notebooks/embedModel-Dev-Train-Test.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "embedModel-Dev-Train-Test.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "MOCAuYXAp0aY",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "%%capture\n",
        "# Install the latest Tensorflow version.\n",
        "!pip3 install --upgrade tensorflow-gpu\n",
        "# Install TF-Hub.\n",
        "!pip3 install tensorflow-hub\n",
        "!pip3 install seaborn"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tSLiH4bgp5In",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from absl import logging\n",
        "import json\n",
        "\n",
        "import tensorflow as tf\n",
        "import tensorflow_hub as hub\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "from scipy.stats import pearsonr"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P1a_3wSh8IPq",
        "colab_type": "code",
        "outputId": "6cf81702-172c-4d99-ec4a-246e99f6e97e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        }
      },
      "source": [
        "module_url = \"https://tfhub.dev/google/universal-sentence-encoder/4\"\n",
        "model = hub.load(module_url)\n",
        "print (\"module %s loaded\" % module_url)\n",
        "def embed(input):\n",
        "  return model(input)\n",
        "\n",
        "print(tf.__version__)"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "module https://tfhub.dev/google/universal-sentence-encoder/4 loaded\n",
            "2.0.0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Qf0zGilMtVBp",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def embed_pipe(batch):\n",
        "  s1_embed = tf.nn.l2_normalize(embed(tf.constant(batch['s1'].tolist())), axis=1)\n",
        "  s2_embed = tf.nn.l2_normalize(embed(tf.constant(batch['s2'].tolist())), axis=1)\n",
        "  cosine_similarities = tf.reduce_sum(tf.multiply(s1_embed, s2_embed), axis=1)\n",
        "  clip_cosine_similarities = tf.clip_by_value(cosine_similarities, -1.0, 1.0)\n",
        "  score = 1.0 - tf.scalar_mul((1 / np.pi), tf.acos(clip_cosine_similarities))  \n",
        "  predics = tf.scalar_mul(4, score) + 1.0\n",
        "\n",
        "  return predics"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SKoD9D9JzoeL",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def rmse(predictions, targets):\n",
        "    assert len(predictions) == len(targets)\n",
        "    return np.sqrt(((predictions - targets) ** 2).mean())\n",
        "\n",
        "\n",
        "def accuracy(predictions, targets):\n",
        "    assert len(predictions) == len(targets)\n",
        "    count_pos = 0\n",
        "    for predic, gold in zip(predictions, targets):\n",
        "        if predic == gold:\n",
        "            count_pos += 1\n",
        "\n",
        "    return float(count_pos) / len(targets)\n",
        "\n",
        "\n",
        "# Scraped from evaluation.py, returns recall, precision, f1\n",
        "def get_scores(predictions, targets, prec=3):\n",
        "    label_set = [1, 2, 3, 4, 5]\n",
        "    classification_report = {}\n",
        "    classification_report[\"micro\"] = {\"recall\": 0.0, \"precision\": 0.0, \"fscore\": 0.0}\n",
        "    for label in label_set:\n",
        "        classification_report[label] = {\"recall\": 0.0, \"precision\": 0.0, \"fscore\": 0.0}\n",
        "        tp, fp, fn = 0, 0, 0\n",
        "        for idx, gold in enumerate(targets):\n",
        "            prediction = predictions[idx]\n",
        "            if gold == prediction:\n",
        "                if prediction == label:\n",
        "                    tp += 1\n",
        "            else:\n",
        "                if prediction == label:\n",
        "                    fp += 1\n",
        "                else:\n",
        "                    fn += 1\n",
        "        try:\n",
        "            recall = float(tp) / (tp + fn)\n",
        "        except ZeroDivisionError:\n",
        "            recall = 0.0\n",
        "        try:\n",
        "            precision = float(tp) / (tp + fp)\n",
        "        except ZeroDivisionError:\n",
        "            precision = 0.0\n",
        "        try:\n",
        "            fscore = 2 * precision * recall / (precision + recall)\n",
        "        except ZeroDivisionError:\n",
        "            fscore = 0.0\n",
        "        classification_report[label][\"recall\"] = round(recall, prec)\n",
        "        classification_report[label][\"precision\"] = round(precision, prec)\n",
        "        classification_report[label][\"fscore\"] = round(fscore, prec)\n",
        "        classification_report[\"micro\"][\"recall\"] += recall\n",
        "        classification_report[\"micro\"][\"precision\"] += precision\n",
        "        classification_report[\"micro\"][\"fscore\"] += fscore\n",
        "\n",
        "    for key in classification_report[\"micro\"].keys():\n",
        "        classification_report[\"micro\"][key] /= len(label_set)\n",
        "        classification_report[\"micro\"][key] = round(\n",
        "            classification_report[\"micro\"][key], prec\n",
        "        )\n",
        "\n",
        "    return classification_report"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "j46nXA4arC-H",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#from google.colab import drive\n",
        "#drive.mount('/content/drive')\n",
