explore.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "5b14eddf-3cbe-4594-94cf-30aa4c10ba83",
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "from operator import index\n",
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "import pickle\n",
    "import time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "7141b8cb-44e0-4395-bad5-dc1da956f1fb",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/xgboost/compat.py:36: FutureWarning: pandas.Int64Index is deprecated and will be removed from pandas in a future version. Use pandas.Index with the appropriate dtype instead.\n",
      "  from pandas import MultiIndex, Int64Index\n"
     ]
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.svm import SVR\n",
    "from sklearn.neural_network import MLPRegressor\n",
    "from sklearn.ensemble import ExtraTreesRegressor\n",
    "import xgboost as xgb\n",
    "#import tensorflow as tf\n",
    "\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.preprocessing import StandardScaler, RobustScaler\n",
    "\n",
    "from sklearn.model_selection import train_test_split, cross_validate, ShuffleSplit\n",
    "\n",
    "from sklearn.metrics import mean_absolute_percentage_error\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "from sklearn.metrics import r2_score\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('once', '.*sliced data.*', )\n",
    "warnings.filterwarnings('once', '.*Maximum iterations*', )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "4fdbfc4d-5c8d-4227-a2f2-71dbbc754008",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def prepare_dataset(path : str, workload_name : str):\n",
    "    dataset = {}\n",
    "    files = os.listdir(path)\n",
    "\n",
    "    for file in files:\n",
    "        with open(path+\"/\"+file, \"r\") as f:\n",
    "            raw_str = f.read()\n",
    "        data = json.loads(raw_str)\n",
    "\n",
    "        if \"layers\" in data.keys():\n",
    "            del data[\"layers\"]\n",
    "\n",
    "        if \"relay\" in data.keys():\n",
    "            del data[\"relay\"]\n",
    "\n",
    "        if \"conv2d\" in workload_name:\n",
    "            data[\"kernel_0\"] = data[\"kernel\"][0]\n",
    "            data[\"kernel_1\"] = data[\"kernel\"][1]\n",
    "            del data[\"kernel\"]\n",
    "            data[\"dilation_0\"] = data[\"dilation\"][0]\n",
    "            data[\"dilation_1\"] = data[\"dilation\"][1]\n",
    "            del data[\"dilation\"]\n",
    "\n",
    "            if not \"kernel layout\" in data.keys():\n",
    "                data[\"kernel layout\"] = \"OIHW\"\n",
    "\n",
    "        elif workload_name == \"dense\":\n",
    "            data[\"features\"] = data[\"input shape\"][1]\n",
    "            del data[\"input shape\"]\n",
    "            del data[\"output shape\"]\n",
    "            #print()\n",
    "\n",
    "        elif workload_name in [\"max_pool2d\", \"avg_pool2d\"]:\n",
    "            data[\"pool_0\"] = data[\"pool_size\"][0]\n",
    "            data[\"pool_1\"] = data[\"pool_size\"][1]\n",
    "            del data[\"pool_size\"]\n",
    "\n",
    "        if workload_name in [\"max_pool2d\", \"avg_pool2d\", \"conv2d\", \"dilated_conv2d\", \"depthwise_conv2d\"]:\n",
    "            del data[\"padding\"]\n",
    "            data[\"C_I\"] = data[\"input shape\"][3]\n",
    "            data[\"H_I\"] = data[\"input shape\"][1]\n",
    "            data[\"W_I\"] = data[\"input shape\"][2]\n",
    "            del data[\"input shape\"]\n",
    "\n",
    "            data[\"C_O\"] = data[\"output shape\"][3]\n",
    "            data[\"H_O\"] = data[\"output shape\"][1]\n",
    "            data[\"W_O\"] = data[\"output shape\"][2]\n",
    "            del data[\"output shape\"]\n",
    "\n",
    "            key = \"strides\"\n",
    "            if \"stride\" in data.keys():\n",
    "                key = \"stride\"\n",
    "            #print(key)\n",
    "            data[\"strides_0\"] = data[key][0]\n",
    "            data[\"strides_1\"] = data[key][1]\n",
    "            del data[\"strides\"]\n",
    "            if \"stride\" in data.keys():\n",
    "                del data[\"stride\"]\n",
    "\n",
    "        dataset[file] = data\n",
    "\n",
    "    return dataset\n",
    "\n",
    "def create_dataframe(dataset : dict, workload_name : str, labels = [\"time\", \"power\", \"memory\"]):\n",
    "    df = pd.DataFrame.from_dict(dataset, orient='index')\n",
    "    categoricals = [\n",
    "        \"output dtype\",\n",
    "        \"compute dtype\",\n",
    "        \"workload\",\n",
    "        ]\n",
    "    if workload_name in [\"conv2d\", \"max_pool2d\", \"avg_pool2d\", \"depthwise_conv2d\", \"dilated_conv2d\"]:\n",
    "        if \"data_layout\" in list(df.columns):\n",
    "            categoricals += [\n",
    "                #\"padding\",\n",
    "                \"data_layout\",\n",
    "            ]\n",
    "        else:    \n",
    "            categoricals += [\n",
    "                #\"padding\",\n",
    "                \"data layout\",\n",
    "            ]\n",
    "        \n",
    "\n",
    "    if \"conv2d\" in workload_name:\n",
    "        if \"kernel_layout\" in list(df.columns):\n",
    "            categoricals += [\n",
    "                \"kernel_layout\",\n",
    "            ]\n",
    "        if \"kernel layout\" in list(df.columns):\n",
    "            categoricals += [\n",
    "                \"kernel layout\",\n",
    "            ]\n",
    "\n",
    "    for col in categoricals:\n",
    "        oh = pd.get_dummies(df[col], prefix=col, drop_first=False)\n",
    "        df = pd.concat([df, oh], axis=1).drop(col, axis=1)\n",
    "    \n",
    "    features = list(df.columns)\n",
    "    #labels = [\"time\", \"power\", \"memory\"]\n",
    "    for label in labels:\n",
    "        idx = features.index(label)\n",
    "        del features[idx]\n",
    "    del idx\n",
    "    df = df.drop_duplicates(subset=features)\n",
    "\n",
    "    label_data = []\n",
    "    for label in labels:\n",
    "        label_data.append(df[label])\n",
    "        df.pop(label)\n",
    "\n",
    "    output = pd.concat(label_data, axis=1)\n",
    "\n",
    "    return df, output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "19ed4b20-7bfd-4cc3-b15c-c77480851952",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " cuda_980ti\t\t      'llvm -mcpu=haswell_E5-2680_faulty'\n",
      " cuda_A100\t\t       opencl_Mali-T628_MP6\n",
      " cuda_K80\t\t       opencl_Mali-T628_MP6_no_power\n",
      "'llvm -mcpu=haswell_E5-2680'   rasp4b\n"
     ]
    }
   ],
   "source": [
    "!ls dataset/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "dc9224c1-1409-4fc2-8c8b-d09596a419b8",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "./dataset/cuda_980ti/avg_pool2d/\n",
      "./dataset/cuda_980ti/conv2d/\n",
      "./dataset/cuda_980ti/dense/\n",
      "./dataset/cuda_980ti/depthwise_conv2d/\n",
      "./dataset/cuda_980ti/dilated_conv2d/\n",
      "./dataset/cuda_980ti/max_pool2d/\n",
      "./dataset/cuda_A100/avg_pool2d/\n",
      "./dataset/cuda_A100/conv2d/\n",
      "./dataset/cuda_A100/dense/\n",
      "./dataset/cuda_A100/depthwise_conv2d/\n",
      "./dataset/cuda_A100/dilated_conv2d/\n",
      "./dataset/cuda_A100/max_pool2d/\n",
      "./dataset/cuda_K80/avg_pool2d/\n",
      "./dataset/cuda_K80/conv2d/\n",
      "./dataset/cuda_K80/dense/\n",
      "./dataset/cuda_K80/depthwise_conv2d/\n",
      "./dataset/cuda_K80/dilated_conv2d/\n",
      "./dataset/cuda_K80/max_pool2d/\n",
      "./dataset/llvm -mcpu=haswell_E5-2680/avg_pool2d/\n",
      "./dataset/llvm -mcpu=haswell_E5-2680/conv2d/\n",
      "./dataset/llvm -mcpu=haswell_E5-2680/dense/\n",
      "./dataset/llvm -mcpu=haswell_E5-2680/depthwise_conv2d/\n",
      "./dataset/llvm -mcpu=haswell_E5-2680/dilated_conv2d/\n",
      "./dataset/llvm -mcpu=haswell_E5-2680/max_pool2d/\n",
      "./dataset/llvm -mcpu=haswell_E5-2680_faulty/conv2d/\n",
      "./dataset/llvm -mcpu=haswell_E5-2680_faulty/dense/\n",
      "./dataset/llvm -mcpu=haswell_E5-2680_faulty/depthwise_conv2d/\n",
      "./dataset/llvm -mcpu=haswell_E5-2680_faulty/dilated_conv2d/\n",
      "./dataset/llvm -mcpu=haswell_i7-4790/avg_pool2d/\n",
      "./dataset/llvm -mcpu=haswell_i7-4790/conv2d/\n",
      "./dataset/llvm -mcpu=haswell_i7-4790/max_pool2d/\n",
      "./dataset/opencl_Mali-T628_MP6/avg_pool2d/\n",
      "./dataset/opencl_Mali-T628_MP6/conv2d/\n",
      "./dataset/opencl_Mali-T628_MP6/dense/\n",
      "./dataset/opencl_Mali-T628_MP6/depthwise_conv2d/\n",
      "./dataset/opencl_Mali-T628_MP6/dilated_conv2d/\n",
      "./dataset/opencl_Mali-T628_MP6/max_pool2d/\n",
      "./dataset/opencl_Mali-T628_MP6_no_power/conv2d/\n",
      "./dataset/opencl_Mali-T628_MP6_no_power/dense/\n",
      "./dataset/opencl_Mali-T628_MP6_no_power/depthwise_conv2d/\n",
      "./dataset/rasp4b/conv2d/\n",
      "./dataset/rasp4b/dilated_conv2d/\n",
      "found 42 folders with samples, going to train models for each of these targets\n",
      "\n"
     ]
    }
   ],
   "source": [
    "dataset_base = \"./dataset\"\n",
    "targets = os.listdir(dataset_base)\n",
    "\n",
    "layer_targets = list()\n",
    "for target in targets:\n",
    "    target_path = dataset_base + \"/\" + target\n",
    "    layers = os.listdir(target_path)\n",
    "    \n",
    "    for layer in layers:\n",
    "        dataset_path = target_path + \"/\" + layer + \"/\"\n",
    "        print(dataset_path)\n",
    "        layer_targets.append(dataset_path)\n",
    "\n",
    "print(\"found {0} folders with samples, going to train models for each of these targets\".format(len(layer_targets)))\n",
    "print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "77e12fd2-8725-4325-a3a2-42a81043adb9",
   "metadata": {},
   "outputs": [],
   "source": [
    "layer_targets = [\n",
    "#\"./dataset/opencl_Mali-T628_MP6/avg_pool2d/\",\n",
    "#\"./dataset/opencl_Mali-T628_MP6/conv2d/\",\n",
    "#\"./dataset/opencl_Mali-T628_MP6/dense/\",\n",
    "#\"./dataset/opencl_Mali-T628_MP6/depthwise_conv2d/\",\n",
    "#\"./dataset/opencl_Mali-T628_MP6/dilated_conv2d/\",\n",
    "#\"./dataset/opencl_Mali-T628_MP6/max_pool2d/\",\n",
    "    \n",
    "#\"./dataset/llvm -mcpu=haswell_E5-2680/avg_pool2d/\",\n",
    "#\"./dataset/llvm -mcpu=haswell_E5-2680/conv2d/\",\n",
    "#\"./dataset/llvm -mcpu=haswell_E5-2680/dense/\",\n",
    "##\"./dataset/llvm -mcpu=haswell_E5-2680/depthwise_conv2d/\",\n",
    "#\"./dataset/llvm -mcpu=haswell_E5-2680/dilated_conv2d/\",\n",
    "#\"./dataset/llvm -mcpu=haswell_E5-2680/max_pool2d/\",\n",
    "    \n",
    "\"./dataset/cuda_980ti/avg_pool2d/\",\n",
    "\"./dataset/cuda_980ti/conv2d/\",\n",
    "\"./dataset/cuda_980ti/dense/\",\n",
    "\"./dataset/cuda_980ti/depthwise_conv2d/\",\n",
    "\"./dataset/cuda_980ti/dilated_conv2d/\",\n",
    "\"./dataset/cuda_980ti/max_pool2d/\",\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "f45d4284-0c3f-4260-8505-30fdbee18a6c",
   "metadata": {},
   "outputs": [],
   "source": [
    "layer_targets = [   \n",
    "\"./dataset/cuda_A100/avg_pool2d/\",\n",
    "\"./dataset/cuda_A100/conv2d/\",\n",
    "\"./dataset/cuda_A100/dense/\",\n",
    "\"./dataset/cuda_A100/depthwise_conv2d/\",\n",
    "\"./dataset/cuda_A100/dilated_conv2d/\",\n",
    "\"./dataset/cuda_A100/max_pool2d/\",\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "b58be0a0-102a-4a2a-92cd-71c1465b353f",
   "metadata": {},
   "outputs": [],
   "source": [
    "layer_targets = [   \n",
    "\"./dataset/cuda_K80/avg_pool2d/\",\n",
    "\"./dataset/cuda_K80/conv2d/\",\n",
    "\"./dataset/cuda_K80/dense/\",\n",
    "\"./dataset/cuda_K80/depthwise_conv2d/\",\n",
    "\"./dataset/cuda_K80/dilated_conv2d/\",\n",
    "\"./dataset/cuda_K80/max_pool2d/\",\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "1b79c7db-56fd-4f40-a865-4bf7ee6208c4",
   "metadata": {},
   "outputs": [],
   "source": [
    "layer_targets = [   \n",
    "\"./dataset/llvm -mcpu=haswell_i7-4790/avg_pool2d/\",\n",
    "\"./dataset/llvm -mcpu=haswell_i7-4790/conv2d/\",\n",
    "\"./dataset/llvm -mcpu=haswell_i7-4790/max_pool2d/\",\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "b1dcb605-36ed-4330-839c-41c0b08995e5",
   "metadata": {},
   "outputs": [],
   "source": [
    "layer_targets = [   \n",
    "\"./dataset/llvm -mcpu=haswell_E5-2680/avg_pool2d/\",\n",
    "\"./dataset/llvm -mcpu=haswell_E5-2680/conv2d/\",\n",
    "\"./dataset/llvm -mcpu=haswell_E5-2680/dense/\",\n",
    "\"./dataset/llvm -mcpu=haswell_E5-2680/depthwise_conv2d/\",\n",
    "\"./dataset/llvm -mcpu=haswell_E5-2680/dilated_conv2d/\",\n",
    "\"./dataset/llvm -mcpu=haswell_E5-2680/max_pool2d/\",\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "a8e6d991-9609-40e6-aec7-a7dc6044b966",
   "metadata": {},
   "outputs": [],
   "source": [
    "layer_targets = [   \n",
    "\"./dataset/rasp4b/conv2d/\",\n",
    "\"./dataset/rasp4b/dilated_conv2d/\",\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "623a93ad-41c2-4077-880b-fa7d3b57cf4d",
   "metadata": {},
   "outputs": [],
   "source": [
    "models = {\n",
    "    \"xgb\": xgb.XGBRegressor,\n",
    "    \"ert\": ExtraTreesRegressor,\n",
    "    \"dTr\": DecisionTreeRegressor,\n",
    "    \"liR\": LinearRegression,\n",
    "    \"kNN\": KNeighborsRegressor,\n",
    "    \"SVR\": SVR,\n",
    "    \"MLP\": MLPRegressor,\n",
    "}\n",
    "\n",
    "kwargs_dict = {\n",
    "    \"xgb\": None,\n",
    "    \"ert\": {\"n_estimators\":175},\n",
    "    \"dTr\": {\"criterion\":\"mae\", },\n",
    "    \"liR\": {\"normalize\": True, \"n_jobs\":-1},\n",
    "    \"kNN\": {\"n_neighbors\":3, \"weights\":\"distance\"},\n",
    "    \"SVR\": {},\n",
    "    \"MLP\": {\"hidden_layer_sizes\":(100,400,200,100,), \"max_iter\":500},\n",
    "}\n",
    "\n",
    "results = {}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "bf7e2cf7-e518-4ff6-a25a-0d27961e359a",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "./dataset/rasp4b/conv2d/\n",
      "conv2d : rasp4b\t:\t contains 9031 samples\n",
      "\t[INFO] Remaining Samples after duplicate elimination:\t1909\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "time\n",
      "xgb\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'time', 'predictor': 'xgb', 'r2': 0.8542990010021295, 'max_error': 9.496242531967162, 'mae': 0.17755613381648575, 'mape': 11.13440119679909, 'medae': 0.025678360572837296, 'fit_time': 0.2931285500526428, 'score_time': 0.008359074592590332, 'dataset_min': 5.57071e-06, 'dataset_max': 40.071, 'dataset_mean': 0.62976961568967, 'dataset_median': 0.14018, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "power\n",
      "xgb\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'power', 'predictor': 'xgb', 'r2': -0.4392164024024604, 'max_error': 3.1424236745834353, 'mae': 0.4027891185057375, 'mape': 0.09288030780630585, 'medae': 0.17896602287292468, 'fit_time': 0.30745065212249756, 'score_time': 0.004167914390563965, 'dataset_min': 2.8355, 'dataset_max': 7.6719, 'dataset_mean': 4.194550656888424, 'dataset_median': 4.066, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "ws_size\n",
      "xgb\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'ws_size', 'predictor': 'xgb', 'r2': 0.8496820306095239, 'max_error': 264114948.0, 'mae': 5244277.843452573, 'mape': 6.243041999036612, 'medae': 964895.5625, 'fit_time': 0.2834956645965576, 'score_time': 0.005350708961486816, 'dataset_min': 288.0, 'dataset_max': 929710144.0, 'dataset_mean': 26467918.43268727, 'dataset_median': 8110192.0, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "io_size\n",
      "xgb\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'io_size', 'predictor': 'xgb', 'r2': 0.9107096914001092, 'max_error': 210737972.0, 'mae': 2204784.125592297, 'mape': 25.38428134540815, 'medae': 226804.80859375, 'fit_time': 0.29191917181015015, 'score_time': 0.004358828067779541, 'dataset_min': 16.0, 'dataset_max': 807469056.0, 'dataset_mean': 14319010.01990571, 'dataset_median': 1881600.0, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "time\n",
      "ert\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'time', 'predictor': 'ert', 'r2': 0.7420098458626402, 'max_error': 14.590472804285701, 'mae': 0.22152475198057955, 'mape': 0.5060807404739898, 'medae': 0.013567394428571546, 'fit_time': 0.5553600788116455, 'score_time': 0.023984193801879883, 'dataset_min': 5.57071e-06, 'dataset_max': 40.071, 'dataset_mean': 0.62976961568967, 'dataset_median': 0.14018, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "power\n",
      "ert\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'power', 'predictor': 'ert', 'r2': -0.3811147837413445, 'max_error': 3.1090282485714265, 'mae': 0.41284937661929655, 'mape': 0.0942312489669681, 'medae': 0.1879327907142841, 'fit_time': 0.5905734300613403, 'score_time': 0.024936139583587646, 'dataset_min': 2.8355, 'dataset_max': 7.6719, 'dataset_mean': 4.194550656888424, 'dataset_median': 4.066, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "ws_size\n",
      "ert\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'ws_size', 'predictor': 'ert', 'r2': 0.802327161215341, 'max_error': 298524929.3485714, 'mae': 6444796.197935676, 'mape': 0.28622649864398014, 'medae': 254064.20571428686, 'fit_time': 0.5541859865188599, 'score_time': 0.0243796706199646, 'dataset_min': 288.0, 'dataset_max': 929710144.0, 'dataset_mean': 26467918.43268727, 'dataset_median': 8110192.0, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "io_size\n",
      "ert\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'io_size', 'predictor': 'ert', 'r2': 0.8732198044156413, 'max_error': 277075610.3314285, 'mae': 2543400.3597606583, 'mape': 0.1460115657187593, 'medae': 18553.000000000007, 'fit_time': 0.47436612844467163, 'score_time': 0.021132349967956543, 'dataset_min': 16.0, 'dataset_max': 807469056.0, 'dataset_mean': 14319010.01990571, 'dataset_median': 1881600.0, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "time\n",
      "dTr\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'time', 'predictor': 'dTr', 'r2': 0.645605896517985, 'max_error': 12.820615, 'mae': 0.26207447252876964, 'mape': 0.5544645008655844, 'medae': 0.015933218749999988, 'fit_time': 0.3540647029876709, 'score_time': 0.0013018250465393066, 'dataset_min': 5.57071e-06, 'dataset_max': 40.071, 'dataset_mean': 0.62976961568967, 'dataset_median': 0.14018, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "power\n",
      "dTr\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'power', 'predictor': 'dTr', 'r2': -0.7563163418255663, 'max_error': 3.3678250000000003, 'mae': 0.4479668331151832, 'mape': 0.10215012303193857, 'medae': 0.18639100000000025, 'fit_time': 0.42327362298965454, 'score_time': 0.0014657974243164062, 'dataset_min': 2.8355, 'dataset_max': 7.6719, 'dataset_mean': 4.194550656888424, 'dataset_median': 4.066, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "ws_size\n",
      "dTr\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'ws_size', 'predictor': 'dTr', 'r2': 0.39173083134210873, 'max_error': 592084544.0, 'mae': 10053686.167539267, 'mape': 0.4026088874012881, 'medae': 542594.0, 'fit_time': 0.32326018810272217, 'score_time': 0.001389622688293457, 'dataset_min': 288.0, 'dataset_max': 929710144.0, 'dataset_mean': 26467918.43268727, 'dataset_median': 8110192.0, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "io_size\n",
      "dTr\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'io_size', 'predictor': 'dTr', 'r2': 0.7599986843203084, 'max_error': 230520000.0, 'mae': 4228680.759162304, 'mape': 0.16838791811701415, 'medae': 34.0, 'fit_time': 0.39001744985580444, 'score_time': 0.0013036727905273438, 'dataset_min': 16.0, 'dataset_max': 807469056.0, 'dataset_mean': 14319010.01990571, 'dataset_median': 1881600.0, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "time\n",
      "liR\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'time', 'predictor': 'liR', 'r2': 0.18000559636185365, 'max_error': 29.95336431063287, 'mae': 0.8428258178764703, 'mape': 528.1224630616955, 'medae': 0.4084131642142731, 'fit_time': 0.00562596321105957, 'score_time': 0.0008167028427124023, 'dataset_min': 5.57071e-06, 'dataset_max': 40.071, 'dataset_mean': 0.62976961568967, 'dataset_median': 0.14018, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "power\n",
      "liR\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'power', 'predictor': 'liR', 'r2': 0.001850251755097565, 'max_error': 3.195243770947134, 'mae': 0.38153612291868444, 'mape': 0.0850151310326595, 'medae': 0.2429661071021491, 'fit_time': 0.006504416465759277, 'score_time': 0.0009511113166809082, 'dataset_min': 2.8355, 'dataset_max': 7.6719, 'dataset_mean': 4.194550656888424, 'dataset_median': 4.066, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "ws_size\n",
      "liR\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'ws_size', 'predictor': 'liR', 'r2': 0.2879054212971805, 'max_error': 633371042.9779876, 'mae': 25485412.676334158, 'mape': 812.5183177020642, 'medae': 14011121.392113801, 'fit_time': 0.005375504493713379, 'score_time': 0.0008761286735534668, 'dataset_min': 288.0, 'dataset_max': 929710144.0, 'dataset_mean': 26467918.43268727, 'dataset_median': 8110192.0, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "io_size\n",
      "liR\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'io_size', 'predictor': 'liR', 'r2': 0.4009079524577309, 'max_error': 340731991.55082214, 'mae': 15304844.914117362, 'mape': 2561.4572262010006, 'medae': 8640092.735502753, 'fit_time': 0.005846977233886719, 'score_time': 0.0009123682975769043, 'dataset_min': 16.0, 'dataset_max': 807469056.0, 'dataset_mean': 14319010.01990571, 'dataset_median': 1881600.0, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "time\n",
      "kNN\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'time', 'predictor': 'kNN', 'r2': 0.5927634281012534, 'max_error': 20.710855047998145, 'mae': 0.3316435890979649, 'mape': 1.6340738748681591, 'medae': 0.019281727858882734, 'fit_time': 0.005319058895111084, 'score_time': 0.009160518646240234, 'dataset_min': 5.57071e-06, 'dataset_max': 40.071, 'dataset_mean': 0.62976961568967, 'dataset_median': 0.14018, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "power\n",
      "kNN\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'power', 'predictor': 'kNN', 'r2': -0.21315526897294945, 'max_error': 3.201265880291106, 'mae': 0.36749834862389386, 'mape': 0.08403182774432932, 'medae': 0.18100367274697088, 'fit_time': 0.005653321743011475, 'score_time': 0.00999826192855835, 'dataset_min': 2.8355, 'dataset_max': 7.6719, 'dataset_mean': 4.194550656888424, 'dataset_median': 4.066, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "ws_size\n",
