diff --git a/examples/MVT (Cholesky).ipynb b/examples/MVT (Cholesky).ipynb
new file mode 100644
index 00000000..b9bed7cf
--- /dev/null
+++ b/examples/MVT (Cholesky).ipynb
@@ -0,0 +1,1496 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "
\n",
+ "
Multivariate Student-T Example
\n",
+ " Cholesky Decomposition of Covariance Matrix $\\Sigma$
\n",
+ "\n",
+ "\n",
+ "
\n",
+ "
"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "In this example, we model and predict all parameters of a trivariate ($Y_{D}=3$) Student-T distribution. As a generalization of the multivariate Normal, the multivariate Student-T is suitable when modelling heavy-tailed data, i.e., when there is more mass in the tails of the distribution. The multivariate Student-T distribution has an additional degrees of freedom parameter $\\mathbf{\\nu}_{\\mathbf{x}} > 2$ that governs the tail behaviour, where for $\\mathbf{\\nu}_{\\mathbf{x}} \\rightarrow \\infty$ the Student-T converges in distribution to the multivariate Normal. The conditional means $\\mathbf{\\mu}(x) \\in \\mathbb{R}^{D}$ and the conditional covariance matrix $\\mathbf{\\nu}(x)\\big(\\mathbf{\\nu}(x)-2\\big)^{-1}\\mathbf{\\Sigma}(x) \\in \\mathbb{R}^{D \\times D}$ are given as follows"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "$$\n",
+ "\\mathbf{\\mu}(x)=\\begin{pmatrix}\\mu_{1}(x) \\\\ \\mu_{2}(x) \\\\ \\mu_{3}(x)\\end{pmatrix}, \\qquad \\qquad \n",
+ "\\mathbf{\\Sigma}(x)= \\begin{pmatrix}\n",
+ "\\sigma^{2}_{11}(x) & \\rho_{1,2}(x)\\sigma_{1}(x)\\sigma_{2}(x) & \\rho_{1,3}(x)\\sigma_{1}(x)\\sigma_{3}(x) \\\\\n",
+ "\\rho_{2,1}(x)\\sigma_{2}(x)\\sigma_{1}(x) & \\sigma^{2}_{22}(x) & \\rho_{2,3}(x)\\sigma_{2}(x)\\sigma_{3}(x) \\\\\n",
+ "\\rho_{3,1}(x)\\sigma_{3}(x)\\sigma_{1}(x) & \\rho_{3,2}(x)\\sigma_{3}(x)\\sigma_{2}(x) & \\sigma^{2}_{33}(x)\n",
+ "\\end{pmatrix}\n",
+ "$$"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "where $\\mathbf{\\mu}(x)$ and $\\mathbf{\\Sigma}(x)$ are defined as for the multivariate Normal. To ensure positive definiteness of $\\Sigma(\\cdot)$, the $D(D + 1)/2$ entries of the covariance matrix must satisfy specific conditions. For the bivariate case, this can be ensured by applying exponential functions to the variances and a suitable transformation to restrict the coefficient of correlation $\\rho \\in [-1,1]$. However, in high-dimensional settings, where all moments are modelled as functions of covariates, ensuring positive definiteness of the covariance matrix becomes challenging, since joint restrictions for the elements are necessary. A computationally more tractable approach to ensure positive definiteness is based on the Cholesky decomposition, that uniquely decomposes the covariance matrix as follows"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "$$\n",
+ "\\mathbf{\\Sigma}(x) = \\mathbf{L}(x) \\mathbf{L}^{\\prime}(x)\n",
+ "$$"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "where $\\mathbf{L}(\\cdot) \\in \\mathbb{R}^{D \\times D}$ is a lower triangular matrix. To ensure $\\mathbf{\\Sigma}(\\cdot)$ to be positive definite, the $D$ diagonal elements $\\ell_{ii}$ of\n",
+ "$\\mathbf{L}(\\cdot)$ need to be strictly positive, whereas all $D(D −1)/2$ off diagonal elements $\\ell_{ij}$ can take on any value. For the trivariate case, the Cholesky factor $\\mathbf{L}(\\cdot)$ is given as follows"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "$$\n",
+ "\\mathbf{L}(x)= \\begin{pmatrix}\n",
+ "\\exp\\big(\\ell_{11}(x)\\big) & 0 & 0 \\\\\n",
+ "\\ell_{21}(x) & \\exp\\big(\\ell_{22}(x)\\big) & 0 \\\\\n",
+ "\\ell_{31}(x) & \\ell_{32}(x) & \\exp\\big(\\ell_{33}(x)\\big)\\\\\n",
+ "\\end{pmatrix}\n",
+ "$$"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "tags": []
+ },
+ "source": [
+ "Given the usefulness of the Cholesky decomposition, instead of estimating the entries of $\\mathbf{\\Sigma}(\\cdot)$ directly, XGboostlss estimates the Cholesky factors $\\mathbf{L}(\\cdot)$ and then uses these for creating $\\mathbf{\\Sigma}(\\cdot)$. However, in contrast to the original formulation of $\\mathbf{\\Sigma}(\\cdot)$, the elements in $\\mathbf{L}(\\cdot)$ *do not have any direct interpretation*. Since XGBoostLSS is based on a *one vs. all estimation strategy*, where a separate tree is grown for each distributional parameter, **estimating many parameters for a large dataset can become computationally extremely expensive**. For more details, we refer to our related paper **[März, Alexander (2022), *Multi-Target XGBoostLSS Regression*](https://arxiv.org/abs/2210.06831)**.\n",
+ "
\n",
+ "
"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Imports"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 17,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from xgboostlss.model import *\n",
+ "from xgboostlss.distributions.MVT import *\n",
+ "from xgboostlss.datasets.data_loader import load_simulated_multivariate_studentT_data\n",
+ "\n",
+ "from sklearn.model_selection import train_test_split\n",
+ "import pandas as pd\n",
+ "import multiprocessing\n",
+ "import plotnine\n",
+ "from plotnine import *\n",
+ "plotnine.options.figure_size = (18, 9)\n",
+ "n_cpu = multiprocessing.cpu_count()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Data"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "data_sim = load_simulated_multivariate_studentT_data()\n",
+ "\n",
+ "# Create 60%, 20%, 20% split for train, validation and test \n",
+ "train, validate, test = np.split(data_sim.sample(frac=1,random_state=123), [int(0.6*len(data_sim)), int(0.8*len(data_sim))])\n",
+ "\n",
+ "# Train\n",
+ "x_train = train.filter(regex=\"x\")\n",
+ "y_train = train.filter(regex=\"y\").values\n",
+ "n_targets = y_train.shape[1]\n",
+ "dtrain = xgb.DMatrix(x_train, label=y_train, nthread=n_cpu)\n",
+ "\n",
+ "# Validation\n",
+ "x_eval = validate.filter(regex=\"x\")\n",
+ "y_eval = validate.filter(regex=\"y\").values\n",
+ "deval = xgb.DMatrix(x_eval, label=y_eval, nthread=n_cpu)\n",
+ "\n",
+ "# Test\n",
+ "x_test = test.filter(regex=\"x\")\n",
+ "y_test = test.filter(regex=\"y\").values\n",
+ "dtest = xgb.DMatrix(x_test, nthread=n_cpu)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Distribution Selection"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "# Specifies a multivariate Student-T distribution, using the Cholesky decompoisition. See ?MVT for details.\n",
+ "xgblss = XGBoostLSS(\n",
+ " MVT(D=n_targets, # Specifies the number of targets\n",
+ " stabilization=\"None\", # Options are \"None\", \"MAD\", \"L2\".\n",
+ " response_fn=\"exp\", # Function to transform the lower-triangular factor of the covariance, e.g., \"exp\" or \"softplus\".\n",
+ " loss_fn=\"nll\" # Loss function, i.e., nll.\n",
+ " ) \n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Hyper-Parameter Optimization\n",
+ "\n",
+ "Any XGBoost hyperparameter can be tuned, where the structure of the parameter dictionary needs to be as follows:\n",
+ "\n",
+ " - Float/Int sample_type\n",
+ " - {\"param_name\": [\"sample_type\", low, high, log]}\n",
+ " - sample_type: str, Type of sampling, e.g., \"float\" or \"int\"\n",
+ " - low: int, Lower endpoint of the range of suggested values\n",
+ " - high: int, Upper endpoint of the range of suggested values\n",
+ " - log: bool, Flag to sample the value from the log domain or not\n",
+ " - Example: {\"eta\": \"float\", low=1e-5, high=1, log=True]}\n",
+ "\n",
+ " - Categorical sample_type\n",
+ " - {\"param_name\": [\"sample_type\", [\"choice1\", \"choice2\", \"choice3\", \"...\"]]}\n",
+ " - sample_type: str, Type of sampling, either \"categorical\"\n",
+ " - choice1, choice2, choice3, ...: str, Possible choices for the parameter\n",
+ " - Example: {\"booster\": [\"categorical\", [\"gbtree\", \"dart\"]]}\n",
+ "\n",
+ " - For parameters without tunable choice (this is needed if tree_method = \"gpu_hist\" and gpu_id needs to be specified)\n",
+ " - {\"param_name\": [\"none\", [value]]},\n",
+ " - param_name: str, Name of the parameter\n",
+ " - value: int, Value of the parameter\n",
+ " - Example: {\"gpu_id\": [\"none\", [0]]}\n",
+ "\n",
+ "Depending on which parameters are optimized, it might happen that some of them are not used, e.g., when {\"booster\": [\"categorical\", [\"gbtree\", \"gblinear\"]]} and {\"max_depth\": [\"int\", 1, 10, False]} are specified, max_depth is not used when gblinear is sampled, since it has no such argument."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "\u001b[32m[I 2023-06-22 09:15:06,410]\u001b[0m A new study created in memory with name: XGBoostLSS Hyper-Parameter Optimization\u001b[0m\n",
+ "C:\\Users\\maerzale\\.virtualenvs\\XGBoostLSS-vIPRRz-M\\lib\\site-packages\\optuna\\progress_bar.py:56: ExperimentalWarning: Progress bar is experimental (supported from v1.2.0). The interface can change in the future.\n"
+ ]
+ },
+ {
+ "data": {
+ "application/vnd.jupyter.widget-view+json": {
+ "model_id": "abff79daa9b24dad80429d4b020b3dcb",
+ "version_major": 2,
+ "version_minor": 0
+ },
+ "text/plain": [
+ " 0%| | 0/20 [00:00, ?it/s]"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\u001b[32m[I 2023-06-22 09:16:05,270]\u001b[0m Trial 0 finished with value: 6413.5888672 and parameters: {'eta': 0.00015015681268702434, 'max_depth': 10, 'gamma': 0.0005522757165865727, 'subsample': 0.5528145862106946, 'colsample_bytree': 0.520626719874306, 'min_child_weight': 0.0008561271530256406, 'booster': 'gbtree'}. Best is trial 0 with value: 6413.5888672.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:17:08,148]\u001b[0m Trial 1 finished with value: 6429.0069336 and parameters: {'eta': 8.987086142669043e-05, 'max_depth': 9, 'gamma': 3.838082616905876e-06, 'subsample': 0.6229266642696729, 'colsample_bytree': 0.4639420830560823, 'min_child_weight': 0.0011999874824879088, 'booster': 'gbtree'}. Best is trial 0 with value: 6413.5888672.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:17:59,658]\u001b[0m Trial 2 finished with value: 5568.4526366 and parameters: {'eta': 0.01908053627257867, 'max_depth': 8, 'gamma': 0.032343331893412645, 'subsample': 0.9097320442665298, 'colsample_bytree': 0.5778612040110485, 'min_child_weight': 0.0013651143838974407, 'booster': 'gbtree'}. Best is trial 2 with value: 5568.4526366.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:18:56,717]\u001b[0m Trial 3 finished with value: 5283.158398199999 and parameters: {'eta': 0.017156445149174616, 'max_depth': 7, 'gamma': 9.3106528704997, 'subsample': 0.7100686661480411, 'colsample_bytree': 0.9770160382003545, 'min_child_weight': 9.966087109371487e-07, 'booster': 'gbtree'}. Best is trial 3 with value: 5283.158398199999.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:19:45,989]\u001b[0m Trial 4 finished with value: 6006.3546876 and parameters: {'eta': 0.002847273698807345, 'max_depth': 3, 'gamma': 4.964165373215629, 'subsample': 0.4550562788155052, 'colsample_bytree': 0.4978966464592219, 'min_child_weight': 1.9584813067320013, 'booster': 'gbtree'}. Best is trial 3 with value: 5283.158398199999.