Copy_of_NLP_Week4_Exercise_Shakespeare_Answer.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "name": "Copy of NLP_Week4_Exercise_Shakespeare_Answer.ipynb",
      "provenance": [],
      "toc_visible": true,
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/RhuthuHegde/Tensorflow-coursera/blob/master/Copy_of_NLP_Week4_Exercise_Shakespeare_Answer.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zX4Kg8DUTKWO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#@title Licensed under the Apache License, Version 2.0 (the \"License\");\n",
        "# you may not use this file except in compliance with the License.\n",
        "# You may obtain a copy of the License at\n",
        "#\n",
        "# https://www.apache.org/licenses/LICENSE-2.0\n",
        "#\n",
        "# Unless required by applicable law or agreed to in writing, software\n",
        "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
        "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
        "# See the License for the specific language governing permissions and\n",
        "# limitations under the License."
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "BOwsuGQQY9OL",
        "colab": {}
      },
      "source": [
        "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
        "from tensorflow.keras.layers import Embedding, LSTM, Dense, Dropout, Bidirectional\n",
        "from tensorflow.keras.preprocessing.text import Tokenizer\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "from tensorflow.keras import regularizers\n",
        "import tensorflow.keras.utils as ku \n",
        "import numpy as np "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "PRnDnCW-Z7qv",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "outputId": "47420402-793e-48aa-80c3-f0949ce89bb7"
      },
      "source": [
        "tokenizer = Tokenizer()\n",
        "!wget --no-check-certificate \\\n",
        "    https://storage.googleapis.com/laurencemoroney-blog.appspot.com/sonnets.txt \\\n",
        "    -O /tmp/sonnets.txt\n",
        "data = open('/tmp/sonnets.txt').read()\n",
        "\n",
        "corpus = data.lower().split(\"\\n\")\n",
        "\n",
        "\n",
        "tokenizer.fit_on_texts(corpus)\n",
        "total_words = len(tokenizer.word_index) + 1\n",
        "\n",
        "# create input sequences using list of tokens\n",
        "input_sequences = []\n",
        "for line in corpus:\n",
        "\ttoken_list = tokenizer.texts_to_sequences([line])[0]\n",
        "\tfor i in range(1, len(token_list)):\n",
        "\t\tn_gram_sequence = token_list[:i+1]\n",
        "\t\tinput_sequences.append(n_gram_sequence)\n",
        "\n",
        "\n",
        "# pad sequences \n",
        "max_sequence_len = max([len(x) for x in input_sequences])\n",
        "input_sequences = np.array(pad_sequences(input_sequences, maxlen=max_sequence_len, padding='pre'))\n",
        "\n",
        "# create predictors and label\n",
        "predictors, label = input_sequences[:,:-1],input_sequences[:,-1]\n",
        "\n",
        "label = ku.to_categorical(label, num_classes=total_words)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2020-08-05 08:22:45--  https://storage.googleapis.com/laurencemoroney-blog.appspot.com/sonnets.txt\n",
            "Resolving storage.googleapis.com (storage.googleapis.com)... 172.217.194.128, 74.125.24.128, 2404:6800:4003:c06::80, ...\n",
            "Connecting to storage.googleapis.com (storage.googleapis.com)|172.217.194.128|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 93578 (91K) [text/plain]\n",
            "Saving to: ‘/tmp/sonnets.txt’\n",
            "\n",
            "\r/tmp/sonnets.txt      0%[                    ]       0  --.-KB/s               \r/tmp/sonnets.txt    100%[===================>]  91.38K  --.-KB/s    in 0.001s  \n",
            "\n",
