Intent_classification_using_Deep_Learning.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Intent Classification using Deep Learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Loading Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "a_WypuUXi92e",
    "outputId": "133d026e-4236-4ff6-f21d-739bfb9640db"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.tokenize import word_tokenize\n",
    "from nltk.stem.lancaster import LancasterStemmer\n",
    "import nltk\n",
    "import re\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "import matplotlib.pyplot as plt\n",
    "from keras.preprocessing.text import Tokenizer\n",
    "from keras.preprocessing.sequence import pad_sequences\n",
    "from keras.utils import to_categorical\n",
    "from keras.models import Sequential, load_model\n",
    "from keras.layers import Dense, GRU, LSTM, Bidirectional, Embedding, Dropout\n",
    "from keras.callbacks import ModelCheckpoint"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Dataset link: https://raw.githubusercontent.com/PolyAI-LDN/task-specific-datasets/master/banking_data/train.csv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "Intent = \"category\"\n",
    "Sentence = \"text\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "LE6wywJrN2ih"
   },
   "outputs": [],
   "source": [
    "def load_dataset(filename, Sentence, Intent):\n",
    "  df = pd.read_csv(filename, names = [Sentence, Intent])\n",
    "  intent = df[Intent]\n",
    "  unique_intent = list(set(intent))\n",
    "  sentences = list(df[Sentence])\n",
    "  \n",
    "  return (df, intent, unique_intent, sentences)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 119
    },
    "colab_type": "code",
    "id": "tF0FQA7gjOCX",
    "outputId": "c609b42a-05da-49f5-8d11-bd670210f635"
   },
   "outputs": [],
   "source": [
    "df, intent, unique_intent, sentences = load_dataset(\"Dataset.csv\", \"text\", \"category\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                                text      category\n",
      "0                                               text      category\n",
      "1                     I am still waiting on my card?  card_arrival\n",
      "2  What can I do if my card still hasn't arrived ...  card_arrival\n",
      "3  I have been waiting over a week. Is the card s...  card_arrival\n",
      "4  Can I track my card while it is in the process...  card_arrival\n",
      "5  How do I know if I will get my card, or if it ...  card_arrival\n",
      "6                  When did you send me my new card?  card_arrival\n",
      "7       Do you have info about the card on delivery?  card_arrival\n",
      "8  What do I do if I still have not received my n...  card_arrival\n",
      "9       Does the package with my card have tracking?  card_arrival\n"
     ]
    }
   ],
   "source": [
    "print(df.head(10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='category', ylabel='count'>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "import tkinter\n",
    "import matplotlib\n",
    "matplotlib.use('agg')\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "sns.countplot(x=Intent, data=df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "O8LLUZlokg0S",
    "outputId": "c15c21dc-2ef2-43b7-b4af-e7ee9e014091"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['text', 'I am still waiting on my card?', \"What can I do if my card still hasn't arrived after 2 weeks?\", 'I have been waiting over a week. Is the card still coming?', 'Can I track my card while it is in the process of delivery?']\n"
     ]
    }
   ],
   "source": [
    "print(sentences[:5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 102
    },
    "colab_type": "code",
    "id": "MhrziINPGHbW",
    "outputId": "0861af1b-4b82-4c92-b8f4-b6b57bb3e380"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to\n",
