-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy path10_Epoch_dump.txt
104 lines (90 loc) · 5.86 KB
/
10_Epoch_dump.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
91
92
93
94
95
96
97
98
99
100
101
102
h5fileHWcalls_1110records.h5 keys are <KeysViewHDF5 ['eval_labels', 'eval_specs', 'test_labels', 'test_specs', 'train_labels', 'train_specs']>
eval_labels length 54
eval_specs length 54
test_labels length 228
test_specs length 228
train_labels length 828
train_specs length 828
top of generate
Confusion matrix fractions for predictions on dataset train dataset of length 100
PREDICT 0 1
Label = 0 TN 0.160 FN 0.350
Label = 1 FP 0.110 TP 0.380
WARNING:tensorflow:`period` argument is deprecated. Please use `save_freq` to specify the frequency in number of batches seen.
start model at time 1649980530.6896343 secs
Epoch 1/10
2022-04-14 16:55:31.324628: W tensorflow/core/framework/cpu_allocator_impl.cc:82] Allocation of 1677721600 exceeds 10% of free system memory.
2022-04-14 16:55:31.766217: W tensorflow/core/framework/cpu_allocator_impl.cc:82] Allocation of 1677721600 exceeds 10% of free system memory.
2022-04-14 16:55:33.074141: W tensorflow/core/framework/cpu_allocator_impl.cc:82] Allocation of 1677721600 exceeds 10% of free system memory.
2022-04-14 16:55:35.251575: W tensorflow/core/framework/cpu_allocator_impl.cc:82] Allocation of 1677721600 exceeds 10% of free system memory.
2022-04-14 16:55:39.085913: W tensorflow/core/framework/cpu_allocator_impl.cc:82] Allocation of 1677721600 exceeds 10% of free system memory.
batchcount= 10 batchsize= 100
top of generate
batchcount= 10 batchsize= 100
Epoch 1: loss improved from inf to 0.47669, saving model to models/Classifier_75_20_5_[256-128-32-8]_Em_h5_best_1100.ckpt
2022-04-14 16:57:43.850463: W tensorflow/python/util/util.cc:368] Sets are not currently considered sequences, but this may change in the future, so consider avoiding using them.
10/10 - 134s - loss: 0.4767 - accuracy: 0.8280 - val_loss: 0.6908 - val_accuracy: 0.5370 - 134s/epoch - 13s/step
Epoch 2/10
batchcount= 20 batchsize= 100
batchcount= 20 batchsize= 100
Epoch 2: loss improved from 0.47669 to 0.24313, saving model to models/Classifier_75_20_5_[256-128-32-8]_Em_h5_best_1100.ckpt
10/10 - 132s - loss: 0.2431 - accuracy: 0.9770 - val_loss: 0.7004 - val_accuracy: 0.5080 - 132s/epoch - 13s/step
Epoch 3/10
batchcount= 30 batchsize= 100
batchcount= 30 batchsize= 100
Epoch 3: loss improved from 0.24313 to 0.15739, saving model to models/Classifier_75_20_5_[256-128-32-8]_Em_h5_best_1100.ckpt
10/10 - 132s - loss: 0.1574 - accuracy: 0.9860 - val_loss: 0.7006 - val_accuracy: 0.4960 - 132s/epoch - 13s/step
Epoch 4/10
batchcount= 40 batchsize= 100
batchcount= 40 batchsize= 100
Epoch 4: loss improved from 0.15739 to 0.11405, saving model to models/Classifier_75_20_5_[256-128-32-8]_Em_h5_best_1100.ckpt
10/10 - 132s - loss: 0.1140 - accuracy: 0.9840 - val_loss: 0.6902 - val_accuracy: 0.4770 - 132s/epoch - 13s/step
Epoch 5/10
batchcount= 50 batchsize= 100
