.gitignore

misc
dask-worker-space
.local

README.md

Active Learning Hacking

Repository for active learning hacks

active-learning-dask.ipynb

active-learning-large.ipynb

active-learning.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Active Learning Notebook\n",
    "\n",
    "Test out using [modAL](https://modal-python.readthedocs.io/en/latest/index.html) with PyMKS."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Todo List:\n",
    "\n",
    " - Make work with all the data already evaluated for speedier evaluation\n",
    " - Implement with multiple runs and plot uncertainty\n",
    " - Implment iGS\n",
    " - Make it all functional to some degree\n",
    " - PR into PyMKS for an active learning notebook"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import dask.array as da\n",
    "from dask_ml.model_selection import train_test_split\n",
    "from sklearn.pipeline import Pipeline\n",
    "from dask_ml.decomposition import IncrementalPCA\n",
    "from dask_ml.preprocessing import PolynomialFeatures\n",
    "from sklearn.linear_model import LinearRegression\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.gaussian_process import GaussianProcessRegressor\n",
    "from sklearn.gaussian_process.kernels import RBF\n",
    "from sklearn.metrics import mean_squared_error as mse\n",
    "from sklearn.metrics import r2_score\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "from pymks import (\n",
    "    generate_multiphase,\n",
    "    plot_microstructures,\n",
    "    PrimitiveTransformer,\n",
    "    TwoPointCorrelation,\n",
    "    GenericTransformer,\n",
    "    solve_fe\n",
    ")\n",
    "\n",
    "from toolz.curried import curry, pipe, valmap, itemmap, iterate, do, merge_with\n",
    "from toolz.curried import map as map_\n",
    "from modAL.models import ActiveLearner, CommitteeRegressor, BayesianOptimizer\n",
    "from modAL.disagreement import max_std_sampling\n",
    "from modAL.models import BayesianOptimizer\n",
    "from modAL.acquisition import max_EI\n",
    "import tqdm\n",
    "import types\n",
    "from pymks.fmks.func import sequence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_data = np.arange(50).reshape((50, 1))\n",
    "y_data = np.arange(50).reshape((50, 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "test_size=-24 should be either positive and smaller than the number of samples 50 or a float in the (0, 1) range",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-5-4a6ca797b8b5>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mx_0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx_1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_1\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtrain_test_split\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_data\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_data\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtrain_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m25\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/nix/store/xh1wxwafz46zf87syxav55mjy43f4pyj-python3.8-dask-ml-1.6.0/lib/python3.8/site-packages/dask_ml/model_selection/_split.py\u001b[0m in \u001b[0;36mtrain_test_split\u001b[0;34m(test_size, train_size, random_state, shuffle, blockwise, convert_mixed_types, *arrays, **options)\u001b[0m\n\u001b[1;32m    510\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mitertools\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mchain\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfrom_iterable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_test_pairs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    511\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 512\u001b[0;31m         return ms.train_test_split(\n\u001b[0m\u001b[1;32m    513\u001b[0m             \u001b[0;34m*\u001b[0m\u001b[0marrays\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    514\u001b[0m             \u001b[0mtest_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtest_size\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/nix/store/h3jg5jg3iv8gxqb2hhfrvxnr3bb4dcah-python3.8-scikit-learn-0.23.2/lib/python3.8/site-packages/sklearn/model_selection/_split.py\u001b[0m in \u001b[0;36mtrain_test_split\u001b[0;34m(*arrays, **options)\u001b[0m\n\u001b[1;32m   2128\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2129\u001b[0m     \u001b[0mn_samples\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_num_samples\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marrays\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2130\u001b[0;31m     n_train, n_test = _validate_shuffle_split(n_samples, test_size, train_size,\n\u001b[0m\u001b[1;32m   2131\u001b[0m                                               default_test_size=0.25)\n\u001b[1;32m   2132\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/nix/store/h3jg5jg3iv8gxqb2hhfrvxnr3bb4dcah-python3.8-scikit-learn-0.23.2/lib/python3.8/site-packages/sklearn/model_selection/_split.py\u001b[0m in \u001b[0;36m_validate_shuffle_split\u001b[0;34m(n_samples, test_size, train_size, default_test_size)\u001b[0m\n\u001b[1;32m   1762\u001b[0m     if (test_size_type == 'i' and (test_size >= n_samples or test_size <= 0)\n\u001b[1;32m   1763\u001b[0m        or test_size_type == 'f' and (test_size <= 0 or test_size >= 1)):\n\u001b[0;32m-> 1764\u001b[0;31m         raise ValueError('test_size={0} should be either positive and smaller'\n\u001b[0m\u001b[1;32m   1765\u001b[0m                          \u001b[0;34m' than the number of samples {1} or a float in the '\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1766\u001b[0m                          '(0, 1) range'.format(test_size, n_samples))\n",
      "\u001b[0;31mValueError\u001b[0m: test_size=-24 should be either positive and smaller than the number of samples 50 or a float in the (0, 1) range"
     ]
    }
   ],
   "source": [
    "x_0, x_1, y_0, y_1 = train_test_split(x_data, y_data, train_size=25)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Functions for evaluating learners\n",
    "\n",
    "Following are functions for evaluating learners. The `oracle_func` generates the `y_data` given some `x_data`. We're assuming that it's expensive to use the oracle. Hence why we need active learning."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "@curry\n",
    "def iterate_times(func, times, value):\n",
    "    iter_ = iterate(func, value)\n",
    "    for _ in tqdm.tqdm(range(times)):\n",
    "        next(iter_)\n",
    "    return next(iter_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "@curry\n",
    "def update_learner(oracle_func, x_pool, x_test, y_test, learner):\n",
    "    query = sequence(\n",
    "        learner.query,\n",
    "        oracle_func,\n",
    "    )\n",
    "    return pipe(\n",
    "        learner,\n",
    "        do(lambda x: x.teach(*query(x_pool))),\n",
    "        do(lambda x: x.scores.append(x.score(x_test, y_test)))\n",
    "    )\n",
    "\n",
    "@curry\n",
    "def evaluate_learner(oracle_func, x_pool, x_test, y_test, n_query, learner):\n",
    "    learner.scores = [learner.score(x_test, y_test)]\n",
    "    return iterate_times(\n",
    "        update_learner(oracle_func, x_pool, x_test, y_test),\n",
    "        n_query,\n",
    "        learner\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generate Data Functions\n",
    "\n",
    "Functions to generate the microstructure. The microstructures are stochastically generated from 6 nominal classes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def shuffle(data):\n",
    "    tmp = np.array(data)\n",
    "    np.random.shuffle(tmp)\n",
    "    return da.from_array(tmp, chunks=data.chunks)\n",
    "\n",
    "def generate_x(shape, seed):\n",
    "    da.random.seed(seed)\n",
    "    np.random.seed(seed)\n",
    "\n",
    "    tmp = [\n",
    "        generate_multiphase(shape=shape, grain_size=x, volume_fraction=(0.5, 0.5), chunks=50, percent_variance=0.15)\n",
    "        for x in [(15, 2), (2, 15), (7, 7), (9, 3), (3, 9), (2, 2)]\n",
    "    ]\n",
    "    return np.array(shuffle(da.concatenate(tmp)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Functions to generate models\n",
    "\n",
    "Here we use the GPR model as it returns a probability that's required by the `ActiveLearner` class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def pca_steps():\n",
    "    return (\n",
    "        (\"reshape\", GenericTransformer(\n",
    "            lambda x: x.reshape(x.shape[0], x_data.shape[1], x_data.shape[2])\n",
    "        )),\n",
    "        (\"discritize\",PrimitiveTransformer(n_state=2, min_=0.0, max_=1.0)),\n",
    "        (\"correlations\",TwoPointCorrelation(periodic_boundary=True, cutoff=31, correlations=[(0, 1), (1, 1)])),\n",
    "        ('flatten', GenericTransformer(lambda x: x.reshape(x.shape[0], -1))),\n",
    "        ('pca', IncrementalPCA(n_components=3, svd_solver='full')),\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_gp_model():\n",
    "    kernel = 1 * RBF(length_scale=1.0, length_scale_bounds=(1e-1, 1e2))\n",
    "    regressor = GaussianProcessRegressor(kernel=kernel, n_restarts_optimizer=9)\n",
    "    return Pipeline(steps=pca_steps() + (\n",
    "        ('poly', PolynomialFeatures(degree=3)),\n",
    "        ('regressor', regressor),\n",
    "    ))\n",
    "\n",
    "def make_linear_model():\n",
    "    return Pipeline(steps=pca_steps() + (\n",
    "        ('poly', PolynomialFeatures(degree=3))\n",
    "        ('regressor', LinearRegression()),\n",
    "    ))\n",
    "\n",
    "def pca_model():\n",
    "    return Pipeline(steps=pca_steps())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## The Oracle\n",
    "\n",
    "The oracle function is an FE simulation on the 2D grid."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "@curry\n",
    "def oracle_from_data(shape, x_data):\n",
    "    y_stress = solve_fe(x_data.reshape((-1,) + shape),\n",
    "                        elastic_modulus=(1.3, 2.5),\n",
    "                        poissons_ratio=(0.42, 0.35),\n",
    "                        macro_strain=0.001)['stress'][..., 0]\n",
    "\n",
    "    return np.array(da.average(y_stress.reshape(y_stress.shape[0], -1), axis=1))\n",
    "\n",
    "@curry \n",
    "def oracle_func(x_data, y_data, query_instance):\n",
    "    idx, query_value = query_instance\n",
    "    return query_value.reshape(1, -1), np.array([y_data[idx]]).reshape(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Helper Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_parity(y_test, y_predict, label='Testing Data'):\n",
    "    pred_data = np.array([y_test, y_predict])\n",
    "    line = np.min(pred_data), np.max(pred_data)\n",
    "    plt.plot(pred_data[0], pred_data[1], 'o', label=label)\n",
    "    plt.plot(line, line, '-', linewidth=3, color='k')\n",
    "    plt.title('Goodness of Fit', fontsize=20)\n",
    "    plt.xlabel('Actual', fontsize=18)\n",
    "    plt.ylabel('Predicted', fontsize=18)\n",
    "    plt.legend(loc=2, fontsize=15)\n",
    "    return plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def three_way_split(x_data, y_data, props, random_state):\n",
    "    x_0, x_, y_0, y_ = train_test_split(x_data, y_data, train_size=props[0], random_state=random_state)\n",
    "    x_1, x_2, y_1, y_2 = train_test_split(x_, y_, train_size=props[1] / (1 - props[0]), random_state=random_state)\n",
    "    return flatten(x_0), flatten(x_1), flatten(x_2), y_0, y_1, y_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def flatten(x_data):\n",
    "    return x_data.reshape(x_data.shape[0], -1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Greedy Sampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "def split_on_ids(arr, ids):\n",
    "    mask = np.zeros(len(arr), dtype=bool)\n",
    "    mask[ids] = True\n",
    "    return arr[mask], arr[~mask]\n",
    "    \n",
    "def calc_distances(d0, d1):\n",
    "    return np.linalg.norm(d0[:, None, :] - d1[None, :, :], ord=2, axis=-1)\n",
    "\n",
    "def calc_distances_nk(labeled_samples, scores):\n",
    "    scores_labeled, scores_unlabeled = split_on_ids(scores, labeled_samples)\n",
    "    if len(scores_labeled) == 0:\n",
    "        scores_labeled = np.mean(scores_unlabeled, axis=0)[None]\n",
    "    return calc_distances(scores_unlabeled, scores_labeled)\n",
    "\n",
    "def next_sample(distances_nk, labeled_samples, scores):\n",
    "    distances_n = np.min(distances_nk, axis=1)\n",
    "    _, unlabeled_ids = split_on_ids(np.arange(len(scores)), labeled_samples)\n",
    "    return unlabeled_ids[np.argmax(distances_n)]\n",
    "\n",
    "def next_sample_gsx(labeled_samples, scores):\n",
    "    distances_nk = calc_distances_nk(labeled_samples, scores)\n",
    "    return next_sample(distances_nk, labeled_samples, scores)\n",
    "        \n",
    "def next_sample_igs(labeled_samples, scores):\n",
