本次通过RNN来实现Mnist,我给每部分代码加上了注释,下面直接上程序:
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets('Mnist_data', one_hot=True)
learning_rate = 0.001
training_iters = 10000
batch_size = 100
n_inputs = 28
n_steps = 28
n_hidden_units = 128
n_classes = 10
x = tf.placeholder(tf.float32, [None, n_steps, n_inputs])
y = tf.placeholder(tf.float32, [None, n_classes])
keep_drop = tf.placeholder(tf.float32)
weights = {
'in':tf.Variable(tf.random_normal([n_inputs, n_hidden_units])),
'out':tf.Variable(tf.random_normal([n_hidden_units, n_classes]))
}
biases = {
'in':tf.Variable(tf.constant(0.1, shape=[n_hidden_units,])),
'out':tf.Variable(tf.constant(0.1, shape=[n_classes,]))
}
def RNN(X, weights, biases):
X = tf.reshape(X, [-1, n_inputs])
X_into_Cell = tf.matmul(X, weights['in']) + biases['in']
X_into_Cell_drop = tf.nn.dropout(X_into_Cell, keep_drop)
X_into_Cell_drop = tf.reshape(X_into_Cell_drop, [-1, n_steps, n_hidden_units])
lstm_cell = tf.nn.rnn_cell.BasicLSTMCell(n_hidden_units, forget_bias=1.0, state_is_tuple=True)
__init__state = lstm_cell.zero_state(batch_size, dtype=tf.float32)
outputs, states = tf.nn.dynamic_rnn(lstm_cell, X_into_Cell_drop, initial_state=__init__state, time_major=False)
results = tf.matmul(states[1], weights['out']) + biases['out']
return results
prediction = RNN(x, weights, biases)
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=y, logits=prediction))
train_step = tf.train.AdamOptimizer(learning_rate).minimize(cost)
correct_pred = tf.equal(tf.argmax(prediction, 1), tf.argmax(y, 1))
accuracy = tf.reduce_mean(tf.cast(correct_pred, tf.float32))
with tf.Session() as sess:
sess.run(tf.initialize_all_variables())
counter = 10
for step in range(training_iters):
batch_xs, batch_ys = mnist.train.next_batch(batch_size)
batch_xs = batch_xs.reshape([batch_size, n_steps, n_inputs])
sess.run(train_step, feed_dict={x: batch_xs, y: batch_ys, keep_drop: 0.5})
if step % 1000 == 0:
counter -= 1
print('the remaining times trained is ', counter * 1000, '.')
print('the current accuracy is ', sess.run(accuracy, feed_dict={x: batch_xs, y: batch_ys, keep_drop: 0.5}), '.')
print()
saver = tf.train.Saver()
save_path = saver.save(sess, 'Mnist_paramater_RNN/save_parameter.ckpt')
print('The training is over!')