            "2020-08-05 08:22:46 (114 MB/s) - ‘/tmp/sonnets.txt’ saved [93578/93578]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "w9vH8Y59ajYL",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 374
        },
        "outputId": "716d3fc7-53b9-45fa-b8d3-708da50a93c1"
      },
      "source": [
        "model = Sequential()\n",
        "model.add(Embedding(total_words, 100, input_length=max_sequence_len-1))\n",
        "model.add(Bidirectional(LSTM(150, return_sequences = True)))\n",
        "model.add(Dropout(0.2))\n",
        "model.add(LSTM(100))\n",
        "model.add(Dense(total_words/2, activation='relu', kernel_regularizer=regularizers.l2(0.01)))\n",
        "model.add(Dense(total_words, activation='softmax'))\n",
        "model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
        "print(model.summary())\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "embedding (Embedding)        (None, 10, 100)           321100    \n",
            "_________________________________________________________________\n",
            "bidirectional (Bidirectional (None, 10, 300)           301200    \n",
            "_________________________________________________________________\n",
            "dropout (Dropout)            (None, 10, 300)           0         \n",
            "_________________________________________________________________\n",
            "lstm_1 (LSTM)                (None, 100)               160400    \n",
            "_________________________________________________________________\n",
            "dense (Dense)                (None, 1605)              162105    \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 3211)              5156866   \n",
            "=================================================================\n",
            "Total params: 6,101,671\n",
            "Trainable params: 6,101,671\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "None\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "AIg2f1HBxqof",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "1b9cbb49-afd4-4c7e-f1c2-fe3914d21f11"
      },
      "source": [
        " history = model.fit(predictors, label, epochs=100, verbose=1)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 6.9080 - accuracy: 0.0212\n",
            "Epoch 2/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 6.5081 - accuracy: 0.0224\n",
            "Epoch 3/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 6.3993 - accuracy: 0.0265\n",
            "Epoch 4/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 6.2776 - accuracy: 0.0292\n",
            "Epoch 5/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 6.1973 - accuracy: 0.0338\n",
            "Epoch 6/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 6.1246 - accuracy: 0.0369\n",
            "Epoch 7/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 6.0504 - accuracy: 0.0391\n",
            "Epoch 8/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 5.9783 - accuracy: 0.0453\n",
            "Epoch 9/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 5.9021 - accuracy: 0.0491\n",
            "Epoch 10/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 5.8227 - accuracy: 0.0503\n",
            "Epoch 11/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 5.7374 - accuracy: 0.0539\n",
            "Epoch 12/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 5.6464 - accuracy: 0.0582\n",
            "Epoch 13/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 5.5688 - accuracy: 0.0626\n",
            "Epoch 14/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 5.4807 - accuracy: 0.0660\n",
            "Epoch 15/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 5.3965 - accuracy: 0.0730\n",
            "Epoch 16/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 5.3082 - accuracy: 0.0767\n",
            "Epoch 17/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 5.2276 - accuracy: 0.0810\n",
