added Bidirectional RNN example

aymericdamien · aymericdamien · commit d77e5271b963 · 2016-01-04T17:40:28.000+08:00
diff --git a/README.md b/README.md
@@ -16,7 +16,8 @@ Code examples for some popular machine learning algorithms, using TensorFlow lib
 - Multilayer Perceptron ([notebook](https://github.com/aymericdamien/TensorFlow-Examples/blob/master/notebooks/3%20-%20Neural%20Networks/multilayer_perceptron.ipynb)) ([code](https://github.com/aymericdamien/TensorFlow-Examples/blob/master/examples/3%20-%20Neural%20Networks/multilayer_perceptron.py))
 - Convolutional Neural Network ([notebook](https://github.com/aymericdamien/TensorFlow-Examples/blob/master/notebooks/3%20-%20Neural%20Networks/convolutional_network.ipynb)) ([code](https://github.com/aymericdamien/TensorFlow-Examples/blob/master/examples/3%20-%20Neural%20Networks/convolutional_network.py))
 - AlexNet ([notebook](https://github.com/aymericdamien/TensorFlow-Examples/blob/master/notebooks/3%20-%20Neural%20Networks/alexnet.ipynb)) ([code](https://github.com/aymericdamien/TensorFlow-Examples/blob/master/examples/3%20-%20Neural%20Networks/alexnet.py))
-- Reccurent Network (LSTM) ([notebook](https://github.com/aymericdamien/TensorFlow-Examples/blob/master/notebooks/3%20-%20Neural%20Networks/reccurent_network.ipynb)) ([code](https://github.com/aymericdamien/TensorFlow-Examples/blob/master/examples/3%20-%20Neural%20Networks/recurrent_network.py))
+- Reccurent Neural Network (LSTM) ([notebook](https://github.com/aymericdamien/TensorFlow-Examples/blob/master/notebooks/3%20-%20Neural%20Networks/reccurent_network.ipynb)) ([code](https://github.com/aymericdamien/TensorFlow-Examples/blob/master/examples/3%20-%20Neural%20Networks/recurrent_network.py))
+- Bidirectional Reccurent Neural Network ([code](https://github.com/aymericdamien/TensorFlow-Examples/blob/master/examples/3%20-%20Neural%20Networks/bidirectional_rnn.py))
 
 #### 4 - Multi GPU
 - Basic Operations on multi-GPU ([notebook](https://github.com/aymericdamien/TensorFlow-Examples/blob/master/notebooks/4%20-%20Multi%20GPU/multigpu_basics.ipynb)) ([code](https://github.com/aymericdamien/TensorFlow-Examples/blob/master/examples/4%20-%20Multi%20GPU/multigpu_basics.py))
diff --git a/examples/3 - Neural Networks/bidirectional_rnn.py b/examples/3 - Neural Networks/bidirectional_rnn.py
@@ -0,0 +1,163 @@
+'''
+A Bidirectional Reccurent Neural Network (LSTM) implementation example using TensorFlow library.
+This example is using the MNIST database of handwritten digits (http://yann.lecun.com/exdb/mnist/)
+Long Short Term Memory paper: http://deeplearning.cs.cmu.edu/pdfs/Hochreiter97_lstm.pdf
+
+Author: Aymeric Damien
+Project: https://github.com/aymericdamien/TensorFlow-Examples/
+'''
+
+# Import MINST data
+import input_data
+mnist = input_data.read_data_sets("/tmp/data/", one_hot=True)
+
+import tensorflow as tf
+from tensorflow.python.ops.constant_op import constant
+from tensorflow.models.rnn import rnn, rnn_cell
+import numpy as np
+
+'''
+To classify images using a bidirectional reccurent neural network, we consider every image row as a sequence of pixels.
+Because MNIST image shape is 28*28px, we will then handle 28 sequences of 28 steps for every sample.
+'''
+
+# Parameters
+learning_rate = 0.001
+training_iters = 100000
+batch_size = 128
+display_step = 10
+
+# Network Parameters
+n_input = 28 # MNIST data input (img shape: 28*28)
+n_steps = 28 # timesteps
+n_hidden = 128 # hidden layer num of features
+n_classes = 10 # MNIST total classes (0-9 digits)
+
+# tf Graph input
+x = tf.placeholder("float", [None, n_steps, n_input])
+# Tensorflow LSTM cell requires 2x n_hidden length (state & cell)
+istate_fw = tf.placeholder("float", [None, 2*n_hidden])
+istate_bw = tf.placeholder("float", [None, 2*n_hidden])
+y = tf.placeholder("float", [None, n_classes])
+
+# Define weights
+weights = {
+    # Hidden layer weights => 2*n_hidden because of foward + backward cells
+    'hidden': tf.Variable(tf.random_normal([n_input, 2*n_hidden])),
+    'out': tf.Variable(tf.random_normal([2*n_hidden, n_classes]))
+}
+biases = {
+    'hidden': tf.Variable(tf.random_normal([2*n_hidden])),
+    'out': tf.Variable(tf.random_normal([n_classes]))
+}
+
+def BiRNN(_X, _istate_fw, _istate_bw, _weights, _biases, _batch_size, _seq_len):
+
+    # BiRNN requires to supply sequence_length as [batch_size, int64]
+    # Note: Tensorflow 0.6.0 requires BiRNN sequence_length parameter to be set
+    # For a better implementation with latest version of tensorflow, check below
+    _seq_len = tf.fill([_batch_size], constant(_seq_len, dtype=tf.int64))
+
+    # input shape: (batch_size, n_steps, n_input)
+    _X = tf.transpose(_X, [1, 0, 2])  # permute n_steps and batch_size
+    # Reshape to prepare input to hidden activation
+    _X = tf.reshape(_X, [-1, n_input]) # (n_steps*batch_size, n_input)
