Deep_Learning/keras_word2vec_lstm.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "virtual-optimum",
   "metadata": {
    "id": "virtual-optimum"
   },
   "outputs": [],
   "source": [
    "# GPU Reset if needed\n",
    "\n",
    "from numba import cuda \n",
    "device = cuda.get_current_device()\n",
    "device.reset()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "iA9Ipq-4zuPT",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "iA9Ipq-4zuPT",
    "outputId": "df79526c-7d47-4cb2-cd2a-4b3adf5bf389"
   },
   "outputs": [],
   "source": [
    "# from google.colab import drive\n",
    "import os\n",
    "\n",
    "# drive.mount('/content/drive')\n",
    "# os.chdir('/content/drive/My Drive/Colab Notebooks/word2vec_run/')\n",
    "# print(os.getcwd())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "MI6QgDjY2g7o",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "MI6QgDjY2g7o",
    "outputId": "16f6f31a-8f70-4650-8d9b-b88fec1322e5"
   },
   "outputs": [],
   "source": [
    "# !ls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "wWbrBVZ7CoPF",
   "metadata": {
    "id": "wWbrBVZ7CoPF"
   },
   "outputs": [],
   "source": [
    "# !rm -rf logs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "distinct-breath",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "distinct-breath",
    "outputId": "e158358f-ff02-499b-deaa-c43ce2dfabba"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TensorFlow version : 2.5.0\n",
      "Numpy version : 1.19.5\n",
      "Pandas version : 1.1.5\n"
     ]
    }
   ],
   "source": [
    "# Import Tensorflow and related libraries\n",
    "\n",
    "import tensorflow\n",
    "\n",
    "from tensorflow.keras.models import Sequential\n",
    "#from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.layers import LSTM\n",
    "from tensorflow.keras.layers import GRU\n",
    "from tensorflow.keras.layers import Dense\n",
    "from tensorflow.keras.layers import Flatten\n",
    "from tensorflow.keras.layers import Dropout\n",
    "from tensorflow.keras.layers import Embedding\n",
    "from tensorflow.keras.layers import TimeDistributed\n",
    "from tensorflow.keras.layers import BatchNormalization\n",
    "\n",
    "\n",
    "from tensorflow.keras.layers import Conv1D\n",
    "from tensorflow.keras.layers import MaxPooling1D\n",
    "from tensorflow.keras.layers import Dropout\n",
    "from tensorflow.keras.layers import GlobalMaxPooling1D\n",
    "\n",
    "from tensorflow.keras.layers import Input\n",
    "\n",
    "from tensorflow.keras.utils import plot_model\n",
    "\n",
    "from tensorflow.keras.preprocessing.text import Tokenizer\n",
    "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
    "\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "print(\"TensorFlow version : \" + tensorflow.__version__)\n",
    "\n",
    "import numpy as np\n",
    "print(\"Numpy version : \" + np.version.version)\n",
    "\n",
    "import pandas as pd\n",
    "print(\"Pandas version : \" + pd.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "circular-angle",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "circular-angle",
    "outputId": "711dfcb1-9d4b-41d6-99ff-e38205573428"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Name: /physical_device:GPU:0   Type: GPU\n"
     ]
    }
   ],
   "source": [
    "# Display available GPUs\n",
    "\n",
    "physical_devices = tensorflow.config.list_physical_devices('GPU')\n",
    "for gpu in physical_devices:\n",
    "    print(\"Name:\", gpu.name, \"  Type:\", gpu.device_type)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "e6dffa8e",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "e6dffa8e",
    "outputId": "e146cc07-813b-49d3-eadd-102481445b5f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GPU 0: NVIDIA GeForce GTX 1060 3GB (UUID: GPU-7e3eed14-c9a0-0a6c-b6f9-798ae952635f)\n"
     ]
    }
   ],
   "source": [
    "!nvidia-smi -L"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d1c6e388",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "d1c6e388",
    "outputId": "7b192348-4451-4e2e-d3b1-0c0b1cb17753"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Thu Sep  2 09:28:20 2021       \n",
      "+-----------------------------------------------------------------------------+\n",
      "| NVIDIA-SMI 471.68       Driver Version: 471.68       CUDA Version: 11.4     |\n",
      "|-------------------------------+----------------------+----------------------+\n",
      "| GPU  Name            TCC/WDDM | Bus-Id        Disp.A | Volatile Uncorr. ECC |\n",
      "| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n",
      "|                               |                      |               MIG M. |\n",
      "|===============================+======================+======================|\n",
      "|   0  NVIDIA GeForce ... WDDM  | 00000000:01:00.0  On |                  N/A |\n",
      "| 25%   34C    P5     9W / 120W |    644MiB /  3072MiB |      7%      Default |\n",
      "|                               |                      |                  N/A |\n",
      "+-------------------------------+----------------------+----------------------+\n",
      "                                                                               \n",
      "+-----------------------------------------------------------------------------+\n",
      "| Processes:                                                                  |\n",
      "|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n",
      "|        ID   ID                                                   Usage      |\n",
      "|=============================================================================|\n",
      "|    0   N/A  N/A       400    C+G   Insufficient Permissions        N/A      |\n",
      "|    0   N/A  N/A      1980    C+G   ... Download Manager\\fdm.exe    N/A      |\n",
      "|    0   N/A  N/A      2492    C+G   ...zilla Firefox\\firefox.exe    N/A      |\n",
      "|    0   N/A  N/A      5224      C   ...a\\envs\\tf2_old\\python.exe    N/A      |\n",
      "|    0   N/A  N/A      6444    C+G   ...zilla Firefox\\firefox.exe    N/A      |\n",
      "|    0   N/A  N/A      6924    C+G   ...y\\ShellExperienceHost.exe    N/A      |\n",
      "|    0   N/A  N/A      7056    C+G   C:\\Windows\\explorer.exe         N/A      |\n",
      "|    0   N/A  N/A      7260    C+G   ...5n1h2txyewy\\SearchApp.exe    N/A      |\n",
      "|    0   N/A  N/A      7400    C+G   ...nputApp\\TextInputHost.exe    N/A      |\n",
      "+-----------------------------------------------------------------------------+\n"
     ]
    }
   ],
   "source": [
    "!nvidia-smi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "preceding-travel",
   "metadata": {
    "id": "preceding-travel"
   },
   "outputs": [],
   "source": [
    "# Enable Memory Growth\n",
    "\n",
