stsievert/a-incremental-model-selection.ipynb

## a-incremental-model-selection.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Incremental Model Selection\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook demos hyperparameter selection, and compares\n",
    "\n",
    "* \"stop on plateau\" random sampling\n",
    "* Successive halving, an adaptive algorithm that spends more time on high-performing models.\n",
    "\n",
    "Successive halving is a fancy early sampling criteria for \"stop on plateau\" random sampling.\n",
    "\n",
    "\n",
    "### Highlights\n",
    "(in order)\n",
    "\n",
    "* model input and output (has a good picture)\n",
    "* description of parameters to tune\n",
    "* Selecting hyperparameters for Hyperband\n",
    "* plots\n",
    "    * best_score vs [partial_fit_calls, wall_time]\n",
    "    * visualization of different parameters"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from sklearn.model_selection import ParameterSampler\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'/Users/ssievert/Developer/dask/dask/dask/__init__.py'"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import dask\n",
    "import dask.array as da\n",
    "dask.__file__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch.nn as nn\n",
    "from pprint import pprint\n",
    "import skorch\n",
    "import torch\n",
    "from toolz import partial"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'/Users/ssievert/Developer/stsievert/dask-ml/dask_ml/__init__.py'"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import dask_ml\n",
    "from dask_ml.model_selection._incremental import fit\n",
    "from dask_ml.model_selection import train_test_split\n",
    "dask_ml.__file__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "from model_selection_algs import stop_on_plateau"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/ssievert/Developer/mrocklin/distributed/distributed/__init__.py\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/ssievert/Developer/mrocklin/distributed/distributed/bokeh/core.py:52: UserWarning: \n",
      "Port 8787 is already in use. Perhaps you already have a cluster running?\n",
      "  warnings.warn('\\n' + msg)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table style=\"border: 2px solid white;\">\n",
       "<tr>\n",
       "<td style=\"vertical-align: top; border: 0px solid white\">\n",
       "<h3>Client</h3>\n",
       "<ul>\n",
       "  <li><b>Scheduler: </b>tcp://127.0.0.1:54602\n",
       "  <li><b>Dashboard: </b><a href='http://127.0.0.1:54603/status' target='_blank'>http://127.0.0.1:54603/status</a>\n",
       "</ul>\n",
       "</td>\n",
       "<td style=\"vertical-align: top; border: 0px solid white\">\n",
       "<h3>Cluster</h3>\n",
       "<ul>\n",
       "  <li><b>Workers: </b>8</li>\n",
       "  <li><b>Cores: </b>8</li>\n",
       "  <li><b>Memory: </b>17.18 GB</li>\n",
       "</ul>\n",
       "</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<Client: scheduler='tcp://127.0.0.1:54602' processes=8 cores=8>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import distributed\n",
    "from distributed import Client\n",
    "print(distributed.__file__)\n",
    "# client = Client(processes=False, n_workers=1)\n",
    "client = Client()\n",
    "client"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import noisy_mnist\n",
    "_X, _y = noisy_mnist.dataset()#n=10 * 1024)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(dask.array<array, shape=(70000, 784), dtype=float32, chunksize=(35000, 784)>,\n",
       " dask.array<array, shape=(70000, 784), dtype=float32, chunksize=(35000, 784)>)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n, d = _X.shape\n",
    "X = da.from_array(_X, chunks=(n // 2, d))\n",
    "y = da.from_array(_y, chunks=n // 2)\n",
    "X, y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The example I show in this notebook is a realistic model. It's an autoencoder that aims to reconstruct a noisy image (and one with a random amount of noise!). Here's an example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import Image\n",
    "Image('./autoencoder-input-output.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I use a deep learning library (PyTorch) for this model, at least through the scikit-learn interface for PyTorch, [skorch].\n",
    "\n",
    "[skorch]:https://github.com/dnouri/skorch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from autoencoder import Autoencoder"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Description of parameters\n",
    "I want to tune several hyperparameters.\n",
    "\n",
    "* `module__init`, which controls the initialization of the neural network.\n",
    "    * This is critical for performance -- a lot of optimization boils down to good initialization\n",
    "* `module__activation`, the type of non-linearity used in the neural net.\n",
    "    * The power of neural nets comes from their activation functions, and the base class I had chose ReLU.\n",
    "* `optimizer`, the type of optimizer used to minimize the loss (e.g., SGD, Adam).\n",
    "    * What optimizer should we choose?\n",
    "    * I choose ones found in [Adam's paper] in Figure 1, as well as ASGD (despite section 7.2). There is a paper titled \"[The marginal value of adaptive gradient methods in machine learning][1]\" by Ben Recht; is this true?\n",
    "* `batch_size`, the number of examples used to approximate the gradient with all examples.\n",
    "    * This is a critical for performance -- smaller batch sizes means more optimization steps, but lower quality steps.\n",
    "    * This is also important for generalization; small batches tend to generalize to unseen data better\n",
    "* `optimizer__lr`, the \"learning rate\" or step size for the optimizer\n",
    "    * How far should the optimizer step? Optimizers can find the direction to step from the gradient, but can't tell how far to step without the expensive function evaluation\n",
    "    * I don't tune this for Adam because it is fully adaptive.\n",
    "* `optimizer__weight_decay`, how fast the weights of the neural net decay to zero\n",
    "    * This is similar to L2 regularization\n",
    "    \n",
    "Then we have one parameter specific to SGD:\n",
    "\n",
    "* `optimizer__momentum`. We can think of optimization as a ball rolling down a hill, and this controls how heavy the ball is.\n",
    "    * Adam's paper compares with Nesterov SGD, so I have the same comparison.\n",
    "\n",
    "[1]:https://papers.nips.cc/paper/7003-the-marginal-value-of-adaptive-gradient-methods-in-machine-learning.pdf\n",
    "[Adam's paper]:https://arxiv.org/pdf/1412.6980.pdf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "params = {\n",
    "    'module__init': ['xavier_uniform_',\n",
    "                     'xavier_normal_',\n",
    "                     'kaiming_uniform_',\n",
    "                     'kaiming_normal_',\n",
    "                    ],\n",
    "    'module__activation': ['ReLU', 'LeakyReLU', 'ELU', 'PReLU'],\n",
    "    'optimizer': ['SGD',\n",
    "                  'ASGD',\n",
    "                  'Adam',\n",
    "                  'Adagrad',\n",
    "                  'RMSprop'],  # optimizers in Adam's paper + ASGD\n",
    "    'batch_size': [32, 64, 128, 256, 512],\n",
    "    'optimizer__lr': np.logspace(0, -3, num=1000),  # all optimizers but Adam\n",
    "    'optimizer__weight_decay': [0] + np.logspace(-7, -3, num=1000).tolist(),  # all optimizers\n",
    "    'optimizer__nesterov': [True],  # only for SGD\n",
    "    'optimizer__momentum': np.linspace(0, 1, num=1000),  # only for SGD\n",
    "    'train_split': [None],\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I think if I did this again (I probably will), I'd reduce `optimizer` to be `SGD` and `Adam`, and keep nesterov and momentum/step size tuning.\n",
    "\n",
    "I might reduce the search space for `weight_decay` a bit.\n",
    "\n",
    "This function enforces the rules I described above (i.e., don't tune step size for Adam)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "from distributed.utils import log_errors\n",
    "\n",
    "def trim_param(*, verbose=False, **kwargs):\n",
    "    if verbose:\n",
    "        pprint(kwargs)\n",
    "    if kwargs['optimizer'] != 'Adam':\n",
    "        kwargs.pop('optimizer__amsgrad', None)\n",
    "    if kwargs['optimizer'] == 'Adam':\n",
    "        kwargs.pop('optimizer__lr', None)\n",
    "    if kwargs['optimizer'] != 'SGD':\n",
    "        kwargs.pop('optimizer__nesterov', None)\n",
    "        kwargs.pop('optimizer__momentum', None)\n",
    "    kwargs['optimizer'] = getattr(torch.optim, kwargs['optimizer'])\n",
    "\n",
    "    if verbose:\n",
    "        pprint({k: v for k, v in kwargs.items() if k != 'module'})\n",
    "    return kwargs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from autoencoder import NegLossScore\n",
    "model = NegLossScore(module=Autoencoder,\n",
    "                     criterion=torch.nn.BCELoss,\n",
    "                     warm_start=True,\n",
    "                     train_split=None,\n",
    "                     max_epochs=1,\n",
    "                     callbacks=[])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This cell is for quick evaluations. The PyTorch model takes a while to run, this doesn't."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# from sklearn.linear_model import SGDClassifier\n",
    "# from sklearn.datasets import make_classification\n",
    "# model = SGDClassifier()\n",
    "# params = {'alpha': np.logspace(-7, 0, num=1000)}\n",
    "\n",
    "# n, d = int(10e3), 700\n",
    "# _X, _y = make_classification(n_samples=n, n_features=d,\n",
    "#                              random_state=1)\n",
    "# X = da.from_array(_X, chunks=(n // 10, d))\n",
    "# y = da.from_array(_y, chunks=n // 10)\n",
    "# X, y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = client.scatter(model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Successive halving"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_history = {}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create the different brackets for successive halving. Every bracket is run \"embarassingly parallel\"."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Selecting hyperparameters for Hyperband\n",
    "We need about 20 epochs from experiment to train until convergence (well, 5 or 6 for a subset of possible parameters).\n",
    "\n",
    "We are searching over 7 parameters. We have 400 combinations for the 5 discrete variables, and are tuning 1 contiuous variable for all optimizers (weight decay) and 1 for 4 out of 5 optimizers (for Adam, don't tune learning rate).\n",
    "\n",
    "Let's sample say 20% of this space – I think that will give a good classifier. This will tune any one of the discrete variables if one value singled out (e.g., `optimizer=Adam` specifies 20% of the discrete space).\n",
    "\n",
    "We need 20 epochs, and want to sample 100 parameters. Because we're using Hyperband, that means we also have 100 `partial_fit` calls on each chunk of the dataset for the most trained model. That means we need `20 epochs = (x epochs/call) * 100 calls`, so `x = 0.2`."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In summary,\n",
    "\n",
    "* select $R = 100$ for the complexity of the search space\n",
    "* guess we need 20 epochs for convergence from some prior experience\n",
    "* divide the dataset with `chunks` so `frac = chunks / len(dataset)`.\n",
    "    * To do this, use the fact that `frac * R = epochs` for Hyperband"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(81.0, 1.0), (34.0, 3.0), (15.0, 11.0), (8.0, 33.0), (5.0, 100.0)]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import math\n",
    "R = 100 # 81 #// 3\n",
    "eta = 3.0\n",
    "s_max = math.floor(math.log(R, eta))\n",
    "B = (s_max + 1) * R\n",
    "# for s in [...]:\n",
    "s = np.arange(s_max + 1)\n",
    "s = s_max - s  # pick the most exploratory bracket\n",
    "n = np.ceil(B / R * eta**s / (s + 1))\n",
    "r = np.floor(R * eta**-s)\n",
    "list(zip(n, r))\n",
    "#n.sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For practicalities sake, let's use `patience=10` with `stop_on_plateau` -- hopefully it finishes quicker."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import ParameterSampler\n",
    "from model_selection_algs import SHA\n",
    "import math\n",
    "from pprint import pprint\n",
    "\n",
    "def test_sha(s=0, eta=3, n=81, r=5, repeat=0):\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y)\n",
    "    \n",
    "    alg = SHA(n, r, patience=10)\n",
    "    params_list = list(ParameterSampler(params, n))\n",
    "    if 'optimizer' in params:\n",
    "        params_list = [trim_param(**param) for param in params_list]\n",
    "        fit_params = {}\n",
    "    else:\n",
    "        fit_params = {'classes': np.unique(_y)}\n",
    "\n",
    "    fn = alg.fit\n",
    "    info, models, history = fit(model, params_list, X_train, y_train, X_test, y_test,\n",
    "                                fn, fit_params=fit_params)\n",
    "    history = [{'alg': f'sh-{s}', 'repeat': repeat, 'bracket': s, **h}\n",
    "               for h in history]\n",
    "    return history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "new iteration. bracket, n, r = 0 81.0 1.0\n",
      "new iteration. bracket, n, r = 1 34.0 3.0\n",
      "new iteration. bracket, n, r = 2 15.0 11.0\n",
      "new iteration. bracket, n, r = 3 8.0 33.0\n",
      "new iteration. bracket, n, r = 4 5.0 100.0\n"
     ]
    }
   ],
   "source": [
    "# _ = test_sha(n=ni, r=ri, s=s)\n",
    "# for s, (ni, ri) in enumerate(reversed(list(zip(n, r)))):\n",
    "for s, (ni, ri) in enumerate(zip(n, r)):\n",
    "    print(\"new iteration. bracket, n, r =\", s, ni, ri)\n",
    "    all_history[f'sh-{s}'] = [test_sha(repeat=repeat, n=ni, r=ri, s=s) \n",
    "                              for repeat in range(1)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Random sampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from toolz import partial\n",
    "\n",
    "def test_rand(i=0, n=5, r=81):\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y)\n",
    "    classes = da.unique(y).compute()\n",
    "\n",
    "    params_list = list(ParameterSampler(params, n))\n",
    "    fit_params = {'classes': np.unique(_y)}\n",
    "    if 'optimizer' in params:\n",
    "        params_list = [trim_param(**param) for param in params_list]\n",
    "        fit_params = {}\n",
    "\n",
    "    fn = partial(stop_on_plateau, max_iter=r)\n",
    "    info, models, history = fit(model, params_list, X_train, y_train, X_test, y_test, fn,\n",
    "                                fit_params=fit_params)\n",
    "    history = [{'alg': 'random sampling', 'repeat': i, 'n': n, 'r': r, **h}\n",
    "               for h in history]\n",
    "    return history"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This cell gets random sampling ready to run the same number of total iterations as all the sum of successive halving brackets.\n",
    "\n",
    "This prepratation assumes that the models will plateau."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1810, 100)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from toolz import concat\n",
    "from pprint import pprint\n",
    "\n",
    "def get_total_calls(history):\n",
    "    total_calls = {}\n",
    "    for item in history:\n",
    "        if total_calls.get(item['model_id'], 0) < item['partial_fit_calls']:\n",
    "            total_calls[item['model_id']] = item['partial_fit_calls']\n",
    "    return sum(total_calls.values()), max(total_calls.values())\n",
    "\n",
    "max_calls = 0\n",
    "total_calls = 0\n",
    "for key, runs in all_history.items():\n",
    "    run = runs[0]\n",
    "    calls, most_calls = get_total_calls(run)\n",
    "    if most_calls > max_calls:\n",
    "        max_calls = most_calls\n",
    "    total_calls += calls\n",
    "    \n",
    "if total_calls == 0:\n",
    "    total_calls = 100\n",
    "if max_calls == 0:\n",
    "    max_calls = 10\n",
    "elif max_calls == 100:\n",
    "    total_calls = 1810\n",
    "total_calls, max_calls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(18, 100)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# r = max_calls\n",
    "r = R\n",
    "n = total_calls // r\n",
    "if n == 0:\n",
    "    n = 1\n",
    "n, r"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# n, r = 5, 81\n",
    "all_history['random'] = [test_rand(n=n, r=r, i=i) for i in range(1)]\n",
    "# all_history['random-2n'] = [test_rand(n=2*n, r=r, i=i) for i in range(20)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['sh-0', 'sh-1', 'sh-2', 'sh-3', 'sh-4', 'random'])"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_history.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import msgpack\n",
    "from distributed.protocol import serialize, deserialize\n",
    "# with open('./2018-08-22-hyperband-run.msgpack', 'w') as f:\n",
    "#     print(serialize(all_history), file=f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualization\n",
    "We want to visualize\n",
    "\n",
    "* comparison with successive halving\n",
    "    * best score vs {wall time, `partial_fit` calls}\n",
    "* the parameters; which ones performed best?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "sh-0 : avg pf_time = 8.2e+03ms\n",
      "sh-1 : avg pf_time = 4.5e+03ms\n",
      "sh-2 : avg pf_time = 1.6e+03ms\n",
      "sh-3 : avg pf_time = 7e+02ms\n",
      "sh-4 : avg pf_time = 1.8e+02ms\n",
      "random : avg pf_time = 1.5e+04ms\n"
     ]
    }
   ],
   "source": [
    "from pprint import pprint\n",
    "out = []\n",
    "for alg, hist in all_history.items():\n",
    "    for repeat, batch in enumerate(hist):\n",
    "        best_score = -np.inf\n",
    "        calls = {}\n",
    "        partial_fit_time = 0\n",
    "        for k, item in enumerate(batch):\n",
    "            if 'score' not in item:\n",
    "                continue\n",
    "            calls[item['model_id']] = item['partial_fit_calls']\n",
    "            partial_fit_time += item['partial_fit_time']\n",
    "            total_calls = sum(calls.values())\n",
    "            best_score = item['score'] if item['score'] > best_score else best_score\n",
    "            if 'random' in alg or alg == 'sh-0':\n",
    "                adaptive = False\n",
    "                bracket = 0\n",
    "            else:\n",
    "                adaptive = True\n",
    "                bracket = int(alg.split('-')[-1])\n",
    "\n",
    "            out += [{'best_score': best_score,\n",
    "                     'total_partial_fit_calls': total_calls,\n",
    "                     'partial_fit_time': partial_fit_time,\n",
    "                     'scored_at': item['scored_at'],\n",
    "                     'alg': alg,\n",
    "                     'epoch': total_calls / n,\n",
    "                     'adaptive': adaptive,\n",
    "                     'bracket': bracket,\n",
    "                     'repeat': repeat}]\n",
    "    print(alg, \": avg pf_time = {:0.2}ms\".format(1e3 * partial_fit_time / sum(calls.values())))\n",
    "          "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "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>bracket</th>\n",
       "      <th>alg</th>\n",
       "      <th>score</th>\n",
       "      <th>model_id</th>\n",
       "      <th>partial_fit_calls</th>\n",
       "      <th>i</th>\n",
       "      <th>adaptive</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>sh-0</td>\n",
       "      <td>-0.199063</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>sh-0</td>\n",
       "      <td>-0.269956</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>sh-0</td>\n",
       "      <td>-0.289043</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>sh-0</td>\n",
       "      <td>-0.155386</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>sh-0</td>\n",
       "      <td>-0.267313</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   bracket   alg     score  model_id  partial_fit_calls    i  adaptive\n",
       "0      0.0  sh-0 -0.199063         0                  1  0.1       0.1\n",
       "1      0.0  sh-0 -0.269956         1                  1  0.1       0.1\n",
       "2      0.0  sh-0 -0.289043         2                  1  0.1       0.1\n",
       "3      0.0  sh-0 -0.155386         3                  1  0.1       0.1\n",
       "4      0.0  sh-0 -0.267313         4                  1  0.1       0.1"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import altair as alt\n",
    "hist = pd.DataFrame(sum(sum(all_history.values(), []), []),\n",
    "                    columns=['bracket', 'alg', 'score', 'model_id',\n",
    "                             'partial_fit_calls', 'i', 'adaptive'])\n",
    "hist = hist.fillna(0.1)\n",
    "hist.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "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>adaptive</th>\n",
       "      <th>alg</th>\n",
       "      <th>best_score</th>\n",
       "      <th>bracket</th>\n",
       "      <th>epoch</th>\n",
       "      <th>partial_fit_time</th>\n",
       "      <th>repeat</th>\n",
       "      <th>scored_at</th>\n",
       "      <th>total_partial_fit_calls</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>False</td>\n",
       "      <td>sh-0</td>\n",
       "      <td>-0.199063</td>\n",
       "      <td>0</td>\n",
       "      <td>0.055556</td>\n",
       "      <td>32.572223</td>\n",
       "      <td>0</td>\n",
       "      <td>104.562557</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>False</td>\n",
       "      <td>sh-0</td>\n",
       "      <td>-0.199063</td>\n",
       "      <td>0</td>\n",
       "      <td>0.111111</td>\n",
       "      <td>36.594511</td>\n",
       "      <td>0</td>\n",
       "      <td>24.980749</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>False</td>\n",
       "      <td>sh-0</td>\n",
       "      <td>-0.199063</td>\n",
       "      <td>0</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>42.498756</td>\n",
       "      <td>0</td>\n",
       "      <td>28.205758</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>False</td>\n",
       "      <td>sh-0</td>\n",
       "      <td>-0.155386</td>\n",
       "      <td>0</td>\n",
       "      <td>0.222222</td>\n",
       "      <td>50.156536</td>\n",
       "      <td>0</td>\n",
       "      <td>128.091547</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>False</td>\n",
       "      <td>sh-0</td>\n",
       "      <td>-0.155386</td>\n",
       "      <td>0</td>\n",
       "      <td>0.277778</td>\n",
       "      <td>59.471542</td>\n",
       "      <td>0</td>\n",
       "      <td>105.600732</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   adaptive   alg  best_score  bracket     epoch  partial_fit_time  repeat  \\\n",
       "0     False  sh-0   -0.199063        0  0.055556         32.572223       0   \n",
       "1     False  sh-0   -0.199063        0  0.111111         36.594511       0   \n",
       "2     False  sh-0   -0.199063        0  0.166667         42.498756       0   \n",
       "3     False  sh-0   -0.155386        0  0.222222         50.156536       0   \n",
       "4     False  sh-0   -0.155386        0  0.277778         59.471542       0   \n",
       "\n",
       "    scored_at  total_partial_fit_calls  \n",
       "0  104.562557                        1  \n",
       "1   24.980749                        2  \n",
       "2   28.205758                        3  \n",
       "3  128.091547                        4  \n",
       "4  105.600732                        5  "
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "df = pd.DataFrame(out)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "alg\n",
       "random    -0.706429\n",
       "sh-0      -0.199063\n",
       "sh-1      -0.360289\n",
       "sh-2      -0.314479\n",
       "sh-3      -0.622155\n",
       "sh-4     -12.406262\n",
       "Name: best_score, dtype: float64"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from datetime import datetime\n",
    "today = datetime.now().isoformat()[:10]\n",
    "# df.to_parquet(today + '-df-even.parquet')\n",
    "# hist.to_parquet(today + '-show-even.parquet')\n",
    "\n",
    "#df.head()\n",
    "df.groupby('alg')['best_score'].min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "plt.style.use('default')\n",
    "\n",
    "_algs = list(df.alg.unique())\n",
    "_algs = sorted(_algs)\n",
    "sns.lineplot(\n",
    "    data=df,\n",
    "    x='total_partial_fit_calls',\n",
    "    y='best_score',\n",
    "    style='adaptive', hue='alg',\n",
    "    hue_order=_algs,\n",
    "    style_order=[True, False],\n",
    "    palette='viridis',\n",
    "    estimator=np.median,\n",
    ")\n",
    "plt.ylim(-0.15, -0.08)\n",
    "plt.grid(linestyle='--')\n",
    "# plt.savefig('autoencoder-pf-calls.png', dpi=300, bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "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>adaptive</th>\n",
       "      <th>alg</th>\n",
       "      <th>best_score</th>\n",
       "      <th>bracket</th>\n",
       "      <th>epoch</th>\n",
       "      <th>partial_fit_time</th>\n",
       "      <th>repeat</th>\n",
       "      <th>scored_at</th>\n",
       "      <th>total_partial_fit_calls</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>False</td>\n",
       "      <td>sh-0</td>\n",
       "      <td>-0.199063</td>\n",
       "      <td>0</td>\n",
       "      <td>0.055556</td>\n",
       "      <td>32.572223</td>\n",
       "      <td>0</td>\n",
       "      <td>104.562557</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>False</td>\n",
       "      <td>sh-0</td>\n",
       "      <td>-0.199063</td>\n",
       "      <td>0</td>\n",
       "      <td>0.111111</td>\n",
       "      <td>36.594511</td>\n",
       "      <td>0</td>\n",
       "      <td>24.980749</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>False</td>\n",
       "      <td>sh-0</td>\n",
       "      <td>-0.199063</td>\n",
       "      <td>0</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>42.498756</td>\n",
       "      <td>0</td>\n",
       "      <td>28.205758</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>False</td>\n",
       "      <td>sh-0</td>\n",
       "      <td>-0.155386</td>\n",
       "      <td>0</td>\n",
       "      <td>0.222222</td>\n",
       "      <td>50.156536</td>\n",
       "      <td>0</td>\n",
       "      <td>128.091547</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>False</td>\n",
       "      <td>sh-0</td>\n",
       "      <td>-0.155386</td>\n",
       "      <td>0</td>\n",
       "      <td>0.277778</td>\n",
       "      <td>59.471542</td>\n",
       "      <td>0</td>\n",
       "      <td>105.600732</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   adaptive   alg  best_score  bracket     epoch  partial_fit_time  repeat  \\\n",
       "0     False  sh-0   -0.199063        0  0.055556         32.572223       0   \n",
       "1     False  sh-0   -0.199063        0  0.111111         36.594511       0   \n",
       "2     False  sh-0   -0.199063        0  0.166667         42.498756       0   \n",
       "3     False  sh-0   -0.155386        0  0.222222         50.156536       0   \n",
       "4     False  sh-0   -0.155386        0  0.277778         59.471542       0   \n",
       "\n",
       "    scored_at  total_partial_fit_calls  \n",
       "0  104.562557                        1  \n",
       "1   24.980749                        2  \n",
       "2   28.205758                        3  \n",
       "3  128.091547                        4  \n",
       "4  105.600732                        5  "
