1038/1038 [==============================] - 7s 7ms/step - loss: 0.0721 - sparse_categorical_accuracy: 0.9771 - val_loss: 0.0869 - val_sparse_categorical_accuracy: 0.9729
Epoch 19/20
1038/1038 [==============================] - 8s 8ms/step - loss: 0.0690 - sparse_categorical_accuracy: 0.9778 - val_loss: 0.0870 - val_sparse_categorical_accuracy: 0.9748
Epoch 20/20
1038/1038 [==============================] - 7s 7ms/step - loss: 0.0674 - sparse_categorical_accuracy: 0.9780 - val_loss: 0.0876 - val_sparse_categorical_accuracy: 0.9729
Model: "sequential"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
flatten (Flatten)            (None, 12288)             0         
_________________________________________________________________
dense (Dense)                (None, 128)               1572992   
_________________________________________________________________
dropout (Dropout)            (None, 128)               0         
_________________________________________________________________
dense_1 (Dense)              (None, 64)                8256      
_________________________________________________________________
dense_2 (Dense)              (None, 32)                2080      
_________________________________________________________________
dense_3 (Dense)              (None, 2)                 66        
=================================================================
Total params: 1,583,394
Trainable params: 1,583,394
Non-trainable params: 0
_________________________________________________________________

6.1

epoch=50后

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
D:\pyAnaconda\envs\pythonProject\python.exe D:/Code/pythonProject/work/sdss_reco.py
Epoch 1/50
1038/1038 [==============================] - 7s 6ms/step - loss: 0.2069 - sparse_categorical_accuracy: 0.9303 - val_loss: 0.1715 - val_sparse_categorical_accuracy: 0.9466
Epoch 2/50
1038/1038 [==============================] - 8s 7ms/step - loss: 0.1347 - sparse_categorical_accuracy: 0.9559 - val_loss: 0.1272 - val_sparse_categorical_accuracy: 0.9596
Epoch 3/50
1038/1038 [==============================] - 7s 6ms/step - loss: 0.1186 - sparse_categorical_accuracy: 0.9611 - val_loss: 0.1345 - val_sparse_categorical_accuracy: 0.9561
Epoch 4/50
1038/1038 [==============================] - 7s 7ms/step - loss: 0.1087 - sparse_categorical_accuracy: 0.9640 - val_loss: 0.0992 - val_sparse_categorical_accuracy: 0.9683
Epoch 5/50
1038/1038 [==============================] - 7s 6ms/step - loss: 0.0996 - sparse_categorical_accuracy: 0.9675 - val_loss: 0.1021 - val_sparse_categorical_accuracy: 0.9661
Epoch 6/50
1038/1038 [==============================] - 7s 7ms/step - loss: 0.0962 - sparse_categorical_accuracy: 0.9671 - val_loss: 0.0946 - val_sparse_categorical_accuracy: 0.9686
Epoch 7/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0908 - sparse_categorical_accuracy: 0.9708 - val_loss: 0.0886 - val_sparse_categorical_accuracy: 0.9694
Epoch 8/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0874 - sparse_categorical_accuracy: 0.9708 - val_loss: 0.0940 - val_sparse_categorical_accuracy: 0.9669
Epoch 9/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0863 - sparse_categorical_accuracy: 0.9716 - val_loss: 0.0917 - val_sparse_categorical_accuracy: 0.9715
Epoch 10/50
1038/1038 [==============================] - 7s 6ms/step - loss: 0.0817 - sparse_categorical_accuracy: 0.9720 - val_loss: 0.0838 - val_sparse_categorical_accuracy: 0.9732
Epoch 11/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0753 - sparse_categorical_accuracy: 0.9751 - val_loss: 0.0930 - val_sparse_categorical_accuracy: 0.9699
Epoch 12/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0752 - sparse_categorical_accuracy: 0.9753 - val_loss: 0.0838 - val_sparse_categorical_accuracy: 0.9737
Epoch 13/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0734 - sparse_categorical_accuracy: 0.9760 - val_loss: 0.0904 - val_sparse_categorical_accuracy: 0.9710
Epoch 14/50
1038/1038 [==============================] - 8s 7ms/step - loss: 0.0712 - sparse_categorical_accuracy: 0.9764 - val_loss: 0.0784 - val_sparse_categorical_accuracy: 0.9745
Epoch 15/50
1038/1038 [==============================] - 7s 7ms/step - loss: 0.0718 - sparse_categorical_accuracy: 0.9765 - val_loss: 0.0915 - val_sparse_categorical_accuracy: 0.9729
Epoch 16/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0691 - sparse_categorical_accuracy: 0.9775 - val_loss: 0.1016 - val_sparse_categorical_accuracy: 0.9694
Epoch 17/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0667 - sparse_categorical_accuracy: 0.9780 - val_loss: 0.0810 - val_sparse_categorical_accuracy: 0.9745
Epoch 18/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0659 - sparse_categorical_accuracy: 0.9780 - val_loss: 0.0839 - val_sparse_categorical_accuracy: 0.9753
Epoch 19/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0655 - sparse_categorical_accuracy: 0.9786 - val_loss: 0.0829 - val_sparse_categorical_accuracy: 0.9756
Epoch 20/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0634 - sparse_categorical_accuracy: 0.9791 - val_loss: 0.0998 - val_sparse_categorical_accuracy: 0.9691
Epoch 21/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0609 - sparse_categorical_accuracy: 0.9794 - val_loss: 0.1092 - val_sparse_categorical_accuracy: 0.9680
Epoch 22/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0627 - sparse_categorical_accuracy: 0.9794 - val_loss: 0.0826 - val_sparse_categorical_accuracy: 0.9753
Epoch 23/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0595 - sparse_categorical_accuracy: 0.9803 - val_loss: 0.0938 - val_sparse_categorical_accuracy: 0.9724