    "if physical_devices != []:\n",
    "    tensorflow.config.experimental.set_memory_growth(physical_devices[0], True)\n",
    "    physical_devices[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "primary-bryan",
   "metadata": {
    "id": "primary-bryan"
   },
   "outputs": [],
   "source": [
    "# Parameters\n",
    "\n",
    "author_list = ['Charles Dickens','Jane Austen','Sir Arthur Conan Doyle','George Eliot','Jules Verne']\n",
    "seed_number = 122\n",
    "dataset_file  = \"dataset.csv\"\n",
    "validation_dataset_file  = \"validation_dataset.csv\"\n",
    "\n",
    "batch_size = 16 # number of images per batch to be processed\n",
    "\n",
    "epochs = 150 # number of maximum epochs to run\n",
    "\n",
    "trunc_type = 'post'\n",
    "padding_type = 'post'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "e41a7202",
   "metadata": {
    "id": "e41a7202"
   },
   "outputs": [],
   "source": [
    "validation_data = pd.read_csv(validation_dataset_file, usecols = ['author','text'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "attractive-oriental",
   "metadata": {
    "id": "attractive-oriental"
   },
   "outputs": [],
   "source": [
    "data = pd.read_csv(dataset_file, usecols = ['author','text'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "dried-aerospace",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "dried-aerospace",
    "outputId": "c01a92c8-1b3e-4eea-e864-2559128fe716"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>author</th>\n",
       "      <th>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>23711</th>\n",
       "      <td>Jules Verne</td>\n",
       "      <td>stations the wire was wound round a magnet tha...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22502</th>\n",
       "      <td>Jules Verne</td>\n",
       "      <td>which by means of a current of electricity wou...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22785</th>\n",
       "      <td>Jules Verne</td>\n",
       "      <td>eye could reach. three hours passed away witho...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1075</th>\n",
       "      <td>Charles Dickens</td>\n",
       "      <td>brace it was not for want of the means of infl...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13135</th>\n",
       "      <td>Sir Arthur Conan Doyle</td>\n",
       "      <td>concerned at the sudden illness of my host to ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       author  \\\n",
       "23711             Jules Verne   \n",
       "22502             Jules Verne   \n",
       "22785             Jules Verne   \n",
       "1075          Charles Dickens   \n",
       "13135  Sir Arthur Conan Doyle   \n",
       "\n",
       "                                                    text  \n",
       "23711  stations the wire was wound round a magnet tha...  \n",
       "22502  which by means of a current of electricity wou...  \n",
       "22785  eye could reach. three hours passed away witho...  \n",
       "1075   brace it was not for want of the means of infl...  \n",
       "13135  concerned at the sudden illness of my host to ...  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "matched-minutes",
   "metadata": {
    "id": "matched-minutes"
   },
   "outputs": [],
   "source": [
    "# max_length = max(data[\"text\"].str.len())\n",
    "# print(\"Max text length: \" + str(max_length))\n",
    "\n",
    "max_length = 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "interim-therapist",
   "metadata": {
    "id": "interim-therapist"
   },
   "outputs": [],
   "source": [
    "# full_text = []\n",
    "# for record in data[\"text\"]:\n",
    "#     full_text += record\n",
    "\n",
    "# unique = set(full_text)\n",
    "# print(unique)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e45096e7",
   "metadata": {
    "id": "e45096e7"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "signal-football",
   "metadata": {
    "id": "signal-football"
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# Train, Test, Validation Spit ratios 60%, 20%, 20% \n",
    "\n",
    "X_train, X_val, y_train, y_val = train_test_split(data[\"text\"], data[\"author\"], test_size=0.2, random_state=2000)\n",
    "X_train, X_test, y_train, y_test = train_test_split(X_train, y_train, test_size=0.25, random_state=seed_number) # 0.25 x 0.8 = 0.2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "original-edition",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "original-edition",
    "outputId": "1fc3087e-28cd-4a99-8d53-63e7ab85a1b2"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "15000"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "psychological-framework",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "psychological-framework",
    "outputId": "e02173b7-440b-41df-8049-5ab94987eb98"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5000"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "aa1f59f8",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "aa1f59f8",
    "outputId": "751e2101-7fc4-4eba-f21a-c1c719b45d91"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5000"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(X_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "4e1f24db",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "4e1f24db",
    "outputId": "9e699168-da38-49cd-ec7d-7045fa5a21a1"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>author</th>\n",
       "      <th>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3795</th>\n",
       "      <td>Jules Verne</td>\n",
       "      <td>be long in lowering they redoubled their effor...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3946</th>\n",
       "      <td>Jules Verne</td>\n",
       "      <td>i now recognized as that john turner who had b...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3401</th>\n",
       "      <td>Jules Verne</td>\n",
       "      <td>know the pym family that you have never met th...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2909</th>\n",
       "      <td>George Eliot</td>\n",
       "      <td>from home with good reasons in his pocket year...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1360</th>\n",
       "      <td>Jane Austen</td>\n",
       "      <td>dearest mother the affair which has given us s...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            author                                               text\n",
       "3795   Jules Verne  be long in lowering they redoubled their effor...\n",
       "3946   Jules Verne  i now recognized as that john turner who had b...\n",
       "3401   Jules Verne  know the pym family that you have never met th...\n",
       "2909  George Eliot  from home with good reasons in his pocket year...\n",
       "1360   Jane Austen  dearest mother the affair which has given us s..."
