Risk_Metric_Research_v2.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load packages\n",
    "import gc\n",
    "import pickle\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from scipy.stats import spearmanr, uniform\n",
    "from xgboost import XGBRegressor\n",
    "from matplotlib import pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# Loading data...\n",
      "Loaded 310 features\n"
     ]
    }
   ],
   "source": [
    "print(\"# Loading data...\")\n",
    "# The training data is used to train your model how to predict the targets.\n",
    "training_data = pd.read_csv(\"https://numerai-public-datasets.s3-us-west-2.amazonaws.com/latest_numerai_training_data.csv.xz\").set_index(\"id\", drop=True).astype(\"float16\", errors=\"ignore\")\n",
    "# The tournament data is the data that Numerai uses to evaluate your model.\n",
    "tournament_data = pd.read_csv(\"https://numerai-public-datasets.s3-us-west-2.amazonaws.com/latest_numerai_tournament_data.csv.xz\").set_index(\"id\", drop=True).astype(\"float16\", errors=\"ignore\")\n",
    "validation_data = tournament_data[tournament_data[\"data_type\"] == \"validation\"]\n",
    "del tournament_data\n",
    "tournament_data = None\n",
    "gc.collect()\n",
    "\n",
    "feature_names = [f for f in training_data.columns if f.startswith(\"feature\")]\n",
    "print(f\"Loaded {len(feature_names)} features\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# prepare data to 2 fold CV\n",
    "era_list = training_data[\"era\"].unique().tolist()\n",
    "h1_eras = era_list[:len(era_list)//2]\n",
    "h2_eras = era_list[len(era_list)//2:]\n",
    "\n",
    "h1_data = training_data[training_data[\"era\"].isin(h1_eras)]\n",
    "h2_data = training_data[training_data[\"era\"].isin(h2_eras)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get all metrics\n",
    "def compute_metrics(predictions):\n",
    "    training_preds_df = pd.DataFrame()\n",
    "    validation_preds_df = pd.DataFrame()\n",
    "    \n",
    "    # create data frame from predictions\n",
    "    experiments = []\n",
    "    colsamples = []\n",
    "    for pred in predictions:\n",
    "        training_preds_df[pred[\"name\"]] = pred[\"training_preds\"]\n",
    "        validation_preds_df[pred[\"name\"]] = pred[\"validation_preds\"]\n",
    "        experiments += [pred[\"experiment\"]]\n",
    "        colsamples += [pred.get(\"colsample_bytree\")]\n",
    "    \n",
    "    # add columns era and target\n",
    "    training_preds_df[\"era\"] = training_data[\"era\"]\n",
    "    validation_preds_df[\"era\"] = validation_data[\"era\"]\n",
    "    training_preds_df[\"target\"] = training_data[\"target_kazutsugi\"]\n",
    "    validation_preds_df[\"target\"] = validation_data[\"target_kazutsugi\"]\n",
    "    \n",
    "    # get scores per era\n",
    "    model_names = [c for c in validation_preds_df if \"era\" != c and \"target\" != c]\n",
    "    val_scores_per_era = validation_preds_df.groupby(\"era\").apply(lambda d: d[model_names].corrwith(d[\"target\"], method=\"spearman\"))\n",
    "    training_scores_per_era = training_preds_df.groupby(\"era\").apply(lambda d: d[model_names].corrwith(d[\"target\"], method=\"spearman\"))\n",
    "    \n",
    "    # Metrics Data Frame\n",
    "    metrics_df = pd.DataFrame(index=model_names, columns=[])\n",
    "    metrics_df[\"experiment\"] = experiments\n",
    "    metrics_df[\"colsample_bytree\"] = colsamples\n",
    "    # Calculate mean\n",
    "    metrics_df[\"val_mean\"] = val_scores_per_era.mean()\n",
    "    metrics_df[\"training_mean\"] = training_scores_per_era.mean()\n",
    "    # Calculate sharpe\n",
    "    metrics_df[\"val_sharpe\"] = val_scores_per_era.mean()/val_scores_per_era.std()\n",
    "    metrics_df[\"training_sharpe\"] = training_scores_per_era.mean()/training_scores_per_era.std()\n",
    "\n",
    "    # validation max drawdown\n",
    "    val_max_drawdown = {}\n",
    "    for model in model_names:\n",
    "        rolling_max = (val_scores_per_era[model]+1).cumprod().rolling(window=100, min_periods=1).max()\n",
    "        daily_value = (val_scores_per_era[model]+1).cumprod()\n",
    "        val_max_drawdown[model] = (rolling_max - daily_value).max()\n",
    "    metrics_df[\"val_max_drawdown\"] = pd.Series(val_max_drawdown)\n",
    "\n",
    "    # training max drawdown\n",
    "    training_max_drawdown = {}\n",
    "    for model in model_names:\n",
    "        rolling_max = (training_scores_per_era[model]+1).cumprod().rolling(window=100, min_periods=1).max()\n",
    "        daily_value = (training_scores_per_era[model]+1).cumprod()\n",
    "        training_max_drawdown[model] = (rolling_max - daily_value).max()\n",
    "    metrics_df[\"training_max_drawdown\"] = pd.Series(training_max_drawdown)\n",
    "    \n",
    "    return metrics_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_predictions(max_depth, learning_rate, n_estimators, colsample_bytree, features, select_eras):\n",
    "    \n",
    "    print(f\"{name}, max_depth={max_depth}, learning_rate={learning_rate}, n_estimators={n_estimators}, colsample_bytree={colsample_bytree}, eras count={len(select_eras)}, features count={len(features)}\")\n",
    "    \n",
    "    model = XGBRegressor(max_depth=max_depth,\n",
    "                         learning_rate=learning_rate,\n",
    "                         n_estimators=n_estimators,\n",
    "                         colsample_bytree=colsample_bytree,\n",
    "                         n_jobs=12,\n",
    "                         verbosity=1,\n",
    "                         random_state=12)\n",
    "\n",
    "    # train on fold 1\n",
    "    model.fit(h1_data[features][h1_data[\"era\"].isin(select_eras)], h1_data[\"target_kazutsugi\"][h1_data[\"era\"].isin(select_eras)])\n",
    "    h2_preds = pd.Series(model.predict(h2_data[features]), index=h2_data.index)\n",
    "    \n",
    "    # train on fold 2\n",
    "    model.fit(h2_data[features][h2_data[\"era\"].isin(select_eras)], h2_data[\"target_kazutsugi\"][h2_data[\"era\"].isin(select_eras)])\n",
    "    h1_preds = pd.Series(model.predict(h1_data[features]), index=h1_data.index)\n",
    "\n",
    "    # train on all training data\n",
    "    model.fit(training_data[features][training_data[\"era\"].isin(select_eras)], training_data[\"target_kazutsugi\"][training_data[\"era\"].isin(select_eras)])\n",
    "    validation_preds = pd.Series(model.predict(validation_data[features]), index=validation_data.index)\n",
    "    \n",
    "    return pd.concat([h1_preds, h2_preds]), validation_preds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "def compare(metrics_df_org, experiments, colsample_plots = True):\n",
    "    \n",
    "    metrics_df = metrics_df_org.copy()\n",
    "    metrics_df.drop(\"ExamplePreds\", inplace=True)\n",
    "    \n",
    "    \n",
    "    if colsample_plots:\n",
    "        fig, axes = plt.subplots(nrows=3, ncols=2, figsize=(12,15))\n",
    "    else:\n",
    "        fig, axes = plt.subplots(nrows=2, ncols=2, figsize=(12,10))\n",
    "        \n",
    "    colors = [\"red\", \"blue\", \"green\"]\n",
    "    \n",
    "    for i, expt in enumerate(experiments):\n",
    "        metrics_df[metrics_df.experiment == expt].plot(x=\"training_sharpe\", y=\"val_sharpe\", \n",
    "                                                           kind=\"scatter\", alpha=0.4, \n",
    "                                                           label=expt, ax=axes[0,1],\n",
    "                                                           color=colors[i])\n",
    "    \n",
    "    metrics = [\"training_mean\", \"val_sharpe\", \"val_max_drawdown\"]\n",
    "    axs = [axes[0,0], axes[1,0], axes[1,1]]\n",
    "    for j, m in enumerate(metrics):\n",
    "        for i, expt in enumerate(experiments):\n",
    "            metrics_df[metrics_df.experiment == expt].plot(x=m, y=\"val_mean\", \n",
    "                                                           kind=\"scatter\", alpha=0.4, \n",
    "                                                           label=expt, ax=axs[j],\n",
    "                                                           color=colors[i])\n",
    "\n",
    "    if colsample_plots:\n",
    "        for i, expt in enumerate(experiments):\n",
    "            metrics_df[metrics_df.experiment == expt].plot(x=\"colsample_bytree\", y=\"val_mean\", \n",
    "                                                           kind=\"scatter\", alpha=0.4, \n",
    "                                                           label=expt, ax=axes[2,0],\n",
    "                                                           color=colors[i])\n",
    "        for i, expt in enumerate(experiments):\n",
    "            metrics_df[metrics_df.experiment == expt].plot(x=\"colsample_bytree\", y=\"val_sharpe\", \n",
    "                                                           kind=\"scatter\", alpha=0.4, \n",
    "                                                           label=expt, ax=axes[2,1],\n",
    "                                                           color=colors[i])\n",
    "\n",
    "    metrics = [\"training_mean\", \"val_mean\", \"training_sharpe\", \"val_sharpe\", \"val_max_drawdown\"]\n",
    "    axs = [axes[0,0], axes[1,0], axes[1,1]]\n",
    "    for f in [\"MAX\", \"MEAN\", \"MIN\"]:\n",
    "        print(\"-\"*22)\n",
    "        for j, m in enumerate(metrics):\n",
    "            for i, expt in enumerate(experiments):\n",
    "                if f == \"MAX\":\n",
    "                    print(m, expt, \"MAX\", np.max(metrics_df[metrics_df.experiment == expt][m]))\n",
    "                elif f == \"MEAN\":\n",
    "                    print(m, expt, \"MEAN\", np.mean(metrics_df[metrics_df.experiment == expt][m]))\n",
    "                else:\n",
    "                    print(m, expt, \"MIN\", np.min(metrics_df[metrics_df.experiment == expt][m]))\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "EXPT_SIZE = 20\n",
    "ESTIMATORS = 50"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ExamplePreds, max_depth=5, learning_rate=0.01, n_estimators=2000, colsample_bytree=0.1, eras count=120, features count=310\n",
      "high_colsample_1, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.8511071462358749, eras count=120, features count=310\n",
      "high_colsample_2, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.770543128337785, eras count=120, features count=310\n",
      "high_colsample_3, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.889568614156339, eras count=120, features count=310\n",
      "high_colsample_4, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.7515124762006702, eras count=120, features count=310\n",
      "high_colsample_5, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.7049786315275602, eras count=120, features count=310\n",
      "high_colsample_6, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.7500209202601426, eras count=120, features count=310\n",
      "high_colsample_7, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.8384068899926187, eras count=120, features count=310\n",
      "high_colsample_8, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.9838573898899379, eras count=120, features count=310\n",
      "high_colsample_9, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.9666478664419185, eras count=120, features count=310\n",
      "high_colsample_10, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.8128566662755887, eras count=120, features count=310\n",
      "high_colsample_11, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.7047583070252617, eras count=120, features count=310\n",
      "high_colsample_12, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.9886191578398008, eras count=120, features count=310\n",
      "high_colsample_13, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.8836883210431771, eras count=120, features count=310\n",
      "high_colsample_14, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.7103473011679383, eras count=120, features count=310\n",
      "high_colsample_15, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.8884659736998711, eras count=120, features count=310\n",
      "high_colsample_16, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.7164860074226469, eras count=120, features count=310\n",
      "high_colsample_17, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.7643721582964933, eras count=120, features count=310\n",
      "high_colsample_18, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.8414777096442064, eras count=120, features count=310\n",
      "high_colsample_19, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.7207919620895099, eras count=120, features count=310\n",
      "high_colsample_20, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.7750243661320597, eras count=120, features count=310\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(1776)\n",
    "\n",
    "predictions = []\n",
    "features = feature_names # default, use all features\n",
    "select_eras = era_list # default, use all eras\n",
