{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Adaptive distances" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext", "tags": [] }, "source": [ "In this notebook, we illustrate the use of adaptive distances implemented in pyABC. \"Adaptive\" means that the distance function is not pre-defined (e.g. after pre-processing), but adjusts over time during the ABC run. Here, we provide adaptation for data scale normalization and outliers. General adaptation concepts can be easily implemented via the :class:`Distance ` class' ``initialize()`` and ``update()`` methods. Adaptive distances are implemented as extensions of the :class:`PNormDistance `." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "It uses a distance between simulations $s$ and observed data $s_\\text{obs}$ (or summary statistics representations thereof)\n", "\n", "$$d(s, s_\\text{obs}) = \\left [\\sum_{j} \\left| w_j ( s_j-s_{\\text{obs},j} ) \\right|^{p} \\right ]^{1/p}$$\n", "\n", "with weights $w$ and exponent $p\\geq 1$. E.g., $p=2$ gives a weighted Euclidean distance, $p=1$ a Manhattan distance, $p=\\inf$ a maximum distance." ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext", "tags": [] }, "source": [ "The notebook can be downloaded :download:`here `." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Scale normalization" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext", "tags": [] }, "source": [ "The :class:`AdaptivePNormDistance ` offers methods to, in each generation (or, if desired, only in selected generations), recalculate the weights in order to rescale the impact of each coordinate. This can be useful if different summary statistics vary on different scales, in which case statistics with higher sample variance would dominate the distance value and thus the analysis, while proper rescaling ensures that each statistic has a roughly similar impact." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The concept is based on this seminal work by [Dennis Prangle 2017](https://doi.org/10.1214/16-BA1002):\n", "\n", " Dennis Prangle.\n", " Adapting the ABC distance function.\n", " Bayesian Analysis 12.1 (2017): 289-309." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "For illustration, we consider a simple Gaussian model:" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "# install if not done yet\n", "!pip install pyabc --quiet" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import logging\n", "import tempfile\n", "\n", "import matplotlib.pyplot as pyplot\n", "import numpy as np\n", "\n", "import pyabc" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "pyabc.settings.set_figure_params(\"pyabc\") # for beautified plots\n", "\n", "\n", "# for debugging\n", "df_logger = logging.getLogger(\"ABC.Distance\")\n", "df_logger.setLevel(logging.DEBUG)\n", "\n", "\n", "# model definition\n", "def model(p):\n", " return {\n", " \"s1\": p[\"theta\"] + 1 + 0.1 * np.random.normal(),\n", " \"s2\": 2 + 10 * np.random.normal(),\n", " }\n", "\n", "\n", "# true model parameter\n", "theta_true = 3\n", "\n", "# observed summary statistics\n", "observation = {\"s1\": theta_true + 1, \"s2\": 2}\n", "\n", "# prior distribution\n", "prior = pyabc.Distribution(theta=pyabc.RV(\"uniform\", 0, 10))\n", "\n", "# database\n", "db_path = pyabc.create_sqlite_db_id(file_=\"adaptive_distance.db\")\n", "\n", "\n", "def plot_history(history):\n", " \"\"\"Plot 1d posteriors over time.