FactoresLatentesInterpretacion.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "FactoresLatentesInterpretacion.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "metadata": {
        "id": "mX49CO0uhVAM",
        "colab_type": "code",
        "outputId": "f2ab8d7b-e7d4-4515-ed0e-f10097926916",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 150
        }
      },
      "cell_type": "code",
      "source": [
        "pip install git+git://github.com/maciejkula/spotlight.git"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting git+git://github.com/maciejkula/spotlight.git\n",
            "  Cloning git://github.com/maciejkula/spotlight.git to /tmp/pip-req-build-3i960z1w\n",
            "Requirement already satisfied (use --upgrade to upgrade): spotlight==0.1.5 from git+git://github.com/maciejkula/spotlight.git in /usr/local/lib/python3.6/dist-packages\n",
            "Requirement already satisfied: torch>=0.4.0 in /usr/local/lib/python3.6/dist-packages (from spotlight==0.1.5) (1.0.1.post2)\n",
            "Building wheels for collected packages: spotlight\n",
            "  Building wheel for spotlight (setup.py) ... \u001b[?25ldone\n",
            "\u001b[?25h  Stored in directory: /tmp/pip-ephem-wheel-cache-skgp99or/wheels/5f/eb/0a/20af178044caa54b089e63026090ab13ca70c45c80a18f7253\n",
            "Successfully built spotlight\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "HhavF2ODkPn6",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "4JLXm5byeRfV",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# Interpretación de los factores latentes\n",
        "\n",
        "Los métodos de factorización matricial para sistemas de recomendación intentan representar a *usuarios* y *items* en un mismo espacio vectorial n-dimensional en función de sus interacciones.\n",
        "\n",
        "En este notebook vamos a explorar si podemos extraer información relevante de estas representaciones además de poder hacer recomendaciones.\n",
        "\n",
        "Para ello vamos a utilizar el dataset *Movielens 100K*, que cuenta con 100K ratings de 943 usuarios en 1682 películas"
      ]
    },
    {
      "metadata": {
        "id": "UXc827a3fGMT",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Carga de datos\n",
        "\n",
        "Vamos a obtener los datos de la [página de Grouplens](http://files.grouplens.org/datasets/movielens/ml-100k/).\n",
        "\n",
        "Leemos primero los metadatos de las películas."
      ]
    },
    {
      "metadata": {
        "id": "RDOXw9UoT3HA",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "movies = pd.read_csv(\n",
        "    \"http://files.grouplens.org/datasets/movielens/ml-100k/u.item\", \n",
        "    sep=\"|\", header = None, encoding = \"ISO-8859-1\"\n",
        ")\n",
        "movies.columns = [\n",
        "    \"item_id\",  \"movie_title\", \"release_date\", \"video_release_date\",\n",
        "    \"IMDb URL\" , \"unknown\", \"Action\", \"Adventure\", \"Animation\",\n",
        "    \"Children\" , \"Comedy\", \"Crime\", \"Documentary\", \"Drama\", \"Fantasy\",\n",
        "    \"Film-Noir\", \"Horror\", \"Musical\", \"Mystery\", \"Romance\", \"Sci-Fi\",\n",
        "    \"Thriller\", \"War\", \"Western\"\n",
        "]\n",
        "movies.head()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "hBfQ2cOcqTgg",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Posteriormente leemos los datos de los usuarios."
      ]
    },
    {
      "metadata": {
        "id": "4qBH5UESqRH7",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "users = pd.read_csv(\n",
        "    \"http://files.grouplens.org/datasets/movielens/ml-100k/u.user\", \n",
        "    sep=\"|\", header = None, encoding = \"ISO-8859-1\"\n",
        ")\n",
        "\n",
        "users.columns = [\n",
        "    \"user_id\", \"age\", \"gender\", \"occupation\", \"zip_code\"\n",
        "]\n",
        "\n",
        "users.head()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "v3xSWY1ufg6b",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Por ultimo leemos las interacciones usuario-película con su rating asociado."
      ]
    },
    {
      "metadata": {
        "id": "37QDNL2qSwMb",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "dataset = pd.read_csv(\"http://files.grouplens.org/datasets/movielens/ml-100k/u.data\", sep = \"\\t\", header = None)\n",
        "dataset.columns = [\"user_id\", \"item_id\", \"rating\", \"timestamp\"]\n",
        "dataset.head()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "zmugQyRhfoMd",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Por último recodificamos los ids de items y usuarios, ya que vienen numerados de 1 a N y nos interesa que estén de 0 a N-1."
      ]
    },
    {
      "metadata": {
        "id": "mtVizRMsSyzO",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "dataset[\"user_id\"] = dataset[\"user_id\"] - 1\n",
        "dataset[\"item_id\"] = dataset[\"item_id\"] - 1\n",
        "movies[\"item_id\"] = movies[\"item_id\"] - 1\n",
        "users[\"user_id\"] = users[\"user_id\"] - 1"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "ZZDkVuVEf2k0",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Entrenamiento del modelo de factorización\n",
        "\n",
        "Para el entrenamiento del modelo vamos a utilizar [spotlight](https://maciejkula.github.io/spotlight/index.html)."
      ]
    },
    {
      "metadata": {
        "id": "H2JEdr-FVf9Q",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from spotlight.interactions import Interactions\n",
        "from spotlight.factorization.implicit import ImplicitFactorizationModel"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "-1OmLuCCgNwR",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Construimos el dataset para que se adecúe al formato de entrada de la librería y entrenamos el modelo.\n",
        "\n",
        "Vamos a elegir 5 factores latentes y un número elevado de epoch, ya que para el análisis nos da un poco lo mismo el overfit."
      ]
    },
    {
      "metadata": {
        "id": "2c0YRQmPXH5i",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "interactions = Interactions(dataset[\"user_id\"].values, dataset[\"item_id\"].values)\n",
        "\n",
        "model = ImplicitFactorizationModel(\n",
        "    loss='pointwise', \n",
        "    embedding_dim=32, \n",
        "    n_iter=50,\n",
        "    random_state=np.random.RandomState(0)\n",
        ")\n",
        "\n",
        "model.fit(interactions, verbose = False)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "mlJmFrFAg2Qr",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Extraemos los *embeddings*, que representan los factores latentes de usuarios e items."
      ]
    },
    {
      "metadata": {
        "id": "DE6PnHUPYBff",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "user_embeddings = model._net.user_embeddings.weight.data.numpy()\n",
        "item_embeddings = model._net.item_embeddings.weight.data.numpy()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "_2FbGDTuhDio",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Exploración de los factores latentes\n",
        "\n",
        "Tenemos los factores $u_{ij}$, la idea es que si elegimos un factor $u_{aj}$ que represente una película típica de, por ejemplo ciencia ficción y otro $u_{bj}$ que represente una película romántica, podemos definir una dirección $\\frac{u_{aj} - u_{bj}}{||u_{aj} - u_{bj}||}$ que represente la tendencia a las películas romanticas o películas de ciencia ficción tanto para usuarios como para items.\n",
        "\n",
        "Para hallar este valor, solo tenemos que proyectar sobre esa dirección los factores $u_{ij}$."
      ]
    },
    {
      "metadata": {
        "id": "Q54DI9IYlX5P",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Dimensión Ciencia Ficción vs Comedia Romántica\n",
        "\n",
        "Como película representativa de ciencia ficción tomamos *Star Trek: Generations*."
      ]
    },
    {
      "metadata": {
        "id": "M6vqb5JIlrrz",
        "colab_type": "code",
        "outputId": "074b3a52-2aed-47f2-ca16-a1508826a3d3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 176
        }
      },
      "cell_type": "code",
      "source": [
        "movies[movies.item_id == 379]"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>item_id</th>\n",
              "      <th>movie_title</th>\n",
              "      <th>release_date</th>\n",
              "      <th>video_release_date</th>\n",
              "      <th>IMDb URL</th>\n",
              "      <th>unknown</th>\n",
              "      <th>Action</th>\n",
              "      <th>Adventure</th>\n",
              "      <th>Animation</th>\n",
              "      <th>Children</th>\n",
              "      <th>...</th>\n",
              "      <th>Fantasy</th>\n",
              "      <th>Film-Noir</th>\n",
              "      <th>Horror</th>\n",
              "      <th>Musical</th>\n",
              "      <th>Mystery</th>\n",
              "      <th>Romance</th>\n",
              "      <th>Sci-Fi</th>\n",
              "      <th>Thriller</th>\n",
              "      <th>War</th>\n",
              "      <th>Western</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>379</th>\n",
              "      <td>379</td>\n",
              "      <td>Star Trek: Generations (1994)</td>\n",
              "      <td>01-Jan-1994</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Star%20Trek:%...</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1 rows × 24 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     item_id                    movie_title release_date  video_release_date  \\\n",
              "379      379  Star Trek: Generations (1994)  01-Jan-1994                 NaN   \n",
              "\n",
              "                                              IMDb URL  unknown  Action  \\\n",
              "379  http://us.imdb.com/M/title-exact?Star%20Trek:%...        0       1   \n",
              "\n",
              "     Adventure  Animation  Children   ...     Fantasy  Film-Noir  Horror  \\\n",
              "379          1          0         0   ...           0          0       0   \n",
              "\n",
              "     Musical  Mystery  Romance  Sci-Fi  Thriller  War  Western  \n",
              "379        0        0        0       1         0    0        0  \n",
              "\n",
              "[1 rows x 24 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "metadata": {
        "id": "vnQq69E3lxvU",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Por otro lado, como película representativa de comedia romántica tomamos *La boda de mi mejor amigo*."
      ]
    },
    {
      "metadata": {
        "id": "vOBb09pKlbyz",
        "colab_type": "code",
        "outputId": "fa9cd8f5-238f-491c-9d11-f4a633fe45b3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 193
        }
      },
      "cell_type": "code",
      "source": [
        "movies[movies.item_id == 254]"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>item_id</th>\n",
              "      <th>movie_title</th>\n",
              "      <th>release_date</th>\n",
              "      <th>video_release_date</th>\n",
              "      <th>IMDb URL</th>\n",
              "      <th>unknown</th>\n",
              "      <th>Action</th>\n",
              "      <th>Adventure</th>\n",
              "      <th>Animation</th>\n",
              "      <th>Children</th>\n",
              "      <th>...</th>\n",
              "      <th>Fantasy</th>\n",
              "      <th>Film-Noir</th>\n",
              "      <th>Horror</th>\n",
              "      <th>Musical</th>\n",
              "      <th>Mystery</th>\n",
              "      <th>Romance</th>\n",
              "      <th>Sci-Fi</th>\n",
              "      <th>Thriller</th>\n",
              "      <th>War</th>\n",
              "      <th>Western</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>254</th>\n",
              "      <td>254</td>\n",
              "      <td>My Best Friend's Wedding (1997)</td>\n",
              "      <td>20-Jun-1997</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?My+Best+Frien...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1 rows × 24 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     item_id                      movie_title release_date  \\\n",
              "254      254  My Best Friend's Wedding (1997)  20-Jun-1997   \n",
              "\n",
              "     video_release_date                                           IMDb URL  \\\n",
              "254                 NaN  http://us.imdb.com/M/title-exact?My+Best+Frien...   \n",
              "\n",
              "     unknown  Action  Adventure  Animation  Children   ...     Fantasy  \\\n",
              "254        0       0          0          0         0   ...           0   \n",
              "\n",
              "     Film-Noir  Horror  Musical  Mystery  Romance  Sci-Fi  Thriller  War  \\\n",
              "254          0       0        0        0        1       0         0    0   \n",
              "\n",
              "     Western  \n",
              "254        0  \n",
              "\n",
              "[1 rows x 24 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "metadata": {
        "id": "ThDCiH2LmBbh",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Hallamos la dirección y proyectamos los factores latentes de cada película sobre ella para calcular un *score*."
      ]
    },
    {
      "metadata": {
        "id": "wpuasiMPXuEb",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "u = item_embeddings[379] - item_embeddings[254]\n",
        "u_hat = u / (u**2).sum()**0.5\n",
        "movies[\"score\"] = np.transpose(np.dot(item_embeddings, np.expand_dims(u_hat, 1)))[0]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "cB0coNLzAz1j",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Si ordenamos las películas de menor a mayor *score* podemos observar que aquellas con un menor *score* son películas románticas mientras que aquellas con mayor son películas de ciencia ficción o fantasía."