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:20:34,724]\u001b[0m Trial 5 finished with value: 6410.0606446 and parameters: {'eta': 0.00019042642156851433, 'max_depth': 3, 'gamma': 0.0049779810602736405, 'subsample': 0.43646319161872954, 'colsample_bytree': 0.6968612539615116, 'min_child_weight': 0.3670679152623085, 'booster': 'gbtree'}. Best is trial 3 with value: 5283.158398199999.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:20:48,833]\u001b[0m Trial 6 finished with value: 5768.8638672 and parameters: {'eta': 0.2481201772276993, 'max_depth': 9, 'gamma': 6.481874848260236e-05, 'subsample': 0.5995539543240552, 'colsample_bytree': 0.774591625394921, 'min_child_weight': 0.016900828792366693, 'booster': 'gbtree'}. Best is trial 3 with value: 5283.158398199999.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:21:27,613]\u001b[0m Trial 7 finished with value: 5230.4515624 and parameters: {'eta': 0.13188558719608656, 'max_depth': 7, 'gamma': 28.53963416261476, 'subsample': 0.7133825226801791, 'colsample_bytree': 0.7860945555989556, 'min_child_weight': 4.099012416582127, 'booster': 'gbtree'}. Best is trial 7 with value: 5230.4515624.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:22:16,097]\u001b[0m Trial 8 finished with value: 5829.6657228 and parameters: {'eta': 0.004707287489510579, 'max_depth': 3, 'gamma': 5.433144587328669e-05, 'subsample': 0.42279809967785414, 'colsample_bytree': 0.5329703752797116, 'min_child_weight': 0.27792105591085403, 'booster': 'gbtree'}. Best is trial 7 with value: 5230.4515624.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:23:02,860]\u001b[0m Trial 9 finished with value: 5276.3163086 and parameters: {'eta': 0.13853174140678592, 'max_depth': 1, 'gamma': 3.544379910172147e-07, 'subsample': 0.9069229763171536, 'colsample_bytree': 0.9119820496327791, 'min_child_weight': 8.227388521257215e-06, 'booster': 'gbtree'}. Best is trial 7 with value: 5230.4515624.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:23:49,933]\u001b[0m Trial 10 finished with value: 5704.2018554 and parameters: {'eta': 0.34201808336947187, 'max_depth': 6, 'gamma': 2.3770120082375134e-08, 'subsample': 0.24722140815244298, 'colsample_bytree': 0.27102299765463517, 'min_child_weight': 350.79417142972017, 'booster': 'gbtree'}. Best is trial 7 with value: 5230.4515624.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:24:11,891]\u001b[0m Trial 11 finished with value: 5307.296093800001 and parameters: {'eta': 0.8368383287338174, 'max_depth': 1, 'gamma': 1.729869262555111e-08, 'subsample': 0.9971343248968885, 'colsample_bytree': 0.9790747520337267, 'min_child_weight': 5.522049459235828e-07, 'booster': 'gbtree'}. Best is trial 7 with value: 5230.4515624.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:24:35,773]\u001b[0m Trial 12 finished with value: 5325.4541994 and parameters: {'eta': 0.08527162754633608, 'max_depth': 5, 'gamma': 0.21026417136851885, 'subsample': 0.8023524195744576, 'colsample_bytree': 0.8296809209307119, 'min_child_weight': 1.4419703384349118e-05, 'booster': 'gbtree'}. Best is trial 7 with value: 5230.4515624.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:25:21,886]\u001b[0m Trial 13 finished with value: 5318.091992199999 and parameters: {'eta': 0.11056590166596084, 'max_depth': 1, 'gamma': 0.40591572664128744, 'subsample': 0.8212147636571172, 'colsample_bytree': 0.8665425081708905, 'min_child_weight': 3.321996703112353e-08, 'booster': 'gbtree'}. Best is trial 7 with value: 5230.4515624.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:25:33,018]\u001b[0m Trial 14 pruned. Trial was pruned at iteration 20.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:26:06,936]\u001b[0m Trial 15 finished with value: 5294.1089844 and parameters: {'eta': 0.05240741421967161, 'max_depth': 7, 'gamma': 1.3563315621816877e-06, 'subsample': 0.9384525445646403, 'colsample_bytree': 0.8914993539488475, 'min_child_weight': 57.609818070560394, 'booster': 'gbtree'}. Best is trial 7 with value: 5230.4515624.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:26:17,793]\u001b[0m Trial 16 pruned. Trial was pruned at iteration 20.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:26:49,356]\u001b[0m Trial 17 finished with value: 5253.330078199999 and parameters: {'eta': 0.1959768979884324, 'max_depth': 6, 'gamma': 37.46828526567046, 'subsample': 0.7138652892491867, 'colsample_bytree': 0.9149240542028265, 'min_child_weight': 0.021649714209434064, 'booster': 'gbtree'}. Best is trial 7 with value: 5230.4515624.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:27:09,885]\u001b[0m Trial 18 finished with value: 5282.545605599999 and parameters: {'eta': 0.36838710685261233, 'max_depth': 7, 'gamma': 36.48739027929726, 'subsample': 0.7434372119969223, 'colsample_bytree': 0.8088085263421839, 'min_child_weight': 0.030622068563904398, 'booster': 'gbtree'}. Best is trial 7 with value: 5230.4515624.\u001b[0m\n",
+ "\u001b[32m[I 2023-06-22 09:27:44,579]\u001b[0m Trial 19 finished with value: 5280.036035200001 and parameters: {'eta': 0.049171821586478794, 'max_depth': 6, 'gamma': 0.7800025149092364, 'subsample': 0.6866784256040568, 'colsample_bytree': 0.6671767737867261, 'min_child_weight': 27.173793414732653, 'booster': 'gbtree'}. Best is trial 7 with value: 5230.4515624.\u001b[0m\n",
+ "\n",
+ "Hyper-Parameter Optimization successfully finished.\n",
+ " Number of finished trials: 20\n",
+ " Best trial:\n",
+ " Value: 5230.4515624\n",
+ " Params: \n",
+ " eta: 0.13188558719608656\n",
+ " max_depth: 7\n",
+ " gamma: 28.53963416261476\n",
+ " subsample: 0.7133825226801791\n",
+ " colsample_bytree: 0.7860945555989556\n",
+ " min_child_weight: 4.099012416582127\n",
+ " booster: gbtree\n",
+ " opt_rounds: 48\n"
+ ]
+ }
+ ],
+ "source": [
+ "param_dict = {\n",
+ " \"eta\": [\"float\", {\"low\": 1e-5, \"high\": 1, \"log\": True}],\n",
+ " \"max_depth\": [\"int\", {\"low\": 1, \"high\": 10, \"log\": False}],\n",
+ " \"gamma\": [\"float\", {\"low\": 1e-8, \"high\": 40, \"log\": True}],\n",
+ " \"subsample\": [\"float\", {\"low\": 0.2, \"high\": 1.0, \"log\": False}],\n",
+ " \"colsample_bytree\": [\"float\", {\"low\": 0.2, \"high\": 1.0, \"log\": False}],\n",
+ " \"min_child_weight\": [\"float\", {\"low\": 1e-8, \"high\": 500, \"log\": True}],\n",
+ "}\n",
+ "\n",
+ "np.random.seed(123)\n",
+ "opt_param = xgblss.hyper_opt(param_dict,\n",
+ " dtrain,\n",
+ " num_boost_round=100, # Number of boosting iterations.\n",
+ " nfold=5, # Number of cv-folds.\n",
+ " early_stopping_rounds=20, # Number of early-stopping rounds\n",
+ " max_minutes=120, # Time budget in minutes, i.e., stop study after the given number of minutes.\n",
+ " n_trials=20, # The number of trials. If this argument is set to None, there is no limitation on the number of trials.\n",
+ " silence=False, # Controls the verbosity of the trail, i.e., user can silence the outputs of the trail.\n",
+ " seed=123, # Seed used to generate cv-folds.\n",
+ " hp_seed=None # Seed for random number generator used in the Bayesian hyperparameter search.\n",
+ " )"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Model Training"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "np.random.seed(123)\n",
+ "\n",
+ "opt_params = opt_param.copy()\n",
+ "n_rounds = opt_params[\"opt_rounds\"]\n",
+ "del opt_params[\"opt_rounds\"]\n",
+ "\n",
+ "# Add evaluation set\n",
+ "eval_set = [(dtrain,\"train\"), (deval,\"evaluation\")]\n",
+ "eval_result = {}\n",
+ " \n",
+ "# Train Model with optimized hyperparameters\n",
+ "xgblss.train(\n",
+ " opt_params,\n",
+ " dtrain,\n",
+ " num_boost_round=n_rounds,\n",
+ " evals=eval_set, \n",
+ " evals_result=eval_result,\n",
+ " verbose_eval=False\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {
+ "image/png": {
+ "height": 900,
+ "width": 1800
+ }
+ },
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/plain": [
+ ""
+ ]
+ },
+ "execution_count": 6,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# Note that XGBoostLSS uses NLL.sum() instead of NLL.mean() for training, so that train-nll and evaluation-nll are not comparable. Hence we manually adjust them.\n",
+ "n_train = y_train.shape[0],\n",
+ "n_eval = y_eval.shape[0]\n",
+ "\n",
+ "eval_df = pd.DataFrame.from_dict({\"train\": np.array(eval_result[\"train\"][\"nll\"]) / n_train,\n",
+ " \"evaluation\": np.array(eval_result[\"evaluation\"][\"nll\"]) / n_eval\n",
+ " })\n",
+ "eval_df[\"iter\"] = eval_df.index + 1\n",
+ "eval_df = eval_df.melt(id_vars=\"iter\")\n",
+ "\n",
+ "(\n",
+ " ggplot(eval_df,\n",
+ " aes(x=\"iter\",\n",
+ " y=\"value\",\n",
+ " color=\"variable\")) + \n",
+ " geom_line() + \n",
+ " labs(title=\"XGBoostLSS Train and Eval Loss\",\n",
+ " x=\"Iteration\",\n",
+ " y=\"NLL\") + \n",
+ " theme_bw(base_size=15) + \n",
+ " theme(legend_position=\"bottom\",\n",
+ " legend_title = element_blank()\n",
+ " )\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Prediction"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "# Set seed for reproducibility\n",
+ "torch.manual_seed(123)\n",
+ "\n",
+ "# Number of samples to draw from predicted distribution\n",
+ "n_samples = 1000\n",
+ "quant_sel = [0.05, 0.95] # Quantiles to calculate from predicted distribution\n",
+ "\n",
+ "# Sample from predicted distribution\n",
+ "pred_samples = xgblss.predict(dtest,\n",
+ " pred_type=\"samples\",\n",
+ " n_samples=n_samples,\n",
+ " seed=123)\n",
+ "\n",
+ "# Calculate quantiles from predicted distribution\n",
+ "pred_quantiles = xgblss.predict(dtest,\n",
+ " pred_type=\"quantiles\",\n",
+ " n_samples=n_samples,\n",
+ " quantiles=quant_sel)\n",
+ "\n",
+ "# Returns predicted distributional parameters\n",
+ "pred_params = xgblss.predict(dtest,\n",
+ " pred_type=\"parameters\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " | \n",
+ " target | \n",
+ " y_sample0 | \n",
+ " y_sample1 | \n",
+ " y_sample2 | \n",
+ " y_sample3 | \n",
+ " y_sample4 | \n",
+ " y_sample5 | \n",
+ " y_sample6 | \n",
+ " y_sample7 | \n",
+ " y_sample8 | \n",
+ " ... | \n",
+ " y_sample990 | \n",
+ " y_sample991 | \n",
+ " y_sample992 | \n",
+ " y_sample993 | \n",
+ " y_sample994 | \n",
+ " y_sample995 | \n",
+ " y_sample996 | \n",
+ " y_sample997 | \n",
+ " y_sample998 | \n",
+ " y_sample999 | \n",
+ "
\n",
+ " \n",
+ " \n",
+ " \n",
+ " 0 | \n",
+ " y1 | \n",
+ " 1.667932 | \n",
+ " 1.631851 | \n",
+ " -2.978731 | \n",
+ " 2.746265 | \n",
+ " 2.059820 | \n",
+ " -1.359539 | \n",
+ " 1.571172 | \n",
+ " 1.759640 | \n",
+ " 2.175634 | \n",
+ " ... | \n",
+ " 1.043452 | \n",
+ " 0.502470 | \n",
+ " 2.391257 | \n",
+ " -0.245706 | \n",
+ " 3.900496 | \n",
+ " 0.572212 | \n",
+ " 0.332272 | \n",
+ " 2.604990 | \n",
+ " 1.990996 | \n",
+ " 2.425229 | \n",
+ "
\n",
+ " \n",
+ " 1 | \n",
+ " y1 | \n",
+ " -0.512730 | \n",
+ " 0.230016 | \n",
+ " 1.467629 | \n",
+ " -0.600538 | \n",
+ " -2.796225 | \n",
+ " 0.919563 | \n",
+ " 0.140832 | \n",
+ " 0.244515 | \n",
+ " -1.500847 | \n",
+ " ... | \n",
+ " 5.274843 | \n",
+ " -0.907351 | \n",
+ " -0.789502 | \n",
+ " 0.573366 | \n",
+ " -1.675156 | \n",
+ " 0.222012 | \n",
+ " -0.917026 | \n",
+ " 0.629812 | \n",
+ " -0.396039 | \n",
+ " 1.318543 | \n",
+ "
\n",
+ " \n",
+ " 2 | \n",