Epoch 24/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0575 - sparse_categorical_accuracy: 0.9813 - val_loss: 0.0962 - val_sparse_categorical_accuracy: 0.9718
Epoch 25/50
1038/1038 [==============================] - 8s 8ms/step - loss: 0.0562 - sparse_categorical_accuracy: 0.9808 - val_loss: 0.0936 - val_sparse_categorical_accuracy: 0.9718
Epoch 26/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0564 - sparse_categorical_accuracy: 0.9812 - val_loss: 0.0908 - val_sparse_categorical_accuracy: 0.9713
Epoch 27/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0545 - sparse_categorical_accuracy: 0.9823 - val_loss: 0.0820 - val_sparse_categorical_accuracy: 0.9753
Epoch 28/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0542 - sparse_categorical_accuracy: 0.9831 - val_loss: 0.0894 - val_sparse_categorical_accuracy: 0.9737
Epoch 29/50
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0548 - sparse_categorical_accuracy: 0.9826 - val_loss: 0.0934 - val_sparse_categorical_accuracy: 0.9740
Epoch 30/50
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0490 - sparse_categorical_accuracy: 0.9835 - val_loss: 0.0948 - val_sparse_categorical_accuracy: 0.9732
Epoch 31/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0528 - sparse_categorical_accuracy: 0.9826 - val_loss: 0.1016 - val_sparse_categorical_accuracy: 0.9707
Epoch 32/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0520 - sparse_categorical_accuracy: 0.9827 - val_loss: 0.1126 - val_sparse_categorical_accuracy: 0.9672
Epoch 33/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0506 - sparse_categorical_accuracy: 0.9834 - val_loss: 0.0868 - val_sparse_categorical_accuracy: 0.9740
Epoch 34/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0513 - sparse_categorical_accuracy: 0.9829 - val_loss: 0.0980 - val_sparse_categorical_accuracy: 0.9710
Epoch 35/50
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0489 - sparse_categorical_accuracy: 0.9836 - val_loss: 0.1010 - val_sparse_categorical_accuracy: 0.9729
Epoch 36/50
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0453 - sparse_categorical_accuracy: 0.9844 - val_loss: 0.0969 - val_sparse_categorical_accuracy: 0.9740
Epoch 37/50
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0492 - sparse_categorical_accuracy: 0.9837 - val_loss: 0.0982 - val_sparse_categorical_accuracy: 0.9737
Epoch 38/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0459 - sparse_categorical_accuracy: 0.9855 - val_loss: 0.1095 - val_sparse_categorical_accuracy: 0.9724
Epoch 39/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0449 - sparse_categorical_accuracy: 0.9858 - val_loss: 0.0932 - val_sparse_categorical_accuracy: 0.9740
Epoch 40/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0426 - sparse_categorical_accuracy: 0.9852 - val_loss: 0.1021 - val_sparse_categorical_accuracy: 0.9726
Epoch 41/50
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0455 - sparse_categorical_accuracy: 0.9851 - val_loss: 0.0996 - val_sparse_categorical_accuracy: 0.9742
Epoch 42/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0429 - sparse_categorical_accuracy: 0.9859 - val_loss: 0.1024 - val_sparse_categorical_accuracy: 0.9734
Epoch 43/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0421 - sparse_categorical_accuracy: 0.9858 - val_loss: 0.1048 - val_sparse_categorical_accuracy: 0.9732
Epoch 44/50
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0436 - sparse_categorical_accuracy: 0.9857 - val_loss: 0.1084 - val_sparse_categorical_accuracy: 0.9702
Epoch 45/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0424 - sparse_categorical_accuracy: 0.9864 - val_loss: 0.1064 - val_sparse_categorical_accuracy: 0.9724
Epoch 46/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0398 - sparse_categorical_accuracy: 0.9869 - val_loss: 0.1104 - val_sparse_categorical_accuracy: 0.9740
Epoch 47/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0414 - sparse_categorical_accuracy: 0.9859 - val_loss: 0.1193 - val_sparse_categorical_accuracy: 0.9718
Epoch 48/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0404 - sparse_categorical_accuracy: 0.9861 - val_loss: 0.1003 - val_sparse_categorical_accuracy: 0.9737
Epoch 49/50
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0408 - sparse_categorical_accuracy: 0.9868 - val_loss: 0.0993 - val_sparse_categorical_accuracy: 0.9734
Epoch 50/50
1038/1038 [==============================] - 7s 7ms/step - loss: 0.0462 - sparse_categorical_accuracy: 0.9848 - val_loss: 0.1090 - val_sparse_categorical_accuracy: 0.9729
Model: "sequential"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
flatten (Flatten)            (None, 12288)             0         
_________________________________________________________________
dense (Dense)                (None, 128)               1572992   
_________________________________________________________________
dropout (Dropout)            (None, 128)               0         
_________________________________________________________________
dense_1 (Dense)              (None, 64)                8256      
_________________________________________________________________
dense_2 (Dense)              (None, 32)                2080      
_________________________________________________________________
dense_3 (Dense)              (None, 2)                 66        
=================================================================
Total params: 1,583,394
Trainable params: 1,583,394
Non-trainable params: 0
_________________________________________________________________

