Character_Level_RNN.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Character_Level_RNN.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/paulgureghian/Google_Colab_Notebooks/blob/master/Character_Level_RNN.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "metadata": {
        "id": "JpDLO5pTYkiq",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Created by Paul A. Gureghian on 12/17/18.**"
      ]
    },
    {
      "metadata": {
        "id": "KsrNbKaxLqGf",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Character-Level LSTM in PyTorch.**\n",
        "\n",
        "**In this notebook, I'll construct a character-level LSTM with PyTorch.**\n",
        "\n",
        "**The network will train character by character on some text, then generate new text character by character.**\n",
        "\n",
        "**As an example, I will train on Anna Karenina.**\n",
        "\n",
        "**This model will be able to generate new text based on the text from the book.**"
      ]
    },
    {
      "metadata": {
        "id": "n2lfzOBdao9B",
        "colab_type": "code",
        "outputId": "eaf60d39-7ef4-4bc1-d977-1662e510aa3a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "### Mount Google drive\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "AZLPTMzafjVE",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "### Install Pillow\n",
        "!pip install -q pillow==4.1.1  "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "fuNVVqu6bBD0",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "### Install PyTorch\n",
        "from os.path import exists\n",
        "from wheel.pep425tags import get_abbr_impl, get_impl_ver, get_abi_tag\n",
        "platform = '{}{}-{}'.format(get_abbr_impl(), get_impl_ver(), get_abi_tag())\n",
        "cuda_output = !ldconfig -p|grep cudart.so|sed -e 's/.*\\.\\([0-9]*\\)\\.\\([0-9]*\\)$/cu\\1\\2/'\n",
        "accelerator = cuda_output[0] if exists('/dev/nvidia0') else 'cpu'\n",
        "\n",
        "!pip install -q http://download.pytorch.org/whl/{accelerator}/torch-0.4.1-{platform}-linux_x86_64.whl torchvision"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "ApTl2758LqGh",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**First let's load in the required resources for data loading and model creation.**"
      ]
    },
    {
      "metadata": {
        "id": "8ssR_PBwLqGi",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "### Import packages\n",
        "import PIL\n",
        "import torch\n",
        "import numpy as np\n",
        "from torch import nn\n",
        "import torch.nn.functional as F\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.image as mpimg"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "y_54taqWcfrO",
        "colab_type": "code",
        "outputId": "354b3d40-217a-41eb-97f6-b256b105d68e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 351
        }
      },
      "cell_type": "code",
      "source": [
        "### Character-wise RNN network image\n",
        "character_sequence = mpimg.imread(\"/content/drive/My Drive/Images/charseq.jpeg\") \n",
        "plt.imshow(character_sequence)\n",
        "plt.grid(False)"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZoAAAFOCAYAAAC/qdtaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fc0271f8588>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "Gx0Wf0F1LqGo",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Load in Data.**\n",
        "\n",
        "**Then, I will load the Anna Karenina text file and convert it into integers for my network to use.** "
      ]
    },
    {
      "metadata": {
        "id": "st84yFvuLqGp",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "### Open text file and read in data as `text`\n",
        "with open(\"/content/drive/My Drive/Images/anna.txt\", 'r') as f:\n",
        "    text = f.read()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "O8HNXxnnLqGu",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Let's check out the first 100 characters, to make sure everything is correct.**"
      ]
    },
    {
      "metadata": {
        "id": "Wc_y3PURLqGv",
        "colab_type": "code",
        "outputId": "7e27c685-7882-4f51-9d6d-765550d51bcf",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "### Check the first 100 characters\n",
        "text[:100]"
      ],
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'Chapter 1\\n\\n\\nHappy families are all alike; every unhappy family is unhappy in its own\\nway.\\n\\nEverythin'"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 40
        }
      ]
    },
    {
      "metadata": {
        "id": "owasevOZLqG3",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Tokenization.**\n",
        "\n",
        "**In the cells below, I'm creating a couple of dictionaries to convert the characters to and from integers.** \n",
        "\n",
        "**Encoding the characters as integers makes it easier to use as input in the network.**"
      ]
    },
    {
      "metadata": {
        "id": "01MDYNfgLqG5",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "### Encode the text and map each character to an integer and vice versa\n",
        "\n",
        "### Create two dictionaries:\n",
        "# 1. int2char, which maps integers to characters\n",
        "# 2. char2int, which maps characters to unique integers\n",
        "chars = tuple(set(text))\n",
        "int2char = dict(enumerate(chars))\n",
        "char2int = {ch: ii for ii, ch in int2char.items()}\n",
        "\n",
        "### Encode the text\n",
        "encoded = np.array([char2int[ch] for ch in text])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "WP9mohaqLqG8",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**And we can see those same characters from above, encoded as integers.**"
      ]
    },
    {
      "metadata": {
        "id": "RvNkv30KLqG-",
        "colab_type": "code",
        "outputId": "2d91f521-6e96-4c70-9191-035aa241dcac",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 119
        }
      },
      "cell_type": "code",
      "source": [
        "### The encoded text\n",
        "encoded[:100]"
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([44, 79, 35, 61, 59,  3, 36, 26, 76, 58, 58, 58, 23, 35, 61, 61, 16,\n",
              "       26, 81, 35, 39, 75, 38, 75,  3, 18, 26, 35, 36,  3, 26, 35, 38, 38,\n",
              "       26, 35, 38, 75, 60,  3,  0, 26,  3, 57,  3, 36, 16, 26, 41, 21, 79,\n",
              "       35, 61, 61, 16, 26, 81, 35, 39, 75, 38, 16, 26, 75, 18, 26, 41, 21,\n",
              "       79, 35, 61, 61, 16, 26, 75, 21, 26, 75, 59, 18, 26, 54, 82, 21, 58,\n",
              "       82, 35, 16, 24, 58, 58, 73, 57,  3, 36, 16, 59, 79, 75, 21])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "metadata": {
        "id": "hZHAw3YALqHE",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Pre-processing the data.**\n",
        "\n",
        "**As you can see in my char-RNN image above, my LSTM expects an input that is one-hot encoded** \n",
        "\n",