+ " y1 | \n",
+ " 0.561088 | \n",
+ " -0.748004 | \n",
+ " 0.152646 | \n",
+ " 2.687419 | \n",
+ " -0.009103 | \n",
+ " 1.737594 | \n",
+ " -0.057163 | \n",
+ " 0.098457 | \n",
+ " 2.692869 | \n",
+ " ... | \n",
+ " 2.186634 | \n",
+ " 2.075422 | \n",
+ " 0.191973 | \n",
+ " 1.258858 | \n",
+ " 3.128220 | \n",
+ " 0.700489 | \n",
+ " 1.632449 | \n",
+ " 0.053039 | \n",
+ " 1.424659 | \n",
+ " 2.253676 | \n",
+ "
\n",
+ " \n",
+ " 3 | \n",
+ " y1 | \n",
+ " -0.553102 | \n",
+ " 1.110543 | \n",
+ " 2.103223 | \n",
+ " -1.816552 | \n",
+ " -0.869258 | \n",
+ " -0.275233 | \n",
+ " 0.267580 | \n",
+ " 0.816164 | \n",
+ " 0.878204 | \n",
+ " ... | \n",
+ " -0.082561 | \n",
+ " -1.204835 | \n",
+ " -0.934094 | \n",
+ " 1.607494 | \n",
+ " 1.880708 | \n",
+ " 0.125096 | \n",
+ " 1.173582 | \n",
+ " 0.309261 | \n",
+ " 1.180910 | \n",
+ " -0.961877 | \n",
+ "
\n",
+ " \n",
+ " 4 | \n",
+ " y1 | \n",
+ " 1.091711 | \n",
+ " -0.437804 | \n",
+ " 2.668892 | \n",
+ " 1.585261 | \n",
+ " 1.041862 | \n",
+ " 1.933763 | \n",
+ " 1.987233 | \n",
+ " -6.064761 | \n",
+ " 0.172949 | \n",
+ " ... | \n",
+ " -0.087848 | \n",
+ " 3.759326 | \n",
+ " -0.996257 | \n",
+ " 5.680374 | \n",
+ " 1.832690 | \n",
+ " 0.178290 | \n",
+ " 2.072301 | \n",
+ " 0.520087 | \n",
+ " 0.358304 | \n",
+ " 1.978452 | \n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
5 rows × 1001 columns
\n",
+ "
"
+ ],
+ "text/plain": [
+ " target y_sample0 y_sample1 y_sample2 y_sample3 y_sample4 y_sample5 \n",
+ "0 y1 1.667932 1.631851 -2.978731 2.746265 2.059820 -1.359539 \\\n",
+ "1 y1 -0.512730 0.230016 1.467629 -0.600538 -2.796225 0.919563 \n",
+ "2 y1 0.561088 -0.748004 0.152646 2.687419 -0.009103 1.737594 \n",
+ "3 y1 -0.553102 1.110543 2.103223 -1.816552 -0.869258 -0.275233 \n",
+ "4 y1 1.091711 -0.437804 2.668892 1.585261 1.041862 1.933763 \n",
+ "\n",
+ " y_sample6 y_sample7 y_sample8 ... y_sample990 y_sample991 \n",
+ "0 1.571172 1.759640 2.175634 ... 1.043452 0.502470 \\\n",
+ "1 0.140832 0.244515 -1.500847 ... 5.274843 -0.907351 \n",
+ "2 -0.057163 0.098457 2.692869 ... 2.186634 2.075422 \n",
+ "3 0.267580 0.816164 0.878204 ... -0.082561 -1.204835 \n",
+ "4 1.987233 -6.064761 0.172949 ... -0.087848 3.759326 \n",
+ "\n",
+ " y_sample992 y_sample993 y_sample994 y_sample995 y_sample996 \n",
+ "0 2.391257 -0.245706 3.900496 0.572212 0.332272 \\\n",
+ "1 -0.789502 0.573366 -1.675156 0.222012 -0.917026 \n",
+ "2 0.191973 1.258858 3.128220 0.700489 1.632449 \n",
+ "3 -0.934094 1.607494 1.880708 0.125096 1.173582 \n",
+ "4 -0.996257 5.680374 1.832690 0.178290 2.072301 \n",
+ "\n",
+ " y_sample997 y_sample998 y_sample999 \n",
+ "0 2.604990 1.990996 2.425229 \n",
+ "1 0.629812 -0.396039 1.318543 \n",
+ "2 0.053039 1.424659 2.253676 \n",
+ "3 0.309261 1.180910 -0.961877 \n",
+ "4 0.520087 0.358304 1.978452 \n",
+ "\n",
+ "[5 rows x 1001 columns]"
+ ]
+ },
+ "execution_count": 8,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "pred_samples.head()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " | \n",
+ " target | \n",
+ " y_sample0 | \n",
+ " y_sample1 | \n",
+ " y_sample2 | \n",
+ " y_sample3 | \n",
+ " y_sample4 | \n",
+ " y_sample5 | \n",
+ " y_sample6 | \n",
+ " y_sample7 | \n",
+ " y_sample8 | \n",
+ " ... | \n",
+ " y_sample990 | \n",
+ " y_sample991 | \n",
+ " y_sample992 | \n",
+ " y_sample993 | \n",
+ " y_sample994 | \n",
+ " y_sample995 | \n",
+ " y_sample996 | \n",
+ " y_sample997 | \n",
+ " y_sample998 | \n",
+ " y_sample999 | \n",
+ "
\n",
+ " \n",
+ " \n",
+ " \n",
+ " 5995 | \n",
+ " y3 | \n",
+ " 2.442450 | \n",
+ " 1.802108 | \n",
+ " 0.294235 | \n",
+ " 2.372050 | \n",
+ " 1.303846 | \n",
+ " 1.788222 | \n",
+ " 3.237095 | \n",
+ " 1.963390 | \n",
+ " 1.678264 | \n",
+ " ... | \n",
+ " 1.518361 | \n",
+ " 1.330022 | \n",
+ " -0.248718 | \n",
+ " 3.776295 | \n",
+ " 0.718719 | \n",
+ " 3.304478 | \n",
+ " 3.345208 | \n",
+ " 1.574800 | \n",
+ " 1.465450 | \n",
+ " -1.156816 | \n",
+ "
\n",
+ " \n",
+ " 5996 | \n",
+ " y3 | \n",
+ " 3.541716 | \n",
+ " 1.994839 | \n",
+ " 5.250106 | \n",
+ " 1.521251 | \n",
+ " 1.237782 | \n",
+ " 0.746012 | \n",
+ " 3.503881 | \n",
+ " 1.513360 | \n",
+ " 2.796339 | \n",
+ " ... | \n",
+ " 0.480239 | \n",
+ " 2.537787 | \n",
+ " -5.611905 | \n",
+ " 5.063916 | \n",
+ " 3.080128 | \n",
+ " -0.095438 | \n",
+ " 3.740144 | \n",
+ " 1.901680 | \n",
+ " 2.199507 | \n",
+ " 2.030381 | \n",
+ "
\n",
+ " \n",
+ " 5997 | \n",
+ " y3 | \n",
+ " -2.000938 | \n",
+ " 3.227403 | \n",
+ " 0.932215 | \n",
+ " -0.159020 | \n",
+ " 5.510897 | \n",
+ " -1.754066 | \n",
+ " -0.925815 | \n",
+ " 1.880578 | \n",
+ " 2.218145 | \n",
+ " ... | \n",
+ " 2.134847 | \n",
+ " 0.070474 | \n",
+ " -3.396575 | \n",
+ " 1.632881 | \n",
+ " 1.128258 | \n",
+ " 1.425672 | \n",
+ " -1.723589 | \n",
+ " -5.826797 | \n",
+ " 1.231843 | \n",
+ " 3.264819 | \n",
+ "
\n",
+ " \n",
+ " 5998 | \n",
+ " y3 | \n",
+ " 3.570446 | \n",
+ " 1.398575 | \n",
+ " 0.566380 | \n",
+ " 0.600210 | \n",
+ " 5.554906 | \n",
+ " -1.641754 | \n",
+ " 1.428654 | \n",
+ " 2.209597 | \n",
+ " 1.929370 | \n",
+ " ... | \n",
+ " 1.033945 | \n",
+ " 2.587965 | \n",
+ " 2.678239 | \n",
+ " -0.107618 | \n",
+ " 0.562443 | \n",
+ " 0.331650 | \n",
+ " -0.232188 | \n",
+ " 1.253137 | \n",
+ " -0.764202 | \n",
+ " 1.695968 | \n",
+ "
\n",
+ " \n",
+ " 5999 | \n",
+ " y3 | \n",
+ " 3.100275 | \n",
+ " 2.419199 | \n",
+ " 1.122570 | \n",
+ " 3.293084 | \n",
+ " 3.787231 | \n",
+ " 1.958845 | \n",
+ " 1.300849 | \n",
+ " 2.526066 | \n",
+ " 2.451000 | \n",
+ " ... | \n",
+ " 0.761709 | \n",
+ " 3.733524 | \n",
+ " 0.129880 | \n",
+ " 2.505081 | \n",
+ " 1.340040 | \n",
+ " 1.808093 | \n",
+ " 1.821813 | \n",
+ " 3.089517 | \n",
+ " 3.722296 | \n",
+ " 3.513244 | \n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
5 rows × 1001 columns
\n",
+ "
"
+ ],
+ "text/plain": [
+ " target y_sample0 y_sample1 y_sample2 y_sample3 y_sample4 y_sample5 \n",
+ "5995 y3 2.442450 1.802108 0.294235 2.372050 1.303846 1.788222 \\\n",
+ "5996 y3 3.541716 1.994839 5.250106 1.521251 1.237782 0.746012 \n",
+ "5997 y3 -2.000938 3.227403 0.932215 -0.159020 5.510897 -1.754066 \n",
+ "5998 y3 3.570446 1.398575 0.566380 0.600210 5.554906 -1.641754 \n",
+ "5999 y3 3.100275 2.419199 1.122570 3.293084 3.787231 1.958845 \n",
+ "\n",
+ " y_sample6 y_sample7 y_sample8 ... y_sample990 y_sample991 \n",
+ "5995 3.237095 1.963390 1.678264 ... 1.518361 1.330022 \\\n",
+ "5996 3.503881 1.513360 2.796339 ... 0.480239 2.537787 \n",
+ "5997 -0.925815 1.880578 2.218145 ... 2.134847 0.070474 \n",
+ "5998 1.428654 2.209597 1.929370 ... 1.033945 2.587965 \n",
+ "5999 1.300849 2.526066 2.451000 ... 0.761709 3.733524 \n",
+ "\n",
+ " y_sample992 y_sample993 y_sample994 y_sample995 y_sample996 \n",
+ "5995 -0.248718 3.776295 0.718719 3.304478 3.345208 \\\n",
+ "5996 -5.611905 5.063916 3.080128 -0.095438 3.740144 \n",
+ "5997 -3.396575 1.632881 1.128258 1.425672 -1.723589 \n",
+ "5998 2.678239 -0.107618 0.562443 0.331650 -0.232188 \n",
+ "5999 0.129880 2.505081 1.340040 1.808093 1.821813 \n",
+ "\n",
+ " y_sample997 y_sample998 y_sample999 \n",
+ "5995 1.574800 1.465450 -1.156816 \n",
+ "5996 1.901680 2.199507 2.030381 \n",
+ "5997 -5.826797 1.231843 3.264819 \n",
+ "5998 1.253137 -0.764202 1.695968 \n",
+ "5999 3.089517 3.722296 3.513244 \n",
+ "\n",
+ "[5 rows x 1001 columns]"
+ ]
+ },
+ "execution_count": 9,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "pred_samples.tail()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " | \n",
+ " target | \n",
+ " quant_0.05 | \n",
+ " quant_0.95 | \n",
+ "
\n",
+ " \n",
+ " \n",
+ " \n",
+ " 0 | \n",
+ " y1 | \n",
+ " -1.392198 | \n",
+ " 3.537370 | \n",
+ "
\n",
+ " \n",
+ " 1 | \n",
+ " y1 | \n",
+ " -2.278994 | \n",
+ " 3.741764 | \n",
+ "
\n",
+ " \n",
+ " 2 | \n",
+ " y1 | \n",
+ " -1.676229 | \n",
+ " 3.074070 | \n",
+ "
\n",
+ " \n",
+ " 3 | \n",
+ " y1 | \n",
+ " -1.688622 | \n",
+ " 2.748100 | \n",
+ "
\n",
+ " \n",
+ " 4 | \n",
+ " y1 | \n",
+ " -1.863512 | \n",
+ " 3.838427 | \n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
"
+ ],
+ "text/plain": [
+ " target quant_0.05 quant_0.95\n",
+ "0 y1 -1.392198 3.537370\n",
+ "1 y1 -2.278994 3.741764\n",
+ "2 y1 -1.676229 3.074070\n",
+ "3 y1 -1.688622 2.748100\n",
+ "4 y1 -1.863512 3.838427"
+ ]
+ },
+ "execution_count": 10,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "pred_quantiles.head()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " | \n",
+ " df | \n",
+ " location_1 | \n",
+ " location_2 | \n",
+ " location_3 | \n",
+ " scale_1 | \n",
+ " scale_2 | \n",
+ " scale_3 | \n",
+ " rho_12 | \n",
+ " rho_13 | \n",
+ " rho_23 | \n",
+ "
\n",
+ " \n",
+ " \n",
+ " \n",
+ " 0 | \n",
+ " 7.865457 | \n",
+ " 0.985213 | \n",
+ " 0.087629 | \n",
+ " 1.068884 | \n",
+ " 1.553091 | \n",
+ " 0.777370 | \n",
+ " 1.867574 | \n",
+ " 0.736455 | \n",
+ " 0.488430 | \n",
+ " 0.303660 | \n",
+ "
\n",
+ " \n",
+ " 1 | \n",
+ " 4.911545 | \n",
+ " 0.594036 | \n",
+ " -0.114491 | \n",
+ " 1.938422 | \n",
+ " 2.098058 | \n",
+ " 1.042931 | \n",
+ " 2.288096 | \n",
+ " 0.569916 | \n",
+ " 0.664168 | \n",
+ " 0.688709 | \n",
+ "
\n",
+ " \n",
+ " 2 | \n",
+ " 4.911545 | \n",
+ " 0.814066 | \n",
+ " 0.483306 | \n",
+ " 1.057014 | \n",
+ " 1.504512 | \n",
+ " 1.366037 | \n",
+ " 2.469075 | \n",
+ " 0.435115 | \n",
+ " 0.924239 | \n",
+ " 0.663905 | \n",
+ "
\n",
+ " \n",
+ " 3 | \n",
+ " 4.911545 | \n",
+ " 0.595148 | \n",
+ " 0.207500 | \n",
+ " 2.642362 | \n",
+ " 1.549827 | \n",
+ " 1.389336 | \n",
+ " 2.296717 | \n",
+ " 0.465067 | \n",
+ " 0.923410 | \n",
+ " 0.674316 | \n",
+ "
\n",
+ " \n",
+ " 4 | \n",
+ " 4.911545 | \n",
+ " 0.814066 | \n",
+ " 0.127430 | \n",
+ " 1.057014 | \n",
+ " 1.886949 | \n",
+ " 1.207472 | \n",
+ " 2.271021 | \n",
+ " 0.492254 | \n",
+ " 0.835167 | \n",
+ " 0.707119 | \n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
"
+ ],
+ "text/plain": [
+ " df location_1 location_2 location_3 scale_1 scale_2 scale_3 \n",
+ "0 7.865457 0.985213 0.087629 1.068884 1.553091 0.777370 1.867574 \\\n",
+ "1 4.911545 0.594036 -0.114491 1.938422 2.098058 1.042931 2.288096 \n",
+ "2 4.911545 0.814066 0.483306 1.057014 1.504512 1.366037 2.469075 \n",
+ "3 4.911545 0.595148 0.207500 2.642362 1.549827 1.389336 2.296717 \n",
+ "4 4.911545 0.814066 0.127430 1.057014 1.886949 1.207472 2.271021 \n",
+ "\n",
+ " rho_12 rho_13 rho_23 \n",
+ "0 0.736455 0.488430 0.303660 \n",
+ "1 0.569916 0.664168 0.688709 \n",
+ "2 0.435115 0.924239 0.663905 \n",
+ "3 0.465067 0.923410 0.674316 \n",
+ "4 0.492254 0.835167 0.707119 "
+ ]
+ },
+ "execution_count": 11,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "pred_params.head()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# SHAP Interpretability\n",
+ "\n",
+ "Since XGboostlss estimates the covariance matrix $\\mathbf{\\Sigma}(x) = \\mathbf{L}(x) \\mathbf{L}^{\\prime}(x)$ via the Cholesky factors, **interpretability is only sensible for the location parameters**. The following parameters have been estimated"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "['df',\n",
+ " 'location_1',\n",
+ " 'location_2',\n",