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "validation_data.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "4d8ea1f0",
   "metadata": {
    "id": "4d8ea1f0"
   },
   "outputs": [],
   "source": [
    "X_val_2 = validation_data.text\n",
    "y_val_2 = validation_data.author"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "51b6b54d",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "51b6b54d",
    "outputId": "e67fb407-a1e8-429a-c1b6-af8b8fb0f189"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4000"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(X_val_2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "c6b38e1e",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "c6b38e1e",
    "outputId": "b4e7557c-31cd-43aa-b219-6e501957dd78"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "318     ferret eyes when the chimes of a neighbouring ...\n",
       "3337    dont stop here urged joe let us go up master l...\n",
       "886     of this one great deviation from the strict ru...\n",
       "387     the infection off though the plump sister trie...\n",
       "2563    by a wave of loving penitence and words burst ...\n",
       "Name: text, dtype: object"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_val_2.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "4012b378",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "4012b378",
    "outputId": "32919464-3508-496a-f994-36d554883d49"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "683     Charles Dickens\n",
       "3398        Jules Verne\n",
       "1562        Jane Austen\n",
       "3877        Jules Verne\n",
       "447     Charles Dickens\n",
       "Name: author, dtype: object"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_val_2.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "continuing-auditor",
   "metadata": {
    "id": "continuing-auditor"
   },
   "outputs": [],
   "source": [
    "text_tokenizer = Tokenizer() #num_words = vocab_size, oov_token=oov_tok)\n",
    "text_tokenizer.fit_on_texts(X_train)\n",
    "text_index = text_tokenizer.word_index\n",
    "\n",
    "X_train_sequences = text_tokenizer.texts_to_sequences(X_train)\n",
    "X_train_sequences = pad_sequences(X_train_sequences, maxlen=max_length, padding=padding_type, truncating=trunc_type)\n",
    "\n",
    "X_test_sequences = text_tokenizer.texts_to_sequences(X_test)\n",
    "X_test_sequences = pad_sequences(X_test_sequences, maxlen=max_length, padding=padding_type, truncating=trunc_type)\n",
    "\n",
    "X_val_sequences = text_tokenizer.texts_to_sequences(X_val)\n",
    "X_val_sequences = pad_sequences(X_val_sequences, maxlen=max_length, padding=padding_type, truncating=trunc_type)\n",
    "\n",
    "X_val_sequences_2 = text_tokenizer.texts_to_sequences(X_val_2)\n",
    "X_val_sequences_2 = pad_sequences(X_val_sequences_2, maxlen=max_length, padding=padding_type, truncating=trunc_type)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "federal-struggle",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "federal-struggle",
    "outputId": "1261fd09-6822-4c95-cc2f-b81eba311722"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "31202\n"
     ]
    }
   ],
   "source": [
    "vocab_size = len(text_index) + 1\n",
    "print(vocab_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "explicit-williams",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "explicit-williams",
    "outputId": "bb0f1abe-8a5d-40fd-97ab-2e8ed67a271c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[  16, 6934,   19, ...,    0,    0,    0],\n",
       "       [  15,   43,   65, ...,    0,    0,    0],\n",
       "       [   9,    1, 6300, ...,    0,    0,    0],\n",
       "       ...,\n",
       "       [   3,   31,    6, ...,    0,    0,    0],\n",
       "       [2193,    2, 1027, ...,    0,    0,    0],\n",
       "       [ 244,   25,   15, ...,    0,    0,    0]])"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_sequences"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "strong-ukraine",
   "metadata": {
    "id": "strong-ukraine"
   },
   "outputs": [],
   "source": [
    "label_tokenizer = LabelEncoder()\n",
    "label_tokenizer.fit(author_list)\n",
    "label_index = label_tokenizer.classes_\n",
    "\n",
    "y_train_sequences = np.array(label_tokenizer.transform(y_train))\n",
    "y_test_sequences = np.array(label_tokenizer.transform(y_test))\n",
    "y_val_sequences = np.array(label_tokenizer.transform(y_val))\n",
    "y_val_sequences_2 = np.array(label_tokenizer.transform(y_val_2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "suited-tobacco",
   "metadata": {
    "id": "suited-tobacco"
   },
   "outputs": [],
   "source": [
    "\n",
    "# label_tokenizer = Tokenizer()\n",
    "# label_tokenizer.fit_on_texts(y_train)\n",
    "# label_index = label_tokenizer.word_index\n",
    "\n",
    "# y_train_sequences = np.array(label_tokenizer.texts_to_sequences(y_train))\n",
    "# y_test_sequences = np.array(label_tokenizer.texts_to_sequences(y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "informed-precipitation",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "informed-precipitation",
    "outputId": "64ede3c4-5311-4b68-9620-df1c9c7a19a9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Charles Dickens', 'George Eliot', 'Jane Austen', 'Jules Verne',\n",
       "       'Sir Arthur Conan Doyle'], dtype='<U22')"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "label_index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "still-baseline",
   "metadata": {
    "id": "still-baseline"
   },
   "outputs": [],
   "source": [
    "vec_size = 250"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "numeric-produce",
   "metadata": {
    "id": "numeric-produce"
   },
   "source": [
    "### Word2Vec vector loading"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "collect-apollo",
   "metadata": {
    "id": "collect-apollo"
   },
   "outputs": [],
   "source": [
    "# Ref : https://tfhub.dev/google/Wiki-words-250/2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "proved-irish",
   "metadata": {
    "id": "proved-irish"
   },
   "outputs": [],
   "source": [
    "import tensorflow_hub as hub\n",
    "embed = hub.load(\"https://tfhub.dev/google/Wiki-words-250/2\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "dried-decade",
   "metadata": {
    "id": "dried-decade"
   },
   "outputs": [],
   "source": [
    "no_vectors = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "legendary-combining",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "legendary-combining",
    "outputId": "28c309e6-a10c-4142-c1ac-b30c3df7079f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wall time: 1min 5s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "word_vector_matrix = np.zeros((vocab_size, vec_size))\n",
    "\n",
    "for word, index in text_index.items():\n",
    "#     print(word,\"-->\",index)\n",
    "    vector = embed([word])[0].numpy()\n",
    "  \n",
    "    if vector is not None:\n",
    "        word_vector_matrix[index] = vector\n",
    "    if (not np.any((vector))) or (vector is None):\n",
    "        no_vectors.append(word) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "b43d2b91",
   "metadata": {
    "id": "b43d2b91"
   },
   "outputs": [],
   "source": [
    "# text_index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "meaningful-particle",
   "metadata": {
    "id": "meaningful-particle"
   },
   "outputs": [],
   "source": [
    "np.savetxt(\"word_vector_matrix.csv\", word_vector_matrix, delimiter=\",\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "arctic-redhead",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "arctic-redhead",
    "outputId": "bcda04ef-7251-47be-9c9e-a7b786a54820"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8711"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(no_vectors)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "invalid-distribution",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "invalid-distribution",
    "outputId": "d5ad9965-d930-4a97-f9ba-2e125df9304e"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "31202"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(word_vector_matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "40760414",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "40760414",
    "outputId": "752e3023-17c4-42bf-a56a-e0465f3b90a7"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "31202"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vocab_size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "special-karaoke",
   "metadata": {
    "id": "special-karaoke"
   },
   "outputs": [],
   "source": [
    "# if text_index.get(word) != None:\n",
    "#     print(\"yes\")\n",
    "# else:\n",
    "#     print(\"no\")    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "historical-solomon",
   "metadata": {
    "id": "historical-solomon"
   },
   "outputs": [],
   "source": [
    "# word_vector_matrix[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "acting-midwest",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "acting-midwest",
    "outputId": "f570e984-4ac4-4981-eea8-da213ff3f727"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"keras_word2vec_lstm\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "embedding (Embedding)        (None, 100, 250)          7800500   \n",
      "_________________________________________________________________\n",
      "lstm (LSTM)                  (None, 100, 250)          501000    \n",
      "_________________________________________________________________\n",
      "batch_normalization (BatchNo (None, 100, 250)          1000      \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 100, 250)          0         \n",
      "_________________________________________________________________\n",
      "lstm_1 (LSTM)                (None, 100, 250)          501000    \n",
      "_________________________________________________________________\n",
      "batch_normalization_1 (Batch (None, 100, 250)          1000      \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 100, 250)          0         \n",
      "_________________________________________________________________\n",
      "lstm_2 (LSTM)                (None, 100, 250)          501000    \n",
      "_________________________________________________________________\n",
      "batch_normalization_2 (Batch (None, 100, 250)          1000      \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 100, 250)          0         \n",
      "_________________________________________________________________\n",
      "lstm_3 (LSTM)                (None, 250)               501000    \n",
      "_________________________________________________________________\n",
      "batch_normalization_3 (Batch (None, 250)               1000      \n",
      "_________________________________________________________________\n",
      "dropout_3 (Dropout)          (None, 250)               0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 5)                 1255      \n",
      "=================================================================\n",
      "Total params: 9,809,755\n",
      "Trainable params: 2,007,255\n",
      "Non-trainable params: 7,802,500\n",
      "_________________________________________________________________\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# define new model\n",
    "\n",
    "# Testing model based on \"https://medium.com/voice-tech-podcast/an-overview-of-rnn-lstm-gru-79ed642751c6\"\n",
    "\n",
    "model = Sequential()\n",
    "model.add(Embedding(vocab_size, vec_size, input_length = max_length, weights = [word_vector_matrix], trainable = False))\n",
    "\n",
    "#model.add(BatchNormalization())\n",
    "model.add(LSTM(units=vec_size, return_sequences=True))\n",
    "model.add(BatchNormalization())\n",
    "model.add(Dropout(0.5))\n",
    "\n",
    "model.add(LSTM(units=vec_size, return_sequences=True))\n",
    "model.add(BatchNormalization())\n",
    "model.add(Dropout(0.5))\n",
    "\n",
    "model.add(LSTM(units=vec_size, return_sequences=True))\n",
    "model.add(BatchNormalization())\n",
    "model.add(Dropout(0.5))\n",
    "\n",
    "model.add(LSTM(units=vec_size))\n",
    "model.add(BatchNormalization())\n",
    "model.add(Dropout(0.5))\n",
    "#model.add(GRU(units=vec_size, return_sequences=False))\n",
    "\n",
    "#model.add(LSTM(units=vec_size, ))\n",
    "\n",
    "#model.add(TimeDistributed(Dense(vocab_size)))\n",
    "\n",
    "#model.add(Conv1D(64, 8, activation = 'relu'))\n",
    "#model.add(MaxPooling1D(2))\n",
    "# model.add(Dropout(0.3))\n",
    "\n",