      "kNN\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'ws_size', 'predictor': 'kNN', 'r2': 0.7356894717223704, 'max_error': 458870307.69659126, 'mae': 10728611.318216974, 'mape': 1.136616674120077, 'medae': 866059.9522951773, 'fit_time': 0.005869448184967041, 'score_time': 0.008855283260345459, 'dataset_min': 288.0, 'dataset_max': 929710144.0, 'dataset_mean': 26467918.43268727, 'dataset_median': 8110192.0, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "io_size\n",
      "kNN\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'io_size', 'predictor': 'kNN', 'r2': 0.6712104573978634, 'max_error': 312202849.41451883, 'mae': 5843106.276920128, 'mape': 1.5788725302438917, 'medae': 154081.32467958934, 'fit_time': 0.005653560161590576, 'score_time': 0.008641600608825684, 'dataset_min': 16.0, 'dataset_max': 807469056.0, 'dataset_mean': 14319010.01990571, 'dataset_median': 1881600.0, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "time\n",
      "SVR\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'time', 'predictor': 'SVR', 'r2': 0.33075821950788054, 'max_error': 18.515143040637557, 'mae': 0.3101508890725539, 'mape': 64.15149195428631, 'medae': 0.05987190808681378, 'fit_time': 0.05347055196762085, 'score_time': 0.011187374591827393, 'dataset_min': 5.57071e-06, 'dataset_max': 40.071, 'dataset_mean': 0.62976961568967, 'dataset_median': 0.14018, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "power\n",
      "SVR\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'power', 'predictor': 'SVR', 'r2': -0.06831805222922127, 'max_error': 3.0867880227464677, 'mae': 0.32776051862465605, 'mape': 0.07157489969939847, 'medae': 0.16430456014779438, 'fit_time': 0.08134907484054565, 'score_time': 0.02205127477645874, 'dataset_min': 2.8355, 'dataset_max': 7.6719, 'dataset_mean': 4.194550656888424, 'dataset_median': 4.066, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "ws_size\n",
      "SVR\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'ws_size', 'predictor': 'SVR', 'r2': -0.0844027329014021, 'max_error': 751689013.4999181, 'mae': 22742364.11032972, 'mape': 185.80541593045962, 'medae': 7294320.948067074, 'fit_time': 0.09510636329650879, 'score_time': 0.031133055686950684, 'dataset_min': 288.0, 'dataset_max': 929710144.0, 'dataset_mean': 26467918.43268727, 'dataset_median': 8110192.0, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "io_size\n",
      "SVR\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'io_size', 'predictor': 'SVR', 'r2': -0.06774407935121995, 'max_error': 655235859.9043412, 'mae': 14133505.83514532, 'mape': 765.0335679761519, 'medae': 1831824.243754217, 'fit_time': 0.09631133079528809, 'score_time': 0.03106468915939331, 'dataset_min': 16.0, 'dataset_max': 807469056.0, 'dataset_mean': 14319010.01990571, 'dataset_median': 1881600.0, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "time\n",
      "MLP\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'time', 'predictor': 'MLP', 'r2': 0.7771818332882445, 'max_error': 14.447590929881358, 'mae': 0.35349524947327216, 'mape': 78.27879959222766, 'medae': 0.11372649787592731, 'fit_time': 2.4777464270591736, 'score_time': 0.002955615520477295, 'dataset_min': 5.57071e-06, 'dataset_max': 40.071, 'dataset_mean': 0.62976961568967, 'dataset_median': 0.14018, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "power\n",
      "MLP\n",
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'power', 'predictor': 'MLP', 'r2': -0.11068957376544061, 'max_error': 3.132652579360863, 'mae': 0.3846780698315057, 'mape': 0.08710012648130122, 'medae': 0.2223992714117381, 'fit_time': 4.246971964836121, 'score_time': 0.0033628344535827637, 'dataset_min': 2.8355, 'dataset_max': 7.6719, 'dataset_mean': 4.194550656888424, 'dataset_median': 4.066, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "ws_size\n",
      "MLP\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:692: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:692: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:692: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:692: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'ws_size', 'predictor': 'MLP', 'r2': 0.41314756054936974, 'max_error': 630487995.0980134, 'mae': 21360638.963042907, 'mape': 31.629955973590068, 'medae': 7122022.559655091, 'fit_time': 23.233202695846558, 'score_time': 0.003695964813232422, 'dataset_min': 288.0, 'dataset_max': 929710144.0, 'dataset_mean': 26467918.43268727, 'dataset_median': 8110192.0, 'dataset_size': 1909}\n",
      "\n",
      "rasp4b\n",
      "conv2d\n",
      "io_size\n",
      "MLP\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:692: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:692: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:692: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:692: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'device': 'rasp4b', 'workload': 'conv2d', 'metric': 'io_size', 'predictor': 'MLP', 'r2': 0.5970118443850748, 'max_error': 237843850.14572507, 'mae': 9674037.39751517, 'mape': 766.0494349949483, 'medae': 2170156.7434019065, 'fit_time': 23.353345930576324, 'score_time': 0.003885924816131592, 'dataset_min': 16.0, 'dataset_max': 807469056.0, 'dataset_mean': 14319010.01990571, 'dataset_median': 1881600.0, 'dataset_size': 1909}\n",
      "./dataset/rasp4b/dilated_conv2d/\n",
      "dilated_conv2d : rasp4b\t:\t contains 9938 samples\n",
      "\t[INFO] Remaining Samples after duplicate elimination:\t2816\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "time\n",
      "xgb\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'time', 'predictor': 'xgb', 'r2': 0.6339297915330648, 'max_error': 15.962560344680785, 'mae': 0.2600144266600351, 'mape': 13.253003213364956, 'medae': 0.0343369836533882, 'fit_time': 0.33446168899536133, 'score_time': 0.013256311416625977, 'dataset_min': 5.57071e-06, 'dataset_max': 40.071, 'dataset_mean': 0.6204575610010015, 'dataset_median': 0.140151, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "power\n",
      "xgb\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'power', 'predictor': 'xgb', 'r2': -0.040104106820317814, 'max_error': 3.19692102098465, 'mae': 0.48351865594569676, 'mape': 0.107323123970382, 'medae': 0.2509033267879486, 'fit_time': 0.3155849575996399, 'score_time': 0.004431545734405518, 'dataset_min': 2.8355, 'dataset_max': 7.6719, 'dataset_mean': 4.322041255681818, 'dataset_median': 4.066, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "ws_size\n",
      "xgb\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'ws_size', 'predictor': 'xgb', 'r2': 0.8430614644794671, 'max_error': 399014872.0, 'mae': 5906399.617403936, 'mape': 7.771635421368178, 'medae': 999841.5087890625, 'fit_time': 0.3752276301383972, 'score_time': 0.004388988018035889, 'dataset_min': 288.0, 'dataset_max': 929710144.0, 'dataset_mean': 28140857.15340909, 'dataset_median': 8266576.0, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "io_size\n",
      "xgb\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'io_size', 'predictor': 'xgb', 'r2': 0.8273887851591097, 'max_error': 461162480.0, 'mae': 3398533.9960702667, 'mape': 69.27506119557695, 'medae': 284949.7744140625, 'fit_time': 0.32327157258987427, 'score_time': 0.00434112548828125, 'dataset_min': 16.0, 'dataset_max': 1070530560.0, 'dataset_mean': 17001619.4375, 'dataset_median': 2158272.0, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "time\n",
      "ert\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'time', 'predictor': 'ert', 'r2': 0.794560608404327, 'max_error': 14.058582199999995, 'mae': 0.21536055475498303, 'mape': 0.7681005917149244, 'medae': 0.011346162714285981, 'fit_time': 0.9103089570999146, 'score_time': 0.030435264110565186, 'dataset_min': 5.57071e-06, 'dataset_max': 40.071, 'dataset_mean': 0.6204575610010015, 'dataset_median': 0.140151, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "power\n",
      "ert\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'power', 'predictor': 'ert', 'r2': -0.23803303116946395, 'max_error': 3.1074464157142927, 'mae': 0.4930822443313067, 'mape': 0.1088485903761933, 'medae': 0.20949622571428517, 'fit_time': 0.9172424077987671, 'score_time': 0.03282850980758667, 'dataset_min': 2.8355, 'dataset_max': 7.6719, 'dataset_mean': 4.322041255681818, 'dataset_median': 4.066, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "ws_size\n",
      "ert\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'ws_size', 'predictor': 'ert', 'r2': 0.8108820871169796, 'max_error': 432547966.30857146, 'mae': 6006734.761134751, 'mape': 0.2465489164061991, 'medae': 183108.71999999945, 'fit_time': 0.8557695150375366, 'score_time': 0.029342174530029297, 'dataset_min': 288.0, 'dataset_max': 929710144.0, 'dataset_mean': 28140857.15340909, 'dataset_median': 8266576.0, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "io_size\n",
      "ert\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'io_size', 'predictor': 'ert', 'r2': 0.7918162048689372, 'max_error': 580626326.6742857, 'mae': 3208941.2614994934, 'mape': 0.11839381117275277, 'medae': 10676.868571428542, 'fit_time': 0.7453855276107788, 'score_time': 0.025822877883911133, 'dataset_min': 16.0, 'dataset_max': 1070530560.0, 'dataset_mean': 17001619.4375, 'dataset_median': 2158272.0, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "time\n",
      "dTr\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'time', 'predictor': 'dTr', 'r2': -0.38262535881111986, 'max_error': 22.795095, 'mae': 0.34253293446738475, 'mape': 0.4894799589793235, 'medae': 0.011087, 'fit_time': 1.00570148229599, 'score_time': 0.001412510871887207, 'dataset_min': 5.57071e-06, 'dataset_max': 40.071, 'dataset_mean': 0.6204575610010015, 'dataset_median': 0.140151, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "power\n",
      "dTr\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'power', 'predictor': 'dTr', 'r2': -0.5604157511271606, 'max_error': 3.4413875, 'mae': 0.5294547163120567, 'mape': 0.11590337503073883, 'medae': 0.17053125000000002, 'fit_time': 1.3374433517456055, 'score_time': 0.0014094114303588867, 'dataset_min': 2.8355, 'dataset_max': 7.6719, 'dataset_mean': 4.322041255681818, 'dataset_median': 4.066, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "ws_size\n",
      "dTr\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'ws_size', 'predictor': 'dTr', 'r2': 0.5655550067655546, 'max_error': 551793732.0, 'mae': 8445244.54609929, 'mape': 0.316813351460582, 'medae': 216064.0, 'fit_time': 0.9033032059669495, 'score_time': 0.0014327764511108398, 'dataset_min': 288.0, 'dataset_max': 929710144.0, 'dataset_mean': 28140857.15340909, 'dataset_median': 8266576.0, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "io_size\n",
      "dTr\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/tree/_classes.py:366: FutureWarning: Criterion 'mae' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='absolute_error'` which is equivalent.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'io_size', 'predictor': 'dTr', 'r2': 0.7480651974178807, 'max_error': 435440640.0, 'mae': 5010294.308510639, 'mape': 0.1713848408137187, 'medae': -0.0, 'fit_time': 1.1392066478729248, 'score_time': 0.0014464855194091797, 'dataset_min': 16.0, 'dataset_max': 1070530560.0, 'dataset_mean': 17001619.4375, 'dataset_median': 2158272.0, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "time\n",
      "liR\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'time', 'predictor': 'liR', 'r2': 0.19550230878059682, 'max_error': 23.371485707169924, 'mae': 0.7986338420390686, 'mape': 502.037581798976, 'medae': 0.45350659989449965, 'fit_time': 0.008403420448303223, 'score_time': 0.0012328028678894043, 'dataset_min': 5.57071e-06, 'dataset_max': 40.071, 'dataset_mean': 0.6204575610010015, 'dataset_median': 0.140151, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "power\n",
      "liR\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'power', 'predictor': 'liR', 'r2': 0.03865313850179533, 'max_error': 3.2830627869410014, 'mae': 0.5302940757881817, 'mape': 0.11435091990342933, 'medae': 0.36444686702018, 'fit_time': 0.008188307285308838, 'score_time': 0.0011314749717712402, 'dataset_min': 2.8355, 'dataset_max': 7.6719, 'dataset_mean': 4.322041255681818, 'dataset_median': 4.066, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "ws_size\n",
      "liR\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'ws_size', 'predictor': 'liR', 'r2': 0.33558416421563647, 'max_error': 724987183.1256964, 'mae': 25337298.212087497, 'mape': 426.05989317610056, 'medae': 12890407.70645788, 'fit_time': 0.008054792881011963, 'score_time': 0.0011208653450012207, 'dataset_min': 288.0, 'dataset_max': 929710144.0, 'dataset_mean': 28140857.15340909, 'dataset_median': 8266576.0, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "io_size\n",
      "liR\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'io_size', 'predictor': 'liR', 'r2': 0.33667589413655263, 'max_error': 632648485.4786937, 'mae': 20499289.953829855, 'mape': 6995.782942091293, 'medae': 12048115.128519315, 'fit_time': 0.00803530216217041, 'score_time': 0.0011363029479980469, 'dataset_min': 16.0, 'dataset_max': 1070530560.0, 'dataset_mean': 17001619.4375, 'dataset_median': 2158272.0, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "time\n",
      "kNN\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/linear_model/_base.py:141: FutureWarning: 'normalize' was deprecated in version 1.0 and will be removed in 1.2.\n",
      "If you wish to scale the data, use Pipeline with a StandardScaler in a preprocessing stage. To reproduce the previous behavior:\n",
      "\n",
      "from sklearn.pipeline import make_pipeline\n",
      "\n",
      "model = make_pipeline(StandardScaler(with_mean=False), LinearRegression())\n",
      "\n",
      "If you wish to pass a sample_weight parameter, you need to pass it as a fit parameter to each step of the pipeline as follows:\n",
      "\n",
      "kwargs = {s[0] + '__sample_weight': sample_weight for s in model.steps}\n",
      "model.fit(X, y, **kwargs)\n",
      "\n",
      "\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'time', 'predictor': 'kNN', 'r2': 0.5442508996778548, 'max_error': 13.147348968661387, 'mae': 0.28160212751654545, 'mape': 3.092006154658183, 'medae': 0.029988770320015394, 'fit_time': 0.006581425666809082, 'score_time': 0.021741151809692383, 'dataset_min': 5.57071e-06, 'dataset_max': 40.071, 'dataset_mean': 0.6204575610010015, 'dataset_median': 0.140151, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "power\n",
      "kNN\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'power', 'predictor': 'kNN', 'r2': -0.15858966747994852, 'max_error': 3.2272803193200525, 'mae': 0.4951681210317665, 'mape': 0.10845837264260559, 'medae': 0.21530551896164246, 'fit_time': 0.006758391857147217, 'score_time': 0.018562138080596924, 'dataset_min': 2.8355, 'dataset_max': 7.6719, 'dataset_mean': 4.322041255681818, 'dataset_median': 4.066, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "ws_size\n",
      "kNN\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'ws_size', 'predictor': 'kNN', 'r2': 0.677402193901387, 'max_error': 576965688.2662809, 'mae': 12147517.900544632, 'mape': 1.1124911802832802, 'medae': 1207405.3326903023, 'fit_time': 0.00674206018447876, 'score_time': 0.0188828706741333, 'dataset_min': 288.0, 'dataset_max': 929710144.0, 'dataset_mean': 28140857.15340909, 'dataset_median': 8266576.0, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "io_size\n",
      "kNN\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'io_size', 'predictor': 'kNN', 'r2': 0.6216315847498883, 'max_error': 411994955.0114497, 'mae': 8073257.007377684, 'mape': 2.4101580249540926, 'medae': 310004.6245160875, 'fit_time': 0.00682675838470459, 'score_time': 0.01876741647720337, 'dataset_min': 16.0, 'dataset_max': 1070530560.0, 'dataset_mean': 17001619.4375, 'dataset_median': 2158272.0, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "time\n",
      "SVR\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'time', 'predictor': 'SVR', 'r2': 0.26749635959759754, 'max_error': 24.60727542452095, 'mae': 0.4013257468544931, 'mape': 72.7377103728972, 'medae': 0.06078735423052328, 'fit_time': 0.10132378339767456, 'score_time': 0.0229833722114563, 'dataset_min': 5.57071e-06, 'dataset_max': 40.071, 'dataset_mean': 0.6204575610010015, 'dataset_median': 0.140151, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "power\n",
      "SVR\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'power', 'predictor': 'SVR', 'r2': -0.05202887652969579, 'max_error': 3.437029037835649, 'mae': 0.4531581181197981, 'mape': 0.09202962865322901, 'medae': 0.20198933463262175, 'fit_time': 0.17384332418441772, 'score_time': 0.04891788959503174, 'dataset_min': 2.8355, 'dataset_max': 7.6719, 'dataset_mean': 4.322041255681818, 'dataset_median': 4.066, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "ws_size\n",
      "SVR\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'ws_size', 'predictor': 'SVR', 'r2': -0.0982497338325099, 'max_error': 824117536.2547828, 'mae': 26516708.30044393, 'mape': 149.81697897909748, 'medae': 7366330.266819723, 'fit_time': 0.2000025510787964, 'score_time': 0.06598812341690063, 'dataset_min': 288.0, 'dataset_max': 929710144.0, 'dataset_mean': 28140857.15340909, 'dataset_median': 8266576.0, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "io_size\n",
      "SVR\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'io_size', 'predictor': 'SVR', 'r2': -0.08573487856274081, 'max_error': 681893163.0442071, 'mae': 16156218.386902167, 'mape': 401.11006541360274, 'medae': 2055840.1113200828, 'fit_time': 0.2065873146057129, 'score_time': 0.06798094511032104, 'dataset_min': 16.0, 'dataset_max': 1070530560.0, 'dataset_mean': 17001619.4375, 'dataset_median': 2158272.0, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "time\n",
      "MLP\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'time', 'predictor': 'MLP', 'r2': 0.6491020398050243, 'max_error': 19.448244502303755, 'mae': 0.3059765391995006, 'mape': 60.45021820690374, 'medae': 0.06410016738068594, 'fit_time': 3.1484957337379456, 'score_time': 0.0034604668617248535, 'dataset_min': 5.57071e-06, 'dataset_max': 40.071, 'dataset_mean': 0.6204575610010015, 'dataset_median': 0.140151, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "power\n",
      "MLP\n",
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'power', 'predictor': 'MLP', 'r2': -0.06262123406406722, 'max_error': 3.373746483760855, 'mae': 0.5228803205916737, 'mape': 0.11236258476762964, 'medae': 0.302038046539312, 'fit_time': 7.782306492328644, 'score_time': 0.004549980163574219, 'dataset_min': 2.8355, 'dataset_max': 7.6719, 'dataset_mean': 4.322041255681818, 'dataset_median': 4.066, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "ws_size\n",
      "MLP\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:692: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:692: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:692: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'ws_size', 'predictor': 'MLP', 'r2': 0.42472833933021625, 'max_error': 571765649.6141946, 'mae': 18635154.129464526, 'mape': 74.76258768977243, 'medae': 5864878.882766908, 'fit_time': 36.565068781375885, 'score_time': 0.00529026985168457, 'dataset_min': 288.0, 'dataset_max': 929710144.0, 'dataset_mean': 28140857.15340909, 'dataset_median': 8266576.0, 'dataset_size': 2816}\n",
      "\n",
      "rasp4b\n",
      "dilated_conv2d\n",
      "io_size\n",
      "MLP\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:692: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:692: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/home/max/.local/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:692: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'device': 'rasp4b', 'workload': 'dilated_conv2d', 'metric': 'io_size', 'predictor': 'MLP', 'r2': 0.5621686080397955, 'max_error': 454548063.26999134, 'mae': 12748752.672050577, 'mape': 772.7357275447278, 'medae': 2932236.4079387933, 'fit_time': 38.68094086647034, 'score_time': 0.006015658378601074, 'dataset_min': 16.0, 'dataset_max': 1070530560.0, 'dataset_mean': 17001619.4375, 'dataset_median': 2158272.0, 'dataset_size': 2816}\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/max/.local/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:692: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n"
     ]
    }
   ],
   "source": [
    "for target in layer_targets:\n",
    "    print(target)\n",
    "    tmp = target.split(\"/\")\n",
    "    workload_name = tmp[-2]\n",
    "    device_name = tmp[-3]\n",
    "    files = os.listdir(target)\n",
    "    dataset = {}\n",
    "    \n",
    "    labels = [\"time\", \"power\", \"memory\"]\n",
    "    if device_name == \"rasp4b\" or device_name == \"opencl_Mali-T628_MP6\":\n",
    "        labels = [\"time\", \"power\", \"ws_size\", \"io_size\"]\n",
    "    \n",
    "    print(\"{} : {}\\t:\\t contains {} samples\".format(workload_name, device_name, len(files)))\n",
    "    dataset = prepare_dataset(target, workload_name)\n",
    "    #print(\"\\tLoading data into memory:\\tcompleted\")\n",
    "    df, output = create_dataframe(dataset, workload_name, labels=labels)\n",
    "    #print(\"\\tCreating dataframe:\\t\\tcompleted\")\n",
    "    print(\"\\t[INFO] Remaining Samples after duplicate elimination:\\t{}\".format(len(df)))\n",
    "    #print(\"\\t[INFO] Features that are used as predictor inputs   :\\n\\t\\t{}\".format(list(df.columns)))\n",
    "    #print(\"\\t[INFO] Metrics that are going to be predicted       :\\n\\t\\t{}\".format(list(output.columns)))\n",
    "    #print()\n",
    "    X = df.to_numpy()\n",
    "    Y = output.to_numpy()\n",
    "    labels = list(output.columns)\n",
    "\n",
    "    #X_train, X_test, y_train, y_test = train_test_split(X,Y, test_size=0.2)\n",
    "    scoring = [\"r2\", \"neg_mean_absolute_error\", \"neg_mean_absolute_percentage_error\", \"max_error\", \"neg_median_absolute_error\"]\n",
    "    for method, constructor in models.items():\n",
    "        for idx, name in enumerate(labels):\n",
    "            print()\n",
    "            print(device_name)\n",
    "            print(workload_name)\n",
    "            print(name)\n",
    "            print(method)\n",
    "            \n",
    "            if kwargs_dict[method] != None:\n",
    "                model = make_pipeline(RobustScaler(), constructor(**kwargs_dict[method]))\n",
    "            else:\n",
    "                model = make_pipeline(RobustScaler(), constructor())\n",
    "                \n",
    "            #model.fit(X, Y[:,idx])\n",
    "            score = cross_validate(model, X, Y[:, idx], cv=ShuffleSplit(n_splits=4, test_size=0.2), scoring=scoring)\n",