      "[nltk_data]     /home/shiningflash/nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n",
      "[nltk_data] Downloading package punkt to\n",
      "[nltk_data]     /home/shiningflash/nltk_data...\n",
      "[nltk_data]   Package punkt is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nltk.download(\"stopwords\")\n",
    "nltk.download(\"punkt\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Data Cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "OmNLu2YSXePb"
   },
   "outputs": [],
   "source": [
    "#define stemmer\n",
    "stemmer = LancasterStemmer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "j-7q3iG5PKYI"
   },
   "outputs": [],
   "source": [
    "def cleaning(sentences):\n",
    "  words = []\n",
    "  for s in sentences:\n",
    "    clean = re.sub(r'[^ a-z A-Z 0-9]', \" \", s)\n",
    "    w = word_tokenize(clean)\n",
    "    words.append([i.lower() for i in w])\n",
    "    \n",
    "  return words "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 51
    },
    "colab_type": "code",
    "id": "p1j2GJgDG6qj",
    "outputId": "c7232a8e-6833-4a1d-e71a-4bc7014084a9"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10004\n",
      "[['text'], ['i', 'am', 'still', 'waiting', 'on', 'my', 'card']]\n"
     ]
    }
   ],
   "source": [
    "cleaned_words = cleaning(sentences)\n",
    "print(len(cleaned_words))\n",
    "print(cleaned_words[:2])  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Texts Tokenization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "SJCQ_YhBJW7t"
   },
   "outputs": [],
   "source": [
    "def create_tokenizer(words, filters = '!\"#$%&()*+,-./:;<=>?@[\\]^_`{|}~'):\n",
    "  token = Tokenizer(filters = filters)\n",
    "  token.fit_on_texts(words)\n",
    "  return token"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "QJhdIJC5Q3Q6"
   },
   "outputs": [],
   "source": [
    "def max_length(words):\n",
    "  return(len(max(words, key = len)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "JWjxPGsZZJNX",
    "outputId": "b02c8f6b-d0df-4e90-fa3a-2ff730c88300"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Vocab Size = 2343 and Maximum length = 84\n"
     ]
    }
   ],
   "source": [
    "word_tokenizer = create_tokenizer(cleaned_words)\n",
    "vocab_size = len(word_tokenizer.word_index) + 1\n",
    "max_length = max_length(cleaned_words)\n",
    "\n",
    "print(\"Vocab Size = %d and Maximum length = %d\" % (vocab_size, max_length))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "D0TXu2xsR8jq"
   },
   "outputs": [],
   "source": [
    "def encoding_doc(token, words):\n",
    "  return(token.texts_to_sequences(words))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "dE92Hk1Va--H"
   },
   "outputs": [],
   "source": [
    "encoded_doc = encoding_doc(word_tokenizer, cleaned_words)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "fyOzLEboc4LZ"
   },
   "outputs": [],
   "source": [
    "def padding_doc(encoded_doc, max_length):\n",
    "  return(pad_sequences(encoded_doc, maxlen = max_length, padding = \"post\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "WdejoJrlc-tc"
   },
   "outputs": [],
   "source": [
    "padded_doc = padding_doc(encoded_doc, max_length)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "3eaSIDi0dNf1",
    "outputId": "4ab6b6dd-ffa4-4061-9e9d-7a01decfa837"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of padded docs =  (10004, 84)\n"
     ]
    }
   ],
   "source": [
    "print(\"Shape of padded docs = \",padded_doc.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "X0rXzenSpgFR"
   },
   "outputs": [],
   "source": [
    "#tokenizer with filter changed\n",
    "output_tokenizer = create_tokenizer(unique_intent, filters = '!\"#$%&()*+,-/:;<=>?@[\\]^`{|}~')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 374
    },
    "colab_type": "code",
    "id": "yNHQtkszskxr",