    "    x_scores, y_scores = scores\n",
    "    distances_nk_x = calc_distances_nk(labeled_samples, x_scores)\n",
    "    distances_nk_y = calc_distances_nk(labeled_samples, y_scores)\n",
    "    return next_sample(distances_nk_x * distances_nk_y, labeled_samples, x_scores)\n",
    "\n",
    "def query_helper(model, x_pool, init_scores, update_scores, next_func):\n",
    "    if not hasattr(model, 'query_data'):\n",
    "        model.query_data = [], init_scores()\n",
    "    labeled_samples, scores = model.query_data\n",
    "    scores = update_scores(model, scores)\n",
    "    next_id = next_func(labeled_samples, scores)\n",
    "    model.query_data = (labeled_samples + [next_id], scores)\n",
    "    return next_id, x_pool[next_id]\n",
    "\n",
    "def gsx_query(model, x_pool):\n",
    "    return query_helper(\n",
    "        model,\n",
    "        x_pool,\n",
    "        lambda: pca_model().fit_transform(x_pool),\n",
    "        lambda m, s: s,\n",
    "        next_sample_gsx\n",
    "    )\n",
    "\n",
    "def gsy_query(model, x_pool):\n",
    "    return query_helper(\n",
    "        model,\n",
    "        x_pool,\n",
    "        lambda: None,\n",
    "        lambda m, s: m.predict(x_pool).reshape(-1, 1),\n",
    "        next_sample_gsx\n",
    "    )\n",
    "\n",
    "def igs_query(model, x_pool):\n",
    "    return query_helper(\n",
    "        model,\n",
    "        x_pool,\n",
    "        lambda: (pca_model().fit_transform(x_pool), None),\n",
    "        lambda m, s: (s[0], m.predict(x_pool).reshape(-1, 1)),\n",
    "        next_sample_igs\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Set up the active learners\n",
    "\n",
    "One is a GPR using the maximum std and the other is random"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "query_std = lambda model, x_: pipe(\n",
    "    model.predict(x_, return_std=True)[1],\n",
    "    np.argmax,\n",
    "    lambda i: (i, x_[i])\n",
    ")\n",
    "\n",
    "query_random = lambda model, x_: pipe(\n",
    "    np.random.randint(0, len(x_)),\n",
    "    lambda i: (i, x_[i])\n",
    ")\n",
    "\n",
    "def make_ensemble(x_train, y_train):\n",
    "    ensemble_learner = CommitteeRegressor(\n",
    "        learner_list=[\n",
    "            ActiveLearner(\n",
    "                estimator=make_gp_model(),\n",
    "                X_training=x_train_,\n",
    "                y_training=y_train_\n",
    "            )\n",
    "            for x_train_, y_train_ in zip(np.array_split(x_train, 5), np.array_split(y_train, 5))\n",
    "        ],\n",
    "        query_strategy=max_std_sampling\n",
    "    )\n",
    "\n",
    "    ## required because CommitteeRegressor does not have a score function\n",
    "\n",
    "    def score(self, x_true, y_true):\n",
    "        y_pred = self.predict(x_true)\n",
    "        return r2_score(y_true, y_pred)\n",
    "\n",
    "    ensemble_learner.score = types.MethodType(score, ensemble_learner)\n",
    "    \n",
    "    return ensemble_learner\n",
    "\n",
    "@curry\n",
    "def make_learner(x_train, y_train, klass, query_func):\n",
    "    return klass(\n",
    "        estimator=make_gp_model(),\n",
    "        query_strategy=query_func,\n",
    "        X_training=x_train,\n",
    "        y_training=y_train,\n",
    "    )\n",
    "\n",
    "def make_bayes(x_train, y_train):\n",
    "    return make_learner(x_train, y_train, BayesianOptimizer, max_EI)\n",
    "\n",
    "def make_learners(x_train, y_train):\n",
    "    make_active = make_learner(x_train, y_train, ActiveLearner)\n",
    "\n",
    "    return dict(\n",
    "        std=make_active(query_std),\n",
    "        random=make_active(query_random),\n",
    "#        ensemble=make_ensemble(x_train, y_train),\n",
    "        bayes=make_bayes(x_train, y_train),\n",
    "        gsx=make_active(gsx_query),\n",
    "        igs=make_active(igs_query),\n",
    "        gsy=make_active(gsy_query)\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "@curry\n",
    "def evaluate_item(oracle, x_pool, x_test, y_test, iterations, item):\n",
    "    name, learner = item\n",
    "    print('evaluating', name)\n",
    "    return name, evaluate_learner(oracle, x_pool, x_test, y_test, iterations, learner)\n",
    "\n",
    "@curry\n",
    "def one_round(x_data, y_data, iterations, seed):\n",
    "    x_pool, x_test, x_train, y_pool, y_test, y_train = three_way_split(x_data, y_data, (0.8, 0.16), seed)\n",
    "    oracle = oracle_func(x_pool, y_pool)\n",
    "    eval_item = evaluate_item(oracle, x_pool, x_test, y_test, iterations)\n",
    "    return itemmap(eval_item, make_learners(x_train, y_train))\n",
    "\n",
    "def multiple_rounds(x_data, y_data, rounds, iterations):\n",
    "    process_scores = sequence(\n",
    "        lambda _: one_round(x_data, y_data, iterations, None),\n",
    "        valmap(lambda x: x.scores)\n",
    "    )\n",
    "    \n",
    "    return pipe(\n",
    "        [None] * rounds,\n",
    "        map_(process_scores),\n",
    "        list,\n",
    "        lambda x: merge_with(np.vstack)(*x),\n",
    "        valmap(lambda x: (np.mean(x, axis=0), np.std(x, axis=0)))\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generate the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "grid_shape = (41, 41)\n",
    "n_sample_per_class = 100\n",
    "x_data = generate_x((n_sample_per_class,) + grid_shape, 5)\n",
    "y_data = oracle_from_data(grid_shape, x_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2880x288 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_microstructures(*x_data[:10], cmap='gray', colorbar=False);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Run the learners"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating std\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [1:17:36<00:00, 46.57s/it]   \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating random\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:35<00:00,  1.04it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating bayes\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsx\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating igs\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [02:11<00:00,  1.31s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsy\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.13s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating std\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:54<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating random\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:35<00:00,  1.05it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating bayes\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsx\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating igs\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [02:11<00:00,  1.31s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsy\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating std\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating random\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:35<00:00,  1.05it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating bayes\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsx\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.13s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating igs\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [02:11<00:00,  1.31s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsy\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.13s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating std\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:54<00:00,  1.15s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating random\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:35<00:00,  1.05it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating bayes\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:54<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsx\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.13s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating igs\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [02:11<00:00,  1.31s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsy\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating std\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.13s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating random\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:35<00:00,  1.05it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating bayes\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsx\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.13s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating igs\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [02:11<00:00,  1.31s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsy\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.13s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating std\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating random\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:35<00:00,  1.05it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating bayes\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.13s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsx\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.13s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating igs\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [02:11<00:00,  1.31s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsy\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.13s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating std\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:54<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating random\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:35<00:00,  1.05it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating bayes\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:54<00:00,  1.15s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsx\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating igs\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [02:11<00:00,  1.32s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsy\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating std\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating random\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:35<00:00,  1.05it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating bayes\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsx\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating igs\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [02:11<00:00,  1.32s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsy\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.13s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating std\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:54<00:00,  1.15s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating random\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:35<00:00,  1.05it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating bayes\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:54<00:00,  1.15s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsx\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating igs\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [02:11<00:00,  1.32s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsy\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.13s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating std\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:54<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating random\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:36<00:00,  1.04it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating bayes\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsx\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:54<00:00,  1.14s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating igs\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [02:12<00:00,  1.32s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evaluating gsy\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:53<00:00,  1.14s/it]\n"
     ]
    }
   ],
   "source": [