    "# model.add(Dense(32, activation = 'relu'))\n",
    "# model.add(Dropout(0.3))          \n",
    "    \n",
    "# model.add(Dense(16, activation = 'relu'))\n",
    "          \n",
    "#model.add(GlobalMaxPooling1D())          \n",
    "          \n",
    "# #model.add(LSTM(64,dropout=0.3)) # , recurrent_dropout =0.3))\n",
    "model.add(Dense(5, activation='softmax'))\n",
    "\n",
    "model._name = \"keras_word2vec_lstm\"\n",
    "model.summary()\n",
    "\n",
    "# Save the model as a png file\n",
    "plot_model(model, to_file=model._name +'.png', show_shapes=True, show_layer_names=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "stuck-dispatch",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "stuck-dispatch",
    "outputId": "b75ed63f-d383-4a05-e2b5-719295308c8b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 embedding False\n",
      "1 lstm True\n",
      "2 batch_normalization True\n",
      "3 dropout True\n",
      "4 lstm_1 True\n",
      "5 batch_normalization_1 True\n",
      "6 dropout_1 True\n",
      "7 lstm_2 True\n",
      "8 batch_normalization_2 True\n",
      "9 dropout_2 True\n",
      "10 lstm_3 True\n",
      "11 batch_normalization_3 True\n",
      "12 dropout_3 True\n",
      "13 dense True\n"
     ]
    }
   ],
   "source": [
    "# Show the trainable status of the layers\n",
    "\n",
    "for i, layer in enumerate(model.layers):\n",
    "    print(i, layer.name, layer.trainable)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "manufactured-niagara",
   "metadata": {
    "id": "manufactured-niagara"
   },
   "outputs": [],
   "source": [
    "# Compile the model\n",
    "\n",
    "model.compile(\n",
    "    #optimizer=tensorflow.keras.optimizers.Adam(learning_rate=0.00001),\n",
    "    optimizer=tensorflow.keras.optimizers.RMSprop(learning_rate=0.0001),\n",
    "    loss=\"sparse_categorical_crossentropy\",\n",
    "    metrics=[\"accuracy\"],\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "passing-chambers",
   "metadata": {
    "id": "passing-chambers"
   },
   "outputs": [],
   "source": [
    "# Tensorboard parameters\n",
    "\n",
    "import datetime\n",
    "log_dir = \"logs/fit/\" + datetime.datetime.now().strftime(\"%Y%m%d-%H%M%S\")\n",
    "tensorboard_callback = tensorflow.keras.callbacks.TensorBoard(log_dir=log_dir, histogram_freq=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "diagnostic-hydrogen",
   "metadata": {
    "id": "diagnostic-hydrogen"
   },
   "outputs": [],
   "source": [
    "# Use best weights for the model\n",
    "\n",
    "early_stopping_monitor = tensorflow.keras.callbacks.EarlyStopping(\n",
    "    monitor='val_accuracy',\n",
    "    min_delta=0,\n",
    "    patience=50,\n",
    "    verbose=0,\n",
    "    mode='max',\n",
    "    baseline=None,\n",
    "    restore_best_weights=True\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "unlike-ballot",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "unlike-ballot",
    "outputId": "fcdad898-9d9b-4c78-ef2e-b8c65d89c450"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/150\n",
      "938/938 [==============================] - 82s 79ms/step - loss: 2.5022 - accuracy: 0.1960 - val_loss: 3.5901 - val_accuracy: 0.2054\n",
      "Epoch 2/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 2.0018 - accuracy: 0.2784 - val_loss: 2.2944 - val_accuracy: 0.3272\n",
      "Epoch 3/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 1.7716 - accuracy: 0.3199 - val_loss: 1.6474 - val_accuracy: 0.3070\n",
      "Epoch 4/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 1.6296 - accuracy: 0.3337 - val_loss: 2.0568 - val_accuracy: 0.2418\n",
      "Epoch 5/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 1.5367 - accuracy: 0.3457 - val_loss: 6.7669 - val_accuracy: 0.1988\n",
      "Epoch 6/150\n",
      "938/938 [==============================] - 72s 76ms/step - loss: 1.4760 - accuracy: 0.3627 - val_loss: 2.7974 - val_accuracy: 0.2398\n",
      "Epoch 7/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 1.4383 - accuracy: 0.3836 - val_loss: 1.5720 - val_accuracy: 0.3404\n",
      "Epoch 8/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 1.3314 - accuracy: 0.4359 - val_loss: 1.6858 - val_accuracy: 0.2502\n",
      "Epoch 9/150\n",
      "938/938 [==============================] - 72s 76ms/step - loss: 1.2698 - accuracy: 0.4663 - val_loss: 2.1265 - val_accuracy: 0.2382\n",
      "Epoch 10/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 1.2722 - accuracy: 0.4604 - val_loss: 1.8222 - val_accuracy: 0.3856\n",
      "Epoch 11/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 1.2230 - accuracy: 0.4857 - val_loss: 2.4267 - val_accuracy: 0.3212\n",
      "Epoch 12/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 1.1569 - accuracy: 0.5233 - val_loss: 2.4352 - val_accuracy: 0.3542\n",
      "Epoch 13/150\n",
      "938/938 [==============================] - 73s 78ms/step - loss: 1.1091 - accuracy: 0.5512 - val_loss: 1.3312 - val_accuracy: 0.4958\n",
      "Epoch 14/150\n",
      "938/938 [==============================] - 73s 77ms/step - loss: 1.0628 - accuracy: 0.5745 - val_loss: 3.4149 - val_accuracy: 0.2010\n",
      "Epoch 15/150\n",
      "938/938 [==============================] - 73s 77ms/step - loss: 1.0339 - accuracy: 0.5859 - val_loss: 2.5272 - val_accuracy: 0.3440\n",
      "Epoch 16/150\n",
      "938/938 [==============================] - 73s 78ms/step - loss: 1.0081 - accuracy: 0.5987 - val_loss: 1.8737 - val_accuracy: 0.4288\n",
      "Epoch 17/150\n",
      "938/938 [==============================] - 73s 77ms/step - loss: 0.9903 - accuracy: 0.6075 - val_loss: 1.3474 - val_accuracy: 0.5364\n",
      "Epoch 18/150\n",
      "938/938 [==============================] - 73s 78ms/step - loss: 0.9640 - accuracy: 0.6141 - val_loss: 1.4660 - val_accuracy: 0.4820\n",
      "Epoch 19/150\n",
      "938/938 [==============================] - 74s 79ms/step - loss: 0.9557 - accuracy: 0.6303 - val_loss: 1.2062 - val_accuracy: 0.5508\n",
      "Epoch 20/150\n",
      "938/938 [==============================] - 75s 80ms/step - loss: 0.9222 - accuracy: 0.6427 - val_loss: 1.4906 - val_accuracy: 0.4850\n",
      "Epoch 21/150\n",
      "938/938 [==============================] - 75s 80ms/step - loss: 0.9062 - accuracy: 0.6438 - val_loss: 2.1802 - val_accuracy: 0.3874\n",