    "\n",
    "for i in range(EXPT_SIZE+1):\n",
    "    name = f\"high_colsample_{i}\"\n",
    "    max_depth = 5\n",
    "    learning_rate = 0.1\n",
    "    n_estimators = ESTIMATORS \n",
    "    colsample_bytree = 0.7+np.random.rand()*0.3\n",
    "\n",
    "    experiment_name = \"high_colsample\"\n",
    "    if i == 0:\n",
    "        name = \"ExamplePreds\"\n",
    "        max_depth = 5\n",
    "        learning_rate = 0.01\n",
    "        n_estimators = 2000\n",
    "        colsample_bytree = 0.1\n",
    "        experiment_name = \"ExamplePreds\"\n",
    "        \n",
    "\n",
    "    training_preds, validation_preds = compute_predictions(max_depth, \n",
    "                                                           learning_rate, \n",
    "                                                           n_estimators, \n",
    "                                                           colsample_bytree, \n",
    "                                                           features, \n",
    "                                                           select_eras)\n",
    "\n",
    "    predictions += [{\n",
    "        \"name\": name,\n",
    "        \"experiment\":  experiment_name,\n",
    "        \"training_preds\": training_preds,\n",
    "        \"validation_preds\": validation_preds,\n",
    "        \"colsample_bytree\": colsample_bytree\n",
    "    }]\n",
    "    pickle.dump(predictions, open(\"backup_v3.pickle\", \"bw\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "low_colsample_0, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.04277146852620036, eras count=120, features count=310\n",
      "low_colsample_1, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.15110714623587498, eras count=120, features count=310\n",
      "low_colsample_2, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.0705431283377851, eras count=120, features count=310\n",
      "low_colsample_3, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.1895686141563391, eras count=120, features count=310\n",
      "low_colsample_4, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.0515124762006702, eras count=120, features count=310\n",
      "low_colsample_5, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.0049786315275602505, eras count=120, features count=310\n",
      "low_colsample_6, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.05002092026014264, eras count=120, features count=310\n",
      "low_colsample_7, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.13840688999261874, eras count=120, features count=310\n",
      "low_colsample_8, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.28385738988993797, eras count=120, features count=310\n",
      "low_colsample_9, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.26664786644191857, eras count=120, features count=310\n",
      "low_colsample_10, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.11285666627558882, eras count=120, features count=310\n",
      "low_colsample_11, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.004758307025261643, eras count=120, features count=310\n",
      "low_colsample_12, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.28861915783980086, eras count=120, features count=310\n",
      "low_colsample_13, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.18368832104317714, eras count=120, features count=310\n",
      "low_colsample_14, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.010347301167938394, eras count=120, features count=310\n",
      "low_colsample_15, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.18846597369987114, eras count=120, features count=310\n",
      "low_colsample_16, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.016486007422646953, eras count=120, features count=310\n",
      "low_colsample_17, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.0643721582964934, eras count=120, features count=310\n",
      "low_colsample_18, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.14147770964420647, eras count=120, features count=310\n",
      "low_colsample_19, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.020791962089509942, eras count=120, features count=310\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(1776)\n",
    "for i in range(EXPT_SIZE):\n",
    "    name = f\"low_colsample_{i}\"\n",
    "    max_depth = 5\n",
    "    learning_rate = 0.1\n",
    "    n_estimators = ESTIMATORS \n",
    "    colsample_bytree = 0.0+np.random.rand()*0.3\n",
    "    training_preds, validation_preds = compute_predictions(max_depth, \n",
    "                                                           learning_rate, \n",
    "                                                           n_estimators, \n",
    "                                                           colsample_bytree, \n",
    "                                                           features, \n",
    "                                                           select_eras)\n",
    "\n",
    "    predictions += [{\n",
    "        \"name\": name,\n",
    "        \"experiment\": \"low_colsample\",\n",
    "        \"training_preds\": training_preds,\n",
    "        \"validation_preds\": validation_preds,\n",
    "        \"colsample_bytree\": colsample_bytree\n",
    "    }]\n",
    "    pickle.dump(predictions, open(\"backup_v3.pickle\", \"bw\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Eras selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "training_preds_df = pd.DataFrame()\n",
    "for pred in predictions[:1]: # only ExamplePreds\n",
    "    training_preds_df[pred[\"name\"]] = pred[\"training_preds\"]\n",
    "training_preds_df[\"era\"] = training_data[\"era\"]\n",
    "training_preds_df[\"target\"] = training_data[\"target_kazutsugi\"]\n",
    "model_names = [\"ExamplePreds\"]\n",
    "training_scores_per_era = training_preds_df.groupby(\"era\").apply(lambda d: d[model_names].corrwith(d[\"target\"], method=\"spearman\"))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "_ = plt.hist(training_scores_per_era[\"ExamplePreds\"], 20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    120.000000\n",
       "mean       0.048693\n",
       "std        0.033873\n",
       "min       -0.073576\n",
       "25%        0.025353\n",
       "50%        0.049109\n",
       "75%        0.070396\n",
       "max        0.123162\n",
       "Name: ExamplePreds, dtype: float64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_scores_per_era[\"ExamplePreds\"].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "split_at = np.percentile(training_scores_per_era[\"ExamplePreds\"], 50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "easy_eras = training_scores_per_era[\"ExamplePreds\"].index[training_scores_per_era[\"ExamplePreds\"] > split_at].tolist()\n",
    "hard_eras = training_scores_per_era[\"ExamplePreds\"].index[training_scores_per_era[\"ExamplePreds\"] <= split_at].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'era1, era10, era105, era108, era109, era11, era115, era118, era12, era120, era13, era14, era15, era16, era2, era20, era22, era23, era24, era26, era28, era30, era36, era37, era38, era4, era43, era44, era45, era47, era48, era5, era51, era55, era56, era6, era61, era63, era64, era7, era71, era72, era75, era76, era77, era78, era79, era8, era83, era86, era88, era90, era92, era93, era94, era95, era96, era97, era98, era99'"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "', '.join(easy_eras)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'era100, era101, era102, era103, era104, era106, era107, era110, era111, era112, era113, era114, era116, era117, era119, era17, era18, era19, era21, era25, era27, era29, era3, era31, era32, era33, era34, era35, era39, era40, era41, era42, era46, era49, era50, era52, era53, era54, era57, era58, era59, era60, era62, era65, era66, era67, era68, era69, era70, era73, era74, era80, era81, era82, era84, era85, era87, era89, era9, era91'"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "', '.join(hard_eras)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fb99fed7278>]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(training_scores_per_era[\"ExamplePreds\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "easy_eras_0, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.35702862470160046, eras count=60, features count=310\n",
      "easy_eras_1, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.5014761949811667, eras count=60, features count=310\n",
      "easy_eras_2, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.3940575044503801, eras count=60, features count=310\n",
      "easy_eras_3, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.5527581522084521, eras count=60, features count=310\n",
      "easy_eras_4, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.3686833016008936, eras count=60, features count=310\n",
      "easy_eras_5, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.3066381753700803, eras count=60, features count=310\n",
      "easy_eras_6, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.3666945603468569, eras count=60, features count=310\n",
      "easy_eras_7, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.48454251999015835, eras count=60, features count=310\n",
      "easy_eras_8, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.6784765198532507, eras count=60, features count=310\n",
      "easy_eras_9, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.6555304885892248, eras count=60, features count=310\n",
      "easy_eras_10, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.45047555503411846, eras count=60, features count=310\n",
      "easy_eras_11, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.3063444093670155, eras count=60, features count=310\n",
      "easy_eras_12, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.6848255437864011, eras count=60, features count=310\n",
      "easy_eras_13, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.5449177613909029, eras count=60, features count=310\n",
      "easy_eras_14, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.3137964015572512, eras count=60, features count=310\n",
      "easy_eras_15, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.5512879649331616, eras count=60, features count=310\n",
      "easy_eras_16, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.32198134323019595, eras count=60, features count=310\n",
      "easy_eras_17, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.3858295443953245, eras count=60, features count=310\n",
      "easy_eras_18, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.4886369461922753, eras count=60, features count=310\n",
      "easy_eras_19, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.32772261611934655, eras count=60, features count=310\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(1776)\n",
    "features = feature_names\n",
    "for i in range(EXPT_SIZE):\n",
    "    name = f\"easy_eras_{i}\"\n",
    "    max_depth = 5\n",
    "    learning_rate = 0.1\n",
    "    n_estimators = ESTIMATORS \n",
    "    colsample_bytree = 0.3+np.random.rand()*0.4\n",
    "    training_preds, validation_preds = compute_predictions(max_depth, \n",
    "                                                           learning_rate, \n",
    "                                                           n_estimators, \n",
    "                                                           colsample_bytree, \n",
    "                                                           features, \n",
    "                                                           easy_eras)\n",
    "\n",
    "    predictions += [{\n",
    "        \"name\": name,\n",
    "        \"experiment\": \"easy_eras\",\n",
    "        \"training_preds\": training_preds,\n",
    "        \"validation_preds\": validation_preds,\n",
    "        \"colsample_bytree\": colsample_bytree\n",
    "    }]\n",
    "    pickle.dump(predictions, open(\"backup_v3.pickle\", \"bw\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "hard_eras_0, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.35702862470160046, eras count=60, features count=310\n",
      "hard_eras_1, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.5014761949811667, eras count=60, features count=310\n",
      "hard_eras_2, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.3940575044503801, eras count=60, features count=310\n",
      "hard_eras_3, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.5527581522084521, eras count=60, features count=310\n",
      "hard_eras_4, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.3686833016008936, eras count=60, features count=310\n",
      "hard_eras_5, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.3066381753700803, eras count=60, features count=310\n",
      "hard_eras_6, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.3666945603468569, eras count=60, features count=310\n",
      "hard_eras_7, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.48454251999015835, eras count=60, features count=310\n",