    "scores = multiple_rounds(x_data, y_data, 10, 100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "## The results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.0, 1.0)"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmMAAAHmCAYAAAA/XYmLAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAC4AUlEQVR4nOz9eZgk13mfib7nxJJbrd3VezeAbuyNxg6QAAmAIAlx02ZL1ljWyLSpkZcrz5hXlq1Hske+oixL5JWlsXUlWzOmZyxZ8h1asmRJpkSRoAiSIIkdjUZj724AjV6rq7u2zKzMjIhz5o+I3CqzsrKqIquqG9/7PNXVlRkZcTIjI+IXv+8736estRZBEARBEARhQ9AbPQBBEARBEIR3MyLGBEEQBEEQNhARY4IgCIIgCBuIiDFBEARBEIQNRMSYIAiCIAjCBiJiTBAEQRAEYQNZNzF24sQJ/vE//sdLPv/cc8/xUz/1U3z605/mD//wD9drWIIgCIIgCBuKux4b+Z3f+R0ee+wxxsfHuz5fqVT4/Oc/zy/90i8xPDzMZz7zGe644w4OHDiwHsMTBEEQBEHYMNbFGfvkJz/JZz/72SWfP3bsGPv372dsbAzHcbjvvvt4/vnn12NogiAIgiAIG8qmyBmbnp5mdHS08ffIyAgzMzMbNyBBEARBEIR1Yl3ClP2glGr7OwzDrss9+uijPProowA93TZBEARBEITLgU0hxsbGxpifn2/8PTc3x9jYWNdlH3nkER555JHG32fOnBnYuIJalfnJ85SKxZ7LKTeLs2VP8wFrcG37a8xChDGLjMhMDnx/0coUfqZdmGZczbBeQGHi1aMJbJ66XlUKMlkN1qCicmO5tLDax+rsCl8UoqIFFM3Wp0ZnQS96vyZAm4UVrXrLli1cunRpZeNpITKW0Nj4t7VEtQCCAEwEUYRKfuc0eNqCjX+scbCRRVdnsZUFqNWSnypUq1CrYpPfhAGEUWOdmAiMBa3jHaYUaAew8fONnxAbRfE2jYl/rImfC5Lt1X8vccPSFVehXI3y6r81ylUoR4ET/1auRvkanXVQmfpvB6VV7KErUPE/oJPfKrmRUjTX5er4/65CqfitNL4G1ibvn2S9CqXr61MoJ/7/4pszoR1rbctn2vpE/fnWP2hbtvUpWlsT227PL97u4oEkv1R8pBsFERAqG/+2YLBEyhJCfLxB8mOTn+YQDRabPBcoQ4AhVJYAEy+r4mVMY1mLVc3X22RACpV8ZePfEZaiDgjUonNjMv6VfNtUsrTq8hiAst2X7b5Mly237LbGqy1tW1UAtmV9XUfYuvts+zrtUu+45XHb7VGFsovG1vUNtK+n+9ZUx/dr6REteoe2YytYVX9vy6+zY/0WNJoP3/f/W9mLV8Hu3bu7Pr5hYqxcLlMul5mYmOD666/nt37rt5idnWVoaIgnnniCH/7hH96ooQ0ERURHVNh0EU128VcMTBSgdHNZhUGZCpBtvsQalElfiMUbiFawrEXZKsrU2h82ESqcx1qNxQOrUFRRNsS2XnhV/QoPtJ5SkvcIYKsVbHWh8ZgNApifTQRRrSmIqhUozmPn5jBzc9j5WWxxHlWtQK2GG9Rwa1VUt/2Q0PHOXYX1NSrjoDMa5TuxwEmECJ6D8n1U1kUPOeis0xA3qn60qZY93BA2iShRCpTTOFnGF7/kdJSIJjTJsq2fEdjkNUpZUBalbLw6vcKzE3Ut6GCtm+yzZFuNC0Wyn1DN7ToeURTfLFg01jrJ/oy/06pxBVLYWIGB0vH/W39wUDp+3iqNxUl+dLJdJ3m9oq4SbcvlGGvjR7SPdTyskwHXxzp+cnwYFFF8TFrTfB+xQmx7n51ndtXy/kHZkCgsUVq4SLkyzUIwT2gVET4RGYzOEpIBPKzRWOOAUVijsMbBRCrW3JElNAZrLVprtNY4WqOUwlpLpVqjUq2xUK1QjcpEdgGtFUqr+LcCpTTKOijronDAuHieT7VWxhBgVA1LgFEhWltwQGuTiOEoXoYaRgVYHWBVGMslaxP5k3yvHJP8RM3/6+THWfl3rSdWQeSCccE6KHTyE/9fKweFit87GqXq59h4zBYDyoDVZIIxhsw4rtmKYydwGMVzXFxHxz/awXVaztGtp6X2P+qbiEUx8W/bolRbPwWd3FxopVHJ/8dGR2MDQiU3Hjb5BluLNQaT/I4fS75vqkUCJce+0rHYpGX9WsXfG+UodLL++Cd+3HEcHK1QWuM4TjK25g2Q1vG7rb8nk3wvrbUYFMYSfxtsctwpndxfKWyyHkdrHKXie08UOvnMFt9j1e8pTPLZmS4CsN/7svrHX19n4/NsfazjbqLJgT07mJ1e/U3+WlkXMfaFL3yBp59+mnPnzvEzP/MzfPKTn2RycpLHHnuMn//5nyebzfKpT32Kz3zmM0RRxIMPPsjBgwfXY2gDonOHKxMB3qLF+jxxmU4HRJkAZR2s8hIhVkHZlQkxW6vGFx/X7elEKBuBqWBxkgums2hFBmtj50aFC2DC+EtvTCzkDDQ+kzBAzc9jF8oQRdi682NM7AhFIQRh8juIHaYg/rHJ/2exRFMXsLOzMDcDpd7OpdUaWxjGDo9gh0YwE9uxXgbr++D7WM8DT+NmIpyMxfFDXC/EcSNcN8D1QlwnxHGDFkHRx+eLwqj4gmxUFqvchtiM91SLoFHxPbytKzVL4iomzpyyyefvYJWT7APd/FytwRobn3RUvJ8sybJojHJBeVjlYvBAe83nVX29LkZlwMljlQ9KxReFLm9Z1x2xFsbGxtLN9azruNbP1Na9k9gVwVoMEaGpshCUuDgzyaXZKebLc0SRxYQKE8bGIlgKow65YYuXD7HuAjU7n6wnPlGb+sXH2OS9J5+pMUQ2TARL/GNVhNXtN0osOjTa6JGhayMVi97QwQYOJnQxgYMNXFDg5gKc0RrKr6Ic07aZXkd9dZlhLH6tAhyrwLgNUacavo5OBJ8fP2e95LeLY3xcPLTycZ0MnuPj6AyO4+JoH1d7ONrFUR5KuTjKQSsXpVx0IqxQCkdpPMfB1T5OImIhvkjX70vqRmrTmQWtWq721hLZzguvxSHSGZp3RZ0o1ZRdbTdMLW5V68P18aC6uFGJEFD1cTWHzJatW3CnLsY3PfQWCYOgvt8bN5ut7mnjS1F/R05z4Itoc6ZafofGEmK73M0KS6Hsen8LUmYzhimVDXFsuX2h4jyhGmp/zHEhX+hYn59tOV1ai2OKbMm1n0CCWhxmM7qANhUyGdN3aMcGNagsNJ05pcD1wHXB8/taj7FAGKLCWiycouZRZ8tlmDyLPXcWe/4sXJrCzs7A/NyywqknSoHnoTJZ7NAwjIxih0eICsPY4WEiP4vKG9xMgJ+t4flVlOfE74vEMcKibYBjazg2QFPDsTU0nYI3VFkClSdUeaw7hHILGJ3D6gxG+c3fyksEjYvSscCxygOdRSXuxuLPNEou9qbH0de4viT/6sUXHuKTeKxfTcvdH9QDNvWjWyevc3TnOCKbfJdM3U9Y/i4SpXAAJ1mnoxXjY2PMzs7E521Vd/ZMHGCyBmMjDCFBVGEhLLIQFamGRSq2SGSqaCxok4hQQ2hqlIMZysEsVTNPSBmrVhCa7YE1YGo+hFmsoeFCJM8m73HRi4wG46Csg8ZDKxdP58l5Q+T9YQq5MYZzo3iOlzgbFmNNLO6woKJYwKkkOKdM030iIjQ1QlulZsrUojKBKVMzJRSKnDtO3h0n546Rd8fJOEMNkRQLpdiliExAZGuEpkZgAnzfpVK1KDwclcHVGbTy0cpF4ybOUiyIXJ2Ln8dD61YHhk6zuuVjWvx8eiHmplvZ5sQ2NqhaluscnKrfyCTujVWLboZTIgotC0VDpWyJQlu/t8SY+P9Kg+PELpF2FNqB8S0jlBbmcV2F44LjJS5Sl88udsuaGQtKgXZp7CMhHdbLGdt0Ycp3FfU7jfqRWSda/rZBEaKIL5atF1JbD9iYUmPJ5bBhAJVKLJ7anrBxHlJQg4UFrKPBcWKxqHX825o4RykKsRensCffhEsXMaViLLBKJWypCNMXYW62uW4/AxPbsKPjmD1XEQ0NEw6NEOXymHpYytGgHZTjYBwP7cbbV76Pdj0iz8O6HkY7KKUZGRnj0uw0LmUKnKdgzpM358mYGXTLvX5NFbC4SRhNNX4b5RGqHDU9ilE+kfIJdYHQGSbUw0TOMJEzBNpF6+b+ynsOWqmGgAkTQdXYV6r98hhHSCyteXNRIsDWeg+klILIEoTxGErhJG+XHmWycjj5bjiN0I1WTnwhVjlcncVVOXydI+dup+DsZMjdha9H2y4ExkYEZp6qmcfYsBHyiZ2pOMsnfg/1zB2DUwmZLp1mIbrIQnSJSnSJyHbzZpZ8V3FYyqpYWEaaqOpiah4mcLG1rTgqS+s11gIah3x2hOH8KKNDWxgdmaCQGYqFkA2ICIlsLHoyaoRgwaM8FzEzN8fc3Dyu5+L7Phk/g5/x8X0fz/NwXRfP9XBcB8/1cD23GdYBnIZ3mRyNGhJLk0bOUoeCoSFcYs0cfx9MT+GbXICTkGSrE7P4qG/9fllrGRsbZ3Z2ZgX7IC1iIVUPR7eFgKGLeGp/bfxrbZP9WyL8yy9rLVNnIqZOh5ioHnqkqcvrQir5jYJq2bBQtATV1RzLF3o+W08rtfXverdlGqfpRNC5CsdVaLflMaf1uUQUtixbf851iXM4lyGKLNWypVIyLJQstQUTu4NJqFwnuzioxZ9LrWIJKpagZtGOwvMVrk/yW+F68Y/j0fy/S/I7fvzdIjpFjKXBUlmti6hntrQtZyws+rJZa5vxexsA8WK9IiBNP7x9PbFr1RL6WxbbTCanFocRT76Jfes49p23se+cjEODdbSO3b3CEOQL2APXY7Ztx2zbSTixAzMySoSKE3dt/W42OTHHFkp8YbERGTWLTlylZojN4tkyrp3DM0U8W6ZQLLMtOImfTJKI8Ck5O5n3r6aix6k5Wwi8CZSTa4YHFl8Pu7hVrXS7DJSD3uK5np/Q8k/q2MTBiuIrLsXwLG8VH2WyehiNy87cPXgq3yaeDCGhqRLZBUJboRzNM2tL1BaebKzXVTkK7g4iG1Azc9RMcVXvQeOTc7aQdbZg5scpzdbiuQvGYsL6b42teZjQxQYeJvAIKhH1L7Dv+4yNjTM6Msro6DDDY8MMjwyTz+fbxHFXlMLRza9X47qe5F5ZAyYDZtSyd1+fb6olxFT/0wJRIyev9RBedCy3/Nv5OA1h5vSpGoyxjWk8yw9b9Z1vsxri28F6Xp9ucbDquX8rwxjLzGTE1On4nDe+w2Vsu4PnD+5NWGuZnow4+WpAec6QyScTqBY5gSaCsGbjzz+ZX5PJKca2O+QKityQJlvQjftXnUxM0bo5D8dENpnPYynkhpm+NEcUQhhaoiD+brblPVma3y9Nkh8YPx6FybpCm5ziY1cuDCzRQutj/X8WDWHmKxwnGYMhcXnjz6BWsR2vgaZrV0c74GcUXkaRH9G4voo/wyAWZtVZQ1izhMHy49I6dg4b4/NahKcTb8tJHMfWsdQ/b9PI9qinI8Sf62IRa+bm2Laz/88rbUSMrQeNb6mhQ1LZLo+1PKdsfDSZRYn93eRfI2ehVo1drjDqsmQfw42iWHy9eBj70gtxeBFgYhvqwHWofVej9l0NEzuJsj4mjFgII6qRIaq/Ve2A69T99E41RBzOHQ5PMhYeYzh8C6dLqLBjbCgiZ5iis4MLzp2UnF1U9JY4HOi4uNpBOxp3E87GM0lOV6uzQ3I3qZ2lx2uMJQirLIRz1KI5qmaOmpljpnacC9UXcVSGq/If5KrCB/D1cN/jCUyJYniWUniWYniWcniBjM4z7O0lo0fJ6BF8PYxWftNlSy60cRJ1M0SG0kyM7GahFKGU4uzZszz27ccYH99LPp+Pk4ZdB8dz0L6GfPtY/IxPIT+OtltZKBaYn9ZE84qZkmUhC7M5SyZnyWTB9WwcVffAdS2ur/CSv51Fn2M9dTEMwETxydfz4rkW9dCPUgpjVZyaWFXUauBoFW/HtTiuxXGT9MVqMqG1qghq8fexHnqqX4DDAGqVeP5IrQLVBRvf29j6RdY2Lgjx65o/jge+b/Ey8URrzwfPt8nv+D163upNoyiEcgnKRUW52PxdrcQmdiYH2azCz0M2p/Gy8UxtPxtfeJVemeCKwtgdqf8o3bxIe5l4f5RmDRdOhVw4HRLWwPXjfTJ1JnZVh8c14zschrc4+BmFn40/77WEQ621zF8yvP1qjflLsQi7/s4ME3ucVa3X0Qo/SZiH7oZcnPcWn4/HxvO4mQWiAWcJxQn4cdpxXbBFiYiLcyktURALwrBm4+BHzRJFsSGg6iJQxeeoTF6RLWiyBUU2r9uEciO/0rJkyLXb+KIwFmlREIuzMLANIRn/n/i5sPn/WtXEN3eRbUxWb6UuYLWjGi5j3WGuO471ddWzdU65F/n4D46m+OmvDBFjA0AtFkDGNB7vOPQiE3vN3dZjm7cNHWGMJY5hW6vCQrn7k8tggwDzlS9in30yDj16HuqmW1CH7kQdPIQaGsEoFc+oUVANIqpBGE8Xdy049StMl5O0NXi2RMbM4ptZCtEZRsK3cAgIVZYZ7waKzt74fROhbIQmzuMJVZ5AFwhUgVDlGR0bZ25+Pp51pjS+1o0ZS5sRa+OTTLfQZF0QENokDAWBLTFbOcVM7R1mKicphqeoMd3xWlfluKbwXezLP4SnC4QBlEr1k1D8tapf7Lvh6QLj/nWM+9et4b3F6YfzM4pTk1nK5QpBEPLCS8+Q8Ye57qqPkStoMjlLNgfZXHwMVBcUlTJUK4pKWVGcVUxOxvsvk7Vs223JFwzViqK6AJUFxfyMIop67GMVpwc6Scgldgra79jrOC6xA+AqgupSoaZugcCen0bbX35WkcnFzoCTlPloXBCSMK81FhPVf8eZAMXZWBBG4dLbdn3IZGN3JpNT+NlY5CQJcNRjuBfP1rh0UVEuKRZKUKs016kUZAuQG9aMbVcE1Vg4Tl+01E7HIeiY+LfW8XtyGyGm+DN0XRXPBg1I3A5LEMThqeWcD6Wb2Rtbdjhs2+cyts1BKSjOGqbPR0yfj50raK5MO+BlVJuw87NxGEw7SdZF1TbCZWHNLnKR4vV4GcWBW322X+W25MktCgEnF+/F52BHKXxX4Tsad4WhtJGsRy3nYpJ0h7B+o7aIVpO9ETW1tjEnarnk/3j2ZHwu8DKDPT+uxolVKvkeeWsbW931hsQVX8FArIm/D3u2TRAGc2sax1oQMTYQFguneoZEt1IWXWZZJug2MbbMFu0ahdjsNNF/+jycOom69Q7UbXehbjwImSyhn40LAhhDFEXUjCWIWvKlktsN15TJR+fwwjKuLeHZEq4p49t5fDPbls8VkmHGu45Z9zqKzp7YbXE0GNv15KKUQjsOvnbI5QvUgj787ZRpTa1r/sT5FvlhS75AwypvvqYpxKw1BLbMfLHExUtlZuZKVM0cKnMJMpew3iUi9xJGNyc5qOo29MJ+vMoDqHAUbUYo+MMM5YbJZfNUL2neKCkWyoqg1v0EpJRN8kWSNEDH4tSrTtBiWjYSmVrCJcQCx8/E7kz990JJMTcTC6TmdgOU0kwXX6JaK3LVzu9i9pLL1LneJ0bPt+SHLNv3GMa2GgrD9fwWHbuJScja4BCGmjAwhLX4rjmsWYIgFi6tF1pjbJwc7TVP9tqJ74YbgqEWLzs0qvGz8cXcz8UXdhvVw0dNR8Hx2i/+dVegHoKqC2vHjYVRPzk47V+W+CakXn7DhoagZghqilrgEASaoKZjB69iqVYslbJh9qJZFI5qlRJVXC8OoY1OaHJDilxBkxuO3Y2l8nGMsdQW2l2tasU0BFZQtSwUk7BYkBjhXlOk5Qqa0a2q8blmcjr+XC1JLpFpiOBsQTOx28VdFI4cHnMYHnO46kaoLsT5WfXX1ZLXBlXLQskwd7FT+Ckd5ybV91Um15In5SkyWcXEHhfHTVJClGIk43RMcmn7XOruD6xYgHVDq9hR83vmoPTGNvIOaUwKqo/TJMKtm5gcNI5WeIlj6OrYooiSsGeUOHZhMnlorah6jcLVvFbHeWyFIZfZznvedUPE2HrQuDXv8qXr9kW0AGGbeOuQcYv1XrUCQWXpIRAfpJGtHwzxgaEUOG8fx/v//59xfOVHfxx7y21YYwgtRDioICSylloU38F1rtwyHr7K7srjOMndq0URqDw1W2C2toX54Frma2PMLowzUx5nvjJCM6EnDna5HoxPWMYmDMOjzbKOjuO2OV/rbYAZA+feUbz9xtKCBwAVC7L8UOwCOa4lUkXm3WeZ0U+zYM/QqPaYS34AZT2caBwdjuMVb0MHW8mzj2FvL4V8jtz2eF3loqI0pyjNKy5eUhijcFxLLm8Z22rJFQx+zm3LKTGRiW380BJFqjHxNQxpZDh3pLglzk1991TKilo1ro3Vip+xjG6xDI9FjIxZtu8Y5dTpk3zpz1/mmv3XcP/9E0BIFNFwt6oLifuVs2RyGj/voBwf0HGVFc/GaYiuwnfj2mIRujFpwngQZbuXLtgo4mtA9+9F/YJUv8C35Z1Zkpms9d/JTNz6i/0kJFuA3DJf+rowbZ/VGJdQKJZmVvyetFZxKKpzsncHrTmu/bP0ldNLpg0HUfOsl8lpMrnerzMmFmfGxCLMcVfmkAz5vYUYJDOZN5kBr5TCUb2/h3XaRJpthk6Nia8x9cfr+7Qh61W9pEjs3jtKNWZoW5r5svVD0m35zjfGSf1mdXEqgSVMjoF4UlT7ON8tiBgbBEuEFLs7Y92/bK2uGPT+UtpqFUwFdPyFrkb1ixW9SxRYi/fU43hf+iPs+FbKf/0fEE5sIypW4pmOmRwEvWuXuabMnupjjIZvUXR2cy5zHzU1QkCWC+cynHwjg7UKpW2c++JZPN8wPJaEbhVo7aCUpVpRnH5bcepNF8exjE3E4mx41JIfXjKa28FCGSbPKKYnVb2MGk49z8ith3diwZTJNpNQu3w8XJyEN1/VLJQU+W3n2LJtmoyXJeP7ye8sUagpzUOpaFkoWeaKlsmFYwSjTxENvwTKoBf24c5/hJw3xMhwnvHRArlMgYwewdNDPS4azX1XGLZs22UbYwuDOIeo/lKLR7RUpwRbL3RqEvdlZfmE9RyLoAa1qmrkb7UOWyl4+qmncV2XO++8E4uHVXGx2+yIIjccJ7xbrbG4bTlHjWn/SjHkO3hOU7y4dHchmmU5aHZTSOEueznq+UGejuu8xTc28W9j4nBz3RFY7uLeSt3hCBe9r37yiuoz0RbjZzSUurwgRdJMD3C0YigTz1quhIpyYPoW3VrH7tdqyHvN79yVTF1M9jthpH9Wvz6lFJ4DXpd1LHb9LIlgbBGPV4poEzGWCt3S6bv93TKVo87izEMAG6JpF2PWQlScj8t8KrCBaiZc1AJCbSlH/V+M1Pwc3pf/GP/Is9Suu5n57/3rmGw2nnUZx6OWXcdIcJy91a+jbcCZzPuY8m4HFbsvb72e5dKkx8h4yP6bKni+7ZzRqDW+78fVwJPPJwph5iJcuqC4NKm4eL5ZUTs/BIURy8S2KgaF79skVyZe4sI5xeRpxdx0vKHhsTjVuLIA4bxKElVh8YnD9ZJ8pnyc05TNxx/B6bfideUKloN3BbwQ/m9MmXJcyHAh+Wml7nZNxH/6eoRd3geYcO4lk9+JuzMWf2mgVDzGlkeIVI99lsx4S8wwsBZtq2hqS79m0fbqSfO5Qvfv2KuvvsqFCxd4z3vvx81tJdIunt//xdrRiuE+3InW5R1Um1liEkEWJcaebtzBx+Oo13eLEtFjbHLH31IpXCu6ugaupkt+kFoiyWDlNByOLu+/NTzWKtrqP2vZpqcVnqOS/KVmDbz1xlHx/q8nwWfd+LMu1qKBiuyMo8l5ayujIQyGfl2/+jHRi6ZzVxd1LceVTfemYjWIGBsAHbu07cS2aPakNYumQhocs9BRW8FWK1SqRbJ+HNKpX0OrJg4fKsfi9HM+KRXxH/8q3lOPQxRReuDDlB/4cNOlcL1lhZhjFthdfZzx8A3Kehvv5D5M1dkCQLmoOf5SlsqCZs81VXZdXesaVtSOg+93Fph1XNi6A7buiHPHKgtQmoPiXByim7ukuHCmxlLlzPNDlmtuMGzfEwusxVgbzxyrLiTJ4wvNEFppHi5ONsNxnm+57hbDrn2W+fAMwbky141+jBF/H5GpEtoKoakSNwmJDX1r4ryYvLONcf8G9GoTGVZIpLIrm2KnVNwVwLo4tkLvOu7LU6tWeeKJ77B9+05uvPmO2D1yNY4HQZJj2Es0eI5myNdtBW1Xg1bx7MalBFJdwA2m/OfgaA+PJaHe5K/WME/dNYiWuTh5yf5pnQHYSqtobbh+A3QetVIMd8nXcrViNONQDgyVMP1Wb45WFHwRYpc7ddHW59JdH00jB3AtiBhLhWWcsRYx5lAlsrlF7pihrqQcW0mWX/R8tUoQWXSthu9nqZk4HFm/g132kr9Qxv/OY3jf+TrUalQP3UHp/R/GbJloX66XELOW0fAYe6qP49gq5/17OO/fDcohCmHqnM+pEz6OC7fcEzK2VaNU0wpq5nz196VXCnL5+GdiZ2MuEUOFUS5MzlKtJqUIKnEe1Pg2y9BI75wypUhm9UH7fmqG/+ptLQtDzRDmpepxAPYU3kvOHe/8aEw8LXyleUz1+lfd0ErhKYXraBxNLGgiS9CojVEfedJyJrmwWpJ6VC0JvNB5CjLWYpVLSAFNFW1rjSVt0v/R1juEt5VVUW1r9JTm8OG/JAgCPvDQh8h58bfRdRWuE4cg8GjMHDMN1ypeh17CDRL6o1eYZ2vBxy7EX+LWr9lyn3cv0doWHra28V2rh5BWM/5e+VpKKQq+Q97TbWHcuuhcbf6gSpy4jXZEBAFEjA2IpcWYwuCyQGhbnIygCk4ObasoQjoum5UFjInQST5YqVwlMu27bslzURjiPflN/G98GVVZoHLTrZQffIRoYkfnsnGBn66rcU2JPdVvMBq+SVlv41Tue6k4E9SqivOnPC6c9YlCxdiE5abbTZyrMiBcT5ErQK4A9c/aRHbFU5q7oVSSU7YonDhdOUHO2cJYZmtj78bTy5Np/OHKLgae0mS8OAxjWi5sNnEeXEd35LBkXEXGjS88gal3ZdBodwi1eBpnH7Q6KpFxiCKfuJ+m7vGFaqKUIudqXnj+SY4ff4MHHvgA4+NbGs8vLqlRnzkmrB/1JtDxH+mss1t4uJWl+pnWn1ucY5f3O7/r3egVxq07eaGpB6GS1/R4092SzAVhoxAxth50FDqqC7JcLMiCGpgInbGdJ8xa3L7IWEMtqjctDuMaWz2qkVtjcI4+R+arf4Yzc4nagRsoPvwxoh3d+2IBWM+nNBtPn2/OqLPs1C9zwHkcRcSJ6P28E92JLWlmLrpcmoxn723dYdl3wDCyyDRyHc1Qckdbd2rCMBYS2o1PlvVaOattE2SieJ3QrK21WJRZm8wurPd2W6bQ6uLXTldPsCN/IxlHNypT19vg+C4EKp7x12iQu8T7cJUi47YnC8ehte7ORjdUXdQ4xKHGVQix+nraHZXmelpnR9XbPoWRjcvyWourNXlP8+KR5zh8+Fluuukg73/fQ0xPT7esf1XDEi5zes04HFSIuOHkrU9WgCCkjoixAdBW9HXJNAfbdMjQuOE82HgGY+PVkYFqHDpqVG8n9oLCMMDr0tTbRBH2zTfIP/pFvLPvEGzfxbm/9g+ZHbuObDZkyHbmcFkL03NDnHu1QGm+eTabyJ/l/n1fYfvQWc7N7+Wbb3+cuWqL8+FYtu8J2HdAkR/qFIZKKfKubkvCDJN6SK5WYMH3mvWYbFI+I67k358os7YpxCCpKB01RZk1cQizVeRZ21lodSlXzVrLfPUCNTPPhH8dYdB9XJ6jGgKr2wwga20yC2/ljmGj4nnyF5g4B1+5WO0v/cI1UJ/WTqsTkVxF631Sj750hKee/g7XXXsDD7z/4bbPLw2XUhAE4d2CiLFB09PpiQWZxQFMLL4qJXALcUJNdYFGGG7ReoyxRFGI68ZXSDtzCXX0GfIvPku1aDm9+x4mP/ITzOidhDMOzMSvcxzD6GiV0dEKIyNVZmaynDs3RLXqkskZrr6hwthwkX18h+32ZQJyvBo+wlTuRvbcothLqVHDKJuz5AqZJR26nKvbwgBhELfZaHn71AKL59GoIxaH4jRBZKlFhlrUO2l3qQrfdVG2HMY022GAbRMhNnHqLlXifLGtmQPLr5CV1f1ZCouOxZZyAae7zbRB07kdrXjt9Vf49re/wdVX7+fhhz/c8R1YayK+IAjCuwkRYynQ0f6o9e9+cm9oUQ3GxM3jfKe9sbdpESXJKsMgIPfyYTIvPoU5fZ5zO9/LqRs+zVwubivkORGjI1VGRqoUCjUWFjxmZ7PMzGS4dKk51TA/FHLttQuMbwvZEr7Crsq3cagx5d/OWfduApMhpwxeSx0jpRR+Zmkh5juajNt8Lgjiti8dWAhqtqOvYOw0OWSNphoaal0q80dh92r9a6G92XfMbPAmvi4w7KXbRdaiQDktzpfCquT//czC3CDBc+LEMb7xjb9kz559PPLhj6G7hEmX6+ctCIIgNBExNhBaxdhqpmMbaJnGbYzpnPdnIkb/4r8yfwFeveqvMHXtLVilyedr7Ns6y9hYhUwmarte53IRW7ZUsBbKZY+5OZ/CUMjwdh+lYTR4g32Vr1F0dnM68xAVPd50npKQm1ZqWSGmlWqr2xPU4hlXvaiH/xY3ena0Iu87ZK2lGhqqiaCLncGl1+mouChnaCzhqvZBk9naCbZmDqQSdrOAVV7ieLmXXWLVG8de47HHvsqO7Tv5yHd9AmepvqoixgRBEPpGxFjadFTfX7tzYxaLiTBk9Iu/zzH3Xk7f/hCuG7FtvMy27WXy+bD7SlpQCgqFgEIhAD8LGrLRBfZVvkbJ2cmbue+NewEuyo+yISgfPH9pIQaQ95r1oqJweSHWeFuBjYtzdpnhFAs8h4xryXsOM5FKWsq0r9tTGt9VeC25WZHRcchzib6XvaiZOcrRFNdmH1jR6xZjUVidTSrPX14CrM7Rl47w7W9/g9279vCRj3w3nrd0KvZSPQ8FQRCETkSMpUAjrKUafzUxq3BlOlbR8kAYMPQnv89LQx9hauI2duyYZ+eO+bgp7kqnKWkNrodjFrhm4c8JVYa3sx+LhViXopHGgutlcHokoWcc3RBCi5Pr+yEILb6/9IVcK4W2sVsGNGYwGmvxHd11qrqjFTkdu2u1JBetnxYzAMXwLQC2Zq5Nam4t0daqBxYwOgfq8jzcrLU89/zTPPvsU1x99X4+/