6.2 epoch=100

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
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
D:\pyAnaconda\envs\pythonProject\python.exe D:/Code/pythonProject/work/sdss_reco.py
Epoch 1/100
1038/1038 [==============================] - 15s 14ms/step - loss: 0.2090 - sparse_categorical_accuracy: 0.9314 - val_loss: 0.1482 - val_sparse_categorical_accuracy: 0.9566
Epoch 2/100
1038/1038 [==============================] - 6s 5ms/step - loss: 0.1369 - sparse_categorical_accuracy: 0.9558 - val_loss: 0.1326 - val_sparse_categorical_accuracy: 0.9574
Epoch 3/100
1038/1038 [==============================] - 5s 5ms/step - loss: 0.1219 - sparse_categorical_accuracy: 0.9597 - val_loss: 0.1185 - val_sparse_categorical_accuracy: 0.9637
Epoch 4/100
1038/1038 [==============================] - 6s 5ms/step - loss: 0.1108 - sparse_categorical_accuracy: 0.9641 - val_loss: 0.1095 - val_sparse_categorical_accuracy: 0.9648
Epoch 5/100
1038/1038 [==============================] - 5s 5ms/step - loss: 0.1016 - sparse_categorical_accuracy: 0.9670 - val_loss: 0.1094 - val_sparse_categorical_accuracy: 0.9637
Epoch 6/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0973 - sparse_categorical_accuracy: 0.9687 - val_loss: 0.1157 - val_sparse_categorical_accuracy: 0.9645
Epoch 7/100
1038/1038 [==============================] - 5s 5ms/step - loss: 0.0963 - sparse_categorical_accuracy: 0.9687 - val_loss: 0.0963 - val_sparse_categorical_accuracy: 0.9702
Epoch 8/100
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0917 - sparse_categorical_accuracy: 0.9711 - val_loss: 0.0938 - val_sparse_categorical_accuracy: 0.9691
Epoch 9/100
1038/1038 [==============================] - 14s 13ms/step - loss: 0.0854 - sparse_categorical_accuracy: 0.9726 - val_loss: 0.0964 - val_sparse_categorical_accuracy: 0.9696
Epoch 10/100
1038/1038 [==============================] - 12s 11ms/step - loss: 0.0843 - sparse_categorical_accuracy: 0.9727 - val_loss: 0.1120 - val_sparse_categorical_accuracy: 0.9607
Epoch 11/100
1038/1038 [==============================] - 5s 5ms/step - loss: 0.0811 - sparse_categorical_accuracy: 0.9730 - val_loss: 0.0926 - val_sparse_categorical_accuracy: 0.9707
Epoch 12/100
1038/1038 [==============================] - 5s 5ms/step - loss: 0.0783 - sparse_categorical_accuracy: 0.9751 - val_loss: 0.0949 - val_sparse_categorical_accuracy: 0.9726
Epoch 13/100
1038/1038 [==============================] - 5s 5ms/step - loss: 0.0771 - sparse_categorical_accuracy: 0.9748 - val_loss: 0.0901 - val_sparse_categorical_accuracy: 0.9713
Epoch 14/100
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0756 - sparse_categorical_accuracy: 0.9758 - val_loss: 0.0988 - val_sparse_categorical_accuracy: 0.9713
Epoch 15/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0734 - sparse_categorical_accuracy: 0.9758 - val_loss: 0.0970 - val_sparse_categorical_accuracy: 0.9715
Epoch 16/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0714 - sparse_categorical_accuracy: 0.9763 - val_loss: 0.1045 - val_sparse_categorical_accuracy: 0.9672
Epoch 17/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0692 - sparse_categorical_accuracy: 0.9778 - val_loss: 0.1101 - val_sparse_categorical_accuracy: 0.9669
Epoch 18/100
1038/1038 [==============================] - 8s 8ms/step - loss: 0.0731 - sparse_categorical_accuracy: 0.9770 - val_loss: 0.0948 - val_sparse_categorical_accuracy: 0.9726
Epoch 19/100
1038/1038 [==============================] - 7s 7ms/step - loss: 0.0659 - sparse_categorical_accuracy: 0.9785 - val_loss: 0.1006 - val_sparse_categorical_accuracy: 0.9696
Epoch 20/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0667 - sparse_categorical_accuracy: 0.9783 - val_loss: 0.1242 - val_sparse_categorical_accuracy: 0.9618
Epoch 21/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0635 - sparse_categorical_accuracy: 0.9788 - val_loss: 0.1007 - val_sparse_categorical_accuracy: 0.9713
Epoch 22/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0705 - sparse_categorical_accuracy: 0.9771 - val_loss: 0.0928 - val_sparse_categorical_accuracy: 0.9745
Epoch 23/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0614 - sparse_categorical_accuracy: 0.9801 - val_loss: 0.0957 - val_sparse_categorical_accuracy: 0.9734
Epoch 24/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0613 - sparse_categorical_accuracy: 0.9805 - val_loss: 0.0967 - val_sparse_categorical_accuracy: 0.9713
Epoch 25/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0621 - sparse_categorical_accuracy: 0.9797 - val_loss: 0.0828 - val_sparse_categorical_accuracy: 0.9726