            "Epoch 18/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 5.1468 - accuracy: 0.0867\n",
            "Epoch 19/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 5.0589 - accuracy: 0.0934\n",
            "Epoch 20/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 4.9677 - accuracy: 0.1001\n",
            "Epoch 21/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 4.8810 - accuracy: 0.1068\n",
            "Epoch 22/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 4.7890 - accuracy: 0.1173\n",
            "Epoch 23/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 4.7053 - accuracy: 0.1206\n",
            "Epoch 24/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 4.6066 - accuracy: 0.1269\n",
            "Epoch 25/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 4.5120 - accuracy: 0.1390\n",
            "Epoch 26/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 4.4206 - accuracy: 0.1518\n",
            "Epoch 27/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 4.3268 - accuracy: 0.1588\n",
            "Epoch 28/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 4.2327 - accuracy: 0.1700\n",
            "Epoch 29/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 4.1386 - accuracy: 0.1819\n",
            "Epoch 30/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 4.0494 - accuracy: 0.1931\n",
            "Epoch 31/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.9619 - accuracy: 0.2099\n",
            "Epoch 32/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.8642 - accuracy: 0.2243\n",
            "Epoch 33/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.7886 - accuracy: 0.2405\n",
            "Epoch 34/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.7036 - accuracy: 0.2579\n",
            "Epoch 35/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 3.6204 - accuracy: 0.2746\n",
            "Epoch 36/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.5414 - accuracy: 0.2879\n",
            "Epoch 37/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.4655 - accuracy: 0.3086\n",
            "Epoch 38/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.3911 - accuracy: 0.3276\n",
            "Epoch 39/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.3205 - accuracy: 0.3406\n",
            "Epoch 40/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.2556 - accuracy: 0.3498\n",
            "Epoch 41/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 3.1734 - accuracy: 0.3721\n",
            "Epoch 42/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.1082 - accuracy: 0.3864\n",
            "Epoch 43/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.0398 - accuracy: 0.4006\n",
            "Epoch 44/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.9789 - accuracy: 0.4117\n",
            "Epoch 45/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 2.9222 - accuracy: 0.4303\n",
            "Epoch 46/100\n",
            "484/484 [==============================] - 6s 13ms/step - loss: 2.8559 - accuracy: 0.4450\n",
            "Epoch 47/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 2.8015 - accuracy: 0.4576\n",
            "Epoch 48/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.7491 - accuracy: 0.4668\n",
            "Epoch 49/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 2.6995 - accuracy: 0.4811\n",
            "Epoch 50/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 2.6375 - accuracy: 0.4957\n",
            "Epoch 51/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.5799 - accuracy: 0.5054\n",
            "Epoch 52/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.5385 - accuracy: 0.5103\n",
            "Epoch 53/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.4865 - accuracy: 0.5250\n",
            "Epoch 54/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.4500 - accuracy: 0.5318\n",
            "Epoch 55/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 2.3962 - accuracy: 0.5470\n",