\"\"\"\n", " fig, ax = pyplot.subplots()\n", " for t in range(history.max_t + 1):\n", " df, w = history.get_distribution(m=0, t=t)\n", " pyabc.visualization.plot_kde_1d(\n", " df,\n", " w,\n", " xmin=0,\n", " xmax=10,\n", " x=\"theta\",\n", " ax=ax,\n", " label=f\"PDF t={t}\",\n", " refval={\"theta\": theta_true},\n", " refval_color=\"grey\",\n", " )\n", " ax.legend()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Summary statistic s2 has a high variance compared to summary statistic s1. In addition, s1 is informative about the model parameters $\\theta$, s2 is not. We expect that the proposal distribution for $\\theta$ iteratively centers around the true value $\\theta=3$. Thus, the variability for the sampled s1 decreases iteratively, while the variability of the sampled s2 stays approximately constant. If both summary statistics are weighted similarly in the calculation of the distance between sample and observation, there is hence an undesirable high impact of s2, so that convergence can be slowed down. In contrast, if we weight s1 higher, we may hope that our estimation of $\\theta$ is improved." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "These informal expectations being stated, let us continue with the implementation. First, we consider a non-adaptive Euclidean distance with uniform weights (i.e. $w\\equiv 1$):" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "ABC.Sampler INFO: Parallelize sampling on 4 processes.\n", "ABC.History INFO: Start \n", "ABC INFO: Calibration sample t = -1.\n", "ABC INFO: t: 0, eps: 7.03999965e+00.\n", "ABC INFO: Accepted: 100 / 220 = 4.5455e-01, ESS: 1.0000e+02.\n", "ABC INFO: t: 1, eps: 4.51503638e+00.\n", "ABC INFO: Accepted: 100 / 425 = 2.3529e-01, ESS: 9.4379e+01.\n", "ABC INFO: t: 2, eps: 3.06944258e+00.\n", "ABC INFO: Accepted: 100 / 613 = 1.6313e-01, ESS: 9.6825e+01.\n", "ABC INFO: t: 3, eps: 2.06373619e+00.\n", "ABC INFO: Accepted: 100 / 926 = 1.0799e-01, ESS: 9.8871e+01.\n", "ABC INFO: t: 4, eps: 1.59843112e+00.\n", "ABC INFO: Accepted: 100 / 1282 = 7.8003e-02, ESS: 9.9531e+01.\n", "ABC INFO: t: 5, eps: 1.03245438e+00.\n", "ABC INFO: Accepted: 100 / 1856 = 5.3879e-02, ESS: 9.8850e+01.\n", "ABC INFO: t: 6, eps: 7.35771496e-01.\n", "ABC INFO: Accepted: 100 / 3165 = 3.1596e-02, ESS: 9.7844e+01.\n", "ABC INFO: Stop: Maximum number of generations.\n", "ABC.History INFO: Done \n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "distance = pyabc.PNormDistance(p=2)\n", "\n", "abc = pyabc.ABCSMC(model, prior, distance)\n", "\n", "abc.new(db_path, observation)\n", "\n", "h_uni = abc.run(max_nr_populations=7)\n", "\n", "plot_history(h_uni)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Second, we consider an adaptive Euclidean distance:" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "ABC.Distance DEBUG: Fit scale ixs: \n", "ABC.Sampler INFO: Parallelize sampling on 4 processes.\n", "ABC.History INFO: Start \n", "ABC INFO: Calibration sample t = -1.\n", "ABC.Distance DEBUG: Scale weights[0] = {'s1': 3.5668e-01, 's2': 1.3377e-01}\n", "ABC.Population INFO: Recording also rejected particles: True\n", "ABC INFO: t: 0, eps: 1.72412322e+00.\n", "ABC INFO: Accepted: 100 / 170 = 5.8824e-01, ESS: 1.0000e+02.\n", "ABC.Distance DEBUG: Scale weights[1] = {'s1': 3.5670e-01, 's2': 1.6313e-01}\n", "ABC INFO: t: 1, eps: 1.18844229e+00.\n", "ABC INFO: Accepted: 100 / 266 = 3.7594e-01, ESS: 9.7254e+01.\n", "ABC.Distance DEBUG: Scale weights[2] = {'s1': 5.6878e-01, 's2': 1.5639e-01}\n", "ABC INFO: t: 2, eps: 1.01577127e+00.\n", "ABC INFO: Accepted: 100 / 406 = 2.4631e-01, ESS: 9.9213e+01.\n", "ABC.Distance DEBUG: Scale weights[3] = {'s1': 7.9713e-01, 's2': 1.4391e-01}\n", "ABC INFO: t: 3, eps: 8.16917955e-01.\n", "ABC INFO: Accepted: 100 / 430 = 2.3256e-01, ESS: 9.8489e+01.\n", "ABC.Distance DEBUG: Scale weights[4] = {'s1': 1.3874e+00, 's2': 1.4831e-01}\n", "ABC INFO: t: 4, eps: 7.41906064e-01.\n", "ABC INFO: Accepted: 100 / 488 = 2.0492e-01, ESS: 9.8404e+01.\n", "ABC.Distance DEBUG: Scale weights[5] = {'s1': 2.4945e+00, 's2': 1.4836e-01}\n", "ABC INFO: t: 5, eps: 6.54733733e-01.