Epoch 26/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0597 - sparse_categorical_accuracy: 0.9807 - val_loss: 0.1076 - val_sparse_categorical_accuracy: 0.9691
Epoch 27/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0593 - sparse_categorical_accuracy: 0.9805 - val_loss: 0.0948 - val_sparse_categorical_accuracy: 0.9732
Epoch 28/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0613 - sparse_categorical_accuracy: 0.9808 - val_loss: 0.0974 - val_sparse_categorical_accuracy: 0.9713
Epoch 29/100
1038/1038 [==============================] - 9s 9ms/step - loss: 0.0552 - sparse_categorical_accuracy: 0.9821 - val_loss: 0.1000 - val_sparse_categorical_accuracy: 0.9688
Epoch 30/100
1038/1038 [==============================] - 7s 6ms/step - loss: 0.0567 - sparse_categorical_accuracy: 0.9810 - val_loss: 0.0922 - val_sparse_categorical_accuracy: 0.9726
Epoch 31/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0575 - sparse_categorical_accuracy: 0.9820 - val_loss: 0.0956 - val_sparse_categorical_accuracy: 0.9740
Epoch 32/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0551 - sparse_categorical_accuracy: 0.9833 - val_loss: 0.1317 - val_sparse_categorical_accuracy: 0.9620
Epoch 33/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0560 - sparse_categorical_accuracy: 0.9822 - val_loss: 0.0863 - val_sparse_categorical_accuracy: 0.9721
Epoch 34/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0571 - sparse_categorical_accuracy: 0.9812 - val_loss: 0.0888 - val_sparse_categorical_accuracy: 0.9748
Epoch 35/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0527 - sparse_categorical_accuracy: 0.9823 - val_loss: 0.1107 - val_sparse_categorical_accuracy: 0.9707
Epoch 36/100
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0534 - sparse_categorical_accuracy: 0.9827 - val_loss: 0.1104 - val_sparse_categorical_accuracy: 0.9721
Epoch 37/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0533 - sparse_categorical_accuracy: 0.9822 - val_loss: 0.1183 - val_sparse_categorical_accuracy: 0.9680
Epoch 38/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0541 - sparse_categorical_accuracy: 0.9828 - val_loss: 0.1079 - val_sparse_categorical_accuracy: 0.9713
Epoch 39/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0522 - sparse_categorical_accuracy: 0.9831 - val_loss: 0.1045 - val_sparse_categorical_accuracy: 0.9732
Epoch 40/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0503 - sparse_categorical_accuracy: 0.9838 - val_loss: 0.0964 - val_sparse_categorical_accuracy: 0.9715
Epoch 41/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0500 - sparse_categorical_accuracy: 0.9836 - val_loss: 0.0952 - val_sparse_categorical_accuracy: 0.9732
Epoch 42/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0529 - sparse_categorical_accuracy: 0.9825 - val_loss: 0.0954 - val_sparse_categorical_accuracy: 0.9715
Epoch 43/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0475 - sparse_categorical_accuracy: 0.9844 - val_loss: 0.1000 - val_sparse_categorical_accuracy: 0.9726
Epoch 44/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0498 - sparse_categorical_accuracy: 0.9839 - val_loss: 0.1020 - val_sparse_categorical_accuracy: 0.9688
Epoch 45/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0486 - sparse_categorical_accuracy: 0.9833 - val_loss: 0.1088 - val_sparse_categorical_accuracy: 0.9694
Epoch 46/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0478 - sparse_categorical_accuracy: 0.9843 - val_loss: 0.0983 - val_sparse_categorical_accuracy: 0.9726
Epoch 47/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0469 - sparse_categorical_accuracy: 0.9852 - val_loss: 0.1149 - val_sparse_categorical_accuracy: 0.9680
Epoch 48/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0462 - sparse_categorical_accuracy: 0.9856 - val_loss: 0.1039 - val_sparse_categorical_accuracy: 0.9721
Epoch 49/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0462 - sparse_categorical_accuracy: 0.9852 - val_loss: 0.1063 - val_sparse_categorical_accuracy: 0.9729
Epoch 50/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0481 - sparse_categorical_accuracy: 0.9846 - val_loss: 0.0931 - val_sparse_categorical_accuracy: 0.9742
Epoch 51/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0461 - sparse_categorical_accuracy: 0.9851 - val_loss: 0.1039 - val_sparse_categorical_accuracy: 0.9707
Epoch 52/100
1038/1038 [==============================] - 7s 7ms/step - loss: 0.0422 - sparse_categorical_accuracy: 0.9858 - val_loss: 0.1228 - val_sparse_categorical_accuracy: 0.9680