      "Epoch 22/150\n",
      "938/938 [==============================] - 75s 80ms/step - loss: 0.8845 - accuracy: 0.6585 - val_loss: 1.1651 - val_accuracy: 0.5828\n",
      "Epoch 23/150\n",
      "938/938 [==============================] - 75s 80ms/step - loss: 0.8592 - accuracy: 0.6691 - val_loss: 3.1389 - val_accuracy: 0.2934\n",
      "Epoch 24/150\n",
      "938/938 [==============================] - 75s 80ms/step - loss: 0.8574 - accuracy: 0.6717 - val_loss: 1.2535 - val_accuracy: 0.5898\n",
      "Epoch 25/150\n",
      "938/938 [==============================] - 75s 80ms/step - loss: 0.8391 - accuracy: 0.6788 - val_loss: 1.1500 - val_accuracy: 0.5716\n",
      "Epoch 26/150\n",
      "938/938 [==============================] - 75s 80ms/step - loss: 0.8287 - accuracy: 0.6861 - val_loss: 1.0277 - val_accuracy: 0.64500s - l\n",
      "Epoch 27/150\n",
      "938/938 [==============================] - 71s 76ms/step - loss: 0.8119 - accuracy: 0.6903 - val_loss: 1.3161 - val_accuracy: 0.5932\n",
      "Epoch 28/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.7984 - accuracy: 0.6965 - val_loss: 1.0153 - val_accuracy: 0.6154\n",
      "Epoch 29/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.7766 - accuracy: 0.7049 - val_loss: 0.9314 - val_accuracy: 0.6500\n",
      "Epoch 30/150\n",
      "938/938 [==============================] - 71s 76ms/step - loss: 0.7650 - accuracy: 0.7104 - val_loss: 1.1114 - val_accuracy: 0.5972\n",
      "Epoch 31/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.7424 - accuracy: 0.7171 - val_loss: 1.2271 - val_accuracy: 0.5780\n",
      "Epoch 32/150\n",
      "938/938 [==============================] - 72s 76ms/step - loss: 0.7264 - accuracy: 0.7245 - val_loss: 1.2866 - val_accuracy: 0.5844\n",
      "Epoch 33/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.7028 - accuracy: 0.7363 - val_loss: 1.3474 - val_accuracy: 0.5684\n",
      "Epoch 34/150\n",
      "938/938 [==============================] - 73s 77ms/step - loss: 0.6953 - accuracy: 0.7445 - val_loss: 1.0202 - val_accuracy: 0.6474\n",
      "Epoch 35/150\n",
      "938/938 [==============================] - 71s 75ms/step - loss: 0.6768 - accuracy: 0.7459 - val_loss: 1.2423 - val_accuracy: 0.5652\n",
      "Epoch 36/150\n",
      "938/938 [==============================] - 73s 78ms/step - loss: 0.6632 - accuracy: 0.7545 - val_loss: 1.3295 - val_accuracy: 0.6046\n",
      "Epoch 37/150\n",
      "938/938 [==============================] - 73s 78ms/step - loss: 0.6334 - accuracy: 0.7703 - val_loss: 1.5281 - val_accuracy: 0.5370\n",
      "Epoch 38/150\n",
      "938/938 [==============================] - 73s 78ms/step - loss: 0.6233 - accuracy: 0.7723 - val_loss: 1.3816 - val_accuracy: 0.6056\n",
      "Epoch 39/150\n",
      "938/938 [==============================] - 72s 76ms/step - loss: 0.6089 - accuracy: 0.7790 - val_loss: 1.0466 - val_accuracy: 0.64060s - loss: 0.6089 - accuracy: 0.77\n",
      "Epoch 40/150\n",
      "938/938 [==============================] - 73s 78ms/step - loss: 0.5824 - accuracy: 0.7900 - val_loss: 1.5455 - val_accuracy: 0.5518\n",
      "Epoch 41/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.5652 - accuracy: 0.7931 - val_loss: 1.1574 - val_accuracy: 0.6398\n",
      "Epoch 42/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.5509 - accuracy: 0.7999 - val_loss: 1.2577 - val_accuracy: 0.6392\n",
      "Epoch 43/150\n",
      "938/938 [==============================] - 73s 78ms/step - loss: 0.5339 - accuracy: 0.8059 - val_loss: 1.3247 - val_accuracy: 0.6384\n",
      "Epoch 44/150\n",
      "938/938 [==============================] - 73s 78ms/step - loss: 0.5225 - accuracy: 0.8119 - val_loss: 1.0979 - val_accuracy: 0.6520\n",
      "Epoch 45/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.4896 - accuracy: 0.8246 - val_loss: 0.9272 - val_accuracy: 0.7102\n",
      "Epoch 46/150\n",
      "938/938 [==============================] - 73s 77ms/step - loss: 0.4786 - accuracy: 0.8283 - val_loss: 1.3210 - val_accuracy: 0.6290\n",
      "Epoch 47/150\n",
      "938/938 [==============================] - 73s 77ms/step - loss: 0.4601 - accuracy: 0.8333 - val_loss: 1.2321 - val_accuracy: 0.6582\n",
      "Epoch 48/150\n",
      "938/938 [==============================] - 73s 78ms/step - loss: 0.4412 - accuracy: 0.8413 - val_loss: 1.2847 - val_accuracy: 0.6452\n",
      "Epoch 49/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.4263 - accuracy: 0.8509 - val_loss: 1.1179 - val_accuracy: 0.6548\n",
      "Epoch 50/150\n",
      "938/938 [==============================] - 72s 76ms/step - loss: 0.4121 - accuracy: 0.8540 - val_loss: 1.2302 - val_accuracy: 0.6558\n",
      "Epoch 51/150\n",
      "938/938 [==============================] - 73s 77ms/step - loss: 0.3893 - accuracy: 0.8625 - val_loss: 1.2277 - val_accuracy: 0.6534\n",
      "Epoch 52/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.3796 - accuracy: 0.8687 - val_loss: 1.4566 - val_accuracy: 0.6434\n",
      "Epoch 53/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.3564 - accuracy: 0.8728 - val_loss: 1.3763 - val_accuracy: 0.6258\n",
      "Epoch 54/150\n",
      "938/938 [==============================] - 74s 79ms/step - loss: 0.3486 - accuracy: 0.8825 - val_loss: 1.3058 - val_accuracy: 0.6474\n",
      "Epoch 55/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.3328 - accuracy: 0.8807 - val_loss: 1.1415 - val_accuracy: 0.6810\n",
      "Epoch 56/150\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "938/938 [==============================] - 72s 77ms/step - loss: 0.3243 - accuracy: 0.8877 - val_loss: 1.1265 - val_accuracy: 0.6840\n",
      "Epoch 57/150\n",
      "938/938 [==============================] - 73s 78ms/step - loss: 0.3087 - accuracy: 0.8937 - val_loss: 1.0938 - val_accuracy: 0.7028\n",
      "Epoch 58/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.2937 - accuracy: 0.9009 - val_loss: 1.2873 - val_accuracy: 0.6748\n",
      "Epoch 59/150\n",
      "938/938 [==============================] - 77s 82ms/step - loss: 0.2878 - accuracy: 0.9036 - val_loss: 1.1525 - val_accuracy: 0.6914\n",
      "Epoch 60/150\n",
      "938/938 [==============================] - 73s 77ms/step - loss: 0.2746 - accuracy: 0.9071 - val_loss: 1.1713 - val_accuracy: 0.7058\n",
