Poetry_Construction.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "name": "Poetry_Construction.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/eddie-coding/Poetry-Construction/blob/master/Poetry_Construction.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": 0,
      "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": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "PRnDnCW-Z7qv",
        "outputId": "e939df19-2ee6-448e-abf4-1685d501e588",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        }
      },
      "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": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2020-04-07 18:19:24--  https://storage.googleapis.com/laurencemoroney-blog.appspot.com/sonnets.txt\n",
            "Resolving storage.googleapis.com (storage.googleapis.com)... 173.194.79.128, 2a00:1450:4013:c00::80\n",
            "Connecting to storage.googleapis.com (storage.googleapis.com)|173.194.79.128|: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",
            "2020-04-07 18:19:25 (162 MB/s) - ‘/tmp/sonnets.txt’ saved [93578/93578]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "w9vH8Y59ajYL",
        "outputId": "e07a4ddd-ea2d-4466-9e77-7d6bef47a9c4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 374
        }
      },
      "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": 0,
      "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",
        "outputId": "95fa6835-d02d-4d5b-d038-963738f5a08d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        " history = model.fit(predictors, label, epochs=100, verbose=1)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 6.9061 - accuracy: 0.0213\n",
            "Epoch 2/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 6.5057 - accuracy: 0.0240\n",
            "Epoch 3/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 6.3980 - accuracy: 0.0268\n",
            "Epoch 4/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 6.2739 - accuracy: 0.0292\n",
            "Epoch 5/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 6.1812 - accuracy: 0.0349\n",
            "Epoch 6/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 6.0957 - accuracy: 0.0387\n",
            "Epoch 7/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 6.0192 - accuracy: 0.0419\n",
            "Epoch 8/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 5.9418 - accuracy: 0.0453\n",
            "Epoch 9/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 5.8458 - accuracy: 0.0518\n",
            "Epoch 10/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 5.7419 - accuracy: 0.0590\n",
            "Epoch 11/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 5.6409 - accuracy: 0.0636\n",
            "Epoch 12/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 5.5437 - accuracy: 0.0695\n",
            "Epoch 13/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 5.4454 - accuracy: 0.0757\n",
            "Epoch 14/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 5.3457 - accuracy: 0.0815\n",
            "Epoch 15/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 5.2517 - accuracy: 0.0878\n",
            "Epoch 16/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 5.1556 - accuracy: 0.0955\n",
            "Epoch 17/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 5.0541 - accuracy: 0.1043\n",
            "Epoch 18/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 4.9541 - accuracy: 0.1148\n",
            "Epoch 19/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 4.8583 - accuracy: 0.1201\n",
            "Epoch 20/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 4.7610 - accuracy: 0.1279\n",
            "Epoch 21/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 4.6610 - accuracy: 0.1416\n",
            "Epoch 22/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 4.5647 - accuracy: 0.1490\n",
            "Epoch 23/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 4.4665 - accuracy: 0.1597\n",
            "Epoch 24/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 4.3684 - accuracy: 0.1716\n",
            "Epoch 25/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 4.2754 - accuracy: 0.1807\n",
            "Epoch 26/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 4.1670 - accuracy: 0.1959\n",
            "Epoch 27/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 4.0796 - accuracy: 0.2044\n",
            "Epoch 28/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.9825 - accuracy: 0.2200\n",
            "Epoch 29/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 3.8880 - accuracy: 0.2306\n",
            "Epoch 30/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.7772 - accuracy: 0.2513\n",
            "Epoch 31/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 3.6908 - accuracy: 0.2674\n",
            "Epoch 32/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 3.5998 - accuracy: 0.2853\n",
            "Epoch 33/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 3.5117 - accuracy: 0.3023\n",
            "Epoch 34/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 3.4298 - accuracy: 0.3220\n",
            "Epoch 35/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.3392 - accuracy: 0.3460\n",
            "Epoch 36/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 3.2523 - accuracy: 0.3627\n",
            "Epoch 37/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.1791 - accuracy: 0.3804\n",
            "Epoch 38/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.1010 - accuracy: 0.3993\n",
            "Epoch 39/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 3.0218 - accuracy: 0.4126\n",
            "Epoch 40/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 2.9648 - accuracy: 0.4248\n",
            "Epoch 41/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 2.8854 - accuracy: 0.4457\n",
            "Epoch 42/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 2.8143 - accuracy: 0.4582\n",
            "Epoch 43/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.7543 - accuracy: 0.4735\n",
            "Epoch 44/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.7011 - accuracy: 0.4860\n",
            "Epoch 45/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 2.6347 - accuracy: 0.5002\n",
            "Epoch 46/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.5795 - accuracy: 0.5109\n",
            "Epoch 47/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 2.5387 - accuracy: 0.5199\n",
            "Epoch 48/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 2.4627 - accuracy: 0.5369\n",
            "Epoch 49/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.4210 - accuracy: 0.5463\n",