KGP4rpLvxcJUQqCIKyMy/PqsKlZpgDsKmgk79dqZP/bf+Xw1h9kbuRq9u2dYevW8spXqBW4mbjaPhFXV/4C1y5wPP9XCXU+3mYXg833fNQSle+VUmQcRXZRnthKsSZ2yLr12QOo1Qy5bEsR3RXUCmqfIGCohoawlyhTirngBI7yGfP3ARqjM2hT7pIn2OM9qexlLcS+88TjHD36AjdcfxMPPfShnq4odHc2BUEQhKW5PK8QVzht+WH1Ogy1Gt4f/THP7Pok1dw4B/ZPMzJSXdmKW0VYcr3cXfkWQ9EZTmYfYcHZTs7TaAutWsxacDwPi0NobEetR9/RZBfNnIwiu7p0ueS12ukMddVqZtXrXIyXOHhBZKmFpqOqPUBGKy5WT7Alcw1aOVgFKAejczimPxFslD+w8hODxlrL17/xVV5//VUOHbqd++97oK+8OXHGBEEQVoaIsZTpcEzWmDNWn92nnjvC03t/DHyf626YJp9bVNNhmc04mSyR47VVWthSe5mJ4CgXvDuY8W4g62o8rQhqlroOUkrhul4jP8hYCw6ESTL/YhFWH/NS9bi6v0fDVPU4xeACO3O3kHNHCQKL39JkOk0h1kp91qa1tq3xcgRoXWU2OM3N+Y/F46x/eMrFqCzaVnq/Lxxsr+bdm5zXXn+F119/lbvuvJe7735P3xMYxBgTBEFYGSLGBs0axZixBoxh9kxEsG+YQ4cmcZ3lk/Tr+FrhDxVwPI/QGMqBxRrDRPACu6rfYd7Zx9nMfWTcuBRFa/kJpTSZjN9WukAndbhymaVLL/QbniwFU5wsPcXbpacohxfrW2Vr5gB7C3exb+gOhnNjAxNircTV6JsVvK21nK+8CVi2Zq5tjK2O1T7GmJZ+ju1YdJIndnkqk1KpyBNPPM6uXbtXJMTUEhMwBEEQhKURMZY6KSTwt73c4p94nTlvNxm1QC4XEnS//mNtcjFUiQhzNSpXQCXlB1ytGXZrjM9/ldHgDWbd/byT/TC+5zZyver5adpxyfidFf4haQVkbdfnWsOToakyF5xjIZwmtBUCUyE0FQJbYbp6kqnqG4Bie/YGDo5+N6P+Hs6Uj3C6/BwvXPp9Xrj0B+zI3cy9E38LP8llWy+UUkxVjqPQbMlcU3+0bRmrs1hjUDYWx8knByiMbuk9eplhreXxb32dKIp46MEPraikx2WqPQVBEDYUEWOpYxf9dxXOWGs3JWMoPPsdpvf8vyiMdWnAm8xqc7BkHIPnxDW2cBzI5lEtCTxuNMO22S/iRRe5kL2Ps+6d+J5DriXp3lqF53ltZQsq0TwKRcYZajwWRY1Nx3+biHPF1zhfPsZc7QxzwVlK4cWu718rj4KzhYNj38NVhXvJu80WS6P+bm4e+xhztbO8U36W12cf5YnJf8/7d/wEjkq3q521hmI4RTm8xNbMflzdHlK8WD3OmL+v5fFOpWFUDpSJn7tMxddi3nzzOG+//Sbvfc/7GB0dW9FrpaSFIAjCyhExlgYtekMt8fhqVmetRV+awpy+QOXAFrYNzbYvqBR+LkfG1SgFrmtRxuAGl3DcEKJ5iBSgcKNZthS/BsDk6PdR8a9hKOkx2NiusTjabRNi1hoeO/uvKEfTbMtez578nezJ305GDaOA6cpZ3pp/gndKT1OJ5lBohrztjPn7uKrwXkb8XRTcrXgqi6tzeDqD7mNm4Yi/i1v872HE28nTU7/NM1P/ifdM/G3UKgWPtZa54CznFo4yWzvNfHCO+WASk0xVGPF2877tf5+8G3c6j2zAdPUk14481PZ5d6AUdExpuHypVBb41re/zsTENm699Y4Vv/4K0aOCIAjrioixtKjHCFfYCqkXxhqyzz3B9Nh1AAwN1dpW6WUzZFsq3aM0OXuG7aU/6toRseZMcGH0uwmdUYAuifd01I+aqh6nHF1id+425oKzHL70BQ5f+i9sy15PYBaYqb2DQrMzdwtXFd7DzvwtqTpY+wr3UIlmeXH6v3FkeoTbxn+w79daa7hUfYszCy9wpnyEUjgFQN7ZwrC/k+25mxnxdqJwOHzpv/DYuX/F/dv+HuOZq5iuvo0hbMkXa0ngv4L5zhOPU6lU+cTHv3/ZEhbdkJmUgiAIK0fEWOqkJ8ZsrUr2yDNcvP3HUcqSzzdnUDquJptdNFPPGsaL3yTSw1wc/lBjDAqLRVH192B7CCWlnI4L8KnSszjK556JT+Ion9ngNKdLhzmz8AKO8rht/AfZW7ibrDO8pvfai+tHPsxCOMux+a+Rc8Z4z/hf67m8tZZ3Ss9wdOa/JW6dw/bsDdww8gi78reSdUY6XjPm7+Xbk7/FN87/G+6d+NvMBWcA2sTYapt+Xy68887bvPHGa9x55z1s3Tqx4tfH+YpX9mckCIIwCESMpU2r/lpj8r774vPoygIzW26kkAnQOl6lUpDNZjqiZkPV1/CjKS4Mf4yKf/WKt+c47V8HYyNOlw+zK3drI29qzN/LmL+XW8a/Z9XvazXcOv5XqESzHJ35Y7aM7GJC3dJ1uYVwhucvfYFzC0cZ96/h1vEfYGfuIJ7OdV2+zoi/i4d3/RTfmfzfeeLCvyfrDDPs7STjFFqWujKFhrWW06ff4Rvf/BpjY+Pcdee9q1qPdq7Mz0cQBGHQiBhLhY423sl/VynGLGAt2We+TW3bborhMNu3lBpP5zwH7bUXElU2YLz8HaruDsqZ61e8Sa10WwkLgMnKq9RMib2Fu1f1NtJEKc3dEz9K5fw83zj5vzPuX82u/CF25W5lxNsFwNulJ3nx0h9iCLl1/K9y3fDDK8oxyzojPLjj0zwz9ducWTjCrtyti0eR4jvaeKy1vP32mzx/+FkuXDhPPl/ggw8/smTz7+WQ5H1BEITVIWIsZVQKYUoLqFNv4547zenv+ttYqxgaiutZZB2F9jpdsfHaYVxbYmro46uqL6Adt2PCwanSc3g6x47cTat6H2njKI/7t/8dTgdP8ualp3l55r/z8sx/J+9sIeuMcKn2FhOZ67hr648w5G1b1TZc7fPebf8TJ0tPsz17Y/uTV0gIzlrLiRPHeO75p5mevsTw8AgPPvAwN9xw86qFGHTmIAqCIAj9IWIsLcIQ/MX5WKvPGfOe+ibGzzC983Y4Gyfvezpu4xMt6gjumBLj1ecoeddS9XaveFtaa5TSbdoxsgFnykfYk78j9ZISa8HTOe7Y8Ve4JvMwC+Es5xZeimdIBme4fcsPcWDogVXPuKyjlObqofe2PXYlJe8fO/46X/vaVxgf28IHP/hdXHvg+lUl6y9GZlIKgiCsDhFjaVEtQ+CAZ5qf6iqdMVUq4r10mMrt72G+ksf3Q3zf4GqF7eKKba0+iSLiYu59q9qe4zhY277ScwsvEdrKpghRLkXOHWX/8PvYP7y6970yrgwxVqks8J3vfJPt23fwfd/7g6mIMADtSPK+IAjCapF72VRIRJeJoFKChQUIV5+8nznyFCqKWLj7Pkolj6GhZBal1nGT7xb86CIjwSvM+LcSOmMr3pZSKk7c7xKizOhhtmVXnn92ZXJlCI0nnvw21WqNhx78YGpCDMBzr4zPRxAEYSMQMZYyCgtRGIuyqP8ekg2sxX/+OwR7r2FhZBe1mkuhEOeLKa+z6fRE9dsYfC7596wqKOq4na5YYCqcWzjKnsIdaHXlFDRdG5tLbNRqNc6dO8P09CUqlQVsHy7smTOneP31V7j9tjvYsmXlpSuWwnWV9KMUBEFYAxKm3GTok8dwL11g7v6HKRbjGZP15H3cdgcrF56iEL7Nhcz7MDqHjlP/V7Q9z/UIw/YL+dnyi0Q2YG9+84Yo1xu7CUJwYRjyzjtvc+z465w8+RZRFDWeU0qRzeYYHh7mvve+n507d3e89pvf/BrDwyPcddd7UhuTUuDIWUQQBGFNyGk0dVaftG+txX/uOxg/Q/WmWyme85vFXrWDozVRi3DaUn2aUOWZ9W9b1ZZdz0VrjV1UguNU+TlyzhhbM/tX/V6uPDZOjE1dvMDRo0d4883jBEGNbDbHTTceZO/eqwjDkIWFMguVBRYWFjh9+h3+9L//EXfcfjd3331vo1zJ4cPPMjs3yyc+/n0dXRbWguspyRUTBEFYIyLG1pkoDNGO0/UCFhbnyL7+Igu33IX1fEoln0IhLvaK2x4uzIanyUenuZB5EFvv9bgCNeZ6Lr7nY4xte10tKnF+4RWuG1lZja4rn/UXHPPzczz9zBMcO/Y6nuex/5prue66G9i9e++S+V61Wo1vf+ebPH/4GU6dPsmHPvgRjDEcfuFZrrvuRvbuvSq18WkNjhR6FQRBWDMixtJkmbwdYwxBGKKiENdx0I7bEGVhGJB56Xl0GFA+dC/GQKnksX17UuzVcVGquYmtDVeseyX6Xriey2T1ZZ45+QU8nWdn7hA7c7ewNXOA0+UXsETszd+14vVe2ayf6KhUFnj++Wd46eUXUUpx++13ccftd5PJdOYMLsb3fR7+wIe56qqr+eY3v8Z//cP/m0JhCM/zuP++B1Idp+eJEBMEQUgDEWOpspwYi3N8rIUgjFCRwXUclIIwjBg7+jTBtl0EO/ZQKXvNYq9aga67aZZseJZ8dIoLmfc3XbEWlFJLJ3Q7ES9c+mOOz36L8cw+PF3g2NxjvDH3VTydQyuPgruNMX/fWj8MoU+CoMbFixeZmppk6uIF3nrrBEEQcMP1N3H33e9laGhoxes8sP86dmzfyWOPPcrpM6f4wAc+TC7XuyXUSpCkfUEQhPQQMZYC/UYHo0W9Kq21BGE849I9fxr//GlmPvR9oBSlUkvyvnbbBFacK5Zj1j/UdSC+7+M4DmEYYowhiiKstRTNOZ6e/D3ma5PcvOUj3Drx3YRVTWAqTFZe5Vz5JSYrr3Lt8EOSB7SItBP4py5e4LXXXuHUqZPMzs40Hs/lcuzdexV33XkvW7ZsXdM2CoUhPvGJ72dmZprx8S1rHHETSdoXBEFIFzmlpsrSsswYgzVLP184+gzWcVm4+Q4AikW/UewVN4MidtQy4TkK0UmmMu/DdqmMH9cNi100L6nUb63l1Utf48jFPyajC3xw7//MzsJNcb4YFk9n2ZO/gz35O9bw3q901i7GKpUKx46/zmuvvcLFixdwHIc9e/Zx/XU3snXrBBMT28jnC6kKYaVUqkIMwHElaV8QBCFNRIyliOopxqIlnyMIyL3yPAs3HMJm80CcL1Yo1Iu9uo2q+1trTxOpLDOLXTFiseZ0mRwwXT3F4an/yu7CIe7b+TfJuEPJmFbw5t71rF58lEpFnnn2KY4de40oipiY2Mb73/cQ1157A9lsNsUxrg9raF8pCIIgdEHEWAr0c5leHKJsJffGUXS1QvnWewEIAk2t5rJjRymOB6l4G354nkL4NlOZ+7DK77qubrPsjs9+K26yvetv4Tv5xuNWxNgKWLkYC4KAF448x5Ejz2OM4aabbuGmmw4ysXV1Tcw3A664YoIgCKkjYmwdWC5EmX/xKcKxrdT2HgCgVG7JF0uSc5RSjJWeJiLTqCu2GKVURw2p0NR4e+5p9g3f0SbEAMwaaqK9++hfgBhjeP2NV3nmmScol8sc2H8d73nP/YyMjA5wfOuD5IoJgiCkj5xaU2SpMGWvEKVz6QKZU28y98DHqMciq5V4t+RyIbhxOYNMdJFC8CYXM+/FLOGKqS6u2DvzzxOYCteOvr/LwHq+HSHBAh3d2ZfAGMOXvvSnnDr9Dtu37+CRRz7Ozh27Bjq+9UJcMUEQhMEgYixVuouxKFpa9WRPvAJA+eCdjcdqNQfXjXA83RAB+egMAHPejUuuy+2SzHN89lsMe9vZlruu7XHTw6kTFtN/8dvnn3+GU6ff4f77H+TQLbddUeJFXDFBEITBICXW06CHrjEm6tnE2SnOYTwfM9wMYdVqDr4ftVXdz0TTGDxCNbz0uhy3re7sXPUcFxaOc2D0fR2iQPLFVkJ/gurM2dM89/zTXH/djdx66PYrS4g54ooJgiAMCrnXTZVO0WWWmbKoS0VMvr2oZ7XmxjMpHQ9rLdaCF05Tc8ZAKQKzwIXaq5yvHmUmfIfbhv8623I3dFwsj89+G4Vm/2hnY+iNNsYWb/5yv8xXKgv85V9+mZHhER544AMbPZzUSbGdpSAIgrAIOcWmSqfC6RWiBHDK85hC0+2yFoLAwfcrBJFqrNKNLnEUxesz/4FLwXEshoweRqF5Ye4/81D2p8iSbSwf2ZA3555kz9Ct5NzOxPFebt2gMSqD1Z2tfZSpom11A0a0DMs4QtZaHnvsUSqVBT72V34Iz+ue03e54jhItX1BEIQBImIsFWJhs/hytVyIEkCX5gm3NEsdBIHGWoWXaTbwVqbK14OTvG4WyDtbuSb3IDsytzDq7qUUTfGdmd/guenf4aGdn8YSF3o9U3yRalTk2tH3dR/xBoQpLQqjc9ClhROAVT7WBqhNNrPAEnc/ePvtNzHGsHv33rb6YC++eJiT77zN+9730GVdtmIpHFeEmCAIwiARMTZAlgtRAjilYqOkBUCtFu8SP9cUcaeqX+Rts8At2XvYO/SDbeHIIXc7t43+Dzw/87u8OP3fuHvHDwFwbOZb5N0xdhYOdmyzV5mNbtSlZq+ithawKgNKo2wANmwTpxYnEWI90hSVwuoMyiysaHz9Mj0zy6kzZ7nh2gNkMv27V5GxfOs7j/HKqy81Htu2bQd79+5jdGSMp57+DtdcfYBbDt46iGFvKFqDFldMEARhoIgYS5V2sbJciJIoRFfKRC1hylotTtrPZOOm4Gdr3+Ht4Flu1QWuzn2IsEvIbG/+Luaj0xyb/xrbcvvZPnSAc+VXObT1Y+gu4mfllfcdjM6jbC3+WfQ+jfIaQgyI2zRZg7IhyoZYpbAq21d5CKs8LDUUPToWrID5YpFjJ97i2Ik3mbo0DcDs3BwP3NeZR9eNaq3Gl7/2VU6dPs0dt9/F1Vfv59Spdzh1+iSHDz+LtZahoWEeeuhDV2SCu7higiAIg0fEWKo0RUqceL9MiLJcAsAUmgn81ZoLWLyc5WJwlDcrX2S3s4MHHZ8TTveioY7jcGj8+5muvc3TF/4zexZih2b/6P1dl1958r6KXSuVwVofZQOUrQEKozPdw45Kx2FHVp4/ZXQGx5RX/LpWKpUKX/7aNzh99hwA27dN8P733su5yUlefu0N7rrtVvL5XM91zM0X+bOvfJXZ2TkeeuhD3HRj7DLu2LGLu+9+D9VqlbNnTzM+vuWybGu0HErFsygFQRCEwSJiLEXawnJ9JMg7pXkAony7M+Z5hqJ5i9cXfp9hZx8f8PYQUgTVWUdMO06jBdJ7Jn6Mvzz3OU7OP8vO/M0MeVu7bnelyfu21fFRatUiq2+Ui1UuyoarXsWRl1/l9NlzvOeuO7j+wH5GRuLP+Op9ezjx1kkOH32J973nniVff27yAl969GtExvCJj383u/dc07FMJpPhmmsOdL74CkGEmCAIwvogdcZSwS76DfQheHQixsyiMKVTmOTV8u+S0WPclPubDDFHzRnvvo4WgZZzR7l/+/+Er/PcuOWDS492xWHK9b8oG5VZdbOmIAx56ZXXuOaqvdx9x20NIQYwOjLC9Qf289Krr7OwUOn6+unZWf70S1/B81x+4Hs+zu7de1c5kssbaQguCIKwPogYGxArccYWi7Fg9ElCqhzM/y18lSFj5wj0lq7rWJxcPZG7jr963efYXbil+7hWVWBsAxwS5SzZDH05XnvjOJVqldsPdf8M7rr9VsIw5IWXXu54LooiHv3aN3Adh+//xMcYHxvl8q+CtnK0lnIWgiAI64WIsVRpyRnrw9fR5XqYMs4ZMyauMVbNvcqws4+s3kLGzqKwSztji/tRWrom7ddZnRbbmK9JHA5dmSAwxnDkpZfZPjHBrh3buy4zPjbKdfuv4ejLr1KptLtjTzzzHFOXpvngg+9jqFBvrP7uEyUSohQEQVg/RIylxSInrB9nTJeKmEy2Ud48CBysU6TmnmbcvR6ArI1nANZ0pxhTSnWKsWVE4Grqi61UEKWG0vHMzBXw1slTzM7Nc8etB3vObrzrjlsJwpAjL7/aeOztd05z5KVXOHTzjVxz1b5VD/tKQEuIUhAEYd0QMTYg+gpTlouLylq4REOxOBhzbwCWEWMdQmw5KQZmVZX3N9Al6TJpoRcvHH2J4aEh9l99Vc/lto6Pc+Caq3jxpVeoVmuUywv85Te/xZbxMe6/9+5FY3h3HSbSh1IQBGF9eXddZQbKotpbfSbwt+aLVWsO0fDLuBQo6F0AZO0MNYa6OkS9wpFLjnJVWfEbd2G29C/Gzk1e4NzkBW4/dHMXx7CTu2+/jVoQcOTlV/jLbz5OEAR818MP4bY0YtwwV3ADcWSOtSAIwroip93UaK8x1s9UQKc0T21Hc6ZetaaItrzChHs9KhFaOTtNpYsrBt0ro/fabL/j6mAjnSGlsei+WiS9cPQlMr7PTddf19eqJ7Zu4Zqr9vLs4SNYa3nofe9ly/jY4gGsfMyXMUpJxX1BEIT1RpyxFLAsrjHWX2JW7Iw1C76W7HlwS4x719dXRNbOUFXdxZjqJpJ6iK2VV97fHM6QbQlVWmu5ND3TEQaenZvnxFsnueWmG/G8/vPM7r7jNqy17L9qHwdvvKHLEhv//tcTqbgvCIKw/ogzlhorqzGmgho6qGFaCr6WvTfAKsacWIz5zKMJqS7pjK1MS6+uOfhmuDg7QIC1lsefeIqjr7zGUCHPgWuu5tr917Bj2wRHXnoZrTWHDt64ojVvn5jgh77/exgbHVkiT2ozvP/1Q2qLCYIgrD8ixgZAv/liAFFrK6Tsa/jhXjxdACBr4uT9bmJMab3iJOvVJe9vvHlqlQMWnj9ylKOvvMbBG2+gVC5z9JXXOPLSKxQKeSqVKjdce4BCPr/8ChcxsbV7Dbd42+8eMSaJ+4IgCBuDiLHUsF3+tzS6VASaBV+rUYUo9xYjtQ83lsnaGQBqekvHjuoaolxm46vRYptCjCiHV984wZPPPs91B67h+z7xUWZnZ6nWarx98hTH33qbqYsXueO27kVe17jxAaxz86EUuCurIiIIgiCkhIixlFCtYqyPyqpOveBrIsYuBifBs4zo6xvLZO00AVkinevYUSudSbnq5P1NIEZOnTrJ1x//Fnt27eRDD76/4d5kfJ8brjvADddduf0h1wvXE1dMEARho9j4GNQVSH8FX5NWSEn1/RlOQJRj1GvOrszZaSpqvKsc6jXjrdv2V5cvBhstxqamJvnKo3/O2Pg4H/3wwzjrntR05QsU7UjFfUEQhI1ExFhqxALIWttnX8oiFoXJF7DWMu8cwyneRCabXBStJVsXY12ukytN3o8uQzFWLM7z51/672QyWT7+0e8h46+uV+XauPJFiudd+e9REARhMyNiLDWaYqwfdHkeky+AdihzgUjP45ZuwvXi17ss4FKlosZQiwRB9zZIHUNp/mktJlpVjHJDc8aOvHiYarXCxz/2vRSGRjakzMamyJlbhNJxWNHz1z42z5fwpCAIwkYjOWNpYFv8k37FWEuNsRl7AoBc7caGC1ZP3q/ozjBltzZIi4bT9poo6mtIS410LS9eNdZa3nzzOPv2Xc34eDLbUTlgw3UeyeYQKkrFoUTHAdUSotYZqAV2VWFoCU8KgiBsDsQZS42VOWNOqUiU1BibsSdwqrvIeiON5+s9KStdCr4um7y/aAjRKl2xmI25WE9OnqNUKnJgf7Oa/kpaI/XCorDo5MfB9jwMNl6sOK4ik9Vxkv2iXEGlFb6vViWqJDwpCIKwORBnLGVsn1MWdWmecMsEka0xxzu48w/jZ5qvzdppIlxqDLE4GrVcWKl1BFG02lmUja2t5cWr5sSJY2itufrqaxqP1euNrRYLGJ0Htehrbw2OKS7xqo0VLK6ncJepiq+UwvNBhRAG/X1AMntSEARh8yDOWEqolThj1uKUi5j8MLO8jcWg5w/iZ5uxpqyZoaLGQCkWT5xcSfL+WlwxC3SdPTBgrLWcePM4e/dehe9nmk8od0260upcpxCDRv/L7mycYPH95YVYK66rcPt0u6TSviAIwuZBxFhqxDKhnyr3qlpBRSFRYTjuRwno8n78TFOM1ctadNMCy4qxZAjWrC6XqGVLa3nxqpmcPN8RomyyOjPXKB+rlq5q2tr/svEYbIgYVQoyGYVeReixH5HluOKKCYIgbCZEjKVMf2UtkhpjhWECSmibRVm/4YxpG+BTpKLGO3aQUsu3QaqPIFxT4j5sWIjyzXqIcn/Hc91E03JY5WJ1dpmluq13/d+/4yj8TGduWL8otbyb5oorJgiCsKkQMZYatu8q97rc7EsZ2DKOiWdVZhJnrFfy/nIzKRujWUM5i+bGNiZE+eabx9m7Zx+ZTKbz+RWKMYvGqNzyy3Vd7/q+f89XqZSacHqYh1qzaqEnCIIgDAYRYykRF3vtLybY6EuZT5yxcBjHtY2LaN5OAbDQpfp+X22Q7FrLWdRXs/5fjwsXJikW5zlwoFuIEmIHqz8xYVEYnetPVKpusyrXR7RoDZns6mZEdkMphbOEO7bU44IgCMLGIWIsTfota9HijNUoQTjcli+2NXqFBTVGVY2xWBD0aoPUGAZrLWdRZ/0v3M0Q5RL9JpXqyx0zyktmTvbvpHWsdx2cQddV+JnlQ88rX2/nY/VaZYIgCMLmYl1KWzz33HP83u/9HmEY8oEPfIAf+IEf6Fjmscce40//9E8Jw5CrrrqKf/AP/gHZ7HJ5PpuLfpL3IXbGrHaw2RyBKeEGIw0xljMXGLKTvOO8P55Jufi1fYQpjVlrOYs663vhttZy4sQx9iwRomzg5DAqg7IhinYL0CgPq/wVibCWFQNBczwDfP9Kx3W++hHXq1q/UjiObXNIxRUTBEHYnAzcGatUKnz+85/n537u5/i1X/s1Dh8+zIkTJ9qWmZmZ4Q/+4A/4xV/8Rf7Nv/k3jI6O8ud//ueDHlrq9Kt/nNI8Jj+EwRBRxVRHG8n726KXMThcdG6MF265fi7bBmmlA1luNeucM3ZharkQZYJysDqDcQpEegijshjlE+lCUr5idRnqnY7bYN6/6yoyGT0wIVZnsfiSchaCIAibk4GLsWPHjrF//37GxsZwHIf77ruP559/vm2ZMAypVqtUKhUAxsbGcLvFWTY51vRf8DUqDBNQBkCFw2QyBm1rbDFvcElfR6RiZ6hVD/WbvJ8e6YsFay3PPPMkf/zHf8Dbb7/ZNvv0xIljKKW5psssyqWHqLHaj2dLrlKENde1OG8s3fevNPiZ/muBrRWtFTr5SKSchSAIwuZl4Ipnenqa0dHRxt8jIyOcPXu2bZmJiQm++7u/m5/8yZ/kvvvuY2Zmhn/0j/7RoIeWOv23QponGh4loBQ/EI7gZy1bzBs4BFxwbun6us6g5aBJ9+JtreXJp77NkSPPk8lk+Isvf5EdO3bxnnvvY+fO3bx54hh79+wlk9m48LRVDqoxEWNt71+pWIBprdC6v3y/tHFdRS2yUs5CEARhE7Mu9tPiO/IwbG/2XC6XeeaZZ/jFX/xF3nnnHX7/93+fo0ePctddd3Ws69FHH+XRRx8F4LOf/SwTExMDG3dQq1KeukBhaGjZZXPZHAuVhb7W6yyUsHuvQWcjKMfO2PhYlh3Rq1Sc7TjD1zKcfGY538F3YhHmZzL43tKFS9PGuqOpJbFba/nLr32FI0ee56677uXDH/oIR148zLe+9Q3+9L//EXt272W+OM8DDz7Mli1beq7LcZxll1k1poqK4v1onTxof8WrUAoyWWdDxFc3gsDgeYMX8q7rDvR4FFaO7JPNieyXzcdG75OBi7GxsTHm5+cbf8/NzTE2Nta2zJEjR9izZw979+5l7969ZLNZ/uIv/qKrGHvkkUd45JFHGn9PTU0NbOxBrUpkDKXiUn0L25et1YJll8MYRkpFqn6W+YWLAKhwCD86Rjaa5G33IeZLze0FrsZPZsD5fmbdwrcWMM6ayvc312UtTzzxOC8efYFbbrmNu+96D7Ozs1x91X727N7HSy8d4fALz+I4DtsmdnDp0qWe69uyZcuyy6x+sBGOiR3LSNe6t09aBs9XOOXNIcTWk4mJiYEej8LKkX2yOZH9svlYr32ye/furo8P/Hb5+uuv5/jx48zOzhJFEU888QS33nor5XK58ca3b9/Oq6++SjERPcePH19ywJuWfmdSVsooa5Kcsfiin/MLbLcvE+FxSV/ftnyrMbWSnpRrJ51tWWv5TiLEDt1yG++7/8E2p9R1XW6//S7+xg9/kh/6az+y8TNolbOmWZRSPkIQBEFYKQO3WbLZLJ/61Kf4zGc+QxRFPPjggxw8eJDHHnuMxx57jJ//+Z/nwIEDfPSjH+Wf/bN/ltSYupq/9/f+3qCHlir9l7VotkKq2XMQZRgbDthijnFR34hR7WEx1SIM1jcBO51tPf/8Mxw9+gKHDt3O/fc9sOR78P1Me1PwDcQqF2UDVvMZrFdyviAIgnDlsC4xr3vuuYd77rmn7bGHH36Yhx9+uPH3Jz7xCT7xiU+sx3AGQp9arNmXMj9E1ZRR4TDXjR1FE3VN3K9f2vvpSZkqKWyrVqvywpHnuOaaAz2F2OajXm9sZeMVV0wQBEFYDVKBPyX6nUmpy3EoNioMUzULqHCYq/IvUlTbWdBLJw+udz/BNAqevvLqSwRBwF133nMZCbGWemP9tJ5qQVwxQRAEYTWIGEuJ/vtSNsOUASVyyiWvZphaqpxFcn1X696aaG1fjSiKOHr0BXbv3svExPaUxrROdO1TucxLxBUTBEEQVomIsZRYSZjSuB7W8wlViZFkxuK0vrbr8nVDTK3QpVk7axMWx0+8QalU4rbb7kxpPOuL7TKL0vcVfkZ1jeCKKyYIgiCsFhFj64wuFTGFYYy1WF2m4GgiXIzqUj+sbSbl+o0x3vbqxYW1liNHnmd8fAv79l6V4