      "Epoch 61/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.2582 - accuracy: 0.9115 - val_loss: 1.2408 - val_accuracy: 0.6870\n",
      "Epoch 62/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.2406 - accuracy: 0.9194 - val_loss: 1.5129 - val_accuracy: 0.6732\n",
      "Epoch 63/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.2464 - accuracy: 0.9180 - val_loss: 1.3715 - val_accuracy: 0.6762\n",
      "Epoch 64/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.2285 - accuracy: 0.9253 - val_loss: 1.6501 - val_accuracy: 0.6592\n",
      "Epoch 65/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.2243 - accuracy: 0.9272 - val_loss: 1.3528 - val_accuracy: 0.6928\n",
      "Epoch 66/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.2093 - accuracy: 0.9321 - val_loss: 1.4098 - val_accuracy: 0.6756\n",
      "Epoch 67/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.2011 - accuracy: 0.9343 - val_loss: 1.5072 - val_accuracy: 0.6796\n",
      "Epoch 68/150\n",
      "938/938 [==============================] - 72s 76ms/step - loss: 0.2025 - accuracy: 0.9358 - val_loss: 1.4284 - val_accuracy: 0.6828\n",
      "Epoch 69/150\n",
      "938/938 [==============================] - 73s 78ms/step - loss: 0.1950 - accuracy: 0.9373 - val_loss: 1.4106 - val_accuracy: 0.6968\n",
      "Epoch 70/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.1941 - accuracy: 0.9370 - val_loss: 1.2890 - val_accuracy: 0.6970\n",
      "Epoch 71/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.1717 - accuracy: 0.9436 - val_loss: 1.8290 - val_accuracy: 0.6512\n",
      "Epoch 72/150\n",
      "938/938 [==============================] - 73s 77ms/step - loss: 0.1767 - accuracy: 0.9433 - val_loss: 1.7351 - val_accuracy: 0.6720\n",
      "Epoch 73/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.1682 - accuracy: 0.9459 - val_loss: 1.4403 - val_accuracy: 0.6896\n",
      "Epoch 74/150\n",
      "938/938 [==============================] - 73s 77ms/step - loss: 0.1695 - accuracy: 0.9456 - val_loss: 1.7200 - val_accuracy: 0.6650\n",
      "Epoch 75/150\n",
      "938/938 [==============================] - 73s 77ms/step - loss: 0.1653 - accuracy: 0.9460 - val_loss: 1.4990 - val_accuracy: 0.6900\n",
      "Epoch 76/150\n",
      "938/938 [==============================] - 75s 80ms/step - loss: 0.1546 - accuracy: 0.9515 - val_loss: 1.8332 - val_accuracy: 0.6580\n",
      "Epoch 77/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.1550 - accuracy: 0.9500 - val_loss: 1.5895 - val_accuracy: 0.6892\n",
      "Epoch 78/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.1509 - accuracy: 0.9498 - val_loss: 1.4576 - val_accuracy: 0.6962\n",
      "Epoch 79/150\n",
      "938/938 [==============================] - 72s 76ms/step - loss: 0.1425 - accuracy: 0.9545 - val_loss: 1.5199 - val_accuracy: 0.6938\n",
      "Epoch 80/150\n",
      "938/938 [==============================] - 73s 78ms/step - loss: 0.1397 - accuracy: 0.9565 - val_loss: 1.8026 - val_accuracy: 0.6592\n",
      "Epoch 81/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.1271 - accuracy: 0.9597 - val_loss: 1.6486 - val_accuracy: 0.6876\n",
      "Epoch 82/150\n",
      "938/938 [==============================] - 73s 78ms/step - loss: 0.1256 - accuracy: 0.9589 - val_loss: 1.5362 - val_accuracy: 0.7028\n",
      "Epoch 83/150\n",
      "938/938 [==============================] - 73s 77ms/step - loss: 0.1247 - accuracy: 0.9613 - val_loss: 1.5392 - val_accuracy: 0.7068\n",
      "Epoch 84/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.1224 - accuracy: 0.9613 - val_loss: 1.7187 - val_accuracy: 0.6904\n",
      "Epoch 85/150\n",
      "938/938 [==============================] - 74s 79ms/step - loss: 0.1232 - accuracy: 0.9618 - val_loss: 1.5780 - val_accuracy: 0.6912\n",
      "Epoch 86/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.1162 - accuracy: 0.9647 - val_loss: 1.7332 - val_accuracy: 0.6788\n",
      "Epoch 87/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.1267 - accuracy: 0.9601 - val_loss: 1.7466 - val_accuracy: 0.6952\n",
      "Epoch 88/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.1190 - accuracy: 0.9642 - val_loss: 1.9082 - val_accuracy: 0.6700\n",
      "Epoch 89/150\n",
      "938/938 [==============================] - 73s 78ms/step - loss: 0.1113 - accuracy: 0.9661 - val_loss: 1.9694 - val_accuracy: 0.6512\n",
      "Epoch 90/150\n",
      "938/938 [==============================] - 72s 76ms/step - loss: 0.1205 - accuracy: 0.9643 - val_loss: 1.6911 - val_accuracy: 0.6942\n",
      "Epoch 91/150\n",
      "938/938 [==============================] - 72s 76ms/step - loss: 0.1070 - accuracy: 0.9677 - val_loss: 1.7284 - val_accuracy: 0.6928\n",
      "Epoch 92/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.1008 - accuracy: 0.9689 - val_loss: 2.1587 - val_accuracy: 0.6738\n",
      "Epoch 93/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.1156 - accuracy: 0.9661 - val_loss: 2.1778 - val_accuracy: 0.6690\n",
      "Epoch 94/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.1017 - accuracy: 0.9699 - val_loss: 1.7338 - val_accuracy: 0.6964\n",
      "Epoch 95/150\n",
      "938/938 [==============================] - 72s 77ms/step - loss: 0.0966 - accuracy: 0.9715 - val_loss: 1.7944 - val_accuracy: 0.6852\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x1fedb0d7cc8>"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Training the model\n",
    "\n",
    "model.fit(\n",
    "    X_train_sequences, y_train_sequences, epochs=epochs, batch_size = batch_size,shuffle = True, callbacks=[early_stopping_monitor, tensorboard_callback], validation_data=(X_test_sequences , y_test_sequences ),\n",
    "    )\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "apart-christian",
   "metadata": {
    "id": "apart-christian"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Reusing TensorBoard on port 6006 (pid 8588), started 12:43:22 ago. (Use '!kill 8588' to kill it.)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "      <iframe id=\"tensorboard-frame-157fb413bebbc17f\" width=\"100%\" height=\"800\" frameborder=\"0\">\n",
       "      </iframe>\n",
       "      <script>\n",
       "        (function() {\n",
       "          const frame = document.getElementById(\"tensorboard-frame-157fb413bebbc17f\");\n",
       "          const url = new URL(\"/\", window.location);\n",