        "**Meaning that each character is converted into an integer (via my created dictionary).** \n",
        "\n",
        "**And then converted into a column vector where only it's corresponding integer index will have the value of 1 and the rest of the vector will be filled with 0's.**\n",
        "\n",
        "**Since I am one-hot encoding the data, I will make a function to do that.**"
      ]
    },
    {
      "metadata": {
        "id": "3yV-PCVdLqHE",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "### Define a 'one_hot_encode' function\n",
        "def one_hot_encode(arr, n_labels):\n",
        "    \n",
        "    ### Initialize the the encoded array\n",
        "    one_hot = np.zeros((np.multiply(*arr.shape), n_labels), dtype=np.float32)\n",
        "    \n",
        "    ### Fill the appropriate elements with ones\n",
        "    one_hot[np.arange(one_hot.shape[0]), arr.flatten()] = 1.\n",
        "    \n",
        "    ### Finally reshape it to get back to the original array\n",
        "    one_hot = one_hot.reshape((*arr.shape, n_labels))\n",
        "    \n",
        "    return one_hot"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "EZTB2bbqLqHI",
        "colab_type": "code",
        "outputId": "ebd200a5-efbb-4d71-866c-713a16d8cbb2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        }
      },
      "cell_type": "code",
      "source": [
        "### Check that the function works as expected\n",
        "test_seq = np.array([[3, 5, 1]])\n",
        "one_hot = one_hot_encode(test_seq, 8)\n",
        "\n",
        "print(one_hot)"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[[0. 0. 0. 1. 0. 0. 0. 0.]\n",
            "  [0. 0. 0. 0. 0. 1. 0. 0.]\n",
            "  [0. 1. 0. 0. 0. 0. 0. 0.]]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "8pQ_UYziLqHO",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Making training mini-batches.**\n",
        "\n",
        "\n",
        "**In this example, I'll take the encoded characters (passed in as the `arr` parameter) and split them into multiple sequences, given by `batch_size`.**\n",
        "\n",
        "**Each of the sequences will be `seq_length` long.**\n",
        "\n",
        "**Creating Batches.**\n",
        "\n",
        "**1. The first thing we need to do is discard some of the text so we only have completely full mini-batches. **\n",
        "\n",
        "Each batch contains $N \\times M$ characters, where $N$ is the batch size (the number of sequences in a batch) and $M$ is the seq_length or number of time steps in a sequence. Then, to get the total number of batches, $K$, that we can make from the array `arr`, you divide the length of `arr` by the number of characters per batch. Once you know the number of batches, you can get the total number of characters to keep from `arr`, $N * M * K$.\n",
        "\n",
        "**2. After that, we need to split `arr` into $N$ batches. ** \n",
        "\n",
        "You can do this using `arr.reshape(size)` where `size` is a tuple containing the dimensions sizes of the reshaped array. We know we want $N$ sequences in a batch, so let's make that the size of the first dimension. For the second dimension, you can use `-1` as a placeholder in the size, it'll fill up the array with the appropriate data for you. After this, you should have an array that is $N \\times (M * K)$.\n",
        "\n",
        "**3. Now that we have this array, we can iterate through it to get our mini-batches. **\n",
        "\n",
        "The idea is each batch is a $N \\times M$ window on the $N \\times (M * K)$ array. For each subsequent batch, the window moves over by `seq_length`. We also want to create both the input and target arrays. Remember that the targets are just the inputs shifted over by one character. The way I like to do this window is use `range` to take steps of size `n_steps` from $0$ to `arr.shape[1]`, the total number of tokens in each sequence. That way, the integers you get from `range` always point to the start of a batch, and each window is `seq_length` wide.\n"
      ]
    },
    {
      "metadata": {
        "id": "oX5PtraJ7orJ",
        "colab_type": "code",
        "outputId": "d5342133-6259-49b7-deb2-fa9d6cde66a9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 340
        }
      },
      "cell_type": "code",
      "source": [
        "### To train on this data, I also want to create mini-batches for training \n",
        "### I want my batches to be multiple sequences of some desired number of sequence steps \n",
        "### Considering a simple example, the batches would look like this\n",
        "sequence_batching = mpimg.imread(\"/content/drive/My Drive/Images/sequence_batching@1x.png\") \n",
        "plt.imshow(sequence_batching)\n",
        "plt.grid(False)"
      ],
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeQAAAFDCAYAAADrt32vAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngiyEEELoASnIQgghhB74f+Gk810Ca7hcAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fc024570978>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "MmwRhL2aLqHP",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "### Define 'get_batches'\n",
        "def get_batches(arr, batch_size, seq_length):\n",
        "    '''Create a generator that returns batches of size\n",
        "       batch_size x seq_length from arr.\n",
        "       \n",
        "       Arguments\n",
        "       ---------\n",
        "       arr: Array you want to make batches from\n",
        "       batch_size: Batch size, the number of sequences per batch\n",
        "       seq_length: Number of encoded chars in a sequence\n",
        "    '''\n",
        "    \n",
        "    batch_size_total = batch_size * seq_length\n",
        "    ### Total number of batches we can make\n",
        "    n_batches = len(arr)//batch_size_total\n",
        "    \n",
        "    ### Keep only enough characters to make full batches\n",
        "    arr = arr[:n_batches * batch_size_total]\n",
        "    \n",
        "    ### Reshape into batch_size rows\n",
        "    arr = arr.reshape((batch_size, -1))\n",
        "    \n",
        "    ### Iterate through the array, one sequence at a time\n",
        "    for n in range(0, arr.shape[1], seq_length):\n",
        "        # The features\n",
        "        x = arr[:, n:n+seq_length]\n",
        "        # The targets, shifted by one\n",
        "        y = np.zeros_like(x)\n",
        "        try:\n",
        "            y[:, :-1], y[:, -1] = x[:, 1:], arr[:, n+seq_length]\n",
        "        except IndexError:\n",
        "            y[:, :-1], y[:, -1] = x[:, 1:], arr[:, 0]\n",
        "        yield x, y"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "T949CoilLqHT",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Test The Implementation.**\n",
        "\n",
        "**Now I'll make some data sets so I can check out what's going on as I batch data.**\n",
        "\n",
        "**Here, as an example, I'm going to use a batch size of 8 and 50 sequence steps.**"