      ]
    },
    {
      "metadata": {
        "id": "O50MCmZ9bAh4",
        "colab_type": "code",
        "outputId": "3420d37b-35f6-4851-f80d-fb8f5da04545",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 396
        }
      },
      "cell_type": "code",
      "source": [
        "movies.sort_values(\"score\").head()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>item_id</th>\n",
              "      <th>movie_title</th>\n",
              "      <th>release_date</th>\n",
              "      <th>video_release_date</th>\n",
              "      <th>IMDb URL</th>\n",
              "      <th>unknown</th>\n",
              "      <th>Action</th>\n",
              "      <th>Adventure</th>\n",
              "      <th>Animation</th>\n",
              "      <th>Children</th>\n",
              "      <th>...</th>\n",
              "      <th>Film-Noir</th>\n",
              "      <th>Horror</th>\n",
              "      <th>Musical</th>\n",
              "      <th>Mystery</th>\n",
              "      <th>Romance</th>\n",
              "      <th>Sci-Fi</th>\n",
              "      <th>Thriller</th>\n",
              "      <th>War</th>\n",
              "      <th>Western</th>\n",
              "      <th>score</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>254</th>\n",
              "      <td>254</td>\n",
              "      <td>My Best Friend's Wedding (1997)</td>\n",
              "      <td>20-Jun-1997</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?My+Best+Frien...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>-8.095102</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>274</th>\n",
              "      <td>274</td>\n",
              "      <td>Sense and Sensibility (1995)</td>\n",
              "      <td>01-Jan-1995</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Sense%20and%2...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>-5.637879</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>814</th>\n",
              "      <td>814</td>\n",
              "      <td>One Fine Day (1996)</td>\n",
              "      <td>30-Nov-1996</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?One%20Fine%20...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>-5.299332</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>236</th>\n",
              "      <td>236</td>\n",
              "      <td>Jerry Maguire (1996)</td>\n",
              "      <td>13-Dec-1996</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Jerry%20Magui...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>-4.990129</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1011</th>\n",
              "      <td>1011</td>\n",
              "      <td>Private Parts (1997)</td>\n",
              "      <td>07-Mar-1997</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Private%20Par...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>-4.892868</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 25 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "      item_id                      movie_title release_date  \\\n",
              "254       254  My Best Friend's Wedding (1997)  20-Jun-1997   \n",
              "274       274     Sense and Sensibility (1995)  01-Jan-1995   \n",
              "814       814              One Fine Day (1996)  30-Nov-1996   \n",
              "236       236             Jerry Maguire (1996)  13-Dec-1996   \n",
              "1011     1011             Private Parts (1997)  07-Mar-1997   \n",
              "\n",
              "      video_release_date                                           IMDb URL  \\\n",
              "254                  NaN  http://us.imdb.com/M/title-exact?My+Best+Frien...   \n",
              "274                  NaN  http://us.imdb.com/M/title-exact?Sense%20and%2...   \n",
              "814                  NaN  http://us.imdb.com/M/title-exact?One%20Fine%20...   \n",
              "236                  NaN  http://us.imdb.com/M/title-exact?Jerry%20Magui...   \n",
              "1011                 NaN  http://us.imdb.com/M/title-exact?Private%20Par...   \n",
              "\n",
              "      unknown  Action  Adventure  Animation  Children    ...     Film-Noir  \\\n",
              "254         0       0          0          0         0    ...             0   \n",
              "274         0       0          0          0         0    ...             0   \n",
              "814         0       0          0          0         0    ...             0   \n",
              "236         0       0          0          0         0    ...             0   \n",
              "1011        0       0          0          0         0    ...             0   \n",
              "\n",
              "      Horror  Musical  Mystery  Romance  Sci-Fi  Thriller  War  Western  \\\n",
              "254        0        0        0        1       0         0    0        0   \n",
              "274        0        0        0        1       0         0    0        0   \n",
              "814        0        0        0        1       0         0    0        0   \n",
              "236        0        0        0        1       0         0    0        0   \n",
              "1011       0        0        0        0       0         0    0        0   \n",
              "\n",
              "         score  \n",
              "254  -8.095102  \n",
              "274  -5.637879  \n",
              "814  -5.299332  \n",
              "236  -4.990129  \n",
              "1011 -4.892868  \n",
              "\n",
              "[5 rows x 25 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "metadata": {
        "id": "kZCOrRbuoydu",
        "colab_type": "code",
        "outputId": "e5286700-efee-4822-97dc-dc196263d808",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 380
        }
      },
      "cell_type": "code",
      "source": [
        "movies.sort_values(\"score\").tail()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>item_id</th>\n",
              "      <th>movie_title</th>\n",
              "      <th>release_date</th>\n",
              "      <th>video_release_date</th>\n",
              "      <th>IMDb URL</th>\n",
              "      <th>unknown</th>\n",
              "      <th>Action</th>\n",
              "      <th>Adventure</th>\n",
              "      <th>Animation</th>\n",
              "      <th>Children</th>\n",
              "      <th>...</th>\n",
              "      <th>Film-Noir</th>\n",
              "      <th>Horror</th>\n",
              "      <th>Musical</th>\n",
              "      <th>Mystery</th>\n",
              "      <th>Romance</th>\n",
              "      <th>Sci-Fi</th>\n",
              "      <th>Thriller</th>\n",
              "      <th>War</th>\n",
              "      <th>Western</th>\n",
              "      <th>score</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>228</th>\n",
              "      <td>228</td>\n",
              "      <td>Star Trek III: The Search for Spock (1984)</td>\n",
              "      <td>01-Jan-1984</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Star%20Trek%2...</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>5.991991</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>402</th>\n",
              "      <td>402</td>\n",
              "      <td>Batman (1989)</td>\n",
              "      <td>01-Jan-1989</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Batman%20(1989)</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6.041560</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>664</th>\n",
              "      <td>664</td>\n",
              "      <td>Alien 3 (1992)</td>\n",
              "      <td>01-Jan-1992</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Alien%203%20(...</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6.254703</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>61</th>\n",
              "      <td>61</td>\n",
              "      <td>Stargate (1994)</td>\n",
              "      <td>01-Jan-1994</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Stargate%20(1...</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6.815407</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>379</th>\n",
              "      <td>379</td>\n",
              "      <td>Star Trek: Generations (1994)</td>\n",
              "      <td>01-Jan-1994</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Star%20Trek:%...</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>8.016078</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 25 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     item_id                                 movie_title release_date  \\\n",
              "228      228  Star Trek III: The Search for Spock (1984)  01-Jan-1984   \n",
              "402      402                               Batman (1989)  01-Jan-1989   \n",
              "664      664                              Alien 3 (1992)  01-Jan-1992   \n",
              "61        61                             Stargate (1994)  01-Jan-1994   \n",
              "379      379               Star Trek: Generations (1994)  01-Jan-1994   \n",
              "\n",
              "     video_release_date                                           IMDb URL  \\\n",
              "228                 NaN  http://us.imdb.com/M/title-exact?Star%20Trek%2...   \n",
              "402                 NaN   http://us.imdb.com/M/title-exact?Batman%20(1989)   \n",
              "664                 NaN  http://us.imdb.com/M/title-exact?Alien%203%20(...   \n",
              "61                  NaN  http://us.imdb.com/M/title-exact?Stargate%20(1...   \n",
              "379                 NaN  http://us.imdb.com/M/title-exact?Star%20Trek:%...   \n",
              "\n",
              "     unknown  Action  Adventure  Animation  Children    ...     Film-Noir  \\\n",
              "228        0       1          1          0         0    ...             0   \n",