Epoch 53/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0446 - sparse_categorical_accuracy: 0.9855 - val_loss: 0.1068 - val_sparse_categorical_accuracy: 0.9702
Epoch 54/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0420 - sparse_categorical_accuracy: 0.9858 - val_loss: 0.1050 - val_sparse_categorical_accuracy: 0.9715
Epoch 55/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0446 - sparse_categorical_accuracy: 0.9851 - val_loss: 0.1147 - val_sparse_categorical_accuracy: 0.9721
Epoch 56/100
1038/1038 [==============================] - 7s 6ms/step - loss: 0.0432 - sparse_categorical_accuracy: 0.9855 - val_loss: 0.1015 - val_sparse_categorical_accuracy: 0.9745
Epoch 57/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0463 - sparse_categorical_accuracy: 0.9851 - val_loss: 0.1018 - val_sparse_categorical_accuracy: 0.9707
Epoch 58/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0434 - sparse_categorical_accuracy: 0.9858 - val_loss: 0.1078 - val_sparse_categorical_accuracy: 0.9696
Epoch 59/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0443 - sparse_categorical_accuracy: 0.9854 - val_loss: 0.1153 - val_sparse_categorical_accuracy: 0.9686
Epoch 60/100
1038/1038 [==============================] - 7s 7ms/step - loss: 0.0419 - sparse_categorical_accuracy: 0.9868 - val_loss: 0.1222 - val_sparse_categorical_accuracy: 0.9664
Epoch 61/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0409 - sparse_categorical_accuracy: 0.9865 - val_loss: 0.1257 - val_sparse_categorical_accuracy: 0.9680
Epoch 62/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0428 - sparse_categorical_accuracy: 0.9859 - val_loss: 0.1155 - val_sparse_categorical_accuracy: 0.9710
Epoch 63/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0440 - sparse_categorical_accuracy: 0.9850 - val_loss: 0.1254 - val_sparse_categorical_accuracy: 0.9667
Epoch 64/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0409 - sparse_categorical_accuracy: 0.9869 - val_loss: 0.1135 - val_sparse_categorical_accuracy: 0.9691
Epoch 65/100
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0423 - sparse_categorical_accuracy: 0.9867 - val_loss: 0.1272 - val_sparse_categorical_accuracy: 0.9696
Epoch 66/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0389 - sparse_categorical_accuracy: 0.9867 - val_loss: 0.1229 - val_sparse_categorical_accuracy: 0.9729
Epoch 67/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0400 - sparse_categorical_accuracy: 0.9870 - val_loss: 0.0986 - val_sparse_categorical_accuracy: 0.9742
Epoch 68/100
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0414 - sparse_categorical_accuracy: 0.9868 - val_loss: 0.1165 - val_sparse_categorical_accuracy: 0.9726
Epoch 69/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0416 - sparse_categorical_accuracy: 0.9866 - val_loss: 0.1248 - val_sparse_categorical_accuracy: 0.9694
Epoch 70/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0374 - sparse_categorical_accuracy: 0.9878 - val_loss: 0.1092 - val_sparse_categorical_accuracy: 0.9713
Epoch 71/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0408 - sparse_categorical_accuracy: 0.9872 - val_loss: 0.1141 - val_sparse_categorical_accuracy: 0.9702
Epoch 72/100
1038/1038 [==============================] - 6s 5ms/step - loss: 0.0393 - sparse_categorical_accuracy: 0.9868 - val_loss: 0.1115 - val_sparse_categorical_accuracy: 0.9718
Epoch 73/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0384 - sparse_categorical_accuracy: 0.9882 - val_loss: 0.1278 - val_sparse_categorical_accuracy: 0.9656
Epoch 74/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0392 - sparse_categorical_accuracy: 0.9875 - val_loss: 0.1134 - val_sparse_categorical_accuracy: 0.9715
Epoch 75/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0371 - sparse_categorical_accuracy: 0.9883 - val_loss: 0.1292 - val_sparse_categorical_accuracy: 0.9713
Epoch 76/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0412 - sparse_categorical_accuracy: 0.9867 - val_loss: 0.1344 - val_sparse_categorical_accuracy: 0.9664
Epoch 77/100
1038/1038 [==============================] - 8s 8ms/step - loss: 0.0351 - sparse_categorical_accuracy: 0.9881 - val_loss: 0.1517 - val_sparse_categorical_accuracy: 0.9672
Epoch 78/100
1038/1038 [==============================] - 7s 7ms/step - loss: 0.0385 - sparse_categorical_accuracy: 0.9876 - val_loss: 0.1131 - val_sparse_categorical_accuracy: 0.9691
Epoch 79/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0387 - sparse_categorical_accuracy: 0.9870 - val_loss: 0.1222 - val_sparse_categorical_accuracy: 0.9705