       "          const port = 6006;\n",
       "          if (port) {\n",
       "            url.port = port;\n",
       "          }\n",
       "          frame.src = url;\n",
       "        })();\n",
       "      </script>\n",
       "    "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show tensorboard\n",
    "\n",
    "%load_ext tensorboard\n",
    "%tensorboard --logdir logs/fit\n",
    "#%reload_ext tensorboard # just in case tensorboard didn't refresh\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "c0aeef60",
   "metadata": {
    "id": "c0aeef60"
   },
   "outputs": [],
   "source": [
    "# Function to generate accuracy report\n",
    "from sklearn.metrics import confusion_matrix\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import ConfusionMatrixDisplay\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "def accuracy_report(X_val_sequences,y_val_sequences):\n",
    "    # Calculate preditions for the validation dataset\n",
    "\n",
    "    # import numpy as np\n",
    "\n",
    "    y_pred = []  # store predicted labels\n",
    "    y_true = []  # store true labels\n",
    "\n",
    "    y_pred = model.predict_classes([X_val_sequences])\n",
    "    len(y_pred)\n",
    "\n",
    "    y_true= y_val_sequences\n",
    "    len(y_pred)\n",
    "    \n",
    "    print(y_pred)\n",
    "\n",
    "    # for image_batch, label_batch in val_ds:\n",
    "    #     y_true.append(label_batch)\n",
    "    #     preds = model.predict(image_batch)\n",
    "    #     y_pred.append(np.argmax(preds, axis = - 1))\n",
    "\n",
    "    # # Convert the true and predicted labels into tensors\n",
    "    true_categories = label_tokenizer.inverse_transform(y_true)#tensorflow.concat([item for item in y_true], axis = 0)\n",
    "    predicted_categories = label_tokenizer.inverse_transform(y_pred)#np.argmax(y_pred, axis = - 1)) #tensorflow.concat([item for item in y_pred], axis = 0)\n",
    "\n",
    "    print(predicted_categories)\n",
    "\n",
    "    # Generate confusion matrix\n",
    "\n",
    "    cnf_matrix = confusion_matrix(true_categories,predicted_categories,  normalize='pred')\n",
    "\n",
    "    # Plot confusion matrix\n",
    "\n",
    "\n",
    "\n",
    "    disp = ConfusionMatrixDisplay(confusion_matrix=cnf_matrix,\n",
    "                                  display_labels=label_index)\n",
    "\n",
    "    disp = disp.plot(cmap='Blues', xticks_rotation=45)\n",
    "\n",
    "    plt.show()\n",
    "\n",
    "\n",
    "\n",
    "    print(\"Accuracy for validation split: \" ,accuracy_score(y_true, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "20ee3a09",
   "metadata": {
    "id": "20ee3a09"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "environmental-transcription",
   "metadata": {
    "id": "environmental-transcription"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Miniconda\\envs\\tf2_old\\lib\\site-packages\\tensorflow\\python\\keras\\engine\\sequential.py:455: UserWarning: `model.predict_classes()` is deprecated and will be removed after 2021-01-01. Please use instead:* `np.argmax(model.predict(x), axis=-1)`,   if your model does multi-class classification   (e.g. if it uses a `softmax` last-layer activation).* `(model.predict(x) > 0.5).astype(\"int32\")`,   if your model does binary classification   (e.g. if it uses a `sigmoid` last-layer activation).\n",
      "  warnings.warn('`model.predict_classes()` is deprecated and '\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[3 0 3 ... 0 2 1]\n",
      "['Jules Verne' 'Charles Dickens' 'Jules Verne' ... 'Charles Dickens'\n",
      " 'Jane Austen' 'George Eliot']\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy for validation split:  0.7062\n"
     ]
    }
   ],
   "source": [
    "# accuracy_report for same books\n",
    "\n",
    "accuracy_report(X_val_sequences,y_val_sequences)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "fundamental-raising",
   "metadata": {
    "id": "fundamental-raising"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Miniconda\\envs\\tf2_old\\lib\\site-packages\\tensorflow\\python\\keras\\engine\\sequential.py:455: UserWarning: `model.predict_classes()` is deprecated and will be removed after 2021-01-01. Please use instead:* `np.argmax(model.predict(x), axis=-1)`,   if your model does multi-class classification   (e.g. if it uses a `softmax` last-layer activation).* `(model.predict(x) > 0.5).astype(\"int32\")`,   if your model does binary classification   (e.g. if it uses a `sigmoid` last-layer activation).\n",
      "  warnings.warn('`model.predict_classes()` is deprecated and '\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 4 2 ... 0 3 3]\n",
      "['Charles Dickens' 'Sir Arthur Conan Doyle' 'Jane Austen' ...\n",
      " 'Charles Dickens' 'Jules Verne' 'Jules Verne']\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy for validation split:  0.55575\n"
     ]
    }
   ],
   "source": [
    "# accuracy_report for seperate books\n",
    "\n",
    "accuracy_report(X_val_sequences_2,y_val_sequences_2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "electoral-arrangement",
   "metadata": {
    "id": "electoral-arrangement"
   },
   "outputs": [],
   "source": [
    "# Saving final model for future use\n",
    "\n",
    "model.save(model._name + \".h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "engaged-heritage",
   "metadata": {
    "id": "engaged-heritage"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "olive-filling",
   "metadata": {
    "id": "olive-filling"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "global-egyptian",
   "metadata": {
    "id": "global-egyptian"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "front-norway",
   "metadata": {
    "id": "front-norway"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "chemical-columbus",
   "metadata": {
    "id": "chemical-columbus"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "understanding-ghana",
   "metadata": {
    "id": "understanding-ghana"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "french-blackberry",
   "metadata": {
    "id": "french-blackberry"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "keras_word2vec_lstm.ipynb",
   "provenance": []
  },
  "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": "3.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}