      ]
    },
    {
      "metadata": {
        "id": "-gABFioiLqHV",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "### Test the batching function\n",
        "batches = get_batches(encoded, 8, 50)\n",
        "x, y = next(batches)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "DCQRn7eVLqHa",
        "colab_type": "code",
        "outputId": "1b8f7695-6ab0-444f-c59f-fcb9f82b475d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 340
        }
      },
      "cell_type": "code",
      "source": [
        "### Print out the first 10 items in a sequence\n",
        "print('x\\n', x[:10, :10])\n",
        "print('\\ny\\n', y[:10, :10])"
      ],
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "x\n",
            " [[44 79 35 61 59  3 36 26 76 58]\n",
            " [18 54 21 26 59 79 35 59 26 35]\n",
            " [ 3 21 13 26 54 36 26 35 26 81]\n",
            " [18 26 59 79  3 26 70 79 75  3]\n",
            " [26 18 35 82 26 79  3 36 26 59]\n",
            " [70 41 18 18 75 54 21 26 35 21]\n",
            " [26 56 21 21 35 26 79 35 13 26]\n",
            " [50 22 38 54 21 18 60 16 24 26]]\n",
            "\n",
            "y\n",
            " [[79 35 61 59  3 36 26 76 58 58]\n",
            " [54 21 26 59 79 35 59 26 35 59]\n",
            " [21 13 26 54 36 26 35 26 81 54]\n",
            " [26 59 79  3 26 70 79 75  3 81]\n",
            " [18 35 82 26 79  3 36 26 59  3]\n",
            " [41 18 18 75 54 21 26 35 21 13]\n",
            " [56 21 21 35 26 79 35 13 26 18]\n",
            " [22 38 54 21 18 60 16 24 26 65]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "FN3ZoldULqHh",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**If I implemented `get_batches` correctly, the above output should look something like.** \n",
        "```\n",
        "x\n",
        " [[25  8 60 11 45 27 28 73  1  2]\n",
        " [17  7 20 73 45  8 60 45 73 60]\n",
        " [27 20 80 73  7 28 73 60 73 65]\n",
        " [17 73 45  8 27 73 66  8 46 27]\n",
        " [73 17 60 12 73  8 27 28 73 45]\n",
        " [66 64 17 17 46  7 20 73 60 20]\n",
        " [73 76 20 20 60 73  8 60 80 73]\n",
        " [47 35 43  7 20 17 24 50 37 73]]\n",
        "\n",
        "y\n",
        " [[ 8 60 11 45 27 28 73  1  2  2]\n",
        " [ 7 20 73 45  8 60 45 73 60 45]\n",
        " [20 80 73  7 28 73 60 73 65  7]\n",
        " [73 45  8 27 73 66  8 46 27 65]\n",
        " [17 60 12 73  8 27 28 73 45 27]\n",
        " [64 17 17 46  7 20 73 60 20 80]\n",
        " [76 20 20 60 73  8 60 80 73 17]\n",
        " [35 43  7 20 17 24 50 37 73 36]]\n",
        " ```\n",
        " **although the exact numbers may be different. Check to make sure the data is shifted over one step for `y`.**"
      ]
    },
    {
      "metadata": {
        "id": "xLcyEsCpLqHj",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "---\n",
        "**Defining the network with PyTorch.**\n",
        "\n",
        "**Below is where I will define the network.**\n",
        "\n",
        "**I will use PyTorch to define the architecture of the network.**\n",
        "\n",
        "**I start by defining the layers and operations I want.**\n",
        "\n",
        "**Then, I will define a method for the forward pass.** "
      ]
    },
    {
      "metadata": {
        "id": "O2KqTK9tIcPE",
        "colab_type": "code",
        "outputId": "504b55b4-135e-45ae-c6cc-0688c4121efe",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 351
        }
      },
      "cell_type": "code",
      "source": [
        "### Layers and activations of an RNN\n",
        "char_rnn = mpimg.imread(\"/content/drive/My Drive/Images/charRNN.png\") \n",
        "plt.imshow(char_rnn)\n",
        "plt.grid(False)"
      ],
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZ4AAAFOCAYAAAC2QnsgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzt3Xl8E3X6B/DP5GppaYGWBixye4BQ\nCgVcQApUKKdK5ZZFPICF5RB/ooDIKrvsuiqKAuKKcsiCrpXiSl1cyiEoq6UK1QqIYlFBoLQp9L6T\nzO+PkoHQpleSOZrP+/XiRTKTzDx5+nznSSaTGUEURRFEREQy0SkdABER+RY2HiIikhUbDxERyYqN\nh4iIZMXGQ0REsmLjISIiWRk8vcDnn38eaWlpEAQBy5YtQ48ePTy9CiIi0jCPNp6vvvoKZ8+eRXx8\nPM6cOYNly5YhPj7ek6sgIiKN8+iutuTkZAwbNgwA0LlzZ+Tl5aGwsNCTqyAiIo3zaOPJzs5GixYt\npPshISGwWCyeXAUREWmcx7/juV5tZ+OxWAq8uXoiIlJIWFiQy3ke/cRjNpuRnZ0t3c/KykJYWJgn\nV0FERBrn0cZz1113ISkpCQBw8uRJmM1mNG3a1JOrICIijfPorraoqCh069YNU6ZMgSAIeO655zy5\neFLAmrc24r6RI9CxXVulQyGiRsLj3/E8+eSTnl4kKejjpH3o36c3Gw8ReYzPnLkgJy9f6RA0z2qz\nKR0C+TiO48bBq0e1qcHRb7/D0pV/k+7379sbK5cuVjAi7cnNzcOw8ZOl+/t38kfB9XV9/t5c9QJu\n6dRRwWi0Z+Ijs5CTf63pfLBxA0JaNFcwIm2JnTAFv58wHu/t/BB3D7wLSxfOVzSeRt14/nvgIF55\n402nDeWkmXMgiiIEQVAwMm15e9u7Ug7LyyswcvJU7Il/T+GotGPY+MlONThs/GSsXrkCPe7oqmBU\n2vHfAwcx/r57MOX+sQCA0rIy3DN1OvYlvM9xXA/bdySo5k1jo97VFv/RLtx2S2enaY9MmYSv09IU\nikib5s96VLptMhlhtXKXW12dv3gRAHAlN1f6F9AkAJvefV/hyLQj/qNdUtMBAH8/PwDgOK6nPj3V\nc97MRv2J5/zFDIy+egofh8DAAFzOvqJQRNrUtk0bpUPQrG+OnwQATJox22n6lZwcJcLRpPMXM6qd\nznFcP82Dg5UOQdKoG0/b8HCc+um007SvUr/FtInjFYpIm058fwqd2rdTOgxN6h1Z+S5TLbs4tKht\neHi103v1iJA5EvKURr2rbcu6V/HL2XN4ce16AMB/93+KPZ8eRGszz6ZQH2s3bsalLAtKSkswavI0\n7Nq2RemQNCO8dSuEtGiBPQcPAag8Kuv6Aw2odlvWvYph4yejtLQUQOUX5aEhLTiONaxRf+IBgMTt\n7+C+aQ9j32efAwA+2LRB4Yi0Z8vaVzHtj5VHwbQND0dgQIDCEWnLBxvfxLDxk/Hy6/8AAOxL4Pc7\n9RUz8C7c8/uHAAA9ut2B1X/hj9O1TBBrO5OnF/EkoUREjZNsJwklIiKqDRsPERHJio2HiIhkxcZD\nRESyYuMhIiJZsfEQEZGsfKbxlJeXKR2C5jGH7mH+3MP8uU8tOfSZxnPyJE8o6C7m0D3Mn3uYP/ep\nJYc+03jOnfsF6ek/KB2GZqWn/4Bz535ROgzNYv7cxzHsHjXVoM80HgA4deq40iFoliN3Nl6FtEGY\nP/c48sYx3HBqyp1PNR673a50CJrlyN2pU98pHIk2MX/uceSNY7jh1JQ7n2g8xcVFAACDwSjdpvox\nGIwAgDNnTtfySKoO8+ceR94ceaT6KS4uknKnhm2gTzSegweTAAAxMSOk21R3oigiJmaE0mFoFvPn\nOTExI6DgeY016+DBJKkG1bAN9InGM2bMOABAQECgdJvqThAEBAQEAgDGjuW1ZOqL+XOfI28BAYEQ\nBEHhaLRnzJhxUg2qYRvoE42HiIjUg42HiIhkxcZDRESyYuMhIiJZsfEQEZGs2HiIiEhWbDxERCQr\nNh4iIpIVGw8REcmKjYeIiGTFxkNERLJi4yEiIlnVqfGcPn0aw4YNw/bt2wEAGRkZePDBBzF16lQs\nXLgQ5eXlAIDExESMHz8eEydOxI4dO7wXNRERaVatjae4uBgrV65E//79pWlr167F1KlT8d5776F9\n+/ZISEhAcXEx1q9fj3feeQfbtm3D1q1bkZub69XgiYhIe2ptPCaTCW+//TbMZrM0LSUlBUOHDgUA\nxMTEIDk5GWlpaYiIiEBQUBD8/f0RFRWF1NRU70VORESaZKj1AQYDDAbnh5WUlMBkMgEAQkNDYbFY\nkJ2djZCQEOkxISEhsFgsHg6XiIi0zu2DC1xdDZBXCSQiourU+omnOgEBASgtLYW/vz8yMzNhNpth\nNpuRnZ0tPSYrKws9e/asdVkPWX5AUokM3wX1ao/Z5454fTUfteqGfn5BXl+Pw5GyAszOPo1MW4X3\nVyZDDkc0aY6tYV28uo4byZZDmWpQiRzKMo5lyl8rvREbWt4m+ziOyzzp/RXJWIOfhPV1Ob9Bn3gG\nDBiApKTK63bv3bsX0dHRiIyMxPHjx5Gfn4+ioiKkpqaiT58+NS7npt9S5Gk6MorLPImUsgJZ1pVy\ntVhlaToySSrJxU2/pci2PubQfY1tHGfaKhQZx41JUkkuDKn7Xc6v9RPPiRMn8OKLL+LChQswGAxI\nSkrCyy+/jKVLlyI+Ph7h4eGIi4uD0WjEokWLMGPGDAiCgHnz5iEoqOZ3DI11d9zsyz/h2/Aor6/n\nD9mnvb4OJchZF8yhttYlJ45j99RUF4KoYNXoj+1TatVed6ldP6+vo7UMH5mVIkf+AObQE5hD9zTm\n/Nl6x1Y7nWcuICIiWbHxEBGRrNh4iIhIVmw8REQkKzYeIiKSFRsPERHJio2HiIhkxcZDRESyUm3j\nOXx7X9h6x8LWOxb5Pe+u8/NWhnd2+aMlX9InIFjKn613LG42+tXpeQIAa9QwjG0e5t0ANeDGGqxP\nDm29Y30+hzfW4Oqbb6vzc61RwziO0fAadGwH1VqDqmw8V3oOQbjRD/pj+2A4tg9HinKRdkf/2p8I\nYNlNnRD1fTIAYFfnnviyy53eDFWV2pn8kdL1d9Af2wf9sX244+QXONtjEAJ1+lqf++xNnSAIAnbl\nVl7SwlcHfzuTv1SD+qs1WJ8cRn2f7PM5TOn6O4w7861Ugwtbta/zOBYEQTqzya7OtZ9suDGqbjtY\n1xpcdkMNqm07qMrG00xvROcT/wMAiACG/5SK7k2aVnmcQRCqfX5aSaE3w1O9dztG4ED