    "            #print(score)\n",
    "\n",
    "            result = {\n",
    "                \"device\" : device_name,\n",
    "                \"workload\" : workload_name,\n",
    "                \"metric\" : name,\n",
    "                \"predictor\" : method,\n",
    "                \"r2\" : np.mean(score[\"test_r2\"]),\n",
    "                \"max_error\" : np.mean(score[\"test_max_error\"])*(-1),\n",
    "                \"mae\" : np.mean(score[\"test_neg_mean_absolute_error\"])*(-1),\n",
    "                \"mape\" : np.mean(score[\"test_neg_mean_absolute_percentage_error\"])*(-1),\n",
    "                \"medae\" : np.mean(score[\"test_neg_median_absolute_error\"])*(-1),\n",
    "                \"fit_time\" : np.mean(score[\"fit_time\"]),\n",
    "                \"score_time\" : np.mean(score[\"score_time\"]),\n",
    "                \"dataset_min\" : Y[:,idx].min(),\n",
    "                \"dataset_max\" : Y[:,idx].max(),\n",
    "                \"dataset_mean\" : Y[:,idx].mean(),\n",
    "                \"dataset_median\" : np.median(Y[:,idx]),\n",
    "                \"dataset_size\": len(X),\n",
    "            }\n",
    "            print(result)\n",
    "            #print(\"R2: {.3f}\\tMAE {.3f}\\tMAPE {.3f}\\tMEDAE {.3f}\\tMaxError {.3f}\".format(result[\"r2\"],result[\"mae\"],result[\"mape\"],result[\"medae\"],result[\"max_error\"]))\n",
    "            results[device_name+\"-\"+workload_name+\"-\"+name+\"-\"+method] = result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "41799cb9-b58a-477b-b7f2-b4f30bc038ec",
   "metadata": {},
   "outputs": [],
   "source": [
    "date_time = \"short_results/results_rbst_scaler_rpi4b_KFold_\"+str(int(time.time()))\n",
    "results = pd.DataFrame.from_dict(results, \"index\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "1f36e0c3-ded3-4234-b8e4-2f5a6c783dd8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done\n"
     ]
    }
   ],
   "source": [
    "results.to_csv(date_time+\"-predictor_results.csv\")\n",
    "results.to_excel(date_time+\"-predictor_results.xlsx\")\n",
    "results.to_html(date_time+\"-predictor_results.html\")\n",
    "results.to_markdown(date_time+\"-predictor_results.md\")\n",
    "print(\"done\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "48eb9893-836b-4314-a030-3e9bca4a8db2",
   "metadata": {},
   "outputs": [],
   "source": [
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "cb4ac1b4-5336-40b5-bde8-d79fea599e37",
   "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>device</th>\n",
       "      <th>workload</th>\n",
       "      <th>metric</th>\n",
       "      <th>predictor</th>\n",
       "      <th>r2</th>\n",
       "      <th>max_error</th>\n",
       "      <th>mae</th>\n",
       "      <th>mape</th>\n",
       "      <th>medae</th>\n",
       "      <th>fit_time</th>\n",
       "      <th>score_time</th>\n",
       "      <th>dataset_min</th>\n",
       "      <th>dataset_max</th>\n",
       "      <th>dataset_mean</th>\n",
       "      <th>dataset_median</th>\n",
       "      <th>dataset_size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>cuda_A100-avg_pool2d-memory-xgb</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>avg_pool2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>xgb</td>\n",
       "      <td>0.964444</td>\n",
       "      <td>2.974870</td>\n",
       "      <td>0.242980</td>\n",
       "      <td>0.127675</td>\n",
       "      <td>0.120853</td>\n",
       "      <td>0.404485</td>\n",
       "      <td>0.006448</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>12.140625</td>\n",
       "      <td>2.228403</td>\n",
       "      <td>1.253906</td>\n",
       "      <td>4510</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-avg_pool2d-memory-ert</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>avg_pool2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>ert</td>\n",
       "      <td>0.958458</td>\n",
       "      <td>3.596574</td>\n",
       "      <td>0.207365</td>\n",
       "      <td>0.082563</td>\n",
       "      <td>0.074877</td>\n",
       "      <td>1.529772</td>\n",
       "      <td>0.046241</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>12.140625</td>\n",
       "      <td>2.228403</td>\n",
       "      <td>1.253906</td>\n",
       "      <td>4510</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-avg_pool2d-memory-dTr</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>avg_pool2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>dTr</td>\n",
       "      <td>0.873550</td>\n",
       "      <td>6.135742</td>\n",
       "      <td>0.432842</td>\n",
       "      <td>0.158502</td>\n",
       "      <td>0.149170</td>\n",
       "      <td>1.644190</td>\n",
       "      <td>0.001529</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>12.140625</td>\n",
       "      <td>2.228403</td>\n",
       "      <td>1.253906</td>\n",
       "      <td>4510</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-avg_pool2d-memory-MLP</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>avg_pool2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>MLP</td>\n",
       "      <td>0.974437</td>\n",
       "      <td>4.530356</td>\n",
       "      <td>0.172801</td>\n",
       "      <td>0.086149</td>\n",
       "      <td>0.088608</td>\n",
       "      <td>8.537318</td>\n",
       "      <td>0.005928</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>12.140625</td>\n",
       "      <td>2.228403</td>\n",
       "      <td>1.253906</td>\n",
       "      <td>4510</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-conv2d-memory-xgb</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>conv2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>xgb</td>\n",
       "      <td>0.875510</td>\n",
       "      <td>2.986231</td>\n",
       "      <td>0.028570</td>\n",
       "      <td>0.023212</td>\n",
       "      <td>0.002882</td>\n",
       "      <td>0.324407</td>\n",
       "      <td>0.004309</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>8.394531</td>\n",
       "      <td>0.526753</td>\n",
       "      <td>0.423828</td>\n",
       "      <td>3227</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-conv2d-memory-ert</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>conv2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>ert</td>\n",
       "      <td>0.903359</td>\n",
       "      <td>1.752483</td>\n",
       "      <td>0.020264</td>\n",
       "      <td>0.018353</td>\n",
       "      <td>0.001090</td>\n",
       "      <td>0.827659</td>\n",
       "      <td>0.030908</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>8.394531</td>\n",
       "      <td>0.526753</td>\n",
       "      <td>0.423828</td>\n",
       "      <td>3227</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-conv2d-memory-dTr</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>conv2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>dTr</td>\n",
       "      <td>0.669372</td>\n",
       "      <td>3.565918</td>\n",
       "      <td>0.045734</td>\n",
       "      <td>0.039183</td>\n",
       "      <td>0.001953</td>\n",
       "      <td>1.303666</td>\n",
       "      <td>0.001400</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>8.394531</td>\n",
       "      <td>0.526753</td>\n",
       "      <td>0.423828</td>\n",
       "      <td>3227</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-conv2d-memory-MLP</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>conv2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>MLP</td>\n",
       "      <td>0.838607</td>\n",
       "      <td>3.382187</td>\n",
       "      <td>0.046896</td>\n",
       "      <td>0.058231</td>\n",
       "      <td>0.015715</td>\n",
       "      <td>6.052887</td>\n",
       "      <td>0.005699</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>8.394531</td>\n",
       "      <td>0.526753</td>\n",
       "      <td>0.423828</td>\n",
       "      <td>3227</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-dense-memory-xgb</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>dense</td>\n",
       "      <td>memory</td>\n",
       "      <td>xgb</td>\n",
       "      <td>0.999811</td>\n",
       "      <td>0.015572</td>\n",
       "      <td>0.001485</td>\n",
       "      <td>0.002752</td>\n",
       "      <td>0.001116</td>\n",
       "      <td>0.588313</td>\n",
       "      <td>0.008258</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>2.193359</td>\n",
       "      <td>0.533942</td>\n",
       "      <td>0.490234</td>\n",
       "      <td>18421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-dense-memory-ert</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>dense</td>\n",
       "      <td>memory</td>\n",
       "      <td>ert</td>\n",
       "      <td>0.999898</td>\n",
       "      <td>0.014718</td>\n",
       "      <td>0.000828</td>\n",
       "      <td>0.001477</td>\n",
       "      <td>0.000472</td>\n",
       "      <td>2.304010</td>\n",
       "      <td>0.154267</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>2.193359</td>\n",
       "      <td>0.533942</td>\n",
       "      <td>0.490234</td>\n",
       "      <td>18421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-dense-memory-dTr</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>dense</td>\n",
       "      <td>memory</td>\n",
       "      <td>dTr</td>\n",
       "      <td>0.999663</td>\n",
       "      <td>0.043457</td>\n",
       "      <td>0.001429</td>\n",
       "      <td>0.002515</td>\n",
       "      <td>-0.000000</td>\n",
       "      <td>8.193096</td>\n",
       "      <td>0.002217</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>2.193359</td>\n",
       "      <td>0.533942</td>\n",
       "      <td>0.490234</td>\n",
       "      <td>18421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-dense-memory-MLP</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>dense</td>\n",
       "      <td>memory</td>\n",
       "      <td>MLP</td>\n",
       "      <td>0.999309</td>\n",
       "      <td>0.066499</td>\n",
       "      <td>0.002488</td>\n",
       "      <td>0.004736</td>\n",
       "      <td>0.002094</td>\n",
       "      <td>7.410416</td>\n",
       "      <td>0.023939</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>2.193359</td>\n",
       "      <td>0.533942</td>\n",
       "      <td>0.490234</td>\n",
       "      <td>18421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-depthwise_conv2d-memory-xgb</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>depthwise_conv2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>xgb</td>\n",
       "      <td>0.922525</td>\n",
       "      <td>1.888447</td>\n",
       "      <td>0.026472</td>\n",
       "      <td>0.027502</td>\n",
       "      <td>0.003893</td>\n",
       "      <td>0.347254</td>\n",
       "      <td>0.007412</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>6.589844</td>\n",
       "      <td>0.532126</td>\n",
       "      <td>0.421875</td>\n",
       "      <td>2537</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-depthwise_conv2d-memory-ert</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>depthwise_conv2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>ert</td>\n",
       "      <td>0.918942</td>\n",
       "      <td>2.068345</td>\n",
       "      <td>0.019919</td>\n",
       "      <td>0.015680</td>\n",
       "      <td>0.001006</td>\n",
       "      <td>0.656936</td>\n",
       "      <td>0.024269</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>6.589844</td>\n",
       "      <td>0.532126</td>\n",
       "      <td>0.421875</td>\n",
       "      <td>2537</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-depthwise_conv2d-memory-dTr</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>depthwise_conv2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>dTr</td>\n",
       "      <td>0.755509</td>\n",
       "      <td>2.483398</td>\n",
       "      <td>0.044538</td>\n",
       "      <td>0.040583</td>\n",
       "      <td>0.002930</td>\n",
       "      <td>0.678993</td>\n",
       "      <td>0.001412</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>6.589844</td>\n",
       "      <td>0.532126</td>\n",
       "      <td>0.421875</td>\n",
       "      <td>2537</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-depthwise_conv2d-memory-MLP</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>depthwise_conv2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>MLP</td>\n",
       "      <td>0.823174</td>\n",
       "      <td>2.042712</td>\n",
       "      <td>0.048897</td>\n",
       "      <td>0.067967</td>\n",
       "      <td>0.017096</td>\n",
       "      <td>4.001454</td>\n",
       "      <td>0.004452</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>6.589844</td>\n",
       "      <td>0.532126</td>\n",
       "      <td>0.421875</td>\n",
       "      <td>2537</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-dilated_conv2d-memory-xgb</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>dilated_conv2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>xgb</td>\n",
       "      <td>0.971693</td>\n",
       "      <td>2.464311</td>\n",
       "      <td>0.009304</td>\n",
       "      <td>0.011247</td>\n",
       "      <td>0.002592</td>\n",
       "      <td>0.492724</td>\n",
       "      <td>0.013705</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>10.021484</td>\n",
       "      <td>0.535396</td>\n",
       "      <td>0.427734</td>\n",
       "      <td>7887</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-dilated_conv2d-memory-ert</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>dilated_conv2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>ert</td>\n",
       "      <td>0.963757</td>\n",
       "      <td>2.342319</td>\n",
       "      <td>0.005222</td>\n",
       "      <td>0.003243</td>\n",
       "      <td>0.000020</td>\n",
       "      <td>1.822019</td>\n",
       "      <td>0.056541</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>10.021484</td>\n",
       "      <td>0.535396</td>\n",
       "      <td>0.427734</td>\n",
       "      <td>7887</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-dilated_conv2d-memory-dTr</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>dilated_conv2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>dTr</td>\n",
       "      <td>0.932958</td>\n",
       "      <td>2.007324</td>\n",
       "      <td>0.007596</td>\n",
       "      <td>0.005566</td>\n",
       "      <td>-0.000000</td>\n",
       "      <td>7.323965</td>\n",
       "      <td>0.001735</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>10.021484</td>\n",
       "      <td>0.535396</td>\n",
       "      <td>0.427734</td>\n",
       "      <td>7887</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-dilated_conv2d-memory-MLP</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>dilated_conv2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>MLP</td>\n",
       "      <td>0.956945</td>\n",
       "      <td>1.997973</td>\n",
       "      <td>0.024355</td>\n",
       "      <td>0.033716</td>\n",
       "      <td>0.008954</td>\n",
       "      <td>11.718817</td>\n",
       "      <td>0.009819</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>10.021484</td>\n",
       "      <td>0.535396</td>\n",
       "      <td>0.427734</td>\n",
       "      <td>7887</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-max_pool2d-memory-xgb</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>max_pool2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>xgb</td>\n",
       "      <td>0.965130</td>\n",
       "      <td>3.065484</td>\n",
       "      <td>0.250141</td>\n",
       "      <td>0.130101</td>\n",
       "      <td>0.131659</td>\n",
       "      <td>0.377533</td>\n",
       "      <td>0.004432</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>11.765625</td>\n",
       "      <td>2.246351</td>\n",
       "      <td>1.267578</td>\n",
       "      <td>4449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-max_pool2d-memory-ert</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>max_pool2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>ert</td>\n",
       "      <td>0.959834</td>\n",
       "      <td>3.938700</td>\n",
       "      <td>0.208198</td>\n",
       "      <td>0.081811</td>\n",
       "      <td>0.076070</td>\n",
       "      <td>1.456737</td>\n",
       "      <td>0.043985</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>11.765625</td>\n",
       "      <td>2.246351</td>\n",
       "      <td>1.267578</td>\n",
       "      <td>4449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-max_pool2d-memory-dTr</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>max_pool2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>dTr</td>\n",
       "      <td>0.868178</td>\n",
       "      <td>5.245605</td>\n",
       "      <td>0.442489</td>\n",
       "      <td>0.163336</td>\n",
       "      <td>0.157471</td>\n",
       "      <td>1.563682</td>\n",
       "      <td>0.001719</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>11.765625</td>\n",
       "      <td>2.246351</td>\n",
       "      <td>1.267578</td>\n",
       "      <td>4449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuda_A100-max_pool2d-memory-MLP</th>\n",
       "      <td>cuda_A100</td>\n",
       "      <td>max_pool2d</td>\n",
       "      <td>memory</td>\n",
       "      <td>MLP</td>\n",
       "      <td>0.990310</td>\n",
       "      <td>2.740154</td>\n",
       "      <td>0.104909</td>\n",
       "      <td>0.055627</td>\n",
       "      <td>0.048776</td>\n",
       "      <td>6.329126</td>\n",
       "      <td>0.006049</td>\n",
       "      <td>0.40625</td>\n",
       "      <td>11.765625</td>\n",
       "      <td>2.246351</td>\n",
       "      <td>1.267578</td>\n",
       "      <td>4449</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                          device          workload  metric  \\\n",
       "cuda_A100-avg_pool2d-memory-xgb        cuda_A100        avg_pool2d  memory   \n",
       "cuda_A100-avg_pool2d-memory-ert        cuda_A100        avg_pool2d  memory   \n",
       "cuda_A100-avg_pool2d-memory-dTr        cuda_A100        avg_pool2d  memory   \n",
       "cuda_A100-avg_pool2d-memory-MLP        cuda_A100        avg_pool2d  memory   \n",
       "cuda_A100-conv2d-memory-xgb            cuda_A100            conv2d  memory   \n",
       "cuda_A100-conv2d-memory-ert            cuda_A100            conv2d  memory   \n",
       "cuda_A100-conv2d-memory-dTr            cuda_A100            conv2d  memory   \n",
       "cuda_A100-conv2d-memory-MLP            cuda_A100            conv2d  memory   \n",
       "cuda_A100-dense-memory-xgb             cuda_A100             dense  memory   \n",
       "cuda_A100-dense-memory-ert             cuda_A100             dense  memory   \n",
       "cuda_A100-dense-memory-dTr             cuda_A100             dense  memory   \n",
       "cuda_A100-dense-memory-MLP             cuda_A100             dense  memory   \n",
       "cuda_A100-depthwise_conv2d-memory-xgb  cuda_A100  depthwise_conv2d  memory   \n",
       "cuda_A100-depthwise_conv2d-memory-ert  cuda_A100  depthwise_conv2d  memory   \n",
       "cuda_A100-depthwise_conv2d-memory-dTr  cuda_A100  depthwise_conv2d  memory   \n",
       "cuda_A100-depthwise_conv2d-memory-MLP  cuda_A100  depthwise_conv2d  memory   \n",
       "cuda_A100-dilated_conv2d-memory-xgb    cuda_A100    dilated_conv2d  memory   \n",
       "cuda_A100-dilated_conv2d-memory-ert    cuda_A100    dilated_conv2d  memory   \n",
       "cuda_A100-dilated_conv2d-memory-dTr    cuda_A100    dilated_conv2d  memory   \n",
       "cuda_A100-dilated_conv2d-memory-MLP    cuda_A100    dilated_conv2d  memory   \n",
       "cuda_A100-max_pool2d-memory-xgb        cuda_A100        max_pool2d  memory   \n",
       "cuda_A100-max_pool2d-memory-ert        cuda_A100        max_pool2d  memory   \n",
       "cuda_A100-max_pool2d-memory-dTr        cuda_A100        max_pool2d  memory   \n",
       "cuda_A100-max_pool2d-memory-MLP        cuda_A100        max_pool2d  memory   \n",
       "\n",
       "                                      predictor        r2  max_error  \\\n",
       "cuda_A100-avg_pool2d-memory-xgb             xgb  0.964444   2.974870   \n",
       "cuda_A100-avg_pool2d-memory-ert             ert  0.958458   3.596574   \n",
       "cuda_A100-avg_pool2d-memory-dTr             dTr  0.873550   6.135742   \n",
       "cuda_A100-avg_pool2d-memory-MLP             MLP  0.974437   4.530356   \n",
       "cuda_A100-conv2d-memory-xgb                 xgb  0.875510   2.986231   \n",
       "cuda_A100-conv2d-memory-ert                 ert  0.903359   1.752483   \n",
       "cuda_A100-conv2d-memory-dTr                 dTr  0.669372   3.565918   \n",
       "cuda_A100-conv2d-memory-MLP                 MLP  0.838607   3.382187   \n",
       "cuda_A100-dense-memory-xgb                  xgb  0.999811   0.015572   \n",
       "cuda_A100-dense-memory-ert                  ert  0.999898   0.014718   \n",
       "cuda_A100-dense-memory-dTr                  dTr  0.999663   0.043457   \n",
       "cuda_A100-dense-memory-MLP                  MLP  0.999309   0.066499   \n",
       "cuda_A100-depthwise_conv2d-memory-xgb       xgb  0.922525   1.888447   \n",
       "cuda_A100-depthwise_conv2d-memory-ert       ert  0.918942   2.068345   \n",
       "cuda_A100-depthwise_conv2d-memory-dTr       dTr  0.755509   2.483398   \n",
       "cuda_A100-depthwise_conv2d-memory-MLP       MLP  0.823174   2.042712   \n",
       "cuda_A100-dilated_conv2d-memory-xgb         xgb  0.971693   2.464311   \n",
       "cuda_A100-dilated_conv2d-memory-ert         ert  0.963757   2.342319   \n",
       "cuda_A100-dilated_conv2d-memory-dTr         dTr  0.932958   2.007324   \n",
       "cuda_A100-dilated_conv2d-memory-MLP         MLP  0.956945   1.997973   \n",
       "cuda_A100-max_pool2d-memory-xgb             xgb  0.965130   3.065484   \n",
       "cuda_A100-max_pool2d-memory-ert             ert  0.959834   3.938700   \n",
       "cuda_A100-max_pool2d-memory-dTr             dTr  0.868178   5.245605   \n",
       "cuda_A100-max_pool2d-memory-MLP             MLP  0.990310   2.740154   \n",
       "\n",
       "                                            mae      mape     medae  \\\n",
       "cuda_A100-avg_pool2d-memory-xgb        0.242980  0.127675  0.120853   \n",
       "cuda_A100-avg_pool2d-memory-ert        0.207365  0.082563  0.074877   \n",
       "cuda_A100-avg_pool2d-memory-dTr        0.432842  0.158502  0.149170   \n",
       "cuda_A100-avg_pool2d-memory-MLP        0.172801  0.086149  0.088608   \n",