    "outputId": "f5babc01-89e3-4392-e8e6-c9f257de3d07"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'verify_source_of_funds': 1,\n",
       " 'topping_up_by_card': 2,\n",
       " 'balance_not_updated_after_cheque_or_cash_deposit': 3,\n",
       " 'visa_or_mastercard': 4,\n",
       " 'request_refund': 5,\n",
       " 'extra_charge_on_statement': 6,\n",
       " 'wrong_amount_of_cash_received': 7,\n",
       " 'atm_support': 8,\n",
       " 'card_payment_fee_charged': 9,\n",
       " 'pending_card_payment': 10,\n",
       " 'transfer_fee_charged': 11,\n",
       " 'getting_spare_card': 12,\n",
       " 'wrong_exchange_rate_for_cash_withdrawal': 13,\n",
       " 'top_up_by_cash_or_cheque': 14,\n",
       " 'category': 15,\n",
       " 'transfer_not_received_by_recipient': 16,\n",
       " 'top_up_by_bank_transfer_charge': 17,\n",
       " 'unable_to_verify_identity': 18,\n",
       " 'refund_not_showing_up': 19,\n",
       " 'pending_transfer': 20,\n",
       " 'pin_blocked': 21,\n",
       " 'top_up_by_card_charge': 22,\n",
       " 'country_support': 23,\n",
       " 'getting_virtual_card': 24,\n",
       " 'pending_cash_withdrawal': 25,\n",
       " 'card_payment_wrong_exchange_rate': 26,\n",
       " 'declined_cash_withdrawal': 27,\n",
       " 'fiat_currency_support': 28,\n",
       " 'lost_or_stolen_phone': 29,\n",
       " 'transfer_into_account': 30,\n",
       " 'passcode_forgotten': 31,\n",
       " 'terminate_account': 32,\n",
       " 'verify_top_up': 33,\n",
       " 'top_up_failed': 34,\n",
       " 'declined_transfer': 35,\n",
       " 'verify_my_identity': 36,\n",
       " 'card_acceptance': 37,\n",
       " 'direct_debit_payment_not_recognised': 38,\n",
       " 'declined_card_payment': 39,\n",
       " 'beneficiary_not_allowed': 40,\n",
       " 'cash_withdrawal_charge': 41,\n",
       " 'card_swallowed': 42,\n",
       " 'automatic_top_up': 43,\n",
       " 'supported_cards_and_currencies': 44,\n",
       " 'compromised_card': 45,\n",
       " 'card_arrival': 46,\n",
       " 'virtual_card_not_working': 47,\n",
       " 'change_pin': 48,\n",
       " 'transfer_timing': 49,\n",
       " 'reverted_card_payment': 50,\n",
       " 'get_disposable_virtual_card': 51,\n",
       " 'balance_not_updated_after_bank_transfer': 52,\n",
       " 'exchange_charge': 53,\n",
       " 'card_not_working': 54,\n",
       " 'lost_or_stolen_card': 55,\n",
       " 'card_payment_not_recognised': 56,\n",
       " 'why_verify_identity': 57,\n",
       " 'contactless_not_working': 58,\n",
       " 'disposable_card_limits': 59,\n",
       " 'activate_my_card': 60,\n",
       " 'card_about_to_expire': 61,\n",
       " 'apple_pay_or_google_pay': 62,\n",
       " 'card_delivery_estimate': 63,\n",
       " 'exchange_rate': 64,\n",
       " 'edit_personal_details': 65,\n",
       " 'card_linking': 66,\n",
       " 'get_physical_card': 67,\n",
       " 'cash_withdrawal_not_recognised': 68,\n",
       " 'order_physical_card': 69,\n",
       " 'transaction_charged_twice': 70,\n",
       " 'cancel_transfer': 71,\n",
       " 'top_up_reverted': 72,\n",
       " 'age_limit': 73,\n",
       " 'failed_transfer': 74,\n",
       " 'receiving_money': 75,\n",
       " 'top_up_limits': 76,\n",
       " 'exchange_via_app': 77,\n",
       " 'pending_top_up': 78}"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "output_tokenizer.word_index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "7OOx9qdBto1-"
   },
   "outputs": [],
   "source": [
    "encoded_output = encoding_doc(output_tokenizer, intent)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "0_5Lv5PiyG-z"
   },
   "outputs": [],
   "source": [
    "encoded_output = np.array(encoded_output).reshape(len(encoded_output), 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "dpM86WrVQlx5",
    "outputId": "71ff52a6-b3d0-4b5c-850d-5dc0a56c8aa9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10004, 1)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoded_output.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "rD3QN-RPzfet"