+Zen+j6XF\nAICnW3es9bmhBhMyK8q8FptWvNsxQqpBoLIGgbrn0NdrEAAO5F+WNnw/lhZjyI9f13kcswar3w4C\ndatBQN3bQVU2HgC40GOQ0/3rz+t2pecQPNmqPW7zC8AXVz+KAsC0kJuk/wc3bYGbTf4IMxgxLeQm\n/LtzJDIjB+NM94FY1rojbL1jcaXnEGxo39VpGUDlO9Q32nVFe5M//hp+C/J6xjjNq+62mnxRmIuh\nwaHoaGoiTdMf24flF9MBAC0NRuT3vBu/C2yGmS3bSK+jX2Az3O4fAAC42ejnlE+g8vWWRw3D6GYt\nUdxrKPJ73o2KqGF43NwOtt6xWHZ1QKy6+TbYesfirqbNcU+zlrD1joXp6sZlY/s78KebOkmPm93y\nZhkyUn9fFObiQo9BLnN4peeQGnM4LeQmlzk8032gUw43tO9aYw7/Gn6LUw5tvWOlHNp6x6o2h0OD\nQ7G4VQfp/uHCXGkctzQYYesdK41jW+9YzAtrC6D6cTwt5CYE6HT4d+dIlEcNw0JzO2kcV0QNw5QW\nrV2O43uatURW5GBpHG9sf4f0OEee1aquNXj9a+8X2AwAnGrQsR0EIG0HHTXo2A46atDBkRvHdtCT\nNajKxmNOO4QQvVHat5nRY7A0r53JH830RryceRbflxbhrh+/luZtv5Ih/f9ZYQ7Ol5fCYq2Qprc0\nmND5xP/w/KVfAABfFeVj9tlTTssAgJMlhZh77hTOlpdi+cV0NNVfO4n30NNHkdzlTpzsNgDZ1nKv\n5cAdSy/8hE3ZF5AeMVDK4e+uFiMAnO8xCMHffoqUojxszL6AvqeOYEV4ZxwpysOPpcXIt1lxvqLM\nKZ8Ow04fwyd52bjzhxQE6vUwpu7Ha1nn8JblPB5pGQ4AiA0OwbDTR/FFYS7+k1d5jaa45pWXTp95\n9nusCO+MAJ0OT7Rqjw3Z5+VKS70svfATQvRGKYcZPQY75bCZ3uiUQwBOOdx+JcNlDjuf+J9TDmef\nPVVjDh0bGkcOh54+KuUw21qu2hxuyr6Av998q1SDb7e/Q5p3vscgrM/6TRrHfU8dwdp2ldcQ2n4l\nA/k2q9M43n4lA8V2O4DKGlyTdU4ax8bU/Xg/55LTOG6i00nj+D952Yg4mSyN45lnv4coikjuciee\naNUe4WmfyZWSenFsB6urQcd20FGDjtfuqEEATjVY3XbQUYOO7aCjBh0cNejYDgJVa9CxHaxvDaqy\n8Vy2VqDJNwegP7YP4858CxtEp081thtOqJ1anF+n5f5UWuR0/985WdU+rsf3yYhrHoY323XFz90H\nOs07VJCDOwOboYt/IFqptGAB4A9nv4f+2D60SjuEDZbz+PLqIAMAo+D8Z08tLsDYZnX7EvLzwhwA\nwBWr8/VrCuxWdPKr/LTU8/sjSCsuxNOtO2LPrZWnlb9+Z8oPpUUo6DUUm69usNXKUYOt0g7BBtEp\nh9XVYF1z6FDXHDpq0JHDQwU5Ug61UIP6Y/uwwXIej173ydAo6PD4bz9Ij00trvu1bxw1WJMSu10a\nx3tujcLFyMFO85t+8ynuDGyGzdkXkKnSN5DVbQcdNehqO1iXGrx+O3jFWuG0HSywW6Xbjhq8fjt4\nYw02dDuousYzKril08fzXbkW3Pzd5wCAHk2aorNfALJuKJQbB7ArNxaYFdVfEcLWOxarbr4N3xQX\noPcp7Z2yfNl1+4CzrRWYe+4U3sj6DatqOKqowoNXxzgXEQ1LzyFoY/LDS5d+BQCnTDuKTs1Xcbkx\nhzd/97lTDqurQW/l0FGD1eVQrUYFt3S6P/fcKWk3W4+r3/PYvbj+5nqDNI5fuvQr2l/dhtxIrTXo\najvoqEFX28G61GBdt4OOGrx+O+ipGlRd/aYW5+PvN9+Kzn5Nqsz7rqQQiy+cxk03HFI4LDjU5fKq\nP/ygZj+XFePWE19gQ/Z5FNptTvN+jYhG67RDmPnrSdXuG17auiPO3PBJLTY4BFkVlQWXb7M6zVt9\n822IO/Otx9bfxuQP47F9mH/uB3xacAXAtb9DsN6AlgYT9Mf2YUbLNh5bp6ctbd2xSg1en8PqatBb\nOXTUoCOHv0ZESzlUaw2mFudXqcGmV4/G+q6kEPk2K/5zSy9p3uqbb0OZ3XOt6MBtfaRx/GnBFWzq\n0E2aF6w3oChqKFqnHcKMlm0Q36mHx9brKa62g44adLUddFWDDdkOOmrw+u3gjTXY0O1grVcglVum\ntRzbLl/E6e4Dca68BDYR6OjXBK9mngVQ+fHzy8JcZPQYjMOFORjfohWSC6u/7O6Wyxews3NPZPQY\njCNFdb80bye/AKR3HwiDICD86h93R6ceePdKBtqa/GGxVmDL5YtY3fZ2tDH64YLKjsAJ/vZTab/6\nZwVXEN20BXTXHZ7aOu0z2HrH4pM8C3o2CUK4yR9PnHd9FcSKqGEw1nAZ2xsV222o6B2LnTmZGN+i\nFdJLi7G67e2Iz8lETs8YKY72332Oy5FDEJp2yK3X6w2OHALAZwVXMDgoBMC1g1y+LMx1ymFyYS4u\n1lAHGT0G46bv6r5L4vocxjU3O+WwrclfiiPfZlVlDjOt5ejg1wS23rE4V16Cm43+0AmCNI5bp32G\n4qihTuO4lYvXsOXyBVREDUPH44frvP4pP3+HH7rfhfTuA9HRrwmWnD+NYcGh2NGpB8a1aIV9+Zdh\nsVag/Xef42yPQaobx9dvBwHgl7ISdLzahKqrwdHNwmqswTsDmzW4Bn8rL0W40U+qwTLR7lSDq9ve\nXu8aVO0VSHUAFrfugFK7HR/nWXCmrMRp/rCgENzmH4BtlzNQcMOnkuu1NflDAHCuvLTOcQXq9JjZ\nsg3evZKBbGsFbvELgJ8g4OQN3xHVRA1XLhzf3Ix+TZsjuTAXH+Y6f5/V0mDEw6HhSC7KwxcuGrdD\n34BgfF3H79EcRgSHwl+nkw6nHd2sJT65eqBBXajh6pk6APdfzeGblt+q1GB9ctjO5F+vGgQaRw57\nNGmK0c1a4nRpcZUaBK6N4zcsNX853ZAadIzjNVnnAECT49hRg21N/tVuBx01uMFy3uPbQaCyBo8V\n5yP76tcZ9a1BV5+GVNt4tE7pgtU6NWw0tY45dB/HsXt46WsiIlIFNh4iIpIVGw8REcmKjYeIiGTF\nxkNERLJi4yEiIlmx8RARkazYeLxgRJPmjWo9jRlzSK5wHHuPoo1nafO2Sq7eKwRBwNawLrKsa2tY\nFwguLoanZXLWBXOorXXJhePYfSvDb3E5T9EzF1gsdT8Vuru++eYr9Oqlrsu/ag1z6B7mzz3Mn/vk\nzGFYWJDLeT6zq+3cuV+Qnv5D7Q+kaqWn/4Bz535ROgzNYv7cxzHsHjXVoM80HgA4deq40iFoliN3\nNpvrExGSa8yfexx54xhuODXlzqcaj92D1/vwNY7cnTr1ncKRaBPz5x5H3jiGG05NufOJxlNcXHka\ndIPBKN2m+jEYjACAM2dcX7eHXGP+3OPImyOPVD/FxUVS7tSwDfSJxnPwYBIAICZmhHSb6k4URcTE\njFA6DM1i/jwnJmYEFDweSrMOHkySalAN20CfaDxjxowDAAQEBEq3qe4EQUBAQCAAYOzYyQpHoz3M\nn/sceQsICGyUhx5725gx46QaVMM2sE6Xvn7ppZdw7NgxWK1WzJ49GxEREVi8eDFsNhvCwsKwatUq\nmEwmJCYmYuvWrdDpdJg0aRImTpzo7fiJiEhjam08R44cwU8//YT4+Hjk5OTg/vvvR//+/TF16lSM\nGjUKq1evRkJCAuLi4rB+/XokJCTAaDRiwoQJiI2NRfPmvverXCIicq3WXW19+/bFmjVrAADBwcEo\nKSlBSkoKhg4dCgCIiYlBcnIy0tLSEBERgaCgIPj7+yMqKgqpqanejZ6IiDSn1saj1+sREBAAAEhI\nSMCgQYNQUlICk8kEAAgNDYXFYkF2djZCQkKk54WEhMBisXgpbCIi0qo6H1ywf/9+JCQk4Nlnn3Wa\n7uoIEx55QkRE1alT4zl8+DDefPNNvP322wgKCkJAQABKS0sBAJmZmTCbzTCbzcjOzpaek5WVBbPZ\n7J2oiYhIs2ptPAUFBXjppZewYcMG6UCBAQMGICmp8ljwvXv3Ijo6GpGRkTh+/Djy8/NRVFSE1NRU\n9OnTx7vRExGR5tR6VNsnn3yCnJwcPP7449K0F154AcuXL0d8fDzCw8MRFxcHo9GIRYsWYcaMGRAE\nAfPmzUNQkOuzkxIRkW+qtfFMnjwZkydX/dHbli1bqkwbOXIkRo4c6ZnIiIioUfKJMxcQEZF6sPEQ\nEZGs2HiIiEhWbDxERCQrNh4iIpIVGw8REcmKjYeIiGTFxkNERLJi4yEiIlmx8RARkazYeIiISFZs\nPEREJCs2HiIikpUgKnip0IJ/vQ2/T/6p1Oq9QxCQ98pu2VbXbNEYoJFd7bVs9HSUDpsiy7qYP/f4\n73+fY9hNjbUGgx6Y5XK+op94Gl3BAoAoosmmP8uyqiab/tzoChaQry6YP22tSzYcw26rrS64q80L\nTCdTGtV6Givmj1zhGPYuNh4iIpIVGw8REcmKjYeIiGTFxkNERLJi4yEiIlmx8RARkazYeIiISFaN\nr/HoDUpHQETu4Bhu9DTzFzZtO4zyB6NrnH89e9oRWF9+qtp5Nyp/MFp6THXrqGmeVpi2HUbFn2ZC\n/PXHaucbFqyE7s4hTtPK/28ikH0J0Olh2nrI5bLLH4yGYcUG6DrfAfuJo7C++H9O8/VzlkN/1wiI\nF35FxdIH3XwlymD+3GN4chV0kf1cj6EWYTCt/dBpUl3HsPXddbDv+YBjuIYarMs20FGD1eXI0zWo\nmcZTE/2k2bB/dQjWdX+Sppm2HYbh8edhfW0ZymfGStMNS16F7V/rIZ5Lr3ZZuqFxsB/4yOsxq43u\nziEon3svUJALADC+ugOmV3dUFqHd5pRD08Z9TvcdRFGErnufKtP1d42AgqcElAXz5x7T2g+dxrBw\neySMy1+vdgybNu5DxV/+eG0MWytqXLbQW7vNpq70k2bXWIM3bgN1t3Z3XYPVbAM9XYONYleb/t5p\nTk0HAGxJO6BzFFxZ6bV/djtQXuY8zfGcXVuhn+78bhNGE2xJO7z9EpQVdhNs/31fKlgAqPi/iQAA\n4dbulRNuzFc1+UNxoet15Fg8HbV6MH8ecf0YFn9Mcz2GAecxbLNJz7Pt2go0C3VarmHO8kY/hvX3\nTqu5Bm/cBgIua7C6bSAAj9Zgo2g8Yu5lGJ5/51qCANi2r633x2pbwkYIOueUGP+xG7btaz0Rplt2\n7/4QxcVF3lm4JQP6UVOg6zfUaXL5g9EQfzpR9+U0CYB15yYgxHxtmsmvckAY/TwUbMMcOXIYZ8/+\n7J2F+0D+AODrr7/0Xg0CHhvDxnX/dpom+AeoZgx7qwbF3Mseq0FBp3OqQeM/dnu8BhtF46lYEAdd\n284wbT4A07bDMG49BKH9rQ1alv2s8y44wc/fEyG6LSZmBA4eTMKuXfH4+OMdSE//waPLF/NzYJi3\nAqZth2Hadhj68TPrvxBBB/tH78Cw8K/SJMPs5bC9tx7QKVtq/fpFIysrA7t2xWPXrnh8+eUhjy6/\nsecPAMLCWjnVYH5+nseWbX39OacxbFi5scFjWBAEp/uiJcMTIbotJmaEUw16cgxXLIhzqkHj1kMN\nr8Gz6U41KPj5e7wGG8V3PMDVLw0FAboh90I/7hEY/7oZ9tPHYV05t17Lsa55Brp7fg/7f96Frt+w\nBseza1d8g59bG7vdjpMn09Dbg8usmHcfAEC4uRN0ox+APu4h6OMeQvkjd9e6D/1Guk5dr92+4cvO\nuvJm/gDAYsn06PJ8LX92ux0HD+7BdE8tL+VTlKd8Ko1hw6NPQdfAMQzAaQxb1/6p9idUw9s59OYY\nNix5pcE1aF3zDEyrvfvaG0XjMb61BxV/GAmIIuwHE2E/mAjg6pEsBmP9Br7lIgyT56D8P+/CMO85\nVDzzSINiGjt2coOe58ru3R8iJmYEAgICr038bKtHlm18fRcq5o8FAIjnf4btrb/B9tbfYHw9Eca3\n96LikZh6LU8URejGPgS48U7T0/k7cuQwWrcOR4cOna9NZP7q5euvv0C3bj29UoOG596E9c9zpDFc\n7sYYLl84HqY1O6Ux3NAj2bwxhrt1i/RODTq2gaiswYoF91dOb0gNWi461WBDt4E1aRSNR2gSWHVi\ncIvK/+v5blNa5tUvhV0d/Sa3MWPGeW3Z9u++gi7mPqlhS4KaQfz5VL2XZ9u2Bobpj0O0VsD20Tue\nCdJN/fp578gmX8gfAPTte5fXlq27pVvViQ0dw1eyAFx3YIdKeHMMC00CvVaDFY/c7aEor9FU49FP\ne6zKNNv2tbDt/xDGLQdh/csciL/8CISaYXptJ8SrBVhf9lPfwPjsPyBe+MXdkFVFN3oykJ/rNM2W\nsBG2t/4G07bDsDYNhj1pB1BeBsPzWyHodKhYv6Le67Hv2wlMfxyCwQjbzk0eil55zJ/7XI1hAE5j\nWOh1F4xPvMAxfAOXNbj/QxgefepaDQY1g2HRKo/UoDdoq/GMmFhlmm37Wti2vgpdj34w/mXjtekf\nbobt31satB7rumdheuNjVKxZ3uBY1Ujfv+px+7aP3wVKi1Gx9k8wPrYSmDQbACDmZKN8QRyQe7lB\n6xLtNgg6vVvxqg3z5z5XY7j8wWgYX4nnGK6Fqxq0bX0V9u+/0UwNCqKCv0zT8q+Ia5O3+hOvr6PZ\nE6O9vg6lMH/ukSN/AHPorsacv5rOllDrJ56SkhIsXboUly9fRllZGebOnYsuXbpg8eLFsNlsCAsL\nw6pVq2AymZCYmIitW7dCp9Nh0qRJmDix6rsbIiLybbU2noMHD6J79+6YNWsWLly4gEcffRRRUVGY\nOnUqRo0ahdWrVyMhIQFxcXFYv349EhISYDQaMWHCBMTGxqJ58+ZyvA4iItKIWn8RNHr0aMyaNQsA\nkJGRgVatWiElJQVDh1b+QjYmJgbJyclIS0tDREQEgoKC4O/vj6ioKKSmpno3eiIi0pw6H1wwZcoU\nXLp0CW+++SYeeeQRmEyVp7YIDQ2FxWJBdnY2QkJCpMeHhITAYmn855ciIqL6qXPjef/993Hq1Ck8\n9dRTTmcpdXVsQmM/my4RETVMrbvaTpw4gYyMyl9Qd+3aFTabDYGBgSgtrTyjaWZmJsxmM8xmM7Kz\ns6XnZWVlwWw2V7tMIiLyXbU2nqNHj2Lz5s0AgOzsbBQXF2PAgAFISkoCAOzduxfR0dGIjIzE8ePH\nkZ+fj6KiIqSmpqJPn6rXFiEiIt9W6662KVOm4JlnnsHUqVNRWlqKZ599Ft27d8eSJUsQHx+P8PBw\nxMXFwWg0YtGiRZgxYwYEQcC8efMQFBQkx2tQHXtwaO0P8tB6dPkN+3EYMX/kGsewdyn7A9Lpg4BG\n+F1Q0YJVsHas5txTHmb4+QQCX1/s9fXIThCQ98pur6+G+XNfs0VjOIbd0Jhr0PTPz13OVvQiH3mv\n7EZ5t98pGYLHFc1/SZaCBQBrp+4omv+SbO/O5FDe7XeybTSZP/c1tjFsDw7lGHZTXWpQ0U88FkuB\nbOvatSseMTEjERzcTLZ1Nib5+Xk4eHCPV0637wuYP/dxDLtH7hoMC3P9VYvylzWU0YkT3ygdgmYx\nd+5h/jyDeWw4NeXOpxqPp6866UscucvMVMdlhLWG+XOPI28cww2nptz5VOMh9508+a3SIWga89cw\nzFvj4hON5+uvvwAAREb2kW5T/URGVv4mq6AgX+FItIn5c48jb448Uv18/fUXUu7UsA30icZjs9kB\nAB06dJZuU92VlZVJ14lv1qyo4mboAAAekElEQVSFwtFoD/PnvhYtKs8D2aFDZ5SVlSkcjfbYbHap\nBtWwDfSJxtOvX3S1t6lu/Pz8pNtDhgxXMBJtYv7cN2jQtStvXp9Pqhu1bQN9ovEQEZF6sPEQEZGs\n2HiIiEhWbDxERCQrNh4iIpIVGw8REcmKjYeIiGTFxkNERLJi4yEiIlmx8RARkazYeIiISFZsPERE\nJCvFL33dZNOfYTqZolQIHlc0/yVYO3WXbX2Gn0+gyT9fhC7/smzr9Kbybr9DyYznZFsf8+e+xjSG\n7cGhKJm+hGPYDY4arOnS14o2nvLpgwDlVu81RQtWwdqxm9fXY/jlJALXPeX19chOEJD3ym6vr4b5\nc1+zRWM4ht3QmGvQ9M/PXc5WdldbIyxYAGiy9YVGtR7ZyVQXzJ/G1iUjjmE31VIX/I7HC+T6yNxY\nPporhfkjVziGvYuNh4iIZMXGQ0REsmLjISIiWbHxEBGRrNh4iIhIVmw8REQkKzYeIiKSlUHpAOpK\nP2EmbAkbXc4XuvSE/oF50HXqAnv6SVj/PMfpuTWxJWyUHlPdOmqapxX6CTNh2/9vINf17wb0c5+F\nLrI/UFEB64v/B/G3M5UzBAH68TNcPs+WsBG6u8dCCAmD/cRRiD986zRfaH8bdH0HQczPhX1vgkde\nj9yYP/foBo6A0Lptncew7dDHsG16SZpX0xi2px2B+NMJjmFcq0Hx4lnYNq+SarAu20BHDVaXI0/X\noHYaz9iHXBeNXg/jM+tg//oz2NKSIfQaANO2wyifNRwoLYF+7EM1LtuWsFF6jG1vApCfW2Xdjsdp\nlX7sQ7AfPQzRRdGath2GWFIM+6FEoHkojM+/A/v5X2B9ejog6GrMoS1hI3SDRkPX+Q7o7p2GioeG\nOM03PLMWQpNAiBd+1eyGk/lzj67fMOgi+7kcQ4a/bISu4+3SGNaNfQj6IffWaQyLhfmVjcfVODWa\nGv8Y1utheueQVINC115ONViXbaCjBqvbBnq6BjXTeGpifOM/KH90KFBRXjnhw80wvrUHxg17UPHQ\nYJQ/GC091rB8PWzbXoN49qcqyxEvnYdh/l9gff4x5+k52RBatPTqa1CaeOk8Kp56QLpv27YWpjc+\nBvz8gbJSpxyath12ui8toyAPQlCzKtOFJoEQr2R5J3CVYP7co+t4u9MYttUwhk3bDqNi+aMux/CN\n9LOWNvoxbHzjP1VqEEHNpRq8cRuou72Hyxqsbhvo6RpsFN/xCAFNrzWdq6yvPwfx19P1Wo51zTLo\nuvZyXnbfwbCue9btGNVMiLgT9mM3nNCvIBf2n3+A0L1v3RdkNNawkkZRatVi/jzEU2N4+ASnafr+\nsY1/DAc09VgNVrcNrLzhuRpsFNVc8fJimLYdhnHzAejiHgb8mkD8LgXW52bVazni+V+qTDM+9leI\nPx33UKQNt2tXPHbtiseJE9/AZrN5dNni8a+gHzMVpm2HYVi2FsIdUQAA63OzIB47XPcF+TVB+VNT\noRs1WZqkuzMGFStmAyY/j8ZcX59/vg+7dsXjwIFPkJmZ4dFl+0L+AC/XoM3qPIaBBo9hw4MLq05X\n0Rj2Rg1WvLzYqQbh1wRAw2oQgFMNGh/7q8drsE672kpLS3HPPfdg7ty56N+/PxYvXgybzYawsDCs\nWrUKJpMJiYmJ2Lp1K3Q6HSZNmoSJEyd6LMjaiGnJsG74G/S/XwDD+BnA1S9yy+fdW2VfZW2sH2wA\nwm4CLBnSH6Ehdu2Kb/Bza3LmzGmcOVP5LnC6B5db/uhQ6P+wDPp+Q6V3PPbUL2B9dWn9FnTpNxim\nzkf5fytfv372M6iYMaze8Xgrf4WFBThypPKdIfPXMN6owYqHY6AbONIjYxiA0xi27X6vQTFpqQbF\ntGSnGjRt3AugYTVo/WCDUw0CgHjmew9FWqlOjecf//gHmjWr3Pe8du1aTJ06FaNGjcLq1auRkJCA\nuLg4rF+/HgkJCTAajZgwYQJiY2PRvHlzjwZbE/v/9sD+vz3SfeM7B2Fa/3G1+zFrXM7H22H462ZY\nlz8Kw5zlEDMvNCiesWMn1/6geiguLsLBg0mwWiug0+nQtWsE8JkHV1BRDtv6FbCtXwEAlQUcPQr6\nyXNgi3+zwYsVGvguydP5A4Cvv/4CFy9WfgcQFtbKswv3gfz9+usZnDyZJtXg4MHDgc+2emz5149h\nof1tMPx5Q8PG8M+nYFj4N2kMW9c806B4vDGGT578VqrBbt0iVTmG7R9vBybNlu43dBtYk1p3tZ05\ncwbp6ekYMmQIACAlJQVDhw4FAMTExCA5ORlpaWmIiIhAUFAQ/P39ERUVhdTUVI8H64rxrT1VplU8\nHFN5w1DDfnMXdO1vrfy/zyBY1y53KzZPOXgwCTExIzB27GTce+9E3HJLF48t2/j6rirTbG89DzEv\nB7qR9R98oihCN/Yh6AYM90R4HnHkyGGEhbXG2LGTMXbsZAwYMMRjy/aF/AGAxXLJqQaDg6seCNFQ\nhuecN4zi2dMNHsPWNcudxrBaHDyY5FSDHh3D1WwDPVWD3tgG1tp4XnzxRSxdeu2jWklJCUwmEwAg\nNDQUFosF2dnZCAkJkR4TEhICi8Xi8WBdEZoEQrg1wmma/qH/q7xhraj38kS7HcYXt1fePpfudnye\nMGbMOAQEBHpl2da3nq+68dTrITRrAfuBf9d7eRUL7odhwkwY/vgnVDztyR1aDdevXzQ6dOjslWX7\nQv4AoG/fu7xWg7pbunluDF/JchrDajFmzDiv1aDQJNBrNeiNbWCNu9o++ugj9OzZE23btq12vqur\nZnvratqmbVW/JCt/MBrljw6FafOBynVnX4LQsnXlvKemNmg9tq2rYXjkSVg/3NzwYFXIuLLqbxjK\n58dB/C4FQrOQyt+i2O1AeRkE/yYQ83Jg2762/ivKu/Y7g+oO2NAq5s99rsaw/dQ3MD77BgCO4Zq4\nqkHHNtBRg4Ku8jOFJ2rQGwSxhi7x+OOP47fffoNer8elS5dgMpkgiiJ2794Nf39/fPXVV9i+fTt+\n//vfIz4+HqtXrwYAPP300xg+fDhiYmJqXHl9993WSG+Aru8QCF17Qvz5FOyfyXPNeVfyVn/i9XU0\ne2K0R5cndLwdQo9+EPR62D7aCtg9e+RSfTB/7pEjf4CHc3jdGLYfOQDx1DeeW3YDaLkGkZsN++E9\nitZgdW8yHGr8xPPaa69Jt9etW4c2bdrgm2++QVJSEsaOHYu9e/ciOjoakZGRWL58OfLz86HX65Ga\nmoply5Z57hXUhc0K+5H9wJH98q63ERF/+RHiLz8qHYZmMX9u4hh2m1ZqsN5nLliwYAGWLFmC+Ph4\nhIeHIy4uDkajEYsWLcKMGTMgCALmzZuHoKAgb8RLREQaV+fGs2DBAun2li1bqswfOXIkRo4c6Zmo\niIio0WoUZy4gIiLtYOMhIiJZsfEQEZGs2HiIiEhWbDxERCQrNh4vKO/2u0a1nsaK+SNXOIa9S9HG\nUzZaPeeh8hhBQMmM52RZVcmM5wBBkGVdcpKrLpg/ba1LNhzDbqutLmo8ZY63WSwFsq3rwoVzaNOm\nnWzra4yYQ/cwf+5h/twnZw7DwlyfRMBndrUdPZqsdAiaxxy6h/lzD/PnPrXk0GcaDwB8+eUhhSPQ\nri+/PKRwBNr25ZeHFI6gcfjyy0MKR6BdX355SOEIrvGpxmOxZCodgmY5cufpa8X7CubPPY68cQw3\nnJpy51ONh9x38uS3SoegacxfwzBvjYtPNJ6vv/4CABAZ2Ue6TfUTGdkHAFBQkK9wJNrE/LnHkTdH\nHql+vv76Cyl3atgG+kTjsdnsAIAOHTpLt6nuysrKpEv2NmvWQuFotIf5c1+LFiEAKsdwWVmZwtFo\nj81ml2pQDdtAn2g8/fpFV3ub6sbPz0+6PWTIcAUj0Sbmz32DBsVKt6/PJ9WN2raBPtF4iIhIPdh4\niIhIVmw8REQkKzYeIiKSFRsPERHJio2HiIhkxcZDRESyYuMhIiJZsfEQEZGs2HiIiEhWbDxERCQr\nNh4iIpIVGw8REcmKjYeIiGTFxkNERLJi4yEiIlmx8RARkazYeIiISFZsPEREJCs2HiIikpWhtgek\npKRg4cKFuPXWWwEAt912G2bOnInFixfDZrMhLCwMq1atgslkQmJiIrZu3QqdTodJkyZh4sSJXn8B\nRESkLbU2HgC48847sXbtWun+008/jalTp2LUqFFYvXo1EhISEBcXh/Xr1yMhIQFGoxETJkxAbGws\nmjdv7rXgiYhIexq0qy0lJQVDhw4FAMTExCA5ORlpaWmIiIhAUFAQ/P39ERUVhdTUVI8GS0RE2len\nTzzp6emYM2cO8vLyMH/+fJSUlMBkMgEAQkNDYbFYkJ2djZCQEOk5ISEhsFgs3omaiIg0q9bG06FD\nB8yfPx+jRo3Cb7/9hunTp8Nms0nzRVGs9nmuphMRkW+rdVdbq1atMHr0aAiCgHbt2qFly5bIy8tD\naWkpACAzMxNmsxlmsxnZ2dnS87KysmA2m70XORERaVKtjScxMRGbNm0CAFgsFly+fBnjxo1DUlIS\nAGDv3r2Ijo5GZGQkjh8/jvz8fBQVFSE1NRV9+vTxbvRERKQ5te5qu/vuu/Hkk0/iwIEDqKiowIoV\nK9C1a1csWbIE8fHxCA8PR1xcHIxGIxYtWoQZM2ZAEATMmzcPQUFBcrwGIiLSkFobT9OmTfHmm29W\nmb5ly5Yq00aOHImRI0d6JjIiImqUeOYCIiKSFRsPERHJio2HiIhkJYgK/uDGYinw2rKF/PcgXH4B\nEMurzmwyAPaw5wFDa6+tX/Osl6CzLANKvqw6TzBBDF0KMXiq/HFpBfPnNo5hNylcg2Fhrg8ua3yN\nx54P4exgCGJJrQ8VhSYQ238G6II9H4eGCVdegZD7dp0eKzafBTFkkZcj0hbmz032fOh+vbNODxWF\nJhA7fuPlgLRH+KVXnbaBAGDv8JVXtoG+03hy3oIuZ3W9n2bv9INn49Aw3S9dgXqWhL3FE0CLP3gp\nIm1h/tzUkDEsCLB3POWdeDRILTXoE41H93MXt54vNr0HovllD0WjPULWkxAK/+PWMny5gTN/7uMY\ndo/aarCmxtMoDi4Qij91fxmF//HIcrTK3YIFrr7T8lHMn3s88dp9eQwLxZ9qqgYbR+O5NFdVy9Ea\n4cornlmQKHpuWRrC/HmAh3a8+OwY9tTrlmkHmOYbj+6XO1S9PNWzXqrzF+F14cllaQLz5zaPjznr\nJc8uT+08/Hrl2AZqvvFAtKt7eSqnsyxTOgRNY/48wMNjztf+Jh5/vTJsAzXdeITC3UqHoH3VHePv\nJp/6uzB/bvHKa/XC30TVNFiD2m48BTuVDoGqwb+Le3wpf770WrXE238XTTcelB5VOgKqDv8u7vGl\n/PnSa9USL/9dtN14dIFKR0DV4d/FPb6UP196rVri5b+LthuPyXd/96BqJh87MtDTfCl/HMPq5OUa\n1HTjEYPGKx0CVUMMGqd0CJrmS/njGFYnb9egthtP0zFKh6B5YstnPb9MH/q7MH/u8cZr9cbfRM20\nWIOabjwAIDb1bGf29PLUjqfmdw/z5z6Pj2Ef+5t4+vXKsQ3UfuMxP6/q5WmB2HyWKpelFcyfe3xx\nzKmZHH8PzTceABBbv6Gq5WiNx64HIwg+eW0Z5s8DBMEji/HZMeyp1+2hv0NtGkfjCbgb8I9yaxn2\njmmVy/FR9o5p7i3AP8qnr4nC/LnH3vEUx7AbxIC7NVWDjaLxAIA9/L3Kixk15LmdfgAEPw9HpDGC\n39U81P8dj73FE7CHv+eFoDSE+XNbg8ewIHAMA5qqwUZzITgJL33tNl662T3Mn5t46Wu38dLXNfBK\n47mevQC6nNeA4sM4nx2Km1teBgKiYQ/9k3fX24gIhbsrz9tUehQX8nqjTSsB9lZrAJ3roqJrrs+f\nHQHQ+d/B/NWHYwwXfoLzVzrg5paXIbZY6FOHnLtLKNwNIWcNYL1UWYPBY2Bv8bjXa9B3G891du2K\nx9ixk2VbX2PEHLqH+XMP8+c+OXPY6C99XVfp6b59TXt3MHfuYf48g3lsODXlzqcaz6lTx5UOQbMc\nubPZbApHok3Mn3sceeMYbjg15c6nGo/d7ltXF/UkR+4++2yvwpFoE/PnHkfeOIYbTk2584nG8+uv\nZwAA4eE3S7epfsLDbwYAFBTkKxyJNjF/7nHkzZFHqp9ffz0j5U4N20CfaDwnT1b+sKpbt57Sbao7\nURTRrVtPpcPQLObPc7p16wkFj4fSrJMn06QaVMM20Ccaz5gxlSe9CwgIlG5T3QmCgICAygtD8aii\n+mP+3OfIW0BAIASZTuvSmIwZM06qQTVsA32i8RARkXqw8RARkazYeIiISFaGujwoMTERGzduhMFg\nwGOPPYbbb78dixcvhs1mQ1hYGFatWgWTyYTExERs3boVOp0OkyZNwsSJE70dPxERaUytjScnJwfr\n16/Hzp07UVxcjHXr1iEpKQlTp07FqFGjsHr1aiQkJCAuLg7r169HQkICjEYjJkyYgNjYWDRv3lyO\n10FERBpR66625ORk9O/fH02bNoXZbMbKlSuRkpKCoUOHAgBiYmKQnJyMtLQ0REREICgoCP7+/oiK\nikJqaqrXXwAREWlLrZ94zp8/j9LSUsyZMwf5+flYsGABSkpKYDKZAAChoaGwWCzIzs5GSEiI9LyQ\nkBBYLBbvRU5ERJpUp+94cnNz8frrr+PixYuYPn260w+4XP2Yiz/yIiKi6tS6qy00NBS9evWCwWBA\nu3btEBgYiMDAQJSWlgIAMjMzYTabYTabkZ2dLT0vKysLZrPZe5ETEZEm1dp4Bg4ciCNHjsButyMn\nJwfFxcUYMGAAkpKSAAB79+5FdHQ0IiMjcfz4ceTn56OoqAipqano06eP118AERFpS6272lq1aoUR\nI0Zg0qRJAIDly5cjIiICS5YsQXx8PMLDwxEXFwej0YhFixZhxowZEAQB8+bNQ1AQr7JIRETO6vQd\nz5QpUzBlyhSnaVu2bKnyuJEjR2LkyJGeiYyIiBolnrmAiIhkxcZDRESyYuMhIiJZsfEQEZGs2HiI\niEhWbDxERCQrQVTw3DYWS4FSqyYiIi8KC3P9O05+4iEiIlmx8RARkazYeIiISFZsPEREJCs2HiIi\nkhUbDxERyYqNh4iIZMXGQ0REsvKZxlNYVKh0COSDho2fjK9Sv1E6DCJV8ZnGEzd9htIhEBERfKjx\nkHusNpvSIWgec0hUyScazwe7djn9T3X31bffVu4uOvYNPti1C8PGT1Y6JM15btUreDfhQ3z0yR4M\nGz8ZOXn5SodEPkhNb3x85iShw8ZPxv6d8bKtr7EYNn4yZk6biin3jwUAlJaV4YE/zMW/t25SODJt\nGDZ+MloEB2PHlrcBAOXlFbjvwYewJ/49hSPTjtz8Akx4ZKZ0f++Of0Gn84n3zB5z/RvG/n17Y+XS\nxV5fZ00nCTV4fe2keY6mAwD+fn4oKOSBGvUxf9aj0m2TyQirVT3vPLVgwiMzpTeN5eUVGD7xAb6J\nrIdZjy/CqytXIOKOrgAqm9DHSftw74hYxWLi2wZy6UpOrtIhNApt27RROgTN+mf8Dqf7JpMRAJB9\n5YoS4WjSL7+dl5oOANw3cgTWvLVRwYjYeKgGIS2aAwB+OXtOmma1WhEWGqpUSJr01j+3Od0PDna9\nC4KcfbxvP4DKd+mOfwDwzfETSoalGdW9eezZ/Q4FInHmU7va8gsKERzUVOkwNKVvr0jMeuIp7N8Z\nj4oKK0Y/MA2739tW+xNJcvTb73Apy4JmwU0x7qFZSNjyltIhaUbviAjs//wwd601kOPN4/W+Sv1W\ngUic+dQnnnEP87c89fX35csQM/AuDBs/GaOm/B4Rd3SVdndQ3WxZ+yqm/XE+7v39w2htDkNgQIDS\nIWnGzAenVpnGIyvds+fTg5g/4xFFY/CZo9qISLsczUan02Hvjn8pHI32XN+sP9i0ASHNq34S8rSa\njmpj4yEiIo+rqfH41K42IiJSHhsPERHJStFdbURE5Hv4iYeIiGTFxkNERLJi4yEiIlmx8RARkazY\neIiISFZsPEREJCvFThL6/PPPIy0tDYIgYNmyZejRo4dSoVRx+vRpzJ07Fw8//DCmTZuGjIwMLF68\nGDabDWFhYVi1ahVMJhMSExOxdetW6HQ6TJo0CRMnTlQk3pdeegnHjh2D1WrF7NmzERERodp4S0pK\nsHTpUly+fBllZWWYO3cuunTpotp4AaC0tBT33HMP5s6di/79+6s61pSUFCxcuBC33norAOC2227D\nzJkzVR1zYmIiNm7cCIPBgMceewy33367auPdsWMHEhMTpfsnTpzAv/71L6xYsQIAcPvtt+PPf/4z\nAGDjxo3Ys2cPBEHA/PnzMXjwYNnjLSoqwpIlS5CXl4eKigrMmzcPYWFhyscrKiAlJUX8wx/+IIqi\nKKanp4uTJk1SIoxqFRUVidOmTROXL18ubtu2TRRFUVy6dKn4ySefiKIoiq+88or47rvvikVFReLw\n4cPF/Px8saSkRBwzZoyYk5Mje7zJycnizJkzRVEUxStXroiDBw9Wdby7d+8W33rrLVEURfH8+fPi\n8OHDVR2vKIri6tWrxXHjxok7d+5UfaxHjhwRFyxY4DRNzTFfuXJFHD58uFhQUCBmZmaKy5cvV3W8\n10tJSRFXrFghTps2TUxLSxNFURSfeOIJ8dChQ+K5c+fE+++/XywrKxMvX74sjhgxQrRarbLHuG3b\nNvHll18WRVEUL126JI4YMUIV8Sqyqy05ORnDhg0DAHTu3Bl5eXkoVMlVLU0mE95++22YzWZpWkpK\nCoYOHQoAiImJQXJyMtLS0hAREYGgoCD4+/sjKioKqampssfbt29frFmzBgAQHByMkpISVcc7evRo\nzJo1CwCQkZGBVq1aqTreM2fOID09HUOGDAGg7lpwRc0xJycno3///mjatCnMZjNWrlyp6nivt379\nesyaNQsXLlyQ9tg44k1JSUF0dDRMJhNCQkLQpk0bpKenyx5jixYtkJtbeU2e/Px8NG/eXBXxKtJ4\nsrOz0aJFC+l+SEgILBaLEqFUYTAY4O/v7zStpKQEJpMJABAaGgqLxYLs7GyEhIRIj1HqNej1egRc\nPc1+QkICBg0apOp4HaZMmYInn3wSy5YtU3W8L774IpYuXSrdV3OsDunp6ZgzZw4eeOABfPHFF6qO\n+fz58ygtLcWcOXMwdepUJCcnqzpeh++++w433XQT9Ho9goODpelqi3fMmDG4ePEiYmNjMW3aNCxe\nvFgV8ariQnCihs7a4ypWpV/D/v37kZCQgM2bN2P48OHSdLXG+/777+PUqVN46qmnnGJRU7wfffQR\nevbsibZt21Y7X02xOnTo0AHz58/HqFGj8Ntvv2H69Omw2Wy1xqZkzLm5uXj99ddx8eJFTJ8+XbX1\ncL2EhATcf//9VaarLd5du3YhPDwcmzZtwg8//IB58+YhKOjaWaOVileRTzxmsxnZ2dnS/aysLISF\nhSkRSp0EBASgtLQUAJCZmQmz2Vzta7h+95ycDh8+jDfffBNvv/02goKCVB3viRMnkJGRAQDo2rUr\nbDYbAgMDVRnvoUOHcODAAUyaNAk7duzAG2+8oercAkCrVq0wevRoCIKAdu3aoWXLlsjLy1NtzKGh\noejVqxcMBgPatWuHwMBA1dbD9VJSUtCrVy+EhIRIu7IA1/E6psstNTUVAwcOBAB06dIFZWVlyMnJ\nUTxeRRrPXXfdhaSkJADAyZMnYTab0bSpei9JPWDAACnevXv3Ijo6GpGRkTh+/Djy8/NRVFSE1NRU\n9OnTR/bYCgoK8NJLL2HDhg1ofvXiTmqO9+jRo9i8eTOAyl2uxcXFqo33tddew86dO/HBBx9g4sSJ\nmDt3rmpjdUhMTMSmTZsAABaLBZcvX8a4ceNUG/PAgQNx5MgR2O125OTkqLoeHDIzMxEYGAiTyQSj\n0YhOnTrh6NGjTvH269cPhw4dQnl5OTIzM5GVlYVbbrlF9ljbt2+PtLQ0AMCFCxcQGBiIzp07Kx6v\nYmenfvnll3H06FEIgoDnnnsOXbp0USKMKk6cOIEXX3wRFy5cgMFgQKtWrfDyyy9j6dKlKCsrQ3h4\nOP7+97/DaDRiz5492LRpEwRBwLRp03DffffJHm98fDzWrVuHjh07StNeeOEFLF++XJXxlpaW4pln\nnkFGRgZKS0sxf/58dO/eHUuWLFFlvA7r1q1DmzZtMHDgQFXHWlhYiCeffBL5+fmoqKjA/Pnz0bVr\nV1XH/P777yMhIQEA8Mc//hERERGqjvfEiRN47bXXsHHjRgCV36k9++yzsNvtiIyMxNNPPw0A2LZt\nGz7++GMIgoDHH38c/fv3lz3WoqIiLFu2DJcvX4bVasXChQsRFhameLy8LAIREcmKZy4gIiJZsfEQ\nEZGs2HiIiEhWbDxERCQrNh4iIpIVGw8REcmKjYeIiGTFxkNERLL6f6yO8F3k4KEVAAAAAElFTkSu\nQmCC\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fc0244ea160>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "xAda8cBNLqHj",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Model Structure.**\n",
        "\n",
        "* Create and store the necessary dictionaries\n",
        "* Define an LSTM layer that takes as params: an input size (the number of characters), a hidden layer size `n_hidden`, a number of layers `n_layers`, a dropout probability `drop_prob`, and a batch_first boolean (True, since we are batching)\n",
        "* Define a dropout layer with `dropout_prob`\n",
        "* Define a fully-connected layer with params: input size `n_hidden` and output size (the number of characters)\n",
        "* Finally, initialize the weights "
      ]
    },
    {
      "metadata": {
        "id": "tHTbG8c2LqHl",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "---\n",
        "**LSTM Inputs/Outputs.**\n",
        "\n",
        "**I will  create a basic [LSTM layer](https://pytorch.org/docs/stable/nn.html#lstm) as follows.**\n",
        "\n",
        "```python\n",
        "self.lstm = nn.LSTM(input_size, n_hidden, n_layers, \n",
        "                            dropout=drop_prob, batch_first=True)\n",
        "```\n",
        "\n",
        "where `input_size` is the number of characters this cell expects to see as sequential input, and `n_hidden` is the number of units in the hidden layers in the cell. And we can add dropout by adding a dropout parameter with a specified probability; this will automatically add dropout to the inputs or outputs. Finally, in the `forward` function, we can stack up the LSTM cells into layers using `.view`. With this, you pass in a list of cells and it will send the output of one cell into the next cell.\n",
        "\n",
        "We also need to create an initial hidden state of all zeros. This is done like so\n",
        "\n",
        "```python\n",
        "self.init_hidden()\n",
        "```"
      ]
    },
    {
      "metadata": {
        "id": "kryTihuhLqHm",
        "colab_type": "code",
        "outputId": "799057ef-9e3b-4269-cee7-090b52c2c7da",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "### Check if GPU is available\n",
        "train_on_gpu = torch.cuda.is_available()\n",
        "if(train_on_gpu):\n",
        "    print('Training on GPU!')\n",
        "else: \n",
        "    print('No GPU available, training on CPU; consider making n_epochs very small.')"
