lstm_cnn.py

import tensorflow as tf
import numpy as np
from IPython import embed

class LSTM_CNN(object):
    def __init__(self, sequence_length, num_classes, vocab_size, embedding_size, filter_sizes, num_filters, l2_reg_lambda=0.0,num_hidden=100):

        # PLACEHOLDERS
        self.input_x = tf.placeholder(tf.int32, [None, sequence_length], name="input_x")    # X - The Data
        self.input_y = tf.placeholder(tf.float32, [None, num_classes], name="input_y")      # Y - The Lables
        self.dropout_keep_prob = tf.placeholder(tf.float32, name="dropout_keep_prob")       # Dropout

        
        l2_loss = tf.constant(0.0) # Keeping track of l2 regularization loss

        #1. EMBEDDING LAYER ################################################################
        with tf.device('/cpu:0'), tf.name_scope("embedding"):
            self.W = tf.Variable(tf.random_uniform([vocab_size, embedding_size], -1.0, 1.0),name="W")
            self.embedded_chars = tf.nn.embedding_lookup(self.W, self.input_x)
            #self.embedded_chars_expanded = tf.expand_dims(self.embedded_chars, -1)


        #2. LSTM LAYER ######################################################################
        self.lstm_cell = tf.contrib.rnn.LSTMCell(32,state_is_tuple=True)
        #self.h_drop_exp = tf.expand_dims(self.h_drop,-1)
        self.lstm_out,self.lstm_state = tf.nn.dynamic_rnn(self.lstm_cell,self.embedded_chars,dtype=tf.float32)
        #embed()

        self.lstm_out_expanded = tf.expand_dims(self.lstm_out, -1)

        #2. CONVOLUTION LAYER + MAXPOOLING LAYER (per filter) ###############################
        pooled_outputs = []
        for i, filter_size in enumerate(filter_sizes):
            with tf.name_scope("conv-maxpool-%s" % filter_size):
                # CONVOLUTION LAYER
                filter_shape = [filter_size, embedding_size, 1, num_filters]
                W = tf.Variable(tf.truncated_normal(filter_shape, stddev=0.1), name="W")
                b = tf.Variable(tf.constant(0.1, shape=[num_filters]), name="b")
                conv = tf.nn.conv2d(self.lstm_out_expanded, W,strides=[1, 1, 1, 1],padding="VALID",name="conv")
                # NON-LINEARITY
                h = tf.nn.relu(tf.nn.bias_add(conv, b), name="relu")
                # MAXPOOLING
                pooled = tf.nn.max_pool(h, ksize=[1, sequence_length - filter_size + 1, 1, 1], strides=[1, 1, 1, 1], padding='VALID', name="pool")
                pooled_outputs.append(pooled)

        # COMBINING POOLED FEATURES
        num_filters_total = num_filters * len(filter_sizes)
        self.h_pool = tf.concat(pooled_outputs, 3)
        self.h_pool_flat = tf.reshape(self.h_pool, [-1, num_filters_total])
        
        # #3. DROPOUT LAYER ###################################################################
        with tf.name_scope("dropout"):
            self.h_drop = tf.nn.dropout(self.h_pool_flat, self.dropout_keep_prob)

        # Final (unnormalized) scores and predictions
        with tf.name_scope("output"):
            W = tf.get_variable(
                "W",
                shape=[num_filters_total, num_classes],
                initializer=tf.contrib.layers.xavier_initializer())
            b = tf.Variable(tf.constant(0.1, shape=[num_classes]), name="b")
            l2_loss += tf.nn.l2_loss(W)
            l2_loss += tf.nn.l2_loss(b)
            self.scores = tf.nn.xw_plus_b(self.h_drop, W, b, name="scores")
            self.predictions = tf.argmax(self.scores, 1, name="predictions")

        # CalculateMean cross-entropy loss
        with tf.name_scope("loss"):
            losses = tf.nn.softmax_cross_entropy_with_logits(logits=self.scores, labels=self.input_y)
            self.loss = tf.reduce_mean(losses) + l2_reg_lambda * l2_loss

        # Accuracy
        with tf.name_scope("accuracy"):
            correct_predictions = tf.equal(self.predictions, tf.argmax(self.input_y, 1))
            self.accuracy = tf.reduce_mean(tf.cast(correct_predictions, "float"), name="accuracy")


        print "(!!) LOADED LSTM-CNN! :)"
        #embed()



# 1. Embed --> LSTM
# 2. LSTM --> CNN
# 3. CNN --> Pooling/Output