            "Epoch 50/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.3846 - accuracy: 0.5592\n",
            "Epoch 51/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 2.3379 - accuracy: 0.5642\n",
            "Epoch 52/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.2899 - accuracy: 0.5743\n",
            "Epoch 53/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.2332 - accuracy: 0.5878\n",
            "Epoch 54/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.1927 - accuracy: 0.5975\n",
            "Epoch 55/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.1492 - accuracy: 0.6059\n",
            "Epoch 56/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 2.1170 - accuracy: 0.6116\n",
            "Epoch 57/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 2.0753 - accuracy: 0.6206\n",
            "Epoch 58/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 2.0289 - accuracy: 0.6308\n",
            "Epoch 59/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.9993 - accuracy: 0.6369\n",
            "Epoch 60/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.9698 - accuracy: 0.6418\n",
            "Epoch 61/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.9315 - accuracy: 0.6553\n",
            "Epoch 62/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.9020 - accuracy: 0.6586\n",
            "Epoch 63/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.8809 - accuracy: 0.6636\n",
            "Epoch 64/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 1.8382 - accuracy: 0.6734\n",
            "Epoch 65/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.8079 - accuracy: 0.6783\n",
            "Epoch 66/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 1.7830 - accuracy: 0.6835\n",
            "Epoch 67/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.7514 - accuracy: 0.6891\n",
            "Epoch 68/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.7171 - accuracy: 0.6993\n",
            "Epoch 69/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.7027 - accuracy: 0.6980\n",
            "Epoch 70/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.6791 - accuracy: 0.7030\n",
            "Epoch 71/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.6521 - accuracy: 0.7078\n",
            "Epoch 72/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.6330 - accuracy: 0.7112\n",
            "Epoch 73/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.6071 - accuracy: 0.7176\n",
            "Epoch 74/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 1.5837 - accuracy: 0.7233\n",
            "Epoch 75/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.5562 - accuracy: 0.7278\n",
            "Epoch 76/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.5288 - accuracy: 0.7325\n",
            "Epoch 77/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 1.5014 - accuracy: 0.7376\n",
            "Epoch 78/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 1.4908 - accuracy: 0.7427\n",
            "Epoch 79/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.4871 - accuracy: 0.7399\n",
            "Epoch 80/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 1.4683 - accuracy: 0.7463\n",
            "Epoch 81/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.4462 - accuracy: 0.7463\n",
            "Epoch 82/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.4250 - accuracy: 0.7516\n",
            "Epoch 83/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.4036 - accuracy: 0.7581\n",
            "Epoch 84/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.3984 - accuracy: 0.7558\n",
            "Epoch 85/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.3780 - accuracy: 0.7606\n",
            "Epoch 86/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.3581 - accuracy: 0.7621\n",
            "Epoch 87/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.3321 - accuracy: 0.7705\n",
            "Epoch 88/100\n",
            "484/484 [==============================] - 6s 12ms/step - loss: 1.3150 - accuracy: 0.7765\n",
            "Epoch 89/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.3219 - accuracy: 0.7707\n",
            "Epoch 90/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.3030 - accuracy: 0.7733\n",
            "Epoch 91/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.2916 - accuracy: 0.7753\n",
            "Epoch 92/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.2769 - accuracy: 0.7791\n",
            "Epoch 93/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.2659 - accuracy: 0.7807\n",
            "Epoch 94/100\n",
            "484/484 [==============================] - 6s 11ms/step - loss: 1.2461 - accuracy: 0.7822\n",
            "Epoch 95/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.2369 - accuracy: 0.7851\n",
            "Epoch 96/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.2166 - accuracy: 0.7897\n",
            "Epoch 97/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.2139 - accuracy: 0.7856\n",
            "Epoch 98/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.2058 - accuracy: 0.7876\n",
            "Epoch 99/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.1914 - accuracy: 0.7930\n",
            "Epoch 100/100\n",
            "484/484 [==============================] - 5s 11ms/step - loss: 1.1769 - accuracy: 0.7945\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "1fXTEO3GJ282",
        "outputId": "c3805d31-b85c-4475-de90-b1d4d0c8008f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 545
        }
      },
      "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": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEICAYAAABPgw/pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWoAAAEICAYAAAB25L6yAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "6Vc6PHgxa6Hm",
        "outputId": "3fe6998b-682f-4bfb-aeb9-541954cc61d0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 105
        }
      },
      "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": 0,
      "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 thy sins will prove his side defence light arising beard torn light dyed brought of sense number'd hours 'tis o'er gone within your days lies writ eye still still find some skill but foes goes some near waste did stand there with his tomb thence commend happy did stand did stand know there there am for mine eye doth shouldst mine eyes but straight skill wanting pain be bright thine own grace desire too eye doth sort dyed dyed find unbred qualify had above a store 'tis rare rare dote light dote burn day rhymers worn gone bred new fired\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}