       "cuda_A100-conv2d-memory-xgb            0.028570  0.023212  0.002882   \n",
       "cuda_A100-conv2d-memory-ert            0.020264  0.018353  0.001090   \n",
       "cuda_A100-conv2d-memory-dTr            0.045734  0.039183  0.001953   \n",
       "cuda_A100-conv2d-memory-MLP            0.046896  0.058231  0.015715   \n",
       "cuda_A100-dense-memory-xgb             0.001485  0.002752  0.001116   \n",
       "cuda_A100-dense-memory-ert             0.000828  0.001477  0.000472   \n",
       "cuda_A100-dense-memory-dTr             0.001429  0.002515 -0.000000   \n",
       "cuda_A100-dense-memory-MLP             0.002488  0.004736  0.002094   \n",
       "cuda_A100-depthwise_conv2d-memory-xgb  0.026472  0.027502  0.003893   \n",
       "cuda_A100-depthwise_conv2d-memory-ert  0.019919  0.015680  0.001006   \n",
       "cuda_A100-depthwise_conv2d-memory-dTr  0.044538  0.040583  0.002930   \n",
       "cuda_A100-depthwise_conv2d-memory-MLP  0.048897  0.067967  0.017096   \n",
       "cuda_A100-dilated_conv2d-memory-xgb    0.009304  0.011247  0.002592   \n",
       "cuda_A100-dilated_conv2d-memory-ert    0.005222  0.003243  0.000020   \n",
       "cuda_A100-dilated_conv2d-memory-dTr    0.007596  0.005566 -0.000000   \n",
       "cuda_A100-dilated_conv2d-memory-MLP    0.024355  0.033716  0.008954   \n",
       "cuda_A100-max_pool2d-memory-xgb        0.250141  0.130101  0.131659   \n",
       "cuda_A100-max_pool2d-memory-ert        0.208198  0.081811  0.076070   \n",
       "cuda_A100-max_pool2d-memory-dTr        0.442489  0.163336  0.157471   \n",
       "cuda_A100-max_pool2d-memory-MLP        0.104909  0.055627  0.048776   \n",
       "\n",
       "                                        fit_time  score_time  dataset_min  \\\n",
       "cuda_A100-avg_pool2d-memory-xgb         0.404485    0.006448      0.40625   \n",
       "cuda_A100-avg_pool2d-memory-ert         1.529772    0.046241      0.40625   \n",
       "cuda_A100-avg_pool2d-memory-dTr         1.644190    0.001529      0.40625   \n",
       "cuda_A100-avg_pool2d-memory-MLP         8.537318    0.005928      0.40625   \n",
       "cuda_A100-conv2d-memory-xgb             0.324407    0.004309      0.40625   \n",
       "cuda_A100-conv2d-memory-ert             0.827659    0.030908      0.40625   \n",
       "cuda_A100-conv2d-memory-dTr             1.303666    0.001400      0.40625   \n",
       "cuda_A100-conv2d-memory-MLP             6.052887    0.005699      0.40625   \n",
       "cuda_A100-dense-memory-xgb              0.588313    0.008258      0.40625   \n",
       "cuda_A100-dense-memory-ert              2.304010    0.154267      0.40625   \n",
       "cuda_A100-dense-memory-dTr              8.193096    0.002217      0.40625   \n",
       "cuda_A100-dense-memory-MLP              7.410416    0.023939      0.40625   \n",
       "cuda_A100-depthwise_conv2d-memory-xgb   0.347254    0.007412      0.40625   \n",
       "cuda_A100-depthwise_conv2d-memory-ert   0.656936    0.024269      0.40625   \n",
       "cuda_A100-depthwise_conv2d-memory-dTr   0.678993    0.001412      0.40625   \n",
       "cuda_A100-depthwise_conv2d-memory-MLP   4.001454    0.004452      0.40625   \n",
       "cuda_A100-dilated_conv2d-memory-xgb     0.492724    0.013705      0.40625   \n",
       "cuda_A100-dilated_conv2d-memory-ert     1.822019    0.056541      0.40625   \n",
       "cuda_A100-dilated_conv2d-memory-dTr     7.323965    0.001735      0.40625   \n",
       "cuda_A100-dilated_conv2d-memory-MLP    11.718817    0.009819      0.40625   \n",
       "cuda_A100-max_pool2d-memory-xgb         0.377533    0.004432      0.40625   \n",
       "cuda_A100-max_pool2d-memory-ert         1.456737    0.043985      0.40625   \n",
       "cuda_A100-max_pool2d-memory-dTr         1.563682    0.001719      0.40625   \n",
       "cuda_A100-max_pool2d-memory-MLP         6.329126    0.006049      0.40625   \n",
       "\n",
       "                                       dataset_max  dataset_mean  \\\n",
       "cuda_A100-avg_pool2d-memory-xgb          12.140625      2.228403   \n",
       "cuda_A100-avg_pool2d-memory-ert          12.140625      2.228403   \n",
       "cuda_A100-avg_pool2d-memory-dTr          12.140625      2.228403   \n",
       "cuda_A100-avg_pool2d-memory-MLP          12.140625      2.228403   \n",
       "cuda_A100-conv2d-memory-xgb               8.394531      0.526753   \n",
       "cuda_A100-conv2d-memory-ert               8.394531      0.526753   \n",
       "cuda_A100-conv2d-memory-dTr               8.394531      0.526753   \n",
       "cuda_A100-conv2d-memory-MLP               8.394531      0.526753   \n",
       "cuda_A100-dense-memory-xgb                2.193359      0.533942   \n",
       "cuda_A100-dense-memory-ert                2.193359      0.533942   \n",
       "cuda_A100-dense-memory-dTr                2.193359      0.533942   \n",
       "cuda_A100-dense-memory-MLP                2.193359      0.533942   \n",
       "cuda_A100-depthwise_conv2d-memory-xgb     6.589844      0.532126   \n",
       "cuda_A100-depthwise_conv2d-memory-ert     6.589844      0.532126   \n",
       "cuda_A100-depthwise_conv2d-memory-dTr     6.589844      0.532126   \n",
       "cuda_A100-depthwise_conv2d-memory-MLP     6.589844      0.532126   \n",
       "cuda_A100-dilated_conv2d-memory-xgb      10.021484      0.535396   \n",
       "cuda_A100-dilated_conv2d-memory-ert      10.021484      0.535396   \n",
       "cuda_A100-dilated_conv2d-memory-dTr      10.021484      0.535396   \n",
       "cuda_A100-dilated_conv2d-memory-MLP      10.021484      0.535396   \n",
       "cuda_A100-max_pool2d-memory-xgb          11.765625      2.246351   \n",
       "cuda_A100-max_pool2d-memory-ert          11.765625      2.246351   \n",
       "cuda_A100-max_pool2d-memory-dTr          11.765625      2.246351   \n",
       "cuda_A100-max_pool2d-memory-MLP          11.765625      2.246351   \n",
       "\n",
       "                                       dataset_median  dataset_size  \n",
       "cuda_A100-avg_pool2d-memory-xgb              1.253906          4510  \n",
       "cuda_A100-avg_pool2d-memory-ert              1.253906          4510  \n",
       "cuda_A100-avg_pool2d-memory-dTr              1.253906          4510  \n",
       "cuda_A100-avg_pool2d-memory-MLP              1.253906          4510  \n",
       "cuda_A100-conv2d-memory-xgb                  0.423828          3227  \n",
       "cuda_A100-conv2d-memory-ert                  0.423828          3227  \n",
       "cuda_A100-conv2d-memory-dTr                  0.423828          3227  \n",
       "cuda_A100-conv2d-memory-MLP                  0.423828          3227  \n",
       "cuda_A100-dense-memory-xgb                   0.490234         18421  \n",
       "cuda_A100-dense-memory-ert                   0.490234         18421  \n",
       "cuda_A100-dense-memory-dTr                   0.490234         18421  \n",
       "cuda_A100-dense-memory-MLP                   0.490234         18421  \n",
       "cuda_A100-depthwise_conv2d-memory-xgb        0.421875          2537  \n",
       "cuda_A100-depthwise_conv2d-memory-ert        0.421875          2537  \n",
       "cuda_A100-depthwise_conv2d-memory-dTr        0.421875          2537  \n",
       "cuda_A100-depthwise_conv2d-memory-MLP        0.421875          2537  \n",
       "cuda_A100-dilated_conv2d-memory-xgb          0.427734          7887  \n",
       "cuda_A100-dilated_conv2d-memory-ert          0.427734          7887  \n",
       "cuda_A100-dilated_conv2d-memory-dTr          0.427734          7887  \n",
       "cuda_A100-dilated_conv2d-memory-MLP          0.427734          7887  \n",
       "cuda_A100-max_pool2d-memory-xgb              1.267578          4449  \n",
       "cuda_A100-max_pool2d-memory-ert              1.267578          4449  \n",
       "cuda_A100-max_pool2d-memory-dTr              1.267578          4449  \n",
       "cuda_A100-max_pool2d-memory-MLP              1.267578          4449  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results[(results.metric ==\"memory\")]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "516abcb8-aaa0-4335-bed3-c0097e9dbb4d",
   "metadata": {},
   "outputs": [],
   "source": [
    "workload = \"conv2d\"\n",
    "metric = \"time\"\n",
    "df = results[(results.workload == workload) & (results.metric == metric)]\n",
    "df_r2 = pd.concat([df[\"device\"], df[\"workload\"], df[\"metric\"], df[\"predictor\"], df[\"r2_train\"], df[\"r2_test\"]], axis=1)\n",
    "df_mae = pd.concat([df[\"device\"], df[\"workload\"], df[\"metric\"], df[\"predictor\"], df[\"mae_train\"], df[\"mae_test\"]], axis=1)\n",
    "df_mape = pd.concat([df[\"device\"], df[\"workload\"], df[\"metric\"], df[\"predictor\"], df[\"mape_train\"], df[\"mape_test\"]], axis=1)\n",
    "df_sets = pd.concat([df[\"device\"], df[\"workload\"], df[\"metric\"], df[\"training set size\"], df[\"validation set size\"]], axis=1)\n",
    "df_sets = df_sets.drop_duplicates()\n",
    "df_stats = pd.concat([df[\"device\"], df[\"workload\"], df[\"metric\"], df[\"minimum\"], df[\"maximum\"], df[\"mean\"], df[\"median\"]], axis=1)\n",
    "df_stats = df_stats.drop_duplicates()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2cc5284e-87bf-485f-aaaa-6388d03739e0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1800x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from decimal import Decimal\n",
    "plt.rcParams[\"figure.figsize\"] = (25,8)\n",
    "axes = df_r2.plot.bar(subplots=False)\n",
    "for p in axes.patches:\n",
    "    axes.annotate('{:.3f}'.format(Decimal(str(p.get_height()))), (p.get_x(), p.get_height()))\n",
    "plt.title(workload + \": \"+metric+\" - R2\")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "e74e8ba0-0b25-4a10-a334-99154776fc6d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABvsAAAI4CAYAAABEJFXWAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Z1A+gAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOzde5zPVeLH8ddnjPulIsoYGrlkzMXQFKpV8qNSKdJFtlxrtZdKV9ttbVtLV9lutii6UShsyipSstUYDAmhTBmsW8gllxmf3x/Dd00zhBk0vJ6Pxzx8v+dzzudzzpcx3/m+P+ecIAxDJEmSJEmSJEmSJBU/UUe6A5IkSZIkSZIkSZIOjmGfJEmSJEmSJEmSVEwZ9kmSJEmSJEmSJEnFlGGfJEmSJEmSJEmSVEwZ9kmSJEmSJEmSJEnFlGGfJEmSJEmSJEmSVEwZ9kmSJEnSUSIIgr5BELxWiPb3BEEwuCj7JEmSJEk6tAz7JEmSJOlXKgiCi4Mg+DQIgvVBEPw3CILBQRBULKJznxcEQdaeZWEY/j0Mw55Fcf7CCoIgMwiC7UEQnPiz8llBEIRBEMT9rLzvrvKmPyvvGgRBThAEm372FXMYhiFJkiRJh5xhnyRJkiT9eh0HPATEAPFADeCxI9qjw2sJ0Gn3kyAIkoByP68UBEEAXA/8sOvPn/ssDMMKP/tafqg6LUmSJEmHk2GfJEmSJO2nIAhqBkHwdhAEq4MgWBsEwTO7yqOCILgvCILvgiBYFQTBK0EQHLfrWNyuGWddgiD4PgiCNUEQ3LvrWEwQBD8FQVB5j2s03lWnZBiGb4RhOCEMwy1hGK4DXgTO3qNu7SAIPg6CYGMQBB8AeWbB7WMc5YH3gZg9Z7rtuQzoHv3uFgTB0iAI1gVB0CsIgjOCIJiza7bhMz87b/cgCObvqvvvIAhOKdQLDq+SN7zrArxSQL3fANWBm4FrgiAoVcjrSpIkSVKxYdgnSZIkSfshCIISwLvAd0AcubPsRuw63HXXV0vgVKAC8MzPTnEOcBrQCnggCIL4XbPLPgOu2KPetcCoMAx3FNCNFsBXezx/A5hBbsj3N3LDsD37PCcIgmt/fpIwDDcDFwHL92OmW1OgHnA18BRwL/B/QAJwVRAE5+661mXAPUAHoCowFRi+l3Pur8+BSkEQxO96/a8BCtqTsAvwL+CtXc8vLeR1JUmSJKnYMOyTJEmSpP1zJrnLad4ZhuHmMAy3hmH46a5jnYEnwzD8NgzDTcCfyZ1hFr1H+7+GYfhTGIazgdlAo13lb7Brqcpdy1Fes6ssjyAIWpMbaj2w63kt4Azg/jAMt4Vh+Am5gVdEGIbJYRjmO9cB+tuusU4ENgPDwzBcFYbhMnIDvca76vUC+oVhOD8Mw2zg70BKEc7uaw3MB5bteTAIgnLAlcAbuwLSUeRfyrPZrpmIu7++KWSfJEmSJOlXw7BPkiRJkvZPTeC7XUHWz8WQO+Nvt++AaOCkPcr+u8fjLeTO/gMYDTQPgqA6uTP3dpIbokUEQdCM3ACwYxiGC/e45rpds/T2vG5RW7nH458KeL57HKcAA3cHauTunxeQOwMyjyAIBu2xfOg9v3D9V8md7diVgpfwbA9kA+/tev46cFEQBFX3qPN5GIbH7/FV5xeuKUmSJEnFRvQvV5EkSZIkAUuBWkEQRBcQ+C0nN+zarRa5AdRKIHZfJw3DcF0QBBPJXSYzHhgRhmG4+3gQBI2BcUD3MAwn7dF0BXBCEATl9wj8agEh+2d/6+2vpcDDYRi+/osXDsNe5M4E/EVhGH4XBMESoC3Qo4AqXcgNHL/PnRhJAJQkNyAcuH9dlyRJkqTiy5l9kiRJkrR/0sgN2PoHQVA+CIIyQRCcvevYcKB3EAS1gyCoQO4Slm/uZRZgQd4gd+nJjuyxhGcQBInABOBPYRj+fInO74B04K9BEJQKguAcDmyvupVAlSAIjjuANvsyCPhzEAQJAEEQHBcEwZVFdO4ewPk/m8VIEAQ1yN0D8RIgZddXI+AR8i/lKUmSJElHJcM+SZIkSdoPYRjmkBum1QW+B7LInY0H8BK5y01+AiwBtgJ/OoDTjwPqAf/dtaffbrcDVYEheyx7+dUex68FmpK7ZOZf+Nkyl0EQfBUEQee9jGcBuSHlt7uW3ow5gP4WdL53yA3ZRgRB8CMwF7ioMOfc49zfhGGYXsCh64CMMAwnhmH4391fwD+A5F1hKeQuk7rpZ19nFEXfJEmSJOlIC/ZYHUaSJEmSJEmSJElSMeLMPkmSJEmSJEmSJKmYMuyTJEmSJEmSJEmSiinDPkmSJEmSJEmSJKmYMuyTJEmSJEmSJEmSiqnoI90BgBNPPDGMi4s70t2QJEmSJEmSJEmSfhVmzJixJgzDqr9U71cR9sXFxZGenn6kuyFJkiRJkiRJkiT9KgRB8N3+1HMZT0mSJEmSJEmSJKmYMuyTJEmSJEmSJEmSiinDPkmSJEmSJEmSJKmY+lXs2SdJkiRJkiRJkqRfvx07dpCVlcXWrVuPdFeOGmXKlCE2NpaSJUseVHvDPkmSJEmSJEmSJO2XrKwsKlasSFxcHEEQHOnuFHthGLJ27VqysrKoXbv2QZ3DZTwlSZIkSXl0796datWqkZiYGCm78847adCgAcnJybRv357169cDuXf1dunShaSkJOLj4+nXr1+B55w8eTJNmjQhMTGRLl26kJ2dDcDrr79OcnIySUlJnHXWWcyePXuf/ZAkSZJ0ZG3dupUqVaoY9BWRIAioUqVKoWZKGvZJkiRJkvLo2rUrEyZMyFPWunVr5s6dy5w5c6hfv34k1Bs5ciTbtm3jyy+/ZMaMGfzzn/8kMzMzT9udO3fSpUsXRowYwdy5cznllFMYNmwYALVr1+bjjz/myy+/5P777+fGG2/cZz8kSZIkHXkGfUWrsK+nYZ8kSZIkKY8WLVpQuXLlPGVt2rQhOjp3J4hmzZqRlZUF5P5SunnzZrKzs/npp58oVaoUlSpVytN27dq1lCpVivr16wO5weHo0aMBOOusszjhhBPynXdv/ZAkSZIk5eWefZIkSZKkA/LSSy9x9dVXA9CxY0fGjh1L9erV2bJlCwMGDMgX0J144olkZ2eTnp5Oamoqo0aNYunSpfnOO2TIEC666KLDMgZJkiRJRSOuz/giPV9m/4uL9HzHAmf2SZIkSZL228MPP0x0dDSdO3cGIC0tjRIlSrB8+XKWLFnCE088wbfffpunTRAEjBgxgt69e3PmmWdSsWJFSpQokafORx99xJAhQ3jkkUcO21gkSZIkaV8yMzN54403DqrtWWedVcS92TvDPkmSJEnSfhk6dCjvvvsur7/+emRPiTfeeIMLL7yQkiVLUq1aNc4++2zS09PztW3evDlTp04lLS2NFi1aRJb0BJgzZw49e/Zk7NixVKlS5bCNR5IkSZL2ZV9hX3Z29j7b/uc//zkUXSqQYZ8kSZIk6RdNmDCBRx99lHHjxlGuXLlIea1atZg8eTIAmzdv5vPPP6dBgwb52q9atQqAbdu28cgjj9CrVy8Avv/+ezp06MCrr76aJwCUJEmSpL3JzMykQYMGdO3alfr169O5c2c+/PBDzj77bOrVq0daWhppaWk0b96cxo0bc9ZZZ/H1118DkJOTw5133skZZ5xBcnIy//znP/d6nT59+jB16lRSUlIYMGAAQ4cOpV27dpx//vm0atWKTZs20apVK5o0aUJSUhJjx46NtK1QoQIAU6ZM4bzzzqNjx440aNCAzp07E4Zhkb4e7tknSZIkSfqfvsfRafQWpmTmsGZLSGylKP56Xmn6fbqNbTnQOulkAJrFlmDQJWX5w/aQbpsuICEhgTAM6datG8nJyQC0bduWwYMHExMTw2OPPca7777Lzp07uemmmzj//PMBePDBB1m7di2///3vAYiOjo7MDOzUqRNTpkxhzZo1xMbG8te//pUePXocgRdFkiRJ0q/N4sWLGTlyJC+99BJnnHEGb7zxBp9++injxo3j73//O6+88gpTp04lOjqaDz/8kHvuuYfRo0czZMgQjjvuOKZPn862bds4++yzadOmDbVr1853jf79+/P444/z7rvvArmrncycOZM5c+ZQuXJlsrOzeeedd6hUqRJr1qyhWbNmtGvXLrISym6zZs3iq6++IiYmhrPPPptp06ZxzjnnFNlrYdgnSZIkScpj+BXl8pX1aFKqwLoVSgWMHDmywGPvvfde5PFjjz3GY489lq/O4MGDGTx4cMH9GD58f7orSZIk6RhUu3ZtkpKSAEhISKBVq1YEQUBSUhKZmZls2LCBLl26sGjRIoIgYMeOHQBMnDiROXPmMGrUKAA2bNjAokWLCgz7CtK6dWsqV64MQBiG3HPPPXzyySdERUWxbNkyVq5cycknn5ynzZlnnklsbCwAKSkpZGZmGvZJkiRJkiRJkiTp2FW6dOnI46ioqMjzqKgosrOzuf/++2nZsiXvvPMOmZmZnHfeeUBuQPf0009zwQUXHNR1y5cvH3n8+uuvs3r1ambMmEHJkiWJi4tj69at++xriRIlfnG/vwNl2CdJkiRJkiRJkqSDktn/4iPdhQJt2LCBGjVqALnLb+52wQUX8Pzzz3P++edTsmRJFi5cSI0aNfKEeLtVrFiRjRs37vMa1apVo2TJknz00Ud89913RT6O/RF1RK4qSZIkSZIkSZIkHSJ33XUXf/7zn2ncuHGemXQ9e/akYcOGNGnShMTERH73u9/tdaZdcnIyJUqUoFGjRgwYMCDf8c6dO5Oenk5SUhKvvPIKDRo0OGTj2ZcgDMMjcuE9paamhrs3YJckSZIkHUF9jzuINhuKvh+SJEmSfpXmz59PfHz8ke7GUaeg1zUIghlhGKb+Ultn9kmSJEmSJEmSJEnFlHv2SZIkSZIkSZIk6Zj15Zdfct111+UpK126NF988cUR6tGBMeyTJEmSJEmSJEnSMSspKYmMjIwj3Y2D5jKekiRJkiRJkiRJUjFl2CdJkiRJkiRJkiQVU4Z9kiRJkiRJkiRJUjHlnn2SJEmSJEmSJEk6OH2PK+LzbSja8xVCZmYm//nPf7j22msPqv3f//537rnnniLuVX7O7JMkSZIkSZIkSZJ+JjMzkzfeeOOg2//9738vwt7snWGfJEmSJEmSJEmSio3MzEwaNGhA165dqV+/Pp07d+bDDz/k7LPPpl69eqSlpZGWlkbz5s1p3LgxZ511Fl9//TUAOTk53HnnnZxxxhkkJyfzz3/+c6/X6dOnD1OnTiUlJYUBAwbste2KFSto0aIFKSkpJCYmMnXqVPr06cNPP/1ESkoKnTt3PqSvh8t4SpIkSZIkSZIkqVhZvHgxI0eO5KWXXuKMM87gjTfe4NNPP2XcuHH8/e9/55VXXmHq1KlER0fz4Ycfcs899zB69GiGDBnCcccdx/Tp09m2bRtnn302bdq0oXbt2vmu0b9/fx5//HHeffddAF544YUC27799ttccMEF3HvvveTk5LBlyxZ+85vf8Mwzz5CRkXHIX4tfDPuCIHgJuARYFYZh4q6yx4BLge3AN0C3MAzX7zr2Z6AHkAPcHIbhvw9N1yVJkiRJkiRJknQsql27NklJSQAkJCTQqlUrgiAgKSmJzMxMNmzYQJcuXVi0aBFBELBjxw4AJk6cyJw5cxg1ahQAGzZsYNGiRQWGfT+3t7ZnnHEG3bt3Z8eOHVx++eWkpKQcmkHvxf4s4zkUuPBnZR8AiWEYJgMLgT8DBEHQELgGSNjV5rkgCEoUWW8lSZIkSZIkSZJ0zCtdunTkcVRUVOR5VFQU2dnZ3H///bRs2ZK5c+fyr3/9i61btwIQhiFPP/00GRkZZGRksGTJEtq0abNf19xb2xYtWvDJJ59Qo0YNunbtyiuvvFL0A96HXwz7wjD8BPjhZ2UTwzDM3vX0cyB21+PLgBFhGG4Lw3AJsBg4swj7K0mSJEmSJEmSJO3Thg0bqFGjBgBDhw6NlF9wwQU8//zzkZl+CxcuZPPmzQWeo2LFimzcuPEX23733XecdNJJ3HDDDfTs2ZOZM2cCULJkyUjdQ6ko9uzrDry563ENcsO/3bJ2lUmSJEmSJEmSJOlo03fDke5Bge666y66dOnCQw89xMUXXxwp79mzJ5mZmTRp0oQwDKlatSpjxowp8BzJycmUKFGCRo0a0bVrV2655ZYC206ZMoXHHnuMkiVLUqFChcjMvhtvvJHk5GSaNGnC66+/fsjGGoRh+MuVgiAOeHf3nn17lN8LpAIdwjAMgyB4Bvg8DMPXdh0fArwfhuGoAs55I3AjQK1atU7/7rvvCjsWSZIkSVIBunfvzrvvvku1atWYO3cuACNHjqRv377Mnz+ftLQ0UlNTcyv3PY45K3P43btb+XFbSFQA028oT5noIM8575+8lbFfZxMVQLXkVgwdOpSYmBjGjh3L/fffT1RUFNHR0Tz11FOcc845ZGRkcNNNN/Hjjz9SokQJ7r33Xq6++urD/VJIkiRJKqT58+cTHx9/pLtx1CnodQ2CYEYYhqm/1HZ/9uwrUBAEXYFLgM7h/xLDZUDNParF7irLJwzDF8IwTA3DMLVq1aoH2w1JkiRJ0i/o2rUrEyZMyFOWmJjI22+/TYsWLfKUZ+8M+e3bPzHo4jJ89fsKTOlSjpIF/OZ459mlmXNTBTJ6VeCSSy7hwQcfBKBVq1bMnj2bjIwMXnrpJXr27AlAuXLleOWVV/jqq6+YMGECt956K+vXrz8k45UkSZKkY8lBLeMZBMGFwF3AuWEYbtnj0DjgjSAIngRigHpAWqF7KUmSJEk6aC1atCAzMzNP2d7uxJ34TTbJJ5Wg0cklAKhSruB7RCuV/t9Mv82bNxMEuc8rVKhQYHn9+vUj5TExMVSrVo3Vq1dz/PHHH/B4JEmSJKkoffnll1x33XV5ykqXLs0XX3xxhHp0YH4x7AuCYDhwHnBiEARZwF+APwOlgQ92/eL2eRiGvcIw/CoIgreAeUA28IcwDHMOVeclSZIkSUVr4dqdBAFc8NpmVm8OuSaxJHedXbrAuvdO2sorc3ZwXM3X+eijjyLl77zzDn/+859ZtWoV48ePz9cuLS2N7du3U6dOnUM2DkmSJEnaX0lJSWRkZBzpbhy0X1zGMwzDTmEYVg/DsGQYhrFhGA4Jw7BuGIY1wzBM2fXVa4/6D4dhWCcMw9PCMHz/0HZfkiRJklSUsnfCp99n83qHsnzavTzvLMhm0rfZBdZ9uFUZlvauSOfOnXnmmWci5e3bt2fBggWMGTOG+++/P0+bFStWcN111/Hyyy8TFXXQO0tIkiRJOoL+t7ubikJhX09/s5IkSZIkRcRWiqLFKdGcWC6KciUD2taNZuaKfS/Y0rlzZ0aPHp2vvEWLFnz77besWbMGgB9//JGLL76Yhx9+mGbNmh2S/kuSJEk6tMqUKcPatWsN/IpIGIasXbuWMmXKHPQ5DmrPPkmSJEnS0emCOtE8Om0bW3aElCoBH3+XTe9mpfLVW7Q2h3pVcvf1Gzt2LA0aNABg8eLF1KlThyAImDlzJtu2baNKlSps376d9u3bc/3119OxY8fDOiZJkiRJRSc2NpasrCxWr159pLty1ChTpgyxsbEH3d6wT5IkSZKOcp06dWLKlCmsWbOG2NhY/vrXv1K5cmX+9Kc/sXr1ai6++GJSUlL497//zQllA25rXoozXtxMALStF83F9UsC0HPcT/RKLUVqTAn6TNrG12t2EhXAKWdOZNCgQQCMHj2aV155hZIlS1K2bFnefPNNgiDgrbfe4pNPPmHt2rUMHToUgKFDh5KSknJkXhRJkiRJB6VkyZLUrl37SHdDewh+DdMsU1NTw/T09CPdDUmSJEk66sT1GX9A9TPLXHvgF+m74cDbSJIkSZL2KQiCGWEYpv5SPffskyRJkiRJkiRJkoopwz5JkiRJkiRJkiSpmDLskyRJkiRJkiRJkoopwz5JkiRJkiRJkiSpmDLskyRJkiRJkiRJkoopwz5JkiRJkiRJkiSpmDLskyRJkiRJkiRJkoopwz5JkiRJkiRJkiSpmDLskyRJkiRJkiRJkoopwz5JkiRJkiRJkiSpmDLskyRJkiRJkiRJkoopwz5JkiRJkiRJkiSpmDLskyRJkiRJkiRJkoopwz5JkiRJkiRJkiSpmDLskyRJkiRJkiRJkoopwz5JkiRJkiRJkiSpmDLskyRJkiRJkiRJkoopwz5JkiRJkiRJkiSpmDLskyRJkiRJkiRJkoopwz5JkiRJkiRJkiSpmDLskyRJkiRJkiRJkoopwz5JkiRJkiRJkiSpmDLskyRJkiRJkiRJkoopwz5JkiRJkiRJkiSpmDLskyRJkiRJkiRJkoopwz5JkiRJkiRJkiSpmDLskyRJkiRJkiRJkoopwz5JkiRJkiRJkiSpmDLskyRJkiRJkiRJkoopwz5JkiRJkiRJkiSpmDLskyRJkiRJkiRJkoqpXwz7giB4KQiCVUEQzN2jrHIQBB8EQbBo158n7CoPgiD4RxAEi4MgmBMEQZND2XlJkiRJkiRJkiTpWLY/M/uGAhf+rKwPMCkMw3rApF3PAS4C6u36uhF4vmi6KUmSJEmSJEmSJOnnfjHsC8PwE+CHnxVfBgzb9XgYcPke5a+EuT4Hjg+CoHoR9VWSJEmSJEmSJEnSHg52z76TwjBcsevxf4GTdj2uASzdo17WrrJ8giC4MQiC9CAI0levXn2Q3ZAkSZIkSZIkSZKOXQcb9kWEYRgC4UG0eyEMw9QwDFOrVq1a2G5IkiRJkiRJkiRJx5yDDftW7l6ec9efq3aVLwNq7lEvdleZJEmSJEmSJEmSpCJ2sGHfOKDLrsddgLF7lF8f5GoGbNhjuU9JkiRJkiRJkiRJRSj6lyoEQTAcOA84MQiCLOAvQH/grSAIegDfAVftqv4e0BZYDGwBuh2CPkuSJEmSJEmSJEliP8K+MAw77eVQqwLqhsAfCtspSZIkSZIkSZIkSb/sYJfxlCRJkiRJkiRJknSEGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkiRJkiRJxZRhnyRJkiRJkiRJklRMGfZJkiRJkvQr9vXXX5OSkhL5qlSpEk899RQZGRk0a9aMlJQUUlNTSUtLy9f2u+++o0mTJqSkpJCQkMCgQYMix+69915q1qxJhQoV8rTp3bt35Fr169fn+OOPP9RDlCRJklQIQRiGB984CHoDPYEQ+BLoBlQHRgBVgBnAdWEYbt/XeVJTU8P09PSD7ockSZIkqWBxfcYfUP3MMtce+EX6bjjwNjooOTk51KhRgy+++IIbbriB3r17c9FFF/Hee+/x6KOPMmXKlDz1t2/fThiGlC5dmk2bNpGYmMh//vMfYmJi+PzzzznllFOoV68emzZtKvB6Tz/9NLNmzeKll146DKOTJEmStKcgCGaEYZj6S/UOemZfEAQ1gJuB1DAME4ESwDXAI8CAMAzrAuuAHgd7DUmSJEmS9D+TJk2iTp06nHLKKQRBwI8//gjAhg0biImJyVe/VKlSlC5dGoBt27axc+fOyLFmzZpRvXr1fV5v+PDhdOrUqQhHIEmSJKmoRRdB+7JBEOwAygErgPOB3beCDgP6As8X8jqSJEmSJB3zRowYEQnfnnrqKS644ALuuOMOdu7cyX/+858C2yxdupSLL76YxYsX89hjjxUYChbku+++Y8mSJZx//vlF1n9JkiRJRe+gZ/aFYbgMeBz4ntyQbwO5y3auD8Mwe1e1LKBGQe2DILgxCIL0IAjSV69efbDdkCRJkiTpmLB9+3bGjRvHlVdeCcDzzz/PgAEDWLp0KQMGDKBHj4IX1qlZsyZz5sxh8eLFDBs2jJUrV+7X9UaMGEHHjh0pUaJEkY1BkiRJUtErzDKeJwCXAbWBGKA8cOH+tg/D8IUwDFPDMEytWrXqwXZDkiRJkqRjwvvvv0+TJk046aSTABg2bBgdOnQA4MorryQtLW2f7WNiYkhMTGTq1Kn7db09ZxFKkiRJ+vU66LAP+D9gSRiGq8Mw3AG8DZwNHB8Ewe7lQWOBZYXsoyRJkiRJx7yf758XExPDxx9/DMDkyZOpV69evjZZWVn89NNPAKxbt45PP/2U00477RevtWDBAtatW0fz5s2LqPeSJEmSDpXChH3fA82CICgXBEEAtALmAR8BHXfV6QKMLVwXJUmSJEk6tm3evJkPPvggMpMP4MUXX+T222+nUaNG3HPPPbzwwgsApKen07NnTwDmz59P06ZNadSoEeeeey533HEHSUlJANx1113ExsayZcsWYmNj6du3b+TcI0aM4JprriH3131JkiRJv2ZBGIYH3zgI/gpcDWQDs4Ce5O7RNwKovKvst2EYbtvXeVJTU8P09PSD7ockSZIkqWBxfcYfUP3MMtce+EX6bjjwNto/fY87wPr+XUiSJElHiyAIZoRhmPpL9aJ/qcK+hGH4F+AvPyv+FjizMOeVJEmSJEmSJEmS9MsKs4ynJEmSJEmSJEmSpCPIsE+SJEmSJEmSJEkqpgz7JEmSJEmSJEmSpGLKsE+SJEmSJEmSJEkqpgz7JEmSJEmSJEmSpGLKsE+SJEmSJEmSJEkqpgz7JEmSJEmSJEmSpGLKsE+SJEmSJEmSJEkqpgz7JEmSJEmSJEmSpGLKsE+SJEmSJEmSJEkqpgz7JEmSJEmSJEmSpGLKsE+SJEmSJEmSJEkqpgz7JEmSJEmSJEmSpGLKsE+SJEmSJEmSJEkqpgz7JEmSJEmSJEmSpGLKsE+SJEmSJEmSJEkqpgz7JEmSJEnSIbd+/Xo6duxIgwYNiI+P57PPPgPg6aefpkGDBiQkJHDXXXfla7d06VJatmxJw4YNSUhIYODAgXmOF9Q+MzOTsmXLkpKSQkpKCr169Tr0A5QkSZKOkOgj3QFJkiRJknT0u+WWW7jwwgsZNWoU27dvZ8uWLXz00UeMHTuW2bNnU7p0aVatWpWvXXR0NE888QRNmjRh48aNnH766bRu3ZqGDRvus32dOnXIyMg4jCOUJEmSjgxn9kmSJEmSpENqw4YNfPLJJ/To0QOAUqVKcfzxx/P888/Tp08fSpcuDUC1atXyta1evTpNmjQBoGLFisTHx7Ns2TKA/WovSZIkHe0M+yRJkiRJ0iG1ZMkSqlatSrdu3WjcuDE9e/Zk8+bNLFy4kKlTp9K0aVPOPfdcpk+fvs/zZGZmMmvWLJo2bQqwz/ZLliyhcePGnHvuuUydOvWQjk+SJEk6klzGU5IkSZIkHVLZ2dnMnDmTp59+mqZNm3LLLbfQv39/srOz+eGHH/j888+ZPn06V111Fd9++y1BEOQ7x6ZNm7jiiit46qmnqFSpUuS8BbWvXr0633//PVWqVGHGjBlcfvnlfPXVV5F2kiRJ0tHEmX2SJEmSJOmQio2NJTY2NjIjr2PHjsycOZPY2Fg6dOhAEASceeaZREVFsWbNmnztd+zYwRVXXEHnzp3p0KFDnvMW1L506dJUqVIFgNNPP506deqwcOHCwzNYSZIk6TAz7JMkSZIkSYfUySefTM2aNfn6668BmDRpEg0bNuTyyy/no48+AnKX5Ny+fTsnnnhinrZhGNKjRw/i4+O57bbb8hzbW/vVq1eTk5MDwLfffsuiRYs49dRTD/UwJUmSpCPCZTwlSZIkSdIh9/TTT9O5c2e2b9/Oqaeeyssvv0z58uXp3r07iYmJlCpVimHDhhEEAcuXL6dnz5689957TJs2jVdffZWkpCRSUlIA+Pvf/07btm3p3r17ge0/+eQTHnjgAUqWLElUVBSDBg2icuXKR/YFkCRJkg6RIAzDI90HUlNTw/T09CPdDUmSJEk66sT1GX9A9TPLXHvgF+m74cDbaP/0Pe4A6/9K/y6OlnFIkiRJh1EQBDPCMEz9pXou4ylJkiRJkiRJkiQVU4Z9kiRJkiRJkiRJUjFl2CcV0vr16+nYsSMNGjQgPj6ezz77jJEjR5KQkEBUVBT7WqI2Li4usu9Eaur/ZuLOnj2b5s2bk5SUxKWXXsqPP/4IwNq1a2nZsiUVKlTgj3/84yEfmyRJkiRJkiRJ+nUz7JMK6ZZbbuHCCy9kwYIFzJ49m/j4eBITE3n77bdp0aLFL7b/6KOPyMjIyBMK9uzZk/79+/Pll1/Svn17HnvsMQDKlCnD3/72Nx5//PFDNh5JkiTpaFLQzXk//PADrVu3pl69erRu3Zp169YV2Pbuu+8mMTGRxMRE3nzzzUj5M888Q926dQmCgDVr1kTKN2zYwKWXXkqjRo1ISEjg5ZdfPuTjkyRJkiTDPqkQNmzYwCeffEKPHj0AKFWqFMcffzzx8fGcdtppB33ehQsXRoLC1q1bM3r0aADKly/POeecQ5kyZQrfeUmSJOkYUNDNef3796dVq1YsWrSIVq1a0b9//3ztxo8fz8yZM8nIyOCLL77g8ccfj6y4cfbZZ/Phhx9yyimn5Gnz7LPP0rBhQ2bPns2UKVO4/fbb2b59+2EZpyRJkqRjl2GfVAhLliyhatWqdOvWjcaNG9OzZ082b9683+2DIKBNmzacfvrpvPDCC5HyhIQExo4dC8DIkSNZunRpkfddkiRJOtrt7ea8sWPH0qVLFwC6dOnCmDFj8rWdN28eLVq0IDo6mvLly5OcnMyECRMAaNy4MXFxcfnaBEHAxo0bCcOQTZs2UblyZaKjo/PVK8xswwtf28zx/X/kkje25CnvOuYnag/cSEpKCikpKWRkZAAwduxYkpOTI1sHfPrpp/v78kmSJEkqJgz7pELIzs5m5syZ3HTTTcyaNYvy5csXeFfw3nz66afMnDmT999/n2effZZPPvkEgJdeeonnnnuO008/nY0bN1KqVKlDNQRJkiTpqLW3m/NWrlxJ9erVATj55JNZuXJlvraNGjViwoQJbNmyhTVr1vDRRx/94k14f/zjH5k/fz4xMTEkJSUxcOBAoqLy/9p9sLMNAe48qzSvti9b4LHHWpchIyODjIwMUlJSAGjVqhWzZ88mIyODl156iZ49e+5zDJIkSZKKH8M+qRBiY2OJjY2ladOmAHTs2JGZM2fud/saNWoAUK1aNdq3b09aWhoADRo0YOLEicyYMYNOnTpRp06dou+8JEmSdJTbn5vzgiAgCIJ8bdu