      "hard_eras_8, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.6784765198532507, eras count=60, features count=310\n",
      "hard_eras_9, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.6555304885892248, eras count=60, features count=310\n",
      "hard_eras_10, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.45047555503411846, eras count=60, features count=310\n",
      "hard_eras_11, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.3063444093670155, eras count=60, features count=310\n",
      "hard_eras_12, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.6848255437864011, eras count=60, features count=310\n",
      "hard_eras_13, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.5449177613909029, eras count=60, features count=310\n",
      "hard_eras_14, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.3137964015572512, eras count=60, features count=310\n",
      "hard_eras_15, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.5512879649331616, eras count=60, features count=310\n",
      "hard_eras_16, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.32198134323019595, eras count=60, features count=310\n",
      "hard_eras_17, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.3858295443953245, eras count=60, features count=310\n",
      "hard_eras_18, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.4886369461922753, eras count=60, features count=310\n",
      "hard_eras_19, max_depth=5, learning_rate=0.1, n_estimators=50, colsample_bytree=0.32772261611934655, eras count=60, features count=310\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(1776)\n",
    "features = feature_names\n",
    "for i in range(EXPT_SIZE):\n",
    "    name = f\"hard_eras_{i}\"\n",
    "    max_depth = 5\n",
    "    learning_rate = 0.1\n",
    "    n_estimators = ESTIMATORS \n",
    "    colsample_bytree = 0.3+np.random.rand()*0.4\n",
    "    training_preds, validation_preds = compute_predictions(max_depth, \n",
    "                                                           learning_rate, \n",
    "                                                           n_estimators, \n",
    "                                                           colsample_bytree, \n",
    "                                                           features, \n",
    "                                                           hard_eras)\n",
    "\n",
    "    predictions += [{\n",
    "        \"name\": name,\n",
    "        \"experiment\": \"hard_eras\",\n",
    "        \"training_preds\": training_preds,\n",
    "        \"validation_preds\": validation_preds,\n",
    "        \"colsample_bytree\": colsample_bytree\n",
    "    }]\n",
    "    pickle.dump(predictions, open(\"backup_v3.pickle\", \"bw\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Neutralize"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "def neutralize(df, scores, by, proportion=1.0):\n",
    "    exposures = df[by].values\n",
    "    # constant column to make sure the series is completely neutral to exposures\n",
    "    exposures = np.hstack((exposures, np.array([np.mean(scores)] * len(exposures)).reshape(-1, 1)))\n",
    "    scores = scores - proportion * exposures.dot(np.linalg.pinv(exposures).dot(scores))\n",
    "    return scores / scores.std()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_preds = []\n",
    "for preds in predictions:\n",
    "    new_preds += [{\n",
    "        \"name\": \"neutralized_\" + preds[\"name\"],\n",
    "        \"experiment\": \"neutralized_\" + preds[\"experiment\"],\n",
    "        \"training_preds\": neutralize(training_data, preds[\"training_preds\"], feature_names),\n",
    "        \"validation_preds\": neutralize(validation_data, preds[\"validation_preds\"], feature_names),\n",
    "        \"colsample_bytree\": preds[\"colsample_bytree\"]\n",
    "    }]\n",
    "predictions += new_preds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "#pickle.dump(predictions, open(\"backup_v3.pickle\", \"bw\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analyze results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "metrics_df = compute_metrics(predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "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>experiment</th>\n",
       "      <th>colsample_bytree</th>\n",
       "      <th>val_mean</th>\n",
       "      <th>training_mean</th>\n",
       "      <th>val_sharpe</th>\n",
       "      <th>training_sharpe</th>\n",
       "      <th>val_max_drawdown</th>\n",
       "      <th>training_max_drawdown</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ExamplePreds</th>\n",
       "      <td>ExamplePreds</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>0.029522</td>\n",
       "      <td>0.048693</td>\n",
       "      <td>0.968562</td>\n",
       "      <td>1.437529</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>5.545741</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>high_colsample_1</th>\n",
       "      <td>high_colsample</td>\n",
       "      <td>0.851107</td>\n",
       "      <td>0.025790</td>\n",
       "      <td>0.041549</td>\n",
       "      <td>0.812189</td>\n",
       "      <td>1.278984</td>\n",
       "      <td>0.071039</td>\n",
       "      <td>3.415843</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>high_colsample_2</th>\n",
       "      <td>high_colsample</td>\n",
       "      <td>0.770543</td>\n",
       "      <td>0.025353</td>\n",
       "      <td>0.040573</td>\n",
       "      <td>0.793394</td>\n",
       "      <td>1.232532</td>\n",
       "      <td>0.072572</td>\n",
       "      <td>2.895796</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>high_colsample_3</th>\n",
       "      <td>high_colsample</td>\n",
       "      <td>0.889569</td>\n",
       "      <td>0.025966</td>\n",
       "      <td>0.041021</td>\n",
       "      <td>0.860114</td>\n",
       "      <td>1.221346</td>\n",
       "      <td>0.064457</td>\n",
       "      <td>2.794750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>high_colsample_4</th>\n",
       "      <td>high_colsample</td>\n",
       "      <td>0.751512</td>\n",
       "      <td>0.025328</td>\n",
       "      <td>0.041761</td>\n",
       "      <td>0.797031</td>\n",
       "      <td>1.202276</td>\n",
       "      <td>0.086917</td>\n",
       "      <td>3.900581</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      experiment  colsample_bytree  val_mean  training_mean  \\\n",
       "ExamplePreds        ExamplePreds          0.100000  0.029522       0.048693   \n",
       "high_colsample_1  high_colsample          0.851107  0.025790       0.041549   \n",
       "high_colsample_2  high_colsample          0.770543  0.025353       0.040573   \n",
       "high_colsample_3  high_colsample          0.889569  0.025966       0.041021   \n",
       "high_colsample_4  high_colsample          0.751512  0.025328       0.041761   \n",
       "\n",
       "                  val_sharpe  training_sharpe  val_max_drawdown  \\\n",
       "ExamplePreds        0.968562         1.437529          0.078947   \n",
       "high_colsample_1    0.812189         1.278984          0.071039   \n",
       "high_colsample_2    0.793394         1.232532          0.072572   \n",
       "high_colsample_3    0.860114         1.221346          0.064457   \n",
       "high_colsample_4    0.797031         1.202276          0.086917   \n",
       "\n",
       "                  training_max_drawdown  \n",
       "ExamplePreds                   5.545741  \n",
       "high_colsample_1               3.415843  \n",
       "high_colsample_2               2.895796  \n",
       "high_colsample_3               2.794750  \n",
       "high_colsample_4               3.900581  "
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## `colsample_bytree` low vs high"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------\n",
      "training_mean high_colsample MAX 0.04247875653736934\n",
      "training_mean low_colsample MAX 0.04374273549826444\n",
      "val_mean high_colsample MAX 0.027972166323851425\n",
      "val_mean low_colsample MAX 0.027336417906790166\n",
      "training_sharpe high_colsample MAX 1.2977492125411518\n",
      "training_sharpe low_colsample MAX 1.3023105530113255\n",
      "val_sharpe high_colsample MAX 0.9314164706073689\n",
      "val_sharpe low_colsample MAX 0.8890930631418525\n",
      "val_max_drawdown high_colsample MAX 0.08691697481093952\n",
      "val_max_drawdown low_colsample MAX 0.10901373776362488\n",
      "----------------------\n",
      "training_mean high_colsample MEAN 0.041579793167087695\n",
      "training_mean low_colsample MEAN 0.03753421511229674\n",
      "val_mean high_colsample MEAN 0.02596817829573487\n",
      "val_mean low_colsample MEAN 0.022188545542988748\n",
      "training_sharpe high_colsample MEAN 1.243156176357713\n",
      "training_sharpe low_colsample MEAN 1.1868366294495514\n",
      "val_sharpe high_colsample MEAN 0.8536104108761521\n",
      "val_sharpe low_colsample MEAN 0.7406634697688222\n",
      "val_max_drawdown high_colsample MEAN 0.07135590480769713\n",
      "val_max_drawdown low_colsample MEAN 0.08130518720290361\n",
      "----------------------\n",
      "training_mean high_colsample MIN 0.04057293819734283\n",
      "training_mean low_colsample MIN 0.021234511530050914\n",
      "val_mean high_colsample MIN 0.024797230152463132\n",
      "val_mean low_colsample MIN 0.01060136181620322\n",
      "training_sharpe high_colsample MIN 1.2022755344808622\n",
      "training_sharpe low_colsample MIN 0.8159735992312431\n",
      "val_sharpe high_colsample MIN 0.7822440727413094\n",
      "val_sharpe low_colsample MIN 0.4404601195670108\n",
      "val_max_drawdown high_colsample MIN 0.05637986261124239\n",
      "val_max_drawdown low_colsample MIN 0.048870000169397576\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x1080 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "compare(metrics_df, [\"high_colsample\", \"low_colsample\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## easy vs hard eras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------\n",
      "training_mean easy_eras MAX 0.04455453364826924\n",
      "training_mean hard_eras MAX 0.016932630176030975\n",
      "val_mean easy_eras MAX 0.026465969189361907\n",
      "val_mean hard_eras MAX 0.015028232293465503\n",
      "training_sharpe easy_eras MAX 1.1031385327545837\n",
      "training_sharpe hard_eras MAX 0.7274669091446736\n",
      "val_sharpe easy_eras MAX 0.6822987665093445\n",
      "val_sharpe hard_eras MAX 0.793544446504416\n",
      "val_max_drawdown easy_eras MAX 0.13958833800405746\n",
      "val_max_drawdown hard_eras MAX 0.11892950683491965\n",
      "----------------------\n",
      "training_mean easy_eras MEAN 0.04363257861481375\n",
      "training_mean hard_eras MEAN 0.014793346088155948\n",
      "val_mean easy_eras MEAN 0.024463272917917757\n",
      "val_mean hard_eras MEAN 0.010827332788950298\n",
      "training_sharpe easy_eras MEAN 1.0610645119776416\n",
      "training_sharpe hard_eras MEAN 0.6268127965567225\n",
      "val_sharpe easy_eras MEAN 0.6163919554437711\n",
      "val_sharpe hard_eras MEAN 0.5519496549443975\n",
      "val_max_drawdown easy_eras MEAN 0.11932491807044661\n",
      "val_max_drawdown hard_eras MEAN 0.07267407932399336\n",
      "----------------------\n",
      "training_mean easy_eras MIN 0.04239281296889773\n",
      "training_mean hard_eras MIN 0.011833546664698973\n",
      "val_mean easy_eras MIN 0.022558190304765798\n",
      "val_mean hard_eras MIN 0.005131411929792869\n",
      "training_sharpe easy_eras MIN 1.0242368318089272\n",
      "training_sharpe hard_eras MIN 0.5085404664201721\n",
      "val_sharpe easy_eras MIN 0.5586086946395253\n",
      "val_sharpe hard_eras MIN 0.24438591033940452\n",
      "val_max_drawdown easy_eras MIN 0.10474998349775344\n",