+ " 'location_3',\n",
+ " 'scale_tril_diag_1',\n",
+ " 'scale_tril_offdiag_12',\n",
+ " 'scale_tril_diag_2',\n",
+ " 'scale_tril_offdiag_13',\n",
+ " 'scale_tril_offdiag_23',\n",
+ " 'scale_tril_diag_3']"
+ ]
+ },
+ "execution_count": 12,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "list(xgblss.dist.param_dict.keys())"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ ""
+ ],
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "C:\\Users\\maerzale\\.virtualenvs\\XGBoostLSS-vIPRRz-M\\lib\\site-packages\\numpy\\lib\\function_base.py:2854: RuntimeWarning: invalid value encountered in divide\n",
+ "C:\\Users\\maerzale\\.virtualenvs\\XGBoostLSS-vIPRRz-M\\lib\\site-packages\\numpy\\lib\\function_base.py:2855: RuntimeWarning: invalid value encountered in divide\n"
+ ]
+ },
+ {
+ "data": {
+ "image/png": "iVBORw0KGgoAAAANSUhEUgAABLAAAAPoCAYAAADOWwfbAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAB7CAAAewgFu0HU+AAChDklEQVR4nOzdd5xdZZ0/8M+5U9JDGoQqBBAQRKSoFAtYsMSCYkVcbNhddO3I6i4qKPtzdVdWcZG1oWtdCwYXG21BUBAQaZGOlEB6z5R7fn9MZjKTzGRK7szcmbzfvIace85zzvO9k+Tm3s88z3OKsizLAAAAAECdqox2AQAAAACwNQIsAAAAAOqaAAsAAACAuibAAgAAAKCuCbAAAAAAqGsCLAAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAuibAAgAAAKCuCbAAAAAAqGsCLAAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAuibAAgAAAKCuCbAAAAAAqGsCLAAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAuibAAgAAAKCuCbAAAAAAqGsCLAAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAuibAAgAAAKCuCbAAAAAAqGsCLAAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAuibAAgAAAKCuCbAAAAAAqGsCLAAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAuibAAgAAAKCuCbAAAAAAqGsCLAAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAuibAAgAAAKCuCbAAAAAAqGsCLAAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAuibAAgAAAKCuCbAAAAAAqGsCLAAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAuibAAgAAAKCuCbAAAAAAqGsCLAAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhrjaNdAONLW1tblixZkiSZPXt2Ghv9EQMAAAC2jRFY1NSSJUsyf/78zJ8/vyvIAgAAANgWAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAuibAAgAAAKCuCbAAAAAAqGsCLAAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAuibAAgAAAKCuCbAAAAAAqGsCLAAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOpa42gXAKPhi9e15YwrkzXtWx4rkpTZlO5OaEgaimR9W9K2WduZzclTdk2WrktuWZys2+x6jUVSLZNqt32VJI2Vjn7ay459U5uSE/dPPnhEQw6YXdTgGQIAAMD4IcBiu/KLO9vz0p+WKbfSpvNYZ+i0eSjV3bKW5Ff39n28rZeOqklaqj33LW9JLrg5ueDm9jxzt+T7L23IzlMEWQAAAJCYQsh25Mr72/OSfsKrenDFg8nTv9uepevqvVIAAAAYGQIsthsnXjR2AqG7ViT/en21/4YAAACwHRBgsV14YGV7Hls32lUMzvl/LtPaPnZCNwAAABguAiy2Cz+8Y7QrGLxH1yb3rxrtKgAAAGD0CbAAAAAAqGsCLLYLr9p/tCsYvJ0mJ4+bNtpVAAAAwOgTYLFd2GN6Q3acNNpVDM6pTyrS1FCMdhkAAAAw6gRYbDd+/JKxEwbtOyP5h8P99QQAAIBEgMV25BmPa8hFJxSp9xjrmbslV76uIbMm1XulAAAAMDIEWGxXXrxvQ6ofbMwXj02mNPTepjM2qmz8mtSQTG1MGntpO7M5ef5eyVN27mi3ucZiy79klSTNlWRipeN4Y5HMnJC89eDktjc15PLXNWbnKcIrAAAA6NTbZ3IY9047ojGnHTHaVQAAAAADYQQWAAAAAHVNgAUAAABAXRNgAQAAAFDXBFgAAAAA1DUBFgAAAAB1TYAFAAAAQF0TYAEAAABQ1wRYAAAAANQ1ARYAAAAAdU2ABQAAAEBdE2ABAAAAUNcEWAAAAADUNQEWAAAAAHVNgAUAAABAXRNgAQAAAFDXBFgAAAAA1DUBFgAAAAB1TYAFAAAAQF0TYAEAAABQ1wRYAAAAANS1xtEuAEiqZZkbF7WnrT1pqCTrWpMdJiRr25N505MVLUlbNdnQlkxsTHaZWskOE+XPAAAAbB8EWDCKfnB7W/7u4mRDdRAnlUlSTaWo5u8OSr56fCXNDcIsAAAAxi+femGUvP93bXnNLwYZXiVJ0fFVLZNv/CXZ/SvVrGkZ7EUAAABg7BBgwSj4v79V88U/beNFio5fHluXPOt7AiwAAADGLwEWjIK3XlLbwOn6RcnDq4RYAAAAjE8CLBhhbdUydyyr0cWKTZtnX1vW6KIAAABQXwRYMMJWtwzPdR9eMzzXBQAAgNEmwIIRNrV5eK6769ThuS4AAACMNgEWjLDGSpGDZtfoYt1mDZ5+ZNF3OwAAABjDBFgwCr72/Nr+1XvaLsncKf46AwAAMD75xAuj4MhdK/nYU7fxIhtHX+08Jbn0Nf4qAwAAMH751Auj5KxnNuYXJySTGwfQuNzyq6FI3nVI8sDbK5nU5K8yAAAA49dAPjoDw2T+vo1Z876O7aVr25IkZZm0l8nEpiIb2srMnFjJ+rYy1bJMWzVpakgmN1XSUBFaAQAAsH0QYEGdmNXbUKwJHb9MbRjZWgAAAKCeGMIBAAAAQF0TYAEAAABQ1wRYAAAAANQ1ARYAAAAAdU2ABQAAAEBdE2ABAAAAUNcaR7uA8e7yyy/PxRdfnFtvvTVLlizJlClTsscee+TYY4/NK17xikydOnVY+r399ttzySWX5Nprr82jjz6aNWvWZMaMGZkzZ06e+MQn5vDDD89xxx2XhoaGYekfAAAAoFaKsizL0S5iPFq7dm3OOOOMXHHFFX22mTt3bj772c/m4IMPrlm/q1evzuc///n84he/SH+/tZdeemmmTZtWs76TZNGiRZk/f36SZMGCBZk7d25Nrw8AAABsf4zAGgbt7e356Ec/mquvvjpJMnv27JxwwgmZN29eVq5cmUsuuSQ33XRTFi1alNNOOy0XXHBB5s2bt839rlixIu9973tz6623Jkl22mmnHHfccXn84x+fqVOnZu3atXnggQdy7bXX5rbbbtvm/hhdty0p86u727LgruTBNcmUpuT585K3H9KQ3aebHQwAAMD4YQTWMPjxj3+cs88+O0my99575ytf+Upmz57do80Xv/jFXHjhhUmSQw89NOeff/429/ve9743v//975MkJ598ct75zndmwoQJvbZ97LHHMnPmzDQ21jbDNAJr+P3izmpO+2177l7Rd5u9d0h+8vKGPGknQRYAAABjn0+3Ndbe3t4jjDrzzDO3CK+SjrBpv/32S5LccMMNueaaa7ap34suuqgrvHrlK1+Z973vfX2GV0my44471jy8Yvidd0N7XvI/Ww+vkuTuFcmh32jPHx5qH5nCAAAAYBgJsGrshhtuyOLFi5Mkhx12WA444IBe2zU0NOS1r31t1+NLLrlkm/r95je/mSSZPHly3vOe92zTtahPdy4r8+5fVwfcvppk/o+qqRpkCQAAwBgnwKqxq666qmv7mGOO2Wrbo48+umu7c72sobjxxhtz7733Jkme9axnDdudDRldX7mhmoHHVx0Wr09+ebcACwAAgLHNHLIau+uuu7q2DzrooK22nTNnTubOnZtFixZlyZIlWbZsWWbOnDnoPv/0pz9t0efvfve7/PSnP80dd9yRlStXZocddsj++++f5zznOXnRi15k+uAYdPHdg42vOnz/tmrm7yOrBgAAYOySYtTYfffd17W966679tt+t912y6JFi7rOHUqA1f2OgrNnz86HPvShXHrppT3aLF68OIsXL85VV12V7373u/n85z+f3XbbbdB9MXrWtg3tvNWtRmABAAAwtgmwamzVqlVd2zNmzOi3/Q477NDruYPRueZWkpx33nm577770tTUlPnz5+fJT35yGhsbs3DhwvzsZz/LihUrcuedd+Yd73hHLrzwwh79D0Rn2DaQWqitfWYUuX/l4MOoA+cUw1ANAAAAjBwBVo2tXbu2a7u5ubnf9t3vFLhmzZoh9bly5cqu7fvuuy/Tp0/Pl7/85R4LyL/gBS/ISSedlHe96125++678/DDD+c//uM/cvrppw+qr/nz5w+pRrbdW59UyaX3D+6ugkWSdzy5YXgKAgAAgBFiYZxxoNzsLnOnnXZar3c/nDNnTj796U93Pf7FL36R1atXD3t91MYr9y/yxDmDO+dVBxTZfZoRWAAAAIxtRmDV2OTJk7tGRLW0tPS7WPqGDRu6tqdMmTLkPjtNmjQpL3rRi/psu99+++Xggw/OzTffnJaWltx000393i2xuwULFmz1+OLFi3PKKacM+HoMXHNDkV+9ujHHf78tf1nSf/vn7pl858VGXwEAADD2CbBqbNq0aV0B1vLly3uES71ZsWJFj3OHYvr06V3b++67b5qamrba/glPeEJuvvnmJMnf/va3QfU1d+7cwRdIzewytchNb27Mz/5azSf+r5pbFyeb35tw/xnJf76wIc/cwwBLAAAAxgcBVo3tueeeefDBB5MkDz30UL93Iuxs23nuUPv8wx/+kCSZOnVqv+27txnquluMnkpR5OX7NeTl+xldBQAAwPbBEI0a22effbq2b7nllq22XbJkSddd/WbNmpWZM2cOqc/99tuva3sga1p1bzOQwAsAAABgNAmwauzoo4/u2r766qu32vaqq67q2h7MOlS99VkUHQt133nnnWltbd1q+9tuu61re6ijvgAAAABGigCrxg477LDMnj07SXL99dfn9ttv77Vde3t7vve973U9Pv7444fc59y5c3PYYYclSdatW5eLL764z7YLFy7sWv9qypQpOeSQQ4bcLwAAAMBIEGDVWENDQ0499dSux5/4xCeydOnSLdqde+65WbhwYZLkkEMOyVFHHdXr9S666KIcccQROeKII/K2t72tz37f/e53d23/27/9W6/B2ZIlS3LGGWd0PX7Na16TiRMn9v+kAAAAAEaRRdyHwQknnJBLL7001157be6+++687nWvy8tf/vLMmzcvK1euzCWXXJIbb7wxScedB08//fRt7vNJT3pSTjnllHzzm9/MypUr86Y3vSkvfvGLc8ghh6SxsTELFy7Mz372s667Hh544IF5y1vess39AgAAAAw3AdYwaGxszDnnnJMzzjgjV155ZZYsWZKvfe1rW7SbO3duzjrrrB4Lv2+L9773valUKvnWt76V1tbW/OQnP8lPfvKTLdodddRR+cxnPpMJEybUpF8AAACA4STAGiZTpkzJF77whVx22WVZsGBBbr311ixbtiyTJ0/O7rvvnuOOOy4nnnhize8C+O53vzvPe97z8rOf/SzXXnttHnvssbS1tWXmzJl50pOelPnz52/TgvEAAAAAI60oy7Ic7SIYPxYtWpT58+cnSRYsWJC5c+eOckUAAADAWGcRdwAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAuibAAgAAAKCuNY52AUBt/Pqeai68pT0rNiTrWju+2qvJATsmHz2yksfPahjtEgEAAGBIBFgwhrVVq3ndz6v50cJy086yZ5urH0n+6+ZqGotqvvPSSl59gCALAACAsUWABWNUa3s1877angdXb3ag2PjrZkFWW5m85mfV/G1FmX94mr/6AAAAjB3WwIIx6pU/rW4ZXnVX9L77A5eVWbquOiw1AQAAwHAQYMEY1F6t5uJ7yv4b9hFivfc37bUtCAAAAIaRAAvGoB/ekbRtwyCqn/+1drUAAADAcBNgwRj02NoBjL7ailYzCAEAABhDBFgwBh2yYx9zAwdoh+YaFQIAAAAjQIAFY9AzH1fJDhOGfv6HnrZtARgAAACMJAEWjFEfedoA/vr2MtOwsUj+4akNtS8IAAAAhokAC8aojx3ZkDc9cSsjqXoJrypJ/vLWhlQKI7AAAAAYOwRYMIb914sac9lrG7L3DukIrLp/beYFeyWr/6Eh+8/y1x4AAICxpXG0CwC2zbMeV8ldbxdKAQAAMH751AsAAABAXRNgAQAAAFDXBFgAAAAA1DUBFgAAAAB1TYAFAAAAQF0TYAEAAABQ1wRYAAAAANQ1ARYAAAAAdU2ABQAAAEBdE2ABAAAAUNcEWAAAAADUNQEWAAAAAHVNgAUAAABAXRNgAQAAAFDXBFgAAAAA1DUBFgAAAAB1TYAFAAAAQF0TYAEAAABQ1wRYAAAAANQ1ARYAAAAAdU2ABQAAAEBdE2ABAAAAUNcEWAAAAADUNQEWAAAAAHVNgAUAAABAXRNgAQAAAFDXBFgAAAAA1DUBFgAAAAB1TYAFAAAAQF0TYAEAAABQ1wRYAAAAANQ1ARYAAAAAdU2ABQAAAEBdE2ABAAAAUNcEWAAAAADUNQEWAAAAAHVNgAUAAABAXRNgAQAAAFDXBFgAAAAA1DUBFgAAAAB1TYAFAAAAQF0TYAEAAABQ1xpHuwCg9h5bU82bLm7PgruTlL23KZI8Y4/ki89pzKE7FyNZHgAAAAyKEVgwzvz+wWrmnrv18CobD13xQHLYN9py9u/bR6o8AAAAGDQBFowjKzeUedZ327eWW/Xq9Mur+c4t1WGpCQAAALaVAAvGkbN+X03rEHOoz15jFBYAAAD1SYAF48h//Xnoo6j+8lhy/SODHbsFAAAAw0+ABePIypZtO//BVQIsAAAA6o8AC8aRCQ3bdv4OE2pTBwAAANSSAAvGkfn7FEM+d5epyTG7D/18AAAAGC4CLBhHPnfs0P9Kv+vQShorAiwAAADqjwALxpE9plfyped2+2s9wDzqZY9PPnaUlwMAAADqk0+sMM685/CGXPzKhswcwHpWU5uSLx9fyY9f3pgGo68AAACoU42jXQBQey/cp5Kl76tk0ZpqLv5re357fzJrYvKEWcmK1mSPaUWe+biG7DFdaAUAAED9E2DBODZ3SiVvenIlb3ryaFcCAAAAQ2cKIQAAAAB1TYAFAAAAQF0TYAEAAABQ1wRYAAAAANQ1ARYAAAAAdU2ABQAAAEBdE2ABAAAAUNcEWAAAAADUNQEWAAAAAHVNgAUAAABAXRNgAQAAAFDXBFgAAAAA1LXG0S5gvLv88stz8cUX59Zbb82SJUsyZcqU7LHHHjn22GPzile8IlOnTh32Gv7pn/4pv/jFL7oen3rqqXn7298+7P0CALW38v5VueYj12fJX5andU1bUi1TFEmlMak0NaS6tj2pJinLrnOKzl+LJGXHr5UJSdaVKds72hbZdKxoStKWpH3jNcoklaRxUkPKDdWUrWXXNdNQprGpIdX1G/vdqGlSQ6rlxtrKIo07NKVtWUuqq1o72hVJw5SGVIpK2le1Je3Vjn46660kjTOb09hcSevilqRadvzodUM1RbXc9Hwak6JMyvYyRffz09G+MrEhKcuUZZlifZmuTpJkQpFKNSnbOvou0nG9KQfNzBN+9NxM2neHbfq9AgBqR4A1TNauXZszzjgjV1xxRY/9LS0tWbZsWf785z/n+9//fj772c/m4IMPHrY6rrrqqh7hFQAwNrWubcuCF/8mK+9evdmRImnvCKKqG9q77d4YMZVlR2SzMbspkqStTNnasV10b5syZZkULZtdo+i4Ttva9k3XKMtUyiTtSbWlW78bj7WuaetoWy1TVstUl7X0DKiqSXVFe8q0d4Rt5aagLem4bttjG9Je3Wz/xotU0nFOWrtds/M5bNyqVJNybXu3/R3PuKvthrIzS9u0ry1Zc9OyXPf4H2bC46bk8L++Og3NDQEARpcphMOgvb09H/3oR7vCq9mzZ+ctb3lLPv3pT+fDH/5wDjnkkCTJokWLctppp+Wee+4ZljpWr16ds846K0kyadKkYekDABh+7S3V/ORZv+wlvEqKskwlm4c83RsUG7+SskjSvpX2ne0GcJ2i3PjVX9tK0RWgdQVF3cO0arnV65SVzespuiKqHsFTt+PdxmBttj+b7S82Xq/3717L/Wvyx12+k7La63cEABhBRmANg5/+9Ke5+uqrkyR77713vvKVr2T27Nldx1/96lfni1/8Yi688MKsXLkyZ511Vs4///ya1/Fv//ZvWbRoUebOnZvnPve5+c53vlPzPhhbqmWZ0y9ty+f/mLR1vhcvk4YiaayUaWnf7M3+xgdNlWTulORtT67k409vSKXo82MSAMPgtm/8Nesf27DlgXLTtLl+FUXSXu3/p5dFR0DU53WLIimrHaOvBtJnypQNxabpiEmPUVwDuUZZ9KynSJGi95it+4npI4rrtW015RbfmzJJ29KW/O1f/pw9PnLIAK8FAAwHI7BqrL29vUcYdeaZZ/YIrzq9973vzX777ZckueGGG3LNNdfUtI4//vGP+elPf5ok+ehHP5rJkyfX9PqMPbc81p7mc9ryuT/0DK+SMu1lsqG96Jpa0WXjZms1+duq5BNXVjPlX1rzhwerAWDk/OU/bu91f+8jkPo24PZFP/HQIH+QUW4cSdXxoFsNfY282qK/nlFUZYDB1NZabXms70r+9tmbBtQfADB8BFg1dsMNN2Tx4sVJksMOOywHHHBAr+0aGhry2te+tuvxJZdcUrMa1q9fn09/+tMpyzLPe97z8oxnPKNm12ZsWrm+mkP+q9rjh9+b3rn39oa920eczQ6vb0ue8e22PLxKiAUwEtpbqmlZ0dr7wUHObBvwaK1+L9T/+KfubZOOqYTb0t8mfU34610tnnL78pb+GwEAw0qAVWNXXXVV1/Yxxxyz1bZHH31013bnlMNaOPfcc/Pggw9mhx12yAc/+MGaXZex67Tf9BVeDdBmnxRaqslpv27vvS0AtTWQaXYAAOOcAKvG7rrrrq7tgw46aKtt58yZk7lz5yZJlixZkmXLlm1z/zfddFN+8IMfJElOO+20Xqcvsv3579uG+uGn759x//yvPlABjISGCQ1pmtbU+8FBDmoqa7WEYTmIUVAbA7hiWxZC7xHiDWL0Vwb9LepVQ1/ffwBgxAiwauy+++7r2t511137bb/bbrv1eu5QbNiwIWeeeWaq1Wqe+tSn5qUvfek2XY/xY0MtBktt9glgQ3vS0i7EAhgJB77t8b3u33gzvwEbcPt+Rn0NNowqyjJFt5nn3WexD3yZ9U2qA4ylttZqy2N9V7Lr+7f+Q0kAYPi5C2GNrVq1qmt7xowZ/bbfYYcdej13KM4777zcd999mTBhQk4//fRtulZfFi1atNXjnet/sX1oFIEDjIiDTt0/t11wZ1o2X4upKFIOdDRUWSaVImW1n/Yb72zYZ5uyTFkUqRZl/3ciLMuOhdq7B17FpkNFUfQ/RXKz8CtJypQp09/oqsFFe739k1YkqUxryuM+cdggrgUADAcBVo2tXbu2a7u5ubnf9hMmTOjaXrNmzZD7veWWW/Ld7343SfL2t789u++++5CvtTXz588flusyvHadmjy0ehsvstnngN2mJpVB3oUKgKFZvnBlmmc0ZcPGAKvHkuaVItXqFveR7dZg4wv4xlCqbChSbe+jfWfg1N91yo7piGW36/bafmN4VZRJKklZ3di2R4i16dK91VNUN99fpuh4Jl3/NG1+PL3uTy/7N05v7CPsapwzIUfc/ZoUDX5iAwCjTYA1DrS2tubMM89Me3t7DjjggLz+9a8f7ZKoM597ViVvWNDtx9cd7/sHoO9G//A0b+YBhtuaR9Zlwct+m3WPru/YsfH1u+vVeePoq7JIioaNx1s7jhebBVcpOu9CWKZs6GhftvS8TteFK0mq3Tra+GuluUhaNwZTRdGxplZZptJQpGwrNyVaSSpNlY3bRYqiSMOESqqr21NuaO/cnUpTkSKVlBuqSbXcIhCrTG1IY2Ml7avaUnYGYu2d4VVHq6JSdgRgXecVG38tO+prKDpuYlgtk/bNwqrKxsfVrpKSIpm419Ts/9/PzvSn7TSo3y8AYPgIsGps8uTJWblyZZKkpaUljY1b/xZv2LCha3vKlClD6vOCCy7IXXfdlYaGhpxxxhlpaGgY0nUGYsGCBVs9vnjx4pxyyinD1j9Dc