   },
   "outputs": [],
   "source": [
    "def one_hot(encode):\n",
    "  o = OneHotEncoder(sparse = False)\n",
    "  return(o.fit_transform(encode))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Z6wP_Xed7RNR"
   },
   "outputs": [],
   "source": [
    "output_one_hot = one_hot(encoded_output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "A6HVslLTHgOM",
    "outputId": "752962df-02d8-409b-fb8f-adb06227161d"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10004, 78)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "output_one_hot.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "EqABUESD7xi9"
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "h8P4HTz6A4E-"
   },
   "outputs": [],
   "source": [
    "train_X, val_X, train_Y, val_Y = train_test_split(padded_doc, output_one_hot, shuffle = True, test_size = 0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 51
    },
    "colab_type": "code",
    "id": "7E0uhC2OCtTx",
    "outputId": "6ce0e215-aa3f-43f1-ba5a-0b584b25a35c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of train_X = (8003, 84) and train_Y = (8003, 78)\n",
      "Shape of val_X = (2001, 84) and val_Y = (2001, 78)\n"
     ]
    }
   ],
   "source": [
    "print(\"Shape of train_X = %s and train_Y = %s\" % (train_X.shape, train_Y.shape))\n",
    "print(\"Shape of val_X = %s and val_Y = %s\" % (val_X.shape, val_Y.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Bidirectional GRU "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "from livelossplot.tf_keras import PlotLossesCallback\n",
    "from livelossplot import PlotLossesKeras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "e5BU_x74DNEb"
   },
   "outputs": [],
   "source": [
    "def create_model(vocab_size, max_length):\n",
    "  model = Sequential()\n",
    "  model.add(Embedding(vocab_size, 128, input_length = max_length, trainable = False))\n",
    "  model.add(Bidirectional(GRU(128)))\n",
    "  model.add(Dense(32, activation = \"relu\"))\n",
    "  model.add(Dropout(0.5))\n",
    "  model.add(Dense(78, activation = \"softmax\"))\n",
    "  \n",
    "  return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 306
    },
    "colab_type": "code",
    "id": "f-NvE0P7MFCe",
    "outputId": "8f07056b-579e-4c15-e1af-bdfa8f681e79",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "embedding (Embedding)        (None, 84, 128)           299904    \n",
      "_________________________________________________________________\n",
      "bidirectional (Bidirectional (None, 256)               198144    \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 32)                8224      \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 32)                0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 78)                2574      \n",
      "=================================================================\n",
      "Total params: 508,846\n",
      "Trainable params: 208,942\n",
      "Non-trainable params: 299,904\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model = create_model(vocab_size, max_length)\n",
    "\n",
    "model.compile(loss = \"categorical_crossentropy\", optimizer = \"adam\", metrics = [\"accuracy\"])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 6834
    },
    "colab_type": "code",
    "id": "_r-dxm2sMQ-d",
    "outputId": "3c37b4f8-fc4e-4c82-ab46-2aa1d8b47ffd"
   },
   "outputs": [],
   "source": [
    "filename = 'model.h5'\n",
    "checkpoint = ModelCheckpoint(filename,\n",
    "                             monitor='val_loss',\n",
    "                             verbose=1,\n",
    "                             save_best_only=True,\n",
    "                             mode='min')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy\n",
      "\ttraining         \t (min:    0.018, max:    0.807, cur:    0.803)\n",