      ],
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Training on GPU!\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "NHrMcxIiLqHs",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "class CharRNN(nn.Module):\n",
        "    \n",
        "    def __init__(self, tokens, n_hidden=256, n_layers=2,\n",
        "                               drop_prob=0.5, lr=0.001):\n",
        "        super().__init__()\n",
        "        self.drop_prob = drop_prob\n",
        "        self.n_layers = n_layers\n",
        "        self.n_hidden = n_hidden\n",
        "        self.lr = lr\n",
        "        \n",
        "        ### Creating character dictionaries\n",
        "        self.chars = tokens\n",
        "        self.int2char = dict(enumerate(self.chars))\n",
        "        self.char2int = {ch: ii for ii, ch in self.int2char.items()}\n",
        "        \n",
        "        ### Define the LSTM\n",
        "        self.lstm = nn.LSTM(len(self.chars), n_hidden, n_layers, \n",
        "                            dropout=drop_prob, batch_first=True)\n",
        "        \n",
        "        ### Define a dropout layer\n",
        "        self.dropout = nn.Dropout(drop_prob)\n",
        "        \n",
        "        ### Define the final, fully-connected output layer\n",
        "        self.fc = nn.Linear(n_hidden, len(self.chars))\n",
        "      \n",
        "    \n",
        "    def forward(self, x, hidden):\n",
        "        ''' Forward pass through the network. \n",
        "            These inputs are x, and the hidden/cell state `hidden`. '''\n",
        "                \n",
        "        ### Get the outputs and the new hidden state from the lstm\n",
        "        r_output, hidden = self.lstm(x, hidden)\n",
        "        \n",
        "        ### Pass through a dropout layer\n",
        "        out = self.dropout(r_output)\n",
        "        \n",
        "        ### Stack up LSTM outputs using view\n",
        "        ### I may need to use contiguous to reshape the output\n",
        "        out = out.contiguous().view(-1, self.n_hidden)\n",
        "        \n",
        "        ### Put x through the fully-connected layer\n",
        "        out = self.fc(out)\n",
        "        \n",
        "        ### Return the final output and the hidden state\n",
        "        return out, hidden\n",
        "    \n",
        "    \n",
        "    def init_hidden(self, batch_size):\n",
        "        ''' Initializes hidden state '''\n",
        "        ### Create two new tensors with sizes n_layers x batch_size x n_hidden,\n",
        "        ### Initialized to zero, for hidden state and cell state of LSTM\n",
        "        weight = next(self.parameters()).data\n",
        "        \n",
        "        if (train_on_gpu):\n",
        "            hidden = (weight.new(self.n_layers, batch_size, self.n_hidden).zero_().cuda(),\n",
        "                  weight.new(self.n_layers, batch_size, self.n_hidden).zero_().cuda())\n",
        "        else:\n",
        "            hidden = (weight.new(self.n_layers, batch_size, self.n_hidden).zero_(),\n",
        "                      weight.new(self.n_layers, batch_size, self.n_hidden).zero_())\n",
        "        \n",
        "        return hidden"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "kE5741mKLqHz",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Time to train\n",
        "\n",
        "The train function gives me the ability to set the number of epochs, the learning rate, and other parameters.\n",
        "\n",
        "Below I am using an Adam optimizer and cross entropy loss since I am looking at character class scores as output. I calculate the loss and perform backpropagation, as usual.\n",
        "\n",
        "A couple of details about training: \n",
        ">* Within the batch loop, I detach the hidden state from its history; this time setting it equal to a new *tuple* variable because an LSTM has a hidden state that is a tuple of the hidden and cell states.\n",
        "* I use [`clip_grad_norm_`](https://pytorch.org/docs/stable/_modules/torch/nn/utils/clip_grad.html) to help prevent exploding gradients."
      ]
    },
    {
      "metadata": {
        "id": "CXxPLk2JLqH2",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def train(net, data, epochs=10, batch_size=10, seq_length=50, lr=0.001, clip=5, val_frac=0.1, print_every=10):\n",
        "    ''' Training a network \n",
        "    \n",
        "        Arguments\n",
        "        ---------\n",
        "        \n",
        "        net: CharRNN network\n",
        "        data: text data to train the network\n",
        "        epochs: Number of epochs to train\n",
        "        batch_size: Number of mini-sequences per mini-batch, aka batch size\n",
        "        seq_length: Number of character steps per mini-batch\n",
        "        lr: learning rate\n",
        "        clip: gradient clipping\n",
        "        val_frac: Fraction of data to hold out for validation\n",
        "        print_every: Number of steps for printing training and validation loss\n",
        "    \n",
        "    '''\n",
        "    net.train()\n",
        "    \n",
        "    opt = torch.optim.Adam(net.parameters(), lr=lr)\n",
        "    criterion = nn.CrossEntropyLoss()\n",
        "    \n",
        "    ### Create training and validation data\n",
        "    val_idx = int(len(data)*(1-val_frac))\n",
        "    data, val_data = data[:val_idx], data[val_idx:]\n",
        "    \n",
        "    if(train_on_gpu):\n",
        "        net.cuda()\n",
        "    \n",
        "    counter = 0\n",
        "    n_chars = len(net.chars)\n",
        "    for e in range(epochs):\n",
        "        ### Initialize hidden state\n",
        "        h = net.init_hidden(batch_size)\n",
        "        \n",
        "        for x, y in get_batches(data, batch_size, seq_length):\n",
        "            counter += 1\n",
        "            \n",
        "            ### One-hot encode the data and make them Torch tensors\n",
        "            x = one_hot_encode(x, n_chars)\n",
        "            inputs, targets = torch.from_numpy(x), torch.from_numpy(y)\n",
        "            \n",
        "            if(train_on_gpu):\n",
        "                inputs, targets = inputs.cuda(), targets.cuda()\n",
        "\n",
        "            ### Creating new variables for the hidden state, otherwise\n",
        "            ### the code would backprop through the entire training history\n",
        "            h = tuple([each.data for each in h])\n",
        "\n",
        "            ### Zero accumulated gradients\n",
        "            net.zero_grad()\n",
        "            \n",
        "            ### Get the output from the model\n",
        "            output, h = net(inputs, h)\n",
        "            \n",
        "            ### Calculate the loss and perform backprop\n",
        "            loss = criterion(output, targets.view(batch_size*seq_length))\n",
        "            loss.backward()\n",
        "            \n",
        "            ### `clip_grad_norm` helps prevent the exploding gradient problem in RNNs / LSTMs.\n",
        "            nn.utils.clip_grad_norm_(net.parameters(), clip)\n",
        "            opt.step()\n",
        "            \n",
        "            ### Loss stats\n",
        "            if counter % print_every == 0:\n",
        "                ### Get validation loss\n",
        "                val_h = net.init_hidden(batch_size)\n",
        "                val_losses = []\n",
        "                net.eval()\n",
        "                for x, y in get_batches(val_data, batch_size, seq_length):\n",
        "                    \n",
        "                    ### One-hot encode our data and make them Torch tensors\n",
        "                    x = one_hot_encode(x, n_chars)\n",
        "                    x, y = torch.from_numpy(x), torch.from_numpy(y)\n",
        "                    \n",
        "                    ### Creating new variables for the hidden state, otherwise\n",
        "                    ### the code would backprop through the entire training history\n",
        "                    val_h = tuple([each.data for each in val_h])\n",
        "                    \n",
        "                    inputs, targets = x, y\n",
        "                    if(train_on_gpu):\n",
        "                        inputs, targets = inputs.cuda(), targets.cuda()\n",
        "\n",
        "                    output, val_h = net(inputs, val_h)\n",
        "                    val_loss = criterion(output, targets.view(batch_size*seq_length))\n",
        "                \n",
        "                    val_losses.append(val_loss.item())\n",
        "                \n",
        "                net.train() # reset to train mode after iterating through validation data\n",
        "                \n",
        "                print(\"Epoch: {}/{}...\".format(e+1, epochs),\n",
        "                      \"Step: {}...\".format(counter),\n",
        "                      \"Loss: {:.4f}...\".format(loss.item()),\n",
        "                      \"Val Loss: {:.4f}\".format(np.mean(val_losses)))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "8C12-e1bLqH8",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Instantiating the model.**\n",
        "\n",
        "**Now I can actually train the network.** \n",
        "\n",
        "**First I'll create the network itself, with some given hyperparameters.**\n",
        "\n",
        "**Then, define the mini-batches sizes, and start training.** "
      ]
    },
    {
      "metadata": {
        "id": "YWbJOzPVLqH-",
        "colab_type": "code",
        "outputId": "8f78e461-c20c-45f9-e0b6-68d5a8be1dd5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 102
        }
      },
      "cell_type": "code",
      "source": [
        "### Define and print the net\n",
        "n_hidden=512\n",
        "n_layers=2\n",
        "\n",
        "net = CharRNN(chars, n_hidden, n_layers)\n",
        "print(net)"
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "CharRNN(\n",
            "  (lstm): LSTM(83, 512, num_layers=2, batch_first=True, dropout=0.5)\n",
            "  (dropout): Dropout(p=0.5)\n",
            "  (fc): Linear(in_features=512, out_features=83, bias=True)\n",
            ")\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "scrolled": true,
        "id": "qiWsQo1PLqIE",
        "colab_type": "code",
        "outputId": "2bff143c-c294-4629-c616-60aff90a012d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 238
        }
      },
      "cell_type": "code",
      "source": [
        "### Define mini_batches\n",
        "batch_size = 128\n",
        "seq_length = 100\n",
        "n_epochs = 1 \n",
        "\n",
        "### Train the model\n",
        "train(net, encoded, epochs=n_epochs, batch_size=batch_size, seq_length=seq_length, lr=0.001, print_every=10)"
      ],
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch: 1/1... Step: 10... Loss: 3.2603... Val Loss: 3.2075\n",
            "Epoch: 1/1... Step: 20... Loss: 3.1482... Val Loss: 3.1394\n",
            "Epoch: 1/1... Step: 30... Loss: 3.1430... Val Loss: 3.1238\n",
            "Epoch: 1/1... Step: 40... Loss: 3.1145... Val Loss: 3.1191\n",
            "Epoch: 1/1... Step: 50... Loss: 3.1409... Val Loss: 3.1172\n",
            "Epoch: 1/1... Step: 60... Loss: 3.1181... Val Loss: 3.1150\n",
            "Epoch: 1/1... Step: 70... Loss: 3.1090... Val Loss: 3.1130\n",
            "Epoch: 1/1... Step: 80... Loss: 3.1180... Val Loss: 3.1062\n",
            "Epoch: 1/1... Step: 90... Loss: 3.1096... Val Loss: 3.0909\n",
            "Epoch: 1/1... Step: 100... Loss: 3.0697... Val Loss: 3.0561\n",
            "Epoch: 1/1... Step: 110... Loss: 3.0181... Val Loss: 2.9963\n",
            "Epoch: 1/1... Step: 120... Loss: 2.8986... Val Loss: 2.8820\n",
            "Epoch: 1/1... Step: 130... Loss: 2.8250... Val Loss: 2.7899\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "fkTUtv0VLqII",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Getting the best model.**\n",
        "\n",
        "**To set the hyperparameters to get the best performance, I'll want to watch the training and validation losses.**\n",
        "\n",
        "**If the training loss is much lower than the validation loss, the model is overfitting.** \n",
        "\n",
        "**Increase regularization (more dropout) or use a smaller network.** \n",
        "\n",
        "**If the training and validation losses are close, the model is underfitting so I can increase the size of the network.**"
      ]
    },
    {
      "metadata": {
        "id": "Q73-XhyALqIK",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Hyperparameters\n",
        "\n",
        "Here are the hyperparameters for the network.\n",
        "\n",
        "In defining the model:\n",
        "* `n_hidden` - The number of units in the hidden layers.\n",
        "* `n_layers` - Number of hidden LSTM layers to use.\n",
        "\n",
        "We assume that dropout probability and learning rate will be kept at the default, in this example.\n",
        "\n",
        "And in training:\n",
        "* `batch_size` - Number of sequences running through the network in one pass.\n",
        "* `seq_length` - Number of characters in the sequence the network is trained on. Larger is better typically, the network will learn more long range dependencies. But it takes longer to train. 100 is typically a good number here.\n",
        "* `lr` - Learning rate for training\n",
        "\n",
        "> ## Tips and Tricks\n",
        "\n",
        ">### Monitoring Validation Loss vs. Training Loss\n",
        ">If you're somewhat new to Machine Learning or Neural Networks it can take a bit of expertise to get good models. The most important quantity to keep track of is the difference between your training loss (printed during training) and the validation loss (printed once in a while when the RNN is run on the validation data (by default every 1000 iterations)). In particular:\n",
        "\n",
        "> - If your training loss is much lower than validation loss then this means the network might be **overfitting**. Solutions to this are to decrease your network size, or to increase dropout. For example you could try dropout of 0.5 and so on.\n",
        "> - If your training/validation loss are about equal then your model is **underfitting**. Increase the size of your model (either number of layers or the raw number of neurons per layer)\n",
        "\n",
        "> ### Approximate number of parameters\n",
        "\n",
        "> The two most important parameters that control the model are `n_hidden` and `n_layers`. I would advise that you always use `n_layers` of either 2/3. The `n_hidden` can be adjusted based on how much data you have. The two important quantities to keep track of here are:\n",
        "\n",
        "> - The number of parameters in your model. This is printed when you start training.\n",
        "> - The size of your dataset. 1MB file is approximately 1 million characters.\n",
        "\n",
        ">These two should be about the same order of magnitude. It's a little tricky to tell. Here are some examples:\n",
        "\n",
        "> - I have a 100MB dataset and I'm using the default parameter settings (which currently print 150K parameters). My data size is significantly larger (100 mil >> 0.15 mil), so I expect to heavily underfit. I am thinking I can comfortably afford to make `n_hidden` larger.\n",
        "> - I have a 10MB dataset and running a 10 million parameter model. I'm slightly nervous and I'm carefully monitoring my validation loss. If it's larger than my training loss then I may want to try to increase dropout a bit and see if that helps the validation loss.\n",
        "\n",
        "> ### Best models strategy\n",
        "\n",
        ">The winning strategy to obtaining very good models (if you have the compute time) is to always err on making the network larger (as large as you're willing to wait for it to compute) and then try different dropout values (between 0,1). Whatever model has the best validation performance (the loss, written in the checkpoint filename, low is good) is the one you should use in the end.\n",
        "\n",
        ">It is very common in deep learning to run many different models with many different hyperparameter settings, and in the end take whatever checkpoint gave the best validation performance.\n",
        "\n",
        ">By the way, the size of your training and validation splits are also parameters. Make sure you have a decent amount of data in your validation set or otherwise the validation performance will be noisy and not very informative."