/LBDfnCddX8aeszQDdT9rqZJIfOTf7haV4+AIbT+mUb8rPjf5XWVa09D3QfVlR0xDFT9picV17xopEsDwBgRPkEWmPTpk3rCrCWL1+eyZMnb7X9ihUrepw7WAsXLsw3vvGNJMlJJ52UAw44YNDXGIzOuyYy9lz/pqac8KPW/OzOjTt6jMLafCWRsscvnRqK5MQDinzvBC8dAMPtL+fdsWV41Yc1D6zNwh/ck/1ePW+YqwIAGB0+hdbYnnvumQcffDBJ8tBDD/V7J8LOtp3nDtZFF12Utra2VCqVNDY25mtf+1qv7W644YYe253t9tprrzz3uc8ddL+MTT99ZVOqZZlPXN6Wr92UrGtPdpmSvOVJRZasq+T2pdXcv6LM5KaksUgmNCZrW5P9ZiVP2qmSdx9eSaN1QACGXbW1mtsvvGtQ59z8H7cLsACAcUuAVWP77LNPrr766iQdC6sfccQRfbZdsmRJ1139Zs2alZkzZw6532q1mq9//esDanvdddfluuuuS5I861nPEmBtZypFkU8f25RPH9vb0eGbfgrAwK1+aG3a17YP6py1j6wbpmoAAEafoRQ1dvTRR3dtdwZZfbnqqqu6to855phhqwkAGFvc5BUAoCcjsGrssMMOy+zZs7NkyZJcf/31uf3223tdl6q9vT3f+973uh4ff/zxQ+rvAx/4QD7wgQ/02+6rX/1qzj///CTJqaeemre//e1D6g8AGH5Tdp2cxqmNaVvdNuBzpu6+9XU3AQDGMiOwaqyhoSGnnnpq1+NPfOITWbp06Rbtzj333CxcuDBJcsghh+Soo47q9XoXXXRRjjjiiBxxxBF529veNjxFAwB1pdJYyYFv3HdQ5zz5fQcOUzUAAKPPCKxhcMIJJ+TSSy/Ntddem7vvvjuve93r8vKXvzzz5s3LypUrc8kll+TGG29M0nHnwdNPP310CwYA6s5Bp+6fO757TzYs3dBv2x0ePy3zXvK4EagKAGB0CLCGQWNjY84555ycccYZufLKK7NkyZJe7w44d+7cnHXWWdlnn31GoUoAoJ41T2/KCb86Phe/8ndZde+aPtvtePisvOAHx45cYQAAo0CANUymTJmSL3zhC7nsssuyYMGC3HrrrVm2bFkmT56c3XffPccdd1xOPPHETJ06dbRLBQDq1MTZE/KKS1+YlfesypX/8IcsvWV5qm1lKs2V7PLMnXL0pw7L5LnWvgIAxr+iLMtytItg/Fi0aFHmz5+fJFmwYEHmzp07yhUBAAAAY51F3AEAAACoawIsAAAAAOqaAAsAAACAumYRd2CrvvXntpx9dTX3r0zWtm7c2cvSecXGrzJJUyWZNTE59bBK/vlZjSmKYgQrBgAAYLwxAgvo1R8fqmbqv7TklF9Uc/vSzvBqY3BVFJu+NiqTVDf+2lJNHlmTfOrKaho/05L/ubVt5J8AAAAA44YAC9jCnUurOfqbbVnTNeKq83+9jKTaLMjatL/jl2qZnPjj9lx+b/vwFAsAAMC4ZwohsIU3L2hP2xazBPuZBlgUW04t7JxTmOR1P2nLQ+9vqFGFAFBbZbXMkp/dm4fPvTVrblmW9mXrkrZs/Les7BhmnKRoSppmT0xZraZc255iQiXTn75L5v3bMZmw57TRfAoAMK4JsIAe1reVuepv3YKoLZe7GpyNIdbDq5O7llWzz0wDPwGoLw/+682596N/TNnaORl+8x/bbPqJTNmatD2yftOh1W1Z9rP7svxn96Vpl0k5+I8npnm3KSNUOQBsPwRYQA+3Plamuq2hVR8uu1eABdCbmy9YmD+e85eUbVv+AKF7kNLQXOTwjzwxT3zTfiNa33h29z9ck4e+8JeNj3oLrzoVXW16Tqovuva2PrwuN8z7bg6976Q07yLEAoBa8kkS6KFhGF8VGiruRgjQXdv69nz78Ivyh7NuTtlWdt3RtUi3JQY7v5K0t5T546duzncP/3nKXu4Iy+Cs+sOjAwyvuuurRZEySdlazR0vvaQm9QEAmwiwgB4O2rFI4zC9MrxwHwEWQHc/Ov7XaVne0j2j6tvGBmWSDcta8z/PFZJsq4f+7ZYejwf6r1R/0eGa6x5L65L1/bQCAAZDgAX00Fgp8vx53d7CD/jdfB9v5zfufvzMZO5ULzkAnZbctjxrHlwzuJO6vSavvGdNNqzYUNuitjPLLvnbEM8s+g2xll98/xCvDQD0xqdJYAvfeElDJtdihbyN7+4rSX722qYaXBBg/Lj2M39OMvCfE3QpNo0AuuIDf6xlSdudcstb7tZMdV3bsF0bALZHAixgC3MmV3LHOxqzx/SNO7o+XfXyRr8sex99tXHXlKbk+rc25glzvNwAdLfi3tXbfo27tv0a27NJ++0wxDP7D76mPmXHIV4bAOiNT5RAr3afXsn972nOjW9uzIv2KbLXjGRSY9GRZXWGVmWZhiKZ1NDxNaUxmdyY7DQpecouRf73pMas/uiEPHmXhlF+NgD1p3Hitr821uIa27Nd//7AHo8HMx5rayPnGudMzJRDBVgAUEu1mCQEjGOH7FzJgtfIugFq7YCT98kfPnVTygxyGmG3lOXJ7zuw73b0a86r984DZ92YdbetSMfvwkAirLKP369N5+715afXpD4AYBOfSgEARsGBJ+2dVIZ2d9YiSdFQZK/n71bborYzleaGPOmKl2TSE2Zs3NOxOHvfMVbZxx0jy41nJ4/7/NMy51X71LpUANjuCbAAAEZBQ3Mlx/37U5IMYupat4bP+8YxNa9pe9Q0Z2IOv/WVOeg3L+wIsjaGih1BVtm11f2bX278r3N/0VBk5gl75ojlb8yu//DkkX4KALBdMIUQAGCU7P3CPdL0X4359duvTtnasa/XMVndgquGpiLPv/AZ2dki4TU18zm75fBbXznaZQAAfRBgAQCMoj2etUvefPuJWbpwef7w2Zuz+OblKRqKNDQXqTQ1pKG5SEOlkim7TM5TTj84M/ae3v9FAQDGGQEWAEAdmLXfjLzgv54x2mUAANQla2ABAAAAUNcEWAAAAADUNQEWAAAAAHVNgAUAAABAXRNgAQAAAFDXBFgAAAAA1DUBFgAAAAB1TYAFAAAAQF0TYAEAAABQ1wRYAAAAANQ1ARYAAAAAdU2ABQAAAEBdE2ABAAAAUNcEWAAAAADUNQEWAAAAAHVNgAUAAABAXRNgAQAAAFDXGke7AGB8aauWueze9nzluvasa02evEsyc2JD9p2ZvOjxlUxolJsDAAAwOAIsoCbWtZZ518Wt+cafyx77f3l3krQnSYq057i9kh+e2JRZkwVZAAAADIxPkMA2W9ta5tDzW7YIrzZXJvndvcm8L7Vm8ZrqiNQGAADA2CfAArbZB3/dmjuWDLz9ypbkuRe2DV9BAAAAjCsCLGCbrNpQ5oIbtz7yqjc3PVrmoZVGYQEAANA/ARawTX5zTzUt7UM798vXC7AAAADonwAL2CarW4Z+7tJ1gx+5BQAAwPZHgAVskz2mF0M+96A5Qz8XAACA7YcAC9gmz9qzyC5TB39eUyV52+ENtS8IAACAcadxtAsAxraiKPKZ4xrz5osGd1fBU55USVODEVgAjF3r71mZRy+4PUu+szCtD65J2rpNjS/LbP6vXLHxf0XZ8WtlWlN2fPdB2f2so0auaAAYowRYwDZ705Mbcveyaj79fwNblP34vYuc/5KmYa4KAIZH28qW3P3my7L0x3enSDYLqjqDq83jq45wqyg3PayubM2is2/IorNvyOO+dmx2fMuBw1s4AIxhphACNfGp45py06mN2W9W78crRbL/7ORnr27MJa9vHtniAKBGquvbcvsLLh50eFXpZW+6tb3/rZdlyfcX1rhaABg/jMACauZJOzfkjndb1wqA8euxbyzM6t8v6iW86rRleNX/hPkiSZkH3np5Zr9mv22uEQDGIyOwAABggBZ95ZZ0TgfcXF9B1cBWfCzSvro1q655eGiFAcA4J8ACAIABqLa2Z+2flyYZ6OirgYZXmyz77l8HWxYAbBcEWAAAUCeKwh16AaA3AiwAABiASlNDJh86J0lfkwi33Nt7u77N+rv9B1sWAGwXBFgAADBAc995YPqaGNhXWDWwEKtMw7SmTDl8p6EVBgDjnAALAAAGaMdT9su0Z+6SMgMdhVVspW3Pc/a68LnbXiAAjFMCLAAAGKBKc0MOuPiFmXPyfr0EU8XGx4MJscqkKDLvO8/JjJfOq33BADBONI52AQAAMJY0TGnKvt9+dh732afl0W/ekSXfXpgN96xMWqtJsTHEai97n2jYkKQ9SSVpmNWcuR98cnb5yOEjWT4AjEkCLAAAGILm3aZk99MPy+6nHzbapQDAuGcKIQAAjJK2lrau7WpbNWW1TLW9OooVAUB9MgILAABGSFmWeeif/pBHP3djyg3t2XJlrE2Pm+ZOyk4fOiw7vf/JKSq93/kQALYXAiwAABgBZXs1tx3y/ay/ZVnnnl5aFV3HWhety0MfvCrLv//X7Hf1K1M0mjwBwPbLv4IAADAC7nvrpf2EV90VXa3W/vHR3Pfm3w5naQBQ9wRYAAAwzKprW7P0m3cM8qxN0waX/ffCtK9uqW1RADCGmEIIADCKHr1xaW78j9uzZtG6NE9rygEnzcs+8/cY7bKoseU/v6fboKv+Rl/1oq3M8h/emdlvOrCWZQHAmCHAAgAYBWsfW59fvObyrLp/TY/9j1y7OFedcUOe/7WjM/fwOaNUHbXW+sDqbb5GSw2uAQBjlSmEAAAjbMPKlvzPC3+zRXjVqXVVWxacdGWW3LJ8ZAtj2DTuPGXbr7Hrtl8DAMYqARYAwAj707/flg3Lt76eUdle5oqPXj9CFTHcZpwwr9uSVsXWmvauocjMV+5by5IAYEwRYAEAjKBqe5mFP7xvQG2X3rYiK/sYpcXY0jCtOTNO3HuQZ21aK2uHE+alccaE2hYFAGOIAAsAYAStXbQubWvaBtx+ya3Lh68YRtRe33lemh43deOj/kZhlV2tJjxhZvb6zvOHszQAqHsCLACAEVQMYfYY40OluSFPvOvkzD71gKSxSO8hVpnO8Kphh+bs9NHD8oSbXpvKhIaRLBUA6o67EAIAjKDJcyelaVpjWlcNbBTWnINnDnNFjKSisZI9//PZ2fM/n53Wpeuy/vblmbDPtDRMbkrL39Zk1VV/y0MfvTblkg2prmjJo5/9Ux797PUbo65y469FGmZPyJy/PyQ7/v2T02BqIQDbASOwAABGUFEpcsBr5w2o7ZyDZ2TabpOHuSJGS9OsSZl29C5pnjs1ZXvy1+N+kgdPvSLlkg3pHJ9VbAytOr42jdpqX7Ihiz55bf6y439m6QW3jN6TAIARIsACABhhT373AZm049ZHzVSaijzznCNGqCJGU1ktc9sTvpP2RevSc2LhpnWwelckbckDb/1tln9/4XCXCQCjSoAFADDCmqc25cRfPi8z95/e6/GJs5vz0h8fl5mP7/0448vSb9+etkfW9npsoEumPfyxq1KWZf8NAWCMsgYWAMAomDCjOa9Y8Nwsu2tlbvzS7VnzyLo0T2/KgW/YO7s/Y+fRLo8R9Mhnrksy8LBqS0Va7lmV1b++P9OO37NWZQFAXRFgAQCMopn7TM9xX3zqaJfBKGp9cE1NrrPhtmUCLADGLVMIAQBgFBWVoY+96qHJW3sAxi//ygEAwCiafOTcmlxn2nP3qMl1AKAeCbAAAGAU7fYvxyTpvOfgUJSZ9pw9MmG/mbUqCQDqjgALAABG0eQn75iZr98vyeYhVjGAUKtMw8zm7HbeccNSGwDUCwEWAACMsr0uPD47vv+QJFuGWNUk5RZRVpmkzORn7prH//G1mbDvjJEoEwBGjbsQAgBAHdj9X5+R3T7/9Dz2pT/nsf/8S9oeWpMiScOsiSmLasqlG1K2J027TMqsU56Qma/bP817Th/tsgFgRAiwAACgThRFkZ3+/pDs9PeHjHYpAFBXBFjAsPrMFRvyicuTapn0tTztxEryj8+q5PRnNI9obQAAAIwN1sAChsW9y6tp/PSGnHFZZ3iVJMVmXx3WV5OPX1pN06fX564l7SNeKwAAAPVNgAXU3LrWMvue25r2an8tix6P2qrJAV9uzYr1Q7+ROAAAAOOPAAuouY9f2pb2AWdQm4VYZfLh37TUvCYAAADGLgEWUHP/cV21r+WuBuTrNxqBBQAAwCYCLKDmWrZxGavWatJWFWIBAADQQYAF1KWGov82AAAAbB8aR7uA8e7yyy/PxRdfnFtvvTVLlizJlClTsscee+TYY4/NK17xikydOrUm/axZsybXXHNNrrvuutxxxx25//77s2rVqkycODFz5szJQQcdlBe84AU56qijUhSSAYbXzlOSR1YP/fxZE+PPKQAAAF2KsizN0xkGa9euzRlnnJErrriizzZz587NZz/72Rx88MHb1NeFF16Yr3zlK9mwYUO/bQ899NB86lOfys4777xNffZl0aJFmT9/fpJkwYIFmTt37rD0Q3278Ob2vOEnbYM4o+fL0PkvbshbD2uqbVEAAACMWUZgDYP29vZ89KMfzdVXX50kmT17dk444YTMmzcvK1euzCWXXJKbbropixYtymmnnZYLLrgg8+bNG3J/999/f1d4tdNOO+WpT31qDjjggMyaNSstLS25+eab88tf/jJr167NDTfckLe//e35+te/nlmzZtXk+cLmTj64If98WVvuXDaQ1j3Dq3kzkrcc6qUJADa3fuGy3P3c/0nbA2u67e3572jRbV+x2f6mA2dknz+8Lg1TJgxvoQAwDIzAGgY//vGPc/bZZydJ9t5773zlK1/J7Nmze7T54he/mAsvvDBJx6io888/f8j9nX322XnwwQdz8skn56lPfWoqlS2XNnv44Yfznve8J/fdd1+S5CUveUk++clPDrnPvhiBRXdHXrAh1z7Y19EtX3qO3C256s0TUjF9EAB6WPKtW/PgKb8ZUNtNIVaZIj2DrCTZ9843ZsI+M2pZHgAMOwFWjbW3t2f+/PlZvHhxko7pfQcccECv7d7whjdk4cKFSZJzzz03Rx555JD6XLFiRXbYYYd+2y1cuDAnnXRSkmTixIn5zW9+k4kTJw6pz74IsNhcS1s1J/2kNRf/tePugk2VpJIyG9qSFMnsicnz96vkP+c3ZUKj4AoY/1bctzq/P/Om/O3/Hk2qPY9N2mlijv7EkzLv+buNTnHUpQ13Ls8dj//WgNtv+te0TNHLaKxUkoPa31eb4gBghJinU2M33HBDV3h12GGH9RpeJUlDQ0Ne+9rX5swzz0ySXHLJJUMOsAYSXiXJfvvtlz333DP33Xdf1q9fnwceeCCPf/zjh9QnDFRzYyU/epWpCgBJcuu378rVZ/65z+PrHl2f377nD9n5KbPzwm88PQ3NbhhN8re3/XZQ7ct0BlbFxu3Nfl5dTZb9118y881PrE2BADACvCuqsauuuqpr+5hjjtlq26OPPrpru3O9rOE2ZcqUru2BLPoOANTG/Zc+stXwqrtH/rgkV37sT8NcEWPFmsv7nI8/YJtPuXjk41f12g4A6pUAq8buuuuuru2DDjpoq23nzJnTNcVuyZIlWbZsQCteD1lra2vuv//+rsfDdSdCAGBL13/x1kG1v/MXD2Tto+uHqRrGlGr/TbZuyyn61eUt23pRABhRphDWWOci6Umy66679tt+t912y6JFi7rOnTlz5rDV9r//+79ZvXp1kuSAAw7InDlzBn2Nzlr70jl9EgDYZMU9q7Lk1hWDO6ma3Pmz+/OkU/cbnqLYjvSy5O2khpEvAwC2gQCrxlatWtW1PWPGjH7bd1+/qvu5tbZs2bJ86Utf6nr8lre8ZUjX6VygHQAYuNUPrxvaeQ+trXEljEXN+89Iyx3Lt+kam4/BmmX9KwDGGFMIa2zt2k1vNJubm/ttP2HCpsWt16xZMyw1tba25sMf/nCWLl2aJDn22GNz3HHHDUtfAMCWmiYP7WeGTVObalwJY9Fu5z17UO173oWwd7v8v2duQ0UAMPKMwBrnqtVqzjzzzNxwww1Jkt133z2f+MQnhny9BQsWbPX44sWLc8oppwz5+gAwHs05eGYmzmrO+qWDW3dor+f1vxwB49+0Y3fPjL/bP8u/dccgziq7ft08xNrtey+sUWUAMHIEWDU2efLkrFy5MknS0tKSxsatf4u73wmw+x0Ca6Esy5x99tn55S9/maRj0fYvf/nLmT59+pCv2bnoPAAwcJWGIk9807657vMDX8h9xuOnZccnDd/amIwtj/vm8zPxSTvmkQ//Xy+LupfpHHdVbBZc9QivGovs+etXZOqxewx3uQBQcwKsGps2bVpXgLV8+fJMnjx5q+1XrNi0oOu0adNqVkdZlvnsZz+bn/zkJ0k6gqfzzjtvQAvLAwC196RT98tDVz+Wh37/WL9tm6Y05vn/edQIVMVYstMHDstOHzgs6+9alqUX3JIN965MZWZzUk3a7l2ZckM1jXtMSdnSnoaJTanMnpC2hcvTvM+M7HjG09I0e9JoPwUAGDIBVo3tueeeefDBB5MkDz30UL+BUWfbznNroSzLfO5zn8uPf/zjJMlOO+2U8847L7vvvntNrg8ADF6locgLvn5Mbjj39tz0nwtTbdliGE2SZJcj5+S4zz8lk3eaOMIVMlZM3Gdmdj3r6aNdBgCMKAFWje2zzz65+uqrkyS33HJLjjjiiD7bLlmyJIsWLUqSzJo1KzNnbvs0gc7w6kc/+lGSZMcdd8x5552XPfYwVBwARlulocjhpz0hh/39AVly64rcf/kjWf7XFWma0pTHHbtLdjtmxzRO8vYMAGBz3iHV2NFHH51vf/vbSZKrr756qwuaX3XVVV3bxxxzzDb3vXl4NWfOnJx33nl53OMet83XBgBqpyiKzDloRuYcNGO0SwEAGBMqo13AeHPYYYdl9uzZSZLrr78+t99+e6/t2tvb873vfa/r8fHHH7/NfZ9zzjld4dXs2bNz3nnn1WxaIgAAAMBoEWDVWENDQ0499dSux5/4xCeydOnSLdqde+65WbhwYZLkkEMOyVFH9b5Q60UXXZQjjjgiRxxxRN72trf12e8555yTH/7wh0k6wquvfvWr2WuvvbbhmQAAAADUB1MIh8EJJ5yQSy+9NNdee23uvvvuvO51r8vLX/7yzJs3LytXrswll1ySG2+8MUnHnQdPP/30bervy1/+cn7wgx8k6ZiS8LrXvS733ntv7r333q2ed8ABB2TnnXfepr4BAAAAhpsAaxg0NjbmnHPOyRlnnJErr7wyS5Ysyde+9rUt2s2dOzdnnXVW9tlnn23qrzMMSzrWwTr33HMHdN4nP/nJvOQlL9mmvgEAAACGmwBrmEyZMiVf+MIXctlll2XBggW59dZbs2zZskyePDm77757jjvuuJx44omZOnXqaJcKAAAAUNeKsizL0S6C8WPRokWZP39+kmTBggWZO3fuKFcEAAAAjHUWcQcAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAuibAAgAAAKCuCbAAAAAAqGsCLAAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOpa42gXAAAw1q15bF0W3bg8a5euT+uq1rSub8+cJ+yQJbevyNKFKzNt10mZuc+0rLx3Tdpb2tLWUqbaWs1uR+6YvZ6za5ome0sGALA13i0BAAzRnf/7YK46+y9Zt3jDlgfLMimTovvjapkim/b99Yf3J7k+0/eakhdecEym7zFlROoGABhrTCEEABiC3/+/W/ObD1zfe3iVJEWRVIqURbdd6RZodbPy3jX5wfN+lSV3LB+GSgEAxj4BFgDAIN3zu0dy09fvHFjjokiZpNg4+qovZTX5xcn/V4vyAADGHQEWAMAg/flbdw3yjK2HV51aVrTm/ssfGUpJAADjmgALAGAQ1q9oycN/XDKoc4py4G3/8o0BjuwCANiOWMQdGFWrN7TnivvKPLq6zOTG9vzpkaShkrz8CQ05Yrem0S4PYAuta9oGfc4g8qu0rB789QEAxjsBFjAqfndPW177o7Y8trb342dd1Z6kPc/Zq8glb2hOQ2Ugk28Aht/EGc1JQ5G0DzyWGswr2JS5kwZfFADAOGcKITDivnpda57zrb7Dq+5+e2+Z2edsSFt1MOMXAIZP0+TG7PP8XQd1TlkMfBTWEe9/wuCLAgAY5wRYwIi6b3mZdyxoH9Q5KzYkz/92yzBVBDB4T37TPh2jsAaiLJMUAxqGNfPx0zJzn+nbVBsAwHgkwAJG1Md/1zqk8353b5kNbUZhAfVhxwNn5Pn/ekT/76TKMik7squyKLY6CmvizOa89IfH1q5IAIBxRIAFjKif3FYd8rn//ZfBjdwCGE7znrtLXn/Jc7PP8bum6FxVtCw3+0qP7TI9v5KkaChyyDsen5N//6I0T7Y8KQBAb7xLAkZUy9Dzqzy8yggsoL5M23VynveFI0a7DACAcc8ILGBEbcvggn1meckCAADYHvk0CIyotxw2tJedSpG88kAvWQAAANsjnwaBEXX6M5oGfOOu7l5/cCWVYggnAgAAMOYJsIARNWdykd/8XeNA7ibf5Qlzkm+9vHnYagIAAKC+CbCAEXfsXo25/31NedpuW4+xGorkn5/VkFvfPXGEKgMAAKAeuQshMCp236Eh17y1YbTLAAAAYAwwAgsAAACAuibAAgAAAKCuCbAAAAAAqGsCLAAAAADqmgALAAAAgLpWtwFWW1vbaJcAAAAAQB2oaYD17W9/uybXuf/++/OmN72pJtcCAAAAYGyraYD17//+7znttNOyfPnyIV9jwYIFecMb3pA77rijdoUBAAAAMGbVfArh73//+7zuda/LddddN6jz1q9fn09+8pP553/+56xdu7bWZQEAAAAwRjUOx0UXL16cd7/73XnjG9+Yt7/97alUtp6T3XHHHTn99NPzwAMPdO2bPXv2cJQGAGwHWta25fIv3pE7frUobevaO3aWZYqNx4uyTMp0fCVd+zc23NS2KFKpJEWlTNqTSkNSba2mbO84v9jYvtJQZMcDd8iR73tCdn3KnJF4igAA25WajsD