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk8AAAG2CAYAAABmsmIiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3Xl8VNX5+PHPbFnJShIgkIR9E1llSVQ0oIBIcWlVSsWN0taKuLZV+63FWqW/tlpbse4bolir1qVgBJVF9jXspOxhCyRkJZNMZu7M74/JTDLJTMhMZrmTPO/Xq686Z869OSdPhjw59ywam81mQwghhBBCtIo21A0QQgghhAgnkjwJIYQQQnhBkichhBBCCC9I8iSEEEII4QVJnoQQQgghvCDJkxBCCCGEFyR5EkIIIYTwgiRPQgghhBBekORJCCGEEMILkjwJIYQQQnghbJKnsrIyZs2aRUJCAgkJCcyaNYvy8vIWrzGZTNx///2kpKQQGxvL9OnTOXnypEudLVu2MHHiRBITE0lKSmLSpEnk5+cHsitCCCGECGNhkzzNnDmT/Px88vLyyMvLIz8/n1mzZrV4zYMPPsh//vMfPvzwQ9auXcuFCxeYNm0aiqIAUFVVxeTJk8nMzGTTpk2sXbuW+Ph4Jk+ejNlsDka3hBBCCBFmNOFwMPD+/fsZPHgwGzduZOzYsQBs3LiR7OxsDhw4wIABA5pdU1FRQWpqKu+99x633XYbAKdPnyYjI4Nly5YxefJktm7dyujRoyksLCQjIwOA3bt3M3ToUA4dOkSfPn2C10khhBBChAV9qBvQGhs2bCAhIcGZOAGMGzeOhIQE1q9f7zZ52rZtG2azmUmTJjnL0tPTGTJkCOvXr2fy5MkMGDCAlJQU3nzzTZ544gkUReHNN9/kkksuISsry2N7TCYTJpPJ+dpqtVJaWkrnzp3RaDR+6rUQQgghAslms1FVVUV6ejpabesfxoVF8lRUVERaWlqz8rS0NIqKijxeExERQVJSkkt5ly5dnNfExcWxatUqbrjhBp5++mkA+vfvz9dff41e7/lbs2DBAp566ilfuyOEEEIIFTlx4gQ9evRodf2QJk/z58+/aBKyZcsWALcjOjabzeuRnsbX1NTUcM8993D55ZezZMkSFEXhr3/9K1OnTmXLli1ER0e7vcfjjz/Oww8/7HxdUVFBZmYmR48eJS4uDgCtVotOp0NRFKxWq7Ouo9xisdD4ialOp0Or1bqUWywWdu7cyWWXXUbTp6uO5M5isbSq3GAwYLVanfO9wP491ev1Hss9tb0tfWpc3nRemdr71DgeQLvoU+O2h1OcLBYL+fn5jBw5ksjIyHbRp6bl4dSnpvFoD31q2vZw6ZPJZCI/P5/hw4djMBjaRZ8CGafS0lJ69erl/N3dWiFNnubOncuMGTNarNOzZ0927drF2bNnm71XXFxMly5d3F7XtWtX6urqKCsrcxl9OnfuHDk5OQB88MEHHDt2jA0bNjiH6z744AOSkpL4/PPPPbYtMjKSyMjIZuXJycnEx8e32B9vmM1mampqiIuLw2Aw+O2+wjcSD/Uwm83U1taSmJgosVABiYd6OGKRnJwssfCCtwMxIU2eUlJSSElJuWi97OxsKioq2Lx5M2PGjAFg06ZNVFRUOBOhpkaNGoXBYGDFihXceuutAJw5c4Y9e/bw5z//GQCj0YhWq3X5pjleN85chRBCCCEcwmKrgkGDBjFlyhTmzJnDxo0b2bhxI3PmzGHatGnOyeKnTp1i4MCBbN68GYCEhARmz57NI488wrfffsuOHTu4/fbbufTSS7nmmmsAuPbaaykrK+O+++5j//797N27l7vvvhu9Xk9ubm7I+iuEEEII9QqL5Ang/fff59JLL2XSpElMmjSJoUOH8t577znfN5vNFBQUYDQanWV/+9vfuPHGG7n11lu5/PLLiYmJ4csvv0Sn0wEwcOBAvvzyS3bt2kV2djZXXnklp0+fJi8vj27dugW9j03pdDqGDx/ubK8ILYmHekgs1EXioR4Si+AIi32e1K6yspKEhAQqKir8OudJCCGEEIHj6+/vsBl56ogsFgvfffdds5UJIjQkHuohsVAXiYd6SCyCQ5InFXNs3iWDg+og8VAPiYW6SDzUQ2IRHJI8CSGEEEJ4QZInIYQQQggvSPKkYjqdjuzsbFk1oRISD/WQWKiLxEM9JBbBIavt/EBW2wkhhBDhR1bbtUNms5mlS5c2OzdIhIbEQz0kFuoi8VAPiUVwSPKkcrLcVF0kHuohsVAXiYd6SCwCT5InIYQQQggvSPIkhBBCCOEFmTDuB4GaMO7Y7CwuLg6NRuO3+wrfSDzUQ2KhLhIP9ZBYeEcmjLdT0dHRoW6CaETioR4SC3WReKiHxCLwJHlSMYvFwrJly2Tyn0pIPNRDYqEuEg/1kFgEhyRPQgghhBBekORJCCGEEMILkjwJIYQQQnhBVtv5QSBX21ksFvR6vayaUAGJh3pILNRF4qEeEgvvyGq7dqqmpibUTRCNSDzUQ2KhLhIP9ZBYBJ4kTypmsVhYuXKlrJpQCYmHekgs1EXioR4Si+CQ5EkIIYQQwguSPAkhhBBCeEGSJ5XT6/WhboJoROKhHhILdZF4qIfEIvBktZ0fBGq1nRBCCCECR1bbtUNWq5Vz585htVpD3RSBxENNJBbqIvFQD4lFcEjypGKKorBhwwYURQl1UwQSDzWRWKiLxEM9JBbBIcmTEEIIIYQXJHkSQgghhPCCJE8qptFoiIuLky32VULioR4SC3WReKiHxCI4ZLWdH8hqOyGEECL8yGq7dshqtXL8+HFZNaESEg/1kFioi8RDPSQWwSHJk4opikJ+fr6smlAJiYd6SCzUReKhHhKL4JDkSQghhBDCC5I8CSGEEEJ4QZInFdNoNKSmpsqqCZWQeKiHxEJdJB7qIbEIDllt5wey2k4IIYQIP7Larh1SFIUDBw7IxD+VkHioh8RCXSQe6iGxCA59qBsgPLNarRQUFNCnTx90Ol2om9PheYqHyWLBpFiC0oa4iEgZjkc+G2oj8VAPiUVwSPIkRBtsPHmCuz77hDprcP7KO3Dfg0TIP4hCCBFS8thOCB/ZbDb+uv77oCVOQggh1EFGnlRMq9WSmZmJVis5rho0jcfmUyfZXnQGvVbLyjtmkxobG/A2GORnAZDPhtpIPNRDYhEcstrOD2S1Xcd052cf833hcWZeOow/5l4T6uYIIYTwkqy2a4cURWHHjh2yakIlGsdj19kivi88jk6j4WcjLwt10zoc+Wyoi8RDPSQWwRE2yVNZWRmzZs0iISGBhIQEZs2aRXl5eYvXmEwm7r//flJSUoiNjWX69OmcPHnSpc63335LTk4OcXFxdOvWjd/85jdYLMFZOXUxVquVwsJCOeBRJRrH4+WtmwH4Qf+BZCYkhrhlHY98NtRF4qEeEovgCJvkaebMmeTn55OXl0deXh75+fnMmjWrxWsefPBB/vOf//Dhhx+ydu1aLly4wLRp05wZ+a5du5g6dSpTpkxhx44dfPjhh3zxxRc89thjweiSCFOHSs/z9eGDANx72dgQt0YIIUSwhcWE8f3795OXl8fGjRsZO9b+y+r1118nOzubgoICBgwY0OyaiooK3nzzTd577z2uucY+H2Xx4sVkZGTwzTffMHnyZD788EOGDh3Kk08+CUDfvn1ZsGABP/7xj/n9739PXFxc8DopwsbrO7YBMKlPX/p17hzi1gghhAi2sEieNmzYQEJCgjNxAhg3bhwJCQmsX7/ebfK0bds2zGYzkyZNcpalp6czZMgQ1q9fz+TJkzGZTERFRblcFx0dTW1tLdu2bePqq6922x6TyYTJZHK+rqysBMBsNmM2mwH7igedToeiKC7Dp45yi8VC47n6Op0OrVbrUq4oCv3790er1Trv66DX20PX9BGjp3KDwYDVanV5Dq7RaNDr9R7LPbW9LX1qXB5ufVIUhYSMHny5bjUAc4aPcvYhXPvUuO3hFCdFUejbt6+zve2hT03Lw6lPTePRHvrUtO3h1Ke+ffuiKEq76lPTcn/2yRdhkTwVFRWRlpbWrDwtLY2ioiKP10RERJCUlORS3qVLF+c1kydP5oUXXmDJkiXceuutFBUV8cc//hGAM2fOeGzPggULeOqpp5qVL1++nJiYGAAyMzMZMWIEu3btorCw0FlnwIABDBw4kM2bN1NcXOwsHz58OFlZWaxZs4aqqipneXZ2Njqdjry8PJcg5+bmEh0dzbJly1zaMHXqVGpqali5cqWzTK/Xc/3111NSUsKGDRuc5XFxcUyYMIETJ06Qn5/vLE9NTSUnJ4eDBw9SUFDgLPdnn9LS0li+fHnY9WmlxopiszE4No6TW7fhmEEXzn0K5zjpdLp216dwjlNdXV2761O4xunQoUPtrk/g/zitW7cOX4R0q4L58+e7TUIa27JlC8uXL+fdd991+UYB9OvXj9mzZ7udo/TBBx9w9913u4wQAVx77bX06dOHV155BYDnn3+ep556iurqaiIjI/nd737H448/zr/+9S9uvfVWt21yN/KUkZFBSUmJc6mjPzJ7i8XC9u3bGTt2LE3DFG6ZfXv4a+VMZQXXfrAIs83Gezf8kDHp3cO+T43bHk5xslgsbNu2jdGjRxMZGdku+tS0PJz61DQe7aFPTdseLn0ymUxs27aNUaNGYTAY2kWfAhmn0tJSOnfu7PVWBSEdeZo7dy4zZsxosU7Pnj3ZtWsXZ8+ebfZecXExXbp0cXtd165dqauro6yszGX06dy5c+Tk5DhfP/zwwzz00EOcOXOGpKQkjh07xuOPP06vXr08tikyMpLIyMhm5QaDAYPB4FKm0+ncni/k+CG6WHlJSQk2m63ZfRt/zdaWa7VatxuneSr31Pa29qmlNnpbHsw+vbd3N2abjRFdupKTmeX2jLlw61NrytXap/PnzzuvbS99aizc+tQ4Hu2lT60pV1uf9Ho958+fR6/XO9sQ7n0KRZwuJqTJU0pKCikpKRetl52dTUVFBZs3b2bMmDEAbNq0iYqKCpdEqDFH1r1ixQrnCNKZM2fYs2cPf/7zn13qajQa0tPTAViyZAkZGRmMHDmyLV0T7Ux5bQ1L9u4G4OcjR8vhvEII0YGFxZynQYMGMWXKFObMmcOrr74KwM9+9jOmTZvmnCx+6tQpJk6cyKJFixgzZgwJCQnMnj2bRx55hM6dO5OcnMyjjz7KpZde6lx9B/CXv/yFKVOmoNVq+fTTT/nTn/7ERx99JKdRCxeLduZjNJtJN0RwdVbPUDdHCCFECIVF8gTw/vvvM2/ePOfquenTp7Nw4ULn+2azmYKCAoxGo7Psb3/7G3q9nltvvZWamhomTpzIO++845IYffXVVzzzzDOYTCaGDRvG559/znXXXRe8jrVAp9MxfPhwSeRCrLqujnd2bgdg9rARPg/zCv+Rz4a6SDzUQ2IRHHK2nR/I2Xbt2xvbt/Ls2tVkJSTyzay70cmBm0II0S7I2XbtkMVi4bvvvlPNcTEdkcli4Y0dWwH42YhRrF61SuKhAvLZUBeJh3pILIJDnj+omM1mo6qqqtk2BSJ4Pj2wj3PV1XTr1Inp/QfyzdGvVRePysoa3n/7e04UnneWRUTomb/glhC2KrDks6EuEg/1kFgEhyRPQnhgsVp5tf4A4J+OHE2ECucQrFtdwN//soyy0mqX8qho90uEhRBCtJ0kT0J4sPRgAYWVFSRHRXPbJZeGujnN/PuDjby28BsAMrI686MZY9Eb7AmeTidP5IUQIlAkeVIxnU7nPJ5FBJfVZuPlLZsAuHvESGLqd8BVUzyumjiIJe+uZeoNI7lj9ngiIjvOx1k+G+oi8VAPiUVwyGo7P5DVdu3PisOH+PnSz+lkiGDtPXOIj4y6+EUBVl5Wzfrv/8fU6SOcZReqaukUF/q2CSFEOJLVdu2Q2Wxm6dKlzc4NEoFls9n451b7qNPtQ4c7E6dQxmPNd/uZ85NX+duflrJ102FneUdNnOSzoS4SD/WQWARHxxnnD1Oy3DT41p8sZOfZIiJ1eu4ZMcrlvWDHo6y0moXP57Hmu/0A9OydSkJiTFDboFby2VAXiYd6SCwCT5InIZr45xb7CrsZQy4lJSY0iYrNZmP1t/tY+PzXVJQb0eo0/PiOy5l55xVERMjHVgghQkn+FRaikR1nTrPhZCF6rZafjrwsZO34+1++Yuln9iNhevdN49Hf/oB+A7qFrD1CCCEaSPKkYnq9ntzcXDlLLYhert/X6caBg+ge5zp5UK/Xc9mocZw+WU5Wr1QArFYbu3Yc93i/hMQYevVJc77O33bMY924+Gj69OsCwKjRvcj7Mp+Zd17Oj++8AoNBVs40Jp8NdWlv8diz7gAmo8nte5HREQy5YpDz9b6N/6OmqsZtXX2EnmFXXeJ8XbDlEBfKq93W1eq0jJjQsCXKwe1HqDxf5bGNo64d5vzvwzuPUX6uAgCbDZJJY+fKfWg09veHTxjiXH13dE8hpWfKPN536FWDMUTY94k7vv8kJSfPe6w75IqBREZHAnCi4BTnCks81h2cM4DoWPsczVOHzlB09JzHugPH9iM23j7qf+boWU4fKiLrkgxS0pM9XhNs7eMnvR2Ljo4OdRM6jAMlxXxz9DAa4BejxjR732q18cRDHzNoSHce+/2NAFgsCr+6f7HHe15+1QCXnb5/88D7WK3uF7iOGtObP70wE4Arcwfx9r+60i09qQ09at/ks6Euao5H2bkK/vng285kZPuKXcR3jkOr1WCus3DzA9fzk9/90JlgPP/TlzlRcNrtvbr2SuO9wy85Xy+c+wYHtx91WzepSwIfnXnD+frVRxex+/v9butGd4rii8r3nK/f+u0HbP16p9u6Go2G5cpHzteLn/6YtZ9uclsXYKnxfWffPvrL53zz3hqPdT8pfgtDZ3vy9J+/L2Ppays81n3/2D9Jy7T/Ibn0tW/45G//9Vj3zX0vkDmwOwArFq3m/T9+4rHuS1v+RP9RfQBY/a/1vPnEB/zq7fuYdOfVHq8JNkmeVMxisbBs2TKmTp2KwSA7RgfaK9vso07X9e1P76Tmf+Hs2XWc06fKqKgwoihWdDotGo2GnvWjUO6kprmOXmX1SsXmIXnq0jXB5bUkTp7JZ0NdQh2PT/72X3avbUhKThacpvRMGfEp8bxT8A9MRhNDrxzEp39fysn/nQFwGdV57w//5tShMzy++AEAMgZ2xxDlvh8p3V3/bejePx1FsbqtG985zuV1ep+uVFca3daNiol0ed2tVxd6D8tyW1fjGFKq1yUrtaGuzb78Pj4+HjTOC5x10zJSPN4X7CNgDqk9OrdYV2doSCE6pye3WNfQaB+65K5JLdaNjI5w/ndiWgK9h2XRKSnWY/1QkH2e/CBQ+zyZzWb5BREkx8vLmfjeW1htNr6ccTuXpHVpVueVfyznkw83c/U1g/ntH24OQSuFg3w21CUQ8agzmblQdoHl767G2CjhWPfZZnoPzeK3Sx4CoKKkkh+lzXZ7j5i4aD6vWATAhfJqNi3d7nxPp9fSc0gmYM8tuvfrht4Q/uMJ8tnwjq+/v8P/J0UIP3ht+xasNhtXZfVymzjZbDbWrfkfADnj+we7eUK0W+fPlFFzodb5Oi2jMxFREWxfsYvfTf+T22t0eh3H958ka1AP4jvHEd0pipoLtcz75xxnnYgoA0OuGOh83Skxlok/uTJwHREdiiRPosMrulDFJ/v2AnDf6LFu6xw5dI6i0+Xo9BouG9MrmM0TImwd2HyQM0fOERMfzdipIwEoO1vOi3Pt84C2rdiFsdJ1svWLG59l4Jh+LmWdEmO5ZtZ45+vI6AjS+9j/yNFoNDy36ikyB3V3Tl4WItAkeVIxvV7P1KlT280KFrV6Y/s26qwKY9J7cFl6d7d11q0+AMDocX3pFCebVIaafDbUwWq1otFonPHYvmI3h+onT586dIYVi1YDkDmouzN5qjWa+P6T5pObY+LtE84dc27GTRvFCuu/W9WOfiN7t7kv7YV8NoJDvrsqV1NTQ1xc3MUrCp+U1hhZsse+ouWXHkadANauLgBg9LiewWiWaAX5bASXoiis+nC9c0n86n+vZ//Gg3xt+RcajYbSc6X88bbnqa1uvsR/0NiGR93xneO4f+FPna9j4qMZf0s2EZEyP8df5LMReJI8qZjFYmHlypUy8S+A3t25gxqLhSGpaVyZ6X71h6JYuWrCIAwGHTV1J7FYRko8Qkw+G8FlrKrhH/e9zreLv2/23tavdzLimiFs3LqR0deN4PuPNzL1pxMB0Gi1TLrzKgZnD3DWj42PYfovJwet7R2NfDaCQ5In0WFVmUy8u3MHAPeOHtts+a+DTqflJ3dfya23j2PZsmXBbKIQAWWz2TDX2c9BO3+6lH//5QtMtXXO9++cfytpmamUnilzSZwcE68NkQYGjO7jLH/w1Z/xu3897PGzJER7IcmT6LA+2LOTSpOJPknJTO7T7+IXCNHOmOssXB890+P7Nz9wPWmZqURER3DplYMwRBm4++kZzSZ0m81mwL7RoyROoiOQ5Enl1DDp72RlBbXt7JRuxWbjzR3bAPjFZWPQevgHv6z0Aju3H2dMdl8MEVpVxEPYSSxap662Dsd2ft+8t4b/vrqCXpdm8ut35rqtf9nkYQzPtR8TktzNvlFrWkYKz6/+Q4tfR+KhHhKLwJNNMv0gUJtkqsEb27fy7NrVoW5GwKTHxbHyjtkYdO7Pjvvik628+Fwelw7P5Pl/3hHk1gnRNq888q7bIzPSMlN4/9jL2Gw2jI3OZTNEGmTituhQZJPMdshqtVJSUkJKSgparfbiFwTAvuKGwxuToqJC0oZAidDpeeKKqz0mTgBr67coyL6inyriIewkFu7tXV/AwW1HuPH+6wD7Ya3uPPLmLwH7HkmOA1jbQuKhHhKL4JDkScUURWHDhg1MnTo1ZB8Ci81+ZtPvxudy9/CRIWlDqFRWGNm54zgAl48foIp4CDuJBdRcqOGt3y6hoqTSWbZyyTpGTRrG+FvGkdw1iSc+eJAbE+/kjb1/Iy2jMwAR0RHOQ2L9ReKhHhKL4JDkSbTIUn/gpa4DTgLduO4gVsVG775ppPdIdk6KFUINvvtgLZ+9+FWz8m3Ld3Kh3Ehy1yRi42P4f8t/R9agHiFooRDtlyRPokVK/chTS4+22qu1q+wbY15x1cCL1BQi+Lr0TCMtM4WeQzIYdc0wZ3n3fl3JHNiwU/7Ia4aGonlCtGuSPKmYRqMhLi4upEt/zdaOOfJUY6xj6+bDAFxxtT15UkM8hF17j8XG/25jyYJPsZgVl/L/bT3M0KsG8+S/H+GyScNI7prIXX+YEfLjSdp7PMKJxCI4ZLWdH7Tn1XZ3/Odj1p44znPXXsdNgwaHujlesdlsLD31FMeqm5+jdTFWqw2LRUGDBkNExxt1E8FVP8ALgEYL62dHUbze/d+2+k42pnxfjS7Kfp3mItNarun6KAMTJvqxtUK0H7Larh2yWq2cOHGCjIyMkE38czy202nD76+YGqWCwxfW+naxBnQGAJvzeyBEIOx+OprjS+wrWZNGWrh8cRWX/FahLF+PRguGRNefv/h+CkTaUGyABrjIn782Av/zq4Z/q4SdxCI4JHlSMUVRyM/PJz09PWQfAsdjO702/EZfyupOABCnT+OWrBfafD+zxcLK774jd8IEDLIJXUiFeyxKTpRxYOMRAFblfQgYAegS1Z97+jwCfYBr/PO1onSBPyBWDf9WCTuJRXCE3786IqiU+uTJEIYfwrK6QgCSI7OIM6S1+jqbzeZ2voAZM3pLHHH6VDlwM8TCPRbbtx3m+dvfcilbuGkB3ft1o5MhNkStEkK0liRPokXOCeNhmDyVmuwjT0kRGV5dt/D5ryk6Xc6P77ycIUNbvnbNxxtYNP8jj+/f+7e7GHWtfSXUpqXbeP03iz3Wnb3gJ2T/4DIA8lfuYeH9b3qsO+vJW7jq1hwA9m0o4Pk5r3ise+uvbmDSnVcDcCj/KH+6/R8e694w9zp+8ItJgH2Dxad++FePda+bPZEfPjQNgHOFxTwx9VmPdSfMvJKZT9wMQHlxBY/mzvdY94qbx3LXH2YA9r2M7h/3RLM6NpuNCxcucGZNGb947i4AFIvCz4c/6vG+Q6+6hHkv/dT5+ucjHkVpMiHbYdDYfs6NJAHm5TyBsbLGbd3ew7J44v0Hna9/NXE+ZWcr3Nbt0b8b8z/9NYmp8QzPvaThHkN70v+yPjLJV4gwIcmTimk0GlJTU0P6D6qlPYw8RWS2+hpFsbLm232Ulxv54Y/HurznLh4Xyo0c33fS4/2MVbXO/66urGmxbnWFsdF1LdetKqt2/ndttanFupXnq5z/XVdT13Ldkoa6ZpOlxbrl5xoSBHNdy3XLisqd/61YrC3WHTSuv/O/rVZbi3VLTpW6vG6pbrc+XVxen9h/CnOd+zMbHWe6OZwsOO3yPW8sJj7ate7/zjRrl4NWZ/8cDbv6EoZdfYnbOuFIDf9WCTuJRXDIajs/aM+r7SYvfoeDpedZfNMt5GS0PglRg3cOz6LCfIYfZj5Hj5hhF78A2LXjOI/c9x5xcVF8tPQh9HrXuV7mOjMPj3+SSy4fyJ1/uI3qCiMnC057vF/PIRkkpiYAUHa2vMVf7pmDupPc1f5Lu6KkkqO7Cz3W7dG/Gynd7TtGV5ZWcWTncY910/t2JS0jBYDqimoObj/qsW7XXml07Wl/xFlzoYaCLYc91k3LTCG9T1cAao0mDmw66LFuSvdkevRPB6DOZGbf+gKPdZO7JTn3KVIsCru/3++xbmJaAj0vsY8OWq1Wdq3e57FufOc4eg/Ncr7euWovnv7565QUS9/hvZyvd3+/H8XifpQqJj6a/qP6OF/vXV+A2eR+Q9Wo2EgGjunnsY1CiOCS1XbtkKIoHDx4kH79+vn9OIXWapgwHl4jTxZrHRXmIgCSvBh5Wrva/kt93JX9myVOiqKQ9+E3HNh8iKJjxfzsL7OIjo0iJT25VfdO6pJIUpfEVtVNSIlneO6QVtWNT45rdd3YhNhW143uFN3qulExka2uGxFpaHVdnV7ntq7js5HRL91ZptVqW33n2EGYAAAgAElEQVRfwKuRn0uvHNTqupfkDGh13fZCDf9WCTuJRXCE12/EDsZqtVJQUIDVGrql8uE6Yby87iRgI1LbiRhd6xIWm83GuvqDgN3tKm61Wln3pX3PqFGThspKlhBSw2dDNJB4qIfEIjhk5Em0yGK1P6oItwnjjm0KkiIyOLzzGH//xWu8uHGB8/2X5r3FthU7Xa6pq7Nw7kwFGq2GocMazgJ7/TeL2fDFFmw2G2cLiwGck8CFEEJ0PJI8iRZZrPY5IeE28lTqSJ4iM8l78zsObD7EofyjznksJadLOeFhrpINiIho+GiUFpW51I3uFMXoKcMD13ghhBCqJsmTimm1WjIzM0P6eChcR55K6+wTqJMjMji6xz7h+OiuQmfydNcfbuPmB653uWb7liNs+P5/5IwfgL5R8jTziZuZ+tNrsFoVDh8+zOXX5DgngYvQUMNnQzSQeKiHxCI4ZLWdH7Tn1XZDX3mRC3V1fDPrbnontW5iNMDZ48U8fetzVJ6/QHVtHRXVNdgAyxV9sPZNBUBbWIb+O8+rrizjemEdaF9erjldgWG551VXymWZKEPsk4c156pIWL0Vrc5KrTEKpcyCxmrD9ONR2NK83235xsuH8MiPrvL6OiGEEOomq+3aIUVR2LVrF0OHDg3ZqomGCePeff3PF37lssxdU/8/k9FEXW2d/Z5GE/EVte5vANQZTZjq6+qNJiJbqFtTbaLWUbfGRORpDaADzGgAa7SBC7EGqK/jjTqzfS8gNcRD2Eks1EXioR4Si+AIm+SprKyMefPm8cUXXwAwffp0XnzxRRITPa+keu211/jggw/Yvn07VVVVlJWVNavvy32DxWq1UlhYyJAhQ0L2IbBYvT8Y2Gq1supf6wHIfWgqnx09gU6r4em7ptCtT1fiU+yjP9Xl1RQdKvJ4n5TMVBLS7H8JGCuNnPnlGY91k3t0JqmrPWalFSf5z23r0aDjyvj5aNHRpU9XOiV5PvZi68bD9B+YTnxidLP3OkVFOPsV6ngIO4mFukg81ENiERxhkzzNnDmTkydPkpeXB8DPfvYzZs2axZdffunxGqPRyJQpU5gyZQqPP/643+7bkVh8GHk6ffgsxqoaYuKjOZgSicUYx91Tx3LNtNGuFVMTGdive+tumprIgD7pF68HWGMOkTRMISmiOxN6j7lo/XNFFfxzwVfo9Vr+vfRhOsVFta5NQgghOqSwSJ72799PXl4eGzduZOxY+5EZr7/+OtnZ2RQUFDBggPtN6R580H7e1KpVq/x6345CsVpxTIjzZuSpR79ufFT0Biu/2cnj//0evVbLj8YPDUwj3SirP9MuuZVn2jk2xhw0pIckTkIIIS4qLJKnDRs2kJCQ4ExwAMaNG0dCQgLr16/3Ocnx9b4mkwmTyeR8XVlZCYDZbMZsth/LoNVq0el0KIrislmZo9xisbgcDaHT6dBqtS7liqLQv39/tFqt874Oer09dBaLpVXlBoMBq9WKojQcMaHRaNDr9VitVlZ/vIHD+cfs5fXtNCsKydvtiUjtbTUoEZHodDrWfraJA5sPudxHq9FgrU+2bn5wKnFJnVhXXALAxJF9SYyJdPbB0ddA9am0/ky7BH0P59dw9NVdPNatsW+MmX1F32bx8xQPT/ELdJzclXv6GWvLz57a+6QoCn379nW2tz30qWl5OPWpaTzaQ5+atj2c+tS3b18URWlXfWpa7s8++SIskqeioiLS0tKalaelpVFU5HnOTKDuu2DBAp566qlm5cuXLycmJgaAzMxMRowYwa5duygsbDijbMCAAQwcOJDNmzdTXFzsLB8+fDhZWVmsWbOGqqqGw1mzs7PR6XTk5eW5BDk3N5fo6GiWLVvm0oapU6dSU1PDypUrnWV6vZ7rr7+ekpISNmzY4CyPi4tjwoQJnDhxgk9f/YIDK48065PjeNRvvlrOpcMGMmLECJZ/8B0bPt7u8fsT2Qt0CdHkbbUnXv3jrS7tzM7OJi0tjeXLlwekT2V97V+3cP95Sivt90pNTSUnJ4eDBw9SUNCwwi85qQt7dtYniJZTLFtmj0lLcdLpdHz33XfN4hTIPjnilJ+f7yz31Cd//uypvU86na7d9Smc41RXV9fu+hSucTp06FC76xP4P07r1q3DFyHdqmD+/Pluk5DGtmzZwvLly3n33XddvlEA/fr1Y/bs2Tz22GMt3mPVqlXk5uY2mzD+7LPP+nRfdyNPGRkZlJSUOJc6+iOzt1gsbN++nbFjxzY7wNQfmf2ZI+ewmCyk9Ehm43+3OlfHaTQaNBoNJouFJXt3A/DVx0+TlBSHTqdj5b/WsWdtw7YBjvo2mw2bzcbM397Mx1sP8OrSTQzp2ZW3HrklqH+tvHPsJ1RbzvPD7i/QJWqAs9zdXytfL93F3//8Ff0GdOXvr93pLHcXp8bxANrlX2Dh0ieLxcK2bdsYPXo0kZGR7aJPTcvDqU9N49Ee+tS07eHSJ5PJxLZt2xg1ahQGg6Fd9CmQcSotLaVz587htVXB3LlzmTFjRot1evbsya5duzh79myz94qLi+nSpYvPX79r164+3TcyMpLIyMhm5QaDAYPB4FKm0+ncrnhw/BBdrLykpASbzdbsvo2/ZmvLtVqty8ZpC+e+yY5vd/P44nlMuiOXSXfkutQ/bzSy8I2XAUjunIBWY5/3lHvb5eTedrnbrwtgVhQ+WbsHgBm5wz32NRB9MikXqLacByA1picGXcvxWP/9/wC44uqBbu8fyHh4Km/ap4uVe/oZa+vPXktt9LY8EH06f/6889r20qfGwq1PjePRXvrUmnK19Umv13P+/Hn0er2zDeHep1DE6WJCmjylpKSQkpJy0XrZ2dlUVFSwefNmxoyxr57atGkTFRUV5OTk+Pz1A3XfcKFY7Nm8Vud+J1rHSjtt/Zym1vpu+yFKKqpJiY/h2pH9295QL5TVnQQgVt+ZSF2nFuuaTGZ2bbfvRH65m4OAhRBCCHfCYs7ToEGDmDJlCnPmzOHVV18F7FsKTJs2zTmp+9SpU0ycOJFFixY5E6GioiKKioo4dMg+uXn37t3ExcWRmZlJcnJyq+7bnlmV+j2c9O63IXAkT3ovt/lfsmoHAD+8cigGD/duybEjxbz96kpqa8389JcT6DegGwCb1h3k0482e7zujp+OR5tpf9atr03hNw+877HujFk5jLisF4s/vZ8dW4+R1fPiSbwQQggBYZI8Abz//vvMmzePSZMmAfbNLBcuXOh832w2U1BQgNFodJa98sorLnOqxo8fD8Dbb7/NXXfd1ar7hpJOp3NOUA6E1o48eZM87T1WxK4jZ9DrtPzwSt+2J/js35udj9OqKmuc5SXFVWzfctTjdTf88DJsXe2Tv7XGlBbrTr5+GACJSbHkXntJq9oV6HiI1pNYqIvEQz0kFsERNslTcnIyixcv9vh+z549m02qnj9/PvPnz2/TfUNJq9WSlZUVsPtffOTJnlx5kzx9uMq+KmLSqP6kJHje0bslp06UAjB6XB+yeqU6y4eN6sljv7/B43V9B3Rlc/0eT1mp/Xns956Tt8GX9vC6XYGOh2g9iYW6SDzUQ2IRHGGTPHVEFouFNWvWMH78eJ8ntbVEsdTPafI08lSfjBpamTyVVFTz9Vb7ysUf547wuV2nTpUBcPvdV9I5peEg3x4ZyfTIaPlw4rIj9sd2PdMGkNXrUp/b4E6g4yFaT2KhLhIP9ZBYBId3k1lEUNlsNqqqqpqNqPlLw8iTh+SpfnmorpXJ06drd2NRrFzaqxuX9OzqU5tMJjPFZ+2bjqb3SLpIbVeKzUJ53WkAklq5u7g3Ah0P0XoSC3WReKiHxCI4JC3twKbcM4Hzp0vp2qv5RqHQMPLUmsd2ZovCx2t2AvDj3OE+t6nodDkAMbGRJCTGeHVtpbkIKxb0miji9KkXv0AIIYTwgSRPHdhN86a2+L5zzpPm4snTN9sPUlJpJCUhlokj+vncplMn7Y/suvdIQuPF9ggAZSb7I7ukiB5oWtFmIYQQwheSPKmYTqdzHs8SCpb6x3p6D3OiGnNMFL9lvG/bEziMHteHt5bcS21tndfXltbZJ4snRfr/kR2EPh6igcRCXSQe6iGxCA5JnlRMq9W6PXvPX84cPQs26Nw9mYjI5ju+WmyOrQpa/hDuOVbE7qNnMOh13HxF2yZpGww6MrI6+3RtWf2BwMkRmW1qgyeBjodoPYmFukg81ENiERzybEPFzGYzS5cubXZukL88cvXvuaPvXI7uLnT7vnPk6SKPzz5cad8Uc/Ko/nSO9217An8oc4w8BWCyOAQ+HqL1JBbqIvFQD4lFcMjIk8o1PUDRnxxbFXhcbecYeWph+Lekoprl2+wbWs5ow/YEDq+++A2d4qKYduNIryaM22w2Sk2O5CkwI08Q2HgI70gs1EXioR4Si8CT5KkDsNlsLPx8HXuPFbmUV1TZd2N/5oNviVyV2Oy6sxhBC8eLyvjFCx+7vXdplRGLYmVo724MzvL9kGYAs1nh039twmq1MWXaMK+urVHKMVmrAA1JEd3b1A4hhBCiJZI8dQAnSyp4++stzcqTzApaYE/hWZSqqmbv18XZoAdcMJrYXHiixa9x+8SRbW7n2aIKrFYbUVEGkju3fKhvU45HdvGGLui1kW1uixBCCOGJJE8qptfryc3NbfMuscXlF5z//czd1zn/e+HibdSZLDz0o6tIymx+MO6WklO8cyifvt06M2/iOI/3T4qLZuzAtj8qO33SfixLug/bFJTWObYpCNwjO3/FQ7SdxEJdJB7qIbEIDvnuqlx0dHSb73G+0v54bnifdK4bM9BZ/s/6/88d2Y9uvZs/cjPuV+AQdEmMc7kuUE7X7/GU3qPlI1jcKauf75QcoMniDv6Ih/APiYW6SDzUQ2IReLLaTsUsFgvLli1r8+S/85XVAM0O6r3o2XZWx1YFwfkxOeUYeeru3bEsEPg9nsB/8RBtJ7FQF4mHekgsgkNGnjqA4or65KnJNgLT752Euc5CTLz7v1KCnTw5R558SJ4CvceTEEII4SDJUwfgeGzXdA+mXzx/V4vXKUFOns6ccjy28y55MltrqTSfBQK3x5MQQgjhIMlTB+B4bNc5wbuDds31yZMuSMnTK+/O4czpMtK6Jnh1XXndKcBGpDaOaF3zLReEEEIIf5LkScX0ej1Tp05t86oJx8hT48d2NpuN86dL0eq0JKYloHWTIDkOBjYEKXmKiNST1SvV6+ucj+wiM7xepecNf8VDtJ3EQl0kHuohsQgOmTCucjU1NW2+R0mFfauCxhPG62rr+HHGL7gt/WfUVpvcXmex2oDgPbbzVWmAj2VpzB/xEP4hsVAXiYd6SCwCT92/FTs4i8XCypUr27RqQrFaKa2yf5A6xzc8trPWn1sHLRzPUj/yFIzHduu//x/PLfgv36/c7/W1DdsUBHayuD/iIfxDYqEuEg/1kFgEhyRP7Vz5hRqsNhsaDSR1akieHNsUgOetCpT6kadgPLbbteM4eV/ms3fXSa+vLQvCBplCCCGEgyRP7ZxjvlNSpxj0jZIk15En9wf/moM48tSwQaZ3K+1sNitldfaEK5B7PAkhhBAOkjypXJuPZqnwtEGm4vxvd5PFoWGfp2CMPJ0+1XA0izeqLMVYbCa06EkwdAtE01zIJEz1kFioi8RDPSQWgSffYRUzGAxcf/31bbqHc5uCeNdtCpT6kSdPo07QeJ8nz3X8wWq1ceZUOQDdvTyapcxkf2SXGNEdrSaw7fRHPIR/SCzUReKhHhKL4JCRJxWzWq2cO3cOq9V68coeOI9mabJBprV+5MnTfCdovM9T4Jb/A5wvqaKuzoJOpyWti3d7PAVzpZ0/4iH8Q2KhLhIP9ZBYBIckTyqmKAobNmxAUZSLV/agpMKxu7jryFNkTCTXzZ7IpDuv9vz1nY/tAjuic7r+TLuu6YkeV/550rDHU+Ani/sjHsI/JBbqIvFQD4lFcMhju3auxMOhwAkp8Tz8+i9avDZYI0/F56oA7+c7AZQFceRJCCGEAEme2rWX5r3FzkUrSbIofPTJbj43uIb79d3PEZ8c5/H6YI08XTPlUq64eiBGD5t1tqTUJMmTEEKI4JLkScU0Gg1xcXE+HzlyoaIapbIWLVBtNFPd5H1b/T5OnliCeDBwVJSBqCiDV9eYlAsYFfsjv2AkT22Nh/AfiYW6SDzUQ2IRHJI8qZher2fChAk+X3/XH2awTKnGaLLw3M9/QHpKvMv7nRJjPVxpF+yDgb1hs9lYdOQzAKy2GJ7duzQ4XzgFVu/7NCC3To9O4uf9rg3Ivdubtn42hH9JPNRDYhEckjypWGWFkb8++xm1NVY0Gg29+qTxi3kNv1z/8NtPqL5Q2+y68/uOY1As/PLZ26iKi4S4SFavOcyFiqbnHa0HIK1LPI888QNn6XML/su5ogp2dy2DWPjioy3seXMvAElJsTw2/0Zn3Refy+Nk4Xm37Y+NjeTJZ3/kfP3KP1Zw9PA5lzo2m40dW48xdfoI5tw3kU5xUa363hy6UMQ3RavJSYVztXq+PZvfquvUbGB8d0meWslqtXLixAkyMjI87lMmgkfioR4Si+CQ5EnFtmw8xIbvjzhfWyyuqyd27zhOebmx+YUHjkB5FXu3XgZAlEHPgT0nOVdU6fbrZPZMcXm9f89Jjh8toXx8BMTqKDxaQvlR+9fu0tV1K4GCfacp2H/a7X0TEl1X+B0sOMOuHYVu636/cj9zH5ni9j13yuuMJBjsiWPP2AE8MGBqq6/1ldWqsH//fgYNGoQ2APPAOkd28vs92ytFUcjPzyc9PV1+QaiAxEM9JBbBIcmTiilKw5ykx+ffSGKSazIy71fXYTY3X4764W8Xc3RbFUaTGbBvU3Dfg1dQW2t2+3ViYiNdXs/55USqq008X7iVfcbzTL9xFNkJ6QDN5iXdOecqqirdn+BtMLgmGD+56wquv8FNsgf0G9CtWf2WGC0m4uuTp7EpYxiefEWrr/WV2Wxm2Z5ypmbmYDB4Nz9LCCFE+yHJUxgYPiqLCZOGNCu/MneQ2/pfPxfLUeBCbR0AnRNiyRk/oNVfb+zl/QB445N9YDzP0GGZTOg/0G3d0eP6tPq+I0f3bnXdizEqJuIN9qQtKQh7PAkhhBAOMqanYo7FEt6umnAc+utInpruLt5ajoOBg7HazlvV5hriDPatDZKDtE2BRqMhNTVVVrGogMRCXSQe6iGxCA4ZeVIxg0FPVJSByEjvHhE5tiC4YKofefIxeVLq7xOMg4G9dcFShFZjw2bT00mfcvEL/ECv15OTkxOUryVaJrFQF4mHekgsgkOSJxW7MncgXXvo6Nevn1fXOc40co48JcS0VN0jS/3Ikxq3KjAqRQDYSEKjCU77FEXh4MGD9OvXD50usBuHipZJLNRF4qEeEovgUN9vReFktVopKCjw+oBHx8hTVY39sZavI08Wm/0+anxsZ1LsWx7oNMEZdQLf4yH8T2KhLhIP9ZBYBIf6fiuKNvvV2/fxav5fqe1mP3ql6bl2rWWpP1hSH6SRHW8othIADNrUELdECCFER6O+34rCaef243z572O8+8Yar65L79OV3kOzKDNbAEiJ9+2xnWOHcb1OfT8mNuwbc8bouoa4JUIIIToan38rfv/999x+++1kZ2dz6tQpAN577z3Wrl3rt8Z1dKXnqzl+pIqDB854fa3VaqO00r6nks8Txm3BORjYWzabDZ2mAoBO+vSgfV2tVktmZqZsPKcCEgt1kXioh8QiOHz67n7yySdMnjyZ6OhoduzYgclkn1tTVVXFs88+69cGdmQNP/zeLTld9vo3vPn7D7GW2Y8CTvZx5MmiqPNsO6NShk5Th80G8RHBS550Oh0jRoyQSZgqILFQF4mHekgsgsOn34p//OMfeeWVV3j99ddddlrOyclh+/btfmtcY2VlZcyaNYuEhAQSEhKYNWsW5eXlLV7z2muvcfXVVxMfH49Go3Fb/5lnniEnJ4eYmBgSExMD0nZfNUz4s7VYr6kvX1nOR898iq6ilsRO0Rh8/BBZ6kee1DZhvMxkP+LlgiWSTvrgHWmiKAo7duxAUZrv6i6CS2KhLhIP9ZBYBIdPvxULCgoYP358s/L4+PiLJjS+mjlzJvn5+eTl5ZGXl0d+fj6zZs1q8Rqj0ciUKVN44oknPNapq6vjlltu4d577/V3k9vMVr/azeZd7tSQdGk1Ps93ArA45jypbMJ4ad0JACrNUcTqIy9S23+sViuFhYWyikUFJBbqIvFQD4lFcPi0z1O3bt04dOgQPXv2dClfu3YtvXv77wgOh/3795OXl8fGjRsZO3YsAK+//jrZ2dkUFBQwYID7o0cefPBBAFatWuXx3k899RQA77zzjl/bHEqOrQps+D7fCRolTyqbMF7mTJ6iidZHhLg1QgghOhqfkqef//znPPDAA7z11ltoNBpOnz7Nhg0bePTRR3nyySf93UY2bNhAQkKCM3ECGDduHAkJCaxfv95j8hQoJpPJOc8LoLKyErAfHGs22w/f1Wq16HQ6FEVx+QvAUW6xWJwjS2B/Tq3Val3KFcXifN9xXwe93h46i8XSrNxanzyh0ZAcF43ZbMZgMGC1Wl2GcjUaTX199+WO5AnFitls9kufGpd706fG5edrjwNQYY4iWmNwuc/F+uSp7a3pU+Ov4+8+OfgSp7b0qXHbw6lPjjYpioLBYGgXfWpaHk59ahqP9tCnpm0Plz457m82m9tNnwIdJ1/4lDz9+te/pqKigtzcXGpraxk/fjyRkZE8+uijzJ0716eGtKSoqIi0tLRm5WlpaRQVFfn9613MggULnCNWjS1fvpyYGPtjsszMTEaMGMGuXbsoLCx01hkwYAADBw5k8+bNFBcXO8uHDx9OVlYWa9asoaqqCoAvNp6hok8U2yzl3Db/TZfAJyYlotNqOX++1KUNnTsnc6q4/tGpBipKili+fDnXX389JSUlbNiwwVk3Li6OCRMmcOLECfLz853lqamp5OTkYK7/gV2zahWJer1f+gSQnZ1NWloay5cvd/nBzc3NJTo6mmXLlrn0aerUqdTU1LBy5UoATvf+HxjsI081ldUs27Ku1X06ePAgBQUFznJv+9SlSxe0Wi2rV6/2a5/A/g+LL3Fqa58CFadg9OnIkSMMHjy4XfUpnOO0b98+Ro4c2a76FK5xWrFiRbvrE/g/TuvWNfz+8IbGZvN2Rk0Do9HIvn37sFqtDB48mE6dvJu8O3/+fLdJSGNbtmxh+fLlvPvuuy7fKIB+/foxe/ZsHnvssRbvsWrVKnJzcykrK/M4Kfydd97hwQcfbNWcLXcjTxkZGZSUlBAfHw/4J7P/y79X8/H3uy/anqYS/7UDXXkNFdOH8MSjN3H92EFeZ/Y6nY4+Lz4PwLo7f0pKTIwq/loxW2t5/ehNAHxcOIIVE55BZ2tYjdje/6qUPkmfpE/SJ+mT//pUWlpK586dqaiocP7+bg2vR54sFgtRUVHk5+czZMgQLrvsMm9v4TR37lxmzJjRYp2ePXuya9cuzp492+y94uJiunTp4vPX91VkZCSRkc0nKhsMBpfVh2APkLslo44fopbKp2cPJp4a+vTti86LeUdvfFVAWXkND//oKn6QM8S51YBWq3W794e7ckujH76oiAiXfrWlT401/V61pry8/pFdraJHsUUR5WHOk6e+emp7a/pksVjYuHEjY8aM8WufWtv2QPSpNeVq7JPFYmHz5s2MGTOmxbaHU5+aCqc+NY1He+hTa8vV1ier1eqMhaMN4d6nUMTpYry+Sq/Xk5WV5ZdlkCkpKaSkXPxssuzsbCoqKlw+nJs2baKioqJdnx49oEcqh2OtXDOir8cfQnd6f/Yb6mrqyLokw+c9mpRGyZOatipovNIuJsiTxW02G8XFxbRhsFb4icRCXSQe6iGxCA6ffiv+3//9H48//jilpaUXr+wHgwYNYsqUKcyZM4eNGzeyceNG5syZw7Rp05yTxU+dOuV8rulQVFREfn4+hw4dAmD37t3k5+e7tLuwsJD8/HwKCwtRFIX8/Hzy8/O5cOFCUPrWkt07T/DVZ8dZsmi9V9f1Hd6LwdkDiG3DNgVmlSZPjj2eKs3RxOqCt02BEEII4eDTeNU//vEPDh06RHp6OllZWcTGui6HD8RGme+//z7z5s1j0qRJAEyfPp2FCxc63zebzRQUFGA0Gp1lr7zyisucKsfeVG+//TZ33XUXAE8++STvvvuus86IESMAWLlyJVdffbXf++GNc0UVHC6oJDHhZNC/tsXaMLKoquTJZeRJkichhBDB51PydOONN/q7HReVnJzM4sWLPb7fs2fPZsOU8+fPZ/78+S3e95133lHtHk++Hs/y5SvLMRlNTPzJlSR18W3XdIu14XuppuSp8WO7ZDfzzgJJp9MxfPhwOfZABSQW6iLxUA+JRXD4lDz9/ve/93c7hBsajab+/727bvHTH1N6pozhuUPakDzZR550Go2zHaFms1mdI08V5mh6xAQ3edJqtWRlZQX1awr3JBbqIvFQD4lFcPg2zbzetm3b2L9/PxqNhsGDBzsfeQn/cCy39Hbin63+Om2TFXr7i8/xu5XfcKHJslJ3HHs86bXq+eulynwOxVYH6Ki2RAZ9wrjFYmHNmjWMHz/e5xUawj8kFuoi8VAPiUVw+PSdPXfuHDNmzGDVqlUkJiZis9mcm2Z++OGHpKam+rudHZKvqyUcx7NotK4jRh/s2cX2ojNe3SsrIcGnNgRCaZ19srhWk4wNDTFBnjBus9moqqqSVSwqILFQF4mHekgsgsOn5On++++nsrKSvXv3MmjQIMC+s+ydd97JvHnzWLJkiV8bKbzjOJ5F2yR52ld8DoB5Y7IZ071Hq+41xM3O7qHieGRnsyUBBH3kSQghhAAfk6e8vDy++eYbZ+IEMHjwYF566SXnajgROo7HdppGE70VqwjzFksAACAASURBVJUDJfat6X/QfwB9kjuHpG1t4Zgsrtjso2GxstpOCCFECPiUPFmtVrebNjq2WBf+MXHSpQwemkaal49B3Y08HS0vo8ZiIcZgoGdikl/bGSyOPZ5MSjxQFfTHdjqdjuzsbFnFogISC3WReKiHxCI4fFqDPmHCBB544AFOnz7tLDt16hQPPfQQEydO9FvjOjpDhJ7MzO5ERXv3eMox56nxhPE95+yP7AalpPq863ioOR7bVVvs+4oFe58nrVZLWlqa26MCRHBJLNRF4qEeEovg8Om7u3DhQqqqqujZsyd9+vShb9++9OrVi6qqKl588UV/t7HDMpvNLF26tNmhixfzhy9+w/9b8STJ3RpGmPYV288GHJyqnjlM3qhVqjAqZQBUWqKA4CdPvsZD+J/EQl0kHuohsQgOnx7bZWRksH37dlasWMGBAwew2WwMHjyYa665xt/t69D27j5J3udHuVC2mdtuv7zV1w276pLm9yq2z3cK1+TJMerUSZ9CtcX+aDhWF/wJ401PAxehI7FQF4mHekgsAq9Nm0Bce+21XHvttf5qi2ii6FQ5B/aUExtzzKvkqSmbzeZcaTckXJMnkz15SorIwGipA4I/8iSEEEKAj4/t5s2bxz/+8Y9m5QsXLuTBBx9sc6NE2/z31RX899UV1FTXAnCqqpIKUy0GrZZ+nVNC3DrfOPZ4SorIxKiYAIiWg4GFEEKEgE/J0yeffMLllzcfCcnJyeHjjz9uc6OEnWPCd9PNLi/mxfte5+/3voaxsgaAvfWjTv2SOxMRpiswHI/tkiMzMFrsyVNskPd50uv15Obmyq69KiCxUBeJh3pILILDp+Tp/PnzJLjZeTo+Pp6SkpI2N0q0TdOtChyP7AaraMNLbzmSp6SIDIxK6B7bRUdHB/1rCvckFuoi8VAPiUXg+ZQ89e3bl7y8vGblX331Fb17925zo4SdVak/287a+m32G2/J7xix2nvOMd+pix9bFzyKzUJFnX1bjHh9d+qs9smQwd7nyWKxsGzZMpmMqQISC3WReKiHxCI4fBrXe/jhh5k7dy7FxcVMmDABgG+//ZbnnnuOF154wa8NFN5pvEmp47Gf47FduK60q6g7jRUFgyYKiHOWyw7jQgghQsGn5Omee+7BZDLxzDPP8PTTTwPQs2dPXn75Ze644w6/NlB4p/EolVarpcRo5Gz1BTTYN8gMR87J4pGZ1Fjtj+witHr02vCcvyWEECK8+Tyj7N577+Xee++luLiY6OhoOnXq5M92CeDK3IGUXzjC5MmtPy/Q8agP7I/tHPOdeiUlERsRngfpljlX2jVMFo8JwR5PQgghBPg456mmpgaj0QhAamoq58+f54UXXmD58uV+bVxHFxMbxQ9/NJ3EpNYnplaXkScNe8N8Z3Fo2OMpOSKD6hDu8aTX65k6daqsYlEBiYW6SDzUQ2IRHD4lTzfccAOLFi0CoLy8nDFjxvDcc89xww038PLLL/u1gR1dTU2NV/UNEXqe/uIxnvrs10RERTgni18SxslTqXOlXcMeT6HaINPbeIjAkVioi8RDPSQWgedT8rR9+3auvPJKAD7++GO6du3K8ePHWbRokdvNM4Vv9uw8zvzHF/Off29u9TU6vY5x00aRM300Or0u7CeL22w252O75MjMhj2eQrBBpsViYeXKlbKKRQUkFuoi8VAPiUVw+DSuZzQaiYuzr3pavnw5N998M1qtlnHjxnH8+HG/NrAjO360hN3LD1G68yyG8+X84N7Jzvfy3vqOsrMVbq+Lio3kpnlTqTKZOF5RDoTvyJNRKaXOakSDlgRDOkZlDwDRQd4gUwghhHDwKXnq27cvn332GTfddBNff/01Dz30EADnzp0jPj7erw3syE7sPg6HT3DqMPy76LxL8vT5S3kc2nHU7XXJXRO5ad5U9pfYDwPu1imO5OiYoLTZ30rr5zvFG7qh10Y0mjAu2xQIIYQIDZ+SpyeffJKZM2fy0EMPMXHiRLKzswH7KNSIESP82sCOzFR/Np0+JpKrbsl2eS/nhtH0HdHL7XWxCfZEyfHILlxHnaBhpV1yRAZAo6NZQpM8ySRM9ZBYqIvEQz0kFoHn03f4Rz/6EVdccQVnzpxh2LBhzvKJEydy0003OV+fPHmS9PR0tFqfplZ1eI7vW3RyHLMX/MTlvVlP3nLR6/eF+XwnaHQsS2R98uScMB78x3YGg4Hrr78+6F9XNCexUBeJh3pILILD56yma9eujBgxwiUxGjNmDAMHDnS+Hjx4MMeOHWtTAzsy50krrT+dxYVj5GlIGJ9p59wgMyITgOoQPrazWq2cO3fOZRd3ERoSC3WReKiHxCI4Ajok1PicNeE9m63+bDsfrjVZLBwqPQ+E+chToz2eAIwh3OdJURQ2bNiAoihB/9rClcRCXSQe6iGxCA55ntZOFZwvwWK1khQVRbdOcRe/QIXM1hqqLPbRs6aP7eRcOyGEEKEis8pUbOYD1xGZZmbipFyvr20830mj0fi7aUFRVncSgGhdAtG6BABZbSeEECLkZORJxZLT4hkwIpPeg7p7fa1zpV1aF383K2ick8XrH9lBozlPIRh50mg0xMXFhW0y2p5ILNRF4qEeEovgCOjIkwSvbfR6PRMmTPDp2n3tYJsCxx5PjZOnUK62a0s8hH9JLNRF4qEeEovgkAnjKvbNp5uYM/53LPjlG15dp1itzg0ywzl5atjjKdNZVuOYMB6i1XbHjx+XVSwqILFQF4mHekgsgsOn5Omee+6hqqqqWXl1dTX33HOP8/W+ffvIysryvXUd3K4NBzm29gBblm736rojZWXUWizEGAz0TEwKUOsCr+keTwDVIRx5UhSF/Px8WcWiAhILdZF4qIfEIjh8Sp7effddt6c219TUsGjRIufrjIwMdDqd760TPnHMdxqUkoo2TB+dWm2Kc8J4UqORJ8dWBaE4GFgIIYQAL+c8VVZWYrPZsNlsVFVVERUV5XxPURSWLVtGWhhvyNhe7Cs+C4T3I7sq8zkUWx06jYF4g33Se53VgsVm/2sqFBPGhRBCCPAyeUpMTESj0aDRaOjfv3+z9zUaDU899ZTfGtfR+TpotLcdHcuSGNEdrcY+eunYpgAgWhf8x3YajYbU1FRZCKECEgt1kXioh8QiOLxKnlauXInNZmPChAl88sknJCcnO9+LiIggKyuL9PR0vzeyo9Jo7E9VvfkI2Gy2RseyhO82BU2PZYGGbQoitQb02uA/Dtbr9eTk5AT964rmJBbqIvFQD4lFcHiVPF111VUAHD16lMzMTMlsA61+taI3axZPVVVSaTJh0Grpm9w5MO0KAsfIU7KbbQpiQzBZHOyPpg8ePEi/fv1kLl+ISSzUReKhHhKL4PBpwvj+/ftZt26d8/VLL73E8OHDmTlzJmVlZX5rXEdn9WGrB8eoU//OKUSE8QenzLnHU/PJ4qGa72S1WikoKJAlwCogsVAXiYd6SCyCw6fk6Ve/+hWVlZUA7N69m4cffpipU6dy5MgRHn74Yb82sCO79b7/z97bx0VV5v//r3NmhmFAZrgTgeQuTUVFQV0VKRU1RDQ3y4+lLbv8vi7llrlmu63drOFWWq53m37KtNas3LQy+3RjhCmKKTdmEpTKijdpCopyKwPD3JzfHzgHRmZgGAbONeP7+XjwEK5znXPe7/Oacd5zXe/rfd2LlIx7sezDP9t9zs9XXT/fCWg1bdeqTAFtzUIQBEGwgEMVxs+dO4fBgwcDAHbt2oX77rsPK1aswA8//ICUlBSnGng70+cOf9w5vA+G/OZOu8/52Q0qizcaa9FgrAZgq7o4BU8EQRCEdDg08uTh4QGtVgsA+Pbbb5GUlAQA8Pf3F0ekiK7D8zzCw8PB8/bLJG7L4sIlI8z1nXrJe8ODV4nt4rSdBCvtAMf0ILoH0oItSA92IC16BodGnu6++24sWbIECQkJKCgowM6dOwEA//3vf9G3b1+nGng7892eH/H5lm8RNvBnLP7n7zrsX6Gtx5X6G+AADAro3f0GdhOVOvNKuzCLdq2EmwIDgEwmQ1xcnCT3JiwhLdiC9GAH0qJncCg03bhxI+RyOT755BO8+eabuOOOOwAAX3/9NZKTk51qoJmqqiqkpqZCo9FAo9EgNTUV1dXV7Z6zefNmTJw4EWq1GhzHtel//vx5zJ8/H1FRUVCpVOjXrx9efPFFNDU1dYsPneXYwZ9R9OVRHProiF39T9zMd4ry84O3hzSjM86gZU87y+CpXuJpO6PRiOPHj9O2BwxAWrAF6cEOpEXP4NDIU3h4OL788ss27evWreuyQbaYN28efv31V2RmZgIAHn30UaSmpuKLL76weY5Wq0VycjKSk5Px7LPPtjl+6tQpmEwmvPXWW+jfvz9++uknpKeno76+HqtXr+42X+yls4vtWvKdXLe+EwBUinvahVu0S50wbjKZcOHCBQwdOpSWAEsMacEWpAc7kBY9g0PBEwCcOXMGW7duxZkzZ/Cvf/0LQUFByMzMRFhYGIYMGeJMG3Hy5ElkZmYiLy8PY8aMAQBs2bIF8fHxKCkpwcCBA62et3jxYgDAgQMHrB43B1Zm7rzzTpSUlODNN99kInjqLCfcIFkcAKpsTdtJXOeJIAiCIAAHg6eDBw9i2rRpSEhIQE5ODl555RUEBQWhqKgIb7/9Nj755BOnGpmbmwuNRiMGTgAwduxYaDQaHDlyxGbw5Ag1NTUWldOtodPpoNO1bBViTpLX6/XQ6/UAmpP2ZDIZjEajRb0Nc7vBYIDQamhJJpOB53mLdqHVeebrmpHLm6UzGAxi288397QbHNi7TX+FQgGTyWQxlMtxHORyuc12W7Z3xafW7dZ8MpiaUKMvAwD48CHQ6/WirzeaGgEASk4OvV7f4z61trczPpn729PuKjpJ7ZPZJqPRCIVC4RY+3druSj7dqoc7+HSr7a7ik/n6er3ebXzqbp0cwaHgaenSpXj55ZexZMkS+Pj4iO2JiYn417/+5ZAh7VFeXm51w+GgoCCUl5c77T5nzpzBhg0bsGbNmnb7rVy50uoefllZWfDy8gLQPLUZFxeHoqIiXLhwQewzcOBADBo0CAUFBaioqBDbY2NjERERgZycHNTV1QEALl5qXnUGrvnarUVOTEyESqXCnj17AAANJiN+qalpvrfKS2wHml+w06dPx7Vr15Cbmyu2+/j4YNKkSbh48SIKCwvF9t69e2PcuHE4ffo0SkpKxHZn+AQA8fHxCAoKsupTg+wqBJjAGRU4mJUPDhxSUlLQ0NCAs5cuADLg7MnTyCrRSuJTnz59wPM8Dh48aLdPrXUyY/YpOzvbJXVixaezZ89i8ODBbuWTK+t04sQJjBgxwq18clWd9u7d63Y+Ac7XqXXB787ACULny1j36tULxcXFiIqKgo+PD3788UfceeedOH/+PAYNGoTGxka7rpORkdHhRsJHjx5FVlYWtm3bZvGgAOCuu+7C/PnzsXTp0navceDAASQmJqKqqgq+vr5W+1y+fBkTJkzAhAkT8Pbbb7d7PWsjT2FhYbh27RrUajUA50T2ry/dgcx1n0MdHoQdp9db2HBrBH/08iX87v92IdTHB4fS0pmL7O39tlJadwh7Lv8DQcoBmN33Xxa+PlHwDr6vOotlQx7E1OBhLuMT4DrfwMgn8ol8Ip9uJ58qKysREBCAmpoa8fPbHhwaefL19UVZWRmioqIs2o8fPy6uvLOHhQsX4uGHH263T2RkJIqKinDlypU2xyoqKtCnT9eToy9fvozExETEx8dj8+bNHfZXKpVQKtsmLSsUCigUCos2mUxmNWnP/CJqt13cOlBoc93W9wSAkqpKAM35ThzHWe3P87zV2h+22m3Z3iWfrNjeGnONJ39leJvjDabmVZBqpZd4rCd9MhgMyMvLw+jRozvlU2fbXUGnzrY72yeDwYCCggKMHj26XdtdyadbcSWfbtXDHXyyt501n0wmk6iF2QZX90kKnTrCobPmzZuHv/3tb/j444/BcRxMJhMOHz6Mv/zlL/j9739v93UCAwMRGBjYYb/4+HjU1NRYvDnz8/NRU1PT5d2jL126hMTERIwcORJbt25lqrDYA+mToJfXYvzkuzvsa04Wd/VtWcxlClrvaWdGa2wOnlQSFckUBAEVFRVwYLCWcDKkBVuQHuxAWvQMDkUKr7zyCsLDw3HHHXfgxo0bGDx4MO655x6MGzcOL7zwgrNtRHR0NJKTk5Geno68vDzk5eUhPT0dM2bMEJPFL126JM5rmikvL0dhYSFKS0sBNO/DV1hYiMrK5lGay5cvY+LEiQgLC8Pq1atRUVGB8vJyp+ZRdYW+dwYhOr4vfjNxcId9f7raPDI31MXLFFTdLFPg32pPOzPmUgXetD0LQRAEISEOjTwpFAps374dL730Er7//ntwHIe4uDj079/f2faJbN++HYsWLRK3gpk5cyY2btwoHtfr9SgpKRG3jQGATZs2WeRUjR8/HgCwdetWpKWlISsrC6WlpSgtLW1TGd2VonadwYDSyusAXHvkSRAEMXiyOvIk1nmiUgUEQRCEdDhc5+mdd97BunXrcPr0aQDNyduLFy/GH//4R6cZ1xp/f3988MEHNo9HRka2CXgyMjKQkZFh85y0tDSkpaU5yULnk/ftzzj47ilc/kGHPy2fY7NfyfVrMAoC/D1VCO7VqwctdC71hutoMmnBgYdGEWJxTBAEcdpOyu1ZYmNjqfAcA5AWbEF6sANp0TM4FDz9/e9/x7p16/Dkk08iPj4eQHMtpqeeegrnz5/Hyy+/7FQjb1eO7vsJxf/3PS4UXmg3eGqd78RxnM1+rGMeddIoQiDnLUeXdCYDjELzCgqpgiee5xERESHJvQlLSAu2ID3YgbToGRzKeXrzzTexZcsWrFy5EjNnzsTMmTOxcuVKbN68GZs2bXK2jbct5iKZHc0g/mSuLG6lFpYrUWlOFlfanrIDpJu2MxgM2L9/v8NF1QjnQVqwBenBDqRFz+BQ8GQ0GjFq1Kg27SNHjiTBnIi9WVfmDYFdflsWc7K4h5Vk8Ztbs6hkHuA5aVZECoKAuro6l8qHc1dIC7YgPdiBtOgZHPoU+t3vfoc333yzTfvmzZvxyCOPdNko4iZ2vPiNJhNOXW+umurKyeIAUKUzJ4vbXmkn1ZQdQRAEQZixO+dpyZIl4u8cx+Htt99GVlYWxo4dCwDIy8vDxYsXO1Xnieg6Z6uq0GgwwFuhQKSvn9TmdInKdmo81dNKO4IgCIIR7A6ejh8/bvH3yJEjATTvBwc07zfTu3dv/Pzzz0407/aG45uTv9tLATdvBhzdOwi8CyeLN5kacMPQPILmp+zb5rh5pZ2UNZ5kMhni4+NpFQsDkBZsQXqwA2nRM9gdPLXeyI/oGThz2NROTPRzhXvkO1Xf3JZFJfOFSqZpc9w8bSdVdXGgeRWLtQ2qiZ6HtGAL0oMdSIuegZ29SIg2zEgbj/GLxuPxdWk2+7jLtiyVOvOUXdt8JwCS13gCmguxfvXVV202wSR6HtKCLUgPdiAtegaHi2QS3U/UoFAMSwzHPSmxVo8LguA2I0/trbQDWm3NIpM2YZxWk7IDacEWpAc7kBbdD408uTC/1taiVqeDBy9Df/8Aqc3pElXt1HgCaLUdQRAEwQ4UPDFM/r6fkPnWD3h/zZdWj5tHnQYEBMDDxZMDK5tslykAgHojBU8EQRAEG1DwxDB53xSjNPME9mzeZ/W4u+Q7mQSjmDDe0bSdlKUK5HI5EhMTIZfTbLfUkBZsQXqwA2nRM1Dw5ML87CbBU53+CoyCHjJOAR9FH6t9tAbpSxUAgEqlkvT+RAukBVuQHuxAWnQ/FJoyzE/a5rpH1zx0mLnjg7bHrzbXeBoaZD3gcBXMU3a+Hn3Bc9anH7UMTNsZDAbs2bMHKSkpUCgUktlBkBasQXqwA2nRM1DwxDD1puYVE3pOEAOlW9EoPREd2LsnzXI6Ha20A1pXGKecJ4Ig2MVoNEpaJkCv10Mul6OxsRFGo1EyO1hBoVB0S8FQCp4YZoDSFwUA/PUKvDxzlvU+/oFQufi3iyqd7W1ZzDQwUOeJIAjCFoIgoLy8HNXV1ZLbERwcjIsXL4Jz4V0nnImvry+Cg4Od+jwoeGIY35uBgtLIIzHyTomt6T5a9rSzPfLEQsI4QRCELcyBU1BQELy8vCQLXEwmE27cuIFevXqB52/vtGZBEKDVanH1anN+cEhIiNOuTcETw3Acb/5FWkO6GXHazkaNJ4CNOk9yuRwpKSm0ioUBSAu2uN31MBqNYuAUECBtzT1BEODp6QmO42jkCS3J81evXkVQUJDTpvBuz1e6izA99R5oArwQ3t950TJrNBhq0GCsAdCcMG4LVuo8NTQ0wMfHR1IbiGZIC7a4nfUw5zh5eXlJbEkzgiBQ4NQKsy56vd5pwdPtPabHOP1j+iIoRoFJD46W2pRuwzzq1EveGx689eW1giCgwVyqQMJpO4PBgOzsbNr6gAFIC7YgPZphIWARBAF1dXUQBEFqU5ihO3Sh4ImQlEo7puwajXqY0PwfgdQjTwRBEARBwRPD/HDoFPa9V4SPN30rtSndhnlPO/92VtqZp+w4cFBRwjhBEES3k5aWhvvvv19qM5iFgieGOfDZ9zi5uwi7130ltSndRpWu/T3tAMuVdlIPi9+uCbEsQlqwBelBAIBOp8OTTz6JwMBAeHt7Y+bMmfj111+lNsvpUPDEMPxtsNquZUNgtlfaAc3F1qZPn05VexmAtGAL0oMdeJ6Hr6+vZGUKFi9ejN27d2PHjh347rvvcOPGDcyYMcPtCnZS8MQ07p3wZzA1oVZfBgDwV7Yz8iSutJN2ys5kMuHq1aswmUyS2kGQFqxBetimoaHJ5k+TzmB3X51O32FfoDlhXK/X250w/sknnyAmJgYqlQoBAQGYMmUK6uvrxeOrV69GSEgIAgIC8MQTT7RbPb2mpgbvvPMO1qxZgylTpiAuLg4ffPABiouL8e237pV+QuOsDGMyv/jddNVEjf4yBJjgwXvBS+Zvs595U2Cpt2YxGo3Izc1FSkrKbV98TmpIC7YgPWwzc/Iqm8dGx/fHK2seFv+eM30dGhutByfD4sKx5n9/L/6d+uBG1FRrLfrsPfICBEFAfX091Gp1h2kOZWVlmDt3LlatWoVZs2ahrq4Ohw4dEgOv7OxshISEIDs7G6WlpXjooYcQGxuL9PR0q9c7duwY9Ho9kpKSxLbQ0FAMHToUR44cwdSpU9u1x5Wg4ImQjMpW27K09yZnZeSJIAjCnSgrK4PBYMADDzyAiIgIAEBMTIx43M/PDxs3boRMJsOgQYMwffp07Nu3z2bwVF5eDg8PD/j5+Vm09+nTB+Xl5d3niARQ8ERIhj0bAgPsjDwRBEF0ls/3PWPzmOyWUbqPvnrKZl+et/yC+f6uhV0zDMDw4cMxefJkxMTEYOrUqUhKSsLs2bPF4GfIkCEWRSVDQkJQXFwMAFixYgVWrFghHjtx4oTN+7hj0U4KnpiGs/jH3TAHT37t5DsBQD0jCeMcx8HHx8ft/hNwRUgLtiA9bKNS2T9i7oy+HMeB53m7tJDJZNi7dy+OHDmCrKwsbNiwAc8//zzy8/MBoM0CAI7jxLy2BQsWYM6cOeKx0NBQBAcHo6mpCVVVVRajT1evXsW4cePs9s0VoOCJYabNS4Dazxuhkb2lNqVbaNkQ2PZKO6D1tJ20wZNcLsekSZMktYFohrRgC9KDHTiOg1qt7lT/hIQEJCQkYNmyZYiIiMDu3bs7PM/f3x/+/pa5qiNHjoRCocDevXvFwKqsrAw//fQTVq2ynfvlilDwxDBDR/eDJsQDYWHtj8y4IoIgdGLarjl48pZ42s5kMuHixYsICwujpFiJIS3YgvRgB0EQ0NTUBA+Pjuvi5efnY9++fUhKSkJQUBDy8/NRUVGB6OhoFBUVdfreGo0G8+fPx9NPP42AgAD4+/vjL3/5C2JiYjBlyhRHXWISepUzjNFoRGFhodvVxwCAesN16E0N4MBD4xHabt+WOk/SJoy7sx6uBmnBFqQHOwiCgIaGBrtKFajVauTk5CAlJQUDBgzACy+8gDVr1mDatGkO33/dunW4//77MWfOHCQkJMDLywtffPGF0zbkZQUaeWKY4vxSfPfxSZiu+2DW/3OvIXHzlJ3GIxQyrv3CevWMTNsRBEG4E9HR0cjMzLR67N13323Ttn79+g6v6enpiQ0bNmDDhg1dNY9paOSJYbJ25KLwP8fwn1c6nn92NeydsgNotR1BEATBFhQ8MYw7r1up0tmXLA60ynliYLVd7969aUURA5AWbEF6sAPHcZDL5aRFN0PTdgzD3azr4Y7vgZY97ewYeTI2jzypZNLmPMnlcrdbbuuqkBZsQXqwA8dx6NWrl9RmuD008sQw4u4s0prRLYjTdkr7R56kznkyGo04deoUJcUyAGnBFqQHO3QmYZxwHAqeGEZw0+ipyajFDUMFAMDPo2+H/c11nqSetjOZTCgpKaHNTxmAtGAL0oMdBEGATqej4KmboeCJ6HGq9L8CAFQyX3jKOi7mJo48UcI4QRAEwQAUPBE9TpXOvNKu4yk7k2ASc56krvNEEARBEAAljDPNpAdHo7a2BsN+Ey21KU6lyrwtSwd72gFAg1Ev/i71tB3P8wgPD6cKygxAWrAF6cEOHMfZVV2c6BoUPDHM6MShGJ04VGoznE6nVtrdnLLjwUHJt19Ms7uRyWSIi4uT1AaiGdKCLUgPduA4Dl5eXlKb4fa4zNeEqqoqpKamQqPRQKPRIDU1FdXV1e2es3nzZkycOBFqtRocx1ntP3PmTISHh8PT0xMhISFITU3F5cuXu8uNTmE0GnH8+HG3W8FiHnmyZ9qu9Uo7qb9JuaserghpwRakBzsIggCtVtvlhPG0tDTcf//9TrLK/XCZ4GnevHkoLCxEZmYmMjMzVOEfDgAAIABJREFUUVhYiNTU1HbP0Wq1SE5OxnPPPWezT2JiIj766COUlJRg165dOHPmDGbPnu1s8x3ip+/P4OP/3Yesj/OkNsVpmAQjqpsuAbBv5ImlrVlMJhMuXLhAK4oYgLRgC9KDHcwbA0u12s6eQQt3wCWm7U6ePInMzEzk5eVhzJgxAIAtW7YgPj4eJSUlGDhwoNXzFi9eDAA4cOCAzWs/9dRT4u8RERFYunQp7r//fuj1eigU0k4T7dl2CEf/nY/SAxcw7eEESW1xFrX6chgFPWScB3wUQR32F6uLS1wgkyAIgugY86BFcnIynn32WanN6TZcInjKzc2FRqMRAycAGDt2LDQaDY4cOWIzeOoslZWV2L59O8aNG9du4KTT6aDT6cS/a2trAQB6vR56fXOCM8/zkMlkMBqNFt/GzO0Gg8Him4FMJgPP8xbtrc8zX9eMXN4sncFgsKtdoVDAZDJZDKuby/jbardle1d8qtCeA9Bc38loMMGIlutYs71O1wCgubr4rc+gp31qfX9bvrqLTqz7ZLbJaDRCoVC4hU+3truST7fq4Q4+3Wp7ez7p9XoIggCTyQRBEMBxHEwmExp0eovrm9tbY05HuHWkyFY7z/MQBMGi3XxtQRDgIefF88y2tNf/448/xksvvYTS0lJ4eXkhLi4On332mXjOP//5T6xduxZNTU146KGHsG7dOvHz0ZpPixYtAsdx4qCFyWQSjzvqU1fbzffU6/WQyWQALF97juASwVN5eTmCgtqOUgQFBaG8vLzL1//b3/6GjRs3QqvVYuzYsfjyyy/b7b9y5UosX768TXtWVpaYqBceHo64uDgUFRXhwoULYp+BAwdi0KBBKCgoQEVFhdgeGxuLiIgI5OTkoK6uDgBQVl7WfJBrvnZrkRMTE6FSqbBnzx4LG1JSUtDQ0IDs7GyxTS6XY/r06bh27Rpyc3PFdh8fH0yaNAkXL15EYWGh2N67d2+MGzcOp0+fRklJidjuDJ9q/H4Agprznezx6Sf+GqAAlJzcol0qn/r06QOe53Hw4EHRJwCIj49HUFCQ2+jkKj6dPXsWgwcPdiufXFmnEydOYMSIEW7lkz06VVVVITg4GDdu3IBMJoNSqURFZRWm/f099DRfLf8dVB4K1NXVQS6Xo1evXmhsbLT4wu/h4QEvLy+cPXsWjzzyCJYvX44ZM2ZAp9Ph6NGjuHHjBvR6PbKzsxEQEIBvvvkGFy5cwEMPPYSBAwfiD3/4AwDA29sbCoVCHEAw4+PjIwYwdXV14ipMtVoNQRAstAYAX19fGAwG1NfXi208z0OtVqOpqQkNDQ1ie0c+NTQ0oKmpSWxXKpvzZZuampCTkyO+nsyvvcOHDzv0nDlBwjKkGRkZVoOQ1hw9ehRZWVnYtm2bxQsaAO666y7Mnz8fS5cubfcaBw4cQGJiIqqqquDr69vm+LVr11BZWYlffvkFy5cvh0ajwZdffmkzQdnayFNYWBiuXbsGtbq56KMzvlX+88/vIXtTJvz6hWD7iTUWNrD6Dawjn7KvrsfJum8wJiAVI33ndWj7Z78exT9LvsT4oGisjHmYSZ9at7vKN2XyiXwin5znU0NDAy5evIjIyEioVCpwHAdtYxPufup/0dMcWvs4VErLkSFbozTHjh3DqFGjcPbsWURERFj0T0tLw8GDB3H69Glxo+E5c+aA4zh8+OGHFtexNpp24MABTJo0CdevXxc/d6UaedLpdDh79izCwsLg6ekJoOU1VllZiYCAANTU1Iif3/Yg6cjTwoUL8fDDD7fbJzIyEkVFRbhy5UqbYxUVFejTp0+X7QgMDERgYCAGDBiA6OhohIWFIS8vD/Hx8Vb7K5VKKJVtE5gVCkWb6T6ZTCYOE7bG/Ga3q12AzWnEzrTzPG+1Doutdlu2d8WnakNzdXE/ZbhdtuvQ/J+bt1wpuU8GgwF5eXkYPXq0TV/dRaeObOxsu7N9MhgMKCgowOjRo9u13ZV8uhVX8ulWPdzBJ3vb5XI5FAoFOI4Dz/NigKBSKnB4/UKrdnYnSoUMWq0W3t7eoi3mwOJWYmNjMXnyZAwfPhxTp05FUlISZs+eDT8/P3AchyFDhlg805CQEBQXF4PneaxYsQIrVqwQj504cQLh4S0rqM33s/acrdliy0Zntlv7jLb12usISYMnc9DSEfHx8aipqbF4c+bn56OmpsbpO3mbI9bWI0uE8xA3BLZjpR3A1tYsgiCgoqKC9oxiANKCLUiPtnAcJ44A9SQmk0kcyeuovItMJsPevXtx5MgRZGVlYcOGDXj++eeRn58PoG0w2nqUacGCBZgzZ454LDQ01MmesI1L5DxFR0cjOTkZ6enpeOuttwAAjz76KGbMmCEmi1+6dAmTJ0/Ge++9JwZY5eXlKC8vR2lpKQCguLgYPj4+CA8Ph7+/PwoKClBQUIC7774bfn5+OHv2LJYtW4Z+/frZHHUiHKfBUINGY/PcuK8dGwIDQL2BnVIFBEEQ7gbHcUhISEBCQgKWLVuGiIgI7N69u8Pz/P394e/v3wMWsolLBE8AsH37dixatAhJSUkAmotbbty4UTyu1+tRUlICrVYrtm3atMkip2r8+PEAgK1btyItLQ0qlQqffvopXnzxRdTX1yMkJATJycnYsWOH1Wm5nubuGXG4eP4CRk8YIbUpTqHyZnFMH3kfKHhPu87RinWeqFQBQRCEM8nPz8e+ffuQlJSEoKAg5Ofno6KiAtHR0SgqKnLomh0NWrgLLhM8+fv744MPPrB5PDIyss2QcUZGBjIyMmyeExMTg/379zvLRKdzd3IsIoYEICzMviku1hGn7OzY085MS50n6YNZmUyG2NhYq3kPRM9CWrAF6cEOHMeJSesdoVarkZOTg/Xr16O2thYRERFYs2YNpk2bhp07dzp0/44GLdwFlwmebkd4nkdERITUZjiNKnFPu463ZTGjNTYvOWVh2s7d9HBlSAu2ID3YgeM4u2dOoqOjkZmZafXYu+++26Zt/fr1HV6zo0ELd8Fltme5HTlVeA6v/nUTcr48JrUpTqFS1zxtZ8+2LGa0DOU8GQwG7N+/3+GiaoTzIC3YgvRgB0EQUFtbS8n73QwFTwyza9M+7FuzD/+7pOcLrXUHnV1pB7QaeWJgexZzcTf6T0l6SAu2ID3YoXWlc6L7oOCJ6BEMpibU6purwfspOzFtx9DIE0EQBEEAFDwRPUS1/hIEmODBe8NL5mf3efUM1XkiCIIgCICCJ6bh+Y5XS7gKVbqWKTt7VoGYMZcq8GZg5EkmkyE+Pp5WFDEAacEWpAc7cBxnUV2c6B5otR3RI1Q1mZPF7Z+yMwomNBqb97Zioc4Tz/NWN6gmeh7Sgi1ID3Ywb0NCdC808sQwJpP7JPxVmssUdKLGU4OhZWdsFqbt9Ho9vvrqqzablRI9D2nBFqQHO5hMJlRXV7fZrJdwLhQ8ET1Cy0o7+0ee6m9O2ck4Hh48G4OktBSbHUgLtiA9iNsJNj6RCKuMTRqGc6fP4TcT4qQ2pUsIgoCqLtR48pYraf6eIAiCYAYKnhhm/H1xuCErQ0pKitSmdIkbhmvQC43gIYPGw/6dt7W00o4gCEIS0tLSUF1djc8++0xqU5iEpu0YRi6XIzExEXK5a8e45mRxjUcoZJz9vrRszSJ9sjjgPnq4A6QFW5Ae7MBxHHx8fCQZra+srMSTTz6JgQMHwsvLC+Hh4Vi0aBFqamp63Jbuhl7pDPPLfy/j5LFz6N3XH6PGD5baHIdp2dOucxscszjypFKppDaBuAlpwRakBztIleZw+fJlXL58GatXr8bgwYPxyy+/YMGCBbh8+TI++eQTSWzqLmjkiWE+WLsHa3/3L/zzj29JbUqXqNQ5GDwZ2aoubjAYsGfPHkqMZQDSgi1Ij7YIggCtXt/jPyaTqVN7233yySeIiYmBSqVCQEAApkyZgvr6evH46tWrERISgoCAADzxxBPtrqgcOnQodu3ahfvuuw/9+vXDpEmT8Morr+CLL75wu9cGjTwR3Y552q4zK+2AVtXFGZm2IwiCsJcGgwFD33y9x+9b9NhCu/uWlZVh7ty5WLVqFWbNmoW6ujocOnRIDLyys7MREhKC7OxslJaW4qGHHkJsbCzS09PtvkdNTQ3UarXbTem6lzcEk1Q5UOMJALQG86bAbIw8EQRBuBNlZWUwGAx44IEHEBERAQCIiYkRj/v5+WHjxo2QyWQYNGgQpk+fjn379tkdPF2/fh0vvfQSHnvssW6xX0ooeCK6lSajFjcM1wA4Pm3HwtYsBEEQnUEll+OnPy3q8fsqeR51dvYdPnw4Jk+ejJiYGEydOhVJSUmYPXs2/Pya9x8dMmSIxZY7ISEhKC4uBgCsWLECK1asEI+dOHEC4eEtswu1tbWYPn06Bg8ejBdffLHrjjEGBU8Mw99M+uNceI+7yptTdl4yP3jKfDp1LmsJ43K5HCkpKW43/OyKkBZsQXq0heM4eEmwTYogCFCr1XYljctkMuzduxdHjhxBVlYWNmzYgOeffx75+fkA0GabF47jxMrlCxYswJw5c8RjoaEtZWjq6uqQnJyMXr16Yffu3W65XQy90oluxdGVdkDrnCc2gicAaGhogI9P54JAonsgLdiC9GAHQRDsXnHHcRwSEhKQkJCAZcuWISIiArt37+7wPH9/f/j7+7dpr62txdSpU6FUKvH555/D09Oz0/a7ArTajmFMN5P27F01wSIt+U6dSxYHWq+2YyNh3GAwIDs72+1WjbgipAVbkB7sIAgC6urq7PrcyM/Px4oVK/D999/jwoUL+PTTT1FRUYHo6GiH7l1XV4ekpCTU19fjnXfeQW1tLcrLy1FeXg6j0ejQNVmFRp4YZmTiYJz+qRS/mRArtSkOU6lzbKUd0JIw7s3ItB1BEIQ7oVarkZOTg/Xr16O2thYRERFYs2YNpk2bhp07d3b6eseOHROn/Pr3729x7Ny5c4iMjHSG2UxAwRPDTHlwNJpU11x6e5aWDYE7P21nHnlSMTLyRBAE4U5ER0cjMzPT6rF33323Tdv69evbvd7EiRNdeqakM9C0HeO4cgKmSTCiuukSAAen7RgsVeDKergbpAVbkB7E7QQFTwxz9VIVfGV9cbrootSmOESNvgwmGCDnlPCR9+70+ebVdqyUKlAoFJg+fbpbrhxxNUgLtiA92IHnefj6+oLn6eO9O6GnyzBbV3yGjJSX8Y9Her5KrTOoErdl6QuO6/xLjbXtWUwmE65evSou1SWkg7RgC9KDHQRBgF6vv22mz6SCgieGcfXXvnlbFj8HksUB9uo8GY1G5Obmut2qEVeEtGAL0oMdBEFAfX09BU/dDAVPRLdR6eC2LABgMBmhMzUve2Zl5IkgCIIgAAqeGMe1vzl0baVdk/i7N622IwiCIBiCgieiWxAEQazx5Mi0nXnKTsHJoODZWMXDcRx8fHzsrtxLdB+kBVuQHuzAcRx4nictuhk2PpUIq7QkWbvem6DBWAOdqQ4AB1+POzp9Potbs8jlckyaNElqMwiQFqxBerADx3FQq9VSm+H20MgT0wi3/Os6mKfs1Io+UPCd39uo4ea0HUvBk8lkwi+//EIrihiAtGAL0oMdBEGATqejhPFuhoInhokZNwBBI8Lxm2lxUpvSaVpW2nU+3wlovdKOnXwno9GIwsJCWlHEAKQFW5Ae7CAIAhoaGrocPKWlpeH+++93klXuBwVPDJPySALm/H08Fq2aJ7UpnaZSrPHkWPBUz1iNJ4IgCKJjHnvsMfTr1w8qlQq9e/fGb3/7W5w6dUpqs5wOBU9Et2AeeXJkQ2CAvRpPBEEQRMeMHDkSW7duxcmTJ/HNN99AEAQkJSW53agkBU8Mc/1KDeqvmXCh9IrUpnSaqi7UeAJab83CzrQdx3Ho3bs3rWJhANKCLUiPtgiCgAZDU4//AM0J/PZq8cknnyAmJgYqlQoBAQGYMmUK6uvrxeOrV69GSEgIAgIC8MQTT0Cv17d7vUcffRTjx49HZGQkRowYgZdffhkXL17E+fPnHX6WLEKr7Rhmy/JPkfPOXnwxKAcfnlgntTl2YzA1oUZfDqAL1cUZTBiXy+UYN26c1GYQIC1Yg/RoS6NRjwnfZvT4fQ9OyUCvXr3s6ltWVoa5c+di1apVmDVrFurq6nDo0CExXyo7OxshISHIzs5GaWkpHnroIcTGxiI9Pd2u69fX12Pr1q2IiopCWJhjX6RZhUaeGEZM+HOxRRPVTb8CEKDke8FL5uvQNeoZnLYzGo04deqU2w0/uyKkBVuQHuzQmYTxsrIyGAwGPPDAA4iMjERMTAwef/xxMfjy8/PDxo0bMWjQIMyYMQPTp0/Hvn37OrzuG2+8gV69eqFXr17IzMzE3r174eHBziyCM6CRJ5Zx0aWm4pSdR5jDw/haBus8mUwmlJSUoF+/fpDJZFKbc1tDWrAF6dEWT5kCB6dk9Ph9PTgZ6nR1UCqVHf7/O3z4cEyePBkxMTGYOnUqkpKSMHv2bPj5+QEAhgwZYqFnSEgIiouLAQArVqzAihUrxGMnTpxAeHjzTMMjjzyCe++9F2VlZVi9ejXmzJmDw4cPw9Oz82VrWIWCJ8LptOxp59iUHQBojeacJ3aCJ4IgCHvhOA4qCXI2O1NrSyaTYe/evThy5AiysrKwYcMGPP/888jPzwcAKBQKi/4cx4nXX7BgAebMmSMeCw0NFX/XaDTQaDS46667MHbsWPj5+WH37t2YO3duV1xjCgqeCKfTstLO8TluFus8EQRBuBscxyEhIQEJCQlYtmwZIiIisHv37g7P8/f3h7+/v133MBfudCcoeGIZF125UtXFGk8AmwnjPM8jPDwcPE+pglJDWrAF6cEOHMfBw8PDrpSJ/Px87Nu3D0lJSQgKCkJ+fj4qKioQHR2NoqKiTt/77Nmz2LlzJ5KSktC7d29cunQJr732GlQqFVJSUhxxh1nolc4w4ovfhWIoQTCJ03aO1ngC2Bx5kslkiIuLo5wOBiAt2IL0YAeO4+Dl5WVX8KRWq5GTk4OUlBQMGDAAL7zwAtasWYNp06Y5dG9PT08cOnQIKSkp6N+/P+bMmQNvb28cOXIEQUFBDl2TVVxm5KmqqgqLFi3C559/DgCYOXMmNmzYAF9f26u5Nm/ejP/85z/44YcfUFdXh6qqKpv9dTodxowZgx9//BHHjx9HbGxst/jRGQb/ph9KfjiNoWMHSW2K3dwwXINBaAQPGdQeIQ5fR2tgb+TJaDSiqKgIw4YNow8JiSEt2IL0YAfzajuVStVhABUdHY3MzEyrx9599902bevXr2/3eqGhodizZ4/dtroyLjPyNG/ePBQWFiIzMxOZmZkoLCxEampqu+dotVokJyfjueee6/D6zzzzjEXCGwvM/P/G43+W3Y2n1/9ealPsxjzqpPG4AzLO8dicxe1ZTCYTLly4QJufMgBpwRakBzsIgoCmpibaGLibcYmRp5MnTyIzMxN5eXkYM2YMAGDLli2Ij49HSUkJBg4caPW8xYsXAwAOHDjQ7vW//vprZGVlYdeuXfj666+davvtRpWu68niAG3PQhAEQbCLSwRPubm50Gg0YuAEAGPHjoVGo8GRI0dsBk/2cOXKFaSnp+Ozzz6Dl5eXXefodDqLlQO1tbUAAL1eL5au53keMpkMRqPR4tuYud1gMFh8M5DJZOB53qK98mo1KstvoKKsCr1D/CxskMubpTMYDHa1KxQKmEwmiyJ2HMdBLpfbbLdle3s+XWs8DwBQy++AyWRq41NrX28t89/adnPw5AFePFcqn8z3b22vIz7Z097TPrW23ZV8MttkNBqhUCjcwqdb213Jp1v1cAefbrW9PZ/0ej0EQYDJZIIgCOA4DoIgWPjEcZzFUv/W7QDajBTZaud53ua1W7ebf2/PFme195RPjrab76nX68Vp5davPUdwieCpvLzcarJZUFAQysvLHb6uIAhIS0vDggULMGrUKLv33lm5ciWWL1/epj0rK0sMwMLDwxEXF4eioiJcuHBB7DNw4EAMGjQIBQUFqKioENtjY2MRERGBnJwc1NXVAQC+3vQDznxzAnujj+GRVZMtRE5MTIRKpWozv5ySkoKGhgZkZ2eLbXK5HNOnT8e1a9eQm5srtvv4+GDSpEm4ePEiCgsLxfbevXtj3LhxOH36NEpKSsR2e3w6o/wR8AYunqzERd3FNj4BQHx8PIKCgpCVlWXVpy/2fAm9svk/tcP7c/BgykxJfWqtU58+fcDzPA4ePNgpn1jTqaPXnqv4dPbsWQwePNitfHJlnU6cOIERI0a4lU/26FRVVYXg4GDcuHEDMpkMSqUSdXV1FkGFt7c3FAqF+GW7tT0cx7VpV6vVEATB4rkAgK+vLwwGg8X+czzPQ61Wo6mpCQ0NDQCAuro6yOVy9OrVC42NjRZf+D08PODl5YWGhgY0NTWJ7UqlEiqVCvX19RbPXaVSMeETAId84jgOTU1NyMnJEf0yv/YOHz4MR+AECSdGMzIyrAYhrTl69CiysrKwbds2ixc0ANx1112YP38+li5d2u41Dhw4gMTExDYJ46+//jp27tyJnJwcyGQynD9/HlFRUR0mjFsbeQoLC8O1a9egVqsBOOdb5SuPvY3D7+5H7+gwbCt8zcIGVr+BbT03D1pjJR68Yz1CvQc79K3yekMtUnKa/T2YuAyeHkpJfXLHb8rkE/lEPjnPp4aGBly8eBGRkZFiorarjtK448iTTqfD2bNnERYWJlY5N7/GKisrERAQgJqaGvHz2x4kHXlauHAhHn744Xb7REZGoqioCFeuXGlzrKKiAn369HH4/vv370deXh6USsu8mlGjRuGRRx7Btm3brJ6nVCrbnAM0v1Fvrcgqk8msrj4xv9nba2+9TuLW6zrSzvO81Tosttpt2W6r3cg1QmusBAD09ooUr2nLV1u2N6H5jajk5VApPdvt390+tbbdYDAgLy8Po0eP7rRPLOlkz2uvIxs72+5snwwGAwoKCjB69Oh2bXcln27FlXy6VQ938MnedrlcDoVCAY7jwPO8GCCYP8yt3dcatlbGWWu3dW1zW319Pby9vTu0xVnt3e2Ts9qtfUbbeu11hKTBU2BgIAIDAzvsFx8fj5qaGos3Z35+Pmpqarq0k/frr7+Ol19+Wfz78uXLmDp1Knbu3GmRXyUV5tjZVRZNVDX9CgDwlgdAKbNvV29raBlcaQc0f1uqqKigVSwMQFqwBenBDoIgiCN5ju4tSnSMS+Q8RUdHIzk5Genp6XjrrbcAAI8++ihmzJghJotfunQJkydPxnvvvScGWOXl5SgvL0dpaSkAoLi4GD4+PggPD4e/v7+4iaEZ807S/fr1Q9++fXvKPbfBvC1LVyqLA61qPNFKO4IgCIJBXKbO0/bt2xETE4OkpCQkJSVh2LBheP/998Xjer0eJSUl0Gq1YtumTZsQFxeH9PR0AMD48eMRFxcnFtoknIu4IXBXg6ebI09SbKpJEARBEB3hEiNPQPMmhB988IHN45GRkW2GjDMyMpCRkWH3PaxdQ0pa5qslNsROzHvadWVbFqClxpM3YyNPMpkMsbGxVEGZAUgLtiA92IHjOLuqi3dEWloaqqur8dlnnznJMvfCZUaebkcGxkUhJPZODE5wje1ZxGk7pZMKZDKW88TzPCIiImjzUwYgLdiC9GAHjuPE5flSIggCpk2bBo7j3DIAo1c6w8xeMBlpq6di6Rv/T2pTOsQoGFDddBlA16ftWrZmYWvazmAwYP/+/Q4XVSOcB2nBFqQHOwiCgNraWslnUdavXy95ANedUPDEMOZiYlK/CeyhVl8OEwyQc57wkffu0rVYTRh3JT3cHdKCLUiPtgiCAL2pocd/TCaTWOncHj755BPExMRApVIhICAAU6ZMsShWuXr1aoSEhCAgIABPPPFEm9pb1vjxxx+xdu1a/Pvf/3b4+bGOy+Q83Y40NuigrW1CfW0DfAOs1zVhBfOedn4efcFxXYvJxZwnxqbtCIIg7MUgNOKN/97X4/dd0P//7O5bVlaGuXPnYtWqVZg1axbq6upw6NAhMfDKzs5GSEgIsrOzUVpaioceegixsbHiIixraLVazJ07Fxs3bkRwcHCX/WEVCp4YZvWi93DkvWx8EZ2L//y8Vmpz2kVcadfFfCeA3TpPBEEQ7kRZWRkMBgMeeOABREREAABiYmLE435+fti4cSNkMhkGDRqE6dOnY9++fe0GT0899RTGjRuH3/72t91uv5RQ8MQwLfPF7M8bm5PFu7rSDgDqzQnjjE3byWQyxMfH04oiBiAt2IL0aIuc88TjA77o8fvKoATv7WFXvtHw4cMxefJkxMTEYOrUqUhKSsLs2bPh59e8Ef2QIUMsNA0JCUFxcTEAYMWKFVixYoV47MSJEygsLMT+/ftx/PhxJ3vFHpTzxDBi6MR+7IQqJ9V4AlqPPLGVMM7zPIKCgmhFEQOQFmxBerSF4zgoeFWP//A8L24X0xEymQx79+7F119/jcGDB2PDhg0YOHAgzp07B6Dtdjet97FbsGABCgsLxZ/Q0FDs378fZ86cga+vL+Ryubj1yYMPPoiJEyc69wFLDL3SGcZ0c96Z9SRMQRBQqTMHT10feWpgNGFcr9fjq6++sithkuheSAu2ID3YwWQyobq6us1mvbbgOA4JCQlYvnw5jh8/Dg8PD+zevbvD8/z9/dG/f3/xRy6XY+nSpSgqKrIIqgBg3bp12Lp1a5f8Yg2atnMFGB96ajBWQ2eqA8DBz+OOLl+vntE6T0DbXdsJ6SAt2IL0cD3y8/Oxb98+JCUlISgoCPn5+aioqEB0dDSKioo6fb3g4GCrSeLh4eGIiopyhsnMQCNPRJcxT9mpFX0g57se8LA6bUcQBOFOqNVq5OTkICUlBQMGDMALL7yANWvWYNq0aVKbxjw08kR0mUpxQ+CuT9kBLXWeWNuehSAIwp2Ijo6OQ5/iAAAeAElEQVRGZmam1WPvvvtum7b169d3+h6sp504Co08McydQ8PgP+AO3DXyTqlNaZeWPe26niwOsFuqQC6XIzExUUyCJKSDtGAL0oMdOI6Dj4+PW1f3ZgF6pTPM75+egXl/Tmb+PyRn1ngSBIHpnCeVSiW1CcRNSAu2ID3YgQKn7odGnhjGYDBgz549zCdiOrPGU5PJAKPQvEqEtWk7V9HjdoC0YAvSgx1Y2dvO3aHgiegSBpMOtforAJxb4wkAVJQwThAEQTAIBU8M89L8zdg46wOkxS2V2hSbVDVdAiBAyftAJfPt8vXMyeKeMgVkXdwjjyAIgiC6A/p0IrqEOGWnDHPKPLuYLM7YlB1BEARBmKHgiWE4/mYwwnDynzO3ZQFaRp5YrPEkl8uRkpLCfAL/7QBpwRakBztwHAe1Wk1J490MBU8s4wL5fpU6Z9d4YnvkqaGhQWoTiJuQFmxBerADJYt3PxQ8MYzp5qozMPxGaFlp55yRp3pGazwBzSuKsrOzaUURA5AWbEF6sIMgCKirq6MAqpuh4IlwGEEwoarpVwCAn9K5I0/eDAZPBEEQhOOkpaXh/vvvl9oMp0DBE+EwdYYKGAQdeMihUYQ45Zot03bs5TwRBEEQBEDBE9OE3xUCVagfwoc6Z0rM2VTdzHfy9bgDPCdzyjVZnrYDQAmxDEFasAXp4R40NTVJbYJLQMETw6T//UF8/utmrNyxWGpTrFLp5JV2QKvVdgwmjCsUCkyfPh0KhUJqU257SAu2ID1s01DfaPOnqbHJ7r66Bl2HfQGA53n4+vqC5+37eJ84cSIWLlyIJUuWIDAwEPfeey/Wrl2LmJgYeHt7IywsDI8//jhu3LghnvPuu+/C19cX33zzDaKjo9GrVy8kJyejrKxM7GM0GrFkyRL4+voiICAAzzzzTJs8LJ1Oh0WLFiEoKAienp64++67cfToUfH4gQMHwHEcvvnmG8TFxUGlUmHSpEm4evUqvv76a0RHR0OtVmPu3LnQarX2CeIk6KsCw5hMJly7dg2BgYF2vxF6kpYaT87JdwLYznliXY/bCdKCLUgP28z0SbV5bHRKHF758jnx7zl9/ohGrc5q32ETBmNN9nLx79Sox1Fzrc6iz17TxxAEAQaDAXK53O5yBdu2bcOf/vQnHD58GIIgIDMzE6+//joiIyNx7tw5PP7443jmmWfwxhtviOdotVqsXr0a77//Pniex+9+9zv85S9/wfbt2wEAa9aswb///W+88847GDx4MNasWYPdu3dj0qRJ4jWeeeYZ7Nq1C9u2bUNERARWrVqFqVOnorS0FP7+/mK/jIwMbNy4EV5eXpgzZw7mzJkDpVKJ//znP7hx4wZmzZqFDRs24G9/+5td/joDepUzjNFoRG5uLoxGo9SmWMXZNZ6AVkUyGQyeWNfjdoK0YAvSgx0EQUB9fX2nVtv1798fq1atwsCBAzFo0CAsXrwYiYmJiIqKwqRJk/DSSy/ho48+sjhHr9dj06ZNGDVqFEaMGIGFCxdi37594vH169fj2WefxYMPPojo6Ghs2rQJGo1GPF5fX48333wT//znPzFt2jQMHjwYW7ZsgUqlwjvvvGNxr5dffhkJCQmIi4vD/PnzcfDgQbz55puIi4vDPffcg9mzZyM7O9vBJ+YYNPLEMCseexvfvXcAnw4/gm3HXpPanDZU6rpz2o4SxgmCcH0+r3vf5jGZzHL84qMrb9vsy/OWo0jvn3vDRs/OM2rUKIu/s7OzsWLFCpw4cQK1tbUwGAxobGxEfX09vL29AQBeXl7o16+feE5ISAiuXr0KAKipqUFZWRni4+PF43K5HKNGjRKDujNnzkCv1yMhIUHso1AoMHr0aJw8edLCnmHDhom/9+nTB15eXrjzzjst2goKCrr6GDoFBU8MY9AbAZMJRj173+Z0xhvQGisB3D4jTwRBEJ1F5e0ped+OMAdEAPDLL78gJSUFCxYswEsvvQR/f3989913mD9/PvR6vdjv1vw2juM6Ndpl7nvr1KIgCG3aWt+L4zir9zaZTHbf2xnQtB3hEOYpO295AJQy7w56249YqoDB4InjOPj4+NC2BwxAWrAF6cEOHMeB53mHtfj+++9hMBiwZs0ajB07FgMGDMDly5c7dQ2NRoOQkBDk5eWJbQaDAceOHRP/7t+/Pzw8PPDdd9+JbXq9Ht9//z2io6Mdsr0noZEnhuE4c2zL3n9ILSvtnJcsDrBd50kul1skOxLSQVqwBenBDua97RylX79+MBgM2LBhA+677z4cPnwYmzZt6vR1/vznP+PVV1/FXXfdhejoaKxduxbV1dXicW9vb/zpT3/CX//6V/j7+yM8PByrVq2CVqvF/PnzHba/p6CRJ6YRbvmXHcw1npy1LYuZeqN5Y2D2Rp5MJhN++eWXHh8eJtpCWrAF6cEOgiBAp9M5vD1LbGws1q5di9deew1Dhw7F9u3bsXLlyk5f5+mnn8bvf/97pKWlIT4+Hj4+Ppg1a5ZFn1dffRUPPvggUlNTMWLECJSWluKbb76Bn5+fQ7b3JJxAG+B0mdraWmg0GtTU1HQp4r+Vv6duRN72g+gzNAIfFK122nWdwZe/vogzNw5jQtATiPWf1fEJdjLp23/ghqERH9/9FCJ69XbadZ2BXq/Hnj17kJKSQvVsJIa0YIvbXY/GxkacO3cOUVFR8PR0Xi6SI5hMJtTW1kKtVlPZiJu0p4+jn9/0ZAmHEKftnFjjSRAEpnOeCIIgCAKg4Ilp+kQEwqO3GiH9gqU2xQKjYEBN0yUAzp2205n0MN2coqTgiSAIgmAVShhnmAUZszEiJRyjR4+W2hQLapvKYIIRCs4TveSBTrtuvaGlsq5Kxt7QP8dx6N27N60oYgDSgi1ID3bgOK5T1cUJx6DgiWHkcjnGjRsntRltqLy5LYufR1irFYFdp3WBTN6J13UWrOpxO0JasAXpwQ4cx6FXr15Sm+H2sPcJRYgYjUacOnWKuS0PxG1ZlM5dadfAeIFMVvW4HSEt2IL0aIaF9VeCIKChoYEJW1ihO54FBU8Ms2LB23hy+HI8evcyqU2xoLtqPJmn7VQM1ngCmlexlJSU0HJsBiAt2OJ218O8wlCr1UpsSddLFbgjZl2cuRKUpu0YplGrA/QG6LRNUptiQXfVeNLerPHkzejIE0EQhDVkMhl8fX3Fvd28vLwkyzkymUxoampCY2PjbV+qQBAEaLVaXL16Fb6+vpDJZE67NgVPRKcQBKFl2s7ZwZNYXZyCJ4IgXIvg4OZV0eYASirM03YqlYqSxm/i6+sr6uMsKHhiGO7mtiwsvfy1xmroTDcAcPD16OvUa9czXuOJ53mEh4ff9t/mWIC0YAvSozlROyQkBEFBQRYb6PY0RqMRJSUliIiIcOpIi6uiUCi65TlQ8MQw4rcGhqKnqpsr7TSKYMh55+YmacWEcTZznmQyGeLi4qQ2gwBpwRqkRwsymUzyoGXEiBGS3v92wGW+JlRVVSE1NRUajQYajQapqakWmwxaY/PmzZg4cSLUajU4jrPaPzIyEhzHWfwsXbq0u9zoFGLCH0N5f1W67pmyA1qm7bwZnbYzGo04fvz4bb+iiAVIC7YgPdiBtOgZXCZ4mjdvHgoLC5GZmYnMzEwUFhYiNTW13XO0Wi2Sk5Px3HPPtdvvH//4B8rKysSfF154wZmmO4zAUtR0k5YaT85daQe0qvPE6LSdyWTChQsXbtsVRSxBWrAF6cEOpEXP4BLTdidPnkRmZiby8vIwZswYAMCWLVsQHx+PkpISDBw40Op5ixcvBgAcOHCg3ev7+Pg4PZnMGQSG+kPu642AsACpTRExJ4v7O7nGEwDUM17niSAIgiAAFwmecnNzodFoxMAJAMaOHQuNRoMjR47YDJ7s5bXXXsNLL72EsLAw/M///A/++te/wsPDdt6NTqeDTteylUhNTQ0AoLKyUkwU5HkeMpkMRqPR4huAud1gMFjU4ZDJZOB53qJ97tNTEDxChqSkJFy/ft3CBrm8WTqDwWBXu0KhgMlkshjKNZfxt9VuzfbL18+i0WAAbqhx3Xi90z61br81qbK6ugrGeh2EGzoLf7vbJ3t10uv10Gq1qK2tBcdxdvkklU5dfe2x7pNZi6qqKnh6erqFT7e2u5JPt+rhDj7darur+NTY2AitVovr16/Dw8PDLXzqTp0qKysBdL6QpksET+Xl5QgKCmrTHhQUhPLy8i5d+89//jNGjBgBPz8/FBQU4Nlnn8W5c+fw9ttv2zxn5cqVWL58eZv2qKioLtniSizF+G679mNYi8e67eoEQRAEYUldXR00Go3d/SUNnjIyMqwGIa05evQoAFitVyEIQpfrWDz11FPi78OGDYOfnx9mz56N1157DQEB1qfLnn32WSxZskT822QyobKyEgEBAU6tq1FbW4uwsDBcvHgRarXaadclHIP0YAfSgi1ID3YgLTqHIAioq6tDaGhop86TNHhauHAhHn744Xb7REZGoqioCFeuXGlzrKKiAn369HGqTWPHjgUAlJaW2gyelEollErLvBxfX1+n2tEatVpNbwKGID3YgbRgC9KDHUgL++nMiJMZSYOnwMBABAYGdtgvPj4eNTU1KCgowOjRowEA+fn5qKmpcfpO3sePHwcAhISEOPW6BEEQBEG4By5RqiA6OhrJyclIT09HXl4e8vLykJ6ejhkzZojJ4pcuXcKgQYNQUFAgnldeXo7CwkKUlpYCAIqLi1FYWCgmiOXm5mLdunUoLCzEuXPn8NFHH+Gxxx7DzJkzER7u/KX4BEEQBEG4Pi4RPAHA9u3bERMTg6SkJCQlJWHYsGF4//33xeN6vR4lJSUWu1pv2rQJcXFxSE9PBwCMHz8ecXFx+PzzzwE0T7/t3LkTEydOxODBg7Fs2TKkp6fjww8/7FnnbKBUKvHiiy+2mSIkpIH0YAfSgi1ID3YgLXoGTujs+jyCIAiCIIjbGJcZeSIIgiAIgmABCp4IgiAIgiA6AQVPBEEQBEEQnYCCJ4IgCIIgiE5AwRPDvPHGG4iKioKnpydGjhyJQ4cOSW2S25GRkQGO4yx+Wm8SLQgCMjIyEBoaCpVKhYkTJ+Lnn3+2uEZVVRVSU1Oh0Wig0WiQmpqK6urqnnbF5cjJycF9992H0NBQcByHzz77zOK4s559cXExJkyYAJVKhTvuuAP/+Mc/Or2P1e1AR3qkpaW1ea+Yiwqb0el0ePLJJxEYGAhvb2/MnDkTv/76q0WfCxcu4L777oO3tzcCAwOxaNEiNDU1dbt/rsLKlSvxm9/8Bj4+PggKCsL999+PkpISiz7Oes4HDx7EyJEj4enpiTvvvBObNm3qdv/cBQqeGGXnzp1YvHgxnn/+eRw/fhz33HMPpk2bhgsXLkhtmtsxZMgQlJWViT/FxcXisVWrVmHt2rXYuHEjjh49iuDgYNx7772oq6sT+8ybNw+FhYXIzMxEZmYmCgsLkZqaKoUrLkV9fT2GDx+OjRs3Wj3ujGdfW1uLe++9F6GhoTh69Cg2bNiA1atXY+3atd3un6vRkR4AkJycbPFe2bNnj8XxxYsXY/fu3dixYwe+++473LhxAzNmzBA3djUajZg+fTrq6+vx3XffYceOHdi1axeefvrpbvXNlTh48CCeeOIJ5OXlYe/evTAYDEhKSkJ9fb3YxxnP+dy5c0hJScE999yD48eP47nnnsOiRYuwa9euHvfZJREIJhk9erSwYMECi7ZBgwYJS5culcgi9+TFF18Uhg8fbvWYyWQSgoODhVdffVVsa2xsFDQajbBp0yZBEAThxIkTAgAhLy9P7JObmysAEE6dOtW9xrsRAITdu3eLfzvr2b/xxhuCRqMRGhsbxT4rV64UQkNDBZPJ1N1uuSy36iEIgvCHP/xB+O1vf2vznOrqakGhUAg7duwQ2y5duiTwPC9kZmYKgiAIe/bsEXieFy5duiT2+fDDDwWlUinU1NQ42Qv34OrVqwIA4eDBg4IgOO85P/PMM8KgQYMs7vXYY48JY8eO7W6X3AIaeWKQpqYmHDt2DElJSRbtSUlJOHLkiERWuS+nT59GaGgooqKi8PDDD+Ps2bMAmr+ZlZeXW+igVCoxYcIEUYfc3FxoNBqMGTNG7DN27FhoNBrSqgs469nn5uZiwoQJFgUDp06disuXL+P8+fM944wbceDAAQQFBWHAgAFIT0/H1atXxWPHjh2DXq+30Cw0NBRDhw610GPo0KEWm7BOnToVOp0Ox44d6zlHXIiamhoAgL+/PwDnPefc3Nw2nzFTp07F999/D71e360+uQMUPDHItWvXYDQa22x63KdPH5SXl0tklXsyZswYvPfee/jmm2+wZcsWlJeXY9y4cbh+/br4rNvToby8HEFBQW2uGxQURFp1AWc9+/LycqvXaH0Pwj6mTZuG7du3Y//+/VizZg2OHj2KSZMmQafTAWh+nh4eHvDz87M471bNbtXDz88PHh4epIcVBEHAkiVLcPfdd2Po0KEAnPecbb03DAYDrl271l0uuQ2SbgxMtA/HcRZ/C4LQpo3oGtOmTRN/j4mJQXx8PPr164dt27aJybAd6WBNE9LKOTjj2Vu7hq1zCds89NBD4u9Dhw7FqFGjEBERga+++goPPPCAzfPo/eI4CxcuRFFREb777rsO+9J7o2ehkScGCQwMhEwma/NN7OrVq22+KRDOxdvbGzExMTh9+rS46q49HYKDg3HlypU216moqCCtuoCznn1wcLDVawBtR7WIzhESEoKIiAicPn0aQPOzbmpqQlVVlUW/WzW7VY+qqiro9XrS4xaefPJJfP7558jOzkbfvn3Fdmc9Z1vvDblcjoCAgO5wya2g4IlBPDw8MHLkSOzdu9eife/evRg3bpxEVt0e6HQ6nDx5EiEhIYiKikJwcLCFDk1NTTh48KCoQ3x8PGpqalBQUCD2yc/PR01NDWnVBZz17OPj45GTk2OxRDsrKwuhoaGIjIzsGWfclOvXr+PixYsICQkBAIwcORIKhcJCs7KyMvz0008Wevz0008oKysT+2RlZUGpVGLkyJE96wCjCIKAhQsX4tNPP8X+/fsRFRVlcdxZzzk+Pr7NZ0xWVhZGjRoFhULRXe65D5KkqRMdsmPHDkGhUAjvvPOOcOLECWHx4sWCt7e3cP78ealNcyuefvpp4cCBA8LZs2eFvLw8YcaMGYKPj4/4nF999VVBo9EIn376qVBcXCzMnTtXCAkJEWpra8VrJCcnC8OGDRNyc3OF3NxcISYmRpgxY4ZULrkMdXV1wvHjx4Xjx48LAIS1a9cKx48fF3755RdBEJzz7Kurq4U+ffoIc+fOFYqLi4VPP/1UUKvVwurVq3vcX9ZpT4+6ujrh6aefFo4cOSKcO3dOyM7OFuLj44U77rjDQo8FCxYIffv2Fb799lvhhx9+ECZNmiQMHz5cMBgMgiAIgsFg+P/bu7uQqLY+DODPaDOmjV8zmpNaTmaZ2pQfEUXojCjZF2pCiFgYQmphZaBBlqQXeWdZEUVlc3NEoaigutHKGbVAKk3ThKxMrdSivJA0Ql3nItzvmZP5NqdR+3h+MBd77bXX/u89IA9rr9mKZcuWiZiYGNHY2Chu3bolfH19RXZ29kxd9k9n165dwtXVVZhMJtHb2yt9hoaGpD62uM8vXrwQTk5OYv/+/eLJkyeirKxMyOVycfny5Wm/5l8Rw9NP7PTp08LPz08oFAoRHh4u/VSVbCc5OVnMmzdPyOVy4e3tLZKSkkRbW5u0f2xsTBw5ckRoNBrh4OAgoqKixOPHjy3GeP/+vUhNTRXOzs7C2dlZpKamioGBgem+lF9OTU2NAPDVJy0tTQhhu3vf0tIiIiMjhYODg9BoNKKwsJCvKZjAZN/H0NCQWLdunfD09BRyuVwsWLBApKWlie7ubosxhoeHRXZ2tlCpVMLR0VFs3rz5qz5dXV1i06ZNwtHRUahUKpGdnW3xKok/3UTfAQBhNBqlPra6zyaTSYSFhQmFQiG0Wq04c+bMdFzib0EmBF+1S0RERPS9uOaJiIiIyAoMT0RERERWYHgiIiIisgLDExEREZEVGJ6IiIiIrMDwRERERGQFhiciIiIiKzA8EREREVmB4YmIfgsGgwE5OTnStlarRWlp6X8er7CwEKGhobYo7T8pKChARkbGd/XNzc3F3r17p7giIhrH8EREfzyZTIZr165ZtOXm5uL27dszUk9/fz9OnDiB/Pz87+p/4MABGI1GdHZ2TnFlRAQwPBERTUipVEKtVs/IucvKyrBmzRpotdrv6j937lysW7cOZ8+endrCiAgAwxMRzYDr16/Dzc0NY2NjAIBHjx5BJpMhLy9P6pOZmYmUlBQAwPv375GSkgJfX184OTlBp9OhoqLCJrWMB5QtW7ZAJpNJ2/9+bLdjxw4kJiaiuLgYXl5ecHNzQ1FREUZGRpCXlweVSgVfX19cvHjRYvzXr18jOTkZ7u7uUKvVSEhIwMuXLyetqbKyEvHx8RZtly9fhk6ng6OjI9RqNWJjY/Hx40dpf3x8vM3uCRFNjuGJiKZdVFQUBgcH0dTUBAAwm83w8PCA2WyW+phMJuj1egDAp0+fEBERgRs3bqC1tRUZGRnYvn07GhoafriW+/fvAwCMRiN6e3ul7YncuXMHb968QW1tLY4dO4bCwkJs3rwZ7u7uaGhoQFZWFrKystDT0wMAGBoaQnR0NJRKJWpra1FfXw+lUon169fj8+fPE55jYGAAra2tWLlypdTW29uLlJQUpKeno729HSaTCUlJSfjn/3VftWoVenp60NXV9cP3hIgmx/BERNPO1dUVoaGhMJlMAL4Epf3796O5uRmDg4Po6+vD06dPYTAYAAA+Pj7Izc1FaGgo/P39sWfPHsTFxeHSpUs/XIunpycAwM3NDRqNRtqeiEqlwsmTJxEYGIj09HQEBgZiaGgI+fn5WLx4MQ4ePAiFQoG7d+8C+DKDZGdnhwsXLkCn0yEoKAhGoxHd3d3Stf9bV1cXhBDw9vaW2np7ezEyMoKkpCRotVrodDrs3r0bSqVS6uPj4wMA/3dWi4h+HMMTEc0Ig8EAk8kEIQTq6uqQkJCAZcuWob6+HjU1NfDy8sLSpUsBAKOjozh69CiWL18OtVoNpVKJqqoqdHd3T2vNISEhsLP7359NLy8v6HQ6adve3h5qtRpv374FADx8+BDPnj2Ds7MzlEollEolVCoVPn36hOfPn094juHhYQDA7NmzpbYVK1YgJiYGOp0OW7duxfnz5zEwMGBxnKOjI4Avs11ENLVmzXQBRPRnMhgMKCsrQ3NzM+zs7BAcHAy9Xg+z2YyBgQHpkR0AlJSU4Pjx4ygtLYVOp8OcOXOQk5PzzUdfU0Uul1tsy2SyCdvG13KNjY0hIiIC5eXlX431rRkuDw8PAF8e3433sbe3R3V1Ne7du4eqqiqcOnUKhw4dQkNDAxYuXAgA+PDhw6TjEpHtcOaJiGbE+Lqn0tJS6PV6yGQy6PV6mEwmi/VOAKSZqW3btmHFihXw9/dHR0eHzWqRy+UYHR212XjjwsPD0dHRgblz5yIgIMDi4+rqOuExixYtgouLC548eWLRLpPJsHbtWhQVFaGpqQkKhQJXr16V9re2tkIulyMkJMTm10FElhieiGhGjK97+uuvv6S1TVFRUWhsbLRY7wQAAQEB0sxLe3s7MjMz0dfXZ7NatFotbt++jb6+vq8eh/2I1NRUeHh4ICEhAXV1dejs7ITZbMa+ffvw6tWrCY+xs7NDbGws6uvrpbaGhgYUFxfjwYMH6O7uxpUrV/Du3TsEBQVJferq6hAZGSk9viOiqcPwREQzJjo6GqOjo1JQcnd3R3BwMDw9PS2CQUFBAcLDwxEXFweDwQCNRoPExESb1VFSUoLq6mrMnz8fYWFhNhvXyckJtbW1WLBgAZKSkhAUFIT09HQMDw/DxcXlm8dlZGSgsrJSevzn4uKC2tpabNy4EUuWLMHhw4dRUlKCDRs2SMdUVFRg586dNqudiL5NJv75W1ciIppxQgisXr0aOTk50ruuJnPz5k3k5eWhpaUFs2ZxKSvRVOPMExHRT0Ymk+HcuXMYGRn5rv4fP36E0WhkcCKaJpx5IqLfWnl5OTIzMyfc5+fnh7a2tmmuiIh+dQxPRPRbGxwcRH9//4T75HI5/Pz8prkiIvrVMTwRERERWYFrnoiIiIiswPBEREREZAWGJyIiIiIrMDwRERERWYHhiYiIiMgKDE9EREREVmB4IiIiIrLC3whKc5qDIfvhAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import altair as alt\n",
    "import matplotlib\n",
    "show = df.copy()\n",
    "\n",
    "def median_filter(scores, frac=0.05):\n",
    "    bounds = [(s.index.min(), s.index.max()) for s in scores]\n",
    "    bounds = (min(b[0] for b in bounds), max(b[1] for b in bounds))\n",
    "    frac = 0.005\n",
    "    width = (bounds[1] - bounds[0]) * frac\n",
    "    num = int((bounds[1] - bounds[0]) / frac)\n",
    "    times = np.linspace(*bounds, num=num)\n",
    "\n",
    "    # len(times[:-1]), len(times[1:])\n",
    "    plot = []\n",
    "    for beg, end in zip(times[:-1], times[1:]):\n",
    "        # get scores in beg, end\n",
    "        valid = []\n",
    "        for score in scores:\n",
    "            idx = (beg <= score.index) & (score.index < end)\n",
    "            v = score[idx]\n",
    "            valid += [v.values]\n",
    "        valid = np.concatenate(valid)\n",
    "        if len(valid) > 0:\n",
    "            plot += [{'scored_at': beg, 'best_score': np.mean(valid)}]\n",
    "    return pd.DataFrame(plot)\n",
    "    \n",
    "\n",
    "cmap = 'viridis'\n",
    "cmap = matplotlib.cm.get_cmap(cmap)\n",
    "\n",
    "colors = {alg: i for i, alg in enumerate(_algs)}\n",
    "colors = {k: cmap(v / len(colors)) for k, v in colors.items()}\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "for alg in show.alg.unique():\n",
    "    score_times = []\n",
    "    for repeat in show.repeat.unique():\n",
    "        s = show[(show.alg == alg) & (show.repeat == repeat)]\n",
    "        t = s.pivot_table(index='scored_at', values='best_score')\n",
    "        t = t.sort_values(by='scored_at')\n",
    "#         diffs = np.diff(t.index.flat[:])\n",
    "#         diffs = diffs[np.abs(diffs) > 1e-10]\n",
    "#         if not all(diffs >= 0):\n",
    "#             print(diffs)\n",
    "        if len(t.values) == 0:\n",
    "            continue\n",
    "        t['best_score'] = np.maximum.accumulate(t.values)\n",
    "        t.index -= t.index.min()\n",
    "        score_times = [t['best_score']]\n",