        "#!ls \"/content/drive/My Drive/log\""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7luCXw8xs3Sz",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "dev = pd.read_csv(\"/content/drive/My Drive/log/dev.csv\", index_col=0)\n",
        "train = pd.read_csv(\"/content/drive/My Drive/log/train.csv\", index_col=0)\n",
        "test = pd.read_csv(\"/content/drive/My Drive/log/test.csv\", index_col=0)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ye-Rctb0tWSe",
        "colab_type": "code",
        "outputId": "a732ae72-45b7-430c-b459-30798b395070",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        }
      },
      "source": [
        "%%time\n",
        "dev_predics = []\n",
        "for batch in np.array_split(dev, 10):\n",
        "  predic = embed_pipe(batch)\n",
        "  dev_predics.extend(predic)\n",
        "\n",
        "train_predics = []\n",
        "for batch in np.array_split(train, 10):\n",
        "  predic = embed_pipe(batch)\n",
        "  train_predics.extend(predic)\n",
        "\n",
        "test_predics = []\n",
        "for batch in np.array_split(test, 10):\n",
        "  predic = embed_pipe(batch)\n",
        "  test_predics.extend(predic)"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "CPU times: user 2.77 s, sys: 110 ms, total: 2.88 s\n",
            "Wall time: 2.14 s\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EI9VcFM7w649",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "dev[\"prediction\"] = [int(elem) for elem in np.round(dev_predics)]\n",
        "dev[\"noRound\"] = [elem.numpy() for elem in dev_predics]\n",
        "\n",
        "train[\"prediction\"] = [int(elem) for elem in np.round(train_predics)]\n",
        "train[\"noRound\"] = [elem.numpy() for elem in train_predics]\n",
        "\n",
        "test[\"prediction\"] = [int(elem) for elem in np.round(test_predics)]\n",
        "test[\"noRound\"] = [elem.numpy() for elem in test_predics]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PNArJVJP00vT",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "dev.to_csv(\"/content/drive/My Drive/log/new_dev_embed_predics.csv\", index=False)\n",
        "res = dev[[\"id\", \"prediction\"]]\n",
        "res.to_csv(\"/content/drive/My Drive/log/new_dev_embedPredic.txt\", sep=\"\\t\", index=False)\n",
        "\n",
        "train.to_csv(\"/content/drive/My Drive/log/new_train_embed_predics.csv\", index=False)\n",
        "res = train[[\"id\", \"prediction\"]]\n",
        "res.to_csv(\"/content/drive/My Drive/log/new_train_embedPredic.txt\", sep=\"\\t\", index=False)\n",
        "\n",
        "test.to_csv(\"/content/drive/My Drive/log/new_test_embed_predics.csv\", index=False)\n",
        "res = test[[\"id\", \"prediction\"]]\n",
        "res.to_csv(\"/content/drive/My Drive/log/new_test_embedPredic.txt\", sep=\"\\t\", index=False)\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MeU2YogW5cQo",
        "colab_type": "code",
        "outputId": "e0c04701-6dab-4a4b-c2c6-e7a5ba9a08ed",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "for df, name in zip([dev, train], [\"Dev\", \"Train\"]):\n",
        "    acc = accuracy(df[\"prediction\"], df[\"gold\"])\n",
        "    _rmse = rmse(df[\"prediction\"], df[\"gold\"])\n",
        "    pear_corr = pearsonr(list(df[\"prediction\"]), list(df[\"gold\"]))\n",
        "    cols = [\"RMSE\", \"Accuracy\", \"Pearson's R\", \"Pearson's R p-val\"]\n",
        "    vals = [_rmse, acc, pear_corr[0], pear_corr[1]]\n",
        "    stats = pd.DataFrame(list(df[\"prediction\"]), columns=[\"Predic_Label\"]).describe()\n",
        "    extra = pd.DataFrame(vals, index=cols, columns=[\"Predic_Label\"])\n",
        "    print(f\"\\n{name} Gold stats: \")\n",
        "    print(pd.DataFrame(list(df[\"gold\"]), columns=[\"Gold_Label\"]).describe())\n",
        "    print(f\"\\n{name} Embedding Model Prediction stats: \")\n",
        "    print(stats.append(extra))\n",
        "    print(\"\\n------\")\n",
        "\n",
        "for df, name in zip([dev, train], [\"Dev\", \"Train\"]):\n",
        "    print(f\"\\n{name} Prediction Metrics:\")\n",
        "    metrics = get_scores(list(df[\"prediction\"]), list(df[\"gold\"]))\n",
        "    print(json.dumps(metrics, indent=2))"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Dev Gold stats: \n",
            "        Gold_Label\n",
            "count  1209.000000\n",
            "mean      3.816377\n",
            "std       1.063869\n",
            "min       1.000000\n",
            "25%       3.000000\n",
            "50%       4.000000\n",
            "75%       5.000000\n",
            "max       5.000000\n",
            "\n",
            "Dev Embedding Model Prediction stats: \n",
            "                   Predic_Label\n",
            "count              1.209000e+03\n",
            "mean               4.062862e+00\n",
            "std                2.865970e-01\n",
            "min                3.000000e+00\n",
            "25%                4.000000e+00\n",
            "50%                4.000000e+00\n",
            "75%                4.000000e+00\n",
            "max                5.000000e+00\n",
            "RMSE               1.044898e+00\n",
            "Accuracy           3.581472e-01\n",
            "Pearson's R        2.985311e-01\n",
            "Pearson's R p-val  2.612130e-26\n",
            "\n",
            "------\n",
            "\n",
            "Train Gold stats: \n",
            "        Gold_Label\n",
            "count  1484.000000\n",
            "mean      3.853774\n",
            "std       0.999416\n",
            "min       1.000000\n",
            "25%       3.000000\n",