            "Epoch 56/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.3490 - accuracy: 0.5566\n",
            "Epoch 57/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.3165 - accuracy: 0.5665\n",
            "Epoch 58/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.2735 - accuracy: 0.5727\n",
            "Epoch 59/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 2.2302 - accuracy: 0.5834\n",
            "Epoch 60/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.2036 - accuracy: 0.5909\n",
            "Epoch 61/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 2.1459 - accuracy: 0.6010\n",
            "Epoch 62/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.1115 - accuracy: 0.6120\n",
            "Epoch 63/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.0714 - accuracy: 0.6201\n",
            "Epoch 64/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 2.0416 - accuracy: 0.6248\n",
            "Epoch 65/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 2.0154 - accuracy: 0.6299\n",
            "Epoch 66/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.9741 - accuracy: 0.6380\n",
            "Epoch 67/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.9524 - accuracy: 0.6486\n",
            "Epoch 68/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 1.9196 - accuracy: 0.6541\n",
            "Epoch 69/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 1.8905 - accuracy: 0.6643\n",
            "Epoch 70/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.8572 - accuracy: 0.6671\n",
            "Epoch 71/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.8299 - accuracy: 0.6726\n",
            "Epoch 72/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.7873 - accuracy: 0.6837\n",
            "Epoch 73/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.7672 - accuracy: 0.6866\n",
            "Epoch 74/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.7438 - accuracy: 0.6895\n",
            "Epoch 75/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 1.7311 - accuracy: 0.6937\n",
            "Epoch 76/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.7040 - accuracy: 0.6947\n",
            "Epoch 77/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.6809 - accuracy: 0.7024\n",
            "Epoch 78/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.6607 - accuracy: 0.7059\n",
            "Epoch 79/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.6441 - accuracy: 0.7092\n",
            "Epoch 80/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.6141 - accuracy: 0.7172\n",
            "Epoch 81/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.5813 - accuracy: 0.7242\n",
            "Epoch 82/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.5704 - accuracy: 0.7239\n",
            "Epoch 83/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.5575 - accuracy: 0.7247\n",
            "Epoch 84/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.5470 - accuracy: 0.7288\n",
            "Epoch 85/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.5287 - accuracy: 0.7314\n",
            "Epoch 86/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.5081 - accuracy: 0.7374\n",
            "Epoch 87/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.4920 - accuracy: 0.7366\n",
            "Epoch 88/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.4808 - accuracy: 0.7386\n",
            "Epoch 89/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.4594 - accuracy: 0.7435\n",
            "Epoch 90/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.4379 - accuracy: 0.7514\n",
            "Epoch 91/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.4209 - accuracy: 0.7501\n",
            "Epoch 92/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.4068 - accuracy: 0.7545\n",