              "402        0       1          1          0         0    ...             0   \n",
              "664        0       1          0          0         0    ...             0   \n",
              "61         0       1          1          0         0    ...             0   \n",
              "379        0       1          1          0         0    ...             0   \n",
              "\n",
              "     Horror  Musical  Mystery  Romance  Sci-Fi  Thriller  War  Western  \\\n",
              "228       0        0        0        0       1         0    0        0   \n",
              "402       0        0        0        0       0         0    0        0   \n",
              "664       1        0        0        0       1         1    0        0   \n",
              "61        0        0        0        0       1         0    0        0   \n",
              "379       0        0        0        0       1         0    0        0   \n",
              "\n",
              "        score  \n",
              "228  5.991991  \n",
              "402  6.041560  \n",
              "664  6.254703  \n",
              "61   6.815407  \n",
              "379  8.016078  \n",
              "\n",
              "[5 rows x 25 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "metadata": {
        "id": "R_b5539HCF7t",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Vamos a hacer el mismo ejercicio per con los usuarios, proyectando sus factores latentes sobre la dimensión calculada."
      ]
    },
    {
      "metadata": {
        "id": "l8TsJglFq2ts",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "users[\"score\"] = np.transpose(np.dot(user_embeddings, np.expand_dims(u_hat, 1)))[0]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "kz7GmpoOD8Gl",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Si calculamos el score medio por género vemos que las mujeres están más inclinadas a ver películas románticas mientras que los hombres a películas de ciencia ficción.\n",
        "\n",
        "Por otro lado vemos que los ingenieros y programadores tienen un score mayor mientras que las amas de casa tienen el score mínimo."
      ]
    },
    {
      "metadata": {
        "id": "ToAz-M-qruuE",
        "colab_type": "code",
        "outputId": "70a4dde6-00a5-4a09-9321-01a942b37a53",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 371
        }
      },
      "cell_type": "code",
      "source": [
        "fig, (ax_1, ax_2) = plt.subplots(1, 2, figsize = (20, 5))\n",
        "\n",
        "sns.barplot(\n",
        "    x = \"gender\", y = \"score\", \n",
        "    data = users.groupby(\"gender\").agg({\"score\": \"mean\"}).reset_index(),\n",
        "    ax = ax_1\n",
        ")\n",
        "\n",
        "sns.barplot(\n",
        "    y = \"occupation\", x = \"score\", \n",
        "    data = users.groupby(\"occupation\").agg({\"score\": \"mean\"})\\\n",
        "      .reset_index().sort_values('score'),\n",
        "    ax = ax_2\n",
        ")\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/seaborn/categorical.py:1428: FutureWarning: remove_na is deprecated and is a private function. Do not use.\n",
            "  stat_data = remove_na(group_data)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABJ0AAAFBCAYAAADddN60AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3X98zfX///HbOWdnkt/L8asUeYts\nLAqxL5F5+/nuh4zNmBApTEpZmLdmTH4mVIgwM71pb2yKskR+bIQMSeONxsgw2vzYz/P9o0vn8957\naGw7x3bu18vFpfN6vZ4/Ho/TZS4vjz1fz5fBarVaERERERERERERKUJGRwcgIiIiIiIiIiKlj4pO\nIiIiIiIiIiJS5FR0EhERERERERGRIqeik4iIiIiIiIiIFDkVnUREREREREREpMip6CQiIiIiIiIi\nIkXO7kWnKVOm0Lt3b3x9fUlISMhzbefOnfTs2ZPevXszf/58e4cmIiIiIiIiIiJFxK5Fp927d3Pq\n1Ck+//xzJk+ezOTJk/NcDw0NZe7cuURGRrJjxw6OHTtmz/BERERERERERKSI2LXotGvXLry9vQGo\nV68eV65cIT09HYCkpCQqVapEzZo1MRqNPPPMM+zatcue4YmIiIiIiIiISBGxa9HpwoULVKlSxXbs\n5uZGSkoKACkpKbi5ud30moiIiIiIiIiIlCwujpzcarUWeozs7BxcXExFEM2t9XknoljHFyktVk7z\nd3QIReblz0Y6OgSREmHpgDmODkEcJCUlzdEh3FKVKveTmnrN0WHYhbPk6ix5gvPk6ix5gvPk6ix5\ngvPkarFUKPQYdi06VatWjQsXLtiOz58/j8Viuem13377jWrVqv3lmM7wP1qkpLiX/wEiIsXDHj/3\nRXHDIyXQmnV33TW1CMO41zlLrs6SJzhPrs6SJzhPrs6SJzhRrq/1LfQQdn28zsvLi02bNgFw+PBh\nqlWrRvny5QF46KGHSE9P5/Tp02RnZ7Nlyxa8vLzsGZ6IiIiIiIiIiBQRu650atasGe7u7vj6+mIw\nGPjnP/9JVFQUFSpUoGPHjkycOJG33noLgK5du1K3bl17hiciIiJSKPHx8URERPDhhx86OpQ8goKC\n6NSpE+3bt3d0KCIiIuJE7L6n0+jRo/McN2zY0Pa5efPmfP755/YOSUREREREREREiphDNxIXERER\nKW2uXr3K6NGjOXr0KJ06daJjx46EhIRgNBopV64cU6dO5ejRoyxfvhyTycRPP/3E0KFD+f777zly\n5AjvvPMO3t7efP311yxZsgQXFxc8PDwICgoiKiqKPXv2kJqaSmJiIqNGjSImJobjx48zY8YMPD09\nCQsLIyEhgYyMDPz8/PDx8bHFlpWVxeDBgxk6dCh169Zl3LhxZGVlYTKZCA0NpVatWvz973+nUaNG\neHl55ekrIiIicqdUdBIREREpQsePH+err74iNzeXDh06sHv3bt555x08PT1ZvHgxy5cvp2XLlhw5\ncoSNGzeyZ88eRo8eTWxsLAcOHCA8PJxWrVrx8ccf8/nnn+Pq6srIkSPZu3cvACdPnmTlypWsXr2a\nBQsWsHbtWqKiooiJiaFhw4Y8+OCDvPvuu9y4cQNvb+88haOwsDC6dOnC008/zdixYxk4cCCtW7dm\n69atfPTRR4SGhpKUlMT8+fOpX7++o75CERERKSVUdBIREREpQo0aNaJs2bIAWK1Wjh8/jqenJwAt\nW7Zk3rx5tGzZkoYNG+Lq6orFYqFOnTrcf//9PPDAA6SlpXHs2DGSk5MZNGgQAGlpaSQnJwPg4eGB\nwWDAYrHQoEEDTCYTVatWZd++fZQpU4YrV67g6+uL2WwmNfX/3q/z73//m8zMTCZMmADA/v37OXHi\nBB9//DE5OTm4ubkBULZsWRWcREREpEio6CQiIiJShFxcbn17lZWVhdFozNfuf/uYzWY8PDxYvHhx\nnvNRUVG37Ge1Wtm9ezdxcXGEh4djNptp2rRpnuunT5/m5MmT1KlTB7PZzJw5c6hWrVq+uUVERESK\ngtHRAYiIiIiUZvXr12f//v0A7NmzBw8Pj7/sU7duXY4fP87FixcB+PDDD/ntt9/+sl9qaio1atTA\nbDYTGxtLTk4OmZmZAPTo0YNx48Yxbtw4rFYrnp6ebN68GYBdu3YRHR19tymKiIiI3JSKTiIiIiLF\naPz48cyaNYuAgAAOHjxIQEDAX/YpW7YsY8eOZfDgwfj6+nL58uV8K5JupnXr1pw6dYq+ffuSlJRE\nu3btmDhxou16q1atqFevHsuXL2f48OHExsbi7+/P/PnzeeKJJwqTpoiIiEg+BqvVanV0EIWRkpJW\n7HOMnL6+2OcQKQ3mvP2co0MoMm/HjHd0CCIlwvTuocU+h8VSodjnkDtX7Pdga9YV7/giIiJyW5bX\n+hZ6DO3pJCIiIiL3np7P33VXi6WCXX4xeS9wllydJU9wnlydJU9wnlydJU9wrlwLS4/XiYiIiIiI\niIhIkdNKJxERERERB8hdM7fQY/z19vKlg7PkCc6Tq7PkCc6Tq7PkCU6U62tjCz2EVjqJiIiIiIiI\niEiRU9FJREREpJRKTk4mISEBgMmTJ5OUlOTgiERERMSZ6PE6ERERkVIqLi6Oa9eu0aRJE8aNG+fo\ncERERMTJaKWTiIiIiJ3l5OQwduxY+vXrh5+fH9u3b6dHjx6cPXuW7OxsXnzxRZKSkjh27BgBAQH0\n79+f119/nd9//x2ARYsW0bNnT3r16kVcXBynT5+mR48etvF79OhBQkIC8+bNY/ny5cTGxtKvXz9+\n/vln2rdvT0ZGBgC7d+9m+PDhpKenExgYSP/+/enbty8///yzQ74XERERKV200klERETEzqKjo7FY\nLEyZMoVLly7Rv39/xo0bx6xZs2jSpAmdOnWidu3a9O/fn5CQEOrUqUNERAQRERF06dKFTZs28a9/\n/YukpCQWLlzIa6+9lm8ONzc3XnzxRapUqUKHDh1YunQpRqORVq1asWvXLtq1a0dsbCydOnVi2bJl\ntGnTBh8fH44dO8bkyZP57LPPHPDNiIiISGmiopOIiIiIne3fv5+9e/eyb98+ADIyMmjWrBlRUVGs\nX7+elStXApCQkEBwcDAAmZmZNG7cmJ9++glPT0+MRiOPPPIIkydP5vTp0wWe++9//zvffvst7dq1\nY/v27YwYMYI33niDS5cusX79egCuX79exBmLiIiIM1LRSURERMTOzGYzQ4cOpXv37nnOX758mZyc\nHK5fv47ZbKZs2bIsX74cg8Fga7Np0yZyc3Pz9Pvv6wDZ2dm3nLt169ZMmzaNo0ePUrt2bcqXL4/Z\nbCY4OJimTZsWQXYiIiIif9CeTiIiIiJ25unpSWxsLAAXL15k1qxZbNiwgXr16jF48GBmzpwJQMOG\nDdm2bRsAGzZsYNeuXbi7u7Nv3z6ys7O5cOECw4YNo3z58ly8eBGr1UpKSortLXUGgyFfAcrV1ZWG\nDRuyePFiOnfubItn8+bNABw7dkyP1omIiEiR0EonERERETvr0qULcXFx+Pr6kpOTw6uvvsrcuXNZ\nsWIFFSpUYOXKlSQkJDBu3DiCg4NZtGgRZcqUYebMmVSuXJnnn3+evn37YrVaGTVqFJUqVaJ169a8\n9NJLNGzYkMcffxyApk2bMmbMGNzc3PLM37FjR4KCghg/fjwAffv25d1336VPnz7k5ubqTXciIiJS\nJAxWq9Xq6CAKIyUlrdjnGDl9fbHPIVIazHn7OUeHUGTejhnv6BBESoTp3UOLfQ6LpUKxzyF3zh73\nYHfLYqlwT8f3p9w1cx0dgoiIyC1Vf21socfQSicREREREQcw9hxR6DFKSoGtsJwlT3CeXJ0lT3Ce\nXJ0lT3CuXAtLezqJiIiIiIiIiEiR00onEREREZFS5vLnfR0dQpG67OgA7MhZcnWWPMF5cnWWPMF5\ncrUMX1foMbTSSUREREREREREipyKTiIiIiLFLD4+nsDAwLvuHxQUxJYtWwDYtGkTAFFRUbz//vtF\nEp+IiIhIcVDRSURERKSEOH36NBs2bHB0GCIiIiIFoj2dREREROzg6tWrjB49mqNHj9KpUyc6d+5M\nSEgIBoOBcuXKMXXqVCpWrEhYWBgJCQlkZGTg5+eHj4+PbYyQkBASEhKYN28etWrV4vz584wYMYJj\nx44xaNAgevbsyY4dO5g1axYmk4muXbvy8ssvs379elasWIHRaKR+/fpMmjSJqKgotm3bxvnz55k9\nezabN28mOjoao9GIt7c3AwcOdOC3JSIiIqWBik4iIiIidnD8+HG++uorcnNz6dChA3v27CEkJIQ6\ndeoQERFBREQEAwcO5MEHH+Tdd9/lxo0beHt75yk6DRo0iIiICIYPH05UVBRJSUlERkZy6tQpRo0a\nxUsvvcR7773HqlWrqFSpEq+//jq+vr5cv36dTz/9lIoVK+Lv78/Ro0cBOHv2LKtWreL06dNs3LiR\nyMhIAPz8/OjcuTO1atVyyHclIiIipYOKTiIiIiJ20KhRI8qWLQuA1WolISGB4OBgADIzM2ncuDFl\nypThypUr+Pr6YjabSU1Nve2Ynp6emEwmqlevTlpaGpcuXaJMmTK4ubkBsGDBAgBbAQr+KH5dvvzH\ne3caN26MwWDg4MGDnDp1ioCAAOCPVVlnzpxR0UlEREQKxa5Fp6ysLIKCgkhOTsZkMhEWFkbt2rXz\ntPnyyy9ZsmQJRqORVq1aMWrUKHuGKCIiIlIsXFzy3naVLVuW5cuXYzAYbOd2795NXFwc4eHhmM1m\nmjZtekdjGo1GcnNz85zLzMwkJCSEdevWYbFYePXVV23XzGaz7b/t2rUjJCTkrnITERERuRm7biQe\nExNDxYoViYyMZOjQocycOTPP9evXrzNjxgyWLl3K559/zs6dOzl27Jg9QxQRERGxi4YNG7Jt2zYA\nNmzYwK5du0hNTaVGjRqYzWZiY2PJyckhMzPT1sdoNJKdnX3LMatUqUJOTg6//fYbVquVV199latX\nr2IymbBYLJw9e5ZDhw6RlZWVp5+7uzvx8fFcv34dq9VKaGgoN27cKJ7ERURExGnYtei0a9cuOnbs\nCEDr1q3Zt29fnutly5Zl/fr1lC9fHoPBQOXKlW3Lv0VERERKk3HjxrFgwQL69u1LVFQUjz/+OK1b\nt+bUqVP07duXpKQk2rVrx8SJE2196tWrx08//cSUKVNuOe4///lPAgMD8fX1pVWrVlSpUgUvLy9e\neukl5s2bxyuvvEJYWFie4lWtWrUICAjA39+fXr16YbFYuO+++4ozfREREXECdn287sKFC7Y9BoxG\nIwaDgczMTFxdXW1typcvD8DRo0c5c+YMnp6e9gxRREREpMi1bNmSli1b2o7j4+MBWLlyZb62a9as\nsX1++eWX813/7rvv8p0rV64c3377LQCtWrWiVatWea5PnTo1z/GAAQPyjeHv74+/v/+tkxARERG5\nQ8VWdFq9ejWrV6/Oc+7AgQN5jq1W6037njx5ktGjRzNz5kzbXgO3UqXK/bi4mAoXrIgUCYulgqND\nEBE708+9yL2pcu8Vjg6hSFksFUhJSXN0GHbhLLk6S57gPLk6S57gXLkWVrEVnXx8fPK84hcgKCiI\nlJQUGjZsSFZWFlarNc8qJ4Bz584xbNgwpk2bxuOPP/6X86SmXivSuEXk7ukvXhHnY4+fexW2RERE\nREomuz5e5+XlxcaNG2nTpg1btmzJs8z8T+PGjWPixIm4u7vbMzQREREREafw81o/R4dwx352dAB2\n5Cy5Okue4Dy5Okue4Dy5WgbHFHoMu24k3rVrV3Jzc/Hz8yMiIoK33noLgIULF7J//35OnDjBDz/8\nwIcffki/fv3o168fsbGx9gxRREREpMS4evUqzz777B312bRpUzFFIyIiIpKXXVc6mUwmwsLC8p0f\nMmSI7fP/7vskIiIiIkXj9OnTbNiwgU6dOjk6FBEREXECdi06iYiIiEjhpKenM2LECDIyMnjyySeB\nP96GN3v2bFxcXKhevTphYWEYDAaCgoI4c+YMZcqUYdq0aYSEhJCQkMC8efPo378/QUFB/P7772Rn\nZzN+/Hjc3d35+9//TqNGjfDy8sq3P6eIiIjInbDr43UiIiIiUjjr1q2jfv36rFy50vbSlX/+85/M\nnj2bFStWUKlSJaKjo1m7di1Vq1Zl1apV9OrVi9jYWAYNGkSLFi0YPnw4y5Ytw9PTk/DwcMaOHWtb\njZ6UlMSwYcNUcBIREZFCU9FJREREpAQ5fvw4TZs2BaBFixZcvnwZg8FAzZo1AWjZsiVHjhzh8OHD\nNGvWDIBu3brRp0+fPOMcOnTI9lKXxo0bc+rUKQDKli1L/fr17ZWOiIiIlGIqOomIiIiUIFarFaPx\nj1u43NxcDAYDVqvVdj0rKwuDwYDJZCI3N/eW4/xvvz/bms3mYopcREREnI2KTiIiIiIlSN26dTl0\n6BDwx15OlSpVwmAwkJycDMDu3bvx8PCgcePGxMXFAbBlyxY++eQTjEYj2dnZwB+rm+Lj4wH48ccf\ntbpJREREipw2EhcREREpQV544QWGDRtG//79bRuJT5o0ibfeegsXFxdq165Nt27dyM3NZefOnfTt\n2xcXFxfef/99zGYzP/30E1OmTCEwMJCxY8cSEBCA1WplwoQJDs5MREREShsVnURERERKkIoVKxIe\nHm47DgwMBCAyMjJf22nTpuU7991339k+f/jhh/mu/7n6SURERKSwVHQSEREREXEiDV/IX6C811ks\nFUhJSXN0GHbhLLk6S57gPLk6S57gXLkWlvZ0EhERERERERGRIqeik4iIiIiIiIiIFDk9XiciIiIi\nIiXejuhejg5BRKRUeWHgV4UeQyudRERERO4hUVFRjBs3zvY2uWeffZarV68Wasxt27axcuXKoghP\nREREpMC00klERETkHlOxYkXGjBlTZOO1bdu2yMYSERERKSgVnURERETuMWfOnKFHjx5ERUUBsGDB\nAn744QdMJhPz589n8+bNbNu2jfPnzzN79myWLFlCQkICGRkZ+Pn54ePjQ1BQEGazmcuXL9O+fXsS\nExMZM2YMYWFhN21brVo1Dh8+THJyMjNmzMDd3d3B34KIiIiUdHq8TkREROQe16BBA1auXImHhwfr\n1q0D4OzZs0RERFC5cmUefPBBIiMjWblyJXPmzLH1q1SpEnPnzrUdZ2Rk3LJtZmYmixcvJiAggLVr\n19ovORERESm1tNJJRERE5B7XsmVLABo3bswPP/yAh4cHjRs3xmAwUKZMGa5cuYKvry9ms5nU1FRb\nvyZNmuQZ53Ztn3rqKQBq1KhBQkKCHbISERGR0k5FJxEREZF7nMFgyPfZbDYDsHv3buLi4ggPD8ds\nNtO0aVNb2z/b/Ol2bU0mk+2z1WotljxERETEuejxOhEREZF73A8//ADAgQMHePTRR/NcS01NpUaN\nGpjNZmJjY8nJySEzM/Om49xJWxEREZHCUtFJRERE5B6XmJjIyy+/zNGjR3n++efzXGvdujWnTp2i\nb9++JCUl0a5dOyZOnHjTce6krYiIiEhhGawlfP10Skpasc8xcvr6Yp9DpDSY8/Zzjg6hyLwdM97R\nIYiUCNO7hxb7HBZLhWKfQ+6cPe7B7pbFUuGejq8oOUuuBclzR3QvO0UjIuIcXhj4VaHH0J5OIiIi\nIiJS4nn941+ODqFIqJBY+jhLrs6SJzhXroWlx+tERERERERERKTIqegkIiIiIiIiIiJFTo/XiYiI\niIiI3IX1X73k6BBERIrNoICvCz2GVjqJiIiI3IOeffZZrl696ugwRERERO6aik4iIiIiIiIiIlLk\n9HidiIiIiB0lJyfz9ttvYzQaycnJYfr06YSEhHDt2jVu3LhBcHAwTZo0sbX/7bffGDduHFlZWZhM\nJkJDQ6lVqxahoaEcOnSInJwc/Pz86NGjB97e3jz77LPs2rWLNm3aYLVa2bFjB23btmX06NHs3LmT\nOXPmYDabqVixIh988AH79+8nIiICg8HAf/7zHzp16sTw4cMd+A2JiIhIaaGVTiIiIiJ2tGnTJlq3\nbk14eDjjxo0jOTkZHx8fwsPDefPNN1m0aFGe9nPmzGHgwIEsW7aM/v3789FHH3H58mW+++47Vq1a\nxcqVK8nOzgbg9OnT9O7dm3/961+Eh4fTuXNn/vWvf/HFF18AcOXKFWbMmMGKFSsoX74827dvByAh\nIYGpU6eyatUqwsPD7fuFiIiISKmllU4iIiIiduTl5cXw4cNJS0ujU6dONGzYkJCQEBYvXkxmZib3\n339/nvb79+/nxIkTfPzxx+Tk5ODm5kblypWpU6cOr732Gp07d+aFF14AoHz58tSrVw+A+++/H3d3\nd1xcXMjNzQXAzc2N8ePHk5OTQ1JSEk8//TTlypWjUaNGlC1b1r5fhIiIiJR6di06ZWVlERQURHJy\nMiaTibCwMGrXrn3Ttm+++Saurq5MnTrVniGKiIiIFKvHHnuMdevWsWPHDmbNmkXLli2pXr0606dP\n5+DBg0ybNi1Pe7PZzJw5c6hWrVqe859++imHDx8mJiaGdevWsWTJEkwmU542Li55b/XGjh3LwoUL\nqVevHiEhIbdsJyIiIlIU7Pp4XUxMDBUrViQyMpKhQ4cyc+bMm7bbsWMHv/76qz1DExEREbGLDRs2\nkJiYiLe3NyNHjiQ1NZWHH34YgM2bN5OVlZWnvaenJ5s3bwZg165dREdHc/r0aZYvX467uztjxozh\n8uXLBZo7PT2dmjVr8vvvvxMfH59vLhEREZGiZNei065du+jYsSMArVu3Zt++ffnaZGZm8vHHH/Pa\na6/ZMzQRERERu6hTpw4hISEEBAQwf/58unTpwmeffcbAgQNp0qQJKSkptj2YAIYPH05sbCz+/v7M\nnz+fJ554gmrVqrF//358fX3p168fL730UoHm7tOnD35+fgQHB/PKK6+wYMECUlJSiitVERERcXJ2\nXUt94cIF3NzcADAajRgMBjIzM3F1dbW1WbBgAX5+fpQvX96eoYmIiIjYhbu7O2vWrMlz7quvvrJ9\n7tChA4CtkFSuXDkWL16cb5zZs2fnOxcfH3/bzyNHjmTkyJG28y+++CIA3bt3v2k/ERERkcIotqLT\n6tWrWb16dZ5zBw4cyHNstVrzHJ88eZJDhw4xYsSIAt/wVKlyPy4upr9uKCLFzmKp4OgQRMTO9HMv\nIs7suS5f/HWjO2SxVCAlJa3Ix73XOEue4Dy5Okue4Fy5FlaxFZ18fHzw8fHJcy4oKIiUlBQaNmxI\nVlYWVqs1zyqn7777juTkZHr16kV6ejqXLl1i0aJFDB48+JbzpKZeK64UROQO6S9eEedjj597FbZE\nRERESia7Pl7n5eXFxo0badOmDVu2bKFly5Z5rr/88su8/PLLwB9Lu//973/ftuAkIiIiIiIiIiL3\nJrsWnbp27crOnTvx8/PD1dWVqVOnArBw4UKaN29O06ZN7RmOiIiIiIhIoS2J7enoEEREitwY302F\nHsOuRSeTyURYWFi+80OGDMl3rmXLlvlWQomIiIiIiIiISMlgdHQAIiIiIiIiIiJS+th1pZOIiIiI\nFL2oqCj27t3LpUuXOHHiBIMGDeLhhx9m9uzZuLi4UL16dcLCwoiJicnXzsfHhx9++IFZs2bh4uJC\nzZo1mTRpUp6XvYiIiIjcDRWdREREREqBX375hVWrVnHy5EnefPNNMjIy+Oyzz6hZsyYhISFER0dj\nMBjytfPx8SE0NJSlS5dSuXJlpk2bxsaNG3nuueccnZKIiIiUcCo6iYiIiJQCTzzxBCaTiRo1apCW\nlkaZMmWoWbMm8MdemXv27KFRo0b52l24cIFTp04xYsQIAK5du0aVKlUcmYqIiIiUEio6iYiIiJQC\nLi7/d1t35coVLBaL7TgrKwuDwZCvHYDZbKZatWqEh4fbJ1ARERFxGtpIXERERKSUqVSpEgaDgeTk\nZAB2796Nh4fHLdsCHDt2DIDw8HB+/vln+wQqIiIipZpWOomIiIiUQpMmTeKtt97CxcWF2rVr061b\nN9avX3/TtpMnT+bdd9+1rXrq3bu3naMVERGR0khFJxEREZESrkePHrbP5cqV49tvvwUgMjKyQO2e\neuopVq9ebYdIRURExJmo6CQiIiIiIlIIAzusKbKxLJYKpKSkFdl49ypnyROcJ1dnyROcK9fC0p5O\nIiIiIiIiIiJS5FR0EhERERERERGRIqfH60RERERERIrBO9t9HB2CiMhd++zFjYUeQyudREREREqJ\n9PR0tm/fDsDChQvZv3//HY/xyy+/0K9fv6IOTURERJyQik4iIiIipcThw4fZsWMHAEOGDKFp06YO\njkhEREScmR6vExERESkhoqKi2LZtG+fPn6dNmzZs3boVo9GIt7c3AwcOJCQkhPT0dOrUqcP+/fvp\n1KkTqamptj6zZ89m8+bNREdH5+l37tw5Ro4ciaurKw0aNHB0miIiIlJKaKWTiIiISAly9uxZ3n//\nfXbu3ElkZCQRERF8/fXXJCcnM2jQILp27Urv3r3z9YmIiCAzM5ONGzfm67d8+XK6du1KeHg41apV\nc1BmIiIiUtpopZOIiIhICdK4cWMOHjzIqVOnCAgIAODq1aucOXPmtn0MBsMt+x0/fpzOnTsD0LJl\nS77//vviT0RERERKPRWdREREREoQs9mM2WymXbt2hISE5LmWlJR0yz5//vdm/RYtWoTR+McC+Nzc\n3GKIWkRERJyRHq8TERERKWHc3d2Jj4/n+vXrWK1WQkNDuXHjBkajkezs7DvuV7duXQ4dOgRAfHy8\nvdIQERGRUk4rnURERERKmFq1ahEQEIC/vz8mkwlvb2/uu+8+GjVqxIwZM6hRo8Yd9QsICOCNN97g\nm2++4bHHHrNzNiIiIlJaqegkIiIiUkL06NHD9tnf3x9/f/881x977DG2b99+2zFu1u/BBx9k9erV\nRReoiIiICCo6iYiIiIiIFItp/+/Oi7kWSwVSUtKKIZp7i7PkCc6Tq7PkCc6Va2FpTycRERERERER\nESlyKjqJiIiIiIiIiEiR0+N1IiIiIiIiJUD/HZMdHYKIOJEvX5ha6DG00klERESkBNm2bRsrV668\n436xsbFkZmaSkpLChAkTbtnwkWL8AAAgAElEQVRuz549XLx4sTAhioiIiAB3UHS6cOECCQkJJCQk\ncOHCheKMSURERERuoW3btvTp0+eO+y1dupSsrCwsFgshISG3bPfFF1+o6CQiIiJF4i8fr/vyyy9Z\nuHAhKSkp1KhRA4CzZ89SvXp1hgwZQpcuXYo9SBERERFnkJyczNtvv43RaCQnJ4fp06cza9Yszpw5\nQ5kyZZg2bRo7duwgMTGRMWPGEBERQXR0NEajEW9vbwYOHMjcuXNJS0vjxIkT/Prrr4wdO5bU1FR+\n/PFHBg8ezOTJk3nrrbeIiopi4cKFfPPNNxiNRtq3b0/jxo3ZvHkziYmJzJ07l1q1ajn6KxEREZES\n7LZFp6CgILKzs5k6dSoNGzbMc+3nn3/m008/ZevWrUydWvjn/ERERESc3aZNm2jdujXDhg3j8OHD\nrFu3jqpVqzJz5kw2bNhAbGws9913HwBJSUls3LiRyMhIAPz8/OjcuTMA586dY9GiRWzbto1Vq1bx\n0Ucf8eGHH7Jo0SJSU1Nt8y1ZsoTt27djMpmIjIzEy8uLxx9/nODgYBWcREREpNBuW3Ty9vbG29v7\nptcaNmzIjBkz2Lx5c4Eny8rKIigoiOTkZEwmE2FhYdSuXTtPm59//pmxY8cC0KFDB4YNG1bg8UVE\nRERKMi8vL4YPH05aWhqdOnXi/PnztGrVCoBu3boBEBUVBcDBgwc5deoUAQEBAFy9epUzZ84A0KxZ\nMwBq1KhBWlraLefr1KkTAwYMoHv37jz33HPFlpeIiIg4p9vu6fRnwenKlSu8//77jB49GoBvv/2W\nS5cu5WlTEDExMVSsWJHIyEiGDh3KzJkz87UJDg5m0qRJrFmzhuPHj3P9+vUCjy8iIiJSkj322GOs\nW7eOp556ilmzZrFt2zZyc3Nv2tZsNtOuXTvCw8MJDw8nOjqa5s2bA+DiUrAXFL/33ntMnDiRlJQU\n+vXrR3Z2dpHlIiIiIlKgjcTHjx9PzZo1OX36NACZmZmMGTPmjifbtWsXHTt2BKB169bs27cvz/UL\nFy5w7do13N3dMRqNzJo1i7Jly97xPCIiIiK3k5iYyJQpUxgzZgzvvPOO7Y+jbdiwgcTERLy9vRk5\nciQGg4G4uDgAtmzZwieffGJr6+7uTnx8PNevX8dqtRIaGsqNGzduObbBYCAnJ8d2nJaWxrx586hX\nrx7Dhw+nUqVKpKen52snIiIicrcKVHS6dOkSAQEBmM1mADp37nzbm5pbuXDhAm5ubn9MbDRiMBjI\nzMy0XT9z5gyVKlUiKCgIX19fli5desdziIiIiPyVN954g4oVK/L000/TqlUr2x9Hq1OnDiEhIQQE\nBDB//nzmz5/P9evX6du3L8uWLePFF1+0ta1VqxYBAQH4+/vTq1cvLBaLbb+nm2nRogV9+vSx7elU\noUIFUlNT6dmzJwEBAXh6elK5cmVatGhBYGAgiYmJxZ6viIiIlG4FW3vNH/sxGQwG4P9WJN3O6tWr\nWb16dZ5zBw4cyHNstVrzHZ8+fZr58+dz33330bt3b7y8vKhfv/4t56lS5X5cXEwFTUNEipHFUsHR\nIYiInZXUn/uqVasyfPhwR4eRj7u7O2vWrMlzbtq0aXmOe/ToYfvs7++Pv79/nusjRoywfX7ssccI\nDw8HICwszHb+z32hgoOD88UwfPjwe/K7ERERkZKnQEUnf39/evbsSUpKCkOHDuXgwYOMGzfutn18\nfHzw8fHJcy4oKIiUlBQaNmxIVlYWVqsVV1dX2/UHHniA+vXrU6VKFQCefPJJEhMTb1t0Sk29ffFL\nROwnJeXWm9WKSOlkj5/74ihstW3blu3bt9OiRYs8+x8ZjQVaBC4i4hDLvG7/b7CSxGKp4DT3js6S\nq7PkCc6Va2EVqOjUtWtXmjVrxv79+3F1dSUkJIRq1ard8WReXl5s3LiRNm3asGXLFlq2bJnneu3a\ntbl69SqXL1+mYsWKHDlyhN69e9/xPCIiIiK38/HHH5Oenp7nnMFg4MiRIw6KSERERKT0KVDR6Y03\n3uCDDz6gS5cuhZqsa9eu7Ny5Ez8/P1xdXZk6dSoACxcupHnz5jRt2pR3332XwYMHYzAYaNOmDQ0b\nNizUnCIiIiL/64cffnB0CCIiIiKlXoGKTg899BBr1qyhadOmeR6Hq1279h1NZjKZ8uwn8KchQ4bY\nPnt6eubbC0pERESkKF29epWlS5dy8OBBDAYDTZs2JSAg4LYbcYuI3Ate/n6xo0MQESexoccbhR6j\nQEWnL7/8Mt85g8FAbGxsoQMQERERsbfg4GCqV6+Or68vVquVnTt3Mn78eGbMmOHo0P5SVFQUiYmJ\njBkz5i/b/vzzz5QpU4a6desyatQowsLCVFgTERERuylQ0enbb78t7jhERERE7ObChQvMmjXLdty+\nfXv69evnwIiKxzfffIOHhwd169Zl9uzZjg5HREREnEyBik7nz5/ngw8+sC1Bf+KJJ3jjjTdwc3Mr\n7vhEREREitz169e5fv06ZcuWBeDatWtkZGTYZe6oqCj27NlDamoqiYmJjBo1ipiYGI4fP86MGTP4\n8ssvSUhIICMjAz8/P3x8fAgKCsJsNnP58mXat29vG2vmzJmULVuWV199leDgYJKSksjOziYwMBA3\nNzdWrVqFm5sbDzzwAG+88QbR0dFMmjSJatWqcfjwYZKTk5kxYwbu7u6Ehoayb98+6tevz4kTJ5g1\naxYPPfSQXb4TERERKZ0KVHSaMGECbdq0YcCAAbYl6GPHjuWTTz4p7vhEREREilzv3r3p0qULHh4e\nWK1WfvrpJ0aOHGm3+U+ePMnKlStZvXo1CxYsYO3atURFRfHFF1/wt7/9jXfffZcbN27g7e2Nj48P\nAJUqVWLSpElERUUB8NVXX3H27FlmzJjB2rVrsVgsTJkyhUuXLtG/f3+io6Np06YNnTp1okmTJnnm\nz8zMZPHixURGRrJ27VpcXFzYu3cvX3zxBYmJibz44ot2+y5ERESk9CpQ0en69ev4+/vbjh977DE9\nciciIiIlVs+ePfHy8uLw4cMYDAYmTJhA9erV7Ta/h4cHBoMBi8VCgwYNMJlMVK1alaysLK5cuYKv\nry9ms5nU1FRbn/8uHCUmJvL111/b9t3cv38/e/fuZd++fQBkZGSQmZl5y/mfeuopAGrUqEFCQgLH\njx/H09MTo9FIgwYNePDBB4sjbREREXEyBS46nT9/nmrVqgFw7ty5297IiIiIiNyLtm7dyjPPPMOa\nNWvynP/++++BP4pR9uDi4nLTz6dPn+bXX38lPDwcs9lM06ZNbdfMZrPt85kzZ6hfvz4bN27k+eef\nx2w2M3ToULp3716g+U0mk+2z1WoFwGg02s4ZDIY7T0pERETkfxSo6PT666/To0cPLBYLVquVS5cu\nMXny5OKOTURERKRIHT16lGeeeYa9e/fe9Lq9ik63cujQIZ599lnMZjOxsbHk5OTc9Bd97dq1Y/Dg\nwfj5+eHl5YWnpyexsbF0796dixcvsmzZMt58800MBgM5OTl/OW/t2rVZtmwZVquV//znPyQnJxdH\neiIiIuJkClR0ateuHZs3b+bkyZMA1K1blzJlyhRnXCIiIiJFbsiQIQD8v//3/+jWrVuea5GRkY4I\nKY/WrVtz6tQp+vbti7e3N+3atWPixIk3bevm5kZgYCATJ07kgw8+IC4uDl9fX3Jychg+fDjwx2N0\noaGhlCtX7rbzNm7cmDp16uDj40OjRo2oV69entVQIiIiInfDYP1zTfVt/PDDD6xevZr3338fgAED\nBvD666/TvHnzYg/wr6SkpBX7HCOnry/2OURKgzlvP+foEIrM2zHjHR2CSIkwvXtosc9hsVQosrGO\nHDnCoUOHWLJkCYMGDbKdz87OZt68eWzfvr3I5ipJMjMz+fLLL3nhhRe4du0aXbp0ITY2Ns+jf//L\nHvdgd8tiqXBPx1eUnCVXZ8kT/jrXl79fbMdoRMSZbejxRqHHKNBKp5kzZzJ16lTb8aRJk3j77bfv\nid8IioiIiBSUq6srFy9eJC0tLc8jdgaDgXfeeceBkTmWq6srBw8eZPny5RiNRkaOHHnbgpOIOM7S\nNoP+ulEJoEJi6eMseYJz5VpYBbqbsFqtPPLII7bjhx56KM9mkyIiIiIlQb169ahXrx5PP/00Tzzx\nRJ5rmzZtclBU94bg4GBHhyAiIiKlTIGKTrVq1WL69Om0aNECq9XK999/T40aNYo7NhEREZFiUa1a\nNaZNm0Zqairwx+Nl8fHxdOrUycGRiYiIiJQeBSo6hYWFsXjxYtvjdM2aNWP06NHFGpiIiIhIcXnn\nnXdo27YtW7ZsoW/fvsTGxjJt2jRHhyUiUuQGbFvj6BBEpISKeWlAocco0DNyZcqUISAggE8++YRJ\nkybRrFkzvb1ORERESiyTycSQIUOoWrUq/v7+fPzxx0RERDg6rGKTnJxMQkICAJMnTyYpKemm7dLT\n0512M3UREREpegUqOk2aNImvvvqKy5cv06dPH1asWHHL1/eKiIiI3OsyMjI4d+4cBoOBpKQkXFxc\nOHPmjKPDKjZxcXG2otO4ceOoXbv2TdsdPnyYHTt22DM0ERERKcUK9HjdTz/9RHBwMJGRkbzwwgsM\nGzaM/v37F3dsIiIiIsXilVdeYefOnQwaNIjnn38ek8lE9+7dHR3WXUtPT+ett97i2rVr3Lhxg+Dg\nYEaPHk3btm2pXLkyUVFRuLi4ULNmTZYuXUpwcDDZ2dm89957uLq64urqyuzZswkJCSE9PZ06derQ\nu3dvR6clIiIiJVyB314H8N133/HGG28Af2y4KSIiIlISeXt72z7v3r2bq1evUqlSJQdGVDgpKSn4\n+Pjg7e3Nrl27WLRoEdnZ2bRt25a2bdtitVqpUqUKHTp0YOnSpQBERUXh5+fHCy+8wK5du0hJSWHQ\noEEkJiaq4CQiIiJFokBFp7p169KtWzeqVKnC448/ztq1a0v0jZmIiIg4t2PHjjFnzhyOHz+OwWCg\nQYMGjBgxgrp16zo6tLtStWpVPvroIxYvXkxmZib3338/AE2aNLllnw4dOjBx4kROnjxJ165dqVev\nHgcOHLBXyCIiIuIEbrunU1ZWFgChoaHMmDGDJUuWAPC3v/3N9oaXP9uIiIiIlBRBQUE888wzzJs3\njw8//JCnn36aMWPGODqsu7Zs2TKqV69OZGRknn03zWbzLfu0atWKNWvW8OijjxIUFERcXJwdIhUR\nERFnctui0yuvvMKJEycwmUw8/vjjuLq6AuDh4UHFihU5fvw4r7zyil0CFRERESkqZcuWpWfPnjz6\n6KPUq1ePXr16UaFCBUeHdddSU1N5+OGHAdi8eXO