Epoch 80/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0367 - sparse_categorical_accuracy: 0.9880 - val_loss: 0.1233 - val_sparse_categorical_accuracy: 0.9718
Epoch 81/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0340 - sparse_categorical_accuracy: 0.9888 - val_loss: 0.1138 - val_sparse_categorical_accuracy: 0.9702
Epoch 82/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0427 - sparse_categorical_accuracy: 0.9870 - val_loss: 0.1517 - val_sparse_categorical_accuracy: 0.9593
Epoch 83/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0424 - sparse_categorical_accuracy: 0.9863 - val_loss: 0.1133 - val_sparse_categorical_accuracy: 0.9702
Epoch 84/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0399 - sparse_categorical_accuracy: 0.9871 - val_loss: 0.1268 - val_sparse_categorical_accuracy: 0.9724
Epoch 85/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0394 - sparse_categorical_accuracy: 0.9874 - val_loss: 0.1281 - val_sparse_categorical_accuracy: 0.9691
Epoch 86/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0382 - sparse_categorical_accuracy: 0.9884 - val_loss: 0.1268 - val_sparse_categorical_accuracy: 0.9710
Epoch 87/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0375 - sparse_categorical_accuracy: 0.9879 - val_loss: 0.1293 - val_sparse_categorical_accuracy: 0.9702
Epoch 88/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0390 - sparse_categorical_accuracy: 0.9876 - val_loss: 0.1538 - val_sparse_categorical_accuracy: 0.9694
Epoch 89/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0361 - sparse_categorical_accuracy: 0.9883 - val_loss: 0.1266 - val_sparse_categorical_accuracy: 0.9699
Epoch 90/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0383 - sparse_categorical_accuracy: 0.9874 - val_loss: 0.1244 - val_sparse_categorical_accuracy: 0.9718
Epoch 91/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0323 - sparse_categorical_accuracy: 0.9901 - val_loss: 0.1377 - val_sparse_categorical_accuracy: 0.9713
Epoch 92/100
1038/1038 [==============================] - 7s 7ms/step - loss: 0.0347 - sparse_categorical_accuracy: 0.9882 - val_loss: 0.1412 - val_sparse_categorical_accuracy: 0.9705
Epoch 93/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0339 - sparse_categorical_accuracy: 0.9894 - val_loss: 0.1389 - val_sparse_categorical_accuracy: 0.9658
Epoch 94/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0354 - sparse_categorical_accuracy: 0.9886 - val_loss: 0.1223 - val_sparse_categorical_accuracy: 0.9699
Epoch 95/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0352 - sparse_categorical_accuracy: 0.9887 - val_loss: 0.1688 - val_sparse_categorical_accuracy: 0.9604
Epoch 96/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0334 - sparse_categorical_accuracy: 0.9892 - val_loss: 0.1383 - val_sparse_categorical_accuracy: 0.9661
Epoch 97/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0318 - sparse_categorical_accuracy: 0.9902 - val_loss: 0.1281 - val_sparse_categorical_accuracy: 0.9694
Epoch 98/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0372 - sparse_categorical_accuracy: 0.9884 - val_loss: 0.1387 - val_sparse_categorical_accuracy: 0.9702
Epoch 99/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0400 - sparse_categorical_accuracy: 0.9871 - val_loss: 0.1396 - val_sparse_categorical_accuracy: 0.9702
Epoch 100/100
1038/1038 [==============================] - 6s 6ms/step - loss: 0.0339 - sparse_categorical_accuracy: 0.9890 - val_loss: 0.1252 - val_sparse_categorical_accuracy: 0.9710
Model: "sequential"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
flatten (Flatten)            (None, 12288)             0         
_________________________________________________________________
dense (Dense)                (None, 128)               1572992   
_________________________________________________________________
dropout (Dropout)            (None, 128)               0         
_________________________________________________________________
dense_1 (Dense)              (None, 64)                8256      
_________________________________________________________________
dense_2 (Dense)              (None, 32)                2080      
_________________________________________________________________
dense_3 (Dense)              (None, 2)                 66        
=================================================================
Total params: 1,583,394
Trainable params: 1,583,394
Non-trainable params: 0
_________________________________________________________________