            "Epoch 93/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.3907 - accuracy: 0.7579\n",
            "Epoch 94/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.3842 - accuracy: 0.7557\n",
            "Epoch 95/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.3793 - accuracy: 0.7596\n",
            "Epoch 96/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.3652 - accuracy: 0.7601\n",
            "Epoch 97/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.3462 - accuracy: 0.7646\n",
            "Epoch 98/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.3213 - accuracy: 0.7701\n",
            "Epoch 99/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.3139 - accuracy: 0.7706\n",
            "Epoch 100/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.3032 - accuracy: 0.7686\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "1fXTEO3GJ282",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 545
        },
        "outputId": "01e36337-807e-457f-f776-5540a9615c19"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "acc = history.history['accuracy']\n",
        "loss = history.history['loss']\n",
        "\n",
        "epochs = range(len(acc))\n",
        "\n",
        "plt.plot(epochs, acc, 'b', label='Training accuracy')\n",
        "plt.title('Training accuracy')\n",
        "\n",
        "plt.figure()\n",
        "\n",
        "plt.plot(epochs, loss, 'b', label='Training Loss')\n",
        "plt.title('Training loss')\n",
        "plt.legend()\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEICAYAAABPgw/pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deZgU1dXH8e+RRUGNCOIGKKi4xSX6jqBi1IhGUAH3gLtiMEZwwQ01bhjiEkXQAIqCuKMQF0QCkUVRQGUMBgXEjCTAsMgICCoIDnPeP26RtOPANNA9NV39+zxPP3RV3ek6RQ2H27fuYu6OiIjkvq3iDkBERDJDCV1EJCGU0EVEEkIJXUQkIZTQRUQSQgldRCQhlNClWjGzv5nZxZkuK5IPTP3QZUuZ2bcpm3WBNcC6aPsKd3++6qMSyT9K6JJRZvYf4HJ3H1vBsZruXlr1UeUW/T3J5lKTi2SNmR1vZsVmdrOZLQaeMrMdzWykmZWY2fLofeOUn3nbzC6P3l9iZu+Z2YNR2X+bWdvNLNvMzCaa2TdmNtbM+pnZcxuIu7IY65vZU2a2MDr+WsqxDmb2sZmtNLMvzKxNtP8/ZnZiSrm71p/fzJqamZtZZzObB4yP9g8zs8VmtiKK/ecpP1/HzB4ys7nR8feifW+aWbdy1zPdzM7Y1PsnuUcJXbJtV6A+sCfQhfA791S0vQewGvjLRn6+JTAb2Al4ABhkZrYZZV8APgQaAHcBF27knJXF+CyhaennwM7AwwBm1gJ4BrgRqAccC/xnI+cp7zjgAODkaPtvQPPoHP8AUpuuHgT+Dzia8Pd7E1AGPA1csL6QmR0KNALe3IQ4JFe5u156ZexFSGAnRu+PB9YC22yk/C+A5SnbbxOabAAuAYpSjtUFHNh1U8oSknIpUDfl+HPAc2le039jBHYjJM4dKyj3OPBwZX8v0fZd688PNI1i3WsjMdSLyuxA+A9nNXBoBeW2AZYDzaPtB4H+cf9e6FU1L9XQJdtK3P379RtmVtfMHo+aClYCE4F6ZlZjAz+/eP0bd18Vvd1uE8vuDixL2Qcwf0MBVxJjk+izllfwo02ALzb0uWn4b0xmVsPM7ouabVbyv5r+TtFrm4rOFf1dvwRcYGZbAZ0I3ygkDyihS7aVf+p+PbAf0NLdf0ZolgDYUDNKJiwC6ptZ3ZR9TTZSfmMxzo8+q14FPzcf2HsDn/kd4VvDertWUCb17+o8oANwIqFW3jQlhq+A7zdyrqeB84HWwCp3n7KBcpIwSuhS1bYnNBd8bWb1gTuzfUJ3nwsUAneZWW0zOwpotzkxuvsiQtt2/+jhaS0zW5/wBwGXmllrM9vKzBqZ2f7RsY+BjlH5AuDsSsLentD9cynhP4I/pcRQBgwGepvZ7lFt/igz2zo6PoXQLPQQqp3nFSV0qWp9gDqEWub7wOgqOu/5wFGEBPlHQrPEmg2UrSzGC4EfgM+AJcC1AO7+IXAp4SHpCuAdwoNVgNsJNerlwN2Eh7Qb8wwwF1gAzIziSHUD8AkwFVgG3M+P/z0/AxxMeFYgeUL90CUvmdlLwGfunvVvCHEws4uALu5+TNyxSNVRDV3ygpkdYWZ7R00hbQjt069V9nO5KHpW8HtgYNyxSNVSQpd8sSuhm+O3wCPAle4+LdaIssDMTgZKgC+pvFlHEkZNLiIiCZFWDd3M2pjZbDMrMrMeFRzfw8wmmNm0aJjxKZkPVURENqbSGno0mOJz4CSgmPBUvZO7z0wpMxCY5u4DzOxAYJS7N93Y5+60007etOlGi4iISDkfffTRV+7esKJjNdP4+RaEIdVzAMxsKOGB0syUMg78LHq/A7Cwsg9t2rQphYWFaZxeRETWM7O5GzqWTpNLI348TLo42pfqLsJQ42JgFNCNCphZFzMrNLPCkpKSNE4tIiLpylQvl07AEHdvDJwCPBvNI/Ej7j7Q3QvcvaBhwwq/MYiIyGZKJ6Ev4MfzXjSO9qXqDLwM/x12vA1hAiEREaki6ST0qUDzaIGA2kBHYES5MvMIEwFhZgcQErraVEREqlClCd3DUlhdgTHALOBld59hZj3NrH1U7Hrgt2b2T+BF4BJXB3cRkSqVTi8X3H0U4WFn6r47Ut7PBFplNjQREdkUGvovIpIQadXQRUQkfd98A2++CcuWwbffwpo10LQpHHQQHHAAbLNNds6rhC4ikiFLlsAjj0C/fvD11xWX2WqrcPx3v8v8+ZXQRUS2gDtMmgRPPAEvvxxq42ecAd27Q/PmsO22UKsWfPEFfPopfPIJFBRkJxYldBGRCqxeHZpOdt75f/uWLoVXX4UpU+C772DVKvj8c5g9G7bfHi6+GK67Dvbb76efd8AB4XXOOdmLWQldRCRSXAyvvw6jRsH48fD99yGhH3IImIV969ZBw4aw445Qty7ssQfcdBOcey5st1288Suhi0jeGTsW5s+Hxo2hUSP45z/hqafCfnfYZx/o0iU8yPzkk3B89Wq48cZQwz7ssJDgqxsldBHJG8uWQbdu8EIFazntuSfcfjucfz7su2/Vx5YJSugiknhlZfDaa9C1K5SUwN13h8S9YEGoqTdqBMceG3qg5DIldBFJrK++giFDYMAAmDMn9AN/883QZAKw996xhpdxSugikrPcYdw46N8/1MLr1QvdBOfODV0E50ZLQfzyl9CrF5x5JtSuHW/M2aSELiI5Z906mDgR7ror/LnrrqE3ytdfh66GjRvD0UeHB5unnRZ6qeQDJXQRqfbcYdq00Ad88mT48MMwpH633eDRR+G3v4Wtt447yvgpoYtI7L75BqZOhZo1Q2I2Cz1Sli0Lg3aGDg0DeGrUCLXtiy6CVq3CiMw6deKOvvpQQheR2LjD88+H/t2LF1dcxgyOOw5uuAHOOgvq16/aGHOJErqIxGLmzDBB1bvvwhFHwJNPhlkI16wJDzjr14cGDWCXXcLDTqmcErqIVKl166BPH7jttjBUfuBA6Nw59/uAVwdK6CJSJdzh44/h2mtDz5QOHUIyT538SrZMWgndzNoAfYEawJPufl+54w8Dv4o26wI7u7u+JInkue++C/OjvPlmmPBqwYIwK+FTT4WZCavjfCi5rNKEbmY1gH7ASUAxMNXMRkTriALg7tellO8GHJaFWEUkB7iHecGffjrMTrhmTUjiv/41nHJK6BeuWnl2pFNDbwEUufscADMbCnQAZm6gfCfgzsyEJyK5ZN48uOIKGD0a9toLrrwS2rULIzVr1Yo7uuRLJ6E3AuanbBcDLSsqaGZ7As2A8Rs43gXoArDHHntsUqAiUv3861+hf/iSJWGulL59w0PPvn3hqqtCv3GpOpl+KNoRGO7u6yo66O4DgYEABQUFnuFzi0gVWLgwDPR57rkwejPVSSfBY4+F2rlUvXQS+gKgScp242hfRToCV21pUCJSfSxeHHqlvPMOvP126D8OYV3MPn2gZcvQV3yXXcIKPhKfdBL6VKC5mTUjJPKOwHnlC5nZ/sCOwJSMRigiVe7rr+EPf4C33gpNKhD6jB9zTBh236ED7L9/vDHKT1Wa0N291My6AmMI3RYHu/sMM+sJFLr7iKhoR2Cou6spRSSHLV0aeqRMnw5t2sDll4fFHw4/XA82q7u02tDdfRQwqty+O8pt35W5sEQkDl9+CSeeGB52vv566GYouUODbUUEd5gwIdTE58wJA4GUzHOPErpIHvv++zDH+JFHwgknwMqVMGYMtG4dd2SyOTSXi0ieWbQo9BN/910oLIS1a6FZs7Du5iWXhBkPJTcpoYvkkZkzoW3b0Jf8iCPgmmtCM0ubNmFxCcltuoUieeKdd+D000MN/IMPQq8VSRa1oYsk3PLlcM89oSvibrvBlClK5kmlGrpIQpWUwH33hTnHv/02DAYaPFhLuCWZErpIAhUVhRr5vHnQqRPcdBMcfHDcUUm2KaGLJMw//hEefJaVweTJ0KJF3BFJVVEbukiCjB4Nxx8PderAe+8pmecbJXSRBFi9OnRBbNsWmjaFSZNgv/3ijkqqmhK6SI6bPDlMZfvII9CtW+iS2KhR3FFJHJTQRXJQaWlYt/Ooo6BVq9A1cfTokNTr1Ik7OomLHoqK5JB582DQoPBasAD23jsk8UsuCQsxS35TQhfJAUuWhOaUYcPCdps20L8/nHqq1u2U/1FCF6nmRo6Ezp1hxQq45Rbo0gX23DPuqKQ6UkIXqaZKS0PPlf794dBDYdw4OOiguKOS6iyth6Jm1sbMZptZkZn12ECZc81sppnNMLMXMhumSH5ZswbOOSck8+7dQ88VJXOpTKU1dDOrAfQDTgKKgalmNsLdZ6aUaQ7cArRy9+VmtnO2AhZJum+/hTPOgLFjw7zlV18dd0SSK9JpcmkBFLn7HAAzGwp0AGamlPkt0M/dlwO4+5JMByqSD776Ctq1gw8/hCFD4OKL445Ickk6TS6NgPkp28XRvlT7Avua2SQze9/M2lT0QWbWxcwKzaywpKRk8yIWSaiiotCvfNo0GD5cyVw2XaYeitYEmgPHA42BiWZ2sLt/nVrI3QcCAwEKCgo8Q+cWyXmTJ0P79uH9+PFw9NHxxiO5KZ0a+gKgScp242hfqmJghLv/4O7/Bj4nJHgR2Yi1a8PiE7/6Fey4Y1h8QslcNlc6CX0q0NzMmplZbaAjMKJcmdcItXPMbCdCE8ycDMYpkjjvvx9WDrrjjrA03JQp0FzVINkClSZ0dy8FugJjgFnAy+4+w8x6mln0JZExwFIzmwlMAG5096XZClok1z3xRJiDZcUKeOMNeOkl2GmnuKOSXGfu8TRlFxQUeGFhYSznFomLO9x/fxjx2bZtSOSag0U2hZl95O4FFR3TbIsiVaS0FG68MSTz886D119XMpfMUkIXyTJ3+Otfw0jPhx6Crl3h2WehVq24I5OkUUIXyaLPPoOWLeHss2GrreCVV8J0t1vpX55kgSbnEsmScePgrLOgdm0YPBguukhT3Up2KaGLZMETT8Dvfx/W9Rw5MqzzKZJt+uInkkHr1sH114c5y1u3Dos1K5lLVVENXSRDVqyAjh3D2p7dukHv3lBT/8KkCunXTSQD5s4Ny8IVFcHjj4caukhVU0IX2UKrV4eh+4sXhznMjzsu7ogkXymhi2yhq6+Gjz8ODz+VzCVOeigqsgWGDIEnn4Rbb4VTT407Gsl3Sugim2nSJLjyyjD17d13xx2NiBK6yCZbtAguvRSOOQZ23hlefFG9WaR6UEIX2QRDh8K++8ILL8DNN8Mnn8Auu8QdlUigeoVImj78MKzzecQRoe18n33ijkjkx5TQRdLw5Zdw5pmw++5h2tsGDeKOSOSnlNBFKvHDD3DuubB0aVjMWclcqqu02tDNrI2ZzTazIjPrUcHxS8ysxMw+jl6XZz5UkarlDhMmwGmnwcSJYcKtww6LOyqRDau0hm5mNYB+wElAMTDVzEa4+8xyRV9y965ZiFGkyg0bFroizpgRauS9e8MFF8QdlcjGpVNDbwEUufscd18LDAU6ZDcskXiUlYUl4s49N3RFHDIEiovhuuvijkykcum0oTcC5qdsFwMtKyh3lpkdC3wOXOfu8ysoI1JtffcdXHghvPoqXHEFPPqolomT3JKpfuhvAE3d/RDgLeDpigqZWRczKzSzwpKSkgydWmTLff89nHRS6MHSpw8MGKBkLrknnYS+AGiSst042vdf7r7U3ddEm08C/1fRB7n7QHcvcPeChg0bbk68IhnnHlYXmjIlDBy65howizsqkU2XTkKfCjQ3s2ZmVhvoCIxILWBmu6VstgdmZS5EkewaMACeegpuvx3OOSfuaEQ2X6Vt6O5eamZdgTFADWCwu88ws55AobuPAK42s/ZAKbAMuCSLMYtkzLvvhhr5aafBXXfFHY3IljF3j+XEBQUFXlhYGMu5RQDmzQvD+OvVC8P6d9gh7ohEKmdmH7l7QUXHNDmX5KVVq+CMM8LD0NdfVzKXZNDQf8k77tC5M0ybBm+8AfvvH3dEIpmhhC555777Qm+We+/VKkOSLErokjd++AGuvz4MGOrUKcxnLpIkSuiSFxYtCsP533svDOO//371NZfkUUKXxJszB449FpYvDysNdeoUd0Qi2aGELom2eHEY0r96dZjL/NBD445IJHuU0CWxvv4a2rQJqw2NG6dkLsmnhC6JtGYNtG8PM2fCyJHQsqL5QUUSRgldEscdrroqDOt/8UX49a/jjkikamikqCTO44/DoEFw223QsWPc0YhUHSV0SZRJk+Dqq6Ft27CEnEg+UUKXxFiyBM4+G/bcM3RPrFEj7ohEqpba0CUxrr8eli6FMWPCDIoi+UY1dEmEsWPhueegRw845JC4oxGJhxK65LzVq+HKK2GffeDWW+OORiQ+anKRnNerFxQVhVr6NtvEHY1IfFRDl5z23nvwwANw4YXQunXc0YjEK62EbmZtzGy2mRWZWY+NlDvLzNzMKlweSSSTxo6Fk0+GZs3goYfijkYkfpUmdDOrAfQD2gIHAp3M7MAKym0PXAN8kOkgRcp7442wsPNee8HEidCwYdwRicQvnRp6C6DI3ee4+1pgKNChgnL3APcD32cwPpGfGD0azjwTDj4Y3n4bdtkl7ohEqod0EnojYH7KdnG077/M7HCgibu/ubEPMrMuZlZoZoUlJSWbHKzIrFnwm9/AQQeFJpcGDeKOSKT62OKHoma2FdAbuL6ysu4+0N0L3L2gob4jyyZavjzMoLjNNvD667DDDnFHJFK9pNNtcQHQJGW7cbRvve2Bg4C3LazptSswwszau3thpgKV/FZaGpaQmzsXJkyAPfaIOyKR6iedhD4VaG5mzQiJvCNw3vqD7r4C2Gn9tpm9DdygZC6ZdMMNoYll8GBo1SruaESqp0qbXNy9FOgKjAFmAS+7+wwz62lm7bMdoMiQIdC3L1xzDVx6adzRiFRf5u6xnLigoMALC1WJl4374IOwwPMvfxl6t9TU2GbJc2b2kbtXONZHI0Wl2lq4EM44Axo1gpdeUjIXqYz+iUi1dcMNYaHnDz5Q90SRdKiGLtXS9OlhPdDrrgsDiESkckroUi3dfnvoZ37DDXFHIpI7lNCl2nn/fRgxAm66CXbcMe5oRHKHErpUO7fdBjvvHBZ7FpH06aGoVCujR8P48aHf+XbbxR2NSG5RDV2qjaeegtNPD0vJXXFF3NGI5B4ldIndmjUhgV92GRxzDEyeDFtvHXdUIrlHCV1itXYttGsHAwdCjx6hyUUTcYpsHrWhS2zKykKt/K23YNCg8F5ENp9q6BKbm2+G55+HP/1JyVwkE5TQJRb9+sGDD0LXrqGpRUS2nBK6VLl//zuMAD31VOjTB8K6KCKypZTQpUq5Q7duYebExx6DGjXijkgkOfRQVKrUa6/Bm2/CQw9B48ZxRyOSLKqhS5X59tswnP+QQzSsXyQb0kroZtbGzGabWZGZ/eQRlpn9zsw+MbOPzew9Mzsw86FKruvZE4qLYcAALVYhkg2VJnQzqwH0A9oCBwKdKkjYL7j7we7+C+ABoHfGI5WcVlQUHoBedhkcfXTc0YgkUzo19BZAkbvPcfe1wFCgQ2oBd1+ZsrktEM9CpVJt3Xwz1K4NvXrFHYlIcqXzxbcRMD9luxhoWb6QmV0FdAdqAydU9EFm1gXoArDHHntsaqySoyZOhFdegXvugV13jTsakeTK2ENRd+/n7nsDNwN/2ECZge5e4O4FDTVhR14oK4Pu3UOPlu7d445GJNnSqaEvAJqkbDeO9m3IUGDAlgQlyfHCC/DRR/Dss1C3btzRiCRbOjX0qUBzM2tmZrWBjsCI1AJm1jxl81TgX5kLUXLVnDmhVl5QAOedF3c0IslXaQ3d3UvNrCswBqgBDHb3GWbWEyh09xFAVzM7EfgBWA5cnM2gpfpbuhTatoXS0lA730ojHkSyLq3ewO4+ChhVbt8dKe+vyXBcksNWr4YOHWDu3DA17v77xx2RSH7Q8A7JKHe45BKYNAleegl++cu4IxLJH/oiLBk1eDC8/DLcey+ce27c0YjkFyV0yZgvvoBrroETToCbboo7GpH8o4QuGVFaChdeGOZoGTJED0FF4qA2dMmIBx6AKVPCknJNmlReXkQyT/Uo2WL//CfceWdoM+/UKe5oRPKXErpskbVr4eKLoUED6N9fy8mJxElNLrJF/vSnUEN/7bWQ1EUkPqqhy2abNi1Mh3vBBWEgkYjESwldNsuaNWEAUcOG0Ldv3NGICKjJRTbTrbfC9OkwciTUrx93NCICqqHLZhg9Gnr3hq5d4dRT445GRNZTQpdN8uWXoVfLQQeFvuciUn2oyUXSVlYGl14KK1fCuHFQp07cEYlIKiV0SVuvXvC3v0G/fqGGLiLVi5pcJC0jRsAdd4T5Wq68Mu5oRKQiSuhSqVmzQl/zggJ4/HGNBhWprpTQZaNWrAiDhurWhVdfVbu5SHWWVkI3szZmNtvMisysRwXHu5vZTDObbmbjzGzPzIcqcejWLSz2PHw4NG4cdzQisjGVJnQzqwH0A9oCBwKdzOzAcsWmAQXufggwHFCHtgQYNiws8PyHP8Axx8QdjYhUJp0aegugyN3nuPtaYCjwo5k73H2Cu6+KNt8HVJfLcQsXwu9+B0ccAbfdFnc0IpKOdBJ6I2B+ynZxtG9DOgN/q+iAmXUxs0IzKywpKUk/SqlS7qG/+erVoYZeq1bcEYlIOjL6UNTMLgAKgD9XdNzdB7p7gbsXNGzYMJOnlgxxh+7d4e9/hwcfhP32izsiEUlXOgOLFgCpi4o1jvb9iJmdCNwGHOfuazITnlSlsjL4/e9D18Srr1Z/c5Fck04NfSrQ3MyamVltoCMwIrWAmR0GPA60d/clmQ9Tsm3dOujcOSTzHj2gTx/1NxfJNZUmdHcvBboCY4BZwMvuPsPMeppZ+6jYn4HtgGFm9rGZjdjAx0k1tG4dXHYZDBkCd98dViFSMhfJPWnN5eLuo4BR5fbdkfL+xAzHJVWkrAy6dIFnnoF77gldFEUkN2mkaB5zD23mgweHeVqUzEVymxJ6niorC/3MH38cbrkF7ror7ohEZEtp+tw8VFoa+pk/91xI5r16qc1cJAmU0PPM2rVw3nnw17+GRH7rrXFHJCKZoiaXPDJ5MrRoEZL5ww8rmYskjRJ6Hli6NHRLbNUKli2D116Da6+NOyoRyTQ1uSTc8uVw/PHw2Wdw881w++2w7bZxRyUi2aCEnmCrVkG7dvD55zB6NLRuHXdEIpJNSugJVVoKv/lNaDd/+WUlc5F8oISeQGVlYV6WkSOhf384++y4IxKRqqCHognjDlddFYby9+ypGRNF8okSeoK4w/XXw2OPhRkTNZRfJL8ooSfEsmWhNv7ww2Euc82YKJJ/1Iae41atgkcfhXvvhZUrw2pDDz6oZC6Sj5TQc9CSJfDGGzBqFLz1FnzzDZx2WqiVH3xw3NGJSFyU0HPMO+9Ahw6wYgU0bgydOsGFF8Ixx8QdmYjETQk9hwwbBhdcAHvvDRMmwC9+oaYVEfmftB6KmlkbM5ttZkVm1qOC48ea2T/MrNTM1Os5w8rKoHfvMFDoiCPgvffgsMOUzEXkxypN6GZWA+gHtAUOBDqZ2YHlis0DLgFeyHSA+W7yZGjZMnRHPP300GZev37cUYlIdZRODb0FUOTuc9x9LTAU6JBawN3/4+7TgbIsxJiX5s8P85a3agULF4bFKIYPhzp14o5MRKqrdBJ6I2B+ynZxtG+TmVkXMys0s8KSkpLN+YjEW7UqjPDcbz945RW47TaYPRvOPx+20qgBEdmIKk0R7j7Q3QvcvaBhw4ZVeepqzx2GDoUDDoA77wzdED/7DP74R9huu7ijE5FckE5CXwA0SdluHO2TDJk6NXQ77NQptI+//XaYIbFp07gjE5Fckk5Cnwo0N7NmZlYb6AiMyG5Y+WHaNDjjjLAs3BdfwJNPQmEhHHdc3JGJSC6qNKG7eynQFRgDzAJedvcZZtbTzNoDmNkRZlYMnAM8bmYzshl0ListDSM827WDww8PtfE77wyLUHTuDDVqxB2hiOSqtAYWufsoYFS5fXekvJ9KaIqRCpSVwYcfhmaU558PQ/cbNIB77oFu3WCHHeKOUESSQCNFs6ioCPr2hVdfhQULoFatUDO/6CJo2xZq1447QhFJEiX0LFi1Ksx++MADoath27Zw5pmh50q9enFHJyJJpYSeIe4wa1ZoH//LX2Du3NB3/M9/ht12izs6EckHSuhb6NNPYciQMIpz7tywr6AAnn5avVVEpGopoW+Gr76Cl14KibywEGrWhDZt4NZbQ/NKkyaVfoSISMYpoadp5UoYPRpeeCE0q/zwAxx6KPTpE+Zc0cBXEYmbEnoFFi2CmTNDz5TiYpg4EcaPD0l8113Dmp0XXQSHHBJ3pCIi/6OEHpk/PywgMXw4TJny42N77x2S+Omnw1FHafCPiFRPeZ/QFy4MA3yefDKM4jzsMOjVKyTuRo3Ca9tt445SRKRyeZfQ160LA34+/jis/DNoUGhK6dIFuncPtXERkVyUFwn9s8/g73+HsWPD3CnffBP216oVlnW7+27Ya69YQxQR2WKJTejffAMvvghPPBG6FgLss0/okXLkkWGB5QMOgK23jjdOEZFMSUxCX7s29EZ5912YNCmsxbl6Nfz85/Dww+GBpuYXF5Eky9mE7g4rVsD06aFv+LBhsGxZmDvl0EPh8stDbbxlSzCLO1oRkezLuYQ+aFCY+GrhwlADB6hbFzp0CCv+HHcc/Oxn8cYoIhKHnEvoO+8cVvjZffcw6VXTpnDyyVp3U0Qk5xJ6u3bhJSIiP5bOmqKYWRszm21mRWbWo4LjW5vZS9HxD8ysaaYDFRGRjas0oZtZDaAf0BY4EOhkZgeWK9YZWO7u+wAPA/dnOlAREdm4dGroLYAid5/j7muBoUCHcmU6AE9H74cDrc3Ut0REpCqlk9AbAfNTtoujfRWWcfdSYAXQIBMBiohIetJqQ88UM+tiZoVmVlhSUlKVpxYRSbx0EvoCIHUNnsbRvgrLmFlNYAdgafkPcveB7l7g7gUNtSKEiEhGpZPQpwLNzayZmdUGOgIjypUZAVwcvT8bGO/unrkwRUSkMpX2Q3f3UjPrCo1S+z8AAAPQSURBVIwBagCD3X2GmfUECt19BDAIeNbMioBlhKQvIiJVyOKqSJtZCTB3M398J+CrDIaTK/LxuvPxmiE/rzsfrxk2/br3dPcK26xjS+hbwswK3b0g7jiqWj5edz5eM+TndefjNUNmr7tKe7mIiEj2KKGLiCRErib0gXEHEJN8vO58vGbIz+vOx2uGDF53Trahi4jIT+VqDV1ERMpRQhcRSYicS+iVzc2eBGbWxMwmmNlMM5thZtdE++ub2Vtm9q/ozx3jjjXTzKyGmU0zs5HRdrNojv2iaM792nHHmGlmVs/MhpvZZ2Y2y8yOypN7fV30+/2pmb1oZtsk7X6b2WAzW2Jmn6bsq/DeWvBIdO3TzezwTT1fTiX0NOdmT4JS4Hp3PxA4Ergqus4ewDh3bw6Mi7aT5hpgVsr2/cDD0Vz7ywlz7ydNX2C0u+8PHEq4/kTfazNrBFwNFLj7QYRR6B1J3v0eArQpt29D97Yt0Dx6dQEGbOrJciqhk97c7DnP3Re5+z+i998Q/oE34sfzzj8NnB5PhNlhZo2BU4Eno20DTiDMsQ/JvOYdgGMJ02fg7mvd/WsSfq8jNYE60YR+dYFFJOx+u/tEwnQoqTZ0bzsAz3jwPlDPzHbblPPlWkJPZ272RImW8zsM+ADYxd0XRYcWA7vEFFa29AFuAsqi7QbA19Ec+5DM+90MKAGeipqanjSzbUn4vXb3BcCDwDxCIl8BfETy7zds+N5ucX7LtYSeV8xsO+CvwLXuvjL1WDSbZWL6nJrZacASd/8o7liqWE3gcGCAux8GfEe55pWk3WuAqN24A+E/tN2Bbflp00TiZfre5lpCT2du9kQws1qEZP68u78S7f5y/Vew6M8lccWXBa2A9mb2H0JT2gmEtuV60VdySOb9LgaK3f2DaHs4IcEn+V4DnAj8291L3P0H4BXC70DS7zds+N5ucX7LtYSeztzsOS9qOx4EzHL33imHUuedvxh4vapjyxZ3v8XdG7t7U8J9He/u5wMTCHPsQ8KuGcDdFwPzzWy/aFdrYCYJvteRecCRZlY3+n1ff92Jvt+RDd3bEcBFUW+XI4EVKU0z6XH3nHoBpwCfA18At8UdT5au8RjC17DpwMfR6xRCm/I44F/AWKB+3LFm6fqPB0ZG7/cCPgSKgGHA1nHHl4Xr/QVQGN3v14Ad8+FeA3cDnwGfAs8CWyftfgMvEp4R/ED4NtZ5Q/cWMEIvvi+ATwg9gDbpfBr6LyKSELnW5CIiIhughC4ikhBK6CIiCaGELiKSEEroIiIJoYQuIpIQSugiIgnx/7UhFOT1jfmqAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "6Vc6PHgxa6Hm",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 105
        },
        "outputId": "516df603-ca94-4fef-d0c6-9a5782b2ddfe"
      },
      "source": [
        "seed_text = \"Help me Obi Wan Kenobi, you're my only hope\"\n",
        "next_words = 100\n",
        "  \n",
        "for _ in range(next_words):\n",
        "\ttoken_list = tokenizer.texts_to_sequences([seed_text])[0]\n",
        "\ttoken_list = pad_sequences([token_list], maxlen=max_sequence_len-1, padding='pre')\n",
        "\tpredicted = model.predict_classes(token_list, verbose=0)\n",
        "\toutput_word = \"\"\n",
        "\tfor word, index in tokenizer.word_index.items():\n",
        "\t\tif index == predicted:\n",
        "\t\t\toutput_word = word\n",
        "\t\t\tbreak\n",
        "\tseed_text += \" \" + output_word\n",
        "print(seed_text)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From <ipython-input-7-622d307fa19a>:7: Sequential.predict_classes (from tensorflow.python.keras.engine.sequential) is deprecated and will be removed after 2021-01-01.\n",
            "Instructions for updating:\n",
            "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",
            "Help me Obi Wan Kenobi, you're my only hope but are not mine own rhyme might all thee thine thee such life thy 'will ' ' ' are in me alone alone still much deem'd ' ill love to thee thine such thee less in hue one tongue away none hate none none hate best grow are ill 'no ' are blind hast more her hate of such shade torn well ' new shouldst grow ill new ' 'tis still still thine alone hast her that i prove you blind survey ' lies still still thine thine own up disgrace in me more skill more worth of none none\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tgXrDQdxwU3x",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}