+XwoaDAays7PznFuxYgWXL1/mueeeo3///hw5\ncgSj0ZivnYiIiMjduu3jdePHj+fNN9+kRo0atGnThpo1awJw9uxZvv/+e3777Tfef/99uwQqIiIi\nUlSefvppNm/ejJeXF7m5ucTFxdG0aVOsVitWqxWjsUAv+L1nPP/884wZM4aNGzfi7+9PTEyMbU9O\ngKZNmzJmzBjc3Nxs5x5++GFGjhxJhQoVcHV1JSwsjEuXLjFjxgxq1KjBoEGDHJGKiIiIlCIG63/f\nkdyE1WolNjaWbdu2ce7cOQBq1KhB27Zt6dChAwaDwS6B3kpKSlqxzzFy+vpin0OkNJjz9nOODqHI\nvB0z3tEhiJQI07uHFvscFkvRr0Byd3cnJyfnptcMBgNHjhwp8jlLG3vcg90ti6XCPR1fUXKWXJ0l\nTyj6XAdsW1NkY4mIc4l5aUChx/jLjcQNBgPe3t553vIiIiIiUpIdPnzY0SGIiNjFZ217OjqEm1Ih\nsfRxljzBuXItrAK9vS4mJoZPP/2UK1eu5Fmq/d133xVXXCIiIiLFZs6cOTc9P3LkSDtHIiIiIlJ6\nFajoNHfuXEJDQ6lVq1ZxxyMiIiJS7Ewmk+1zVlYWe/bsoVGjRg6MSERERKT0KVDR6ZFHHqF58+bF\nHYuIiIiIXQwfPjzPcU5ODiNGjHBQNCIiJdeArRsdHYKIFJOYnj6FHqNARaemTZsya9YsWrRokec3\ng61atSp0ACIiIiKOlp2dza+//uroMIrMnj17ePTRR3nggQd49tlniY6Oply5co4OS0RERJxMgYpO\nO3fuBGD//v22cwaDQUUnERERKZGeeeaZPG/gvXLlCi+++KIDIypaX3zxBQMHDuSBBx5wdCgiIiLi\nxApUdAoPD893btOmTUUejIiIiIg9rFy50vbZYDBQvnx5Klas6MCI7l5WVhYTJkwgKSmJzMxMRowY\nwebNm0lMTGTu3LkAREREsHXrVnJycvj0008pW7YswcHBJCUlkZ2dTWBgIK1ataJfv37Ur18fgAkT\nJjgyLRERESkFClR0Sk5OZsWKFaSmpgKQmZlJfHw8nTp1uqPJsrKyCAoKIjk5GZPJRFhYGLVr187T\nZvbs2cTHx2O1WvH29mbw4MF3NIeIiIjIX7l+/Trr1q3jrbfeAuDdd99l4MCBtoJLSbJhwwZcXV1Z\nsWIFv/32GwEBATz++OMEBwfbXgJTv359hgwZwptvvklcXBzp6elYLBamTJnCpUuX6N+/P9HR0ba2\nfn5+jkxJRERESgljQRq98847VK5cmR9//BEPDw9SU1OZNm3aHU8WExNDxYoViYyMZOjQocycOTPP\n9V9++YX4+HhWrVpFZGQkUVFRpKSk3PE8IiIiIrfz3nvv8cwzz9iOX3rpJUJCQhwY0d07dOgQLVu2\nBKB69eq4urpy+fLlPG2efPJJ2/W0tDT2799PbGws/fr1Y+TIkWRkZJCZmQlAkyZN7JuAiIiIlFoF\nKjqZTCaGDBlC1apV8ff35+OPPyYiIuKOJ9u1axcdO3YEoHXr1uzbty/P9QoVKthuejIyMjAajZQt\nW/aO5xERERG5nZycHJ566inb8VNPPYXVanVgRIXz37FnZmZiNOa9xfvvF8FYrVbMZjNDhw4lPDyc\n8PBwvv76a1xdXQEwm832CVpERERKvQIVnTIyMjh37hwGg4GkpCRcXFw4c+bMHU924cIF3Nzc/pjY\naMRgMNh+qwZQs2ZNOnfuTPv27Wnfvj2+vr6UL1/+jucRERERuZ0KFSqwcuVKjh8/TmJiIkuWLCmx\nb3dr3Lgx8fHxAJw9exaj0UilSpXIycm5ZR9PT09iY2MBuHjxIrNmzbJLrCIiIuJcCrSn0yuvvMLO\nnTsZNGgQzz//PCaTie7du9+2z+rVq1m9enWecwcOHMhz/L+/UUxKSuKbb75h8+bNZGdn4+vrS9eu\nXW/75pUqVe7HxcV0y+siYj8WSwVHhyAidlZSf+7DwsKYOXMmkZGRADRr1oywsDAHR3V3unXrxu7d\nu+nXrx9ZWVmEhIQQHx9PYGAgH3300U37dOnShbi4OHx9fcnJyWH48OF2jlpEREScQYGKTt7e3rbP\nu3fv5urVq1SqVOm2fXx8fPDx8clzLigoiJSUFBo2bEhWVhZWq9W2lBvg4MGDeHp62h6pa9CgAb/8\n8gutWrW65TypqdcKkoKI2EFKSpqjQxARO7PHz31xFLbc3NwYPHgwderUAeCnn36yrcYuaVxcXJg8\neXKec82bN7cVkr799lvb+TFjxtg+/28fuPkbi0VERETuVoGKTmfOnOH9998nNTXV9tx/8+bNbTdq\nBeXl5cXGjRtp06YNW7ZssW16+aeHH36YZcuWkZubS05ODr/88ku+t9uJiIiIFNbs2bM5f/68bXXT\nwoULeeihhxg9erSDIxMRKVk+e6bzXfWzWCo4zS8snSVXZ8kTnCvXwirQnk7BwcE8//zztsfh6tSp\nQ3Bw8B1P1rVrV3Jzc/Hz8yMiIsL2muKFCxeyf/9+PDw88PLyok+fPvTr14+ePXvy0EMP3fE8IiIi\nIrcTHx+f53G6Dz74gL179zowIhEREZHSp0ArnbKysujQoQNLly4F/liyfTdMJtNN90sYMmSI7XNg\nYCCBgYF3Nb6IiIhIQWRlZZGZmWl7zP/q1atkZ2c7OCoRERGR0qVARSeA33//HYPBAEBiYiIZGRnF\nFpSIiIhIcfL19aVbt264u7uTm5vLwYMHtZm2iMhdeGXrTkeHICLFZF3PToUeo0CP1w0bNoxevXpx\n+PBh/vGPfzBgwABGjRpV6MlFREREHKFHjx4MGzaM48ePc/LkSUaOHMlLL71UZONv3LixwG3T09PZ\nvn37La+npKQwYcKEogirQGJjY8nMzLTbfCIiIlJ6FajoVLduXV588UUGDBjAI488wgsvvKB9D0RE\nRKTECg0NZcuWLfj6+tKrVy+++eabm77N7W5kZmbatiQoiMOHD7Njx45bXrdYLISEhBRBZAWzdOlS\nsrKy7DafiIiIlF4Ferxu8ODBuLu7U716df72t78BaN8DERERKbESExNZsWKF7bhv37706dPntn1y\ncnIIDg4mKSmJ7OxsAgMD+eijj2jdujVxcXGkpqbyySefsGjRIo4ePcrEiRMJDg7O16dVq1b069eP\n+vXrA39sap6enk6dOnXw9PTkvffew8XFBaPRyJw5c0hPTycwMJCoqCg6duxI79692bJlC5mZmXz2\n2Wd8/fXX7Nmzh9TUVBITExk1ahQxMTEcP36cGTNm4OnpSUREBNHR0RiNRry9vRk4cCBz584lLS2N\nEydO8OuvvzJ27FhSU1P58ccfGTx4MEuXLrXteSUiIiJyNwpUdKpcufJNNwAXERERKYmysrLIzc3F\naPxj0XdOTg45OTm37RMdHY3FYmHKlClcunSJ/v37U7lyZcqXL8+yZcuYMWMGX3/9NYMGDeLAgQNM\nnDiRtWvX5usTHR0NQP369fHz8yMqKorExER69+7Njh07CA4OplGjRsyZM4fo6Gjat29viyEnJ4dH\nH32UV155hVGjRhEXFwfAyZMnWblyJatXr2bBggWsXbuWqKgoYmJicHNzY+PGjURGRgLg5+dH585/\nvOL83LlzLFq0iG3btrFq1So++ugjPvzwQxYtWqSCk4iIiBRagYpOHTt2ZP369TRt2hSTyWQ7X6tW\nrWILTERERKS4PPPMM/Ts2dP2Rt74+Hi6du162z779+9n79697Nu3D4CMjAyysrJ46qmnAKhRowaX\nL1/+yz5/7pfUpEmTfHM88MADzJgxgxs3bnD+/Hn+8Y9/5Gvz3/OlpaUB4OHhgcFgwGKx0KBBA0wm\nE1WrVmXfvn0cPHiQU6dOERAQAPzxpr4zZ84A0KxZs3xjiYiIiBSVAhWdjh49SnR0NJUrV7adMxgM\nfPfdd8UVl4iIiEixef3112ndujUHDhzAYDAQEhJy0yLQfzObzQwdOpTu3bvbzvXr1y/PL+SsVutf\n9vnva/9r8uTJDB48mLZt27J48WKuXbuWr83N5nNx+b9buv/+bLVaMZvNtGvXLt++UHFxcXnaioiI\niBS1Am0kfuDAAfbs2cPWrVttf1RwEhERkZLsiSeeoH///gQEBPxlwQnA09OT2NhYAC5evMisWbNu\n2s5oNNoe1StIH6PRaNsr8/Llyzz88MNkZmaydevWItnQ293dnfj4eK5fv47VaiU0NJQbN27csr3B\nYPjLRw1FRERECqJARScPDw8yMjKKOxYRERGRe1aXLl24//778fX1ZejQoTz55JM3bWexWMjKyiIw\nMLBAfRo1asRXX33F4sWL6du3L8OGDSMwMJB+/frx73//m/T09ELFXatWLQICAvD396dXr15YLBbu\nu+++W7Zv0aIFffr04dKlS4WaV0RERMRg/d914DcxcOBADh48SL169fIs6Y6IiCjW4AoiJaX49x8Y\nOX19sc8hUhrMefs5R4dQZN6OGe/oEERKhOndQ4t9DoulQrHPIXfOHvdgd8tiqXBPx1eUnCVXZ8kT\nSl6ur2zd6egQRKSYrOvZqdBjFOhB/qFDhxZ6IhERERERESldPn2m9V31K2nFtcJwllydJU9wrlwL\nq0BFpxYtWhR3HCIiIiIiIiIiUooUaE8nERERERERERGRO6H35IqI3IQ99qkRERERKY2GbvvZ0SGI\nSBH44qXmhR5DK51ERERESohNmzYVuG1gYCDx8fHFOoeIiIjI7ajoJCIiIlICnD59mg0bNpT4Of5/\ne3ceF2W99nH8M2yKigk6gutjmaWiEuaSlbtWFlmpGJioRYuVQmmmiaThUqmZC2ZpqISWRlKPppbJ\nwacsxRLFpU7uiaJGAQKaIsM8f/RyThxBUYYZGL7vf87Mff+W65rpPq/bi9/9GxEREak69HidiIiI\nSAWUnp7OuHHjcHJywmQy4ezszMGDB4mOjsZsNuPp6cnQoUM5cOAAU6dOJS4ujiVLlrB+/XoaNmxI\nXl4eAHl5eUycOJGzZ89iMpmYNGkSLVu2pG/fvjz++OMkJSWRn5/PsmXLiIqKYs+ePURHRzNq1Cg7\nfwIiIiJS2Wmlk4iIiEgF9PXXX3P33XcTFxdHREQEXbt2pVOnTiUWg3Jycvjkk09YvXo1M2fO5ODB\ngwDExsbStWtXYmNjmTJlCm+//TYAJpOJW265hZUrV9K4cWO2b99OaGjoVecQERERuR5a6SQiIiJS\nAd1zzz2MGjWK3Nxc7r//fvz8/Ni3b1+J7X/77TduvfVWqlWrRrVq1fD19QVg165dZGZmsnbtWgD+\n+usvS58OHToA4OPjQ25uLh4eHuWYkYiIiFQ1KjqJiIiIVEC33XYb//u//8v333/PnDlzGDhwoOWc\nwWCwvC4oKADAbDbj5PSfRexmsxkAV1dXIiMj8ff3v2IOZ2fnK9qLiIiIWIserxMRERGpgNavX8/B\ngwfp06cP4eHhJCQkWApMtWrVIiMjA4CdO3cC0LRpUw4fPkx+fj55eXmWVVF+fn5s3rwZgEOHDrFs\n2bIS53RycrLMISIiIlJWKjqJiIiIVEDNmjUjKiqKYcOGsXDhQsLCwvj555+ZMWMGffv2JTExkSef\nfJKcnBwA6tSpw6OPPoAx6oQAACAASURBVEpQUBATJ06kbdu2AAwdOpTjx48zZMgQJk2aZHmkrjjN\nmze3zCEiIiJSVgZzJV9LnZGRW+5zhM9aW+5ziDiCeeP62zsEEXFARqP2GaqIbHEPdqOMRo8KHZ81\nVZVcq0qe4Bi5jvz23/YOQUSsYM3AjmUeQ3s6iYiIiIiIiNW8363lNds4QnGttKpKrlUlT6hauZaV\nHq8TERERERERERGr00onERERERERqVSmbs22dwjXoTLFWhZVJU+oKrnOf6zsWxxopZOIiIiIiIiI\niFidik4iIiIilURCQgJvv/22vcMQERERKRUVnURERERERERExOpsXnTasWMHXbp0ISkpqdjza9eu\nZeDAgQQGBhIfH2/j6EREREQqvjfffJPg4GAGDBhAfHw8v/32G08//TQAKSkpdOjQgcLCQgoKCggI\nCCAwMJDjx48DcPr0aQYMGIDJZGLixImEhIQQHBzMtm3bAAgJCSEqKoqoqCi75SciIiKOwaYbiR8/\nfpxly5bRvn37Ys+fP3+ehQsX8tlnn+Hq6sqgQYPo27cvderUsWWYIiIiIhVao0aNeO2117hw4QJ9\n+vRh69atnDlzBrPZTEpKCq1ateLgwYPk5+fTtm1bfH192bBhAyNHjiQxMZGHHnqIdevWYTQamTFj\nBpmZmQwfPpx169YB0KJFC4KDg+2cpYiIiFR2Nl3pZDQaiY6OxsOj+B3QU1NTadu2LR4eHlSvXp32\n7duTkpJiyxBFREREKryzZ88SFBTEM888Q1ZWFgC33XYbR48eZc+ePQwZMoTdu3eTkpJC586deeih\nh9i0aRMAW7ZsISAggF27dpGYmEhISAjh4eFcvHiR/Px8ANq1a2e33ERERMRx2HSlk7u7+1XP//HH\nH3h5eVnee3l5kZGRcdU+np41cHFxtkp8IlI2RmPZf1JTRESubt++fRQWFhIXF4erqyv+/v4AdOrU\nidTUVC5cuEDnzp2ZNWsW58+fZ8KECXh6euLj48OePXsoLCzE29sbV1dXRo4cSUBAwBVzuLq62jot\nERERcUDlVnSKj4+/Yk+m0aNH07Vr11KPYTabr9kmK+v8dccmIuUjIyPX3iGIiANSQbuokydP4u/v\nj6urK4mJiZhMJvLz8+nYsSNRUVHceuuteHl5kZWVxblz52jQoAEAjzzyCFFRUTz++OMA+Pn5kZiY\nSEBAAH/++SexsbGMGTPGnqmJiIiIgym3olNgYCCBgYHX1ad+/fr88ccflve///47d9xxh7VDExER\nEam0+vTpQ0pKCkOHDqVPnz706NGDKVOmMGPGDA4dOmS5/6pduzb16tWz9OvZsyeRkZHcf//9APTr\n14/t27cTFBSEyWRi1KhRdslHREREHJdNH6+7Fj8/PyZNmkROTg7Ozs6kpKQwceJEe4clIiIiUiEM\nGDDgimMjRoywvN66davl9cyZM4u0S0lJoWfPntSuXRsAFxcXpk+ffsV4cXFxVopWREREqjqbFp22\nbNlCTEwMR44cYf/+/cTFxbF06VIWL15Mx44d8ff3Z+zYsYSGhmIwGHjxxRdL3HRcREREREpn/vz5\nbN26lQULFtg7FBERq4i8t/L8wrnR6FEltqGoKnlC1cq1rAzm0mycVIHZ4osOn7W23OcQcQTzxvW3\ndwgi4oC0p1PFVJFvtqvSPwaqSq5VJU+oOrlWlTyh6uRaVfKEqpOrNe7BKtTjdSIiIiIiIiLXY8N3\nF+wdwjVU9PisparkCVUl1+EDyl50crJCHCIiIiIiIiIiIkWo6CQiIiJSga1YscJqezF99dVXVhlH\nREREpDRUdBIRERGpAvLz81m+fLm9wxAREZEqRHs6iYiIiNiRyWQiMjKStLQ0CgoKCAsLA2DGjBnU\nq1cPo9FIkyZNSE5OZuXKlcyfPx+Azp07k5yczM8//8wbb7yBwWDA39+f8ePH88MPPzBv3jxcXV2p\nXbs2c+fO5c033+TXX39lypQpRERE8Prrr5OWlkZ+fj5hYWHce++93HfffXTr1o26devy/PPP2/Nj\nEREREQegopOIiIiIHa1btw6j0ciMGTPIzMxk+PDhVKtWjVmzZtGyZUueeeYZmjRpUmL/adOm8cYb\nb9CyZUteffVVTp48ydmzZ5k9ezZNmjTh1VdfZevWrYSGhpKamsqUKVP44osvcHNzY8WKFZw5c4Zh\nw4bx9ddfU1BQQLdu3ejWrZsNPwERERFxVCo6iYiIiNjRrl272LlzJykpKQBcvHiRM2fO0LJlSwA6\nduzIxYsXS+x/9OhRS9uZM2cCcOLECSZNmoTJZCItLY277rqrSJ99+/bRuXNnALy9vXFzcyM7OxuA\ndu3aWTdBERERqbJUdBIRERGxI1dXV0aOHElAQIDl2D333GN5bTabATAYDEX6FRQUAODkdOUWnRMn\nTmTx4sU0b96cqKioYue9PC78vd/T5XFcXV1vMBMRERGRorSRuIiIiIgd+fn5kZiYCMCff/7JnDlz\n8Pb25siRI5jNZnbs2AFArVq1+P333wH497//zblz5wBo3rw5qampwN/FpsOHD5OXl0eDBg3Iyckh\nOTmZS5cu4eTkhMlkAqBt27YkJycDcOrUKZycnKhdu7ZN8xYRERHHp5VOIiIiInbUr18/tm/fTlBQ\nECaTiVGjRtGhQwfCw8Np2LAhPj4+ALRs2ZIaNWoQFBSEv78/jRo1AiAiIoIpU6YAcMcdd9C8eXOG\nDBlCcHAwzZo14+mnn2bBggV069aNS5cuERYWxpw5c9ixYwchISFcunSpxNVQIiIiImVhMP9zbXUl\nlJGRW+5zhM9aW+5ziDiCeeP62zsEEXFARqOHvUOQYtjiHuxGGY0eFTo+a6oquVaVPKHq5GrNPDd8\nd8Eq44hIUcMHGMs8hlY6iYiIiIiISKX1YNfq9g7hqlRIdDxVKdey0p5OIiIiIiIiIiJidVrpJCIi\nIiIiIpXanm8q8iN2FTk2a6oqeUJVybX3kLJvcaCVTiIiIiIiIiIiYnUqOomIiIg4uJdffpkLFy6Q\nnp7Onj177B2OiIiIVBEqOomIiIg4uHfffZfq1auzfft2FZ1ERETEZlR0EhEREamkHnjgAUwmEwUF\nBfj7+7N3714AQkND6dWrF6GhoSQlJdGrVy9OnTpFdHQ0H330EYmJiRw6dIhhw4YxfPhwXnjhBXJy\ncjhx4gTBwcGWfiIiIiJloY3ERURERCopX19fDh48SH5+Pm3atGH37t34+vqSmppKQUEBa9aswdPT\nk6lTp1K7dm0ee+wxPD096d27N8OHDycqKopmzZqxcuVKVq5cycMPP8wvv/xCUlISnp6e9k5PRERE\nKjkVnUREREQqqU6dOrF7924uXLhASEgImzZtomPHjrRu3ZqsrKyrFo727NlDZGQkAPn5+bRt2xaA\nJk2aqOAkIiIiVqGik4iIiEgl1alTJxYvXsyFCxcYNGgQCQkJ7Ny5k86dO5OYmHjVvu7u7nz00UcY\nDAbLsRMnTuDq6lreYYuIiEgVoT2dRERERCqpm2++mVOnTpGbm0utWrWoV68eiYmJdO7cudj2BoOB\ngoICAFq2bMm3334LwPr169m2bZvN4hYREZGqQUUnERERkUqsbt26NGzYEAA/Pz9OnjyJj49PsW39\n/f358MMPWbt2LREREXzwwQcMHTqUhIQEWrVqZcuwRUREpAowmM1ms72DKIuMjNxynyN81tpyn0PE\nEcwb19/eIYiIAzIaPewdghTDFvdgN8po9KjQ8VlTVcm1quQJVSdXa+e555sLVhtLRP7We4ixzGNo\nTycRERERERGp1Nr1rW7vEEqkQqLjqUq5lpUerxMREREREREREavTSicRERERERGpUs4knLfdXNhu\nLnuqKnlC1cnV+FzZtzjQSicREREREREREbE6FZ1EREREysHbb79NQkLCNdtlZGTw+uuvl3j+22+/\n5eOPPy7x/FdffVXqmPLy8ti6dWup24uIiIiUhc2LTjt27KBLly4kJSUVe37Dhg0MGjSIwYMH8+67\n79o4OhERERHbMhqNREVFlXi+W7duDBkypMTzixcvLvVc+/fv5/vvv7+u+ERERERulE33dDp+/DjL\nli2jffv2xZ7/66+/mD17NmvXrqVmzZoMHjyYhx9+mFtvvdWWYYqIiIhcU15eHmPHjuX8+