qDWj8VizHUV2ol7SfoZ1dZqSGlxxQRBEITVI2JsnXHKcV/KYrUKyjLkKCK6l3Noa8yzzs7YWnLGTp9+h0uXLnLbbXduqlwxrVfQn7FFjCnd7nwppfA8hefHLpkrDbgFQRCENXD5NYC8zNGlIuH4BPPVKvgw7EColq+ttf6iZvXbe+HI8+Tzea679oYUx7M6tANO0qNRKYW1lihaWSd1f4kQpOPEbY42k+AUBEEQLj/EGVtndGkeUximVKsCMKwtId3razXLWqh1LvjauvWVMXXxAqdPv8OhW27H6duGGgyup/B93dYkW6nuOV9L4fmq50xWEWKCIAjCWhExtp5EIU6lTFQYohLG/Q+HlOnhjDWy99dnfG2bXt1X48iRw3iex803H0p5QCtnKQ3Vb89Ix5FcMEEQBGHwiBhbR3Q5bn8U5kep2jIAQxiiJZyx+s7p5b6stARD+2vTXWexOM/x429w040HyWQ2vpr+UnpS92HYKSUzJAVBEIT1QcTYOlKvvl/M7sS68yjrkiNY2hlLtIBeQlVYFEavvpejVdklEvVXLkKCIODJp74NWA4dun3VY0oLpZYWsf1EfB1HSQhSEARBWBckgX8dqVffn3O3Y93X8MijNYRLzKZsFHxdQhRY5YNysSjUkj5Xdywaq30wFmWri7fc/3qs5cSJYzzx5LcolYrcccfdDA+PrGgsg6CX4FJKobSlV53eDU53EwRBEN5FiBhbR+rOWIkx8ObIqBwAoeod0usmxixgk9pkzcbW/WOThuRWebBIjNk+HaFLl6b41re/ydmzp9m6dYIPf+gj7Ny5e0XjGBTLtY9ytCI03QWs0uvffkoQBEF49yJibB2pt0IqBQWUN0eGDGCWrDNWF2HdEs5jVyyxf5QLKxBjDVcMQGmM8tBtr+8tRGZmpjny4vO89tor+H6GBx54mJtuPLgBMz6XZrmh9HpekvYFQRCE9UTE2DriTl+gOjxBpephnXmyaidQWTJnrFnaol05xK6Y3/L3ymJqra+N//YWibku4s9azp49zZEXD3Py5Fs4jsPBmw9x993vJZtdfd7aoFjO3NOOYqkpDM7m0ZSCIAjCuwARY+uIN3mGqb13YrFEukQWH6gsXWdM1etitSuLNlcMQGksDopo2TG0uWKN17vtr18k/k6ffocnnvw2Fy9eIJvNctdd93Lw5lvJ5/PLbm8j6JW834rWYEznYxKiFARBENYTEWPrRRjgXrzAzE03gVMCZcmp2NHq6Ywt44o1HlcOyvYhxrq8Nn7ca7x+8QzLx77+VZRSPPjgB7n+uhtx3c39tek3WqodhVmUN6YlRCkIgiCsMxKQWSe8qfMoa5gt7MPNzQCQVy4WutYZs2gUGr2cK9Z4fHmBZNGNpP/O13stIqy5zSiKKJWK3HDDTdx80y2bXohB/85WN9EmsygFQRCE9UbE2DrhTZ4BoMgW/Pw0AHkUEd3FVUSGUBWwOtsowrqUKwYkocbexEJuCaGiVItQay5TTgrVDhWGl1n75qFvZ0y3t0aSPpOCIAjCRiBibJ3wJs9QKWylFvkNZ6yglq4xZtEorVBuFuMMEekcVmV7tynq4Y71csUay9Sfb9lGsRjXRhsaGur52s3ESvRUq3CTEKUgCIKwEYgYWye8yTNc3HsXADozC0ABu0QrJAVKoWiZSam8zsT7RdgeKYBWZ5ZXKcrBLBJspaQcR6FweYgxtUJ3q7VsiIQoBUEQhI1AxNh6YAzuhbPMbr0x/tufR6HJq7CrM2bQsRjrc1ZgnaXyxqxyl3XFmsu2i8PLzRlbnGO37PKJAJMQpSAIgrBRiBhbB9zpKXQYUC1sRWtDqIp4FPBU0LX6flxmImZFhVSV7mjybVGYpXpfLrGOVoqlIr6fwfN6u3LrhdaQzS5tYfWK4nZdPhG9UuhVEARB2ChEjK0D7oWzAAS5EbS2BLaERwGXatcaY3Gvyc5ir/2w2B2zepk8s2UoFuc3jSumFHi+QjtqyST91TQBcBzVcMgEQRAEYb0RMbYOeJNnsNoh8vKxGKOMRx5HBURdXKu6M9atDdJyWOW0/L//8ORSlEpFhjZJvpjnNwvgul73z2Y1n5njSohSEARB2DhEjK0D3uQZgokdGHQixkpkEhHWdTalqueMrWb3xCUuVhyeXIJisUhhaOPLWni+ahNaWne6Y6s1AEWICYIgCBuJiLFBY20sxrbvxkQKpQ0B5aRJeKcYq+d8rdYZi2dMunEi/hrCkwBhGFCtVjbcGXNd1TWny3XbH1tp8r4gCIKweZleCPmHX3yTf/n1U8xWwo0ezkARMTZgdHEOZ6FEuG0XxiiUW8YSJX0pO1shNcSYWl3OGIDR/rJlMPqhPpOysIE5Y1r3CEkuyh1bTb6YIAiCsPko1SI+87V3ODtf47kzJf7ff/YWL54vbfSwBoZcvgZMvfJ+sH0Pxipw47pduSTRfnHOWF2MabVKZwx6Fn9dCaVSUtZiA52xxe5Xr+fXaAQKgiAIm4BaZPilr5/i5EyVn31oD7/y0avJupqfe/Qd/vORC0RmuX4zlx9y+Row3uQZLIpg+y6M0Vh3DoAcSZPwRbMpW0tTrNYZS4tiMRaOQxuYM7Zc5LHVHVu1eBUEQRA2BZGx/KvHz3B0coFP37+Lu3YPcWBLll/7+DV88MAIX3jxIv/royeZKgcbPdRUETE2YLwLZ4jGt2L9DMYobOKM5ROhFapc2/LNMKXa8MTyYuKMbWT1/X6afruuEldMEARhnZlLOY/LWsu/feocT54q8uN3b+cD+0cbz+U8zafv381Pvm8XJ6ar/PSX3ubUXDXV7W8kcgkbMN7kWYJtuwBiMebUxRhYqzraD9VVhbMJEqBKxSK5XB5ng/oE9atFtaOWDWcKgiAI6fHa1AJ/878e41tvz6W2zt99YYpHj8/yPxzayvfetKXrMg/vH+VzH7mK0Fr+6VdO8tZ0petyobH88SuX+KOXL2Lt5g9rbvwV/wpGVRZwZy8RbN8NxGLMOPOAIo+NQ5QtiqM1RLkZQm7FDa4xthJjUCroC4IgrB+vTy0A8O+ePs/Mwtodsq8cm+EPXrrIR68b40dum+i57DXjWX7pu67CVYp/9uhJ3ri40Pb8GxcX+Mdfeov/87lJ/uPzF/ijVy6teXyDRsTYAPHqlfe378baxAlzi3jk8VXQJV+s6UDpTRB3KxWLGzqTcqPDtIIgCEJ3Ts3VyDiKSmD4t0+dW5P7dPR8md96+hx37Crw9+7d0de5f+9Ihl/+yFUUfIefe/QdXposUwkN/9dzk/z0X7zNbCXiZx7aw4NXD/Pbz1/g62/Ornp860E60+6ErjRnUu7GmPjLFenikq2QLK1FTTdWjFlrKZbm2bNn74aNYRPoUUEQBKELp2ar7B/Pct++If7j8xf4+ltzPNyS49UvZ+drfPYbp9g55PNPHtiNs4Ko0I4hn1/+rqv45199h5//y3cYy7pMlgI+et0Yf+vObRR8h3t2F5iuRPz6E2cZy7ncvrOw4jGuB3K5GyDe5BmiwjCmMNwQY0YXk1ZIVaIezpjrbOyuqdVqBEGwwc7Yhm1aEARB6ME7czX2jvp8301buGkix//xzHkurnCGY6kW8YuPnQLgf314L0P+yvOTt+Y9/uV3XcW+UR/PUfzSI1fxE+/dSSFZl+dofvahPewZzvDLXz/Nm11yzOYqIYdPb6xzJs7YAPEunGnLF4PYGfPVHlyqlGmPi7fOpNyopPk6pVJS1qKwectaCIIgCOvPXDVithKxb9TH0YpP37+LT//Zm/zmk+f4uYf3NsKMQWR58tQ83z45z44hj9t3Frh5W46Mq4mM5VceP8PZ+Rq/8OGr2DW8+kLlY1mXX/noNWjVPb1lyHf45x/ay0//xdt85mun+Gcf2MO5+YCXJsu8NFnm5GyNrKv5vR+6HneD8rVFjA2KMMS9OEnlwM1AMpMSS6hKSZiyuKgVkmqoD60dNlqH1KvvD4kzJgiCILRwajYuKbF3JI7u7B7x+eQd2/j8s5N89cQst+8s8BdvzPCV4zPMVCJGsw5PvBPxhy9fwtWKm7blyDiK58+W+J/fu5NDO/JrHtNy4c2JvMf/54P7+Nkvv80//tLbAGRdzc3bcnzgmlHed8MuFBtXKkPE2IDwLp5DGdNW1gJdxaoAjxxaRYS2eSfQOpPScd0NT15v1hjbGGdMLXGHIwiCIGwsp+ZqAOwbbV7DvvvGcZ54Z57feuo8kbVYC3fvLvDxG8a5c1eBWmR5ebLMkfNljpwr8dJ0lb968xa+67qxdRv31WMZPvPhfbw8ucDN23JcuyXbEHETEyNMTU2t21gWI2JsQHjnm8n7QFvB13pfyqjFGWsTYxscooR4JqVSinx+7Xcsq0J0mCAIwqbkndkqvqPYVmjWydRK8Q/v38W//vZZDm7P85HrRtkx1BRrOa24e88Qd++Joy3V0OBvQEmi67fmuH5rbvkF1xkRYwPCnTqH8XyisbhwXSzGklZISe/IUDUT+Bs9KbVGbwJHqFicJ58vbNiszk1QZk0QBEHowqnZGntH/I5r1Y4hn1/+yNV9rSPjyvzBVuTTGBC6VsVkc436DG19KetijG5izNkUO6VYKm5wT0pRY4IgCJuRU3NV9o5mll9Q6JvNcN2/MolCcJrGY2uYsluT8IYYc5xNkbleKhUleV8QBEFooxIaJksh+0ZWP/tR6ETE2IBQUYhdLMayp8FCPkmIqueM2baZlHrDd4q1llKpuLENwkWMCYIgbDpOzcbJ+3tHRYylieSMDQgVRW1ibMZ7jnD4W2zjdnwVF8arO2OtrphSasOT1yuVBaIo2ti+lButSAVBEIQOTs0lZS0kTJkqcskbECoMIZkVOW1PcGHov6GLN3JAfRRXVYmsg020cL3yfn0W5UYnrzdrjEnOmCAIgtDkndkaWsGuIXHG0kTE2KCIQqzrUrTneM38EV64g9ypH8PRDg5JK6REb9R7Umq98SUtIJ5JCWxYmFJ0mCAIwubk1FyV3cNx6yEhPUSMDQgVhZSGFa+Y/4JLlm0XP4WThCWbTcLrX2aFUqpRRmKjXaFSaWOdMRFjgiAIm5NTszXJFxsAIsYGROAEPHn/DIaIg/p/QEWjaG0BcFWN0GZaVIeKZ1E2/tpYiqUijuOQzWaXX3gAqI2O0wqCIAgdBJHl7Hyt0QZJSA8RYwPijZtrlAshN+kfJK+2YYxqirEOZwyclhDlRjtDxWI8k3KjHLqNfv+CIAhCJ2eLNSLb3gZJSAcRYwOikjVkKy6j6ioATNQUY91yxjaTM1YqzW/sTMqN/gAEQRCEDhY3CBfSQ8TYgIhci2OaH6+xdTFmcakR4jdUl9ZOmwu10TljxWKRghR8FQRBEFqQGmODQ8TYgIgci2Oablc9TOkQoJSNc8bqsygXNQbfSC1ijKFcLjFU2MiyFhu2aUEQBGEJ3pmrsb3gkpW+kqkjn+iACBc7Y0ajlcUhtnlbWyF1iLENFCPlcglr7Ya1QlJq451BQRAEoZNTs1UJUQ4IEWMDInLp6oy5iRiLaP1Ct4sPvYFipJiUtShIWQtBEAQhwVjLqTkpazEoRIwNAmMSMdbem1Jrg6sSZ6wlTNmmPza6xli9+r4UfBUEQRASLpQCapFln7RBGggixgaAikJCT+HYxWKs6YyF1OuMtfei3OgdUizF1fc3LkwpakwQBGGz8U49eX9EnLFBsNHX/iuTRIzppOektWBtPYG/LsbiL3Rc7GLzCJBSsYjnefj+xtz9iBYTBEHYfEiD8MEiYmwQRCGRB471gNgVA2JnTNWwFiLqdxeqTYpt9A4pFosbO5Nyoz8AQRAEoYN3ZmuMZh1GMpujh/KVhlz6BkFUw2qFposYo5oIMd0SpmyRYxtdY6w0v2EhStjwty8IgiB04dRsjX0SohwYIsYGQBQtAOAsIcYaZS1UPUzZZCN3iDGG+fn5DS74KmpMEARho4iMZaocYG3z6mSt5dRcVUKUA8RdfhFhpRgTx9a7OWNOwxmro9oFyAZqkZMn36JarbBn974N2b7oMEEQhI3h3HyNR4/P8tUTs1xaCNk55HHv3iHes2eIXcM+xZqR5P0BImJsAFhTAcBR8Re3PWesSmizSyoPvYGC5MiLhxkaGmb//ms3ZPuSLyYIgjB4ImMp1SKKNcPxSxW+cnyGF86V0Qru2lXg+28e58VzZb70+gx/+uo0XnJhkrIWg0PE2AAwUTwFWNNFjFGjwmhjWbVJFMjkhfOcO3eG++57AK03ZkwSohQEQRgMj705y++9MMV8NWIhNG3PbS+4/MhtE3z42lEm8nFE56/cvJVKaHjhbImnThc5N1/j+q3ZjRj6uwIRYwMgskmYspszVs8ZW0J3bJQgefHFw3iez003HlyX7SkVl/xY/JggCIKQLhfLAf/uqfPsGvZ4794hhnyHgq8Z8h22Fzxu3p7r2vkl62reu2+Y9+7buBn27xZEjA0AYxNnTMeWbl2MOTrCUQGRyTSUh110AGyEHikW5zlx4hi3Hrod3x98ToDrKVw3fqfG2LgOm4ENMuQEQRCuaP7Ds5MYa/mZB/ewc1jyvjYjIsYGgEkKuzbFWKwyfB3nkjWr78Ni+bUR7tDRl44AcOjQ7QPflus2hRiArifJSekaQRCE1Dl8tsS3Ts7zN26bECG2iREvYgBENgDA0XF8ve6MeU5djDUPiMXaS62zN1ar1XjllZc4cOA6hgbcHNx1Fa4nsUhBEIT1IIgM//vT59k55PEDB7ds9HCEHogYGwCGWIwtDlO2OWMJdoOdsddef4UgqHHrrXcMdDuOCDFBEIR15Y9eucSZ+Rp/794d+I5c7jczsncGQF2MqTZnzOKpOHwZ2ZYw5aLZlOspV4wxHH3xMDt37mb7th0D247jKDwRYoIgCOvG+WKN3z96kfv3DXHX7o0r5C30h4ixARAlYsxxmmKsXmMM2p2xjjDlOmqWt946wXxxntsG7Ip5vggxQRCE9eTzz06iFfxPdw/uRltIDxFjAyBSATq04CalLaxqlLWAdjG2eBd0m148CKy1HHnxeUZGRrnqqmsGth3HESEmCIKwnjx9qshTp4r89UMTbCt4Gz0coQ9EjA0AQ4gbWKwTTxGsO2OOqmGtwuDS8MRatco62mLHj7/B5OR5brvtzoEWeXVklqQgCMK6ERnLf3x+kj0jPt97kyTtXy6IGBsAkY5wA0C3izGXWjKTsk2BNf63XjujUlng29/5Btu37xhokVelQIszJgiC0OBnv/w2f/Hq5MDW/8235zg1V+N/vG0CT86/lw0ixgZApEKckIbT1XDGqC1qEg4bUeb1O088TrVa46EHPzhQV0yEmCAIQpPIWF6+sMD/9eRJzOIWJCmt/wsvTnHNWIb7r5Kq+ZcTIsYGQKSiWIwlGKPQapEYa9R8XV9n7NSpk7zxxmvcccddbNkyMdBtSYhSEAShSWBiAfbOTIUj58qpr/9rb85yZj7gb9w2sW75x0I6iBgbAJGOcMPmgdCaMxbZujMWP9/WKHzAB08QBHzz8ccYHR3jzjvuGei2lGqpri8IgiBQi5pu2JfemE513aGxfOHFi1y7Jct790opi8sNEWMDIHIMTtRFjDVyxmipM9Z83aB3xrPPPcX8/BwPPfhBXHewnbBkFqUgCEI7QWQAGM95PHmqyMVykNq6v3p8lslSwI/cNoESV+yyQ8TYAIi0wWlzxnSSwB8QUZ9m3HmwDPL4uXBhkhdfPMzNN93Crl17BrehBC0hSkEQhDbCJEz5V27dibXw5WMzqaw3iAz/5egUN05kuXt3IZV1CuvLujQKf+655/i93/s9wjDkAx/4AD/wAz/QsUwYhvzn//yfefrppwnDkF/5lV9haOjytFpjZ6ypc7sm8Kv4n9ZelIMSY8YYvvHNr5HL5Xjve983mI20ICFKQRCETuphymu25LlzV4EvH5vlhw5N4K7xfPnlY7NMlUP+l/t2iSt2mTJwZ6xSqfD5z3+en/u5n+PXfu3XOHz4MCdOnOhY7vOf/zyZTIZf//Vf59/+239LoXD5qvvIMTimXYy5uoZWUTNnTIFNftcZVJPwV197mYsXL/C++x/C9zPLv2CNSIhSEAShkyARY56j+PgNY1xaCHn6VHFN66yGht9/6SIHt+W4fWc+jWEKG8DAxdixY8fYv38/Y2NjOI7Dfffdx/PPP9+2zMzMDK+//jo/9EM/hFKq8XO5EjkWN8kZsxasVfhu0peS1gT++k/LQylTq9V49tkn2blzN/v3X5v+BrogsygFQRA6qc+m9BzN3buH2JZ3+fM1JvL/xbEZphdC/sfbt13W1813OwMPU05PTzM6Otr4e2RkhLNnz7Ytc/LkSZRS/MIv/ALT09Nce+21/N2/+3fJZrODHl7qWGswDjimWfAVIOPUWyEtrjPWZBCH0eEXnmVhYYGPffR71uVA1RqUhCgFQRA6CBNnLONqHG34yPVj/N4LU5yeq7FnZOlrQy++cmyGm7flOLRDXLHLmXXJGVssAsIwbPt7bm6OXbt28elPfxrHcfhP/+k/8Qd/8Af86I/+aMe6Hn30UR599FEAPvvZzzIxMbhaWUGtSnnqAoUV5K6FtgJz4CqPwtAQtVr83gu5eBaNlx2h4Ayhc3msk8ErjOMkVtJQxiXjpmdWzs3NcvToC9xy8FZuvPHm1NbbC8/XeN5gDVfXdQe634XVIftl8yH7ZHORLcUuWNb3mJgY46/fO8L//eJFvn6qwj98aPeK13dmtsLJ2Rr/y4P7ZT+vkY0+VgYuxsbGxpifn2/8PTc3x9jYWNsyhUKBTCaD58UzDe+9917+5E/+pOv6HnnkER555JHG31NTU+kPOiGoVYmMoVTsP6Zfs/GyKtSUikWqVQcYRpn4MygtGEoUITQYp4obOuhk6mHou6m2r/jLr30Fay23334Xly5dSm29vchkBx9inpiYGOh+F1aH7JfNh+yTzcXF6fg6oDGN/XL/viG++NI5fvCGoRXfjH/ltVjcHRxTsp/XyHodK7t3dxfdA88Zu/766zl+/Dizs7NEUcQTTzzBrbfeSrlcbrzxG2+8kVdeeYXJybhf1+HDh7n++usHPbSBEFEDOsOUvl6UM6Y6c8bS1DAXLpzn2LHXuPXWOxgaGnxbDKXAz1zeuX6CIAiDpJ4z5jvNS+/Hrx+nWDN88+25Fa/vqdNF9oz47F5liFPYPAzcGctms3zqU5/iM5/5DFEU8eCDD3Lw4EEee+wxHnvsMX7+53+efD7P3//7f59f+ZVfIQxDbrjhBn78x3980EMbCIa4iJ+TfLR1MeY1csa8lqXVEi3D14a1lu888S1yuRx33H53SmtdGqXA95XkigmCIPSgPpvSdzRE8WO3bM9x1ajPF1+b5sMHRvu+oS0HEUfPl/meG8cHNVxhHVmXnLF77rmHe+5pb7/z8MMP8/DDDzf+vu222/iVX/mV9RjOQIkSMabtIjG22Bmro9J3xt566wTnzp3hgfc/jO8P9o5J6USIiSMmCILQk9bSFnUxppTie27cwr996hwvX1jglu39JeK/cK5MaCz37Ll8y0AJTaQCf8o0nbHYAWsVY8ZqLEndB6Wwi7ywNBq7GhPx5FPfZnxsCzfddHDN6+uFFiEmCILQN93ClAAP7x9h2Nf86av9l7l4+lSRgqe5eZvMorwSEDGWMnVnzOlwxurV99t7UjaETEqCZnZ2lrm5WW677Q60HtzuVQo8EWKCIAh90whTLkrUz7ia77pujCdPzTNZXL5fpbGWZ84UuXN3Yc3V+4XNgYixlDE2TuDXKg4P1sWYq6qdIcoW0toR9bIh2WwupTV2R2sRYoIgCCuhGabsPON/4oY49+uLry/vjh27WGG2EnHvnsuzZaDQiYixlInqYqwRpow/YlfVmk3CExGjVPPjT0vWRFGciOAMuAy+NAIXBEFYGYFJ6k12KWG0reBx/75hvnJ8hkpoeq7n6dNFtIK7dosYu1IQMZYyxsSJ+k6HM1YjtK1NwpPelLQ9tGaiKHbGHGewczMGGAEVBEG4Igkii6uXzg/+3pvGKdUMXzsx23M9T58uctNEjpGM3BVfKcglNWWMjcWYTkKSxirA4qhal5mULc5YSiG/hjPmDu4g1Tq98QqCILxbCIzF63Ene9NEjuu2ZPnT16Yx1nZdZqoc8OZ0lXskRHlFIWIsZYytoiKLcprOmNYWl9qiJuHtVcbSkjZh4oy5AwxT6hS7BAiCILxbCCLbs8uKUorvvWmc03M1Dp8tdV3mmdNxlxfJF7uyEDGWMsbUcAOwTrMCv9YWp1WMdXGV0jKaTCNnbHBhyi65p4IgCMIyxM5Y75P9+68aYTznLlnm4pnTRXYMeewblar7VxJyWU2ZiBpuYMFtlrZwnQCtTDNnLMG2hSnT2X444AR+pZBK+4IgCKtgOWcM4uT+j18/xnNnS5yarbY9Vw0NL5wrc8+eIUkVucIQMZYyxtZwAot1mmIs6y0AXarvt6BSClQOejalhCgFQRBWR60PMQbw0evHcLXiX379FP/p8AWOno+r7R85V6YWWQlRXoGsSzukdxMRAW5ImxgrOJXkuUWlLVoFWEoaJwoHO5tSQpSCIAirIzRm2TAlwFjW5Sfft4svvjbNH758kT946SI5VzOc0WRdzaHtg60jKaw/IsZSxhDgBBZaxFjGr4uxxc5Y+gn84owJgiBsTvoJU9Z54OoRHrh6hFIt4sj5Ms+fKXH4XIkPHxjpWjRWuLwRMZYykQrJBBabaSbwZ9ykSfiinDFa7pDSCv/XxdggWiFJoVdBEITVExi7YiFV8B3u3zfM/fuGBzQqYTMg8jplDEEym7LpjPlu7IyFHaUtmqSVMxZGIY7jDiS505HEfUEQhFVTi5afTSm8OxExljKRCjsS+DNO4oy1hSkVrXIsreMziiLcARV8FWdMEARh9YSm/zCl8O5CxFjKRCrCDVtzxnTDGWvWGUtaIQ3gmIyicCD5Ykqq7guCIKyJQJwxYQlEjKVMpEOcRUVffaeKsRpLq0hSbaHJ1JyxMBrITEoJUQqCIKyNfktbCO8+RIyliLUWow1uEqa0FqxVeLraJURJmzOWZm/KQThjEqIUBEFYG/1U4BfenYgYSxFDABCLMdfFmPig851FYky154ulGf6LzGCcMS0nEEEQhDURRkacMaErIsZSJErEmBMATlOMebraMpMSQGFpzqBM89CMwjD1JuGSKiYIgrB2AmPxpUaY0AX5VqRIwxkL4zBlqxhrqzGmkn86o5VrJooidNphShFjgiAIa6YWWVyJMghdEDGWIm3OmNYNMeaq2qIwZft/0jw0wyhK3xlLdW2CIAjvPiJjMRYJUwpdETGWIoYaAI7RoFS7M9alxlj9kEw1Zywp+pomEqYUBEFYG6GxAJLAL3RFxFiK1MOUThR/rLEYs7iqRlhvEp5gBximdFIv+ionD0EQhLVQi2Ix5oszJnRBxFiKNMKUplljzFEhWplFfSlV2+80nadBlLYQZ0wQBGFtBIkzJjljQjdEjKWIsZ1iLLO4+j50qJt0xVj6YUoxxgRBENZGEBlAcsaE7ogYS5GGM0ZLk/CufSmhLWcsRbUThoMp+ioIgiCsnrozJqUthG7ItyJFmgn8dWdM4ztdnDHqvSkHEaaUOmOCIAibjSCSBH5haUSMpUjdGdM2ccZs0xkLrbdo6fQPSGstxpiBVOAXBEEQVk9DjEmYUuiCiLEUMQRgQatlwpQNRyz+nVqT8CgCkAR+QRCETYYk8Au9WFaMfeUrX+Hf//t/z5//+Z9TqVTanvsbf+NvDGxglyMRAW4IOE0xlvUWkue6VOBP0Ck2CQdw3JTrjKW6NkEQhHcfgZS2EHrQU4z97u/+Ll/60pfI5XI8+eST/JN/8k84c+bMeo3tssMQ4IQK2yLGus6mBFDNj95J6U4pisJkfdIOSRAEYTMhYUqhFz0tlMcff5x/8S/+Bdu2bQPgS1/6Ep/5zGf4zGc+w86dO9dlgJcTETXckDYx5jsVjHWwJAJpkQuWlisGrc5Y2mFKOXkIgiCshcAkpS0kTCl0oaczVi6XGR0dbfz9sY99jL/5N/8m/+Jf/AsuXbo08MFdbhgb4AS2LUyZcavd+1ImzthAxJiUthAEQdhUNJ0xSdUWOunpjO3YsYPXX3+dQ4cONR574IEHmJ+f55d/+ZcHPrjLjYgANwDrtocpw46+lE1NluZNUpiEKd0UZ1OKKSYIgrB2AulNKfSgp0T/+Mc/3jVH7OMf/zh33nknW7ZsGdjALkcMsTPWHqZc7Iyptt9pHpdRKM6YIAjCZkRyxoRe9LRQPvShDy353I/8yI/wIz/yI6kP6HImdsYs1mm2Q/KdClFHjTFaxFiaYcokgT9FMSbOmCAIwtoRZ0zohQSvU8RQw62ZtpwxX1e7tEJqDVMOImdMir4KgiBsJsQZE3rRtxhbWFjgN37jNwY5lsueiACnZtpyxjynuihnDNprjKW4/QHMphRnTBAEYe2IGBN60ZcYO3r0KD/90z/dKHEhdMd0hCnB07WOnLH4kFSgVGo1xmAwCfyixgRBENZOYCyOSjcaIlw59Lxqh2HI7/7u7/LCCy/w6U9/muuuu269xnXZYa0lsrEYI3HGNBFaGSLTxRlT6ceIB1HaQk4bgiAIayeIjLhiwpL01AM//dM/Ta1W43Of+5wIsWWwRKBso+irtbErBhDR2SRcAU7Kd0gDqTMm5w5BEIQ1U4us1BgTlqTnN+OOO+7g1Vdf5eTJk+s1nsuWiFh41Utb1GdSxs+11xmrhynTnlQziAR+0WKCIAhrJzBWZlIKS9JTjH3yk5/kx3/8x/n1X/91vvCFLzQu9kInhgAgzhlz3UaNMaB7Av8AcgeiMP3SFqLGBEEQ1k4YWQlTCkuyrGd68OBBPve5z3Hx4kV+9md/dj3GdFkSNcQYkDhjmUSMRbazHZJCoVN2rCVnTBAEYXMizpjQi77iWblcjp/4iZ/gmWeeaXu8Wq2SyWQGMrDLjboz5iSzKY1R+G73MOXAnLEoRGudbmNvOXcIgiCsmZo4Y0IPVuTN3HPPPY3/v/322/zMz/xM6gO6XIlaw5QdOWMtCfwtQintm6QwilIv+CqnDkEQhLUjzpjQi55X7oWFBX77t3+b6elp7r33Xh555BEAvvzlL/M7v/M7PPTQQ+syyMsBU0/gD+kMUy7KGbMolNIDcMYi3BQLvgKixgRBEFIglNIWQg96irH/8B/+AzMzM9xzzz189atfpVKp8Prrr/PSSy/x6U9/mnvvvXe9xrnpaXXGAtfB2HpfSherXLA2WTI+GAdxTEZRmL4zJgUKBUEQ1kwtsuT9lG+WhSuGnlfuZ599ln/9r/81o6Oj3HbbbfzkT/4kt99+O7/6q7/K2NjYOg3x8sDYphirOS4mUvhulZAMZLJQWYgXVPE/WqVfbyaKonRnUgqCIAipIGFKoRc9xVilUmF0dBSA3bt34zgO//Sf/tN1GdjlRtRI4KeZM+ZX4hCl44CfgVoVFFhgELX/ojBdMSammCAIQjoEksAv9KCnGLPWcvTo0cbfSqm2vwEOHTo0mJFdZrTWGcNxkzBllUgls009D0xEfTalHsAd0iDClIIgCMLaCY3FF2dMWIKeV+6tW7fy7/7dv2v8PTY21va3Uorf+I3fGNzoLiOilgR+67qYqk7EWEvyvp9t5I4NollsFEW44owJgiBsOmqRxRVnTFiCnmLsN3/zN9drHJc9hgBtNIpmmDLjVAjUcHMhRaJwVOp9KSEWY563uA+mIAiCsNFIzpjQC+lamhIRAU4Uu1LNoq/VjrIWdQZxTIZRhBZnTBAEYdMhOWNCL0SMpYQhwDHJx+m4GENc2kJ16VCgNDrtXkjEOWNumjljosYEQRBSIYgM/iBmbglXBPLNSInIBjhR/HFax4UoQivb1Rkb1IeedmkLkWKCIAhrx1hLZJEwpbAkIsZSwlDDjRQWBVqjkrpj3ZwxrVLuH5kQRSFOmhX45bwhCIKwZoIonrglCfzCUogYS4k4Z0yD64BS6ESMhXQRYwM6IKOUe1PKaUMQBGHtBCYWY76IMWEJRIylhCHAiZIQJeBQd8Y6w5TOAKrvA4QpF30VNSYIgrB26s6YhCmFpRAxlhIRAW6osI6LteCqWvJ4F2dsAJ+6tTb1BH45bQiCIKydhhgTZ0xYAhFjKWEI4oKvjksYajJOBeiWM6YGUvDVGAMgzpggCMImox6mdMUZE5ZAxFhKRAQ4AeDGYsx3qgCEi2ZTxn0pB9MkHNIVY3LaEARBWDtBFN8sS86YsBQixlLCUMMN44KvQejg152xRWIsnkmZ/vYbYsxNs85YeqsSBEF4t1J3xrxB5KgIVwTyzUgBYyMsBiewjTCl71aJrAuq3akaREkLiMtaQMrOmBR9FQRBWDOSMyYsh4ixFDDJzEk3sJCIsbxXJFC5jmVV0psybQYRphQEQRDWTtMZEzEmdEfEWApEiRhzagbrOISBZjw3RUVt6VhWKzWgMGW6zpiYYoIgCOkgzpiwHCLGUsAQl7FwA4t1PaIQRjOXqOhOMTYoZyxMnLHUSlvIOUMQBCEVRIwJyyFiLAXqzpibOGNZPY+jIxbUeMeyA3PGwnTDlHLKEARBSAcJUwrLIWIsBeo5Y14tAsdlyLkEwEKXMOWgZE4zTCnOmCAIwmaiXtpCnDFhKUSMpUAjZ6waYR2XIe8S1kKlizOmWv5NdQwpJ/DLKUMQBCEdao0wpVxyhe7INyMFGrMpqwbjuIxmLlEKR7Gqm0s1uCbhkOJsSsngFwRBSIVQwpTCMogYS4HIxgn8Xi0kdLOMZacoRd1ClHEF/kHonDAJU7opFX2VU4YgCEI6SAK/sBwixlKg7ow5lYjQKzCanaZkOkOUMZdJnTE5ZwiCIKRCTZwxYRlEjKVAPWfMqwToYQ+tDAt2bImlB1VnTHLGBEEQNiNBZNEKHBFjwhKIGEuBRs5YzeIOx49V1GiPVwzCGZPZlIIgCJuR0FhxxYSeiBhLgYgAZR20hWwhxFhF4Ix0XXZgvSmlzpggCMKmJIgMvuSLCT0QMZYChhoOsSOVK1SYq4zjuN0PPIsaiCCLogilFFqntEvlvCEIgpAKtcjiSlkLoQfr8u147rnn+Kmf+ik+/elP84d/+Ic9l/2TP/kTfuqnfmo9hpUaEQHaegDkcyVmKhPoJe+CBlf0NbUQJVLZQhAEIS0CCVMKyzBwMVapVPj85z/Pz/3cz/Frv/ZrHD58mBMnTnRd9tVXX+Xxxx8f9JBSxxDi4ICrKGRKzNW2LrmsGlA7pDCK0ptJKQiCIKRGEFkpayH0ZOBi7NixY+zfv5+xsTEcx+G+++7j+eef71hubm6O3/7t3+bv/J2/M+ghpY61BmUV7pYMSlnmg+5lLWzDFRtMmNJ10xNjg8ptEwRBeLcRGCs5Y0JPBi7GpqenGR1tziwcGRlhZmambRlrLb/5m7/Jj/7oj7Yte1lhwZvIAFCMlqoxpgcW/ks7TCkIgiCkQxBZXAlTCj1Yl6v3YpclDMO2v7/4xS9y4403cssttzA5OdlzXY8++iiPPvooAJ/97GeZmJhId7AtBLUq5akLFIaGei7nlDQ60rjbMkRGE7lbGRqyqEyubTmjPEaHRhkaGkrdeXK0g+/7bNnSvfL/SlAKcvnNK+xc1x3ofhdWh+yXzYfsk02Cc5Z8xmnsC9kvm4+N3icDv+KOjY0xPz/f+Htubo6xsbG2ZSYnJ3nhhRf4xje+QRRFXLx4kX/+z/85v/ALv9CxvkceeYRHHnmk8ffU1NTAxh7UqkTGUCoWey4XRSHWGJyJLLPVLRgbUiwuQNAuOg0eNjTUarXUxdjCQhmAS5curXldSkOmvHln/kxMTAx0vwurQ/bL5kP2yeagXKmRc1VjX8h+2Xys1z7ZvXt318cHLsauv/56fuu3fovZ2VmGhoZ44okn+OEf/mHK5TLlcpmJiQl+7Md+rLH85OQkn/vc57oKsc2KBZQFd1uOmYUJXDfqvpwanMCJoii1MKWY6YIgCOkRRIaRzOaNNggbz8Dtj2w2y6c+9Sk+85nP8I/+0T/itttu4+DBgzz11FP8xm/8xqA3v05YFBZv1GV6YRuea5ZcCgaTHB9GEa70pRQEQdh0BMbiplUDUrgiWRepfs8993DPPfe0Pfbwww/z8MMPdyy7fft2fvVXf3U9hpUiFkUswKYrE3iF7mJMoQc2S9FEEY7vp7IuJWpMEAQhNaS0hbAcItVTwGLRKhFjCxO4rqGbvWQHKHLCKMRxpS+lIAjCZiOIpLSF0BsRYymhlcFEMF8dWzJnLHacBlWBP8LR0pdSEARhsyEV+IXlEDGWChZHW0pzGSwKx7HdFxtQ9X2oJ/BLzpggCMJmI4gsrjhjQg9EjKWAxaK1Yb40hOtEXQWXbXzUqzsgrbWcO3dmyeejFMOUcsoQBEFID3HGhOUQMZYCCoNSMFsa72Mm5eq28corR/mTP/1DLl3qXgdlOWdsRRFMOWcIgiCkgrGWUNohCcsgYiwFNCEKmK5sw/USMdZx3K3+QLTW8uLRFwAol8tdlwnDpcWYdsD3Nf3OrJa2lIIgCOkQmjhtxZPSFkIP5NuRAkoFAFyo7UlmUnZi0XVvbMXrf+fUSWZnZwCoVqsdzxtjsNbgdin6qhR4XrxN1xWVJQiCsJ4EUSLGxBkTeiBiLAU0IViYtrubzlgHqw9THj36Ap7nAVCtVjqej6J49mY3Z8z3VaO2mXZUX+6YOGOCIAjpEBgRY8LyiBhLAUUERhE5uRZnrP3Aa9YYW9kBOT19iVOnTnLroTsAqHRxxqIo7oG5WIz5vkItShoVd0wQBGH9aDhjksAv9EDEWGrUQ4Hda4yt1hk7+tIRHMfhlltuw3Vdar2csZbZlK6n0F3uxPpxxwbVJUAQBOHdhoQphX4QMZYClmZdsV45Y7ES6/+ArFQqvPHGq1x37Q3kcjkymewSzlh7mFLr3g5Yr+dEhwmCIKRHI0wpzpjQAxFjqWCpi6xGaYvFqmYVKufV114mDEMOHbodgEwmQ62HGKsn8C/nbGlHoWTPC4IgDJxaFF8TxBkTeiGX5JSoH2bdnbGVH4TGGF566Qi7d+1h69YJIBZj3RP445wxvYIK/N3cMddV+Bk5YQiCIKRF2AhTyuVWWBr5dqRAfKgtnTNmW4pa9GuQvfXWCUqlYsMVA8j4WSpdxFjYcMYSMdbHNpwWd8xxFJmswvWU5IsJgiCkiIQphX5Ip3/OuxybhCmdqNKSHK9anm89CPs7IF88+gLDwyNcddU1jccymQy1WpcwZbh0aYteeF7SulxOEoIgCANBEviFfhBnLDUUftS9On79Y+7XGZuamuT8+bMcuuU2dMvUx0w2S6XSq7TFyrS11p2lLwRBEIT0qIkzJvSBiLG0sArPtIixluNupc7YhQuTAOzff23b4xk/QxSFhGHY9nivoq+CIAjCxiHOmNAPIsZSwNo4TOmbpZyxlR2EQSK2PM9vezyTyQJ0hCoXF32VQ14QBGFzEIozJvSBiLG0UArPLHR9quGMqf4KqtadL9dtDztmMhmgsz9lI4HflRRAQRCEzYSUthD6QcRYKsTOmGdbZjq2hSlX9jGHYYBSuiPs2BRj7TMqJUwpCIKwOQmktIXQB/LtSIF6aQufzuR6kufq//aTwB8GAZ7X6XLVxdjiKvxRuLoEfkEQBGGwSGkLoR9EjKVAvRmSu5QYUytL4A/CENf1Oh5v5IwtdsaMOGOCIAjdeOrUPP/y66eIjF1+4QFQL/rqS5hS6IGIsZRQKHxVa3sE2mdS9ltPNQyDrvlfdTHW4YwlYUq9XAdwQRCEdxl/9voMT50q8vTp4oZsvxZZtAJHnDGhB3L1ToF6o3C3TYzVaf+I+03g97o4Y74fz65c3J8yCiMcx5Xq+YIgCC0sBIYXz8ez3P/s9ekNGUNgLK4IMWEZRIylgIW4zpgKujy38oMwDALcLjljSikymUxHS6QoCttDlHLcC4Ig8MK5EqGx3LEzzwvnypyaWyqvd3AExspMSmFZRIylQZKK4DrpNAkPwqBrzhjEocrFzlgYRZIvJgiCsIinTxfJe5r/5f5duBq+9PrMuo8hiAy+OGPCMogYSwGLRVkLbufHudgZ6zdMuVTNsEwm06W0RdhsEi4IgiBgrOXZ00Xu3FVgIu/xvn0j/OWJWSpht5vmwRFE4owJyyNiLCUUYLuUlmhL4O/TJVsqZwzilkjdEvgdKfgqCILQ4PilCtOViHv3DAHwiRvGKAWGb7w1t67jiHPG5FIr9Ea+ISlQnzDdLsbqwku3PNKfGAuWyBmDuFl4R2kLCVMKgiC08czpIgq4e3cBgJu25dg/nuHPXp9OWtitD+KMCf0gYiwVEv9rGWes3087XCZnrFvRVxFjgiBcqTx/tsSL50sres3Tp0vcOJFjJBufl5VSfPz6cd6crvLqVPfWdYMgiKzUGBOWRcRYWliFbQ0VNnLDVnYQWmt754z5GWq1atudXeyMNZeXw14QhCuB2UrIv3r8ND//l+/wi4+d4kKpc8Z6Ny6WA45fqjRClHU+sH+EvKf5s3VM5A+Mler7wrKIGEuBep2xbjljrZVedR/J+8YYrLVL54xlMlhrCYJmTbMwiiSBXxCEKwZrLd94a47/+b+/yXfemeev3rwFa+HfP3O+r9c/eyZ20e7ZU2h7POtqPnRglG+fnGOmEqY+7m4EkcUVZ0xYBhFjqaGwiwTRShuEQ5wvBuB6S4sxgEqlGao0i5wxQRCEy5WL5YBf+sZpfvVbZ9gx5PG/fXw/f/uu7fzwrRM8earIk6fml13HM6eLbMu7XD2W6Xju4zeMERr4yrGZAYy+k8BIaQtheUSMpYAlSc7vEESLD8B+ylrEYsxbsrRF0p+y1kziD6MQx5Wir4IgXP7862+f5fDZEj9213Y+95GruSoRVN938xauGvX5P54+z0KwdHmKWmQ4fLbEPXuGupYS2juS4badeb70xsy69KuUBH6hH0SMpYXtDFOuqvp+GFvnSyfwJ85YSxK/zKYUBOFK4e3ZKg/vH+H7b97S1s/R1YqfeM9OpsohX3hxasnXHz1fphrZjnyxVj58YJSpcsjbM4OvyB/njMmlVuiNfENSwCbe2HJhyn5aR9adsaWLvibO2GIxpkWMCYJweRNEhtlKxES++83ozdvzfNe1o/zxq5d4a7rSdZmnTxfJOIpbd+aX3M4t2+PnXposr33QyyA5Y0I/iBhLiaVKW7Qt08fxGASJM7ZczlhLrbEoCqXoqyAIlz1T5fj8N5Ff+nz2yTu3M+Q7/LunzmMW1Quz1vLM6SK37yrgO0tf3rYVPLblXV65MPgSF1LaQugHEWMpYJN/rLs4TLn4400xZ0zClIIgXGFMlePz30Sh+80owEjG4VN3befVqQX+4KWLvDNbpRxEWGs5OVtjshT2DFHWObg9z8uT5YEXgJXSFkI/iJ2SCkuVtmg/APs5HoNGmLL7ych1XRzHaThj1loRY4IgXBFcKMXO2LYlwpR1Prh/hK+dmOX3Xpji916I88eyriKTuGH1qvu9OLg9x9ffmuNcMWDXsL/GkS+NJPAL/SBiLAViKabaxZhSXRL4+2sSDkvnjEG9WXjsjEVRFC8vpS0EQbjMqTtjW3uEKSGupv/PP7iXV6cWuFQOubQQ/1wsh+wa9tm6jJgDOLgtzht7ebI8MDFmrY2dMRFjwjLIFTxNOgRU+wHYVwJ/kjPmLZEzBnGospo4Y3Ux1uqMyWEvCMLlyFQpZCTjkHGXz6DxHM2tO5Z3wJZi76jPsK95+cICH752bNXr6UWYlM6QMKWwHJIzlhLKLppN2fXY6z9nbKkwJcQtkZrOWCzepOirIAiXO1PloGfyfppopbhpW56XJweXxB/UxZg4Y8IyiBhLgXo7JBaXl+iwwpY/IINlSlsAZLLZRgJ/wxlzJWdMEITLm6ly2DN5P20Obs9xZr7GzMJgWiMFUd0Zk0ut0Bv5hqTG8kKrn+MxDEO01ugeC2cymUYCf7cwpSAIwuXIVClg2zo5Y9CsN/byhcHUG6tF4owJ/SFiLAXSnBgdBkHPfDFYHKbsksAvx70gCJcZ5SCiFJglC74OggPjWXxH8fKA6o1JzpjQLyLGUsBi496Uy9DPMkEY9swXgziBPwwDoihqyRkTZ0wQhMuXRsHXdQxTeo7ihoncwPLGAnHGhD4RMZYGii722OoOvjAMeuaLQbMKf7VaJZQwpSAIVwBTpaTg6zqGKQEObsvx5nSFchB1PHduvsb/95unma92PtcPksAv9IuIsRSo1xlroFTX0KXqo7ZF2KczBlCrVYlCccYEQbj8abZCWj9nDOJK/MbC61PtvS6ttfzW0+f51sl5nj5dXNW6a5EBJEwpLI+IsdRY+czJboRBsGQrpDqt/SmbCfxS2kIQhMuXqXKAVrBlnZ2xGyeyaNXZNPyJU0WeP1sC4MXzq0vwlzCl0C9yBU+JznywzoOvr0bhYdgQW0vR2p9SZlMKgnAlcKEUMp51cdfZRcp7DvvHM21Nw6uh4f989jxXj2bYMexx9HxpVetuJvCL7yH0Rr4hKWAXByVV9xmW/YUp+88Zi52xzqKvcg8mCMLlxlQ5YKKwMf7AwW15XptaaDhZf/DSRSZLIX/33h3csbPAZCnkfLG24vXWS1v44owJyyBiLAVs459WVpvA30/OWGcCvytFXwVBuIyZKoXrni9W5+D2HLXIcmK6wtn5Gn/08iUeunqEQzvy3LojrkW2mlBlXdy5IsaEZRAxlhp9FH3t0xlbLmfM9+tirCJhSkEQLnustevaCmkxrU3D/8Oz53G04m/ftQ2AfaM+o1mHF8+tQoxJnTGhTyRnLAUsi3PGFKt1xoIgxF2m6KvWGt/3qVarjWKvjhR9FQThMmW+ZqhFlm3rWGOslbGcy+5hjz99dZqLCyF/685tbE1cOqUUh7bnefF8GWttX+kmdUIpbSH0iThjKZFGFX5rbeKMLX9CymSyVCWBXxCEK4BmjbGNEWMQl7i4uBCyd8Tne2/c0vbcrTvi587OBytaZ720hS/OmLAMIsbWiLUdVcbix7vYU736TUJLa6NlwpQQ541VqxUiE6G1XtHdmiAIwmbiQjkRYxuUwA9wW5Ib9nfu2dHhZNXzxo5OrixUKTljQr9ImDItbHvR19UQhvEJqT8xlo1LW4Sh1BgTBOGyZqq0MQVfW3nwmhFumMixa9jveG7PiM94kjf2kevG+l5nIKUthD6Rb8ia6S9Aqfo4GIMgEWPL5IxB7IzVi77KTEpBEC5npsoBrobR7Mady7RSXYUYxHljt+4o8OL5UiMa0g9BFMdIXLnSCssgX5GU6Kfo63KESWujvnLG/EyjtIXWIsYEQbh8mSqHbM17fc043yhu3ZlnuhJxeq7/emNBZPEcJWkkwrKIGFsj9YKv1qpFj7fTKdY6WVGYMltP4A9xJXlfEITLmKlSwLYNKmvRL6upNxYYK2UthL4QMbZm6gn8yzhjfbZCApYt+gqxM2atoVKp4CwSb3IXJgjC5URcY2zj8sX6YeeQx9a8uzIxFllJ3hf6QsTYGulohbQE/RyOYZIz5nn9JfADlMslKWshCMJlS2QsF8shExtUY6xf4ryxPEeTemP9EBgjZS2EvhAxlhbLtEPqry/lCpyxpCVSLMY2t70vCIKwFDOVkMiyYdX3V8KtO/LMViNOzvaXN1bPGROE5RAxtmbqKmztB9zKSls0+1OKMyYIwuXKVHnjy1r0S6PeWJ+hyjhnTC6zwvLIt2TNdBFjSnUt+rocK8oZS8KUINX3BUG4fKlX39+2gQVf+2XHkM/2gsuL50t9LS/OmNAvIsbWSL8VZ/qaTbminLFM4/+uhCkFQbhMuZycMYBDOwocPV/G9JE3JmJM6BcRY2umv9mU/UxwDBphSnHGBOH/ae/Oo6Mqs7WBP6eGDJV5xCABwhQMYRBRAkkgagQa2w8RYcmUJVyhVVCkRWwVHID0AgEZVOy+0i2CDKIXr5A2lxaZDBAbGpCOgBAChEkyz6nhVJ3vj6SKhNQUqipVKZ7fWgqpnFTe1Ms52bXfffZLd4fiOh38FAICfNrHr6O+HVSo1hpwuUJj81i2tiB7tY9//R5NavaHI0RRhFwut6vYX6FQmPa6lLMDPxG1UyW1IiJVynbTkqdPtD8A4HRRvc1jmRkjezEYc5BkqYD/tguLzK6mr6JdWbGGpxdM2THeTUlE7VVDj7H2cw2LDlAizE+OX0sYjJHzMBhzGsHC3y0/dDtRp4PSjn0pjYx1Y1ymJKL2qqRW5/E9xpoSBAHxUf72BWMGA5cpyS4MxhxmaPjD5jKl7RNSJ+rsamthZC4YayeZfiLyYBfL1Xjum3ysPXIDBWVql30fnV5ChVqPqHZSvG8UH+mP32p0qFCLVo9jZozs1X5ywx7K/DJly5PP3qavrQvGuExJRM4lSRL+9u8iVGsNyLlchR8KKnFflD8e7xWGIZ2DoHBipqesXgcJQGQ7aGvRVO/IhrqxX4vrMTg2yOJx3thnTJIkqNVqGAyGdlPnZ4+bN29Co7F9U4Y9JEmCTCaDn5+f3a9R+zoDPJBgzIw1CcDM9hizZ5lS1EFpZ80Y0LA/JQBuFE5ETnP0Wg3+c7MOMwd1wPC4YPxwoRLfnSvHikPXEX5cgUWPxiI2xNf2E9mhpLZ9tbUw6h7uB7kAnC2xEYx5YWZMrVZDqVS2KnHQHigUCqeW/IiiCLVaDX9/f7uO966Q3Y1snW729RkToWhNzZifMTPGYIzobnH0ag0+O150R18rSRKKG5usmiMaJGw4UYx7g30wsmcoAn3kGHNfOD75f92wMK0TarV67Dpbfkffu0arh07fvJ6juK5hLO2pgB8AfBUydAv3s1k35o2tLQwGg9cFYq6gUChgMBhsH9iIwZiDTMuUko0Cfju0umbMx1gzxhOD6G5x5Eo1/vdMmV0F5LfbeLIYz/3vBfwzv8Ls53efr8C1Ki2evT+q2XKkTBAw6N5AJMUG4VBhFXR6+3/JAA2bgb+cdRFzvruIc03GbWr42o4K+I3iI/1xvlQN0WC+YFiSJK/MjHnT0qSrtea1YjDmIHPLlGaPs2uZsrU1Y43BGPuMEd01tI2B0Ldnylr1dSdu1GLH6TIE+cqx7qffcPBSVbPP12j12PqfEvTtoMKD9waafY60uGDUaA04ds2+7YCMrlRqUFovoqhGh9f/eRmbfy6GTi+hpFaHIB8Z/BTt71dRfKQ/tHoJl8rN1xmJhob7urwtGPNUa9eutfi5V155BVlZWW04mtZrf2eAx2n5rshczZhdy5StrRnz5TIl0d1G27jUd+RKNW7WaO36mop6EasPX0fnEB988kQ39OmgwqrD1/HTlWrTMV/nlaJGo8f0gdEW39H3vycAYX5y7LtY2aoxnyluyIa9P7IL0uJCsD2vFK/tvoRfS+rbZVYMaFLEbyFDqWtcovK2ZUpP9eGHH7p7CA5hMOYgyYkd+HWtrRnzNRbwc5mS6G6h0UuIVCkgAMj61Xb9lkGSsObIDdTpDJiXci+CfOV4a/i96BHuh/dzruPEjVrcrNFi16/leLhbCLqF+1l8LrlMwLCuwfj39RpUafR2j/lMcT3C/OSIC/PFnCExeHPYvSirF1FQrml39WJGUQEKhPkrcNZCMHahsSVIRDu7OaE9qKqqQkZGBpKTk5GcnIzhw4ejvr4ew4YNw9SpUwEAW7ZsQXJyMkaMGIGcnBw3j9i2NjkLjh8/js2bN0MURQwfPhxPPfVUs89rtVosW7YMRUVFkMlkZo/xXGZaW5h5VynYCHslSYJe37plytDQcCiVSgQHh9j9NUTUvmlFA+4JVCIhWoXv8yvxTN9IBPhYzo7vPFuG4zdq8fyDHdAltOENnEopxzsPx2LBD4X484GriAtruDtwSv9Im98/LS4E354tx6HLVfhdrzC7xnymuB69o1SmjNvg2CD0jvLHllMlGBATYNdzeBpBENA70nIR/+7zFQjwkWFwJ/NLvt7AsO1TSFcuOvU5hdg4yJ6ZYfWY7du3o2vXrti4cSNqampw8eJFPPXUUzh48CAA4Pz581i/fj2ys7MRHByMF154waljdAWXZ8bUajXWr1+PhQsX4oMPPsDJkydRUFDQ4rgxY8bgww8/xPLly3HkyBFcunTJ1UNzCsHSdkhmjrRGbMUm4UbBwcGY9uwfEBpq3wWRiNo/nUGCj1yGJ+8LR71owPcXKiwee+ZmNTadLEZSbCBG9Qxt9rlAXznefSQW0QFK/FpSj7EJ4XZlceLCfNElxBf7LlbZPBYASut0KKrV4b6o5rf4h/gp8MJD92CIldYQni4+0h83a3SoqG/e/LVSLeLIlWo8HBcC33ZYD+fpevfujezsbLz//vs4fvw4EhMTm30+JycHo0aNQnBwMIBbq0iezOWZsfz8fMTFxSE0NBQAkJSUhBMnTqBbt26mY3x8fNCvXz/T3zt06ICKigpXD80pjMuUgo27KW2FaqLYcDIrHb1lmOUJRF5NK0rwUQnoHu6HxA4q7Dpbjt/Hh7doxlqn0+Pd3b8ixE+B2YNjzNaBhfo19A3bV1CFJ3rb96ZOEASkxQXj85PFuFGtRUyQj9XjzzbWi90ejHkDY93Y2ZJ6JDUJKn8oqIRoAEbeFgB7G1sZLFdJSUnBV199hb1792LlypU4dOhQs89rNJpWbS3oCVwespeXlyMk5NYyWnBwsNVAq6KiAufPn0fPnj1dPTQnsbOA38btlDpdQzDWmpoxIrr7aPQG+MgbLt1jeoehpE7E4cLqZsdcr9Lize8Lcb1KjVeHdkSQr+VlzAiVEk8nRrQqgzMsLhgCgP12FPKfKa6Hj1ywWovWXnWP8INC1ryI3yBJ+Gd+BRKi/NHZSc1xqbnc3FwEBARg+vTpePHFF1FQUAAfHx+UlpZCkiT0798f+/btgyiKqKmpaRcrbW1SM3Z7IGLMAt1Op9Nh1apVmDhxIgICzNcR7NmzB3v27AEALF26FJGRtmsc7pROq0FdSTECAi2v+ctFf6AWAGSm4yTfIEi+wc2OCw8JRoCf5UBLr294TcJCwxAeHn7HY5bJBPj5e/fdlQqFwqXzTneG89I2RKkAwQH+iIyMxKiICGw6VYZ/nK/C2AfiIAgC9pwrxtI9l6GUC1j+ZF8kdXZ+TWkkgAdiS/BjYS1mP9zb6pvNc+VX0ScmGPdERzl9HJ6gV9R1XKgQTf/2jxVW4Ea1DjOGxlk8H9rzuXLz5k23N32trKzEuHHjIAgCQkJCsGLFCnz//fcYMWIEEhISsHXrVmRnZ2PYsGGIiYmBTqeDXC63OW5n/1y+vr52z7PLX9HQ0FBUV99611ZVVWVasmxKFEWsXLkSAwYMQFpamsXnS09PR3p6uunjkpISZw63GZ1WA73BgNqaGovHaKXGfjsSTMfptTIYNM0zZnIYoKmzHCSVljb8HGq1GmVlresf1JQgA3x9vbtGITIy0qXzTneG89I21DoRkk5jeq0f7xmMT/51Ez/kFeLIlWr83/kKxEf647WUjrivc4jL5iS5kwprjlQi5+wV3BelMntMvc6A88U1GJcQ4bX/NrqHKrE7vwK/FRVDIROw/d/XEOQrR98wy7+f2vO5otFo3N5OaeTIkRg5cmSzx3r16oVZs2YBaIgnFi1ahEWLFjU7xlIiCGgIxKx9/k5oNJoW89yxY0ezx7r8t3bPnj1x4cIFVFZWQq/XIzc3F3379kVdXZ1pkBqNBsuWLUPv3r0xduxYVw/Jyewt4LfO+I/A0cicJWNE3k2rl0zLlADwcFwIgnzleG/fFfzf+QqMvS8cf36sM6Jc3L8rKTYQvnIB+wosF/KfL62HQfLOejEjY/PXi+VqlNeLyL1SjUfigpvNEZEtLs+M+fn5Ydq0aXjvvfeg1+uRmpqKhIQE7N+/H/v378e7776L/Px8/PLLLyguLsb+/fsBAA899BAmTZrk6uE5gX01Y7bodI13U7JmjIgskCSpIRhT3LrG+CpkGJcQjm/OlOGlwTF4sI1aKaiUctP2SDMGRUNpJvg4U1wPAUC8FwdjvaNuNX/9WVcHvQSM8PLCfXK+Nln4HTRoEAYNGtTssbS0NNNyZJ8+fbBly5a2GIoL2N6b0p79qW7dTclgjIjMM3bfvz3rMjYhAk/eF97m+wamxQXjwKUq/HS1Bildglt8/kxxPTqH+CLQSh+09i5SpUC4vwJniutxvlSNxA4qdApm4T61DvOoDrNvmdLWdkg6U5+x9tmNmohczxiM+ZrZ79AdGzj3vycAHYN8sP0/pdDftmG23iDh15J6U+bIWwmCgPhIfxwprMbNGh1G9gh195CoHWIw5iDJbDB220WxFZmx1jR9JaK7i3GTcE+pR5LLBEzuH4nLlRr8eLl57diVSg3qdAavrhcz6h3lB70EBPvKMSTWezvuk+t4xhndjgkwmHuw5Yc24jGxsWZMqWRmjIjMu7VM6Tm36gztHIS4MF9sPVUCnf5Wdux0Y7PXhOi7IBiLbLib9NFuIWZr54hs4b8aB0lOKuA3Zsbkjm767TnXaCJyMo1ozIx5zokuEwRM7R+F32p02NNka6YzxfUI81cg2sV3dXqCXpF+mDmoA8b1iXD3UKiVDh8+jIyMDHcPg8GY00hNXsrbr5N2LFPqRB0UCqVb6j6IqH2wVMDvbgM7BiAhyh9f5pWaAsazxXW4L8r/rrimyQQBj8eHWd3pgMgarok5rGVmrMXdlC0eaUkURRbvE5FVxmXApq0tPIEgCJg6IApvfF+IrF/LkRYXjKJaEU/09v4lSmp7V65cwbhx4zB48GDk5eWhsLAQMTExkCQJI0aMwDvvvAMA6N69O8aOHYtjx47B398fn3/+OaKjo3HhwgXMnTsXlZWV8PX1xT333AOgobP/vHnz8OuvvyI0NBQrVqxAr169sHLlSvz8888oKytDWVkZpk+fjlOnTuHkyZOIjIzEF198AZXKfONje/G3v8PsuZvSjgJ+nc7xTcKJyKtp9J63TGmUEK3CAx0DsON0qamVxd1QvH83W3/sJi6Wq536nHFhfnhuUAebx9XV1eGll15Cr169UFJSgsjISIiiiBEjRiA/Px89evSAXq/HxIkTsWLFCrz77rv48ssv8dJLL2Hu3Ln44x//iLS0NOzbtw+fffYZAOCDDz5AQkICPv30U+Tm5mLOnDnIzs4GAAQEBODTTz/F1atXkZ6ejm+++QYDBgzAiy++iN27dzvcsN6zct3tUkMwJjTtMya0/mXViTqnNHz1vEs0ETnLrdYWnnnpntI/CjVaAz47XgRfuYC4MO/bHJw8Q3R0NHr16gUAOHjwICZMmID09HRcvXoVxcXFAAClUokHHngAABAfH4+bN2+itrYWN27cMPU59fW91RPuyJEjGDduHAAgKSkJpaWlpu0c+/fvD19fX8TFxUGhUGDAgAEAgLi4ONy8edPhn4epGAeZb21xGzsipIZlSu8vdCWiO+eJd1M21S3cD6ldgvDj5WokdlBBIfPMcZJz2JPBcrWffvoJq1atwmeffYYePXogIyMDktSyfEgmk0GSJGg0GrtLgiw9j7WP75Rnvr1qR4QWwZiZZoywXcPPmjEissXT+oyZM6lfFBQyoG8Hx2poiOxRXl6O+Ph49OjRA2VlZbh27ZrV48PDG3aqOHv2LAAgLy/P9LmkpCR88803AIB//etfiIiIQFBQkOsG3wR/+ztKauwzZq6O38Sepq86qPydcfHiO1Eib6URPbOAv6mOwT746PfdEO7PXy/keg8//DC++OILJCcno1OnTnbdvbt06VL813/9F/z9/dG9e3fT43/84x8xb948DBs2DCEhIVi9erULR96cIJnLw7Uj169fd9lz67QaVBfdRG1NjcVj1IbTOC59i/uKUhAWkwoJMoj+EYD81i3OCqUSoQH+Vt/Nfrn9C0RERCL90VEOjVkuF6D08dwLtTNERkaipKTE3cOg23BeXO+b06XYcKIYWyf0hEppu40C58Qzted5qaurc/jOQU+kUChM/T6dxdxr1bFjR7PHem6uu72QzG0U3pxgc2dK1owRkW2eXsBPRHeGZ7TD9I1/Wtub0vaziKLOOTVj3p0UI7qrafUS5ELDnpBE5D0YjDnMdgG/9ccbiKIIJTNjRGSFRm/w6OJ9IrozPKsdZSrgbwi2JKDFrZO2FikNBgP0ej0U3CSciKzQipJHF+8T0Z1hMOaw2+9/MHOhFKy3tjAWDTIzRkTWaPUG+HCJksjrMBhzmNT4/zu/QIqiDgCcUjPGyzSR99LqJfgoeNkm8jY8qx3VeDelzIG9KY2ZMd5NSUTWaPWSx3bfJ2orV65cwSOPPOLuYTgVgzFHNdaM3VqstNCB38pT6HSNmTHWjBGRFVoW8BN5JZ7VDjP2GZM1/ag5m1shNQRjrBkjImu0egm+zIwRoaqqCjNmzMCwYcMwdepU1NbWYv78+UhOTkZycjJmz54NURSxdetWzJs3z/R1GzZswKJFiwAAmzdvxogRI5Camoq33noLAHDt2jWMGzcOycnJSEtLw+HDh9vk52EqxmGNd1NabW1h7zIlp4OILNPqDQhgBp08RN7xOlRV6G0f2ArBoXIkDrTd4V+v1+Ptt99GbGws3njjDWzatAnz589HZGQkAGDy5Mn48ccfMWbMGCxfvhy1tbUICAjAtm3b8NFHH+Hnn39GTk4OvvvuO8hkMsyYMQMHDhzA3r17MWrUKMyYMQOlpaUoKytz6s9nCc9qh93eZ6wlW+9jdawZIyI7aEQW8BMBQFhYGGJjYwEA6enp2LFjB+Lj4/H3v/8dV65cQXFxMYqLi6FSqfDoo49i165dSExMRGBgIHr06IHVq1cjNzfXVHumVqtRWFiIPn36YO3atairq0NycjIGDRrUJj8PgzFH2bEdEgTBemuLxpoxpdIJwRhXMIi8Fgv4yZPYk8FqCzqdDkVFRXjllVewceNG9OvXDwsWLIBx6+2JEyfivffeQ0JCAiZPngwAkCQJzz33HGbNmgWg+d6UiYmJOHjwIF5//XU8++yzmDp1qst/Br7FcpSx6atgbPpqo6mYGc5sbUFE3kurN3BfSiIAWq0Wer0eBoMB//M//4PU1FTExMSgX79+qK+vR35+vunYgQMHora2Fnv27MHo0aMBACkpKdi2bZtpw/ZLly7hwoUL2L9/P2JjY/H8889j4sSJKCgoaJOfh7/9HSY1+8McwUZwpmPNGBHZgZkxogaVlZV48sknUVpaikceeQSzZ89GQUEBhgwZgujoaMjl8mbHjxw5ErW1tfD19QUAPPjgg5g5cyaefvppaLVaBAcH44MPPsClS5fw5ptvQqlUIjo6GmvXrm2Tn4e//R0kNUZhginJaP5Cae3yeSszxpoxIrKsobUFgzG6u8XGxuI///lPi8dXr15t8Wv27duHNWvWNHts6tSppiVI4zJlQkICnn32WWcO1y7MdztMavL/Rq28Voo6EYIgtIjkiYiM9AYJogEs4CdqpePHj8PPzw89e/Z091As4lntMOsF/MYlSmtLlTpRB4VCYXM5k4juXlp9w7WGmTGi1hk4cCB27Njh7mFYxWDMUY0F/MZlyjvZo1IUdVyiJCKrtPqGaw2DMSLvw2DMUZKZZcqmBNt3V4qi6LTifV6mibzTrcwYL9tE3oZntaOE25u+Ng+H7AmORJ3onB5jROS1uExJ5L0YjDmoxQbhd1D3ZawZIyKyxLhMyT5jRN6HZ7WjJBt7UwqCzeyYU2vG+KaZyCsxM0bkvRiMOUgwtraQzL+Udi1TOrFmjIi8k0ZsLOBXMBgj8jYMxhwkme6mbPx/i+ukHZkxnY41Y0RkFQv4iW7ZunUrUlJSkJ6ejqSkJGRlZWHx4sUYOnQoBg8ejCVLlqC8vByDBw/GqVOnAAAbNmzAyy+/7OaRm8d0jKNMpWJ3foHUiSJbWxCRVbdqxpgZI89w8OBBFBcXO/U5o6KiMGzYMKvHnD9/HuvXr0d2djaCgoIwc+ZMnDt3Djk5OTh06BAkScKxY8cQFhaG5cuX45VXXsEnn3yCzZs345tvvnHqeJ2FwZjDjMuUFnqM2XHdFFnAT0Q2MDNG1ODQoUMYOXIkgoKCAAAqlQrdu3dHSUkJFixYgMGDB+Oxxx4DAAwbNgwpKSl46qmnsHnzZgQGBrpz6BYxAnBU4zKlBONWRnfQ2kIUoWQwRkRWmIIx1oyRh7CVwXIVtVrdorRHLpdjz5492LNnD3bv3o3169dj586dAGBKdtTV1bX5WO3Ft1hOIlhs+2r9wmkw6GEwGLhMSURWmQr4mRmju1xiYiJycnKg1+tRW1uLy5cvo7CwENeuXcP48eORmZmJixcvAmhYSs3Ly8O2bdvw2muvobKy0s2jN4/pGIcZ96aUw2zgZa6mvwmdTgQAKJxUwM/3zETeyZgZY80Y3e1SUlKQnZ2NlJQUREREQBRFBAYGYv78+aioqIBKpUJmZibKysqwYMECbN68GbGxsXj66afx5ptv4uOPP3b3j9ACgzEHSWhcphQawzLh9mVKW1sh6QCANWNEZJWxgF/JYIwICxYsQGZmJtRqNcaNG4fHHnsMGRkZLY47ePCg6e9z585tyyG2CiMAp5HhTjJjotiQGVNymZKIrNDqJShlAmR3sMsHkbcZP348ysvL4e/vj+nTpyMmJsbdQ3IIgzFHNW4ULljqwG+DztmZMV6nibySVi+x+z5Ro6ysLHcPwalYCeqwxtYWFqIgAYLV/SrFxpoxNn0lImu0egODMSIvxWDMYca7KAVYWqa0hjVjRGQPrSjBR8FLNpE34pntIMl0N6Wl5ha2Cvgb76ZkzRgRWaHhMiWR12Iw5jBjCGahgN/iow2cXjNGRF6pYZmSl2wib8Qz20kEC3Vh9mwSDrBmjIis0+ol9hgjAnDjxg1MmDDB3cNwKgZjDpIk496Uph3Dmx9gZ2sLZsaIyBoW8BM1iImJwfbt2909DKdiMOYgQbhVwG/+jkrrF0+9Xg+gYV8tp4zHKc9CRJ5Gq2cBPxEAlJWVYfDgwdDr9ViwYAGSk5MxevRo9O7dG2VlZaiqqkJGRgaSk5ORnJyMb7/91t1DtonpGAdJzWrGWhJM/zPP0JhZk8l4kSUiyzQiC/jJswQW74JCc8Opzyn6xqAm6gm7jt26dSsqKyuRk5MDnU6HIUOGAAC2b9+Orl27YuPGjaipqTHtU+nJGIw5zFZrC9sbhTccxmCMiCxjAT9RcwcPHsTUqVMhCAJ8fHxM5T69e/fGX//6VwQGBiIpKQmpqaluHqltDMYcZKoZs7Liay0ckwwSBEGweANAq/GNM5FXYgE/eRp7M1iuotFo4OPj0+LxlJQUfPXVV9i7dy9WrlyJQ4cO4Y033nDDCO3Ht1lOYmlDcFtBlsFg4BIlEdnEAn6i5hITE7Fv3z4AQGFhIWpqagAAubm5CAgIwPTp0/Hiiy+ioKDAncO0CzNjDmtcphQaC/hbea00SAYuURKRVZIkNe5NyWsFkdFzzz2HF154AampqejWrRsEQYC/vz/Kysrw2muvQRAEhISE4P3333f3UG1iMOYwy3tTGrNi1pJjEjNjRGSDXgIMEpgZIwIQHh6On376CTqdDl988QUUCgWuXbuGadOmwd/fH6NHj8bo0aPdPcxWYTDmoFvbIRk78LfuYtmwTMkLLBFZptUbAAA+Cl4riIyuX7+OiRMnQqlUIigoCH/+85/dPaQ7xmDMae7sImmQmBkjIuu0YsObPi5TEt3SpUsXHD582N3DcAqe2Q4yZcYEM3tT2nGHpMHAmjEisk5jzIxxmZLIKzEKcBozNWN2fJXBIDk1M+a0FhlE5DG0embGiLwZz2yH2dNnzHKAxNYWRGSLMRhjnzEi78QowGFNO/Dfxo4slcSaMSKyQSsaC/h5rSDyRjyzHWQMxSA0bvTdJP4SWvylJYPBABlrxojICo1pmZKZMSJvxCjAYZKVz9lZwM/WFkRkhZYF/ERejcGYEzRcHs28lEKzP8xi01cisuVWzRivFUQAsHXrVqSkpCA9PR1JSUnIysrC4sWLMXToUAwePBhLlizBpUuXkJKSYtpD+ty5cx7bDJZ9xhwkNfbev5UfE8z8zTIuUxKRLVouU5IHOn7jC1SoLzv1OUP9umBgzBSrx5w/fx7r169HdnY2goKCMHPmTJw7dw45OTk4dOgQJEnCsWPH0LVrV9xzzz3Izc3FkCFDsGXLFkyaNMmp43UWRgHOcocBlUFybmsLIvI+GhbwE5kcOnQII0eORFBQEABApVKhe/fuKCkpwYIFC5CVlYW+ffsCACZNmoStW7dCq9Vi9+7dePLJJ904csuYGXPY7ZmxpuypGdNDrvRx8piIyJswM0aeyFYGy1XUajWUSmWzx+RyOfbs2YM9e/Zg9+7dWL9+PXbu3InRo0cjMzMTX331FVJTUxEYGOiWMdvCt1kOkozXRmNmTDB3O6WVr3dy01ci8j46BmNEJomJicjJyYFer0dtbS0uX76MwsJCXLt2DePHj0dmZiYuXrwIAPDz88Pvfvc7vPPOO5g8ebKbR24ZM2MOkyw2dTU+bq3dmDObvrL5PpF30ugNEAAoeec1EVJSUpCdnY2UlBRERERAFEUEBgZi/vz5qKiogEqlQmZmpun4UaNG4ejRo+jfv78bR20dgzEHSWhMgJlqxlqXGeNG4URki1YvQSkXuN0ZUaMFCxYgMzMTarUa48aNw2OPPYaMjAyzx+7du9ejs2IAgzEnMG4Ubvkiae3yyY3CicgWrd7ArZCImhg/fjzKy8vh7++P6dOnIyYmxuxx9fX12LVrF/bu3dvGI2wdBmNO0FDAb26jcPuavsq49EBEVmj1EjcJJ2oiKyvLruP8/f1x9OhRF4/GcTy7HXbbfZSCub9bDrYktrYgIhu0ogQfBd+0EXkrRgEOaqgZEywsU9rX2oJNX4nIGo3ewMwYeQRjN3uyrTWvFc9uZzETjNnXgZ+ZMSKyrmGZkpkxcj+ZTAZRFN09DI8nimKrfrezZsxB0u3VYmaul7ZaWwgMxojIChbwk6fw8/ODWq2GRqPxqrt7fX19odFonPJcxvIjPz8/u7+mTYKx48ePY/PmzRBFEcOHD8dTTz11R8d4JmMa0kwffnuavrK1BRHZoNVLCPKRu3sYRBAEAf7+/u4ehtNFRkaipKTEbd/f5VGAWq3G+vXrsXDhQnzwwQc4efIkCgoKWn2MJ7NUM2ZsWWGrtQVrxojIGhbwE3k3l0cB+fn5iIuLQ2hoKORyOZKSknDixIlWH+OpJFOfMeMjjV33ZTK7UrjO7MBvV5EaEbU7LOAn8m4uP7vLy8sREhJi+jg4OBgVFRWtPsZjCY0LlLcFXvZkuyRJgiRJENhnjIis0LGAn8irtUnN2O0ZInN3YthzDADTruwAsHTpUnTs2NFJo7Sga5yNA9IdevqlS5c69PV3K5fPO90Rzotr/N/sO39dOSeeifPiedw5Jy7PjIWGhqK6utr0cVVVFUJDQ1t9jFF6ejqWLl3aZkHMn/70pzb5PmQ/zoln4rx4Hs6JZ+K8eB53z4nLg7GePXviwoULqKyshF6vR25uLvr27Yu6ujrTnQuWjiEiIiLydi5fpvTz88O0adPw3nvvQa/XIzU1FQkJCdi/fz/279+Pd9991+IxRERERN6uTWrGBg0ahEGDBjV7LC0tDWlpaVaP8QTp6Y7VhJHzcU48E+fF83BOPBPnxfO4e04EiRtNEREREbkNG9cQERERuRH3prSg/W7P5F20Wi2WLVuGoqIiyGQy01xUV1dj7dq1KCoqQnR0NObMmYPAwEB3D/eus3PnThw4cAArV67knHgAURSxZcsWHD16FKIoYvny5ZAkifPiRvv378euXbsgiiI6d+6MWbNmQafTcU7coKCgAOvWrcOKFSsAwOo1a8eOHThw4AAUCgWmTJmC+++/36VjY2bMjPa+PZO3GTNmDD788EMsX74cR44cwaVLl7Bp0yY89NBDWLNmDR566CFs377d3cO865w9exY5OTmmjzkn7rd+/Xr4+vpi7dq1WLduHQICAjgvblRRUYGvv/4aS5YswZo1axASEoLs7GzOiRts3LgRS5YsQdPKLEvzcPr0aZw4cQKrVq3CggULsGHDBou9T52FwZgZ7Xl7Jm/j4+ODfv36mf7eoUMHVFRUIC8vD0OHDgUAJCcnc37aWFVVFT7//HPMmDHD9BjnxL0qKipw7tw5jB8/HoIgmP7jvLiPKIrQaDRQq9UAGnpqKhQKzokbZGRktOhPamke8vLyMGTIEMhkMoSFhaFTp07Iz8936fgYjJnRrrdn8mIVFRU4f/48evbsierqagQEBAAAVCoVampq3Dy6u4ckSfj4448xZcqUZucJ58S9CgsLIQgCFi1ahDlz5mDt2rVQq9WcFzeKjIzE448/jrlz5+Ivf/kL8vPzMXLkSM6Jh7A0D2VlZQgODjYd1xYxAIMxC+zdnonahk6nw6pVqzBx4kQEBARwftzoH//4B+Lj49GnT59mj3NO3KuqqgoxMTF46623sGrVKoSEhODrr7/mvLhRXV0djh07hiVLlqB///4oKipCXl4e58RDWJsHmUxm8XOuwAJ+M1qzPRO5niiKWLlyJQYMGGDqTRcQEAC1Wg0/Pz/U1dWx+LUNFRUV4eeff8bBgweh1+tRWlqKt99+m3PiZgEBAfD19YVSqQQAPPjgg9i5cyfnxY1OnTqFe++9F506dUKnTp3g5+eH3bt3c048hKV5CA0NRVVVlem4togBmBkzg9szeQ6NRoNly5ahd+/eGDt2rOnxxMREHD58GABw6NAhzk8bmj59OtasWYPVq1dj4cKFiImJwaJFizgnbhYfH48zZ86gqKgIAHDy5En07NmT8+JG0dHROHv2rGn568KFC+jYsSPnxENYmoe+ffsiNzcXBoMB5eXluHjxInr06OHSsbDpqwXHjh3Dli1bTNszPf300+4e0l3pl19+QWZmJqKjo02PPfTQQ/j973+PtWvXori4GFFRUXj55ZebrfFT2ygqKsKyZcuwcuVKVFVVcU7c7NSpU9i0aRNEUUSvXr3w3HPPob6+nvPiRt999x12794NmUyGLl264A9/+IOptQXnpO18+eWXOHr0KG7cuIHY2FhkZGSgU6dOFufhq6++Qk5ODmQyGaZMmYIHHnjApeNjMEZERETkRlymJCIiInIjBmNEREREbsRgjIiIiMiNGIwRERERuRGDMSIiIiI3YjBGRB7vmWeewenTpwEAH3/8MbZt2+ay77V//34sXLiw2WOLFi3C119/7bLvSUR3N3bgJyKP58rgyx5vv/22W78/EXk39hkjIo83YcIEfPTRRzhz5gzWrVsHoGFfuZ49e2Lx4sUoKyvD3//+d/zyyy9QqVR44oknMGrUKAANmTSdTgeNRoPTp09jzJgxpn0bq6qqoFKpkJKSgqlTp6K0tBSzZ8+GJEmmvek2bdqEzMxMpKWlIS0tDZIk4dtvv8U///lPaDQaDBw4ENOmTYNKpUJRURFmz56NyZMn44cffkBVVRUeffRRTJ06FQCQn5+PDRs24PLlywgMDMSQIUOQkZHhnheViDwGM2NE1G4MHz4ceXl5iIiIwDPPPAMAMBgMWLZsGR544AHMnj0bv/32G5YsWYK4uDjEx8cDAAoKCjBz5kzMmzcPer0e5eXluP/++xEaGoqamhpkZmYiNzcXQ4cOxQsvvIAffvgBixcvNjuGAwcO4MCBA1i4cCGCg4Px17/+FZ999hlmzZplOqaurg6LFy9GdXU1Xn/9dSQnJ6Nbt25Ys2YNxowZg9TUVBQXF+Pf//636180IvJ4DMaIqF27cOECqqurMX78eAiCgK5duyIpKQknTpwwBWNDhw5FYmIiAEAul0Mul2PHjh345ZdfUFFRgfr6ely9etWu7/fjjz/i8ccfR0xMDABg0qRJmDt3Lp5//nnTMRMmTIBcLkdISAhiY2Nx/fp1dOvWDRqNBsXFxRBF0bR5NBERgzEiateKi4tRWlqKKVOmmB4zGAwYPny42eMlSUJmZia6dOmCefPmITo6Gv/93/8Ng8Fg1/crKytDeHi46eOIiAjo9XpUVlaaPV6pVEIURQDAnDlz8OWXXyIrKwvR0dEYO3Yshg0bZu+PSkReisEYEbUrMpkMTUtdw8PDER0djQ8//NCur6+srMS1a9eQmZkJlUpl9vmtCQ8PR2lpqenj0tJSUxas6ePm9OnTB4sWLYIoijh8+DA++eQTDBo0yOw4iOjuwdYWRNSuRERE4OzZs6itrUVZWRl69OgBHx8fbN26FTU1Naivr8fp06dNrTBuFxQUBJVKhVOnTkGv1+PYsWM4efJks+e/evUqioqKUFZWhtvvcUpNTUVWVhauX7+O2tpabNmyBSkpKZDL5VbHLYoiPv30U1y5cgWCICA4OBg+Pj7w8fFx+DUhovaNmTEialdGjBiBU6dOYcaMGRgwYADmz5+PP/3pT9i0aRPmzp0LrVaLzp07Y/LkyWa/Xi6XY9asWfjb3/6GdevWYcCAAYiNjTV9PiEhAQMHDsSrr74KPz8/fPLJJ82+fvjw4SgvL8fixYub3U1pi0wmg0ajQWZmJmpqanDvvffi1VdfhULByzDR3Y6tLYiIiIjciMuURERERG7EYIyIiIjIjRiMEREREbkRgzEiIiIiN2IwRkRERORGDMaIiIiI3IjBGBEREZEbMRgjIiIiciMGY0RERERu9P8BkFfxkYGZMbcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.style.use('ggplot')\n",
    "plt.figure(figsize=(10, 8))\n",
    "ax = plt.gca()\n",
    "\n",
    "for k, v in scores.items():\n",
    "    y = v[0]\n",
    "    e = v[1]\n",
    "    x = np.arange(len(y))\n",
    "    ax.plot(x, y, label=k)\n",
    "    if k in ['std', 'bayes', 'gsx', 'gsy', 'igs']:\n",
    "        ax.fill_between(x, y - e, y + e, alpha=0.1)\n",
    "plt.legend()\n",
    "plt.xlabel('iterations')\n",
    "plt.ylabel('R^2');\n",
    "plt.ylim(0., 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.9, 1.0)"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.style.use('ggplot')\n",
    "plt.figure(figsize=(10, 8))\n",
    "ax = plt.gca()\n",
    "\n",
    "for k, v in scores.items():\n",
    "    y = v[0]\n",
    "    e = v[1]\n",
    "    x = np.arange(len(y))\n",
    "    ax.plot(x, y, label=k)\n",
    "    if k in ['std', 'bayes', 'gsx', 'gsy', 'igs']:\n",
    "        ax.fill_between(x, y - e, y + e, alpha=0.1)\n",
    "plt.legend()\n",
    "plt.xlabel('iterations')\n",
    "plt.ylabel('R^2');\n",
    "plt.ylim(0.9, 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Check what the accuracy actually looks like"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_std = learner_accuracy['std'][1].predict(x_test)\n",
    "y_pred_random = learner_accuracy['random'][1].predict(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<module 'matplotlib.pyplot' from '/nix/store/c8sgkmibi2vyfw75w9vai2917j5smvq7-python3.8-matplotlib-3.3.1/lib/python3.8/site-packages/matplotlib/pyplot.py'>"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_parity(y_test, y_pred_random, label='random')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<module 'matplotlib.pyplot' from '/nix/store/c8sgkmibi2vyfw75w9vai2917j5smvq7-python3.8-matplotlib-3.3.1/lib/python3.8/site-packages/matplotlib/pyplot.py'>"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_parity(y_test, y_pred_std, label='std')"
   ]
  },
  {
   "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}