      ]
    },
    {
      "metadata": {
        "id": "6L3UJmj8LqIK",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Checkpoint\n",
        "\n",
        "**After training, I'll save the model so I can load it again later if I need too.** \n",
        "\n",
        "**Here I'm saving the parameters needed to create the same architecture, the hidden layer hyperparameters and the text characters.**"
      ]
    },
    {
      "metadata": {
        "id": "xr6L9bJOLqIP",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "### Change the name, for saving multiple files\n",
        "model_name = 'rnn_20_epoch.net'\n",
        "\n",
        "checkpoint = {'n_hidden': net.n_hidden,\n",
        "              'n_layers': net.n_layers,\n",
        "              'state_dict': net.state_dict(),\n",
        "              'tokens': net.chars}\n",
        "\n",
        "with open(model_name, 'wb') as f:\n",
        "    torch.save(checkpoint, f)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "KU10gQNJLqIT",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "---\n",
        "## Making Predictions\n",
        "\n",
        "Now that the model is trained, we'll want to sample from it and make predictions about next characters! To sample, we pass in a character and have the network predict the next character. Then we take that character, pass it back in, and get another predicted character. Just keep doing this and you'll generate a bunch of text!\n",
        "\n",
        "### A note on the `predict`  function\n",
        "\n",
        "The output of our RNN is from a fully-connected layer and it outputs a **distribution of next-character scores**.\n",
        "\n",
        "> To actually get the next character, we apply a softmax function, which gives us a *probability* distribution that we can then sample to predict the next character.\n",
        "\n",
        "### Top K sampling\n",
        "\n",
        "Our predictions come from a categorical probability distribution over all the possible characters. We can make the sample text and make it more reasonable to handle (with less variables) by only considering some $K$ most probable characters. This will prevent the network from giving us completely absurd characters while allowing it to introduce some noise and randomness into the sampled text. Read more about [topk, here](https://pytorch.org/docs/stable/torch.html#torch.topk).\n"
      ]
    },
    {
      "metadata": {
        "id": "XFrQvVQZLqIT",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def predict(net, char, h=None, top_k=None):\n",
        "        ''' Given a character, predict the next character.\n",
        "            Returns the predicted character and the hidden state.\n",
        "        '''\n",
        "        \n",
        "        # tensor inputs\n",
        "        x = np.array([[net.char2int[char]]])\n",
        "        x = one_hot_encode(x, len(net.chars))\n",
        "        inputs = torch.from_numpy(x)\n",
        "        \n",
        "        if(train_on_gpu):\n",
        "            inputs = inputs.cuda()\n",
        "        \n",
        "        # detach hidden state from history\n",
        "        h = tuple([each.data for each in h])\n",
        "        # get the output of the model\n",
        "        out, h = net(inputs, h)\n",
        "\n",
        "        # get the character probabilities\n",
        "        p = F.softmax(out, dim=1).data\n",
        "        if(train_on_gpu):\n",
        "            p = p.cpu() # move to cpu\n",
        "        \n",
        "        # get top characters\n",
        "        if top_k is None:\n",
        "            top_ch = np.arange(len(net.chars))\n",
        "        else:\n",
        "            p, top_ch = p.topk(top_k)\n",
        "            top_ch = top_ch.numpy().squeeze()\n",
        "        \n",
        "        # select the likely next character with some element of randomness\n",
        "        p = p.numpy().squeeze()\n",
        "        char = np.random.choice(top_ch, p=p/p.sum())\n",
        "        \n",
        "        # return the encoded value of the predicted char and the hidden state\n",
        "        return net.int2char[char], h"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "Mh6jZ9mmLqIX",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Priming and generating text \n",
        "\n",
        "Typically you'll want to prime the network so you can build up a hidden state. Otherwise the network will start out generating characters at random. In general the first bunch of characters will be a little rough since it hasn't built up a long history of characters to predict from."
      ]
    },
    {
      "metadata": {
        "id": "WzbNu_GYLqIX",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def sample(net, size, prime='The', top_k=None):\n",
        "        \n",
        "    if(train_on_gpu):\n",
        "        net.cuda()\n",
        "    else:\n",
        "        net.cpu()\n",
        "    \n",
        "    net.eval() # eval mode\n",
        "    \n",
        "    # First off, run through the prime characters\n",
        "    chars = [ch for ch in prime]\n",
        "    h = net.init_hidden(1)\n",
        "    for ch in prime:\n",
        "        char, h = predict(net, ch, h, top_k=top_k)\n",
        "\n",
        "    chars.append(char)\n",
        "    \n",
        "    # Now pass in the previous character and get a new one\n",
        "    for ii in range(size):\n",
        "        char, h = predict(net, chars[-1], h, top_k=top_k)\n",
        "        chars.append(char)\n",
        "\n",
        "    return ''.join(chars)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "Nvv9w871LqIc",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 88
        },
        "outputId": "2dbf13c2-6338-4360-aa31-bcf32c9a5896"
      },
      "cell_type": "code",
      "source": [
        "print(sample(net, 1000, prime='Anna', top_k=5))"
      ],
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Annan  aed tare he e ee tired\n",
            " he tirt had har tie asta a aad ant hare aed an tar aon aa sied ia  atthes tees ho  ho thes ittat oan  aet oed an etestes ot etitt har hhe etang ae het to  hhi thon  he  hit tirethod at eed annt arind ton tond tart ae hher io  ao  oir the had hh shon he thard hanten torttis tond oat teredd an ha thir thhe  hhi ta attar iod thodss hin torteses oe thes an atirens\n",
            " horens onn ho  hi te artits her eadset oedd tin at and horss ata ethat tote tint inns hh ass atan as the  ied at hat hhe sit te ee tan thetsin ha soest hen ha te entestit an  har thare ha  aon ann  anredtas aa a aet on ae  he the tartta hantthhe  oong atatin hhi sens atis on iostir ee aes ae ans ie end en tar ther hotit oa hhe then ied an ian  hhe ton thins he hate at ont an ha tee ata  in eters hhrrist on tae hiti hor on the  ha  earet on and itte tho te ae tares oan ae  at he aadst are tha hirtare oe hartan aos on e etat hot onn  oond, ie  aot on atit hid tan etins ior enti ae ae ther ans annte thes a\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "zRboYuR2LqIf",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Loading a checkpoint"
      ]
    },
    {
      "metadata": {
        "id": "QQyZzOk7LqIg",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "### Here we have loaded in a model that trained over 20 epochs `rnn_20_epoch.net`\n",
        "with open('rnn_20_epoch.net', 'rb') as f:\n",
        "    checkpoint = torch.load(f)\n",
        "    \n",
        "loaded = CharRNN(checkpoint['tokens'], n_hidden=checkpoint['n_hidden'], n_layers=checkpoint['n_layers'])\n",
        "loaded.load_state_dict(checkpoint['state_dict'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "SSFDBKMvLqIl",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 139
        },
        "outputId": "0150eb80-4270-4267-d439-cb1ef3cdf6eb"
      },
      "cell_type": "code",
      "source": [
        "### Sample using a loaded model\n",
        "print(sample(loaded, 2000, top_k=5, prime=\"And Levin said\"))"
      ],
      "execution_count": 60,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "And Levin said  eose hod io an hh tetthad hens aad hhr hed tons ae een had hhedse oente eesend ta thod tho shatiss tar ea and it end tis atthe tees hhed ian tand, on to ennd ann tee aod eteretthe aa  hat onte hh sate thhrri no et arint at ete iong aesed hhe an thet ta  hot an tot tae the tin thi tonn  ao  east aen at hentetes\n",
            "sotes inndd her at oo  iars ins tors aa  oa ten ten hhed an aa  oeneg ta a ta enn te ast hed tar he te aret iod hi  eeritestin atsirn ont ar hots ao  hond oer oos ae e hor ethed tho thon  herti hirtin tho  han tor etenn at ant hh es tan tha toe hod aot eend an tan anse on iti tan he hod aans hat ho  eas oer annte at hhrrendd hin aa sathon atarin hant oot the ate ees an hond ao toet he atit herenn heret aod thar hed ote tat ant hort oen attin ha esit ei  ar eis ao tand atethen\n",
            "nat he aod iots oed thi etan  ae an in he tho  ae an ho  hit ot tie hat hand thin terind he hhe anttan toes hh  aor his thin the  en aidd aod hett an hed thed tonn ao hh  on toe hor hir ho tha tit aes an it oe end ind tanse ae har ie  aed toristhed oo  an ho  ee ten arte at ie thi tor oan han hods her oetter teren ao  aed iat hi te hhrtti  oote hod eo ta tin toe  oo ethang ooss ansi tindt tons io  eee tored ae ae hor oat eed,thhs aad an tere hhetit hot oe tho e eed taet herthens, hhe tons hed oed ante he toe thord oe ae thod aed ann tore an and toring ha ha tha  aa hote tond\n",
            "\n",
            "hhing tha e ens an hin an ein ated ha tor tore he aa terttan an at arend io ens aos ie aed an he hate an ete ha tet and had it ae tho toe  aed hid antta he hir oa tin hhr ee tir aa  ea e etes hid eesend ant eresedd hir and thit hod an te hiret the aet at han enn hetet hh shongs tons toe  ha  hon tee  oed the  hir thes,stes oos tho sho ha es the tha se eat oe tari  har anse tho  aes hengt tant aed hhe en ete end atanss tise ia tor on arr oo the  hort on at ans oot ars oet ho  horte ons, ont hid hodstas tettirs hher hirtetistite  eet tha terte eo tor hin hhe  ann and iat tie he ha esind hir ee to  on iod inn ethor\n",
            "nth\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}