fNcuHCB\nyMhIjh49yocffoi3tzfVq1enRYsWJCQk8OOPP5KVlcXBgwd5+eWX+fLLLzl8+DCzZ8+mbt26hIWF\nkZCQQN++fXn88cdJSkoiPz+fZcuWsWnTJg4ePMiYMWMYN24cGRkZ5OfnM3r0aA4cOMCvv/7KqFGj\nCAkJYenSpZw/f57x48ezY8cOvv76awoLC+nevTujRo0iKiqKvLw8mjVrRvfu3Zk4cSKXLl3CYDAw\nffp0DAYD48aNo0aNGgwdOpSePXva+2MWERGRSsymK52MRiPR0dF4eBS/GZW7uztr166lVq1aGAwG\n6tSpQ3Z2ti1DFBERESmVjIwMAgMDiYuLY8yYMSxZsoR3332X5cuXs2jRIn777TdL22PHjrFo0SKe\ne+45PvjgAxYuXMizzz7Ll19+WWRMk8nELbfcwsqVK2ncuDHbt2+3nDtw4ABZWVmsXLmSmJgYzp49\ny9NPP02tWrWIjo62tImJiaFNmzYAfPzxx3z66ackJCSQl5dHaGgoDz74II8//jjz5s1j0KBBxMXF\nMWTIEMsYv/zyC7Nnz1bBSURERMrMpiud3N3dr9mmVq1aAPz666+cPHkSPz+/q7b39KyBi4uzVeIT\nkbIxGsv+6wYiIpVFvXr1eO+994iJiSE/P5+LFy9Ss2ZN6tatC1BkZXebNm0wGAwYjUZuv/12nJ2d\nqVevHikpKVeM26FDBwB8fHzIzc21HL/llls4d+4c48aNo2/fvjz00ENX9L399ttxc3MDoHr16gwd\nOhQXFxeysrKu+EPevn37GDt2LACdO3dm4cKFADRp0gRPT8+yfDQiIiIiQDkWneLj44mPjy9ybPTo\n0XTt2vWafY8dO8Yrr7zCO++8g6ur61XbZmVVjZ8qFKkMMjJyr91IROQ6VdSCdmxsLN7e3syaNYu9\ne/cyYcIEnJz+s4jcbDZbXru4uBT7+p9tLnN2di72vLu7O59++ikpKSl8/vnnJCUl8eabbxbpe7ng\ndPLkSZYvX87nn39OzZo1CQgIuGIeg8FgGf/SpUuW2K917yUiIiJSWuVWdAoMDCQwMPC6+50+fZoX\nX3yRmTNn0qpVq3KITERERKTssrKyuP322wHYvHkzHh4enD59mpycHNzd3UlJSeGOO+6w2nz79+/n\n0KFDPPLII/j5+fHEE08AxReusrKy8PLyombNmuzfv5+TJ09aCksFBQUAtG3bluTkZAICAvjxxx8t\nj+SJiIiIWIvNf73uWiIiIpgyZQq+vr72DkVERESkRI888gjLli3jqaeeol27dmRkZPDiiy8ydOhQ\nwsLCaNGihVXna9y4MWvXrmXIkCE89dRThIaGAtCqVSsGDRpUpG2rVq2oWbMmQUFBbNiwgaCgIN54\n4w1at27Nxo0biYmJISwsjC+++IJhw4aRkJBAWFiYVeMVERERMZiL+/NYOdmyZQsxMTEcOXIELy8v\njEYjS5cuZfHixXTs2JE6derw6KOP0q5dO0ufESNG0Lt37xLHtMXjPOGz1pb7HCKOYN64/vYOQUQc\nUEV9vK6qq8iPVBuNHhU6PmuqKrlWlTyh6uRq7zzPJGibFpFrafOcd5nHsOlG4j169KBHjx5XHH/2\n2Wctr1NTU20YkYiIiIiIiFQ13gNq2GwuexfYbKWq5AlVK9eyqnCP14mIiIiIiIiISOVn05VOIiIi\nIiIiIlJ2puWZ9g6hiNNUrHjKU5XJdVzZtzjQSicREREREREREbE6FZ1ERERERERERMTqVHQSERER\nERERERGr055OIiIiIlaQkJDAd999R15eHqdPn2bEiBF88MEHdOvWjbp16/LYY48xceJELl26hMFg\nYPr06TRp0oRp06aRkpJCixYtOHr0KHPmzCE6OhpXV1eys7N58803GTt2LOfPn+fChQtERkbSrl07\n+vTpw+DBg/nqq6/4n//5H3x9fS2v33nnHSZMmICXlxf79+8nMzOTZ555hoSEBLKyslixYgU1atQg\nMjKStLQ0CgoKCAsLo0uXLoSEhNCiRQsAXn/9dTt/qiIiIlKZaaWTiIiIiJUcOnSIRYsWERsby9y5\nc8nPz6dbt248//zzzJs3j0GDBhEXF8eQIUOIjo7m119/ZefOnXz22Wc89dRT7Nu3zzLWTTfdxIIF\nC8jIyCAwMJC4uDjGjBnDkiVLACgsLKR169asWbOGlJQUGjVqxGeffcbOnTvJyckBwMXFhdjYWG67\n7TZ27drF8uXLue2220hOTmbdunUYjUbi4uJYuHAhM2bMsMzdokULFZxERESkzLTSSURERMRKOnbs\niIuLC15eXtx0002kpaXRrl07APbt28fYsWMB6Ny5MwsXLuTw4cP4+fnh5OTE7bffTqNGjSxjXe5X\nr1493nvvPWJiYsjPz6dGjRpF2hgMBurWrUvr1q0B8PLyIjc3t8gY9evX55ZbbrGMl5uby+7du9m5\ncycpKSkAXLx4kfz8/CL9RERERMpCRScRERERKyksLLS8NpvNGAwGXF1dATAYDJjNZgAuXbqEk9Pf\nC84v/+/lNpdd7hcbG4u3tzezZs1i7969zJw509LG2dm52NeX57naeVdXV0aOHElAQMAVeVyeW0RE\nRKQs9HidiIiIiJXs3r0bk8lEZmYm586do06dOpZzbdu2JTk5GYAff/yRNm3a0KRJE/bv34/ZbObw\n4cOkp6dfMWZWVhZNmzYFYPPmzVy6dMkqsfr5+ZGYmAjAn3/+yZw5c6wyroiIiMhlKjqJiIiIWEmj\nRo0IDw9n+PDhvPTSS0VWMYWFhfHFF18wbNgwEhISCAsLo23btjRr1ozAwEBiY2Np3rx5kRVJAI88\n8gjLli3jqaeeol27dmRkZLBmzZoyx9qvXz9q1KhBUFAQI0eO5M477yzzmCIiIiL/ZDBfXn9dSWVk\n5Jb7HOGz1pb7HCKOYN64/vYOQUQckNHoYe8QSiUhIYGDBw8yfvz4UvfJz89nw4YNPProo5w/f55+\n/fqRmJiIi0vF3wHBFvdgN8po9KjQ8VlTVcm1quQJVSfXqpInlF+upuWZVh9T5J98xv1Pmceo+Hc0\nIiIiIg7Kzc2NvXv38tFHH+Hk5ER4eHilKDiJiIj9OY/wsncIRaiQKMXRXY2IiIiIFQwYMOCG+kVG\nRlo5EhEREZGKQUUnEREREREREQdTuPLfNp3vjE1ns68qk+tLHcs8hDYSFxERERERERERq1PRSURE\nRKQSmz59OmlpafYOQ0REROQKerxOREREpBKLiIiwdwgiIiIixVLRSURERMSOTCYTkZGRpKWlUVBQ\nQFhYGO+99x53330327dvJysri/fffx+j0ci4ceNIT0/H39+fjRs38u233xISEkJkZCRff/01ubm5\nHD16lOPHjzNx4kS6d+/Opk2bWLp0KS4uLrRp04YJEyYUO2eXLl0ICQmhRYsWALz++ut2/mRERESk\nslPRSURERMSO1q1bh9FoZMaMGWRmZjJ8+HDq1KlDrVq1iI2NZfbs2WzatImmTZty8eJFPv30U5KS\nkoiNjb1irNOnT7NkyRK+/fZbVq1aRYcOHVi0aBGrV6/Gzc2N8PBwdu7cSVpa2hVzrlu3DoAWLVoQ\nHBxs649BREREHJCKTiIiIiJ2tGvXLnbu3ElKSgoAFy9e5NKlS3To0AEAHx8fsrOzOXz4MO3btweg\ne/fuuLhceRt3+byPjw+5ubkcOnSI9PR0QkNDAcjNzSU9Pb3YOfPz8wFo165d+SYsIiIiVYaKTiIi\nIiJ25OrqysiRIwkICLAcCwkJwdnZ2fLebDZjNpstxwwGQ7Fj/XchytXVlTZt2hATE1PkeGpq6hVz\n/rOPiIiIiDXo1+tERERE7MjPz4/ExEQA/vzzT+bMmVNsu6ZNm7Jv3z4Atm7dislkuubYN998M4cP\nH+bPP/8EYP78+Zw5c6bUc4qIiIiUhYpOIiIiInbUr18/atSoQVBQECNHjuTOO+8stl3Pnj3Jy8sj\nODiYn376iTp16lxzbHd3dyZOnMgzzzxDUFAQ2dnZ1K9fv9RzioiIiJSFwWw2m+0dRFlkZOSW+xzh\ns9aW+xwijmDeuP72DkFEHJDR6GHvECqE7OxskpOTuf/++zlz5gzDhw/nq6++sls8trgHu1FGo0eF\njs+aqkquVSVPqDq5VpU8wX65Fq78t83nFMfi/VLHMo+hPZ1EREREKoGaNWuyceNGYmJiKCws5LXX\nXrN3SCIiUoE5PdHSpvOpkCjFUdGpFLR6Q0REROzN1dWVuXPn2jsMERERkVJT0UlEREREREREKhzz\n6h/sHUKxfrd3ALYy6v4yD2HzjcR37NhBly5dSEpKumq7MWPGMGHCBBtFJSIiIiIiIiIi1mTTotPx\n48dZtmwZ7du3v2q777//nuPHj9soKhERERHb+Prrr6+rfUhICAcOHChV2+nTp5OWllbsuYSEBL75\n5pvrmltERESkrGxadDIajURHR+PhUfKv0OTn57No0SKef/55G0YmIiIiUr5OnDjB+vXry238iIgI\nmjRpUuy5AQMG0Ldv33KbW0RERKQ4Nt3Tyd3d/ZptPvjgA4KDg6lVq5YNIhIRERGxjaioKPbs2UN0\ndDQHDhzg7NmzmEwmJk2aRMuWLfn++++ZM2cOzs7OPPjgg4wYMQKAjRs3Mn36dLKzs1m0aBFpaWms\nXLkSg8HAkSNHuP/++xk1ahQhISFERkbi4+PDK6+8Ql5eHh4eHsyZM4elS5fi6elJUFAQ48eP58yZ\nM5w/f57Ro0fTs2dPQkJCuPvuu9m+fTtZWVm8//77NGzY0L4fmIiIiFR65VZ0io+PJz4+vsix0aNH\n07Vr1xL7HDt2jH379jF69GiSk5NLNY+nZw1cXJzLFKuIiIhIeQsNDbUUi7p27UpgYCCHDh1i+vTp\nLF26lDfeeINVq1Zx00038cILLxAUFARA3bp1iY2N5Z133mHTpk20atWKPXv2sHHjRgoLC+nVqxej\nRo2yzBMTE8O9997LsGHDWL58Odu2bbOcO3v2LPfeey+PPfYYaWlphIeH07NnTwBq1apFbGwss2fP\nZtOmTZail4iIiMiNKreiU2BgIIGBgdfVZ8uWLaSnpzN48GDy8vLIzMxkyZIlPPPMMyX2yco6X9ZQ\nRUREpAIzGkt+LL8y2rVrF5mZmaxduxaAv/76i8zMTKpVq4aXlxfw98rvy+68804AvL29yc7OBqB1\n69YlriD/+eefCQ8PB7AUjn755RcAateuzd69e1m9ejVOTk6W8QA6dOgAgI+PT5HjIiIiIjfKpo/X\nXcuIESMsN0fJycl8/vnnVy04iYiIiFQ2rq6uREZG4u/vbzmWlZVFYWFhse2dnf+zottsNgPg4lLy\nLZyzs3OJY3355ZecPXuWjz/+mOzsbAYNGnTVeURERETKwqYbiW/ZsoWQkBC+++475syZw1NPPQXA\n4sWL2bVrly1DEREREbEpJycnCgoK8PPzY/PmzQAcOnSIZcuW4enpiclk4syZM5jNZp577jlycnJu\naJ42bdqwfft2AFatWsXnn39uOZeVlUXjxo1xcnLim2++IT8/v+yJiYiIiJTApiudevToQY8ePa44\n/uyzz15xrHPnznTu3NkGUYmIiIiUv+bNm/Pzzz/TuHFjTp06xZAhQygsLCQiIgKAyZMnExYWBkC/\nfv2oXbv2Dc0zfPhwXn31VUJCQqhZsyazZ89m2bJlANx33308//zz7N69m4EDB+Lj40N0dLR1EhQR\nERH5LwZzJV8/nZGRa+8QREREpBw52p5OjqIi34MZjR4VOj5rqiq5VpU8oerkWlXyhKqTa3nkaV79\ng1XHk+tTf9T9ZR6jQu3pJCIiIiIiIiICYHj8bnuHUKyqUki0hkq/0klERERERERERCoem24kLiIi\nIiIiIiIiVYOKTiIiIiIiIiIiYnUqOomIiIiIiIiIiNWp6CQiIiIiIiIiIlanopOIiIiIiIiIiFid\nik4iIiIiIiIiImJ1LvYOQOR6nThxgocffpg2bdpYjrVs2ZKIiAg7RiUi5eHEiRP07t2b1atXc8cd\nd1iODxw4kBYtWvDWW2/ZMToRsaeCggIiIiI4fvw4JpOJV199lQ4dOhRps3btWmJjY3FycmLw4MEE\nBgZy6dIlJkyYQHp6Os7Ozrz55ps0adLETlmUzo4dOwgPD2fGjBn07NmzyLl9+/bx9ttvW94fOnSI\nhQsX8v3337Nu3Tq8vb0B6N+/P4GBgTaN+0ZcLVcAX19f2rdvb3m/fPlyCgsLHeo7BdiwYQNLly7F\nycmJLl268PLLL5OQkMC8efNo2rQpAHfffTfPP/+8rUO/btfK1RGu02vF60jXaWm+G0e4TkuTpyNc\npzNmzCA1NRWDwcDEiRNp166d5dwPP/zAnDlzcHZ2plu3brz44ovX7FMss0glk5aWZn7sscfsHYaI\n2EBaWpq5d+/e5qlTp1qOHTt2zNynTx/z+PHj7RiZiNjbZ599Zp48ebLZbDabDxw4YB44cGCR8+fO\nnTPfd9995pycHPNff/1lfuihh8xZWVnmhIQE85QpU8xms9n83XffmcPDw20d+nX57bffzCNHjjS/\n8MIL5n/9619XbXv27FnzE088YTaZTOb58+eb4+LibBSldZQm106dOl1xzNG+0/Pnz5t79uxpzs3N\nNRcWFpoHDRpkPnjwoHnNmjXmt956yw4R37hr5eoo1+n1xFvZr9PS5OoI1+m14nWE6zQ5Odn87LPP\nms1ms/nQoUPmwYMHFznfr18/c3p6utlkMpmDg4PNBw8evGaf4ujxOhERqdD8/Pz44YcfMJlMAKxf\nv5577rnHzlGJE6dOygAADK5JREFUiL3179+f1157DQAvLy+ys7OLnE9NTaVt27Z4eHhQvXp12rdv\nT0pKCtu2baNv377A33+BTklJsXns18NoNBIdHY2Hh8c128bExDB8+HCcnCrnLf715PpPjvaduru7\ns3btWmrVqoXBYKBOnTpX/PddWVwrV0e5Tq8n3sp+nd7od+No36kjXKfbtm2jT58+ADRv3pyzZ8+S\nl5cHQFpaGjfddBMNGjTAycmJ7t27s23btqv2KUnl/C9dRESqDFdXV/z8/EhOTgYgMTGR7t272zkq\nEbE3V1dXqlWrBkBsbCwBAQFFzv/xxx94eXlZ3nt5eZGRkVHkuJOTEwaDgfz8fNsFfp3c3d1xdna+\nZrsLFy6wdetWevfubTn21Vdf8eSTT/Lcc8+RlpZWnmFaRWlyzc/PZ+zYsQQFBbFs2TIAh/xOa9Wq\nBcCvv/7KyZMn8fPzA/5+VC00NJThw4fz888/l3usZXWtXB3lOi1tvI5wnZYmV0e4TksTb2W/Tv/4\n4w88PT0t7y9ffwAZGRklXpsl9SmJ9nSSSuno0aOEhIRY3lf0Z2VFpGweeOABvvzyS+rVq4e3tzc1\natSwd0giYkPx8fHEx8cXOTZ69Gi6du3KypUr2b9/P++///5VxzCbzdd13B6ulue1bN68mR49elhW\nT3Tv3p277rqLjh07sn79eqZNm8YHH3xQLnHfiBvN9dVXX6V///4YDAaGDh16xT5e4Djf6bFjx3jl\nlVd45513LH+A8fLyokePHuzatYvx48ezbt268gr9upUl18sq63Wamppa5H1J8TrCdVqaXB3hOi3t\nd1rZrtOruZHvpDR9VHSSSunmm28mLi7O3mGIiI106dKFqKgojEYj999/v73DEREbCwwMLHZz3fj4\neP71r3/x3nvv4erqWuRc/fr1+eOPPyzvf//9d+644w7q169PRkYGLVu25NKlS5jNZtzc3Mo9h9Io\nKc/SSEpKIjg42PL+nxu79urVi9mzZ5c5Pmu60Vz/meNdd93FgQMHHPI7PX36NC+++CIzZ86kVatW\nwN+PsjRv3hwAf39/MjMzMZlMpVoJZws3kqujXKcTJkwoVbyOcJ2WJldHuE5Lk2dlvE7/qbjrz2g0\nFnvuzJkz1K9fH1dX1xL7lESP14mISIXn5uZGx44dWbNmDb169bJ3OCJSAaSlpbFq1Sqio6Mtj9n9\nk5+fH3v37iUnJ4dz586RkpJChw4duOeee/jqq6+Av/8B2LlzZ1uHXi727dtHy5YtLe+nTZvGTz/9\nBPz9qEeLFi3sFZrVHDlyhLFjx2I2mykoKCAlJYUWLVo45HcaERHBlClT8PX1tRxbsmQJX375JQAH\nDhzAy8urQv5D9no4ynVa2ngd4Tq9Vq6Ocp2WJt7Kfp3ec889fP311wDs37+f+vXrWx4ZbNy4MXl5\neZw4cYKCggKSkpK45557rtqnJFrpJCIilcIDDzxAZmbmdW8wKyKOKT4+nuzsbJ599lnLsZiYGJYv\nX07Hjh3x9/dn7NixhIaGYjAYePHFF/Hw8ODBBx/khx9+IDg4GDc3N9566y07ZnFtW7ZsISYmhiNH\njrB//37i4uJYunQpixcvtuQJkJOTU+TGPzAwkMmTJ+Pi4oLBYGDatGn2SqHUSpOrj48PgwYNwsnJ\niV69etGuXTt8fX0d6jutU6cOP/30E/Pnz7f0GTFiBA8//DDjxo1j1apVFBQUMH36dDtmUTql+U4d\n4TotKV5HvE5Lk6sjXKfXytMRrtP27dvj6+tLUFAQBoOByZMnk5CQgIeHB3379mXKlCmMHTsW+Pvz\nuPnmm7n55puv6HMtBnNFephSREREREREREQcgh6vExERERERERERq1PRSURERERERERErE5FJxER\nERERERERsToVnURERERERERExOpUdBIREREREREREatT0UlEKr3ffvuNXr162TsMEREREYdz8OBB\nQkJCyM/Pt9mcmzdvZvz48TabT0TKj4pOIiIiIiIicoXCwkLGjRvHlClTcHNzs9m8ffr0oaCggA0b\nNthsThEpHy72DkBEqhaz2UxUVBSpqanUq1cPHx8fPD096dKlCwsXLsRsNuPi4sLUqVNp0qQJvXr1\nYtiwYXz77becOHGCN954gy5dupCSksLkyZPx8vLC19fXMv7Zs2eZPHkymZmZ5OXl8eSTT/Lwww+z\nYMECTpw4QXp6OuPHj6dNmzZ2/BREREREbkxycjLvv/8+Pj4+7N27Fz8/P26//Xa++eYbsrOzWbJk\nCceOHSv2vuqbb77hww8/xM3NDZPJxMyZM2ncuDEhISF06dKFXbt2cezYMUaPHk3//v1JTEzEx8eH\n5s2bU1BQwKRJkzh69CgGg4FWrVoxefJkLly4wGuvvcapU6cAGDNmDJ06dWLLli0sXLiQ6tWr4+7u\nztSpU/H29qZXr17069ePtLQ05s+fz4YNG1ixYgVmsxkvLy+mTZuGp6cnTz/9NBMmTODBBx+08ycu\nImWhlU4iYlPbtm1jz549xMfHM3fuXLZv386FCxeYPHkyCxYsYMWKFQwdOpSZM2da+lSrVo2lS5fy\n/PPP89FHHwEwc+ZMXnnlFWJjYzEajZa2c+fOpWvXrnz00UesWLGC+fPnk5mZCcCJEyf46KOPVHAS\nERGRSm3Pnj2MHz+eNWvWsG7dOmrXrk1cXBy+vr588cUXJd5X5eTk8O677xIXF0f37t1ZuXKlZczz\n58+zZMkSpk+fzocffgjAd999R9euXQE4cOAAqamprF69mlWrVtGqVStyc3OJiYnBx8eHVatW8dZb\nbxEfH89ff/3FpEmTWLBgAXFxcXTr1o25c+da5mrWrBnz58/n1KlTvP/++yxfvpxPPvmETp068cEH\nHwDQqlUrfv/9d37//XdbfawiUg600klEbOqXX36hQ4cOODs7U6NGDbp27cqRI0fIyMhg9OjRAJhM\nJgwGg6VPp06dAGjYsCFnz54F4Ndff+XOO+8E4K677iIuLg74+69/e/fu5YsvvgDAxcWFEydOAODn\n51dkXBEREZHKqHnz5tSpUweAOnXq4O/vD4C3tzcFBQUl3lfVq1eP8ePHYzabycjIsPSD4u+3Tp06\nRffu3S1zenp68swzz9CzZ0/69euHh4cHe/bsITg4GPi7mDRr1ix++eUX6tati4+Pj2XsVatWWea6\nPO+uXbvIyMggNDQUgPz8fBo3bmxp16BBA9LT06lfv741Pz4RsSEVnUTEpgoLC3Fy+s8iSycnJ9zc\n3GjYsKGlcPTfXFz+839VZrO5SF/4+2bqMjc3NyZPnkzbtm2LjPF///d/uLq6WiUHEREREXtydnYu\n8f0vv/xS7H3VpUuXeOmll/j8889p1qwZK1asYN++fZbzJd1vXVatWjU+/vhj9u/fT1JSEoMGDeKT\nTz7BYDBQWFhYpO1//5HPbDYXOXb5nszNzY127dpZVjeJiOPR43UiYlO33HILu3fvxmw289dff7F1\n61aaNGlCVlYWBw4cAODHH39k9erVVx2nefPm7N69G4AffvjBcvzOO+9k48aNAFy4cIEpU6ZQUFBQ\nTtmIiIiIVCzNmjUr9r7q3LlzODk50ahRIy5evEhiYuI1f5GuQYMGnD59GoC9e/fy+eef4+vry6hR\no/D19eXYsWP4+/vz3XffAX9vZTB8+HCaNWvGn3/+SXp6OvD39gp+fn5XjN+2bVv27NlDRkYGABs3\nbmTz5s2W86dOnaJhw4Zl/1BExG600klEbKp79+6sX7+egQMH0qBBA/z9/alZsyazZs0iIiKCatWq\nARAVFXXVccaNG8fUqVNp0KABrVu3thwfNWoUkyZNIjg4mPz8fB5//PEif7kTERERcWTVq1cv9r6q\nTp06BAQEMGjQIBo2bEhoaCivvvqq5Y91xenatStr1qzhiSeeoGnTpixcuJDVq1fj5uZG06ZNad++\nPW3atCEyMpIhQ4ZQWFjISy+9RPXq1Zk+fTovv/wybm5u1KhRg+nTp18xvre3NxERETz33HO4u7tT\nvXp13n77bQD+/e9/U79+fT1aJ1LJGczFrZ0UESknubm5bN68mUcffRSDwcDIkSMJCAggICDA3qGJ\niIiIyD8UFhYyYMAA3nnnHZo3b27TuceOHUvv3r3163UilZwerxMRm6pZsyYpKSkMGDCAoKAgPD09\neeCBB+wdloiIiIj8FycnJ2bOnMmUKVOu+SieNW3evBlnZ2cVnEQcgFY6iYiIiIiIiIiI1Wmlk4iI\niIiIiIiIWJ2KTiIiIiIiIiIiYnUqOomIiIiIiIiIiNWp6CQiIiIiIiIiIlanopOIiIiIiIiIiFid\nik4iIiIiIiIiImJ1/w9sKZfY6a7/dgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 