65Cc/mUmTJqUoilSr1Xz961/P2972tixatKjPc/77v/87b37zm7vCq7Isc9RRR+W73/1uLUsDALYTvz771nz5uEtzy88eSvu69hTpCJqKokg2fpVF0Zld9aLjjDJFyjKptiftLUl7a5m29dWU7Z1tOvOvItX2ZNGfl+dnb/p9vnnsJVl216rhfpoAANuVmgZYL37xi/Otb30r++67b9e+m266KSeddFIuu+yyHm1XrFiR97///fnCF76QlpaWlGWZhoaGnHbaafn3f//3zJw5s5alAQDbgV995pbc8tMHk2w+qmozRZFUir7bdYZd2RhSbWxfplvw1RmG9XicrF3ckh+9+oqsuH/NNj4bAAA61XwNrD333DPf/OY386pXvSplWaYoiqxcuTIf/vCH87nPfS6tra25/vrrc9JJJ+Wqq67qOm+33XbLBRdckJNPPrnWJQEA24FVi9bn1oseTtJPeNWpKFJW+mlZFD23+2i/eYjVtqGa3/+/WwZSBQAAAzAsa2A1NTXlwx/+cJ761KfmU5/6VFauXJmyLPPjH/84V199dR555JFUq9Wu9s9//vPzsY99LFOmTBmOcgCA7cAfv3lPUvY9MbBXRUf4NKDAK9k44qpIjwmIRdFrv/deuihrF2/I5DkTBlcTAABbGJYAq9Oxxx6bAw44IGeccUZuuummlGWZhx56qOv4xIkT86EPfSgvfelLh7MMAGA78LcblicZRBiVbBwxVWYrC2L1ck623r4z0CqTJXesyOQ5Ow2mIqiZ1qXr8sAJP8/6Pz6aVMs0PG5qJuw9I9XHOu74XezQnKK1mvZl61NprKThcdOSajXt969K26K1SVt7KpObMvGQHTP97w7K1Fftn6K/UYsAMEyGNcBKkp133jknn3xybr755pQbfzrZObXwDW94g/AKAABqqH1tS+7Y9WvJipZue8u037ki6+5c0WNf0nnLgiR/fqzH6MIiSXXZhqx9cHXWXHxPipMXZOZHnpo5n37GCDwLAOip5mtgddfW1pbPf/7z+fCHP7xFeFWWZb72ta/ln//5n7N+/frhLAMA2A7sfviMJIMbTNU5UmpQ+mvfOZ2wkszef4dBXhy2TXV9W+6Y9uUtwqvOu3H2tOmOm9m4tWnvli3LtjLLPnNtHj55wTBUDgBbN2wB1t/+9re8+c1vzve///2UZZmyLDNt2rS85z3vyaxZs5J0hFkLFizIG97whvz1r38drlIAgO3AU/5uXs9F1weiHOSUw7Lsuf7Vxn29mffsna1/xYi7+2n/nVS77ykH8Gd805/qrbXtPLb6O7dl7aX3D61AABiiYQmwLrnkkpx88sm5/fbbu/Ydcsgh+e53v5tTTjkl//3f/50jjzyy69i9996bN77xjfn+978/HOUAANuBaTtNzBNfumuSAQ6qKssU1X5adg+nyjLpo33RvU2ZNE6s5KgPHDiQKqBmymqZlj8vGdq5g4tys+Sfrh5SPwAwVDUNsDZs2JBPfepT+cd//MesWbOma7rgm970pnz1q1/N3LlzkyQzZ87Ml770pbz3ve9NQ0NDiqJIS0tLPv/5z+eDH/xgVq5cWcuyAIDtxHNPPzAHn7h7kn5CrG5hVK/tyrIrvCq6tS+6f8zf2Gbz8GryjhPyqh8+K9P3cHdlRta6PzyyDWcPLsBaf81D/TcCgBqqaYB18skn56KLLupa72r27Nk599xz8653vSsNDQ1btP+7v/u7nH/++dlll1269l1xxRU56aSTcsMNN9SyNABgO/GcDz8h77782Tn4FbuncUpDynSEVGVnKFVuFkRtoeOMIkmlkjQ0Jo0TijQ2F2mcVEnR0Nlm00f+hqZkl8Nm5YRvHpNTLj0+M+ZNHdbnCL1pX9nSf6NaaRvs4nEAsG2Ksuxj4YYheMpTntK1QPuRRx6ZM888MzNnzuz3vDVr1uQzn/lMfv3rX3ed39DQkGuuuaZWpTFCFi1alPnz5ydJFixY0DXqDgCA4dX26NosnPufm+0d6OTAMg2pDngcVsMuU7L3Q+8cVH0AsC1qvgZWpVLJu9/97nzpS18aUHiVJFOmTMlZZ52Vj3/845kwoWOx02q12s9ZAABAp8adJqeY2jSkc7e4OUE/Zrzv8CH1AwBD1VjLi+2yyy75zGc+k4MPPnhI559wwgk55JBD8rGPfSx33313LUsDxogbHm7PP/6uNb+7p8y6tvRYnGZCQ/LKAyv5rxOa09w4yDuNAcB2YJcLnpuHXvPLHvvK9LfC1aZRWv23TSqzJ2bGO588xAoBYGhqOgLrO9/5zpDDq07z5s3Lt771rZx44ok1qgoYCx5dU+bwr67PYV9tyYK/llnXmi1WVt7Qnnzn5momfGp9PnP5CK7zAQBjxIxX7585Zzy1256OOKrv8VVlt9FX/bVNitkT8rgb/i6Vac3bVigADFJN18ACa2AxFKs2lHnyeRty97KNL0cDfFX6jxc15F1P8wYaADbXunxd7j3mB2m9ddnGPd3/ce25Llb32GrL7Q6Nu07J7E8/PdNPeWKKilHQAIy8mk4hBBiK865r3xReDcL7/rc973xqmaLwRhoAumuaMSmPv+WU0S4DAGqm5ou4AwzWv13bNqTzWqvJL+4Y2rkAAACMHQIsYFSt3FDmwZVDn8l8yV3uWAoAADDeCbCAUbWtk/8swwEAADD+CbCAUTVtQpHHzx56CvWKJ3gZAwAAGO988gNG3fuPHNr9JCY1JsfOa6pxNQAAANQbARYw6t5yWEOO2HXwo7B++Go3UgUAANgeCLCAUdfcUOSyN07ICQc0dOzoJ8uqJPnhqxoyf3+jrwAAALYHhi8AdWFKc5GfvLY5j64p869Xt+Y7N7fnkdVJtZqUZdJYSWZOSj75rIa862nNo10uAAAAI0iABdSVnaYU+ezzmvPZ5412JQAAANQLUwgBAAAAqGsCLAAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhr7kIIAAD0qbquNau+e1uqyzdkwuFzM/nYx412SQBshwRYAADAFtpXb8jDL/1J1l3+t6Radu2vzJiQmR8/MrM++NRRrA6A7Y0ACwAA6KF99Ybcu9d/prpk/RbHqss3ZMmHLk/bfSuz05eeOwrVAbA9sgYWAADQw8Mv+2mv4VV3K869IetvXDRCFQGwvTMCCwAYVx64bkl+9elbs25Za4pKkV0P2SEvOuvgTJjcNNqlwZjQvq416y57YEBtF//Dpdn9d68d5ooAQIAFAIwTa5e15BuvvCota9p77L/vmqX5yrMvz77P3ikvPutJo1QdjB1rfnBHjzWvtmb9tQ8PczUA0MEUQgBgzNuwpi0XvOzKLcKr7u783aO5+B9vHsGqYGxqX75hwG3L9oEFXQCwrQRYAMCYd/VX7kx7S/8fpBf+elHWr2gdgYpg7Jr4lLkDbtuww4RhrAQANhFgAQBjWlktc/NPHxxw+yvOXTiM1cDYN+no3VOZNXFAbae/5eBhrgYAOgiwAIAxbe2yllTbBj6N6aEblw9fMTBOzPrk0f22KSY3ZvY/HTMC1QCAAAsAGOOKohjtEmDcmfn3h2fGB47o83gxpSl7/OENKZobRrAqALZn7kIIjBnL1rbl/OvactX9ZSY3JbtNS6ZNTKZPqKRSFNl5WpHn7F3JnCneTMP2ZNLMpjQ0FWlvHdgorD2fNmuYK4LxYcf/d1ymv/ngLH7/pVn3+weT1moqMyZmh1OflJmfOCqVRv/eAjByBFjD7PLLL8/FF1+cW2+9NUuWLMmUKVOyxx575Nhjj80rXvGKTJ06dVz0CcPp13e2Zf6Frenzs2lZ7f4gMycmXz+hKS87sGkkygNGWVEUOeTVe+RP37l/QO2Pfsfjh7kiGD8mHDgnu13yqtEuAwBSlGXp3rfDYO3atTnjjDNyxRVX9Nlm7ty5+exnP5uDD67N4pej0efmFi1alPnz5ydJFixYkLlzB34XG+jN169vzZt/3jawxmXPB599bmM+8szm4SgLqDOt69rz1RddkbZ17Vtt98QTds1zP3rgCFUFAECtCLCGQXt7e97//vfn6quvTpLMnj07J5xwQubNm5eVK1fmkksuyU033ZQkmT59ei644ILMmzdvzPXZGwEWtdTaXqb5zPWDO2mzEOu2907MATta7g+2B+tXteabr7k665a29nr8iS/bNc/9mPAKAGAsEmANgx//+Mc5++yzkyR77713vvKVr2T27Nk92nzxi1/MhRdemCQ59NBDc/7554+5PnsjwKKWTv3ZhnztT9X+G3a3WYD1kv0r+fnrB3YrcGB8WHz36vzyH2/OqkfWp2goMu/pc/K80w9MQ6MwGwBgrBJg1Vh7e3vmz5+fxYsXJ0kuvPDCHHDAAb22e8Mb3pCFCxcmSc4999wceeSRY6bPvgiwqKUpn1qXtQOcPdhDt1e15kqZDf80uWY1AQAAMPL8KLLGbrjhhq4g6bDDDus1SEqShoaGvPa1r+16fMkll4ypPmEktA5y8NVwXQMAAIDRJcCqsauuuqpr+5hjjtlq26OPPrpru3PtqrHSJ4yESTW4ieAk91oFAAAY8wRYNXbXXXd1bR900EFbbTtnzpyuKXZLlizJsmXLxkyfMBJOf0bD4E/abA2sNzx5CNcAAACgrhibUGP33Xdf1/auu+7ab/vddtstixYt6jp35syZdd1n53l96ZzKCLXw4ac35fTftGeoswArRfLJ45prWhMAAAAjzwisGlu1alXX9owZM/ptv8MOO/R6br32OX/+/K1+nXLKKV1tlyxZ0ud1Wltb88gjj6SlpaXPNosXL87KlSv7PL569eo89thj+hjHfRRFkZveOYgAqtvoqyJl/vfkpuwyrdhqH5sbq98rfehDH/rQhz70oQ996EMfjGcCrBpbu3Zt13Zzc/8fvCdMmNC1vWbNmjHT50D84he/6PPY8uXL881vfnOrLzg//elP88c//rHP4zfeeGN+8IMf6GOc9/HEnRuy9CPNOXBO0pFQlV2/pCx7fqVMUs1Tdm7J/R+YkOc9fstFtMbz90of+tCHPvShD33oQx/60AfjVVGWZdl/MwbqyCOPTFtbW5LkmmuuSWPj1mdpnnHGGfnf//3fJMmnP/3pvOAFL6jrPgcyhbBzFNa3vvWtHHjggb22a21tzZIlSzJr1qw+Q7fFixenubk506dP7/X46tWrs27duuy444760Ic+9KEPfehDH/rQhz70oQ99MI4JsGrs2c9+dtcwxyuuuCKTJ0/eavsPfehDufTSS5MkX/jCF/KMZzxjTPTZl0WLFmX+/PlJkgULFnQtGA8AAAAwVKYQ1ti0adO6tpcvX95v+xUrVvR6br33CQAAADBSBFg1tueee3ZtP/TQQ/22f/DBB3s9t977BAAAABgpAqwa22effbq2b7nllq22XbJkSdeaUrNmzcrMmTPHTJ8AAAAAI2Xrq30zaEcffXS+/e1vJ0muvvrqrgXNe3PVVVd1bR9zzDFjqk8AGG33Xbc0vzj95mxY3b7xTqRJyjJFZ4OiyC5PnJaXf/HQTJiy5V1JAQAYO4zAqrHDDjsss2fPTpJcf/31uf3223tt197enu9973tdj48//vgx1ScAjJb21mq++pIr8+O/v7EjvOqmKIqk8yvJw39ZlS8/5/L8/oK7RqNUAABqRIBVYw0NDTn11FO7Hn/iE5/I0qVLt2h37rnnZuHChUmSQw45JEcddVSv17voootyxBFH5Igjjsjb3va2EekTAOrZf73q91mzpHXTjo2jr4o+2qcocs359+TWBf2vEwkAQH0yhXAYnHDCCbn00ktz7bXX5u67787rXve6vPzlL8+8efOycuXKXHLJJbnxxhuTdNwF8PTTTx+TfQLASHvoL8uz6tENW+zvM7zqalDkt/9yew6cv+uw1AUAwPASYA2DxsbGnHPOOTnjjDNy5ZVXZsmSJfna1762Rbu5c+fmrLPO6rEI+1jqEwBG2m//ZWHPHZ1rXw1A2/pq/nbDsux+qBuYAACMNQKsYTJlypR84QtfyGWXXZYFCxbk1ltvzbJlyzJ58uTsvvvuOe6443LiiSdm6tSpY7pPABhJyx9ct8W+fkdfdfO3PwmwAADGoqIsB/GjS+jHokWLMn/+/CTJggULMnfu3FGuCIDx5Nzjr0jL6rZNO7rfdbA/ZZmnv2vfPOXv9hqGygAAGE4WcQcAxow9n7rl6KnB/CTugOfvXLtiAAAYMQIsAGDMOO59+/XcUQxw/FVZZurc5kybO7H2RQEAMOysgQWMK39+uC3fuqkt9y+rZlJTcsjOjTn5yQ3ZaWrDaJcG1MDUORPyxJfskr9c9HCP/WX6WQurSE780uHDWRoAAMNIgAWMC+dc2ZKP/7Ytbe2bHbipLR+4pC2zJyU/P7k5Rz/Oyx6Mdcd/7AlpmlTJDT94sGNHUSRl2TWVsEeQVZapNBV5w7eOyqzHTRnhSgEAqBWLuFNTFnFnNLz5Jxvy9T9tnlz17mcnNeWlT2ga5oqAkVCWZa6+4J78+ScPpnVde8q29lTbkpRJUUmm7zoxLzn74Oy47/TRLhUAgG1kKAIwpl19X1u+fkN7x5CLAcTxr/jv1qz7ZGOaGgZ83zKgThVFkWPeuneOeeveo10KbBfK9mpW//Svabl9aTb8aVE2/OWxVFdtSKW5MZUdmlOtVlMsW580VtK429Q07jwl7WtaUz66JmVrNZUdJ6VMUr1nZarL1yXt1RTNlWTmxBRJKi3tKac0pWhuSLFsfaoNRZrmzUi5ckOyqiXt1bZkyfpkTVvSWKaYOiGVx+2Qxh0mpmGnyUmlSHXJ2lSXr0sxZUKK6c1pPmDHZEIl6396e7K6JZnUlLS1JW3VNOw7K+Xq1pQPrUwqSdOL9s/UDz89jbsIvQHqkQALGNPe+YvWjo0BjiVtL5MvXNWaDz+zefiKAoBxpGxtzyNvWJDVP/5r0lbtfqRrq5qkTJli477yvpXZ+C90177exkqXKTtCqY3HO38e1fljptZ7VvTop+vHTxuScs3atC9amzJJay9tylTTmtt6WR+vo2114dIed7TasPCatHzxmmT2xMy87M1peqK7lo5b7dXkF39K/uePyd2Lkr88kKxe37G/8w9hWXZu9HGR3vZv3r7OJjtVNv5tqPZRV1NDssuspLUtqVaTVeuSdS0dT6Oxksw/Ivne+5OJE0asZOhOgAWMaX9eNPg3Bl/5Y3s+/MxhKAYAxpmypT33Hfhfab1reS9HOz4MV1OmkqRIsXFftdvR/v6d3tSq6Iizer1Neu/jpjv7r27sv0flW7nd+qaIq3tY1nX+kvVZdvCXM/OPb0/TEbv1Uz9jzp/vT17+rx3BVW/Kzv9tLbwaqO5/KusgzOoruOrU2p7c/1jvx9qqyc/+kEx6XXLe25K3v6D29UE/+n5dB6hzre1DeyOwtrUO3kAAwBiw6F2/7iO86q7o8dG8M8YauM4Qa/Ozyl739nb25uHVwMKCYqutlh37XwO4BmPKvY8lz/lMcvejW2k00PCq9z+vA28/hr3jP5Nf/mm0q2A7JMACxqymhiJDWcpq7tRx9AYCAIZJ2VbN6u/cNrC23baG9q/spnFQ1V72br3vzYOowdSwlRBrTWs2/O6uAV+JMeBzP08Wr0rtRkMNdoTVOHoPesqXRrsCtkMCLGBMO/HAwb+Mfe54s6cBoD8b/rQo5fq2Abfv/PhebLFnqAZ6/raFAlvrZe1ZV27TtakjazckF141gIZD+XO7HY7uf2xFck8f0zBhmAiwgDHtX56/cTH2Ab53nTkxeeF+TcNXEACME+2rWwbRehyNLOmmHNT3gLr2wJKOhdprbnz+2R+QP/x1tCtgOyPAAsa0x82o5JcnDyyQmtSY3HHaxGGuCADGh+Z5OwyiddnL1kjZth63Fj80PHGnbbo2daSpYbQrGH8muqs3I0uABYx5L9ivKY98aGKO2LX3442V5K2HVbLijEnZcaqXPQAYiKZ5M9I4iBCr6Pb/bLE9UN3XrxrMSlZ9P+qvv62Z9tnnDeJa1LW9dkzm7TgMF+5+L8vtzPMOGe0K2M5YCAYYF+ZOq+SP75w82mUAwLgy5/89K4+c+PN+WvVcNL2ajp+SD+4j/aYgafNlsfu/C+GWIdRgIoW+2jUeuXsqc6YM8CrUvUolecdzk4/8dz8NiwxtVN9QzxujDpuXTJ4w2lWwnTEUAQAA6NW0V+yfWWc/YystOsKrYrPHHVvJQO8j2NFy87sHFtnqXQI3nltm89hgIOf1rHVzlX1nZeb/vbXfKzDG/P3zk2cckK3/uSy6ffVnsIHVOAm4Gorkfz8x2lWwHRJgAQAAfZr90SOz551vzZSX7ptM7FxHqGdsVG78r3Nf51Y1HTFR2bWv7HZ08xbdr9x5vc172/z8TeNeqlscSY9+N++72OLaSaY2Zeo3Tsicv74vRYOPSuPOxObkfz+SvPt5yZRtWRd1y9h064FXb+3HqB0mJw98LdlxMGvkQW0UZVmOk79J1INFixZl/vz5SZIFCxZk7ty5o1wRAADAZlauTS75c7JyXfKnu5Ob7k8eXp7MntoxwmjNhmTF2mTZ6qSlNakUyaypydqWZN2GpKGSTGhOijJpaesIxKZMSDa0JmWZrFmXtLV3fLW2Je0bP3ZXiqTYGHY1bNwuiqShYePjjfVNaE7Wbug4tmFj/5OakxmTk8WrOo5Vy46pkc0NyfTJHeFSc2Py4JJk9YaO60xu7gjuGhqSA3fraH/z/cljKztq67TDpOTlT0sO3ye54Z6OcyY3Jz++pqOvvecm3zot2XeXkfodgi1YAwsAAIDty/TJyauO7Nh+y3GjW0s9+5c3jnYF0EWABQAAbLfablmUVR/+VVr/9FCyeE3SVt14ZMs1srqv9ZVu292nQBbdG5ZbrgTWsV3d9KiSZPqEpL09xaoNW7RPQ5E0VZL1bVvpf9O+njVvbNdYSTF7csfRskyWrk3a2nq271qDvPsUyy0n6xTFpufVvY8t2/dRz8TGZH1rz/0TGpN5s1MsWpksW9vzOs0NKR4/N9l9ZnLj/cnq9R2jn1pak/WtffTTW+1limwc4VRJx4im6ZOSo/ZN/vEVyVP33eIKQH0xhZCaMoUQAICxoNrSmuVHnp+2Gx7OliFI/6FIX8d6BjlbLnLf+f8tz6/2spD90PrdtN3f8+rro2BniLVlPWXXc9rauUnfi+RXN6thy3N7Huutfc91zHq/Tv/fsx7bT9g1+dXHkt1n93lVYHRZmRAAANjuLDvo3EGGV517+4pNtoyQ+mq95ciuair9hFd999x7dLVl64GGV52tKr1eo/97PG66i9+Wrfq7N+Xm47j6GlW2tfCqew1b/5712L7toeSIM5LFK7d6ZWD0CLAAAIDtyrpv3pD2O5d2PR5YeNVT7xFO0W1r85Bl66Odtqb/enoJZPo90r+ylxBq4NfoPzrq/7zevof9hVfdr7O1qK2X78yiFcn/WzCgqwMjT4AFAABsV9aedfnGrV7WeBrQFbbequwnkNq89WCDpb6vtPl20cuR3h73pfu4r6GsPNP7KLCB6Ot7WKvvVZ9XPu83HXcNBOqOAAsAANiutD+womu79oHI1qau9da6VksS9zbWatPVh/o8u1fX18LuAz1/4Ab3PezvOoM6vmJd8relW+4HRp0ACwAAADoVwzfOCxg6ARYAALBdadh3053man9L9rLb/wfSevgmEHa/+lCfZ88VqTbfM7jzB25w38P+rjOo4ztOS/ZwJ0KoRwIsAABguzLlU8/ZuLVlvDKw0KS/Rde3XMJ9a62HbxJhXxMIB1pb9wXmhxJFbcsS8L3fv7H2gWOnjVd+z/OTBh+ToR75mwkAAGxXJr7sCWk8Yteux5tCkYGs8rS1cUg9R0H1vE5f4U3/YVf/oU1vo6/6O9K/3iobzDLsA9/bV4veAquBhlj93bGwl+/MPjsl73vBgK4OjDwBFgAAsN2Z+Yd3pPn4fdM9tOoeTfUWkvR1Z7yy17OLrr39B2SVVPsMZjavr2fPvVVd9NK67OOM3pVJqr30WPQbC6Xbs96y1dYCwt4q6619ka2HWD2/61v7nqV7i2cekPzh08n0yX1eGRhdjaNdAAAAwEgriiIzLjkl1cVrsvrjv0nr7+9P+73LUq5rS9p6C2+6T2vbGGU1dOwp2qs9G05s7MhZ1rVujLKSsui4ZJEiRaodC4VXiqSpkmL3HVK0VVM+sjLl+taewc/0CcnkpmTp6hQt3WOodG1vMXWwqdJx/bb2pFKknNSYYu85KVo2Pn5wWcoV67vqSWOxsch0nNN1zaJnaFckxYSmpL2asrW9Ry29RVrl5vubimTHHVI8sjKplpsKnjUlecpeKe5+LLn7sY3f/43HZk5J8ZS9kyfumlz852Tx6mRKc9LamvKRlUl7dYvvQbfvRFJsrLDS8XuehqJjimBjJdl5RvLsA5OPvjTZc8dengFQTwRYwLi0dE01H/tNay75azWPrq5mXVu3g33+yK7s68ebSZKpzckPT5qQF+zfVNtiAYBRU5kzJdO/+rLRLoOB+H+vG+0KgFEkwALGnQ//b0v+3/+19TnMv+/b8RSdo9KTstzi+OqW5IXf2JAdJmzIvR+ZkhmT3GIZAABgJFgDCxhX/um3LfmX/2vb9jvUFEWfS0Os2JDsdc6atLQN331wAAAA2ESABYwba1rKnHV551zBfsKlgWRPWwux1iffuaF1MOUBAAAwRKYQAuPGBde1pbXaf7tBKYqO6YS9+OglLXnTU5pr3CFsf1Y+uiGXfH5h7rlmWUe2XHZbELjay8K85aZ9RZKG5iJHv3VenvK6x6WomNoLADAeGYEFjBvXPFDr9GrrHlszot3BuHTfDctz3muuzd3XLEuycRm6oui4U1TRcaeosrJxu9tXWWy6tXp7S5krv3x3vvXGP6baNrKvAwAAjAwBFjBuNI7wK1phoAdskxWPrM/3P/TnlNU+Z+t2KIqes347w6vN/hIuvmtNfvnp22pdJgAAdUCABYwbLzmgYUT723/OiHYH484NP304ZWs/4VWnIr2GWEnP/bf/5tFsWN0WAADGFwEWMG684qCGTJ9Q44v2