\n", "ABC INFO: Accepted: 100 / 775 = 1.2903e-01, ESS: 9.8508e+01.\n", "ABC.Distance DEBUG: Scale weights[6] = {'s1': 3.7143e+00, 's2': 1.5201e-01}\n", "ABC INFO: t: 6, eps: 6.09213308e-01.\n", "ABC INFO: Accepted: 100 / 736 = 1.3587e-01, ESS: 9.1398e+01.\n", "ABC.Distance DEBUG: Scale weights[7] = {'s1': 7.6311e+00, 's2': 1.4161e-01}\n", "ABC INFO: Stop: Maximum number of generations.\n", "ABC.History INFO: Done \n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "scale_log_file = tempfile.mkstemp(suffix=\".json\")[1]\n", "\n", "distance_adaptive = pyabc.AdaptivePNormDistance(\n", " p=2,\n", " scale_function=pyabc.distance.mad, # method by which to scale\n", " scale_log_file=scale_log_file,\n", ")\n", "\n", "abc = pyabc.ABCSMC(model, prior, distance_adaptive)\n", "\n", "abc.new(db_path, observation)\n", "\n", "h_ada = abc.run(max_nr_populations=7)\n", "\n", "plot_history(h_ada)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In the debug output of `abc.run` above, it can be seen how the weights evolve over time. In the posterior plot, we observe how, compared to the non-adaptive distance, the densitities tend to be narrower around the true parameter $\\theta=3$. In addition, despite the better convergence, the required number of samples in total is lower, as can be seen from the below sample numbers plot, as not so much time was wasted trying to match an uninformative summary statistic." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "histories = [h_uni, h_ada]\n", "labels = [\"Uniform weights\", \"Adaptive weights\"]\n", "pyabc.visualization.plot_sample_numbers(histories, labels)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Indeed, the acceptance rates for the adaptive distance function are consistently higher:" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "pyabc.visualization.plot_acceptance_rates_trajectory(histories, labels)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The log file stores the weights of each generation, allowing for further analysis. We can see that indeed ss1 is assigned a higher weight than ss2, the ratio increasing over time." ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "ts = list(range(h_ada.max_t + 1))\n", "labels = [f\"t={t}\" for t in ts]\n", "\n", "pyabc.visualization.plot_distance_weights(\n", " log_files=scale_log_file,\n", " ts=ts,\n", " labels=labels,\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In detail, the adaptive distance feature works as follows: In each iteration of the ABCSMC run, after having obtained the desired number of accepted particles (and once at the beginning using a calibration sample from the prior), the method `update()` is called. It is given a set of summary statistics which can be used to e.g. compute weights for the distance measure in the next iteration. In order to avoid bias, via `configure_sampler()`, the distance function tells the sampler to not only record accepted particles, but all that were generated during the sampling process.\n", "\n", "Above, the acceptance step uses in each generation only the distance based on the current weights. Sometimes it may be preferable to have nested acceptance regions, by also checking all previous acceptance criteria. This can be realized by passing a `acceptor=pyabc.UniformAcceptor(use_complete_history=True)` to the `ABCSMC` class.\n", "\n", "For scale normalization, we used as a measure of sample variability above the median absolute deviation (MAD), a robust alternative to the standard deviation. Other measures are possible and implemented in `pyabc.distance.scale`." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Robustify against model error\n", "\n", "One problem not addressed by the previous adaptive weighting scheme is that if there are outliers in the data, things can go wrong. We informally denote by an outlier a data point corrupted by large errors that cannot be expected under the given experimental setup, already accounting for measurement\n", "noise. To circumvent this, robust methods have been developed that are less sensitive to outliers and can even detect and down-weight such points. The methods presented in the following are based on [this work](https://www.biorxiv.org/content/10.1101/2021.07.29.454327v1):\n", "\n", " Yannik Schaelte, Emad Alamoudi, Jan Hasenauer.\n", " Robust adaptive distance functions for approximate Bayesian inference on outlier-corrupted data.\n", " bioRxiv 2021.07.29.454327; doi: https://doi.org/10.1101/2021.07.29.454327.\n", " \n", "As an example, consider the previous model, but with an additional summary statistic of small variance, whose observation however is an outlier, i.e. a deviant value that cannot be explained under the model. This is a scenario that can typically occur in data where some coordinates hardly depend on the inputs, some may however be subject to errors not accounted for by the model due to e.g. external perturbations, or human errors such as wrong or missing labels. Also other outlier scenarios are possible, for a more extensive discussion see the publication." ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "import logging\n", "import os\n", "import tempfile\n", "\n", "import matplotlib.pyplot as pyplot\n", "import numpy as np\n", "\n", "import pyabc.visualization\n", "\n", "# for debugging\n", "df_logger = logging.getLogger(\"Distance\")\n", "df_logger.setLevel(logging.DEBUG)\n", "\n", "\n", "# model definition\n", "def model(p):\n", " return {\n", " \"s1\": p[\"theta\"] + 1 + 0.1 * np.random.normal(),\n", " \"s2\": 2 + 10 * np.random.normal(),\n", " \"s3\": 2 + 0.1 * np.random.normal(),\n", " }\n", "\n", "\n", "# true model parameter\n", "theta_true = 3\n", "\n", "# observed summary statistics\n", "observation = {\"s1\": theta_true + 1, \"s2\": 2, \"s3\": 5}\n", "\n", "# prior distribution\n", "prior = pyabc.Distribution(theta=pyabc.RV(\"uniform\", 0, 10))\n", "\n", "# database\n", "db_path = pyabc.create_sqlite_db_id(file_=\"adaptive_distance.db\")\n", "\n", "\n", "def plot_history(history):\n", " \"\"\"Plot 1d posteriors over time.\"\"\"\n", " fig, ax = pyplot.subplots()\n", " for t in range(history.max_t + 1):\n", " df, w = history.get_distribution(m=0, t=t)\n", " pyabc.visualization.plot_kde_1d(\n", " df,\n", " w,\n", " xmin=0,\n", " xmax=10,\n", " x=\"theta\",\n", " ax=ax,\n", " label=f\"PDF t={t}\",\n", " refval={\"theta\": theta_true},\n", " refval_color=\"grey\",\n", " )\n", " ax.legend()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "An distance with uniform weights suffers from the same problems as before." ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "ABC.Sampler INFO: Parallelize sampling on 4 processes.\n", "ABC.History INFO: Start \n", "ABC INFO: Calibration sample t = -1.\n", "ABC INFO: t: 0, eps: 8.83875673e+00.\n", "ABC INFO: Accepted: 100 / 176 = 5.6818e-01, ESS: 1.0000e+02.\n", "ABC INFO: t: 1, eps: 5.92014462e+00.\n", "ABC INFO: Accepted: 100 / 292 = 3.4247e-01, ESS: 9.4200e+01.\n", "ABC INFO: t: 2, eps: 4.52756560e+00.\n", "ABC INFO: Accepted: 100 / 537 = 1.8622e-01, ESS: 9.7015e+01.\n", "ABC INFO: t: 3, eps: 3.79064700e+00.\n", "ABC INFO: Accepted: 100 / 872 = 1.1468e-01, ESS: 9.7878e+01.\n", "ABC INFO: t: 4, eps: 3.46070308e+00.\n", "ABC INFO: Accepted: 100 / 1222 = 8.1833e-02, ESS: 9.8035e+01.\n", "ABC INFO: t: 5, eps: 3.20079151e+00.\n", "ABC INFO: Accepted: 100 / 1404 = 7.1225e-02, ESS: 9.7536e+01.\n", "ABC INFO: t: 6, eps: 3.08406668e+00.