    "        style = '--' if alg in {'sh-0', 'random'} else '-'\n",
    "    w = median_filter(score_times)\n",
    "    w.plot(x='scored_at', y='best_score', ax=ax,\n",
    "           label=alg, color=colors[alg], style=style)\n",
    "plt.legend(loc='best')\n",
    "plt.ylabel('best_score')\n",
    "plt.xlabel('wall_time (s)')\n",
    "plt.grid(linestyle='--')\n",
    "plt.ylim(-0.15, -0.08)\n",
    "# plt.savefig('autoencoder_wall_time.png', dpi=300, bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [],
   "source": [
    "# scores = {k: }\n",
    "# all_history['sh-0']\n",
    "import toolz\n",
    "hist = sum(all_history.values(), [])\n",
    "hist = list(toolz.concat(hist))\n",
    "scores = {}\n",
    "for h in hist:\n",
    "    calls, score, params, model_id = scores.get(h['model_id'], (-1, -np.inf, {}, 0))\n",
    "    if h['partial_fit_calls'] > calls and h['score'] > score:\n",
    "        scores[h['model_id']] = (h['partial_fit_calls'], h['score'], h['params'], h['model_id'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "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>model_id</th>\n",
       "      <th>param__batch_size</th>\n",
       "      <th>param__module__activation</th>\n",
       "      <th>param__module__init</th>\n",
       "      <th>param__optimizer</th>\n",
       "      <th>param__optimizer__lr</th>\n",
       "      <th>param__optimizer__momentum</th>\n",
       "      <th>param__optimizer__weight_decay</th>\n",
       "      <th>partial_fit_calls</th>\n",
       "      <th>score</th>\n",
       "      <th>loss</th>\n",
       "      <th>rank</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>11</td>\n",
       "      <td>512</td>\n",
       "      <td>PReLU</td>\n",
       "      <td>kaiming_normal_</td>\n",
       "      <td>&lt;class 'torch.optim.adam.Adam'&gt;</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.452959e-07</td>\n",
       "      <td>33</td>\n",
       "      <td>-0.088522</td>\n",
       "      <td>0.088522</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>29</td>\n",
       "      <td>64</td>\n",
       "      <td>PReLU</td>\n",
       "      <td>xavier_normal_</td>\n",
       "      <td>&lt;class 'torch.optim.sgd.SGD'&gt;</td>\n",
       "      <td>0.551749</td>\n",
       "      <td>0.852853</td>\n",
       "      <td>4.117320e-05</td>\n",
       "      <td>27</td>\n",
       "      <td>-0.091809</td>\n",
       "      <td>0.091809</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>32</td>\n",
       "      <td>PReLU</td>\n",
       "      <td>kaiming_normal_</td>\n",
       "      <td>&lt;class 'torch.optim.rmsprop.RMSprop'&gt;</td>\n",
       "      <td>0.002900</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.224865e-07</td>\n",
       "      <td>33</td>\n",
       "      <td>-0.092263</td>\n",
       "      <td>0.092263</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>10</td>\n",
       "      <td>128</td>\n",
       "      <td>LeakyReLU</td>\n",
       "      <td>xavier_normal_</td>\n",
       "      <td>&lt;class 'torch.optim.sgd.SGD'&gt;</td>\n",
       "      <td>0.051489</td>\n",
       "      <td>0.982983</td>\n",
       "      <td>2.608654e-07</td>\n",
       "      <td>55</td>\n",
       "      <td>-0.093597</td>\n",
       "      <td>0.093597</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>34</td>\n",
       "      <td>256</td>\n",
       "      <td>PReLU</td>\n",
       "      <td>kaiming_normal_</td>\n",
       "      <td>&lt;class 'torch.optim.adam.Adam'&gt;</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.024447e-05</td>\n",
       "      <td>27</td>\n",
       "      <td>-0.095980</td>\n",
       "      <td>0.095980</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    model_id  param__batch_size param__module__activation param__module__init  \\\n",
       "11        11                512                     PReLU     kaiming_normal_   \n",
       "29        29                 64                     PReLU      xavier_normal_   \n",
       "6          6                 32                     PReLU     kaiming_normal_   \n",
       "10        10                128                 LeakyReLU      xavier_normal_   \n",
       "34        34                256                     PReLU     kaiming_normal_   \n",
       "\n",
       "                         param__optimizer  param__optimizer__lr  \\\n",
       "11        <class 'torch.optim.adam.Adam'>                   NaN   \n",
       "29          <class 'torch.optim.sgd.SGD'>              0.551749   \n",
       "6   <class 'torch.optim.rmsprop.RMSprop'>              0.002900   \n",
       "10          <class 'torch.optim.sgd.SGD'>              0.051489   \n",
       "34        <class 'torch.optim.adam.Adam'>                   NaN   \n",
       "\n",
       "    param__optimizer__momentum  param__optimizer__weight_decay  \\\n",
       "11                         NaN                    4.452959e-07   \n",
       "29                    0.852853                    4.117320e-05   \n",
       "6                          NaN                    1.224865e-07   \n",
       "10                    0.982983                    2.608654e-07   \n",
       "34                         NaN                    2.024447e-05   \n",
       "\n",
       "    partial_fit_calls     score      loss  rank  \n",
       "11                 33 -0.088522  0.088522     0  \n",
       "29                 27 -0.091809  0.091809     1  \n",
       "6                  33 -0.092263  0.092263     2  \n",
       "10                 55 -0.093597  0.093597     3  \n",
       "34                 27 -0.095980  0.095980     4  "
      ]
     },
     "execution_count": 164,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d = list(scores.values())\n",
    "out = []\n",
    "for di in d:\n",
    "    params = {'param__' + k: v for k, v in di[2].items()}\n",
    "    params['param__optimizer'] = str(params['param__optimizer'])\n",
    "    for key in ['param__optimizer__nesterov', 'param__train_split']:\n",
    "        if key in params:\n",
    "            params.pop(key)\n",
    "    out += [{'partial_fit_calls': di[0],\n",
    "             'score': di[1],\n",
    "             'model_id': di[3],\n",
    "             **params}]\n",
    "show = pd.DataFrame(out)\n",
    "show['loss'] = show['score'] * -1\n",
    "show = show.sort_values(by='loss')\n",
    "show['rank'] = np.arange(len(show), dtype=int)\n",
    "# show.to_csv('')\n",
    "show.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "model_id                                                       11\n",
       "param__batch_size                                             512\n",
       "param__module__activation                                   PReLU\n",
       "param__module__init                               kaiming_normal_\n",
       "param__optimizer                  <class 'torch.optim.adam.Adam'>\n",
       "param__optimizer__lr                                          NaN\n",
       "param__optimizer__momentum                                    NaN\n",
       "param__optimizer__weight_decay                        4.45296e-07\n",
       "partial_fit_calls                                              33\n",
       "score                                                  -0.0885225\n",
       "loss                                                    0.0885225\n",
       "rank                                                            0\n",
       "Name: 11, dtype: object"
      ]
     },
     "execution_count": 165,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_idx = 0  # because sorted\n",
    "show.iloc[best_idx]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/ssievert/anaconda3/envs/dask-master/lib/python3.6/site-packages/matplotlib/figure.py:459: UserWarning: matplotlib is currently using a non-GUI backend, so cannot show the figure\n",
      "  \"matplotlib is currently using a non-GUI backend, \"\n",
      "/Users/ssievert/anaconda3/envs/dask-master/lib/python3.6/site-packages/matplotlib/figure.py:459: UserWarning: matplotlib is currently using a non-GUI backend, so cannot show the figure\n",
      "  \"matplotlib is currently using a non-GUI backend, \"\n",
      "/Users/ssievert/anaconda3/envs/dask-master/lib/python3.6/site-packages/matplotlib/figure.py:459: UserWarning: matplotlib is currently using a non-GUI backend, so cannot show the figure\n",
      "  \"matplotlib is currently using a non-GUI backend, \"\n",
      "/Users/ssievert/anaconda3/envs/dask-master/lib/python3.6/site-packages/matplotlib/figure.py:459: UserWarning: matplotlib is currently using a non-GUI backend, so cannot show the figure\n",
      "  \"matplotlib is currently using a non-GUI backend, \"\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkgAAAHFCAYAAAAJ2AY0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzsnXt8VNW5/p+955IECfd7yAW8EKxoIng0QUBQIgQrtVU42oa2QpWDWIEeT6VIq1ZFWn82tvVST09rL1YR7y0BiYcAWiJWIMSjgrQFAgiSoMagSWZm7/X7I9k7syczk0myd7Jn9fl+PvlAVva8z3rWu/ead9asPaMIIQQIIYQQQoiJ2tsdIIQQQghxGyyQCCGEEEIiYIFECCGEEBIBCyRCCCGEkAhYIBFCCCGERMACiRBCCCEkAhZIhBBCCCERsEAihBBCCImABRIhhBBCSAQskAghtnDXXXdBURTU1dXZEq+srAx33XVXlx+fk5ODq666ypa+dMShQ4egKAqefPLJHtEjhDgPCyRCiCspKyvD3Xff3dvdSIiRI0eisrISc+bM6e2uEEJswtvbHSCEkGQnJSUFl1xySW93gxBiI1xBIoTYypEjR/DVr34V/fr1Q//+/fGNb3wDtbW15t/XrVuHoqIijBw5EmlpaRg/fjzuuOMOfP755+Yx3/rWt/DII48AABRFMX8OHToEANB1Hb/4xS+Ql5eHtLQ0DBgwAJdccgleeeWVdv3ZtGkTLrzwQqSlpSE3Nxe/+c1vOu1p/fr1uPjii9G/f3/06dMHY8eOxY033mj+PdpbbOH9jvwxfADA22+/jauvvhqDBg1Camoq8vPz8eyzz3a6j4QQe+EKEiHEVq655hrMmzcPixcvxrvvvovVq1fjvffew86dO+Hz+XDgwAEUFxdj2bJlOOOMM7Bv3z6sXbsWb731FrZs2QIAWL16NT7//HM899xzqKysNGOPHDkSQEsB9cc//hELFy7EPffcA7/fj927d1sKDwDYu3cvvve97+GOO+7A8OHD8etf/xoLFy7EWWedhalTpybkp7KyEvPnz8f8+fNx1113ITU1FYcPHzb7Gu9x4TQ2NqKkpASapmHQoEEAgIqKCsyaNQsXX3wxHn/8cfTv3x/PPPMM5s+fjy+++ALf+ta3EuojIcQBBCGE2MCPfvQjAUAsX77c0v7UU08JAOKPf/xju8foui6CwaDYtm2bACD27t1r/u2WW24R0aao7du3CwBi1apVcfuTnZ0tUlNTxeHDh822xsZGMWjQIHHzzTcn7OvBBx8UAMSnn34a85iDBw8KAOK3v/1t1L+HQiExd+5c0bdvX7Fr1y6zPTc3V+Tn54tgMGg5/qqrrhIjR44UmqYl3E9CiL3wLTZCiK18/etft/w+b948eL1eVFRUAAD++c9/4oYbbsCIESPg8Xjg8/kwbdo0AMD777/fYfyNGzcCAG655ZYOj83Ly0NWVpb5e2pqKs455xwcPnw4YT8XXXSR6ePZZ5/FsWPHEn6swdKlS7FhwwasX78eF154IQDg73//O/bt22eOVygUMn+Ki4tx/Phx7N+/v9NahBB7YIFECLGVESNGWH73er0YPHgwTp06hdOnT2PKlCnYuXMn7r33XmzduhV/+9vf8MILLwBoeRuqI2pra+HxeNrpRGPw4MHt2lJSUhLSMZg6dSpeeuklhEIhLFiwAKNHj8Z5552Hp59+OqHH33vvvXj88cfxq1/9CrNmzTLbP/roIwDAf/7nf8Ln81l+lixZAgC2fWQCIaTzcA8SIcRWTpw4gYyMDPP3UCiEU6dOYfDgwdiyZQs+/PBDbN261Vw1AoBPP/004fhDhw6Fpmk4ceKEuSfJaebOnYu5c+eiubkZb775JtasWYMbbrgBOTk5KCgoiPm4J598EqtXr8Zdd91l2dQNAEOGDAEArFy5El/96lejPn7cuHH2mSCEdAquIBFCbOWpp56y/P7ss88iFArhsssug6IoAFpWccL51a9+1S6OcUzkas/s2bMBAI899phtfU6UlJQUTJs2DWvXrgUA7NmzJ+axmzZtwne+8x3ceOON+NGPftTu7+PGjcPZZ5+NvXv3YtKkSVF/0tPTHfNCCIkPV5AIIbbywgsvwOv1YubMmeZdbBdccAHmzZuHhoYGDBw4EIsXL8aPfvQj+Hw+PPXUU9i7d2+7OBMmTAAArF27FrNnz4bH48H555+PKVOmoKSkBPfeey8++ugjXHXVVUhJScGePXvQp08f3Hrrrbb6+eEPf4ijR4/i8ssvx+jRo/Hpp5/i4YcftuydiuTgwYO47rrrMHbsWHz729/Gm2++afl7fn4+UlJS8Ktf/QqzZ8/GlVdeiW9961vIyMjAxx9/jPfffx+7d+/G+vXrbfVCCOkEvb1LnBAiB8ZdbLt27RJf/vKXRd++fUV6erq4/vrrxUcffWQet2PHDlFQUCD69Okjhg4dKhYtWiR2797d7i6w5uZmsWjRIjF06FChKIoAIA4ePCiEEELTNPGzn/1MnHfeecLv94v+/fuLgoIC8ec//9l8fHZ2tpgzZ067fk6bNk1MmzYtYV9/+ctfxOzZs0VGRobw+/1i2LBhori4WLz++uvmMZF3sVVUVAgAMX8MH0IIsXfvXjFv3jwxbNgw4fP5xIgRI8SMGTPE448/nnAfCSH2owghRC/VZoQQQgghroR7kAghhBBCIuAeJELIvyyhUCju31VVharydSQh/4rwyieE/MsS+flDkT+Rt+YTQv516PUC6dFHH8WYMWOQmpqKiRMn4vXXX4957Lvvvouvfe1ryMnJgaIoKC0tbXfMmjVrcNFFFyE9PR3Dhg3DV77yFX4aLSEkKn/729/i/tx111293UVCSC/Rq2+xrVu3DsuWLcOjjz6KyZMnm7e8vvfee5avBzD44osvMHbsWFx33XVYvnx51Jjbtm3DLbfcgosuugihUAirVq1CUVER3nvvPZxxxhlOWyKEJBGTJk3q7S4QQlxKr97FdvHFF+PCCy+0fODb+PHj8ZWvfAVr1qyJ+9icnBwsW7YMy5Yti3tcbW0thg0bhm3btiX87d2EEEII+dem11aQAoEAdu3ahTvuuMPSXlRUhB07dtimU19fDwAYNGhQzGOam5vR3Nxs/q7rOj7++GMMHjzY/ORfQgghhLgbIQQaGhowatSobt9g0WsFUl1dHTRNw/Dhwy3tw4cPx4kTJ2zREEJgxYoVuPTSS3HeeefFPG7NmjW4++67bdEkhBBCSO9y5MgRjB49ulsxev02/8gVGiGEbas2S5cuRXV1Nd544424x61cuRIrVqwwf6+vr0dWVhYOHDiAgQMHAmi53dfj8UDTNOi6bh5rtIdCIYS/W+nxeKCqasz2xsZGVFVVIS8vD16vF15vSyoibzuO1e7z+aDrOjRNM9sURYHX6zXbQ6EQqqqqkJ+fj9TU1Jh976onI/7EiRPh9/sRDAYT6ntnPGmahr179yI/P99yXhhe7fCk6zrefvttXHDBBWbfDK92eWpubjbz7fP5LHmyy1NkPmLlr6uejPgXXXQRVFWNe+511ZOu66iqqsIFF1wAj8fTru92eAqFQnj77bfNaw9I7HrqjKdAIGDm2+/3d2mO6MiTkY9JkyaZvsLprqdQKGR+T5yiKLbMe5GeAGDXrl2Way+a1+54am5uxu7du8182z2XB4NBMxd5eXlITU2N2vfuekokH93xFB4/cueNXc9PANrNtXbO5aqqora2FmeddZYt32PYawXSkCFD4PF42q0WnTx5st2qUle49dZb8corr2D79u0dVpEpKSntvjwTAAYOHIjBgwd3uy/RSEtLQ1NTEwYNGgSfz+eIRjAYRFNTEwYOHOiIhhG/f//+jnr44osvHNdobGzskVz0hIZTY2XET09PdzzfTp2zhkZP5aInNPr16+dYvhsbGx3Pt0zXntMaTuZDpnwD7RdfukKv3ebv9/sxceJElJeXW9rLy8tRWFjY5bhCCCxduhQvvPACtmzZgjFjxnS3q4QQQgj5F6NX32JbsWIFSkpKMGnSJBQUFOCJJ55ATU0NFi9eDABYsGABMjIyzDvaAoEA3nvvPfP/x44dQ1VVFfr27YuzzjoLAHDLLbfgT3/6E15++WWkp6ebK1T9+/dHWlpaL7gkhBBCSNLRU9+KG4tHHnlEZGdnC7/fLy688EKxbds282/Tpk0T3/zmN83fjW/MjvwJ/2buaH9HxLeEd0R9fb0AID755BMbHEZH0zRx6NAhoWla0mrI4IEa7olPDXdpyOCBGu6J31Man3zyiQAg6uvrux2rVz8Hya189tln6N+/P+rr69GvX7/e7g4hhBBCEsDO5+9e/6oRN9PRF1l2N/aWLVuSWkMGD9RwT3xquEtDBg/UcE/8ntSwCxZIcXBycU20fphVMmvI4IEa7olPDXdpyOCBGu6J35MadsECiRBCCCEkAhZIhBBCCCERcJN2FIxNXp988gkGDBjgiIau66irq8OQIUO6/X0xvaUhgwdquCc+NdylIYMHargnfk9pfPrppxg4cKAtm7RZIEWBd7ERQgghyQfvYushon1vkJ2xN2zYkNQaMnighnviU8NdGjJ4oIZ74vekhl2wQOpFnLzVsac0ZPBADffEp4a7NGTwQA33xO8pDbtggUQIIYQQEgELJEIIIYSQCLhJOwrGJq9PP/0U/fv3d0TD+MCs9PR0KIqSlBoyeKCGe+JTw10aMnighnviO6FRc88Ey+9ZP3wH9fX1GDBgADdpJztpaWlJryGDB2q4Jz413KUhgwdquCd+T2nYBQukODj9fTFlZWVJrSGDB2q4Jz413KUhgwdquCd+T2rYBQskQgghhJAIvL3dAUIIIYT8a7Ft6jTL79O2b+ulnsSGK0iEEEIIIRHwLrYo9NRdbKFQCF6v19E7BpzUkMEDNdwTnxru0pDBAzXcEz9SY/u0yyx/68oKEu9ik5jGxsak15DBAzXcE58a7tKQwQM13BO/pzTsggVSHJzeaV9RUZHUGjJ4oIZ74lPDXRoyeKCGe+L3pIZdcJM2IYQQ8i9EtLemSHu4gkQIIYQQEgELpF7E63V+Ac9pDRk8UMM98anhLg0ZPFDDPfF7SsMueBdbFIy72OzYBU8IIYR0BqffAnPDW2x2fA5SNB92Pn9zBSkOuq47GvvkyZNJrSGDB2q4Jz413KUhgwdquCd+T2rYBQukOGia5mjsysrKpNaQwQM13BOfGu7SkMEDNdwTvyc17IIFEiGEEEJIBCyQCCGEEEIiSJ7t5L2AUx+3bsROT09Pag0ZPFDDPfGp4S6NZPQQbdNuT/tw6ktYme/ENWyLxbvY2sO72AghJPmQ+e4sO5F5nHgXWw/h9E77w4cPJ7WGDB6o4Z741HCXhgweqOGe+D2pYRcskOLg9E77qqqqpNaQwQM13BOfGu7SkMEDNdwTvyc17IJ7kAghhNjOxNt/b/l9108X9FJPCOkaXEEihBBCCImAK0hxcHqn/dChQ5NaQwYP1HBPfGq4SyM8vpN3ZiX7OMmiIYMHQ8MuWCDFwckv1fN6vSgsLHQsfk9oyOCBGu6JTw13acjggRruid+TGnbBt9ji4PRGsn379iW1hgweqOGe+NRwl4YMHqjhnvg9qWEXLJDi4PStiPv3709qDRk8UMM98anhLg0ZPFDDPfF7UsMuWCARQgghhETAAokQQgghJAIWSHFQVeeGR1VVZGVlJbWGDB6o4Z741HCXhgweqOGe+D2pYRe8iy0OHo/H0dj5+fmOxe8JDRk8UMM98anhLg0ZPFDDPfF7UsMuuIIUB6d32u/ZsyepNWTwQA33xKeGuzRk8EAN98TvSQ27YIEUB6d32tfU1CS1hgweqOGe+NRwl4YMHqjhnvg9qWEXLJAIIYQQQiJggUQIIYQQEgELpDg4vdN+3LhxSa0hgwdquCc+NdylIYMHargnfk9q2AXvYouD03ex5ebmOha/JzRk8EAN98Snhrs0ZPBADffE70kNu+AKUhxCoZCjsXfs2JHUGjJ4oIZ74ierRs09Eyw/yaoRSTLmghrujd+TGnbBAikOQghHY9fW1ia1hgweqOGe+NRwl4YMHqjhnvg9qWEXLJAIIYQQQiLo9QLp0UcfxZgxY5CamoqJEyfi9ddfj3nsu+++i6997WvIycmBoigoLS3tdkxCCCGEkEh6tUBat24dli1bhlWrVmHPnj2YMmUKZs+ejZqamqjHf/HFFxg7diweeOABjBgxwpaY8XB6k3ZeXl5Sa8jggRruiU8Nd2nI4IEa7onfkxp20asF0kMPPYSFCxdi0aJFGD9+PEpLS5GZmYnHHnss6vEXXXQRfvrTn+Lf//3fkZKSYkvMeDh9K2J2dnZSa8jggRruiU8Nd2nI4IEa7onfkxq2xbItUicJBALYtWsXioqKLO1FRUXYsWOHK2I6vdN+y5YtSa0hgwdquCc+NdylIYMHargnfk9q2EWvfQ5SXV0dNE3D8OHDLe3Dhw/HiRMnejRmc3Mzmpubzd8/++wzAC0FVzAYBNBSlXo8HmiaZvmuF6M9FApZds97PB6oqhqzPRAIoKGhAYFAAEIIeL0tqYhMbqx2n88HXdctX8ynKAq8Xq/ZHgwG0dDQgGAwCK/XG7PvXfVkxNc0DV6v1xyrjvreGU+hUMjUCO+L4dUOT0IISy7CvdrlKTzfxvGx8tdVT5H5iJW/rnoy4gshOjz3uupJ0zTznO3M9WR4mrr66Za+aoAA8Ob917fzpOt6u3wncj3F8qSpLavZitCgihA0TbPkW1GULs0R4XnS1BSoegCAgK6mIBgMmvnQdR1CiC7NEfHyFJ5v4fUCYa/MNU3r0JM/7J2OYKtU5LkX7doz8gR0Yd5Tw99Z0M0Y4Rp2z+XhuQgEAhCKAggB+P0Wz53xpKl+ePQAdMUDoXgtGkKIbs97uuKBKjRoqh+A0i6+E3O5oijWfLeOD3QdSijUJU/h+VZE2zxlF73+QZGKolh+F0K0a3M65po1a3D33Xe3a6+oqECfPn0AAFlZWcjPz0d1dbVlP9O4ceOQm5uLt956C7W1tWZ7Xl4esrOzsX37djQ0NJjtBQUFGDZsGCoqKgAA5eXlAIDp06cjLS0NZWVllj4UFxejsbHRPB5oOSnnzJmDuro6VFZWmu3p6emYMWMGjhw5gqqqKrN99+7dmDx5Mg4cOID9+/eb7XZ5OnXqFEaNGoXNmzdbLhQ7PPXt2xcAcOzYMbzzzjtm+9ChQ1FYWGiLp1GjRgFoywXQlie7PZWXl8fMk12ejh07hrFjx8Y897rjySDRc6+zns466ywALedsXV2d2d7R9WR4uqOgPwDgsd0NqG/Wo3o6ffo0gLZ8d/Z6audpwmoAwOBTbyPr6EsWT+Xl5V2eIyx5mrAauft+Dn+wHtUTVqM6zFcoFEJTU1O35ohYeTJonDEdgQkTzN8PHDjQoScjFwDw1P99DgDtzr0pU6ZYcmHQ1XmvujUXAJDecABjAPzjH/+waNg9l4d7Ki8vR/qgQVAbGlB/61IAMM/BznhKPftmjN//C3w8MA9HMq+x5NsY/+7MEZkD8zDk41344Oyb0ZQ6vF18J+by9PR0a75bx8d76BD6Pv9C1zyF5TvzyIvIAbBz507YhSKc/ECCOAQCAfTp0wfr16/HNddcY7bfdtttqKqqwrZt2+I+PicnB8uWLcOyZcu6HTPaClJmZiaOHz+OwYMHA7B/BemLL75AeXk5Zs6cCZ/P59gKUnl5OYqKipCWlubIClJ5eTlmzZqFlJQUx1aQNm/ejFmzZlneW7ZzBUnTNJSVlZm5CPdql6empiYz336/37EVpPB8OLGCVF5ejuLiYrM/kX23YwXp1VdfRVFRkakf3nc7VpCCwSA2btxoyXd3VluOri1oaW9dQcpYVYXm5mYz3ykpKd1eQTq6tsCygjT6+5VmPmbPng2fz+fICpKR779eOcuygjTl1U0deppy55/M9qAGvP3TBVFXkCJzYeQJ6Py8d/j+fws7WseYO3ejqakJr776qqnh1AqSke+3ZhdbVpAmbyzrtKejay+xrCCF57u4uBiqqnZr3jv2wEWWFaTI+JHYuYJkzLVvXT23NU0tK0iXVmzptKcjD1zSFl+EkLO6Ch999BFGjBiB+vp69OvXr52XztBrK0h+vx8TJ05EeXm5pZgpLy/H3LlzezRmSkpK1E3fqamplosWaLkoou2SD5/ME2lPTU1FQUEBUlNTLU/8kXrx2lVVjbohzWj3eDwoKCgwvcXqe1c9GfHDn2QS7XuinsI1onm1y1O0XHS277HaVVWNmu9Y+euqp8h8xPLaVU9GfOOJId65l0jfjU+ENsj64TuWczZanI48BbTo7ZExouW7q548enO79mj57uwcEd73cA2P3gyfz2eOldfrhaIoXZojYnky/m/kW4l4AjSOiecpMheRngBA1/WY11604zvyFJkLoOV5IZqGXXN5eC5SU1OhGIVI69vpkR4Smvf0lseqQgOEZtEwrr/uzHuq0Cw60eLH8ppoe7Q8hedbaR2fjvoez1O0fMe6gasr9OpbbCtWrEBJSQkmTZqEgoICPPHEE6ipqcHixYsBAAsWLEBGRgbWrFkDoGWF6L333jP/f+zYMVRVVaFv377m0nxHMTuD0zvthw0b5lj8ntCQwQM17In/y+/92fL70v/3Zds17IIa7ohPDXdpyODB0LAtlm2RusD8+fNRWlqKe+65B3l5edi+fTvKysqQnZ0NAKipqcHx48fN4z/88EPk5+cjPz8fx48fx4MPPoj8/HwsWrQo4Zidwc7NXtFib9iwIak1ZPBADffEp4a7NGTwQA33xO9JDbvo9U3aS5YswZIlS6L+bevWrZbfc3JyEvqelXgx3YSTtzr2lIYMHqjhnvjUcJeGDB6o4Z74PaVhF73+VSOEEEIIIW6j11eQCCGdJ9oGZ0IIIfbBFaQ4xLprwa7Y06dPT2oNGTxQwz3xqeEuDRk8UMM98XtSw7ZYtkUinSYtLS3pNWTw0F2Nibf/3vL7rp8usF0jUZhvaiRTfGq4S0MGD3bCFaQ4OP19MWVlZUmtkQweJt7+e8uPExqJIIOGDB6o4Z741HCXhgweDA27YIFECCGEEBIB32IjScO2qdMsv0/bHv/raAghhJCuwhUkQgghhJAIWCDFwemd9sXFxUmtIYMHargnPjXcpSGDB2q4J35PatgFC6RepLGxMek1ZPBADffEp4a7NGTwQA33xO8pDbtggRQHp3faV1RUJLWGDB6o4Z741HCXhgweqOGe+D2pYRcskAghhBBCImCBRAghhBASAQukXsTJjWpOa0z+xWQU/aoIzXozin5V5IiGQTKPk2waMnighnviU8NdGjJ4sJPk6Wkv4PP5HI09Z84cx+L3hEZABPDw0Ycdiw/IMU6yaMjggRruiU8Nd2nI4MHQsAuuIMVB13VHY588eTKpNRQoyEnNgQLFkfiAHOMki4YMHqjhnvjUcJeGDB4MDbtggRQHTdMcjV1ZWZnUGj7Fh3nD5sGnOLfSJsM4yaIhgwdquCc+NdylIYMHQ8MuWCARQgghhETAPUhJSuQ30+/66YJe6gkhhBAiHyyQ4qAo9uytqblnguX3rB++A0VRkJ6ebptGNJzWEBCoC9RBQDgSH3DeAzXcE58a7tKQwQM13BO/JzXsggVSHJz+vpgZM2Y4Fr8nNIIiiN+c+I1j8QE5xkkWDRk8UMM98anhLg0ZPBgadsE9SHFweqf94cOHk1pDhYrzzzgfqoOnkQzjJIuGDB6o4Z741HCXhgweDA27YIEUB6d32ldVVSW1hlfxYtbgWfAqzq20yTBOsmjI4IEa7olPDXdpyODB0LALFkiEEEIIIRGwQCKEEEIIiYAFUhyc3mk/dOjQpNYQEDjYeNDxu9iSfZxk0ZDBAzXcE58a7tKQwYOhYRe8iy0OTt/FVlhY6Fj8ntAIiiDW1653LD4gxzjJoiGDB2q4Jz413KUhgwdDwy64ghQHpzeS7du3L6k1PPBgcv/J8MDjSHxAjnGSRUMGD9RwT3xquEtDBg+Ghl2wQIqD07ci7t+/P6k1PEprgaQ4VyDJME6yaMjggRruiU8Nd2nI4MHQsAu+xdZJfvm9P1t+X/r/vtxLPSGEEEKIU3AFiRBCCCEkAhZIcVBV54ZHVVVkZWUltYYOHdWnq6HDueVSGcZJFg0ZPFDDPfGp4S4NGTwYGnbBt9ji4PE4t7fG4/EgPz/fsfg9oRESIWz6eJNj8QE5xkkWDRk8UMM98anhLg0ZPBgadsEVpDg4vdN+z549Sa3hVbyYNcj5rxpJ9nGSRUMGD9RwT3xquEtDBg+Ghl2wQIqD0zvta2pqklpDhYrz+zr/ZbXJPk6yaMjggRruiU8Nd2nI4MHQsAsWSIQQQgghEbBAIoQQQgiJgAVSHJzeaT9u3Lik1tCEhr/W/xWacO79ZBnGSRYNGTxQwz3xqeEuDRk8GBp2wbvY4uD0XWy5ubmOxe8JDQ0tBZKTyDBOsmjI4IEa7olPDXdpyODB0LALriDFIRQKORp7x44dSa3hU3y4buh18Ck+R+IDcoyTLBoyeKCGe+JTw10aMngwNOyCBVIchBCOxq6trU1qDQUKxqSNgQLFkfiAHOMki4YMHqjhnvjUcJeGDB4MDbtggUQIIYQQEgELJEIIIYSQCFggxcHpTdp5eXlJrRESIWw6tQkh4dz7yTKMkywaMnighnviU8NdGjJ4MDTsgnexxcHpWxGzs7Mdi98TGjp0VH9e7Vh8QI5xkkVDBg/UcE98arhLQwYPhoZtsWyLJCFO77TfsmVLUmv4FB9uHHGj43exJfs4yaIhgwdquCc+NdylIYMHQ8MuWCDFwemd9g0NDUmtoUDBEP8Qx+9iS/ZxkkVDBg/UcE98arhLQwYPhoZdsEAihBBCCImABRIhhBBCSAQskOLg9E77goKCpNYIiiCePfksgiLoSHxAjnGSRUMGD9RwT3xquEtDBg+Ghl3wLrY4OH0X27BhwxyL3xMaAgKHmg45Fh+QY5xk0ZDBAzXcE58a7tKQwYOhYVss2yJ1kUcffRRjxoxBamoqJk6ciNdffz3u8c8//zzOPfdcpKSk4Nz+kDK7AAAgAElEQVRzz8WLL75o+fvp06exdOlSjB49GmlpaRg/fjwee+yxLvUtGHRuZSQYDGLDhg1JreFX/Lht9G3wK35H4gNyjJMsGjJ4oIZ74lPDXRoyeDA07KJXC6R169Zh2bJlWLVqFfbs2YMpU6Zg9uzZqKmpiXp8ZWUl5s+fj5KSEuzduxclJSWYN28edu7caR6zfPlybNq0CX/84x/x/vvvY/ny5bj11lvx8ssv95SthHHyVsee0khRUxyND8gxTrJoyOCBGu6JTw13acjgwU56tUB66KGHsHDhQixatAjjx49HaWkpMjMzY674lJaWYubMmVi5ciVyc3OxcuVKXH755SgtLTWPqaysxDe/+U1cdtllyMnJwU033YQLLrgAb7/9dk/ZIoQQQkiS02t7kAKBAHbt2oU77rjD0l5UVIQdO3ZEfUxlZSWWL19uabvyyistBdKll16KV155BTfeeCNGjRqFrVu34oMPPsDDDz8csy/Nzc1obm42f//ss88AtCzVGct1qqrC4/FAUQXCP/ZH0zR4PB6EQiHL5y94PB6oqopQKAQtbJVF1YNm7PB/vd6WVERW17HagZZu+Fr3owWDQSiKAq/XC13XoWmaGTsUCsHn80HTNOi63taXVk+x2uN58it+860143OQIpc2O+vJ5/OZfTcwjtF1HcIf9lZea3878uRvHR9NBzSBqJ4MwvtveE3Ek98DBLS2fIQ/xvAUnu/IPBkY7YnkSVNToIgQVKFBU32Wc9U4Jlb+upqn8Me167sqIHQFUAQUte3YeJ4AQFe8EIrHjG/EjOxL+LkXz5OR76AGCEQ/J43HR8tTtHx0lCfj+laEBlWELNdeMBhM6HrqKE+amgJVDwAQ0NUUS76FEBBCJHQ9debcC9cXXi8QtrcjkXnPH7ZXNtgqFZmPaLkAOjdHWDxZVrTb5ohwje7Me7HyFJ5voSiAEEDrfNXRHB/Nk6b64dED0BUPhOK1aBieujOX64qnde7wA1DaxXdiLleUtifOYDDYNp/rOpRQqEuewvOttH7llZ0rVL1WINXV1UHTNAwfPtzSPnz4cJw4cSLqY06cONHh8T//+c/xne98B6NHj4bX64Wqqvj1r3+NSy+9NGZf1qxZg7vvvrtde0VFBfr06QMAyMrKQn5+PgacKdB3ZNsxBw4cQG5uLt566y3U1taa7Xl5ecjOzsb27dvRMGG12X7mP580YwNAeXk5AGD69OlIS0tDWVmZpQ/FxcVobGw0jwfaTsqxA7z4+nlnAADKysqQnp6OGTNm4MiRI6iqqjKP37NnDwoLC3HgwAHs37/fbDc8VVdXW97WHDduXIeelmUuM9tHpYwCAGzevNlycnbW05w5c1BXV4fKykqzPT09HdOnT8fx48dRf+vStuMPHQKADj3dUdAfALCtpgnbapqjesrKykKfPn3MXABAQUEBhg0blpCnOwr644HKevRPUfEfF6abf4vmqby8PGaehg4dmnieJqzGiBNbMPKjLTiYcwOqw/pz/Phx5OTktJx7DQ1d8hQrTx6PB16vF7W1tZY8Dc8HTuxScMZwYNA5wowVz9NgAEczrsKpwZMAANVlZTjnnHMwffp07NmzJ/b1FMeTke/HdjegvlmP6cnIRaw8AUg8T63X9+BTbyPr6EuWPJWXlyd0PXWYpwmrkbvv5/AH61E9YbUl3wDQ0NCQ8PXUmXNv5MiR8Hq9aJwxHYEJE8z2ROY9IxcA8NT/fQ6g/Rxx2WWXYcqUKZZrLzxPnfVUHTbXpjccwBgABw8eNHMBdG/e6yhP5eXlSB80CGpDgzlfGedgZzylnn0zxu//BT4emIcjmdeY+R44cCC8Xi/279/frbk8c2Aehny8Cx+cfTOaUoeb8c8//3x4vV6UlZU5NpePGzeuJRet4+M9dAh9n3+ha89PYfnOPPIicgC89dZbsAtFOPmRlnH48MMPkZGRgR07dqCgoMBsv++++/CHP/wB+/bta/cYv9+P3/3ud7j++uvNtqeeegoLFy5EU1MTAODBBx/Ef//3f+PBBx80T+qVK1fixRdfxBVXXBG1L9FWkDIzM1FbW4v+/VsucqNqfeT2VywrSIvXzOnwVceRBy4x21U9iOwf7kUgEEAoFILX6zVfAQGJV+iX/OBpywrS9h9f3+7VofGq0ufz2b6CNPPxmS05Ufz4XP8cb9z6hiOvOsL79PqVs9oadB2XbfnfDj1NXf00gLYVpJ1rbmjnSVEUNDU1mf8P95qIp6mrn7asIG3/cdv5aXgKhUJmvlVV7fYK0tG1BZYVpNHf32nmOyUlJW7+uponI35aWhqEEJa+P/GDjZYVpJvum92hp2P35VlWkEZ/vxKKokT9JNxEV1uMfBsrSG/ef70ljrGC1NTUZF574XnqymrL0bUt85exgpSxqgqappn59ng83V5BOrq2wLKCNPr7lWY+UlNToSiK7StIQgjouo6UlBRsu/wKywrSlFc3dehpyp1/MtuDGvD2Txe0O/eMGEYfwvMEdH4F6fD9/xZ2tI4xd+5GKBRCIBAw8+3ECpKRC6/Xix0ziywrSJM3tr1gStTT0bWXWFaQjHxrmobU1FTout6tufzYAxdZVpCM+EII+P3+hOfszubJ4/EgEAhAURTsKJ7TmqaWFaRLK7Z02lP4c6siQshZXYVTp05hyJAhqK+vR79+/dAdem0P0pAhQ+DxeNqtFp08ebLdKpHBiBEj4h7f2NiIH/zgB3jooYfw5S9/Geeffz6WLl2K+fPn48EHH4zZl5SUFPTr18/yA7Qk1CgujLdihK5AaG0/RrvX6zWP9fl85lsIXq8XHr3Z/FFal30VRUF5ebmpoSiKRc/4idUOtDwBBLSWH5/PZ56sqqqajw1/ZebxeCwxjL7Hao/nKSACAIAlo5eY38WWaN/jeTL6Ht5WVlYGXdehBAJtP60XZEeejPHRWueGaJ5CoRA2b95s6ZPhNZG+B1rnACMf0TyF5zsyT8aP0Z5Injx6M1TRIuzRg5Z8G5NJrPx1NU9G/FAo1K7vQm99chMt10WinlQRMq8Nn88HIdpWn2Kde/E8Gfk2ngqiedI0zXLtxTr3Es2T0X9VtJ2T4flO5HrqKE8tc4eAAphjZWhompbw9dSZc09RFLz66qsIhUJQQiHL9ZfQHKGhw3xomoaNGze2639n5ohwT+FzrUdvewsyPN/dmfdi5Sk834poyZMxVl3x5NEDrdeHZsm3sVrV7bncnDsClvibNm2yvKju7BzRUZ5CoRA2bdrU8rhOzuXRPIXn2/BkJ71WIPn9fkycOLHd0mp5eTkKCwujPqagoKDd8Zs3bzaPN95HjfwcBI/HY6lACSGEEELi0asfFLlixQqUlJRg0qRJKCgowBNPPIGamhosXrwYALBgwQJkZGRgzZo1AIDbbrsNU6dOxdq1azF37ly8/PLLeO211/DGG28AAPr164dp06bh9ttvR1paGrKzs7Ft2zb8/ve/x0MPPdRrPgkhhBCSXPRqgTR//nycOnUK99xzD44fP47zzjsPZWVlyM7OBgDU1NRYVoMKCwvxzDPP4M4778Tq1atx5plnYt26dbj44ovNY5555hmsXLkSX//61/Hxxx8jOzsb9913n1l0EUIIIYR0RK9/1ciSJUuwZMmSqH/bunVru7Zrr70W1157bcx4I0aMwG9/+1tb+ma8t+0EXq8XxcXFSa0REAGUHik19yM5gQzjJIuGDB6o4Z741HCXhgweDA276PWvGvlXxrjdOFk1FChI96abn4PkFMk+TjJpyOCBGu6JTw13acjgwU5YIMXByY9ED4VCqKioSGoNn+LDwpELzbvYnECGcZJFQwYP1HBPfGq4S0MGD4aGXbBAIoQQQgiJgAUSIYQQQkgELJB6ESc3qvWURrPe3PFB3USGcZJFQwYP1HBPfGq4S0MGD3aSPD3tBYxPBHUq9pw5cxyL3xMaARHAw0djfwmwHcgwTrJoyOCBGu6JTw13acjgwdCwC64gxcHJT9/WdR0nT55Mag0FCnJScxy9i02GcZJFQwYP1HBPfGq4S0MGD4aGXbBAikO0L021M3ZlZWVSa/gUH+YNm+foXWwyjJMsGjJ4oIZ74lPDXRoyeDA07IIFEiGEEEJIBCyQCCGEEEIiYIEUB0Vxbm+NoihIT09Pag0BgbpAHQSEI/EBOcZJFg0ZPFDDPfGp4S4NGTwYGnbBu9ji4PT3xcyYMcOx+D2hERRB/ObEbxyLD8gxTrJoyOCBGu6JTw13acjgwdCwC64gxcHpnfaHDx9Oag0VKs4/43yoDp5GMoyTLBoyeKCGe+JTw10aMngwNOyCBVIcnN5pX1VVldQaXsWLWYNnwas4t9ImwzjJoiGDB2q4Jz413KUhgwdDwy5YIBFCCCGERMACiRBCCCEkAhZIcXB6p/3QoUOTWkNA4GDjQcfvYkv2cZJFQwYP1HBPfGq4S0MGD4aGXfAutjg4fRdbYWGhY/F7QiMoglhfu96x+IAc4ySLhgweqOGe+NRwl4YMHgwNu+AKUhyc3ki2b9++pNbwwIPJ/SfDA48j8QE5xkkWDRk8UMM98anhLg0ZPBgadsECKQ5O34q4f//+pNbwKK0FkuJcgSTDOMmiIYMHargnPjXcpSGDB0PDLlggEUIIIYREwAKJEEIIISQCFkhxUFXnhkdVVWRlZSW1hg4d1aerocO55VIZxkkWDRk8UMM98anhLg0ZPBgadsG72OLg8Ti3t8bj8SA/P9+x+D2hERIhbPp4k2PxATnGSRYNGTxQwz3xqeEuDRk8GBp2wRWkODi9037Pnj1JreFVvJg1yPmvGkn2cZJFQwYP1HBPfGq4S0MGD4aGXbBAioPTO+1ramqSWkOFivP7Ov9ltck+TrJoyOCBGu6JTw13acjgwdCwCxZIhBBCCCERsEAihBBCCImABVIcnN5pP27cuKTW0ISGv9b/FZpw7v1kGcZJFg0ZPFDDPfGp4S4NGTwYGnbBu9ji4PRdbLm5uY7F7wkNDS0FkpPIME6yaMjggRruiU8Nd2nI4MHQsAuuIMUhFAo5GnvHjh1JreFTfLhu6HXwKT5H4gNyjJMsGjJ4oIZ74lPDXRoyeDA07IIFUhyEEI7Grq2tTWoNBQrGpI2BAsWR+IAc4ySLhgweqOGe+NRwl4YMHgwNu2CBRAghhBASAQskQgghhJAIWCDFwelN2nl5eUmtERIhbDq1CSHh3PvJMoyTLBoyeKCGe+JTw10aMngwNOyCd7HFwelbEbOzsx2L3xMaOnRUf17tWHxAjnGSRUMGD9RwT3xquEtDBg+Ghm2xbIskIU7vtN+yZUtSa/gUH24ccaPjd7El+zjJoiGDB2q4Jz413KUhgwdDwy5YIMXB6Z32DQ0NSa2hQMEQ/xDH72JL9nGSRUMGD9RwT3xquEtDBg+Ghl2wQCKEEEIIiYAFEiGEEEJIBCyQ4uD0TvuCgoKk1giKIJ49+SyCIuhIfECOcZJFQwYP1HBPfGq4S0MGD4aGXfAutjg4fRfbsGHDHIvfExoCAoeaDjkWH5BjnGTRkMEDNdwTnxru0pDBg6FhWyzbIklIMOjcykgwGMSGDRuSWsOv+HHb6NvgV/yOxAfkGCdZNGTwQA33xKeGuzRk8GBo2AULpF7EyVsde0ojRU1xND4gxzjJoiGDB2q4Jz413KUhgwc7YYFECCGEEBJBlwqk3/3ud9iwYYP5+3/9139hwIABKCwsxOHDh23rHCGEEEJIb9ClAun+++9HWloaAKCyshK//OUv8ZOf/ARDhgzB8uXLbe1gb+L1OreH3ev1Yvr06UmtERRB/M/x/3H0LjYZxkkWDRk8UMM98anhLg0ZPBgatsXqyoOOHDmCs846CwDw0ksv4dprr8VNN92EyZMn47LLLrOtc7JjFJnJqiEg0BBqgIBzn4oKJP84yaQhgwdquCc+NdylIYMHO+nSClLfvn1x6tQpAMDmzZtxxRVXAABSU1PR2NhoX+96Gae/L6asrCypNfyKH8sylzl6F5sM4ySLhgweqOGe+NRwl4YMHgwNu+jSCtLMmTOxaNEi5Ofn44MPPsCcOXMAAO+++y5ycnJs6xwhhBBCSG/QpRWkRx55BAUFBaitrcXzzz+PwYMHAwB27dqF66+/3tYOEkIIIYT0NF0qkAYMGIBf/vKXePnllzFr1iyz/e6778aqVas6FevRRx/FmDFjkJqaiokTJ+L111+Pe/zzzz+Pc889FykpKTj33HPx4osvtjvm/fffx9VXX43+/fsjPT0dl1xyCWpqajrVL0IIIYT869KlAmnTpk144403zN8feeQR5OXl4YYbbsAnn3yScJx169Zh2bJlWLVqFfbs2YMpU6Zg9uzZMYuZyspKzJ8/HyUlJdi7dy9KSkowb9487Ny50zzmH//4By699FLk5uZi69at2Lt3L1avXo3U1NRO+3R6p31xcXFSawREAKVHShEQAUfiA3KMkywaMnighnviU8NdGjJ4MDTsoksF0u23347PPvsMAPDOO+/ge9/7HoqLi/HPf/4TK1asSDjOQw89hIULF2LRokUYP348SktLkZmZicceeyzq8aWlpZg5cyZWrlyJ3NxcrFy5EpdffjlKS0vNY1atWoXi4mL85Cc/QX5+PsaOHYs5c+Y4/v0vXaEnNrQ7qaFAQbo3HQoUxzSA5B8nmTRk8EAN98Snhrs0ZPBgJ10qkA4ePIhzzz0XQMtbXldddRXuv/9+PProo9i4cWNCMQKBAHbt2oWioiJLe1FREXbs2BH1MZWVle2Ov/LKK83jdV3Hhg0bcM455+DKK6/EsGHDcPHFF+Oll16K25fm5mZ89tlnlh+gJZHBYBDBYBCapgEAFFVA8bT9GO2hUMg8NhgMQtd1s11TU8wf0TrkjY2NqKioMDWEEBBCWGLEawcABYDf0/ITDAbN3fu6riMYDJoaTU1NAABN0ywxjL7Hao/nya/4cYZ6BhaOXGjexZZo3+N5Mvpu/DQ1NaGiogKBQADC72/7aX2V0JEnY3w8rTVcNE+hUMiSi3CvifTd3/rl0UY+onkKz3dknowfoz2RPGlqCnSlRVhTfZZ8BwKBuPnrap6M+KFQqF3fFbX1ox6UlusiUU+64jWvjWAwiObmZvOcjXXuxfNk5Nso2aN5CgaD7fIdLx8d5cnov660nZPh+U7keuooTy1zhwIBmGNlaIT76uh66sy5F55v4fVarr+E5ggPOsxHtFx0do4I9xQ+12qqDwDMc8rQ6M68FytP4fkWSkuejLHqiidN9bdeH552+Q6FQt2fy825wx81flfniI7yFD7XdnYuj+YpPN+GJ+P5zg66tBbl9/vxxRdfAABee+01LFiwAAAwaNAgs7joiLq6OmiahuHDh1vahw8fjhMnTkR9zIkTJ+Ief/LkSZw+fRoPPPAA7r33XqxduxabNm3CV7/6VVRUVGDatGlR465ZswZ33313u/aKigr06dMHAJCVlYX8/HwMOFOg78i2Yw4cOIDc3Fy89dZbqK2tNdvz8vKQnZ2N7du3o2HCarP9zH8+acYGgPLycgDA9OnTkZaWhrKyMksfiouLzRPXwFhCHDvAi6+fdwYAoKysDOnp6ZgxYwaOHDmCqqoq8/jdu3dj8uTJOHDgAPbv32+2G56qq6stb2uOGzeuQ0/LMpe1xUnNAtDykQ/ht1h21tOcOXNQV1eHyspKs71v374AgGPHjqH+1qVtxx86BAAderqjoD8AYFtNE7bVNEf1NGrUKABtuQCAgoICDBs2LCFPdxT0xwOV9eifouI/Lkw3/xbNU3l5ecw8DR06FIWFhYnlacJqjDixBSM/2oKDOTegOqw/x44dw9ixY1vOvYaGLnmKlSeDyDwNzwdO7FJwxnBg0DnCjBXP02AARzOuwqnBkwAA1WVl5uer7d69G3V1dZY8mddTHE9Gvh/b3YD6Zj2qp9OnT5u5iJUnAInnqfX6HnzqbWQdfcmSp/Ly8oSupw7zNGE1cvf9HP5gPaonrLbkOxQKmS8kDLrtKYLGGdMRmDDB/D2Rec/IBQA89X+fA2g/R0yZMsWSC4POzBHhnqrD5tr0hgMYg5ZtF+Ea3Zn3OspTeXk50gcNgtrQYM5XxjnYGU+pZ9+M8ft/gY8H5uFI5jWWfBvj3525PHNgHoZ8vAsfnH0zmlKHt4vvxFyenp5uzXfr+HgPHULf51/omqewfGceeRE5gGXLTXdRhBCd/pS/q6++GoFAAJMnT8aPf/xjHDx4EBkZGdi8eTOWLl2KDz74oMMYH374ITIyMrBjxw4UFBSY7ffddx/+8Ic/YN++fe0e4/f78bvf/c5yp9xTTz2FhQsXoqmpyYx5/fXX409/+pOlv2eccQaefvrpqH1pbm5Gc3Oz+ftnn32GzMxMHD9+3LxDT1VVeDwePHL7Kwh/R2nxmjnweDwtr7LChtLj8UBVVYRCIRx54BKzXdWDyP7hXnzxxRcoLy/HzJkz4fP5zKIn8jMcYrVf8oOnoQDwta5ebP/x9VAUBV6vt+WVVGvVXV5ejqKiIqSlpUHTNPNVT7inWO3xPM18fCb8ih9LRi/Bw0cexpalW8xXDh31PVa7z+cz+24QCoWwefNmzJo1C5Vzrmo7WNdx2Zb/7dDT1NUtOdd0QBPAzjU3tPOkaRrKysrMXIR7TcTT1NVPI6DBzMf2H7edn4anpqYmM99+v9+SJwMjf4nk6ejaAigiBFVo0FQfRn9/p5nvWbNmISUlJWb+uponI35xcbHZH4MnfrARQldaVpBU4Kb7Znfo6dh9edAVL0TrK7/R36+Epml49dVXUVRUZNlLEH7uxfNk5DuoAQLAm/db76r1er0IBoPYuHGjJd/Rzr3I6ylWno6ubZm/FKFBFSFkrKpCc3Ozme+UlJQOr6eO8nR0bQFUPQBAQFdTMPr7lWY+Zs+eDZ/Pl9D1lKinyHz/9cpZgNr2hsOUVzd16GnKnW1zcFAD3v7pgnbnnhCiXS6MPAGJzRHhng7f/29hR+sYc+duNDU14dVXXzU1ujPvxcqTMVYzZ87EW7OLASEAf8sq0OSNbS+YEvV0dO0l8OgB6IoHQvFa8l1cXAxVVbs1lx974KLWucMPQGkXPxI75nJFUSCEMOfat66e25omHUoohEsrtnTaU/hzqyJCyFldhY8++ggjRoxAfX09+vXr185LZ+jSCtIvf/lLLFmyBM899xwee+wxZGRkAAA2btxouastHkOGDIHH42m3WnTy5Ml2q0QGI0aMiHv8kCFD4PV6zbf/DMaPH2/ZVB5JSkoKUlLafyu9z+ezXLQAWp4EwvB4Wib3WBvDvF4vPHpzu3ajKIrUiNSL1y4ABLT2f1dVFWrrZOb1es2+eTwes7+RHqK1x/NkbMxu1pvNT9LuTN9jtYf3PVxPVVUogfabwTvyFNCs7dE8aZoWNReJ9t3QMPIR+RhVVS35NvoQzWsingBYzimPHjQ1jbGK5TVRT7HaY/XdvC6EAhFlDGJ5UkUIECGLjnHORutPR54i8x0thvGEGpnvWPnoKE+R17fH47HkO5E5Ip4nIDLfzZaxUhQFiqIkfD0l4imyb0rEE2AiniJzEekJaCnCYl170Y6P13dVVaPOtR6PJ6pGV+a9aITnwufzQTEKkdb5KpE5JdKTR295rCo0QGgWjXh9T9STKjSLTmR8p+by8HxHzudd8RQt373+VSNZWVn4y1/+0q79Zz/7WcIx/H4/Jk6ciPLyclxzzTVme3l5OebOnRv1MQUFBSgvL7d839vmzZtRWFhoxrzooovaLRF/8MEHyM7OTrhvBrFOBjvw+XzmB2wmq0ZABPDw0Ycdiw/IMU6yaMjggRruiU8Nd2nI4MHQsIsul1qapuGll17C+++/D0VRMH78eMydOzdqpReLFStWoKSkBJMmTUJBQQGeeOIJ1NTUYPHixQCABQsWICMjA2vWrAEA3HbbbZg6dSrWrl2LuXPn4uWXX8Zrr71mWR26/fbbMX/+fEydOhXTp0/Hpk2b8Oc//xlbt27ttMfwZT270XUddXV1GDJkSNRXQ8mgoUBBdmo2Djcdtj22QbiHntBI1lz0hIYMHqjRtfhOIcM4RWo4hWzXt1PY+bzdpVH4+9//jvHjx2PBggV44YUX8Nxzz6GkpARf+tKXzA1xiTB//nyUlpbinnvuQV5eHrZv346ysjJztaempgbHjx83jy8sLMQzzzyD3/72tzj//PPx5JNPYt26dbj44ovNY6655ho8/vjj+MlPfoIJEybg17/+NZ5//nlceumlnfYZ/v6p3WiahsrKyqTW8Ck+zBs2Dz7FuZU2GcZJFg0ZPFDDPfGp4S4NGTwYGnbRpRWk7373uzjzzDPx5ptvYtCgQQCAU6dO4Rvf+Aa++93vYsOGDQnHWrJkCZYsWRL1b9FWfa699lpce+21cWPeeOONuPHGGxPuAyGEEEJIOF0qkLZt22YpjgBg8ODBeOCBBzB58mTbOkcIIYQQ0ht06S22lJQUy+dBGJw+fRr+1lsbZUBRnPuEaEVRkJ6entQaAgJ1gTrzLjYnkGGcZNGQwQM13BOfGu7SkMGDoWEXXSqQrrrqKtx0003YuXOn+Umab775JhYvXoyrr77ats71Nk5/X8yMGTOSWiMogvjNid8gKIIdH9xFZBgnWTRk8EAN98Snhrs0ZPBgaNhFlwqkn//85zjzzDNRUFCA1NRUpKamorCwEGeddZble9GSHafvYjt8+HBSa6hQcf4Z50Pt2mmUEDKMkywaMnighnviU8NdGjJ4MDTsokvPbAMGDMDLL7+MDz74AM899xzWr1+PDz74AC+++CIGDBhgW+d6G6d32ldVVSW1hlfxYtbgWfAqzr0akGGcZNGQwQM13BOfGu7SkMGDoWEXCT+zrVixIu7fw+84e+ihh7rcIUIIIYSQ3ibhAmnPnj0JHefk5itCCCGEkJ4g4QIp2jd6y47TO+2HDp1TdXQAACAASURBVB2a1BoCAgcbDzp+F1uyj5MsGjJ4oIZ74lPDXRoyeDA07MK5zSMS4PROe+M75JJVIyiCWF+73rH4gBzjJIuGDB6o4Z741HCXhgweDA27cO72IwlweiPZvn37klrDAw8m958MDxL//r3OIsM4yaIhgwdquCc+NdylIYMHQ8MuWCDFwelbEffv35/UGh6ltUBSnCuQZBgnWTRk8EAN98Snhrs0ZPBgaNgFCyRCCCGEkAhYIBFCCCGERMACKQ6q6tzwqKqKrKyspNbQoaP6dDV0OLdcKsM4yaIhgwdquCc+NdylIYMHQ8MueBdbHDwe5/bWeDwe5OfnOxa/JzRCIoRNH29yLD4gxzjJoiGDB2q4Jz413KUhgwdDwy64ghQHp3fa79mzJ6k1vIoXswY5/1UjyT5OsmjI4IEa7olPDXdpyODB0LALFkhxcHqnfU1NTVJrqFBxfl/nv6w22cdJFg0ZPFDDPfGp4S4NGTwYGnbBAokQQgghJAIWSIQQQgghEbBAioPTO+3HjRuX1Bqa0PDX+r9CE869nyzDOMmiIYMHargnPjXcpSGDB0PDLngXWxycvostNzfXsfg9oaGhpUByEhnGSRYNGTxQwz3xqeEuDRk8GBp2wRWkOIRCIUdj79ixI6k1fIoP1w29Dj7F50h8QI5xkkVDBg/UcE98arhLQwYPhoZdsECKgxDC0di1tbVJraFAwZi0MVCgOBIfkGOcZNGQwQM13BOfGu7SkMGDoWEXLJAIIYQQQiJggUQIIYQQEgELpDg4vUk7Ly8vqTVCIoRNpzYhJJx7P1mGcZJFQwYP1HBPfGq4S0MGD4aGXfAutjg4fStidna2Y/F7QkOHjurPqx2LD8gxTrJoyOCBGu6JTw13acjgwdCwLZZtkSTE6Z32W7ZsSWoNn+LDjSNudPwutmQfJ1k0ZPBADffEp4a7NGTwYGjYBQukODi9076hoSGpNRQoGOIf4vhdbMk+TrJoyOCBGu6JTw13acjgwdCwCxZIhBBCCCERsEAihBBCCImABVIcnN5pX1BQkNQaQRHEsyefRVAEHYkPyDFOsmjI4IEa7olPDXdpyODB0LAL3sUWB6fvYhs2bJhj8XtCQ0DgUNMhx+IDcoyTLBoyeKCGe+JTw10aMngwNGyLZVskCQkGnVsZCQaD2LBhQ1Jr+BU/bht9G/yK35H4gBzjJIuGDB6o4Z741HCXhgweDA27YIHUizh5q2NPaaSoKY7GB+QYJ1k0ZPBADffEp4a7NGTwYCcskAghhBBCImCBRAghhBASAQukOHi9zu1h93q9mD59elJrBEUQ/3P8fxy9i02GcZJFQwYP1HBPfGq4S0MGD4aGXbBA6kXS0tKSWkNAoCHUAAHnPhUVSP5xkklDBg/UcE98arhLQwYPdsICKQ5Of19MWVlZUmv4FT+WZS5z9C42GcZJFg0ZPFDDPfGp4S4NGTwYGnbBAokQQgghJAIWSIQQQgghEbBAIoQQQgiJgAVSHJzeaV9cXJzUGgERQOmRUgREwJH4gBzjJIuGDB6o4Z741HCXhgweDA27YIHUizQ2Nia1hgIF6d50KFAc0wCSf5xk0pDBAzXcE58a7tKQwYOdsECKg9M77SsqKpJaw6f4sHDkQvgUnyPxATnGSRYNGTxQwz3xqeEuDRk8GBp2wQKJEEIIISQCFkiEEEIIIRGwQOpFnNyo1lMazXqzo/EBOcZJFg0ZPFDDPfGp4S4NGTzYSfL0tBfw+ZzbW+Pz+TBnzhzH4veERkAE8PDRhx2LD8gxTrJoyOCBGu6JTw13acjgwdCwC1esID366KMYM2YMUlNTMXHiRLz++utxj3/++edx7rnnIiUlBeeeey5efPHFmMfefPPNUBQFpaWlne6XruudfkxnYp88eTKpNRQoyEnNcfQuNhnGSRYNGTxQwz3xqeEuDRk8GBp20esF0rp167Bs2TKsWrUKe/bswZQpUzB79mzU1NREPb6yshLz589HSUkJ9u7di5KSEsybNw87d+5sd+xLL72EnTt3YtSoUV3qm6ZpXXpcorErKyuTWsOn+DBv2DxH72KTYZxk0ZDBAzXcE58a7tKQwYOhYRe9XiA99NBDWLhwIRYtWoTx48ejtLQUmZmZeOyxx6IeX1paipkzZ2LlypXIzc3FypUrcfnll7dbITp27BiWLl2Kp556ytG3ygghhBAiH71aIAUCAezatQtFRUWW9qKiIuzYsSPqYyorK9sdf+WVV1qO13UdJSUluP322/GlL33J/o4TQgghRGp6dZN2XV0dNE3D8OHDLe3Dhw/HiRMnoj7mxIkTHR6/du1aeL1efPe7302oH83NzWhubrsb67PPPgPQ8oFTwWAQAKCqKjweDxRVIHzLjaZp8Hg8CIVCEEKY7R6PB6qqIhQKQVNTzHZVD5qx+/bta36olbGzP/JDrmK1Ay3d8Hla/h8MBqEoCrxeL3Rdh6ZppoamafD5fNA0zfL+rOEpVns8T37FD5/iw6nAKfPvxlh11PdY7T6fz+y7gaZpSE9PhxACwu9vO7i1vx158reOj6YDmkBUT4qiWHIR7jURT34PENDa8hH+GMNTeL4j82RgtCeSJ01NgSJCUIUGTfUhGAyaGoa/WPnrap6M+IqitO+7KiB0BVAEFLVtDOJ5AgBd8UIoLUkKBoPQdR3p6entlsnDz714nox8BzVAIPY5GZnvaOdeonkyrm9FaFBFyHLthUKhhK6njvKkqSlQ9QAAAV1NseQbAIQQCV1PnTn3wvMtvF5AbXs9nci8Z+TCyAei5CNaLoDOzREWT2FzLdDiQ9d1i0Z35r1YeQrPt1AUQAigdb4yPHdq3lP98OgB6IoHQvFaNBRF6fZcriue1rnDD0BpF9+JuVxRFMtca87nug4lFOqSp/B8K6KlD3a+xeaKu9gUxbrJVwjRri3R43ft2oWHH34Yu3fvjhsjnDVr1uDuu+9u175lyxb06dMHAJCVlYX8/HwMOFOg78i2Yw4cOIDc3Fy89dZbqK2tNdvz8vKQnZ2N7du3o2HCarP9zH8+acYOhULYvHkzAGD69OlIS0tDWVmZpQ/FxcVobGxERUWF2WaclGMHePH1884AAJSVlSE9PR0zZszAkSNHUFVVZR6/e/duFBYW4sCBA9i/f7/Zbniqrq627PkaN25ch56WZS4z20eltOzx2rx5s+VC6aynOXPmoK6uDpWVlWa74enw4cOov3Vp2/GHDpnjH8/THQX9AQDbapqwraY5pidFUcxcAEBBQQGGDRuWkKc7Cvrjgcp69E9R8R8Xppt/i+Zp8+bNMfM0dOjQxPM0YTVGnNiCkR9twcGcG1Ad1p/jx4+3nXsNDV3yFC9PXq8XJ0+etORpeD5wYpeCM4YDg84RZqx4ngYDOJpxFU4NngQAqC4rw7hx4zBjxgzs2LEj9vUUx5OR78d2N6C+WY/p6fTp02a+Ozr3OsxT6/U9+NTbyDr6kiVPmzdvTuh66jBPE1Yjd9/P4Q/Wo3rCaku+AaChoaFT11Nnzj2v14vGGdMRmDDBbE9k3jNyAQBP/d/n5nhEnntTp061ZY44cuQIqsPm2vSGAxgD4ODBg5Z8d2fe6yhPmzdvRvqgQVAbGsz5yvDWGU+pZ9+M8ft/gY8H5uFI5jVmvocOHQqv14t9+/Z1ay7PHJiHIR/vwgdn34ym1OFm/IKCAni9XmzYsMGxufyss85qyUXr+HgPHULf51/o2vNTWL4zj7yIHCDqfuSuoojwsrKHCQQC6NOnD9avX49rrrnGbL/ttttQVVWFbdu2tXtMVlYWli9fjuXLl5ttP/vZz1BaWorDhw+jtLQUK1asMF+hAi0VpaqqyMzMxKHWJ9Zwoq0gZWZm4uTJkxgwYACAtqr1kdtfsawgLV4zp8NXHUceuMRsV/Ugsn+4F83NzTh27BgyMjKgqmqnK/RLfvC0ZQVp+4+vb/fqUNd1HDt2DKNHj4bf77d1BWnm4zOhQsX4M8aj+nQ1tt+63ZFXHUIIHD9+HBkZGXhj1uy2g3Udl2353w49TV39NIC2FaSda25o5wkADh06ZOYi3GsinqauftqygrT9x9e38xQMBs18ezyebq8gHV1bYFlBGv39nWa+jSc1u1eQjPg5OTktYxrW9yd+sNGygnTTfbM79HTsvjzLCtLo77dMph9++CFGjhxpeYGT6GqLkW9jBenN+9tyYXjSdR2HDx+25Ls7qy1H1xa0tLeuIGWsqkIoFDLz7fV6u72CdHRtgWUFafT3K818ZGdnm3HC6e4Kkq7r+PDDD1uKgytmWlaQpry6qUNPU+78k9ke1IC3f7qg3bmnqiqOHDmCkSNHWubsrq4gHb7/38KO1jHmzt0IBoM4cuSImW8nVpCMXGRkZKCy6ErLCtLkjW0vmBL1dHTtJZYVpPB85+TkQAjRrbn82AMXWVaQjPjHjx9HVlZWu1UYu1aQVFXF4cOHMWrUKFTOuao1TS0rSJdWbOm0p/DnVkWEkLO6CnV1dRg6dCjq6+vRr18/dIdeXUHy+/2YOHEiysvLLQVSeXk55s6dG/UxBQUFKC8vtxRImzdvRmFhIQCgpKQEV1xxheUxV155JUpKSvDtb387asyUlBSkpKS0a1dVtd0Gb6FbV6WMJ9hYH37l9XrhifJhiqqq4p133kFmZqZFI9aG8mjtAi1v7UT+3ZgEgsGgqWH01ehvpIdo7fE8BUQAfsWPmYNm4t3P3+1032O1G303CAaDqKqqwqhRo6AEAgn33WgPRKy2RvMUPk6RfUqk74aGkY/IxxieDA2jD5FeE/UEwHJOefQgfD5fu3zHyl9X8xQ5TuF9N68LoUBEGYNYnlQRAlqXxg0PRr6j9acjT5H5jhZD1/Wo+Y6Vj47yFHl9ezwei0Yic0Q8T0Bkvpvb5dvj8SR0PSXqCWjJd3V1NUaPHg0l4gkwEU+RuYj0ZGjs3bsXGRkZUfvfWU/R5loAUfPdlXkvGpG5UIxCpHW+SmROifTk0VseqwoNEFo7DZ/P1625XBWaRceIHy8XsfqeqCcg4pyKmM+78vwULd+JvnOUCL3+FtuKFStQUlKCSZMmoaCgAE888QRqamqwePFiAMCCBQuQkZGBNWvWAGhZXZo6dSrWrl2LuXPn4uWXX8Zrr72GN954AwAwePBgDB482KLh8/kwYsQIjBs3rmfNEUIIISQp6fUCaf78+Th16hTuueceHD9+HOeddx7KysqQnZ0NAKipqbFUoYWFhXjmmWdw5513YvXq1TjzzDOxbt06XHzxxb1lgRBCCCGS0esFEgAsWbIES5Ysifq3rVu3tmu79tprce211yYcP9q+o0Swc6kuWuyhQ4cmtYaAwMHGgxBwbhubDOMki4YMHqjhnvjUcJeGDB4MDbtwRYHkVpz8Uj2v12vum0pWjaAIYn3tesfiA3KMkywaMnighnviU6M9uq4jEAggdMZIS3tTUxMA4MILL0QoFIr6sS+dwen48TA0lGHDovahI2LtvzKw83mbBVIcnP449AMHDuDss8+Om2w3a3jg+f/tnXt8DPf+/1+z10QkcYlIUhIRJCTulLhTQlBUi9IGh/ZXrbQupSgOVYJWneiFlqPqUuXroHLcIhqirUsJKVWNIMQlkQrJ5r63z++P2LW7uW4ys9md834+Hvsgs7Ov17w/7/3MvPcz85lBd/fuOJtzlndtA6Yx2MLDUXNhCw8xxEAe1dMXCjG0k6VHdVGr1UhNTS25b1qPeWbvpaamGu91JZPJajxKIrR+eZh6uE43P2uUmppaZZ169erBy8urzPdEdx8ke0XoB+olJycjICBAsE4rtIeUk6Kne0+cV53nXduAaQy28HDUXNjCQwwxkEf19IVCDO1k6VEdDLczkUqlaNq0KbQu5iM4Ck9/6PV65OXloW7dumXO4rMGdaaw+uVh6lFo8Z6Lv3+ln2eMoaCgAJmZmRV68AUVSARBEARRi2i1WhQUFMDHxwd16tRBscy8QFE6ORlPvzk5OdW4gOEE1i8PUw+dhYeTk1OVNJydnQEAmZmZcJY6QaKr2qm56lDrD6slCIIgiP9lDKeFFKaPUyLKxfCEC72TeyVr1gwqkCpAqCraoO3r6+vQHnrocTnvMvQQ7lSkGNpJLB5iiIE87EefPEpT2SO2FAqFoLPYhNTn0+PZ58u+YShf0Cm2ChDqfLVBu2PHjoLp28JDy7Q4+vioYPqAONpJLB5iiIE87EefPKyD4zjjyIkj6tvKg8/jNo0gVYDQs9guXbrk0B4yToYhDYZAxglXZ4uhncTiIYYYyMN+9MnDOgwXKAv1+FSh9W3lQbPYbITQs9jS0tIQEhIi6MwKIT0kkKBd3XaIfxLPu7YB0xhs4eGoubCFhxhiII/q6QuFGNrJ0oNPevwr8en/Eitcr+aU6P+0cAScnJyqfApsw4YN2LBhg/FmzMHBwfjnP/+J8PBwPH78GEuWLMGxY8dw9+5deHh4YOTIkZgzZ47xeZFCQLPYCIIgCIKoVZo0aYJVq1ahRYsWAICtW7di5MiRuHTpEhhjePDgAdasWYM2bdrgzp07mDZtGtLS0rB///5a3vKqQQUSQRAEQRBW8+KLL5r9vWLFCmzYsAFnz57F1KlTsXfvXuN7AQEB+PjjjzFx4kRB79TNJ1QgVYDQsx4CAwMd2kPHdPg151fomHDn3cXQTmLxEEMM5GE/+uRhfyiVymrPMNPpdNizZw/y8/MRGhpa5joqlQpubm6Qy+UorsmGVgDNYrMRQs9iCwoKEkzfFh46lBRIQiKGdhKLhxhiIA/70ScP+8NwE0ZruHLlCkJDQ1FUVIS6deti//79aNOmTan1srKysHz5crz11luC3kqAZrHZCCGHAbVaLU6fPu3QHnJOjjGNxkDOyQXRB8TRTmLxEEMM5GE/+uRhf+Tl5Vk9wywwMBBJSUk4e/Ys3n77bUyaNAl//vmn2ToqlQrDhg1DmzZt8P777ws6i43PHNAIUgUIPd3x77//dmgPDhz8nf3BQbhfA2JoJ7F4iCEG8rAfffKwP7RaLRhjVo3wKBQK40XaXbp0wfnz57Fu3Tp88803AIDc3FwMGTIEdevWxd69e6FWqwXPBV/QCBJBEARBELzAGENxcckVRiqVCmFhYVAoFIiJiany89bsBRpBIgiCIAjCaj788EOEh4ejadOmyM3Nxa5du3Dy5EkcPXoUubm5CAsLQ0FBAXbs2AGVSoXs7Gzk5ubCxcWltje9SlCBVAFCX6TdoUMHh/bQMi2OZh2Flgl33l0M7SQWDzHEQB72o08elXN6VmcAgNInGIwxqNVqXp5lVvzgqtnfpvrWaD98+BARERFIT0+Hu7s72rVrh6NHj2LQoEE4efIkzp07BwDGU3AGbt26BY8aRVA+fOaACqQKEHpaqJ+fn2D6tvDQQ4/L+ZcF0wfE0U5i8RBDDORhP/rkYR0cx0GpVNqV/ubNm8t9r1+/fhVeD5T7119WeVUVPo/bdA1SBQg96yE+Pt6hPeScHFO8pgg+i83R20ksHmKIgTzsR588rIMxBpVKJehF+ULq28qDzxxQgVQBQn9RcnNzHdqDAwcPhYfgs9gcvZ3E4iGGGMjDfvTJw3oPvV4vaL6F1LelB19QgUQQBEEQBGEBFUgEQRAEQRAWUIFUAULPeggNDXVoDw3T4P8y/w8aphFEHxBHO4nFQwwxkIf96JOHdXAcBxcXF8Ee0yG0vq08aBabjRB6Fpunp6dg+rbwYGC4XXRbMH1AHO0kFg8xxEAe9qNPHtbBcRzkcuEmxAitbysPmsVmIzQa4UZGNBoNDh065NAeCk6BGU1mQMEpBNEHxNFOYvEQQwzkYT/65GEder0e2dnZ0Ov1DqlvKw8+c0AFUi1iiwcbCu2hlAh3Xw4DYmgnsXiIIQbysB998iDsGSqQCIIgCIIgLKACiSAIgiAIwgK6SLsCZDLhmkcmk6F///4O7aFhGmxO3yzoLDYxtJNYPMQQA3nYjz55VM7Df79aapmKVwdznluYZPUMs/v372PevHk4cuQICgsL0apVK2zevBmdO3cute60adOwadMmrF27Fm+Eh/O12WbwmQMqkGoRZ2dnh/ZgYMjV5oJBuLuiAo7fTmLyEEMM5GE/+uRhX1hbHD158gQ9e/ZE//79ceTIEXh6euLmzZuoV69eqXV//PFH/Pbbb/Dx8eFrcwWHTrFVgNDP7jl8+LBDeyg4BWY2nSnoLDYxtJNYPMQQA3nYjz552B/WPidt9erVaNq0KbZs2YLnn38ezZo1wwsvvICAgACz9e7fv4/IyEhs375d0HtFAfQsNoIgCIIgapmYmBh06dIFY8aMgaenJzp27IhNmzaZraPX6xEREYG5c+ciODi4lra0elCBRBAEQRCE1dy6dQsbNmxAy5YtERsbi2nTpuG9997Dtm3bjOusXr0aMpkM7733Xi1uafWga5AIgiAIgrAavV6PLl26ICoqCgDQsWNHXL16FRs2bMDEiRORmJiIdevW4eLFi+A4zqrTd/YAjSBVgNCzHoYOHerQHmqmRvTdaKiZWhB9QBztJBYPMcRAHvajTx72h5ubm1UXant7e6NNmzZmy1q3bo20tDQAwM8//4zMzEz4+vpCJpNBoVDg7t27mDNnDkJeeIHXbTfAZw6oQKpFCgsLHdqDAwdXmSs4CPfgQcDx20lMHmKIgTzsR5887AtrR3h69uyJ5ORks2XXr1+Hn58fACAiIgKXL19GUlISkpKScOnSJfj4+GDOnDnY/+9/87bdQkEFUgUIPevhxIkTDu0h5+SY6j0Vck64hw+KoZ3E4iGGGMjDfvTJw/7Izc21qkiaNWsWzp49i6ioKNy4cQM7d+7Exo0bMX36dABAw4YNERISYnwFBwdDKpXCy8sLLf39BYmBzxw49nggQRAEQYiUxm/sAgAofYKh1+uhUqng5uZW4yfWFz+4ava3qb41dO3aFfv378eCBQuwbNky+Pv7Izo6Gq+99lqNts9eoAKJIAiCIIhqMXz4cAwfPrzK61++fBlubm7Iv35dwK3iBzrFVovY4oI+oT2K9cWC6gPiaCexeIghBvKwH33yIOwZymgFyOXCXVsjl8sxbNgwwfRt4aFmaqy7t04wfUAc7SQWDzHEQB72o08e1iGRSMp8hIej6NvKg8/jNo0gVYBerxdUOzMz06E9OHBo5tRM0FlsYmgnsXiIIQbysB998rAOxhg0Go1g9xISWt9WHnzmgAqkCtDpdIJqnzlzxqE95JwcYz3HCjqLTQztJBYPMcRAHvajTx7WwRhDfn6+oAWSkPq28uAzB1QgEQRBEARBWEAFEkEQBEEQhAVUIFWANbdcr462q6urQ3swMDxSPwKDcMOlYmgnsXiIIQbysB998rDeQyKRCJpvIfVt6cEXNIutAoR+ds+AAQME07eFh4Zp8G3Gt4LpA+JoJ7F4iCEG8rAfffKwDo7j4Obm5rD6tvKgZ7HZCKFnPdy5c8ehPSSQoJ1LO0gE/BqJoZ3E4iGGGMjDfvTJwzoYYyguLhb0Im0h9W3lQbPYbITQsx6SkpIc2kPGyTCk4RDIOOFG2sTQTmLxEEMM5GE/+uRhHYwxFBYWClogCalvKw8+c0Cn2AiCIAjCDhmw9//Z1O9QxCGrP3Pq1Cl8+umnSExMRHp6Ovbv349Ro0YBADQaDRYtWoTDhw/j1q1bcHd3R58+fbBmzRq4m2ikpKbio3nz8Ouvv0KtVqNt27ZYvnw5+vfvz1Nk1cMuRpDWr18Pf39/ODk5oXPnzvj5558rXH/v3r1o06YNlEol2rRpg/379xvf02g0mDdvHtq2bQsXFxf4+Phg4sSJePDggdBhEARBEMT/FPn5+Wjfvj2+/PLLUu8VFBTg4sWLWLx4MS5evIj//Oc/uHnzprGAMjBm2jRotVrEx8cjMTERHTp0wPDhw5GRkWGrMMqk1guk3bt3Y+bMmVi4cCEuXbqE3r17Izw8HGlpaWWuf+bMGYwbNw4RERH4/fffERERgbFjx+LcuXMASidk3759uH79OkaMGGH1tgl9pX2jRo0c2oOBIbUwVfBZbI7eTmLxEEMM5GE/+uRhf8hkMqtjCA8Px/LlyzF69OhS77m7uyMuLg5jx45FYGAgQkNDsWbNGiQmJuLu00GLrCdPcCstDfPnz0e7du3QsmVLrFq1CgUFBbh69arVMYhqFtvatWsxdepUvPHGGwCA6OhoxMbGYsOGDVi5cmWp9aOjozFo0CAsWLAAALBgwQIkJCQgOjoaP/zwgzEhpnzxxRd4/vnnkZaWBl9f3ypvm9Cz2Hr06CGYvi08NEyDPX/vEUwfEEc7icVDDDGQh/3ok4f9UbduXUH1OY6DRqMBx3FwfzqbrUG9eggMCMC2bdvQqVMnKJVKfPPNN2jcuDE6d+5stQefx+1aLZDUajUSExMxf/58s+VhYWE4ffp0mZ85c+YMZs2aZbZs8ODBiI6OLtcnJycHHMeV+5C84uJiFBc/eyq9SqUCABQVFcHZ2RlAyUP2pFIpOAmD6aPHdDodpFIptFqt2YVnUqkUEokEWq0WOonSuFyi1xi1b968iYCAAEilUmNStVqt2baVtxwo2Qy5tOT/hi+dTCaDXq+HTqeDTqfDzZs30aJFCyiVSuh0OrMr/A0xlbe8opgUnAJSSNHVrStO55w2bkNVtr285XK53LjtBvR6PVJTUxEQEACmUMDkDQCoNCbF0/bR6QEdQ5kxMcaQnJyM5s2bQyqVmsValZgUUkCte5YP088YYlKr1cZ8y2QyszwZMOSvKnnSSZTgmBYSpoNOIodGozHmu1WrVpDL5eXmr7p5MugHBQWB4zjzbZcwMD0HcAyc5FkbVBQTAOg5GRhX0uaGZzTdunUL/v7+xnVMt72ymAz51ugAhrK/kzqdDsnJyca+Z5qnsvJRWZ4M/ZtjOkiYFjqdDhqNxphvuVxeaX+qLE86iRIS+YfNrAAAIABJREFUvRoAg16iNMt3YGCgUd+UmsRkyPetW7cQGBgIJpMBJvmoyn7PkAtDPlBGPjiOQ0pKCvz9/Y25MOQJqNo+wiwmk30tUBKHWq1GSkqKMd812e+VlydDLgICAsA4DmAMeLq/MsRcWUyMMej1+qfbxAECjsyXR2FhIZycnEpdRG0YlanKckObMsbMlnMch6KiInzwwQcYP3483Fzdni5nOLB5M157/324urpCIpGgcePGOHLkCOrVq1emjkFfL5Gb9L+SdjU9lteUWi2QHj16BJ1Oh8aNG5stb9y4cbnnHjMyMqxav6ioCPPnz8eECRPKvf/CypUr8dFHH5Va/tNPP6FOnToAAF9fX3Ts2BH1Ahjqej9bJyUlBUFBQfjtt9/w999/G5d36NABfn5+OHXqFHLbLjYuD7j1nVFbq9Xixo0bAID+/fvD2dkZhw8fNtuGoUOHorCwECdOnDAuM3S05vVkeC3EBQBw+PBhuLq6YsCAAbh79y6SkpKM62dnZ6Nnz55ISUlBcnKycbkhpsuXL5ud0gwMDKw0pplNZxqXp6vTAQDHjh0z6/zWxjRs2DA8evQIZ86cMS6vW7cu8vLyoFAokPNu5LP1b982tn9FMc0PLbkUMCGtCAlpxWXG5OPjg5SUFKSkpBiXh4aGwtPTs0oxzQ91x6ozOXBXSvB2J1fje2XFdOPGjXLz1KhRI/To0aNqeWq7GF4Z8fB+GI/UZhNw2WR7nJ2d0bx585LvXm5utWIqL08A0KpVKzx58sQsT407AhmJHFwaAw1aMaNWRTE1BHDvueHIatgFAHD58GG0aNECN27cQFZWFh49emSWJ2N/qiAmQ743XMxFTrG+zJhyc3Nx48YNY98r77tX5Tw97d8Nsy7A996PZnm6ceNGlfpTpXlquxhBf30OhSYHl9suNsu3v78/tFptlfqTtd89AGjZsiUKB/SHum1b47Kq7PcMuQCA7//IB1B6H9G7d29cv34d169fL5Wnqu4jTGO6bLKvdc1Ngf/TbTXNd032e5Xl6caNG3Bt0ACS3Fzj/srwHawopidPnqCwsBB5eXlQq9XgnBrBqSgTtqa4uBhKpdI4SGDAcBNMy+Vubm5gjJm1S2FhIYCSQjA/P9+4XKfTYcqUKdDpdFi5ciV0T4/HrKgIs995Bw0bNsThw4fh7OyMbdu2Yfjw4bhw4QLc3d2hVquNOkqlEhzHQa1WI7XVO8a2b3p3P5oBZt+NmsIxIefbVcKDBw/w3HPP4fTp0wgNDTUuX7FiBbZv346//vqr1GcUCgW2bt2K8ePHG5d9//33mDp1KoqKiszW1Wg0GDNmDNLS0nDy5MlyC6SyRpCaNm2K9PR0NGzYEMCzXxdfzY0xG0GatnJYpb867q7qblwu0Wvg98/fUVBQgLi4OAwaNAhyudzq0ZbuH/5gNoJ06uPxpX4dajQaxMXFISwsDM7OzryOIA36ehAUnALvNHkH6+6uQ3xkvCAjSFqtFseOHcOQIUNwZtjwZyvr9egX/1OlMfVZ/AOAZyNI51ZOKBWTTqfD4cOHjbkwjbUqMfVZ/IPZCNKpj599Nw0xFRUVGfOtUChqPIJ0b3Wo2QhSk3nnjPkeMmQIlEol7yNIBv2hQ4cat8fAxg+PmI0g/b8V4ZXGdH9FB7MRpCbzSh72GRsbi7CwMLOh8qqOthjybRhBOhv1LBeGmDQaDY4cOWKW75qMttxbXbLvMowgPbcwCcXFxcZ8K5XKGo8g3VsdajaC1GTeGWM+wsPDjSOGptR0BMk0378OHmI2gtQ79milMfVetNO4XKMDLnw6sdR3jzFWKheGPAHWjyDdiXreZG09/BddRFFREWJjY40eQowgGdpq0KBB+C18qNkIUs8jz34wlRdTQUEBbt++jWbNmsHJyQmajL8AsFqZxVbWcbKqI0hSqRR79+7F6NGjzUZ+NBoNXn31Vdy6dQv79++Hn58fCm7cBACcPHMao6ZOxePHj+Hq6mrUDgwMxNSpUzFv3rxSI0jFxcW4desWFLFzIc1/WLKcadFscRIePnwILy8v5OTk1PimlLU6guTh4QGpVFpq9CczM7PUKJEBLy+vKq2v0WgwduxYpKamIj4+vsKGUiqVUCqVpZbL5XKzTgug5CBggmFYuLzznjKZDFJ96SE/0x2zqYelX0XLGUpO7Vi+L5FIzE5PGLZNKpWaDWObxlDW8opiUrNnFb3hIm1rtr285ZbbbrqcM/kVUdm2G5arLW6JUVZMhp1tWfmuyrYbPAz5sPyMRCIxy7dhG8qLtSp5Mv1OSfWaUvkvL9aqxlTZcsttN/YLxoGV0QblxSRhWuDp0LjlAbIs38pissx3WRqGnbplviv67lWUJ8v+LZVKzfJdlX1ERTEBlvkuNnuP4zhwHGd1f7Lmu8dZHNSrEpNlLoDS+TAUTGX1vbLWr2jbJRJJmfta09OoZm1ajf1eWVjuvznDwfzp/qoq+xTD4zeexVZr4xbG7SmL8i6ANl1u+Kzhe2kojlJSUvDTTz8ZR4DASopTw4iToeA01dHr9Uadsjwlek2pnIvmTtoKhQKdO3cudVF1XFxcuRe8hYaGllr/2LFjZusbiqOUlBQcP37cOApkLeV9SfhAIpHA19fXoT300ONy3mXoIdzdY8XQTmLxEEMM5GE/+uRhfygUCqtngeXl5SEpKcl4yjY1NRVJSUlIS0uDVqvFK6+8ggsXLuD777+HXq/H48eP8fDhQ+Nps+c7dEA9NzdMmjQJv//+O65fv465c+ciNTUVw4YNszoGPnNQ67PYZs+ejYiICHTp0gWhoaHYuHEj0tLSMG3aNADAxIkT8dxzzxlntM2YMQN9+vTB6tWrMXLkSBw4cADHjx/HL7/8AgDGhFy8eBEHDx6ETqczjjg1aNAACtMLfSuhrF8XfCGVStGxY0fB9G3hoWVaHH18VDB9QBztJBYPMcRAHvajTx6VE//yRgCA0ieYV93iB+bT52uif+HCBbMbOs6ePRsAMGnSJCxduhQxMTEASq7lMuXQ1q3o/fzzaFi/PvZt2oSoTZswYMAAaDQaBAcH48CBA2jfvr3V28PncbvWy91x48YhOjoay5YtQ4cOHXDq1CkcPnwYfn5+AIC0tDSkp6cb1+/Rowd27dqFLVu2oF27dvjuu++we/dudOvWDQBw7949xMTE4N69e+jQoQO8vb2Nr/JmxpWH0Lemv3TpkkN7yDgZhjQQ/lEjjt5OYvEQQwzkYT/65GEdjDEUFBQI+qiR6uj369fPeL2R6eu7775Ds2bNzJbp9Xrk5+dDr9ej9/PPrhfrFBKC2NhYZGVlQaVS4cyZMwgPD69WHHzmoNYLJAB45513cPv2bRQXFyMxMRF9+vQxvnfy5El89913Zuu/8sor+Ouvv6BWq3Ht2jWzG1RZJsT01a9fP6u2S+iHG6alpTm0hwQStKsr/MNqHb2dxOIhhhjIw370ycM6GGNQq9WCFkhC6tvKgx5WSxAEQRAEISBUIBEEQRAEQVhABVIFCD3rITAw0KE9dEyHX3N+hY4Jd95dDO0kFg8xxEAe9qNPHtbBcdyzKfIOqG8rD1HNYrNnhJ7FFhQUJJi+LTx0KCmQhEQM7SQWDzHEQB72o08e1sFxnPHRV46obysPUc1is2fKev4Zn9qnT592aA85J8eYRmMg58q+wSAfiKGdxOIhhhjIw370ycM6GGPIy8sT9CJtIfVt5cFnDqhAqgChvyh///23Q3tw4ODv7A8Owg2XiqGdxOIhhhjIw370ycN6D8vHnTiSvi09+IIKJIIgCIIgCAuoQCIIgiAIgrCACqQKEPoi7Q4dOji0h5ZpcTTrKLRMuPPuYmgnsXiIIQbysB998rAOwwXOQs5is1Z/5cqV6Nq1K1xdXeHp6YlRo0YhOTnZbJ1+/foZHzgrlUpRv359TJgwoZTWoUOH0K1bNzg7O8PDw8PsBtDWwGcOaBZbBQg9LdTwOBVH9dBDj8v5lwXTB8TRTmLxEEMM5GE/+uRROWdffYd3zYroeyrBqvUTEhIwffp0dO3aFVqtFgsXLkRYWBj+/PNPuLi4GNd78803sWzZMuPfzs7OgMkjxA4cO4b3li5FVFQUBgwYAMYYrly5Uq0Y+Dxu0whSBQg96yE+Pt6hPeScHFO8pgg+i83R20ksHmKIgTzsR5887A+VSmXVRc5Hjx7F5MmTERwcjPbt22PLli1IS0tDYmKi2Xp16tSBl5cXGjdujDp16sDNzc34nlarxbyoKHz66aeYNm0aWrVqhcDAQLzyyivVioFmsdkIoa+0z83NdWgPDhw8FB6Cz2Jz9HYSi4cYYiAP+9EnD/tDr9fXKIacnBwAQIMGDcyWf//99/Dw8EBISAgWLlwIlUplfC/pzz/x4OFDSCQSdOzYEd7e3ggPD8fVq1ertQ00i40gCIIgCLuBMYbZs2ejV69eCAkJMS5/7bXX8MMPP+DkyZNYuHAhYmJi8PLLLxvfv333LgBg6dKlWLRoEQ4ePIj69eujb9++ePz4sc3jMIWuQSIIgiAIokZERkbi8uXL+OWXX8yWv/nmm8b/t2nTBj4+Pujfvz+Srl5Fh+Bg6J+O+CxcuNBYOG3ZsgVNmjTBnj178NZbb9kuCAtoBKkChJ71EBoa6tAeGqbB/2X+HzRMI4g+II52EouHGGIgD/vRJw/7w8XFpVqz5N59913ExMTgxIkTaNKkSbnrcRyHHj16QC6X4+adOwAAr0aNAJQUTwaUSiWaN2+OtLQ0q7eFHjViI4Sexebp6enQHgwMt4tug0G48+5iaCexeIghBvKwH33ysD/kcrlVBRJjDJGRkdi3bx/i4+Ph7+9f4focx+H69evQaDTGwqhDcDCUCoXZ7QE0Gg1u375drZmBNIvNRmg0wo2MaDQaHDp0yKE9FJwCM5rMgIJTCKIPiKOdxOIhhhjIw370ycP+yM7Ohl6vr/L606dPx44dO7Bz5064uroiIyMDGRkZKCwsBADcvHkTy5Ytw4ULF3D79m0cPHgQL7/8Mjp27IjunToBANzq1sWUceOwZMkSHDt2DMnJyXj77bcBAGPGjLE6Bj5zQNcg1SK2mBIqtIdSohRUHxBHO4nFQwwxkIf96JOHY7NhwwYAJTeDNGXLli2YPHkyFAoFfvrpJ6xbtw55eXlo2rQpBg4ciOXLl0P66JFx/eVz56KOpyciIiJQWFiIbt26IT4+HvXr17dlOKWgAokgCIIg7JDuu9YDAJQ+wdDr9VCpVHBzc6vxaaTiB+ZT6E31raGyKfVNmzZFQsKzm0+axpBvUiDJ5XKsWbMGa9asscpfaOgUG0EQBEEQhAVUIFWATCbcAJtMJkP//v0d2kPDNNicvlnQWWxiaCexeIghBvKwH33ysA6O4+Dq6iros9iE1LeVB585oAKpFnF2dnZoDwaGXG2uoLPYAMdvJzF5iCEG8rAfffKwDiELC1vo28qDL6hAqgChn91z+PBhh/ZQcArMbDpT0FlsYmgnsXiIIQbysB998rAOxpjVz0qzJ31bedCz2AiCIAiCIASECiSCIAiCIAgLqEAiCIIgCIKwgAqkChB61sPQoUMd2kPN1Ii+Gw01UwuiD4ijncTiIYYYyMN+9MnDOjiOg5ubm6Cz2ITUt5UHzWITCYbbsTuqBwcOrjJXcBB2VoKjt5OYPMQQA3nYjz55WIeQFzfbQt9WHnxBBVIFCD3r4cSJEw7tIefkmOo9FXJOLog+II52EouHGGIgD/vRJw/rYIwhNzdX0FlsQurbyoNmsREEQRAEUassXboUHMeZvby8vIzv79u3D4MHD4aHhwc4jkNSUpLZ5x9nZ2PO8uXoFB6OOnXqwNfXF++99x5ycnJsHUqZ0LPYCIIgCMIO2fTZraf/u1XhejWnRP/1xb2t/mRwcDCOHz9u/FsqlRr/n5+fj549e2LMmDF48803S302IzMTGZmZWPHBB+gcFoY7d+5g2rRpePDgAf7zn/9UIw5+oQKpFhHygj5beRTriwXVB8TRTmLxEEMM5GE/+uTh+MhkMrNRI1MiIiIAALdv3y7z/TatWmHH558DAFwDAhAQEIAVK1bg9ddfh1arrfU2/d/MaBWRy4W7tkYul2PYsGGC6dvCQ83UWHdvnWD6gDjaSSweYoiBPOxHnzzsj3r16ln9mZSUFPj4+ECpVKJbt26IiopC8+bNy1xXIpFU6pGTkwM3N7dqF0d8HrfpGqQK0Ov1gmpnZmY6tAcHDs2cmgk6i00M7SQWDzHEQB72o08e9odGo7HqAupu3bph27ZtiI2NxaZNm5CRkYEePXogKyurzPUZYxV6ZGVl4eOPP8Zbb71Vre0H+D1uU4FUATqdTlDtM2fOOLSHnJNjrOdYQWexiaGdxOIhhhjIw370ycP+yM/Pt6pACg8Px8svv4y2bdti4MCBOHToEABg69atZa7PGCvXQ6VSYdiwYWjTpg2WLFlSvQDA73GbCiSCIAiCIGqMi4sL2rZti5SUFKs+l5ufjyFDhqBu3brYv3+/oJe3WAMVSARBEARB1Jji4mJcu3YN3t7eVf6MKi8Po6ZOhUKhQExMDJycnATcQuugi7QrQOhbrru6ujq0BwPDI/UjMAh30y8xtJNYPMQQA3nYjz552B8SicSqGObMmYMXX3wRvr6+yMzMxPLly6FSqTBp0iQAwOPHj5GWloYHDx4AAK5fv46CggIEBASgLkpGjkZNnYrCoiL8sHcvVCoVVCoVAKBRo0ZmtwyoKnzmgAqkChD62T0DBgwQTN8WHhqmwbcZ3wqmD4ijncTiIYYYyMN+9MnD/nBzc7Nq/Xv37mH8+PF49OgRGjVqhO7du+Ps2bPw8/MDAMTExOAf//iHcf3x48cDAJYsWYL3X30VSVev4sLlywCAFi1amGmnpqaiWbNmVsfA53GbCqQKEHrWw927d9G0aVNIJMKc6RTaQwIJQlxC8Ef+H7xrGzCNwRYejpoLW3iIIQbyqJ6+UIihnSw9+OTN90umyyt9gsEYg1qthkKhqPEoSfGDq2Z/G/SLi4ut0t+1a1eF70+ePBmTJ082/m0aQ15yMno//zxU164BAFyDgqwLohxoFpuNEHrWQ1JSkkN7yDgZhjQcAhknXJ0thnYSi4cYYiAP+9EnD+tgjKGwsFDQZ7EJqW8rD5rFRhAEQRAEISBUIBEEQRAEQVhABVIFCD3roVGjRg7twcCQWpgq+Cw2R28nsXiIIQbysB998rDeQyaTCZpvIfVt6cEXdJF2BQg9i61Hjx6C6dvCQ8M02PP3HsH0AXG0k1g8xBADediPPnlYB8dxqFu3rsPq28qDz+M2jSBVgNAX9f31118O7SGFFD3de0IK6+9VUVXE0E5i8RBDDORhP/rkUZqKLl6mi7TNdUooPWONLtK2EUJP809OTnZoDyn3tEDihCuQxNBOYvEQQwzkYT/65PEMww0R1Wp1uesYpuELWSAJqc+nR0FBAQBAUpRT6j0+80yn2AiCIAiiFpHJZKhTpw7+/vtvyOVyaLXmB3lWVAS9Xg+1Wo2ioqIa389JLbB+eZh6qC0KmaKioko/zxhDQUEBMjMzUa9ePWh0lX+mJlCBRBAEQRC1CMdx8Pb2RmpqKu7cuQNtdqbZ+7J8mfH0lLOzc40vRBZavzxMPYofPjR7z8kKz3r16sHLywt3+d5AC6hAqgChqmiDtq+vr0N76KHH5bzL0JdxHpgvxNBOYvEQQwzkYT/65GGOQqFAy5YtoVar8eCrGWbv+UyPgU6nQ3JyMvz8/Kr1jDJThNYvD1OPxH8uMXuv9Y7tVdKQy+UVbh+febaLAmn9+vX49NNPkZ6ejuDgYERHR6N3797lrr93714sXrwYN2/eREBAAFasWIGXXnrJ+D5jDB999BE2btyIJ0+eoFu3bvjqq68QHBxs1XYJ9SUxaHfs2FEwfVt4aJkWRx8fFUwfEEc7icVDDDGQh/3ok0dpJBIJnJycIMtPN1tueMJ9p06dauwBQHD9ijB4sEzzUSzDNtQUPo/btX6R9u7duzFz5kwsXLgQly5dQu/evREeHo60tLQy1z9z5gzGjRuHiIgI/P7774iIiMDYsWNx7tw54zqffPIJ1q5diy+//BLnz5+Hl5cXBg0ahNzcXKu2TehZD5cuXXJoDxknw5AGwj9qxNHbSSweYoiBPOxHnzzsy0MMMRg8+KLWC6S1a9di6tSpeOONN9C6dWtER0ejadOm2LBhQ5nrR0dHY9CgQViwYAGCgoKwYMECvPDCC4iOjgZQMnoUHR2NhQsXYvTo0QgJCcHWrVtRUFCAnTt3WrVtQs96SEtLc2gPCSRoV7cdJAJ+jcTQTmLxEEMM5GE/+uRhXx5iiMHgwRe1WiCp1WokJiYiLCzMbHlYWBhOnz5d5mfOnDlTav3Bgwcb109NTUVGRobZOkqlEn379i1XkyAIgiAIwpRavQbp0aNH0Ol0aNy4sdnyxo0bIyMjo8zPZGRkVLi+4d+y1rlz506ZmsXFxSguLjb+nZNTcm+Fx48fG5dJJBJIpVIUafIBk4vtnzx5AqlUCq1Wa3ZvB6lUColEAq1Wi2z1s2aW6DVQqVQoKChAQUEBsrKyIJfLjXf/1Gq1ZttW3nJdcSE4ALKnp1uzsrKMt3HX6/XQ6XTQaDQoKCjA48eP4ezsDJ1OZ1ZdG2Iqb3lFMUmKJJBwEhQUFEBXqINKpYJGo6nStpe3XC6XG7fdgFarRUFBAbKzs5FnevGdXg+VSlVpTBJtoWF16FhJTi1j0ul0ZrkwjbUqMUm0hdDoYMxHVlZWqZiKioqMHgqFwixPBgz5q0qestUycEwLCdNBJ5HDJSvLmO/s7Gwolcpy81fdPBn0VSqVcXsMFGnywfQcwDFwkmdtUFFMuUU66DkZ2NP7aLlkZRlz8fjxY7M74pp+9yqKyZBvrQ5gMM+FISZDHKb5Luu7Z9mfysuToX9zTAcJ0+LJkycoLi42eiiVykr7U2V5ylbLINGrATDoJUqzfOfk5DydGl55f6pqTJb5zitJmnH9quz3DLkw5KOsfYRh2rZpLgx5Aqq2jzCNyXRfC+hRT6Uy63tyubxG+73y8mT6ncrT6QDGAIUCwLPvoDUxZaulkOrV0HNSME5mlm+VSgWJRFKjfXlOMZ7uOxQAuFL6lvCxL+c4zizfxv25Xg9OW9JvrI3JNN8c00KlUuHJkycAKr7pZpVhtcj9+/cZAHb69Gmz5cuXL2eBgYFlfkYul7OdO3eaLduxYwdTKpWMMcZ+/fVXBoA9ePDAbJ033niDDR48uEzNJUuWMJTsT+lFL3rRi170opeDv27evFnd0sRIrY4geXh4QCqVlhotyszMLDUCZMDLy6vC9b28vACUjCR5e3tXSXPBggWYPXu28e/s7Gz4+fkhLS0N7u7u1gdWBVQqFZo2bYq7d+/Czc3NIT3EEAN52I8+ediXhxhiIA/70beVR05ODnx9fdGgQYMaa9VqgaRQKNC5c2fExcWZTdOPi4vDyJEjy/xMaGgo4uLiMGvWLOOyY8eOGR8U6O/vDy8vL8TFxRmnXarVaiQkJGD16tVlaiqVSiiVylLL3d3dBUuiATc3N4f3EEMM5GE/+uRhXx5iiIE87EffVh583A+p1u+DNHv2bERERKBLly4IDQ3Fxo0bkZaWhmnTpgEAJk6ciOeeew4rV64EAMyYMQN9+vTB6tWrMXLkSBw4cADHjx/HL7/8AqDkPOfMmTMRFRWFli1bomXLloiKikKdOnUwYcKEWouTIAiCIAjHodYLpHHjxiErKwvLli1Deno6QkJCcPjwYfj5+QEA0tLSzCrBHj16YNeuXVi0aBEWL16MgIAA7N69G926dTOu88EHH6CwsBDvvPOO8UaRx44dg6urq83jIwiCIAjC8ZAuXbp0aW1vRNeuXTFz5kwsXrwYb731lrE4AoDJkydj1KhRZuu3adMGkZGRWLx4MSIjI9G6dWuz9zmOQ79+/TBnzhwsWrQI//jHP+Dp6WnVNkmlUvTr189sJg3fiMFDDDGQh/3ok4d9eYghBvKwH31H8+AY42MuHEEQBEEQhHio9TtpEwRBEARB2BtUIBEEQRAEQVhABRJBEARBEIQFVCARBEEQBEFYQAVSGaxfvx7+/v5wcnJC586d8fPPP/OmferUKbz44ovw8fEBx3H48ccfedMGgJUrV6Jr165wdXWFp6cnRo0aheTkZF49NmzYgHbt2hlv9hUaGoojR47w6mHJypUrjfe44oulS5eC4zizl+FO7Hxx//59vP7662jYsCHq1KmDDh06IDExkTf9Zs2alYqB4zhMnz6dNw+tVotFixbB398fzs7OaN68OZYtW8b7E7lzc3Mxc+ZM+Pn5wdnZGT169MD58+errVdZX2OMYenSpfDx8YGzszP69euHq1ev8uqxb98+DB48GB4eHuA4DklJSbzpazQazJs3D23btoWLiwt8fHwwceJEPHjwgNcYli5diqCgILi4uKB+/foYOHAgzp07x6uHKW+99RY4jkN0dDSvHpMnTy7VT7p37857HNeuXcOIESPg7u4OV1dXdO/eHWlpabzol9XXOY7Dp59+ylsMeXl5iIyMRJMmTeDs7IzWrVtjw4YNVdavisfDhw8xefJk+Pj4oE6dOhgyZAhSUlKqrF+V41xxcTHeffddeHh4wMXFBSNGjMC9e/esioMKJAt2796NmTNnYuHChbh06RJ69+6N8PDwKn/BKyM/Px/t27fHl19+yYueJQkJCZg+fTrOnj2LuLg4aLVahIWFIT8/nzePJk2aYNWqVbhw4QIuXLiAAQMGYOTIkVYfXKrK+fPnsXHjRrRr14537eDgYKSnpxtfV65c4U37yZMn6NmzJ+RyOY4cOYI///wTn332GerVq8ebx/nz5822Py4uDgAwZswY3jxWr16Nr7/+Gl9++SWuXbuGTz75BJ9++im++OIL3jwA4I033kBcXBxVDWd1AAAQV0lEQVS2b9+OK1euICwsDAMHDsT9+/erpVdZX/vkk0+wdu1afPnllzh//jy8vLwwaNAg5Obm8uaRn5+Pnj17YtWqVbzHUFBQgIsXL2Lx4sW4ePEi9u3bh+vXr2PEiBG8eQBAq1at8OWXX+LKlSv45Zdf0KxZM4SFheHvv//mzcPAjz/+iHPnzsHHx8eqGKrqMWTIELP+cvjwYV49bt68iV69eiEoKAgnT57E77//jsWLF8PJyYkXfdNtT09Px7fffguO4/Dyyy/zFsOsWbNw9OhR7NixA9euXcOsWbPw7rvv4sCBA7x4MMYwatQo3Lp1CwcOHMClS5fg5+eHgQMHVvk4VZXj3MyZM7F//37s2rULv/zyC/Ly8jB8+HCzB+hWSo2f5iYynn/+eTZt2jSzZUFBQWz+/Pm8ewFg+/fv513XlMzMTAaAJSQkCOpTv3599u9//5t33dzcXNayZUsWFxfH+vbty2bMmMGb9pIlS1j79u1507Nk3rx5rFevXoLpl8WMGTNYQEAA0+v1vGkOGzaMTZkyxWzZ6NGj2euvv86bR0FBAZNKpezgwYNmy9u3b88WLlxYY33LvqbX65mXlxdbtWqVcVlRURFzd3dnX3/9NS8epqSmpjIA7NKlS9XSrkzfwG+//cYAsDt37gjmkZOTwwCw48eP8+px79499txzz7E//viD+fn5sX/961/V0i/PY9KkSWzkyJHV1qyKx7hx43jrF1XJxciRI9mAAQN49QgODmbLli0zW9apUye2aNEiXjySk5MZAPbHH38Yl2m1WtagQQO2adOmanlYHueys7OZXC5nu3btMq5z//59JpFI2NGjR6usSyNIJqjVaiQmJiIsLMxseVhYGE6fPl1LW1UzcnJyAICXB/eVhU6nw65du5Cfn4/Q0FDe9adPn45hw4Zh4MCBvGsDQEpKCnx8fODv749XX30Vt27d4k07JiYGXbp0wZgxY+Dp6YmOHTti06ZNvOlbolarsWPHDkyZMgUcx/Gm26tXL/z000+4fv06AOD333/HL7/8gqFDh/LmodVqodPpSv3SdnZ2Nj5GiE9SU1ORkZFh1teVSiX69u3rsH0dKOnvHMfxOkppilqtxsaNG+Hu7o727dvzpqvX6xEREYG5c+ciODiYN11LTp48CU9PT7Rq1QpvvvkmMjMzedPW6/U4dOgQWrVqhcGDB8PT0xPdunXj/TIKAw8fPsShQ4cwdepUXnV79eqFmJgY3L9/H4wxnDhxAtevX8fgwYN50S8uLgYAs74ulUqhUCiq3dctj3OJiYnQaDRm/dvHxwchISFW9W8qkEx49OgRdDodGjdubLa8cePGyMjIqKWtqj6MMcyePRu9evVCSEgIr9pXrlxB3bp1oVQqMW3aNOzfvx9t2rTh1WPXrl24ePGi8Tl8fNOtWzds27YNsbGx2LRpEzIyMtCjRw9kZWXxon/r1i1s2LABLVu2RGxsLKZNm4b33nsP27Zt40Xfkh9//BHZ2dmYPHkyr7rz5s3D+PHjERQUBLlcjo4dO2LmzJkYP348bx6urq4IDQ3Fxx9/jAcPHkCn02HHjh04d+4c0tPTefMxYOjPYunrAFBUVIT58+djwoQJvD8I9ODBg6hbty6cnJzwr3/9C3FxcfDw8OBNf/Xq1ZDJZHjvvfd407QkPDwc33//PeLj4/HZZ5/h/PnzGDBggPGAXVMyMzORl5eHVatWYciQITh27BheeukljB49GgkJCbx4mLJ161a4urpi9OjRvOp+/vnnaNOmDZo0aQKFQoEhQ4Zg/fr16NWrFy/6QUFB8PPzw4IFC/DkyROo1WqsWrUKGRkZ1errZR3nMjIyoFAoUL9+fbN1re3ftf4sNnvE8tc3Y4zXX+S2IjIyEpcvXxbkF3hgYCCSkpKQnZ2NvXv3YtKkSUhISOCtSLp79y5mzJiBY8eOVfn8vbWEh4cb/9+2bVuEhoYiICAAW7duxezZs2usr9fr0aVLF0RFRQEAOnbsiKtXr2LDhg2YOHFijfUt2bx5M8LDw6t1/UZF7N69Gzt27MDOnTsRHByMpKQkzJw5Ez4+Ppg0aRJvPtu3b8eUKVPw3HPPQSqVolOnTpgwYQIuXrzIm4clYunrGo0Gr776KvR6PdavX8+7fv/+/ZGUlIRHjx5h06ZNGDt2LM6dO2f1I5zKIjExEevWrcPFixcFbftx48YZ/x8SEoIuXbrAz88Phw4d4qXIMExaGDlyJGbNmgUA6NChA06fPo2vv/4affv2rbGHKd9++y1ee+013vePn3/+Oc6ePYuYmBj4+fnh1KlTeOedd+Dt7c3LSL5cLsfevXsxdepUNGjQAFKpFAMHDjTbH1uDNcc5a/s3jSCZ4OHhAalUWqrCzMzMLPVL09559913ERMTgxMnTqBJkya86ysUCrRo0QJdunTBypUr0b59e6xbt443/cTERGRmZqJz586QyWSQyWRISEjA559/DplMZt2FdlXExcUFbdu2tWo2RUV4e3uXKhhbt27N2wX/pty5cwfHjx/HG2+8wbv23LlzMX/+fLz66qto27YtIiIiMGvWLN5H9gICApCQkIC8vDzcvXsXv/32GzQaDfz9/Xn1AWCcrSiGvq7RaDB27FikpqYiLi6O99EjoKRvtGjRAt27d8fmzZshk8mwefNmXrR//vlnZGZmwtfX19jX79y5g/fffx/NmjXjxaMsvL294efnx1t/9/DwgEwms0mf//nnn5GcnMx7fy8sLMSHH36ItWvX4sUXX0S7du0QGRmJcePGYc2aNbz5dO7c2fgDOz09HUePHkVWVpbVfb2845yXlxfUajWePHlitr61/ZsKJBMUCgU6d+5snAlkIC4uDj169KilrbIOxhgiIyOxb98+xMfHC3JwKc+Xr6FqAHjhhRdw5coVJCUlGV9dunTBa6+9hqSkJEilUt68DBQXF+PatWvw9vbmRa9nz56lpp5ev37d7GHMfLFlyxZ4enpi2LBhvGsXFBRAIjHfVUilUt6n+RtwcXGBt7c3njx5gtjYWIwcOZJ3D39/f3h5eZn1dbVajYSEBIfp68Cz4iglJQXHjx9Hw4YNbeLLZ3+PiIjA5cuXzfq6j48P5s6di9jYWF48yiIrKwt3797lrb8rFAp07drVJn1+8+bN6Ny5M6/XgQEl3yeNRmOz/u7u7o5GjRohJSUFFy5cqHJfr+w417lzZ8jlcrP+nZ6ejj/++MOq/k2n2CyYPXs2IiIi0KVLF4SGhmLjxo1IS0vDtGnTeNHPy8vDjRs3jH+npqYiKSkJDRo0gK+vb431p0+fjp07d+LAgQNwdXU1/kJ2d3eHs7NzjfUB4MMPP0R4eDiaNm2K3Nxc7Nq1CydPnsTRo0d50QdKrkmxvG7KxcUFDRs25O16qjlz5uDFF1+Er68vMjMzsXz5cqhUKt5OG82aNQs9evRAVFQUxo4di99++w0bN27Exo0bedE3oNfrsWXLFkyaNEmQJ2S/+OKLWLFiBXx9fREcHIxLly5h7dq1mDJlCq8+sbGxYIwhMDAQN27cwNy5cxEYGIh//OMf1dKrrK/NnDkTUVFRaNmyJVq2bImoqCjUqVMHEyZM4M3j8ePHSEtLM96byHDw9PLyqtI9tyrS9/HxwSuvvIKLFy/i4MGD0Ol0xv7eoEEDKBSKGsfQsGFDrFixAiNGjIC3tzeysrKwfv163Lt3z6pbSVTWTpaFnVwuh5eXFwIDA3nxaNCgAZYuXYqXX34Z3t7euH37Nj788EN4eHjgpZde4i2OuXPnYty4cejTpw/69++Po0eP4r///S9OnjzJiz4AqFQq7NmzB5999lmVt9saj759+2Lu3LlwdnaGn58fEhISsG3bNqxdu5Y3jz179qBRo0bw9fXFlStXMGPGDIwaNarUBKnyqOw45+7ujqlTp+L9999Hw4YN0aBBA8yZMwdt27a17jRhtebUiZyvvvqK+fn5MYVCwTp16sTrFPkTJ04wAKVekyZN4kW/LG0AbMuWLbzoM8bYlClTjO3TqFEj9sILL7Bjx47xpl8efE/zHzduHPP29mZyuZz5+Piw0aNHs6tXr/Kmzxhj//3vf1lISAhTKpUsKCiIbdy4kVd9xhiLjY1lAFhycjLv2owxplKp2IwZM5ivry9zcnJizZs3ZwsXLmTFxcW8+uzevZs1b96cKRQK5uXlxaZPn86ys7OrrVdZX9Pr9WzJkiXMy8uLKZVK1qdPH3blyhVePbZs2VLm+0uWLKmxvuHWAWW9Tpw4wUsMhYWF7KWXXmI+Pj5MoVAwb29vNmLECPbbb7/x2k6WVGeaf0UeBQUFLCwsjDVq1IjJ5XLm6+vLJk2axNLS0niPY/PmzaxFixbMycmJtW/fnv3444+86n/zzTfM2dm52n2jMo/09HQ2efJk5uPjw5ycnFhgYCD77LPPrLp1SGUe69atY02aNDHmYtGiRVbtT6pynCssLGSRkZGsQYMGzNnZmQ0fPtzqfHNPzQiCIAiCIIin0DVIBEEQBEEQFlCBRBAEQRAEYQEVSARBEARBEBZQgUQQBEEQBGEBFUgEQRAEQRAWUIFEEARBEARhARVIBEEQBEEQFlCBRBAEYcHkyZMxatSo2t4MgiBqESqQCIIgCIIgLKACiSAI0aBWq2t7EwiCEAlUIBEE4bD069cPkZGRmD17Njw8PDBo0CCsXbsWbdu2hYuLC5o2bYp33nkHeXl5xs989913qFevHmJjY9G6dWvUrVsXQ4YMQXp6erk+iYmJ8PT0xIoVK2wRFkEQdgAVSARBODRbt26FTCbDr7/+im+++QYSiQSff/45/vjjD2zduhXx8fH44IMPzD5TUFCANWvWYPv27Th16hTS0tIwZ86cMvVPnjyJF154AR999BEWLlxoi5AIgrADZLW9AQRBEDWhRYsW+OSTT4x/BwUFGf/v7++Pjz/+GG+//TbWr19vXK7RaPD1118jICAAABAZGYlly5aV0j5w4AAiIiLwzTffYPz48QJGQRCEvUEFEkEQDk2XLl3M/j5x4gSioqLw559/QqVSQavVoqioCPn5+XBxcQEA1KlTx1gcAYC3tzcyMzPNdM6dO4eDBw9iz549eOmll4QPhCAIu4JOsREE4dAYih4AuHPnDoYOHYqQkBDs3bsXiYmJ+OqrrwCUjBoZkMvlZhocx4ExZrYsICAAQUFB+Pbbb+nib4L4H4QKJIIgRMOFCxeg1Wrx2WefoXv37mjVqhUePHhQLS0PDw/Ex8fj5s2bGDdunFmBRRCE+KECiSAI0RAQEACtVosvvvgCt27dwvbt2/H1119XW8/T0xPx8fH466+/MH78eGi1Wh63liAIe4YKJIIgREOHDh2wdu1arF69GiEhIfj++++xcuXKGml6eXkhPj4eV65cwWuvvQadTsfT1hIEYc9wzPLEO0EQBEEQxP84NIJEEARBEARhARVIBEEQBEEQFlCBRBAEQRAEYQEVSARBEARBEBZQgUQQBEEQBGEBFUgEQRAEQRAWUIFEEARBEARhARVIBEEQBEEQFlCBRBAEQRAEYQEVSARBEARBEBZQgUQQBEEQBGEBFUgEQRAEQRAW/H9Cc/ZJnrcBHAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkgAAAHFCAYAAAAJ2AY0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzsnXl8FdX9/p+ZuyQBQpAlIEsCSCXBggloNYkICYIQrNYNaivYCi1IqUZ+tYVSa8EFQUtjq2BtrVvVonWXiIkSFkmEFghYlaV+gSCLJKghSHKXmfP7I8xwt9zMhDmTO6ef9+vFS3My9zznOc+5c0/OPedeiTHGQBAEQRAEQejIHd0AgiAIgiCIRIMmSARBEARBEBHQBIkgCIIgCCICmiARBEEQBEFEQBMkgiAIgiCICGiCRBAEQRAEEQFNkAiCIAiCICKgCRJBEARBEEQENEEiCIIgCIKIgCZIBEGYZv/+/ZAkCU8//XS7Hi9JEn73u99Z2qaO4oEHHsDrr78eVb5u3TpIkoR169Zx066qqsLvfvc7fP3111G/Gzt2LMaOHctNmyBEhyZIBEEQZ0FrE6SRI0eiuroaI0eO5KZdVVWFRYsWxZwgrVixAitWrOCmTRCi4+7oBhAEQYhI165dcemll3aY/rBhwzpMmyBEgFaQCMKh/O53v4MkSdi5cyduvPFGpKWloXv37pg3bx6CwSB2796NiRMnIjU1FQMHDsSyZcvCHl9bW4ubb74Z6enpSEpKQnZ2Nn7/+99DVdWw6w4fPowpU6YgNTUVaWlpmDp1Ko4ePRrVntbe0vnRj36EgQMHtunn6NGjmDVrFvr37w+v14tBgwZh0aJFCAaDpvqlLerq6jBnzhwMGzYMXbp0QXp6OoqKirBx48aoa30+HxYvXozs7GwkJyejR48eKCwsRFVVFYCWtwq/+eYbPPPMM5AkCZIk6X0Q+RZbaWkpJEnCf//73yidX/3qV/B6vaivrwcAVFRU4JprrkH//v2RnJyMIUOGYNasWfrvgZb877rrLgDAoEGDdH1NL1YeX375JebMmYN+/frB6/Vi8ODBWLhwIXw+X9h1kiRh7ty5eO6555CdnY1OnTrhwgsvxNtvv226vwnCqdAKEkE4nClTpuDmm2/GrFmzUFFRgWXLliEQCOC9997DnDlz8Itf/AIvvPACfvWrX2HIkCG47rrrUFdXh/z8fPj9ftx7770YOHAg3n77bfziF7/AZ599pr8109TUhCuuuAKHDx/GkiVLcP7552P16tWYOnWqpR6OHj2K73znO5BlGb/97W9x3nnnobq6Gvfddx/279+Pp556yjKtL7/8EgBwzz33oE+fPjh58iRee+01jB07Fu+//74+qQgGg5g0aRI2btyIkpISFBUVIRgM4sMPP0RtbS3y8/NRXV2NoqIiFBYW4u677wbQsnIUi5tvvhm/+tWv8PTTT+O+++7TyxVFwd///nd897vfRc+ePQEAn332GfLy8jBz5kykpaVh//79WL58OS677DJ89NFH8Hg8mDlzJr788kv86U9/wquvvopzzz0XQOsrR83NzSgsLMRnn32GRYsWYcSIEdi4cSOWLFmCmpoarF69Ouz61atX41//+hcWL16MLl26YNmyZbj22muxe/duDB48uP0BEIRTYARBOJJ77rmHAWC///3vw8pzcnIYAPbqq6/qZYFAgPXq1Ytdd911jDHG5s+fzwCwzZs3hz32tttuY5Iksd27dzPGGFu5ciUDwN54442w637yk58wAOypp57Sy8aMGcPGjBkT1c5bbrmFZWZmhpUBYPfcc4/+86xZs1iXLl3YgQMHwq57+OGHGQD28ccfx+2LsyEYDLJAIMDGjRvHrr32Wr382WefZQDYX/7yl7iP79y5M7vllluiyisrKxkAVllZqZddd911rH///kxRFL2srKyMAWBvvfVWzPpVVWWBQIAdOHAgKouHHnqIAWD79u2LelxkHo8//jgDwF566aWw65YuXcoAsPLycr0MAOvduzc7ceKEXnb06FEmyzJbsmRJa11BEEJBb7ERhMO56qqrwn7Ozs6GJEmYNGmSXuZ2uzFkyBAcOHAAALB27VoMGzYM3/nOd8Ie+6Mf/QiMMaxduxYAUFlZidTUVFx99dVh1/3gBz+w1MPbb7+NwsJC9O3bF8FgUP+neVi/fr2leo8//jhGjhyJ5ORkuN1ueDwevP/++/j000/1a9555x0kJyfj1ltvtUz3xz/+MT7//HO89957etlTTz2FPn36hOV17NgxzJ49GwMGDNDbl5mZCQBhbTTD2rVr0blzZ9xwww1h5T/60Y8AAO+//35YeWFhIVJTU/Wfe/fujfT0dH0MEYTo0ASJIBxO9+7dw372er3o1KkTkpOTo8qbm5sBAMePH9ffkgmlb9+++u+1//bu3Tvquj59+ljSdo0vvvgCb731FjweT9i/Cy64AADC9t6cLcuXL8dtt92GSy65BK+88go+/PBD/Otf/8LEiRPR1NSkX1dXV4e+fftClq27TU6aNAnnnnuu/pbhV199hTfffBPTp0+Hy+UCAKiqigkTJuDVV1/FL3/5S7z//vvYsmULPvzwQwAIa6MZjh8/jj59+kCSpLDy9PR0uN1uPXONHj16RNWRlJTUbn2CcBq0B4kg/gfp0aMHjhw5ElV++PBhAND3wvTo0QNbtmyJui7WJu3k5GQ0NDRElRuZ3PTs2RMjRozA/fffH/P32sTNCv7+979j7NixWLlyZVh5Y2Nj2M+9evXCBx98AFVVLZskuVwuTJs2DX/84x/x9ddf44UXXoDP58OPf/xj/Zr//Oc/2LFjB55++mnccsstenmszd1m6NGjBzZv3gzGWNgk6dixYwgGg3rmBEG0QCtIBPE/yLhx4/DJJ59g27ZtYeXPPvssJElCYWEhgJa3WRobG/Hmm2+GXffCCy9E1Tlw4EDs2bMn7ETU8ePH9RNf8bjqqqvwn//8B+eddx4uuuiiqH9WTpAkSUJSUlJY2c6dO1FdXR1WNmnSJDQ3N7f5YZhmV1V+/OMfo7m5GS+++CKefvpp5OXlISsrK6x9Wr2h/PnPf46pDRhbVRo3bhxOnjwZ9ZlNzz77rP57giDOQCtIBPE/yJ133olnn30WkydPxuLFi5GZmYnVq1djxYoVuO2223D++ecDAKZPn44//OEPmD59Ou6//35861vfQllZGd59992oOqdNm4Y///nPuPnmm/GTn/wEx48fx7Jly1o91RXK4sWLUVFRgfz8fNx+++0YOnQompubsX//fpSVleHxxx9H//79LfF+1VVX4d5778U999yDMWPGYPfu3Vi8eDEGDRoU9pECN910E5566inMnj0bu3fvRmFhIVRVxebNm5GdnY3vf//7AIDhw4dj3bp1eOutt3DuueciNTUVQ4cObVU/KysLeXl5WLJkCQ4ePIgnnngi6vfnnXce5s+fD8YYunfvjrfeegsVFRVRdQ0fPhwA8Mgjj+CWW26Bx+PB0KFDw/YOaUyfPh2PPfYYbrnlFuzfvx/Dhw/HBx98gAceeADFxcW44oor2tWfBCEsHbxJnCCIdqKdYqurqwsrv+WWW1jnzp2jrh8zZgy74IIL9J8PHDjAfvCDH7AePXowj8fDhg4dyh566KGwE1aMMfb555+z66+/nnXp0oWlpqay66+/nlVVVUWdYmOMsWeeeYZlZ2ez5ORkNmzYMLZq1SpDp9gYY6yuro7dfvvtbNCgQczj8bDu3buzUaNGsYULF7KTJ0+a76BW8Pl87Be/+AXr168fS05OZiNHjmSvv/56zHY2NTWx3/72t+xb3/oW83q9rEePHqyoqIhVVVXp19TU1LCCggLWqVMnBkA/ORbrFJvGE088wQCwlJQU1tDQEPX7Tz75hI0fP56lpqayc845h914442strY2Zr8tWLCA9e3bl8myHKYX61Th8ePH2ezZs9m5557L3G43y8zMZAsWLGDNzc1h1wFgP/vZz6LalZmZGfPEHkGIiMQYYx02OyMIgiAIgkhAaA8SQRAEQRBEBLQHiSAIR6AoCuIteEuSFPf3QMspsshj7gRBELGgFSSCIBzBuHHjoj4nKfTfeeedF/f3Ho/H8g+cJAhCXDp8D9KKFSvw0EMP4ciRI7jgggtQWlqK0aNHx7z2448/xm9/+1ts3boVBw4cwB/+8AeUlJSEXbNkyRK8+uqr2LVrF1JSUpCfn4+lS5fGPVVCEETis3v37qjPKgrlm2++QefOnePW0doJL4IgiEg69C22VatWoaSkBCtWrEBBQQH+/Oc/Y9KkSfjkk0+QkZERdf2pU6cwePBg3Hjjjbjzzjtj1rl+/Xr87Gc/w8UXX4xgMIiFCxdiwoQJ+OSTT9q8eRIEkbjQHzkEQdhJh64gXXLJJRg5cmTYJ9pmZ2fje9/7HpYsWRL3sQMHDkRJSUnUClIkdXV1SE9Px/r163H55Zdb0m6CIAiCIMSmw1aQ/H4/tm7divnz54eVT5gwwdAn7xpF++qDyO+rCsXn84V9+q+qqvjyyy/Ro0cP2tBJEARBEA6BMYbGxkZLvkexwyZI9fX1UBQl6oswe/fuHfN7ntoDYwzz5s3DZZddhm9/+9utXrdkyRIsWrTIEk2CIAiCIDqWgwcPnvWn73f4Mf/IFRoW8UWKZ8PcuXOxc+dOfPDBB3GvW7BgAebNm6f/3NDQgIyMDOzduxfnnHMOAECWZbhcLiiKAlVV9Wu18mAwGHbE2OVyQZblVsubmppQU1ODnJwcuN1uuN0tUYR+1QGAVss9Hg9UVYWiKHqZJElwu916eTAYRE1NDXJzc5GcnNxq29vrSat/1KhR8Hq9CAQChtpuxpOiKNixYwdyc3PDxoXm1QpPqqri3//+Ny688EK9bZpXqzz5fD49b4/HE5aTVZ4i82gtv/Z60uq/+OKLIcty3LHXXk+qqqKmpgYXXnih/u32oW23wlMwGMS///1v/bkHGHs+mfHk9/v1vL1eb7vuEW150vK46KKLdF+hnK2nYDCIHTt24KKLLoIkSZbc9yI9AcDWrVvDnnuxvJ6NJ5/Ph23btul5W30vDwQCehY5OTlITk6O2faz9WQkj7PxFFp/5M4bq16fAETda628l8uyjLq6OgwZMsSSwxgdNkHq2bMnXC5X1GrRsWPHolaV2sPPf/5zvPnmm9iwYUObs8ikpKSoL4YEgHPOOQc9evQ467bEIiUlBc3NzejevTs8Hg8XjUAggObmZpxzzjlcNLT609LSuHo4deoUd42mpiZbsrBDg1dfafWnpqZyz5vXmNU07MrCDo2uXbtyy7upqYl73iI993hr8MxDpLyB6MWX9tBhn4Pk9XoxatSoqC9g1L6wsr0wxjB37ly8+uqrWLt2LQYNGnS2TSUIgiAI4n+MDn2Lbd68eZg2bRouuugi5OXl4YknnkBtbS1mz54NoOXbp/v166efaPP7/fjkk0/0/z906BBqamrQpUsXDBkyBADws5/9DC+88ALeeOMNpKam6itUaWlpSElJ6QCXBEEQBEE4Dju+ETcejz32GMvMzGRer5eNHDmSrV+/Xv/dmDFjwr45et++fQxA1L/Qb6yO9XvE+NbxeDQ0NDAA7KuvvrLAYWwURWH79++P+uZ0J2mI4IE0Eqd+0kgsDRE8kEbi1G+XxldffcUAsIaGhrOuq8M/STsROXHiBNLS0tDQ0ICuXbt2dHMIgiAIgjCAla/f9F1scYjcmW913WvXrnW0hggeSCNx6ieNxNIQwQNpJE79dmpYBU2Q4sBzcY2d/jArJ2uI4IE0Eqd+0kgsDRE8kEbi1G+nhlXQBIkgCIIgCCICmiARBEEQBEFEQJu0Y6Bt8vrqq6/QrVs3LhqqqqK+vh49e/Y86++L6SgNETyQRuLUTxqJpSGCB9JInPrt0vj6669xzjnnWLJJmyZIMaBTbARBEAThPOgUm03E+t4gK+tevXq1ozVE8EAaiVM/aSSWhggeSCNx6rdTwypogtSB8DzqaJeGCB5II3HqJ43E0hDBA2kkTv12aVgFTZAIgiAIgiAioAkSQRAEQRBEBLRJOwbaJq+vv/4aaWlpXDS0D8xKTU2FJEmO1BDBA2kkTv2kkVgaInggjcSp3y6NhoYGdOvWjTZpO52UlBTHa4jggTQSp37SSCwNETyQRuLUb5eGVdAEKQ68vy+mrKzM0RoieCCNxKmfNBJLQwQPpJE49dupYRU0QSIIgiAIgojA3dENIAiCIAjif4cJf54AP/PrP2/6+aYObE3r0AoSQRAEQRBEBHSKLQZ2nWILBoNwu91cTwzw1BDBA2kkTv2kkVgaInggjcSpX9MoeqwobPUIsHYFiU6xCUJTU5PjNUTwQBqJUz9pJJaGCB5II3HqB4BUdyok8JmAWQ1NkOLAe6d9ZWWlozVE8EAaiVM/aSSWhggeSCNx6tc0Zpw7Ax7Jw1XDKmiTNkEQBEEIzqi7ng37eetD0zuoJc6BVpAIgiAIgiAioAlSB+J281/A460hggfSSJz6SSOxNETwQBqJUz8A+FQfdw2roFNsMdBOsVmxC54gCIIgrKDgTwVRZUZPgCXKW2xn48EIVr5+0wpSHFRV5Vr3sWPHHK0hggfSSJz6SSOxNETwIJrGwOSB3E6AieBB07AKmiDFQVEUrnVXV1c7WkMED6SROPWTRmJpiOBBNI0p6VO4nQATwYOmYRU0QSIIgiAIgoiAJkgEQRAEQRAR0OcgxYHXx61rdaempjpaQwQPpJE49ZNGYmmI4CGexvrLx0RdO2bD+nZr1PvrMf7P4xFgAb3cys3HmgYDn3NVdmXB04OmYVlddIotGjrFRhAEITZWTpAA/qezzlaDTrGZh95iiwPv3fwHDhxwtIYIHkgjceonjcTSEMGDnRojOo+AzPEllbeGSP1kFTRBigPv3fw1NTWO1hDBA2kkTv2kkVgaIniwU2Nij4lwS/x2rfDWEKmfrIL2IBEEQRC2ULt4eNjPipwEDL+7g1pDEPGhFSSCIAiCIIgIaAUpDrx38/fq1cvRGiJ4II3EqZ80OlYjenXHg15Fj3L1AKiO66fWNPY17eN+Oounhkj9ZFlddIotGjrFRhDE/xqREyQAyPjtR47TMAqdYqNTbG1Bb7HFgfdmtV27djlaQwQPpJE49ZNGYmmokou7Bzs07MqiIK0ALrgcqyFSP1kFTZDiwPu44+7dux2tIYIH0kic+kkjsTSY5ObuwQ4Nu7IoSCuAS+L3ws9bQ6R+sgqaIBEEQRAEQURAEySCIAiCIIgIaIIUB1nm1z2yLCMjI8PRGiJ4II3EqZ80EktDYgp3D3Zo2JXFzpM7oYLf21O8NUTqJ6ugY/5xcLn4vU/qcrmQm5vLrX47NETwQBqJUz9pJJaGzILcPdihYVcWa75c42gNkfrJKmgFKQ68d/Nv377d0RoieCCNxKmfNBJLQ5Xc3D3YoWFXFhO72/BVIxw1ROonq6AJUhx47+avra11tIYIHkgjceonjcTSYJKLuwc7NOzKYkQXG76slqOGSP1kFTRBIgiCIAiCiIAmSARBEARBEBHQBCkOvHfzDx061NEaInggjcSpnzQSS0NiQe4e7NCwK4tNDZugMH77d3hriNRPVkGn2OLA+xRbVlYWt/rt0BDBA2kkTv2kkVgaMlO4e7BDw64sNjVY+71rdmuI1E9WQStIcQgGg1zrrqqqcrSGCB5II3HqF02j9B+lGPvoWBT8qUD/Z7VGLB/rLx8T9a89KLKHez/ZoWFX3jf2uhEeyeNYDZH6ySpoghQHxhjXuuvq6hytIYIH0kic+kXTGJQyCBIkrhp8fcjc+8kODZHy5qkhUj9ZBU2QCIIgCIIgIujwCdKKFSswaNAgJCcnY9SoUdi4cWOr13788ce4/vrrMXDgQEiShNLS0rOukyAIgiAIIpIOnSCtWrUKJSUlWLhwIbZv347Ro0dj0qRJqK2tjXn9qVOnMHjwYDz44IPo06ePJXXGg/cm7ZycHEdriOCBNBKnftE01hxfgyDjt5+Dtw+JBbn3kx0aIuXNU0OkfrKKDp0gLV++HDNmzMDMmTORnZ2N0tJSDBgwACtXrox5/cUXX4yHHnoI3//+95GUlGRJnfHgfdwxMzPT0RoieCCNxKlfNI2d3/D/Uk6ueTOFfz/ZoSFQ3jw1ROony+qyrCaT+P1+bN26FRMmTAgrnzBhAqqqqhKiTt67+deuXetoDRE8kEbi1C+axq19buV+WoenD0X2cu8nOzREypunhkj9ZBUd9jlI9fX1UBQFvXv3Divv3bs3jh49amudPp8PPp9P//nEiRMAWiZcgUAAQMus1OVyQVGUsO960cqDwWDY7nmXywVZllst9/v9aGxshN/vB2MMbndLFJHhtlbu8XigqmrYF/NJkgS3262XBwIBNDY2IhAIwO12t9r29nrS6lcUBW63W++rttpuxlMwGNQ1QtuiebXCE2MsLItQr1Z5Cs1bu761/NrrKTKP1vJrryetfsZYm2OvvZ4URdHHrJnnk+bp8rtfbGmrAjAAXhew4d6bwjypqhqVt5HnkxlPfr8fPb09kSQlQWUqFCjwSJ6wvjfqqa08VFUFY0wvZ15vSzv9fjBJAjxndON5AgBVcoGd/iJRRfbqeVt131Pk8JV/Bikqi1heNdqTk3YP0TS0tjO3GwhdbTj9WLOeAoEAAoGAnneQBcHA4JW8YRmerSdNQ4IEF1xwSS693Mg9whvyzlPw9CWhnkKf3zzu5ZIkgTGm95N2ko2Bne5+a16fItt+NnT4B0VqT0wNxlhUGe86lyxZgkWLFkWVV1ZWolOnTgCAjIwM5ObmYufOnWH7mYYOHYqsrCxs2bIFdXV1enlOTg4yMzOxYcMGNDY26uV5eXlIT09HZWUlAKCiogIAUFhYiJSUFJSVlYW1obi4GE1NTfr1QMugnDx5Murr61FdXa2Xp6amoqioCAcPHkRNTY1evm3bNhQUFGDv3r3YvXu3Xm6Vp+PHj6Nv374oLy8Pe6JY4alLly4AgEOHDuGjjz7Sy3v16oX8/HxLPPXt2xfAmSyAMzlZ7amioqLVnKzydOjQIQwePLjVsXc2njSMjj2znoYMGQKgZczW19fr5W09nzRP8/PSAAArtzWiwadifl5amK/i4mKcPHlSzwIw/3wy42lO/znY1LAJmxo24Xs9vxfWFqOeNFrLKRgMorm5+UxOP58L+Hzo9uhjCGZk4JsbrtcfE89TfwBfpI/B0T5FiMSq+95/ht0F1ZWsl5+/5/GwLDSsuO9pOX322WdhGlpOTUWF8A8frl+fVNVSp1lPoTnN6T8HTx55Eo3BRpQMKIkae2frSePStEtRkNbyuVplZWWG7hHacwMA3tp7CgCiPGnwuJenpqZi9OjRej9p7GvaBwCWvT5t3rw5yk97kRjfD7loFb/fj06dOuHll1/Gtddeq5ffcccdqKmpwfr16+M+fuDAgSgpKUFJSclZ1xlrBWnAgAE4cuQIevToAcD6FaRTp06hoqIC48ePh8fj4baCVFFRgQkTJiAlJYXLClJFRQUmTpyIpKQkbitI5eXlmDhxYth7y1auICmKgrKyMj2LUK9WeWpubtbz9nq93FaQQvPgsYJUUVGB4uJivT2RbbdiBendd9/FhAkTdP3QtluxghQIBPDOO++E5W31CpLP50NFRQVWfL4CTWqTvoJUMevMRMCKFaSKigpMmjQJHo9HL980qbilnSErSAXvlLXp6fADuVErSB9f8CsUFxdDlmVL7nv77rsozBODhI+G/yYsi1heNdqTU3NzM959911dQ2v7uqJxUStIYyvXtnsFScv7G/UbfQWpfFa5ZZ40jdKDpVCYoq8glc8qN3SPGP2bF/TyoAr8a9n0qBUk7fkdiZUrSGVlZVjx+Qr4WctqOgPDurnrLHt9+uKLL9CnTx80NDSga9euUV7M0GErSF6vF6NGjUJFRUXYZKaiogLXXHONrXUmJSXF3PSdnJwc9qQFWkKItUs+9GZupDw5ORl5eXlITk4Oe+GP1ItXLstyzA1pWrnL5UJeXp7urbW2t9eTVn/oi4zRthv1FKoRy6tVnmJlYbbtrZXLshwz79bya6+nyDxa89peT1r92o0o3tiL1fbvzH8+qnzrQ9PDPIWO2Vj1tOXJH/EVT34l2pfb7Y6Zd3s8tZZTcnIyXjr2kv5iCQABFojZx+3NSesrt9sNSZL0cun027gAIDEG+P1RdbXqlSnA6e/JktVAWN5W3Pdcqi/sZwa51edeqFdDbW+l3Ov1xtSQWtmrYtaTx+OBy+WKytvPovv9bDxpGgEWAAPTv88stL5494jI50akp8jnd2tejZbH8qSqalQ/tdV2s/fy1g5wtYcOfYtt3rx5mDZtGi666CLk5eXhiSeeQG1tLWbPng0AmD59Ovr164clS5YAaFkh+uSTT/T/P3ToEGpqatClSxd9ab6tOs3Aezd/eno6t/rt0BDBA2m0Tu3i4VFlGb/9KMaV1uHEfmpNY3/zfu4aPH1IULn3kx0aIuXNU0OkfrKsLstqagdTp05FaWkpFi9ejJycHGzYsAFlZWXIzMwEANTW1uLIkSP69YcPH0Zubi5yc3Nx5MgRPPzww8jNzcXMmTMN12kGKzd7xap79erVjtYQwQNpGEeRkxzvwU6NO/rfAa/k5arh9LxFGlN25M1TQ6R+sooO36Q9Z84czJkzJ+bv1q1bF/bzwIEDDX3PSrw6Ewmexynt0hDBA2kkTv0iaSTJ1i31twblnTgaduTNW0OUfrKKDv+qEYIgCIIgiESjw1eQCIJom1F3PRtVtvWh6R3QEoIgiP8NaAUpDq3tkreq7sLCQkdriOCBNIwjq37He7BT48kjTyLA+O3nECFvkcaUHXnz1BCpnyyry7KaCNOkpKQ4XkMED2erYXR1J9F9tA0TwIN9Go3BxqijzFbj/LzFGVN25M1bQ5R+sgpaQYoD7++kKSsrc7SGEzyMuuvZqH9WaxhBBA1VTnK8Bzs1SgaUcD2tI0LeIo0pO/LmqSFSP1kFTZAIgiAIgiAioLfYiA6DNh4TBEEQiQqtIBEEQRAEQURAE6Q48N7NX1xc7GgNETyQhnFk1ed4D3ZqlB4s1b9E4XNQAAAgAElEQVSQk5eG0/MWaUzZkTdPDZH6ySpogtSBNDU1OV5DBA+kYRRJAA/2aaS6UyFB4qrh/LzFGVN25M1bQ5R+sgqaIMWB927+yspKR2uI4IE0jKPKXsd7sFNjxrkz4JFif/u5VRpOz1ukMWVH3jw1ROonq6AJEkEQBEEQRAQ0QSIIgiAIgoiAjvl3IDw3w/HQqF08POxnRU6CO3eRZfW3htP6SWQNETzYpeFTfdw1KO/E0bAjb94aovSTVdAEKQ4eD7/3ST0eDyZPnsytfjs0XKrP8R5IwziUtzmNRz5/hLuG0/MWaUzZkTdPDZH6ySroLbY4qKrKte5jx445WoNBdrwH0jAO5W1OY2DyQK6ndUTIW6QxZUfePDVE6ieroAlSHBRF4Vp3dXW1ozVU2eN4D6RhHMrbnMaU9ClcT+uIkLdIY8qOvHlqiNRPVkETJIIgCIIgiAhoD5JDoe8xIwiCIAh+0AQpDpLUvvdJjUxeJElCampquzWMwF+DCeCBNIxDeZvRqPfXg4Fx1XB63iKNKTvy5qkhUj9ZBU2Q4sD7O2mKioq41W+Hhkv1O94DaRiH8jan8bejf+Ou4fS8RRpTduTNU0OkfrIK2oMUB967+Q8cOOBoDVVyOd4DaZion/I2pTGi8wjIHG+xQuQt0JiyI2+eGiL1k1XQBCkOvHfz19TUOFqDSW7HeyAN41De5jQm9pgIt8RvFVqEvEUaU3bkzVNDpH6yCpogEQRBEARBREATJIIgCIIgiAhoghQH3rv5e/Xq5XANVQAPpGEcytuMxr6mfdxP6zg9b5HGlB1589QQqZ+sgk6xxYH3Kbb8/Hxu9duh4VIDjvdAGsahvM1pvFz3MncNp+ct0piyI2+eGiL1k1XQClIceG9W27Vrl6M1VMnleA+kYRzK25xGQVoBXHBx1XB63iKNKTvy5qkhUj9ZBU2Q4sD7uOPu3bsdrcEkt+M9kIZxKG9zGgVpBXBJ/F4IRMhbpDFlR948NUTqJ6ugCRJBEARBEEQENEEiCIIgCIKIgCZIcZBlft0jyzIyMjIcrSExxfEeSMM4lLc5jZ0nd0IFv7crRMhbpDFlR948NUTqJ6ugU2xxcLn4vU/qcrmQm5vLrX47NGQWdLwH0jAO5W1OY82Xa7hrOD1vkcaUHXnz1BCpn6yCVpDiwHs3//bt2x2toUpux3sgDeNQ3uY0Jnbn/5UKTs9bpDFlR948NUTqJ6ugCVIceO/mr62tdbQGk1yO90AaxqG8zWmM6ML/SzmdnrdIY8qOvHlqiNRPVkETJIIgCIIgiAhogkQQBEEQBBEBTZDiwHs3/9ChQx2tIbGg4z2QhnEob3Mamxo2QWH89nOIkLdIY8qOvHlqiNRPVkGn2OLA+xRbVlYWt/rt0JCZ4ngPpGEcytucxqaGTdw1nJ63SGPKjrx5aojUT1ZBK0hxCAaDXOuuqqpytIYiexzvgTSMQ3mb07ix143wSB6uGk7PW6QxZUfePDVE6ieroAlSHBhjXOuuq6tzuIYsgAfSMA7lbUZjUMogSJC4ajg9b5HGlB1589QQqZ+sgiZIBEEQBEEQEdAEiSAIgiAIIgKaIMWB9ybtnJwcR2tILOh4D6RhHMrbnMaa42sQZPz2c4iQt0hjyo68eWqI1E9WQafY4sD7uGNmZia3+u3QkJmCDId7IA0T9VPepjR2frOTu4bj8xZoTNmRN08NkfrJsrosq0lAeO/mX7t2raM1FNnreA+kYRzK25zGrX1u5X5ax+l5izSm7Mibp4ZI/WQVNEGKA+/d/I2NjQ7XkATwQBrGobzNaPT09uR+WsfpeYs0puzIm6eGSP1kFTRBIgiCIAiCiIAmSARBEARBEBHQBCkOvHfz5+XlOVpDVgOO90AaxqG8zWm8dOwlBFiAq4bT8xZpTNmRN08NkfrJKugUWxx4n2JLT0/nVr8dGhJUx3sgDeNQ3uY09jfv567h9LxFGlN25M1TQ6R+sqwuy2pqJytWrMCgQYOQnJyMUaNGYePGjXGvf+WVVzBs2DAkJSVh2LBheO2118J+f/LkScydOxf9+/dHSkoKsrOzsXLlyna1LRDgN8sNBAJYvXq1ozUUOcnxHkjDOJS3OY07+t8Br+TlquH0vEUaU3bkzVNDpH6yig6dIK1atQolJSVYuHAhtm/fjtGjR2PSpEmora2NeX11dTWmTp2KadOmYceOHZg2bRqmTJmCzZs369fceeedWLNmDf7+97/j008/xZ133omf//zneOONN+yyZRiexynt0hDBA2kkTv0iaSTJSdw1KO/E0bAjb94aovSTVXToBGn58uWYMWMGZs6ciezsbJSWlmLAgAGtrviUlpZi/PjxWLBgAbKysrBgwQKMGzcOpaWl+jXV1dW45ZZbMHbsWAwcOBA//elPceGFF+Lf//63XbYIgiAIgnA4HbYHye/3Y+vWrZg/f35Y+YQJE1BVVRXzMdXV1bjzzjvDyq688sqwCdJll12GN998E7feeiv69u2LdevWYc+ePXjkkUdabYvP54PP59N/PnHiBICWpTptuU6WZbhcLiiKAlVV9Wu18mAwqH/+gtcFBFVAZYBHBiTpzLKfy+WCLMv6z9p/3e6WKCJn8K2VA4AEwBOyHy0YDMLtdkNVVSiKotcdDAbh8XhabbsRTwCgSi7ITIEiewFIp/8L/bGRS5ttefJG7KXzK9DbHupJ0witX5IkuN3uNj1pGooKKKfzCK0ndENfZHloTvE8eV0tbQ/NQ3ucx+MJa3sgENDbHuk1nicAcMuAHPLxIYqihOWkaWiPjczPjCcAaHmkBPV0zlremkastsfzFJq3ylqeI5FetcdGtkVre1ueNI2A0tJ+rys8V7fbrT8+tFzLyayn1sae3h7JC4UpUKDAI3lijrH25qT9njEGxphezrwtOUl+P5gkAZ4zuvE8AS3PcSa16ITmbfQe0ZYnJWL1gJ3+PByjY7K9OYVqaG1nbjcQul8lZOyZ8RT6OuGVvAiwABhYy/9HjL2z8RQ2duCCS3Lp5Ubu5aHPv+DpS0I9hdZv9l5u1JNG6Fts7PSd5mxfn0JzsooOmyDV19dDURT07t07rLx37944evRozMccPXq0zev/+Mc/4ic/+Qn69+8Pt9sNWZbx17/+FZdddlmrbVmyZAkWLVoUVV5ZWYlOnToBADIyMpCbm4udO3eGvQU4dOhQZGVlYcuWLairqwMAzM9Lw1t7T2H7FwHMuLAL0ju7UFZWBgDIy8tDeno6KisrAQAVFRUAgMLCQqSkpOjXaRQXF6OpqUm/HjgzKAd3c+OH3+6sl2/YsAFFRUU4ePAgampq9PLt27cjPz8fe/fuxe7du/VyM54AYMA5Oej55Vbs+dYsNCefyeHrr79G7969UV5eHjY42/I0Py9NL/MFGZZ+eAL19fWorq7Wy1NTU1FYWIgjR45gx44denmvXr0MedI01tc2Y32tD1OyO4W1JycnBxkZGejUqZOeBXAmJyOe5uel4cHqBqQlybhtZCoAoKysDG63G5MnTw7zVFFRgdTU1Jg5xfMEABMHp2BknzM3lr1798bM6ciRIxg4cCA2bNiAxsbGdnkCgBFyEvyeNOzKul0vc7lccLvdqKuri8qpLU+heW876sfb/22KGnvnn38+CgsLsX379jBPOTk5yMzMbNOTprFyWyMafCrm56WF+dLGnpYFgJg5GfXU1vNpTv852NSwCZsaNuF7Pb8XNfaMeGorJwBobGw8c4/4+VzA50O3Rx9DMCMD39xwvf6YeJ76A/gifQyO9inSy88991y43W7U1NQYuke05ek/w+6C6ko+U8+uP2H06NFhz73QnCLve+3Jad++fQDO5K3l1FRUCP/w4fr1SVUtdZr1FJrTnP5z8OSRJ9EYbETJgJKYY+9sPH3u+xwBFkB+Wj4K0goAtNxrjNzLQ59/b+09BQBRnkaMGAG3242ysjJT93KjngoLC7Hx642Y03+OXr6vqSWfs3190nLasmULrEJifD8mtVUOHz6Mfv36oaqqCnl5eXr5/fffj+eeew67du2KeozX68UzzzyDm266SS97/vnnMWPGDDQ3NwMAHn74YfzlL3/Bww8/rA/qBQsW4LXXXsMVV1wRsy2xVpAGDBiAuro6pKW1DCozs9nL734xagVpw70tbdZmuX6/X1/x0f5aAIzP0C/99YtRK0gb7/tB2F8d2l+VHo/HkhWkQw9eHLaCxAD0mbcBSUlJcLlcpv/qyF/wXFi5XwH+tfTmsL86QtsU2kajK0iX3/0igPAVpI33nRk/LpcLkiShublZ/3+t3Ohqy+V3vxi1gqTlrf0lFQwG9bxlWTa9MvGd+c9HrSB9cP8Pw3LS8tbyONsVpMNLRiJ0BUnLOyUlBYwx03/FX7bweb1cW0Ha8uAPo3KNdUsyutqi5R26gqRloXlljKG5uVl/7oXmZNUKkqIo+O5fvgs/84etIFXMOjMRONsVJC3v5ORkSJKkl2+aVNzSzpAVpIJ3ytr0dPiB3LAVJAag7y8+QFJSElRVtWQFad99F4V5klQ/+s7/t96G1rxqtCenYDAIv9+v5621fV3RuKgVpLGVa9u1gsQY0/MOXUEqn1VumSfGGCb/ZTJOqafCVpDKZ5UbupeP/s0LenlQBf61bHqYJ+0e4vV6Db8OmfXkcrlwxWNXQMWZNjIwrJu7zrIVpOPHj6Nnz55oaGhA165dcTZ02B6knj17wuVyRa0WHTt2LGqVSKNPnz5xr29qasKvf/1rLF++HN/97ncxYsQIzJ07F1OnTsXDDz/caluSkpLQtWvXsH9AS6Da5EJ7K8blculloeVut1sv8ystLwAAEFBbXvi132lvl0iShIqKCl1DkqQwPe1fa+VAyw3Mr5z5pw1WWZb1x4b+ZdZa24148ng8kFnLYHepfrhUHyS0/FWmDV6jbdfKQ9vuP/080toe6rOsrAyqqoaVa17b8qTVrYTkEebp9I2vvLw8rK1aTkY8aW0PzSO0/bIsh+UdmZNRT0E1vL8ic9I0tDyi8jPhyePxQELLNNil+sLyDgaDrbY9nqfQtmtL/JFeGWP6X6mx2t6WJ61+7bYZmoXmVVGUsOdeaE5mPbU29iRJ0v9KVnD6LRIWaJen1nLS8lYUJbzc74fk97fcZxiD5Pcb9iQzJSzvd999F8Fg0Pg9og1PWt3aPyZ78c4770SNv3j3PbM5McbC8tbaLgWDel9Jfj+k0y/mZj1p7dXy1t4y8jO/pZ4kScLc/nNb3raFAj/z6xpG7uWhzz/t9SnUkyRJWLNmTdgf1W3eI0x6CgaDuH3A7Xr/aBPKtp5PZsaelXTYBMnr9WLUqFFRS6sVFRXIz8+P+Zi8vLyo68vLy/XrtfeCIzvK5XKFzUAJgiAIgiDi0aEfFDlv3jxMmzYNF110EfLy8vDEE0+gtrYWs2fPBgBMnz4d/fr1w5IlSwAAd9xxBy6//HIsXboU11xzDd544w289957+OCDDwAAXbt2xZgxY3DXXXchJSUFmZmZWL9+PZ599lksX768w3wSBEEQBOEsOnSCNHXqVBw/fhyLFy/GkSNH8O1vfxtlZWXIzMwEANTW1oatBuXn5+Mf//gHfvOb3+Duu+/Geeedh1WrVuGSSy7Rr/nHP/6BBQsW4Ic//CG+/PJLZGZm4v7779cnXQRBEARBEG3R4V81MmfOHMyZMyfm79atWxdVdsMNN+CGG25otb4+ffrgqaeesqRt2vvAPHC73SguLna0hqz6HO+BNIxDeZvTKD1YCj/zc9Vwet4ijSk78uapIVI/WUWHf9XI/zLacWPnakgCeCAN41DeZkh1p0KC1PaFZ4Hz8xZnTNmRN28NUfrJKmiCFAeeH7seDAZRWVnpaA1V9jreA2kYh/I2pzHj3BnwSB6uGk7PW6QxZUfePDVE6ieroAkSQRAEQRBEBDRBIgiCIAiCiIAmSB0Iz81wdmmI4IE0Eqd+kTR8qq/ti84SyjtxNOzIm7eGKP1kFR1+ii2R0T4RlFfdkydP5la/HRou1ed4D6RhHMrbnMYjn7f+BdlWaTg9b5HGlB1589QQqZ+sglaQ4sDz07dVVcWxY8ccrcEgO94DaRiH8janMTB5INfTOiLkLdKYsiNvnhoi9ZNV0AQpDrG+NNXKuqurqx2tocoex3sgDeNQ3uY0pqRP4XpaR4S8RRpTduTNU0OkfrIKmiARBEEQBEFEQBMkgiAIgiCICGiCFAdJ4vc+qSRJSE1NdbgGE8ADaRiH8jajUe+vBwPjquH0vEUaU3bkzVNDpH6yCjrFFgfe30lTVFTErX47NFyq3/EeSMM4lLc5jb8d/Rt3DafnLdKYsiNvnhoi9ZNV0ApSHHjv5j9w4ICjNVTJ5XgPpGGifsrblMaIziMgc7zFCpG3QGPKjrx5aojUT1ZBE6Q48N7NX1NT42gNJrkd74E0jEN5m9OY2GMi3BK/VWgR8hZpTNmRN08NkfrJKmiCRBAEQRAEEQFNkAiCIAiCICKgCVIceO/m79Wrl8M1VAE8kIZxKG8zGvua9nE/reP0vEUaU3bkzVNDpH6yCjrFFgfep9jy8/O51W+HhksNON4DaRiH8jan8XLdy9w1nJ63SGPKjrx5aojUT1ZBK0hx4L1ZbdeuXY7WUCWX4z2QhnEob3MaBWkFcMHFVcPpeYs0puzIm6eGSP1kFTRBigPv4467d+92tAaT3I73QBrGobzNaRSkFcAl8XshECFvkcaUHXnz1BCpn6yCJkgEQRAEQRAR0ASJIAiCIAgiApogxUGW+XWPLMvIyMhwtIbEFMd7IA3jUN7mNHae3AkV/N6uECFvkcaUHXnz1BCpn6yCTrHFweXi9z6py+VCbm4ut/rt0JBZ0PEeSMM4lLc5jTVfruGu4fS8RRpTduTNU0OkfrIKWkGKA+/d/Nu3b3e0hiq5He+BNIxDeZvTmNid/1cqOD1vkcaUHXnz1BCpn6yCJkhx4L2bv7a21tEaTHI53gNpGIfyNqcxogv/L+V0et4ijSk78uapIVI/WQVNkAiCIAiCICKgCRJBEARBEEQENEGKA+/d/EOHDnW0hsSCjvdAGsahvM1pbGrYBIXx288hQt4ijSk78uapIVI/WQWdYosD71NsWVlZ3Oq3Q0NmiuM9kIZxKG9zGpsaNnHXcHreIo0pO/LmqSFSP1kFrSDFIRgMcq27qqrK0RqK7HG8B9IwDuVtTuPGXjfCI3m4ajg9b5HGlB1589QQqZ+sgiZIcWCMca27rq7O4RqyAB5IwziUtxmNQSmDIEHiquH0vEUaU3bkzVNDpH6yCpogEQRBEARBREATJIIgCIIgiAhoghQH3pu0c3JyHK0hsaDjPZCGcShvcxprjq9BkPHbzyFC3iKNKTvy5qkhUj9ZBZ1iiwPv446ZmZnc6rdDQ2YKMhzugTRM1E95m9LY+c1O7hqOz1ugMWVH3jw1ROony+qyrCYB4b2bf+3atY7WUGSv4z2QhnEob3Mat/a5lftpHafnLdKYsiNvnhoi9ZNV0AQpDrx38zc2NjpcQxLAA2kYh/I2o9HT25P7aR2n5y3SmLIjb54aIvWTVdAEiSAIgiAIIgKaIBEEQRAEQURAE6Q48N7Nn5eX52gNWQ043gNpGIfyNqfx0rGXEGABrhpOz1ukMWVH3jw1ROonq6BTbHHgfYotPT2dW/12aEhQHe+BNIxDeZvT2N+8n7uG0/MWaUzZkTdPDZH6ybK6LKtJQAIBfrPcQCCA1atXO1pDkZMc74E0jEN5m9O4o/8d8EperhpOz1ukMWVH3jw1ROonq6AJUgfC8zilXRoieCCNxKlfJI0kOYm7BuWdOBp25M1bQ5R+sgqaIBEEQRAEQUTQrgnSM888g9WrV+s///KXv0S3bt2Qn5+PAwcOWNY4giAIgiCIjqBdE6QHHngAKSkpAIDq6mo8+uijWLZsGXr27Ik777zT0gZ2JG43vz3sbrcbhYWFjtaQVb/jPZCGcShvcxpPHnmS62kdEfIWaUzZkTdPDZH6ybK62vOggwcPYsiQIQCA119/HTfccAN++tOfoqCgAGPHjrWscaKjTTKdq8EE8EAaxqG8zdAYbAQDz0+hFiFvccaUHXnz1hCln6yiXStIXbp0wfHjxwEA5eXluOKKKwAAycnJaGpqsq51HQzv76QpKytztIYqJzneA2kYh/I2p1EyoITraR0R8hZpTNmRN08NkfrJKtq1gjR+/HjMnDkTubm52LNnDyZPngwA+PjjjzFw4EDLGkcQBEEQBNERtGsF6bHHHkNeXh7q6urwyiuvoEePHgCArVu34qabbrK0gQRBEARBEHbTrglSt27d8Oijj+KNN97AxIkT9fJFixZh4cKFpupasWIFBg0ahOTkZIwaNQobN26Me/0rr7yCYcOGISkpCcOGDcNrr70Wdc2nn36Kq6++GmlpaUhNTcWll16K2tpaU+0iCIIgCOJ/l3ZNkNasWYMPPvhA//mxxx5DTk4OfvCDH+Crr74yXM+qVatQUlKChQsXYvv27Rg9ejQmTZrU6mSmuroaU6dOxbRp07Bjxw5MmzYNU6ZMwebNm/VrPvvsM1x22WXIysrCunXrsGPHDtx9991ITk427ZP3bv7i4mJHa8iqz/EeSMM4lLc5jdKDpfAzP1cNp+ct0piyI2+eGiL1k1W0a4J011134cSJEwCAjz76CP/v//0/FBcX4//+7/8wb948w/UsX74cM2bMwMyZM5GdnY3S0lIMGDAAK1eujHl9aWkpxo8fjwULFiArKwsLFizAuHHjUFpaql+zcOFCFBcXY9myZcjNzcXgwYMxefJk7t8x0x7s2NDOV0MSwANpGIfyNkOqOxUSJK4azs9bnDFlR968NUTpJ6to1wRp3759GDZsGICWt7yuuuoqPPDAA1ixYgXeeecdQ3X4/X5s3boVEyZMCCufMGECqqqqYj6muro66vorr7xSv15VVaxevRrnn38+rrzySqSnp+OSSy7B66+/HrctPp8PJ06cCPsHtAyWQCCAQCAARVEAAIqi6GWh5cFgUC/zugD5dP4eGfC6oP9OVVW97srKSl2DMQbGWFjd8coBQEJL3do/bfe+qqoIBAK6RnNzc9y2G/EUCASgSi3fkqzIXihyEgLuLqisrITf3/LXgNG2a+Whbfee/gJmre3av+bmZl0jtFzz2pYnrW5XSB5hnlQVwWAwLIvQnIx40toemkdoTqqqhuUdmZNRT245vL8ic9I0tDyi8jPhKRAIgAFgkKDISWF5B4PBVtsez1No292n7zyRXn0+nz5mY7W9LU9a/drtNzQLzWsgEIjKO17bzeakKAqampow49wZ6Cx3hgstA8QjedrlqbWctLxDfQUCATCvF8zbckqISRKY12vYkyq5YuZt+B7Rhiet7tB/kVm0dd8zm5M2pjQNre3M7db7inm9YKe/Bd6sJy0LLW/txd8reS31pGl4JA9ccMEreXUNI/fy0Oef9voU6kkbT5E+494jTHoKBoN6P2nt90ieNp9PZsae9npnBe1ai/J6vTh16hQA4L333sP06dMBAN27d9cnF21RX18PRVHQu3fvsPLevXvj6NGjMR9z9OjRuNcfO3YMJ0+exIMPPoj77rsPS5cuxZo1a3DdddehsrISY8aMiVnvkiVLsGjRoqjyyspKdOrUCQCQkZGB3Nxc7Ny5M+wtwKFDhyIrKwtbtmxBXV0dAGB+Xhre2nsK278IYMaFXZDe2YWysjIAQF5eHtLT01FZWQkAqKioAAAUFhYiJSVFv06juLhYH7ga2hLi4G5u/PDbnfXyDRs2oKioCAcPHkRNTY1evm3bNhQUFGDv3r3YvXu3Xm7GEwAMOCcHPb/cij3fmoXm5DM5HD9+HH379kV5eXnYEcu2PM3PS9PLfEGGpR+eQH19Paqrq/XyLl26AAAOHTqEjz76SC/v1asX8vPz2/Skaayvbcb6Wh+mZHcKa09OTg769u0blgVwJicjnubnpeHB6gakJcm4bWQqAKCsrAxutxuTJ08O81RRUYHU1NSYOcXzBAATB6dgZJ8zx2P37t0bM6dDhw5h8ODB2LBhAxobG9vlCQBGyEnwe9KwK+t2RBKZkxFPoXlvO+rH2/9tihp72uerbdu2DfX19Xp5Tk4OMjMz2/Skaazc1ogGn4r5eWlhvoqLi3Hy5EkAZ/KOlZNRT209n+b0n4NNDZuwqWETvtfze1Fjz4gnjdZyCgaD+h8SAICfzwV8PnR79DEEMzLwzQ3X64+J56k/gC/Sx+BonyJEYvQe0Zan/wy7C6rrzHaH8/c8HpaFRmv3vfbk9Nlnn4VpaDk1FRXCP3y4fn1SVUudZj2F5jSn/xw8eeRJNAYbUTKgJGrsna0njUvTLkVBWgGAlnuNkXt56PPvrb0tr9+RnjTM3suNeho9erTeTxr7mvYBwFm/Pmk5hW65OVskxpjpT2y6+uqr4ff7UVBQgHvvvRf79u1Dv379UF5ejrlz52LPnj1t1nH48GH069cPVVVVyMvL08vvv/9+PPfcc9i1a1fUY7xeL5555pmwk3LPP/88ZsyYgebmZr3Om266CS+88EJYezt37owXX3wxZlt8Ph98Pp/+84kTJzBgwAAcOXJEP6EnyzJcLhcURdFnqqHlwWAQWldefveLCKqAylpWLCQJ2HBvS5tdLhdkWcapU6dQUVGB8ePHw+Px6JOeyM9waK380l+/CAmAx3WmbON9P4Db7Yaqqvqsu6KiAhMmTEBKSkqrbTfiCQAOPXgxZKZAkb0AJCiyFx9f8CtMnDgRSUlJ+l8ObbVdK89f8FxYuV8B/rX0Zv0vBO2x5eXlmDhxImT5zIKnJElwu91terr87pbMFRVQTuex8b4z40e7tqysTM9CK5dl2ZCny+9+EX4FYXloeXs8HqiqiubmZj1vr9cblpMRT9+Z/zzc8pm//ADgg/t/GJaTlreWR2R+ZjwBwBT1vkcAACAASURBVOElIwFIUOWWSZmWd3Fxsd5vkW2P5+myhc/r5SoDgiqw5cEfhnlVFAXvvvsuJkyYELaXQGt7W560vAMKwNDy17KWheY1EAjgnXfeCctby8msp9bGns/nQ0VFBVZ8vgJNahMUKPBIHlTMOjMRMOqptZy0vCdNmgSPx6OXb5pU3NJOvx9MkgCPBwXvlLXp6fADuVAlF5jUohOatyzLhu4RbXnad99FYZ4YJHw0/DdhWcTyqtGenJqbm/Huu+/qGlrb1xWNA0LuKVAUjK1ca9qTtoKh5f2N+g0YGLySF+Wzyi3zpGmUHiyFwhS4Tq/ol88qN3QvH/2bM6+JQRX417LpYZ60+ouLixGJ2den1jwxxlBWVoYVn6/Q9yExMKybu+6sX5+0PL744gv06dMHDQ0N6Nq1a5QXM7RrBenRRx/FnDlz8M9//hMrV65Ev379AADvvPNO2Km2ePTs2RMulytqtejYsWNRq0Qaffr0iXt9z5494Xa79bf/NLKzs8M2lUeSlJSEpKTobxj2eDxhT1qgJQSXyxV1bejN3H9mTCCgnqkrsm632x2lEXldvHIWoaW1QZZlfTLhdrv18tbabsQTAMisRcyl+sOu0bTMtB0Rbdc1QtoeqRGrnrY8RWoE1Oj2KIoSM4t4bQ8t1zRC8wj9vdZ2TSNWTkY8BdXo64DwnELzaG2zotGcpNOuXOqZPx7aans8T7HyjuVVG7Ox2tmWp0gNvxLD1+kXn8i82+OptbHn8XjgU33wMz8UnH6BY4F2eYpX7na7IUkSJEnSyyX/meenxBjg90fV1apXpgDsTCe2de9ore2tlYeOJaDlLbfWnntA7D5oT06xNKRWPlDQrCetTi1v7VOi/Sy630OvN9L2yHLf6f5ToEA5nVNoffFyivX8i/Sk/Wz2Xm7UUyAQ0PspcqP22b4+tVXeHtpVU0ZGBt5+++2o8j/84Q+G6/B6vRg1ahQqKipw7bXX6uUVFRW45pprYj4mLy8PFRUVYd/3Vl5ejvz8fL3Oiy++OGyZDgD27NmDzMxMw23TaG0wWIHH49E/YNOpGi7V53gPpGEcytucxiOfP8Jdw+l5izSm7Mibp4ZI/WQV7dqkDbT81f3KK6/gvvvuw/33349XX301bDnNCPPmzcNf//pX/O1vf8Onn36KO++8E7W1tZg9ezYAYPr06ViwYIF+/R133IHy8nIsXboUu3btwtKlS/Hee++hpKREv+auu+7CqlWr8Je//AX//e9/8eijj+Ktt97CnDlzovTbInRZz2pUVcWxY8ccrcEgO94DaRiH8janMTB5INfTOiLkLdKYsiNvnhoi9ZNVtGuC9N///hfZ2dmYPn06Xn31Vfzzn//EtGnTcMEFF+gb4owwdepUlJaWYvHixcjJycGGDRtQVlamr/bU1tbiyJEj+vX5+fn4xz/+gaeeegojRozA008/jVWrVuGSSy7Rr7n22mvx+OOPY9myZRg+fDj++te/4pVXXsFll11m2qfZCZ/Zuqurqx2tocoex3sgDeNQ3uY0pqRP0U/o8NJwet4ijSk78uapIVI/WUW73mK7/fbbcd555+HDDz9E9+7dAbScZLr55ptx++23Y/Xq1YbrmjNnTqurO+vWrYsqu+GGG3DDDTfErfPWW2/FrbfeargNBEEQBEEQobRrgrR+/fqwyREA9OjRAw8++CAKCgosaxxBEARBEERH0K632JKSkmJ+dsLJkyfh9XpjPMKZSBK/90klSUJqaqrDNZgAHkjDOJS3GY16f71+oomXhtPzFmlM2ZE3Tw2R+skq2jVBuuqqq/DTn/4Umzdv1j9J88MPP8Ts2bNx9dVXW9a4job3d9IUFRU5WsOl+h3vgTSMQ3mb0/jb0b8hwAJtX3wWGk7PW6QxZUfePDVE6ieraNcE6Y9//CPOO+885OXlITk5GcnJycjPz8eQIUPCvhfN6fDezX/gwAFHa6iSy/EeSMNE/ZS3KY0RnUdAbv9BYUMajs9boDFlR948NUTqJ6toVyu7deuGN954A3v27ME///lPvPzyy9izZw9ee+01dOvWzbLGdTS8d/PX1NQ4WoNJbsd7IA3jUN7mNCb2mAi3xO+vcRHyFmlM2ZE3Tw2R+skqDLdy3rx5cX8feuJs+fLl7W4QQRAEQRBER2N4grR9+3ZD1/Hd0EcQBEEQBMEfwxOk0G/r/V+B927+Xr16OVxDFcADaRiH8jajsa9pH/fTOk7PW6QxZUfePDVE6ier4PdGoADw3s2vfYecUzVcasDxHkjDOJS3OY2X617mruH0vEUaU3bkzVNDpH6yCn5byQWA92a1Xbt2OVpDlVyO90AaxqG8zWkUpBXAhehvIbdSw+l5izSm7Mibp4ZI/WQVNEGKA+/jjrt373a0BpPcjvdAGsahvM1pFKQVwCXxeyEQIW+RxpQdefPUEKmfrIImSARBEARBEBHQBIkgCIIgCCICmiDFQZb5dY8sy8jIyHC0hsQUx3sgDeNQ3uY0dp7cCRX83q4QIW+RxpQdefPUEKmfrIJOscXB5eL3PqnL5UJubi63+u3QkFnQ8R5IwziUtzmNNV+u4a7h9LxFGlN25M1TQ6R+sgpaQYoD793827dvd7SGKrkd74E0jEN5m9OY2J3/Vyo4PW+RxpQdefPUEKmfrIImSHHgvZu/trbW0RpMcjneA2kYh/I2pzGiC/8v5XR63iKNKTvy5qkhUj9ZBU2QCIIgCIIgIqAJEkEQBEEQRAQ0QYoD7938Q4cOdbSGxIKO90AaxqG8zWlsatgEhfHbzyFC3iKNKTvy5qkhUj9ZBZ1iiwPvU2xZWVnc6rdDQ2aK4z2QhnEob3Mamxo2cddwet4ijSk78uapIVI/WQWtIMUhGAxyrbuqqsrRGorscbwH0jAO5W1O48ZeN8IjebhqOD1vkcaUHXnz1BCpn6yCJkhxYIxxrbuurs7hGrIAHkjDOJS3GY1BKYMgQeKq4fS8RRpTduTNU0OkfrIKmiARBEEQBEFEQBMkgiAIgiCICGiCFAfem7RzcnIcrSGxoOM9kIZxKG9zGmuOr0GQ8dvPIULeIo0pO/LmqSFSP1kFnWKLA+/jjpmZmdzqt0NDZgoyHO6BNEzUT3mb0tj5zU7uGo7PW6AxZUfePDVE6ifL6rKsJgHhvZt/7dq1jtZQZK/jPZCGcShvcxq39rmV+2kdp+ct0piyI2+eGiL1k1XQBCkOvHfzNzY2OlxDEsADaRiH8jaj0dPbk/tpHafnLdKYsiNvnhoi9ZNV0ASJIAiCIAgiApogEQRBEARBREATpDjw3s2fl5fnaA1ZDTjeA2kYh/I2p/HSsZcQYAGuGk7PW6QxZUfePDVE6ieroFNsceB9ii09PZ1b/XZoSFAd74E0jEN5m9PY37yfu4bT8xZpTNmRN08NkfrJsrosq0lAAgF+s9xAIIDVq1c7WkORkxzvgTSMQ3mb07ij/x3wSl6uGk7PW6QxZUfePDVE6ieroAlSB8LzOKVdGiJ4II3EqV8kjSQ5ibsG5Z04GnbkzVtDlH6yCpogEQRBEARBREATJIIgCIIgiAhoghQHt5vfHna3243CwkJHa8iq3/EeSMM4lLc5jSePPMn1tI4IeYs0puzIm6eGSP1kFTRB6kBSUlIcrsEE8EAaxqG8zdAYbAQDz0+hFiFvccaUHXnz1hCln6yCJkhx4P2dNGVlZY7WUOUkx3sgDeNQ3uY0SgaUcD2tI0LeIo0pO/LmqSFSP1kFTZAIgiAIgiAioAkSQRAEQRBEBDRBIgiCIAiCiIAmSHHgvZu/uLjY0Rqy6nO8B9IwDuVtTqP0YCn8zM9Vw+l5izSm7Mibp4ZI/WQVNEHqQJqamhyuIQnggTSMQ3mbIdWdCgkSVw3n5y3OmLIjb94aovSTVdAEKQ68d/NXVlY6WkOVvY73QBrGobzNacw4dwY8koerhtPzFmlM2ZE3Tw2R+skqaIJEEARBEAQRAU2QCIIgCIIgIqAJUgfCczOcXRoieCCNxKlfJA2f6uOuQXknjoYdefPWEKWfrIJ/bzgYj4ff+6QejweTJ0/mVr8dGi7V53gPpGEcytucxiOfP8Jdw+l5izSm7Mibp4ZI/WQVCbGCtGLFCgwaNAjJyckYNWoUNm7cGPf6V155BcOGDUNSUhKGDRuG1157rdVrZ82aBUmSUFpaarpdqqqafoyZuo8dO+ZoDQbZ8R5IwziUtzmNgckDuZ7WESFvkcaUHXnz1BCpn6yiwydIq1atQklJCRYuXIjt27dj9OjRmDRpEmpra2NeX11djalTp2LatGnYsWMHpk2bhilTpmDz5s1R177++uvYvHkz+vbt2662KYrSrscZrbu6utrRGqrscbwH0jAO5W1OY0r6FK6ndUTIW6QxZUfePDVE6ier6PAJ0vLlyzFjxgzMnDkT2dnZKC0txYABA7By5cqY15eWlmL8+PFYsGABsrKysGDBAowbNy5qhejQoUOYO3cunn/+ea5vlREEQRAEIR4dOkHy+/3YunUrJkyYEFY+YcIEVFVVxXxMdXV11PVXXnll2PWqqmLatGm46667cMEFF1jfcIIgCIIghKZDN2nX19dDURT07t07rLx37944evRozMccPXq0zeuXLl0Kt9uN22+/3VA7fD4ffL4zO+tPnDgBoOUDpwKBAABAlmW4XC4oihL2HqdWHgwGwRgDAHhdQFAFVAZ4ZECSoNfjcrkgyzKCwSC6dOmif6iVdnog8kOuWisHAAmAx3Xm52AwCLfbDVVVoSiKrqEoCjweT6ttN+IJAFTJBZkpUGQvAAmK7EGXLl30azSPbbVdK/e6worhV6C3XUNRFKSmpoIxFla/JElwu91tetI0FBVQTucRWo/L5YIkSWFZaOWyLBvy5HW1tD00D+1xHo8HqqqG5a21PdJrPE8A4JYBOeSte0VRwnLSNLQ8IvMz4wkAWh4pQZW9LXqn85YkqdW2x/MUmrfKWp4jkV5VVUVqamrUMnno8yaeJ00joLS03+sKz1vzGpm3lpNZT62NvWAwiOP+4/BIHihMgQIFHskTNfaMeIpsu9ZuLW8Aev4AwLwteUl+P5gkAZ4zuvE8AS3PcSa16ITmbfQe0ZYnRU4K88RiZBHLq0Z7clJVNUxDaztzuwE5ZI3g9GPNegoEAmF5B1gADAxeyRtz7LXXUzAYRL2/HgwMLrjgkloGeyAQMHQvD33+BU9fEupJG0+SJJm+lxv1JEkS6v31YW+xsdN3mrN9fdLHmIVvsSXEKTbtyanBGIsqM3r91q1b8cgjj2Dbtm1x6whlyZIlWLRoUVT52rVr0alTJwBARkYGcnNzsXPnzrD9UUOHDkVWVha2bNmCuro6AMD8vDS8tfcUtn8RwIwLuyC9swtlZWUAgLy8PKSnp2Pt2rUIBoMoLy8HABQWFiIlJUW/TqO4uBhNTU2orKzUy7RBObibGz/8dme9fMOGDSgqKsLBgwdRU1Ojl2/btg35+fnYu3cvdu/erZeb8QQAA87JQc8vt2LPt2ahOfn0JPXkSXz99ddIT09HeXl52BOlLU/z89L0Ml+QYemHJ1BfX4/q6mq9PDU1FUVFRThw4ECYp169ehnypGmsr23G+lofpmR3CmtPTk4OMjMzIUmSnkVoTkY8zc9Lw4PVDUhLknHbyFQAQFlZGdxuNyZPnhzmqby8XPcUmVM8TwAwcXAKRvbx6uV79+6NmdORI0eQmZmJDRs2oLGxsV2eAGCEnAS/Jw27ss78oeFubobb7caxY8di5hTPU2je/7+9M4+v4fr//2vulkQ2JSIJ2WyJff20EmoXQWttURoUn1bRCqWl+FFqa1VTVVQ/ii7Kx7ct/ZAiiN0HDbHvQjQSsWVf7jLn90fcK3Nzk9ybzNyY+3k/H4/7IOfOvF/nfd5nZt73zDkzp9O02HEj32Lf69atG44dOybwyRin8nwyaqw+nY3MQh4zwjwFfhn7Xk5OjineluJkrU9l9b2ampqYWHcijmYexdHMoxjgNcBi36tsnAAgOzv72TnivUlAYSGqr/wG+oAA5L422LRPWT7VBXDfuzPSfLo986lGDahUKpw5c8aqc0R5Pl1oMh280tlUHnplBTp16mT1ea8icUpKShLE2xin/G5doW3e3LS907Eim7b6ZIyTMd7rUtchW5+NaP9oi32vMj5lG7KhYzp08OyADp4dABSda6w5lxc//v5zPQ8ALPqkUqmwc+dOm87ltvj0V/ZfmFh3oqk8KT8JACp9fTLGydJ85IrCseIpmJ3RarWoVq0atm7dioEDB5rKJ0+ejMTERBw8eLDEPgEBAZgyZQqmTJliKvvyyy8RExODO3fuICYmBlOnTjX96gaKMlOFQgF/f3/cvn27hE1LI0j+/v5IT09H9erVAdiWzXaa80uJEaRDC94A8CzLLSwsREpKCurUqQOFQmFzht7+419KjCAd/nS44FcHz/NISUlB3bp1odFoKp2hpyz5h2AEieeU4N74BQEBAVCpVDb/6gif+aOgXGsATi19U/ALgDGG1NRU1KlTR1AXa0eQOs35BYBwBOnwp2+Ytlcqixrw9u3bplgYy60dbek055cSI0jGeBt/Sel0OlO8lUqlzSMTL874ucQI0pGFIwRxMsbbGI/KjiDdW9wGxUeQjPEOCgoqalMbR1s6zvrZVG4cQTq5ZESJVSf37t2Dr6+v4AeOtaMtxngXH0EyxsLoK8/zuHPnjiDeUowgfbDpA1zOvQwd05lGkOLeibPZp+J1B57FyRjvwMBAkx0AONq7T1E9i40gdfgztlyf7i1qLRhB4jklFMM3IzAw0NS/zH21dbQl6dN2Ap/ADOCifoevr6/gnC3mCJJOp8Pdu3dN8TbW/UC37iVGkLrE76/QCBLP86Z4F7JC0wjSnnee/eiqrE88z+ODnz/A2dyz4MCZRpD2vLPHqnP5y7M3mcr1PHDqs5ECn3ieR2pqKgICAkqMwog1gqRQKDBpwyRczbsKHkX1ZGA4MOmAaCNIDx8+RK1atZCZmQkPDw9UhiodQdJoNGjbti3i4uIECVJcXBz69+9vcZ+wsDDExcUJEqQ9e/YgPDwcABAVFYUePXoI9unVqxeioqLw1ltvWbTp5OQEJyenEuUKhaLEBG+lUmm6qBan+AO2tMX6lu5pXM3tKBQKnD9/Hv7+/oLvSptQbqmcmWkZ62A8Ceh0OpNGWXW3xicAULAiMSX/9E3MCiecK2bflrrDrO4mjad1N6LT6ZCYmAg/Pz+LdsrzyVxDx5esT/F2Mv/OGp+MGsXjUfx7o09GDfM4WeuTni+5HfAsTubxLu2hb9bGiXvqldL4YLdi8Var1RbrXpZPluJt7mt58S7PJ3MNraGkXzzPW4x3aXW3NU5KpRI8z6NnjZ64mHsRBhRVSsd0FfKptPLi8VYqlaZyTvvsTekcY4BWa/H8Y9FXZgCeHudQOOHcuXOoW7cu1Gq1VeeI8sqVZg8JNCiccPbsWdSpU8eiv5bKbI0TAIvx5kp5Z5etPqmf3sI0xtt4y0jLSra7cXtr6168XKfToWfNnriYdxFapoXhaZyK2yvrXG7p+Cvuk06nKzMWpdXdFp90Oh161eyFy3mXoWVawXeVvT4ZsfbOkTVU+S22qVOnIioqCu3atUNYWBjWrl2L5ORkjB8/HgAwcuRI1KlTB4sXLwZQNLrUqVMnLF26FP3798f27duxd+9eHDlyBABQs2ZN1KxZU6ChVqvh4+ODkJAQ+zpHEARBEIQsqfIEaejQoXj06BHmz5+P1NRUNGvWDLGxsQgMDAQAJCcnC7LQ8PBwbN68GbNnz8acOXNQv359bNmyBS+99FJVuUAQBEEQhINR5QkSAEyYMAETJkyw+N2BAwdKlL322mt47bXXrLZvad6RNYg5VGfJdq1atWSuwTuAD6RhPRRvWzSS8pNMt1uk0pB7vB2pT9kj3lJqOFI7icVzkSA9r0j54j6VSmWaNyVXDSWvk70PpGE9FG/bNLY+2Cq5htzj7Uh9yh7xllLDkdpJLKr8SdrPM1I/cv3KlSuy1uA5pex9IA3roXjbptHBswOUKDm5VEwNucfbkfqUPeItpYYjtZNYUIJUBlK/tO/q1auy1mCcSvY+kIb1ULxt0+jg2cG0FFsqDbnH25H6lD3iLaWGI7WTWFCCRBAEQRAEYQYlSARBEARBEGZQglQGpT10TCzbAQEBstbgmEH2PpCG9VC8bdM4l3PO9LRgqTTkHm9H6lP2iLeUGo7UTmJBq9jKwNLTO8W03bp1a8ns20NDwfSy94E0rIfibZvGrse7JNeQe7wdqU/ZI95SajhSO4kFjSCVgdSz+c+cOSNrDZ5Tyd4H0rAeirdtGpE1IqHipPsN6gjxdqQ+ZY94S6nhSO0kFpQglYHUs/mTk5NlrcE4pex9IA3roXjbptHCrQUUEp5iHSHejtSn7BFvKTUcqZ3EghIkgiAIgiAIMyhBIgiCIAiCMIMSpDKQejZ/SEiIrDU4ppe9D6RhPRRv2zSOZh6FgUk3n8MR4u1Ifcoe8ZZSw5HaSSxoFVsZSL2KLTQ0VDL79tBQMIPsfSAN66F426ZxNPOo5Bpyj7cj9Sl7xFtKDUdqJ7GgEaQy0Ov1kto+duyYrDUMCrXsfSAN66F426bxeq3XoebUkmrIPd6O1KfsEW8pNRypncSCEqQyYIxJavvBgwcy11A4gA+kYT0Ub1s0gl2CwYGTVEPu8XakPmWPeEup4UjtJBaUIBEEQRAEQZhBCRJBEARBEIQZlCCVgdSTtFu1aiVrDY7pZe8DaVgPxds2jV2PdkHPpJvP4QjxdqQ+ZY94S6nhSO0kFrSKrQykXu4YGBgomX17aCiYAQEy94E0bLBP8bZJ41zuOck1ZB9vB+pT9oi3lBqO1E6i2RLNkgMi9Wz+/fv3y1rDoNDI3gfSsB6Kt20aY3zGSL5aR+7xdqQ+ZY94S6nhSO0kFpQglYHUs/mzs7NlrsE5gA+kYT0Ub1s0vDRekq/WkXu8HalP2SPeUmo4UjuJBd1iIwiCIGQLz/PQarWlfq/T6aBSqVBQUCB40zvn7V1i24KCggrVwahRy7kWdNBV2p4UGj7uGsHf5vuV1k5iIkY7qdVqSedJFYcSJIIgCEKWaLVaJCUllfkGd8YYfHx8cPfuXXDcs5EL94kTSmyblJRUoXoYNSbUmiAYwaioPSk0PuhRX/C3+X6ltZOYiNVO1atXh4+Pj2T1NEIJUhlIPZs/LCxM1hoKXid7H0jDeijetmn8O/3f0DFd+RtXQkPu8a6MBmMMqampUCqV8Pf3L3VyLmMMer0eKpVKcEHNtXArxjU42OZ6GDX+fvI3XHgXQXmwV8XsSaFhcH4i3M/3hRL2LbWTmFTWB8YY8vLykJ6eDgDw9fUtsQ2tYrMTUq9i87YwxCsnDQ687H0gDeuheNumcbvgtuQaco93ZTT0ej3y8vLg5+eHatWq2by/zsL53dnZuUJ1AQBDlgEKpdBmZeyJraFQCSdGi103a6lsO7m4FCVX6enp8Pb2LpEQ0So2O6HTSffrT6fTYefOnbLWMCicZO8DaVgPxds2jcl1J0PDacrfuBIaco93ZTSM82Q0mrLbmOd5ZGRklHkbrrLwPA9vjbekt3yk1pBTOxkTYkv9Rsz+SglSFSLlckp7aTiCD6Tx/Nh3JA0nhZPkGhRvSD4PxVoUnPSXU3toSI0YPtgr5vJvbYIgCIIgCJGhBIkgCIIgCMIMmqRdBiqVdM2jUqnQtWtXWWsoeK3sfSAN66F426axLnWdpKvYHCHeUmi0nf5DJfY+afMeCZ+PBMdxeKB9YNVDCkePHo2NGzcCKIqhv78/Bg0ahE8++QQPHjxAcLGVdB4eHmjcuDFmzZqFV155xWoNAAgKCkJ0dDSio6NLfHfy+FE0DRiAJ0+eoHr16gCKblu5u7ujTZs2GDBgAObNm2eVji3Y0k4VRcy+RCNIVYhxNr58NZgD+EAa1kPxtoVsfTYYpHwKtSPE2z59yh7wsH5yc2RkJFJTU3Hr1i18+umnWLVqFaZNm2b6fu/evUhNTcWJEyfw4osvYvDgwbhw4YJNGhXBHnN7pPZBTChBKgOp30kTGxsraw1e4SR7H0jDeijetmlE+0dLuorNEeJtDw17wBhDbU1tqxMMJycn+Pj4wN/fH8OHD8eIESOwbds20/c1a9aEj48PQkNDsXDhQuh0OsTHxws04vfEY3DEYDg7O6NevXr45JNPKtWOjDFkZWVVeH9rNWxpp4pA72IjCIIgCAfBxcWl1CXr3333HQDhraMj8Ufw4cQPETU2CpcuXcK3336LDRs2YOHChXar8/8CNAeJIAiCIKqIkydPYtOmTejevbupLDw8HAqFAvn5+eB5HkFBQRgyZIjp+zVfrcE/3/snBgwdgHq166FevXpYsGABPvzwQ8ydO7cq3HBIKEEiCIIgCDuyY8cOuLm5Qa/XQ6fToX///vj666+Rl5cHANiyZQtCQ0Nx7do1REdHY82aNahRo4bpFtils5dwIfECvo351nS7ymAwoKCgAHl5eRV6sjhREkqQykDqVS59+vSRtYaCL5S9D6RhPRRv2zRi7sZAy0p/y7wYGnKPtz007AHHcbivvW/16qyuXbti9erVUKvV8PPzg1pd9BqQ27dvAwD8/f3RsGFDNGzYEG5ubhg8eDAuXryIDC4DjDHwjMekaZPQs29P1PcSvoTWmtd2uLm5AQAyMzMFq9g8PDyQkZEBT09Pa123CVvbqSLQKjYHIT8/X+YanAP4QBrWQ/G2BXeVOzhIuypI/vG2T5+yBwobLqeurq5o0KABAgMDTclRaXTu3BnNmjXDwoULTRpNmjdB0s0kBAYHokGDBoKPNe8iCwyuD4VCgVOnTgnK7927h5SUFISEhFjti63Y0k5VjXxqWgVIvcolPj5e1hq8QiN7H0jDeijetmmM9R0LNVf2xa+yGnKPt0H15QAAIABJREFUtz007AFjDLU0tSRbnfXBBx9g7dq10D3UgeM4TJg6AX9s/QMrP1+Jixcv4vLly9iyZQtmz54t2C8lJQWJiYlITEzE5YvncfnieWRkPIGrmxveeecdfPDBB9i2bRuSkpJw5MgRDBs2DI0bN0ZERIQkfkjdToC41215j2sSBEEQRDESPh8p+JvneWRlZcHDw0MwupJ95UqJfd1DQyWvX0V45ZVXEBQUhC+++AIfLv4QHbt2xOofV2PV8lX4ftX3UKvVCA0Nxbhx4wT7LVu2DMuWLROUffrFCgx8/Q18+eWX8PX1xccff4zbt2/D29sbHTp0wOeffy77W55iQa1AEARBEHZiw4YNpX4XFBRkcX4Ox3G4dOkSsrKycF97HwDQsWtHdOzaEaG1LSd1xvlMRi7dfSj428nJCXPmzMGcOXMACBNJogi6xVaF2CNLl1rDEXwgjefHviNpFPKFkmtQvJ8feCb9E6LtoSE1cvLBMXqmRJQ3ea6ytvv27SuZfXtoKPlC2ftAGtZD8bZN46u/v5JcQ+7xtoeGQqEwrdSSUiNdmy5rDUdpJzGv2zSCVAY8L12my/M80tPTZa3BoJC9D6RhPRRv2zSCnIMkXcXmCPG2iwZj0Ol0ki4tZ4xBo5DutTL20HCUdhKzL1GCVAYGg0FS28ePH5e1Bq9Qy94H0rAeirdtGkO8h0i6is0R4m0PDcYYcnNzJb/w11DXkHR1ltQajtJOYvYlSpAIgiAIgiDMoASJIAiCIAjCDEqQykDKYUCO4+Du7i5zDeYAPpCG9VC8bdF4qH0IBuluVzhCvO3VpxQKheQaOl4nmX17aDhSO4kFrWIrA6nf1dStWzfJ7NtDQ8lrZe8DaVgPxds2je/TvpdcQ+7xtoeG8R1jUms80j2StYajtBO9i81OSL3K5c6dO7LW4Dml7H0gDRvsU7xt0mjh2kLS9045RLztoMEYQ2FhoeSTj12ULpLZt4eGo7QTrWKzE1KvcklMTJS1BuNUsveBNKyH4m2bRmTNSKg46UahHSHedtFgDPn5+ZJf+D1VnpKvYrNGIygoCDExMRWy7wjtJGZfoltsBEEQhMOQPL+5xfIsK/Z9UgG9gP933qbtR48ejYyMDGzbtq0CatKzYcMGjB071vS3t7c3XnzxRSxZsgRNmza12s7t27cRHByMM2fOoFWrViW+X/n5SuzbtQ+/7/tdUJ6RkYEXXngB8fHx6NKlS4X9EIPnYgRp1apVCA4OhrOzM9q2bYvDhw+Xuf2vv/6KJk2awMnJCU2aNMHvvz9rYJ1Oh48++gjNmzeHq6sr/Pz8MHLkSNy7d09qNwiCIAhC9nh4eCA1NRX37t3Dzp07kZubi759+0Kr1VZ11exKlSdIW7ZsQXR0NGbNmoUzZ87g5ZdfRu/evZGcnGxx++PHj2Po0KGIiorC2bNnERUVhSFDhuDEiRMAgLy8PJw+fRpz5szB6dOn8dtvv+HatWvo16+fzXWTejZ/rVq1ZK7BO4APpGE9FG9bNJLykyRfxSb3eNtHQ3o4jhPl3XuZmZl4++234e3tDQ8PD3Tr1g1nz541aVy9eRUTRk5Ax2Yd0bZeW7ze63Xs3bu3TJvr16+Hp6cnjh06gO3/twXhLRpBWyis6+DBgzF69GioVCpwHAeO4+Dj4wNfX1+0a9cOU6ZMwZ07d3D16lXTPowxfPbZZ6hXrx5cXFzQsmVL/N///V+ZdRGrncrTEIsqT5CWL1+OsWPHYty4cWjcuDFiYmLg7++P1atXW9w+JiYGPXv2xMyZMxEaGoqZM2eie/fupnuunp6eiIuLw5AhQxASEoL27dvj66+/RkJCQqlJV2lIvYotPDxc1hpKXid7H0jDeijetmlsfbAVOibdkmZHiLc9NOwBx3F4ontSqfk7jDH07dsXaWlpiI2NRUJCAtq0aYPu3bvj8ePH4DgOqZmp6NS9E77f+j1+3fsrOnTtgFdffbXUa9uyZcswbdo07N69G+GduqDXK/3A8wbEx+0ybfPw4UPs2LEDb731Ftzc3EokGBkZGdi0aRMA4XvOZs+ejfXr12P16tW4ePEipkyZgjfffBMHDx6UtJ3KQ8y+VKW9UqvVIiEhATNmzBCUR0RE4NixYxb3OX78OKZMmSIo69WrV5mT0jIzM8FxXKkv4issLERhsYw6K6vobnVBQQFcXIpm3CsUCiiVShgMBsEseWO5Xq83BV2jBPQ8wDNArQA4rujWHwAolUooFAoUFBTg5s2bqF+/PpRKpSmoer1eULfSygGAA6BWPvtbr9dDpVKB53kYDAYYDAbcvHkTDRo0gJOTU6l1t8YnoGjFiYIZYFBoAHDgOSUuXryIRo0aQa1Wm3wsr+7Gco1SUAytAaa6mzR5HklJSahfv77Qd46DSqUq1yejhoEHDE/jUbyeSqUSjDFcvXoV9erVg1KpNJUrFAqrfNIoi+pePB7G/dRqNXieh1arNcVbpVIJ4mSNTwCgUgCKYucug8EgiJMx3sZ4mMfPFp8APB374MA/fXeSMd6hoaHgOM5i3cvyqXi8eVZ0jJj7yhjDrVu3EBwcbPK7eN3L88mooTMU1V+jFMbbWJerV6+ajr3icbLVp9L6nk6nQyfPTjiVdQpapoUBBqg5dYm+Z41PpcXJGO+QkBBTPwAApimKF6fVgnEcUOzYLMsnoCjG7OnEcp5T4tKlSwgJCSn624pzRHk+GRROAp/ADLh8+TKCg4NNsbDkq5HicTK+N8yoU/z/9sTYLm5KN+Tyz17VwXGcoM2MbWysp/E7hUIBxhj27duH8+fPIy0tDc7OzuA4Dp9//jm2bduGf//733j77bfRrkU7hDYNNdmbMnMKDu85jO3bt+O9994TtMWMGTPwww8/4MCBA2jWrBmupDyCi4sL+vYfjN+3/oIp744FYww//fQT6tati06dOpkmaWdmZsLNzQ2MMeTl5QEA+vXrh0aNGoHneeTm5mL58uXYt28fwsLCwBhDUFAQDh8+jDVr1qBz586muvA8D57nTf4b28n4msLS2qW0uBbfXqfTlThnFxaKN0JVpQnSw4cPYTAYULt2bUF57dq1kZaWZnGftLQ0m7YvKCjAjBkzMHz48FKf8bB48WJ88sknJcr37duHatWqAQACAgLQunVrnDt3TpCth4SEIDQ0FCdPnsSDBw8AADPCPPGf63k4c1+HsS3d4O2qRGxsLAAgLCwM3t7e2LdvH/R6PW7cuAEA6Nq1K1xcXEzbGenTpw/y8/MRHx9vKjOePOpVV2FEM1dT+aFDh9CtWzfcvXsXiYmJpvKMjAx06NAB169fFwyR2uITAPi/0ApejxNwreE7KHB+GoMbN/DCCy/Az88Pe/bsEZzQyvNpRpinqaxQz7D0v1l4+PAhjh8/bip3c3NDTk4ONBoNzp9/NhmyVq1aCA8PL9cno8bB5AIcTC7EkMbVBPVp1aoV/Pz8cP36dVy/ft1UboyTNT7NCPPEkuOZ8HRS4N027gCA2NhYqFQq9O3bV+DTjRs34O7ubjFOZfkEAJH1XNDG59mLHq9fv24xTi4uLqhXrx4OHTqE7OzsCvkEAC0UTtCqPXEl9P1nhTduoFGjRnjy5IkgTtb4VDzep9O02HEjv0Tfa9CgAW7cuIFHjx7h4cOHgjgFBgaW65NRY/XpbGQW8pgR5inwq0+fPsjOzsaNGzdMx56lOFnrU1l9r71ne7T3bI+jmUdxNPMoBngNKNH3rPGpvDgFBwdDr9c/O0e8NwkoLET1ld9AHxCA3NcGm/Ypy6e6AO57d0aaT7HnEl2/joYNG+LChQtWnSPK8+lCk+nglc6m8kbX1uDatWu4du2awKfSznvF46RSqeDj44Pc3Fy4uLhAq9UiPz8f9sb4g9pN5QY3uOGB9gF48KitqW36Diia12O8qOv1etN31atXh16vx7Fjx5CTk4NatWoJ7Ofn5+Py5cvIysoCV8hh2eJlOBR3CPfT7sNgMCA/Px+3bt0ybcvzPJYtW4a8vDwcPXoUTZo0QU5ODnzdiq4b744bjW7duiElJQXu7u5Yt24dhg0bZooXY0UP7zxw4AD0ej2OHj2Kb775BqtWrTLV+fTp0ygoKEBERISgrlqtFi1atADwLLnNzc1FVlYWFAoF3NzcTO3kpnSDilOhuqpo4KKgoACFhYWmehgTnfz8fMHcJycnJ3AcB61Wi0OHDpl0jH2v+DFcWThWFSn3U+7du4c6derg2LFjCAsLM5UvXLgQP/74I65cuVJiH41Gg40bN+KNN94wlf38888YO3YsCgoKBNvqdDq8/vrrSE5OxoEDB0pNkCyNIPn7+yM1NRU1a9YEYNtoS6c5v5QYQTq0oKi+xiw3Ly8PcXFx6NmzJ9Rqtc0jSO0//qXECNLhT4cLfh3qdDrExcUhIiICLi4ulR5BSlnyD8EIkkGhwcWmHyEyMhJOTk42jyCFz/xRUK41AKeWvin4ZavX67Fnzx5ERkYKRhSsHUHqNOcXAMIRpMOfPus7xm1jY2NNsTCWWzva0mnOLyVGkIzxNv7iLSgoMMVbo9HYPDLx4oyfS4wgHVk4QhAnY7yN8ajsCNK9xW1QfATJGO8+ffqY2s287mX51HHWz6Zy4wjSySUjBL4aDAbs3r0bERERgqFya0dbjPEuPoJkjIXRV51Ohz///FMQb7FHkAoLCxEXF4dVf69CPp9vGkGKeyfOZp9Ki5Mx3r179zaNGALA0d59iupZbASpw5+x5fp0b1FrwQhS8XgrFApRRpCSPm0n8ImBw/nmswWxsOSrkeJxKigowN27dxEcHAwXFxfTSMPfn7aEPak7+ywYY8jOzka6Nh08K2onjuPQ0KuhaTvjyIdxFZtxcZFxBGnp0qVYuXIl9u/fb5oHZPSpevXqqFmzJt5++23E7YvD9LnTERgcCGcXZ3z4zofo3LkzvvrqKzDGEBwcjLCwMMTGxmLGjBmYOXMmGGO4kvLsAY2v9emON98YhoiICLz44ou4desW6tati+zsbPz222+YOnUqHj9+bNp+/vz52L9/Pw4cOAAAOHHiBMLDwxEfH486deoIYu3k5ITAwEAkJSWhXr16SEhIQKtWrUz+ZGVlIV2bju/XfI+f/vUT9v21DyHeISZfb9++jfr165tuMVoaQSosLMStW7fg7+8PZ+eihNvYx+7fvw8fHx9kZmZW+sGXVTqC5OXlBaVSWWL0Jz09vcQokREfHx+rttfpdBgyZAiSkpKwf//+MhvKyckJTk5OJcrVarXgoAWKglB8KNhI8ZO5tthjGHT8M1vmti1pmG9XVjkz0zLWQaFQCJIJY3lpdbfGJwBQsCIxJS9cyWDUsqXuMKt7cVvF61683JKd8nwy19DxJetjvFBYirc1Phk1isej+PfF6148GS7N19J80vMltwMsxMl4S66Ue/HWxunpQDaUFiZVlhWn0nyyFO+y+p6lepbnk7mG1mDBr6cXKvN4V8Sn0vqeqT5Pb68BgI7pKuRTeeXGi6mxnCv2a5tjDNBqS+xTqq/MALCSgbL2HFFeuXlfMt5ys3TsGctL1PFp3Q0Gg8l3AIL/25PiyWPxSfmMMYttbKyn+Q++tm3bIi0tDRqNBkFBQSX243kex48fx8BhA9GzT08ARaMzt2/fNi2HN/r/0ksv4f3330evXr2gUqkwffp0FB8KGTzsTaxf/x1SUlLQo0cPBAYGlnjAYvH6TZ06FV9++SW2b9+OgQMHolmzZnBycsLdu3dLXYpvrEvxvma6bQaG4PrBuJ96H+n30xHiHWJql4SEBCgUCjRs2FDQXpbsW+o3DjMHSaPRoG3btoiLi8PAgQNN5XFxcejfv7/FfcLCwhAXFyeYh7Rnzx6Eh4eb/jYmR9evX0d8fLxpFMhWLHVusVAoFAgICJC1BscMsveBNKyH4m2bxrmcc+Ah3ROiHSHe9tCwBxzHIc+QZ/WqxczMTMEtTgBo1KgRwsLCMGDAACxduhQhISG4d+8eYmNjMWDAALRt2xaBwYGI2xmHLj27gOM4rFi6otQnR4eFheHPP/9EZGQkVCoVer0WZfrulQGv4YuF8/Ddd9/hhx9+MPmg0WgsJiMeHh4YN24c5s6diwEDBsDd3R3Tpk3DlClTwPM8OnbsiKysLBw7dgxubm4YNWqUaV/zlW8PMh/Ap4EPOnTpgHoN6+GDdz7Al59/CT8/P5w7dw7Tpk3D+PHj4e7ublVbmiNmX6rypQNTp05FVFQU2rVrh7CwMKxduxbJyckYP348AGDkyJGoU6cOFi9eDACYPHkyOnXqhKVLl6J///7Yvn079u7diyNHjgAoGpJ97bXXcPr0aezYsQMGg8E04lSjRg1oNBrLFbGApV9MYqFUKtG6dWvJ7NtDQ8H0sveBNKyH4m2bxq7Hu8rfsJIaco+3FBrWPrgx28IUDvfQ0AppchyHLL01j6Is4sCBAyX8HjVqFGJjYzFr1iyMGTMGDx48gI+PDzp16oTatWuD4zhMmzcNs6bMwvBXh+OFGi9g7MSx4AtLT8I7dOiAnTt3ok+fPniQlY83x7wNAHBzd8fgwYOxc+dODBgwwORDtWrVSh2Fmzx5MlasWIGtW7diyJAhWLBgAby9vbF48WLcunUL1atXR5s2bfDxxx8L9hs2bFgJW3tP7kWdgDr415Z/IWZRDEaMGIH09HQEBgZi3Lhx+PDDD61uS3PEvG5XeYI0dOhQPHr0CPPnz0dqaiqaNWuG2NhYBAYGAgCSk5MFGWF4eDg2b96M2bNnY86cOahfvz62bNmCl156CQDw999/448//gCAEk/vtPXJnFK/iuDcuXNo0aKFZImY1Bo8p8KZM2dk7QNpWA/F2zaNyBqR2PtkL/Ss5ApUsTTkHm97aBhfoeHi4iLZLTjGGDxUHsgyZKG8QaQNGzZgw4YNpX6/YsUKrFixwqJG43qNseG3DQKNBTMXCLa7ffu24O9OnTohJycHl+4+FJSnpqZixIgRpuklxnYaNWoURo8eXUI/ICBAMC+O4zi8//77eP/990tsCxS98sR8ijNjDKmPU03tVMu7FhbGLERo7YolppYQ87r9XIxrTpgwAbdv30ZhYSESEhLQqVMn03cHDhwo0Zlee+01XLlyBVqtFpcvX8agQYNM3xmDYulj62PLpX6ZZXJysqw1GKeUvQ+kYT0Ub9s0WrhJ/7JaucfbLhqMQavVSv6OsWrKauAg3RwosTQyMp4g9o/fsX//fkycOFFg3xHaScy+VOUjSARBEARB2IfX+3RHVmaGaZ4TUTqUIBEEQRDE/whxx04DAJr4e1VxTZ5/notbbM8rUq9yCQkJkbUGx/Sy94E0rIfibZvG0cyjMFhYMi+mhtzjbRcNjjM9XFBKjRx9jqS3p6TWcJR2cqhVbM8zUq9iC63gionnRUPBDLL3gTSsh+Jtm8bRzKOSa8g93vbQ4DjO9MooKTVyDDmy1nCUdhLzuk0jSGVg6f1nYto+duyYrDUMCrXsfSAN66F426bxeq3XoeYsP+xRLA25x9seGowx5ORIO2rBGMML6hckHX2RWsNR2knMvkQJUhlI3VEePHggcw2FA/hAGtZD8bZFI9glWPJVTXKPt736lPmrT6TQcDJ/Ea/MNBypncSCEiSCIAiCIAgzKEEiCIIgCIIwgxKkMpB6knarVq1krcExvex9IA3roXjbprHr0S7JnqJt1JB7vO2i8XTysdSrszL1mVbd3hk9erTpBawqlQoBAQF499138eTJE6s1ft/8O15s9GKp23bp0gXR0dElyvftjkXTgFql2n+e2qmiONSrRp5npF4GbHydilw1FMyAAJn7QBo22Kd426RxLvec5Bqyj7cEGh2+7lDxneNs3+Xoe0fBcRzyDflW7xMZGYn169dDr9fj0qVLGDNmDDIyMvDLL7+Uuo+tGrZiXOYvJVL7AIh73aYRpDKQepXL/v37Za1hUGhk7wNpWA/F2zaNMT5jJF/FJvd420PDHjDGUFNd0+rRFycnJ/j4+KBu3bqIiIjA0KFDsWfPHtP3mZmZePvtt+Ht7Q0PDw9069YNiYmJNmlUxIesrCzJJ2lL6QNAq9jshtQdJTs7W+YanAP4QBrWQ/G2RcNL4yX5Kja5x9s+GtLDGINaUbFk+NatW9i1axfUarXJVt++fZGWlobY2FgkJCSgTZs26NGjB3IypXuGEGMMPM9LflxUtJ1s0RALusVGEARBEHZkx44dcHNzg8FgQEFBAQBg+fLlAID4+HicP38e6enppltey5Ytw7Zt27B9+3b0Ht67yur9vwYlSARBEARhR7p27YrVq1cjLy8P//rXv3Dt2jW89957AICEhATk5OSgZs2agn3y8/ORlJRUFdX9n4USpDKQepVLWFiYrDUUvE72PpCG9VC8bdP4d/q/oWM6STXkHm97aNgDjuPwWPfY6ts7rq6uaNCgAQBgxYoV6Nq1Kz755BMsWLAAPM/D19cXBw4cEOzDGEOmwboVYB4eHsjMzCxRnpWVCTd391J9cHV1lXwVmy3tVBFoFZudkHoVm7e3t2T27aHBgZe9D6RhPRRv2zRuF9yWXEPu8baHhj3gOA5aXlvh/efOnYvevXvj3XffRZs2bZCWlgaVSoWgoCDBdlfuX7HKXmhoKP78888S5RfOnkFQvQYW9+E4zjQPSioq207WQKvY7IROJ92vP51Oh507d8paw6Bwkr0PpGE9FG/bNCbXnQwNp5FUQ+7xtoeGPeB5Ht4a7wqPvnTp0gVNmzbFokWL0KNHD4SFhWHAgAHYvXs3bt++jWPHjmHWrFlIuZhi0jAYDLh84TISExNNn0uXLgEAJkyYgJs3b2LixIk4e/Ysrl27hk0b1+G3LZsw5p2JpfqQkZEBnucr1ghWUNl2sgYx+xKNIFUh9ljaKrWGI/hAGs+PfUfSkPqdUwDF+3lCwVVuvGHq1Kl466238NFHHyE2NhazZs3CmDFj8ODBA/j4+ODll19Gbe/apu3zcvMwqMcggY3AwEDcvn0bQUFBOHz4MGbNmoWIiAgUFBTAP6geFn7xNXq90r9S9awslW0ne0IJEkEQBOEwHH3vqOBvnueRlZUFDw8Pwe2X7Cslb1e5h4ZKXr8NGzZYLB8+fDiGDx9u+nvFihVYsWKF6W+jH/e19zFw2EAMHDYQABBa23Kd27Zti127dpn+vnT3oQi1/99CPqkcQRAEQRCEnaAEqQxUKukG2FQqFbp27SprDQWvlb0PpGE9FG/bNNalrpN0FZsjxNseGhzHwd3dXfLVWQ+0DyRdnSW1hqO0k5h9iRKkKsTFxUXmGswBfCAN66F420K2PhsM0j4hWv7xtk+fkvKib4SHdJOb7aXhKO0kFpQglYHU72qKjY2VtQavcJK9D6RhPRRv2zSi/aMlXcXmCPG2h4a93jFWW1Nb0gRDag1HaSd6FxtBEARBEISEUIJEEARBEARhBiVIBEEQBEEQZlCCVAZSr3Lp06ePrDUUfKHsfSAN66F426YRczcGWibdaxUcId720OA4Dh4eHpKvzrqvvS/5KjYpNRylnWgVm4OQn58vcw3OAXwgDeuheNuCu8odHKRdFST/eNunT0l5QTaisMPlVGoNR2knsZBPTasAqVe5xMfHy1qDV2hk7wNpWA/F2zaNsb5joeake/mnI8TbHhqMMWRnZ0u+OquWppbkq9ik1HCUdqJVbARBEAQhQ0aPHg2O40p8IiMjAQBBQUGIiYmxuO+RI0cQWjsUWZlZJb5r1aoV5s2bJ2XV/+egd7ERBEEQDsPBTp3tqtf50EGb94mMjMT69esFZU5O0r/cmLANSpCqECknJtpLwxF8II3nx74jaRTyhZJrULyfH3hm/ROinZyc4OPjI6nG84qcfHCMnikRarV08wfUajX69u0rmX17aCj5Qtn7QBrWQ/G2TeOrv7+SXEPu8baHhj1QKBRI16ZLrpGhy5DUfvXq1SWzb9SQup3EvG7THKQy4HnpMl2e55Geni5rDQaF7H0gDeuheNumEeQcJOkqNkeItz007AFjDBqF9a+V2bFjB9zc3ASfBQsWlKuhUkg3psEYg06nk3ySti3tVBHE7EuUIJWBwWCQ1Pbx48dlrcEr1LL3gTSsh+Jtm8YQ7yGSrmJzhHjbQ8MeMMZQQ13D6tVZXbt2RWJiouAzceLEcjU8VB5iVLdU+7m5uZInSLa0U0UQsy/RLTaCIAiCsCOurq5o0KCBzfu5u7sDALKzsuHhKUyWMjIy4OnpKUr9iCJoBIkgCIIgZEC9evWgUChwIfGCoDw1NRUpKSkICQmpopo5JjSCVAZSP3Ld3d1d5hrMAXwgDeuheNui8VD7EAzSvnpC7vG2j4b0cBwHHa+zevvCwkKkpaUJylQqFby8vAAAKSkpSExMFHzv7+8PZ1dnDB05FEvnLYVSpURok1A8uvEIs2bNQuPGjREREVEpHxQKheTHhS3tVFENsaAEqQykfldTt27dJLNvDw0lr5W9D6RhPRRv2zS+T/tecg25x9seGvaA4zg80j2yevtdu3bB19dXUBYSEoIrV64AAJYtW4Zly5YJvl+/fj3a926PGZ/MgJe3F2IWxSDl7xTU9q6Nrl27YvPmzZW6ZhnfxSYltrZTRRDzuk0JUhlIvcrl7t278Pf3h0IhzZ1OqTV4Tok7d+7I2gfSsME+xdsmjRauLXAh9wJ4SHMecYh4S6Bh/uBGxhi0Wi00Go1gdCH7aTJSHPfQ0AppMsbgonRBvqH898pt2LABGzZsKPX727dvl6qR/CgZcAImTJ2ACVMnAABCa1eszpbsW2onMbGlnSoKrWKzE1KvcklMTJS1BuNUsveBNKyH4m2bRmTNSKg46X6DOkK87aLBGPLz8yVfneWp8pT8XWxSajhKO4nZlyhBIgiCIAiCMIMSJIIgCIIgCDMoQSoDqWfz16pVS+YavAP4QBrWQ/G2RSMpP0nyVWxyj7e9+pRKpZJcQ+p4mgZoAAAbm0lEQVR370mt4UjtJBY0SbsMpF7FFh4eLpl9e2goeZ3sfSAN66F426ax9cFWyTXkHm97aHAcBzc3N8k1nuieyFrDUdpJzOs2jSCVgdSTOK9cuSJrDZ5Tyt4H0rAeirdtGh08O0AJpaQaco+3GBrlTSq21+RjN6W0yYXUGnJqp7LqSJO07YTUy/yvXr0qaw3GqWTvA2lYD8XbNo0Onh2g5KRLkBwh3pXRUCqL2lar1ZatwRgKCwulv/Cr3CRfxSalhpzaKS8vDwCgVpd816GY/ZVusREEQRCyQ6VSoVq1anjw4AHUanWpz1HieR5arRYFBQWCbbQWLqQFBQUVqotRg9fxggSjovak0OD1widYm+9XWjuJSWV9YIwhLy8P6enpqF69uilJlgpKkAiCIAjZwXEcfH19kZSUhDt37pS6nfHWkYuLi2DkouD+/RLbOldwZMOokW3IFlz4uWzxRnsqq5H+JEfwt7Igw6J983YSE7HaqXr16vDx8RG7eiWgBKkMpMqijbYDAgJkrcExg+x9IA3roXjbpnEu55xkT9E2asg93pXV0Gg0aNiwYZm32QwGA65evYrAwEDBiMPJOf+vxLaNf/qxQvUwGAz4NvZbHM88Dj30pvJf3vylQvak0Ji2dZvg718/HFDCvqV2EhMx2kmtVpdZPzH763ORIK1atQqff/45UlNT0bRpU8TExODll18udftff/0Vc+bMwc2bN1G/fn0sXLgQAwcONH3PGMMnn3yCtWvX4smTJ3jppZfwzTffoGnTpjbVS8rhO6VSidatW0tm3x4aCqaXvQ+kYT0Ub9s0dj3eJbmG3OMthoZCoYCzs3OZ27Rp06ZEGUtPL1FWnp2y+DXtV1Htia2Rli1MIi3tZ6mdxEbqdhLzul3lk7S3bNmC6OhozJo1C2fOnMHLL7+M3r17Izk52eL2x48fx9ChQxEVFYWzZ88iKioKQ4YMwYkTJ0zbfPbZZ1i+fDlWrlyJU6dOwcfHBz179kR2drZNdZN6lcuZM2dkrcFzKtn7QBrWQ/G2TSOyhvSvGpF7vB2pT9kj3lJqOFI7iUWVJ0jLly/H2LFjMW7cODRu3BgxMTHw9/fH6tWrLW4fExODnj17YubMmQgNDcXMmTPRvXt3xMTEACgaPYqJicGsWbMwaNAgNGvWDBs3bkReXh42bdpkU92kXuWSnJwsaw3GKWXvA2lYD8XbNo0Wbi2gkPAU6wjxdqQ+ZY94S6nhSO0kFlWaIGm1WiQkJCAiIkJQHhERgWPHjlnc5/jx4yW279Wrl2n7pKQkpKWlCbZxcnJC586dS7VJEARBEARRnCqdg/Tw4UMYDAbUrl1bUF67dm2kpaVZ3CctLa3M7Y3/WtqmtJUOhYWFKCx89vjzzMxMAMDjx49NZQqFAkqlEgaDQZChGsv1er1pVr5Cnw8DD/AMUCkAjgMePXoEoOj+qEKhQF5eHvLy8vDo0SOo1WrT0z/1+mcT1wCUWm4ozAcHQFXsduvjx4+hUqnA8zwMBgN0Oh3y8vLw+PFjuLi4lFp3a3wCgMxCQMEMMCg0ADgYnvqRkZEBJycn6HTCZaTl+aTQ5wvKdQYgIyNDMESq1+tNGsUn3xkfi1+eT0YNngcMT+NhjAUA07bFY2EsVygUVvmk0OdDZ4AgHkYNtVoNnudRUFBg0tBoNII4WeOToTDf1JeMPHnyRBAnY7yN8TCPny0+AUBWgQEAB16hAQBTvLOyskztZl73snwqHm/GAD1f5ENxX42xMPZl87qX55NRQ28AGAC1UhhvlUplaqfi8TbGyVafSut7hYWFyMvLg6JAAcYzGGCAmlOX6HvW+FRanIx+ZGZmQq1Wm8pznh4nnFYLxnGA+pluWT5lFxjAc0qwp7c/isdboVBYdY4oz6cMrfCSw8CXiIUlX41UJE7Fjz3j4wCUSiVyihx5ZtxgQFZWls0+6XQ6UywUBQoYeAMYGDScpkTfq4xPRg19flE9jM/YevTokVXn8uLHn4FHCV+N9rOysmCOrden0nwyLtNXFCigYEVtz8CQlZVV6euTMR5PnhQ9qVuU5zmxKiQlJYUBYMeOHROUf/rppywkJMTiPmq1mm3atElQ9tNPPzEnJyfGGGNHjx5lANi9e/cE24wbN4716tXLos25c+cyFJ1P6UMf+tCHPvShj8w/N2/erGhqYqJKR5C8vLygVCpLjBalp6eXGAEy4uPjU+b2xmcjpKWlwdfX1yqbM2fOxNSpU01/Z2RkIDAwEMnJyfD09LTdMSvIysqCv78/7t69Cw8PD1lqOIIPpPH82CeN50vDEXwgjefHvr00MjMzERAQgBo1alTaVpUmSBqNBm3btkVcXJxgmX5cXBz69+9vcZ+wsDDExcVhypQpprI9e/aYXngYHBwMHx8fxMXFmZaParVaHDx4EEuXLrVo08nJCU5OTiXKPT09JQuiEQ8PD9lrOIIPpPH82CeN50vDEXwgjefHvr00xHgeUpU/B2nq1KmIiopCu3btEBYWhrVr1yI5ORnjx48HAIwcORJ16tTB4sWLAQCTJ09Gp06dsHTpUvTv3x/bt2/H3r17ceTIEQBF9zmjo6OxaNEiNGzYEA0bNsSiRYtQrVo1DB8+vMr8JAiCIAhCPlR5gjR06FA8evQI8+fPR2pqKpo1a4bY2FgEBgYCAJKTkwWZYHh4ODZv3ozZs2djzpw5qF+/PrZs2YKXXnrJtM2HH36I/Px8TJgwwfSgyD179sDd3d3u/hEEQRAEIT+U8+bNm1fVlfjHP/6B6OhozJkzB++8844pOQKA0aNHY8AA4SPRmzRpgkmTJmHOnDmYNGkSGjduLPie4zh06dIF06ZNw+zZs/HWW2/B29vbpjoplUp06dJFsJJGbBxBwxF8II3nxz5pPF8ajuADaTw/9uWmwTEmxlo4giAIgiAIx6HKn6RNEARBEATxvEEJEkEQBEEQhBmUIBEEQRAEQZhBCRJBEARBEIQZlCBZYNWqVQgODoazszPatm2Lw4cPi2b70KFDePXVV+Hn5weO47Bt2zbRbAPA4sWL8Y9//APu7u7w9vbGgAEDcPXqVVE1Vq9ejRYtWpge9hUWFoY///xTVA1zFi9ebHrGlVjMmzcPHMcJPsYnsYtFSkoK3nzzTdSsWRPVqlVDq1atkJCQIJr9oKCgEj5wHIeJEyeKpqHX6zF79mwEBwfDxcUF9erVw/z580V/63d2djaio6MRGBgIFxcXhIeH49SpUxW2V96xxhjDvHnz4OfnBxcXF3Tp0gUXL14UVeO3335Dr1694OXlBY7jkJiYKJp9nU6Hjz76CM2bN4erqyv8/PwwcuRI3Lt3T1Qf5s2bh9DQULi6uuKFF15Ajx49cOLECVE1ivPOO++A4zjExMSIqjF69OgSx0n79u1F9+Py5cvo168fPD094e7ujvbt2yM5OVkU+5aOdY7j8Pnnn4vmQ05ODiZNmoS6devCxcUFjRs3xurVq622b43G/fv3MXr0aPj5+aFatWqIjIzE9evXrbZvzXWusLAQ7733Hry8vODq6op+/frh77//tskPSpDM2LJlC6KjozFr1iycOXMGL7/8Mnr37m11By+P3NxctGzZEitXrhTFnjkHDx7ExIkT8d///hdxcXHQ6/WIiIhAbm6uaBp169bFkiVL8Ndff+Gvv/5Ct27d0L9/f5svLtZy6tQprF27Fi1atBDddtOmTZGammr6nD9/XjTbT548QYcOHaBWq/Hnn3/i0qVL+OKLL1C9enXRNE6dOiWof1xcHADg9ddfF01j6dKlWLNmDVauXInLly/js88+w+eff46vv/5aNA0AGDduHOLi4vDjjz/i/PnziIiIQI8ePZCSklIhe+Uda5999hmWL1+OlStX4tSpU/Dx8UHPnj2RnZ0tmkZubi46dOiAJUuWiO5DXl4eTp8+jTlz5uD06dP47bffcO3aNfTr1080DQBo1KgRVq5cifPnz+PIkSMICgpCREQEHjx4IJqGkW3btuHEiRPw8/OzyQdrNSIjIwXHS2xsrKgaN2/eRMeOHREaGooDBw7g7NmzmDNnDpydnUWxX7zuqamp+P7778FxHAYPHiyaD1OmTMGuXbvw008/4fLly5gyZQree+89bN++XRQNxhgGDBiAW7duYfv27Thz5gwCAwPRo0cPq69T1lznoqOj8fvvv2Pz5s04cuQIcnJy8MorrwheoFsulX6bm4Px4osvsvHjxwvKQkND2YwZM0TXAsB+//130e0WJz09nQFgBw8elFTnhRdeYP/6179Et5udnc0aNmzI4uLiWOfOndnkyZNFsz137lzWsmVL0eyZ89FHH7GOHTtKZt8SkydPZvXr12c8z4tms2/fvmzMmDGCskGDBrE333xTNI28vDymVCrZjh07BOUtW7Zks2bNqrR982ON53nm4+PDlixZYiorKChgnp6ebM2aNaJoFCcpKYkBYGfOnKmQ7fLsGzl58iQDwO7cuSOZRmZmJgPA9u7dK6rG33//zerUqcMuXLjAAgMD2Zdfflkh+6VpjBo1ivXv37/CNq3RGDp0qGjHhTWx6N+/P+vWrZuoGk2bNmXz588XlLVp04bNnj1bFI2rV68yAOzChQumMr1ez2rUqMG+++67CmmYX+cyMjKYWq1mmzdvNm2TkpLCFAoF27Vrl9V2aQSpGFqtFgkJCYiIiBCUR0RE4NixY1VUq8qRmZkJAKK8uM8SBoMBmzdvRm5uLsLCwkS3P3HiRPTt2xc9evQQ3TYAXL9+HX5+fggODsawYcNw69Yt0Wz/8ccfaNeuHV5//XV4e3ujdevW+O6770Szb45Wq8VPP/2EMWPGgOM40ex27NgR+/btw7Vr1wAAZ8+exZEjR9CnTx/RNPR6PQwGQ4lf2i4uLqbXCIlJUlIS0tLSBMe6k5MTOnfuLNtjHSg63jmOE3WUsjharRZr166Fp6cnWrZsKZpdnucRFRWF6dOno2nTpqLZNefAgQPw9vZGo0aN8M9//hPp6emi2eZ5Hjt37kSjRo3Qq1cveHt746WXXhJ9GoWR+/fvY+fOnRg7dqyodjt27Ig//vgDKSkpYIwhPj4e165dQ69evUSxX1hYCACCY12pVEKj0VT4WDe/ziUkJECn0wmObz8/PzRr1sym45sSpGI8fPgQBoMBtWvXFpTXrl0baWlpVVSrisMYw9SpU9GxY0c0a9ZMVNvnz5+Hm5sbnJycMH78ePz+++9o0qSJqBqbN2/G6dOnTe/hE5uXXnoJP/zwA3bv3o3vvvsOaWlpCA8Px6NHj0Sxf+vWLaxevRoNGzbE7t27MX78eLz//vv44YcfRLFvzrZt25CRkYHRo0eLavejjz7CG2+8gdDQUKjVarRu3RrR0dF44403RNNwd3dHWFgYFixYgHv37sFgMOCnn37CiRMnkJqaKpqOEePx7CjHOgAUFBRgxowZGD58uOgvAt2xYwfc3Nzg7OyML7/8EnFxcfDy8hLN/tKlS6FSqfD++++LZtOc3r174+eff8b+/fvxxRdf4NSpU+jWrZvpgl1Z0tPTkZOTgyVLliAyMhJ79uzBwIEDMWjQIBw8eFAUjeJs3LgR7u7uGDRokKh2V6xYgSZNmqBu3brQaDSIjIzEqlWr0LFjR1Hsh4aGIjAwEDNnzsSTJ0+g1WqxZMkSpKWlVehYt3SdS0tLg0ajwQsvvCDY1tbju8rfxfY8Yv7rmzEm6i9yezFp0iScO3dOkl/gISEhSExMREZGBn799VeMGjUKBw8eFC1Junv3LiZPnow9e/ZYff/eVnr37m36f/PmzREWFob69etj48aNmDp1aqXt8zyPdu3aYdGiRQCA1q1b4+LFi1i9ejVGjhxZafvmrFu3Dr17967Q/I2y2LJlC3766Sds2rQJTZs2RWJiIqKjo+Hn54dRo0aJpvPjjz9izJgxqFOnDpRKJdq0aYPhw4fj9OnTommY4yjHuk6nw7Bhw8DzPFatWiW6/a5duyIxMREPHz7Ed999hyFDhuDEiRM2v8LJEgkJCfjqq69w+vRpSdt+6NChpv83a9YM7dq1Q2BgIHbu3ClKkmFctNC/f39MmTIFANCqVSscO3YMa9asQefOnSutUZzvv/8eI0aMEP38uGLFCvz3v//FH3/8gcDAQBw6dAgTJkyAr6+vKCP5arUav/76K8aOHYsaNWpAqVSiR48egvOxLdhynbP1+KYRpGJ4eXlBqVSWyDDT09NL/NJ83nnvvffwxx9/ID4+HnXr1hXdvkajQYMGDdCuXTssXrwYLVu2xFdffSWa/YSEBKSnp6Nt27ZQqVRQqVQ4ePAgVqxYAZVKZdtEOytxdXVF8+bNbVpNURa+vr4lEsbGjRuLNuG/OHfu3MHevXsxbtw40W1Pnz4dM2bMwLBhw9C8eXNERUVhypQpoo/s1a9fHwcPHkROTg7u3r2LkydPQqfTITg4WFQdAKbVio5wrOt0OgwZMgRJSUmIi4sTffQIKDo2GjRogPbt22PdunVQqVRYt26dKLYPHz6M9PR0BAQEmI71O3fu4IMPPkBQUJAoGpbw9fVFYGCgaMe7l5cXVCqVXY75w4cP4+rVq6If7/n5+fj444+xfPlyvPrqq2jRogUmTZqEoUOHYtmyZaLptG3b1vQDOzU1Fbt27cKjR49sPtZLu875+PhAq9XiyZMngu1tPb4pQSqGRqNB27ZtTSuBjMTFxSE8PLyKamUbjDFMmjQJv/32G/bv3y/JxaU0XbGGqgGge/fuOH/+PBITE02fdu3aYcSIEUhMTIRSqRRNy0hhYSEuX74MX19fUex16NChxNLTa9euCV7GLBbr16+Ht7c3+vbtK7rtvLw8KBTCU4VSqRR9mb8RV1dX+Pr64smTJ9i9ezf69+8vukZwcDB8fHwEx7pWq8XBgwdlc6wDz5Kj69evY+/evahZs6ZddMU83qOionDu3DnBse7n54fp06dj9+7domhY4tGjR7h7965ox7tGo8E//vEPuxzz69atQ9u2bUWdBwYU9SedTme3493T0xO1atXC9evX8ddff1l9rJd3nWvbti3UarXg+E5NTcWFCxdsOr7pFpsZU6dORVRUFNq1a4ewsDCsXbsWycnJGD9+vCj2c3JycOPGDdPfSUlJSExMRI0aNRAQEFBp+xMnTsSmTZuwfft2uLu7m34he3p6wsXFpdL2AeDjjz9G79694e/vj+zsbGzevBkHDhzArl27RLEPFM1JMZ835erqipo1a4o2n2ratGl49dVXERAQgPT0dHz66afIysoS7bbRlClTEB4ejkWLFmHIkCE4efIk1q5di7Vr14pi3wjP81i/fj1GjRolyRuyX331VSxcuBABAQFo2rQpzpw5g+XLl2PMmDGi6uzevRuMMYSEhODGjRuYPn06QkJC8NZbb1XIXnnHWnR0NBYtWoSGDRuiYcOGWLRoEapVq4bhw4eLpvH48WMkJyebnk1kvHj6+PhY9cytsuz7+fnhtddew+nTp7Fjxw4YDAbT8V6jRg1oNJpK+1CzZk0sXLgQ/fr1g6+vLx49eoRVq1bh77//tulREuW1k3lip1ar4ePjg5CQEFE0atSogXnz5mHw4MHw9fXF7du38fHHH8PLywsDBw4UzY/p06dj6NCh6NSpE7p27Ypdu3bhP//5Dw4cOCCKfQDIysrC1q1b8cUXX1hdb1s0OnfujOnTp8PFxQWBgYE4ePAgfvjhByxfvlw0ja1bt6JWrVoICAjA+fPnMXnyZAwYMKDEAqnSKO865+npibFjx+KDDz5AzZo1UaNGDUybNg3Nmze37TZhhdbUOTjffPMNCwwMZBqNhrVp00bUJfLx8fEMQInPqFGjRLFvyTYAtn79elHsM8bYmDFjTO1Tq1Yt1r17d7Znzx7R7JeG2Mv8hw4dynx9fZlarWZ+fn5s0KBB7OLFi6LZZ4yx//znP6xZs2bMycmJhYaGsrVr14pqnzHGdu/ezQCwq1evim6bMcaysrLY5MmTWUBAAHN2dmb16tVjs2bNYoWFhaLqbNmyhdWrV49pNBrm4+PDJk6cyDIyMipsr7xjjed5NnfuXObj48OcnJxYp06d2Pnz50XVWL9+vcXv586dW2n7xkcHWPrEx8eL4kN+fj4bOHAg8/PzYxqNhvn6+rJ+/fqxkydPitpO5lRkmX9ZGnl5eSwiIoLVqlWLqdVqFhAQwEaNGsWSk5NF92PdunWsQYMGzNnZmbVs2ZJt27ZNVPvffvstc3FxqfCxUZ5GamoqGz16NPPz82POzs4sJCSEffHFFzY9OqQ8ja+++orVrVvXFIvZs2fbdD6x5jqXn5/PJk2axGrUqMFcXFzYK6+8YnO8uadiBEEQBEEQxFNoDhJBEARBEIQZlCARBEEQBEGYQQkSQRAEQRCEGZQgEQRBEARBmEEJEkEQBEEQhBmUIBEEQRAEQZhBCRJBEARBEIQZlCARBEGYMXr0aAwYMKCqq0EQRBVCCRJBEARBEIQZlCARBOEwaLXaqq4CQRAOAiVIBEHIli5dumDSpEmYOnUqvLy80LNnTyxfvhzNmzeHq6sr/P39MWHCBOTk5Jj22bBhA6pXr47du3ejcePGcHNzQ2RkJFJTU0vVSUhIgLe3NxYuXGgPtwiCeA6gBIkgCFmzceNGqFQqHD16FN9++y0UCgVWrFiBCxcuYOPGjdi/fz8+/PBDwT55eXlYtmwZfvzxRxw6dAjJycmYNm2aRfsHDhxA9+7d8cknn2DWrFn2cIkgiOcAVVVXgCAIojI0aNAAn332menv0NBQ0/+Dg4OxYMECvPvuu1i1apWpXKfTYc2aNahfvz4AYNKkSZg/f34J29u3b0dUVBS+/fZbvPHGGxJ6QRDE8wYlSARByJp27doJ/o6Pj8eiRYtw6dIlZGVlQa/Xo6CgALm5uXB1dQUAVKtWzZQcAYCvry/S09MFdk6cOIEdO3Zg69atGDhwoPSOEATxXEG32AiCkDXGpAcA7ty5gz59+qBZs2b49ddfkZCQgG+++QZA0aiREbVaLbDBcRwYY4Ky+vXrIzQ0FN9//z1N/iaI/0EoQSIIwmH466+/oNfr8cUXX6B9+/Zo1KgR7t27VyFbXl5e2L9/P27evImhQ4cKEiyCIBwfSpAIgnAY6tevD71ej6+//hq3bt3Cjz/+iDVr1lTYnre3N/bv348rV67gjTfegF6vF7G2BEE8z1CCRBCEw9CqVSssX74cS5cuRbNmzfDzzz9j8eLFlbLp4+OD/fv34/z58xgxYgQMBoNItSUI4nmGY+Y33gmCIAiCIP7HoREkgiAIgiAIMyhBIgiCIAiCMIMSJIIgCIIgCDMoQSIIgiAIgjCDEiSCIAiCIAgzKEEiCIIgCIIwgxIkgiAIgiAIMyhBIgiCIAiCMIMSJIIgCIIgCDMoQSIIgiAIgjCDEiSCIAiCIAgzKEEiCIIgCIIw4/8DLYpGH5wnQ14AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "\n",
    "for col in show.columns:\n",
    "    if 'param__' not in col:\n",
    "        continue\n",
    "    if any(x in col for x in ['lr', 'momentum', 'weight_decay']):\n",
    "        continue\n",
    "    fig, ax = plt.subplots()\n",
    "    s = show[show.loss < show.loss.median()]\n",
    "    sns.barplot(\n",
    "        data=s,\n",
    "        x='rank',\n",
    "        y='loss',\n",
    "        hue=col,\n",
    "        ax=ax, \n",
    "    )\n",
    "    ax.legend(loc='lower right')\n",
    "    ax.set_ylim(0.0, 0.12)\n",
    "    ax.set_xlim(0, 20)\n",
    "    ax.set_title(col.replace('param__', ''))\n",
    "    plt.grid(linestyle='--')\n",
    "    fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}