sf5VknzxxRNr3BlsX/78y0cG3rivIY+b7y+T2369aOhFAQBQlwRYwLjRUCnyrRObN47m6GdMx0CGfJRln3crPHzXIsfv5z4YsC3WrxyekVIrH14/LNcFAGD0CLCAceVlBzbmopObM3NSstWUqu+BVRuP9x1ePWfvSv7w7slDrBDo1DRpeKb9TpnVNCzXBQBg9Bg+AIw78w9ozNKPN+biO9pz4U2tuXNxex5YnixvSVraN2VTRToWfp81KVmyNmkvO2YjTSjKtFWT1m4BVlMledmBlXzr1ZMyqcnq7VALT3jOjrnp5w8PrHFf03l72X/A8TtvQ1UAANQjARYwbr1o/4a8aP+RXdgdGLgjTtw1f77o4ZTlwGb19mjTLbjqvv9xR8zMlFnNNaoQAIB6YQohADAq5uw1JS86ff8k/czq3XxK78bwqths9NXkWU152dlPrG2RAADUBQEWADBqnnj83LzhvCdn5m4TU2azIKssk2o1qZYdo6zKsuurM7zqHH11wHN3yqk/OirNkw0uBwAYj7zLAwBG1a5PmJ63ffepad3QnsfuXp1Fd6zJhKkN2XHfyWlvSZb/bXUeXbg669e2Za/DZmbnA6fnwZtWpNpaZpeDd8jM3SenKKxNBwAwngmwAIC60DShIbs+YYfs+oQdeuzfeb9pOeDZPdtOf547gQIAbE9MIQQAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAuibAAgAAAKCuCbAAAAAAqGsCLAAAAADqmgALAAAAgLomwAIAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAutY42gUA1LOWtjKv+u7aXHxHNW3VjTvLTccbK8nTdi/yizdOyozJDaNSIwAAwHgnwALow1t/vC4XXNe25YFi469l0lZNrrq/zMwz1+boxxW56l1TR7RGAACA7YEphAC9ePdP1/YeXnVXZFOYleTq+8s89dzVw1oXAADA9kiABbCZFevLfPna9oGf0C3E+uPfyix8bBDnAgAA0C8BFsBmPn/5hm06/+9+sK5GlQAAAJAIsAC28JPbWgd/UrdRWLcsKvtuBwAAwKAJsABqrJRfAQAA1JQAC2AzLz1gCDdo7RZaHTi36LsdAAAAgybAAtjMh541cZvO/9arJ9WoEgAAABIBFsAWZkwqcuoRDQM/odvoq8N2LXLAToM4FwAAgH4JsAB68Z8nTs4bD+1nKmGZHuHVU3cvcv3fTx3WugAAALZHAiyAPnz91ZOy7sypOf7xlTR0X9aqW3DVUEmO2K3IY2dMzrXvEV4BAAAMhyGsVAyw/ZjYVOSSN08Z7TIAAAC2a0ZgAQAAAFDXBFgAAAAA1DUBFgAAAAB1TYAFAAAAQF0TYAEAAABQ1wRYAAAAANQ1ARYAAAAAdU2ABQAAAEBdE2ABAAAAUNcaR7sAgLGmpa3MR3+5Pv/x+7a0tJe9ttllWvLll03MCQc3j3B1AAAA448AaxgtXrw4P/7xj3PllVfm4Ycfzvr16zNnzpwceuiheelLX5rDDjusZn399a9/zTXXXJObbropd911Vx577LG0tbVl+vTpmTdvXp761KfmZS97WebMmVOzPmF7dPeSag78/JpsaO/cU2zWoiPQenhV8vIL1+fQXVvyp9OmjmSJAAAA405RlmXvwwfYJpdddlnOPPPMrFy5ss82r3jFK/KRj3wkDQ0NQ+5n1apVOeWUU3L//ff323bixIn5+7//+7z61a8ecn/9WbRoUebPn58kWbBgQebOnTtsfcFIW7WhzNwzV2ddW38te76svnC/hlz8linDVhcAAMB4ZwTWMLjuuuvysY99LK2trUmSpz/96XnmM5+ZSZMm5fbbb8/PfvazrF69Ov/zP/+ToijysY99bMh9bdiwoSu8amhoyJOf/OQceuih2W233TJp0qQ8/PDD+c1vfpNbbrkl69evzznnnJMNGzbkDW94Q02eK2xPzv9DywDCq6RjVNamEOuXC9uztqXM5ObNR2sBAAAwEEZg1VhLS0te+cpX5qGHHkqSfOhDH8prXvOaHm3uu+++vO1tb8uSJUuSJF/5ylfylKc8ZUj9LV68OK95zWty0kknbXWK4De+8Y2ce+65SZKmpqb84Ac/yB577DGkPrfGCCzGs50/tSqLVg+0dc+X1o8+qylnv2hSzWsCAADYHrgLYY39/Oc/7wqvnvGMZ2wRXiXJnnvumY985CNdj7/yla8Mub8ZM2bkoosuylve8patrm/1xje+Mccee2ySpLW1NRdffPGQ+4Tt1eI1Qz/3hoertSsEAABgOyPAqrFf/epXXduvf/3r+2x37LHHZtddd02S/PnPf87DDz88pP4aGxszefLkAbV93vOe17V95513Dqk/2J41Dn25ukyfULs6AAAAtjcCrBpas2ZNbrzxxiTJlClTcuihh/bZtlKp5Kijjup6fNVVVw13eZkyZdMi0hs2bBj2/mC8edF+Q3/J/MixE2tYCQAAwPZFgFVD99xzT6rVjmlC+++/f793FzzwwAO7tu+6665hrS3pOepq5513Hvb+YLz55PMGGkL1XP9qxqTk8N23YfgWAADAds5dCGvo3nvv7drunB64NbvttlvX9n333TccJXVpa2vLz3/+867HT3/604d0nUWLFm31+OLFi4d0XRgLDtm1IR9+ZmPOuWJrtyLsGV4VSa5658Cm+QIAANA7AVYNrV696fZkM2bM6Lf9Djvs0LW9atWq4Sipy9e//vXcf//9SZL99ttvyAFW5x0GYXv1ufmTstes9fn7n7embYt12XuGVztPSy5/++Tst6OXWgAAgG3hU1UNrV27tmu7ubm53/YTJmxa1XnNmm24vVk//u///i/nn39+ko5F3z/+8Y+nUjF7FIbqnUdNzDuPmpir72nNl37fkkdWltl5epGUyZrW5Ak7FXn/0ydk5+mmDQIAANTCdhdgve1tb8uf/vSnmlzrM5/5TJ7//OfX5FrD5dZbb83HP/7xrrW53ve+9+Wggw4a8vUWLFiw1eOLFy/OKaecMuTrw1hy9LymHD2vabTLAAAAGPe2uwBrOE2evGmdm5aWln7bd78TYPc7BNbKX//617z3ve/tGt311re+Na997Wu36Zpz586tRWkAAAAAA7bdBVjPec5zsv/++9fkWnvuuWePx1OnTu3aXr58eb/nr1ixomt72rRpNamp05133pl3vvOdXX286U1vyjve8Y6a9gEAAAAwEra7AOs1r3nNsF17r7326tp+6KGH+m3/4IMPdm1vHoZti87wqjNEO+WUU/Lud7+7ZtcHAAAAGElW8q6hefPmdS2Ofscdd6S9vX2r7W+99dau7X322acmNXSGV8uWLUuSvOENb8h73/vemlwbAAAAYDQIsGpoypQpOeSQQ5J03FXwxhtv7LNttVrNNddc0/X46KOP3ub+Nw+vXv/61+e0007b5usCAAAAjCYBVo0df/zxXdsXXnhhn+0uu+yyrimEBx98cHbddddt6veuu+7qEV6ddNJJef/7379N1wQAAACoBwKsGnvpS1+anXfeOUly5ZVX5gc/+MEWbe6///587nOf63q8tcXVX/KSl+SII47IEUcckeuuu67XNnffffcW4dU//MM/bMvTAAAAAKgb290i7sNtwoQJ+cd//MecdtppaWtryznnnJPf//73eeYzn5lJkybl9ttvz09/+tOsXr06SfLyl788T3va04bc36JFi/LOd74zS5cuTZLsv//+Oeyww3LZZZdt9byJEyfmyCOPHHK/AAAAACNFgDUMnva0p+Xss8/OmWeemVWrVuXKK6/MlVdeuUW7l7/85fnoRz+6TX098MADWbJkSdfjO+64Ix/84Af7PW+XXXbJRRddtE19AwAAAIwEAdYwOe6443LwwQfnRz/6Ua644oo8/PDDaWlpyZw5c3LIIYfkZS97WQ4//PDRLhMAAACg7hVlWZajXQTjx6JFizJ//vwkyYIFCzJ37txRrggAAAAY6yziDgAAAEBdM4UQABhRf/7fh/O78+7N+tXtSVvPgeDNE4oc8tKd85x37ztK1QEAUI8EWADAiLj7D0vzg4/c2vGg2woGRbc2LRvK/PGHD+e6Hz6cI1+/W5516t4jWyQAAHXJFEIAYNjdf9PyTeHVRkV6hlfd95VJrvnOg7nigntHpD4AAOqbAAsAGHY/+Ogtmx4M4P4xncHW77/9QKrt7jcDALC9E2ABAMPq0btWp219zxBq85FXW3P1t++rbUEAAIw5AiwAYFjd8tvHtun82363uEaVAAAwVgmwAIBhVW2rbtP5ZdUUQgCA7Z0ACwAYVvscOXubzt/9kOk1qgQAgLFKgAUADKu9DpuRSsPQz3/e3+9bu2IAABiTBFgAwLB77nv23vSgKNLfpMDO4/scNTNNE7Yh/QIAYFwQYAEAw+6wE3bNMW/co8e+Muk1yCrTcZfCPZ68Q1559hNHoDoAAOpd42gXAABsH55xyp456nV75Edn3JJ7r1+RtHfEV91DrCLJrMdNzEn/9qRMnTlhVOoEAKD+CLAAgBHT2FzJa885eLTLAABgjDGFEAAAAIC6JsACAAAAoK4JsAAAAACoawIsAAAAAOqaAAsAAACAuuYuhAA19McH2vKf17ZmdUuZJ+1cyfueMSGTmorRLgsAAGBME2AB1MAND7blpd9cl7+tKLv2fe+m5B9/1ZLXPbkx33j1pDRUBFkAAABDYQohwDb604Nteeq5a3uEV53ay+TCG9ry3PPXpiy3PA4AAED/BFgA2+hVF65LW3XrbS67uz0/vrl1ZAoCAAAYZwRYANvgTw+25+6lAxtZdeZvW4a5GgAAgPFJgAWwDX61cOCjqm5Z1M8wLQAAAHolwALYBi3tA29blrEOFgAAwBAIsAC2wRG7DfxldOdpRYrCnQgBAAAGS4AFsA1esH9TZkwcWNv3P6N5eIsBAAAYpwRYANugUinyry/uP8HaeVqRdx0lwAIAABgKARbANnrTU5rzry+ekEofswP3nFnkpvdNyZRm0wcBAACGonG0CwAYD97/jAk59anN+eSvN+QXt7VlfVuZ3Xco8onnTMjz928a7fIAAADGNAEWQI1MnVDk8y+emM+/eLQrAQAAGF9MIQQAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6ZhF3AGCrWta15xdfuie3XLok1WqSskySFJ0NOh6m6NzofNy9XVluar9x35x5k/LyTxyQOXtOHtb6AQAY+4zAAgD6tPhv63LOq67Pzb9dkrLaEUoV6XgDUXR+FR1fnXuKouMrlUpSFCnKsmf7zmvfsy7nv+mGXP/Th0bhmQEAMJYIsACAXpXVMv/57r+kvbXsGkWVpMdIqh42Hii779ps5FVvfvXv92TRXWu2qVYAAMY3ARYA0Ktrf7YoreurW0z926ruDcqyaxphfxacs3CQ1QEAsD0RYAEAvbrmJ49sejDAIGpz/QZeGy3669ohXR8AgO2DAAsA6NX61W2DG33VrWGZDHj0Vadq+9BCMgAAxj8BFgDQq8bmnm8TBhwvlR1hV1kMOPJKklQaBtceAIDthwALAOjVE46ZuSm0GmQY1WmgodfU2U1Duj4AANsHARYA0KvnvnmPwedWPW5BWAx4FNaz3zFvkB0BALA9EWABAL2aMKUxLz5tXo9RWAMZUdXjRoTZ+iisMsm8p8zIQc/ZcYhVAgCwPWgc7QIAgPp12At3yg47Tcj/fO7OrF3RlqIoUm5cnL0zqOpYsD0995XlFkHW5uc0NCRPfdWuOe5tRl8BALB1AiwAYKv2OXyHfOgHh2fV0pYsvHZZ7r1pZdatak2lUmTa7MZMmtqUNctas+LRDVm7rDXNkxsyZ48JmbPHlKxe2pL1q9rTPCkp25OyvZqpcybm8UfOzC4HTB/tpwYAwBghwAIABmTarOYc/sK5OfyFc0e7FAAAtjPWwAIAAACgrgmwAAAAAKhrAiwAAAAA6poACwAAAIC6JsACAAAAoK65CyEAbKdWLW3Jr7/xYB7869qsXt6apEyloUhDkZTVpGlSkaKabFjfntZ17Wme0JAd95yUl39g7+yw04TRLh8AgO2IAAsAtjMrFm/I+R+8Iysebe27UVmmKKspkq6vtnVtuW/5qnzh727KxMmVvOMrT8yMuRNHqGoAALZnphACwHZk+WMb8q9vvmXr4VWSFEXKFF3hVY9DSTasrebf3vjnLH9kwzBVCgAAmwiwAGA7cuGZd6XaVvbfsCzTkHKL8Kpnm+SbH7mtVqUBAECfTCEEqLHr/9aWj/1yXa65r0xLe9LSXqbsnhds3C6SNFaSsizTttm+qc3JYbs35F9fOilP2tVLNbWx7NENWXTXugG17W3kVW+WL2rJkr+ty+zdJ21TbQAAsDU+FQHUyMLH2nP8+Wty3/LOPWVXWJWi42G6BVllktZqz2t07lu2LvntX9tzyOdXZ6+ZRa5497TsMdOgWbbN/2/vzuOjqg7+j3/vzCQhCZCQRIKsohSiFFBEBRWUH24t7rWlbuXRFpFHhapVhAe3sAmWqr9SpL+nraXFvVQQcOsilVKwUhG1gAjKYoiUsCchJJm5vz8mc0kgk0ySmblnJp/365WXd2bOvefccJy58805527+4GDEZS07glFaNbasPUiABQAAgJji2xAARMEXewMa+HTt8ErBNCo0jOW48Kquesa61Hpq235bfZ88pOJDgeN3BJqkqiLyUKopKivomwAAAIgtAiwAiIL7llWorM6a2HZk86/qqGeHmqcOH5UmLots6hcQTn7PNuFz1OPYTei/HXsw+goAAACxRYAFAC2062BASzdUN1woCgNfXlpXpYNHYjOCBq1Dr7PaKzU9so9+W1ZE3dbjs/SNc7Nb1C4AAACgMQRYANBCH3zlV+D4b/r1LNrebDUjYaoC0qdf+1t4MLRmlmXp8tu6RNYlLauxexBKkobf2kUeb5OHGwIAAABNQoAFAC3UhLWuW+yEoAxoovOu7Khh38uPqGxAdoNh18Bv5enCUZ2j0zAAAACgAdyFEABa6MzOXmeddod13HZLgqeafT2WdEY+f3dAy132X101+KqOennWl9r579JjIaxdd+igZVkK2Had2wxYknI6p+nWmX2Und8mru0GAABA60WABQAtdEqORyO+4dWfP29gel9LQyxJ1/T1KTeTAAvR0T43VWNm93G7GQAAAEBE+CYEAFHwsyvbKLXOO6rVjMCqnh1qnkrzSj+9mju9AQAAAGidCLAAIAr6nezVP+7OVHZarSdDo67smu2w61yHCtXztKTsdGnd/e10aq43eg0GAAAAgARCgAUAUXJ2V6/2FbbTktHp6nOSlOKx5AkFV6F86rgQy6odXNXKsdK80ukdPVp6e4b2T8vW6fmEVwAAAABaL9bAAoAosixLV/dN0dV9U9xuCgAAAAAkDUZgAQAAAAAAwGgEWAAAAAAAADAaARYAAAAAAACMRoAFAAAAAAAAoxFgAQAAAAAAwGgEWAAAAAAAADAaARYAAAAAAACMRoAFAAAAAAAAoxFgAQAAAAAAwGgEWAAAAAAAADAaARYAAAAAAACMRoAFAAAAAAAAoxFgAQAAAAAAwGg+txsAAKjrq/1+3fZyudYV+VVRLaV5pbKjtir9wdfTU6Qxg1M055pMeT2Wu40FAAAAgDggwAIAQ1T7bQ39xWGt2R6o83yZJNnHHpdXSc+srNIzKw9o1sg2enBEelzbCQAAAADxxhRCADDEoKdODK8cVv0jrSYur9DclRUxbBUAAAAAuI8ACwAM8JfPq7S+OEx4FRImxLr/9SOybbve1wAAAAAgGRBgAYABfvL6kcgK1hNiVfqlFz+sjHKLAAAAAMAcBFgAYIDNexoZfdWI5RsIsAAAAAAkLwIsAEgC3IsQAAAAQDLjLoQxVFJSokWLFmnlypUqLi5WRUWF8vLydNZZZ+nqq6/WwIEDY96Gbdu26eabb9bRo0ed59auXRvzegE0Td9OHn2ws/mjsK7pnxrF1gAAAACAWQiwYmTFihUqLCzUoUOH6jxfVFSkoqIiLVu2TNdff70mTpwor9cbkzYEAgEVFhbWCa8AmGnOVekaNq+s8YL1LNae5pNuIMACAAAAkMQIsGJg7dq1mjRpkqqqqiRJF154oYYNG6b09HRt2rRJS5YsUWlpqf74xz/KsixNmjQpJu148cUX9fHHHys9PV1HjkS4QDQAVww9LUWDu3u1Zoc/fKEwdxp89jvpssLcoRAAAAAAkgFrYEVZZWWlCgsLnfDqgQce0NNPP63rr79e3/rWt3TvvfdqwYIFys3NlSQtWrRIH3zwQdTb8dVXX+nZZ5+VJI0bNy7qxwcQfavuaasR3wgzIrOB8Oq289rEsFUAAAAA4D4CrCh7/fXXtWvXLknS0KFDNWrUqBPK9OjRQxMnTnQeh4KmaLFtW1OnTlVFRYX69u2r73//+1E9PoDY8Hgs/fnOdtpb2F439PepS3tLHdKlru2l9qmSxwr+ZLeRpn2rjQJzsnXnBYRXAAAAAJIfUwij7J133nG2b7755rDlLr74YnXu3Fm7du3Sxx9/rOLiYp188slRacOiRYv0r3/9Sz6fTw8//LA8HnJKIJHkZHr06ui2bjcDAAAAAIxBgBVFZWVl+uijjyRJmZmZOuuss8KW9Xg8GjJkiBYtWiRJWrVqlW644YYWt+Hrr7/Wz3/+c0nSrbfeql69erX4mACAxLCnqELLf7NLWz46rOoqW7JthVZHs46bhepLkc4akaNv395VaRmxuZkIAAAAEC0MzYmiL7/8UoFAQJLUp0+fRu8ueMYZZzjbW7dujUobpk+frrKyMnXv3l1jxoyJyjEBAOb7yytfa864Tdr0waE64VXop+4DqbpKWvvWPk278WNt31DqVrMBAACAiDACK4q2bdvmbHfu3LnR8l26dHG2t2/f3uL6X3/9da1evVqWZWnKlClKTU1t8TGPt3v37gZfLykpiXqdAICG/esv+/SnhV8fe6JWeFUvS5It2ZICfulXk7food/1VWb7lJi3FQAAAGgOAqwoKi099hfs7OzsRstnZWU524cPH25R3Xv27NFTTz0lSbr22ms1cODAFh0vnJEjR8bkuACA5rFtW+88X1znuQbDq9qFaqYVBqptvf3bXbp+fI8YtBAAAABoOaYQRlF5ebmzHcnop7S0NGe7rKysRXXPnDlThw8f1kknnaQJEya06FgAgMTxxaelOlhSdeyJWuteNcWnKw9Eq0kAAABA1LW6EVh33HGHPvzww6gca/r06br88sujcqyWeOutt/Tee+9Jkh588EG1bRu7u5ctX768wddLSko0evTomNUPAKjrwH+qGi8UgcqKQFSOAwAAAMRCqwuwYikjI8PZrqysbLT80aNHne3MzMxm1blv3z799Kc/lSQNHz5cw4cPb9ZxIpWfnx/T4wNonG3bqg5IVX5bbXzStpIqtUuVKqotdczy6UCZXx7L1uGjAZ2SlyqPh8G2ySy1TXT+fT3e5ozbAgAAAOKj1QVYI0aMUJ8+faJyrB496q4VUnvk04EDBxrd/+DBg852u3btmtWG2bNn68CBA2rbtq0mTpzYrGMASAxvbqzUE38+ove+8DtrFx1jS7Jk2bbSJKUoOEfcVplsSd/u69VLd2THt8GIi14D2sqbYslfdaxTBHtD05x8WnpU2wUAAABEU6sLsEaNGhWzY59yyinO9q5duxotX1RU5GwfH4ZForS0VH/+858lSaeeeqoWL14c0X6/+tWvnO3rrrtOubm5Ta4bQHyN/2OZfr6yZtTmCeGVJFny2LYyVXdxw9Bi3m/926/uE/dqxyz+f0826W19Gji8gz54Z1/wCcuS7Ho7SV3HFbnitsbvngsAAAC4pdUFWLHUs2dPeTweBQIBffbZZ/L7/fJ6vWHLb9iwwdk+7bTTmlyfXesLyscff6yPP/44ov3mz5/vbA8dOpQACzDcvL9XNBJeSaonvDreoQrpsqcP6J0fZ0e3gXDdlbd30faNZfrPzmA/CXWTsKOw7LqvD7o8Vz2/2byRwAAAAEA8sDBKFGVmZmrAgAGSgncV/Oijj8KWDQQCWrNmjfP4/PPPj3XzACQg27Y1Z0VFo+VCUwYb8/6X/ha3CeZJSfMo5+Q02VYwm7ItSwFJAR2XedqqE16lpFm64vbOuu6e7nFuMQAAANA0jMCKsssuu0zr1q2TJC1cuFBnn312veVWrFjhTCHs16+fOndu+tSNdu3aae3atRGVHTRokLMd6T4A3Ldme7W+2Ftzd7gGZoWlRng8S9LvVh/RD4aw3lGy+M9XFZpz92c1/cM6NqzKsmRL8qXauvi6jqo6GlD10YDKDlUrJz9NBedlq3tB824gAgAAAMQbAVaUXX311VqwYIG+/vprrVy5Uq+88oq+973v1SmzY8cOzZo1y3l85513hj3eVVddpeLiYknBqX+1gygAyW9PaQRrGalpC3b/exejsJJFwG/rZ054Vb+qSktrVxzQQ//vjPg1DAAAAIgyAqwoS0tL08MPP6wJEyaourpas2fP1urVqzVs2DClp6dr06ZNWrx4sUpLSyUFF1E/77zzXG41AFPlZUYWTUUWcwUVdAq/Nh8Sy1u/L45ovfb9u6tUerBKbbNSYt8oAAAAIAYIsGLgvPPO08yZM1VYWKjDhw9r5cqVWrly5QnlrrvuOj300EMutBBAohjcw6eeOR59uS8QHGYVJqyoUmRv6Lak2y5g+mCyWLW8JOKyL87ZoTGFTb9hCAAAAGACAqwYGT58uPr166c//OEPeu+991RcXKzKykrl5eVpwIABuuaaa8KujwUAIR6PpR9f1EYTXitvsFylpDQ1vpD72d0ZfZVMqqsiH3tXsutoDFsCAAAAxJZl25FMPgAis3v3bo0cOVKStHz5cuXn57vcIiA53PlKmX65uiaACPOu7bVtZSh8iJWZKhU9mRuL5sElD123XnYgsrIndUnTT+YVxLZBAAAAQIxEctd1AIDL5n8vU4tvb6vzunvCrNhuy29ZKpV0VMcyLltSQNJFvT2EV