/
mnist_lstm.py
53 lines (47 loc) · 2.06 KB
/
mnist_lstm.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
from __future__ import print_function
from tensorflow.examples.tutorials.mnist import input_data
import tensorflow as tf
import tflearn
import numpy as np
import sys, os
import time
class TflearnLSTM():
def __init__(self, h_size=128, n_inputs=28, n_steps=28, n_classes=10, l_r=0.001):
# parameters init
l_r = l_r
self.n_inputs = n_inputs
self.n_steps = n_steps
n_classes = n_classes
self.model_dir = 'model/tflearn/lstm'
## build graph
tf.reset_default_graph()
tflearn.init_graph(gpu_memory_fraction=0.1)
X = tflearn.input_data(shape=[None, n_steps, n_inputs], name='input')
lstm = tflearn.lstm(X, h_size, dynamic=True, name='lstm')
dense = tflearn.fully_connected(lstm, n_classes, activation='softmax', name='dense')
classifier = tflearn.regression(dense, optimizer='adam', loss='categorical_crossentropy', metric='R2', learning_rate=l_r)
self.estimators = tflearn.DNN(classifier)
def fit(self, X_data, Y_data, n_epoch=10, batch_size=128):
self.estimators.fit(X_data, Y_data, n_epoch, show_metric=True, snapshot_epoch=False, batch_size=batch_size)
if not os.path.exists(self.model_dir):
os.makedirs(self.model_dir)
self.estimators.save('%s/model.ckpt' % self.model_dir)
print("Model saved in file: %s" % self.model_dir)
def predict(self, X_test):
self.estimators.load('%s/model.ckpt' % self.model_dir)
return self.estimators.predict(X_test)
def main():
#load mnist data
mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)
tfl_lstm = TflearnLSTM()
t1 = time.time()
tfl_lstm.fit(mnist.train.images.reshape(-1, 28, 28), mnist.train.labels)
t2 = time.time()
print('training time: %s' % (t2-t1))
pred = tfl_lstm.predict(mnist.test.images.reshape(-1, 28, 28))
t3 = time.time()
print('predict time: %s' % (t3-t2))
test_lab = mnist.test.labels
print("accuracy: ", np.mean(np.equal(np.argmax(pred,1), np.argmax(test_lab,1)))*100)
if __name__ == '__main__':
main()