neural-machine-translation/17.lstm-seq2seq-bahdanau.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from sklearn.utils import shuffle\n",
    "import re\n",
    "import time\n",
    "import collections\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_dataset(words, n_words, atleast=1):\n",
    "    count = [['PAD', 0], ['GO', 1], ['EOS', 2], ['UNK', 3]]\n",
    "    counter = collections.Counter(words).most_common(n_words)\n",
    "    counter = [i for i in counter if i[1] >= atleast]\n",
    "    count.extend(counter)\n",
    "    dictionary = dict()\n",
    "    for word, _ in count:\n",
    "        dictionary[word] = len(dictionary)\n",
    "    data = list()\n",
    "    unk_count = 0\n",
    "    for word in words:\n",
    "        index = dictionary.get(word, 0)\n",
    "        if index == 0:\n",
    "            unk_count += 1\n",
    "        data.append(index)\n",
    "    count[0][1] = unk_count\n",
    "    reversed_dictionary = dict(zip(dictionary.values(), dictionary.keys()))\n",
    "    return data, count, dictionary, reversed_dictionary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('english-train', 'r') as fopen:\n",
    "    text_from = fopen.read().lower().split('\\n')[:-1]\n",
    "with open('vietnam-train', 'r') as fopen:\n",
    "    text_to = fopen.read().lower().split('\\n')[:-1]\n",
    "print('len from: %d, len to: %d'%(len(text_from), len(text_to)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "vocab from size: 657\n",
      "Most common words [('you', 132), ('is', 78), ('i', 68), ('what', 51), ('it', 50), ('that', 49)]\n",
      "Sample data [7, 28, 129, 35, 61, 42, 12, 22, 82, 225] ['what', 'good', 'stuff', 'she', 'okay', 'they', 'do', 'to', 'hey', 'sweet']\n",
      "filtered vocab size: 661\n",
      "% of vocab used: 100.61%\n"
     ]
    }
   ],
   "source": [
    "concat_from = ' '.join(text_from).split()\n",
    "vocabulary_size_from = len(list(set(concat_from)))\n",
    "data_from, count_from, dictionary_from, rev_dictionary_from = build_dataset(concat_from, vocabulary_size_from)\n",
    "print('vocab from size: %d'%(vocabulary_size_from))\n",
    "print('Most common words', count_from[4:10])\n",
    "print('Sample data', data_from[:10], [rev_dictionary_from[i] for i in data_from[:10]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "vocab from size: 660\n",
      "Most common words [('i', 97), ('you', 91), ('is', 62), ('it', 58), ('not', 47), ('what', 39)]\n",
      "Sample data [12, 216, 5, 4, 94, 25, 59, 10, 8, 79] ['the', 'real', 'you', 'i', 'hope', 'so', 'they', 'do', 'not', 'hi']\n",
      "filtered vocab size: 664\n",
      "% of vocab used: 100.61%\n"
     ]
    }
   ],
   "source": [
    "concat_to = ' '.join(text_to).split()\n",
    "vocabulary_size_to = len(list(set(concat_to)))\n",
    "data_to, count_to, dictionary_to, rev_dictionary_to = build_dataset(concat_to, vocabulary_size_to)\n",
    "print('vocab to size: %d'%(vocabulary_size_to))\n",
    "print('Most common words', count_to[4:10])\n",
    "print('Sample data', data_to[:10], [rev_dictionary_to[i] for i in data_to[:10]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "GO = dictionary_from['GO']\n",
    "PAD = dictionary_from['PAD']\n",
    "EOS = dictionary_from['EOS']\n",
    "UNK = dictionary_from['UNK']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in range(len(text_to)):\n",
    "    text_to[i] += ' EOS'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Chatbot:\n",
    "    def __init__(self, size_layer, num_layers, embedded_size,\n",
    "                 from_dict_size, to_dict_size, learning_rate, batch_size):\n",
    "        \n",
    "        def cells(reuse=False):\n",
    "            return tf.nn.rnn_cell.LSTMCell(size_layer,initializer=tf.orthogonal_initializer(),\n",
    "                                           reuse=reuse)\n",
    "        \n",
    "        self.X = tf.placeholder(tf.int32, [None, None])\n",
    "        self.Y = tf.placeholder(tf.int32, [None, None])\n",
    "        self.X_seq_len = tf.placeholder(tf.int32, [None])\n",