1440x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "hRqEH-7n9sJI",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Dimensión *películas antiguas*"
      ]
    },
    {
      "metadata": {
        "id": "o_2WI4D99AGA",
        "colab_type": "code",
        "outputId": "21cc3cc0-3a63-4e53-f552-f73960e85b3f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 70
        }
      },
      "cell_type": "code",
      "source": [
        "movies[\"release_year\"] = movies[\"release_date\"].str.extract(\"(\\d{4})$\").astype(float)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:1: FutureWarning: currently extract(expand=None) means expand=False (return Index/Series/DataFrame) but in a future version of pandas this will be changed to expand=True (return DataFrame)\n",
            "  \"\"\"Entry point for launching an IPython kernel.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "metadata": {
        "id": "e7uQdoCG9y7t",
        "colab_type": "code",
        "outputId": "00332466-60c1-4915-d131-75e3a196a997",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 210
        }
      },
      "cell_type": "code",
      "source": [
        "movies[movies[\"item_id\"] == 674]"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>item_id</th>\n",
              "      <th>movie_title</th>\n",
              "      <th>release_date</th>\n",
              "      <th>video_release_date</th>\n",
              "      <th>IMDb URL</th>\n",
              "      <th>unknown</th>\n",
              "      <th>Action</th>\n",
              "      <th>Adventure</th>\n",
              "      <th>Animation</th>\n",
              "      <th>Children</th>\n",
              "      <th>...</th>\n",
              "      <th>Horror</th>\n",
              "      <th>Musical</th>\n",
              "      <th>Mystery</th>\n",
              "      <th>Romance</th>\n",
              "      <th>Sci-Fi</th>\n",
              "      <th>Thriller</th>\n",
              "      <th>War</th>\n",
              "      <th>Western</th>\n",
              "      <th>score</th>\n",
              "      <th>release_year</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>674</th>\n",
              "      <td>674</td>\n",
              "      <td>Nosferatu (Nosferatu, eine Symphonie des Graue...</td>\n",
              "      <td>01-Jan-1922</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Nosferatu,%20...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3.758008</td>\n",
              "      <td>1922.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1 rows × 26 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     item_id                                        movie_title release_date  \\\n",
              "674      674  Nosferatu (Nosferatu, eine Symphonie des Graue...  01-Jan-1922   \n",
              "\n",
              "     video_release_date                                           IMDb URL  \\\n",
              "674                 NaN  http://us.imdb.com/M/title-exact?Nosferatu,%20...   \n",
              "\n",
              "     unknown  Action  Adventure  Animation  Children      ...       Horror  \\\n",
              "674        0       0          0          0         0      ...            1   \n",
              "\n",
              "     Musical  Mystery  Romance  Sci-Fi  Thriller  War  Western     score  \\\n",
              "674        0        0        0       0         0    0        0  3.758008   \n",
              "\n",
              "     release_year  \n",
              "674        1922.0  \n",
              "\n",
              "[1 rows x 26 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "metadata": {
        "id": "9F1VjV8T9iO7",
        "colab_type": "code",
        "outputId": "52bc2f3a-b9eb-4890-990c-322fc8eca374",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 160
        }
      },
      "cell_type": "code",
      "source": [
        "movies[movies[\"item_id\"] == 884]"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>item_id</th>\n",
              "      <th>movie_title</th>\n",
              "      <th>release_date</th>\n",
              "      <th>video_release_date</th>\n",
              "      <th>IMDb URL</th>\n",
              "      <th>unknown</th>\n",
              "      <th>Action</th>\n",
              "      <th>Adventure</th>\n",
              "      <th>Animation</th>\n",
              "      <th>Children</th>\n",
              "      <th>...</th>\n",
              "      <th>Horror</th>\n",
              "      <th>Musical</th>\n",
              "      <th>Mystery</th>\n",
              "      <th>Romance</th>\n",
              "      <th>Sci-Fi</th>\n",
              "      <th>Thriller</th>\n",
              "      <th>War</th>\n",
              "      <th>Western</th>\n",
              "      <th>score</th>\n",
              "      <th>release_year</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>884</th>\n",
              "      <td>884</td>\n",
              "      <td>Phantoms (1998)</td>\n",
              "      <td>01-Jan-1998</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Phantoms+(1998)</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.354297</td>\n",
              "      <td>1998.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1 rows × 26 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     item_id      movie_title release_date  video_release_date  \\\n",
              "884      884  Phantoms (1998)  01-Jan-1998                 NaN   \n",
              "\n",
              "                                             IMDb URL  unknown  Action  \\\n",
              "884  http://us.imdb.com/M/title-exact?Phantoms+(1998)        0       0   \n",
              "\n",
              "     Adventure  Animation  Children      ...       Horror  Musical  Mystery  \\\n",
              "884          0          0         0      ...            1        0        0   \n",
              "\n",
              "     Romance  Sci-Fi  Thriller  War  Western     score  release_year  \n",
              "884        0       0         0    0        0  0.354297        1998.0  \n",
              "\n",
              "[1 rows x 26 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 19
        }
      ]
    },
    {
      "metadata": {
        "id": "Mj6XsP-6-TqG",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "u = item_embeddings[674] - item_embeddings[884]\n",
        "u_hat = u / (u**2).sum()**0.5\n",
        "movies[\"score\"] = np.transpose(np.dot(item_embeddings, np.expand_dims(u_hat, 1)))[0]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "-ajNuYPH-f8B",
        "colab_type": "code",
        "outputId": "a7a98a25-b035-41b1-b8db-9c93d5d92215",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 396
        }
      },
      "cell_type": "code",
      "source": [
        "movies.sort_values(\"score\").head()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>item_id</th>\n",
              "      <th>movie_title</th>\n",
              "      <th>release_date</th>\n",
              "      <th>video_release_date</th>\n",
              "      <th>IMDb URL</th>\n",
              "      <th>unknown</th>\n",
              "      <th>Action</th>\n",
              "      <th>Adventure</th>\n",
              "      <th>Animation</th>\n",
              "      <th>Children</th>\n",
              "      <th>...</th>\n",
              "      <th>Horror</th>\n",
              "      <th>Musical</th>\n",
              "      <th>Mystery</th>\n",
              "      <th>Romance</th>\n",
              "      <th>Sci-Fi</th>\n",
              "      <th>Thriller</th>\n",
              "      <th>War</th>\n",
              "      <th>Western</th>\n",
              "      <th>score</th>\n",
              "      <th>release_year</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>305</th>\n",
              "      <td>305</td>\n",
              "      <td>Mrs. Brown (Her Majesty, Mrs. Brown) (1997)</td>\n",
              "      <td>01-Jan-1997</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Her+Majesty%2...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>-5.291117</td>\n",
              "      <td>1997.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>884</th>\n",
              "      <td>884</td>\n",
              "      <td>Phantoms (1998)</td>\n",
              "      <td>01-Jan-1998</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Phantoms+(1998)</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>-4.846834</td>\n",