0aUPbtm0566yz6NSpE82bN6dEiRL7vN6///1vUlJSWL58ORkZGfzxj3+MLP25W2FmGwK0OjWaiqXz93dvKlSoEBnf5s2bCxyrJEmSpOLNsE8qhJNPPpmaNWvy9ddfAzBp0iQaNmy4X203b97Mxo0bI48nTpxIYmIiAKtWrQJg586dPPTQQ/Tq1esQ9F6SJEk6uu3t5ryTTjqJFStWALBixQqqVatWYPt7772XjIwMPvjgA8IwpH79+vu83ssvv0yHDh0IgoC6detSu3ZtFixYkKdOYWYb/pJ7J28jOTmZ3r17s23btkj5O++8Q4MGDbj44ot56aWXDvi8kiRJkn7dChX2BUFwfBAEo4IgWBAEwfwgCJoHQVA5CIIPgiBYtOvPE4qqs9KvTt/jeDphFp1bNiT5pBJkjHiYe6L+yTtXlyO2UhSfffoxF593JhfUjYa+x7F8+XLatm0LwMqVKznnnHNo1KgRZ555JhdffDEXXnghAMOHD6d+/fo0aNCAmJgYunXrFrlkXFwct912G0OHDiU2NpZ58+YdkaFLkiRJv3Z7uzmvXbt2DBs2DIBhw4Zx2WWX5Wubk5PD2rVrAZgzZw5z5syhTZs2+7xerVq1mDRpEpD7fv/rr7/m1FNPzVOnMLMN96Vfq9Is+EN5pk+fzg8//MAjjzwSOda+fXsWLFjAmDFjuP/++w/ovJIkSZJ+/fLvFH5gBgITwjDsGARBKaAccA8wKQzD/kEQ9AH6AHcX8jrSr1bKySVIv7FCnrL28SVpH18yX92YmBjee+89AE499VRmz55d4DlvueUWbrnllgKPZWZmFq7DkiRJ0rGg73EAPJ2QQ+eWDdmeA6eeEMXLl5VlZxhy1Qs/MeSxeznluIC3riwHPEp6ejqDBg1i8ODB7Nixg9/85jcAVKpUiddee43o6Nxfof/xj3/w6KOP8t///pfk5GTatm3L4MGDuf/+++natStJSUmEYcgjjzzCiSeemKdbBc027N+/f2S2YfXq1fc523BvqlfMvZe3dOnSdOvWjccffzxfnRYtWvDtt9+yZs2afP2SJEmSVHwddNgXBMFxQAugK0AYhtuB7UEQXAact6vaMGAKhn2SJEmSpCOgoJvzIGDS9eXz1U1NTWXw4MEAlClTZq+raNx8883cfPPN+cpjYmKYOHHiPvuz52zD0047LTLbsGHDhgwbNow+ffrsdbbhvqzYuJPqFaMIw5AxY8ZEtghYvHgxderUIQgCZs6cybZt26hSpcoBnVuSJEnSr1thZvbVBlYDLwdB0AiYAdwCnBSG4Ypddf4LnFRQ4yAIbgRuhNylTiRJkiRJOprF9RkPwPa6V5F83iWEOdlEH38yVdreCuFO1gzqz/2P/oPoStU48bI+vNVnPKN+yGFQ+nYGtysLwG9e3syCNTvZtD0k9smNDGlXlgvqRtP57Z9YvSUkHJVESkoKgwYNAmD06NG88sorlCxZkrJly/Lmm28e8BKhkiRJkn7dgjAMD65hEKQCnwNnh2H4RRAEA4EfgT+FYXj8HvXWhWG4z337UlNTw/T09IPqh3RE7VoaaP/rbzg0/ZAkSZL2YnfAtL8yy1x74Bf5Nb7P/RW+Vz/Qvws4iL+PX+PfBfwq/z4kSZKkX7sgCGaEYZj6S/WiCnGNLCArDMMvdj0fBTQBVgZBUH1XJ6oDqwpxDUmSJEmSJEmSJEl7cdBhXxiG/wWWBkFw2q6iVsA8YBzQZVdZF2BsoXooSZIkSZIkSZIkqUCF2bMP4E/A60EQlAK+BbqRGyC+FQRBD+A74KpCXkOSJEmSJEmSJElSAQoV9oVhmAEUtFZoq8KcV5IkSZIkSZIkSdIvK8yefZIkSZIkSZIkSZKOIMM+SZIkSZIkSZIkqZgy7JMkSZIkSZIkSZKKKcM+SZIkSZIkSZIkqZgy7JMkSZIkSZIkSZKKKcM+6Wfi4uJISkoiJSWF1NRUAEaOHElCQgJRUVGkp6fvve1TG0l6fhMpgzaR+sKmSPn9k7eS/PwmUlJSaNOmDcuXL8/Tbvr06URHRzNq1KhDMyhJkiRJkiRJknRUMuyTCvDRRx+RkZERCfYSExN5++23adGixS+37VKOjF4VSL+xQqTszrNLM+emCmRkZHDJJZfw4IMPRo7l5ORw991306ZNm6IfiCRJkiRJkiRJOqpFH+kOSMVBfHx8odpXKh1EHm/evJkg+N/zp59+miuuuILp06cX6hqSJEmSJEmSJOnYY9gn/UwQBLRp04YgCPjd737HjTfeeABtoc2rWwgC+N3ppbjx9FKRY/dO2sorQ2py3HHH8dFHHwGwbNky3nnnHT766CPDPkmSJEmSJEmSdMBcxlP6mU8//ZSZM2fy/vvv8+yzz/LJJ5/sf9tu5Zn5uwq837kcz07fziffZUeOPdyqDEuXLqVz584888wzANx666088sgjREX5rShJkiRJkiRJkg6cCYP0MzVq1ACgWrVqtG/fnrS0tP1vWyn3W6pa+SjaN4gmbVlOvjqdO3dm9OjRAKSnp3PNNdcQFxfHqFGj+P3vf8+YMWMKPwhJkiRJkiRJknRMMOyT9rB582Y2btwYeTxx4kQSExP3r+32kI3bwsjjid/kkFitBACL1v4v9Bs7diwNGjQAYMmSJWRmZpKZmUnHjh157rnnuPzyy4twRJIkSZIkSZIk6Wjmnn3SLnF9xrNj/X9Z/fZDuQU7d1K+4bn0mpLDlhfu5YcP/knOTxtoem5rSlWrzUlX/43/7NhJz3Fbea9zOVZuDmn/5hYAsnfCtYklubBu7rdYn0nb+HrNTqLeTuaUU05h0KBBR2qYkiRJkiRJkiTpKGLYJ+2h5PEnE9P9mXzl5eqfRbn6Z+Urj6kYxXudywFw6glRzO5VocDzjr4qtw595+z12kOHDj3wDkuSJEmSJEmSpGOay3hKkiRJkiRJkiRJxZRhnyRJkiRJklSEcnJyaNy4MZdccgkAYRhy7733Ur9+feLj4/nHP/5RYLu7776bxMREEhMTefPNNyPlXbt2pXbt2qSkpJCSkkJGRgYACxYsoHnz5pQuXZrHH3/8kI9LkiT9Ohn2SZIkSfrVOJgPR7/77juaNGlCSkoKCQkJkf2RN27cGPlQNCUlhRNPPJFbb70VgE8++YQmTZoQHR3NqFGjDtv4pKPJwYYZJR78kZRBm0gZtIl2w7fkO37z+1upUOF/WyRs27aNq6++mrp169K0aVMyMzMPyXikojRw4EDi4+Mjz4cOHcrSpUtZsGAB8+fP55prrsnXZvz48cycOZOMjAy++OILHn/8cX788cfI8ccee4yMjAwyMjJISUkBoHLlyvzjH//gjjvuOORjkiRJv17u2SdJkiTpV2P3h6O7P9zc88PRqKgoVq1ala9N9erV+eyzzyhdujSbNm0iMTGRdu3aERMTE5n5AHD66afToUMHAGrVqsXQoUOdBSEVwsF8vwKUjYaMvex3nr48h3VbwzxlQ4YM4YQTTmDx4sWMGDGCu+++O8+MJ+nXJisri/Hjx3Pvvffy5JNPAvD888/zxhtvEBWVe999tWrV8rWbN28eLVq0IDo6mujoaJKTk5kwYQJXXXXVXq9VrVo1qlWrxvjx4w/NYCRJUrHgzD5JkiRJvwq7Pxzt2bNnpOz555/ngQce2OeHo6VKlaJ06dJA7gygnTt35quzcOFCVq1axW9+8xsA4uLiSE5OjpxX0oE52O/XfcnZGXLnB1t59P9K5ykfO3YsXbp0AaBjx45MmjSJMAwLOoX0q3Drrbfy6KOP5vkZ88033/Dmm2+SmprKRRddxKJFi/K1a9SoERMmTGDLli2sWbOGjz76iKVLl0aO33vvvSQnJ9O7d2+2bdt2WMYiSZKKB3+zlSRJkvSrcLAfjgIsXbqU5ORkatasyd13301MTEye4yNGjODqq68mCIJDOgbpWFGY79et2ZD6wiaaDd7MmAU7IuXPpG2nXf1oqlfM+1HFsmXLqFmzJgDR0dEcd9xxrF279hCMSiq8d999l2rVqnH66afnKd+2bRtlypQhPT2dG264ge7du+dr26ZNG9q2bctZZ51Fp06daN68OSVKlACgX79+LFiwgOnTp/PDDz/wyCOPHJbxSJKk4sGwT5IkSToK/HzvrN1uvvnmPHtf7Wn79u1069aNpKQkGjVqxJQpUyLHLrzwQho1akRCQgK9evUiJycHgB9++IHWrVtTr149Wrduzbp164qk/4X5cBSgZs2azJkzh8WLFzNs2DBWrlyZ5/iIESPo1KlTkfRVOtYV9vv1u1srkH5jBd64oiy3TtjKNz/sZPnGnYycl82fmpY6HEOQDplp06Yxbtw44uLiuOaaa5g8eTK//e1viY2NjSwl3b59e+bMmZOn3e6f45999hkZGRl88MEHhGHI5MmTqVChAtWrVycIAkqXLk23bt1IS0sjMzOTsmXLkpKSwqBBgxg9enTkfDNmzCApKYm6dety8803R2bD9u3blxo1akT2s33vvfcO34sjSZIOGcM+SZIk6Siwe++sPaWnp+8zjHvxxRcB+PLLL/nggw+4/fbbI0tgvvXWW8yePZu5c+eyevVqRo4cCUD//v1p1aoVixYtolWrVvTv379I+n+wH47+XExMDImJiUydOjVSNnv2bLKzs/MFE4fSwYSvaWlpkQ9fGzVqxDvvvBM5tn79ejp27EiDBg2Ij4/ns88+A/zQVkdGYb9fa1TK/Sji1BOiOC8umln/zWHWihwW/7CTuv/YRNxTG9myZQt169bNrV+jRmQpw+zsbDZs2ECVKlUOw0ilA9evXz+ysrLIzMxkxIgRnH/++bz22mtcfvnlfPTRRwB8/PHH1K9fP0+7gQMH0qBBA7Zv3w7AnDlz+OKLL6hYsSIAK1asACAMQ8aMGUNiYiIAderUISMjg169enHFFVdEznfTTTfx4osvsmjRIhYtWsSECRMix3r37k1GRgYZGRm0bdv20L0YkiTpsDHskyRJkoq5gvbOysnJ4c477+TRRx/da7t58+Zx/vnnA7l7ax1//PGkp6cDUKlSJSD3g/Xt27dHlr/cc++sLl26MGbMmCIZw8F+OAq54//pp58AWLduHZ9++imnnXZa5Pjw4cMP+6y+gwlfExMTSU9PJyMjgwkTJvC73/2O7OxsAG655RYuvPBCFixYwOzZs/Oc2w9tdbgV5vt13U8h27JzZxit2bKTaUtzaFg1iovrl+S/d1Qk89bcr3LlyrF48WIA2rVrx7BhwwAYNWoU559/vkvy6lcprs/4PF/XvPAZkxesIq7PeN7Y2pg/Pvw8parGcUHnXixLup64PuNJT0+nU6dOjB8/ni5duvDZZ5/RsGFDbrjhBipXrszjjz8OQOfOnUlKSiIpKYk1a9Zw3333Abk/p2NjY3nyySd56KGHiI2NZeHChfz44480a9aMIAi4/vrri+zntSRJ+nWKPtIdkCRJklQ4u/fO2rhxY6TsmWeeoV27dlSvXn2v7Ro1asS4cePo1KkTS5cuZcaMGSxdupQzzzwTgAsuuIC0tDQuuugiOnbsCMDKlSsj5zz55JPzLZdZ1Pr06UPnzp0ZMGAAFSpUYPDgwUBucDZo0CAGDx7M/Pnzuf322wmCgDAMueOOO0hKSoqc46233so342369Om0b9+edevW8a9//Yu//OUvfPXVV0XS593h67333suTTz4J/C98feONN/LM2NtTuXLlIo+3bt0aCTM2bNjAJ598wtChQwEoVaoUpUq51KGOnLg+4yOPt34/hx93hRk7tzZmzcOP0+vuBwlKlaHKBX8grs94tq1YxMVf/sTgdmWZvyaH3727lagAdobQ5+xSNKxaYp/X69GjB9dddx1169alcuXKjBgx4lAPUSoSZWolU6ZWMgBRZSpQ7cq++eqkpqayY8eOyM/xc889l3fffZeBAweyc+fOyM/cyZMn52u7Zs0ali5dSv369alTpw4PPfQQv/nNb0hPTyc2NjZSLzY2lmXLlkWeP/PMM7zyyiukpqbyxBNPcMIJJxTxyCVJ0uFm2CdJkiQVY3vunbV7z73ly5czcuTIPHvwFaR79+7Mnz+f1NRUTjnlFM466yxKlPjfh+7//ve/2bp1K507d2by5Mm0bt06T/sgCIpmdk3f4/I8PQ84LzW3/Hhg/BnAGbsOvtMC3oHUvhsiwV/r1q33ubznt99+m6/sjDPOICsrq/B9L8DBhq8AX3zxBd27d+e7777j1VdfJTo6miVLllC1alW6devG7NmzOf300xk4cCDly5ePnNsPbXWk7E+YUbp6PQbXLgvAWTWj+fKmgpey3dOmTZv+d40yZSJLCUtHm8L8HK9evTrff/89VapUYcaMGVx++eW/eOPKTTfdxP33308QBNx///3cfvvtvPTSS0U0GkmSdKS4jKckSZJUjBW0d1ZCQgKLFy+mbt26xMXF5dn7ak/R0dEMGDCAjIwMxo4dy/r16/Mtu1emTBkuu+wyxo4dC8BJJ50U2TdoxYoVVKtW7dAPshjZ80Pb3XZ/aPunP/3pF9s3bdqUr776iunTp9OvXz+2bt1KdnY2M2fO5KabbmLWrFmUL18+slfiTTfdxDfffENGRgbVq1fn9ttvP2RjkyQVvcL8HC9dunRk/8rTTz+dOnXqsHDhQmrUqJHnhpasrCxq1KgB5P4cL1GiBFFRUdxwww2kpaUdnoFKkqRDyrBPkiRJKsYK2jtr3bp1/Pe//yUzM5PMzMw8e1/tacuWLWzevBmADz74gOjoaBo2bMimTZsigV52djbjx4+nQYMGQN69s4YNG8Zll112mEZaPBTmQ9s9xcfHU6FCBebOnUtsbCyxsbE0bdoUgI4dOzJz5kzAD20lqbgrzM/x1atXk5OTA+TOYl+0aBGnnnoq1atXp1KlSnz++eeEYcgrr7wS+Xm9++c7wDvvvENiYuLhGagkSTqkXMZTkiRJKs72XAIzMxsWbs+3LCbbN0fKxjV5lfT0dB588EFWrVrFBRdcQFRUFDVq1ODVV18FYPPmzbRr145t27axc+dOWrZsSa9evYDcPfSuuuoqhgwZwimnnMJbb711WIZZXPTr149+/foBMGXKFB5//HHefffdPHUqVKhQ4Ie2S5YsoWbNmkRHR/Pdd9+xYMEC4uLiOPHEE6lZsyZff/01p512GpMmTaJhw4ZA7oe2u5cG9UNbSTr6jRs3LvJz/JNPPuGBBx6gZMmSREVFMWjQICpXrgzAc889R9euXfnpp5+46KKLuOiiiwC46667yMjIIAgC4uLi+Oc//3kkhyNJkoqIYZ8kSZJ0lDgvLprz4vK/xd90T6XI43bt2tGuXTsA4uLi+Prrr/PVP+mkk5g+fXqB16hSpQqTJk0qoh4fZX4esh5g+Prpp5/Sv3//yIe2zz33HCeeeCIATz/9NJ07d2b79u2ceuqpvPzyy4Af2kpSsbfHz4jz+N+etXvadEeJSFm7vhsiP8evuOIKrrjiigJPm5qayty5c/OV776xR5IkHV0M+yRJkiTpEDjQ8PW6667juuuuK/BcKSkppKen5yv3Q1tJkiRJknv2SYrIycmhcePGXHLJJUDuUlJNmzalbt26XH311Wzfvj1fmw8++IDTTz+dpKQkTj/9dCZPnhw5duGFF9KoUSMSEhLo1atXZC+Bvn37UqNGDVJSUkhJSeG99947PAOUJEmSJEmSJOkoY9gnKWLgwIHEx8dHnt9999307t2bxYsXc8IJJzBkyJB8bU488UT+9a9/8eWXXzJs2LA8d6O/9dZbzJ49m7lz57J69WpGjhwZOda7d28yMjLIyMigbdu2h3ZgkiRJkiRJkiQdpQz7JAGQlZXF+PHj6dmzJwBhGDJ58mQ6duwIQJcuXRgzZky+do0bNyYmJgaAhIQEfvrpJ7Zt2wZApUq5S1RlZ2ezfft2giA4DCORJEmSJEmSJOnYYdgnCYBbb72VRx99lKio3P8W1q5dy/HHH090dO4+M7GxsSxbtmyf5xg9ejRNmjShdOnSkbILLriAatWqUbFixUhwCPDMM8+QnJxM9+7dWbdu3SEYkSRJkiRJkiRJR79Ch31BEJQIgmBWEATv7npeOwiCL4IgWBwEwZtBEJQqfDclHUrvvvsu1apV4/TTTz/oc3z11Vfcfffd/POf/8xT/u9//5sVK1awbdu2yH5+N910E9988w0ZGRlUr16d22+/vVD9lyRJkiRJkiTpWFUUM/tuAebv8fwRYEAYhnWBdUCPIriGpENo2rRpjBs3jri4OK655homT57MLbfcwvr168nOzgZyl/msUaNGge2zsrJo3749r7zyCnXq1Ml3vEyZMlx22WWMHTsWgJNOOokSJUoQFRXFDTfcQFpa2qEbnCRJkiRJkiRJR7FChX1BEMQCFwODdz0PgPOBUbuqDAMuL8w1JB16/fr1Iysri8zMTEaMGMH555/P66+/TsuWLRk1KvfbediwYVx22WX52q5fv56LL76Y/v37c/bZZ0fKN23axIoVK4DcPfvGjx9PgwYNACLlAO+88w6JiYmHcniSJEmSJEmSJB21ogvZ/ingLqDirudVgPVhGGbvep4FFDgVKAiCG4EbAWrVqlXIbkg6aH2Py/s8MxsWboe+x/FIzE6uuf1t7vv9tTSuXoIeJ5aFvn9mXJNXSU9P58EHH+SZZ55h8eLFPPjggzz44IMATJw4kTAMadeuHdu2bWPnzp20bNmSXr16AXDXXXeRkZFBEATExcXlW/pTkiRJkiRJkiTtn4MO+4IguARYFYbhjCAIzjvQ9mEYvgC8AJCamhoebD8kFa3z4qI5Ly73v4ZTT4gi7YYK+eq0a9eOdu3aAXDfffdx3333FXiu6dOnF1j+6quvFlFv89q6dSstWrRg27ZtZGdn07FjR/76178ShiH33XcfI0eOpESJEtx0003cfPPNBZ7jxx9/pGHDhlx++eU888wzALz55ps8/PDD5OTkcMkll/DII48A8P3339OlSxfWr19PTk4O/fv3p23btodkbJIkSZIkSZIkFaQwM/vOBtoFQdAWKANUAgYCxwdBEL1rdl8ssKzw3ZSkX1a6dGkmT55MhQoV2LFjB+eccw4XXXQR8+fPZ+nSpSxYsICoqChWrVq113Pcf//9tGjRIvJ87dq13HnnncyYMYOqVavSpUsXJk2aRKtWrXjooYe46qqruOmmm5g3bx5t27YlMzPzMIxUkiRJkiRJkqRcB71nXxiGfw7DMDYMwzjgGmByGIadgY+AjruqdQHGFrqXkrQfgiCgQoXcmYg7duxgx44dBEHA888/zwMPPEBUVO5/edWqVSuw/YwZM1i5ciVt2rSJlH377bfUq1ePqlWrAvB///d/jB49OnK9H3/8EYANGzYQExNzyMYmSZIkSZIkSVJBDjrs24e7gduCIFhM7h5+Qw7BNSSpQDk5OaSkpFCtWjVat25N06ZN+eabb3jzzTdJTU3loosuYtGiRfna7dy5k9tvv53HH388T3ndunX5+uuvyczMJDs7mzFjxrB06VIA+vbty2uvvUZsbCxt27bl6aefPixjlCRJkiRJkiRptyIJ+8IwnBKG4SW7Hn8bhuGZYRjWDcPwyjAMtxXFNSRpf5QoUYKMjAyysrJIS0tj7ty5bNu2jTJlypCens4NN9xA9+7d87V77rnnaNu2LbGxsXnKTzjhBJ5//nmuvvpqfvOb3xAXF0eJEiUAGD58OF27diUrK4v33nuP6667jp07dx6WcUqSJEmSJEmSBIXbs0+SfrWOP/54WrZsyYQJE4iNjaVDhw4AtG/fnm7duuWr/9lnnzF16lSee+45Nm3axPbt26lQoQL9+/fn0ksv5dJLLwXghRdeiIR9Q4YMYcKECQA0b96crVu3smbNmr0uEypJkiRJkiRJUlE7FMt4StIRsXr1atavXw/ATz/9xAcffECDBg24/PLL+eijjwD4+OOPqV+/fr62r7/+Ot9//z2ZmZk8/vjjXH/99fTv3x+AVatWAbBu3Tqee+45evbsCUCtWrWYNGkSAPPnz2fr1q2Rvf0kSZIkSZIkSTocnNkn6ejQ9zhWrMyhy5ifyNkJO0O4KqEkl6R35hxCOvf/iQF/voEKpQIGX1wG+h5H+iWTGDRoEIMHD97nqW+55RZmz54NwAMPPBAJC5944gluuOEGBgwYQBAEDB06lCAIDvlQJUmSJEmSJEnazbBP0lEj+aQSzPpdhXzlx5cJGH9tuXzlqampBQZ9Xbt2pWvXrpHnw4cPL/B6DRs2ZNq0aQffYUmSJEmSJEmSCsllPCVJkiRJkiRJkqRiyrBPkiRJkiRJ0lFn69atnHnmmTRq1IiEhAT+8pe/APDMM89Qt25dgiBgzZo1e21fokQJUlJSSElJoV27dpHyyZMn06RJExITE+nSpQvZ2dkAPPbYY5H6iYmJlChRgh9++OHQDlKSJAz7dIQU5s1WRkYGzZs3JyEhgeTkZN58883IsSVLltC0aVPq1q3L1Vdfzfbt2yPH3nrrLRo2bEhCQgLXXnvtoR2gJEmSJEmSjqjSpUszefJkZs+eTUZGBhMmTODzzz/n7LPP5sMPP+SUU07ZZ/uyZcuSkZFBRkYG48aNA2Dnzp106dKFESNGMHfuXE455RSGDRsGwJ133hmp369fP84991wqV658yMcpSZJhn46IwrzZKleuHK+88gpfffUVEyZM4NZbb2X9+vUA3H333fTu3ZvFixdzwgknMGTIEAAWLVpEv379mDZtGl999RVPPfXUYRilJEmSJEmSjpQgCKhQoQIAO3bsYMeOHQRBQOPGjYmLi8tX/+c3p+++iXzPm8s7dOhAyZIlqV+/PgCtW7dm9OjR7Nixgy5dupCUlER8fDz33XcfnTp1AmDChAmcdtpp1K1bl/79+0eu17VrV2rXrh2ZDZiRkXFoXxBJ0lHLsE9HxIG+2dpT/fr1qVevHgAxMTFUq1aN1atXE4YhkydPpmPHjgB06dKFMWPGAPDiiy/yhz/8gRNOOAGAatWqHZqBSZIkSZIk6VcjJyeHlJQUqlWrRuvWrWnatOle6/785vQdO3bQoEEDUlNTOeecc1i8eDEnnXQS69evJz09HYBRo0axdOlSRo4cybZt2/jyyy+ZOnUqc+fO5YwzziAnJ4c//OEPvP/++8ybN4/hw4czb968yDUfe+yxyGzAlJSUQ/1ySJKOUoZ9OmIO5M3W3qSlpbF9+3bq1KnD2rVrOf7444mOjgYgNjaWZcuWAbBw4UIWLlzI2WefTbNmzZgwYUKRjkUqSkuXLqVly5aRZWcHDhwIwOzZs2nevDlJSUlceuml/PjjjwW2HzhwIImJiSQkJOSZxTpy5EgSEhKIioqK/FICsH37drp160ZSUhKNGjViypQph3J4kiRJkiQdNiVKlCAjI4OsrCzS0tKYO3fuXuv+/Ob0xMREhg0bRhiGjBo1im+++YauXbtSr149evfuzZlnnknFihUpUaIEQRCwefNmsrOzGTNmDGXLlqVWrVqkpaVRt25dTj31VEqVKsU111zD2LFjD9fwJUnHCMM+HTEH8marICtWrOC6667j5ZdfJipq3/+Us7OzWbRoEVOmTGH48OHccMMNkaU/pV+b6OhonnjiCebNm8fnn3/Os88+y7x58+jZsyf9+/fnyy+/pH379jz22GP52s6dO5cXX3yRtLQ0Zs+ezbvvvsvixYsBSExM5O2336ZFixZ52rz44osAfPnll3zwwQfcfvvt7Ny589APVJIkSZKkw+T444+nZcuWv3gD+J43p7dt25Y6depQuXJlWrZsyaxZs4iNjWXz5s1MnTqVtLQ0WrRoQf369enYsSPly5enevXq9OrVi2uuuYbKlSuzbNkyatasGTn/njenA9x7770kJyfTu3dvtm3bdsjGL0k6uhn26Yjb3zdbe/rxxx+5+OKLefjhh2nWrBkAVapUYf369WRnZwOQlZVFjRo1gNw3Uu3ataNkyZLUrl2b+vXrs2jRoqIfjFQEqlevTpMmTQCoWLEi8fHxLFu2jIULF0aCut17Avzc/Pnzadq0KeXKlSM6Oppzzz2Xt99+G4D4+HhOO+20fG3mzZvH+eefD+QucXv88cfnmfknSZIkSVJxtHr16sjN3j/99BMffPABDRo02Geb3Tenz507l88//5wFCxaQk5PDtGnTaNiwIUDks6dt27bxyCOP0KtXL9LS0ihRogTz58+nQoUKfPLJJ3z77bf7vFa/fv1YsGAB06dP54cffuCRRx4p/KAlScckwz4dEQfzZmu37du30759e66//vrI/nyQu9RCy5YtGTVqFADDhg3jsssuA+Dyyy+PLE24Zs0aFi5cyKmnnlp0A5IOkczMTGbNmkXTpk1JSEiILPUxcuRIli5dmq9+YmIiU6dOZe3atWzZsoX33nuvwHp7atSoEePGjSM7O5slS5YwY8aMX2wjSZIkSdKv3YoVK2jZsiXJycmcccYZtG7dmksuuYR//OMfxMbGkpWVRXJyMj179gQgPT098nj58uUsWLCAK6+8kqVLl3LnnXfSsGFDsrKy2LZtG/Hx8SQnJ3PppZdy/vnn88Ybb3DhhRfy7rvvcuGFF3LOOeeQnp5OjRo18vyOvefN6dWrVycIAkqXLk23bt1IS0s7/C+SJOmoEH2kO6Bj04oHatNlzE/k7ISdIVyVUJJL0jvzj2e38ei07fx3U0hy7Wq0rRfN4HZlSb9kEoMGDWLw4MG89dZbfPLJJ6xdu5ahQ4cCMHToUFJSUnjkkUe45ppruO+++2jcuDE9evQA4IILLmDixIk0bNiQEiVK8Nhjj1GlSpUj+ApIv2zTpk1cccUVPPXUU1SqVImXXnqJm2++mb/97W+0a9eOUqVK5WsTHx/P3XffTZs2bShfvjwpKSmUKFFin9fp3r078+fPJzU1lVNOOYWzzjrrF9tIkiRJkvRrFtdnfO6DCx6KlL20BV7qMx6oQ/Rv/0nsrvIPd/15yimn8PjjjwPQuHFj6tevz913382wYcOoVKkSkHtz+Z133snvf//7PNerVasWkydP5uWXX+bKK6/kjDPO4NZbb6Vhw4YsWrSIJUuWUKNGDUaMGMEbb7wB5IaR1atXJwxDxowZQ2Ji4qF6OSRJRznDPh0RySeVYNbvKuQrv7lpaW5uWjpfeWpqKoMHDwbgt7/9Lb/97W8LPO+pp55a4F1QQRDw5JNP8uSTTxay59LhsWPHDq644go6d+5Mhw4dAGjQoAETJ04EYOHChYwfP77Atj169IgE3ffccw+xsbEF1tstOjqaAQMGRJ6fddZZ1K9fvyiGIUmSJElS8dD3OFaszCnw5vSGMTu55va3ue/319K4egl6nFgW+v6ZcU1eJT09nQcffJA//OEPdOvWjYSEBMIwpFu3biQnJwPwzDPPcMEFF5CTk0P37t1JSEgAoHPnzqxevZowDElJSWHQoEFH8hWQJBVjhn2S9CsThiE9evQgPj6e2267LVK+atUqqlWrxs6dO3nooYfo1atXge131/v+++95++23+fzzz/d5vS1bthCGIeXLl+eDDz4gOjo6sg+BJEmSJEnHir3dnH7qCVGk3ZC/vF27drRr1w6AChUqMHLkyALP27ZtW9q2bZuvfPLkyYXssSRJuQz7JOnXpO9xTPs+m1df3UJStShSRjwNwN9blWbR2p08O30HAB3io+lWewzQneXLl9OzZ0/ee+89AK644grWrl1LyZIlefbZZzn++OMBeOedd/jTn/7E6tWrufjii0lJSeHf//43q1at4oILLiAqKooaNWrw6quvHoGBS5IkSZIkSZIOhmGfJP3KnFMrmvAvlfIfqAe3NMu/zG1MTEwk6AOYOnVqgedt37497du3z1ceFxfH119/ffAdliRJkiRJkiQdMVFHugM6cEuXLqVly5Y0bNiQhIQEBg4cCEDfvn2pUaMGKSkppKSk5Pnwf0/r16+nY8eONGjQgPj4eD777DMAfvjhB1q3bk29evVo3bo169atA3KXFLz55pupW7cuycnJzJw58/AMVJIkSZIkSZIkSftk2FcMRUdH88QTTzBv3jw+//xznn32WebNmwdA7969ycjIICMjo8C1wAFuueUWLrzwQhYsWMDs2bOJj48HoH///rRq1YpFixbRqlUr+vfvD8D777/PokWLWLRoES+88AI33XTT4RmoJEmSJEmSJEmS9smwrxiqXr06TZo0AaBixYrEx8ezbNmy/Wq7YcMGPvnkE3r06AFAqVKlIvt5jR07li5dugDQpUsXxowZEym//vrrCYKAZs2asX79elasWFG0g5IKsLdZrLs98cQTBEHAmjVrCmx/1wdbSXhuE/HPbuLm97cShiFbdoRc/MYWGjyziYSEBPr06ROpP3ToUKpWrRqZHTt48OBDOj5JkiRJkiRJkgrrmAr7ChMcfPfddzRp0oSUlBQSEhIYNGhQ5Nh5553HaaedFgkIVq1aBcCTTz5Jw4YNSU5OplWrVnz33XdFPqbMzExmzZpF06ZNAXjmmWdITk6me/fukWU497RkyRKqVq1Kt27daNy4MT179mTz5s0ArFy5kurVqwNw8skns3LlSgCWLVtGzZo1I+eIjY3d73BRKox9zWJdunQpEydOpFatWgW2/c/SbKYtzWFOr/LMvak805fn8PF3OQDc0bwUC/5YgVmzZjFt2jTef//9SLurr746Mju2Z8+eh36QkiRJkiRJkiQVwjEV9hUmOKhevTqfffYZGRkZfPHFF/Tv35/ly5dHjr/++uuRgKBatWoANG7cmPT0dObMmUPHjh256667inQ8mzZt4oorruCpp56iUqVK3HTTTXzzzTdkZGRQvXp1br/99nxtsrOzmTlzJjfddBOzZs2ifPnykeU69xQEAUEQFGl/pQO1r1msvXv35tFHH93rv9MA2Jodsj0HtuXAjpyQk8oHlCsZ0LJ2NJA7s7VJkyZkZWUdlvFIkiRJkiRJklTUjqmwrzDBQalSpShdujQA27ZtY+fOnb94vZYtW1KuXDkAmjVrVqSBwo4dO7jiiivo3LkzHTp0AOCkk06iRIkSREVFccMNN5CWlpavXWxsLLGxsZGZgB07dmTmzJmR9ruX51yxYkUktKxRowZLly6NnCMrK4saNWoU2Vik/bHnLNaxY8dSo0YNGjVqtNf6zWtG0zIumupPbKT6Exu5oE408VVL5Kmzfv16/vWvf9GqVatI2ejRo0lOTqZjx455/t1LkiRJkiRJkvRrdEyFfXs60OAAcmf/JScnU7NmTe6++25iYmIix7p160ZKSgp/+9vfCMMwX9shQ4Zw0UUXFUnfwzCkR48exMfHc9ttt0XK99xH75133iExMTFf25NPPpmaNWvy9ddfAzBp0iQaNmwIQLt27Rg2bBgAw4YN47LLLouUv/LKK4RhyOeff85xxx0XWe5TOhz2nMUaHR3N3//+dx588MF9tln8w07mr9lJ1m0VWXZbRSZn5jD1u+zI8eydIZ06deLmm2/m1FNPBeDSSy8lMzOTOXPm0Lp168gelpIkSZIkSZIk/VpFH+kOHAkFBQcTJ078xXY1a9Zkzpw5LF++nMsvv5yOHTty0kkn8frrr1OjRg02btzIFVdcwauvvsr1118faffaa6+Rnp7Oxx9/XCT9nzZtGq+++ipJSUmkpKQA8Pe//53hw4eTkZFBEATExcXxz3/+E4Dly5fTs2dP3nvvPQCefvppOnfuzPbt2zn11FN5+eWXAejTpw9XXXUVQ4YM4ZRTTuGtt94CoG3btrz33nvUrVuXcuXKRepLh8PPZ7F++eWXLFmyJBLOZ2Vl0aRJE9Ku2snJFf53/8I783fQrEYJKpTKna17Ud1oPsvK4Ten5P63d+O/tlLvN/W49dZbI22qVKkSedyzZ88iX3pXkiRJkiRJkqSidsyFffsdHKSlcfLJJxd4jpiYGBITE5k6dSodO3aMLGlZsWJFrr32WtLS0iJh34cffsjDDz/Mxx9/HFkGtFD6Hsc5QPiXSsB3/ytP60TbOkCd3QXfwT8bQN8NxMTERII+gJSUFNLT0/OdukqVKkyaNClfeRAEPPvss4Xvu3SACprFmpSUxKpVqyJ14uLiSE9P58Rn6uRpW+u4KF6cuZ0/7yxFGMLH32Vza9NSANw3eSsbtoUMfuqpPG1WrFgRmbU6btw44uPjD+HoJEmSJEmSJEkqvGMq7Dug4ODEE/O0zcrKokqVKpQtW5Z169bx6aef0rt3b7Kzs1m/fj0nnngiO3bs4N133+X//u//AJg1axa/+93vmDBhQmT/O0n7J67PeLZmfcXK11+lZNU4nhs+DoATWlxP2TpnROplrfuJxg9OZGllSF+ew6D07QxuV5aODaOZvCSbpOc3EwAX1o3m0tNKkvXjTh6eup0GJ0ZF9vD84x//SM+ePfnHP/7BuHHjiI6OpnLlygwdOvQIjFySJEmSJEmSpP137IR9fY9j2vfZvPrqFpKqRZEy4mkA/t6qNG3rlfxfvfUb4dHaUC6K9EsmMWjQIAYPHsz8+fO5/fbbCYKAMAy54447SEpKYvPmzVxwwQXs2LGDnJwc/u///o8bbrgBgDvvvJNNmzZx5ZVXAlCrVi3GjRt32IcuFVdlYhM45e5391kn9qaXIo9TY0owuF1ZAEpEBfzz0rL561eK2jUzFuibkedYv3796NevX+E6LUmSJEmSJEnSYXTshH3AObWi//ch/15k3lox8jg1NZXBgwcD0Lp1a+bMmZOvfvny5ZkxY0aB5/rwww8L0VtJkiRJkiRJkiRp36KOdAckSZIkSZIkSZIkHRzDPkmSJEmSJEmSJKmYMuyTJEmSJElSsdO9e3eqVatGYmJipGz27Nk0b96cpKQkLr30Un788ce9ts/JyaFx48ZccsklkbLf/OY3pKSkkJKSQkxMDJdffjkAr7/+OsnJySQlJXHWWWcxe/bsQzYuSZKkA2XYJ0mSJEmSpGKna9euTJgwIU9Zz5496d+/P19++SXt27fnscce22v7gQMHEh8fn6ds6tSpZGRkkJGRQfPmzenQoQMAtWvX5uOPP+bLL7/k/vvv58Ybbyz6AUmSJB2kgw77giCoGQTBR0EQzAuC4KsgCG7ZVV45CIIPgiBYtOvPE4quu5Ik6VhV0J3bGRkZNGvWjJSUFFJTU0lLSyuw7YUXXsjxxx+f565tyP2AqHbt2pG7tzMyMgDYsGEDl156KY0aNSIhIYGXX375kI1LkiRJB6dFixZUrlw5T9nChQtp0aIFAK1bt2b06NEFts3KymL8+PH07NmzwOM//vgjkydPjszsO+usszjhhNyPuJo1a0ZWVlYRjUKSJKnwCjOzLxu4PQzDhkAz4A9BEDQE+gCTwjCsB0za9VySJCmPAw3v9rxze9iwYdSrV4+zzjqLs88+m4yMDB588EHuuusuANq1a5fnvB06dODkk0/mk08+ybec02OPPRa5ezslJQWAZ599loYNGzJ79mymTJnC7bffzvbt2w/lyyFJkqQikJCQwNixYwEYOXIkS5cuLbDerbfeyqOPPkpUVMEfjY0ZM4ZWrVpRqVKlfMeGDBnCRRddVHSdPkoU5ua83e/v69Wrx7BhwyLl9957LzVr1qRChQoFths9ejRBEJCenl60g5EkqZg56LAvDMMVYRjO3PV4IzAfqAFcBuz+qTwMuLyQfZQkSUehgpZduuuuu/jLX/6SL7yD/925nZOTw1//+le++OILmjVrxquvvsq6devYsGEDMTExvP322/k+DHjxxRfp1asXLVq0+MXlnACCIGDjxo2EYcimTZuoXLky0dHRRTd4SZIkHRIvvfQSzz33HKeffjobN26kVKlS+eq8++67VKtWjdNPP32v5xk+fDidOnXKV/7RRx8xZMgQHnnkkSLt99HgQN/f7/bDDz9E3t+npaXx17/+lXXr1gFw6aWX7jUg3LhxIwMHDqRp06ZFPxhJkoqZItmzLwiCOKAx8AVwUhiGK3Yd+i9wUlFcQ5IkHV0KWnYpCILIrLvd4d3Pbdq0idatW1O5cmWeffZZtmzZQt26dbnjjju47777ePLJJ7nvvvvytFm4cCGNGjUC8i/ndO+995KcnEzv3r3Ztm0bAH/84x+ZP38+MTExJCUlMXDgwL3e9S1JkqRfjwYNGjBx4kRmzJhBp06dqFOnTr4606ZNY9y4ccTFxXHNNdcwefJkfvvb30aOr1mzhrS0NC6++OI87ebMmUPPnj0ZO3YsVapUOeRjKW4O9v39v//978j7+xNOOIHWrVtHQsNmzZpRvXr1Aq93//33c/fdd1OmTJkiHokkScVPoT+1CoKgAjAauDUMwx/3PBaGYQiEe2l3YxAE6UEQpK9evbqw3ZAkSUeBp556ijvvvJOaNWtyxx130K9fv3x1duzYQc2aNQF4/vnnueSSS/jzn//MgAEDuPDCC7n99tspV65cnjYJCQlMmzYNyLucU79+/ViwYAHTp0/nhx9+iNyh/e9//5uUlBSWL19ORkYGf/zjH/Ms/SlJkqRfp1WrVgGwc+dOHnroIXr16pWvTr9+/cjKyiIzM5MRI0Zw/vnn89prr0WOjxo1iksuuSRPiPT999/ToUMHXn31VerXr3/oB3KU2J/398uWLYu8vweIjY1l2bJl+zzvzJkzWbp0ab5AVpKkY1Whwr4gCEqSG/S9Hobh27uKVwZBUH3X8erAqoLahmH4QhiGqWEYplatWrUw3ZAkSUeJ559/ngEDBrB06VIGDBhAjx499ll/2LBhNGzYEIB69eqxYsUK2rdvn6/eSy+9xJgxY5g6dWqe5ZyqV69OEASULl2abt26RZYIevnll+nQoQNBEFC3bl1q167NggULini0kiRJOmh9j6NTUkmaJ53K1/O/IrZSFEPalWX49bWoX6UEDaqWJObbN+n23a3Q9ziWL19O27Zt9+vUI0aMyLeE54MPPsjatWv5/e9/H9l/Tr/sQN/f74+dO3dy22238cQTTxRBDyVJOjoc9OYzQRAEwBBgfhiGT+5xaBzQBei/68+xheqhJEk6ZgwbNoyBAwcCcOWVV9KzZ898dUqWLBmZmRcTE8P06dO59tprGTp0KABxcXFkZ2ezatUqzjvvPKZMmUKDBg14/PHHefzxx+nUqRPjx48HYMWKFVSvXp0wDBkzZgyJiYkA1KpVi0mTJvGb3/yGlStX8vXXX3PqqacehldAkiRJ+2v4FeUKLL+lWel8ZTExMbz33nv5ys877zzOO++8PGVTpkzJV2/w4MEMHjz4oPp5LNuf9/c1atTI85pnZWXl+zvZ08aNG5k7d26kzn//+1/atWvHuHHjDGElScesgw77gLOB64AvgyDI2FV2D7kh31tBEPQAvgOuKlQPJUnSMSMmJoaPP/6Y8847j8mTJ1OvXr3IsU6dOjFlyhTWrFnDkCFDSEhI4Mknn+Tyyy9nyZIllC9fns8//5zTTz+dcePG0aVLl8iHBs2aNeObb75h06ZNNG7cmN/97ncAdO7cmdWrVxOGISkpKQwaNAjI3f+ja9euJCUlEYYhjzzyCCeeeOJhfz0kSZKk4mxf7+93u+CCC7jnnntYt24dABMnTixwuc/djjvuONasWRN5ft555/H4448b9EmSjmkHHfaFYfgpEOzlcKuDPa8kSTr6xfUZz+pxj7Lt+y/J+elHoiueyHHndKbk6d244JqehDtzCKJLUbnN74nrM55RP1xN+a+3s+J3ZYFyvDRrO3/veysAgy4sTbfGuTP9+Nf58C9IXr+TGjVOiVyvU6dOPPvss5xwwgl06NAh8uHB5MmTC+xfTEwMEydOPJQvgSRJknT06HscnUZvYUpmDmu2hMRWiuKv55XmxeZR3HJtK7J3QploeKFtWeh7HOmXTGLQoEEMHjyYypUrc//993PGGWcA8MADD1C5cmUA7rrrLt544w22bNlCbGwsPXv2pG/fvkdwoJIk/ToVZmafJEnSQava7q4Cy6t3HZivLDWmBIPblY087964FN0bl9rrueOOj2Lu3LmR57fccgu33HJLIXorSZIkaV/2tqzqjBsr5CtLTU3Nsyxq9+7d6d69e756jz76KI8++ug+r1vQsquSJB1roo50ByTpaNe9e3eqVasW2QsMcpcITE5OJiUlhTZt2rB8+fIC25Z48EdSBm0iZdAm2g3fEimf9G02Tf65iZSUFM455xwWL14MwPfff0/Lli1p3LgxycnJBe5JIUmSJEmSJEk6ehj2SdIh1rVrVyZMmJCn7M4772TOnDlkZGRwySWX8OCDDxbYtmw0ZPSqQEavCozr9L+7JG8av5XXO5QlIyODa6+9loceegiAhx56iKuuuopZs2YxYsQIfv/73x+6gUmSJEmSJEmSjjjDPkk6xFq0aBHZb2C3SpUqRR5v3ryZINjbFqgFCwL4cVsIwIYNG4iJidlVHvDjjz/mK5ckSZIkSZIkHZ0M+yTpCLn33nupWbMmr7/++l5n9m3NhtQXNtFs8GbGLNgRKR98aRnavvETsbGxvPrqq/Tp0weAvn378tprrxEbG0vbtm15+umnD8tYJEmSpGNBQUv033nnnTRo0IDk5GTat2/P+vXr87VbunQpLVu2pGHDhiQkJDBw4P/2KM7IyKBZs2akpKSQmppKWlpanrbTp08nOjqaUaNGHbJxSZIkqXgz7JOkI+Thhx9m6dKldO7cmWeeeabAOt/dWoH0GyvwxhVluXXCVr75YScAAz7fznvXliUrK4tu3bpx2223ATB8+HC6du1KVlYW7733Htdddx07d+48bGOSJEmSjmYFLdHfunVr5s6dy5w5c6hfvz79+vXL1y46OponnniCefPm8fnnn/Pss88yb948AO666y7+8pe/kJGRwYMPPshdd90VaZeTk8Pdd99NmzZtDu3AJEmSVKwZ9knSEda5c2dGjx5d4LEalXL/mz71hCjOi4tm1n9zWL15J7NX5tA0NhqAq6++mv/85z8ADBkyhKuuugqA5s2bs3XrVtasWXMYRiFJkiQd/Qpaor9NmzZER+e+N2/WrBlZWVn52lWvXp0mTZoAULFiReLj41m2bBmw76X4n376aa644gqqVat2SMYjSZKko4NhnyQdAYsWLYo8Hjt2LA0aNMhXZ91PIduyc/flW7NlJ9OW5tCwahQnlA3YsBUWrs0B4IMPPiA+Ph6AWrVqMWnSJADmz5/P1q1bqVq16qEejiRJkiTgpZde4qKLLtpnnczMTGbNmkXTpk0BeOqpp7jzzjupWbMmd9xxR2Rm4LJly3jnnXe46aabiryfBS1H+sMPP9C6dWvq1atH69atWbduXb52H330ESkpKZGvMmXKMGbMGAB69OhBo0aNSE5OpmPHjmzatAmA7777jlatWpGcnMx5551XYBgqSZKkwjHsk6RDKK7PeMo3PJe6iU34av4CoiueSJWLbiHlkq6UqnoKparV5u6Br/L5SZcS12c86ctz6DnuJwDmr8kh9cXNNBq0iZbDttDn7FI0rFqC6KiAFy8twxVv/USjRo149dVXeeyxxwB44oknePHFF2nUqBGdOnVi6NChBEFwJF8CSZIk6Zjw8MMPEx0dTefOnfdaZ9OmTVxxxRU89dRTVKpUCYDnn3+eAQMGsHTpUgYMGECPHj0AuPXWW3nkkUeIiir6j24KWo60f//+tGrVikWLFtGqVSv69++fr13Lli3JyMggIyODyZMnU65cucgSowMGDGD27NnMmTOHWrVqRbYquOOOO7j++uuZM2cODzzwAH/+85+LfDyS8jrYQD8jI4PmzZuTkJBAcnIyb775ZuRY165dqV27diTsz8jIyNPW/UUl6ciKPtIdkKSjXdV2d+Urq9io4D03UmNKMLhdWQDOqhnNlzdVKLBe+/iStI8vCX1n5ylv2LAh06ZNK2SPJUmSJB2IoUOH8u677zJp0qS93my3Y8cOrrjiCjp37kyHDh0i5cOGDWPgwIEAXHnllfTs2ROA9PR0rrnmGgDWrFnDe++9R3R0NJdffnmh+9uiRQsyMzPzlI0dO5YpU6YA0KVLF8477zweeeSRvZ5j1KhRXHTRRZQrVw4gEl6GYchPP/0UeR3mzZvHk08+CeSGhUXRf+lY0717d959912qVavG3Llzgdzw7uqrryYzM5O4uDjeeustTjjhBCA3mPvjH//I9ddfz7Bhw3jooYdYs2YN//d//8eiRYvo378/Dz/8MD/++CNTpkwhKiqKhx9+mKSkJC677DKeeOIJqlWrxm9/+1tWrlzJzTffDMBjjz1Gx44d8/XP/UUl6chzZp8kSZIkSdJBmjBhAo8++ijjxo2LBF8/F4YhPXr0ID4+nttuuy3PsZiYGD7++GMAJk+eTL169QBYsmQJmZmZZGZm0rFjR5577rlDGpStXLmS6tWrA3DyySezcuXKfdYfMWIEnTp1ylPWrVs3Tj75ZBYsWMCf/vQnABo1asTbb78NwDvvvMPGjRtZu3btIRiBdPQ60Nm4u/cXzcnJ4a9//StffPEFJ554Il988QXr1q2jS5cuDB06lGrVqrFw4ULmzZvHueeeS/369Tn55JO5+uqr+eqrr2jYsOEvLk0M7i8qSb8Ghn2SJEmSJEm/pO9xdEoqSfOkU/l6/lfEVopiSLuy/PHatmxctoDWSSeTcnIJeqWWgr7Hsfz2irRt2xaAadOm8eqrrzJ58uTIEnjvvfceAC+++CK33347jRo14p577uGFF144kqMEIAiCfW4HsGLFCr788ksuuOCCPOUvv/wyy5cvJz4+PrL83+OPP87HH39M48aN+fjjj6lRowYlSpQ4pP2Xjja7w7s9jR07li5dugC5s3F375+5p02bNtG6dWsqV67M6tWrueiii5gwYQInn3wyP/zwQ2RZ3aioKE488cQ8bdPS0ti+fTt16tSJlN17770kJyfTu3dvtm3