6.4

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
D:\pyAnaconda\envs\pythonProject\python.exe D:/Code/pythonProject/work/sdss_reco.py
Epoch 1/50
519/519 [==============================] - 4s 7ms/step - loss: 0.2239 - sparse_categorical_accuracy: 0.9276 - val_loss: 0.1582 - val_sparse_categorical_accuracy: 0.9504
Epoch 2/50
519/519 [==============================] - 3s 6ms/step - loss: 0.1384 - sparse_categorical_accuracy: 0.9546 - val_loss: 0.1269 - val_sparse_categorical_accuracy: 0.9604
Epoch 3/50
519/519 [==============================] - 3s 6ms/step - loss: 0.1186 - sparse_categorical_accuracy: 0.9612 - val_loss: 0.1119 - val_sparse_categorical_accuracy: 0.9631
Epoch 4/50
519/519 [==============================] - 3s 7ms/step - loss: 0.1159 - sparse_categorical_accuracy: 0.9615 - val_loss: 0.1103 - val_sparse_categorical_accuracy: 0.9639
Epoch 5/50
519/519 [==============================] - 3s 6ms/step - loss: 0.1049 - sparse_categorical_accuracy: 0.9646 - val_loss: 0.0992 - val_sparse_categorical_accuracy: 0.9677
Epoch 6/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0989 - sparse_categorical_accuracy: 0.9675 - val_loss: 0.1057 - val_sparse_categorical_accuracy: 0.9650
Epoch 7/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0922 - sparse_categorical_accuracy: 0.9708 - val_loss: 0.1037 - val_sparse_categorical_accuracy: 0.9658
Epoch 8/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0916 - sparse_categorical_accuracy: 0.9698 - val_loss: 0.1100 - val_sparse_categorical_accuracy: 0.9686
Epoch 9/50
519/519 [==============================] - 4s 7ms/step - loss: 0.0864 - sparse_categorical_accuracy: 0.9719 - val_loss: 0.0980 - val_sparse_categorical_accuracy: 0.9688
Epoch 10/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0812 - sparse_categorical_accuracy: 0.9736 - val_loss: 0.0850 - val_sparse_categorical_accuracy: 0.9729
Epoch 11/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0768 - sparse_categorical_accuracy: 0.9757 - val_loss: 0.0868 - val_sparse_categorical_accuracy: 0.9724
Epoch 12/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0743 - sparse_categorical_accuracy: 0.9756 - val_loss: 0.0987 - val_sparse_categorical_accuracy: 0.9680
Epoch 13/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0731 - sparse_categorical_accuracy: 0.9761 - val_loss: 0.0906 - val_sparse_categorical_accuracy: 0.9718
Epoch 14/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0711 - sparse_categorical_accuracy: 0.9771 - val_loss: 0.0782 - val_sparse_categorical_accuracy: 0.9751
Epoch 15/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0669 - sparse_categorical_accuracy: 0.9784 - val_loss: 0.0904 - val_sparse_categorical_accuracy: 0.9734
Epoch 16/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0750 - sparse_categorical_accuracy: 0.9755 - val_loss: 0.0949 - val_sparse_categorical_accuracy: 0.9705
Epoch 17/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0723 - sparse_categorical_accuracy: 0.9764 - val_loss: 0.0831 - val_sparse_categorical_accuracy: 0.9761
Epoch 18/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0639 - sparse_categorical_accuracy: 0.9794 - val_loss: 0.0987 - val_sparse_categorical_accuracy: 0.9710
Epoch 19/50
519/519 [==============================] - 3s 7ms/step - loss: 0.0648 - sparse_categorical_accuracy: 0.9787 - val_loss: 0.0897 - val_sparse_categorical_accuracy: 0.9734
Epoch 20/50
519/519 [==============================] - 4s 7ms/step - loss: 0.0637 - sparse_categorical_accuracy: 0.9793 - val_loss: 0.0803 - val_sparse_categorical_accuracy: 0.9759
Epoch 21/50
519/519 [==============================] - 4s 7ms/step - loss: 0.0612 - sparse_categorical_accuracy: 0.9798 - val_loss: 0.0770 - val_sparse_categorical_accuracy: 0.9783
Epoch 22/50
519/519 [==============================] - 4s 7ms/step - loss: 0.0574 - sparse_categorical_accuracy: 0.9820 - val_loss: 0.0890 - val_sparse_categorical_accuracy: 0.9753
Epoch 23/50
519/519 [==============================] - 3s 7ms/step - loss: 0.0590 - sparse_categorical_accuracy: 0.9801 - val_loss: 0.0872 - val_sparse_categorical_accuracy: 0.9729
Epoch 24/50
519/519 [==============================] - 4s 8ms/step - loss: 0.0581 - sparse_categorical_accuracy: 0.9806 - val_loss: 0.1062 - val_sparse_categorical_accuracy: 0.9699
Epoch 25/50
519/519 [==============================] - 4s 7ms/step - loss: 0.0578 - sparse_categorical_accuracy: 0.9810 - val_loss: 0.0894 - val_sparse_categorical_accuracy: 0.9721
Epoch 26/50
519/519 [==============================] - 4s 8ms/step - loss: 0.0529 - sparse_categorical_accuracy: 0.9828 - val_loss: 0.0863 - val_sparse_categorical_accuracy: 0.9745
Epoch 27/50
519/519 [==============================] - 4s 7ms/step - loss: 0.0524 - sparse_categorical_accuracy: 0.9833 - val_loss: 0.0928 - val_sparse_categorical_accuracy: 0.9729
Epoch 28/50
519/519 [==============================] - 4s 7ms/step - loss: 0.0540 - sparse_categorical_accuracy: 0.9820 - val_loss: 0.0815 - val_sparse_categorical_accuracy: 0.9759
Epoch 29/50
519/519 [==============================] - 4s 8ms/step - loss: 0.0536 - sparse_categorical_accuracy: 0.9829 - val_loss: 0.1091 - val_sparse_categorical_accuracy: 0.9686