            "50%       4.000000\n",
            "75%       5.000000\n",
            "max       5.000000\n",
            "\n",
            "Train Embedding Model Prediction stats: \n",
            "                   Predic_Label\n",
            "count              1.484000e+03\n",
            "mean               4.045148e+00\n",
            "std                2.673199e-01\n",
            "min                3.000000e+00\n",
            "25%                4.000000e+00\n",
            "50%                4.000000e+00\n",
            "75%                4.000000e+00\n",
            "max                5.000000e+00\n",
            "RMSE               9.671143e-01\n",
            "Accuracy           4.595687e-01\n",
            "Pearson's R        3.200301e-01\n",
            "Pearson's R p-val  1.086964e-36\n",
            "\n",
            "------\n",
            "\n",
            "Dev Prediction Metrics:\n",
            "{\n",
            "  \"micro\": {\n",
            "    \"recall\": 0.211,\n",
            "    \"precision\": 0.326,\n",
            "    \"fscore\": 0.132\n",
            "  },\n",
            "  \"1\": {\n",
            "    \"recall\": 0.0,\n",
            "    \"precision\": 0.0,\n",
            "    \"fscore\": 0.0\n",
            "  },\n",
            "  \"2\": {\n",
            "    \"recall\": 0.0,\n",
            "    \"precision\": 0.0,\n",
            "    \"fscore\": 0.0\n",
            "  },\n",
            "  \"3\": {\n",
            "    \"recall\": 0.008,\n",
            "    \"precision\": 0.429,\n",
            "    \"fscore\": 0.015\n",
            "  },\n",
            "  \"4\": {\n",
            "    \"recall\": 0.951,\n",
            "    \"precision\": 0.314,\n",
            "    \"fscore\": 0.472\n",
            "  },\n",
            "  \"5\": {\n",
            "    \"recall\": 0.095,\n",
            "    \"precision\": 0.889,\n",
            "    \"fscore\": 0.171\n",
            "  }\n",
            "}\n",
            "\n",
            "Train Prediction Metrics:\n",
            "{\n",
            "  \"micro\": {\n",
            "    \"recall\": 0.208,\n",
            "    \"precision\": 0.301,\n",
            "    \"fscore\": 0.153\n",
            "  },\n",
            "  \"1\": {\n",
            "    \"recall\": 0.0,\n",
            "    \"precision\": 0.0,\n",
            "    \"fscore\": 0.0\n",
            "  },\n",
            "  \"2\": {\n",
            "    \"recall\": 0.0,\n",
            "    \"precision\": 0.0,\n",
            "    \"fscore\": 0.0\n",
            "  },\n",
            "  \"3\": {\n",
            "    \"recall\": 0.006,\n",
            "    \"precision\": 0.238,\n",
            "    \"fscore\": 0.012\n",
            "  },\n",
            "  \"4\": {\n",
            "    \"recall\": 0.951,\n",
            "    \"precision\": 0.439,\n",
            "    \"fscore\": 0.601\n",
            "  },\n",
            "  \"5\": {\n",
            "    \"recall\": 0.085,\n",
            "    \"precision\": 0.83,\n",
            "    \"fscore\": 0.154\n",
            "  }\n",
            "}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Vn9hVwQG8myb",
        "colab_type": "code",
        "outputId": "82d3994c-aedb-45dc-919b-8c75925633e3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 465
        }
      },
      "source": [
        "from sklearn.metrics import confusion_matrix\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "%matplotlib inline\n",
        "\n",
        "labels = [1, 2, 3, 4, 5]\n",
        "\n",
        "for df, name in zip([dev, train], [\"Dev\", \"Train\"]):\n",
        "    cm = confusion_matrix(list(df[\"gold\"]), list(df[\"prediction\"]))\n",
        "    df_cm = pd.DataFrame(cm, index=labels, columns=labels)\n",
        "\n",
        "    f,(ax1,ax2) = plt.subplots(1,2,sharey=False, figsize=(10,3))\n",
        "    g1 = sns.heatmap(df_cm,annot=True, fmt='d', ax=ax1)\n",
        "    g1.set_ylabel('True Label')\n",
        "    g1.set_xlabel('Predicted Label')\n",
        "    g1.set_title(f'{name} Confusion Matrix')\n",
        "\n",
        "    cm_norm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
        "    df_cm_norm = pd.DataFrame(cm_norm, index=labels, columns=labels)\n",
        "    g2 = sns.heatmap(df_cm_norm,annot=True, vmin=0, vmax=1, ax=ax2)\n",
        "    g2.set_ylabel('True Label')\n",
        "    g2.set_xlabel('Predicted Label')\n",
        "    g2.set_title(f'{name} Normed Confusion Matrix')"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlQAAADgCAYAAADFXPhjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\na6iMMcYYY+rp/wF2jlFyDYgn3AAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 720x216 with 4 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlQAAADgCAYAAADFXPhjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjIsIGh0\ndHA6Ly9tYXRwbG90bGliLm9yZy8li6FKAAAgAElEQVR4nOzdd3gUVffA8e/ZTUKRXkMSBBQsIIoI\nWACp0iSAimDBrvjaG9grr/X92UCxoCIISBELvRfpUhSkQyhCGp2A1GRzf3/MJGxCshuSTXaSnM/z\nzEN25u6ds+1w586dO2KMQSmllFJK5Z4r2AEopZRSShV22qBSSimllMojbVAppZRSSuWRNqiUUkop\npfJIG1RKKaWUUnmkDSqllFJKqTzSBlUBEhG3iPwrIucHO5bMRKS+iKwRkaMi8mge6vlWRF4OZGwF\nzcmfk1JZKW7fWREZKSJvBmnfPUUk1n6/G+ahns0i0jKQsRU0EWktIuuDHYdTaIPKB/sHk7akisgJ\nr8d3nmt9xhiPMaaMMWZXLuMpISIDRCRGRI6JyE67AROIJPoCMNMYU9YY80VuKzHGPGiMeTcA8WQg\nIm+LiBGRxzKtf85e/2oO61kkIvf6KpPXz0kpf5yUW0Skrv0bmphp/Zic/q6cRkQiReR7EUkUkSMi\nslFE3hCRUgGo/iPgYfv9XpvbSowxFxtjFgYgngzsHGdEpEGm9ZPs9S1yUEeIXba2r3LGmPnGmAa+\nyhQn2qDywf7BlDHGlAF2AdFe60ZlLi8iIfkVi4gI8AvQGegNlAcaAX8DbQOwi1qA0480tgB3Z1p3\nt70+IPLzM1QqjZNyi5frRKRZXisJ9m9IRKoAS4EQ4GpjTDmgE1AVuCCPdbuAmhSyXCki1YAmwMFA\n7SDYn7MTaYMqD+xek7EiMlpEjgJ9RORaEVkmIodFJEFEBolIqF0+Q6vf7rYeJCLT7FNtS0WkTja7\n6wi0AXoYY1YZY1KMMYeNMYOMMcPs+qJEZLKIHBSRrSJyf6ZYR9v7PCoi60Sksb1tAdAS+Mo+Qr4g\nc0+OiDwoIvPtv1123HtFJElE/haR+l6v6U2v5/3H7lE7ICK/iUiNTO/Fw/b2QyIyyM9bvhSoJCIX\n23U0wvoO/+W1v8oiMlVE9tl1ThKRSHvbB8C1Xq/zU684HhWRGGCT9+ckVq/gWhF5xCvuZVLIT2sq\nZyvg3JLm/4B3fMTk77ec1W/oERHZZsfwhojUs1/DEfu1hXrV302sYQeH7fxzmde2q0RktV3PaKCE\nj9fRDzgA3G2M+QfAGPOPMeZxY8x6u74WIrLSzl/LReRqr30tEpG3RGSJvb/pIlJJRM4DjgACrBfr\nlN1ZPTneOVBEqtn56LBYeXmBV7lYEWlt/13S/rwSRCRORD4WkTB7W3uxzkY8b+e1eBHJfGCZ2Sjg\ndrEagAB3AOOBZK/9Z/t9AtLiXG/nylu84nhZRBKBb9LW2fXVs1/jFfbjmvZ3pVCf1jwX2qDKu5uA\nH7F6jMYCKcBTQBWgOdaR0cM+nn8H8BpQCetI9b/ZlGsPLDXGxPmoayywA4jA6sX6n4i08treAxgB\nVACmAYMAjDHXYzVW/mMfIW/3sQ+wesmuAeoBFYHbyOLIR0Q6AAOAnkAkEI/1Q/fWBbgKuBLrP432\nfvY9gjNHXncDP2Ta7gK+Ac7H6nVLBgbar/OFTK/zaa/ndQOaAhnGRBhjTgF9gHdF5CLgFcADfOAn\nTqXyqqByS5rPgIZp/8l7y+FvOavf0A1YPenNsX47X2Dli1pYv/ledv1NsX63DwKVgaHABBEJE5ES\nwAR7XSX77x4+Xkd74BeTzX3VxOrBmoJ16q6y/bqnikhFr2J3APcA1YHzgGeNMcewcidAA2PMxT5i\nSNMf2I7VOxYOZHcK9XWsHqTLsd6X5sBLXtujgFJYuf0/wJciUs7HfncBMUA7+3FWudLX9+l6+98G\ndq782SuOMlj5NcNYW2PMVqzPeJRYp1aHAd/kx2lNp9IGVd4tMsZMMsakGmNOGGNWGGP+sHuQtgND\ngFY+nj/eGLPSGJOMlaAaZVOuMpCQXSX20Wcz4EVjzEljzJ/A98BdXsV+N8bMMMZ4sBom2e3Ln2Sg\nHHAJgDFmgzEmMYtydwLfGmNWG2NOAi8CrUQkyqvMe8aYJGPMTmB+DmIaAdxpH0n1JlNSN8bsM8b8\nan8WR4B38f3+p3nXGHPIGHMi8wZjzBrgfaxE/hRwl/0eKpWfCiq3AGCMOY71e3k7i805+S1n9Rv6\nwBhz1BjzN7ARmG6M2WmMOQTMwGo8APQFvrBfo8cYM9Re3xTrP3sDfGaMSTbGjMGrVzoLPnMlEA2s\nN8aMtt/LEViNnhu9ynxnjNlqvyc/kbdcGQGcb4w5bYxZkE25O4E37fy1F6vx6p27TwJv269/InAK\nuMjPvn8A7hZrLFVJY8wK7425+D6B1Qh7034tWeXKL7Eac8uxGr+v+6mvSNEGVd7t9n4gIpeIyBSx\nB0Ni/TCq+Hi+d0PkOFbrPysHgBo+6okA9ttHUWn+wTqazG5f5/moL1vGmJnAV8CXwB4R+UpEymYT\n0z9ezzsCHPITU3avP62OHVg/2HeBdcaYeO/tIlJGrIH6u+z3fy6+3/80u/1sHwZcCEzKQQ+eUoFQ\nULnF29fA+SLSOdP6nPyWs/oN7fH6+0QWj9NiqgW8YJ9+Oiwih7HyXaS979hMPU7/kL2c5MrMz/eX\nK3Py3mXlfbvuOfapz/45jClzPPszHcTlJKbxQAfgMawD0Qxy8X0C2GOMOe2nzDfAZcCgHJQtUrRB\nlXeZu5W/BtYBde3BkK9jnXPPq9nAtWnjFrIQD1Sxz/OnOR/wdYrQl2NAaa/H4d4bjTGfGmMaY/1w\n6gPPZhNTrbQHdqOrYh5iSvMD8Bxnd2GD1cVeB2hmv/+ZB+xneRrAx/o0XwK/AV1F5JpziFWp3Cqo\n3HJmh9Yp7gFYvVTedefkt+zvN+TLbuAtY0wFr6W0MWYcVm9TVKbyvq5sng3cJCLZvTcZXotXfeec\nl4wxKVi9RVnmSmPMEWPMM8aY2linKV/INAwju5jykrvT9v0vMAur929kFkV8fZ9ylSft05CfAN8B\nAzKdRi3ytEEVeGWBJOCYiFyK7zEO52IGMA/4TUSuFGvemXJiDQS91+65WYk11qeEWAO27yPrH1JO\nrAZuEZFS9tgh7wHuzewlBKvhdRpIzaKO0cADInK5PQ7iPWChMSY2lzGl+RHryOvnLLaVxTp6OyQi\nlTm7y3kP53ilj4jch9VwvBd4BhiRqeGqVEHIr9yS2TCsU/re4xnz67ec5hvgMRFpKpYyIhJt/84W\nAS4RedweBN4LaOyjrg+xelq+F3tKGbEu2Blon/6aDDQQkd52fXcAdbHGVeXGGqxhCG4RuRFIn5bA\nfg0X2o27JKzxl9nlytdFpIqIVMUa+5bb3O3tBaCVMSar3sNsv092b9gBzv2qyM+AxcaYB7Eac4Nz\nFXUhpQ2qwHsOazDjUawjgLGBqNTu7r4ZmInVlXsEWIt1bn+OXaw31kDxRLvMy8aY+bnc5YdYRyN7\nsQaDev+4K2AdgRwGdmIdQX6cRczTsY52f7XLnI81ViBPjDHHjTGz7bEcmX2MNYj3ALAEa/C9t0+x\nrn45LCJnxZyZWFfvfIR1xdBxY8wPWFNVfJiHl6BUbuRLbsnM7nV5A2sMTNq6fPkte9W/DHgEqyf4\nENZl/33sbaewBug/ZG+7Cau3OLu69mNdzQuwQqyrJGcB+4Htxph9WAPoX8DKE88AXe1xXbnxpB3T\nYeBWwHs+r4uxhh38CywGBmYzSPstrIbZOqz88gdWozVPjDFxxpjF2Wz29316A/jRzpU3+9uXXaYt\n1ilGsMabXisivXMVfCEkJusLIZRSSimlVA5pD5VSSimlVB5pg0oppZRSxYaIDBVrYup12WwXsSY6\njRFr4mpfY/bSaYNKKZUrIlJBRMaLyCax7pV2rVgzSs8Sa6b+WWlX+eQ2QSmlVD4YhjWRaXY6Y41H\nrod1leSXOalUG1RKqdwaiDVR4yXAFVgTN74IzDHG1MO6WOJFu2yuEpRSSgWaPcGqr/sadgd+MJZl\nQAUfUxal0waVUuqciUh5rNtTfAdgz5x8GCsRDbeLDefMLUJylaCUUioIIsk4UW0sGSdazZJj7xYd\nEhaplx/mUYjLHewQfCpforT/QkF24MTRYIfgV8rpuHOe3DF5/3afv6/QKhf4q7MOsA9rrp8rgFVY\nl0lXN8ak3fYjEeteaJB9gvJ1i5BCTXNY3txUo0mwQ/Br+PM1gx2CT+We+TXYIfiVH/krrOqFD2P1\nhKcZYowZcq77OVeObVAppfKRJ9nnZhHpi++EFII1ueITxpg/RGQgZ07vAdbcaSKijQqlVGD5yV92\nrspLAyoO8G4tR5GDmeu1QaVUcZSa1WTNZ+QgIcVi3V/tD/vxeKwG1R4RqWGMSbBP6e21t+cqQSml\n1Fn85K8AmAg8LiJjgKuBJK+e92xpg0qpYsh4UvL2fGMSRWS3iFxsjNkMtAM22Ms9WDeFvQeYYD8l\nVwlKKaUyy2v+EpHRQGus+9/GYs0KHwpgjPkKmAp0AWKwbmV2X07q1QaVUsWRCcgR3hPAKBEJA7Zj\nJR0XME5EHgD+AXrZZXOVoJRS6ix5zF/GmNv9bDecuYVOjmmDSqniyM8YhJwwxqwGsho53C6LsrlK\nUEopdZYA5K/8oA0qpYqj/B+DoJRS+cOh+UsbVEoVQ3kdg6CUUsHi1PylDSqliiOHdpkrpZRfDs1f\n2qBSqjgKzKB0pZQqeA7NX9qgUqo4cmiXuVJK+eXQ/KUNKqWKI4cO6lRKKb8cmr+Kxc2RO3Zozfp1\nC9i0YRHP93fmldtOi/Hrr/+PXbv+ZNWqWenrKlYsz5Qpo1i37nemTBlFhQrlgxghlCgRxvS545i7\n6Dd+XzaJ/i89AcDAL95jxd+zmbPwV+Ys/JUGDS8JapxpnPQZm9Rkn4tyDid9b7LjhBgbtbqSgXO/\n4LPfv6LHI7ectb3rg934ZPbnfDh9IK//OIAqkVUBqF2/Du/8+gEfz/qMD6cP5LquLfIlvsU799Pj\nh8V0G76IoSt3nLX9wwWb6f3jUnr/uJTuPyym5Vfz0rd9umgLt4xcws0jlvDB75uwZiEpeE74nMG5\n+avIN6hcLheDBr5D1+g+NLyiDb179+DSS+sFO6wMnBjjiBE/0a3b3RnW9ev3GPPmLeayy1oxb95i\n+vV7NEjRWU6dOs3N0ffStkUP2rW4ibbtW3BVkysAeOu1/6Ndy5to1/Im1q/dFNQ4wYGfcWqq70U5\nguO+N1lwQowul4sH/vsw79zzFs+0f5zm3VoSVS/jjYt3rN/BC12fpV+np1g2dQl3vXQvAKdOnOKz\nZz7l2Rue4J273+LeNx6gdLnzAhqfJ9Xw/vxNfN79Sn7ucx3TtySy7cC/Gcr0u/5ixt5xLWPvuJbb\nrqhJu7rVAFidcJjVCYcZd8e1/HTntazfc4RVcYcCGl9OOOFzTufQ/FXkG1TNml7Jtm072bFjF8nJ\nyYwbN4Fu0R2DHVYGToxx0aLlHDp0OMO66OgbGDlyPAAjR46nW7cOwQgtg+PHjgMQGhpCSGhI0I7c\n/HHcZ+xJ9r0oR3Dc9yYLToixbqN6JO5MZO/uPaQkp7B40kKa3NAsQ5n1S9dy+uRpALb8tZlKNSoD\nkLAjnsSd1l2QDu09SNL+JMpVKhfQ+NbtSaJmhdJElS9NqNtFx3rhzN++L9vy0zcn0umicAAEOJ2S\nSnJqKqc9qaSkplKpdFhA48sJJ3zO6Ryavwq8QSUiBXrLiYjIcHbHxqc/jo1LICIivCBD8KswxAhQ\nrVoVEhOte90mJu6lWrUqQY7IOmqas/BX1scs5vd5S/hz1d8AvPTa08xbPIEB775IWFhokKN04Gds\nUn0vKlsFmcMc973JghNirBRemQMJ+9MfH0w4QOXwytmWb9f7Bv6av+qs9XWvqEdIWAh7/kkMaHx7\n/z1F9TIl0h9XL1OCfcdOZVk2/sgJ4o+coGlUJQCuqFGBJlGVuOHbBXT4bgHXnV+FCyqVCWh8OeGE\nzzmdQ/NXMHqo3spug4j0FZGVIrIyNfVYQcakcsEJnUGpqam0a3kTjeq3pnHjy7nk0nq889bHNG/S\nmY5telKhYgUef/qhYIfpPJ4U34vyRXNYIdbyplZc0LAuE7/+NcP6CtUq8sQnz/BFv0FB7emesSWR\ndnWr43YJALsOH2fHoWPMuL8lM+5vyfLYg/wZhFN+juLQ/JUvV/mJyN/ZbQKqZ/c8Y8wQYAhASFhk