      "\tvalidation       \t (min:    0.025, max:    0.694, cur:    0.681)\n",
      "Loss\n",
      "\ttraining         \t (min:    0.562, max:    4.331, cur:    0.568)\n",
      "\tvalidation       \t (min:    1.468, max:    4.558, cur:    2.376)\n",
      "\n",
      "Epoch 00100: val_loss did not improve from 1.46752\n",
      "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
      "251/251 [==============================] - 27s 107ms/step - loss: 0.5683 - accuracy: 0.8031 - val_loss: 2.3763 - val_accuracy: 0.6812\n"
     ]
    }
   ],
   "source": [
    "hist = model.fit(train_X, train_Y,\n",
    "                 epochs = 100,\n",
    "                 batch_size = 32,\n",
    "                 validation_data = (val_X, val_Y),\n",
    "                 callbacks = [PlotLossesKeras(), checkpoint])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5. Bidirectional LSTM "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "embedding_1 (Embedding)      (None, 84, 128)           299904    \n",
      "_________________________________________________________________\n",
      "bidirectional_1 (Bidirection (None, 256)               263168    \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 32)                8224      \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 32)                0         \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 78)                2574      \n",
      "=================================================================\n",
      "Total params: 573,870\n",
      "Trainable params: 273,966\n",
      "Non-trainable params: 299,904\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "def create_model(vocab_size, max_length):\n",
    "  model = Sequential()\n",
    "  model.add(Embedding(vocab_size, 128, input_length = max_length, trainable = False))\n",
    "  model.add(Bidirectional(LSTM(128)))\n",
    "  model.add(Dense(32, activation = \"relu\"))\n",
    "  model.add(Dropout(0.5))\n",
    "  model.add(Dense(78, activation = \"softmax\"))\n",
    "  \n",
    "  return model\n",
    "\n",
    "model_lstm = create_model(vocab_size, max_length)\n",
    "\n",
    "model_lstm.compile(loss = \"categorical_crossentropy\", optimizer = \"adam\", metrics = [\"accuracy\"])\n",
    "model_lstm.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy\n",
      "\ttraining         \t (min:    0.017, max:    0.774, cur:    0.768)\n",
      "\tvalidation       \t (min:    0.034, max:    0.711, cur:    0.688)\n",
      "Loss\n",
      "\ttraining         \t (min:    0.651, max:    4.339, cur:    0.695)\n",
      "\tvalidation       \t (min:    1.337, max:    4.285, cur:    1.775)\n",
      "\n",
      "Epoch 00100: val_loss did not improve from 1.33676\n",
      "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
      "251/251 [==============================] - 27s 108ms/step - loss: 0.6949 - accuracy: 0.7677 - val_loss: 1.7749 - val_accuracy: 0.6882\n"
     ]
    }
   ],
   "source": [
    "filename = 'model.h5'\n",
    "checkpoint = ModelCheckpoint(filename,\n",
    "                             monitor='val_loss',\n",
    "                             verbose=1,\n",
    "                             save_best_only=True,\n",
    "                             mode='min')\n",
    "\n",
    "hist = model_lstm.fit(train_X, train_Y,\n",
    "                 epochs = 100,\n",
    "                 batch_size = 32,\n",
    "                 validation_data = (val_X, val_Y),\n",
    "                 callbacks = [PlotLossesKeras(), checkpoint])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "YjXKos8ocXvw"
   },
   "outputs": [],
   "source": [
    "model_lstm = load_model(\"model.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "qSTEzrlzcuya"
   },
   "outputs": [],
   "source": [
    "def predictions(text):\n",
    "  clean = re.sub(r'[^ a-z A-Z 0-9]', \" \", text)\n",