## b-autoencoder.py
import skorch.utils
from skorch import NeuralNetRegressor
import torch.nn as nn
import torch
import skorch
from distributed.utils import log_errors

def _initialize(method, layer, gain=1):
    weight = layer.weight.data
    _before = weight.data.clone()
    kwargs = {'gain': gain} if 'xavier' in str(method) else {}
    method(weight.data, **kwargs)
    assert torch.all(weight.data != _before)


class Autoencoder(nn.Module):
    def __init__(self, activation='ReLU', init='xavier_uniform_',
                 **kwargs):
        super().__init__()

        self.activation = activation
        self.init = init
        self._iters = 0

        init_method = getattr(torch.nn.init, init)
        act_layer = getattr(nn, activation)
        act_kwargs = {'inplace': True} if self.activation != 'PReLU' else {}

        gain = 1
        if self.activation in ['LeakyReLU', 'ReLU']:
            name = 'leaky_relu' if self.activation == 'LeakyReLU' else 'relu'
            gain = torch.nn.init.calculate_gain(name)

        inter_dim = 28 * 28 // 4
        latent_dim = inter_dim // 4
        layers = [
            nn.Linear(28 * 28, inter_dim),
            act_layer(**act_kwargs),
            nn.Linear(inter_dim, latent_dim),
            act_layer(**act_kwargs)
        ]
        for layer in layers:
            if hasattr(layer, 'weight') and layer.weight.data.dim() > 1:
                _initialize(init_method, layer)
        self.encoder = nn.Sequential(*layers)
        layers = [
            nn.Linear(latent_dim, inter_dim),
            act_layer(**act_kwargs),
            nn.Linear(inter_dim, 28 * 28),
            nn.Sigmoid()
        ]
        layers = [
            nn.Linear(latent_dim, 28 * 28),
            nn.Sigmoid()
        ]
        for layer in layers:
            if hasattr(layer, 'weight') and layer.weight.data.dim() > 1:
                _initialize(init_method, layer)
        self.decoder = nn.Sequential(*layers)

    def forward(self, x):
        self._iters += 1
        shape = x.size()
        x = x.view(x.shape[0], -1)
        x = self.encoder(x)
        x = self.decoder(x)
        return x.view(shape)

class NegLossScore(NeuralNetRegressor):
    steps = 0
    def partial_fit(self, *args, **kwargs):
        super().partial_fit(*args, **kwargs)
        self.steps += 1

    def score(self, X, y):
        X = skorch.utils.to_tensor(X, device=self.device)
        y = skorch.utils.to_tensor(y, device=self.device)

        self.initialize_criterion()
        y_hat = self.predict(X)
        y_hat = skorch.utils.to_tensor(y_hat, device=self.device)
        loss = super().get_loss(y_hat, y, X=X, training=False).item()
        print(f'steps = {self.steps}, loss = {loss}')
        return -1 * loss

    def initialize(self, *args, **kwargs):
        super().initialize(*args, **kwargs)
        self.callbacks_ = []


## model_selection_algs.py
import math
import toolz
import numpy as np
from time import time


def stop_on_plateau(info, patience=10, tol=0.001, max_iter=None):
    out = {}
    for ident, records in info.items():
        pf_calls = records[-1]['partial_fit_calls']
        if max_iter is not None and pf_calls > max_iter:
            out[ident] = 0

        elif pf_calls > patience:
            # old = records[-patience]['score']
            plateau = {d['partial_fit_calls']: d['score']
                       for d in records
                       if pf_calls - patience <= d['partial_fit_calls']}
            plateau_start = plateau[min(plateau)]
            if all(score < plateau_start + tol for score in plateau.values()):
                out[ident] = 0
            else:
                out[ident] = 1
        else:
            out[ident] = 1
    return out

def _hyperband_paper_alg(R, eta=3):
    """
    Algorithm 1 from the Hyperband paper [1]_.

    References
    ----------
    1. "Hyperband: A novel bandit-based approach to hyperparameter
       optimization", 2016 by L. Li, K. Jamieson, G. DeSalvo, A. Rostamizadeh,
       and A. Talwalkar.  https://arxiv.org/abs/1603.06560
    """
    s_max = math.floor(math.log(R, eta))
    B = (s_max + 1) * R
    brackets = reversed(range(int(s_max + 1)))
    hists = {}
    for s in brackets:
        n = int(math.ceil(B / R * eta ** s / (s + 1)))
        r = int(R * eta ** -s)
        T = set(range(n))
        hist = {
            "num_models": n,
            "models": {n: 0 for n in range(n)},
            "iters": [],
        }
        for i in range(s + 1):
            n_i = math.floor(n * eta ** -i)
            r_i = np.round(r * eta ** i).astype(int)
            L = {model: r_i for model in T}
            hist["models"].update(L)
            hist["iters"] += [r_i]
            to_keep = math.floor(n_i / eta)
            T = {model for i, model in enumerate(T) if i < to_keep}
        hists["bracket={s}".format(s=s)] = hist
    info = [
        {
            "bracket": k,
            "num_models": hist["num_models"],
            "num_partial_fit_calls": sum(hist["models"].values()),
            "iters": {int(h) for h in hist["iters"]},
        }
        for k, hist in hists.items()
    ]
    return info


class SHA:
    def __init__(self, n, r, eta=3, limit=None,
                 patience=np.inf, tol=0.001):
        """
        Perform the successive halving algorithm.

        Parameters
        ----------
        n : int
            Number of models to evaluate initially
        r : int
            Number of times to call partial fit initially
        eta : float, default=3
            How aggressive to be in culling off the models. Higher
            values correspond to being more aggressive in killing off
            models. The "infinite horizon" theory suggests eta=np.e=2.718...
            is optimal.
        patience : int
            Passed to `stop_on_plateau`
        tol : int
            Passed to `stop_on_plateau`
        """
        self.steps = 0
        self.n = n
        self.r = r
        self.eta = eta
        self.meta = []
        self.start = time()
        self.patience = patience
        self.tol = tol
        self.limit = limit

    def fit(self, info):
        n, r, eta = self.n, self.r, self.eta
        n_i = math.floor(n * eta ** -self.steps)
        r_i = np.round(r * eta**self.steps).astype(int)

        # Initial case
        # partial fit has already been called once
        if r_i == 1:
            # if r_i == 1, a step has already been completed for us
            assert self.steps == 0
            self.steps = 1
            pf_calls = {k: info[k][-1]['partial_fit_calls'] for k in info}
            return self.fit(info)
        # this ordering is important; typically r_i==1 when steps==0
        if self.steps == 0:
            # we have r_i - 1 more steps to train to
            self.steps = 1
            return {k: r_i - 1 for k in info}

        keep_training = stop_on_plateau(info,
                                        patience=self.patience,
                                        tol=self.tol)
        if sum(keep_training.values()) == 0:
            return keep_training
        info = {k: info[k] for k in keep_training}

        best = toolz.topk(n_i, info, key=lambda k: info[k][-1]['score'])
        self.steps += 1

        if len(best) in {0, 1} and self.steps > self.limit:
            return {0: 0}

        pf_calls = {k: info[k][-1]['partial_fit_calls'] for k in best}
        addtl_pf_calls = {k: r_i - pf_calls[k]
                          for k in best}
        return addtl_pf_calls


from sklearn.base import BaseEstimator
class _Constant(BaseEstimator):
    def __init__(self, value=0, meta=None):
        self.value = value
        if meta is None:
            meta = {}
        self.meta = meta
        super().__init__()

    def partial_fit(self, *args, **kwargs):
        pass
        return self

    def score(self, *args, **kwargs):
        return self.value

from sklearn.model_selection import ParameterSampler
from sklearn.base import clone
from dask_ml.model_selection._incremental import fit
from dask_ml.datasets import make_classification
from distributed import Client

if __name__ == "__main__":
    client = Client('localhost:8786')
    X, y = make_classification(n_features=5, n_samples=200, chunks=10)
    R = 100
    eta = 3.0
    # def hyperband(R, eta=3):
    info = _hyperband_paper_alg(R, eta=eta)
    # Because we call `partial_fit` before
    for i in info:
        i['iters'].update({1})

    sh_info = []
    s_max = math.floor(math.log(R, eta))
    B = (s_max + 1) * R
    for s in reversed(np.arange(s_max + 1)):
        n = np.ceil(B / R * eta**s / (s + 1))
        r = np.floor(R * eta**-s)
        alg = SHA(n, r, limit=s+1)
        model = _Constant()
        params = {'value': np.linspace(0, 1, num=1000)}
        params_list = list(ParameterSampler(params, n))

        _, _, hist = fit(model, params_list, X, y, X, y, alg.fit)
        ids = {h['model_id'] for h in hist}
        info_hist = {i: [] for i in ids}
        for h in hist:
            info_hist[h['model_id']] += [h]
        hist = info_hist

        calls = {k: max(hi['partial_fit_calls'] for hi in h)
                 for k, h in hist.items()}
        iters = {hi['partial_fit_calls'] for h in hist.values() for hi in h}
        sh_info += [{'bracket': f'bracket={s}',
                     'iters': iters,
                     'num_models': len(hist),
                     'num_partial_fit_calls': sum(calls.values())}]

    assert sh_info == info

## noisy_mnist.py
from keras.datasets import mnist
import numpy as np
import skimage.util
import random

import skimage.filters
import skimage
import scipy.signal


def noise_img(x):
    noises = [
        {"mode": "s&p", "amount": np.random.uniform(0.0, 0.2)},
        {"mode": "gaussian", "var": np.random.uniform(0.0, 0.15)},
    ]
    # noise = random.choice(noises)
    noise = noises[1]
    return skimage.util.random_noise(x, **noise)


def train_formatting(img):
    img = img.reshape(28, 28).astype("float32")
    return img.flat[:]


def blur_img(img):
    assert img.ndim == 1
    n = int(np.sqrt(img.shape[0]))
    img = img.reshape(n, n)
    h = np.zeros((n, n))
    angle = np.random.uniform(-5, 5)
    w = random.choice(range(1, 3))
    h[n // 2, n // 2 - w : n // 2 + w] = 1
    h = skimage.transform.rotate(h, angle)
    h /= h.sum()
    y = scipy.signal.convolve(img, h, mode="same")
    return y.flat[:]


def dataset(n=None):
    (x_train, _), (x_test, _) = mnist.load_data()
    x = np.concatenate((x_train, x_test))
    if n:
        x = x[:n]
    else:
        n = int(70e3)
    x = x.astype("float32") / 255.
    x = np.reshape(x, (len(x), 28 * 28))

    y = np.apply_along_axis(train_formatting, 1, x)

    clean = y.copy()

    noisy = y.copy()
    # order = [noise_img, blur_img]
    # order = [blur_img]
    order = [noise_img]

    random.shuffle(order)

    for fn in order:
        noisy = np.apply_along_axis(fn, 1, noisy)

    noisy = noisy.astype("float32")
    clean = clean.astype("float32")
#     noisy = noisy.reshape(-1, 1, 28, 28).astype("float32")
#     clean = clean.reshape(-1, 1, 28, 28).astype("float32")
    return noisy, clean
	import skorch.utils
	from skorch import NeuralNetRegressor
	import torch.nn as nn
	import torch
	import skorch
	from distributed.utils import log_errors

	def _initialize(method, layer, gain=1):
	weight = layer.weight.data
	_before = weight.data.clone()
	kwargs = {'gain': gain} if 'xavier' in str(method) else {}
	method(weight.data, **kwargs)
	assert torch.all(weight.data != _before)


	class Autoencoder(nn.Module):
	def __init__(self, activation='ReLU', init='xavier_uniform_',
	**kwargs):
	super().__init__()

	self.activation = activation
	self.init = init
	self._iters = 0

	init_method = getattr(torch.nn.init, init)
	act_layer = getattr(nn, activation)
	act_kwargs = {'inplace': True} if self.activation != 'PReLU' else {}

	gain = 1
	if self.activation in ['LeakyReLU', 'ReLU']:
	name = 'leaky_relu' if self.activation == 'LeakyReLU' else 'relu'
	gain = torch.nn.init.calculate_gain(name)

	inter_dim = 28 * 28 // 4
	latent_dim = inter_dim // 4
	layers = [
	nn.Linear(28 * 28, inter_dim),
	act_layer(**act_kwargs),
	nn.Linear(inter_dim, latent_dim),
	act_layer(**act_kwargs)
	]
	for layer in layers:
	if hasattr(layer, 'weight') and layer.weight.data.dim() > 1:
	_initialize(init_method, layer)
	self.encoder = nn.Sequential(*layers)
	layers = [
	nn.Linear(latent_dim, inter_dim),
	act_layer(**act_kwargs),
	nn.Linear(inter_dim, 28 * 28),
	nn.Sigmoid()
	]
	layers = [
	nn.Linear(latent_dim, 28 * 28),
	nn.Sigmoid()
	]
	for layer in layers:
	if hasattr(layer, 'weight') and layer.weight.data.dim() > 1:
	_initialize(init_method, layer)
	self.decoder = nn.Sequential(*layers)

	def forward(self, x):
	self._iters += 1
	shape = x.size()
	x = x.view(x.shape[0], -1)
	x = self.encoder(x)
	x = self.decoder(x)
	return x.view(shape)

	class NegLossScore(NeuralNetRegressor):
	steps = 0
	def partial_fit(self, args, *kwargs):
	super().partial_fit(args, *kwargs)
	self.steps += 1

	def score(self, X, y):
	X = skorch.utils.to_tensor(X, device=self.device)
	y = skorch.utils.to_tensor(y, device=self.device)

	self.initialize_criterion()
	y_hat = self.predict(X)
	y_hat = skorch.utils.to_tensor(y_hat, device=self.device)
	loss = super().get_loss(y_hat, y, X=X, training=False).item()
	print(f'steps = {self.steps}, loss = {loss}')
	return -1 * loss

	def initialize(self, args, *kwargs):
	super().initialize(args, *kwargs)
	self.callbacks_ = []
	import math
	import toolz
	import numpy as np
	from time import time


	def stop_on_plateau(info, patience=10, tol=0.001, max_iter=None):
	out = {}
	for ident, records in info.items():
	pf_calls = records[-1]['partial_fit_calls']
	if max_iter is not None and pf_calls > max_iter:
	out[ident] = 0

	elif pf_calls > patience:
	# old = records[-patience]['score']
	plateau = {d['partial_fit_calls']: d['score']
	for d in records
	if pf_calls - patience <= d['partial_fit_calls']}
	plateau_start = plateau[min(plateau)]
	if all(score < plateau_start + tol for score in plateau.values()):
	out[ident] = 0
	else:
	out[ident] = 1
	else:
	out[ident] = 1
	return out

	def _hyperband_paper_alg(R, eta=3):
	"""
	Algorithm 1 from the Hyperband paper [1]_.

	References
	----------
	1. "Hyperband: A novel bandit-based approach to hyperparameter
	optimization", 2016 by L. Li, K. Jamieson, G. DeSalvo, A. Rostamizadeh,
	and A. Talwalkar. https://arxiv.org/abs/1603.06560
	"""
	s_max = math.floor(math.log(R, eta))
	B = (s_max + 1) * R
	brackets = reversed(range(int(s_max + 1)))
	hists = {}
	for s in brackets:
	n = int(math.ceil(B / R * eta ** s / (s + 1)))
	r = int(R * eta ** -s)
	T = set(range(n))
	hist = {
	"num_models": n,
	"models": {n: 0 for n in range(n)},
	"iters": [],
	}
	for i in range(s + 1):
	n_i = math.floor(n * eta ** -i)
	r_i = np.round(r * eta ** i).astype(int)
	L = {model: r_i for model in T}
	hist["models"].update(L)
	hist["iters"] += [r_i]
	to_keep = math.floor(n_i / eta)
	T = {model for i, model in enumerate(T) if i < to_keep}
	hists["bracket={s}".format(s=s)] = hist
	info = [
	{
	"bracket": k,
	"num_models": hist["num_models"],
	"num_partial_fit_calls": sum(hist["models"].values()),
	"iters": {int(h) for h in hist["iters"]},
	}
	for k, hist in hists.items()
	]
	return info


	class SHA:
	def __init__(self, n, r, eta=3, limit=None,
	patience=np.inf, tol=0.001):
	"""
	Perform the successive halving algorithm.

	Parameters
	----------
	n : int
	Number of models to evaluate initially
	r : int
	Number of times to call partial fit initially
	eta : float, default=3
	How aggressive to be in culling off the models. Higher
	values correspond to being more aggressive in killing off
	models. The "infinite horizon" theory suggests eta=np.e=2.718...
	is optimal.
	patience : int
	Passed to `stop_on_plateau`
	tol : int
	Passed to `stop_on_plateau`
	"""
	self.steps = 0
	self.n = n
	self.r = r
	self.eta = eta
	self.meta = []
	self.start = time()
	self.patience = patience
	self.tol = tol
	self.limit = limit

	def fit(self, info):
	n, r, eta = self.n, self.r, self.eta
	n_i = math.floor(n * eta ** -self.steps)
	r_i = np.round(r * eta**self.steps).astype(int)

	# Initial case
	# partial fit has already been called once
	if r_i == 1:
	# if r_i == 1, a step has already been completed for us
	assert self.steps == 0
	self.steps = 1
	pf_calls = {k: info[k][-1]['partial_fit_calls'] for k in info}
	return self.fit(info)
	# this ordering is important; typically r_i==1 when steps==0
	if self.steps == 0:
	# we have r_i - 1 more steps to train to
	self.steps = 1
	return {k: r_i - 1 for k in info}

	keep_training = stop_on_plateau(info,
	patience=self.patience,
	tol=self.tol)
	if sum(keep_training.values()) == 0:
	return keep_training
	info = {k: info[k] for k in keep_training}

	best = toolz.topk(n_i, info, key=lambda k: info[k][-1]['score'])
	self.steps += 1

	if len(best) in {0, 1} and self.steps > self.limit:
	return {0: 0}

	pf_calls = {k: info[k][-1]['partial_fit_calls'] for k in best}
	addtl_pf_calls = {k: r_i - pf_calls[k]
	for k in best}
	return addtl_pf_calls


	from sklearn.base import BaseEstimator
	class _Constant(BaseEstimator):
	def __init__(self, value=0, meta=None):
	self.value = value
	if meta is None:
	meta = {}
	self.meta = meta
	super().__init__()

	def partial_fit(self, args, *kwargs):
	pass
	return self

	def score(self, args, *kwargs):
	return self.value

	from sklearn.model_selection import ParameterSampler
	from sklearn.base import clone
	from dask_ml.model_selection._incremental import fit
	from dask_ml.datasets import make_classification
	from distributed import Client

	if __name__ == "__main__":
	client = Client('localhost:8786')
	X, y = make_classification(n_features=5, n_samples=200, chunks=10)
	R = 100
	eta = 3.0
	# def hyperband(R, eta=3):
	info = _hyperband_paper_alg(R, eta=eta)
	# Because we call `partial_fit` before
	for i in info:
	i['iters'].update({1})

	sh_info = []
	s_max = math.floor(math.log(R, eta))
	B = (s_max + 1) * R
	for s in reversed(np.arange(s_max + 1)):
	n = np.ceil(B / R * eta**s / (s + 1))
	r = np.floor(R * eta**-s)
	alg = SHA(n, r, limit=s+1)
	model = _Constant()
	params = {'value': np.linspace(0, 1, num=1000)}
	params_list = list(ParameterSampler(params, n))

	_, _, hist = fit(model, params_list, X, y, X, y, alg.fit)
	ids = {h['model_id'] for h in hist}
	info_hist = {i: [] for i in ids}
	for h in hist:
	info_hist[h['model_id']] += [h]
	hist = info_hist

	calls = {k: max(hi['partial_fit_calls'] for hi in h)
	for k, h in hist.items()}
	iters = {hi['partial_fit_calls'] for h in hist.values() for hi in h}
	sh_info += [{'bracket': f'bracket={s}',
	'iters': iters,
	'num_models': len(hist),
	'num_partial_fit_calls': sum(calls.values())}]

	assert sh_info == info
	from keras.datasets import mnist
	import numpy as np
	import skimage.util
	import random

	import skimage.filters
	import skimage
	import scipy.signal


	def noise_img(x):
	noises = [
	{"mode": "s&p", "amount": np.random.uniform(0.0, 0.2)},
	{"mode": "gaussian", "var": np.random.uniform(0.0, 0.15)},
	]
	# noise = random.choice(noises)
	noise = noises[1]
	return skimage.util.random_noise(x, **noise)


	def train_formatting(img):
	img = img.reshape(28, 28).astype("float32")
	return img.flat[:]


	def blur_img(img):
	assert img.ndim == 1
	n = int(np.sqrt(img.shape[0]))
	img = img.reshape(n, n)
	h = np.zeros((n, n))
	angle = np.random.uniform(-5, 5)
	w = random.choice(range(1, 3))
	h[n // 2, n // 2 - w : n // 2 + w] = 1
	h = skimage.transform.rotate(h, angle)
	h /= h.sum()
	y = scipy.signal.convolve(img, h, mode="same")
	return y.flat[:]


	def dataset(n=None):
	(x_train, _), (x_test, _) = mnist.load_data()
	x = np.concatenate((x_train, x_test))
	if n:
	x = x[:n]
	else:
	n = int(70e3)
	x = x.astype("float32") / 255.
	x = np.reshape(x, (len(x), 28 * 28))

	y = np.apply_along_axis(train_formatting, 1, x)

	clean = y.copy()

	noisy = y.copy()
	# order = [noise_img, blur_img]
	# order = [blur_img]
	order = [noise_img]

	random.shuffle(order)

	for fn in order:
	noisy = np.apply_along_axis(fn, 1, noisy)

	noisy = noisy.astype("float32")
	clean = clean.astype("float32")
	# noisy = noisy.reshape(-1, 1, 28, 28).astype("float32")
	# clean = clean.reshape(-1, 1, 28, 28).astype("float32")
	return noisy, clean