1
2
3
>>> git clone https://github.com/fchollet/keras.git
>>> cd keras/examples/
>>> python mnist_mlp.py

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
root@astron:/asdata# git clone https://github.com/fchollet/keras.git
Cloning into 'keras'...
remote: Counting objects: 23274, done.
remote: Compressing objects: 100% (7/7), done.
remote: Total 23274 (delta 1), reused 0 (delta 0), pack-reused 23267
Receiving objects: 100% (23274/23274), 8.84 MiB | 1.12 MiB/s, done.
Resolving deltas: 100% (16738/16738), done.
Checking connectivity... done.
root@astron:/asdata# cd keras/
root@astron:/asdata/keras# ls
CONTRIBUTING.md
docs
ISSUE_TEMPLATE.md
LICENSE
pytest.ini
setup.cfg
tests
docker
examples
keras
MANIFEST.in
README.md
setup.py
root@astron:/asdata/keras# cd examples/
root@astron:/asdata/keras/examples# ls
addition_rnn.py
imdb_bidirectional_lstm.py
mnist_hierarchical_rnn.py
neural_doodle.py
antirectifier.py
imdb_cnn_lstm.py
mnist_irnn.py
neural_style_transfer.py
babi_memnn.py
imdb_cnn.py
mnist_mlp.py
pretrained_word_embeddings.py
babi_rnn.py
imdb_fasttext.py
mnist_net2net.py
README.md
cifar10_cnn.py
imdb_lstm.py
mnist_siamese_graph.py
reuters_mlp.py
conv_filter_visualization.py
lstm_benchmark.py
mnist_sklearn_wrapper.py
reuters_mlp_relu_vs_selu.py
conv_lstm.py
lstm_text_generation.py
mnist_swwae.py
stateful_lstm.py
deep_dream.py
mnist_acgan.py
mnist_tfrecord.py
variational_autoencoder_deconv.py
image_ocr.py
mnist_cnn.py
mnist_transfer_cnn.py
variational_autoencoder.py
root@astron:/asdata/keras/examples# python mnist_mlp.py
Using TensorFlow backend.
I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcublas.so.8.0 locally
I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcudnn.so.5 locally
I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcufft.so.8.0 locally
I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcuda.so.1 locally
I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcurand.so.8.0 locally
60000 train samples
10000 test samples
_________________________________________________________________
Layer (type)
Output Shape
Param #
=================================================================
dense_1 (Dense)
(None, 512)
401920
_________________________________________________________________
dropout_1 (Dropout)
(None, 512)
0
_________________________________________________________________
dense_2 (Dense)
(None, 512)
262656
_________________________________________________________________
dropout_2 (Dropout)
(None, 512)
0
_________________________________________________________________
dense_3 (Dense)
(None, 10)
5130
=================================================================
Total params: 669,706.0
Trainable params: 669,706.0
Non-trainable params: 0.0
_________________________________________________________________
Train on 60000 samples, validate on 10000 samples
Epoch 1/20
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE3 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.
I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:910] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
I tensorflow/core/common_runtime/gpu/gpu_device.cc:885] Found device 0 with properties:
name: Tesla M60
major: 5 minor: 2 memoryClockRate (GHz) 1.1775
pciBusID 0000:00:15.0
Total memory: 7.93GiB
Free memory: 7.86GiB
I tensorflow/core/common_runtime/gpu/gpu_device.cc:906] DMA: 0
I tensorflow/core/common_runtime/gpu/gpu_device.cc:916] 0:
Y
I tensorflow/core/common_runtime/gpu/gpu_device.cc:975] Creating TensorFlow device (/gpu:0) -> (device: 0, name: Tesla M60, pci bus id: 0000:00:15.0)
60000/60000 [==============================] - 3s - loss: 0.2442 - acc: 0.9246 - val_loss: 0.0994 - val_acc: 0.9692
Epoch 2/20
60000/60000 [==============================] - 1s - loss: 0.1041 - acc: 0.9684 - val_loss: 0.0818 - val_acc: 0.9751
Epoch 3/20
60000/60000 [==============================] - 1s - loss: 0.0751 - acc: 0.9766 - val_loss: 0.0821 - val_acc: 0.9762
Epoch 4/20
60000/60000 [==============================] - 1s - loss: 0.0596 - acc: 0.9821 - val_loss: 0.0688 - val_acc: 0.9809
Epoch 5/20
60000/60000 [==============================] - 1s - loss: 0.0501 - acc: 0.9845 - val_loss: 0.0789 - val_acc: 0.9801
Epoch 6/20
60000/60000 [==============================] - 1s - loss: 0.0429 - acc: 0.9874 - val_loss: 0.0918 - val_acc: 0.9796
Epoch 7/20
60000/60000 [==============================] - 1s - loss: 0.0367 - acc: 0.9889 - val_loss: 0.0879 - val_acc: 0.9803
Epoch 8/20
60000/60000 [==============================] - 1s - loss: 0.0336 - acc: 0.9900 - val_loss: 0.0799 - val_acc: 0.9828
Epoch 9/20
60000/60000 [==============================] - 1s - loss: 0.0324 - acc: 0.9907 - val_loss: 0.0896 - val_acc: 0.9826
Epoch 10/20
60000/60000 [==============================] - 1s - loss: 0.0285 - acc: 0.9913 - val_loss: 0.0860 - val_acc: 0.9829
Epoch 11/20
60000/60000 [==============================] - 1s - loss: 0.0265 - acc: 0.9923 - val_loss: 0.0994 - val_acc: 0.9822
Epoch 12/20
60000/60000 [==============================] - 1s - loss: 0.0237 - acc: 0.9933 - val_loss: 0.1013 - val_acc: 0.9844
Epoch 13/20
60000/60000 [==============================] - 1s - loss: 0.0226 - acc: 0.9933 - val_loss: 0.1026 - val_acc: 0.9818
Epoch 14/20
60000/60000 [==============================] - 1s - loss: 0.0229 - acc: 0.9938 - val_loss: 0.1056 - val_acc: 0.9830
Epoch 15/20
60000/60000 [==============================] - 1s - loss: 0.0219 - acc: 0.9942 - val_loss: 0.0991 - val_acc: 0.9825
Epoch 16/20
60000/60000 [==============================] - 1s - loss: 0.0210 - acc: 0.9943 - val_loss: 0.1119 - val_acc: 0.9827
Epoch 17/20
60000/60000 [==============================] - 1s - loss: 0.0206 - acc: 0.9948 - val_loss: 0.1041 - val_acc: 0.9837
Epoch 18/20
60000/60000 [==============================] - 1s - loss: 0.0206 - acc: 0.9947 - val_loss: 0.1147 - val_acc: 0.9836
Epoch 19/20
60000/60000 [==============================] - 1s - loss: 0.0179 - acc: 0.9953 - val_loss: 0.1231 - val_acc: 0.9807
Epoch 20/20
60000/60000 [==============================] - 1s - loss: 0.0174 - acc: 0.9955 - val_loss: 0.1126 - val_acc: 0.9823
Test loss: 0.112580380508
Test accuracy: 0.9823