    "  test_word = word_tokenize(clean)\n",
    "  test_word = [w.lower() for w in test_word]\n",
    "  test_ls = word_tokenizer.texts_to_sequences(test_word)\n",
    "  \n",
    "  #Check for unknown words\n",
    "  if [] in test_ls:\n",
    "    test_ls = list(filter(None, test_ls))\n",
    "    \n",
    "  test_ls = np.array(test_ls).reshape(1, len(test_ls))\n",
    "  x = padding_doc(test_ls, max_length)\n",
    "\n",
    "  pred = model_lstm.predict(x)\n",
    "  return pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "P1ddofshmdzK"
   },
   "outputs": [],
   "source": [
    "def get_final_output(pred, classes):\n",
    "  predictions = pred[0]\n",
    "  classes = np.array(classes)\n",
    "  ids = np.argsort(-predictions)\n",
    "  classes = classes[ids]\n",
    "  predictions = -np.sort(-predictions)\n",
    " \n",
    "  for i in range(pred.shape[1]):\n",
    "    print(\"%s has confidence = %s\" % (classes[i], (predictions[i])))\n",
    "  \n",
    "  return classes[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 391
    },
    "colab_type": "code",
    "id": "23VpGuihMdEU",
    "outputId": "cd36c932-0fb0-4166-92ae-546a7676e645"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "card_arrival has confidence = 0.5400851\n",
      "lost_or_stolen_card has confidence = 0.13254714\n",
      "card_not_working has confidence = 0.099424854\n",
      "contactless_not_working has confidence = 0.05694259\n",
      "compromised_card has confidence = 0.02806151\n",
      "card_about_to_expire has confidence = 0.024988813\n",
      "card_delivery_estimate has confidence = 0.024908025\n",
      "declined_card_payment has confidence = 0.023217054\n",
      "card_linking has confidence = 0.019666543\n",
      "activate_my_card has confidence = 0.019631721\n",
      "card_swallowed has confidence = 0.015140002\n",
      "reverted_card_payment? has confidence = 0.011102826\n",
      "atm_support has confidence = 0.0015607645\n",
      "declined_cash_withdrawal has confidence = 0.0007894527\n",
      "lost_or_stolen_phone has confidence = 0.00065130007\n",
      "card_payment_not_recognised has confidence = 0.00042219783\n",
      "cash_withdrawal_not_recognised has confidence = 0.0002289453\n",
      "card_acceptance has confidence = 0.0001782307\n",
      "declined_transfer has confidence = 0.00015152607\n",
      "pin_blocked has confidence = 9.299157e-05\n",
      "passcode_forgotten has confidence = 5.496411e-05\n",
      "order_physical_card has confidence = 3.1216285e-05\n",
      "unable_to_verify_identity has confidence = 2.856871e-05\n",
      "transaction_charged_twice has confidence = 1.5831929e-05\n",
      "beneficiary_not_allowed has confidence = 1.5821604e-05\n",
      "country_support has confidence = 1.3478491e-05\n",
      "pending_card_payment has confidence = 1.0269475e-05\n",
      "why_verify_identity has confidence = 9.724789e-06\n",
      "receiving_money has confidence = 6.8202044e-06\n",
      "direct_debit_payment_not_recognised has confidence = 5.4109573e-06\n",
      "virtual_card_not_working has confidence = 2.912666e-06\n",
      "supported_cards_and_currencies has confidence = 2.4180426e-06\n",
      "pending_cash_withdrawal has confidence = 2.1170988e-06\n",
      "topping_up_by_card has confidence = 2.0330606e-06\n",
      "request_refund has confidence = 1.1766249e-06\n",
      "transfer_not_received_by_recipient has confidence = 1.0827664e-06\n",
      "card_payment_fee_charged has confidence = 1.0126768e-06\n",
      "visa_or_mastercard has confidence = 6.679286e-07\n",
      "terminate_account has confidence = 5.908272e-07\n",
      "Refund_not_showing_up has confidence = 3.619633e-07\n",
      "top_up_failed has confidence = 3.1810998e-07\n",
      "cancel_transfer has confidence = 2.8714103e-07\n",
      "verify_top_up has confidence = 2.622885e-07\n",