    "        self.Y_seq_len = tf.placeholder(tf.int32, [None])\n",
    "        batch_size = tf.shape(self.X)[0]\n",
    "        \n",
    "        encoder_embeddings = tf.Variable(tf.random_uniform([from_dict_size, embedded_size], -1, 1))\n",
    "        decoder_embeddings = tf.Variable(tf.random_uniform([to_dict_size, embedded_size], -1, 1))\n",
    "        encoder_embedded = tf.nn.embedding_lookup(encoder_embeddings, self.X)\n",
    "        main = tf.strided_slice(self.X, [0, 0], [batch_size, -1], [1, 1])\n",
    "        decoder_input = tf.concat([tf.fill([batch_size, 1], GO), main], 1)\n",
    "        decoder_embedded = tf.nn.embedding_lookup(encoder_embeddings, decoder_input)\n",
    "        \n",
    "        attention_mechanism = tf.contrib.seq2seq.BahdanauAttention(num_units = size_layer, \n",
    "                                                                    memory = encoder_embedded)\n",
    "        rnn_cells = tf.contrib.seq2seq.AttentionWrapper(cell = tf.nn.rnn_cell.MultiRNNCell([cells() for _ in range(num_layers)]), \n",
    "                                                        attention_mechanism = attention_mechanism,\n",
    "                                                        attention_layer_size = size_layer)\n",
    "        _, last_state = tf.nn.dynamic_rnn(rnn_cells, encoder_embedded,\n",
    "                                          dtype = tf.float32)\n",
    "        last_state = tuple(last_state[0][-1] for _ in range(num_layers))\n",
    "        with tf.variable_scope(\"decoder\"):\n",
    "            rnn_cells_dec = tf.nn.rnn_cell.MultiRNNCell([cells() for _ in range(num_layers)])\n",
    "            outputs, _ = tf.nn.dynamic_rnn(rnn_cells_dec, decoder_embedded, \n",
    "                                           initial_state = last_state,\n",
    "                                           dtype = tf.float32)\n",
    "        self.logits = tf.layers.dense(outputs,to_dict_size)\n",
    "        masks = tf.sequence_mask(self.Y_seq_len, tf.reduce_max(self.Y_seq_len), dtype=tf.float32)\n",
    "        self.cost = tf.contrib.seq2seq.sequence_loss(logits = self.logits,\n",
    "                                                     targets = self.Y,\n",
    "                                                     weights = masks)\n",
    "        self.optimizer = tf.train.AdamOptimizer(learning_rate = learning_rate).minimize(self.cost)\n",
    "        y_t = tf.argmax(self.logits,axis=2)\n",
    "        y_t = tf.cast(y_t, tf.int32)\n",
    "        self.prediction = tf.boolean_mask(y_t, masks)\n",
    "        mask_label = tf.boolean_mask(self.Y, masks)\n",
    "        correct_pred = tf.equal(self.prediction, mask_label)\n",
    "        correct_index = tf.cast(correct_pred, tf.float32)\n",
    "        self.accuracy = tf.reduce_mean(tf.cast(correct_pred, tf.float32))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "size_layer = 128\n",
    "num_layers = 2\n",
    "embedded_size = 128\n",
    "learning_rate = 0.001\n",
    "batch_size = 16\n",
    "epoch = 20"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "sess = tf.InteractiveSession()\n",
    "model = Chatbot(size_layer, num_layers, embedded_size, len(dictionary_from), \n",
    "                len(dictionary_to), learning_rate,batch_size)\n",
    "sess.run(tf.global_variables_initializer())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def str_idx(corpus, dic):\n",
    "    X = []\n",
    "    for i in corpus:\n",
    "        ints = []\n",
    "        for k in i.split():\n",
    "            ints.append(dic.get(k,UNK))\n",
    "        X.append(ints)\n",
    "    return X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = str_idx(text_from, dictionary_from)\n",
    "Y = str_idx(text_to, dictionary_to)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10, 12)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "maxlen_question = max([len(x) for x in X]) * 2\n",
    "maxlen_answer = max([len(y) for y in Y]) * 2\n",
    "\n",