batchcount= 50 batchsize= 100
Epoch 5: loss improved from 0.11405 to 0.09953, saving model to models/Classifier_75_20_5_[256-128-32-8]_Em_h5_best_1100.ckpt
10/10 - 132s - loss: 0.0995 - accuracy: 0.9850 - val_loss: 0.6843 - val_accuracy: 0.5770 - 132s/epoch - 13s/step
Epoch 6/10
batchcount= 60 batchsize= 100
batchcount= 60 batchsize= 100
Epoch 6: loss did not improve from 0.09953
10/10 - 131s - loss: 0.1194 - accuracy: 0.9720 - val_loss: 0.6870 - val_accuracy: 0.5070 - 131s/epoch - 13s/step
Epoch 7/10
batchcount= 70 batchsize= 100
batchcount= 70 batchsize= 100
Epoch 7: loss did not improve from 0.09953
10/10 - 129s - loss: 0.1140 - accuracy: 0.9810 - val_loss: 0.6934 - val_accuracy: 0.5070 - 129s/epoch - 13s/step
Epoch 8/10
batchcount= 80 batchsize= 100
batchcount= 80 batchsize= 100
Epoch 8: loss improved from 0.09953 to 0.09590, saving model to models/Classifier_75_20_5_[256-128-32-8]_Em_h5_best_1100.ckpt
10/10 - 120s - loss: 0.0959 - accuracy: 0.9870 - val_loss: 0.7055 - val_accuracy: 0.5240 - 120s/epoch - 12s/step
Epoch 9/10
batchcount= 90 batchsize= 100
batchcount= 90 batchsize= 100
Epoch 9: loss improved from 0.09590 to 0.08181, saving model to models/Classifier_75_20_5_[256-128-32-8]_Em_h5_best_1100.ckpt
10/10 - 104s - loss: 0.0818 - accuracy: 0.9890 - val_loss: 0.7126 - val_accuracy: 0.4780 - 104s/epoch - 10s/step
Epoch 10/10
batchcount= 100 batchsize= 100
batchcount= 100 batchsize= 100
batchcount= 110 batchsize= 100
Epoch 10: loss improved from 0.08181 to 0.08073, saving model to models/Classifier_75_20_5_[256-128-32-8]_Em_h5_best_1100.ckpt
10/10 - 103s - loss: 0.0807 - accuracy: 0.9860 - val_loss: 0.7634 - val_accuracy: 0.4660 - 103s/epoch - 10s/step
save classifyAE_Model Classifier_75_20_5_[256-128-32-8]_Em_h5 at directory
models/Classifier_75_20_5_[256-128-32-8]_Em_h5_0_10_epochs/
Elapsed time s 1250, m 20, hr 0.35 s/epoch 125.08
history keys dict_keys(['loss', 'accuracy', 'val_loss', 'val_accuracy']) models/history_Classifier_75_20_5_[256-128-32-8]_Em_h5_[]-0_1100.pkl
Confusion matrix fractions for predictions on dataset train dataset of length 828
PREDICT 0 1
Label = 0 TN 0.390 FN 0.136
Label = 1 FP 0.278 TP 0.196
WARNING:tensorflow:Keras is training/fitting/evaluating on array-like data. Keras may not be optimized for this format, so if your input data format is supported by TensorFlow I/O (https://github.com/tensorflow/io) we recommend using that to load a Dataset instead.
Confusion matrix fractions for predictions on dataset test dataset of length 228
PREDICT 0 1
Label = 0 TN 0.377 FN 0.105
Label = 1 FP 0.360 TP 0.158
WARNING:tensorflow:Keras is training/fitting/evaluating on array-like data. Keras may not be optimized for this format, so if your input data format is supported by TensorFlow I/O (https://github.com/tensorflow/io) we recommend using that to load a Dataset instead.
Confusion matrix fractions for predictions on dataset eval dataset of length 54
PREDICT 0 1
Label = 0 TN 0.333 FN 0.019
Label = 1 FP 0.519 TP 0.130