\nQL7R8XGJ1IyKSH8cFVmD+PjAHn3kVWGIEWDv3v2Eh1cjMXEv4eHV2Ldvv/8nFZAjSUdZtPAP2rRv\nyZefDQXg9Olkxoz6hUefuD/I0TnwM07RRpMvTslhjvveZMEJMR5MPEDlGmd6zCvVqMyBxANnlWvY\n/ApufvxW3uj1Cimnz/wGSpUpxUvfv8boD0ey9a8tAY+vWpkS7Pn3TI/Unn9PUfW8ElmWnbFlDy+2\nOXMhzbxte2kYXp7SYdZ/181rVebvxCQaR1YMeJy+OOFzTufQ/JVfPVTVgbuB6CyWs7/l+WjFytXU\nrVuH2rVrEhoaSq9e3Zk0eWZBhuBXYYgRYPLkWfTp0xOAPn16MmnSLD/PyF+VK1ekXPmyAJQsWYJW\nba4jZst2qlWvml6m843t2LQx8AnyXDntMzbG43NRzshhTvveZMUJMcas2UqNOjWoVrMaIaEhNI9u\nycpZyzOUqd2gDn3fe4QPHniHIweS0teHhIbQf8hL/P7zPJZNXZIv8TWoXo5dh48Tl3SCZE8qM7Ym\n0vqCqmeV23HwGEdOJXNF+JkrqMPLlmRV3CFSUlNJ9qTyZ9xh6lQM7KD5nHDC55zGqfkrv+ahmgyU\nse9Gn4GIzM+nfWbJ4/Hw1NOvMnXKj7hdLoYNH8uGDcH/D9abE2P84YfPaNnyWqpUqUhMzB+8/fbH\nfPjhF4wa9SX33tubXbviuPPOR4IaY/Xwqgz66n3cLjculzDh1+nMmjGfnycNo3LlSojAurWb6P/M\nm0GNExz4GetpPX8ckcMc973JghNiTPWk8t3rQ3jlhzdxuV3MGzeH2K276f3sHWz7O4aVs5dz18v3\nUbJ0KZ774nkA9sfv54MH3+Hars25tFkDylYoS5uebQEY3G8QOzecPbVBboW4XLzQ+mIenfAnqamG\n7g0iuLByGb5YFkP9auVofYF1Rd+MLYl0vCgcEUl/bvu61VkRe5Beo5YBcF2tyrTKojGW35zwOZ8J\nxpn5S5x6VVSgTvkVZyEud7BD8Kl8idLBDsGvAyeOBjsEv1JOx4n/UhmdmDPE5++rVLu+51ynykhz\nWN7cVKNJsEPwa/jzNf0XCqJyz/zqv1CQFaX8pTOlK1UcOfQITyml/HJo/tIGlVLFkU6NoJQqrBya\nv7RBpVRx5NCrZJRSyi+H5i9tUClVHDm0y1wppfxyaP7SBpVSxZFDu8yVUsovh+YvbVApVRw59AhP\nKaX8cmj+0gaVUsVREO/IrpRSeeLQ/KUNKqWKI4/Ohq6UKqQcmr+CcXNkpVSwpab6XpRSyqkCkL9E\npJOIbBaRGBF5MYvt54vIPBH5S0T+FpEu/urUHiqliiOHjkFQSim/8pi/RMQNDAZuAGKBFSIy0Riz\nwavYq8A4Y8yXIlIfmArU9lWvNqiUKo4c2mWulFJ+5T1/NQNijDHbAURkDNAd8G5QGaCc/Xd5IN5f\npdqgKsJSHXppaZo654UHOwS/CsO9/HJFT+sph1uUtDXYIfjlbvV4sEPww/n38ssVP/lLRPoCfb1W\nDTHGDPF6HAns9nocC1ydqZo3gZki8gRwHtDeX1jaoFKqONJTfkqpwspP/rIbT0N8FvLvdmCYMeYj\nEbkWGCEilxmTfU+FNqiUKoZMqs+btSullGMFIH/FATW9HkfZ67w9AHQCMMYsFZGSQBVgb3aV6lV+\nShVHnhTfi1JKOVXe89cKoJ6I1BGRMOA2YGKmMruAdgAicilQEtjnq1LtoVKqONIeKqVUYZXH/GWM\nSRGRx4EZgBsYaoxZLyIDgJXGmInAc8A3IvIM1gD1e40xPnesDSqliiOH3q1dKaX8CkD+MsZMxZoK\nwXvd615/bwCan0udespPqeLI4/G95ICIuO1J7ybbj4eJyA4RWW0vjez1IiKD7An0/haRxvn4ypRS\nRV0A8ld+0B4qpYqjwJzyewrYyJm5WgD6G2PGZyrXGahnL1cDX3L2JcpKKZUzDh2yoD1UShVHeTzC\nE5Eo4Ebg2xzsrTvwg7EsAyqISI28vQClVLHl0B4qbVApVQyZ1FSfi4j0FZGVXkvfTFV8CjwPZJ6T\n5R37tN4nIlLCXpfVJHqR+fPKlFJFnb/8FSx6yk+p4sjPUZyvifFEpCuw1xizSkRae216CUgEwuzn\nvgAMCES4SimVzqG3ztIGlVLFUd7GIDQHutl3Xy8JlBORkcaYPvb2UyLyPdDPfpyTSfSUUipndAxV\n8HTs0Jr16xawacMinu//WLDDyZLTYyxfvhxjRn/N2r/n8/eaeVx9tTMu1LrtoZ78OO97Rs39ngFf\nvEZYiTCuan4lw2cMYdTc73nt0xdxu93BDhNw2Gec4vG9+GCMeckYE2WMqY01Id5cY0yftHFRIiJA\nD2Cd/ZSJwN321X7XAEnGmIR8e21FjKO+N9kIRoxt2rVg4YopLPlzOo8//eBZ28PCQvlq6Ecs+XM6\nU2aPIer8iPRtlza4iEkzf2T+0onMXfwbJUqEAfDj+K+ZvegX5i+dyAcfv4HLFbj/Ihf9tYHoJwZw\n42Nv8t0vM8/aHr/3IA++OYhbnnmX+1//lMQDhzJs//f4Cdo/9CrvfjMuYDGdK8d8F/OQv/JTkW9Q\nuVwuBg18h67RfWh4RRt69+7BpZfWC3ZYGRSGGD/+6C1mzJxPw8tbc1WTDmzaFBPskKgaXoVeD9zC\nfZ0f5s629+FyuehwUzteH/gSrz4ygDvb3kdi3B669OoY7FCd9xnnz6DOUSKyFliLdYuGt+31U4Ht\nQAzwDfBoXsMvLhz3vclCMGJ0uVy8++Gr3NnzYVpdHU2Pnl246OILM5S5/a5bSDp8hOsad2LIF8N5\n9c3nAHC73Xw+5ANeePYtWl/bjVu63kNysjWvUd/7nqV9i5tpfW03KlepSHSPwOQOjyeVd78Zx5ev\nPMpvn77KtEWr2LY74zHFRz/8SnSrZvz8ycs8fGtnBo3MOHH356OncFX9jK+xIDnqu1jcBqWLyCUi\n0k5EymRa3ym/9pmVZk2vZNu2nezYsYvk5GTGjZtAt+jg/wfrzekxlitXlhYtr+b770cDkJycTFLS\nkSBHZXGHuClRsgRut5uSpUpy8vhJkk8ns3t7LADLf19Jmy7XBzlK533GgRrUaYyZb4zpav/d1hjT\n0BhzmTGmjzHmX3u9McY8Zoy50N6+Mp9eVsBo/sq5YMR45VUN2bl9F7v+iSU5OZkJP0+jY5e2Gcp0\n6tKWcaN/A2DyhJm0bHUNAK3aNmfjui1sWLcZgEOHkki1v/P/Hj0GQEhICKFhofiZGDvH1sXs5Pzw\nKkSFVyE0NIROLRozb8XfGcps353A1Q0vBqDZZRcxb8Xa9G0btu3iYNIRrrvi0oDEkxtO+i46dVB6\nvjSoRORJYALwBLBORLp7bX43P/aZnYjIcHbHxqc/jo1LICIivCBD8MvpMdapXZP9+w7y7Tcfs/yP\n6Xz15f9RunSpYIfFvsT9jPpyLL+tGMfk1T9z7Oi/zJ44D3eIm0sutxJT266tqBZRLciROvAzTkn1\nvRRjmr/OTTBiDK9Rnbi4xPTHCfGJhNeodlaZeLuMx+PhyJGjVKpUgQvr1sJgGP3zEGb+Pp5Hn7w/\nw/NG/zyEtTEL+ffoMSZPOPvUXG7sOZhE9SoV0x9Xr1SRvQeSMpS5qHYks5etBmDOH2s4duIkh4/+\nS2pqKh8O/4Vn77kpILHklqO+iw7NX/nVQ/UQcJUxpgfQGnhNRJ6yt0k+7VPlE3dICFdeeRlfDxlB\ns6s7cez4cUeM5ShbvgzXd2zOzVffRtcrb6Fk6VJ0uvkGXntkAE+/9RjfTfmS4/+eSD/6VF5Mqu+l\neNP8VYS53SE0u6Yxjz30PN079aFz1/a0uP6a9O2339KXRhe3okSJMFpcX3Dzzz53z02s2hBDr37v\ns3J9DNUqVcDlcjF2+kJaNG5AeOWK/ispLhyav/LrKj+XV3f/TvvS6vEiUgsfCcme66YvgLjL43Kd\nl+dA4uMSqRl1ZjBiVGQN4uMTfTyj4Dk9xri4BGJjE1ix4i8AfvllCv0d0KBq2vIq4ncncPigdaQ3\nf+oCGjZpwPRfZvGfm54EoFmrJtS8ICqYYQLO+4xNMe+F8iNX+QsCn8Oc9r3JSjBiTEzYQ2Tkmd6R\nGhHhJCbsPatMRGQ4CfF7cLvdlCtXloMHD5MQn8iyJSs5ePAwAHNnLaDhFfVZtGBZ+nNPnTrNjKlz\n6dilLQvmL81zvNUrlWfP/jODzPccPES1yuUzlKlWqQKfPP8QAMdPnGL2stWUO680a7bs4M+N2xg3\nfSHHT54iOcVD6ZIlePqu7hQkJ30XnZq/8quHak/afbwA7OTUFWugasPsnmSMGWKMaWKMaRKIxhTA\nipWrqVu3DrVr1yQ0NJRevbozaXJgunEDxekx7tmzj9jYeC666AIA2rZpwcaNW4McFeyJ28tljetT\nopQ1f2STFo3ZGfMPFStXACA0LJS7Hr2dX0dM9FVNgXDcZ5xqfC/FW67yl102oDnMcd+bLAQjxtV/\nrqPOhbWoWSuS0NBQut/SmRnT5mUoM2PaPHrd3gOArt07sGjBHwDMn7OYS+tfRKlSJXG73VzTvClb\nNsdQ+rzSVKteBbAGrrfr0IqYrTsCEm+DurX4J2EfsXv2k5ycwvRFf9K6yeUZyhw68m96b/q3v8zg\nprZWr9n7T9/LzK//y/SvBvDc3TcR3apZgTemwGHfRYfmr/zqobobyHA7aGNMCtal01/n0z6z5PF4\neOrpV5k65UfcLhfDho9lw4YtBRmCX4UhxmeeeY3hwz4jLCyMHTv+4cGHngt2SKz/ayNzp/zO8Bnf\n4EnxsGXdVn4bOZmHX3iAFrLJM+cAACAASURBVO2vRVzCL8MnsmrxX8EO1XmfcRAvLS4ENH+dg2DE\n6PF4eLn/O4z++RvcbhdjRv7Klk0x9H/5cdb8tZ6Z0+YxesTPfPb1Byz5czqHDx3mP/db06IlJR3h\n68HDmTZ3HMYY5sxawJyZC6hStTLDRw8mrEQYLnGxeNFyfhg6NiDxhrjdvPxgLx7572A8qYYeba+h\n7vk1GDx6MvXrnk+bppezYv1WBo2ciAg0rl+XVx7qFZB9B4qjvosOzV8SqKsYAi0kLNKZgRUiLnH2\ncI/GlesGOwS/Vu4Pfk+cPymn4875gz7ycEefv69yX89w9penENAcljdVS5f3XyjIdi4aGOwQfDqv\n8b3BDsGvopS/dKZ0pYojPa2nlCqsHJq/tEGlVDHk1EGdSinlj1PzlzaolCqOHHqEp5RSfjk0f2mD\nSqliyKQ4MyEppZQ/Ts1f2TaoRORXINuojTE350tESqn859AjvEDR/KVUEebQ/OWrh+rzAotCKVWg\nnHqEF0Cav5Qqopyav7JtUBlj5qT9LSJhwPnGmJgCiUopla+cmpACRfOXUkWXU/OX35nSReRGYC0w\ny37cyO5OV0oVVql+liJC85dSRZBD81dObj0zALgaOAxgjFkNOH9GRqVUtkyK76UI0fylVBETiPwl\nIp1EZLOIxIjIi9mU6SUiG0RkvYj86K/OnFzll2yMOSwZZ912Zn+bUipHgnhD9oKm+UupIiav+UtE\n3MBg4AYgFlghIhONMRu8ytQDXgKaG2MOiUg1f/XmpEG1UUR6AS4RqQM8CSzz8xyllIMVsV4oXzR/\nKVXEBCB/NQNijDHbAURkDNAd2OBV5iFgsDHmEIAxZq+/SnPSoHoceB3rzOSvwAzglXMKXaks/Hkg\nhiVVmwQ7DJ+uDXYA+aQY9VBp/iqkDp38N9gh+CXl/XZaBJXT7+eaWwHIX5HAbq/HsVhDA7xdBCAi\niwE38KYxZrqvSv02qIwxx4AXROQt66E5cS5RK5UdpzemijLjKZqJNjPNX0oVPf7yl4j0Bfp6rRpi\njBlyjrsJAeoBrYEoYIGINDTGHPb1BH+BNQa+A6raj/cADxlj/jzH4JRSDpGaUjwaVJq/lCp6/OUv\nu/HkqwEVB9T0ehxlr/MWC/xhjEkGdojIFqwG1orsKs3JVX7fA88aY6KMMVHAc/Y6pVQhZVJ9L0WI\n5i+lipgA5K8VQD0RqWPPU3cbMDFTmd+weqcQkSpYpwC3+6o0J2OoUo0x89JfiDHzRaRopVylipnU\nYnLKD81fShU5ec1fxpgUEXkca0ylGxhqjFkvIgOAlcaYifa2DiKyAfAA/Y0xB3zV6+tefpfbf84X\nkcHAaKzLjXsDc/P0apRSQWVSi3aDSvOXUkVXIPKXMWYqMDXTute9/jbAs/aSI756qAZneny51986\nj4tShVgx6KHS/KVUEeXU/OXrXn4tCzIQpVTBKeo9VJq/lCq6nJq/cjKGChHpCDQASqatM8a8m19B\nKaXyl1OP8PKD5i+lihan5q+cTJvwBVABuB7r6phb0JmGlSrUnJqQAk3zl1JFj1PzV06mTWhhjLkD\nOGCMeQ1rNlG9uahShViqEZ9LEaL5S6kixqn5Kyen/NJmFj4pIuHAASAi/0JSSuW3VE9OjqWKBM1f\nShUxTs1fOWlQTRORCsCHwGqs+RiG52tUSql8ZYrPdW6av5QqYpyav/w284wxbxpjDhtjfgLqAA2B\nn/M9sgDq2KE169ctYNOGRTzf/7Fgh5Mlp8e4ZfNS/lw1mxXLZ7B0yZSgxSElQrl08v+oP/MTGswZ\nRMRztwEQVrMal0z6H5ct+pILvuiHhFrHCtUf6kaDuZ9Rf9anXDRmAGGRVYMSd4kSJViyeDKrVs5i\n9eq5vP76c0GJI43H4/K5FBWavwpGMGK84YZW/P33PNavX0C/fo+etT0sLIwRIwazfv0CFiyYQK1a\nUQBUqlSBGTPGsH//Rj75ZECG54SGhjJ48PusXTufNWvm0qNH54DFu2j5aqLvfYoudz/Bt6N/O2t7\n/J59PNh/ADc/1I/7nn2TxH3WHJLLV6+j58P905erOt/JnMXLcx1Hhw6tWbf2dzZsWET/fmd/VmFh\nYYwa+QUbNixi0cJJ6e8bwPP9H2PDhkWsW/s7N9zQKn39448/wF9/zmb1X3N44okH0tePGvkFK5bP\nYMXyGWzZvJQVy2fkOm5vTs1fObrKL419Y9ETIrIaOD9/Qgosl8vFoIHv0KnL7cTGJrBs6VQmTZ7J\nxo1bgx1ausIQI8ANHW7lwIFDQY3BnEpmc6/XST1+Eglxc/Gv75E070+qP9SdPd9M5NDERZz/3n+o\nclt79o2YzvH129nY5TlST56m6l2diHrlHrY/+mGBx33q1Clu6NCLY8eOExISwu/zf2XG9Hn8sTw4\nt5QzeRxnICIlgQVACaw8Mt4Y84aI1AHGAJWBVcBdxpjTIlIC+AG4Cuu0W29jzM48BXGONH/lj2DE\n6HK5GDjwbW688U5iYxNYvHgSkyfPYtOmM/u8997eHD6cRIMG13PrrdG8/fZL3HXXY5w8eYq33vqI\n+vUvpkGDizLU++KLT7Bv334aNmyNiFCpUoWAxOvxpPLOZ98x5INXCa9amdsee4k21zXhQq/Gyodf\njyD6huvp3qE1f/y1joHf/ch7Lz5Bs0aXMf7r/wMg6ci/dLnnCa676opcxZH2vnXpcgexsQksXTKF\nyZNnstHrfbvvvts4dDiJ+vVb0OvWbrz7zsvc2edRLr2kHr16dadRo7ZERFRn2rTRNGhwPZdeUo8H\n7r+d65p35fTpZCZPHsnUqXPYtm0nd/Y509D94IPXOJJ0NJfvYEZ5zV/5JbdNOb+vRkSaiUhT++/6\nIvKsiHTJ5f5yrVnTK9m2bSc7duwiOTmZceMm0C26Y0GH4VNhiNFJUo+fBEBC3EiIG4yhbPOGHJqy\nBIADP82jQserATi6ZB2pJ08D8O+fmwmrUTk4QQPHjh0HIDQ0hNDQUEwQ+609qeJzyYFTQFtjzBVA\nI6CTiFwDfAB8YoypCxwC0g5XHwAO2es/scsFi+avAApGjE2bNsqwz59+mkR0dIcMZaKjOzBy5HgA\nfvllKm3aNAfg+PETLFmyglOnTp5V7z339OJ//7PmhDXGBOwAcu3mGM6PCKdmRHVCQ0Po3Po65i3O\neI/d7f/EcnWjywBo1qgB85asPKuemQuW0aLplZQqWSJXcWR+38aNm5Dl+zZixE8A/PzLFNq0aZG+\nfty4CZw+fZqdO3ezbdtOmjZtxCWX1GX58tWcOHESj8fDwgXLsuzZ63lLNGPHTchV3JkFIH/li9w2\nqHz+TyAibwCDgC9F5D3gc+A84EUReSWX+8yViMhwdsfGpz+OjUsgIiK8IEPwqzDEaDBMnfIjy5ZO\n5YEH7gxuMC4X9Wd8whVrhnNk4RpO7UzEc+QYeKxbtJ1OOEBYeKWznlb19vYkzQtOjxBYR4crV8wk\nPu5vZs9ZwPIVfwUtltRU8bn4Yyz/2g9D7cUAbYHx9vrhQA/77+6cGbs0HmgnIsHKfJq/AigYMUZE\nhBPrtc+4uAQiIqpnW8bj8XDkyFEqV66YbZ3ly5cD4I03+rF06RRGjfqSatWqBCTevfsPEl7tzMFc\n9aqV2XPgYIYyF11Qi9mLrFN5cxYt59jxExzO1KMzff5iurRtnus4IiNqELs7If1xXFwiEZE1MpUJ\nJzbWKuPxeEg6coTKlSsSEVkjfT1AXGwikRE1WL9hMy1aNKNSpQqUKlWSTp3aEhWV8bqPFi2uZu/e\nfcTE7Mh17N7ymr/yi697+f1K1olHsLrzfemJddRaAkgEoowxR0TkQ+AP4J1s9tkX6Asg7vK4XOf5\nfQGqYLRpczPx8YlUrVqZaVNHs3lzDIsW/RGcYFJT2dDxGdzlzuPCb1+kZN0ov0+pdHMrSl9el809\nC/T/wwxSU1Np0rQD5cuXY/xP39GgwcWsX785OLEEoMtcRNxYp/XqYt3qZRtw2BiTYheJBSLtvyOB\n3ZB+Y9IkrDyyP8+BZB1bgecve7+awwqpkBA3UVERLFu2ihde+C9PPvkg77//Kvff/3SB7L/fw3fx\n7udDmTBjPlddfinVqlTC5T7T57HvwCG27tjFdU1yd7ovv2zaFMP/ffgFU6f8yLFjx1nz93o8Hk+G\nMr17dw9Y7xQEJn/lB19jqD7P5TaAFGOMBzguItuMMUfAGsPg607vxpghwBCAkLDIgJwPiY9LpKZX\nazkqsgbx8YmBqDpgCkWMdjz79h1gwoTpNG3aKHgNKpvnyDGOLlnLeVddjLvceeB2gSeVsBqVOZ14\n5uivbIvLqfFETzb3fBVzOsVHjQUjKekI839fTIcOrYPWoPKk+u6c9m4Y2IbYv8909m+8kX0V3a/A\nJYGOMw8KPH/ZZQKawwpFbghCjPHxiRl6QSIjaxAfvyfLMnFxibjdbsqVK+vzFN6BA4c4duw4v/02\nDYBffpnCvffeFpB4q1WpROLeA+mP9+w7QPXKlc4q8+mb/QA4fuIksxb+QbkyZxrkM35fStvmzQgN\nOaehzxnExScQVfNMj1RkZDjxcQmZyiQSFVWDuLgE3G435cuV48CBQ8THJRAV5fXcqHDi4q3nDhs2\nhmHDxgDw3wEvEOtVp9vtpkf3zlxzbeDOmPvLX8GSbVTGmDm+Fj/1nhaR0vbfV6WtFJHygM+EFGgr\nVq6mbt061K5dk9DQUHr16s6kyTMLMgS/nB5j6dKlKGP/sEuXLkX79tcHrSEQUqmc1XgCpGQY5Vo2\n4uTWWI4uWUvFG68DoPKtbTg80+o6L9WgDrXef5SY+98l5UBSUGIGqFKlUvophZIlS9K+3fVs3rwt\naPEYf4sxQ4wxTbyWIdnWZcxhYB5wLVBBRNIyfhQQZ/8dB9QEsLeXxxqcni80fxWcYMS4cuWaDPu8\n9dZoJk+elaHM5Mmz6NOnJwA339yF+fOX+K13ypTZtGp1LQBt2jQP2MD6yy6+kH/iEohN2EtycgrT\n5i+h9XVNMpQ5lHSE1FTr6/Xt6F+5qVObDNunzc3b6T44+33r1at7lu/bXXfdCsAtN9/I/PmL09f3\n6tWdsLAwateuSd26dVixYjUAVatanb41a0bQo0dnxow5cxVju3Yt2bx5G3GZGm554S9/BUvum7q+\nXW+MOQVgjPFOQKHAPfm0zyx5PB6eevpVpk75EbfLxbDhY9mwYUtBhuCX02OsXr0qP437FrC6xceM\n+Y2ZM+cHJZbQ6hWp88lT4HYhIhycvJikOSs5sXU3F37xHJHP38nxddvZP8ZKEjVfvRf3eSW58Kvn\nATgdt4+Y+wv+Nm41alRn6Hef4na7EJeL8eMnMXXq7AKPI01ej/BEpCqQbIw5LCKlgBuwBprPwzpl\nNgbrt57Wzz/RfrzU3j7XBHNUvm+av85BMGL0eDw8/fRrTJo0ArfbzfDhY9m4cQuvv/4sq1atZcqU\nWQwbNpahQz9l/foFHDx4mLvvfjz9+Zs3L6Zs2bKEhYUSHd2Rrl37sGnTVl599T2GDv2U//u/N9i/\n/yB9+wZmepMQt5uXn7if/7z4Dp7UVG7q1Ia6tWvy+bCxNLjoQtpc14QVazYw8LsfEYSrLr+UV7ym\nH4hL3Evivv00ubx+nuJIe9+mTB6Fy+1i+LCxbNi4hTde78eqP9cwefIsvv9+DMO+H8iGDYs4dPAw\nfe6yrtTbsHEL48dPYs2auXhSPDz11KvpDcCxY4ZQuXJFkpNTePKpV0hKOpK+z163dmPsuLOnicjT\n63BoD5U4NacF6pRfceYK2pjfnFlStYn/QkF27d4V/gsFWfLpuHP+oBeG9/T5+2qZON5nnSJyOdYg\nczdWT/c4Y8wAEbkAqzFVCfgL6GOMOWVPszACuBI4CNxmjNl+rnEXJprD8ibE5Q52CH4d2Top2CH4\nVKbujcEOwa/Tp2ILPH/llxz3UIlIibSjNqVU4ebJ46BOY8zfWI2jzOu3A82yWH8SuDVPO80DzV9K\nFR15zV/5xW+/mT0fy1pgq/34ChH5LN8jU0rlGw8un0tRoflLqaLHqfkrJ3seBHTFHkBqjFkDtPH5\nDKWUo6X6WYoQzV9KFTFOzV85OeXnMsb8k2kOPk92hZVSzufxP1l4UaH5S6kixqn5KycNqt0i0gww\n9kR+TwDOusxEKXVOilgvlC+av5QqYpyav3LSoHoEq9v8fGAPMNtep5QqpDwOvwI0gDR/KVXEODV/\n+W1QGWP2AoGZLlYp5QipDu0yDzTNX0oVPU7NX34bVCLyDVlMPmqM6ZtFcaVUIVBcBhFp/lKq6HFq\n/srJKT/v6ZxLAjdh3+RUKVU4ObXLPB9o/lKqiAlE/hKRTsBArMmJvzXGvJ9NuVuA8UBTY8xKX3Xm\n5JTf2EyVjwAW5TRopZTzOHVQZ6Bp/lKq6Mlr/rIvUBmMdcusWGCFiEw0xmzIVK4s8BTwR07qzc0M\nWHWA6rl4nlLKIVJEfC5FmOYvpQq5AOSvZkCMMWa7MeY01u2yumdR7r9Y9yg9mZNKczKG6hBnxiC4\nsO7D9WJOKlfBlerQ+zSmuWL1x8EOwa+QmkVzDkhnfzMCR/NX4ZWS6tSRMl7cOb57W1C4pOjc9cBb\nAPJXJBlP/ccCV3sXEJHGQE1jzBQR6Z+TSn1+G8SaDe8KIM5elergO8QrpXIopUh3Qlk0fylVNPnL\nXyLSF/C+8GSIMWZITusXERfwMXDvucTls0FljDEiMtUYc9m5VKqUcrbi0KrQ/KVU0eQvf9mNJ18N\nqDigptfjKM4ceAGUBS4D5tt3WQgHJopIN18D03PSH7haRM66q7xSqvBKEd9LEaL5S6kiJgD5awVQ\nT0TqiEgY1lx1E9M2GmOSjDFVjDG1jTG1gWWAz8YU+OihEpEQY0wKcCXWCPhtwDFArP2ZxjkKWynl\nOEX9Kj/NX0oVXXnNX8aYFBF5HJiBNW3CUGPMehEZAKw0xkz0XUPWfJ3yWw40BrrlpmKllHN5ilYv\nVFY0fylVRAUifxljpgJTM617PZuyrXNSp68GldgVbcthfEqpQqIQXD+VV5q/lCqinJq/fDWoqorI\ns9ltNMY4/5p3pVSWUot+D5XmL6WKKKfmL18NKjdQBhx6F0KlVK6lBDuA/Kf5S6kiyqn5y1eDKsEY\nM6DAIlFKFZhiMG2C5i+liiin5i+/Y6iUUkVPEZsaIStF/xUqVUw5NX/5alC1K7AolFIFyqlHeAGk\n+UupIsqp+SvbiT2NMQcLMpD81LFDa9avW8CmDYt4vv9jwQ4nS06P0UnxHTn6L8+88jbRtz9E9B19\nWb1uI0lHjvLgUy/TpfcDPPjUyyQdOZrhOWs3buaK629k5ryFBRrrV1/9H//8s4qVK2emr3v33ZdZ\nvXoOy5dPZ+zYrylfvlyBxgSQgvG5FHaavwqWxujfouV/En3343S581G+/fGXs7bHJ+7lwWff4OYH\nnuG+p18jcd/+9G0ff/UDPe59im73PMF7g74lkHdQuuGGVvz99zzWr19Av36PnrU9LCyMESMGs379\nAhYsmECtWlEAVKpUgRkzxrB//0Y++STj2fWePaNZsWIGf/45m7fffilgsaZxav4qmndO9OJyuRg0\n8B26Rveh4RVt6N27B5deWi/YYWXg9BidFt/7n35F86ubMGn0N/wyfDAX1KrJtyPGcU2TRkwd+x3X\nNGnEdyPHpZf3eDx88sX3XNe04OdyHDHiJ7p3vyfDujlzFnLVVR1o1qwTW7fuoH//s5NYfvP4WZQz\nOO23lxWN0T+Px8M7A7/hi/dfZcKwgUybs5BtO3dnKPPhV8OJ7tCaX777hP/c3YuB34wCYPW6Tfy1\nbiM/f/cxvw79lHWbY1i5Zn1A4nK5XAwc+Dbdu99Do0bt6NWrG5dckvF9uffe3hw+nESDBtfz2Wff\npjeQTp48xVtvfcSLL76ToXylShV4772X6dz5dho3bk94eFXatGkekHjTODV/FfkGVbOmV7Jt2052\n7NhFcnIy48ZNoFt0x2CHlYHTY3RSfEf/PcaqNeu4xd5/aGgo5cqWYd7CpXTv3B6A7p3bM3fB0vTn\n/Dh+Ije0bk6lihUKPN7Fi5dz8ODhDOvmzFmIx2P97Jcv/4vIyBoFHleq+F6UMzjpt5cdjdG/tZti\nOD+iBjUjwgkNDaVz2xbMW7w8Q5ntO2O5unFDK94rLzuzXYRTp5NJTknhdHIKKSkeKgcolzVt2ijD\n+/LTT5OIju6QoUx0dAdGjhwPwC+/TE1vHB0/foIlS1Zw6tTJDOXr1DmfmJid7N9vdRLPnbuIHj06\nByTeNE7NXwXWoBKRHwpqX94iIsPZHRuf/jg2LoGIiPBghJItp8fopPji4hOpWKE8r77zMT3vfYzX\n3/uU4ydOcuDQYapWqQRAlcoVOXDIasTs2befOQuW0PumG4MSrz93392LGTPmF/h+PRifi8pI81f2\nNEb/9u4/QHi1yumPq1etzJ79Gc9KX3RhbWYvWAbAnIV/cOz4CQ4nHaVRg4tpduVltL3lAdr2fIDm\nTRtxgX3aLa8iIsKJ9Xpf4uISiIionm0Zj8fDkSNHqVy5YrZ1btv2D/XqXUCtWlG43W6iozsQFRUR\nkHjTODV/+RqUnmsikvk+OAK0EZEKAMYYvR2EypUUj4eNW2J4+ZlHuLzBJbz36Vd8N2JchjIign2H\ncD4Y+DXPPHI/LpfzOmOff/5xPJ4Uxoz5tcD3XdTv5ZcXmr9UMPR75B7eHfQNE2bM46rL61OtSiVc\nbhe74hLY/k8ss3/6BoCH+r3Fqr83cNXl9YMccdYOH07iySdfYcSIwaSmprJs2SouuKBWQPfh1PyV\nLw0qIArYAHyLNSBfgCbAR76eJCJ9gb4A4i6Py3VengOJj0ukplfrOCqyBvHxiXmuN5CcHqOT4guv\nVoXqVatweYNLAOjQugXfjhxH5YoV2Lf/IFWrVGLf/oNUqlAegPWbttL/jfcBOJR0hIVLV+B2u2l3\n/XVBiT9Nnz496dKlHZ073x6U/WsvlE+5yl8Q+BzmpN9edjRG/6pVqUzi3gPpj/fsO0B1u0f9TJlK\nfDrgBQCOnzjBrAVLKVfmPH6ePIvL619E6VKlAGjRrDFr1m8OSIMqPj4xQ+9RZGQN4uP3ZFkmLi4R\nt9tNuXJlOXDgkM96p06dzdSpswF44IE78HgC2wRyav7Kr8P2JsAq4BUgyRgzHzhhjPndGPN7dk8y\nxgwxxjQxxjQJRGMKYMXK1dStW4fatWsSGhpKr17dmTR5pv8nFiCnx+ik+KpUrkR4tars+CcWgGWr\nVnNh7fNp3eIaJkyzfsATps2mTctrAZgxfhgzfx7OzJ+H06F1C17t91jQG1M33NCKZ5/9Dz17PsCJ\nEyf9PyEfpPpZirlc5S8IfA5z0m8vOxqjf5ddUpd/4hKITdhDcnIy0+YuovV1TTOUOZR0hNRU69f3\n7ahfuKmzNfNHjWpVWLlmAykeD8kpKaxasz5gp/xWrlyT4X259dZoJk+elaHM5Mmz6NOnJwA339yF\n+fOX+K23alXr9GaFCuXp2/cuvv9+dEDiTePU/JUvPVTGmFTgExH5yf53T37tyx+Px8NTT7/K1Ck/\n4na5GDZ8LBs2bAlGKNlyeoxOi+/lZx7hhbf+R3JKMjUjavDfl5/BGMNzr73LL5NnEBFejY/++3LQ\n4vM2fPggWra8lipVKhITs4z//vcT+vd/lBIlwpg8eSRgDUx/8slXCjQupx7hOYHmr3OjMfoX4nbz\n8pMP8p/nB+BJTeWmzu2oW+d8Ph86mgYXX0ib5s1YsXodA78ZhQhcdXl9XnmqLwA3tLqWP/5ay833\nP42I0LzplWc1xnLL4/Hw9NOvMWnSCNxuN8OHj2Xjxi28/vqzrFq1lilTZjFs2FiGDv2U9esXcPDg\nYe6++/H052/evJiyZcsSFhZKdHRHunbtw6ZNW/noozdp2NDqQXv33U+JidkRkHjT43Zo/pJAzmeR\n7U5EbgSaG2Ny/L9cSFikM98xFTAn4gt2TqjcKFezTbBD8OvEiX/O+bqWJ2v39vn7GrRzrF7rZ8tN\n/gLNYcXB8Z3O6onLrNwFgb26Lj+cPLmryOSvAjnqMsZMAaYUxL6UUv7pab2c0/yllLM4NX8579In\npVS+y+tlxyIyVET2isg6r3VvikiciKy2ly5e214SkRgR2SwizpqkSClVqBSraROUUs6WmvekMwz4\nHMg8P9MnxpgPvVeISH3gNqABEAHMFpGLjDE6KbtS6pwFIH/lC+2hUqoYyusRnjFmAZDT++V1B8YY\nY04ZY3YAMUCz3EevlCrOnNpDpQ0qpYohf5cdi0hfEVnptfTNYdWPi8jf9inBtOmUIwHvG5fF2uuU\nUuqcOXXaBG1QKVUM+TvC855PyV6G5KDaL4ELgUZAAjmYCFMppc6VU3uodAyVUsWQJx+mSzHGpE+x\nLCLfAJPth3FATa+iUfY6pZQ6Z/mRvwJBe6iUKoZSMT6X3BCRGl4PbwLSrgCcCNwmIiVEpA5QD1ie\npxeglCq28iN/BYL2UClVDOW1W1xERgOtgSoiEgu8AbQWkUZY97/bCTwMYIxZLyLjsO6PlwI8plf4\nKaVyy6kzpWuDSqliKK9HccaYrO7q/J2P8u8A7+Rpp0opRWCmTRCRTsBAwA18a4x5P9P2Z4EHsQ4C\n9wH3G2P+8VWnnvJTqhhy6qBOpZTyJwATE7uBwUBnoD5wuz1fnre/gCbGmMuB8cD//NWrDSqliiFj\njM9FKaWcKgD5qxkQY4zZbow5DYzBmi/Pex/zjDHH7YfLsC6m8UlP+amgSej0ULBD8Kte+Yhgh5Av\nUrQXSqk88ywcH+wQfGpe5ZJgh5Av/OUve94877nzhmSa+iWrufGu9lHlA8A0f3Fpg0qpYsjj2NuL\nKqWUb/7yl914ysnceX6JSB+gCdDKX1ltUClVDOlpPaVUYRWA/JWjufFEpD3wCtDKGHPKX6XaoFKq\nGNKB50qpwioA+WsFVZQa+QAAESZJREFUUM+eFy8O6+btd3gXEJErga+BTsaYvTmpVBtUShVDTr1b\nu1JK+ROAaV9SRORxYAbWtAlD7fnyBgArjTETgf8DygA/iQjALmNMN1/1aoNKqWLIY3QMlVKqcApE\n/jLGTAWmZlr3utff7c+1Tm1QKVUMGe2hUkoVUk7NX9qgUqoYcurNRZVSyh+n5i9tUClVDKXotAlK\nqULKqflLG1RKFUM6bYJSqrByav7SBpVSxZBO7KmUKqycmr+0QaVUMeTUIzyllPLHqfmrWNwcuWOH\n1qxft4BNGxbxfP/Hgh1Olpweo2PiCwul+vDBhP84hPCx31G+7z0ZNlfs9xhRCyanPz6va0ciZ/1M\n+KivCR/1Ned171Jgoda+8HzGzR6evizZOps+D/Xmsef7Mn7uCMbNHs5XYz6lavUqBRZTGo9J9bko\n53DMb88HjdG/xVvj6T5wEtGfTmTogvVnbU84fIwHh86m9xfTuHXwVBZusSbuXhu7n15fTLWWwVOZ\nu2H3Wc8NlKatmzDs9+/4YdH33PZY77O2N7y6IV9NG8zMndO4/saW6esbXXcFX8/4Mn2ZFjOZ5h2v\ny7c4nZq/inwPlcvlYtDAd+jU5XZiYxNYtnQqkybPZOPGrcEOLZ3TY3RUfKeT2fuf5zAnToLbTfXv\nBnJiyXJOr9tI2KUX4SpX9qynHJ81n0P/+6zAQ925bRe92lsNPpfLxezVE5kz7XeOHD7C4P9Zt5m6\n44FbefjZ+3n7hf8VaGw6sWfh4KjfXjY0Rv88qam8N3klX93TlurlSnHn1zNodUkUF1Yrn17mm9/X\n0eGyWvRqVo9te5N4fOR8pj0bSd1qFfjx4U6EuF3sO3qCXl9M5fqLIwlxB7Y/xOVy8eTbj/P8HS+y\nL2E/X0z5jKUzl/LP1l3pZfbG7eV/z37IrQ/3zPDc1UvW8HDHRwAoW6EsPyz6npW/rwpofN6cmr8K\npIdKRFqIyLMi0qEg9uetWdMr2bZtJzt27CI5OZlx4ybQLbpjQYfhk9NjdFp85sRJACQkBAkJAWPA\n5aLCUw9zaGBA7ocZcFe3bMLunXEkxCZy7N/j6etLlS4FQUgOTj3CcyLNX75pjP6tiz1AzUpliKpU\nhtAQNx0b1mL+ptgMZUSEY6eSAfj35Gmqli0FQKmwkPTG0+kUD4LkS4yXNLqYuJ3xJOxKJCU5hXkT\nfue6Dhl7mfbE7mH7xh2Y1Oxz1vU3tmT5vJWcOun31ne55tT8lS8NKhFZ7vX3Q8DnQFngDRF5MT/2\nmZ2IyHB2x8anP46NSyAiIrwgQ/DL6TE6Lj6Xi/BRXxM562f+v707D5KyvvM4/v4Mww2Ccl8uKkZW\nY0QhrlcIGGWJKEhWTVwSY2ktJtkYLVddUnFXMVVZ3WxtttajKhqtjUeMlSiGDR5QhChEIAJyiFdc\nlYQrHB5cIjL93T/6cTKAM80w3dO/nv68qC6ffvrXz/PtkfnwfZ6nu3+7Fy9lz+pX6X7JhXzw3PPk\ntr5zwPAuZ3+O/o/cS+/bb6Zdvz5lKBjGX3guTz0xp/7+1dOuYvbSJ5jwd+O469/vbfV6Ug2kFDi/\nmsc1FrZp+wf079G1/n6/w7qwaduufcZ8Y+yJzFrxFuP+Ywbffui3TJswqv6xVX/awpfumMVFdz3J\nTRd8tuhnpwB6D+jN5g2b6+9v3riZ3gN6NXs7YyeOYd4T84pZ2gFSza9SnaFq32B5KnBuREwHxgFT\nGnuSpKmSlkhaksvtLFFpVvFyOTZOuYp1532ZDicMp+PJJ9L5nNFsf3TGAUM/mL+QdRdMYeOl/8Du\nxUvpdcs/t3q5te1rGTPuLGbPnFu/7o7bfsy4kRcy67HZXHrFRU08uzSiwJ8qd0j5Bc4wO3RPr3yb\niScfzezrJ3PnV8dw02PPk8vOBJ04pDePXz2Bh6/6W+6bv5oPP6orc7Wf7Ii+R3DU8KG88OySku4n\n1fwqVUNVI+lwSb0ARcRmgIjYCext7EkRcU9EjIqIUTU1XRsb1izr121kyOCB9fcHDxrA+vUbi7Lt\nYkm9xlTrix072b1kOR1HjaD94EEMnPEgA2c+jDp1ZMCMBwDIvb8NPspOoz/xJB3++thWr/Oss0/n\nlVWv8c6Wdw94bNbjz3DOhDGtXlOqR3iJOKT8ysYUNcNS/d1ryDUW1rd7Zza+/5cG+8/bdtH3sC77\njJmx7E3GffpIAE46sg8f7q3jvV37XjY7uk8PunRozxub3it6jVs2bKHPgL+cwe/Tvw9bNmxt1jbG\nXDCaBU8/T93e0jZ8qeZXqRqqHsBSYAlwhKQBAJK6QYkuADfihSXLGTbsKIYOHUL79u255JJJ/O+v\nZ7dmCQWlXmNK9dX07IG65f+hUscOdPqbkex55Q+sG38x6ydOYf3EKcTuD9kw+bL8+F5H1D+38+jT\n+eitP37idkvpi5P3vdx35FGD65fHjv8cb72xptVrykU0eatyzq9mcI2FnTCoF398Zzvr3t3BR3vr\neGbVGj4/fNA+Ywb06MLiN/8MwJub32fP3hyHd+3Iund3sLcu3ySsf28nb2/ZxsCexTnh0NCrK15j\n0FGD6D+kP7Xtaxk76fM8P2dhs7YxdtJY5v2qtJf7IN38Ksmn/CJiaCMP5YDJpdhnY+rq6rjm2pt4\nctbPaFdTw//89FFefvn11iyhoNRrTKm+dr170Wv6jVDTDmrErjnPsnvBokbHd//KZDqPPgPq6sht\n287WW1r303Sdu3Ti9NGn8v0bbq9fd+33vsXQYUeSywUb1m7k+ze2bk0AuUjzkkEKnF/N4xoLq21X\nw7QJo/jmA/PI5YJJpxzNsL49uXvuSo4fdARjhg/muvGncOuvFvPw86+CYPrk05DEi2s2c//8l6lt\nJ2okvnv+KA7v2qnoNebqctzxL3dy+8M/oKamhqcefYY1r6/h8usv47UVr7NwziKOO+lTTP/JzXTr\n0Z3Tzz2Nr1/3Na78wlQA+g3uR9+BfVixcGXRazug1kTzS6l+QVZth0FpFmZF8+Znhpe7hILOX7ur\n8KAyW7lxYbPPmvxVr880+fu1ZuvKVj0T0xY5w9q+7Q9OLXcJTTr/n35X7hIKmrt2dpvJrzb/PVRm\ndiC/T8rMKlWq+eWGyqwK1eXSDCQzs0JSzS83VGZVyF+NYGaVKtX8ckNlVoVSPWVuZlZIqvnlhsqs\nCqX6YRQzs0JSzS83VGZVKNX3IJiZFZJqfrmhMqtCqc7WbmZWSKr55YbKrAqleoRnZlZIqvlVqqln\nzCxhxZgLS9J4Sa9JekPStBKXbGYGtE5+Seoo6dHs8cWShhbaphsqsyoUEU3eCpHUDrgL+CJwPHCp\npONLXLaZWWvl15XAuxExDPgRcDsFuKEyq0K5yDV5OwinAm9ExJsRsQf4OTCppEWbmdFq+TUJ+Gm2\n/EvgC5KanNLGDZVZFWrpER4wCPhTg/trs3VmZiXVSvlVPyYi9gLvA72a2miyb0rfu2ddUSc3lDQ1\nIu4p5jaLzTW2XLHrK8W86Sn8DD8q8PslaSrQcObXe8pdc6WptgxLvT6ovhrnfrkYW9lXCj/DVPOr\nms5QpT0teJ5rbLnU64MKqDEi7omIUQ1u+4fROmBIg/uDs3VWOqn/vUm9PnCNxZB6fcXKr/oxkmqB\nHsDWpvZbTQ2VmRXPC8Cxko6S1AH4CjCzzDWZmR2Mg8mvmcDXs+WLgN9EgeuJyV7yM7N0RcReSd8G\nngHaAfdHxOoyl2VmVlBj+SXpVmBJRMwE7gMelPQG8A75pqtJ1dRQJX3dPOMaWy71+qAyaiwoIp4E\nnix3HVUk9b83qdcHrrEYUq/voHxSfkXEvzZY3g1c3JxtKtVJBs3MzMwqhd9DZWZmZtZCbb6hknS/\npE2SXip3LY2RNETSPEkvS1ot6Zpy19SQpE6Sfi9pRVbf9HLX1BhJ7SS9KOnX5a5lf5LelrRK0nJJ\nS8pdj1WG1DMs9fyCysmwlPMLnGGFtPlLfpJGAzuAByLi0+Wu55NIGgAMiIhlkroDS4ELI+LlMpcG\nQPbtsF0jYoek9sAC4JqIWFTm0g4g6TpgFHBYRJxf7noakvQ2MCoitpS7FqscqWdY6vkFlZNhKecX\nOMMKafNnqCLiOfLv0E9WRGyIiGXZ8nbgFRL61unI25HdbZ/dkuvEJQ0GJgA/KXctZsWSeoalnl9Q\nGRnm/Kp8bb6hqjTZjNYnA4vLW8m+slPRy4FNwJyISKq+zH8BNwIHN9146wtgtqSl2Tf5mrUpqeYX\nVESGpZ5f4AxrkhuqhEjqBjwGXBsR28pdT0MRURcRI8h/o+ypkpK69CDpfGBTRCwtdy1NOCsiTiE/\nw/k/ZpdyzNqElPML0s6wCskvcIY1yQ1VIrLr+o8BD0fE4+WupzER8R4wDxhf7lr2cyYwMbvG/3Pg\nbEkPlbekfUXEuuy/m4AZ5Gc8N6t4lZJfkGyGJZ9f4AwrxA1VArI3TN4HvBIR/1nuevYnqY+kntly\nZ+Bc4NXyVrWviPhuRAyOiKHkv9H2NxHx1TKXVU9S1+wNu0jqCowDkvzUlllzpJ5fkH6GpZ5f4Aw7\nGG2+oZL0CLAQOE7SWklXlrumT3Am8DXyRyXLs9t55S6qgQHAPEkryc+BNCcikvxYb8L6AQskrQB+\nD8yKiKfLXJNVgArIsNTzC5xhxeAMK6DNf22CmZmZWam1+TNUZmZmZqXmhsrMzMyshdxQmZmZmbWQ\nGyozMzOzFnJDZWZmZtZCbqgSJaku+/jxS5J+IalLC7Y15uPZyyVNlDStibE9JX3rEPZxi6TrD3Z9\nE9vZUXjUoW/fzErP+dWy/VplckOVrg8iYkQ2u/we4BsNH1Res///RcTMiLitiSE9gWYHkplZA84v\nqzpuqCrDfGCYpKGSXpP0APlvqB0iaZykhZKWZUeC3QAkjZf0qqRlwJc+3pCkyyXdmS33kzRD0ors\ndgZwG3BMdnT5w2zcDZJekLRS0vQG2/qepNclLQCOa84LkvRENsHm6v0n2ZT0o2z9XEl9snXHSHo6\ne858ScMP4edoZq3P+eX8qgpuqBInqZb8RJSrslXHAndHxAnATuAm4JxswsolwHWSOgH3AhcAI4H+\njWz+v4FnI+Ik4BRgNTAN+L/s6PIGSeOyfZ4KjABGShotaST5KRJGAOcBn23mS7siIkYCo4DvSOqV\nre8KLMle37PAzdn6e4Crs+dcD9zdzP2ZWStzfjm/qkltuQuwRnWWtDxbnk9+rqyBwJqIWJStPw04\nHvidJIAO5KeoGA68FRF/AFB+ks19jqIyZwOXQX4mduB9SYfvN2Zcdnsxu9+NfEB1B2ZExK5sHzOb\n+fq+I2lytjwk2+ZWIAc8mq1/CHg8O2o9A/hF9joBOjZzf2bWepxfzq+q44YqXR9ExIiGK7Jfxp0N\nV5Gfk+rS/cbt87wWEvBvEfHj/fZx7SFvUBoDnAOcHhG7JP0W6NTI8CB/JvW9/X8eZpYs51ee86uK\n+JJfZVsEnClpGNTPBv4p8rOoD5V0TDbu0kaePxf4ZvbcdpJ6ANvJH7197BngigbvbRgkqS/wHHCh\npM7Kz0B+QTPq7gG8m4XRcPJHqh+rAS7Klv8eWBAR24C3JF2c1SBJJzVjf2aWHueXtSluqCpYRGwG\nLgceUX4W9YXA8IjYTf4U+azsTZ2bGtnENcBYSauApcDxEbGV/Cn4lyT9MCJmAz8DFmbjfgl0j4hl\n5E9trwCeIj+De2NukrT24xvwNFAr6RXybyJd1GDsTuBUSS+RP6V/a7Z+CnCl8jOdrwYmHezPyczS\n4/yytkYRUe4azMzMzCqaz1CZmZmZtZAbKjMzM7MWckNlZmZm1kJuqMzMzMxayA2VmZmZWQu5oTIz\nMzNrITdUZmZmZi3khsrMzMyshf4fNMfQV0qAG0YAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 720x216 with 4 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}