    "maxlen_question, maxlen_answer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def pad_sentence_batch(sentence_batch, pad_int, maxlen):\n",
    "    padded_seqs = []\n",
    "    seq_lens = []\n",
    "    max_sentence_len = maxlen\n",
    "    for sentence in sentence_batch:\n",
    "        padded_seqs.append(sentence + [pad_int] * (max_sentence_len - len(sentence)))\n",
    "        seq_lens.append(maxlen)\n",
    "    return padded_seqs, seq_lens"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch: 1, avg loss: 3.317611, avg accuracy: 0.620000\n",
      "epoch: 2, avg loss: 2.050228, avg accuracy: 0.660167\n",
      "epoch: 3, avg loss: 1.981821, avg accuracy: 0.673667\n",
      "epoch: 4, avg loss: 1.960332, avg accuracy: 0.671333\n",
      "epoch: 5, avg loss: 1.922090, avg accuracy: 0.674333\n",
      "epoch: 6, avg loss: 1.912903, avg accuracy: 0.675667\n",
      "epoch: 7, avg loss: 1.893768, avg accuracy: 0.674167\n",
      "epoch: 8, avg loss: 1.877391, avg accuracy: 0.676833\n",
      "epoch: 9, avg loss: 1.846027, avg accuracy: 0.682833\n",
      "epoch: 10, avg loss: 1.825397, avg accuracy: 0.686333\n",
      "epoch: 11, avg loss: 1.801734, avg accuracy: 0.691333\n",
      "epoch: 12, avg loss: 1.796578, avg accuracy: 0.694333\n",
      "epoch: 13, avg loss: 1.764410, avg accuracy: 0.700667\n",
      "epoch: 14, avg loss: 1.743519, avg accuracy: 0.703667\n",
      "epoch: 15, avg loss: 1.718304, avg accuracy: 0.703333\n",
      "epoch: 16, avg loss: 1.686815, avg accuracy: 0.713833\n",
      "epoch: 17, avg loss: 1.670385, avg accuracy: 0.714833\n",
      "epoch: 18, avg loss: 1.651643, avg accuracy: 0.717167\n",
      "epoch: 19, avg loss: 1.631121, avg accuracy: 0.721833\n",
      "epoch: 20, avg loss: 1.629793, avg accuracy: 0.721833\n"
     ]
    }
   ],
   "source": [
    "for i in range(epoch):\n",
    "    total_loss, total_accuracy = 0, 0\n",
    "    X, Y = shuffle(X, Y)\n",
    "    for k in range(0, len(text_to), batch_size):\n",
    "        index = min(k + batch_size, len(text_to))\n",
    "        batch_x, seq_x = pad_sentence_batch(X[k: index], PAD, maxlen_answer)\n",
    "        batch_y, seq_y = pad_sentence_batch(Y[k: index], PAD, maxlen_answer)\n",
    "        predicted, accuracy, loss, _ = sess.run([tf.argmax(model.logits,2),\n",
    "                                      model.accuracy, model.cost, model.optimizer], \n",
    "                                      feed_dict={model.X:batch_x,\n",
    "                                                model.Y:batch_y,\n",
    "                                                model.X_seq_len:seq_x,\n",
    "                                                model.Y_seq_len:seq_y})\n",
    "        total_loss += loss\n",
    "        total_accuracy += accuracy\n",
    "    total_loss /= (len(text_to) / batch_size)\n",
    "    total_accuracy /= (len(text_to) / batch_size)\n",
    "    print('epoch: %d, avg loss: %f, avg accuracy: %f'%(i+1, total_loss, total_accuracy))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "row 1\n",
      "QUESTION: yes my love\n",
      "REAL ANSWER: these gentlemen are from salzburg\n",
      "PREDICTED ANSWER: i is not it \n",
      "\n",
      "row 2\n",
      "QUESTION: so which dakota you from\n",
      "REAL ANSWER: north actually how would you\n",
      "PREDICTED ANSWER: i is not it \n",
      "\n",
      "row 3\n",
      "QUESTION: i used an accelerant\n",
      "REAL ANSWER: yeah what kind\n",
      "PREDICTED ANSWER: i is \n",
      "\n",
      "row 4\n",
      "QUESTION: what anger\n",
      "REAL ANSWER: about the ballet\n",
      "PREDICTED ANSWER: i is \n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(batch_x)):\n",
    "    print('row %d'%(i+1))\n",
    "    print('QUESTION:',' '.join([rev_dictionary_from[n] for n in batch_x[i] if n not in [0,1,2,3]]))\n",
    "    print('REAL ANSWER:',' '.join([rev_dictionary_to[n] for n in batch_y[i] if n not in[0,1,2,3]]))\n",
    "    print('PREDICTED ANSWER:',' '.join([rev_dictionary_to[n] for n in predicted[i] if n not in[0,1,2,3]]),'\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}