shell.nix

#
# $ nix-shell --pure --arg withBoost false --argstr tag 20.09
#

{
  tag ? "20.09",
  pymksVersion ? "cf653e004848c9c68ca31a85add0d1ac8611a93f"
}:
let
  pkgs = import (builtins.fetchTarball "https://github.com/NixOS/nixpkgs/archive/${tag}.tar.gz") {};
  pymkssrc = builtins.fetchTarball "https://github.com/materialsinnovation/pymks/archive/${pymksVersion}.tar.gz";
  pymks = pypkgs.callPackage "${pymkssrc}/default.nix" { graspi = null; };
  pypkgs = pkgs.python3Packages;
  extra = with pypkgs; [ black pylint flake8 ipywidgets zarr pymks h5py ];
in
  (pymks.overridePythonAttrs (old: rec {

    propagatedBuildInputs = old.propagatedBuildInputs;

    nativeBuildInputs = propagatedBuildInputs ++ extra;

    postShellHook = ''
      export OMPI_MCA_plm_rsh_agent=${pkgs.openssh}/bin/ssh

      SOURCE_DATE_EPOCH=$(date +%s)
      export PYTHONUSERBASE=$PWD/.local
      export USER_SITE=`python -c "import site; print(site.USER_SITE)"`
      export PYTHONPATH=$PYTHONPATH:$USER_SITE
      export PATH=$PATH:$PYTHONUSERBASE/bin

      jupyter nbextension install --py widgetsnbextension --user > /dev/null 2>&1
      jupyter nbextension enable widgetsnbextension --user --py > /dev/null 2>&1
      pip install jupyter_contrib_nbextensions --user > /dev/null 2>&1
      jupyter contrib nbextension install --user > /dev/null 2>&1
      jupyter nbextension enable spellchecker/main > /dev/null 2>&1

      pip install --user nbqa
      pip install --user tqdm
      pip install --user modAL
    '';
  }))