Epoch 30/50
519/519 [==============================] - 5s 9ms/step - loss: 0.0483 - sparse_categorical_accuracy: 0.9837 - val_loss: 0.0886 - val_sparse_categorical_accuracy: 0.9756
Epoch 31/50
519/519 [==============================] - 4s 8ms/step - loss: 0.0484 - sparse_categorical_accuracy: 0.9843 - val_loss: 0.0877 - val_sparse_categorical_accuracy: 0.9729
Epoch 32/50
519/519 [==============================] - 4s 8ms/step - loss: 0.0490 - sparse_categorical_accuracy: 0.9836 - val_loss: 0.0902 - val_sparse_categorical_accuracy: 0.9740
Epoch 33/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0491 - sparse_categorical_accuracy: 0.9839 - val_loss: 0.0917 - val_sparse_categorical_accuracy: 0.9734
Epoch 34/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0517 - sparse_categorical_accuracy: 0.9833 - val_loss: 0.0899 - val_sparse_categorical_accuracy: 0.9729
Epoch 35/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0478 - sparse_categorical_accuracy: 0.9840 - val_loss: 0.0999 - val_sparse_categorical_accuracy: 0.9724
Epoch 36/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0444 - sparse_categorical_accuracy: 0.9855 - val_loss: 0.0949 - val_sparse_categorical_accuracy: 0.9726
Epoch 37/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0480 - sparse_categorical_accuracy: 0.9843 - val_loss: 0.0861 - val_sparse_categorical_accuracy: 0.9751
Epoch 38/50
519/519 [==============================] - 4s 7ms/step - loss: 0.0430 - sparse_categorical_accuracy: 0.9857 - val_loss: 0.0805 - val_sparse_categorical_accuracy: 0.9756
Epoch 39/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0415 - sparse_categorical_accuracy: 0.9868 - val_loss: 0.0850 - val_sparse_categorical_accuracy: 0.9775
Epoch 40/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0417 - sparse_categorical_accuracy: 0.9864 - val_loss: 0.1631 - val_sparse_categorical_accuracy: 0.9547
Epoch 41/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0526 - sparse_categorical_accuracy: 0.9826 - val_loss: 0.0935 - val_sparse_categorical_accuracy: 0.9751
Epoch 42/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0423 - sparse_categorical_accuracy: 0.9863 - val_loss: 0.0992 - val_sparse_categorical_accuracy: 0.9718
Epoch 43/50
519/519 [==============================] - 3s 7ms/step - loss: 0.0403 - sparse_categorical_accuracy: 0.9861 - val_loss: 0.0881 - val_sparse_categorical_accuracy: 0.9745
Epoch 44/50
519/519 [==============================] - 3s 7ms/step - loss: 0.0407 - sparse_categorical_accuracy: 0.9864 - val_loss: 0.0972 - val_sparse_categorical_accuracy: 0.9729
Epoch 45/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0413 - sparse_categorical_accuracy: 0.9863 - val_loss: 0.0901 - val_sparse_categorical_accuracy: 0.9753
Epoch 46/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0397 - sparse_categorical_accuracy: 0.9867 - val_loss: 0.0934 - val_sparse_categorical_accuracy: 0.9745
Epoch 47/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0408 - sparse_categorical_accuracy: 0.9865 - val_loss: 0.1015 - val_sparse_categorical_accuracy: 0.9715
Epoch 48/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0377 - sparse_categorical_accuracy: 0.9878 - val_loss: 0.0932 - val_sparse_categorical_accuracy: 0.9748
Epoch 49/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0384 - sparse_categorical_accuracy: 0.9871 - val_loss: 0.0941 - val_sparse_categorical_accuracy: 0.9751
Epoch 50/50
519/519 [==============================] - 3s 6ms/step - loss: 0.0395 - sparse_categorical_accuracy: 0.9871 - val_loss: 0.0971 - val_sparse_categorical_accuracy: 0.9753
Model: "sequential"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
flatten (Flatten)            (None, 12288)             0         
_________________________________________________________________
dense (Dense)                (None, 128)               1572992   
_________________________________________________________________
dropout (Dropout)            (None, 128)               0         
_________________________________________________________________
dense_1 (Dense)              (None, 64)                8256      
_________________________________________________________________
dense_2 (Dense)              (None, 32)                2080      
_________________________________________________________________
dense_3 (Dense)              (None, 2)                 66        
=================================================================
Total params: 1,583,394
Trainable params: 1,583,394
Non-trainable params: 0
_________________________________________________________________

auc = 0.992495791163834

21

loss: 0.0612 - sparse_categorical_accuracy: 0.9798 - val_loss: 0.0770 - val_sparse_categorical_accuracy: 0.9783

39

519/519 [==============================] - 3s 6ms/step - loss: 0.0415 - sparse_categorical_accuracy: 0.9868 - val_loss: 0.0850 - val_sparse_categorical_accuracy: 0.9775

STERNE

STERNE

SDUer，大一入门，大二入土，大三老东西！

63
1
20
NOTICE

这里还什么都没有呢喵~

CATEGORYS
TAGS
ACM Arc CCPC HPC bot c cpp deeplearning games hexo java maths pyqt python qt recommends review sql srtp test