              "      <td>1998.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>279</th>\n",
              "      <td>279</td>\n",
              "      <td>Up Close and Personal (1996)</td>\n",
              "      <td>01-Mar-1996</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Up%20Close%20...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>-4.591436</td>\n",
              "      <td>1996.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>250</th>\n",
              "      <td>250</td>\n",
              "      <td>Shall We Dance? (1996)</td>\n",
              "      <td>11-Jul-1997</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Shall%20we%20...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>-4.537306</td>\n",
              "      <td>1997.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>534</th>\n",
              "      <td>534</td>\n",
              "      <td>Addicted to Love (1997)</td>\n",
              "      <td>23-May-1997</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Addicted%20to...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>-4.499990</td>\n",
              "      <td>1997.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 26 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     item_id                                  movie_title release_date  \\\n",
              "305      305  Mrs. Brown (Her Majesty, Mrs. Brown) (1997)  01-Jan-1997   \n",
              "884      884                              Phantoms (1998)  01-Jan-1998   \n",
              "279      279                 Up Close and Personal (1996)  01-Mar-1996   \n",
              "250      250                       Shall We Dance? (1996)  11-Jul-1997   \n",
              "534      534                      Addicted to Love (1997)  23-May-1997   \n",
              "\n",
              "     video_release_date                                           IMDb URL  \\\n",
              "305                 NaN  http://us.imdb.com/M/title-exact?Her+Majesty%2...   \n",
              "884                 NaN   http://us.imdb.com/M/title-exact?Phantoms+(1998)   \n",
              "279                 NaN  http://us.imdb.com/M/title-exact?Up%20Close%20...   \n",
              "250                 NaN  http://us.imdb.com/M/title-exact?Shall%20we%20...   \n",
              "534                 NaN  http://us.imdb.com/M/title-exact?Addicted%20to...   \n",
              "\n",
              "     unknown  Action  Adventure  Animation  Children      ...       Horror  \\\n",
              "305        0       0          0          0         0      ...            0   \n",
              "884        0       0          0          0         0      ...            1   \n",
              "279        0       0          0          0         0      ...            0   \n",
              "250        0       0          0          0         0      ...            0   \n",
              "534        0       0          0          0         0      ...            0   \n",
              "\n",
              "     Musical  Mystery  Romance  Sci-Fi  Thriller  War  Western     score  \\\n",
              "305        0        0        1       0         0    0        0 -5.291117   \n",
              "884        0        0        0       0         0    0        0 -4.846834   \n",
              "279        0        0        1       0         0    0        0 -4.591436   \n",
              "250        0        0        0       0         0    0        0 -4.537306   \n",
              "534        0        0        1       0         0    0        0 -4.499990   \n",
              "\n",
              "     release_year  \n",
              "305        1997.0  \n",
              "884        1998.0  \n",
              "279        1996.0  \n",
              "250        1997.0  \n",
              "534        1997.0  \n",
              "\n",
              "[5 rows x 26 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        }
      ]
    },
    {
      "metadata": {
        "id": "N8MSqXJM-iOw",
        "colab_type": "code",
        "outputId": "1524b07a-4f58-4f4a-9be8-3c12cadb28b9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 446
        }
      },
      "cell_type": "code",
      "source": [
        "movies.sort_values(\"score\").tail()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>item_id</th>\n",
              "      <th>movie_title</th>\n",
              "      <th>release_date</th>\n",
              "      <th>video_release_date</th>\n",
              "      <th>IMDb URL</th>\n",
              "      <th>unknown</th>\n",
              "      <th>Action</th>\n",
              "      <th>Adventure</th>\n",
              "      <th>Animation</th>\n",
              "      <th>Children</th>\n",
              "      <th>...</th>\n",
              "      <th>Horror</th>\n",
              "      <th>Musical</th>\n",
              "      <th>Mystery</th>\n",
              "      <th>Romance</th>\n",
              "      <th>Sci-Fi</th>\n",
              "      <th>Thriller</th>\n",
              "      <th>War</th>\n",
              "      <th>Western</th>\n",
              "      <th>score</th>\n",
              "      <th>release_year</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>435</th>\n",
              "      <td>435</td>\n",
              "      <td>American Werewolf in London, An (1981)</td>\n",
              "      <td>01-Jan-1981</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?American%20We...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6.073152</td>\n",
              "      <td>1981.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>431</th>\n",
              "      <td>431</td>\n",
              "      <td>Fantasia (1940)</td>\n",
              "      <td>01-Jan-1940</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Fantasia%20(1...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6.151361</td>\n",
              "      <td>1940.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>495</th>\n",
              "      <td>495</td>\n",
              "      <td>It's a Wonderful Life (1946)</td>\n",
              "      <td>01-Jan-1946</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?It's%20a%20Wo...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6.235801</td>\n",
              "      <td>1946.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>442</th>\n",
              "      <td>442</td>\n",
              "      <td>Birds, The (1963)</td>\n",
              "      <td>01-Jan-1963</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Birds,%20The%...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6.864280</td>\n",
              "      <td>1963.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>674</th>\n",
              "      <td>674</td>\n",
              "      <td>Nosferatu (Nosferatu, eine Symphonie des Graue...</td>\n",
              "      <td>01-Jan-1922</td>\n",
              "      <td>NaN</td>\n",
              "      <td>http://us.imdb.com/M/title-exact?Nosferatu,%20...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>7.636474</td>\n",
              "      <td>1922.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 26 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     item_id                                        movie_title release_date  \\\n",
              "435      435             American Werewolf in London, An (1981)  01-Jan-1981   \n",
              "431      431                                    Fantasia (1940)  01-Jan-1940   \n",
              "495      495                       It's a Wonderful Life (1946)  01-Jan-1946   \n",
              "442      442                                  Birds, The (1963)  01-Jan-1963   \n",
              "674      674  Nosferatu (Nosferatu, eine Symphonie des Graue...  01-Jan-1922   \n",
              "\n",
              "     video_release_date                                           IMDb URL  \\\n",
              "435                 NaN  http://us.imdb.com/M/title-exact?American%20We...   \n",
              "431                 NaN  http://us.imdb.com/M/title-exact?Fantasia%20(1...   \n",
              "495                 NaN  http://us.imdb.com/M/title-exact?It's%20a%20Wo...   \n",
              "442                 NaN  http://us.imdb.com/M/title-exact?Birds,%20The%...   \n",
              "674                 NaN  http://us.imdb.com/M/title-exact?Nosferatu,%20...   \n",
              "\n",
              "     unknown  Action  Adventure  Animation  Children      ...       Horror  \\\n",
              "435        0       0          0          0         0      ...            1   \n",
              "431        0       0          0          1         1      ...            0   \n",
              "495        0       0          0          0         0      ...            0   \n",
              "442        0       0          0          0         0      ...            1   \n",
              "674        0       0          0          0         0      ...            1   \n",
              "\n",
              "     Musical  Mystery  Romance  Sci-Fi  Thriller  War  Western     score  \\\n",
              "435        0        0        0       0         0    0        0  6.073152   \n",
              "431        1        0        0       0         0    0        0  6.151361   \n",
              "495        0        0        0       0         0    0        0  6.235801   \n",
              "442        0        0        0       0         0    0        0  6.864280   \n",
              "674        0        0        0       0         0    0        0  7.636474   \n",
              "\n",
              "     release_year  \n",
              "435        1981.0  \n",
              "431        1940.0  \n",
              "495        1946.0  \n",
              "442        1963.0  \n",
              "674        1922.0  \n",
              "\n",
              "[5 rows x 26 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    }
  ]
}