0moU7e0iMte8n3+oAAAAIDExRRCAEgQ1/RL1TX9UlXlt3W0Wio9astrV2vtzoDSU6vk9aSqW7a0t8xSRWWVDh316IJTU9Q+g7f6ZPVfD5+qmT/a2Gg5yyOdOaxDHFoEAAAAxAbfagAgwaR4LaV4pbZplqRUfStLkto4r5+SJ0mp7jQOcZV9UqoGDM3S+pUHGyx328M949QiAAAAIDaYQggAQAK76SenaOg1ubLqmVrqS7V0x7RT1Wdg+/g3DAAAAIgiRmABAJDgrry9q668vavW/32/Pv77AflSPRr+3Y7q1C3d7aYBAAAAUUGABQBAkhhwYQcNuJC1rgAAAJB8mEIIAAAAAAAAoxFgAQAAAAAAwGgEWAAAAAAAADAaARYAAAAAAACMRoAFAAAAAAAAoxFgAQAAAAAAwGgEWAAAAAAAADAaARYAAAAAAACMRoAFAAAAAAAAoxFgAQAAAAAAwGgEWAAAAAAAADAaARYAAAAAAACMRoAFAAAAAAAAoxFgAQAAAAAAwGgEWAAAAAAAADAaARYAAAAAAACMRoAFAAAAAAAAoxFgAQAAAAAAwGgEWAAAAAAAADAaARYAAAAAAACMRoAFAAAAAAAAoxFgAQAAAAAAwGgEWAAAAAAAADAaARYAAAAAAACMRoAFAAAAAAAAoxFgAQAAAAAAwGgEWAAAAAAAADAaARYAAAAAAACMRoAFAAAAAAAAoxFgAQAAAAAAwGgEWAAAAAAAADAaARYAAAAAAACMRoAFAAAAAAAAoxFgAQAAAAAAwGgEWAAAAAAAADAaARYAAAAAAACMRoAFAAAAAAAAoxFgAQAAAAAAwGgEWAAAAAAAADAaARYAAAAAAACMRoAFAAAAAAAAoxFgAQAAAAAAwGgEWAAAAAAAADAaARYAAAAAAACMRoAFAAAAAAAAoxFgAQAAAAAAwGgEWAAAAAAAADCaz+0GAACib3+pX//nZwf15d6AZEuWpLSaH4+kQE05jyV5bVvegOSRLUuSt6aAXbOf17blUXBbNf/1KFheds1xJFm27bxuSZItpXik9hmW8nN9uu6iDFVWBvTuB0dUtLta1QHJX2WrqtKWbQfrTkmRUrxSaoolr2XrSLlUVWXLtuWcR/DHrmmPLaumDbKD7fB5pYBfzgl4al60bcljBw9g2cFzt6zgOdgB22mzZMtrBV8PhH5Rtc4tdK6h34+c/9rHzj3UtpryXq/k8yl4rv5j/05WzbGOnU+wXilUzq51vJp9nPO1ZdW00+uV7OrgC2ltPOrWO0PfndBNWTmpAgAAAJKBZdu23XgxIDK7d+/WyJEjJUnLly9Xfn6+yy0CWhfbtjVk5n5t2l3PW7ttK1VSG6lOKOIJ2EqTLev4qMS2lWbb8h5XPsR53rblPSG8CR2jJvSpCcE89ZapHZDZsmxb3lphVe1jScHgLOxx7HqGFtf6mLNqtaO+MsHXw52LLSsQ/H3ouNft487VauAYdfa1a+q0TwyqnDoVLiCrv0zt43fvk66x03vJl8KAawAAACQ2rmgBIImEDa8kybJUaVmqqPWUx7aVJp0YXik4RNen+sMrSfIrGNx47DCBkoJP2qoZhRS2jKWAjoVAXjtM0FWT0gSOf772cTzWia9bVvBHkl27rnrKWOHCsZoydnB41ImjomoOYdeMnmroGKpdv7NP+HMKtbne8KpWGbue4+/47Ij+732bFQjwtyoAAAAkNgIsAEgSnxZVhQ+vaqm0LAVnp9lKtcOEIrat1AgG6PprwquGWDo2ail8IcsJuhr9YLLqCalqCVj1BFQ1+zW4vx1ZO/1W+MTJiuD3oeP29wTCjL5qoppZlieEWLt3HNWGfx6MQg0AAACAewiwACBJjHu+NOKylWo4LGpo5FUdoZE/DfBGOPgnNJorEg0FWJG0yQl7aom0bjtcQCaFnwoYpg2KtLxlKdBAcBYqY9faru2d57+OsFUAAACAmQiwACBJbN7dYKxTh18NfwBEGuZIjYRJjdRTh1XfRMbwZRtqYYPT8mr2P+GpRmOvxutuSvtD5ZsUeDVTya6jLdgbAAAAcB8BFgAAAAAAAIxGgAUASaJ3fuRv6V41PHKq0elqtTRWa8Tjwmw78lFGjUy9sxqdQ3higYgn/zVQd1PaHyof6T4tWScrr3NaC/YGAAAA3EeABQBJ4tmb20ZcNlXBYClcuFStCIOVCNZw8keYvFiKPDhr8MMrgjbVN3Uv0rqtcAvf1xyjSYFUpOVtu/FpnbXP+7iyl93cKcJWAQAAAGYiwAKAJPHNLikqyG88hEkN3W3PslTZwB37KiMIdLyN3BFQCo5s8jd2oJrwpaFQrXbZhtfvCn9nxQb3tyJrp7eBIMmO4PdxfLgU8EQeejXECeVqjh/6HXTq0UZnnJsVhRoAAAAA9xBgAUASWT2pg07vFCZ4sm2l2bba1HoqYFk6Kqm+yXvVangkllfHRk0FwpWrCZMCCi4cX3+ZYKAUHJEUHLFV7/Fq5tuF/eCybXkC9YRTtn0s1KldVz1lbDV0LraswLGpfyc0zQ5NXWz4GKpdv93IdEfbdtocdrphTZn6wqvufdJ1z5xvyONpyQREAAAAwH0+txsAAIgey7K0elKO9pf5NeJnB/VFScAJkdIkpVnBAKgmh5HXkjweyQ5YsmpWgfJ6JAV3k9+y6gZMNf8NPradRMWuKWer9kggKcUjtc+w1CkvRdcOy1BlZUDvfnBERf+plt8vVVfZqqqU7JoAJjVF8nktpaZY8npsHSmXqiptZ9BS6N59xwI3+1gAVPNfj1cK+FWTKAXP0VbwGKEcJ7RtWcFzCNUfOozHCv4EQsOpQg2oGS1l2bYC9rHzDIVHx44RbGPod+X1ST6vVF0lBapr/XvpWHAX4vEEj2BXh450rLBd6zxl27Jq/v28XsmuDjYkrY1H3fpk6Hvju6l9TqoAAACAZECABQBJqEOmVx8+nON2M+p19fDI1+oCAAAAAIkphAAAAAAAADAcARYAAAAAAACMRoAFAAAAAAAAo7EGFgAAAIC42rVrV9SO1blz56gdCwBgLkZgAQAAAAAAwGgEWAAAAAAAADAaUwgBAAAAABGL1hRQpn8CaAoCLACtCmtuRI6LUwBAa8I1AgCYjQALAIAW4AsPkPhM/P/YxDYBAOAmAiwAQMLgC138RfN3biL6AQAgVrhuAaKLAAtAzCT7h7aJ55fsYQPQGpj4/7GJ78EAAKB1IcACkoSJYQpgMhNDAtYdA4DkYOJnDAAkOgIsAAkh2S8Ek/38AFMRGiYu3jfjj9854B4+rwACLKBZkv0CLtnPDwCijfdNhJjYF0xsUzQl+/khMgQ8QPIjwEKrwcUNAADu4/MYAAA0BwEWAAAAGkToBKC14P0OMBcBFgAAAAAg7giLADQFARYAAAAAAK0Ady5HIiPAAgDEFH9djT9+5wAAAEg2HrcbAAAAAAAAADSEAAsAAAAAAABGI8ACAAAAAACA0QiwAAAAAAAAYDQCLAAAAAAAABiNAAsAAAAAAABGI8ACAAAAAACA0XxuNwDJxe/3O9slJSUutuREe/bscbsJAAAAAJAUvF6v202oV25urnw+oo5kxL8qomr//v3O9ujRo11syYkqKyvdbgIAAAAAJIXU1FS3m1Cv5cuXKz8/3+1mIAYIsNBqmPoGCwAAAAAAGmbZtm273Qgkj4qKCm3dulWS1KFDByOHlZaUlDijwxYsWKC8vDyXWwSciH4K09FHYTr6KBIB/RSmS8Q+yhTC5MW/KqKqTZs26tu3r9vNiFheXh7DS2E8+ilMRx+F6eijSAT0U5iOPgq3cRdCAAAAAAAAGI0ACwAAAAAAAEYjwAIAAAAAAIDRCLAAAAAAAABgNAIsAAAAAAAAGI0ACwAAAAAAAEYjwAIAAAAAAIDRLNu2bbcbAQAAAAAAAITDCCwAAAAAAAAYjQALAAAAAAAARiPAAgAAAAAAgNEIsAAAAAAAAGA0AiwAAAAAAAAYjQALAAAAAAAARiPAAgAAAAAAgNEIsAAAAAAAAGA0AiwAAAAAAAAYzed2A4Cm+tvf/qY33nhDGzZs0N69e5WZmalu3brp4osv1vXXX6+2bdsmRZ1IXPHqL2VlZVqzZo3Wrl2rzz77TDt27NDhw4fVpk0b5eXlqW/fvrriiis0ZMgQWZYVlTqRPEx4X3vssce0bNky5/GYMWM0duzYmNeLxOBWH920aZPefvttvf/++/rPf/6jsrIyZWdnKy8vT9/85jd19tlna/jw4fJ6vTGpH4kj3n10165dWrJkidauXatt27aptLRUqampysnJUe/evTV8+HBddtll8vn4itfa+f1+ffnll9q4caPzs3nzZh09elSSdOWVV+qxxx6LSd0mXF8geVm2bdtuNwKIRHl5uaZMmaL33nsvbJn8/Hw98cQT6tevX8LWicQVz/6ycOFCPfvss86FSEPOOussTZ06VZ06dWpRnUgOpryvrVq1ShMmTKjzHAEWJPf6aGlpqebMmaNly5apscvjd999V+3atYta3UgsbvTRhQsXat68eaqsrGywXI8ePTRr1iz16tUrKvUiMT3wwAN69913w74eiwDLlOsLJDcCLCQEv9+ve++9V//4xz8kSbm5ubr22mvVs2dPHTp0SG+//bbWr18vSWrfvr1+/etfq2fPnglXJxJXvPvLjBkz9Mc//lGS1LFjR5177rkqKChQTk6OKisr9cknn+jNN99UeXm5JKlLly567rnnlJOT08IzRSIz5X2ttLRUo0aN0u7du5Wenq4jR45IIsCCe3304MGDuueee7RhwwZJwffV4cOH6xvf+Ibatm2r8vJy7dy5U++//742btyov/71rwRYrZQbffTll1/Wk08+6Tzu37+/hg0bpk6dOqm0tFRffPGFli1b5nzmZ2dn66WXXlJeXl6L6kXiuv/++/W3v/3NeZyVlaWsrCzt2LFDUvQDLFOuL5D8CLCQEBYtWqSZM2dKkk499VQ9++yzys3NrVPm6aef1sKFCyUFR5z87//+b8LVicQV7/4yc+ZMFRUV6ZZbbtG5554rj+fEJQ2Li4t19913a/v27ZKkq666So8++miz60TiM+V9bfr06XrttdeUn5+vSy65RM8//7wkAiy410fvuecerV69WpJ0yy23aNy4cUpLS6u37J49e9ShQwemabVS8e6jFRUVuvzyy1VWViZJmjJliq699toTyu3fv1/jxo3Tli1bJEk33XST7rvvvmbXi8T2m9/8RuXl5SooKNDpp5+uLl26aOnSpXr88cclRT/AMuX6AsmPRdxhPL/fX+cNrrCw8IQ3RCl48dm7d29J0rp167RmzZqEqhOJy43+8t///d+aO3euBg8eXG94JUknn3yyczEhSX/6059UUVHR7DqR2Ex5X/vggw+0ePFiSdJDDz2kjIyMqB4ficutPrp06VInvLrhhhv04x//OGx4JUknnXQS4VUr5UYfXb9+vRNenXHGGfWGV5LUoUMH3X333c7jdevWNbtOJL7bb79dd999ty655BJ16dIlpnWZcn2B1oEAC8Zbt26dSkpKJEkDBw5UQUFBveW8Xq++//3vO4/ffvvthKoTicuN/pKVlRVRud69e6tHjx6Sgn/F3blzZ7PrRGIz4X2toqJC06ZNk23buvTSSzV06NCoHRuJz60+umDBAklSRkZGnQAAOJ4bfXT//v3Odvfu3Rss261bN2c7NJ0QiDUTri/QehBgwXirVq1yti+44IIGy55//vnOdmgOdqLUicRlen/JzMx0tiNZ9B3JyYR+OnfuXBUVFSkrK0s/+clPonZcJAc3+uhHH32kbdu2SZIuuugi7o6FBrnRRzt06OBsh9YvCqf266eeemqz6wSawoTrC7QeBFgw3tatW53tvn37Nlg2Ly9P+fn5kqS9e/fW+auV6XUicZncX6qqqupc0HInwtbL7X66fv16vfLKK5KkCRMm1Du9AK2bG330ww8/PKHOv/71rxo/frwuv/xyDRkyRFdccYUmTJig119/XdXV1c2qB8nBjT565plnKjs7W5K0YcMGZwr28fbv369f/OIXkiSPx6Obb765WfUBTeX29QVaFybww3ihBaglqXPnzo2W79Kli3bv3u3sW/svVybXicRlcn956623VFpaKkkqKCjgjkStmJv99OjRoyosLFQgENC5556rq6++utnHQvJyo49u3LjR2c7Nza331vMlJSUqKSnRqlWr9MILL2jOnDkxX1MGZnKjj6alpWnSpEmaPHmy/H6/pk2bpqVLl2rYsGHKz89XWVmZtm7dquXLl6usrEwZGRmaMmWKzjzzzCbXBTSHydfBSD4EWDDe4cOHne3QX6AaUnttoNr7ml4nEpep/WX//v36+c9/7jz+4Q9/GLO6YD43++n8+fO1fft2paWlafLkyS06FpKXG300tG6LdKyfpqSkaOTIkTrzzDPl8/m0efNmLVmyRAcPHtSWLVt05513auHChRGvRYjk4db76IgRIzRv3jzNmjVLX3zxhdavX6/169fXKePz+XT77bfr+uuvZ7Q14srU62AkJ6YQwni1F6FMTU1ttHztOweF7tqSCHUicZnYX6qqqvTggw9q3759kqSLL75Yw4cPj0ldSAxu9dN///vfeuGFFyRJY8eOVdeuXZt9LCQ3N/rooUOHnO3t27erffv2eu655zRlyhRdeeWVuuKKKzR+/Hi9/PLLzppCxcXFzlQttC5uft4PHDhQDz74oPr06VPv69XV1Xr11Vf1/PPPc8dhxJWJ18FIXgRYAJBkAoGACgsLnVtod+3aVY888ojLrUJrVFVVpcLCQvn9fhUUFLAmC4xj23adxxMmTKj3Dlp5eXmaNm2a83jZsmXO9Gwg1g4cOKBx48bpzjvvVHFxse677z4tXrxYa9as0YoVK/Tss8/qggsu0OHDh/Xiiy9q7NixOnDggNvNBoCoI8CC8TIyMpztysrKRsvXvsta7buvmV4nEpdJ/cW2bc2cOVNvvvmmpOCi7fPmzVP79u2jWg8Sjxv99Ne//rW2bt0qr9erKVOmyOv1Nus4aB3c/rxPT0/Xt7/97bBle/furX79+jntO34KF5KfG320oqJCP/rRj7R27Vq1b99ev/3tb3XTTTepa9eu8vl8atu2rc455xw988wz+u53vyspOPL1ySefbFZ9QFOZdB2M5EeABeO1a9fO2Y7kr0kHDx6sd1/T60TiMqW/2LatJ554Qq+99pokKT8/X/Pnz49oQU0kv3j3082bN+u3v/2tJOmmm26qd1QLUJsb76W1w/1evXopJSWlwfKnn366s/3VV181q04kLjf66Kuvvqpt27ZJkm655RZ17949bNnx48c79fzpT3+qs8YbECumXAejdWARdxivR48eKioqkiTt2rWr0S/jobKhfROlTiQuE/qLbduaNWuWFi1aJEnq2LGj5s+fz3pDcMS7ny5dulTV1dXyeDzy+Xz61a9+VW+50FTX0Hao3CmnnKJLLrmkyfUicbn1ef/Pf/5TktS2bdtGy9cuw9otrY8bfXTlypXO9uDBgxssm56erv79+2vVqlUKBALasGGDhg0b1qx6gUiZcB2M1oMAC8Y77bTT9I9//ENScEj0oEGDwpbdu3evc1vWnJycZt+W1Y06kbjc7i+h8OoPf/iDJOmkk07S/Pnz1a1btxYfG8nDrX4aCAT03HPPRVR27dq1Wrt2rSTpoosuIsBqZdzoo71793a2I1nTqnaZSAIvJBc3+mjtUVSR9LnaI1pqL64NxIrb18FoXZhCCOOdf/75znbozTGcVatWOdsXXHBBQtWJxOVmfzk+vMrLy9P8+fMbnGKA1on3NZjOrc97y7IkSVu2bFFVVVWD5Tdu3OhsM3Kg9XGjj9ZeXyj0xb8hxcXFznZ2dnaz6wUixfUF4okAC8YbOHCgcnNzJUn/+te/tGnTpnrL+f1+vfTSS87jyy67LKHqROJys7/Mnj3bCa9yc3M1f/58vlShXvHup/fff78zoqqhnzFjxjj7jBkzxnl+zpw5zaoXicuN99L8/HwNHDhQknTkyBG98cYbYctu3rxZn3zyiaTgwsMDBgxodr1ITG700V69ejnboRu0hLNz5059+umnkiSPx1NnzTYgVvjehHgiwILxvF5vnS84jzzyiPbt23dCublz52rz5s2SpAEDBmjIkCH1Hm/p0qUaNGiQBg0apDvuuCMudSK5udFHpWB49eqrr0oKhle//OUvdcopp7TgTJDM3OqnQKTc6qN33XWXs/3MM8/U++Vr7969mjJlivN41KhRatOmTeMnhaTiRh+94oor6pRfvHhxveVKSkr00EMPye/3S5IuvPBCZWVlRXReQDh8b4JpWAMLCeHaa6/Vu+++q/fff19ffPGFbrzxRl133XXq2bOnDh06pLffflsfffSRpODc/8mTJydknUhc8e4v8+bN0yuvvCJJsixLN954o7Zt2+bcqSicgoICderUqUV1I3HxvgbTudFH+/fvr9GjR2vBggU6dOiQbrvtNl155ZUaMGCAfD6fNm/erCVLljh3zjrjjDP0wx/+sMX1IjHFu48OHjxYI0aM0F/+8hfZtq1p06bpjTfe0EUXXaT8/HxVVFRo48aNeuONN3T48GFJUlZWlu69996WnioSWFFRkZYsWVLnuc8//9zZ/uyzzzRv3rw6r59zzjk655xzmlUf1xeIFwIsJASfz6fZs2drypQpWrlypfbu3VvvHa3y8/M1Y8YMnXbaaQlZJxJXvPtL6CJACq6DNXfu3Ij2e/TRR3XVVVe1qG4kLt7XYDq3+ug999wjj8ej3/3ud6qqqtJrr72m11577YRyQ4YM0fTp05WWlhaVepF43OijU6dOVWZmpl5//XVJ0ocffqgPP/yw3rI9evTQjBkzuJFLK1dcXKzf/OY3YV///PPP6wRaUnAkVXMDLK4vEC8EWEgYmZmZeuqpp7RixQotX75cGzZs0P79+5WRkaGuXbtq+PDh+s53vhPVuwK5UScSF/0FiYB+CtO51UfvuusuXXrppVqyZInef/997dmzR9XV1erQoYP69++vkSNHsugwJMW/j6ampuqRRx7RqFGjtGzZMq1fv15FRUUqLS1VSkqKcnJyVFBQoIsvvliXXnqpUlJSolIv0BRcXyAeLNu2bbcbAQAAAAAAAITDIu4AAAAAAAAwGgEWAAAAAAAAjEaABQAAAAAAAKMRYAEAAAAAAMBoBFgAAAAAAAAwGgEWAAAAAAAAjEaABQAAAAAAAKMRYAEAAAAAAMBoBFgAAAAAAAAwGgEWAAAAAAAAjEaABQAAAAAAAKMRYAEAAAAAAMBoBFgAAAAAAAAwGgEWAAAAAAAAjEaABQAAAAAAAKMRYAEAAAAAAMBoBFgAAAAAAAAwGgEWAAAAAAAAjEaABQAAAAAAAKMRYAEAAAAAAMBoBFgAAAAAAAAwGgEWAAAAAAAAjEaABQAAAAAAAKMRYAEAAAAAAMBoBFgAAABxcOjQIV111VUaNGiQBg0apBtuuEEVFRWN7rds2TJnn0GDBmnBggVxaC0AAIBZCLAAAADioH379po2bZq8Xq8kadu2bZo1a1aD+2zfvl2zZ892Hg8ZMkQ/+MEPYtpOAAAAExFgAQAAxMmAAQM0btw45/HSpUv11ltv1Vu2srJSkyZNUnl5uSQpNzdXjz/+uCzLiktbAQAATEKABQAAEEejR4/W4MGDncczZ87Uzp07Tyj39NNPa/PmzZIkj8ejwsJC5eTkxK2dAAAAJiHAAgAAiCPLslRYWKjc3FxJUllZmSZNmqSqqiqnzIoVK/TKK684j0ePHq3zzjsv7m0FAAAwBQEWAABAnOXk5Gjq1KnyeIKXYps2bdIzzzwjSfr66681depUp2z//v01duxYV9oJAABgCgIsAAAAF5x77rkaPXq08/ill17Su+++qylTpujgwYOSpHbt2mn69Ony+XxuNRMAAMAIlm3bttuNAAAAaI2qq6s1duxYrV+/XlJwratAIOC8PmvWLI0YMcKt5gEAABiDEVgAAAAu8fl8mj59urKysiSpTnj1ne98h/AKAACgBgEWAACAizp16qRbb721znMnn3yy7rvvPpdaBAAAYB4CLAAAABft27dPL774Yp3ndu/erU8++cSlFgEAAJiHAAsAAMAltm3r0Ucf1d69e+s8HwgE9PDDD+vAgQPuNAwAAMAwBFgAAAAu+f3vf6/Vq1c7j8eMGaM2bdpIkvbs2aPHHntM3G8HAACAAAsAAMAVn376qebNm+c8HjlypMaOHasHHnjAee7vf/+7XnjhBTeaBwAAYBQCLAAAgDgrLS3V//zP/6i6ulqS1L17d02cOFGSdM011+jyyy93ys6dO1ebNm1ypZ0AAACmIMACAACIsxkzZqioqEiSlJKSohkzZigjI8N5fdKkSerSpYskqaqqSpMnT1Z5ebkrbQUAADABARYAAEAcLV68WO+8847zePz48SooKKhTpm3btpoxY4Z8Pp8kaceOHZo5c2Zc2wkAAGASAiwAAIA4+fLLL/XTn/7UeTx06FDdeOON9Zbt27ev7rrrLufxm2++qWXLlsW8jQAAACaybG5tAwAAEHNHjx7V6NGjtWXLFklSx44d9cILLyg7OzvsPrZta/z48c6dCjMyMvT73/9ePXr0iEeTAQAAjMEILAAAgDh46qmnnPDK4/GosLCwwfBKkizL0uOPP67c3FxJUnl5uSZPnqzKyspYNxcAAMAojMACAAAAAACA0RiBBQAAAAAAAKMRYAEAAAAAAMBoBFgAAAAAAAAwGgEWAAAAAAAAjEaABQAAAAAAAKMRYAEAAAAAAMBoBFgAAAAAAAAwGgEWAAAAAAAAjEaABQAAAAAAAKMRYAEAAAAAAMBoBFgAAAAAAAAwGgEWAAAAAAAAjEaABQAAAAAAAKMRYAEAAAAAAMBoBFgAAAAAAAAwGgEWAAAAAAAAjEaABQAAAAAAAKMRYAEAAAAAAMBoBFgAAAAAAAAwGgEWAAAAAAAAjEaABQAAAAAAAKMRYAEAAAAAAMBo/x8m3dNDGtgCCgAAAABJRU5ErkJggg==",
+ "text/plain": [
+ "