      "val_max_drawdown hard_eras MIN 0.048146376896759824\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAuYAAANdCAYAAADY47lRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOzdeZxU1Zn/8c9j09DYNqIIaliECFEBWbQF18liNMRxxHEZFRMg4wR3ncVhNJOoMfE3kWR03CZ54RKX2IMGx4QkbhmNMXFjSRBp0dAqCY0LLbg0SEkDz++PcxuKtrq7uruq7q2q7/v1qlfVvXXr1tNFceqpU885x9wdERERERGJ1y5xByAiIiIiIkrMRUREREQSQYm5iIiIiEgCKDEXEREREUkAJeYiIiIiIgnQK+4ACmWvvfby4cOHxx2GiEiXLVmy5F13Hxh3HIWkNltEilVP2uyyScyHDx/O4sWL4w5DRKTLzOzPccdQaGqzRaRY9aTNVimLiIiIiEgCKDEXEREREUkAJeYiIiIiIglQNjXmmbS0tNDY2EgqlYo7lJJRVVXFkCFDqKysjDsUERERiSjnyb185DxlnZg3NjZSU1PD8OHDMbO4wyl67s66detobGxkxIgRcYcjIgViZlOAG4EK4HZ3/16b+4cBdwP9o2Mud/eHCx6oSBlTzpNb+cp5yrqUJZVKMWDAAL1Bc8TMGDBggL6Ni5QRM6sAbgW+DIwGzjKz0W0O+ybwgLtPBM4E/ruwUYqIcp7cylfOU9aJOaA3aI7p9RTJIJWCpqZwXXomAQ3u/rq7bwbmAVPbHONAv+j27sCbBYwvcUr77SBJps/o3MrH61nWpSwiInnX0AB1ddDSApWVMG0ajBwZd1S5NBhYnbbdCExuc8zVwONmdjFQDXwx04nMbBYwC2DYsGE5DzQJSv/tICI9UfY95iIieZNKhSyspgaGDg3XdXXl2FV6FnCXuw8BTgDuNbNPfP64+1x3r3X32oEDS2+hU70dRKQzSsxlJ1u2bIk7BJHk6moNQnNz6Bqtrg7b1dVhu7k5fzEW3hpgaNr2kGhfunOABwDc/TmgCtirINElSHm8HUSKS9LyHiXmXZXj4sCf/OQnTJo0iQkTJnDuueeydetWzj//fGpraxkzZgxXXXXV9mMvv/xyRo8ezbhx47jssstobm5mxIgRtLS0APDhhx/utN3Wa6+9xpQpUzj00EM55phjeOWVVwCYOXMm5513HpMnT2b27NksXLiQI444gokTJ3LkkUfy6quvAlBfX7891nHjxrFy5cqcvAYiRaGhAebMgZtuCtcNDZ0/pqYm1Cts3Bi2N24M2zU1+Y21sBYBo8xshJn1JgzuXNDmmL8AxwKY2UGExLypoFEmQHm8HaSk5GFAhPKejqnGvCtyXBy4YsUK7r//fp555hkqKyu54IILuO+++7j22mvZc8892bp1K8ceeyzLli1j8ODBPPTQQ7zyyiuYGe+//z41NTV87nOf41e/+hUnn3wy8+bN45RTTml3Ps1Zs2bxox/9iFGjRvHCCy9wwQUX8OSTTwJhGqVnn32WiooKPvzwQ373u9/Rq1cv/u///o9vfOMbPPjgg/zoRz/i0ksv5eyzz2bz5s1s3bq123+7SFFJr0Gorg4ZVV0dzJ4NVVXtP66qKrQTdXWwfv2OdqOjxxQZd99iZhcBjxGmQrzT3evN7BpgsbsvAP4FuM3M/okwEHSmu3t8UcejDN4OUkryMCBCeU/nlJhnq7sfzB144oknWLJkCYcddhgAmzZtYtCgQTzwwAPMnTuXLVu28NZbb/Hyyy8zevRoqqqqOOecczjxxBM58cQTAfiHf/gH5syZw8knn8yPf/xjbrvttozPtWHDBp599llOP/307fs+/vjj7bdPP/10KioqAPjggw+YMWMGK1euxMy2fxM94ogjuPbaa2lsbOSUU05h1KhR3fq7RYpOphqE9evD/s7+/48cGdqJ5ubQfpRgFhbNSf5wm31Xpt1+GTiq0HElURm8HaQU5CHnAeU92VApS7byUBzo7syYMYOlS5eydOlSXn31VWbMmMEPfvADnnjiCZYtW8Zf//Vfk0ql6NWrFwsXLuS0007jl7/8JVOmTAHgqKOOYtWqVTz11FNs3bqVsWPHZnyubdu20b9//+3PtXTpUlasWLH9/urWvwv41re+xec//3mWL1/OL37xi+1zdE6bNo0FCxbQt29fTjjhhO3fOkVKXldqEDL99FtVBQMHKgsTQG8HKQJ5GhChvKdzSsyzlYfiwGOPPZb58+ezdu1aANavX89f/vIXqqur2X333XnnnXd45JFHgPDN74MPPuCEE07ghhtu4MUXX9x+nunTpzNt2jS+9rWvtftc/fr1Y8SIEfz0pz8Fwn+O9HOk++CDDxg8eDAAd9111/b9r7/+Op/+9Ke55JJLmDp1KsuWLev23y5SVFprEJqbYfXqcJ2pBqE7degiIkmTpwERyns6p8Q8W9l+MHfB6NGj+e53v8vxxx/PuHHjOO644+jTpw8TJ07kwAMPZNq0aRx1VPj1t7m5mRNPPJFx48Zx9NFHc/31128/z9lnn817773HWWed1eHz3Xfffdxxxx2MHz+eMWPG8POf/zzjcbNnz+aKK65g4sSJO41WfuCBBxg7diwTJkxg+fLlTJ8+vdt/u0jRaa1BuOSScN221lJz4YlIqchDzgPKe7Jh5TL+pra21hcvXrzTvhUrVnDQQQd17USpVOKKA+fPn8/Pf/5z7r333rhDAbr5uooUu6am0FM+NG3mwNWrQyLfwzm5zWyJu9f2MMKikqnNFpHuK5WcB5KV92R6XXvSZmvwZ1dVVSXqzXnxxRfzyCOP8PDDD3d+sIjkx/vvw5o1sGVL+Mm3dbCU5sITkWKWsJwHSj/vUWJe5G6++eZP7Lvwwgt55plndtp36aWXdliLJSLd9MQTcO21YWBUSwsccAAMG6a58ERE8qDU8x4l5iXo1ltvjTsEkfLw/vshKe/XD3bbDTZsgFWr4JvfDCUtSspFRPKulPIeDf4UEemut94KveS77Ra2d9sNtm0LJS1KykVEpIvynpib2RQze9XMGszs8gz39zGz+6P7XzCz4dH+48xsiZm9FF1/Ie0xT0XnXBpdBuX77xAR+YR994VddoF162Dr1tBjXlkZ9ouIiHRRXktZzKwCuBU4DmgEFpnZgmgVuFbnAO+5+0gzOxO4DjgDeBf4G3d/08zGEpZ7Hpz2uLPdXUP2RSQ+774Lw4fDb38L7iEhv/Za6N8/7shERKQI5bvGfBLQ4O6vA5jZPGAqkJ6YTwWujm7PB24xM3P3P6YdUw/0NbM+7v4xIiJxa523fMIEmDgxlLVUVMBRWnleRES6J9+lLIOB1Wnbjezc673TMe6+BfgAGNDmmFOBP7RJyn8clbF8y8ws05Ob2SwzW2xmi5uamnryd4iI7Cx9yerKShg0KPSaNzWFS9uFhVKpzPtFREQiiZ+VxczGEMpbjk/bfba7rzGzGuBB4KvAPW0f6+5zgbkQFqsoQLhdsmrVKk488USWL1+ek/PttttubNiwISfnEpFOtC5ZvWoVLF8OmzaFZP3DD2HPPXdMlzhyJDQ0hN71lpad94uIlBHlPZ3Ld4/5GiBtGTyGRPsyHmNmvYDdgXXR9hDgIWC6u7/W+gB3XxNdNwN1hJKZgkhKp1f6krG54u5s27Yt5+cVKUlVVXDqqfD88yEp7907JN2vvQb77BMS97q6MKViXV04vn//cF1XF38jIiLSiaTkPFA+eU++E/NFwCgzG2FmvYEzgQVtjlkAzIhunwY86e5uZv2BXwGXu/v2WePNrJeZ7RXdrgROBHLz1asTDQ0wZ05YdXvOnLDdU1u3buXrX/86Y8aM4fjjj2fTpk3cdtttHHbYYYwfP55TTz2Vjz76CICZM2dy3nnnMXnyZGbPns0bb7zBEUccwcEHH8w3v/nNTp/r+9//Pocddhjjxo3jqquuAsK31wMOOIDp06czduxYVq9ezfnnn09tbS1jxozZfhzA5ZdfzujRoxk3bhyXXXZZz/94kWI3aBBMngxTpsAxx4RkvKIifIpVV4ce8rfegnfegd//Hp58Mly/807oXRcRSah85DygvKczeU3Mo5rxiwgzqqwAHnD3ejO7xsxOig67AxhgZg3APwOtUypeBIwErmwzLWIf4DEzWwYsJfS435bPvwN2jPOqqQnrhrR2hvX0W+TKlSu58MILqa+vp3///jz44IOccsopLFq0iBdffJGDDjqIO+64Y/vxjY2NPPvss1x//fVceumlnH/++bz00kvs28n0bI8//jgrV65k4cKFLF26lCVLlvD0009vj+GCCy6gvr6e/fbbj2uvvZbFixezbNkyfvvb37Js2TLWrVvHQw89RH19PcuWLcvqP4RIyaupgb59w5SJNTVhDvOtW0Ov+MaNoQd9jz1gxYpwzB57hOsVK8J9IiIJlK+cB5T3dCbv85i7+8Pu/hl339/dr432XenuC6LbKXc/3d1Huvuk1hlc3P277l7t7hPSLmvdfaO7H+ru49x9jLtf6u5b8/13pI/zgh2dYT3t9BoxYgQTJkwA4NBDD2XVqlUsX76cY445hoMPPpj77ruP+vr67ceffvrpVFRUAPDMM89w1llnAfDVr361w+d5/PHHefzxx5k4cSKHHHIIr7zyCitXrgRgv/324/DDD99+7AMPPMAhhxzCxIkTqa+v5+WXX2b33XenqqqKc845h//93/9l11137dkfLlIKWstZ3n4bGhvhgAPC5e23Yf16+NKXQkNx0EEhaV+/PlwfdFDYLyKSQPnKeUB5T2cSP/gzKVrHeW3cGN6grZ1hNTU9O2+fPn22366oqGDTpk3MnDmTn/3sZ4wfP5677rqLp556avsx1a3/SyLtTEjzCe7OFVdcwbnnnrvT/lWrVu10zjfeeIMf/OAHLFq0iD322IOZM2eSSqXo1asXCxcu5IknnmD+/PnccsstPPnkk934i0VKSEMDzJsXupF69YJZs2D//WHZMvjFL+CXvwzH9eoFRx8dyly2bg3H97TxEBHJk3zlPKC8pzN57zEvFVVVYSKF5mZYvTpcT5uWn1W3m5ub2XfffWlpaeG+++5r97ijjjqKefPmAXR4HMCXvvQl7rzzzu2jl9esWcPatWs/cdyHH35IdXU1u+++O++88w6PPPIIABs2bOCDDz7ghBNO4IYbbuDFF1/s7p8nUlzaG/2USsHNN8OiRbByJdTXw9y5Yf+jj4aZWYYODdcQGo333w/356vxEBHJgULmPKC8J516zLtg5EiYPTu8QWtq8vcG/c53vsPkyZMZOHAgkydPprmd345uvPFGpk2bxnXXXcfUqVM7POfxxx/PihUrOOKII4AwxdBPfvKT7T8PtRo/fjwTJ07kwAMPZOjQoRwVLZbS3NzM1KlTSaVSuDvXX399Dv5SkYTraJrDpiZYujQMAO3TBz7+OGyvWvXJ34D79Qu96VVV+W08RERypFA5DyjvSWfuiZveOy9qa2t98eLFO+1bsWIFBx10UEwRlS69rlISUqkwFUFNzY7fcpubwydVVVXoRvrKV2DvvcNUiZs3h9lW7rhjx6ipTI/rBjNb4u61Of4LEy1Tmy0i3afP5vzI9Lr2pM1WKYuISCadjX4aOBAmTAgLCr33XrieMAGGDCnsb8AiIlIyVMpSYl566aVPjFTu06cPL7zwQkwRiRSpzkY/VVXBxRfD3XfDRx/BrrvCjBlh/5Ah8LWvheMGDlRSLiKSJ6WW95R9Yu7uWY/wLQYHH3wwS5cuje35y6U0SspA6+inurowzWFrjXl6kj1yJPz7v+9chNlRXbqISIxKLeeBePOefOQ8ZZ2YV1VVsW7dOgYMGFByb9Q4uDvr1q2jSr2DUiqyGf1UVRUuqVQoXbn77jATS2sve11dj+rLRURyQTlPbuUr5ynrxHzIkCE0NjbS1NQUdyglo6qqiiFDhsQdhkjutCbe7Uml4MUXw5zlH30EixfDcceFxLy6OvS2NzcrMReRWCnnyb185DxlnZhXVlYyYsSIuMMQkWLV0BB6yJ9+OkyZOG5cmKHl97+Hv/mbMFNLrlblEBHpAeU8xaGsE3MRkW5LpUKZSu/eYAZr1sAbb8CnPhV6zl97Dfr314wsZSqVKsz8z3E9n4jkhxJzEZHuaJ1OccAAePvtkA3tskuYnQXg/PPDyp/KkspOocf/aryxSOnQPOYiIt3ROp1iczPss0/IijZtCtcHHhh6y5WUl53WH1JqasL3spqasJ1KlcbziUh+KTEXEemO1ukUN28Gd9hjj5Cgu8PKlbB2bdwRFoyZTTGzV82swcwuz3D/DWa2NLr8yczejyPOQuhsXapifz4RyS8l5iIi3dU6j/l3vwsVFdC3b8iMDj8cHnywLLotzawCuBX4MjAaOMvMRqcf4+7/5O4T3H0CcDPwv4WPtDDS16WCT65LVezPJyL5pcRcRKQnqqpC6cqRR8KUKfClL8Hw4eXUbTkJaHD31919MzAPmNrB8WcB/1OQyGLQ+kNKc3OY1r65Ob/jfwv9fCKSXxr8KSLSU5WVsG0bbN26o/uyfLotBwOr07YbgcmZDjSz/YARwJMFiCs22axLVczPJyL5o8RcRKQnWqfE2LABFi2Cgw6CvfdWt2VmZwLz3X1rpjvNbBYwC2DYsGGFjCtn0qctHDiwcM/b2TpYIlIclJiLiHRX+pQY++wD++0H69bBJZeEWVnKwxpgaNr2kGhfJmcCF7Z3InefC8wFqK2t9VwFWCiatlBEeko15iIi3dV2Soz+/cOCQy0t8cZVWIuAUWY2wsx6E5LvBW0PMrMDgT2A5wocX0Fo2kIRyQUl5iIi3aUpMXD3LcBFwGPACuABd683s2vM7KS0Q88E5rl70fWEZ0PTFopILqiURUSku1qnxKirg/Xrd9QvlFmxr7s/DDzcZt+VbbavLmRMhZb+Ha26uiy/o4kkR/pgjyJrj5WYi4j0hKbEEPQdTSQxinywhxJzEZGe0pQYgr6jicQufbBH609XdXXhP2aR/IdUjbmIiEiOVFWFaRKLJAcQKS2dDfZIpaCpKdGjstVjLiIiIiLFr6PBHkVS4qIecxEREREpfq2DPZqbYfXqcD1tWrivSOYzVY+5iIiIiJSGTIM9mpo+WeKyfn04JmF1Z0rMRURERKR0tB2QX0TzmaqURURERESSI9eDNNsrcUlYbzmox1xEREREkiJfgzSLZD5T9ZiLiIiISPzS5yHPxyDNIpjPVIm5iIiIiMSvs3nIy4AScxERERGJX/ogTUj0IM18UWIuIiIiIvErokGa+aLBnyIiIgmXSiV+zJpIbhTJIM18UWIuIiKSYB1NUqGEXUpS23nI8y3Tf6SY/nMpMRcREUmo9EkqWtdFqasLHYqNjfmZVU6krGT65gux/edSjbmIiEhCtTdJRVNTfmeVE0m8XCxClGl6xrvvDpeY/nOpx1xERCSh2ltJHGDTph2JenU1rF8fEnmVtEjJy9UiRJm++TY2gntIylv3FfA/l3rMRUREEqq9SSo+/BBeeAEefRQeewxWrSq7WeWkXGW7CFE2PeqZpmfcddcd34Jb9xXwP5d6zEVERBKs7SQVAHPmwOGHw/Lloef8+efh5pvVWy5lIFMvd9se7Wx71Fu/+dbVhXNUVsKMGeG+9H0FnLIx74m5mU0BbgQqgNvd/Xtt7u8D3AMcCqwDznD3VWZ2HPA9oDewGfhXd38yesyhwF1AX+Bh4FJ393z/LSIiInFIn6SiqSnkG8OHw+DBoUPw3Xdh0KBYQxTpmWxnQWmvvqv1W2tHI6Yznbe96RljmrIxr4m5mVUAtwLHAY3AIjNb4O4vpx12DvCeu480szOB64AzgHeBv3H3N81sLPAYMDh6zA+BrwMvEBLzKcAj+fxbREREOlOIGdba5iWbN0PfvipjkSLWlZrxTL3c6T3a2fSoZzpn2/sKPWVjJN895pOABnd/HcDM5gFTgfTEfCpwdXR7PnCLmZm7/zHtmHqgb9S7vifQz92fj855D3AySsxFRCRGuRqP1pnO8hKRotLVHm4I/7EuuQTeegv23Rf6999xX2c96gmX78R8MLA6bbsRmNzeMe6+xcw+AAYQesxbnQr8wd0/NrPB0XnSzzmYDMxsFjALYNiwYT34M0RERNrXndyiJ8p8cUQpJd3p4W79FrxpE2zbBtOnw5gx4b4i/+aa+MGfZjaGUN5yfFcf6+5zgbkAtbW1qkEXEZG86E5u0VMx/dIukltd7eFu/RacSkF9fUjOFy0Ko59bk/Mi/uaa7+kS1wBD07aHRPsyHmNmvYDdCYNAMbMhwEPAdHd/Le34IZ2cU0REpGAyzbpWRL+ei8SnvTlB20umm5tDMr58OfTpA3vvDRUVcM89O0+NWFUFAwcWVVIO+e8xXwSMMrMRhOT5TGBam2MWADOA54DTgCfd3c2sP/Ar4HJ3f6b1YHd/y8w+NLPDCYM/pwM35/nvEBERaVeR/3ouEq+u9HDX1ITylU2boF+/HaOfd9nlkz9RZTMauxAjtrsgr4l5VDN+EWFGlQrgTnevN7NrgMXuvgC4A7jXzBqA9YTkHeAiYCRwpZldGe073t3XAhewY7rER9DATxERiVkR/3ouEr9sa7OqqkJN+aJF8M47ISkfOzbsT/+JKpvR2IUasd0FVi7Tf9fW1vrixYvjDkNEpMvMbIm718YdRyGpzRaRDtXXh/KVXXYJyXl6Up1KhVW40kdjNzfvPBo7m2O6qSdtduIHf4qIiIiI7GTMGPj2tz/5E1UqBW+8EUpd9tkn7Ms0GjuOEdtZUGIuIiIiOZOwkl0pZW3LX9KnUXz22XA9alSoQ287Gjuh850rMRcRkR4xsynAjYSxRLe7+/cyHPN3hMXkHHjR3dtOBCAlIIElu1IuWqdRrKoKCXlLC/zsZzB0KBxyCFx88c5JfEJHbCsxFxGRbjOzCuBW4DjCgm+LzGyBu7+cdswo4ArgKHd/z8wGxROt5FOhF1kS2UlzcxgMumoVrFwZEu1hw+DII8PtIUM++ZgEjtjO9zzmIiJS2iYBDe7+urtvBuYBU9sc83XgVnd/DyCaXUtKTKaS3ZaWsF8kJ1IpaGraeb7yVpWVsGJFmEqxd++w/c47O+rM23sjJmy+c/WYi4hITwwGVqdtNwKT2xzzGQAze4ZQ7nK1uz/a9kRmNguYBTBs2LC8BCv5k9CSXSkVndVJtbTAQQeFHvPNm8E9JOXNzUX1RlSPuYiI5FsvYBTwOeAs4LZoEbmduPtcd69199qBAwcWOETpqa4u4CiStfQ6qaFDw3Vd3c495zU1YRXQz34WTj0VPvWpsH/z5qJ6I6rHXEREemINMDRte0i0L10j8IK7twBvmNmfCIn6osKEKIWSwJJdKQXZTG2YPpizVy84/HA48UQYP76o3ohKzEVEpCcWAaPMbAQhIT8TaDvjys8IPeU/NrO9CKUtrxc0SimYbBdwFMlatnVSJfDNUKUsIiLSbe6+BbgIeAxYATzg7vVmdo2ZnRQd9hiwzsxeBn4D/Ku7r4sn4uTqaFybSFnrSp1UwgZzdpV6zEVEpEfc/WHg4Tb7rky77cA/RxfJQPN/i3SiBHrDs6EecxERkRhlM65NRCj63vBsKDEXERGJkeb/FpFWSsxFRAQLvmJmV0bbw8xsUtxxlYP0cW2g+b9FypkScxERAfhv4AjC7CkAzcCt8YVTPjT/twga/RzR4E8REQGY7O6HmNkfAdz9PTPrHXdQ5SIf49pSqZIfJyelQqOft8s6MTezwcB+6Y9x96fzEZSIiBRci5lVAA5gZgOBbfGGVF5yNf93KgXLlsEvfhG2yzzPkaRLH/3cOkd5XV34plqG3yizSszN7DrgDOBlYGu02wEl5iIipeEm4CFgbzO7FjgN+Ga8IUlXNTTA3XfD009D375w9NEh1ynjPEeSLptVPbuqiH8uyrbH/GTgAHf/OJ/BiIhIPNz9PjNbAhwb7TrZ3VfEGZN0TWvHY+/eIbfZbTdYuBC+9KUds7wUWY4i5SDbVT2z1bYs5tRTYdCgoknSsx38+TpQmc9AREQkdrsCFYTPhr4xxyJd1NrxuOee0KsXuMPWraHzUbO8SGJlM/o524GhbRcFSKXg4ovh+uthzpyQtCdctj3mHwFLzewJYHuvubtfkpeoRESkoKJpEk8HHgQM+LGZ/dTdvxtvZJKt1o7HzZth0iT4/e9h06awPWNGUXQWSrnqaPRzVwaGppfFtLTA8uVQUQF77QW77FIUNV3ZJuYLoouIiJSms4Hx7p4CMLPvAUsBJeZForXjsTWHOewwOPFEGD8+0XmISJBp9HNXB4aml8Vs2xa+mfbtG46trOx57XoBZJWYu/vd+Q5ERERi9SZQBbT+VtwHWBNfONId+Zh2USQ2XR0Ymv7tdNOmUMs1duyOZL0IarqynZVlFPAfwGhCww2Au386T3GJiEhhfQDUm9mvCbNuHQcsNLObQKWLxSRX0y6KxK47A0PTv52uXQsPPhhq11vLYBL+nyPbUpYfA1cBNwCfB76GVg0VESklD0WXVk/FFIeISJDeA946ijmb5Lr12+nAgbD//kX1E1K2iXlfd3/CzMzd/wxcHU2rdWUeYxMRkQKIFhY63t3PjjsWEZGd9LQ+q8h+Qso2Mf/YzHYBVprZRYS6w93yF5aIiBSKu281s/3MrLe7b447HhGRnRRZct0T2SbmlxLmt70E+A6hnGVGvoISEZGCex14xswWABtbd7r79fGFJCJSXrKdlWURgJltc/ev5TckERGJwWvRZRcg2dMWiIiUqGxnZTkCuINQvjLMzMYD57r7BfkMTkRECsPdvx13DCIi5S7bUpb/Ar5EtMiQu79oZn+Vt6hERKSgzGwgMBsYw87T4n4htqBERMpM1lMeuvvqNru25jgWERGJz33AK8AI4NvAKmBRnAGJiJSbbBPz1WZ2JOBmVmlmlwEr8hiXiIgU1gB3vwNocfffuvvfA+otFxEpoGwT8/OAC4HBhKkSJ0TbIiJSGlqi67fM7K/NbCKwZ5wBiYiUm2xnZXkX0MITIiKl67tmtjvwL8DNQD/gn56dHlQAACAASURBVOINSUSkvGQ7K8sI4GJgePpj3P2k/IQlIiKF5O6/jG5+QFirQkRECizbWVl+Rpgu8RfAtvyFIyIicYhmZfk6n+yA+fu4YhIRKTfZJuYpd78pr5GIiEicfg78Dvg/NOuWiEgssk3MbzSzq4DHgY9bd7r7H/ISlYiIFNqu7v5vcQchIlLOsk3MDwa+Spg6q7WUxdFUWiIipeKXZnaCuz8cdyAiIuUq28T8dODT7r45n8GIiEhhmVkzoaPFgG+Y2ceEqRMNcHfvF2d8IiLlJNvEfDnQH1ibx1hERKTA3L0m7hikdKVS0NwMNTVQVRV3NCLJl+0CQ/2BV8zsMTNb0HrJ5oFmNsXMXjWzBjO7PMP9fczs/uj+F8xseLR/gJn9xsw2mNktbR7zVHTOpdFlUJZ/R6xSKWhqCtciIkliZkeZWXV0+ytmdr2ZDcvysZ218zPNrCmtzf6HXMcPamOTpqEB5syBm24K1w0NcUckknzZ9phf1Z2Tm1kFcCtwHNAILDKzBe7+ctph5wDvuftIMzsTuA44A0gB3wLGRpe2znb3xd2JKw4NDVBXBy0tUFkJ06bByJFxRyUist0PgfFmNp6wyNDtwL3AZzt6UJbtPMD97n5R7sMO2mtj1WMbj1Qq/HvU1EB1NWzcGLZnz9a/g0hHsl3587cd3W9mz7n7ERnumgQ0uPvr0XHzgKlAeoM9Fbg6uj0fuMXMzN03Ar83s6JPX9VAiUgR2OLubmZTgVvc/Q4zOyeLx2XTzudVe23sqafCgw+qQyQOzc3hda+uDtvV1bB+fdivzz2R9mVbytKZ9v6bDQZWp203RvsyHuPuWwirzg3I4jl/HP0k+i0zsy7GW1CZGqiWlrBfRCQhms3sCuArwK/MbBegMovHZdPOA5xqZsvMbL6ZDc10IjObZWaLzWxxU1NT9oFnaGM3bYJ77gnJ+tCh4bquTmUuhVJTE74MbdwYtjduDNs1GtEg0qFcJeaeo/Nk62x3Pxg4Jrp8NdNB3W3kc00NlIgUgTMI61Sc4+5vA0OA7+fo3L8Ahrv7OODXwN2ZDnL3ue5e6+61AwcOzPrkmdrYbdtgl13UIRKXqqrwC0VzM6xeHa6nTVNvuUhncpWYt2cNkN4zMiTal/EYM+sF7A6s6+ik7r4mum4G6gg/pWY6rluNfK6pgRKRpHP3t939enf/XbT9F3e/p/V+M3uunYd22s67+zp3b12c7nbg0NxFnrmNnT4d+vZVh0icRo4MJZuXXBKuVUYk0rlsB392pr1SkkXAKDMbQWiozwSmtTlmATADeA44DXjS3dvtgY+S9/7u/q6ZVQInEpaQTrTWBkqDkESkSLXXanXazpvZvu7+VrR5ErAi18FlamP79AnlK+vX76gxV9tbWFVVes1FuiJXiXnGUhJ332JmFwGPARXAne5eb2bXAIvdfQFwB3CvmTUA6wmNOgBmtgroB/Q2s5OB44E/A49FSXkFISm/LUd/R16pgRKRIpaxwyTLdv4SMzsJ2EJo52fmI8C2baw6RESk2FgHndPpK8J94i6KbEW42tpaX7y4aGZXFBHZzsyWuHttzDH8wd0PKdTzqc0WkWLVkza7wx5zrQgnIiKRRM9+JSJSCrpUyhKtsLn9x0B3/0vOIypxWuxCRIpUxpJFERHJnawS86g28D+BTwFrgf0Ig3fG5C+00qPVP0UkabItWXT35QUNTESkDGU7XeJ3gMOBP7n7COBY4Pm8RVWC0lem02IXIpIU7l7j7v0yXGqKaRyRiEgpyDYxb3H3dcAuZraLu/8GiHUgUrHR6p8iUgzMbJCZDWu9xB2PiEg5ybbG/H0z2w34HXCfma0FNuYvrNKTvjJddbUWuxCRZFHJoohI/LLtMf8NYUXOS4FHgdeAv8lXUKVIq3+KSMKpZFFEJGbZ9pj3Ah4nLAxxP3B/VNoiXaDFLkQkwVrcfZ2ZbS9ZNLP/ijsoEZFyklVi7u7fBr5tZuOAM4Dfmlmju38xr9GVIK3+KSIJpZJFEZGYZVvK0mot8DawDhiU+3BERCQmKlkUEYlZVom5mV1gZk8BTwADgK+7+7h8BibZSaWgqUnTLopIj7WWLD4F1KCSRRGRgsu2xnwo8I/uvjSfwUjXaMEiEckVlSyKiMQvqx5zd79CSXmyaMEiEckTlSyKiMSkqzXmkhBasEhEckkliyIi8cu2lEUSRgsWiUiOqWRRRCRm6jEvUlqwSERySSWLIiLxU495gaRSuV9YSAsWiYiIiJQOJeYF0NAAd98NH30Eu+4KM2bkbvYULVgkIiIiUhqUmOdZKgU33wyvvgoVFbB5M7z9Nnz/+9C/f9zRiYhIHPLxK6qIFD8l5nnW1ARLl8Lee4eG+M03Q5Lerx+cf77mHRcRKTdag0JE2qPBnwWyZQusWgW9eoXekepqzTsuIlJutAaFiHREiXmeDRwIEybA+++HGvMtW0JjvPfemndcRKTcaA0KEemIEvM8q6qCiy+GyZNh0CD41Kfg858Pteaad1xEpLykr0EBWoNCRHamxLwARo6Eq66C//ovOPxw+PjjjucdT6VCbbp+2hQRKS1ag0JEOqLBnwU0YgT8y7+Eny3bG4nfOiho0ybYtg2mT4cxYwofq4iI5IfWoBCR9igxL4BMI/AHDvzkca2DglIpqK8PyfmiRWG6RSXnIiKlQ2tQiEgmKmXJs66MwG9uDsn48uWhwd577zD3+T33qKxFREREpNQpMc+zrozAr6kJ5SubNkHv3qEWvW9f2GUXjdgXERERKXVKzPOsKyPwq6pCTfnWrfDOOyExHzs2JOcasS8iIiJS2pSY51lnI/DbzsAyZgz84Adw4IFwwAE7Hq9aRBEREZHSpsGfBdDeCPxMg0IBfvlL2G23UNZy6qlaqllEks3MpgA3AhXA7e7+vXaOOxWYDxzm7osLGKKISFFQYl4gbUfgpw8Kra4OJS533x3u23NP2GefsO/BB2H//dVjLiLJZGYVwK3AcUAjsMjMFrj7y22OqwEuBV4ofJQiIsVBpSwxyTQo9KOPQjKupZpFpIhMAhrc/XV33wzMA6ZmOO47wHWA5pgSEWmHEvOYZBoUuuuuO3rPAd5/HzZvDseJiCTUYGB12nZjtG87MzsEGOruvypkYCIixUaJeUwyDQqdMSNcmpvhxRfh0Udhwwa46aZQjy4iUmzMbBfgeuBfsjh2lpktNrPFTU1N+Q9ORCRhVGMeo/YGhV5yCfzHf8CUKdC/f+hBr6sLx6rWXEQSZg0wNG17SLSvVQ0wFnjKzAD2ARaY2UltB4C6+1xgLkBtba3nM2gRkSRSj3nMqqpg4MCdE+6WlrDAUP/+YVu15iKSYIuAUWY2wsx6A2cCC1rvdPcP3H0vdx/u7sOB54FPJOUiIqLEPJHa1p+r1lxEksrdtwAXAY8BK4AH3L3ezK4xs5PijU5EpLgoMU+g9Ppz1ZqLSNK5+8Pu/hl339/dr432XenuCzIc+zn1lu/QdpE5ESlvSswTauTIUGu+226h1nz8+NCTXlenBlxEpBQ0NMCcOaHTZc4cdbyIiBLzRFOtuYhIaUpfZG7oUHW8iEigxDzBMs11XlkZ9ouISPHKtMhcITteVEIjkkx5T8zNbIqZvWpmDWZ2eYb7+5jZ/dH9L5jZ8Gj/ADP7jZltMLNb2jzmUDN7KXrMTRbNwVVqMs11Pm2apkwUESl2cXa8qIRGJLnympibWQVwK/BlYDRwlpmNbnPYOcB77j4SuIGwZDOEZZu/BVyW4dQ/BL4OjIouU3IffTK0znV+ySXheuTIuCMSEZGeiqvjRSU0IsmW7wWGJgEN7v46gJnNA6YCL6cdMxW4Oro9H7jFzMzdNwK/N7OdUlEz2xfo5+7PR9v3ACcDj+TzD4lTVZV6yUVESk17i8zlU6YSmvXrw359zojEL9+lLIOB1WnbjdG+jMdE8+F+AAzo5JyNnZxTREQk8TItMpdPGrskkmwlPfjTzGaZ2WIzW9zU1JSz82rQjIiIFCONXRJJtnyXsqwBhqZtD4n2ZTqm0cx6AbsD6zo555BOzgmAu88F5gLU1tZ6VwJPpTL/vNjQEOrxWlpCL8O0aar7FhGR4hFHCY2IZCffPeaLgFFmNsLMegNnAm1XglsAzIhunwY86e7tJtHu/hbwoZkdHs3GMh34eS6Dbm/EemeDZtSTLiIixaDQJTQikp289pi7+xYzuwh4DKgA7nT3ejO7BlgcLdd8B3CvmTUA6wnJOwBmtgroB/Q2s5OB4939ZeAC4C6gL2HQZ84GfqYn39XVof6urm5H70J7g2YaG9WTLiIiIiLdl+9SFtz9YeDhNvuuTLudAk5v57HD29m/GBibuyh36Cj5Th8005q0V1aGS3vJvHojRERERCQbJT34szs6GrHe3qCZlpZ4V3ATERERkeKX9x7zYtOafNfVhZ7y1rKU1p7vTINmUqnMPemafkpEREREsqXEPIPORqy3XfCns2ReRERERKQzSszb0dXVNjX9lIiIiIj0hBLzHOpqMi8iIiIi0kqDPxNCc6CLiEh36PNDpHSoxzwBtJqoiIh0hz4/REqLesxj1tlqoiIiIpno80Ok9Cgxj1mmBY00B7qIiHRGnx8ipUeJecw6WtBIRESkPfr8ECk9Ssxj1t5qoprdRUREOqLPD5HSo8GfCaA50EVEpDv0+SFSWpSYJ4TmQBcRke7Q54dI6VApi4iIiIhIAigxFxERERFJACXmIiIiIiIJoMRcRERERCQBlJiLiIiIiCSAEvOES6WgqUlLLIuIiIiUOk2XmGANDVBXF5ZYrqwMC0eMHBl3VCIiOzOzKcCNQAVwu7t/r8395wEXAluBDcAsd3+54IGKiCSceswTKpUKSXlNDQwdGq7r6tRzLiLJYmYVwK3Al4HRwFlmNrrNYXXufrC7TwDmANcXOEwRkaKgxDwBMpWrNDeHnvLq6rBdXR22m5vjiVFEpB2TgAZ3f93dNwPzgKnpB7j7h2mb1YAXMD4RkaKhUpaYtVeuUlMTtjduDEn5xo1hu6Ym7ohFRHYyGFidtt0ITG57kJldCPwz0Bv4QqYTmdksYBbAsGHDch6oiEjSqcc8Rh2Vq1RVhSS9uRlWrw7X06Zp2WURKU7ufqu77w/8G/DNdo6Z6+617l47cODAwgYoIpIA6jGPUaZylfXrw/6qqtBzPnt22K6pUVIuIom0Bhiatj0k2teeecAP8xqRiEiRUo95jNLLVSBzuUpVFQwcqKRcRBJrETDKzEaYWW/gTGBB+gFmNipt86+BlQWMT0SkaCgxj5HKVUSk2Ln7FuAi4DFgBfCAu9eb2TVmdlJ02EVmVm9mSwl15jNiCldEJNFUyhIzlauISLFz94eBh9vsuzLt9qUFD0pEpAgpMU+Aqiol5CIiIiLlTqUsIiIiIiIJoMRcRERERCQBlJiLiIiIiCSAEnMRERERkQRQYi4iIiIikgBKzEVEREREEkCJuYiIiIhIAigxFxERERFJACXmIiIiIiIJoMRcRERERCQBlJiLiIiIiCSAEnMRERERkQRQYl4iUiloagrXIiIiSaHPJ5Hs9cr3E5jZFOBGoAK43d2/1+b+PsA9wKHAOuAMd18V3XcFcA6wFbjE3R+L9q8CmqP9W9y9Nt9/R5I1NEBdHbS0QGUlTJsGI0fGHZWIiJQ7fT6JdE1ee8zNrAK4FfgyMBo4y8xGtznsHOA9dx8J3ABcFz12NHAmMAaYAvx3dL5Wn3f3CeWelKdSodGrqYGhQ8N1XZ16JkREJF76fBLpunyXskwCGtz9dXffDMwDprY5Zipwd3R7PnCsmVm0f567f+zubwAN0fkkTXNz6Imorg7b1dVhu7k53rhERKS86fNJpOvynZgPBlanbTdG+zIe4+5bgA+AAZ081oHHzWyJmc3KQ9xFo6Ym/Dy4cWPY3rgxbNfUxBuXiIiUN30+iXRdsQ7+PNrdDyGUyFxoZn+V6SAzm2Vmi81scVNTU2EjLJCqqlCz19wMq1eH62nTwn4REZG46PNJpOvyPfhzDTA0bXtItC/TMY1m1gvYnTAItN3Hunvr9Voze4hQ4vJ02yd397nAXIDa2lrPwd+TSCNHwuzZodGrqVGjJyIiyaDPJ5GuyXeP+SJglJmNMLPehMGcC9ocswCYEd0+DXjS3T3af6aZ9TGzEcAoYKGZVZtZDYCZVQPHA8vz/HckXlUVDByoRk9ERJJFn08i2ctrj7m7bzGzi4DHCNMl3unu9WZ2DbDY3RcAdwD3mlkDsJ6QvBMd9wDwMrAFuNDdt5rZ3sBDYXwovYA6d380n3+HiIiIiEi+5X0ec3d/GHi4zb4r026ngNPbeey1wLVt9r0OjM99pCIiIiIi8SnWwZ8iIiIiIiVFibmIiIiISAIoMRcRERERSQAl5iIiIiIiCWBhZsLSZ2ZNwJ/jjqMDewHvxh1EFoolTiieWBVn7hVLrNnGuZ+7D8x3MEnShTa7WP6t2yrWuKF4Yy/WuKF4Yy/WuKFnsXe7zS6bxDzpzGyxu9fGHUdniiVOKJ5YFWfuFUusxRJnkhXra1iscUPxxl6scUPxxl6scUN8sauURUREREQkAZSYi4iIiIgkgBLz5JgbdwBZKpY4oXhiVZy5VyyxFkucSVasr2Gxxg3FG3uxxg3FG3uxxg0xxa4acxERERGRBFCPuYiIiIhIAigxFxERERFJACXmeWJmU8zsVTNrMLPLM9zfx8zuj+5/wcyGR/sHmNlvzGyDmd3S5jFPRedcGl0GxRjncWa2xMxeiq6/kPaYQ6P9DWZ2k5lZQuPM+evZw1gnpcXyopn9bbbnTFCcq6LXeqmZLY4zzrT7h0X/ny7L9pwJijPnr2cxyuK1nWlmTWnvy3+II85MsnmvmdnfmdnLZlZvZnWFjjGTLF7zG9Je7z+Z2ftxxJlJFrEPs/A5+0czW2ZmJ8QRZyZZxL6fmT0Rxf2UmQ2JI862zOxOM1trZsvbud8s5AMNUeyHFDrGTLKI+0Aze87MPk5vm/PK3XXJ8QWoAF4DPg30Bl4ERrc55gLgR9HtM4H7o9vVwNHAecAtbR7zFFCbkDgnAp+Kbo8F1qQ9ZiFwOGDAI8CXExpnTl/PHMS6K9Arur0vsBbolc05kxBntL0K2CsJr2fa/fOBnwKXZXvOJMSZj9ezGC9ZvrYzadNeJuGSZeyjgD8Ce0Tbg4oh7jbHXwzcGXfcXXjN5wLnR7dHA6vijrsLsf8UmBHd/gJwb9xxR7H8FXAIsLyd+08g5ANGyA9eiDvmLOMeBBwGXJveNufzoh7z/JgENLj76+6+GZgHTG1zzFTg7uj2fOBYMzN33+juvwdSCY/zj+7+ZrS/Hugb9QjuC/Rz9+c9vKvvAU5OWpw9jCdfsX7k7lui/VVA68jsbM6ZhDjzodtxApjZycAbhH/7rpwzCXFKkI9/r0LJJvavA7e6+3sA7r62wDFm0tXX/CzgfwoSWeeyid2BftHt3YE3SYZsYh8NPBnd/k2G+2Ph7k8D6zs4ZCpwjwfPA/2jfCFWncXt7mvdfRHQUqiYlJjnx2Bgddp2Y7Qv4zFRkvMBMCCLc/84+unwW60f6gmI81TgD+7+cXR8YyfnTEKcrXL5evY4VjObbGb1wEvAedH92ZwzCXFC+MB73ELZ0KwextijOM1sN+DfgG9345xJiBNy/3oWo2z/vU6NfiKfb2ZDCxNap7KJ/TPAZ8zsGTN73symFCy69mX9f8TM9gNGsCNZjFs2sV8NfMXMGoGHCT3+SZBN7C8Cp0S3/xaoMbNscoe45aPdLUlKzIvL2e5+MHBMdPlqzPFgZmOA64Bz446lI+3EmbjX091fcPcxhJ/OrjCzqrhjyqSDOI9290OALwMXmtlfxRZk+PC9wd03xBhDNq6m/TiT9Hom2S+A4e4+Dvg1O36ZKAa9COUsnyP0PN9mZv1jjahrzgTmu/vWuAPpgrOAu9x9CKHE4l4zK5Z86DLgs2b2R+CzwBqgmF576USxvBGLzRogvcdmSLQv4zFm1ovwc9q6jk7q7mui62agjvCzV2xxRoNOHgKmu/tracenD0bJdM4kxJmP17PHsabFtgLYQFQXn8U5kxBn+mu6lvCax/kenQzMMbNVwD8C3zCzi7I8ZxLizMfrWYw6fW3dfV3aL2G3A4cWKLbOZPO+aAQWuHuLu78B/ImQqMepK/9HziQ5ZSyQXeznAA8AuPtzhJK8vQoSXceyea+/6e6nuPtE4N+jfYkZeNuBfLS7JUmJeX4sAkaZ2Qgz601ouBa0OWYBMCO6fRrwZFSTnZGZ9TKzvaLblcCJQMZRxIWIM+rR+RVwubs/03qwu78FfGhmh0elIdOBnyctzjy9nj2NdUSUsLX+PHwgYfBfNueMPU4zqzazmmh/NXA8Mb5H3f0Ydx/u7sOB/wL+n7vfkuU5Y48zT69nMer0tW1Tq3oSsKKA8XUkm/fFzwi95URt0meA1wsZZAZZ/R8xswOBPYDnChxfR7KJ/S/AsQBmdhAhMW8qaJSZZfNe3yutd/8K4M4Cx9hdC4DpFhwOfBDlC9KWJ2BUbCleCD+P/Ykwwvrfo33XACdFt6sIo6sbCLOYfDrtsasIgxE2EHpTRhNma1kCLCMMELsRqIgrTuCbwEZgadplUHRfLSGBeA24hWiF2STFma/Xs4exfjWKZSnwB+Dkjs6ZtDgJMwm8GF3q446zzTmuZufZThLzerYXZ75ez2K8ZPHa/kf0Gr1IGBB3YNwxdyF2A64HXiaM2Tgz7piziTvavhr4XtyxduM1Hw08E71flgLHxx1zF2I/DVgZHXM70CfumKO4/gd4izBIspHwq8R5hDFIre/zW6O/6yVyPCNaHuPeJ9r/IfB+dLtfPmOy6IlFRERERCRGKmUREREREUkAJeYiIiIiIgmgxFxEREREJAGUmIuIiIiIJIAScxERERGRBFBiLiIiIiKSAErMpSiZWX8zu6Abj3u4s+WuzewaM/ti96MTEZFSaKfNbFXrYnQihaB5zKUomdlw4JfuPrbN/l7uviWWoEREZLtSaKfNbBVhMZx3u/HYovk7JTnUYy7F6nvA/ma21MwWmdnvzGwBYfU8zOxnZrbEzOrNbFbrg1p7P8xsuJmtMLPbomMeN7O+0TF3mdlpacd/28z+YGYvRUtQY2YDzezX0WNvN7M/t9erEj3XK9F5/2Rm95nZF83sGTNbaWaTouOqzexOM1toZn80s6lpj/9dFMMfzOzIaP/nzOwpM5sfnf8+M7P8veQiIl1STO10tZn9ysxeNLPlZnZG2t0XZzj3JDN7LmqrnzWzA6L9M81sgZk9CTwRtdNPR+d+1cx+ZGa7RMceH53jD2b2UzPbLdf/AFJ8lJhLsboceM3dJwD/ChwCXOrun4nu/3t3PxSoBS4xswEZzjEKuNXdxxCW2j21ned6190PAX4IXBbtuwp4MnrsfGBYJ/GOBP4TODC6TAOOjs73jeiYf4/OOQn4PPB9M6sG1gLHRTGcAdyUdt6JwD8Slpj+NHBUJ3GIiBRKMbXTU4A33X181MP/aCfnfgU4xt0nAlcC/y/t+EOA09z9s9H2JOBiQju9P3BK9AXhm8AXo3MvBv65g/ikTPSKOwCRHFno7m+kbV9iZn8b3R5KaNzXtXnMG+6+NLq9BBjezrn/N+2YU6LbRwN/C+Duj5rZe53E94a7vwRgZvXAE+7uZvZS2vMeD5xkZq0NfxXhg+RN4BYzmwBsBT6Tdt6F7t4YnXdpdK7fdxKLiEgcktxOvwT8p5ldRyi/+V0n594duNvMRgEOVKYd/2t3X5+2vdDdXwcws/+J4koREvVnoh86ewPPdRCflAkl5lIqNrbeMLPPAV8EjnD3j8zsKUKS29bHabe3An3bOffHacd09/9M+nNtS9velnZOA05191fTH2hmVwPvAOMJv3Kl2jlvT+ITEcm3xLbT7v4nMzsEOAH4rpk94e7XdHDu7wC/cfe/tVBL/1Ta6Tays7aD+ZzQ3v/a3c/qaqxS2lTKIsWqGahp577dgfeixv5A4PA8PP8zwN9BqBME9sjBOR8j1DJadN6J0f7dgbfcfRvwVaAiB88lIpJvRdNOm9mngI/c/SfA9wnlKB3ZHVgT3Z7ZybGTzGxEVFt+BuFXzeeBo8xsZPT81Wb2mY5OIuVBibkUJXdfR/gJcDmhEU33KNDLzFYQBh89n4cQvg0cHz3/6cDbhA+hnvgO4efQZVG5y3ei/f8NzDCzFwn16W17Y0REEqfI2umDgYVRSeBVwHc7Ofcc4D/M7I903kO/CLgFWAG8ATzk7k2EhP5/zGwZoYzlwE7/Iil5mi5RpBvMrA+w1d23mNkRwA+jAU4iIpIASWino5Kdy9z9xEI+rxQv1aOKdM8w4IHop8nNwNdjjkdERHamdlqKjnrMRXIkmurriQx3HRv9pCsiIjFSOy1Jp8RcRERERCQBNPhTRERERCQBlJiLiIiIiCSAEnMRERERkQRQYi4iIiIikgBKzEVEREREEkCJuYiIiIhIAigxFxERERFJACXmIiIiIiIJoMRcRERERCQBlJiLiIiIiCSAEnMRERERkQRQYi4iIiIikgBKzEVEREREEqBX3AEUyl577eXDhw+POwwRkS5bsmTJu+4+MO44CklttogUq5602WWTmA8fPpzFixfHHYaISJeZ2Z/jjqHQ1GaLSLHqSZutUhYRERERkQRQYi4iIiIikgBKzEVEREREEqBsaswzaWlpobGxkVQqFXcoJaOqqoohQ4ZQWVkZdygiUmLUZuee2myRZCnrxLyxsZGamhqGDx+OmcUdTtFzd9atW0djYyMjRoyIOxwRKTFqs3NLbbZI8pR1KUsqlWLAgAFq4HPEzBgwYIB6s0QkL9Rm55babJHkKevEHFADn2N6PaXbUiloagrXIu1QG5Nbej1LgNrOklLWpSwikhANDVBXBy0tUFkJ06bByJFxRyUikmxqqLlDVgAAIABJREFUO0tO2feYi0jMUqnwwVJTA0OHhuu6OvX+iIh0RG1nSVJiLjvZsmVL3CFIKcnmJ9bm5tDbU10dtqurw3Zzc2FiFCliarPLWBxtp8pm8k6JeVfl+E35k5/8hEmTJjFhwgTOPfdctm7dyvnnn09tbS1jxozhqquu2n7s5ZdfzujRoxk3bhyXXXYZzc3NjBgxgpaWFgA+/PDDnbbbeu2115gyZQqHHnooxxxzDK+88goAM2fO5LzzzmPy5MnMnj2bhQsXcsQRRzBx4kSOPPJIXn31VQD+P3t3HydFfeb9/nM5DIw7jqCIkQUUXNjIgyDrKLIkOSYekTUkeKucIETAQ3xCI+fe9SaaaPRO5GU0rt4a3bgQ4yOziLBGEnVlAzEaTIAh4RkNI5IwxMgIAgMy8uB1/qiasRl7Znp6uruqu7/v16tf3V1dXf2r6a6rrvnVVb/asGFDU1uHDh3K5s2bM/I3kAJVUwP33gsPPRTc19Qkn6+iIjgEu39/8Hz//uB5RUXu2iqFSzFbMbtQ5Tp2phrTpUNUY94eGa7l2rRpE88++yzLli2jtLSU6dOnM3fuXGbNmsWJJ57IkSNHuOCCC1i7di29evXi+eef580338TM2L17NxUVFZx//vm8+OKLXHLJJcybN49LL720xfFor7nmGh599FEGDBjA8uXLmT59OkuXLgWCYcjeeOMNSkpK2Lt3L6+//jqdOnXil7/8Jd/+9rdZuHAhjz76KDNmzGDSpEkcPHiQI0eOpL3uUuASD7GWlwc7jKoqmDkTysqOnresLNiWqqpg165Ptq3m84m0l2K2YnYhy2XsbE9Mlw5RYp6qLPwolyxZwqpVqzjnnHMAOHDgACeffDLz589n9uzZHD58mHfffZeNGzcyaNAgysrKmDZtGmPHjmXs2LEAfOMb3+Dee+/lkksu4fHHH2fOnDlJP2vfvn288cYbjB8/vmnaRx991PR4/PjxlJSUALBnzx6mTJnC5s2bMbOm3pyRI0cya9YsamtrufTSSxkwYEBa6y1FINkh1l27gunJtpf+/YNtqb4+2MYU6KWjFLMVs4tBrmJne2O6pE2lLKnKQi2XuzNlyhRWr17N6tWreeutt5gyZQr33XcfS5YsYe3atXz5y1+moaGBTp06sWLFCi6//HJ+8YtfMGbMGABGjRrF1q1befXVVzly5AhDhgxJ+lkff/wx3bp1a/qs1atXs2nTpqbXyxvXC7j99tv54he/yPr16/n5z3/eNMbtxIkTWbRoEcceeywXX3xxU8+NyKekcoi1eYlBWRn06KEgL5mhmK2YHWeZLLHKRexUyWHOKDFPVRZ+lBdccAELFixgx44dAOzatYs///nPlJeX07VrV9577z1efvllIOg92bNnDxdffDEPPPAAa9asaVrO5MmTmThxIldddVWLn3X88cfTr18/nnvuOSDYwSQuI9GePXvo1asXAE888UTT9C1btnD66adz0003MW7cONauXZv2ukuBazzEWl8P27YF94mHWFWrKNmmmK2YHVf5GP/aiumSMUrMU5WFH+WgQYO46667GD16NEOHDuXCCy+kS5cuDB8+nDPOOIOJEycyatQoAOrr6xk7dixDhw7lc5/7HPfff3/TciZNmsQHH3zAFVdc0ernzZ07l8cee4xhw4YxePBgXnjhhaTzzZw5k1tvvZXhw4cfdcb//PnzGTJkCGeddRbr169n8uTJaa+7FIHGQ6w33RTcN9b2aogvyQXFbMXsOMrn+NdSTJeMMnePug05UVlZ6dXV1UdN27RpEwMHDmzfghoaYlcHu2DBAl544QWefvrpqJsCpPl3lfyT7rZQVxf0FPXp88m0bduCYN+jR+bbWQDMbJW7V0bdjlxSzM4dxewcysf4F8NtKO46ErN18md7lZXF6of5zW9+k5dffpmXXnop6qZIMWgM0Dt2wMKF6Y12kVhi0HhSnmoVJVsUsyVO8i3+pTKykRL3jFJinud+9KMffWraDTfcwLJly46aNmPGjFbrGUXa1BigDxyA5cvhvPOgb9/2j3ah4RGliClmF7l8in+pjGyU4SFJRYl5QXrkkUeiboIUmuYBuqQE1q+HXr3SGzZLwyOKNFHMLjL5Ev/aGiJRY5tnhRJzkWKTzmHHxAB96BAce2zQc97QAAcPpncoNmYlBiIiOZOr+NeRMpO2ym40tnlWZH1UFjMbY2ZvmVmNmd2S5PUuZvZs+PpyM+sbTr/QzFaZ2brw/ksJ73k1XObq8HZyttdDpCCkO0xXYoAuLYUhQ+DIEXj/fQ2bVWAUs0UKREeHZWxrZCONbZ4VWe0xN7MS4BHgQqAWWGlmi9x9Y8Js04AP3L2/mU0A7gG+BrwPfMXd/2JmQ4BXgF4J75vk7kefsi8iLevIYcfmdZFlZfCjH8HJJ8f7UKy0i2K2SIHIVJlJa2U3+VQvn0eyXcpyLlDj7lsAzGweMA5IDPLjgDvDxwuAh83M3P0PCfNsAI41sy7u/hEi0n4dPeyYL3WR0hGK2SKFIJNlJq2V3Wi/kHHZLmXpBWxLeF7L0T0oR83j7oeBPUD3ZvNcBvy+WYB/PDwkeruZWbIPN7NrzKzazKrr6uo6sh4i+S+Vw46Nl4nevTv55aITL/2cyUtKS1woZovEWapxNxdlJo1tgU/2C+kuQ/uRJrE/+dPMBhMcKh2dMHmSu283swpgIXAl8FTz97r7bGA2BBeryEFz22Xr1q2MHTuW9evXZ2R5xx13HPv27cvIsqQAtXXYsXHYq/feg02bYOBA+Mxnkg9/pSGypAWFHLNBcVsi1J64m+0yk0zsA7QfSSrbPebbgYTLW9E7nJZ0HjPrBHQFdobPewPPA5Pd/e3GN7j79vC+HqgiOPyaE3H55y7xssuZ4u58/PHHGV+uxEjzSyr37v1JD3lVVRC0t22D448P7svKPn256Hy+pLS0RTE7ixS3JW3pxN3eveGqq+Caa4J4n6mkNxP7AO1HWpTtxHwlMMDM+plZZ2ACsKjZPIuAKeHjy4Gl7u5m1g14EbjF3ZuuvGBmnczspPBxKTAWyEzXRRs6eoJzMkeOHOHqq69m8ODBjB49mgMHDjBnzhzOOecchg0bxmWXXcaHH34IwNSpU7nuuusYMWIEM2fO5J133mHkyJGceeaZ3HbbbW1+1g9/+EPOOecchg4dyh133AEEvT+f/exnmTx5MkOGDGHbtm1cf/31VFZWMnjw4Kb5AG655RYGDRrE0KFDufnmmzu+8hKNxnKU2tpPftB33x30lJeUBKOtHHccHD4cPD90KKgfbJSsdrH5PJKvFLNToLgtOdfeuNv44589Gx5/PIj3UbUlW8soUFlNzMP6wxsJzs7fBMx39w1m9j0z+2o422NAdzOrAf4ZaBye60agP/DdZkNsdQFeMbO1wGqC3ps52VwPyN4/d5s3b+aGG25gw4YNdOvWjYULF3LppZeycuVK1qxZw8CBA3nsscea5q+treWNN97g/vvvZ8aMGVx//fWsW7eOnj17tvo5ixcvZvPmzaxYsYLVq1ezatUqXnvttaY2TJ8+nQ0bNnDaaacxa9YsqqurWbt2Lb/+9a9Zu3YtO3fu5Pnnn2fDhg2sXbs2pR2KxExi12HzH3T37kH5SkNDkIzv2wedOgVJemlpcGt8r4bIKliK2alR3Jaca0/cbevH39HDSKm2pbXP0X6kRVmvMXf3l4CXmk37bsLjBmB8kvfdBdzVwmLPzmQbU5GtcfT79evHWWedBcDZZ5/N1q1bWb9+Pbfddhu7d+9m3759XHTRRU3zjx8/npKSEgCWLVvGwoULAbjyyiv51re+1eLnLF68mMWLFzN8+HAA9u3bx+bNmzn11FM57bTTOO+885rmnT9/PrNnz+bw4cO8++67bNy4kUGDBlFWVsa0adMYO3YsY8eOTX+lJfea1/KNGXP0D7pbt6CmfO/eIJA31pg3NMCoUUGXY2IdoIbIKliK2W1T3Jaca0/NeGs//trajtd1p9KWturHNdRii2J/8mdctHUBrHR16dKl6XFJSQkHDhxg6tSp/OxnP2PYsGE88cQTvPrqq03zlDduaKEWBjf4FHfn1ltv5dprrz1q+tatW49a5jvvvMN9993HypUrOeGEE5g6dSoNDQ106tSJFStWsGTJEhYsWMDDDz/M0qVL01hjSUtHrt6WbDzbn/88eC3xB/2ZzwS1542BtPH+oYeSj4WrIbIkxrIVs0FxWyKS6tCELf34S0szM7Z5W21JdQz1OA612JF9bYZk/cqfhaKtC2BlUn19PT179uTQoUPMnTu3xflGjRrFvHnzAFqdD+Ciiy7ipz/9adPZ/9u3b2fHjh2fmm/v3r2Ul5fTtWtX3nvvPV5++WUg6KnZs2cPF198MQ888ABr1qxJd/WkvTpaKJus9wRg7NhP/6C7dQvqzxvv6+uDE0M7d/7kvY11gIlDJ4rETC5jNihuS46kEndb+vEfOpRaXXeqpS4ttaU99eNx2o9k66SUdlKPeTvk6p+773//+4wYMYIePXowYsQI6ls4GeLBBx9k4sSJ3HPPPYwbN67VZY4ePZpNmzYxcuRIIBii65lnnmk6vNpo2LBhDB8+nDPOOIM+ffowatQoINjpjBs3joaGBtyd+++/PwNrKm3KxNXbWuo9GTYsuLX0g66pgSefhFWrYONG+NzngverDlDyRC475BS3JVaS/fgbGto+jJSJIQyzebgqWzJ1pdQMMPdYDhWbcZWVlV5dffTVoDdt2sTAgQMjalHh0t81g+rqgv/e+ySMYLdtW1By0qNH6stpb7BtaAh6DCoqggD1m9/AgQPwhS/AlCkaazbHzGyVu1dG3Y5cUszOHf1di0hr+4LEuN+YnNbXp5ec5tsY5Zna14Y6ErPVYy4SZ5nqeWhv12HiocjycvjKV2DLlmA83MTAJSIi+aO1fUEmz5iOY/14a2LUy6/EvMCsW7eOK6+88qhpXbp0Yfny5RG1SNqt+cknmTpzvaws/fKXgweha9e0eg5EpHWK25JTLe0LMp2ctmefE7Vk+9rLLvukLj6H61H0ibm7p3yGfD4488wzWb16dWSfXyylUVnT0uG/dHse0j3DXENZSUwVWsyGaOO2YrY0aS3ux2C0kqxL3Nfu2AELF0ZSilPUiXlZWRk7d+6ke/fuBRfoo+Du7Ny5k7JC3Wizra2TT3Jd45dvhyKl4ClmZ5ZitnxKsrifb/XiHdG4Lfz4x5GdCFrUiXnv3r2pra2lrq4u6qYUjLKyMnr37h11M/JTJuv7mif5u3fDnDlw663BUIiJ87WWeOfToUgpeIrZmaeYHRPJYnFUvdSJcT9Go5XkTDavTpaCok7MS0tL6devX9TNEAlksr4vMbDs2AErVsAHH8Ddd8PVVwe9HcXUCyIFQTFbClKyWAzxiM8RJ6mRiPhEUF1gSCQuMnlFlMbAsnt3kJQfcwyccAJ07x4E+927P+kF6dMnuK+qavuCEiIikjmJPdKNsfjJJ4NbHOJzYpIK+TEmeUfl+upkzRR1j7lI7CTW95WWBj0VDQ3tDwiNgWXOnKCn/IQT4NxzgzKWbdvg3XeLrxdERCRukvVIb9sGZp8MTRtlfI7DQABRlPREeI6VEnORuCkrg9rajh/G7N8/qCm/++6gp7xbt096O3r2jM2YrSIiRStZ2URjkh6X+BzlQABRllxGdI6VSllE4ibZoc10D2N26xbUlDc0HH1Irlu3SA/ViYgIycsmpkwJbnGKz2VlwXUsct1TXoQll+oxF4mbTJ9s01Jvh4ZDFBGJXkuxuNjjczGeeIoSc5H4ycYZ4S0dktNwiCIi0UsWi4s9Pkc8OkpUVMoiEkdjxgQ9A3E5jCkiIpJLiWU+77wDf/0rXHZZwe8L1WMuEieJJ7oAjB0Lw4YVfCASERH5lP79g2T8qaeCYX8XLoQuXQr6mhvqMReJi+Ynupx4IrzyStStEhERiUZDQ5CMn3IK9OtXFCeAKjEXiYtkJ7ocOhRMFxERKTZFuF9UYi4SF8V4hTUREZGWFOF+UYm5SFxEfBlgERGRWCnC/aJO/hSJE40tLiIi8oki2y8qMReJm2Ifu1ZERCRREe0XVcoiIiIiIhIDSsxFRERERGJAibmIiIiISAwoMRcRERERiQEl5iIiIiIiMaDEXEREREQkBpSYi4iIiIjEgBJzEREREZEYUGIuIiIiIhIDSsxFRERERGJAibmIiIiISAwoMRcRERERiQEl5iIiIiIiMaDEXEREREQkBpSYi4iIiIjEgBJzEREREZEYUGIuIiIiIhIDSsxFRERERGIg64m5mY0xs7fMrMbMbknyehczezZ8fbmZ9Q2nX2hmq8xsXXj/pYT3nB1OrzGzh8zMsr0eIiLFQDFbRCQ6WU3MzawEeAT4J2AQcIWZDWo22zTgA3fvDzwA3BNOfx/4irufCUwBnk54z4+Bq4EB4W1M1lZCRKRIKGaLiEQr2z3m5wI17r7F3Q8C84BxzeYZBzwZPl4AXGBm5u5/cPe/hNM3AMeGPTU9gePd/Xfu7sBTwCVZXg8RkWKgmC0iEqFsJ+a9gG0Jz2vDaUnncffDwB6ge7N5LgN+7+4fhfPXtrFMAMzsGjOrNrPqurq6tFdCRKRIKGaLiEQo9id/mtlggkOl17b3ve4+290r3b2yR48emW+ciIgcRTFbRCR92U7MtwN9Ep73DqclncfMOgFdgZ3h897A88Bkd387Yf7ebSxTRETaTzFbRCRC2U7MVwIDzKyfmXUGJgCLms2ziOBEIYDLgaXu7mbWDXgRuMXdlzXO7O7vAnvN7LzwzP7JwAtZXg8RkWKgmC0iEqGsJuZh/eGNwCvAJmC+u28ws++Z2VfD2R4DuptZDfDPQOPwXDcC/YHvmtnq8HZy+Np04CdADfA28HI210NEpBgoZouIRMuCk+QLX2VlpVdXV0fdDBGRdjOzVe5eGXU7ckkxW0TyVUdiduxP/hQRERERKQZKzEVEREREYkCJuYiIiIhIDCgxFxERERGJASXmIiIiIiIxoMRcRERERCQGlJiLiIiIiMSAEnMRERERkRhQYi4iIiIiEgNKzEVEREREYkCJuYiIiIhIDCgxFxERERGJASXmIiIiIiIxoMRcRERERCQGlJiLiIiIiMSAEnMRERERkRjolOqMZtYLOC3xPe7+WjYaJSIiHae4LSKSX1JKzM3sHuBrwEbgSDjZAQV4EZEYUtwWEck/qfaYXwJ81t0/ymZjREQkYxS3RUTyTKo15luA0mw2REREMkpxW0Qkz6TaY/4hsNrMlgBNvS/uflNWWiUiIh2luC0ikmdSTcwXhTcREckPitsiInkmpcTc3Z/MdkNERCRzFLdFRPJPqqOyDADuBgYBZY3T3f30LLVLREQ6QHFbRCT/pHry5+PAj4HDwBeBp4BnstUoERHpMMVtEZE8k2pifqy7LwHM3f/k7ncCX85es0REpIMUt0VE8kyqJ39+ZGbHAJvN7EZgO3Bc9polIiIdpLgtIpJnUu0xnwH8DXATcDbwdWBKtholIiIdprgtIpJnUh2VZSWAmX3s7ldlt0kiItJRitsiIvknpR5zMxtpZhuBN8Pnw8zs37LaMhERSZvitohI/km1lOX/ABcBOwHcfQ3whWw1SkREOkxxW0Qkz6SamOPu25pNOpLhtoiISAYpbouI5JdUR2XZZmb/CLiZlRKcVLQpe80SEZEOUtwWEckzqfaYXwfcAPQiGHLrrPC5iIjEk+K2iEieSXVUlveBSVlui4iIZIjitohI/kkpMTezfsA3gb6J73H3r2anWSIi0hGK2yIi+SfVGvOfAY8BPwc+zl5zREQkQxS3RUTyTKqJeYO7P5TVloiISCYpbouI5JlUE/MHzewOYDHwUeNEd/99VlolIiIdpbgtIpJnUk3MzwSuBL7EJ4dEPXwuIiLxo7gtIpJnUk3MxwOnu/vBbDZGREQyRnFbRCTPpDqO+XqgWzYbIiIiGaW4LSKSZ1JNzLsBb5rZK2a2qPGWyhvNbIyZvWVmNWZ2S5LXu5jZs+Hry82sbzi9u5n9ysz2mdnDzd7zarjM1eHt5BTXQ0TyREMD1NUF95KWtOJ2McVs/cZEJG5SLWW5I52Fm1kJ8AhwIVALrDSzRe6+MWG2acAH7t7fzCYA9wBfAxqA24Eh4a25Se5enU67RCTeamqgqgoOHYLSUpg4Efr3j7pVeafdcbuYYrZ+YyISRyn1mLv7r5PdGl83s9+28NZzgRp33xLWOc4DxjWbZxzwZPh4AXCBmZm773f33xAEexEpEg0NQcJUUQF9+gT3VVXq1WyvNON2UcRs/cZEJK5SLWVpS1kL03sB2xKe14bTks7j7oeBPUD3FD7z8fCQ6O1mZslmMLNrzKzazKrr6upSWKSIRK2+PujFLC8PnpeXB8/r66NtVwFKFreLImbrNyYicZWpxNwztJxUTXL3M4HPh7crk83k7rPdvdLdK3v06JHTBopIeioqgtKC/fuD5/v3B88rKqJtVwHKZdyOVczWb0xE4ipTiXlLtgN9Ep73DqclncfMOgFdgZ2tLdTdt4f39UAVweFXESkAZWVBvW99PWzbFtxPnBhMl6wripit35iIxFWqJ3+2JelhSWAlMMDM+hEE8wnAxGbzLAKmAL8FLgeWunuLPTnhjqCbu79vZqXAWOCXHWy/iMRI//4wc2aQMFVUKGHKkmRxu2hitn5jIhJHmUrMWzosedjMbgReAUqAn7r7BjP7HlDt7ouAx4CnzawG2EWwIwDAzLYCxwOdzewSYDTwJ+CVMMCXEAT4ORlaDxGJibIyJUtZ9qm4XWwxW78xEYkba6WjAzOrJ3kdogHu7sdnq2GZVllZ6dXVsRmpS0QkZWa2yt0rU5y3IOK2YraI5Kv2xOzmWu0xd3edCiMikkcUt0VE8le7SlnCq7U1Hfhz9z9nvEUiIgWooSGaeuZiittR/Y1FRDIlpcTczL4K/Cvwt8AO4DRgEzA4e00TESkMUVxlstjitq7kKSKFINXhEr8PnAf80d37ARcAv8taq0RECkSEV5ksmritK3mKSKFINTE/5O47gWPM7Bh3/xWQVlG7iEgxifAqk0UTt3UlTxEpFKnWmO82s+OA14G5ZrYD2J+9ZomIFIbEq0yWl+f0KpNFE7cj/BuLiGRUqj3mvyK4utsM4L+At4GvZKtRIiKFIsKrTBZN3NaVPEWkUKTaY94JWExwMYlngWfDQ6QiItKGiK4yWVRxW1fyFJFCkFKPubv/b3cfDNwA9AR+bWaRX1JZRCRflJVBjx65SxiLMW7n+m8sIpJpqZayNNoB/BXYCZyc+eaIiEiGKW6LiOSJlBJzM5tuZq8CS4DuwNXuPjSbDRORzGtogLo6DSNXDBS3JVsUR0SyJ9Ua8z7A/+fuq7PZGBHJHl2ApegobkvGKY6IZFeqNea3KriL5C9dgKX4KG5LpimOiGRfe2vMRSQP6QIsItJRiiMi2afEXKQIJF6ABXQBFhFpP8URkexTYi5SBHQBFhHpKMURkexL9eRPEYmhhobUL6hSqBdgac/fQApbMf8WcrXuhRpHROJCiblInkpndISyssLakWqECGlUzL+FXK97ocURkThRKYtIHorL6AhRjmccl7+BRK+Yfwv5uO4aB12kZeoxF8lDyUZH2LUrmJ6rnqyoeyjj8DeQeCjm30K+rXvUcUMk7tRjLpKHoh4dIQ69dFH/DSQ+ivm3kE/rHoe4IRJ3SsxF8lDUoyPEYTzjqP8GEh/F/FvIp3WPQ9wQiTuVsojkqShHR2jspdu9G0pK4MiRaHrpNEKENMrH30KmRlLJl3VP7N0vL493775IVJSYi+SxqEZHKCuDUaNg1qxPakW/853o2hLXRERyK59+C5mutc6HdW/s3a+qCurgG9c77u0WySUl5iKSVGu9eQ0NsGwZjBnzSY/5smVBsp7LnWwxj1st0eno7y6x1rqx57iqKuj1bu/y8m0byJfefZGoKDEXkU9pqzevsVb0lFM+mdZY31oso8JIccrE7y5TI6nk6zaQD737IlHRyZ8icpRURk6IeiQIje4gUcjU7y4T24+2AZHCpMRcRI6SysgJUY8EodEdJAqZ+t1lYvvRNiBSmFTKIiJHaW3khMR61jiMCqPRHSSXEn93nTsH5SeN09uro9uPtgGRwqQec5E8kMtLWLfUm1dbC/feCw89FNzX1ATz9uiR+3rRqHvspTg1/u62boX58+GXv4S9e4NtI93ltbX9tLTtaxsQKUzm7lG3IScqKyu9uro66maItFtUJ3gl9o5DkIwnjiJRX5/eKBLZamMhJyRmtsrdK6NuRy7FNWY3NATDhHbuDCeeCAcPZm9bSGXbL5ZtQCSfdCRmq8dcJMaiPMErsTcvrvWsUfXYS/Fq/M1/5jNBspytbSHVbV/bgEhhUWIuEmNxSYijHoVFJC5ytS3EZdsXkdxSYi4SY6WlwaHy3buD51ElxKpnlUKT7nkbudoW9M+wSHHSqCwiMdVYX7pvH6xcCQMHBofPo0qIdcU+KRQdPW8jF9uCLl8vUpyUmIvEUGJ96SmnwGmnwc6dcNNN0K1bdO3SFfsk3yVuW40nMldVtf/kzVxsC/pnWKT4qJRFJIaa15d26xaMAnHoULTtEsl3+Va7rZM7RYqLEnORVuRy/PBEqi8VyY64nLchIpKMSllEWhDV+OGg+lKRbIjbeRsiIs0pMRdJIlN1qB2h+lKRzInreRsiIolUyiKSRFzqUFVfKpIZOm9DRPKBEnORJFTjLVJYtE2LSD7IemJuZmPM7C0zqzGzW5K83sXMng1fX25mfcPp3c3sV2a2z8webvaes81sXfieh8zMsr0eUlx0QR0pVoUas7VNi0g+yGqNuZmVAI8AFwK1wEozW+TuGxNmmwZ84O79zWwCcA/wNaABuB0YEt4S/Ri4GlgOvASMAV7O5rpI8VGNtxSbQo9kDYSYAAAgAElEQVTZ2qZFJO6y3WN+LlDj7lvc/SAwDxjXbJ5xwJPh4wXABWZm7r7f3X9DEOybmFlP4Hh3/527O/AUcElW10KKlmq8pcgUfMzWNi0icZbtxLwXsC3heW04Lek87n4Y2AN0b2OZtW0sEwAzu8bMqs2suq6urp1NFxEpOorZIiIRKuiTP919trtXuntljx49om6O5ImoLiokUuwUsyUTFMMln2V7HPPtQJ+E573DacnmqTWzTkBXYGcby+zdxjJF0pJ4USGAr3wFhg7VYW8pGorZklcaGo4+ZyDKC8OJZEK2e8xXAgPMrJ+ZdQYmAIuazbMImBI+vhxYGtYhJuXu7wJ7zey88Mz+ycALmW+6RCWq3o7EC5B06RJcGfB//S+YNSsI9iJFQDFb8kZNDdx7Lzz0UHC/YcMnMbxPn+C+qko95x2lIxC5ldUec3c/bGY3Aq8AJcBP3X2DmX0PqHb3RcBjwNNmVgPsItgRAGBmW4Hjgc5mdgkwOhwdYDrwBHAswZn9GpGlQETZ29F4AZLOneFXv4Ljj4djjgme5/qqnyJRUMyWfJHs6sxPPRXE7FNOCeYpL4ddu4LYrtidHh2ByL1sl7Lg7i8RDI+VOO27CY8bgPEtvLdvC9Or+fRwXJLnkgXaXCbEjRcg2bULjhwBd+jUCU48Ef76VwV3KQ6K2ZIPkl2d+Zhj4OOPg31H4z5EF5FKX9T75GJV0Cd/Sn5JFmgPHQqm50LjBUgOHoR9+4LPPffc4LmCu4hIfCS7kuuxx8LkybqIVKZEvU8uVlnvMRdJVWKgjaq3o39/+M53YOxY+MUv4KOPgh4YBXcRkfho7EipqgqOciaWWfzd3+kiUpkQh31yMVJiLrHRUqDNdWAtK4MRI2DYMAV3EZG4aulKrmVlitmZEJd9crFRYi6xEqdLZiu4i4jEm+J0dsVpn1wslJhL7CjQioiIxIP2ybmlkz8lZRrLVESkfRQ3RaQ91GMuKdFYpiIi7aO4KSLtpR5zaVPiWKa6mpqISNsUN0UkHUrMpU0ay1REpH0UN0UkHUrMpU3JLuSgsUxFRFqmuCki6VBiLm1qHMtUV1MTEUmN4qaIpEMnf0pKNJapiEj7KG6KSHspMZeUaSxTEZH2UdwUkfZQKYuIiIiISAwoMRcRERERiQEl5iIiIiIiMaDEXEREREQkBpSYi4iIiIjEgBLzAtTQAHV1uvSziIhIMVIekL80XGKBqamBqqrg0s+lpcEFLfr3j7pVIiIikgvKA/KbeswLSENDsDFWVECfPsF9VZX+YxYRESkGygPynxLzAlJfH/yHXF4ePC8vD57X16e3PB0KExGRdGj/EY1M5wGSeyplKSAVFcFhq/37g41x//7geUVF+5elQ2EiIpIO7T+ik8k8QKKhHvMCUlYWBMD6eti2LbifOLH9l4PWoTAREUmH9h/RylQeINFRj3mB6d8fZs4MNsaKivQ2xmSHwnbtCqZr4xYRkZZo/xG9TOQBEh0l5gWorKxjG6IOhYmISDq0/4iHjuYBEh2Vssin6FCYiIikQ/sPkY5Rj7kkpUNhIiKSDu0/RNKnxFxapENhIiKSDu0/RNKjUhYRERERkRhQYi4iIiIiEgNKzEVEREREYkCJuYiIiIhIDCgxFxERERGJASXmIiIiIiIxoMRcRERERCQGlJiLiIiIiMSAEnMRERERkRhQYi4iIiIiEgNKzEVEREREYkCJuYiIiIhIDCgxb0FDA9TVBfciIiIi2aS8QwA6ZfsDzGwM8CBQAvzE3X/Q7PUuwFPA2cBO4GvuvjV87VZgGnAEuMndXwmnbwXqw+mH3b0yk22uqYGqKjh0CEpLYeJE6N8/k58gIhJP+RizRfKd8g5plNUeczMrAR4B/gkYBFxhZoOazTYN+MDd+wMPAPeE7x0ETAAGA2OAfwuX1+iL7n5WpgN8Q0OwcVRUQJ8+wX1Vlf6DFZHCl48xWyTfKe+QRNkuZTkXqHH3Le5+EJgHjGs2zzjgyfDxAuACM7Nw+jx3/8jd3wFqwuVlVX198B9reXnwvLw8eF5fn+1PFhGJXN7FbJF8p7xDEmU7Me8FbEt4XhtOSzqPux8G9gDd23ivA4vNbJWZXZPJBldUBIeR9u8Pnu/fHzyvqMjkp4iIxFLexWyRfKe8QxLl68mfn3P3fyA43HqDmX0h2Uxmdo2ZVZtZdV1dXUoLLisLarvq62HbtuB+4sRguoiIpCVrMVsk3ynvkETZPvlzO9An4XnvcFqyeWrNrBPQleCEohbf6+6N9zvM7HmCw6WvNf9wd58NzAaorKz0VBvdvz/MnBlsHBUV2jhEpGjkZcwWyXfKO6RRtnvMVwIDzKyfmXUmODFoUbN5FgFTwseXA0vd3cPpE8ysi5n1AwYAK8ys3MwqAMysHBgNrM90w8vKoEcPbRwiUlTyNmaL5DvlHQJZ7jF398NmdiPwCsHQWz919w1m9j2g2t0XAY8BT5tZDbCLYEdAON98YCNwGLjB3Y+Y2WeA54NzjegEVLn7f2VzPUREioFitohItCzo6Ch8lZWVXl1dHXUzRETazcxWFdswg4rZIpKvOhKz8/XkTxERERGRgqLEXEREREQkBpSYi4iIiIjEgBJzEREREZEYKJqTP82sDvhT1O1ow0nA+1E3IkMKZV0KZT2gcNalGNfjNHfvkc3GxE2exOxUFMrvNVXFtr5QfOtcbOsL7V/ntGN20STm+cDMqgtl5IVCWZdCWQ8onHXRekg+KbbvudjWF4pvnYttfSG366xSFhERERGRGFBiLiIiIiISA0rM42V21A3IoEJZl0JZDyicddF6SD4ptu+52NYXim+di219IYfrrBpzEREREZEYUI+5iIiIiEgMKDEXEREREYkBJeYRMLMxZvaWmdWY2S1JXr/OzNaZ2Woz+42ZDYqinaloa10S5rvMzNzMYjnEUgrfyVQzqwu/k9Vm9o0o2tmWVL4PM/t/zGyjmW0ws6pctzFVKXwnDyR8H380s91RtLMtKazHqWb2KzP7g5mtNbOLo2intE8K32sXM3s2fH25mfVNeG2omf023AbXmVlZLtuernTX2cxKzezJcF03mdmtuW57OlJY3y+Y2e/N7LCZXd7stSlmtjm8Tcldqzsm3XU2s7MSftNrzexruW15ejryHYevH29mtWb2cMYa5e665fAGlABvA6cDnYE1wKBm8xyf8PirwH9F3e501yWcrwJ4DfgdUBl1u9P8TqYCD0fd1gysxwDgD8AJ4fOTo253R35bCfN/E/hp1O1O8zuZDVwfPh4EbI263bpl5HudDjwaPp4APBs+7gSsBYaFz7sDJVGvU5bXeSIwL3z8N8BWoG/U65SB9e0LDAWeAi5PmH4isCW8PyF8fELU65Tldf57YED4+G+Bd4FuUa9TttY34fUHgSoymB+oxzz3zgVq3H2Lux8E5gHjEmdw970JT8uBuJ6h2+a6hL4P3AM05LJx7ZDqesRdKutxNfCIu38A4O47ctzGVLX3O7kC+I+ctKx9UlkPB44PH3cF/pLD9kl6UvlexwFPho8XABeYmQGjgbXuvgbA3Xe6+5EctbsjOrLODpSbWSfgWOAgsJd4S2VfvdXd1wIfN3vvRcB/u/uuMNb+NzAmF43uoLTX2d3/6O6bw8d/AXYAcb9acUe+Y8zsbOAzwOJMNkqJee71ArYlPK8Npx3FzG4ws7eBe4GbctS29mpzXczsH4A+7v5iLhvWTil9J8Bl4SG6BWbWJzdNa5dU1uPvgb83s2Vm9jszi+vOItXvBDM7DegHLM1Bu9orlfW4E/i6mdUCLxH0/ku8pfK9Ns3j7oeBPQS9438PuJm9Eh4in5mD9mZCR9Z5AbCfoBf1z8B97r4r2w3uoJRjUIbfG6WMtNvMziXogX47Q+3KlrTX18yOAf4VuDnTjVJiHlPu/oi7/x3wLeC2qNuTjvCHez/wL1G3JQN+TnDodShB78eTbcwfV50IylnOJ+hlnmNm3SJtUcdNABbkSa9jMlcAT7h7b+Bi4Olw25HC1An4HDApvP8fZnZBtE3KunOBIwQlDv2AfzGz06NtkmSDmfUEngaucvdP9TIXkOnAS+5em+kFK/jn3nYgsbe1dzitJfOAS7LaovS1tS4VwBDgVTPbCpwHLLL4nQDa5ncSHm7+KHz6E+DsHLWtPVL5bdUCi9z9kLu/A/yRIFGPm/ZsJxOIZxkLpLYe04D5AO7+W6AMOCknrZN0pfK9Ns0TlnB0BXYSbIOvufv77v4hwVGSf8h6izuuI+s8keBcqUNh+dwyIG77gebau6/O1Huj1KF2m9nxwIvAd9z9dxluWzZ0ZH1HAjeGuc19wGQz+0EmGqXEPPdWAgPMrJ+ZdSZIKhYlzmBmiYnSl4HNOWxfe7S6Lu6+x91Pcve+7t6X4OTPr7p7dTTNbVEq30nPhKdfBTblsH2panM9gJ8R9JZjZicRHFbfkstGpiiVdcHMziA4ueq3OW5fqlJZjz8DFwCY2UCCxLwup62U9krle10ENI7GcTmw1IOzxV4BzjSzvwmT1/8L2JijdndER9b5z8CXAMysnKCT5s2ctDp9KcWgFrwCjDazE8zsBILzCl7JUjszKe11Dud/HnjK3RdksY2ZlPb6uvskdz81zG1uJljvFkema5cozoQt9hvB4eo/EtRffSec9j2CpBWCs3w3AKuBXwGDo25zuuvSbN5XieGoLCl+J3eH38ma8Ds5I+o2p7keRlBetBFYB0yIus0d+W0R1Gf/IOq2dvA7GUTQg7gm3OZHR91m3TLyvZYBzwE1wArg9IT3fj2MJ+uBe6Nel2yvM3BcOH1DGHv+V9TrkqH1PYfgCMh+giMDGxLe+/+Gf4cagrKOyNcnm+sc/qYPhTGs8XZW1OuTze84YRlTyeCoLBYuVEREREREIqRSFhERERGRGFBiLiIiIiISA0rMRURERERiQIm5iIiIiEgMKDEXEREREYkBJeYiIiIiIjGgxFwkZGb7cvEeEZFiEtc4me12mdn5ZvaLbH6GFB4l5iIRCa/6JyIiMaG4LFFTYi4Fy8x+YGY3JDy/08xuM7MlZvZ7M1tnZuNSXFZPM3vNzFab2Xoz+3zCa7PMbI2Z/c7MPhNO+4qZLTezP5jZLxOm32lmT5vZMuBpM5tqZi+Y2atmttnM7khY7tfNbEX4mf9uZiUZ++OIiKQpw7H1fDP7dRgHt4TLnhTGvnVm9nfhfC3F1AfN7Lvh44vCOJ00twkvvf7bcLl3NWvD62a2iODKpJjZz8xslZltMLNrwmnjzez+8PEMM9sSPj49jOmY2Rgze9PMfg9cmvAZJ4bLXBvuK4aG09eZWTcL7DSzyeH0p8zswnAf8Z9m9l/hPuLe1L4lyVtRXw5VN92ydQOGA79OeL4R6AMcHz4/ieByyY1XwN3XyrL+hU8u11sCVISPHfhK+Phe4Lbw8QkJy/0G8K/h4zuBVcCx4fOpwLtAd+BYgkt0VwIDgZ8DpeF8/wZMjvpvqptuuumW4dh6PrAb6Al0AbYD/zt8bQbwf8LHLcXUvwE2AF8E3gL+rpXPWtQYR4EbGtsVtmE/0C9h3hPD+8a43B04BVgZTl8ArAR6AVOAu4EyYBswADBgPvCLcP4fAXeEj78ErA4fPwp8GRgSLm9OOH0zUB7uI7YAXcPl/wnoE/VvQLfs3XTIRgqWu//BzE42s78FegAfAH8FHjCzLwAfEwTVz4TTW7MS+KmZlQI/c/fV4fSDQGMN4SrgwvBxb+BZM+sJdAbeSVjWInc/kPD8v919J4CZ/SfwOeAwcDaw0swg2DnsaM/6i4hkQ4ZjKwTJ7rsAZvY2sDicvo4g4YYWYqq7f2hmVwOvAf/T3d9u5XNGAZeFj58G7kl4bYW7J8bpm8zsf4SP+wAD3P13ZnacmVWE06qALwCfB/4TOAN4x903h+vyDHBNuIzPNX62uy81s+5mdjzweriMPwE/Bq4xs17AB+6+P4z/S9x9T7jMjcBpBP8ASAFSKYsUuueAy4GvAc8Ckwh2JGe7+1nAewS9EK1y99cIgud24InGw43AIfegewM4Ak3/7P4IeNjdzwSubfYZ+5svPslzA55097PC22fd/c622ikikiMZia2hjxIef5zw/GNSi6lnAjuBv03hs5rH20ZNcdnMzgf+b2Ckuw8D/pDweW8AVxH0zr9OkJSPBJal8NnJvBYu4/PAq0Adwd/19YR5Ev8+ifsZKUBKzKXQPQtMIAh0zxEcDtzh7ofM7IsEPQ9tMrPTgPfcfQ7wE+Af2nhLV4IkHoLDnK25MKw/PBa4hCDALwEuN7OTw88/MWyDiEgcZCS2tkPSmBrGxX8hKK/5JzMb0coyloVthuAfidY+64OwN/4M4LyE114HbiZIqP9A0KP/Udij/SbQt7EuHrii2fsmhW0+H3jf3fe6+zaC0p8B7r4F+E3C8qUIKTGXgubuG4AKYHt4qHQuUGlm64DJBIE0FecDa8zsDwQ9RA+2Mf+dwHNmtgp4v415VwALgbXAQnevdveNwG3AYjNbC/w3QQ2miEjkMhhbU3UnzWKqBXUejwE3u/tfgGnAT8yspZ76GcANYRt7tfJZ/wV0MrNNwA+A3yW89jpBGctr7n6EoKTkNwDu3kBQuvJiePJnYvnhncDZYTz/AUd32CwH/piw/F6Ny5TiY58chReRXDOzqUClu98YdVtEREQkWuoxFxERERGJAfWYiyQwszMJztZP9JG7t1a3KCIirchlbDWz7wDjm01+zt1nZfqzRDJNibmIiIiISAyolEVEREREJAaUmIuIiIiIxIAScxERERGRGFBiLiIiIiISA0rMRURERERiQIm5iIiIiEgMKDEXEREREYkBJeYiIiIiIjGgxFxEREREJAaUmIuIiIiIxIAScxERERGRGFBiLiIiIiISA0rMRURERERioFPUDciVk046yfv27Rt1M0RE2m3VqlXvu3uPqNuRS4rZIpKvOhKziyYx79u3L9XV1VE3Q0Sk3czsT1G3IdcUs0UkX3UkZquURUREREQkBpSYi4iIiIjEgBJzEREREZEYKJoacxERyQ4zGwM8CJQAP3H3HzR7/VTgSaBbOM8t7v5SzhsqUsQOHTpEbW0tDQ0NUTelYJSVldG7d29KS0sztkwl5iIikjYzKwEeAS4EaoGVZrbI3TcmzHYbMN/df2xmg4CXgL45b6xIEautraWiooK+fftiZlE3J++5Ozt37qS2tpZ+/fplbLkqZRERkY44F6hx9y3ufhCYB4xrNo8Dx4ePuwJ/yWH7RARoaGige/fuSsozxMzo3r17xo9AKDEvJA0NUFcX3IuI5EYvYFvC89pwWqI7ga+bWS1Bb/k3ky3IzK4xs2ozq66rq2t3QxQCRVqnpDyzsvH3VClLoaipgaoqOHQISkth4kTo3z/qVomIAFwBPOHu/2pmI4GnzWyIu3+cOJO7zwZmA1RWVnp7PkAhUEQKgXrMC0FDQ7BHqqiAPn2C+6oqdRuJSC5sB/okPO8dTks0DZgP4O6/BcqAkzLVAIVAESkUSswLQX190E1UXh48Ly8PntfXR9sukXylmoj2WAkMMLN+ZtYZmAAsajbPn4ELAMxsIEFi3v5alRYoBIpkSRHEwsOHD0fdhKMoMS8EFRXBsdv9+4Pn+/cHzysqom2XSD6qqYF774WHHgrua2qiblGsufth4EbgFWATwegrG8zse2b21XC2fwGuNrM1wH8AU929XaUqrVEIFMmCLMXCZ555hnPPPZezzjqLa6+9liNHjnD99ddTWVnJ4MGDueOOO5rmveWWWxg0aBBDhw7l5ptvpr6+nn79+nHo0CEA9u7de9Tz5t5++23GjBnD2Wefzec//3nefPNNAKZOncp1113HiBEjmDlzJitWrGDkyJEMHz6cf/zHf+Stt94CYMOGDU1tHTp0KJs3b87I36A1qjEvBGVlQUFlVRXs2vVJgWVZWdQtE8kviTUR5eVBhldVBTNnantqRTgm+UvNpn034fFGYFS2Pl8hUCTDshQLN23axLPPPsuyZcsoLS1l+vTpzJ07l1mzZnHiiSdy5MgRLrjgAtauXUuvXr14/vnnefPNNzEzdu/eTUVFBeeffz4vvvgil1xyCfPmzePSSy9tcRzxa665hkcffZQBAwawfPlypk+fztKlS4Fg+Mg33niDkpIS9u7dy+uvv06nTp345S9/ybe//W0WLlzIo48+yowZM5g0aRIHDx7kyJEjaa97qpSYF4r+/YMNpr4+2JC0RxJpv2Q1Ebt2BdO1TcWaQqBIBmUpFi5ZsoRVq1ZxzjnnAHDgwAFOPvlk5s+fz+zZszl8+DDvvvsuGzduZNCgQZSVlTFt2jTGjh3L2LFjAfjGN77BvffeyyWXXMLjjz/OnDlzkn7Wvn37eOONNxg/fnzTtI8++qjp8fjx4ykpKQFgz549TJkyhc2bN2NmTT3wI0eOZNasWdTW1nLppZcyYMCAtNc9VUrMC0lZmfZGEvR0KDtJT2JNRGMvkWoi8oZCoEiGZCkWujtTpkzh7rvvbpr2zjvvcOGFF7Jy5UpOOOEEpk6dSkNDA506dWLFihUsWbKEBQsW8PDDD7N06VJGjRrF1q1befXVVzly5AhDhgxJ+lkff/wx3bp1Y/Xq1UlfL2/8pwO4/fbb+eIXv8jzzz/P1q1bOf/88wGYOHEiI0aM4MUXX+Tiiy/m3//93/nSl77Uob9BW1RjLlJIVB/dMY01EfX1sG1bcK+aCBEpNlmKhRdccAELFixgx44dAOzatYs///nPlJeX07VrV9577z1efvllIOjx3rNnDxdffDEPPPAAa9asaVrO5MmTmThxIldddVWLn3X88cfTr18/nnvuOSD4pyBxGYn27NlDr17B5ReeeOKJpulbtmzh9NNP56abbmLcuHGsXbu2Q+ufCvWYixQK1UdnhmoiRESyEgsHDRrEXXfdxejRo/n4448pLS3lkUceYfjw4Zxxxhn06dOHUaOC01Hq6+sZN24cDQ0NuDv3339/03ImTZrEbbfdxhVXXNHq582dO5frr7+eu+66i0OHDjFhwgSGDRv2qflmzpzJlClTuOuuu/jyl7/cNH3+/Pk8/fTTlJaWcsopp/Dtb3+7w3+DtlgGT4yPtcrKSq+uro66GSLZU1cX9JT3SRhSets2uOmmIKgq0cxbZrbK3SujbkcuKWaLZNamTZsYOHBg1M3IiAULFvDCCy/w9NNPR92UpH/XjsRs9ZiLZFpUNd4t1QTu2AE//rEuiSgiInnvm9/8Ji+//DIvvfRS2zPnISXmIpkU5XXBk40Zd9llsHChyltERKQg/OhHP/rUtBtuuIFly5YdNW3GjBmt1qDHlRJzkUyJQ41385pADf8nIiIF7pFHHom6CRmjUVmkKC65mxNxuS54WRn06BHc65KIIiIieSPribmZjTGzt8ysxsxuSfJ6FzN7Nnx9uZn1DadfaGarzGxdeP+lhPe8Gi5zdXg7OdvrUbA0vF7mxDEJ1vB/IiIieSOrpSxmVgI8AlwI1AIrzWxReHnmRtOAD9y9v5lNAO4Bvga8D3zF3f9iZkOAV4BeCe+b5O46Zb8j4lB6UUjiel1wDf8nIiKSF7JdY34uUOPuWwDMbB4wDkhMzMcBd4aPFwAPm5m5+x8S5tkAHGtmXdz9IyQzVH+ceXFNgnVJRBERkdjLdilLL2BbwvNaju71Pmoedz8M7AG6N5vnMuD3zZLyx8MyltvNzJJ9uJldY2bVZlZdV1fXvpYXQ911HEsvCkFijXecFMNvOi70txYRkTTEflQWMxtMUN4yOmHyJHffbmYVwELgSuCp5u9199nAbAguVpHyh0Y55F0uxbX0QjKvWH7TcaC/tYgUiExflmPr1q2MHTuW9evXd3xhwHHHHce+ffsysqy4yHZivh1IuAwhvcNpyeapNbNOQFdgJ4CZ9QaeBya7+9uNb3D37eF9vZlVEZTMfCoxT0ux1V23VnoR1YVyJLOK7TcdJf2tRaRAxK2P4fDhw3TqlNm01d1xd445Jj6DFGa7JSuBAWbWz8w6AxOARc3mWQRMCR9fDix1dzezbsCLwC3u3jRqvJl1MrOTwselwFggM/96QXyGvMulZKUXGq2lcBTjbzoq+luLSAFI7GPo0ye4r6rKTHXekSNHuPrqqxk8eDCjR4/mwIEDzJkzh3POOYdhw4Zx2WWX8eGHHwIwdepUrrvuOkaMGMHMmTN55513GDlyJGeeeSa33XZbm5/1wx/+kHPOOYehQ4dyxx13AEGv/Wc/+1kmT57MkCFD2LZtG9dffz2VlZUMHjy4aT6AW265hUGDBjF06FBuvvnmjq98CrKamIc14zcSjKiyCZjv7hvM7Htm9tVwtseA7mZWA/wz0Dik4o1Af+C7zYZF7AK8YmZrgdUEPe5zMtbodOuuC6mmNJtbZD7K9+9W5xLkjv7WIlIAstnHsHnzZm644QY2bNhAt27dWLhwIZdeeikrV65kzZo1DBw4kMcee6xp/traWt544w3uv/9+ZsyYwfXXX8+6devo2bNnq5+zePFiNm/ezIoVK1i9ejWrVq3itddea2rD9OnT2bBhA6eddhqzZs2iurqatWvX8utf/5q1a9eyc+dOnn/+eTZs2MDatWtT+kcgE7JeY+7uLwEvNZv23YTHDcD4JO+7C7irhcWenck2HiWduuu4He/pKI3W8olC+G51LkHu6G8tIgUgsY+hsSovU30M/fr146yzzgLg7LPPZuvWraxfv57bbruN3bt3s2/fPi666KKm+cePH09JSQkAy5YtY+HChQBceeWVfOtb32rxcxYvXszixYsZPnw4APv27WPz5s2ceuqpnHbaaZx33nlN886fP5/Zs2dz+PBh3n33XTZu3MigQYMoKytj2rRpjB07lrFjx3Z85VMQ+5M/I9GeIe9aqim96aYgmcvH+uxsbpH5pJDqheM6jGMh0v49khwAACAASURBVN9aRPJcNvsYunTp0vS4pKSEAwcOMHXqVH72s58xbNgwnnjiCV599dWmecobOwlDLQzE9ynuzq233sq111571PStW7cetcx33nmH++67j5UrV3LCCScwdepUGhoa6NSpEytWrGDJkiUsWLCAhx9+mKVLl6axxu2jxLwlqY77nKx3uaYG7r4bOnfOz15W9foFCu3IgcYyz51kf2udTC0ieSSXfQz19fX07NmTQ4cOMXfuXHr1aj6ydmDUqFHMmzePr3/968ydO7fVZV500UXcfvvtTJo0ieOOO47t27dTWlr6qfn27t1LeXk5Xbt25b333uPll1/m/PPPZ9++fXz44YdcfPHFjBo1itNPPz0j69oWJeYd1bx3efdu2LQJxoyBbt3yt5dVvX46ciCZUwglUSJSdHLVn/P973+fESNG0KNHD0aMGEF9C8XsDz74IBMnTuSee+5h3LhxrS5z9OjRbNq0iZEjRwLB0IrPPPNMU1lMo2HDhjF8+HDOOOMM+vTpw6hRo4Dgn4Vx48bR0NCAu3P//fdnYE3bZu6pD++dzyorK726ujo7C0/c6R48CPv2wbBhn7y+bVtQ2tKjR3Y+X7JHCZV0VENDMLJRYklUfX27/lk3s1XuXpnllsZKVmO2SBHatGkTAwcOjLoZBSfZ37UjMVs95pmQ2LtcWhoMMahe1sKgIwfSUYVWEiUiIlmjxDxTEo/3ZKs+WzWq0VBttnSESqJERLJm3bp1XHnllUdN69KlC8uXL4+oRR2jxDwbstHLmmpJhZJ3kXjRydQiEhPunvKoJvnizDPPZPXq1ZF8djbKwZWYZ0sme1lTHbZP9dAi8aSSKBGJWFlZGTt37qR79+4Fl5xHwd3ZuXMnZRmO50rM80EqNaqFNOZ2e+T6CIGOSEi6VBIlIhHq3bs3tbW11NXVRd2UglFWVkbv3r0zukwl5vkglRrVYjzBLNdHCHREQkRE8lRpaSn9+vWLuhnShmOiboCkoLFGtb4+GHqxvv7TNaqJyTsU/glmiUcI+vQJ7quqgumF8HkiUhAaGqCuTqEizvQdSZyox7y9oiplaKtGtdhOMMv1EYJiPCIhIh2ig2zxp+9I4kaJeXtEvQW3VaNaTCeY5XoIOg15JyLtUKyn/eQTfUcSRyplSVW+lDKUlQVXGC30qJJKeU8+f55IHjGzMWb2lpnVmNktSV5/wMxWh7c/mtnuKNqZS8kOsh06FEyXeNB3JHGkHvNUqZQhfnJ9hKCYjkiIpMjMSoBHgAuBWmClmS1y942N87j7/0yY/5vA8Jw3NMd0kC3+9B1JHKnHPFXFdnJlvsj1EYJiOSIhkrpzgRp33+LuB4F5wLhW5r8C+I+ctCxCOsgWf/qOJI7UY96a5id6FtPJlSIiqekFbEt4XguMSDajmZ0G9AOWtvD6NcA1AKeeempmWxkBHWSLP31HEjdKzFvS0ome2oJFRNI1AVjg7keSvejus4HZAJWVlZm/1nUrsjXglq4rFX/6jiROlJgn09ap2tqCRUQabQf6JDzvHU5LZgJwQ9Zb1E5RD7glItJINebJ6FRtEZFUrQQGmFk/M+tMkHwvaj6TmZ0BnAD8Nsfta1W+DLglIsVBiXky/397dx8fVXnmf/xzGaKxMWBFunUFha60CgKi8YFau+3aUqpVXB9WxFawVutTdX9dl9WuVevDb5X2hSuVrT98rEqWWlwtbbHaqmgXqxK3iAa0RKAl1K0pKETKaMDr98c5wSFOkkkyZ849M9/36zWvmXPmPFxzkty5zj33gzp6iojkxd23ARcDjwIrgQfcvcnMrjWzE7M2nQLMd/eiNlHpSar1MJpyUkQ6UVOWXNTRU0Qkb+6+CFjUad1VnZavKWZM+UptyDy1nxGRHJSYd0UdPUVEyl4q9TCaclJEuqDEvDvq6CkiUvaKXg+jCeskW1JDAklJUmIuIiIVr6j1MJpyUjqoSZN0os6fIiIixaQpJwU0JJDkpBpzERGRYlM/JlGTJslBibmIiEga1I+psqlJk+SgpiwiIiIixaYmTZKDasxFRERE0qAmTYVRRiPbKDEXERERSYuaNPVPmY1so6YsIiIiIlJ6ynBkGyXmIiIiIlJ6co1s094erS9RSsxFREREpPRkj2wDZTGyjRJzERERESk9ZTiyjTp/iohI2Qp9sIbQ4xMJXpmNbKPEXEREylLogzWEHp+kSHdsvVOIkW0CueZKzEVEpOxkD9bQMaliQ0NUsRZCnhN6fJIi3bEVX0DXXG3MRUSk7IQ+WEN/4stkoLW1pEeEC0tIF7QMh/8LXmDXXDXmIiJSdrIHa+iokQ5psIa+xhdQxV55CO2C5rpj27gxWq+vUpIR2DVXjbmIiJSd0Adr6Et8gVXshSvfGvAQL2gZDv8XvMCuuWrMRUSkLIU+WENv4wusYi9MvakBD/GCdtyxNTREsXR8Bv2AkxPYNU88MTezScAtQBVwh7vf2On93YB7gcOADcDp7r7WzD4P3AjsCrwL/LO7PxHvcxhwD7A7sAi41N096c8iIiKlpavBGgIZgKFXg0mE3jwndb3tURvqBQ39jrIc5brmKRUSiSbmZlYFzAE+D7QAS81sobuvyNrsHOBNdz/AzKYANwGnA38GTnD3P5rZwcCjwL7xPj8AzgWeI0rMJwGPJPlZRESkPITWrDhfgVXshae3NeAhX9COGDp6A4cQU7nLvktOsZBIusb8CKDZ3VcDmNl8YDKQnZhPBq6JXy8AbjUzc/ffZm3TBOwe167vBQx092fjY94LnIQScxGRypZHDVepD1OoytRu9KUGPNQLWqp3j+Ug5UIi6c6f+wLrspZbeL/W+wPbuPs2YBMwuNM2pwD/4+7vxNu39HBMAMzsPDNrNLPG1tbWPn8IEREJXHMzzJwJs2dHz83NOTcLfRjFfNTUwJAh4eSQwehrj9/QLmiInVIrScqFRPCdP81sNFHzlom93dfd5wJzAerr69UGXUSkHPWihivUZsVSIKHWgPdGiJ1SK0nKhUTSNebrgWFZy0PjdTm3MbMBwCCiTqCY2VDgIeAsd38ta/uhPRxTREQqRS9quEIfRlEKILQa8N4KbPi+ipNyIZF0jflSYKSZjSBKnqcAUzttsxCYBvwGOBV4wt3dzPYEfg5c7u5LOjZ299fNbLOZHUXU+fMs4PsJfw4REQlVL2u4yqFSNadQhpoJSSlek5A7pVaKFAuJRBNzd99mZhcTjahSBdzl7k1mdi3Q6O4LgTuB+8ysGdhIlLwDXAwcAFxlZlfF6ya6+xvAhbw/XOIjqOOniEjl6kMi05thCkuCOgt+UClfk7K9eywhKRUSVinDf9fX13tjY2PaYYiI9JqZveDu9WnHUUx9KrNLsXa0EDKZqMNrdhv7trbSGWomCbomkqL+lNlJtzEXEREpjlJvW5yvzlPOl8NQMwWUyUDrmrfJbHVdEyk5wY/KIiIiUnClWrueq3nG0KGlM9RMwtd9x+XZOojq5z7NVFvDAcO3hX1NRLIoMRcRkX4xs0nALUR9ie5w9xtzbPMPRJPJOfCiu3ceCKB4SrXtcXfDQpZCZ8GEr/tOl+ej1Wyxg2l4Fmb409TsbmFeE5FOlJiLiEifmVkVMAf4PNGEb0vNbKG7r8jaZiRwBXC0u79pZh9JJ1pSn9UvZzz51iB3N7516J0Fi3DdP3B5hg9ho3+KtrPGUDNij/CuiUgOSsxFRKQ/jgCa3X01gJnNByYDK7K2OReY4+5vAsSja6UjpMlbeluD3NOwkCEPNVOE657z8uxeTd2IvSHQyyLSmTp/iohIf+wLrMtabonXZfs48HEzW2Jmz8ZNXz7AzM4zs0Yza2xtbU0m2lAmb+nLtOulPDtSEa57KV8ekQ6qMRcRkaQNAEYCnyGarflpMxvj7m9lb+Tuc4G5EA2XmEgkoUze0tca5NCbrHSlSNe9VC+PSAcl5iIi0h/rgWFZy0PjddlagOfcvR1YY2a/I0rUlxYnxE5CyN56OVvpTkJustKdIl33Ur08IqCmLCIi0j9LgZFmNsLMdiWavXlhp20eJqotx8z2JmrasrqYQX5A2mOeV2q7i7Svu0jgVGMuIiJ95u7bzOxi4FGi4RLvcvcmM7sWaHT3hfF7E81sBbAd+Gd335Be1IEIoeZeglWqQ+1L/ygxFxGRfnH3RcCiTuuuynrtwDfjh2RTuwvJoVSH2pf+U1MWERERkUD0ZcAeKR9KzEVEREQCkWvAnvb2aL2UPyXmIiKCRb5sZlfFy/uZ2RFpxyVSaUIZal/SocRcREQA/gOYAJwRL7cBc9ILp8gyGWhtVXsBKbrOv3qVOmCPRNT5U0REAI5090PN7LcA7v5mPPxhyejzKBbqaScp6epXTwP2VK68E3Mz2xfYP3sfd386iaBERKTo2s2sCnAAMxsCvJduSPnrMrfuKVvP7mnXMdFPQ0OUFSkbkgT19KunAXsqU16JuZndBJwOdIxBC1HhrcRcRKQ8zAYeAv7KzG4ATgWuTDek/HSZ4JzyGjUPzuu+JjxXT7uNG6P1yookQfrVk1zyrTE/CfiEu7+TZDAiIpIOd59nZi8Ax8arTnL3lWnGlK8dCc6u7dCWobamho1boe3eh6j5aA814dk97Tq2K2RPO80SI11I+ldPSlO+iflqoBpQYi4iUr4+RDR7pwO7pxxL3urqoHrzBrYsbaS2KsOW7TVU7zeKusFboHafaKOuqiM7eto1NETvd9SsFyKJVtt16UaSv3pSuvJNzP8CLDOzx8lKzt39kkSiEhGRooqHSTwNeBAw4G4z+7G7X59uZD2rIcNUGmjgMDb63lTzDlOrf0zNgG35VUcm0dNObdclD+rkKZ3lm5gvjB8iIlKezgTGuXsGwMxuBJYBwSfmtLVxwMA3mHHCStoy1dTVtFPzv2/CCSfAL36RX3VkoXvaqQGx5EmdPCVbXom5u/8w6UBERCRVfwRqgI6BvHcD1qcXTi/EjXVr3t1MTV1W7fjYsdEjjepINSAWkT7Ia4IhMxtpZgvMbIWZre54JB2ciIgUzSagyczuMbO7gZeBt8xstpnNTjm27nU3I0tNDQwZUvwqSc0SIyJ9kG9TlruBq4Gbgc8CZ6NZQ0VEyslD8aPD4pTi6JsQG+uGGJOIBC3fxHx3d3/czMzdfw9cEw+rdVWCsYmISBHEEwtNdPcz046lX0JsrBtiTCISrHwT83fMbBdglZldTNTucI/kwhIRkWJx9+1mtr+Z7eru76Ydj4hIpco3Mb+UaHzbS4DriJqzTEsqKBERKbrVwBIzWwhs6Vjp7rPSC0lEpLLkOyrLUgAze8/dz042JBERScFr8WMXQEOHiIikIK/E3MwmAHcSNV/Zz8zGAV939wuTDE5ERIrD3b+TdgwiIpUu36Ys/w58gXiSIXd/0cw+nVhUIiJSVGY2BJgBjCYazxwAd/+71IISEakweQ956O7rOq3aXuBYREQkPfOAV4ARwHeAtcDSNAMSEak0+Sbm68zsk4CbWbWZXQasTDAuEREprsHufifQ7u5PuftXAdWWi4gUUb6J+fnARcC+REMlHhIvi4hIeWiPn183s+PNbDywV5oBiYhUmnxHZfkzUNoTT4iISHeuN7NBwD8B3wcGAv8n3ZBERCpLvqOyjAC+AQzP3sfdT0wmLBERKSZ3/1n8chPRXBUiIlJk+Y7K8jDRcIk/Bd5LLhwREUlDPCrLuXywAuaracUkIlJp8k3MM+4+O9FIREQkTT8Bfg38Co26JSKSinwT81vM7GrgMeCdjpXu/j+JRCUiIsX2IXf/l7SDEBGpZPkm5mOArxANndXRlMXRUFoiIuXiZ2Z2nLsvSjsQEZFKlW9ifhrwMXd/N8lgRESkuMysjaiixYBvmdk7REMnGuDuPjDN+EREKkm+45i/DOyZZCAiIlJ87l7n7gPj513cffes5byScjObZGavmlmzmV2e4/3pZtZqZsvix9cK/0lEykcmA62t0bNUlnxrzPcEXjGzpezcxrzH4RLNbBJwC1AF3OHuN3Z6fzfgXuAwYANwuruvNbPBwALgcOAed784a5/FwD7A1njVRHd/I8/PUhSZDLS1QV0d1NSkHY2ISPfM7GhgmbtvMbMvA4cC/+7uf+hhvypgDvB5oAVYamYL3X1Fp01/lF2Oi/5PSG7NzdDQAO3tUF0NU6fCAQekHZUUS76J+dV9OXieBfY5wJvufoCZTQFuAk4HMsC3gYPjR2dnuntjX+JKmv6oRKQE/QAYZ2bjiCYZugO4D/jbHvY7Amh299UAZjYfmAx0TswlS6X+n9DNSPcymej3oq4Oamthy5ZoecYMXa9KkVdTFnd/Ktej430z+00Xu+4osOP26R0FdrbJwA/j1wuAY83M3H2Lu/83UYJeMrL/qIYNi54bGvR1lIgEb5u7O1GZfKu7zwHq8thvX2Bd1nJLvK6zU8xsuZktMLNh/Q+3dFXq/4nmZpg5E2bPjp6bm9OOKDxtbdHNWm1ttFxbGy23taUblxRPvm3Me9LVfVw+BfaObdx9G9Gsc4PzOOfdcVvFb5uZ5drAzM4zs0Yza2xtbc3jkP2nPyoRKVFtZnYF8GXg52a2C1BdoGP/FBju7mOBX/J+ZcxO0iiz01CJ/ycq9Wakt+rqom9QtmyJlrdsiZbr8rlFlrJQqMTcC3ScfJ3p7mOAY+LHV3Jt5O5z3b3e3euHDBlSlMD0RyUiJep0oj5E57j7/wJDge/msd96ILsGfGi8bgd33+DuHf2T7iDqU/QBaZTZaajE/xOVeDPSFzU1UbOmtjZYty56njpVzVgqSb5tzPuqxwI7a5sWMxsADCLqBNold18fP7eZWQNRk5l7CxV0f3T8UTU0wMaN77cd1B+ViIQsTsZnZS3/gaxy1cx+4+4Tcuy6FBhpZiOIyvMpwNTsDcxsH3d/PV48EVhZ4PBLSiX+n8i+GeloO13uNyN9dcABUZtytcWvTIVKzHM2JSGPAhtYCEwDfgOcCjwRt3PMfaIoed/T3f9sZtXAl4imkA6G/qhEpAzlLMncfZuZXQw8SjT61l3u3mRm1wKN7r4QuMTMTgS2ARuB6UWKOViV9n+iEm9G+qOmRtemUhUqMe+qKUk+BfadwH1m1kxUYE/p2N/M1gIDgV3N7CRgIvB74NE4Ka8iSspvL9DnKBj9UYlImemywiSeLXRRp3VXZb2+ArgiudBKU6X9n6i0mxGRvug2Mc+aEe4Db5E1I5y7v9zVMfIosDNEM4vm2nd4F4fN2T5RREREwlVpNyMivdVtYu7uFd36S+Otiojs0FWTRRERKZBeNWUxs4+Q1c6wpxnhSlmlTv4gItKFnE0WRUSkcPIaLtHMTjSzVcAa4ClgLfBIgnGlSuOtikilMLM2M9uc49FmZps7tuuuyaKIiBRGvuOYXwccBfzO3UcAxwLPJhZVyjTeqohUCnevc/eBOR51Hf2IRESkOPJNzNvdfQOwi5nt4u5PAvUJxpWqSpz8QUQEoiaLZrZfxyPteEREKkm+iflbZrYH8GtgnpndAmxJLqx0aeYtEak0ldZkUUQkRPl2/nySaEbOS4Evx6+vTSqoEGi8VRGpMB1NFn/l7uPN7LNE5b2IiBRJvjXmA4DHgMVAHfCjuGlLWaupgSFDlJSLSEWoqCaLIiIhyisxd/fvuPto4CJgH+ApM/tVopGJiEgxVVSTRRGREOVbY97hDeB/gQ3ARwofjoiIpCS7yeIvgNeAE1KNSESkwuQ7jvmFZrYYeBwYDJzr7mOTDEwkFJkMtLZqHHspexXZZFFEJCT5dv4cBvyjuy9LMhiR0GgGWKkU7v4d4DtmNhY4najJYou7fy7l0EREKka+bcyvUFIulUYzwEqFUpNFEZGU9LaNuUjF0AywUknUZFFEJH35NmURqTjZM8DW1moGWCl7arIoIpIy1ZiLdEEzwEolUZNFEZH0qcZcpBt9mQE2kwlzxthQ4xIREZGIEnORHtTU5J/IhjqKS6hxiUjh6OZbpPSpKUuZ0Zjb6Ql1FJdQ4xKRwmluhpkzYfbs6Lm5Oe2IRKQvVGNeRlQrmq5co7hs3BitT7P2KtS4RKQwsm++OzqqNzREzfD0Ny5SWlRjXiZUK5q+7FFcIJxRXEKNS0QKQ0O7ipQPJeZlQgVz+kIdxSXUuESkMHTzLVI+1JSll0LtXKMxt8PQl1FciiHUuESk/zpuvhsaomZqHU0Z9XcuUnqUmPdCyG24VTCHozejuBRTqHGJSP/p5lukPCgxz1NPnWuKUZPe0zlUMIuIVC7dfIuUPiXmeepuZIuWluRr0vOtrVfBLCIiIlKa1PkzT111rqmuTn40FI24IiIiIlL+lJjnqauRLdrbkx8NRSOuiIiIiJQ/NWXphVxtuDOZ5EdDSXrElVBHmhERERGpJErMe6lzG+5ijIaS5DlCHmlG+kY3WiIiIqVJiXkBFGM0lCTOoWmcy49utCQNZjYJuAWoAu5w9xu72O4UYAFwuLs3FjFEEZGSoDbmBVJTA0OGJJvQFvocarteXtRJWNJgZlXAHOCLwCjgDDMblWO7OuBS4LniRigiUjqUmFcwTeNcXnSjJSk5Amh299Xu/i4wH5icY7vrgJsA3SqKiHRBiXkXMhlobS3v2sauRppRM5bSpBstScm+wLqs5ZZ43Q5mdigwzN1/3t2BzOw8M2s0s8bW1tbCRyoiEji1Mc+hktrparbQ8lGMjsgivWVmuwCzgOk9bevuc4G5APX19Z5sZCIi4VFi3kkldojUbKHlQzdakoL1wLCs5aHxug51wMHAYjMD+Ciw0MxOVAdQEZGdqSlLJ2qnK6WuGB2RRbIsBUaa2Qgz2xWYAizseNPdN7n73u4+3N2HA88CSspFRHJQYt6J2ulKOaqEPhOSDnffBlwMPAqsBB5w9yYzu9bMTkw3OhGR0qKmLJ2ona6Um0rqMyHpcPdFwKJO667qYtvPFCMmEZHeCGVyPiXmOaidrpSLSuwzISIi0hshVWCpKUsX1E5XyoH6TIiIiHQttMn5lJiLlDH1mRARkXLXn35UoVVgJZ6Ym9kkM3vVzJrN7PIc7+9mZj+K33/OzIbH6web2ZNm9raZ3dppn8PM7KV4n9kWj8ElIjvTJFIiIlLOmpth5kyYPTt6bm7u3f6hVWAlmpibWRUwB/giMAo4w8xGddrsHOBNdz8AuJloymaIpm3+NnBZjkP/ADgXGBk/JhU+epHy0NFn4pJLomd1/BQRkXJQiGYooVVgJd358wig2d1XA5jZfGAysCJrm8nANfHrBcCtZmbuvgX4bzPbKY0ws32Age7+bLx8L3AS8EiSH0SklGkSKRERKTe5mqFs3Bit783/vJAG/Ui6Kcu+wLqs5ZZ4Xc5t4vFwNwGDezhmSw/HBMDMzjOzRjNrbG1t7WXoIiIiIhKqQjZDCWXQj7Lu/Onuc9293t3rhwwZknY4kgBNnCMiIlKZQmuGUghJN2VZDwzLWh4ar8u1TYuZDQAGARt6OObQHo5ZkQo9OH4og+13JaRxR0VERKT4QmqGUghJJ+ZLgZFmNoIoeZ4CTO20zUJgGvAb4FTgCXf3rg7o7q+b2WYzOwp4DjgL+H4SwZeSQiepoSe9mjhHREREoLz6USXalCVuM34x8CiwEnjA3ZvM7FozOzHe7E5gsJk1A98EdgypaGZrgVnAdDNryRrR5ULgDqAZeI0K7/hZ6MHxQxtsP5fQxh0VERER6a+ka8xx90XAok7rrsp6nQFO62Lf4V2sbwQOLlyUpa1QvZKTOl4Ssjt8dNSYa+IcERERKWVl3fmzUhR6cPzQBtvPpRw7fIiIiEhlS7zGXJLXkaQ2NEQ12x1twvuapBb6eEkptw4fIiIiUtmUmJeJQieppZL0llOHDxEREalsSszLSKGTVCW9IiIiIsWjNuYiIiIiIlnSmsBQNeYiIgUW+uRcIiLStTTnclFiLiJSQKFPziUiIl1LewJDNWURESmQUpicS0REupb2BIZKzEVECiTtAl1ERPon7blclJiLiBRI2gW6iIj0T9oTGKqNuYhIgZTK5FwiItK1NOdyUWIuIlJApTI5l4iIdC2tuVyUmIuIFJgm5xIRkb5QG3MRERERkQAoMRcRERERCUBFN2Vpb2+npaWFjAYZLpiamhqGDh1KdXV12qGIiIiIlJSKTsxbWlqoq6tj+PDhmFna4ZQ8d2fDhg20tLQwYsSItMMRERERKSkV3ZQlk8kwePBgJeUFYmYMHjxY30CIiIiI9EFFJ+aAkvIC0/UUyS2TgdbW6FlERCSXim7KIiJSDM3N0aRD7e3vTzp0wAFpR1U4ZjYJuAWoAu5w9xs7vX8+cBGwHXgbOM/dVxQ9UBGRwFV8jbmISJIymSgpr6uDYcOi54aG8qk5N7MqYA7wRWAUcIaZjeq0WYO7j3H3Q4CZwKwihykiUhKUmPdWmX8fvW3btrRDEEldIf/M29qimvLa2mi5tjZabmvr/7EDcQTQ7O6r3f1dYD4wOXsDd9+ctVgLeBHjExEpGUrMe6O5GWbOhNmzo+fm5n4f8v777+eII47gkEMO4etf/zrbt2/nggsuoL6+ntGjR3P11Vfv2Pbyyy9n1KhRjB07lssuu4y2tjZGjBhBe3s7AJs3b95pubPXXnuNSZMmcdhhh3HMMcfwyiuvADB9+nTOP/98jjzySGbMmMHzzz/PhAkTGD9+PJ/85Cd59dVXAWhqatoR69ixY1m1alW/P79IaAr9Z15XFzVf2bIlWt6yJVquq+t/rIHYF1iXtdwSr9uJmV1kZq8R1ZhfkutAZnaemTWaWWNra2siwYqIhExtzPOV/X10bW3037WhAWbM6PPc2ytXruRHP/oRS5Ysobq6mgsvvJB58+Zxww03sNdee7F9+3aOPfZYli9fzr777stDDz3EK6+8gpnx1ltvUVdXx2c+8xl+/vOfc9JJJzF//nxOPvnkLscQP++887jtttsYYy2bdAAAFChJREFUOXIkzz33HBdeeCFPPPEEEA0d+cwzz1BVVcXmzZv59a9/zYABA/jVr37Ft771LR588EFuu+02Lr30Us4880zeffddtm/f3ufLKRKiBP7MqamJ2pQ3NMDGje+3Me/r8UqVu88B5pjZVOBKYFqObeYCcwHq6+tVqy4iFUeJeb5yfR+9cWO0vo//YR9//HFeeOEFDj/8cAC2bt3KRz7yER544AHmzp3Ltm3beP3111mxYgWjRo2ipqaGc845hy996Ut86UtfAuBrX/saM2fO5KSTTuLuu+/m9ttvz3mut99+m2eeeYbTTjttx7p33nlnx+vTTjuNqqoqADZt2sS0adNYtWoVZrajBn7ChAnccMMNtLS0cPLJJzNy5Mg+fW6RUCXwZw5EHT1nzIiOU1dXdkn5emBY1vLQeF1X5gM/SDQiEZESpaYs+Urg+2h3Z9q0aSxbtoxly5bx6quvMm3aNL73ve/x+OOPs3z5co4//ngymQwDBgzg+eef59RTT+VnP/sZkyZNAuDoo49m7dq1LF68mO3bt3PwwQfnPNd7773HnnvuueNcy5YtY+XKlTver+3IRIBvf/vbfPazn+Xll1/mpz/96Y5xyadOncrChQvZfffdOe6443bUtouUiySbndTUwJAhZZeUAywFRprZCDPbFZgCLMzewMyy7+KPB9QOTkQkByXm+er4PrqtDdati577+X30sccey4IFC3jjjTcA2LhxI3/4wx+ora1l0KBB/OlPf+KRRx4BohrvTZs2cdxxx3HzzTfz4osv7jjOWWedxdSpUzn77LO7PNfAgQMZMWIEP/7xj4HopiD7GNk2bdrEvvtGTUTvueeeHetXr17Nxz72MS655BImT57M8uXL+/zZRUKUwJ952XP3bcDFwKPASuABd28ys2vN7MR4s4vNrMnMlgHfJEczFhERUVOW3inw99GjRo3i+uuvZ+LEibz33ntUV1czZ84cxo8fz4EHHsiwYcM4+uijAWhra2Py5MlkMhncnVmz3h9t7Mwzz+TKK6/kjDPO6PZ88+bN44ILLuD666+nvb2dKVOmMG7cuA9sN2PGDKZNm8b111/P8ccfv2P9Aw88wH333Ud1dTUf/ehH+da3vtWvzy8SojJvdpIId18ELOq07qqs15cWPSgRkRJk7pXRv6a+vt4bGxt3Wrdy5UoOOuiglCIqnAULFvCTn/yE++67L+1QgPK5riKhMLMX3L0+7TiKKVeZLSJSCvpTZqvGvMR94xvf4JFHHmHRokU9bywiIiIiwVJiXuK+//3vf2DdRRddxJIlS3Zad+mll3bbBl1ERERE0qXEvAzNmTMn7RBEREREpJc0KouIiIiISACUmIuIiIiIBECJuYiIiIhIAJSYi4iIiIgEQIm5iIiIiEgAlJj3UiYDra3Rc3+tXbuWgw8+uP8Hiu2xxx4FO5aIiIiIFJeGS+yF5mZoaID2dqiuhqlTo+m707Bt2zYGDCjsj8/dcXd22UX3ayIiIiLFpgwsT5lMlJTX1cGwYdFzQ0P/a863b9/Oueeey+jRo5k4cSJbt27l9ttv5/DDD2fcuHGccsop/OUvfwFg+vTpnH/++Rx55JHMmDGDNWvWMGHCBMaMGcOVV17Z47m++93vcvjhhzN27FiuvvpqIKq1/8QnPsFZZ53FwQcfzLp167jggguor69n9OjRO7YDuPzyyxk1ahRjx47lsssu698HFxEREZGdKDHPU1tbVFNeWxst19ZGy21t/TvuqlWruOiii2hqamLPPffkwQcf5OSTT2bp0qW8+OKLHHTQQdx55507tm9paeGZZ55h1qxZXHrppVxwwQW89NJL7LPPPt2e57HHHmPVqlU8//zzLFu2jBdeeIGnn356RwwXXnghTU1N7L///txwww00NjayfPlynnrqKZYvX86GDRt46KGHaGpqYvny5XndCIiIiIhI/pSY56muLmq+smVLtLxlS7RcV9e/444YMYJDDjkEgMMOO4y1a9fy8ssvc8wxxzBmzBjmzZtHU1PTju1PO+00qqqqAFiyZAlnnHEGAF/5yle6Pc9jjz3GY489xvjx4zn00EN55ZVXWLVqFQD7778/Rx111I5tH3jgAQ499FDGjx9PU1MTK1asYNCgQdTU1HDOOefwX//1X3zoQx/q3wcvUYXsYyAiIslSmS2lJvE25mY2CbgFqALucPcbO72/G3AvcBiwATjd3dfG710BnANsBy5x90fj9WuBtnj9NnevT/pz1NREbcobGmDjxvfbmNfU9O+4u+22247XVVVVbN26lenTp/Pwww8zbtw47rnnHhYvXrxjm9qOKvuYmeV1Hnfniiuu4Otf//pO69euXbvTMdesWcP3vvc9li5dyoc//GGmT59OJpNhwIABPP/88zz++OMsWLCAW2+9lSeeeKIPn7h0hdTHQEREuqcyW0pRojXmZlYFzAG+CIwCzjCzUZ02Owd4090PAG4Gbor3HQVMAUYDk4D/iI/X4bPufkgxkvIOBxwAM2bAJZdEz0n9gbe1tbHPPvvQ3t7OvHnzutzu6KOPZv78+QDdbgfwhS98gbvuuou3334bgPXr1/PGG298YLvNmzdTW1vLoEGD+NOf/sQjjzwCwNtvv82mTZs47rjjuPnmm3nxxRf7+vFKUlJ9DEREpPBUZkupSrrG/Aig2d1XA5jZfGAysCJrm8nANfHrBcCtFlUDTwbmu/s7wBoza46P95uEY+5WTU3/a8l7ct1113HkkUcyZMgQjjzySNq6aMh+yy23MHXqVG666SYmT57c7TEnTpzIypUrmTBhAhANrXj//ffvaBbTYdy4cYwfP54DDzyQYcOGcfTRRwPRzcLkyZPJZDK4O7NmzSrAJy0dufoYbNwYrU/690FERHpHZbaUKnP35A5udiowyd2/Fi9/BTjS3S/O2ubleJuWePk14EiiZP1Zd78/Xn8n8Ii7LzCzNcCbgAP/z93ndnH+84DzAPbbb7/Dfv/73+/0/sqVKznooIMK+IkFyvO6ZjIwc2ZU61JbG/UxaGuLvjlRIS9JM7MXivntYAjq6+u9sbEx7TCkRKnMljT1p8wu1c6fn3L3Q4mayFxkZp/OtZG7z3X3enevHzJkSHEjlLLS0cegrQ3WrYueC9HHQERECk9ltpSqpJuyrAeGZS0Pjdfl2qbFzAYAg4g6gXa5r7t3PL9hZg8RNXF5OokPUGpeeumlD4zQsttuu/Hcc8+lFFH56Ohj0NYW1cKogBcRCZfKbClFSSfmS4GRZjaCKKmeAkzttM1CYBpR2/FTgSfc3c1sIdBgZrOAvwZGAs+bWS2wi7u3xa8nAtf2NUB3z3tkk1IwZswYli1bltr5k2waFYJi9DEQEZHCUJktpSbRxNzdt5nZxcCjRMMl3uXuTWZ2LdDo7guBO4H74s6dG4mSd+LtHiDqKLoNuMjdt5vZXwEPxcn0AKDB3X/Rl/hqamrYsGEDgwcPLqvkPC3uzoYNG6hRKSgiIiLSa4mPY+7ui4BFndZdlfU6A5zWxb43ADd0WrcaGFeI2IYOHUpLSwutra2FOJwQ3ewMHTo07TBERERESk7iiXnIqqurGTFiRNphiIiIiIiU7KgsIiIiIiJlRYm5iIiIiEgAlJiLiIiIiAQg0Zk/Q2JmrcDve9hsb+DPRQgnX6HFA+HFFFo8oJjyEVo8EF5M2fHs7+4VNUtaiZbZEF5MocUD4cUUWjwQXkyhxQPhxVSQMrtiEvN8mFljSNNehxYPhBdTaPGAYspHaPFAeDGFFk+IQrxGocUUWjwQXkyhxQPhxRRaPBBeTIWKR01ZREREREQCoMRcRERERCQASsx3NjftADoJLR4IL6bQ4gHFlI/Q4oHwYgotnhCFeI1Ciym0eCC8mEKLB8KLKbR4ILyYChKP2piLiIiIiARANeYiIiIiIgFQYi4iIiIiEoCKSMzNbJKZvWpmzWZ2eY73zzezl8xsmZn9t5mNynrvini/V83sC2nHZGbDzWxrvH6Zmd1WrJiytjvFzNzM6rPWFfw69TWeNK+RmU03s9asc38t671pZrYqfkwLIJ7tWesXFiKefGKKt/kHM1thZk1m1pC1vujXqId4UrlGZnZz1nl/Z2ZvZb1X8GsUotDKbZXZycWjMjvveFRm9xxP6ZfZ7l7WD6AKeA34GLAr8CIwqtM2A7Nenwj8In49Kt5+N2BEfJyqlGMaDrycxnWKt6sDngaeBeqTuk79jCe1awRMB27Nse9ewOr4+cPx6w+nFU/83tspXaORwG87Pj/wkZSvUc540rxGnbb/BnBXUtcoxEeeP7eildv9jCe18ijeTmW2yuz+xqQyu4hldiXUmB8BNLv7and/F5gPTM7ewN03Zy3WAh09YicD8939HXdfAzTHx0szpqT0GFPsOuAmIJO1Lonr1J94kpJvTLl8Afilu2909zeBXwKTUownKfnEdC4wJ74OuPsb8fq0rlFX8SSltz+3M4D/jF8ncY1CFFq5rTI72XiSojK7MDGpzC5imV0Jifm+wLqs5ZZ43U7M7CIzew2YCVzSm32LHBPACDP7rZk9ZWbHFCCevGIys0OBYe7+897uW+R4IKVrFDvFzJab2QIzG9bLfYsVD0CNmTWa2bNmdlI/Y+lNTB8HPm5mS+JzT+rFvsWMB9K7RgCY2f5ENZpP9HbfEhdaua0yO9l4QGV2T/GAyuye4oEyKLMrITHPi7vPcfe/Af4FuDLteKDLmF4H9nP38cA3gQYzG5h0LGa2CzAL+Kekz5WPHuJJ5RrFfgoMd/exRHfGPyzSefsSz/4eTR88Ffh3M/ubIsU0gOiryM8Q1SzcbmZ7FuncvY0nrWvUYQqwwN23F/m8JSG0cltldtdUZhckHpXZPcdT8mV2JSTm64HsO86h8bquzAc67rJ6u2/iMcVfPW6IX79A1O7p40WIqQ44GFhsZmuBo4CFFnXeSeI69TmeFK8R7r7B3d+JF+8ADst33yLHg7uvj59XA4uB8f2MJ6+YiGoMFrp7e/w1+u+ICtlUrlE38aR5jTpM4f2vRHu7bykLrdxWmZ1gPCqz84pHZXbP8ZRHme0FbiQf2oPozmo10VcLHY32R3faZmTW6xOAxvj1aHbuILOawnT+7E9MQzpiIOqIsB7Yqxgxddp+Me933Cn4depnPKldI2CfrNd/Dzwbv94LWEPU+ePD8et+xdTPeD4M7Ba/3htYRTedWQoc0yTgh1nnXgcMTvEadRVPatco3u5AYC3xRHBJ/R6F+Mjz51a0cruf8ajM7jkeldk9x6Myu+d4yqLM7lewpfIAjiO6o3oN+Nd43bXAifHrW4AmYBnwZPYFB/413u9V4ItpxwSckrX+f4ATihVTp20XExeqSV2nvsaT5jUC/i0+94vxz+3ArH2/StTJqhk4O814gE8CL8XrXwLOKeI1MqKvtFfE556S8jXKGU+a1yhevga4Mce+Bb9GIT7y+LkVtdzuazyozFaZXYB4UJldMWW2xTuJiIiIiEiKKqGNuYiIiIhI8JSYi4iIiIgEQIm5iIiIiEgAlJiLiIiIiARAibmIiIiISACUmIuIiIiIBECJuZQ8M7vGzC5L4bzTzezWPuzXq3jN7BAzO6635xERCZXKbZHclJiLhO8QoskNPsDMBhQ5FhER6ZnKbekTJeYSLDM7y8yWm9mLZnafmQ03syfidY+b2X459rnEzFbE28yP1x1hZr8xs9+a2TNm9ol4/XQze9jMfmlma83sYjP7Zrzds2a2V7zdYjO7xcyWmdnLZnZEjvMOMbMHzWxp/Di6h483Lo5plZmdGx/jXjM7KeuY88xsMtHsYqfH5z89rrm5z8yWAPd1dW4zqzWzu8zs+fgzTe7bT0JEJD8qt1VuSz8VarpSPfQo5AMYTTT97d7x8l7AT4Fp8fJXgYfj19cAl8Wv/wjsFr/eM34eCAyIX38OeDB+PZ1oitw6YAiwCTg/fu9m4B/j14uB2+PXnwZeztr/1vh1A/Cp+PV+wMpuPts1RFMG7w7sDawD/hr426zPNAhYAwzIPk/W/i8Au3d3buD/Al/uuBbx9axN+2erhx56lOdD5bbKbT36/9DXKRKqvwN+7O5/BnD3jWY2ATg5fv8+YGaO/ZYD88zsYeDheN0g4IdmNhJwoDpr+yfdvQ1oM7NNRP9EAF4CxmZt959xHE+b2UAz27PTeT8HjDKzjuWBZraHu7/dxef7ibtvBbaa2ZPAEe7+sJn9h5kNAU4h+ke0LeuY2RbG+3d5bmAicKK93y6yhvgfQBcxiYj0h8ptldvST0rMpdwcT1Q7cgLwr2Y2BriOqCD/ezMbTlST0uGdrNfvZS2/x85/H97pPJ2XdwGOcvdMnnF2dbx7gS8DU4Czu9l/S0/ntqjEP8XdX80zJhGRNKjcjqncFrUxl1A9AZxmZoMB4naDzxAVfABnAr/O3sHMdgGGufuTwL8Q1bjsET+vjzeb3sd4To/P8Slgk7tv6vT+Y8A3smI5pIfjTTazmvjzfQZYGq+/B/hHAHdfEa9rI/ratitdnftR4BtxQY+Zje8hJhGR/lC5rXJb+kmJuQTJ3ZuAG4CnzOxFYBZRIXa2mS0HvgJc2mm3KuB+M3sJ+C0w293fIvrq9N/M7Lf0/VuiTLz/bcA5Od6/BKiPOy+tAM7v4XjLgSeBZ4Hr3P2PAO7+J6KvLO/O2vZJoq88l5nZ6b0493VEX/8uN7OmeFlEJBEqt1VuS/+Ze+dvZkQkm5ktJuqk1FiEc32IqJ3koTlqd0REJA8qt6VUqcZcJBBm9jmiWpfvq3AXEQmfym0pNNWYiyTEzM7mg1/bLnH3i9KIR0REuqdyW9KmxFxEREREJABqyiIiIiIiEgAl5iIiIiIiAVBiLiIiIiISACXmIiIiIiIB+P8XH6BzX3FZ8gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 864x1080 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "compare(metrics_df, [\"easy_eras\", \"hard_eras\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## neutralized high colsample "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------\n",
      "training_mean high_colsample MAX 0.04247875653736934\n",
      "training_mean neutralized_high_colsample MAX 0.01267862412039768\n",
      "val_mean high_colsample MAX 0.027972166323851425\n",
      "val_mean neutralized_high_colsample MAX 0.02268279849914095\n",
      "training_sharpe high_colsample MAX 1.2977492125411518\n",
      "training_sharpe neutralized_high_colsample MAX 0.6276165994334757\n",
      "val_sharpe high_colsample MAX 0.9314164706073689\n",
      "val_sharpe neutralized_high_colsample MAX 1.067454592046834\n",
      "val_max_drawdown high_colsample MAX 0.08691697481093952\n",
      "val_max_drawdown neutralized_high_colsample MAX 0.06617891746570836\n",
      "----------------------\n",
      "training_mean high_colsample MEAN 0.041579793167087695\n",
      "training_mean neutralized_high_colsample MEAN 0.01112359867936002\n",
      "val_mean high_colsample MEAN 0.02596817829573487\n",
      "val_mean neutralized_high_colsample MEAN 0.019560598126851864\n",
      "training_sharpe high_colsample MEAN 1.243156176357713\n",
      "training_sharpe neutralized_high_colsample MEAN 0.5272275933951578\n",
      "val_sharpe high_colsample MEAN 0.8536104108761521\n",
      "val_sharpe neutralized_high_colsample MEAN 0.9665359995641006\n",
      "val_max_drawdown high_colsample MEAN 0.07135590480769713\n",
      "val_max_drawdown neutralized_high_colsample MEAN 0.03573593213123112\n",
      "----------------------\n",
      "training_mean high_colsample MIN 0.04057293819734283\n",
      "training_mean neutralized_high_colsample MIN 0.00978310866617043\n",
      "val_mean high_colsample MIN 0.024797230152463132\n",
      "val_mean neutralized_high_colsample MIN 0.01769923096779493\n",
      "training_sharpe high_colsample MIN 1.2022755344808622\n",
      "training_sharpe neutralized_high_colsample MIN 0.4534592858651852\n",
      "val_sharpe high_colsample MIN 0.7822440727413094\n",
      "val_sharpe neutralized_high_colsample MIN 0.8483624233876996\n",
      "val_max_drawdown high_colsample MIN 0.05637986261124239\n",
      "val_max_drawdown neutralized_high_colsample MIN 0.021780577920696986\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x1080 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "compare(metrics_df, [\"high_colsample\", \"neutralized_high_colsample\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## neutralized easy eras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------\n",
      "training_mean easy_eras MAX 0.04455453364826924\n",
      "training_mean neutralized_easy_eras MAX 0.016918657179867666\n",
      "val_mean easy_eras MAX 0.026465969189361907\n",
      "val_mean neutralized_easy_eras MAX 0.01858620149873445\n",
      "training_sharpe easy_eras MAX 1.1031385327545837\n",
      "training_sharpe neutralized_easy_eras MAX 0.8533257544885557\n",
      "val_sharpe easy_eras MAX 0.6822987665093445\n",
      "val_sharpe neutralized_easy_eras MAX 1.195949372876037\n",
      "val_max_drawdown easy_eras MAX 0.13958833800405746\n",
      "val_max_drawdown neutralized_easy_eras MAX 0.04106514399054295\n",
      "----------------------\n",
      "training_mean easy_eras MEAN 0.04363257861481375\n",
      "training_mean neutralized_easy_eras MEAN 0.015210229921274711\n",
      "val_mean easy_eras MEAN 0.024463272917917757\n",
      "val_mean neutralized_easy_eras MEAN 0.015250147273694883\n",
      "training_sharpe easy_eras MEAN 1.0610645119776416\n",
      "training_sharpe neutralized_easy_eras MEAN 0.7555310870494019\n",
      "val_sharpe easy_eras MEAN 0.6163919554437711\n",
      "val_sharpe neutralized_easy_eras MEAN 0.8819791128342706\n",
      "val_max_drawdown easy_eras MEAN 0.11932491807044661\n",
      "val_max_drawdown neutralized_easy_eras MEAN 0.02602868141214648\n",
      "----------------------\n",
      "training_mean easy_eras MIN 0.04239281296889773\n",
      "training_mean neutralized_easy_eras MIN 0.013479592552521987\n",
      "val_mean easy_eras MIN 0.022558190304765798\n",
      "val_mean neutralized_easy_eras MIN 0.0130343580437055\n",
      "training_sharpe easy_eras MIN 1.0242368318089272\n",
      "training_sharpe neutralized_easy_eras MIN 0.6751398821528647\n",
      "val_sharpe easy_eras MIN 0.5586086946395253\n",
      "val_sharpe neutralized_easy_eras MIN 0.6813194386693266\n",
      "val_max_drawdown easy_eras MIN 0.10474998349775344\n",
      "val_max_drawdown neutralized_easy_eras MIN 0.011352238158261585\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x1080 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "compare(metrics_df, [\"easy_eras\", \"neutralized_easy_eras\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv",
   "language": "python",
   "name": "venv"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}