      "category has confidence = 2.5423205e-07\n",
      "getting_spare_card has confidence = 1.889249e-07\n",
      "transfer_fee_charged has confidence = 1.5563825e-07\n",
      "age_limit has confidence = 1.2111296e-07\n",
      "pending_top_up has confidence = 6.930501e-08\n",
      "get_physical_card has confidence = 5.6756654e-08\n",
      "getting_virtual_card has confidence = 5.3940063e-08\n",
      "top_up_by_cash_or_cheque has confidence = 4.5198856e-08\n",
      "failed_transfer has confidence = 1.5451935e-08\n",
      "edit_personal_details has confidence = 1.4790701e-08\n",
      "exchange_rate has confidence = 1.3748348e-08\n",
      "wrong_amount_of_cash_received has confidence = 1.2084236e-08\n",
      "top_up_reverted has confidence = 1.031402e-08\n",
      "change_pin has confidence = 7.595862e-09\n",
      "verify_my_identity has confidence = 5.57865e-09\n",
      "balance_not_updated_after_bank_transfer has confidence = 2.6290028e-09\n",
      "verify_source_of_funds has confidence = 2.390937e-09\n",
      "automatic_top_up has confidence = 1.8031706e-09\n",
      "card_payment_wrong_exchange_rate has confidence = 1.7656182e-09\n",
      "top_up_by_card_charge has confidence = 1.2948175e-09\n",
      "wrong_exchange_rate_for_cash_withdrawal has confidence = 9.181001e-10\n",
      "fiat_currency_support has confidence = 8.16949e-10\n",
      "balance_not_updated_after_cheque_or_cash_deposit has confidence = 6.584936e-10\n",
      "transfer_timing has confidence = 4.593117e-10\n",
      "apple_pay_or_google_pay has confidence = 2.880023e-10\n",
      "top_up_by_bank_transfer_charge has confidence = 2.7394342e-10\n",
      "cash_withdrawal_charge has confidence = 1.2649745e-10\n",
      "exchange_charge has confidence = 9.780896e-11\n",
      "extra_charge_on_statement has confidence = 8.268214e-11\n",
      "transfer_into_account has confidence = 5.788913e-11\n",
      "get_disposable_virtual_card has confidence = 4.7355724e-11\n",
      "pending_transfer has confidence = 1.560852e-11\n",
      "exchange_via_app has confidence = 5.856411e-12\n",
      "disposable_card_limits has confidence = 1.6419753e-12\n",
      "top_up_limits has confidence = 7.180127e-14\n",
      "\n",
      "ans: card_arrival\n",
      "\n"
     ]
    }
   ],
   "source": [
    "text = \"I am still waiting on my card?\"\n",
    "pred = predictions(text)\n",
    "result = get_final_output(pred, unique_intent)\n",
    "print('\\nans: {}\\n'.format(result))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_only_final_output(pred, classes):\n",
    "  predictions = pred[0]\n",
    "  classes = np.array(classes)\n",
    "  ids = np.argsort(-predictions)\n",
    "  classes = classes[ids]\n",
    "  predictions = -np.sort(-predictions)\n",
    "  \n",
    "  return classes[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_intent(text):\n",
    "  pred = predictions(text)\n",
    "  result = get_only_final_output(pred, unique_intent)\n",
    "  print('ans: {}'.format(result))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ans: card_arrival\n"
     ]
    }
   ],
   "source": [
    "get_intent(\"I am still waiting on my card?\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ans: exchange_rate\n"
     ]
    }
   ],
   "source": [
    "get_intent(\"What are you exchange rates?\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ans: country_support\n"
     ]
    }
   ],
   "source": [
    "get_intent(\"Which countries are represented?\")"
   ]
  }
 ],
 "metadata": {
  "accelerator": "TPU",
  "colab": {
   "collapsed_sections": [],
   "include_colab_link": true,
   "name": "Intent_classification_final.ipynb",
   "provenance": [],
   "version": "0.3.2"
  },
  "kernelspec": {
   "display_name": "Python 3.6.9 64-bit",
   "language": "python",
   "name": "python36964bitf6eaa932bd364e6c99622ad728a40cf7"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}