bBhza/UUlSfvPZTwlSZLEgAEDGDx4MEEQkJSUxMsvv0yZMmXy1Hnrrbfo27cvQRDQqFEj3njjDQDuuusuxo8fz86dO2ndujUDBw7kp59+4sorr+Sbb76hRIkSXHrppQXu/SNJUnEy/Ir8M/d6NClVYN2YilGRQO+cc84hDMMC651zzjnMmDFjn9cdOnTogXX0IJx00kmsWLGC6tWrs2LFin3O0Hnrrbdo3749JUuWzHesRIkSXHPNNTz66KN069aNmJiYyMy+TZs2MXr0aI4//vhDNQzpmLE/s3F37NhBzZo1I89jY2NZtmwZGzZsAOD+++9nypQp1KlTh2eeeYaTTjoJyP0ef/HFFznrrLNYtmwZNWvWpF+/fpx88sls376dG2+8kUceeYQHHnjgkO4vKknaf/4vLEmSdIxbtmwZ//jHP0hPT2fu3Lnk5OQwYsSIPHUWLVpEv379mDZtGl999RVPPfUUAP/5z3+YNm0ac+bMYe7cuUyfPj2yFNkdd9zBggULmDVrFtOmTeP9998/3EOTJEn7qV27dgwbNgzI3Ufwsssu22vd4cOH51nCMwxDFi9eHHk8btw4GjRoAOTuN7hz504A+vXrR/fu3Q/VEKRj1i/NxoXcQH/jxo0AZGVlEYYhZ511FjNnzqR58+bccccdAJx33nnUqFGD119/nWuuuSYye7B69eoEQUDp0qXp1q0baWlpwP/2F42Li2PUqFH8/ve/L3CWoSTp0DLskyRJEtnZ2fz0009kZ2ezZcsWYmJi8hx/8cUX+cMf/sAJJ5wAwLp160hJSaFr167MmjWLE044gSeeeIIdO3Zw0kknUa5cOVq2bMmUKVM488wzmTdvHrfccgsAX3/9dWQJs5SUFCpVqhQJD3d74oknCIKANWvWHJbxS5J0LOnUqRPNmzfn66+/JjY2liFDhtCnTx8++OAD6tWrx4cffkifPn2A3A/ye/bsGWmbmZnJ0qVLOffccyNlYRjSpUsXkpKSSEpKYsWKFTzwwAMATJkyhdNOO4369euzcuVK7r333sM7WOkotXs2LrDX2bglS5Zk6dKlQG6gv3sp3X/9619ER0fToUMHAK688kpmzpzJ9u3b6dGjB126dKFjx4707NkzMvN497XCMGTMmDEkJiYCh39/UUlSwVzGU5Ik6RhXo0YN7rjjDmrVqkXZsmVp06YNbdq0yVNn4cKFAJx99tnk5OTQt29fMjIyALjtttt46qmn+Nvf/sbNN99MfHw8AOvXr+f3v/89b775JpdddhmvvfYaAKeddlqkbU5ODjVq1KB9+/aRay1dupSJEydSq1atQzxySZKOQX2PY/hpwGkA5YCNsPQ2ePo2Jv1md6VV8I/aAKQCgwdviDSPi4tj2bJleU4ZFRXFtGnTCrxcx44d6dixYxEPQtLu2bh9+vTJNxu3U6dOTJkyhTVr1jBkyBASEhLo1asXzzzzDPfddx+1a9fmkksuYcqUKVSqVInbbruNhg0b8tZbb/HJJ5+wdu1ahg4dyoYNGyLvyTt37szq1asJw5CUlBQGDRp0pIYuSSqAM/skSToGrF+/no4dO9KgQQPi4+P57LPP8hxft24d7du3Jzk5mTPPPJO5c+fmOZ6Tk0Pjxo255JJLImWTJ0+mSZMmJCYm0qVLF7Kzsw/LWFT01q1bx9ixY1myZAnLly9n8+bNkWBut+zsbBYtWsSUKVMYPnw4N9xwA+vXr2fx4sV8+umnnHnmmaxYsYLJkyczdepUAN544w0uv/xy7rrrLm6++WbOPPPMfNeeNGkSderU4ZRTTomU9e7dm0cfffQXlyKSJEmSjnZxfcZTvuG51E1swlfzFxBd8USqXHQLI7Y34a+DhlOycgx/HTSC4duaENdnPOk3VqD816NZ8bst7Li3HC9cXJJ/9L2VC86ox3Nt4Jvr1vDhOdN56pQP6dutDd0vbkrJkiV54okn+O1vf8sdd9zBjh07CMOQuLg4Ro4cCeT+/vfll18yd+5cXnvt/9k77zCpiqwPvz+GnMSEIogIKAoSzWvEiKisAeOac9g1pzXrfoY1rtk1ArqCGVBXDKgYEZEk4AqIqJhFkSCZ8/1Rt5ueYQamexhuV1nv88zj3NvdwznevlV1z6nfOY/TsGHD5Wzt06dPTO5HIpFISsRkXyQSiUQifwDOPvtsevTowf/+9z/Gjh2bVV5luP766+nSpQvjxo2jX79+2XKLGe64445Sn1m6dCnHHnssAwYMYPz48Wy00UbZHi+RlbOy5OugQYPo1KkTXbp0YauttuLdd9/NvnbxxRezxRZbsMUWW/Dkk09mz9999920bdu2oNKXr7/+OhtvvDHrrrsutWrV4qCDDuL9998v9Z4WLVrQq1cvatWqxcYbb8ymm27K5MmTef7551m8eDFHHXUUDRs2ZJ999sn6M2nSJJ5//nk++eQTHnvsMfr167fcvz1gwIBSPX8GDRpE8+bN6dy5c14+VIWVXY8MH330ETVr1uSZZ57Jnrvooovo0KEDm2++OWeddRZmVuozvXr1ypY4qjabrp3FMxMXZc/1eHwuTW6cxX5P/F7uZ84666xygzORSCQSiUQikeJk3V4X0eKvj7HRhYNocWZfGnXei5J6jVnv8OtpfsqDrHf4dZTUawTAVhuU8FCvetnPntC1NlPOasSUsxpxfNfa2fMbNanB28c3YNzpDRk6dGhWwXfDDTcwYcIExo4dy5tvvpntvxmJRCKR4iYm+yKRSCQSCZzffvuNt99+mxNPPBGA2rVr06RJk1LvmThxIrvtthsAm222GdOmTeOHH34AXPP2l156qVSvlhkzZlC7dm023XRTAPbcc0+effbZ1eBNGKws+br77rszduxYxowZwyOPPJL9f//SSy8xatQoxowZw4cffsgtt9zCrFmzAFde8/XXXy+lkKssLVu2ZPjw4fz++++YGUOHDl3OpgMOOIC33noLgJ9//plJkybRunVrNthgA8aPH8+BBx7IokWLGDZsWPaz7777Lj/88AOffvopr7zyCv/4xz+y5UABFi5cyODBgznkkEMA+P3337n++uu59tpr8/ahKqzseoBTt1588cWlypu+//77vPfee4wbN47x48fz0UcfMWzYsOzrzz33XMFJtbxsalO6Mv+Ff6rDYwfWW+79ACO/XcKvv/5akE2RSCQSiUQikUgkEolEipOY7ItEIpFIJHC++OIL1l13XY4//ni6du3KSSedxNy5c0u9p3Pnzjz33HMAjBgxgi+//JLp06cDcM4553DTTTdRo8ayZcM666zD4sWLGTlyJADPPPNMtvF7ZMVUJvnasGHDbAnLuXPnZn+fOHEiO++8MzVr1qRBgwZ06tSJIUOGANC1a1datWqVtz2tLnmJw57/mW+adGaNDdtRp2krnv34a275ckOa7HAETQ++klaXvMSpby7m+Ym/UXudlnTv3p2bb76Ztddem/r169O0aVP22GMPOnfuTOfOndl///2ZPn06H3/8MTVq1GCnnXZijz32YL311mPs2LHZf/vll1+mW7durLfeegB8/vnnfPHFF3Tu3JlWrVoxffp0unXrxvfff5+3X5WlMtcD4K677uLggw+madOm2XOSmD9/PgsXLmTBggUsWrQo68ucOXO47bbbuPzyy6vfpgaly53u3romjeosXwJ1yVLjwtfmc9NNN+VtUyQSiUQikUgkEolEIpHiJSb7IpFIJBIJnMWLFzNq1ChOP/10Ro8eTYMGDbjxxhtLveeSSy5h5syZdOnShbvuuouuXbtSUlLCiy++SNOmTdlyyy1LvV8SAwYM4Nxzz2WbbbahUaNGlJSUrE63vKUyyVeA559/ns0224x9992XRx55BHBJ2SFDhvD777/z888/8+abb66yJGuTnf5C85PvZ4MT72Wd/c5HNWvRZKejqL/JtoC75mvtfjIbnHQfn3zyCYcffjgATz75JP/4xz/49NNPmThxIrfddhvgyn5OnDiRzp07M3LkSN5//31mzpxZSqHWv3//UiU8O3bsyI8//si0adOYNm0aLVq0YNSoUay//vqrxMfyqMz1+Oabb3j++ec5/fTTS53ffvvt6d69O82aNaNZs2bsvffeWf+uuOIKzj//fOrXr79abVoRd49YSK9Na9KsWbO8bYpEIpFIJBKJRCKRSCRSvMRkXyQSiUQigdOiRQtatGjBttu6pE3v3r0ZNWpUqfc0btyYRx99lDFjxtCvXz9++uknWrduzXvvvcfgwYNp1aoVhx9+OG+88QZHHXUU4BId77zzDiNGjGDnnXfOlvSMrJjKJF8BDjzwQP73v/8xcOBArrjiCgD22msvevbsyZ/+9CeOOOIItt9++3SSrFevAVevwdxLG/Pa4Kc46LOzs+fu368e9+9XD65eg82f3I4eNd6jU6dObLPNNpx00knZ/nVz587ltdde46CDDlrl5i1ZsoSuXbuy3377Lfdanz59WHfddenSpQtdunThmWeeYdSoURxwwAFIYtCgQbRq1Yr7778/+5mzzjqLJk2asNlmm/H8888zfPhwAKZMmcKnn37K9OnT+eabb3jjjTd45513GDNmDJ9//jkHHnhgQfZX5jtyzjnn8M9//rOU4nZFfDt7KU9PXMzftq298jdHIpFIJBKJRCKRSCQS8YqY7ItEIpFIZCWsKHHw9ttv061bN2rWrMkzzzyz3OuzZs2iRYsW/PWvf82e69GjB507d6ZDhw6cdtppLFmypFrtX3/99dlwww357LPPABg6dCjt27cv9Z6ZM2eycOFCAB566CF23nlnGjduzA033MD06dOZNm0aAwYMYLfdduPxxx8H4McffwRgwYIF/POf/+S0006rVj9CoTLJ11x23nlnpk6dys8//wzAZZddxpgxY3jttdcws1STrA1qixkXNWKNustKRp62VW1O22pZQunCHeowceJExo8fzznnnLPssw0aMGPGDNZYY40K//60adNYZ5118rbrjjvuKLfHXYbDDjuMMWPGMGbMGP7617/SokULevbsyQcffMBzzz1Ht27duPHGG/n2228BeOONNxg2bBgLFy5kyZIl9O3bl4EDB/L888+z3Xbb0bBhQxo2bMg+++zDBx98wAcffMDIkSNp1aoVO+64I5MmTWLXXXettP2V+Y6MHDmSww8/nFatWvHMxEWc8dJ8Bv5vUYV/c/R3S5jyy1La3jmHVq1a8fvvv9O2bdtK2xSJRCKRSCQSiUQikUikeInJvkgkEolUC/Pnz2ebbbbJJrWuuuqq5d5z7rnnZtU1m266aameVH379mWTTTZhk002oW/fvst9tlevXlmFUHWzosRBy5Yt6dOnD0ceeWS5r19xxRXsvPPOpc499dRTjB07lvHjx/PTTz/x9NNPr3KbsyRqq7s6jOYv3dvTab0Sxgy4jktr/LuUAuvT85uzRbN6tFunhJdffpk77rhjpX/65ptvZvPNN6dTp07sv//+7LbbbtXnR0BUJvk6ZcoUzAyAUaNGsWDBAtZee22WLFnCjBkzABg3bhzjxo1jr732Wr0OFDnTp0/npZde4qSTTqrU+zPX44svvqBOnToMHTqUTTbZhKVLl2bf07BhQ7755humTZtG7969ue+++zjggANo2bIlw4YNY/HixSxatIhhw4ax+eabc/rpp/Ptt98ybdo03n33XTbddFPeeuutSvtQme/IF198kS132rt9Le7dty4HbFarwr+576a1+P6CRkw7pxHTpk2jfv36TJkypdI2RSKRSCQSiUQikUgkEileaqZtQCQSiUTCpE6dOrzxxhs0bNiQRYsWseOOO7LPPvuw3XbbZd9z++23Z3+/6667GD16NAC//PIL11xzDSNHjkQSW265Jb169WLNNdcE4LnnnqNhw4arxY9M4uCyyy7L9iLLpVWrVgDlltL7+OOP+eGHH+jRowcjR47Mnm/cuDHgSvUtXLgQSct9dlXTZf0SRp5S+v9Zrvpq+w1rMulvyetXP1fu39h1111LqZNuvvlmbr755lVua9Bc7VRsd3VYwl+6t2fhEmi9Zg0e/XM97t/vbsBdl2ffXUC/cYuoVQPqbbwNTz75JJJYtGgRO+20E+C+R48//jg1a7rl3J133slNN93E999/T6dOnejZsycPPfRQOn6myDnnnMNNN93E7NmzK3zPs88+y9tvv82miz/l9r3rclcH4y/d2zN3EXw7aykLl8CfN6vJ4FPacPgWtWDWHK7YYy3eWtCBX375hR133BFwirs33niDjh07IokePXqw//77V8n+Vpe8BMDCtofSadf9sCWLqdlkfdbueQ6P7H0mAI269iz1mV3L/I2dHp3L/35eypyFRovbZvNwr3rs3TYu+yORSKQ8vv76a4455hh++OEHJHHKKadw9tlnl3rPr7/+ygknnMDn78+hbk145M/12KKpK6Pd6l+zaVRHlAhq1qDUeuuuDxdyz2abUVJSwr777stNN920Wn2LRCKRSCQSifxxiE/9kUgkEqkWJGUTcosWLWLRokUrTGr179+fa665BoBXXnmFPffck7XWWguAPffckyFDhnDEEUcwZ84cbrvtNh544AEOPfTQavejMomD8li6dCnnn38+jz/+OK+//vpyr++9996MGDGCffbZh969e68qcyOesLLk68U71uHiHeu4g6s/yJ6vW7cuEydOLPdvnnXWWZx11lmr3liPePHFF2natClbbrllhUq6/fffnyOOOII6derw7/3qcezAebxxbINS1+Pb2Us5YMDv3NGjJouXGtNnGX/asITbnhrFbbfdxrBhwzjppJMoKSnh3//+9wptatWqFePHj8/bl9rrtabZsf8qda5ski9DnwPqlTp+5/gGK/37c+bMydumSCQSCZGaNWty66230q1bN2bPns2WW27JnnvuWUpRff3119OlSxee7/wG//t5CWf+dz5Dj1k21r55bH3WqV9649ebXyxm0GeLGPvpWOrUqZMtfx6JRCKRSCQSiVQH3pbx/Prrr+nevTvt27enQ4cO5ZYbMzPOOuss2rZtS6f75jDqO9cT6c0vFtPl/jnZn7r/Nyvb48TMuGzofDa9aw6bb745d95552r1KxKJREJiyZIldOnShaZNm7Lnnntm+0+V5csvv+SLL77IloH85ptv2HDDDbOvt2jRgm+++QZwZTHPP/986tevX+325yYO8uXee++lZ8+etGjRotzXX3nlFb777jsWLFjAG2+8UVVTI5EI8N577zF48GBatWrF4YcfzhtvvMFRRx1V6j1rr702deq4ROpJ3Wrx8XfL98zcoFENtmhawjtfLWHteqJ+LThoc7dH7pBDDllhj8VIJBKJ+EWzZs3o1q0bAI0aNWLzzTfPrjszTJw4MbtO3WydEqbNXMoPc5Yu97dyuW/kQi7ZsU52zmnatGk1WB+JRCKRSCQSiTi8TfZldt9NnDiR4cOHc8899yy30/3ll19m8uTJTJ48mQf2r8vpL80DoPvGNRlzWkPGnNaQN45tQP1asFcbF8DpM2YRX88y/vfXBnz66accfvjhq923SCQSCYWSkhLGjBnD9OnTGTFiRIXqlgEDBtC7d29KSkpW+PfGjBnD559/zoEHHlgd5i5HZRIHFfHBBx9w991306pVKy644AL69evHJZdcUuo9devW5c9//jODBg2qDvMjkT8cN9xwA9OnT2fatGkMGDCA3Xbbjccff7zUe7777rvs74M/W8zm67hxZ/qspcxb5Pok/jrPePerJbRbuwaS2H/Tmrw1zSUFy+ufF4lEIpEwmDZtGqNHj15ug1rnzp157jlX5nzEN0v4cqZTfQNIsNdjv7PlA3N44OOF2c9MmrGUd75czLbbbssuu+zCRx99tPociUQikUgkEon84fC2jGezZs1o1qwZUHr3XW7wZdCgQRxzzDFIYrsWNZk5H76bvZRmjZblOJ+ZuIh9NqlJ/VqutNx9IxfyxMH1qZGUmou77yKRSKTqNGnShO7duzNkyBC22GKL5V4fMGAA99xzT/a4efPmpUrwTZ8+nV133ZUPPviAkSNH0qpVKxYvXsyPP/7IrrvuWmG5vqpyww03cMMNNwDw1ltvccsttyyXOKiI//znP9nf+/Tpw8iRI7nxxhuZM2cOs2fPplmzZixevJiXXnop24MtEokUTqbXXYb5X41j1v9+pNUlLzHzncepvf4m1N9kW34d1od5k0dAjRrs0GAhfQ6oC8CnPy3l/FfnI4EZXPCn2nRczyUC/7lHXY5+fh7ndOrEuuuuy6OPPrra/YtEIpFI9TJnzhwOPvhg/vWvf2X7K2e45JJLOPvss+nSfw4d1yuha7MalCRhhXePb0DzxjX4ce5S9nzsdzZbpwY7b1STxUvhl3nG8A+H89FHH3HooYcyderU1dKrORKJRCKRSCTyx8NbZV8uFe2+W64MXGPxzWwr9Z4B4xdxxBa1ssef/2o8OX4RWz0wh3322YfJkydXr/GRSCTiCUOGDKFdu3a0bduWG2+8sdz3PPXUU9nyygcddBAzZ84EYNKkSdx+++3ccccdtG/fnmnTpgGw0047sdlmmzFhwgQOPvhgDjjgAMD1s3v11Vf59ddf+fXXX3n11VfZe++9Of300/n222+ZNm0a7777Lptuumm1JfoAuHqNZT999oVJr8DVa3DlLnUYfER9uHoNPjq5IS0a1+DpJ/py6qmn0qFDhxX+yblz59KrVy86deqULXF62mmnVZ8PkcgflLotO9G091UANNnpKOpv4taJa+5yHBucdC8bnHA3bx7bgM0SZd+ebWoy7vSGjD2tIeNOb8gpWy7robhRkxq8fXwDxo0bx9ChQ2nZsuXqdygSiUQi1caiRYs4+OCD+ctf/sJBBx203OuNGzfm0UcfZcxpDel3QF1+mmu0XtOFU5o3dv9t2qAGB25WkxHfOCV4i8bioM1rIYltttmGGjVq8PPPP68+pyKRSCQS+YOwsnjVV199Rffu3enatSudOnXi2muvpV27djRv3pzmzZvTsWNHttxyy2yLla+++ooOHTpQr1496tatS+/evQGXg6hXrx5dunShS5cupWI5H3/8MR07dqRt27acddZZmNlydkQi1Y33yb4V7b5bGd/NXsonPy5l7zbLBI4LFht1a8LIUxpy8sknc8IJJ6xqkyORSMQ7lixZwplnnsnLL7/MxIkT6d+//3KlkydPnswNN9zAe++9x4RDpvO3xkPovvnadFqvhK5btKNXy7l8feIseq31OR9d2B6uXoN3dh/H4et/wXnnncef/vSnbHBlrbXW4oorrmDrrbdm66235sorr2SttdZKw/Usu7aqyYtHuj6B13avS692bqPI1s1LmH5eI+Ze2pgZM2YwYcKE5T573HHHcffddwOw3nrr8dFHHzFu3DjGjx/PXXfdRc2a3grtI5FIJBKJRLzGzDjxxBPZfPPNOe+888p9z8yZM1m40JXofGjUInbeqCaN64i5C43ZC1wwb+5C49XPl7BFU7eJ5IDNavHmtMWA2/i2cOFC1llnndXgUSQSiYTFyhI5t912G+3bt6dTp07svvvufPnll4BrA7L99tvToUMHOnXqxJNPPpn9zN13303btm2RtNo2YhTqx5dffkm3bt3o0qULHTp04P77789+pkePHnTu3JkOHTpw2mmnsWTJ8v3IV6cP999/Px07dqRLly7suOOO2bjRjBkz6N69Ow0bNuSvf/1rqc88+eSTdOrUiQ4dOnDxxRfnbVNl4lX/93//x6GHHsro0aP5z3/+wz/+8Q9efvllnn/+edZYYw2efPJJ+vbty9FHHw3A5ZdfznfffcdXX33FqFGjGDJkCEOHDgWgTZs2jBkzhjFjxpS6FqeffjoPPvhgtqXYkCFD8vYlXwq9Hq+99hpbbrnlcklOqHrSshCbhgwZwqabbkrjxo1Zf/312XzzzbNVtgDWXntt6tatS7169WjYsGGp6ztu3Ljsfd6xY0fmz58PQP/+/enYsSOdOnWiR48eed/nK/Pj7bffplu3btSsWZNnnnmm1GdatWrFJptswuabb15K8NC6deusH3Xq1GGvvfbK/r2SkpJsErlXr17Z8zvttBNdunQBaC/pW0kDV2S318m+le2+a968OV9//XX2ePoso3mjZSUznpqwiAM3q0mtkmXnWjSuwUGbuwDugQceyLhx46rRA8fKvjwLFizgsMMOo23btmz70BymzVzWCPyGdxbQ9s7ZtLt7Dq9MWbzsb05ZTLu756xQgRP9CJtKX487Z8frEVkpI0aMoG3btrRu3ZratWtz+OGHL9dn7sEHH+TMM89kzTXXBKB7q1qMPrUhA3rXo+v6JfQ9sB4AN+5Rl0M6LFNUX71rXS699FLeeOONrLIP4IQTTmDKlClMmTKF448/fjmbWrVqVWEPwBURylgV7/HioVq/U3fOjtci4jUhjFXxHi8uQpnHI8VBq0teotnRN/PYY49x7xODqL1ea2qv15r1Drmatfc+k7X3PpNWl7zE5n97iAbrb0y7u+fw8pTF3NHDlYD+Ya6x46Nz6Xz/HLZ5aC77blKTHm3dJq4TutZi6q/GFltsweGHH07fvn2rtYRnKGNVvMeLi3g9IquavL5T227L559/zplnnslll11Go0aNuOaaa2jfvn2p5ETfvn1ZtGgRNWrU4LPPPuOss84CoH79+vTr148JEyYwZMgQzjnnnGz1oR122IHXX3+djTbaaLX68fLLL3Psscdy1VVX0apVK1555ZXsZxYtWsSiRYv4/fffadSoERdddBHgWll98MEHjBkzhg8//JAbb7yRb7/9FnDVlcaOHcv48eP56aefePrpp6vNj2222YZTTz01m1S74447aNmyJe3atcv6ceSRR3LzzTczb948Pv/882yMp27duvzjH//glltuKfX3Z8yYwYUXXsjQoUOZMGEC33//fTapVlkqE6+SxKxZswD44IMPaNCgAa1bt2abbbbh6KOPZtCgQXTo0IF58+axYMEC5syZw1prrcW6667Lb7/9RrNmzXj22WcrtOG7775j1qxZbLfddkjimGOOYeDAgXn5Aavveqyzzjq88MILfPLJJ6WSnFC1pGVlEq9HHnkkn3zyCWPGjOGiiy7i3HPP5cwzz+SMM85g7733Zr311uPxxx/n3//+dzZJ1qBBA6ZPn868efN44oknspuzFi9ezFFHHcX999/PhAkTeOutt6hVqxaLFy/m7LPP5s0332TcuHF06tQpuwF/VfnRsmVL+vTpw5FHHrncZ1q2bMnSpUt59tlnGTFiRLZN3MMPP8wvv/zCvHnzaN++Pb/99lv279WrVy+bRB48eHD2/DvvvMOYMWMAJgIfAM+tyHZvk32V2X3Xq1cv+vXrh5kxfPpi1qhDqX59/ccvLlXCE+CAzWpmd98NGzaMTTfdtPqcoHJfnocffpg111yTKVOmcO52dbj4dZehnvjTEgZMWMSEMxoy5C/1OeO/81iy1Fiy1Djzv/N4+S/1K/yb0Y+wyet6nNUoXo/ISlmuLHKLFnzzzTel3jNp0iQmTZrEDjvswHYPzWVI8kA3acZSmtQVBz35O13/PYcLX53PkqWldwYNHDiQ3XffPW+Fdr6EMlbFe7x4qPbv1JkN47WIeEsIY1W8x4uLUObxSHFRt0UHNrr4RTY44W42OP4uNjj+Luq12ZpGXXvSqGtPAOo035zmpzzAZ39tyHOH1WfNei5p13rNGow9zZWAnnBGQy7buU7279YuEY8fVI/x48czatQodtttt2rzIZSxKt7jxUW8HpFVTd7fqXPP5ZRTTqFt27Z07NiRl156iSuvvJK99tqrVHJizTXXpH///owZM4YXXnghq+DZdNNN2WSTTQDYYIMNaNq0KT/99BMAXbt2pVWrVqvdj/nz5/PMM89w+eWXc8ghh3DGGWewZMkSlixZwgMPPMArr7zCxIkTmThxIpMmTQKgdu3a1Knj5pcFCxawdOmypHomhrJ48WIWLlyY16aSfP3Yb7/9WLBgAa1bt2bKlCkAnHzyyQwZMiTrR4MGDbJ/86abbuL7779n4sSJNGjQgB133JG6deuW+vtTp05lk002Yd111wVgjz32WGFSrTwqE6+6+uqrefzxx2nRogXnnnsuu+6663Lvf/bZZ+nWrRt16tThuuuu48svv6RZs2b07NmTDTfcMCso+uKLL+jatSu77LIL77zzTtaGFi1arNCGlbE6r0fXrl3ZYIMNAEolOauatKxM4jU37jd37lx+++032rZty3rrrcf8+fM59NBDGThwILVr186+t0aNGqU+k/mev/rqq3Tq1InOnTsDTgFYUlKCmWFmzJ07FzNj1qxZWX9XlR+tWrWiU6dOWdsyn5k/fz5Lly7lpJNOYtCgQTRs2JD69V2VsO7du1O/fn1mzZrFlClT8t0EVgPYDRi4sjd5SWV2353xtvHuDyXUXmsDTn5hPvfuWy/7+Wkzl/L1rKXs0qqk1N+9ZMc6PPvpYjreN4e///3vPPTQQ9XqR2W+PIMGDeLYY48FoHf7mgydugQzY9D/FnN4h1rUqSk2XrMGbdeqwYhvljDimyW0XasGrdesUeHfjH6ETbwekTRYvHgxkydP5q233qL/wfU4+YV5zJxvLF4K73y1mFv2qstHJzdg6syl9BmzqNRn+/fvzxFHHFHtNoZyb4TiRwhU+7UoUbwWEW8JYayK93hxEcJ3KhKpDkIZq+I9XlzE6xFZ1eT9nerdmxEjRtCiRYtscqJFixYsWrQom5woy8MPP8w+++xT7r+9cOFC2rRpk6ofgwYN4vDDD6dVq1bMmzePtm3bMmLEiOX+5tprr12q9PPXX39Np06d2HDDDbn44otLJS723ntvmjZtSqNGjbK95arDj3bt2jFz5kx3jw8axE477cQPP/zAxhtvXMqP2rVrs+eee3LppZdy8sknr/Aeb9u2LZ999hnTpk1j8eLFDBw4sFSVvlVF//79Oe6445g+fTqXXnopb731Vqmk6S+//MLFF1/Mv//9bwBefvlljj76aFq2bEnz5s0ZO3YsNWrUoFmzZnz11VeMHj2a2267jSOPPDKrGKwqaV2P3CRnVZOWlUm8Atxzzz20adOGiy66iL/85S9suOGG9O7dmwYNGnDjjTdy4403csEFF2Tb+Uiia9eu1KlThzPPPJM777wTcKIDSey9995069aNm266CYBatWpx33330bFjRzbYYAMmTpzIiSeeuMr9KO8zkyZNokmTJjz77LPcdtttXHjhhcuV1x04cCBNmzZlv/32y56bP38+W221Fdttt11FCdY1gaFmtsIvnLfJvsrsvpPE2nudTvNTH+KT0xuy1QbLEnutmtTgm/MaUaNMBrVJXfHSkfX55PSGfPDBB9nMcHVRmS9P7ntq1hBr1IUZ84xvZi9lwzVySpA2qsE3s41vZhsbNq6xwr8Z/QibeD0iq5rlyiJPn07z5s1LvadFixb06tWLWrVqsfGaNdh07RpMnrGUFo1Fl/VLaL1mDWrWEAe0q8mo75ZNdD//vpQRI0aw7777VrsfodwbofgRAvFaRCIVE8L9EYIPIRGvRyRSPqHcG6H4EQrxekRWNXl/p2rWpH79+ssl9T7//PNsciLD8ccfz0YbbcSgQYO44IILSr3/u+++4+ijj+bRRx8tpRBKw4+KPpt7/vHHH+enn36ibdu22fdtuOGGjBs3jilTptC3b19++OGH7GuvvPIK3333HQsWLChV3nRV+1FSUkLt2rWZMWMG33zzTTYRU9aPHXbYgc8//5x//vOffPjhhyu8x9dcc03uu+8+DjvsMHbaaSdatWpFSUlJhe8vj8rEqx5++GEOPfRQAHbbbTcWLlyYVYBOmDCBV199lX79+mWTwQ8//DDXXHMNH374IePHj6dGjRq0aNGCOnXqsPbaawOw5ZZb0qZNGyZNmkTz5s2ZPn36Cm1YGWlcjwkTJpRKcq4uzjzzzKxNGSXniBEjKCkp4c477+SYY47h1ltvZerUqQC8++67fP3113z99dc0bNiQv/3tb4ATHbz77rv85z//4d133+X5559n6NChLFq0iPvuu4/Ro0fz7bff0qlTp1I9AKuTxYsX884773DEEUdwyCGHMHXqVPr06VPqPbfeeislJSVceOGF2XNffvklI0eO5IknnuCcc87h888/L/un1wL6r+zfr7Zkn6Qekj6TNEXSJdX170QikUik+tl6662ZPHkyX3zxBQsXLmTAgAGlGsYCHHDAAbz11luAS+BNmrGU1muKrTcoYeZ846e5btfUG9OW0H7dZYu3ZyYuZr/99luunEMkEolEIpFIJBKJRCKRPy41a9YslZz4+OOPGTFiRKnkxH/+8x9uv/126tatS5s2bXjqqaeyr82aNYt9992X6667ju2222612p5LWT8qSga9/vrrXHfddZx77rnlJr022GADtthii2z5yAx169blz3/+c7UqZZs3b87ixcv6b/7yyy8rTGgdfvjhjBo1aqV/d//99+fDDz/kgw8+oF27dnm31KpMvKply5bZXoANGzZk4cKFzJkzhx9//JE77riDyy+/nB122KHU+59//nkAhg8fzsyZMznrrLP46aefsiqtqVOnMnnyZFq3bk2zZs1o3Lgxw4cPx8zo168ff/7zn/PyI1+qej2mT5/OgQceWCrJWdWkZWUSr2Vt+uijj/j666954okn6NGjB99//z1t27Zlhx12YOTIkdm/C9C0aVOOP/54hg0bBrik5s4778w666xD/fr16dmzJ6NGjcr0uKNNmzZI4tBDD+X999+vNj9yP9OiRQu6dOnCwoUL2XDDDTnggANK/X9/9tlnGT9+PK+//nqpDQuZv9+6dWt23XVXRo8enX0tSUw3AF5ame3VkuyTVALcA+wDtAeOkNS+Ov4t36nMlyf3PYuXGr/Nh7XrieaNavD1b8v6Xk2fvZTmjUTzRuLrWUtX+DejH2ETr0dkVdLqkpdoe/krzN3yGDbdckcarLcRX63RmX0fm0aTHY6g6cFX0uqSlzj1zcU8P/E3aq/Tku59f+fmPeuydv0alNQQt+xZl937/U7H++ZgBidvuaxf6oDxi1ZLCU8I594IxY8QiNciEqmYEO6PEHwIiXg9IpHyCeXeCMWPUIjXI7Kqyfs7tXgx8+fP58svv+SLL75g6tSp3Hfffdxxxx2lynH++OOPnHrqqbz44oscd9xxjBgxAoCFCxdy4IEHcswxx+RV3rI6/ahXrx5ffPFFNiGV+Wzz5s2ZMGECp556KoMHD2bWrFnZvzl9+nTmzZsHwK+//sq7775Lu3btmDNnDt99913233jppZfYbLPNqs2Prl27smDBAmbNmsV6663HO++8k02qZT67dOnS7Ptfeukl1lprrZXe4z/++GPWt3vvvZeTTjqp0j5UNl41tnkvTrv8Jmo33Zij9+zC1TvXpMfWm7DZRuuxZOF8+tx4Pus3FBs3qcGPFzbi1tbv8I+/n03dmmLXXXflwgsvpF27drz99tt06tSJLl260Lt3b+6///6soi5je9u2bWnTpk255WRXxOq8HjNnzmTfffflxhtvLJXkrGrSsjKJ18mTJ2d/z3xnJ0+eTMOGDXnttdcYMGAAe+yxB8OHD2ezzTZj7ty52cTX3LlzefLJJ7Pqxr333ptPPvmE33//ncWLFzNs2DDat29P8+bNmThxYrZH52uvvcbmm2++Sv2o6DPrrLMOv/zyC48//ji9evXijTfeoH17lxYbPXo0p512Gr169aJly5bZz/76669ZBfPPP//Me++9l/0MwDPPPAMw08zmr8z2mpX2Mj+2AaaY2VQASQOAPwOx624Zcr88zZs3Z8CAATzxxBOl3tOrVy/69u3L9ttvzzMTF7PbxiVIole7mhz53DzO27423842Js9YyjbNSzBg8oylfPHrUponX8iyfzP6ETZ5XY/1idcjUinqtdma5m22LnWuyU5HZX+XxFq7nwzAJ3WPLPW+PdvUZNzpDcv9u28d1wB69FjF1pZPKGNVvMeLh2r/TjVWvBYRbwlhrIr3eHERyjweiaxqQhmr4j1eXMTrEVnV5P2dOrwxu6+/iOO6zGPPLdvw9Szj4M1rcvTnf+XKXU5iqw1K6LlJTc7p+ztzZiyl906b8eX8Rtkg+lNPPcXbb7/NjBkzsmX0+vTpQ5cuXbjzzju56aab+P777+nUqRM9e/bkoYceqnY/znjwVr6bY1y6Y23qP7AdI4bN4Yd/vcN+7WoyZuRsGtQSvXv3ZtKkSWy33Xb8/e9/59NPP+X8889HEmbGBRdcQMeOHfnhhx/o1asXCxYsYOnSpXTv3p3TTjut2q7HwKPW4k/NocfWmzBvsWFLoG3/7TnrtfmMmbCIbV7+gCdemc+bIxfRft0SmnbYiQYNGmSTJK1atWLWrFksXLiQgQMH8uqrr9K+fXvOPvtsxo4dC8CVV16Zt7IPVh6vqr1OS9Y/6mYAxiTxqst3rkNFNG0AP13Y2B1c/Vv2/MEHH8zBBx9c7me22morxo8fn7ftGfK6Hq/0YOD4RQVfj7vvvpspU6Zw7bXXcu211wLw6quv0rRpU+69916OO+445s2bxz777JNX0rLt5a8wL0m8Yktp2HFP9n1sGjPf+T9qr78J9TfZll9e/zfzp42FkhJ2qP8V/fapy5e/LeWsB2/m+zlG/Vri2J7bsUEjMe2G7WjYtIQ9H5zDnIXu39ig5ca88MILgCsDe95557H11lsjiZ49e2ZbBF111VXsvPPO1KpVi4022mi5UppV9WPBd5P46bnrWLpgDi881Y+rGoq796xDz23bMWehsdTg8N06UUOw3+9Pwk+XcmG/ufz2yxLGjRtHly5daNmyJYMHD+bTTz/l1FNPpUaNGixdupRLLrmkVLJvwIABAL9UxnaZ2crflSeSegM9zOyk5PhoYFsz+2vOe04BTkkO2wGfrXJDSrMO8HM1/xuFsgaQKcr7M/A9sAEwF/gNELAxUD/5/TMg+YqzPs43gK+ATJPG8v5mdROKH/kQwveqEe46fI7f16OYr0U+RD+qn1DGqniPFw+hfKfyoZivRz5EP6qfEMaqeI8XF/F6+EsIfhSzD6HcG6H4kQ/xe1U8FPO1yIdi9qPQ71Qz3Hcqt4HfJGApLq6s5GcW8DXVTyj3Rih+5EMI90cxPzvlQzFfi3xYHX5sZGbrruxN1aXsWylm9gDwwOr69ySNNLOtVte/V11EP4qLEPwIwQeIfhQb0Y/iIQQfIPpRbEQ/iosQ/AjBB4h+FBvRj+IiBD9C8AGiH8VG9KN4CMEHiH4UG9GP4iIEP0LwAaIf1UG19OwDvmFZxhigRXIuEolEIpFIJBKJRCKRSCQSiUQikUgkEomsIqor2fcRsImkjSXVBg4HBlfTvxWJRCKRSCQSiUQikUgkEolEIpFIJBKJ/CGpljKeZrZY0l+BV4AS4BEzm1Ad/1YerLaSodVM9KO4CMGPEHyA6EexEf0oHkLwAaIfxUb0o7gIwY8QfIDoR7ER/SguQvAjBB8g+lFsRD+KhxB8gOhHsRH9KC5C8CMEHyD6scqRmaVtQyQSiUQikUgkEolEIpFIJBKJRCKRSCQSKYDqKuMZiUQikUgkEolEIpFIJBKJRCKRSCQSiUSqmZjsi0QikUgkEolEIpFIJBKJRCKRSCQSiUQ8JSb7IpFIJBKJRCKRSCQSiUQqQFJtSZ0kdZRUO217CkHSxpU5V6xIqiHpT2nbUR0kvv0lbTsikUgkEon4TUz2RVYLkkokvZm2HdWFpAZp2xCJD0mRyB8BSS3TtiESKRZ8Dz77HniGGHyORP4ISNoX+By4E7gbmCJpn3StKohnyzn3zGq3okDMbClwT9p2VAVJjSX9XdLdkvaS42/AVODQtO2LOCTtKem1tO2oLJK2lTRW0hxJH0hqn7ZNq4LkfmmUth2FIqlOZc4VM8kYtWHadlQHknZI24ZIpDoIOtknaW1Jd0kaJeljSXdIWjttu/JF0mOVOVfMmNkSYKmkNdK2pSpIai5pq0xATVJTSdcDk1M2rVJIelTSIxX8PJy2fZUllIckSXUlnZP4caqkmmnbVBUktZb0gqSfJf0oaZCk1mnblS+ShlbmXLGSbK74X9p2VBVJ20vqLalpctxJ0hPAeymbVmkkvSnpjQp+vPlOVYSkJpIuS9uOypLM2f+S9KKkGyQ1TtumqhBI8NnrwDPE4HNk9eFT8DnAwPOtQHcz29XMdgG6A7enbFOlkbSZpIOBNSQdlPNzHFA3ZfPyZaikgyUpbUMK5DGgHfAJcBLwJtAbOMDM/pymYfkgabakWcnP7Jzj3yUtTtu+yiJpN0mTkrHqcbnNUyOBG4H70rYvD+4BLgDWBm7Do/GpPCRtLekTYBwwPplPtkzbrgL4oJLnihYzM+C/adtRKEls5AhJF0jaIjm3n6T3cc9PXiDpC0lTc35yjz9P276qImlTSQ+mbUdlkbRJEu8cL6m/pOZp25SL14HlSjAAeBs4ODn+C/AksEdqFhVGh9wDSSWANxOdpO3MbDgwB/gkeUidm3ndzM5Kzbg8kHQOcBkwBagj6V7gn0A//LkeL5ZzbkPgXKBkNdtSFR4DfsUtlE4CLgWEe0gak6Jd+dIXWAS8A+wDtAfOTtWiqvEE7kHjwOT4cKA/sG1qFuWBpLpAfWAdSWvivlMAjYGimrxXhJktkfSZpJZm9lXa9hSCpJuB/YAxwMWSXsHd6zcAJ6RoWr5cUM657YCLgB9Xsy0FI7eb8wpgA2Ag7r6+Fjg6+d0X+gEfA3fhvl93AselaVAVyQSfpwBIagO8BLycqlWVQNJmuPXtGpIOynmpMf4FniEJPgPPJUER3whiXSVpNpD5/5+Zww33zFvbzLx49pW0G3A/y8bcfwKP4ny6Lj3L8iITeH4b6IULPO+dqkVVY3ZmrE2YCsxOy5gCaIeb95oA++ecnw2cnIZB+SLpejO7FDgVOA9YLGk+7r4wM/NlA09rM+sIIOkh4DugpZnNT9es/DCzUmorSQ2BM3HX5/lUjCqMW4FTcPPfPsl/LzEzb5IACTXMLLMZ5GlJf0/VmqrzMHCGmb0DIGlH3DzYKVWrKomk9XHxg3qSulI6rlA/NcMKZ5Skrc3so7QNKYCHcTHPEcCdkr4FtsLd5wPTNCxPtipzXAO3Ie8CYPTqN6cwJHUCbmHZGvceXNJ1W9x47AuP4GILmXXuXcBBK/zEakR+Po9WDknjzWyLMuc+ySyuip1kgr4UqAf8njkNLAQeMDMvJnBJo8ysm6Rjy3vdzPqubpsKQdJEYEcz+0WujNwkYAcz+zhl0wpCTnF1KbAz7gH8YTNbmK5VlSP3Pk6S314+JJXxoyYwwsy6pWxWwUgaZ2adypwba2ad07IpHySdDZyDW3h8w7JF+SzgQR8e+iQdZGbPSXob6Ipb1OZuruiVmnF5kIy33cxsfpJ4/RrYwsympWtZ4UjaBZcwqwtcZ2ZFn5DJIFeGexguANIj+RkDnGtm36doWl6UHY8y65M0baoKkj4ys61zjoWbR7ZewceKAkl/Bg7APRwNznlpNjDAzN5Pw658yQSfkyRTA2Ax4F3wOZR1VVnKBp/N7PyUTaoUkkbjNuJlgs+P41nwuez46ut4m7MZYU9gI+ApXAL5EOArMzsjLdvyJbm3Lzaz69O2pRB8/Q6VJZR7I4OkJrjnp2NwGz9vN7MZadqUD+Vcj8/MrF2aNhWCpKmU3mR4S+6xmT232o2qApJGm1nXMue8uVeS2OdxuATNR5SOK/T15XpI+quZ3S1XNagt8CUutpBZ5xZ98lXSeKCTmS1NNnd/D7TxaZzKRVIN3IbbC3HP49eb2cRUjcoDSR/iVNOZuMKlOBHElT49d0gaY2Zdco6LanzyYndjFXhV0uG4RTm48givpGhPXpjZDZL+CTxkZj4pGcrFl6TeCphvZr8AmNlXyULQu0Rfspv+clwi4GbgNDPzptRGwqLML4mKabpPE0MOuX4slqfVaCStlfz6sqRLcKpqAw7Do5IPZnaHpLuBS83sH2nbUyCXA8/hkko+Mz9zT5vZr5Im+5rok7Q37roswCX5fOxfu5aZXZ38/oqkQ4C/JOULvaKMarck9zgzxxc7OcHnkZL+S+ngsxc7bs1skKQX8TjwnNADN2d4208mIZR1FVBu8Hlrz4I6ZmZvJb8PlPSNT4m+hCZlVLuljn0JdFJaBfcDsEvy+094pkJO7u0DAF/H3JIyc3gpfJnDgc6SZiW/C6f8mYV/m0TWAc7HPe89AnQ1s9/Staogyo5VNT0dq4ZRerzKPTbc82HRIykTLB8m6d+4CiKZuMJbadmVL2bWV6710hFm9p+07akCJ+BUVz4r8xdmnlmTjcRTPVsTAiCpFu56nAu8i6u+MWXFnypK6phZn+T3zySdbWYXpWlQgdQto9otpeI1s1GpWUagyr6cEi7C7bJdkrxUAszxZQGVwSc1YnlImomTtpaLR0qTH3FJjAyH5x6bB+VIJT2NKzl6Ky44uCT3dV8ekiQtYZlaSSxTv/r2kBSKH1+wbMwti5mZV337yttJ6AvFtqOoUMqZN3bOPfZo3vgIWBe3sWK5/gxpLwIri6SxwK4su8ffzD32aO6YBizF87FK0qMreNl82iAmaYSZbZO2HYVSzr1RCo/ujVDWI2WDz3f5GHwuR6FxM24HN+BH8DmkcSokJN0O1MK1Nsmt/FD06xFJCyhdeSMXb+bwUJA0F5f0fpRyStqa2W2r3agC+COMVZIONrPyeiQXHUk1kYowM9tttRmzCpA00szKll/0hhBiC5J+x7ViAjd/tEmOvVEnAkiajqsg8i9guVYtPqwNARKV6BEsm8v/AxwJxZEkqyzFPlYFmewLDUl9gbvNz/rISJqM6/9RLmY2bDWaUzAVlSHN4INyMQl0Zm765XqaxIekSAQk3YJLzHjXe6nMYnY5PFrM7rKi1z2aN95i2VhbltQXgZWlzNxRliDmDknNzeybtO34o+Fz4Bli8LnYiMFnP/Ap8JxB0p3lnP4NGGlmg1a3PYVSQXDKi/WIz5vxcpHU2Mxm5VRGKYVHm0SupuK1IWZ2zeqzpnAkrW8elaQvBElfmVnLtO34IyLpRuBnll/n+nKfL2ZZS6lSL+HJZjBJG63odTP7cnXZUhUk9WHFz+NerA1DiY+sCEnbmdnwVG3wLI6ZFzkS8Fx+A770qWyhz/WRIZyFeaR4kPQSrizTQDObu7L3+4iPi/IyJVAy/AZ8YmY/rm57CkXLei8tAebh12J2AtCzotd9WcxGIqsbT8dc74PPPgeeIZw1bgw+FxehB589HW8fADYDnk5OHQx8AawNTDWzc1Iy7Q9DQOPti2a2XwWVUeImkdWMpO+B8bhykc+a2cx0LVr1SPrazDZM2458kHReOad/Az42szGr2ZyCSe7zsnhzn4cy7kaKB0m1zGzRyt/pL8Wwzg29Z9+9QDfgk+S4I24iX0PS6Wb2amqW5YfP9ZHBPQh5j6R/mdk5kl6gnECCD2XlJK1wwDGz5eTgRcoDuDKq/0oChf2Bl8xsYbpmrVJ8bOB3IrA9rsQfuNJmHwMbS7rWzB5Ly7B8ML97Ly0MIaEn6SkzO1TSJ5Q/3vqy2WXnFbxsZvbOajOmCkiaiCux0d/MpqZtTzXh45hbl/KDz50ldfch+Gxm3dO2IQK4DVT74ebs5YLPgBdBKVvWW9R3xkgKOfjs43jbCdjBzJYASLoPeAfYkWWxhqJH0nq4nn0bmNk+ktoD25vZwymbVhnuSNuAVYGZ7Zf8d+O0bakKkq5cwctm/vQ/bw7sgYstXC9pOG7sHWRm81K1bNXho8Jjq+TnheR4P2AccJqkp83sptQsywPf7/MQkPSume2Y0+4r+xKebOgGkHTMCl42X2JtwDeSBuOePd70rZJWJUl9nRu6su854Aozm5ActweuBS7ClWfrkqJ5eSFpR2ATM3tU0rpAQzPzIokm6WBWvMvWl9rCW5rZxxWVl/OhrFxO4LxsEGddoKmZlaRiWIFIqo9rPH04Lsn0MvCEmb2WqmGrgGLYDZIvkl4BjjGzH5Lj9YB+uJrcb5vZFmnaV1kkCfgLsLGZ/UPShkAzMxuRsmkrRdLdZvbXtO2oKpKamdl3FZXd8CWhmWwOKYvhgoYb+jLmSuqMG2cPBWbggiBPmtm3qRq2CvF0zB1O6eBzTXKCz2bWPk37KoPngWckHWfLmsxHUiaU4LOkEpYFn3sCQQWfPR1vPwO2saQHpKQ1gBFm1s4n5YOkl3Flbi8zs87JvDHazDqmbNpKqWjDbQYfNt5ChdWnspg/ZazPL+d0A9zmz7XNrOFqNqnKSKoN7IMbe7sDQ83sL+laVTkq2iCJi/tsamZ1VrNJVULS20BPM5uTHDcEXgJ64NR9Rb/GhWy86jygpZmdImkToJ2ZvZiyaZVC0qVmdn3adkRA0l0VvNQLaG5mXoi5JK0N9MaNs5sAz+I2FKda9nJVUgzr3NCTfePLBpcz5ySN8SXZJ+kq3K6Wdma2qaQNgKfNbIeUTasUOX0nmgJ/At5IjrsD72d2t0VWP5JaARfjAgp3mllFE0jRI6kT0Bfo5FEAvbzyFOAW5ZeZWbnltIoVSRNzF95J0myCmbX3LBByH7AU2M3MNpe0JvCqmW2dsmkrJXnwXlEgxIt+RaEiaQfgcmBN4DozKy8ZWNRI2g44DKci+xy3weLBdK2qHMlDUkWBkGN92dmZIYTgs8+BZ4jB52IjBp+LhwADzyfi5u+3cD7sjNuo0B+42swuTM+6yiPpIzPbOneO8CUukrPh9iBgfeDx5PgI4AczOzcVw/KkgvLVGcw8KWOdi6RGwNm4sfYp4FbzqIVDLkky5gjgKGCOma1wfiwWKtogmcGXjZIZ5FoZdbSk3J+kOsBYM9vMlzUugKQncVUTjkni0PVxMdAu6VpWOVbw7ASAmZ21Gs0pCFVQoj6DeVKqPpeczekXAxNxcYVx6VqVP0lu4xDcGrcpMMDMLkvXqsqxgmdA4eKIDVazSaXwIvNbBSYkQdsByfFhwMRkovCpRuyBQFdgFICZfZssqLzAzI4HkPQq0N7MvkuOmwF9UjQtL1bw0Ar4U1YOsovYy4BtgVuBs8zDusmJIuBQ3OTQDPdwcVyaNuXJiu5jH0vVvCXpRUqXlHtLUgNgZmpW5c+2ZtZN0mgAM/s1Cbb5QCaQ2Q7YGhicHO8PFL0yMUM5ZTayL+FRuY0MknYHrsD5dL3P6uNk191wSYOA24G7AS+SfcDIAl8rVm7Clft7i5zgczLmvp6mYXmwjpk9JenvAGa2WNKStI3Kg1uS/5YbfE7FosK4dQWvGeBF8NnMsn7kBJ+Pxz0LrsjHosXMFsqVUv4U2BLYPGWTKktQmznN7GFJ/wW2SU5dmqNu9yLRlzA32VVvkN3A81u6JlWOTBUdSbea2VY5L70gyZs53AIqX50E0c/DBZ37At3M7Nd0rcqfpIrL4bi5uwEuid/LzP6XqmF54FsyrxL8B/gwed4A9yz7RLLGnZieWXnTxswOk3QEgJn9niRqfCEztu4AtAeeTI4PwZ/rUF6J+gzelKqHbBWX44ALcFUfepvZZ6kaVQWS3MbDwK+4ueQkXJzaB24p8LXVQujKvnrAGbhyRgDv4fr4zQfqZyThxY6kEWa2jaRRSQC6AfCBT8klAEmfmtnmOcc1cKofLx5aQ9gtJWkL3ODZARck7J8p/+UTkk7GLcbb4WTfA8zs/XStiiQL14MoPeY+a55NNJI+xKmQP0rG3HVxyj4vdhBCtvTJvmY2OzluhOtruaIecpFVjKR9cWPub7gdd++mbFKVkLQ1buzN9IYbgKs0MCNVw/7AJBunMsHnj8yz0qpJovJg4LVkvN0O+KeZlVsyvViRNLJM8Lncc5Hqp5zg8x0BBZ8H+BR8DglJJ1pOeWG5UquXm9k1KZqVN4mK9y5gC2A8rpVDb58UAZI+xa1xpybHGwP/9SWmUB6SHjCzU9K2Ix8k3Yx77nsAuMeX2FpZJL2P69v3FG6M/ThlkwqinI2SYlmCw7uNkgCStsIlmQDeMzNvkvoZku/X7jj7u0lqg4vBbbOSjxYVcq0DdjSzxclxLeAdM9suXcv+OEg6E7eJbSjuWWlauhYVjqS6uAT+Ebi42xBcXOE1H+PTxUjoyb6mZcsHSGrnW+Zb0gW4WrZ7AjcAJ+DKZnlVclHS3Tg/+ienDgOmmNnf0rOqakjazzyptw2Q7Jb/GlfvfLlB1AcZPoCkR3Dfo6FmtjRtewpF0p0ret2X65FBUnszm1jm3K5m9lZKJhWEpL/gxqduuEBhb1xA5+kVfrCISMr7dTKzBclxHWCcmbVL17LCkXSKmT2Qth35IGkpMB0YSzlKRY9K/F2Puyd+wS3EnzSz6elalT9yzcArxJfrkSGE4HMIgWeIwediIQafi4fQAs+SngCa4MoUroWrTjPMzC5I0ayCSJQB7XDX4jPfqrtI6oG7x6fifNgIOMXMXk3VsCqQ2dSdth35kKxxFwCLKede9+Uel7QzLmnhdWBU0kBchYHncPPGV+laVDUk7WFmr5c5d6yZ9U3LpkKQtCeuBHR74FVc8vI4D+Mjn+F6av+SHK8JDPc1tiDpajO7Om078iEZc38EfqL8MdcLMVCyntoDGIaLK7xkZvPTtSp/JK3wWTXt6xF6su8z4Aozeyo5Ph840Txp5ppLMknshbuRX/G1BJikA3FlpgDeNrPn07Snqvi2MJd07Ipe92XxlKgsZ9qyPkXdgQOAL4G7zWxhiuZVGkkLccHNp4BvKVNawJfrkUHSeKAfcDNQF6ce3crMtk/VsAKQtBluF55wSeVPUzYpLyRdhitx+zzOhz/jEjQ3pGpYFfBtvAVy+8uUS6YkVbEj6UrcLtTJadtSFST9hNvw0h/4kOXHXC+uR4ZQgs++B54hBp+LhRh8Lh5CCzwDSDoMuAeYCxxpZu+lbFKV8DGhnyHZxLZZcvi/zOY2X5E0xMx6pG3HH5GkYtBbZjY5qVLzCK7iwDRcUsaLnrUAcr2bD8KpwuviSi4OMD97kr0NTADOx7U+eQhYYGa9UzWsAJLSydvh1iLDzeznlE3KG0nHA1cDb0K2dcDVvsWrMni6xvW+0hyApGOA5zMVqHxF0hjcs8YTwAvAvNzX074eoSf7muEevOcD6+F6HZzv6y5P8E9JVhEB+THap9J+GSRtbGZflDm3tZl9lJZN+ZCUWTzQXI3nLrjeRDcAnYBFZnZSmvZVlmThdwhOMbMYtyB/xsxmpmlXoSQlhv+J6ynTCFdr/5+eqy+9U5NlSBQzOyWHb5vZ6DTtqSq+jrcAkrYsq8zwaR5Mynd+bWbfJ8fH4AIhX+Ie9LwIIiTKtz1xJUM64VTu/c1sQqqGVYGQgs8+B54hBp8jq45Qgs+BBZ43wVV7+ATXN3EicJ6Z/Z6qYVXAx2BnWXyfNwAk1ff8e1SqykBy7kYzuyQtm/Ih2aza1cwWSToSl1zaC+gKXGVmO63wDxQhcu1yDgfuxPUKvy1lk/ImmfvOB05NTl1pZv1X8JGix0c1WS6S1ge2TQ4/zDwX+ojncYV9zOzlMudOM7P707IpHyTtj6s49WVyfCXL4gpnl41TFzOJQOAIXEnSibjE36uWlLtNkxppG1CdmNl3uNqv2wOtgL4+J/oSrk3bgFVEKH6cuvK3FCXPSGqeOUjUJ4+kaE++1LNlfYmOAh4xs1uB41nWu6joMbMZZna/uUbtx+MUGhMlHZ2uZQWzCLejpR4uqPOFz4m+hNPSNqBQzGyUmd0BzPM90Zewf9oGVIEH5XqmAiDXpP2KFO3Jl38DCyGrOLkRp+L9DbepygvMbImZDTGzY3E7bKcAb0n6a8qmFUQSfD4b17v2S+BoSfXTtapKeN3fzswWmNlY4MwAEn31fU70STqxnHM3pmFLgZyNS+zBss0JG+N6Ed6Rkk15Y2a/mdmjwD64eeRa4LhUjSqcF3AVg04FdgEmA15sklwBP678LUWPt/OGpD9Jmgj8LznuLOnelM0qhIPlWiAAIOkeoGmK9uTL4pyKAvsB/ZJn9Ndx/VK9IflO3QWMwvXBOtDHRF/Cmri4zuc4xf5GSQLQZ7xqF1AWM/vezAbhkuPeJvoStkzbgCpwhaTdMgeSLsJVcfKF63ClSJG0Hy6eewIwGPAiYZnBzP5nZlclG6dewMVHzk3ZLCDwZJ+k13E7D7YA9gX+JemWdK2qMr5PcBm89UNSfUlXSHrQzEZI2iQZpHziNGCgpPUl9cTt+uqZsk35kPv92Q3XpBZfE0uJAuts3ET3MuBdb5aEj3DJvq1xirIjJHnT564CvB2rcvA2YSlpPUkPS3rZzKZLal9eENcDegP9JG2WKDbOwO0a9oWSHCXGYcADZvasmV0BtE3RrryRVEfSQcDjwJm4+c/XkuKhBZ9DCDxDDD4XAzH4XAQEFnjexswyzxuWbDI8MGWbCsb3hH4OPs8btwN7AzMAks0iO6/wE8XJwcBxko6Q1Bc3fp2QtlF5sFRSM0l1cS0ccvvE1UvJpryRNA24F/gGOAW3kXuupG5JrME3hgOZCgNbAxsA3lavSAghrgCeJi0lbSppqKTxZrZUUidJl6dtVwH0Aq6XtJOk63A5D5+SfZajZj8IeNjMPjazh3C9271BUnNJ50t6FxfLPRe4L2WzAKiZtgHVzN1mNjD5faakPwF/T9GeVYGvSrKy+OzHo7hkTKYP2TfA04AX5dgAzOwjSWfhmgTPB/Yws59SNisf3pD0FPAdbtfXG5At3etFvz4ASdfiNiJ8imtO+/dikHxXgRPNbGTy+3fAnz1WKWbwWU2WwecHiz64Mfey5HgSrgzYwxV9oBgxs6mSDgcGAl8Be5nZvBV/qqgokVQzGZ92xwURMnizlpTUD7cB7L/ANWY2PmWTqso2ZjYL3FMTcKukF1K2qSACCjxDGMHnweCCz4ma1zcOBgbL9fDrgevz7F3wGfgVN+Zel/OaF8HnJPA8E7e+PQVXrj6zwQ1fSpFmyIy1sKx0pJlNStOmQkjiIQ8BDYGWkjoDp5rZGelalj8hzBtm9nUZsdKStGzJF0lr5RyehFvjvgdcI2ktj8r1XgmMBEqAwZnS7knlo6lpGpYn03D9o/bGbSjM/WIZboO0T+xhSa/X5JnpLE/XI7n4rCbLxdfYwoPAhbhKA5jZOLn+5/+XqlV5YmY/S+qF25jwMdA7eQ70BUlqCPyOW+Pmbiqsm45J+SNpGK510VO4Km0zkpdqF8McGHTPvlzkUW+csiQlmc4HWprZyUnZpna++ROQHyPNbKvcOs+SxppZ57RtWxlJEDD3pm+PS8r8CmBmXuzSSUo4HAY0A54ys2+S812Bpmb2Spr2VZYkCPUFbqKDZddGuNhtp1QMWwXI71536wHXAxuY2T6S2gPbl+1H4QuSWpjZ9LTtKARJH5nZ1mXG2zFm1iVl0yqFpE8oPeY2xZW+XADgyz0u6TKc+vtnoCXQzcxMUltcifQdUjWwkiRj7tzkMPe6ZMbcxqvfqlWDPO1blBt4NjOvA88QRO+lD81sWx/XuLBc8LkRy4LPVwKk/eBdWZKKIf/GBZ9fMLOTk/O7ABeZ2b5p2lcZJL3FsnHWKBN4NjPfAs9Z5HGvO7m+571xSY3MPT7ezLZY8SeLh1DmDUnPALcBd+OUGWcDW5nZ4akaVkkkfcGye7u8e7x1KoYVgKSaQCMz+zXnXANcvNT3NkDeI4973UnaFKf0Wc/MtpDUCehlZl4lmHKRVMPHiloBxBVmU/r5tTZuI5Xh0XOspBOAS4FZwI+ZTTtJLPcWM9s9TfsqS7KpLXedC8vmwdTnQG92Y68CrsUj5VUZvFeSJYTix0JJ9UhuaEltSAK3HuB7GVsgq2AYkHvO04T+xmkbUI2chke9vMrQB8/VZAElLOdKWptl4+12uGSZL/hW4rlczOw6SUNxGyxezdk9WAP4W3qW5YeZhVy+3tfSkUEoyQJSy3yd+GKSauGCz5+mbFM+fMzywed9kx8DvAg+m9mLkjaiTPAZpz45LCWz8sLMdk3bhmrEZ/Wu12qyhCDmDdyz0h1Ac1xc5FVceXEvMLNgnmOTyhW5iT4vN1CVJRQ/cGULr07biAIJQk0WSNLy5yR2m4kr9MYJH7zAzBqlbcOqwMwekfQKbhP02JyXvscp5LzAzFqlbcOKCDnoURZfpcYAbczsJmARQLJj2Ed/QvHjKmAIsKGk/+D6xV2UrkmVw8yGlf3BBRIyv/vMtWkbkC9m9mXZH6Bjzu8+4+O9nWEdM3sKWArZB0DfAiF9gFdw/Q3AJSzPScuYKnAeLpjTRtJ7uKbHPiWXyrvH9/bxHjez4Wb2vJnNhax6d5J5VoqtLJJCCIKAx8FnM/u6zCnfxlsIp/fSabhgcyb43AXPgs9m1rrMfzM/XiT6MpjZ4jIqkwfMbK7PKhNJvm4CyxJA6chSCX1JF+BXQh8IY94ws5/N7C9mtp6ZNTWzo8xsxso/WbyEcI8n+LqBqiyh+OFzXKG+mY0oc87Hti0P4lpiZWK54wAvVMg5nIlLum4m6RtcbOT0VC2qIpKuTtuGQjCzb8xsdEYhmqh3v7OkdK+vFNP1+CMl+3zuEeezkiyXIPwws9dwjUSPA/rjym28laZNVcS7JFkF+LwIzCWU6+Fzrzvf1WQQRsIy09NnF+BPuHm8Q/Jw4TOnpW3AKiL6USR4HnwOIvAMMfhcrMTgc1HhrQ+S/iRpIvC/5LizpHtX8rFixOuEfkIQ84akjSXdJuk5SYMzP2nbVUW8vcfL4O0GqjKE4ofPve68VpPl4H3S0symmtkewLrAZma2o5lNS9msquJFG6ZKEP1YxQRdxjOUHnEsryTbAZdo8o1Q/AD3gFSCu4d2loSZPZeyTYUSSpLM54R+Lt5ej4BKR5ZVk62L62/iEyEkLJFUgusV1wo33u6VjLe3pWpY1fD2Hi9D9CNlAikd6XUZsxx8L38JuOAzTj3dipznRPOkp3MFxOBz8eCzD0GUjjSzn4G/pG1HFQll3hiIaxHwAsnmvADw+R4HvN9AlcV3PwIpGwlubHqAZWqyL4Cj0jWpILxPWkpqAhxDssbNlLM2s7PSs6rKePscW4boxypGy9quhIekJ3G9G45JJoj6wPvmSQPOXJKg7Xa4L8/wZKHuHSH4IekRoBMwgWULczOzE9KzqnAkbVPOLh0vCCihn8Xz6/EySa87M+ucNDsfbWYdUzYtbxLb2+HGqs/MbFHKJuWFpG7AXcAWwHiShKVvqjhJ/wXmA5+QEwgxs2tSM6qKSGphZtPTtqOqRD/SR9KHuI0Ig21Zo/nxZrZFupb98ZC0Di74vAdu3ngVONs3VZyksbjgc9kx19tS75KG+BzwhGzQ9ve07agKvvsg6UMz21bS6JzxdqyZdU7btnwINKHvJZnvVNp2rCoCuMezG6jMzNcNVCH5MYyk110Ia1xJDYAaZjY7bVsKQVJrXNLyT7j+ll8AR/mkjJP0PjCc5de4fVMzqopIqpEphekz0Y9VT9DKPlyPuMMkHQGuR5zKdKP2iFCUZCH4sZ2ZtU/biKoQUJLsUVxCf/vk+BvgacArPwK6HuuY2VOS/g6udKQk78qYhaAmM7NRknbB44RlQgsz65S2EVUlFNVr9KP4MLOvyyxtvRpzQwk8B6KWAZhvZnembcSqIgBlg/fq3RB8SAhCvUsAarJQ5g3gDklX4TaHZFubmGe9kAO6x4NQ7xKOH/XNbESZNa5XZSMhHDWZmU0F9vA8aVnXzM5L24iqEorqNfpRvYSe7AuiR1xFSjLAqyRZKH4AH0hqb2YT0zakCgSRJCOchH4o1yOI0pG4AMhyajKfCCFhmfCypL3M7NW0DakifUhUr8nxJOBJXMDNJ/oQ/SgmQgg+D8TzwDPE4HOxEYPPRUUIPkA4pSNDSOgPJIB5A+gIHA3sRunYyG6pWVQYodzj3m+gyhCIH96XjUz4L+WoyXwjkKTlY5JOxsXYcte4v6RnUkE8SKJ6BTCzcZKeALxKkhH9qFZCT/aF0iPOeyVZQih+9MMl/L7HTRLClfH0SX0SSpIsiIQ+4VyPEHrdQRhqMu8TlgnDgecl1QAWsWy8bZyuWXkThOqV6EexEULwOYTAM8Tgc7ERg89FRCA+hKLeDSGhH8q8cQjQ2swWpm1IVQnhHieMDVQQjh+h9LoLQk1GGEnLhcDNuM2emX5mBrROzaLCCEL1SvSjWgk62Wdmr0kaxbIecWebhz3iCENJBuH48TAuEOL1RBdIkiyUhH4Q1yOg0pEhqMlCSFgC3IZTvH5i5nWT4VBUr9GPIiKQ4HMIgWeIweeiIwafi4YQfAhJvRtCQj+UeWM80AT4MWU7qkoQ9zhhbKCCQPwIpGwkhKMmCyFpeT7Q1tOcQC6hqF6jH9VI0Mm+hBB6xIWgJINw/PjJzAanbUQVCSJJFlBCP4jrEVDpyBDUZCEkLAG+BsZ7nuiDcFSv0Y8iIpDgcwiBZ4jB52IjBp+LhxB8gHDUuyEk9EOZN5oA/5P0EaXnDZ/mcAjkHg9kA1UwfgRSNhLCUZOFkLScAvyethGrgFBUr9GPakT+x88qpqIecWZ2QnpW5Y+kKbjAVCklmZl9mZpRBRCQH/fiFucvUHqi8yqJnKgaMkmy4Z4myZBrgNqK0oFOr64FhHE9JP2XckpHmtk1qRlVAJK+AP6Mx2oySQcCjwM+JyyR1Af3MPQypcdb3xLISKqJ/6rX6EcRIWksLvhcdswdlppReZKsDdt7HnhG0g244PPnlH7m8Cr4LOkt3LOT18FnSevggs974O7xV3EbwmakaljEWyR9aGbbpm1HVZE0EDjFzLxN6Ac0b+xS3nmf5vCQCGQDVUh+vE85ZSPNrG9qRhWApKnANj7GdnKRdCZwHTCTnKSlmXmTtJT0PNABeJPSa1zfEsgABKB6BaIf1UXoyr5QesSFoCSDcPyoh5sc9so5Z4BvCSbvVa8VJfTx71pAANeDcEpHhqAmC6X85RfJT+3kx0tCUb1GP4qOEEpHhqIkC0EtA67SgPcEpGzwPmgbgg8Joah3m+C/miyIeSOUpF5A9/hAwlDvDiQMP0IoGwnhqMlCKIE5MPnxmlBUr9GP6iX0ZF8oPeJGS3oCz5VkBOKHmR2ftg1VJaAkWRAJ/YCuRyilI6cCb0nyWU0WQsLSO1XoCniBclSvHhL9KC5CCD43wf/AM8Tgc1ERg89FxUD89wHCKR0ZQkK/CR7PG5LeNbMdJc1mmUIGPK3CQTj3eAgbqCAcP0IoGwkwFxgjyXc1mfdJS99UoSvgv5SjevWQ6Ec1EnqyL5QecaEoybz2Q9JFZnaTpLsovTAHvJuwg0iSEU5CP5TrEUKvOwhDTeZ1wlLSv8zsHEkvUP5460VAJ4dQVK/Rj+IihOBzCIFniMHnYmMgMfhcLITgAwSi3g0koe/1vGFmOyb/bZS2LauIUO7xEDZQQTh+hNLrbiABqMnwOGkp6SkzO1TSJ5QfV/DtmTAU1Wv0oxoJvWdfED3iIsWBpP3N7AVJx5b3uk87RSQ9DNzqe5Is6XUwGPA6oR/Q9fC+110oJA95y+GLUk7Slmb2cSj9TCT9Exjqu+o1+lFchNK3KARCGatCIaD+akcCm+Bx0DYEH8D/XncBJvS9R9JjZnb0ys4VOwHd46H03g3FjyB63YWCzzFQSc3M7DtJG5X3um/5AUnnAnPwXPUa/aheQlf2ed0jLhQlWSh+mNkLya+/m9nTua9JOiQFk6pCKKrXh3GL2aKSTBdAKNfD69KRIanJfEnqVYSZfZz82sXM7sh9TdLZgG8B9FBUr9GP4sLb0pGhBZ5DSeqFEnwmHGVDCOrdEHwAz9W7IajJQps3gA65B5JqAlumZEtVCOUeD0K9Szh+eF02MjQ1mQ9JvYows++SX88ws4tzX0s2gF68/KeKmlBUr9GPaiR0Zd+9uIW5lz3iQlGSheJHBkmjzKzbys4VM6GoXiV9YGbbp21HVQnoevTBTWq+lo70Xk0WUsISKhxvR5tZ17RsKoRQVK/Rj+JC0lu4fq9eBp9DILTgc9kxNwk+jzPPSo0HpGzwXr0bgg8Qjno3oIS+t0j6O3AprsVJJpkhXMDwATP7e1q2FUJA9/hAPFbvZgjIj+dxCXHvykZCOGqykJKWFcQVxvnkA4Sjeo1+VC+hK/u87hEXipIsFD8k7QP0BJpLyq1L3xhYnI5VBeO16jWH0ZKewNOEfg6hXA+ve90FoiZ7LPnvLalaUUUkHQEcCWwsKffeaAR4VdohwWvVaw7Rj+LC675F4H/gOQS1DJQOPkualTlNEnxOzbDCCUXZ4K16N4cQfPAuqbcCvFeTBTBv3ADcIOkG3xJ7FRDEPY7n6t0cmhCGHwPxuNddQGqys5P/7peqFVVA0unAGUBrSeNyXmoEvJeOVVXCa9VrDtGPaiRoZV8ohKAkA//9kNQZ6AJcC1yZ89Js4E0z+zUNuwrBd9VrBkmPlnPazOyE1W5MFQjleoRCCGoySWeXl7Ase65YSXZBbgzcAFyS89JsnMrEqw0WvqteM0Q/IquagJRkXgefM4QSfA5I2fAWnqt3ffchFPVuSGqyUOaNXCRdbWZXp21HIfh+j2cISL0bhB+hEJCa7J/lJS3LnitGJK0BrEk5cQXzrD8c+K96zRD9qF6CTPYpkB5xOUqyQ4Enc15qjCuVsE0qhuVJKH5kkFTLzBYlv3cz//p/BJMkCwXfr0copSNz1GQ7Au/kvNQIWGpmu6diWAGEkLAsi6T9zOzFtO0oBLneUcthnvVWjH4UByEEn0MKPEMMPhcbMfhcPITgQ0j4nNAPbd7IxacN0GWJ93hkVRJK2chcNRmupHiGRsB7ZnZUKoYVSChJywySTjEzHytXoHDaY0U/qpFQk31B9IgLRUkWih/l4fPC3GdCSeiHggLodQdhqMlCSliWJY63kUh4+Bx4hhh8LlZ8X49Eio9Q1Lu5+JrQ933eKA/fN+T5TAgbqCAoP0LpdReEmiy0pGUGn9e4kUhlCDLZl0HSIVZOj7iy54qdEJRkEI4fufi2MA8lSRZQQj+I65HB99KR5eGbmiyEhGVF+DbeQlCq1+hHERJa8NnXwDPE4HNk1RJC0DYEH3IJVL3rfbDT53kjF0k1zGxp2nbkQ2j3eKS48LlsZEX4qCYLJWlZFh/XuAGpXqMfq4HQk31e94grD9/tzxCQHweY2cC07agsoSTJMvie0A/weoRYOtL7scq3hGVFSNrGzEakbUc+BKR6jX4UIaEFn0MYbyEGn9MkBp8jq5rA1bter9HBz3mjok2eGXzb7BkKoWygCsiPoMpGgp/jVXn4mLQsi6QWZjY9bTvyISDVa/RjNVAzzX+8utCyHnHNJd2Z81JjwFtVQ4LSNmAV4Z0fksqbmL/KnPdBqWhmLyS//l5ekiwFk6rK34Gyib3yzhUloVyPnNKRG0sanPNSI8DbHV8J3o1V5XAt4FWyT9JBFZxvAWBmz61eiwrDzD5Ofu1SnuoV8CK5FP0oLnKDz5JmZU6TBJ9TM6zqhDDeAvQCrk7biHyoKPgsuUviS/DZzHZM/tsobVtWBSEEbX33wcxuAG4IUb0LbJm2AasAH+eNkcl/dwDaA08mx4cAE1OxqAr4fo/n0CH3INlA5eM94rUfuWUjJY3LeakR8F46Vq0yfByvyuM0PHrekHReBecBMLPbVqtBBWJm3yW/nlGe6hXwQvUa/Vg9BKnsU9g94rxSklWEj35IejP5tS6wFTAWN2F3Akaa2fZp2ZYvvqtecxL6h7LsAQlcQr+9mW2TimEFEsD1CLl0pHdqsrL4uHNb0qPJr02BPwFvJMfdgffNbL9UDCuQUFSv0Y/iIrTgs49KsvLw9LuUqTBQbvDZzE5LxbACCSX4HIJ6NwQfyuKjejdUNZnP84ak4cCOmeckSbWAd8xsu3Qtyw/f7/FQ1LsB+RFk2UjwU01WHr6tcyVdlfzaDtgayGxO3x8YYZ71HQxF9Rr9qF6CTPZlkOc94ipQkmXxxZ9Q/Mgg6TngKjP7JDneArjazHqna9nKCSVJFkpCP5TrUR4+lo6sSE2WwRc1WVl8TlhKehU4NrNzSlIzoI+Z7Z2uZZUjR/W6I/BOzkuNgKVmtnsqhuVJ9KP48S34HGrgGWLwuRiIwef0CcGHivBpQ16GEBL6oc0bkj4Dts8kMCStCQw3s3bpWlY5QrvHQ9lAFYofufhYNrIiNVkGX9Rk5eFr0lLS28C+ZjY7OW4EvGRmO6drWeXIVb0Cn+e81Ah4z5ekZfRj9RB0si8XTxflQSjJQvEjg6QJZla2PMJy54qRUJJkGQJI6Ad1PXLxdMz1Xk0WWsJS0qdmtnnOcQ1gQu65YiYU1Wv0o/jxbcwNIfAMMfhcbMTgc/ERgg9l8U3VkIvPCf1Q5o0Mko7HlXx+EzdO7YzbQOxbz/YQ73GvNlBVREB+eLXGhXDUZCElLZM1biczW5Ac18E9//myxg1C9Rr9WD38kZJ9Pi/KvVWS5RKQH/2BucDjyam/AA3N7Ij0rMoP35Nk5eHjIjBDoNfD5zHXWzVZCAnLXCTdDWwC9E9OHQZMMbO/pWdV1fBR9Voe0Y/iwtcx1+fAM8Tgc7ESg8/FSQg+gPfqXa8T+uD/vJGLpPWBbZPDD83s+zTtqSoB3ePexhVyCcgPL9e4EISaLIikJYCky3AVtZ5PTh0APGmuN6+X+Kh6LY/ox6rnj5Ts865HXAaflWS5BORHXeB0XAAE4G3gPjObn55VhRMXgcVFQNfD59KRXqvJwO+EZVkkHUjOeGtmz6/o/cVOQPd49KOI8DX4HELgGWLwuZiJwefiwUcfAlTvep/QD2jeyG72zDm3jpn9nJZNVcXHe7w8AoorhOKHl2UjwX81WQbfk5YZ5FpM7ZQcvm1mo9O0p6oENOZGP1YxNdM2oDpQ+T3ivsqc91A1M07SQ5RWko1L0Z5CCcKPJKl3e/ITAkrbgFXENWkbsIrw7npUVDpSUgvwr3QkMFTSK5RWk72eoj2FsGEm0ZfwA9AyLWOqQpLc8zrBVwbv7vEKiH6kREXBZ8m54lnw+UZgtFzJ92zgOVWLCmNNXL/dTNmWhsk5r0iCz98Dg3LOeR18Bnrh53eqLN6NVeXgow8jk/+Wq95NxaIqYGaPSnqZZQn9iz1M6Hs9b0jqDjwG1JU0CjjFzKYlL78KFEWgsEB8vMfLY8u0DVhFeOdHRWUjc9a43pSNTOgHjJCUqybrk5o1hbMeriR6hoXJOa9IcgG+5QNWRChjbvRjFROksk/h9YgLQknmux+SGgN/B1oA/zWz/jmv3WtmZ6RmXBXwUfVaQUI/i4cJ/SyeXo+gSkeC/2oy38tfStoQuBloDrwM3JxT6nagmR2QonlVwmfVay7Rj/QIsHSk90oy39UyucFnXBAkG3wupl2qhRCQssFL9W4uPvsQino3FDWZz/OGpI+A48xsgqTeuH4/R5vZcN/HKx/v8VDUuwH5EUzZyAwhqMl8LoEpqSPwIMviCheb2a/JayPMbJs07asKPqtec4l+rHqCTPZlUCA94iLFgaRngcnAcOAEYBFwpJkt8CUQEkqSLJSEfijXI0NIpSNDwOeEpaTXgGdx4+2JuJ2p+5vZDJ8CIRWpXjP4onqNfhQnIQSfQwk8Qww+Fysx+JwOIfiQi++lI0NK6Ps+b0gaa2adc447AM8BFwNX+nItQrnHQ9lAFYofGUIpGxkSviYtJb0L/B8urnAScDzQy8w+92mNW5HqNYMvqtfox+ohyDKeObTLJPoAzGy8JJ96LgWhJAvFD6CNmR2c/D4w2d3yhqReaRqVJ7cm/y03SQZ4kSQzs+6QTeh3K5vQT9G0fAnieuTgdenI0NRknpe/XNfM7k9+/5uko4C3k/HWp11K+yf/LVf1igvu+ED0ozjxtnRkaGXMAih/WdvMJgCY2TOSPgWek3QxHo25AZW4DaF0ZAg+5OJ16UjgJmDvnIT+a5KONrPhFFHZqRUR0LyxSNL6mQ0hyTXZHXgRaJOuaXkRxD2eUeBLOp3SG6juB95J07Z8CMWPHLwuGxmimsz8LYHZyMyGJL/fIuljYIiko/FojQs0Sv5bruo1FYsKI/qxGgg92ed7j7hHcUqyZ4ETkoX5keaau3qza5tw/KiTuzvYzK6T9A2uHGnDdE2rHAElyTJ4ndAP8Hr43uvuEUqryYZJ2t/MZgAbpWpZJQkoYVlLUl1Lyjyb2eOSvgdeARqka1rlMbPjIat6bV9W9ZqiaXkR/ShafA4+ex94hhh8LkJi8LlICMGHXMz/XnchJPSDmDeAS3BJi+z3x8ymS9oF+GtqVuVJaPc4Hm+gKkMofvje6+4+3Jo8oyZ7V1IvM/scqJWmYfkQStJS0hpm9huAmb0p6WBc3GetdC2rPGZ2DWRVr91yVK9XAy+laFpeRD9WD6En+47H9Yg7Ozl+Gzfo+kIISjIIx48XgN3ISV6YWZ8kAH1XalYVhtdJshx8T+hnCOJ6mNlfVbp05APmUelIwlCTeZ+wTHgIF1AbljlhZq9LOgQX7PENr1WvOUQ/igjPg88hBJ4hBp+Lihh8LkpC8CEE9W4ICf0g5g0zex2ypcVfSjZAkwSir0vTtgIJ4h7H7w1UuQThR7Kx/mWWlY083peykQmhqMlCSFr+E9gc5wMAZjYumQOvSM2qwvFa9ZpD9KMaCTrZlygCbk9+fMR7JVlCEH6Y2UUAkkrMbEnO+SHAJqkZVhihJMl8T+hnCOV6+F46MgQ1WQgJS8zsdgBJ65rZTznnRwN7pmZY4fiues0Q/SgiPA8+hxB4hhh8LlZi8Ll48NqHgNS7IST0Q5k3MuwP3J6oAp4EhmQ2KHiG1/d4Bs83UGUJxQ/wumwkEIaajACSlmb2BLieg8l3KnP+K+Dk1AwrHN9VrxmiH9WIzLy4P/NCgfSIk3QT8GrmATznfA/gLjPzIsEUih8ZJH0FDMEtyt8wD28iSXVxSbKMAutt4L5MkiOyevH9eoRSOlLSucAoMxtW5nxX4CYzK/okk6QJwJa53x1JewD3Aw3MrFlqxhWApEnANNx4+1ymbIiPlFG9vu2Z6jVL9CN9coPPuCBINvgsaZSZFX3wORmXfjKzsWXOrwH81cy8SDBJGgnslxtIk9SCJPhsZo0q/HARIulRXBULr4PPko7HBZtLBZ8zyj+fkLQ+y4K2H/oYtPXZB0kfAcflqHdvAI42s+GSRptZ15RNzIuyCX2fCGXeyEVSLWAf3KajHYHXzOykdK3KH5/v8QzJBqpFZc75soEqi+9+BFQ28khgalLpIfd8S+AKM/MiySRpLLBzJmmZnOtEkrQ0s7VTMy5Pkg0J6wPPAE+a2fiUTSoYSd1Ypnp92zPVa5boR/URarLvWVyPuOHACcAikh5xvgRBcimrJPOVgPyoD+wHHI7bzfkibrLwsTyQl4SS0A8FSa9RunTklsD+ZjbD00BIKTWZT4SQsCyLpG1w4+0BuJ5LA8zs8RV+KBIJmJCCzz4HniEGn4uZGHwuDnz3QdJYM+ucc9wBeA64GLjSw7iC9wl93+eNsiRjbg9cxZqdzWydlE3KiwDuce83UEFQfrwL/B/LykYeD/Qys899W+PC8moy3wglaZkhWRseilvjNsbFcf8vXasikVVPjbQNqCbamNklZjbQzHrhJrs3JHmz66AMX0h6QNLuknzq/1GWIPwws9/N7CkzOwjoipsk3krXqsohqbGkGyQ9JumIMq/dm5ZdBfAobqf2s8ARkp6VVCd5bbv0zMqPgK7HumZ2v5mNMbO/AffiSke2wZPyDmV4T9Krkk6U5FXZLzO73cyGSVq3zPnRPib6AMxshJmdB2yDK8nmjTJD0oaSBkh6R9KlSUAn89rAFE3Li+hH0VGqdCQuEd5X0gH4N+buD0xK5sH9JHnVYsDMXjezsZIOylmHYGa/+ZjoA0iCti8DA4CPcd8vr0iCz9+b2aDk53tJ3gTQJXWXNB34LlmPtMp5+dWUzMqLEHxIWJQEBwFXOhLYHacc9ao6DYCZHQ+0BZ4GjgA+l2sl4BNezxsZJO0jqQ9uk/rBuH7V66/wQ0VEQPd4pvfuOsADuN67mXiCTzGrUPxoZGZDzGymmd2CKzM8JPHFtzUuwK2SPpX0D0lbpG1MvpjZE8lmwm5lzn/lW6IPIFkb3gmcBowBrkzXosojqaOk4ZK+TuLqa+a8NiJN2/Ih+rF6CDXZV0dS1rfkYftB3A42HxN+m+F6yZyJS5jdLWmnlXymGAnFDyTtkiRjPsbtnjo0ZZMqSxBJMsJJ6IdyPWrJlSIFXK87XB/FVwCvykYCmNmmwOVAB+BjSS/K9b7zCW8TlrkkCfFj5XpPvA98h0v6+cIjuM0gf8PdC8NyxqmN0jKqAKIfxUUwwedAAs8Qg89FQQw+FxUh+ADLet1lMbPpwC64XmXe4XtCP6B54xhgINDOzI4zs/96prIM5R4PZQNVKH5kqiMArtcdbj3yGH6t1QEws+5Ad+An4N+SPpF0ecpmFYLXSUsASZtLulrSJ8BduNhCi5TNyof7cM96HYFJwLtym+sBalX0oSIk+rE6MLPgfnALjz3KOd8DmJy2fVX0bU1cA8gladvyR/UD1z/qedzDRYO07cnT9jFlji8D3sMlwUelbV8efnwK1Chz7jhgAvBl2vb9Aa/HucAu5Zzviiv/lbqNVfBtHY/Hqm2A24CpuHLDR6VtUwE+fAHcDmyfti0F2j+mzPFRyTjVxrN7PPpRRD/AHkDncs6vAVyWtn0F+lQLlzB7Dvg5bXuq4EMv4D/Al8BDadtUgA/9cQHCOmnbUqD9HwEdkt9745KW2yXHo9O2Lw8/xpY57gB8llwbL8aqEHwoY/9Bvt4XZfzYB+iTPM/2AXoCNdO2q0BfvJ83fP4J5R4HRgLrlznXAqf6mZ22fX9AP47MzNtlzrcEHkzbvir61hGXtFyYti0F2r8+cBYuXvUJcHnaNuVp/we4DekbpG1LgfaXHXO7Z9a5no250Y/V8BNkz74MCqRHHDglGa6ucA/cRP6kmT2brlX5E4Ifkhqb2ay07SgESZ/igiBLc84dB1wINDQzL3ZLSboJeNXMXi9zvgdwl5l5oWwI5XpkkMe97nKR6wl5IK5PXBtccv8pM/s4VcMKJClddhvwFzMrSduefJAk83ihImkCsKWZzc85twdwP26ziBfK1+hHcaIA+hZJyvSG2xWnunwKN7/7pG7IIs97L/mOAumvJmkksJ/l9BmU1AK3caeNmTVKzbhKEoIPuSiAXncAkvrj7H/Z17kjlHkjmcP/CTTFKeEEmJk1TtWwShLKPa5Aeu+G4kcGed7rLoOkzXHj1cHADNz4+6yZ/ZiqYVVAUkfgIuAwM6udtj1/FCSNxT1b/JZzrhOuSthaZuZFlbPox+oh9GTfV8AQ3ID6hq8BQ0nTgNG4hexgM5ubrkWFEZAf6wInA62AbJkmMzshLZsqSyhJsgy+J/QDvB6TcDuFnwSeM7Nf07WoMCR9gSur85SZfZCyOQURSsJS0qbABSw/3u6Wlk35IOlc3M6uYWXOdwVuMk/6KEY/ipMQgs8hBJ4hBp+LhRh8Lh5C8KEsSTI/c6/viKtecVK6Vv3xCGjemALsb2afpm1LIYR2j4ewgQqC8uNNnJLsGZxAYHzKJhWEpA9wJZOfNrNv07anUEJIWkraAVd2cSNcXCGzxm2dpl2VRdKRwFQzG17mfEvgCvOkh2L0Y/UQerKvPrAfLtjZDfeg96SZvZOqYXnis5Isl4D8eB94B9fjIJto8kmh6HuSLENACf0grgeApG1wY+4BwERggLkeft7gu5oMwkhYQnbH1P0sP976lrQMRfUa/SgyYvC5OIjB5+IgBp+LjxB8yMV39a7vCf2QkPSeme2Qth1VJZR7PIQNVBCOHwBy/akPxa1xG+Niuf+XrlV/TEJIWkr6H679TNm4wozUjCqAgFSv0Y9qJOhkXy6S1gTuwM8yZt4qyXIJyI8xZtYlbTuqQkBJslAS+kFcj1w8Lx3ptZoMwkhYAkj62My2TNuOqhKQ6jX6UYT4HHyOgefiIgafi4sQgrYh+ABBqXe9TuhDOPOGpDtwyqWBQHasMrPn0rKpEEK5xyGcDVSh+JHB57KRvqvJQkLSh2a2bdp2VJWAVK/Rj2ok+GSfwugR572SDILy4/+A983sv2nbUiihJMly8TyhH8T1CKh0pPdqshASlgCSrgZ+xH2XcgMhv6RlU6GEoHqF6EcxEULwOYTAM8Tgc7ERg8/FRSA+hKLe9T6hH9C88Wg5p823jdAQxj2ewecNVLn47kcIZSMhKDWZ90lLSTcCJbhezrlr3KJTZa2MUFSv0Y9qtCnkZJ/C6RHnvZIMgvJjNtAAWAgsSk57F9DJ4HOSDMJI6Ofi8/UIqHSk92qyEBKWkP1OlcWrB4uy+Kx6zSX6kT4hBJ9DCDxDDD4XIzH4XFyE4EMIhJDQD2XeCA3f7/EQNlBBUH54XzYSglKTeZ+0TBRYZTHfNkPn4rPqNZfoRzXYEniyL5Qecd4rySAcP0IhhCRZKAl9COZ6hFI68mo8V5OFkLAMiYBUr9GPyColhMAzxOBzsRKDz+kTgg8QlHrX+4R+QPNGC+AuIDN3vAOcbWbT07MqfwK6x73fQAXh+BEKoajJQklahkBAqtfoR3XaFUBctkIUTo+4IJRkofgBIKkXsHNy+JaZvZimPfkSSpIsoIT+NMK4HqGUjvReTRZCwhKygdrTyRlvgX+b2aIKP1SEBKR6jX4UESEEn0MIPEMMPhcbMfhcPITgA4Sj3g2BgOaN14AngMeSU0fhqgzsmZ5V+RPKPR4pLkIoGwnhqMlCSFpKWgO4imVxhWHAtWb2W3pW5U9AqtfoRzUSerIviB5xkeIimei2Bv6TnDoCGGlmf0/PqvwIKEkWSkI/lOsRROnIEAghYQkg6SGgFtA3OXU0sMS3UmwBqV6jH0VEDD4XDzH4XFzE4HNkVROKejeUhH4IlNfiJJS2Jz4SwgYqCMoP78tGhkQISUtJzwLjKR1X6GxmB6VnVSRSPYSe7AtmseS7kixDCH5IGgd0MbOlyXEJMNrMOqVrWeUJKEkWREI/oOsRROnIUNRkISBprJl1Xtm5Yicg1Wv0o4gIIfgcA8/FRQw+FxchBG1D8AGCUu96n9APZd6QNBR4FOifnDoCON7Mdk/PqvwJ6B4PYgNVQH4EUTYyFDVZCISyxg1I9Rr9qEZqrvwtXvOipJ7meY+4cpRkZ0vawSclGYTjR0ITIFMOb40U7SiUQbgHo9fJSZJ5SH0zuzhtI1YBoVyPFySdgeelI4H7cGqye5Pjo5Nz3qjJAkpYLpHUxsw+B5DUGj/vkadxqteH8NP+DNGP4mKkpCfxO/j8KC7wfEhyfFRyzpvAM4QTfAZmSDqK0sFn73bRhxJ8Bm7C/6BtCD4ANAZ+B/bKOWe4kmY+sa6Z5SqR+0g6Jy1jCiSIeQM4ATdv3I77Lr2P6y/qG6Hc4z8E4AOE48ebkm7G47KRCY/g1GSHJsdH48Yrr9RkgSQt50na0czehWySZl7KNhXCw5SjevWQ6Ec1ErqyL4gecSEoySAoP44AbgTexAUPdgYuMbMnUzUsD3zcwVIekv4PeD+AhH4o1yOU0pHeq8kCKn+5O+6BaCpuvN0It+u5vFIiRUtAqtfoRxERQunIgHbZeq+WAZC0ES74vD3Lgs9nmdlXqRqWJwEpG0JQ73rvQ0iEoCYLZd4IhVDu8YDUu6H44X3ZSAhnvAqhBKakLjj7M2KNX4HjzGxsakYVQECq1+hHNRJ0si8UkiTZrhl1jKS1cCUwfUuSBeEHgKRmOJUiwAgz+z5Ne/IloCRZKAn9IK5HKEgaBRxSRk32jJl1S9eyyhNCwjKDpDpAu+TwM/Ow/5Kkq4Ef8Vz1Gv2IrGpCCDxDOMGcUIjB5+IhBB8gHPVuCAn9gOaNvrjv0MzkeE3gVp827EBQ97j3G6ggHD9CQdIHwIVl1GS3mNn26VqWHyGtcyU1BjCzWWnbUghJxbwSPFe9Rj+ql+CTfYH0iPNeSQZB+XEg8EZGsi6pCS6JOTBNu/IhlCRZKIRyPUIpHRmCmiyEhCWApDOB/5QJhBxhZveu8INFRkCq1+hHERFC8DmEwDPE4HOxEYPPxUMIPkA46t0QCGjeGG1mXVd2rtgJ5R6PFBeBlI0MSU3mfdJS0vXATWXWuOeb2eWpGpYnAaleox/VSNDJPi3fI+4IYKR52CPOdyVZhhD8qGBXi3cL81AIIaEfCqGUjgT/1WQhJCwhjreRyIqIwefiIQafi4sYfI6sakJRNYSS0A8BSWNxG4Z/TY7XAoaZWcd0LftjEsIGKgjKD+/LRuYSgJqsC54nLStY447ybTN0JFIZaqZtQDXTk9I94voCowGvkn05SrLByXETSQf4pCSDcPwAapRzzrt7KYQkWTkJ/bMl7eBpQt/76wFsbaXLRL6RPMh6RY6abFxyvKakE31Sk5nZUEmb4HHCMqFEkizZmSTX67V2yjblTUCq1+hHcbGumeUmNfpIOictYwohlMCzmX0J9ErbjlVADUlrlgk+e7fGNbPj07ZhVRBC0DYEHxJmSDqK0urdGSnaUyidMuMtgJn9Ksm3ZH4Q8wZwK/CBpKdxG/N6A9ela1L+BHSPP4rbQHVIcnxUcs63DVSh+NHGzA7OOb5G0pi0jCmUUNRkZjYG6Ox50rJEUp1MTERSPaBOyjblTUCq1+hHNVJe0iI0muT8vkZFbypyrsr9oiQTxVXpmVMwofgxUtJtktokP7cBH6dtVD4kSbKzgYnJz9mSbkjXqoLoCexpZo+Y2SNAD2DflG3Km4CuxxJJbTIHSenIJSnaUygnlw2EACenZ07+JAnLemY2Lkla1pd0Rtp2FcAQ4ElJuydqxf7JOd+4D9gSuDf52TI55xvRj+JihqSjJJUkP0fhX/B5ucAz4FXgGVzwWa6se+Z4TUmPpGhSoWSCz/9Q0k8YuCllm/JGUgtJz0v6Mfl5NglI+8ajwGBgg+TnheScT4TgA8AJwKHA98B3uMSMj0nlGkmwGfA2oR/EvGFm/YCDgB9w36uDzOyxFX+qKAnlHl/XzB41s8XJTx9g3bSNKoBQ/JgnacfMQVI2cl6K9hTKPuWMVz3TM6cwJF0vqYmZzTKzWck69//StitP/gMMlXSipBOB11imHPWJR4DZuDXJocAs/Bxzox/VSOhlPEPpETfOzDqVOfeJbyUeAvKjAXAFsAeuVNNrwHVmNjdVw/JA0jhKq15LgNFlr0+xk/ixq5n9khyvhVPF+ehHCNcjlNKRn+ACCblqsnFm1iFdyypPKOUvJdUATsGNt+DG24fMzKsksqSxZVSv5Z4rdqIfxYUCKB0ZShmzCkoDeTfmAkhqD2T6TLxhZhPTtKcQFEiJ2wrmcq9KR4bgQ0hIOga4FCilJvMpyRTKvJGLpFPM7IG07SiEUO5xhdN7NxQ/uuB52UjIxnm2LqMmG+lTXAHCKYEpqQc5cQUzeyVNewohoDE3+lGN+LaLKy/MrL+kt1jWI+5i87BHHImSDLgnOT4Tz5RkCUH4kST1LgHXg9DMvkvZpEJpAvyS/O6r6vUGYLRcU9RsQj9dkwqmCZ5fDwundGRGTfbv5PhU/FOTBVH+MkmA3w/cL6mbmY1K26YCWSKpjZl9Dl6rXqMfRYSFUToyiDJmBFL+EiBJ7k1Mgs/eJfoSvC9xmxBC6cgQfAimdKSZ9ZM0kmUJ/YM8vM9DmTdyOQ3wMtlHIPc4Tr17F3A7yzZQ+ajeDcKPQMpGwjI1WWZNcjx+qsmCKIFpZkOAIZL28zHRlzBP0o5m9i54rXqNflQjoSv7Mj3ifkuOm+B2gQ1M0658CUFJBuH4kYuPu1kgHNUruIQryxL6I3xM6IdyPbSs193M5HhN4AjzqNcdhKEmk3QzTlmZm7D82szOT8+qquHreAtBqV6jH0VEKMHnQJRk3qtlyuL5mBuKsiEE9a73PkBY6t0MnqvJvJ83cvH5uxTKPR4pLhRIrzsIRk12MbA/y8oTHg8MNjPvSr2D92vcLoSheu1C9KPaCD3ZF0QZs1w8V5JlCcgPb79PgSTJgkjoQzDXI8Qx10s1WQgJy7IE8F2qg/+q1+hHERFa8NnnwDPE4HMxEYPPkVVNoKUjvQ12ZvB93sggqYWZTU/bjj8yAW2gCsWPIMpG5pKoyV5M245CCSFpmcHnNW6GAFSvQPSjuqiRtgHVTHn+eVlSJ4eX0jZgFRGKHw+mbUAhJEmy381ssJkNBuZLOiBlswrhqkyiDyBZ1F6VnjmFEdD1KJGkzIGvpSPL8FDaBhSCmS01s/vNrDdwvZn92+dEX8I1aRtQKInqtZ6ZjTOzcUB9SWekbVe+RD+KjhpJEAfwu3RkwmlpG1AVzGyimd0NLPQ90Zewf9oGFIqZfWlmvcxsXTNramYH+Jjok9Q32ciWOV5T0iMpmpQ3IfiQkCkd+Q9J/4dLIHupaMhBK39L0ePlvCFpb0n3SRosaTBwWRJI946A7vFOmQQZQJLY9zEZEIofJcnGPMDfspFluDZtA6qCmQ0xswuAt3xO9CWcmrYBhSLpeklNzGyWmc1Kxtz/S9uufIl+VC+hJ/tGSrpNUpvk5zY87BFXhhAW5eChH5lMvaS1Mj/AgJzffSKIJBnhJPRDuR6ZXne7y5XJ649/ve7K4t1YVQ5eJiwlrSHpMEnnSToPqJMbTPCMk8t58D45PXMKJvpRXIQWfA5hvIUYfE6dGHwuKkLwATPrBxwE/AB8j+t1522Z3gRvE/o5eDdvSPoXcDYwDDdn35T8fpakO1I0rVCCuMcJZwNVKH5ket2dKOlEXJUaH3vd5eLdeFUBXiYtJdVN4grPARdLOldS3bTtKoB9yhlze6ZnTsFEP6oRHwf9fPgbrkfckyzrEXdmqhZVHS+VZOXgox9PAPvhEsaGm6xz/9s6PdPyJpQk2cgkiX9Pcnwmfib0Q7keF+NKR56eHL+Gp4mmHLxVk+Xg3YOFXP+rq4BXgW+S092B6yVdkwTdfKJEkiypne6x6jX6UUSYWT9JI1lWOvIgzxVlIQSewc8x91/ApkA/IFNKrgUu+LyPmZ2dlm0FslzwWZK3wecypSN9Wx+G4APg1LvARLnSkd6NtZL2Bg4AmienvpE0yMx83pjn47zR08w2LXtS0pPAJFwi0CdCucczG6hK9d5N16SCCMIPM/unXPnkTNnIf0Q1WdHg3To3oR8wG1fmHeBI4DHgkNQsKowSSXUyLSg8Vr1GP6qRoHv25SIPe8RJapzIQMtVjZnZL6vbpkIIxY+QSHY3z6R0kmwtMzsuLZsKQVIDXEJ/D5Yl9K8zs7mpGpYnoVyPXORvr7s1gB7kBEKAV3KDhr4h6QDzrI+lpM+Abcv+f092qn5YXpCkmJF0M7AR8O/k1KnA12Z2fnpW5U/0o3iRh32Lygs8A74Hnr3svSRpUgXBZwGTzGyTFMwqGAXSXy3Z+HIpUCpo65OiLAQfyiIP+0atIKF/DDDZl4R+CPOGpHHAiWb2UZnz2wAPx3EqPRRI791Q/Mggj3vdJcqxM4AdcfGqd4H7zGx+qoZVAUnbmNmItO3IF0kTzaz9ys4VO5Iuxm10eTQ5dTww2My8qu4S/ahe/kjJPh8X5S+a2X6SvqAcJZmZeaEkC8WPDJKGmtnuKztXzISSJMvFx4R+hkCvh49jbnlqshbAnoA3arIQEpaSJgFb55a3Tc6vAYz0MPBcA6d6zTY1Bx4yz/ooRj+KF9/G3FACzxCDz8VIDD4XFyH4kIuk0WbmlVI0hIR+KPOGpG7AfUAjlvmxIfAbcKaZeVelJsB73LsNVOURkB9erXFzkfQUTk32eHLqSKCJmXmlJgshaSnpceBuMxueHG+LG3OPSdey/JErs599jvVV9Rr9qD7+SMk+7xblkeIimeDqA28Cu7JMvt4YGGJmm6VkWpXwOUmWi8+LwFwCuh7ejbkhqMkCSlgeC1yJ8+Pr5HRLnB//MLM+KZlWZXxVvZYl+lFc+DbmhhB4hhh8LmZi8Ln4CMEH8Fa9631CP5R5I4Ok9cnZJGJm36dpz6ogoHs8lLhCKH54tcbNJSA1mbdJS0mf4BKUtYB2wFfJSy2B//l2LXLxWfWaS/Rj1fNHSvadYWb3pm1HIYSgJAP//ZB0NnAOsAHwbc5Ls4AHzezuNOyqKnERWFwEdD18LB3pvZoshIRlhsTmvVleofhrelZVnYDu8ehHEeFb8DmEwDPE4LMPxOBz8eCrD4God71P6IcybwBI2hn4wcw+k7QDsD0w0cz+m7JpVcLXe7wsAcUVQvHDy7KREI6azOekpaSNVvS6mX25umxZ1QQ05kY/VjE+Ns5dKSq/R9yAzLF50iMuR0m2ThL0zFWSNa/wg0VGKH6Y2R3AHZL+ZmZ3rfQD/uBrg92yPJi2AasIL69HeaUjJTXxqXQkrnn5KEnlqslSsyo/MmWSy7IUz75bZvarpDcpHXj2OtGX4NV1WAHRjxQpL/gsyafg83HAfZLKCzwfl5JNhTBf0tZlg8/A1oA3pY2gVPD54yT4fFQSyPE6+AycBnif7MPTsaoM3vmwAvXuWZL28UW9myjYt/U8oX8cAcwbyXdqG6CmpFeA3YGXgfMkdTezC9O0r4p4d49XwP5pG7CK8NaPsmUjJXlVNrKMmux9SaXUZKkZVjijJG1XJmk5MmWbKkV5yTyfxUBlCGXMjX6sYoJU9imQHnGhKMlC8QNA0ma44NoGyanMzk4fJ2zAz4mugoR+Fl8S+uXh6fUIonQk+K8mC6X8paQuwP3AGriAjnDfqZnAGT6XXfRR9Voe0Y/0CKV0JPivJAtBLQOlg89AbvB5F2C0z8HngJQNXql3y8NHH0JS74aiJgtg3pgAbAHUwz1nNDez3yXVwo23W6RqYBXw8R6HMNS7EI4f4HfZSAhHTRZCCUxJ55U9BfwduB7AzG5b7UatInxWveYS/Vj1BJnsC41QlGS++yHpYuAIoD+lExqHAwPM7Ma0bKssoSTJAkroh3I9gikdCSBpPUoHEX5I05588T1hCSBpDHCqmX1Y5vx2wL/NrHMqhhVAeapX3PWYmZpRBRD9KB5CCT6HEniGGHwuZmLwOT0C8SGI0pGhJPRDmDckjTezLRLl0nfABmY2T1IJ8IkPwfMMgdzj/yKADVSh+JHB57KRFeHppm7vk5aSZgP/BSawTHl1DvAvADO7JhXDCqSs6hXwSvWaIfpRzXaFnOyT5z3iIBwlWQh+yPXz6mBmi8qcrw1M8CG4FkqSLBRCuR4KoNcdhKUmCyBhObmi742kKWbWdnXbVAihqF6jH8VFCMHnUALPEIPPxUYMPhcHIfgAQal3vU/ohzJvSPon8CegLvAWsBkwHOfHVDM7LT3rKk9A93goG6iC8CODPO91F7iazKukpaSWwK3AVNzz3u+SpvoSayuL76rXDNGP6iXIZJ+W9Yh7E9gVSvWIG2Jmm6VkWl6EoCSDoPz4H7B32d0ryW6XV82sXTqW/XEJIaEfAgqndOQYPFeThZKwlHQn0AYXQMh8pzbEBRC+MLO/pmVbPoSieo1+FBchBJ9DCDxDUMHnm3BJyhh8LgJCCNqG4EMuAah3vU/ohzJvAEjaHrexc7ikNsCBuPJ4z5jZ0nStqxyh3OMhbKCCoPzwvmwkhKMmCylpKenPwEXA7cBNHif7glC9Rj+ql5pp/uPVyKks6xGXG9icBXjTHw44kfKVZLfhJg0vkmSE48c5wFBJkymd0GgLeBF4zuB7kiwnob9OEqTNTeg3r/CDRYrv18PM+koaTOnSkW8BfzePSkcCDcom+gCSB/EGaRhUAH2oOGH5KFD0CUsAMztL0j7AnymtzrjHJ7UMy9S6ZVkKxdPAuRJEP4qIJGm/refBZzMzk5QJamauy1KgRko2FcKelB98vhEYDXiR7DOziyoIPj8EPJ2udXnRs4Lg85PAJMCLZB8wX9LWZYO2wNaAL+WNQvABKKXe/ThR7x6VBHJ8Wo+8JOkdXEL/IeApSZmE/tupWlZ5Qpk3MLMPcg5/wm1Sn+pLoi8hlHv8OOA+SeVtoDouJZsK4TjC8GO/tA1YRXTAqckasExNdqwvSb4crmH5pGUJbsOhV5jZIElDcVVevCvvnsMoSduVUb2OTNmmQoh+VCNBJvvM7A7gDnneIw63cN0AKFsHuVnymi8E4YeZDZG0KW4Hd27w+SMzW5KeZZUnoCRZEAn9gK4HZvarpDcpHXj2KdEH8LKklyhfTeZL+a8QEpYAmNnLOIWMz1yHWwCWq3pNzar8iX4UGQEEn/8bQOAZAgk+JwGovpljM/scuCVRzDyBq9DhAzH4XDwch/8+lFLvSspV754nqbsv6l0zu3gFCf1n0rWu0oSQsMyUJzzHzH6WKzv8IG4zwiaSLjAzXzZYHEcA93ggG6hC8mO5HnDyrGwkgJl9BRySqMlek3R72jYVSChJywwLgceAfpJqm9nCtA2qLGVUr+9LKqV6Tc2wPIl+rB6CLOMJhNIjrgcucVGuksw86T0Rih+Qrfc8y8xmSmoFbAV8amYT0rWsckg6m2VJsm9zXpoFPGhm3iTKAHxP6IdyPRRI6UiACtRkg30JoCuQ8pcrQtIDZnZK2nZUliSRn6t6/QZ4xbdkePSjeFA4pSMrKmP2tHnygKJwyl+OAu43swdyzjUAnge+NrMTUzMuDxRAidtcfA/agv8+KKDSkblIagxsghunfJr/Qih/+UmmpKKk94EjzWyapHWAoeZB24BcfL/HodQGKm9770IYfiigspEZJDXEqcm2NbOd07anEBRACUxJ++JiVp/jvlcb4yoiebG5WK59VIWUlygvRqIfq4cgk30KpEccgKQaeKwkyxCCH5IuwSnKFgC3ABcA7wHb4eqge7Pw8D1JBmEk9DP4fj0UQK+7kPA9YQkgaa2KXgLGmlmL1WlPVZG0HqUDIT+kaU+hRD+KgxCCz2WVZDnnawH9zMwXJVkoScu1cAr2x83sTknr4so2DTWzS9K1Ln9i8Lk4CMQH73vdwYrVZIBParIsHicsJwDbm9ksSe8CO2cSlZImmFmHdC2sPIHc4/8ijA1U/yIMP4LodZeLpNq4zWAGfOaTmiwX35OWkv4H7GdmU5LjNsBLZrZZupYVjo+q1/KIfqx6Qk32TaL8HnG1gQnmSbNg8F9JliEEP5KF+Va4sovTgNZm9lOy8/lDH4JrEEaSLLCEfgjXY3JF46qkKWbWdnXbtKrxTU3mO5KW4Eo/5/ZRs+S4uZnVTsWwPAlF9Rr9KC5CCD4HpCQLKWnZGBcYfAe3YeR+c60RvCIGn4uDEHwAkPRP4E/4r971Xk0WSsJS0qHAxcA9QDtctaPBQHdghpmdn6J5lSage9z7DVQQlB8tcWUjp7KsbORUH5Vk4L+aLBffk5aSPjKzrXOOBYzIPVfMhKJ6jX6sHoLs2UcgPeJylWSScpVk10jyRkkWih/AkiSYthCYB8wAMLO5bp4ofsokyUYkp1sAAyT5lCQ7kfIT+rfhdoF54UdA1yOEXncrU5P1XJ22VAeeJSynArub63dQCklfl/P+YqUPFateHwWKPriW0IfoRzERQr+7PYAhkup6riQ7W1KdipKW6ZmVH5IOSn59ALgNGAp8nTlvZs+lZVs+KJD+arg+ouUFbW8ERgM++BGCD6H0ugOoIamxmc3CxUO+AkgSZ77EhDqb2c/J71fhFHHZhCXgRbLPzJ5KNrycDGyKi8ltB/Q3s1dSNS4/grjHCaT3LoH4YeH0ustwK9C9rJoMz/rSl5e0lORb0nKkpP8CT+Huj0OAjzxa617D8qrXElzpep+IfqwGQlX2BdEjLiAlWSh+9AFq45rT/g4sxiUydgMamdmh6VlXOUJRvSYS/L2tTB1kubrJr5pZu3Qsy49QrgcEUzrSezXZShKW3pS/lHQm8K6ZjS3nNW/K3oaieo1+FB8VBJ99Kx3pvZJMgZS/lPToCl42MzthtRlTBQJSNoSg3vXeh/KQv6UjvVeTKaDylyEQyj0ekHo3CD9ykedlI8F/NVkGBVAC0/e1biiq1+jH6sGXXVx5YWZDJG2K5z3iCEBJlhCKHyfhdn8YbifntjhV1me4BycfCEL1iqvbPlRSuQn9tIwqgFCuB8muLp92dpVHCGqyn6g4Ydk0FYsKwMxWNKbev9oMqTpBqF6JfhQVKlM60sw+B25JEhpP4NYmRU0oSjIz+0XSHrjv1gZ4mrQ0s+Mr876y370iJAhlA2God0PwYYWlIyV5UzoyEDXZNcCbku7BVQl6WlImYenNHA6QfJdaAK/nbl6VdIKZPZKeZXnxUgj3eCjq3VD8KMNC4DGgn6Ta5lnZyATf1WQZZmcSfQlTgdlpGVMIlV3rFiuhqF6jH6uHIJV9EEyPuD54riSDcPwIgVBUrwCSauB5Qj+k61ER8qh0ZAhqsiQBXmHC0sw2TMGsKpPsgtwNOBK3q3C9lE2qNCGoXiH6UUwogH53vu+uzZCTtGzEsqTlgMzrHgVxKoWkUWbWLW07KkLSTbgefd4rGwJR74bgg/e97kJCUltKJyynAwM9Slgi6XpgR2AUsD/wr8wzRrGPsWVZwT3+TEZ16Ru+qnfL4rsfCqTXXUDr3fuAjSidtPwKeB38WO8ma8T/wwlQhgCdgHPN7PFUDSuAEFSvEP2oToJM9imnRxyQ2yNuO8CbHnFy9fPLU5J9BdxjZnNTNK/SBORHQ+Ai4GDcTryFuMXHfUW+y7kUISTJIIyEPoRxPRRI6cgVIamWlSm3WoyEkLDMRa6X2pHAAcBawJm4xIx3D62RyKpCgZSOrAzFriQLJYhTWSSNNrOuaduxIgJJMJX7vU/Uu/3MzAf1rvc+QFilIwNRk3mPpE+Arma2WFITnCL/MzM714cxtiJ8TS6tSL0LeKPeDcWPDAqgbGRIhLDelTTGzLpIOhDYDzgPeNvHTTtybX82w8XYP/NU9Rr9qEZCTfYF0SMuUlxIGoTbNf86cChOqTgAuBz4xswuTdG8ShNCkiyUhD4Ecz2873VXHj6rycrDl4QlZHc9Z3YM9seNvSPNbONUDVuF+KR6XRHRj3RQAP3uKoNvKoeKgXSngAAALjVJREFUKPakZWUp9usRUIIpBPWu9z5AGL3uIBw1WQgJS0mfmtnmOccluJLWjYH2viSQQ0kuhaLeDcWPDAqn110wajLf0bI+ow/h1MdDJI318N4IRfUa/ahGfOpdkA9LzGweMJMyPeLSNCpfJDWUdK2kCZJ+k/STpOGSjk3btnwIxQ+glZn1MbPpSTKpl5lNBo4HDlrJZ4uCJEk2DBgu6STcgmMfXI3981I1Lj+OBtoDOwC3AzslQYNtgKLfVZQhoOsxFdjVzDbO+WmdJGZ+SNu4fJG0naQ7cQnMQbi+E17uIpRjd0kP40od+cJJuO/OfcBjZjaDZb2XvEHSWhX8rA30TNu+yhL9KC7kSkfugQsOngx8QtLvTsvKSoaCV82dV8DZaRuwiij263G2pFJJ+yTB9BKulYAv7AGcJOksADn17lvAKF+SZIThA2b2FHAYLtG3Ka41RabXnReJvoT9gd3M7BxgS2AfLesvU+z3NZBNWF4GdATekPS3nJd96tn+uaRdMgdmtiS5Jz4DNq/4Y0VHZzP7Ofn9KpzqdQ/c9+vy9MzKmxrJBipw/V2/Akh8q5maVfkTih8ZRkr6r6TjkrjhCyS97jxb6+5lZrNwSrJpuA0jF6ZqUQFIuklSY0m1JA1NYrpHpW1XnrwopxjdEhiarEvmp2xTIdwKdDezXc1sF9zmo6LqF1dJoh/ViI+DfmUYJekJnPJqKNBXUqZH3MRULcuP/+B2P+5NGSWZpHa+KMkIx4+5knY0s3cl9QJ+ATCzpclOIx/IJMnKVb3i+s34wBIzmydpIWUS+v5cCiCc6/EvYE2Sh4oy3LR6TSkcLa8muwanJvNOjaHyy19ekKZNedIM2BNX8vlfkt4E6kmqaWaL0zUtL36iYtVr01QsKozoR3Gxf87vg8ucM6Do+2bkgXdJ/grwanGyAt5L24CVsAcwRFJd87jErZn9ImkP4GVJG+ChejcEHzIkZeQuTtuOKpJdPyXVRPYHHpD0NC6B6QP7s6z85dXAE5Jam9m5+DXGHlLeSTO7XK4vli/UkNQ4SWSUSi7JtXHxhWuANyXdg5vjnpaUUe8OSdWy/AjFjwx1cRs/M4nxn4B6uHHAp7Vu5l7YF1dO/DfP4lUZ9jKzi+RKYE7DiR3eBrxRKJrZJYnS8jczWyJpLm5t4huzM+VtE6YCs9MypgpEP6qRUMt4htIjrpSkWImUXa7H10Rf6lUH5Ecn4CFcaYoJwAlmNikJJBxhZnemamAlkDTOzDrJlQv5DljflvWdGO9LiVtJfXAPpg1wO7UX4xaxuwGNzOzQ9KyrPKFcjxUhv0pH/ogrP/Mv4AUzWyBpqpm1TteyylNOwtL78peS6uB2Qx4B7IQL2h6ZrlWVQ9JkYHczWy4RLulrM9swBbPyJvrhJwqgdKQ87l+Ui/wqk7cv0AEXaAPAzK5Nz6L8UAAlbnNUC41wG7+G4jZKAmBmRR/kDMGHDAqjdOSLwM1mNqzM+f8DLjWzoq/4pEDKX5aHpKvN7Oq07cgHBVLiFkBSW1ylhE1xyZnpwEAzeyVVw/IkFD9CQtKNuM2383BVqJoAL5rZtimalTcKpwTmFrjN9rlr3H7pWZQ/yaaQjYCncDmPTOzndfBnfRX9qGa7Qkz2hYJcre2LcpRkZ5rZ3slrn5lZu3QtrByh+FFZijm4FlCSLJSEfh8CuB5lSZSu3vW6S4IGGTXZ7sCbOJXAhr6oyUJIWK6IJIh7QGZRXszjLYCkM4F3zWxsOa/9zZJ+OcVO9MNP/r+9e4+27S7LO/48J9xi8HCJKFdBQJtQcoWAUAIJitga7gmRotKY1oHtoIggUsQGTOgYSokYo1BEAoVQI8aIUiMBcuNOOCcxIQRQkzAKeIlAIE0ahPD2jznXOfMs176suc/ev/m++/sZ44x9OYTxvmftNdde85nvOzMFTCuxfXZEZFrPtlCW0NL2m9RtGzhe3QVuJ6q7R06KtYtVAibb56zy1xERk19ZX6EHqdS97g6UpOhudTL/dw+IiC9tfVXLqRBYriTTz9IQ4RI2iwvd6872vbV3muy7JO2MiL9rXdcyKoSWtk+TdJy6sO/P1d0+58MRcWLLupZV6Pcr+thEJcM+23eX9HJJz1F3Fd4/qbtZ4hunfFJwXoVJMqlOH+s15V/Wq4RkVVR7PLx4deSfRsTXWtY1RtZpsgqB5TKmfLxdS6ap19XQx3QlCphST5KtR5bQcrBxYPbx7pIujIhjW9e2HlN9w71Zpn7By3pMvQfb12jv6sh7SnqXpM9FxEuyHGNXkm2arEJguZLsP0vZVZjeler0IUm2r4qII92tjTxB0i9KujzbJJlUY5pMyh9a9q/nR0i6MiKOsP19kt4ZEU9pXBqw31UN+96jbnXZBzR3jzhJX4o894hbl6m/SVqvQn3wy/omqxLoV1FxdeRQtmmymayB5TKyHW+zTr3Oo48cMoTh2SfJhiqElrY/ERGPtf1xdfdj+YqkayPi4Y1L26+yvI6vJcNzfC1T76H46shJ/9uvR7bAciW2d0R/K4dMKoRLhaZ3S/Qx4zprI0tMk0n5Q0vvvZ3ULnXvO26RdF0kua3UTJWpV/rYXGlXHazhIRHxtoj4YkScKenpEfFXkk5R98a1mhe3LmA/qdLHZBN023e3/Wu2r7X9dds32f647Re0rm1J56q78elT1d2M+ixJPy3pyf0vuikUejz+vbobaL9R0jsi4iua8PNgWRHxjblfZFMcqyLimxFxfv9m4gc1uDl7wp+xlaT4ObP9w7bPkvQFSe9Rd0PzVG8sJPpIyK0LWIfHR8TPSPpaRLxG0uPUrQNLpQ8tT5b0InX/7iepu39DNu/tp5dep+6E4Y3qLuKpJsXr+DpkeI6vZeo9/I3tJ82+iIg7+osRPifp0JX/sxSm/m+/Hk9vXcCybH/P3Nc/JekNtn+uvwgphf49969IOkzSxbZfNPjryU+yDzxN0pMj4hckPUrSv7b9m/3fpXk8VKePmffa/qy6Xj7YbwW7vXFNY5yobtvO30XEKeomy+7RtqTl9aHlb/d/jpf0G0p0/O2PrVf3v+P+nqRd6n7P/VjLukb6sYj4hrqLum9Ud7/UX2pa0Tj0sYmqhn232n6CJLm7R9xXJam/WirjC91aqvREH5uvREimOoF+lcfjfuquZnmaupMi75B0YL+mtKIpP8cXyhpYrsOkHwvb/832X0l6raSrJR0l6aaIeHum9bb0kdZHWhewDrN1bLfZvr+kb6l7TcmmRGgZEadHxM0Rcb66sPKQiPjV1nVtgkm/diwhxQUva5h6DydJ+uT8NyPiVZIetPXl7FePal3AfpDxuXzR7BPbr1L33m+XujX8Z7YqaoQq4dKdor/VQUTcrK6vnbbfLekuLQtbUpU+JEkR8QpJj5f06OjW7N8q6Rltqxrl9v489Lf7bUH/oJyvHalDy+hWGj6m/x33TeqOty/oe8lmdo7tJyS9OyK+3rKYDaCPTVT1ROwLJb3F9p57xElSfzXI77QsbJNM/U3SelXpY8on1x4SEW/rPz+zH2U/3fYpkj4jKcuK21ttPyEiPjwf6Ge6IlJFHo+IuEPd1NhfDFZHHijpS7bLrY5UjWNVpufJaqZ8vJW6qdfPq5t6/bOI+KbtjD8/9DFRq62OjAT3iNM/nyQLdes8s5kPLb+iRKGl7RUvlLKtiPjjraxnC6R+3g9UeC2fdA+x+P5wr46IV0eie8TZ/p6I+MfB1z8l6TG2Py3p9yLvvV0yBpbDn/lnSzo2Im61/S51r4NZ7BMu2X6apDcnDJf+xvaTIuIyac/72lNtn6HuliFZVOlj6P6SftT23Qbfy7Q2ctE02f9Vzmmy2/tzbZlDy922j4mIKyLixtbFbMBs6vX/Sfr5xFOv9LGJSoZ9EXG1pMcs+P5N6iZnJNW5X4Mm/iZpCWn6SHxyrUpIViXQr/J47BER35R0vqTz+18Enzn7O465k5LmpE7i463Unex/irr7Jr7B9iXqp15nJ0iSoI8J8gr3u2ta1JIi4vT+0/Ntv1fS3aZyReSSsoeWT+s/fq+6K+kv7r8+XtJHJVUL+yq8jkvTv+BlPTL28HRJr25dxJIuknS0tGea7FhJ71J3gd6hkl7SrrT1KRRYHmj7KHVbtg6IiFslKSK+ZfuOtqUtpUq4dNKib0bEq2y/cauL2YAqfUha+V53ShT2RUTYfkw/afkm238haWd/vjqNQqHlYyU93/YX1E2KWt3DdHjbspYTEa9wd5+4r0fEHbZTTr3Sx+Zynt+J9j8nvFHtIrbPTnDCc01Z+ljp5Fp//4ZJs324upr3hGQR8fk+JHteRJy16v9BMlMPl7bh48ExdyJsXxkRR7WuYy2Zj7fzBlOvz1N3ki3l1Ct9TIftqyPi8MHHu0u6MCKObV3bWlabJJOUepKs/9lKGVravkjdWqO/7b++n6S3RcRT21a2f2V6HV/tgpcsKvQwlOV3qKFhzbZ3a+802Z0l7Y6Iw9pWuLbh+4gFgeUXI2LygaUk2b5U+150928j4m9tHyzpfRHx6DaVLcf2gdKK068PyDT5Om82vdu6jo3K3Ifta9StirwyIo6w/X2S3hkRT2lc2lJsv13S2RFxRetaNsL2NbPXCdsPUc7QcuG9tCPiC1tdy0bZfqS6IHz4e1WaIHyGPjbPdg/70vyiXuVNUoU+Mp9cW6+ph2TrVShcqvJ4cMydiCwnOqseb2dTr7NfArM+x+mjLdufiIjH2v64ulVgX5F0bUQ8vHFpa7J9Tv/pwkmyiDihSWFLqhZa2r4uIg4dfL1D3c/Uoav8Z5NU4XW8wgUvFXqYZ3tHdPdgSqNfMfU8ddNkb42IIwZ/d1VEHNmqtvWqEFiuxvYBku4aEbe1rmWszOHSUKFzCGn7cHd7k2Ns71L3+nGLpOsi4pDGpS2lP/Y+XFLqabIqoWUFK029RsSJLetaFn1srpJrPJeQIumssKZJqtOHkt+XZZ1eLGnyJzrXocqapiqPB8fcLbTaic4MQV+v5PE2Ir6hfdfQpHyO00dzaVdHRsQp0p5JskfMT5I1LG1Z1dZfftD2+yT9r/7rkyV9oGE9o1R5HZf0+MEFL6+x/XpJF7Yuakmpe3Cd1ZF/J+nM/vOv2r7fYJosyxrrEusvbf9URLyz//xfRcRHpG4Npu2flXR20wI3JuOK20WqnENI2YddZm2kJFXZjFBiBWYRJ2rv1Osps6nXxjWNQR+baLuHfVle/FK/SRqo0kfak2tLyPLcWEuWN+BrqfJ4ZOkj/bGq0InO7XC8lfI8N9ZCH1soatzv7kGzoK/395K+v1UxyyoUWkrqLgTppxVn09NvjogLWtY0UvrX8V6FC16y95D+XneSFBHHrfBXN0t64tZVsiEVAktJ+kXtPRH42+p/vnrZw74Uvz+tw6NaF7CfpOwjosa97qScKyJXUCW0rOD2iPiO7W/322n+QdKDWhc1An1sou0e9mW5IXj2N0kzJfoocnJtLYRk01Ll8eCYu3VKnOjcJsdbqc5znD62wGqrI21nWx1ZYpJMyUPLof7nJ9PP0CIVXselGhe8ZO9h+F7i2dq7OvJd6vpJocI0WZHAUtr3Z2r+vWr2967pwqUq07tV+hjYbfuYiLgiIm5sXcx2Vyi0TK3K1Ct9bL7yYV+RNWbZ3yTNpO6j2Mm1tWR/ozGTJVxaS5rHg2PuZKQ+0bnNjrdSouf4Guhja5RZHVlokqxEaGn7Fu0Nu+8i6c6Sbo2Ine2qGqXC63iJC14K9FBidaQKTJNVCCx7scLni76erELhUonpXdXpY4a1kcCcKlOv9LH5Sod9VdaYFXiTJKlEH2VOrq1DmpCsSLi0lhSPB8fcScl+onM7HW+lJM/xdaCPLVBwdWT6SbIqoWVEfPfs8/5q1WdI+uF2FY2T/XW8wgUvFXroVVkdWWGaLH1g2TvE9tXq/t0f1n+u/uuHtitraVXCpRLTu6rTxwxrI4HFqky90scmcp4LbpbXry87fPDx7pIujIhj1/yPJ2C1N0mS0rxJqtLHTH9y7QXzJ9ciItUvJKuFZFmsFC5FxKlNCxuhyOPBMXeCbN9VyU50zlQ53ko1nuMSfUyJ7esi4tDB1zskXTv83tQVmiQry/aVEXFU6zrWo8rruO1z+k8XXvASESc0KWwJFXpYje0DJN01Im5rXct62N4dEUfPf77o66kaHovmj0vJjlMPXu3vs6zLm3s8dmtvuHRnSbsj4rC2Fa6P7c9Kep666d23RsQRg7+7KiKObFXbMqr0AWB1/XP94ZJST73Sx+YqPdmn5GvMVGeyoUofM+nvy1JlAktF7ktW6PHgmNtYoSvpZ9Ifb6U6z3H6mJz0qyOrTJJVCS3nXkN2SHq0pNsblTNG+tdxqcb0boUepFKrIytMk5VYf7neMM/2xyLicZtdzwZUWXFbZXq3Sh8AVpfuoucV0Mcmqh72pV5jVuVNUpU+BtKfXFORkEz5w6WZKo8Hx9z2SpzoHKhwvJXqPMfpY0KqrI6ciW7dyJ/YPk3SK1rXs4wqoaX2voZI3cnBGyU9vU0pyyvyOj5U4YKX7D1UWR2ZZuJ7FRUCy2Xcbe3/SVMlwqWIOG6Fv7pZ0hO3rpKNqdIHgNVlmf5eC31srtJhX/b7NQxkf5M0U6KPIifXqoRkqcOlgRKPB8fc9qqd6CxyvJWKPMdFH5MTye93V2CS7J/JHFqqewxeHN2N5mX7XpJery7UyCTt6/icChe8ZO+hwr3uqkyTVQgslzHpacUq4VKV6d0qfQAANq5k2FdwjVn2N0kzVfpIf3JNRUKyQuFS6seDY+4kVTnRWeF4KyV/jg/Qx4QUWR2ZepJsplBoefgs6JOkiPhav6Ytmwqv4yUueCnQQ4nVkUuY7DRZkcCyjELhUpXp3Sp9AAA2yN3Fp7W44A3B594kXZ7sTdIeFfoocnJtD9t3VbKQbLVwSVLGcGmPpI8Hx9yJsX22pB/Uvic6/zoiXtSuquVVO95KOZ/ji9DHtAxXR0ZEmmky22/XgkmyiEg1STZ4HZT2hpZvjoib2lQ0ju2/lHRcRHyt//reki6LiMPaVra87K/jmAbbt0n6a/WrI/vP1X/90Ig4qFVtm8H27og4eu3/5XTZvjIiMl6ksI+p9zH8WZn/ucn0czT8d57/N5/6YzBUpQ8AwMaVnOyrtsZMKjPZUKKPzPdlKTSBVeK+ZFUeD46501PgSnpJuY+3Up3nOH1MX+LVkVUmyaqsv3y9pI/Zfnf/9UmSXtuwntGyv45LNS54KdDDdlsdWUGVq9l/unUBayix4lZ1pner9AEA2KCSYd9AiTVmBd4kSarTx1DCk2slQrJC4VKJx2OAY+6EVDjROZTweCvVeY7TxwQVWR25w/a95ibJMr4/KRFaRsT/tP0pSU/uv/XsiPhMy5rGKPQ6nvqCFyl/D9twdWSmkCaluePTPn+l7tfdneo++fSWFra8KuHSIbavVj+923+u/uuHtitraVX6AABsUMY308uocr+G1G+SZqr0kfnkWqGQbCZ1uFTw8eCYOxFVTnRmPt5KdZ7j9DFZFe53V2WSrEpoqT7cSxfwDVV4HZ+X9IKXfVToYRWTvdfdkqY+TbYekw4sh8en5KqES1Wmd6v0AQDYoJL37Buqer+GKnu3M/ZR4b4stq+LiEMHX++QdO3wexkUui9ZicdD4pg7RVnv5SXVON5KdZ7j9DEthe539wjtnSS7OOkk2c9IeqWkfULLiHhHu6owlPF1fIULXp6UaYKsQg/rMfV7lK13mqwC24+c8lRcfzHIiiLiq1tVy0bYfvBqf7/eqdgsqkzvVukDALCylFecLqPCGrPskw0zVfpQjfuyVJnAKnFfMhV5PCSOuVOU/Er6Csdbqc5znD6mpcrqyAqTZCXWX1ZR6HW8wvRuhR7SqzBNVmj95S51fSyaQAwlmYrbhituq0zvVukDALCC0mFflTVmqvMmqUof6U+uFQrJSoRLVR4PjrnTUehEZ/rjrVTnOU4fk1NmdWQFFULLQtK/jvcqXPBSoYf1mPTqyArTZBUCS0mKiB9oXcMWqxIuVVmJVqUPAMAKSp8QKHS/hipvksr0UeHkWoWQrFC4VOLx4Jg7KWVOdFY43ko1nuMSfUxMlfvdAftbhddxqcYFLxV6WI+p3+su/TRZhcByqH+v9HxJPxARp9v+fkn3jYhPNi5tfyNcAgBgC6U8YTZG8jVmVd4kVekj/cm1KiFZlXCpyuMxxDG3uSonOtMfb6U6z3H6mBZWRwIrqvA6LtW44CV1D1VWRxaZJksfWM75XUnfUfcafrqkWySdL+mYlkVhRZOe3l1ClT4AACtI84v2GIXWmKV+kzRQoo8KJ9eqhGRDmcOlKo8Hx9xJKXGis8LxVqrzHKeP6WF1JLBQhddxqcYFL6l7qLI6cibzNFmRwHLosRFxtO0rpT2/q9+ldVGboEq4NPXp3fWq0gcAYAXuzo/XZPucwZezNWZvjoib2lQ0ju2fkfRKSfu8SYqId7SranlV+qjK9pURkSoMWCFcelKFm4AnfTw45k6E7b+UdNzcic7LIuKwtpVhJuNzfBH6ADA1FV7HZ2w/QnsveLk44wUvmXsouDryjeqnySLi0H7zw0URkWaaLHNgOWT7E5IeL+mKPvS7j7rHotTvIrYfOeXJ1/VO705dlT4AABuX8QrHZZRYY1ZosqFEHxUUmsAqcV+yQo8Hx9zpSH0lfTVVnuP0ASCDIq/jkmpM7ybvodrqyArTZFXWX54l6QJJ32v7tZJOlPSqtiWtX6EVtyWmd6v0AQDYuOphX4k1ZlL6N0l7VOmjgBIhmYqES6rzeHDMnYhKJzqLqPIcpw8AKWR/Hcc0FFwd+S3bB6gPafppsu+0LWlpFQJLRcS5tndJ+hF1AdkzI+K6xmWtW5Vwqcr0bpU+AAAbVz3sq3K/BmB/qxKSVQmXqjweHHMnhBOdk1LmOS76AABsM1VWRyr5NFmvQmAp26dLulzS2yLi1tb1LKtQuFRlerdKHwCADap+EpY1ZsBiZUKyIuFSlceDYy6wWJXnOH0AALajEqsjs0+T9SoElpJ0vaTnSTqrX4n5IUmXR8R72pa1biXCpSrTu1X6AABsXMaT4uvGGjNgRVVCsirhUonHg2MusKISz3HRBwBgeyqxOjL7NJlUJrBURJwj6Rzb95X0XEkvk/RzklKsx6wWLlWZ3q3SBwBgvPInNlhjBixUIiQrFC6VeDwkjrnACqo8x+kDALAdlVgdqfzTZCUCS0my/RZJj5D09+oehxMl7W5a1AiFwqUS07uq0wcAYCRHROsaADRg+xHaG5JdnDQkK4PHA6itynOcPgAA243t50s6WdLRkt6ufnVkRLx71f9wouamye4VESmmySTJ9imSjpX0OHVBRrrAUpJsXyDp/uoukrxMXQ/Xt61qebbfqD5ciohD+/sgXxQRqcIl27tn07sRcVT/vb+MiCNa17aMKn0AAMYj7AMAAAAAAFiB7UO0d3XkBzOujlwwTfZhSbsj4ttNCxshc2A5ZPtQSU+V9BJJB0TEAxuXtJQq4ZLtT0h6vKQr+n7uoy60THVP5yp9AADGK7/GEwAAAAAAYIwqqyMlHSzpAEk3S/qqpH/MFvQVWn95groJxSdKuqeki9X1k02VFbdnSbpA0vfafq366d22JY1SpQ8AwEiEfQAAAAAAAIulv9edJEXEs6R9pskusZ1tmix9YNn7cXU/R78VEV9uXcwGlAiXIuJc27u0d3r3mRmnd6v0AQAYjzWeAAAAAAAAq8i+OnLBNNnHJX0oIt7asq4xCqy/PEjS7RFxh+0fknSIpAsj4luNS1takRW3s+ndj2ae3q3SBwBgPMI+AAAAAACABarc68722erq/1DWabIqgWU/fXWspHtJ+oikKyT9U0Q8v2lhS6oSLtk+Rd3j8ThJaad3q/QBABiPsA8AAAAAAGAB2xdIur+kz0i6TN3J8+vbVrW8CtNkFQJLSbK9OyKOtv0iSQdGxG/Yvioijmxd2zKqhUvZp3dnqvQBAFgeYR8AAAAAAMAqCqyOTD9NViGwlCTbV0r6j5J+U9KpEXGt7Wsi4rDGpY2SPVwqNL1bog8AwHh3al0AAAAAAADAFC1YHXmxuhPp2TgibrN9qqTfnU2TtS5qSZdLOtb2vSRdpC6wPFlSmsCy92JJ/0XSBX3Q91BJlzSuaWkLwqUTJe1uWtQ4B0s6QNLNkr4q6R+TBmRV+gAAjETYBwAAAAAAsNiPqwsyfivz6khJtv04dcHYqf33DmhYzxgVAktFxOXqgsvZ19dL+s/tKhqtRLgUEc+S9pnevcR2uundKn0AAMYj7AMAAAAAAFjsl1VgdaRqTJNVCCxl+z6SXi7pX0q62+z7EfHkZkWNUCVcqjK9W6UPAMB4hH0AAAAAAACLlVgdWWSarEJgKUnnSjpP0gmSXijpBZJualrRCIXCpSrTu1X6AACM5IhoXQMAAAAAAMDk2N4dEUfbfpGkA2erIyPiyNa1LaPKNFkFtndFxKNsXx0Rh/ffuyIijmld2zJsn60uXPpQ5nDJ9kEqML1bpQ8AwHg7WhcAAAAAAAAwUcPVkf+7/1661ZHqpsk+K+kHJL1G0o3qphTTsH0f26+z/ee2L579aV3XCLPw5W9t/4TtoyTdu2VBI/2ypD+KiC/b/iHbT7d959ZFjXC5pLvafoC66d2flvS2phWNU6UPAMBIhH0AAAAAAACLVVkdeXBE/L6kb0XEZRHxs5KyTfWlDyx7Z9i+h6SXSnqZpLdIeknbkkapEi45Im6T9GxJvxsRJ6mbgM2mSh8AgJG4Zx8AAAAAAMACRe51J81Nk0n6svJNkx0cEb9v+8URcZmky2ynC/si4r39p1+XdHzLWjbIEXGb7VPVhUu/Yfuq1kWNMJzePbX/Xsbp3Sp9AABGIuwDAAAAAABYoNC97obTZL8taafyTZNVCCxnP1P/QdJDNDgv109bZlIlXKoyvVulDwDASI6I1jUAAAAAAABMju2LJJ2nbt3iCyW9QNJNEfHLTQvbhmyfIOlDkh6kvYHlayLiT5sWtiTbH1XXxy5Jd8y+HxHnNytqBNtPVPe8+EhE/HofLv1CRGScfAUAID3CPgAAAAAAgAVs74qIR9m+OiIO7793RUQc07q2ZRSaJkvP9lURcWTrOtCpMr1bpQ8AwHis8QQAAAAAAFisxOpISe9RN032AQ2myTIpFFi+1/a/iYg/b13IRhQKl85VN717ggbTu00rGqdKHwCAkZjsAwAAAAAAWKDQ6sj002TZ11/avkVSSLKkgyR9U12YbEkRETsblre0KituC03vlugDADAek30AAAAAAAALRMR7+0+/Lun4lrVsUIVpsu/KFiQNRcR3t65hPzs4In7f9osj4jJJl9m+onVRI1SZ3q3SBwBgJMI+AAAAAACABbKvjpybJnul7czTZBUCS0mS7Z0R8Y3Zx9b1jFQlXDrD9j0kvVR7p3df0rakUar0AQAYiTWeAAAAAAAAC2RfHVlBtfWXkmR7d0QcPfvYup4xqqy4BQCgCib7AAAAAAAAFku9OnIo6zRZwfWXQ25dwFhVVtxmn96dqdIHAGA8wj4AAAAAAIDFyqyOlHSppKMHH9PJGlhWVChceo+6CcUPaDC9m1CVPgAAIxH2AQAAAAAADBS71928tNNkKhBYFlIlXKoyvVulDwDASIR9AAAAAAAAA8VXR1aQObAcitYFbECVcKnK9G6VPgAAIxH2AQAAAAAArIDVkdgEnvuYUepwqcr0bpU+AAAbR9gHAAAAAACwsktVa3Vk5mmyKk6e+5hGlXCpyvRulT4AABtH2AcAAAAAALC2zFNYUo1pspnUgWVEfH74MZOK4VKV6d0qfQAAxtnRugAAAAAAAABsurTTZAPpA0vbN9i+fv5P67rGsL1z+DGxS+c+ZnXp3EcAwDbCZB8AAAAAAEBxmafJBioElo8efH43SSdJunejWjbqUtVacZs2RJ5TpQ8AwBII+wAAAAAAANaWdnWk7Ru0oP6IeGiDckarEFhGxFfmvvUG27sk/dcW9ewnhEsAADRG2AcAAAAAALCy9KsjVWCarEpgaXs4AbdD3WPD+TkAALAh/DIBAAAAAACwsvSrI4tMk6UPLHuvH3z+bUk3Snpum1IwJ+307pwqfQAAlkDYBwAAAAAAsIIKqyMrTJMVCSwVEce3rmETZA+XKkzvSnX6AACMkOoXOwAAAAAAgK1SZXWkCkyTVQgsJcn2wZJOk/QEdT9bH5b0awvCzAyqhEvpp3d7VfoAAIzgiOwX3wAAAAAAAOx/fTAzs2d1ZESkmiarwPYlgy9ngeV/j4jPtaloHNvvl3S5pHf233q+pOMi4kfbVTWO7R+KiM/PPrauBwCA7YywDwAAAAAAYJ1s74qIR7WuYxnFpslSs/3piHjk3PeuiYjDWtW0nVWZ3q3SBwBgvHTrDgAAAAAAALZCldWRkv5A3TTZc/qvny/pPElppskKBZYX2f5JSX/Yf32ipPc1rGeUQuHSowef75nebVTLRlTpAwAwEpN9AAAAAAAACxRaHZl+mqzK+kvbt0g6SNId/bcOkHRr/3lExM4mhS2p8orbjNO7i1TpAwCwPoR9AAAAAAAAhdk+U9Inte802WMi4mXtqlpOhcCyuozh0grTuz8fEUc0KmmUKn0AAMYj7AMAAAAAAFigyurICtNkFQJLSbL9LEkXR8TX+6/vqW5C8U9a1rWsKuFSoendEn0AAMYj7AMAAAAAAFigyurICioElpJk+6qIOHLue1dGxFGNShqFcAkAgGkh7AMAAAAAAFigyurIKtNkFdi+OiIOn/teup+pKgpN75boAwAw3o7WBQAAAAAAAEzURbZ/0vaO/s9zJb2vdVEjnDYL+iQpIm5WFwykYftZtu8x+Pqetp/ZsKSxPmX7TNsP6/+cKWlX66KWZftg22fZ3m17l+3f6gOnbP5A0k2SnqNuNexNks5rWtE4VfoAAIzEZB8AAAAAAMAChVZHpp8mK7T+8iBJvypptgr2/ZLOiIhbV/6vpqfKittC07sl+gAAjHen1gUAAAAAAABMUUR8d+sa9pNP9RNkv9N//Z+Ub5ps0XaqdOe1+lDvFa3r2A/uFxGnD74+w/bJzaoZ7yLbPynpD/uvT1TO6d0qfQAARmKyDwAAAAAAYIEq97qrME1m+62Sbta+geW9I+LftappDNuPlvQrkh6sQVg5P3k5dX14/EntGy49JiJe1q6q5RWa3i3RBwBgPMI+AAAAAACABaqsjqygQmApSbY/J+mXJF0j6Tuz70fEF5oVNQLhEgAA05Ju3QEAAAAAAMAWKbE6ssI0WaH1lzdFxJ+2LmKjqqy4LTS9W6IPAMB4TPYBAAAAAAAsUGh1ZPppsgqBpSTZ/hFJz5P0QUnfnH0/Iv64WVEjVAmXqkzvVukDADBeuqvRAAAAAAAAtsiL1K2OPK//+v3qAr9sKkyTnasFgWVCp0g6RNKdtbePkJQq7JN0WkRcMPsiIm62fZqkP2lX0iglpndVpw8AwEgc9AEAAAAAABYotDryNNtvUe5psgqBpSQdExH/onUR+0GVcOlTts/UvtO7uxrWM1aVPgAAI7HGEwAAAAAAYIFCqyPfqW6a7FoNpski4mfbVbWcQusvz5H0uoj4TOtaNqLQituD1E3v/mj/rfdLOqMP+tOo0gcAYDzCPgAAAAAAgAUq3OtO6vrIPk1WIbCUJNvXSXqYpBvUhZZW10e2AJlwCQCACSHsAwAAAAAAWMD2hyPiCa3r2KgK02QVAktJsv3gRd/PFiBXUWh6t0QfAIDxCPsAAAAAAAAWKLQ6Mv00WYXAcp7tn4uIN7euY4wq4VKl6V0V6AMAMB5hHwAAAAAAwAKFVkemnyarEFjOs707Io5uXccYVcKlQtO7JfoAAIxH2AcAAAAAALBAldWRQ1mnySoElvNsXxkRR7WuY4wq4VKh6d0SfQAAxrvT2v8TAAAAAACAbemjth9RaXWkpBdKShf2DUO9rIHlAk9rXcAGnGb7LcofLp2ibnr3zhpM70qiDwBAKkz2AQAAAAAALFB0dWTaabKZzOsvqyi04rbE9G6VPgAA4zHZBwAAAAAAsNiPty5gE2SeJptx6wKgY4qES1Wmd6v0AQAYick+AAAAAACANRRaHZme7QdGxBdb17Gd2T5H0uuyh0tVpner9AEAGI+wDwAAAAAAYA2sjsT+YPsAST8h6SEabNyKiDNb1TRGlXDJ9oMXfX94j8gMqvQBABiPNZ4AAAAAAABrY3Uk9oc/k3S7pGu09153GZVYcTsMwzJP71bpAwAwHpN9AAAAAAAAa8i8OrLKNFkFtq/ONv22lirhUpXp3Sp9AACWw2QfAAAAAADAGrIGfb3002SFAssLbf9YRFzUupD96IWS0od9qjO9W6UPAMASCPsAAAAAAABqe2CBabL0gWXv45IusL1D0re09153O9uWtSFVwqWntS5gP6nSBwBgCazxBAAAAAAAKMz2r0v6YOZpsirrL23fIOkZkq6JIiflMq+4BQCgCib7AAAAAAAAFii0OrLCNFmV9Zf/R9KnqwR9UvoVtwAAlEDYBwAAAAAAsFiV1ZFnSnqcck+TVQgsJel6SZfavlDSN2ffTBggAwCACSHsAwAAAAAAWKzCve6kGtNkFQJLSbqh/3OX/g8aqjK9W6UPAMB4hH0AAAAAAACLVVkdWWGarEJgKUl/FhG7WxexUYXCpSrTu1X6AACMRNgHAAAAAACwWJXVkRWmySoElpL0etv3lfRHks6LiE+3LmikKuFSlendKn0AAEYi7AMAAAAAAFisyurICtNkFQJLRcTxfdj3XEn/w/ZOdaHfGY1LW1aVcKnK9G6VPgAAIzn376oAAAAAAACbw/blko6LiMyTS7J9iaTU02S2jy4QWO7D9mGSXi7p5IhIFWDa/nVJH8weLtl+lqR3Sko9vVulDwDAeIR9AAAAAAAAC9h+m6SHSsq+OlKDabKTJaWbJqsQWEqS7UPVPQbPkfQVSedJOj8i/qFpYUuqEi7ZvkHSM5R8erdKHwCA8Qj7AAAAAAAAFrB92qLvR8RrtrqW/SX5NFnqwFKSbH9MXcD3hxHx5db1jFUlXCo0vVuiDwDAeIR9AAAAAAAAC1RZHVllmmwmc2BZRZVwqcr0bpU+AADj3al1AQAAAAAAABP1+n6aLPXqSElvVRfwPTXrNNkKgeVLmxa1BNvXSFp0xf1s/eXhW1zSRl0v6VLb2cOlG/o/d+n/ZFWlDwDASEz2AQAAAAAArKDC6sgKsq+/tP3g1f4+Ir6wVbXsD1VW3Baa3i3RBwBgPMI+AAAAAACANWRcHVlwmgwTUSVcsn2JpPTTu1X6AACMR9gHAAAAAACwQPZ73VWYJqsSWNq+Rav3sXOLS9qQSuFSlendKn0AAMYh7AMAAAAAAFgg++rICioEllVVC5cyTu8uUqUPAMByCPsAAAAAAAAKqjZNhmnKHC5ln96dqdIHAGA8wj4AAAAAAICBKqsjKyCwnKYq4VKV6d0qfQAAxiPsAwAAAAAAGGB1JLA6wiUAAKaFsA8AAAAAAADAtlFlerdKHwCAjSPsAwAAAAAAGGB1JLBYlXCpyvRulT4AABtH2AcAAAAAAABgTYRLAABME2EfAAAAAAAAgG2jyvRulT4AABtH2AcAAAAAAABgTYRLAABME2EfAAAAAAAAAAAAkNSO1gUAAAAAAAAAAAAAGIewDwAAAAAAAAAAAEiKsA8AAAAAAAAAAABIirAPAAAAAAAAAAAASIqwDwAAAAAAAAAAAEjq/wNz+d3xWUYFigAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1800x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams[\"figure.figsize\"] = (25,8)\n",
    "axes = df_mae.plot.bar(subplots=False)\n",
    "for p in axes.patches:\n",
    "    axes.annotate('{:.3f}'.format(Decimal(str(p.get_height()))), (p.get_x(), p.get_height()))\n",
    "plt.title(workload + \": \"+metric+\" - MAE\")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "bcadcc94-2715-46b9-9b53-5104e2978c36",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1800x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams[\"figure.figsize\"] = (25,8)\n",
    "axes = df_mape.plot.bar(subplots=False)\n",
    "for p in axes.patches:\n",
    "    axes.annotate('{:.3f}'.format(Decimal(str(p.get_height()))), (p.get_x(), p.get_height()))\n",
    "plt.title(workload + \": \"+metric+\" - MAPE\")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "2e5a3651-0c2e-483a-b7b5-36a78f3c016f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1800x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams[\"figure.figsize\"] = (25,8)\n",
    "axes = df_sets.plot.bar(subplots=False)\n",
    "for p in axes.patches:\n",
    "    axes.annotate(p.get_height(), (p.get_x(), p.get_height()))\n",
    "plt.title(workload + \": \"+metric+\" - Set Sizes\")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "95790730-89ae-449c-849d-8fd8394e26f3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1800x576 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams[\"figure.figsize\"] = (25,8)\n",
    "axes = df_stats.plot.bar(subplots=True)\n",
    "#plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0b3311c5-3606-4207-a079-4f4174df3f26",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6eb6e744-89d8-43cc-a644-b4756709f198",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}