+    # Linear activation
+    _X = tf.matmul(_X, _weights['hidden']) + _biases['hidden']
+
+    # Define lstm cells with tensorflow
+    # Forward direction cell
+    lstm_fw_cell = rnn_cell.BasicLSTMCell(n_hidden, forget_bias=1.0)
+    # Backward direction cell
+    lstm_bw_cell = rnn_cell.BasicLSTMCell(n_hidden, forget_bias=1.0)
+    # Split data because rnn cell needs a list of inputs for the RNN inner loop
+    _X = tf.split(0, n_steps, _X) # n_steps * (batch_size, n_hidden)
+
+    # Get lstm cell output
+    outputs = rnn.bidirectional_rnn(lstm_fw_cell, lstm_bw_cell, _X,
+                                            initial_state_fw=_istate_fw,
+                                            initial_state_bw=_istate_bw,
+                                            sequence_length=_seq_len)
+
+    # Linear activation
+    # Get inner loop last output
+    return tf.matmul(outputs[-1], _weights['out']) + _biases['out']
+
+pred = BiRNN(x, istate_fw, istate_bw, weights, biases, batch_size, n_steps)
+
+
+# NOTE: The following code is working with current master version of tensorflow
+#       BiRNN sequence_length parameter isn't required, so we don't define it
+#
+# def BiRNN(_X, _istate_fw, _istate_bw, _weights, _biases):
+#
+#     # input shape: (batch_size, n_steps, n_input)
+#     _X = tf.transpose(_X, [1, 0, 2])  # permute n_steps and batch_size
+#     # Reshape to prepare input to hidden activation
+#     _X = tf.reshape(_X, [-1, n_input]) # (n_steps*batch_size, n_input)
+#     # Linear activation
+#     _X = tf.matmul(_X, _weights['hidden']) + _biases['hidden']
+#
+#     # Define lstm cells with tensorflow
+#     # Forward direction cell
+#     lstm_fw_cell = rnn_cell.BasicLSTMCell(n_hidden, forget_bias=1.0)
+#     # Backward direction cell
+#     lstm_bw_cell = rnn_cell.BasicLSTMCell(n_hidden, forget_bias=1.0)
+#     # Split data because rnn cell needs a list of inputs for the RNN inner loop
+#     _X = tf.split(0, n_steps, _X) # n_steps * (batch_size, n_hidden)
+#
+#     # Get lstm cell output
+#     outputs = rnn.bidirectional_rnn(lstm_fw_cell, lstm_bw_cell, _X,
+#                                             initial_state_fw=_istate_fw,
+#                                             initial_state_bw=_istate_bw)
+#
+#     # Linear activation
+#     # Get inner loop last output
+#     return tf.matmul(outputs[-1], _weights['out']) + _biases['out']
+#
+# pred = BiRNN(x, istate_fw, istate_bw, weights, biases)
+
+# Define loss and optimizer
+cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(pred, y)) # Softmax loss
+optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost) # Adam Optimizer
+
+# Evaluate model
+correct_pred = tf.equal(tf.argmax(pred,1), tf.argmax(y,1))
+accuracy = tf.reduce_mean(tf.cast(correct_pred, tf.float32))
+
+# Initializing the variables
+init = tf.initialize_all_variables()
+
+# Launch the graph
+with tf.Session() as sess:
+    sess.run(init)
+    step = 1
+    # Keep training until reach max iterations
+    while step * batch_size < training_iters:
+        batch_xs, batch_ys = mnist.train.next_batch(batch_size)
+        # Reshape data to get 28 seq of 28 elements
+        batch_xs = batch_xs.reshape((batch_size, n_steps, n_input))
+        # Fit training using batch data
+        sess.run(optimizer, feed_dict={x: batch_xs, y: batch_ys,
+                                       istate_fw: np.zeros((batch_size, 2*n_hidden)),
+                                       istate_bw: np.zeros((batch_size, 2*n_hidden))})
+        if step % display_step == 0:
+            # Calculate batch accuracy
+            acc = sess.run(accuracy, feed_dict={x: batch_xs, y: batch_ys,
+                                                istate_fw: np.zeros((batch_size, 2*n_hidden)),
+                                                istate_bw: np.zeros((batch_size, 2*n_hidden))})
+            # Calculate batch loss
+            loss = sess.run(cost, feed_dict={x: batch_xs, y: batch_ys,
+                                             istate_fw: np.zeros((batch_size, 2*n_hidden)),
+                                             istate_bw: np.zeros((batch_size, 2*n_hidden))})
+            print "Iter " + str(step*batch_size) + ", Minibatch Loss= " + "{:.6f}".format(loss) + \
+                  ", Training Accuracy= " + "{:.5f}".format(acc)
+        step += 1
+    print "Optimization Finished!"
+    # Calculate accuracy for 128 mnist test images
+    test_len = 128
+    test_data = mnist.test.images[:test_len].reshape((-1, n_steps, n_input))
+    test_label = mnist.test.labels[:test_len]
+    print "Testing Accuracy:", sess.run(accuracy, feed_dict={x: test_data, y: test_label,
+                                                             istate_fw: np.zeros((test_len, 2*n_hidden)),
+                                                             istate_bw: np.zeros((test_len, 2*n_hidden))})