\n", "ABC INFO: Accepted: 100 / 3623 = 2.7601e-02, ESS: 9.6037e+01.\n", "ABC INFO: Stop: Maximum number of generations.\n", "ABC.History INFO: Done \n" ] }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEGCAYAAAB/+QKOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8/fFQqAAAACXBIWXMAAAsTAAALEwEAmpwYAAB4w0lEQVR4nO3dd3hb1fnA8e/RsmzJe287iZ09IM6EEAJJCDMQ9k5L2fArDaOMQqBlFSilpWxK2btQAiSBQBLIJHvveO+9h9b5/XE9Ew85tmyHnM/z6LF0de/VkW3pvWe9R0gpURRFUZQj6fq7AIqiKMrApAKEoiiK0i4VIBRFUZR2qQChKIqitEsFCEVRFKVdhv4uQHfNmTNHLl26tL+LoSiKcrwR3T3guKtBFBcX93cRlFYWL17M4sWL+7sYiqJ4wHEXIJSBZePGjWzcuLG/i6Eoigccd01MysDi6+vb30VQFMVDxPE2kzolJUVu2rSpv4uhKIpyvOl2H4SqQSg98sUXXwAwb968fi6J8mtlt9vJzs6mvr6+v4tyXDCbzcTExGA0Gnt8LhUglB7ZuXMnoAKE4jnZ2dn4+vqSkJCAEN2+CD6hSCkpKSkhOzubxMTEHp9PdVIrPeLv74+/v39/F0P5Fauvryc4OFgFBzcIIQgODu612paqQSg9ctddd/V3EZQTgAoO7uvN35UKEEqPfPTRRyDh/EHT8RkTis6n5+2eiqIMDB5tYhJCzBFC7BdCHBJC3N/O838XQmxrvB0QQpR7sjxK7ztw4ADpB1Ip/99hSt7fi3S6+rtIitLr9Ho948aNY9SoUVx66aXU1ta22T5y5EjGjh3L3/72N1wu7TOwcuVK/P39GTduHOPGjWPmzJlHnXflypWsXbu2W2V55513SEpKIikpiXfeeafnb64THqtBCCH0wEvALCAb2CiEWCSl3NO0j5TyD632vxM4yVPlUTwjJCSE+PJAaICG1ArK/3eYgHlDVJOA8qvi7e3Ntm3bALj66qt59dVXWbBgQZvthYWFXHXVVVRWVvLYY48BMG3aNL755psOz7ty5UqsVitTp051qxylpaU89thjbNq0CSEE48eP54ILLiAwMLBH768jnqxBTAQOSSlTpZQ24GNgbif7Xwl85MHyKB5w++23M8EyFK9B/vjOiKVmYz7Vq3P6u1iK4jHTpk3j0KFDR20PCwvj9ddf51//+hfuzC9LT0/n1Vdf5e9//zvjxo1j1apVXR7z3XffMWvWLIKCgggMDGTWrFl4MjedJ/sgooGsVo+zgUnt7SiEiAcSgeUdPH8TcBNAXFxc75ZS6ZH333yXMQWhJJw1Ct/TYnAU1VKxOA1DiDfew4P7u3jKr8xjX+9mT25lr55zRJQfC88f6da+DoeDJUuWMGfOnHafHzRoEE6nk8LCQgBWrVrFuHHjALj00kt56KGHmvdNSEjglltuwWq1cs899wDwwQcf8Oyzzx513iFDhvD555+Tk5NDbGxs8/aYmBhycjx3QTZQOqmvAD6XUjrbe1JK+TrwOmgzqfuyYErncrKzqTGWMWzIqQidIPCyoThe20HpR/sJvXUspkhLfxdRUXqsrq6u+Yt+2rRp3HDDDW4d11UT05Guvvpqrr766mMpokd4MkDkALGtHsc0bmvPFcDtHiyL4iEnGYYQUGfGGGUFQGfSE3LdCApe2kbJO7sJu30cel9TP5dS+bVw90q/t7Xua+hMamoqer2esLAw9u7d2+3X6aoGER0dzcqVK5u3Z2dnc/rpp3f7ddzlyQCxEUgSQiSiBYYrgKuO3EkIMQwIBNZ5sCyKB0gpGaWLx2uEH0LX0imt9/ci5LoRFL22g5L39hB64xiEUc3JVH7dioqKuOWWW7jjjjvcHqTh6+tLZWVLk1lXNYizzjqLBx98kLKyMgC+//57nnrqqZ4VvBMe+9RKKR3AHcB3wF7gUynlbiHEn4UQF7Ta9QrgY3m8ZQ1UcBTWUlRVwqr8rUc9Z4rxJejyodgyqyj9/IBbnXaKcrxpanoaOXIkM2fOZPbs2SxcuNDt488//3y+/PJLtzupg4KCePjhh5kwYQITJkzgkUceISgoqCdvoVMqm6tyzKrW5PDhd59TqKvg4UcfaXefimUZVP2YSdjvT1b9Ecox2bt3L8OHD+/vYhxXOvid/fpXlFMGjoaD5YwjkciYqA738RkXCoA9p6qviqUoSi8ZKKOYlOOMdLpoSK0g+aThTLrogg73MwR7I0x6bDnVWFL6sICKovSYqkEox8SWWYW0OVmWtpZXX321w/2ETmCMsmDPrenD0imK0htUDUI5JvWHykHAvsp0ZBetR6YoKzUb85Eu2Wa0k6IoA5sKEMoxaThYhjHGlzhLfJf7GqOtyLUuHMV1GMN8+qB0iqL0BhUglG5z1TuwZVfhOz2W+WfN73J/U7Q2ic6eW60ChKIcR1QfhNJtDYfLwQXmpABeeuklXnrppU73N4T6gEGHLae6bwqoKL1sIKX7njNnDgEBAZx33nk9f2NdUAFC6bb6Q+UIow5TnB/FxcUUFxd3ur/QC4wRPthVgFCOU02pNnbt2oXJZGoemNG0fffu3SxbtowlS5Y0p/oGLRfTtm3b2LZtGz/88MNR5z2WAHHvvffy3nvv9ewNuUkFCKXbGg6W4zXIH2HQkZycTHJycpfHmKKt2HJr1Ixq5bjXn+m+Ac4880x8fX27Xe5jofoglG5xlNfjKK7DMikSgCuvvNKt44xRVuQv+TjLGjAEmT1ZROXXbMn9kL+zd88ZMRrOftqtXfs73XdfUwFC6ZaGg+WA1v8A8MILLwBw1113dXpcU0e1LadaBQjluKPSfSuKG+oPlaPzNWII10YjVVRUuHWcMdwCOm0kE6NDPFlE5dfMzSv93jZQ0n33NRUgFLdJl6ThUBnm5KDmdMajR49261hh1GEMs2gBQlF+hfoi3XdfUwFCcZs9rwZXjQOvIQHN2+bNm+f28cZoK/X7S5FSuv0BUpSBrKnpyW63YzAYuPbaa1mwYIHbx59//vlccsklfPXVV7z44otMmzaty2OmTZvGvn37qK6uJiYmhn//+9+cddZZPXkbHVIBQnFbw6FyoKX/AeD5558HcOtDYYqyULu5AFeVDb2flyeKqCgeUV3dfs3X6Wx3lWQATj/99C5Xe0tOTmbHjh3dKou7o516gwoQitvqD5VhCPNp8+VeVeV+Gm9jq45qbxUgFGXAUwFCcYt0uGhIq8Q6KaLN9gkTJrh9DmOkBQTYc6rxHh7c20VUFKWXqQChuMVeWAsOF6a4thN0zjnnHLfPofMyYAjxxqZSfyvKcUHNpFbcYs/TvtSNkdY225977jmee+45t89jjLKqkUyKcpzwaIAQQswRQuwXQhwSQtzfwT6XCSH2CCF2CyE+9GR5lGNnz6tBGHUYQrzbbK+pqaGmxv0agSnKirO8AWeNvbeLqChKL/NYE5MQQg+8BMwCsoGNQohFUso9rfZJAh4ATpFSlgkhwjxVHqVn7HnVGMJ9jlrw57TTTuvWeYzRFu18udXokwJ7rXyKovQ+T9YgJgKHpJSpUkob8DEw94h9bgReklKWAUgpCz1YHuUYSSmx59VgirIe9dyMGTOYMWOG2+dqOodqZlKOJwMl3fe2bduYMmUKI0eOZMyYMXzyySe98wY74MkAEQ1ktXqc3bittWQgWQixRgixXgjRbgYsIcRNQohNQohNRUVFHiqu0hFnpQ1XrUMbhXSEZ555hmeeecbtc+l8jOgDvdTaEMpxZaCk+/bx8eHdd99l9+7dLF26lLvuuovy8vIev7+O9HcntQFIAk4HrgTeEEIEHLmTlPJ1KWWKlDIlNDS0b0uoNF/ttxcg6urqqKur69b5tI5qNZJJOT71Z7rv5ORkkpKSAIiKiiIsLAxPXjR7cphrDhDb6nFM47bWsoFfpJR2IE0IcQAtYGz0YLmUbmoZwXR0gJg1a1a3z2eKslK/uwRXvQOdWY20Vtz31w1/ZV/pvl4957CgYfxx4h/d2ncgpfvesGEDNpuNwYMHu1X2Y+HJT+dGIEkIkYgWGK4Arjpin/+h1Rz+I4QIQWtySvVgmZRjYM+rQR9sRud19L/L1KlTu32+phnV9rwavBL9e1w+RfG0gZbuOy8vj2uvvZZ33nkHnc5zDUEeCxBSSocQ4g7gO0APvCWl3C2E+DOwSUq5qPG52UKIPYATuFdKWeKpMinHxp5bjSni6NoDwNNPa+mX77+/3VHM7Wq9NoQKEEp3uHul39sGUrrvyspKzj33XJ544gkmT57c7dfoDo/W76WUi4HFR2x7pNV9CSxovCkDkKvBgaO0Hp+Tw9t9vqGhodvn1Pua0Pka1Ugm5VelL9J922w2LrroIq677jouueSSHpe5K6oBWOmUPb8WZPv9DwAXXnjhMZ3XFGVVI5mU415fp/v+9NNP+fnnnykpKeHtt98G4O23325u/uptKkAonbLnNY5gimo/QIwdO/aYzmuMtlJ/sAxpdyKM+mMun6L0hYGS7vuaa67hmmuucXv/nurvYa7KAGfPrUGYDej920/P/eSTT/Lkk092+7ymKCu4GmsoiqIMSCpAKJ3SZlBbOmxTtdvt2O3dz6tkjGrpqFYUZWBSTUxKh6RLYs+vwTIxosN9rrvuumM6tz7QC+FtUB3VijKAqQChdMhRUoe0u45K8d1aYmLiMZ1bCIEh3ExNajGBJB1rERVF8SDVxKR0qCkdRkcjmACeeOIJnnjiie6fu6GeQ2mbcRbVU1FQcMxlVBTFc1SAUDpkz6sBncAY7tPhPg6HA4fD0a3zSilZ9sZL5BUfRC8M7FuyvKdFVRTFA1SAUDpkz6vGGOaDMHT8b3Lbbbdx2223deu8W5d+zd5VK0g4XVvPOm/jXhzH0NGtKH1loKT7zsjI4OSTT25+zaassp6iAoTSIVteTafNSwChoaF0J8Nu1p6drHz3TQanTObkyy9ECvB2Wji4fnVPi6soHjNQ0n1HRkaybt06tm3bxi+//MLTTz9Nbm5uz99gB1SAUNrlrLbhqrR1OEGuyeOPP87jjz/u1jkri4v4+u9PExARxdm3L0BnMmAM8SbUN5at33/bG8VWFI/rz3TfJpMJLy9tTlJDQ0NzbcVT1CgmpV2dpfhurbOZpK05bDa+fv5JHDYbc+9+CC8fn+bzB1VFkbf3QwpSDxE+aEjPCq78quU/+SQNe3s33bfX8GFEPPigW/sOhHTfWVlZnHvuuRw6dIhnn32WqKgot99rd6kAobSrJUB0PMQVYOHChV2eS0rJj2+9Qv7hg1xw94MEx7QsE2IMt2DYocdstrLt+28565bf96zgiuIBAyndd2xsLDt27CA3N5cLL7yQSy65hPDw9pNp9pQKEEq77Hk16P1M6C3GHp9rxw9L2LViGZMuupykiW3XjzA2phEfM342W1Z/w2nX/BZvq2+PX1P5dXL3Sr+3DaR0302ioqIYNWoUq1at8lhmVxUglHbZcqub02F0pqn/4U9/+lO7zxemp7L8P6+TMG48Uy87cr0oMEZoTU1JSRPYsOYLdq9YRsr583pQckXpH32R7js7O5vg4GC8vb0pKytj9erV/OEPf+hx2TuiOqmVo0i7C0dRXZf9D6D1QXTWD7FrxTJ0Oh3n3HE3Ot3RWVv1gWaESYfZ7kP0sBFsX7YE6eGON0XpLU1NTyNHjmTmzJnMnj3brWbXJueffz5ffvml253Ue/fuZdKkSYwdO5bp06dzzz33MHr06J68hU6pGoRyFHthLbikWwGisw+DdLk4uGEt8WNPxtvXr919hE5gDLdgz69h3Oxz+fafz5K+fQuJJ6Ucc/kVpbcNlHTfs2bN6tb+PaVqEMpRmhLouRMgOpN/+CDVpSUkTZzS6X7GCC1ADJk4BUtAIFu/c79TT1EUz1EBQjmKPa8GYdJhCPbuct+//OUv/OUvf2n3uQO/rEGn1zN4/KROz2GI8MFV60DUwegz55C2bTPlBfnHVHZFUXqPRwOEEGKOEGK/EOKQEOKoVe2FEPOFEEVCiG2Nt995sjyKe2x51RgjLAhd1x1tLper3ck6UkoObVhH3KixmK2dd3Y3jWSy59cwZuZZCCHYvmxxp8coiuJ5HuuDEELogZeAWUA2sFEIsUhKueeIXT+RUt7hqXIo3SOlxJ5Xg89Y99JndNQHUZSRRnlBHhPmXtzlOVoHCN/kGJImTGHXimVMvexqjKb2V7JTFMXzPFmDmAgcklKmSiltwMfAXA++ntILnGUNyHpnlxPkunJww1qE0DEkZXKX++otRnS+Ruz52uS8cWedS311FfvX/NyjMiiK0jOeDBDRQFarx9mN2450sRBihxDicyFEbDvPI4S4SQixSQixqaioyBNlVRo1z6DuIgdTkz//+c/8+c9/Pmr7wV/WEj18BD7+AW6dp6mjGiBmxGiCY+LYufx79wqtKIpH9Hcn9ddAgpRyDLAMeKe9naSUr0spU6SUKd3JHKp0nz2vGkRLs09XpJRHJSYrycmiJDuTpImnuP26xnAL9sJapFMihCBh3HgK0g7h7OZaE4riCQMl3XeTyspKYmJiuOMOz7bOe3IeRA7QukYQ07itmZSypNXDN4FnPFgexQ22vBoMwd7oTEdPamtPe30QhzasA+hyeGtrxggLOCSOkjqMYT6EJwzCabdTmptNaFyC2+dRFE9onWrj6quv5tVXX2XBggVtthcWFnLVVVdRWVnZnPK7q1xMK1euxGq1MnXq1A73ac/DDz/MaaeddkzvpTs8WYPYCCQJIRKFECbgCmBR6x2EEJGtHl4AdD95idJrpJTYs6vcbl7qyIFf1hA5ZCi+wSFuH9OUcqOpmSkscTAAhWmHe1QWRelt/ZnuG2Dz5s0UFBQwe/bsbpe9uzxWg5BSOoQQdwDfAXrgLSnlbiHEn4FNUspFwP8JIS4AHEApMN9T5VG65iiuw1lhw2twgNvHNPU/PPLIIwBUFOZTmHaY067+Tbde2xjuA6IxQIwJJTAqGoPJi8K0w4ycfma3zqX8eq369ADFWe3Paj5WIbFWpl2W7Na+/Z3u2+Vycffdd/P++++3uwBRb/Noqg0p5WJg8RHbHml1/wHgAU+WQXFfw4EyAMxDAtw+5sgrpYPNzUvdqzILox5DiDf2fK1tV6fTExqXQGF6arfOoyieMFDSfb/88succ845xMTEuH3OnlC5mJRm9YfK0Qeb3ZpB3eTIPoiDv6wlND6RgIjIDo7omDHCgi235eowLHEQe1f/hHS5ELr+Hk+hDATuXun3toGS7nvdunWsWrWKl19+merqamw2G1arlaeffrrbr+UOFSAUAKTDRcPhCnxODjvmc1SXlpB7YC9TL+t8wZOOGMN9qNtVjMvmRGfSE5YwmO3LllBRWHBMAUdR+lJfpPv+4IMPmu+//fbbbNq0yWPBAfp/mKsyQNgyq5A2J+akgG4d13oexKGN6wFInuT+8NbWjBEWkOAo0JqZmjuq01VHtTIw9XW6776mahAKAPUHy0BHtzqooW0fxMENawiKiiE4Ju6YytA65YYp1peQ2HiETkdheirJk089pnMqSm8YKOm+W5s/fz7z588/pmPdpQKEAmj9D6ZYP3Tm7v1LNF0t1VZWkLVnFxPnHvvSh/ogM8Koax7qajCZCI6JU0NdFaWfqCYmBVetHXt2Vbebl1o7vOkXpMvV7dFLrQmdwBDu0xwgAMISBqmRTIrST1QNQqH+UDlI8EoK7PaxTf0Po0wu/ELDm/sNjpUxwkL93tLmx2EJg9nz83Kqy0qxBgb16NyKonSPqkEoNBwqR3jpMcX4dvtYKSXS5SJjxzaSJk5xe/RGR4wRFlw1dpxVNkAb6gqqo1pR+oMKECc4KSX1B8rwGhKA0Hf/y33hwoXMnTIBl9PBsKk9zw1zVMqNhMYAkaaamRSlr6kAcYJzFNfhLG/AfAzNS6AFmG3ff0v4oCGED07qcXlaRjJpQ129fCwEhEeqGoSi9AMVIE5AuQf2krZtMwANB8sBjrmD+s8LF5Kl92bc7HN73LwEoLea0FmNqqNaGVAGUrrvptccN24cF1xwQc/fXCdUJ/UJpigznc8e/xPS6eS6Z/+F62AF+qDupddow+VEGkxIhmCrc2Dy7vm/lDHCgr2gVYBIHMyBX9bQUFuDl0/PMs0qyrEYSOm+3U370RtUDeIEUlddxVfPPY6XjwWDyYvl/36NhsMV3a49SCkpyqzi54+3Yj24E79sEz9/ksbXL27HVt/zBX6M4T44CmqRLm0SXnM/hKpFKANAf6f77kuqBnGCcLmcfPuPZ6gqLubyR5+iIO0wOz74BhnldLv/oSiziv0b8kndWkRVST2OurWAi8nz5uIXGsHKD/bzzb+2c94dYzF1c8Jda8YIC9LuwlFajzHEu83aELEjRh/zeZXj34q3X6cwo3cvFMLiBzFj/k1u7dvf6b4B6uvrSUlJwWAwcP/993PhhRe6+1a7TQWIE8Tqj98jY8dWZt10B1HJw4kYkkzNihykdEGUqcvjs/eXsegf2xA6iB0exMlnxbDq/f9QETWK77Z/zcKFCzGZDXz/7918+9IOzrtjLEYv91alO1JTR7UjvwZjiDeWgEAsAYFqRrXSbwZKum+AjIwMoqOjSU1N5YwzzmD06NEMHtyz+Ucd6TJACCH0wG4p5TCPlEDxuP3rVrHxq88ZM3MOY87Urnx0Oj0JoWMozsrg8DcHOP26Gzs8vqq0nu/f3EVAmDfz7h2P2WJk/7rV1FaUISMSm/cbMj4MKSXL/r2bb1/azrl3jMXo5tKlrRlaLR7kPUpblU51VCuA21f6vW2gpPsGiI6OBrTayumnn87WrVv7L0BIKZ1CiP1CiDgpZaZHSqF4TFFGGktfeYHI5GHMmH9z83ZXrR2KHRBpYMuSrxlx2pnNbf2tOexOlr62E4fdxdm3jMZsMQKw7ftv8AsN5w+PLkSnawkCSSnhSJfkh//s4duXdnDu7WO6HSR0Jj2GIHPbkUyJQ0jfsRW7rQGjyau7vwZF8bi+SPddVlaGj48PXl5eFBcXs2bNGu67774el70j7nZSBwK7hRA/CiEWNd08ViqlV7TulL5gwYMYjMbm5+oPl4OEwRecitnqy4//fgXZODyvtVWfHKQwo4qZ148gsLHppzgznew9uxg76+w2waFJ8sQIzpw/gpwDZSx+eQcOW8cZLztijLA0z4UAbUa1dLkoyczo9rkUxVP6Ot333r17SUlJYezYscyYMYP777+fESNG9OQtdMrdPoiHj+XkQog5wD/Q1qR+U0rZ7soWQoiLgc+BCVLKTcfyWkpbzZ3SJSVc/uhTR+UxajiopdewJocx/ZrfsvTlv7Prpx8YPaNlIfTdq3LYszqX8XPiGXRSaPP2bcuWoDcaGTVjVvNwviM/FEMnRSBdkh/f3cuSV3dyzu1j0OvdHzRnjLRQt6ekzeJBoI1kihjSP6uKKSeugZLue+rUqezcudPt/XvKrU+slPInYB/g23jb27itQ419Fy8BZwMjgCuFEEeFOiGEL/B74JfuFV3pzPbvF5OxYytn3nALUcnD2zzXnF5jcABCr2PEaWcQPWwkP3/wNnVVWnW3IK2Snz85QOyIICZe0NL01FBby56flzN0yjR8/Pw7LcOwKZHMuHoYmXtK2fFjdrfKb4y0gmxJueEfFo6Xj0XNqFaUPuRWgBBCXAZsAC4FLgN+EUJ0lfh/InBISpkqpbQBHwNz29nvL8BfgXq3S610KXPXDgLCI5s7pVtzlNRr6TWSAwAQQjDzhltpqKlm1UfvUFtpY+nrO7H4ezH7tyPR6VraU/esWo69vo5xs88FtJpDZ1XqEadGkTAmhA3fpFJV6v6f2BjVmHIjt6a5jKEJiSonk6L0IXfr/A+hNf9cL6W8Du3Lv6tmp2ggq9Xj7MZtzYQQJwOxUspv3SyH4qbC9NQOU283HCwDwDykZf5DSFwC48+9kJ0/fsf//vYltVU2zr55NGZrS7+FlJLt3y8mfNCQbjXzTLtcy9G06pMDbh+jD/BCmA3Y81qq9mEJgynKSMPVSbVeUZTe426A0EkpC1s9LunGse0SQuiA54G73dj3JiHEJiHEpqKiop687AmhvrqayqKCdkclgbb+gz7QC32wuc32KZdcicknkLx97yEb3mbnjx+SsWMbTocdgOw9OynJzmyTd+mxxx5r7ofoiF+wNxPOSyRtezGp29z7+wkhMEZasOe1zcnksNsoze1ec5WiKMfG3U7qpUKI74CPGh9fDizu4pgcILbV45jGbU18gVHAysYvmwhgkRDigiM7qqWUrwOvA6SkpHQ9h/0E1zRfoL0ahJQSW3oF5mHBRw3Fy9pbhTBeQfy4EgyGDHYu/56tS7/G5O1DwrjxVJcUY7ZYGTp1WrfLNPbMWPavz2fVpweIHR7k1iQ6U5SFmg35SJdE6AThiS0d1SGx8d0ug6Io3eNWgJBS3ts40uiUxk2vSym/7OKwjUCSECIRLTBcAVzV6pwVQEjTYyHESuAeNYqp55o6ctud11BUh6vGgVeCX5vt5YW1LH9nL+GDwrno7rPRG3XYG+rJ3LWdw5t+4fDmDdRWlDPhgosxerXUPNwd0qfX6zj9qqF88dwWNn6bxtR5Q7o8xhjZmHKjpA5jqA9B0bEYjCYK0w4zYtoMt15XUZRj53aqDSnlf4H/dmN/hxDiDuA7tGGub0kpdwsh/gxsklKqeRQeUpieiiUwCEvA0TmWbOnaKCVTqwDhsDn57o1dCJ3grBtHojdqrYdGLzODx09i8PhJ2hyE7EwCIqOPOqe7IocEMPyUSLb/kMXQSREER1s73d8YqT1vz6vBGOqDTq8nJC5ezahW+pxer2f06NE4HA6GDx/OO++8g4+PT/N2u92OwWDguuuu4w9/+AM6nY6VK1cyd+5cEhO1bAMhISH88MMPbc67cuVKTCZTt7K5ZmZm8rvf/Y6srCyEECxevJiEhITefLvNOu1HEEKsbvxZJYSobHWrEkJUdnYsgJRysZQyWUo5WEr5ROO2R9oLDlLK01XtoXcUph3usP+hIb0CncWIIaQlvfeqTw5QnFXNzN+MwK+DtN9CpyMkLqHNZDtwrw+itakXDcHkbeCnD/c3Z2vtiDHcB3TiiH6IwRSmH3YrW6ai9JamVBu7du3CZDLx6quvttm+e/duli1bxpIlS9p8HqZNm8a2bdvYtm3bUcEBjm09iOuuu457772XvXv3smHDBsLCwnr25jrRaYCQUp7a+NNXSunX6uYrpfTr7Filf9htDZTmZjdPLDtSQ0YlpgS/5v6HvWvz2LMmj/Fnx5MwOqTdY3qT2Wpk6sVDyDtcwd51eZ3uKww6jGHe2HNbjWRKHERDTQ2VRYWdHKkontOf6b737NmDw+Fg1qxZAFitVnx8fLr/JtykkvX9ypRkZiBdLsISj65BOCttOEvqsU6OBKA4u5qfP9pP9NBAJp7ffo2jK91JK9Bk2JQI9q7NZe0Xh0gcG4K3teNsssZIq5YWpFHzjOq0w/iHhXf7tZXjW/nXh7Hl1nS9YzeYoiwEnO9esrv+Tvd94MABAgICmDdvHmlpacycOZOnn34avf7YMid3RSXr+5VpHsHUTg2iIb0CAK8Ef2x1Dpa+vhOTj4HZN7SdDOdpQgimXzWUTx/fyNr/HuLM6zvOJWOMtFC7tRBntQ291URIXDxC6ChMP0zSJPfbbRWlJwZKum+Hw8GqVavYunUrcXFxXH755bz99ttul6e73O2kbkrWtwFoDt9SSs8uiKp0W2H6Ybx8LO1eXdvSKxFGHYZIH75/ay+VxfVc+Idx+Ph1vR5ERzrKxdSV4Cgr42bHsWVpBoNPDuuweat5RnVeDfokE0YvM0HRMaqj+gTl7pV+bxso6b5jYmIYN24cgwZpNf4LL7yQ9evX93uAOKZkfUrfK0xLJTQhsd10ww0ZlZjifNn5Uy6HtxQyZd5gotxcTc4TJp6bSMbOYpa/t48rH56It+/Rgar1SKamle9C4xPJ2b+nT8uqKF3pi3TfEyZMoLy8nKKiIkJDQ1m+fDkpKSk9LntHupOsLx0wNt7fCGzxWKmUY+JyOSnKTG+3eclV78CeW02Dr4k1/9Xa/k+aFdfj1+wqF1Nn9EYds347koZaOyve39dux57eYkTvZ2ozkikkLoGq4iLqa9rPsKkofaWv033r9Xqee+45zjzzTEaPHo2Ukhtv7Hixr55yqwYhhLgRuAkIAgaj5VR6FTjTYyVTuq0sNweHraHdIa62zCqQsGlTEQFh3sycP8LtqxxPCo62MuXCwaz5/BB71+Yx4pSoo/bRUm60GskUr40rL85IJ2bEqD4rq3LiGijpvgFmzZrV7WOOlbv5lG5Hm0VdCSClPAh4bvCtckya1mxuL8VG3eFyJFDqlJxz6xhM3r2zHHl350G0Z+wZsUQPDWTVpwepKKo96nljlBV7YR3Sri1oFBKfAEBRZlqPXldRlM65GyAaGlN2AyCEMABqptIAU5Ceit5oJCgqps12KSUFG/KpcLg483cjCQj33LjpYyF0gjOvH45eL1j21h5czrYr2xkjLeCS2Au14GENDMZs9aUoM70fSqsoJw53A8RPQogHAW8hxCzgM+BrzxVLORZF6YcJiU1Ab2hbO9i6NANzrR1Tgj/xI4N79TV70gfRmm+QmelXDqUgrZLNS9suK2qMbBrJpFXzhRCExidSlKFqECcKNXPefb35u3I3QNwPFAE7gZuBxVLKhzo/ROlLUkoK01KPmiCXsauEfd+mYRCCmNOOPY9SX0iaEE7ShHA2fptOQXrLyA5DsDfCpGtePAggNC6B4qwMXC61NsSvndlspqSkRAUJN0gpKSkpwWw2d72zG9xtiL5TSvkP4I2mDUKI3zduUwaAplE9rUcwlRfU8v2/dzMsyAtsTrwSO18i9Fgc6zyIjky/Mpm8Q+X88J89XPbgBIxeeoROYIywYGs9kik+AUdDAxUF+QT2IIGgMvDFxMSQnZ2NWgvGPWazmZiYmK53dIO7AeJ64MhgML+dbUo/KTgixbetzsHiV3ag0wmS4nyhrB59O/MMeqq3R0J5+Rg5c/4IvnphK2s+P8jpV2sZXoyRFmq3FyGlRAhBWLz2Posy0lSA+JUzGo3NGVGVvtVpgBBCXIm2hkOiEKJ1BlY/oNSTBVO6pzAtFSF0hDaO8Fnz+UHKC+u44M4xuD4/gHlY7/Y9NHnkkUd6/ZwxQwM5aVYcW7/PJHpoIEkp4Rgjrchf8nGWN2AINBMUE4sQOooy00mefGqvl0FRlK5rEGuBPLSFff7WansV0DcDcRW3FKYfJjAqGqOXmcKMSvaszWPcmbGEB3tT0M4CQQPdpLmDyDtUwYr39hEa64t3U8qN3BoMgWaMJi8CI6Moykjv34Iqyq9YV+m+M6SUK4GZwKrGWdR5aMuH9v8sK6VZYXoqYQmDkFKy+tODeFuNpJyb2O4CQb2pN+ZBtEev1zH7dyPRG3QsfX0XIsgMgjYT5tRIJkXxLHdHMf0MmIUQ0cD3wLXA254qlNI9tZUVVJcUE5Y4mIObCsg7XMHkuYPx8ja0u0BQbxJCeGxGtm+QmZm/GUFJTjVrvjyMIdi7TUd1aHwilUUFNNT2bvpnRVE07nZSCyllrRDiBuBlKeUzQohtHiyX0g1NmU2DouJZ/flhQuN8GTZVW/OhIb3tAkG9zRN9EK3Fjwpm/Jx4Ni/NIHFkIF5H5GQCKM7MIHpYxynDFUU5Nu7WIIQQYgpwNfBt4zbPrFChdFtTio28VBPVZQ2celkSOp3AWdmAs7T+uOt/ONLE8xOJSgrgUFolztJ6XPUOQKtBAKqZSVE8xN0AcRfwAPCllHK3EGIQsKKrg4QQcxoXGzokhLi/nedvEULsFEJsE0KsFkKoy8BjUJieiiUwmF0/l5A0IZyoIQGAVnsAbYEgT/FUH0RrOr2O2TeMpFav/bvWZVYB4BscgpfFonIyKYqHuJ3uu3FxoJeEEFYpZaqU8v86O6ZxqdKXgLOBEcCV7QSAD6WUo6WU44BngOe7/Q4UCtNT0enDEMCUi1omyjUtENS06I4neLIPojVLgBcnXZEMwN5vUpvnQ4TGJaqcTIriIW4FCCHEaCHEVmA3sEcIsVkIMbKLwyYChxqDiQ34GJjbegcpZWWrhxZUAsBus9XXUZaXQ221PyfPicc3qGWKfUN6BaY4X4Te3Ypi9z3yyCMe74doEnNyGE6Djoasavatywe0ZqbijHSky9XF0YqidJe73xyvAQuklPFSyjjgblql3ehANJDV6nF247Y2hBC3CyEOo9Ug2q2VCCFuEkJsEkJsUtPt2ypMTwcp8QmIbrMAkKvegT2vBpMHm5f6mhAC7wQ/gi0G1n15CHuDk5C4BOwN9VQUFvR38RTlV8fdAGGRUjb3OTTOjeiVdgsp5UtSysHAH4E/dbDP61LKFCllSmhoaG+87K/Gjh+1hf0mXzgZg6ll3EDTAkGe7qDuiz6I1kyRFqxSUl9lZ8eKrOaZ46qjWlF6n7sBIlUI8bAQIqHx9iegq1Xjc4DYVo9jGrd15GPgQjfLowD1NXYOb96NTu/NqOlD2zzXkFEJAkxxvh4tQ1/1QTQxRlnBKUkeFsDW7zPxDY4GIVRHtaJ4gLvzIH4LPAZ8gdZPsKpxW2c2AklCiES0wHAFWl6nZkKIpMbV6QDOBQ6iuG3XTznY6wsIHzQYna5trLfn1WAI9Ubn1Tsrx3Wkr/ofmpga14YYNTqE/fvK2fVzAYERKuWGonhCV8n6zMAtwBC0tSDullLa3TmxlNIhhLgD+A5tzsRbjUNk/wxsklIuAu4QQswE7EAZWtZYxQ0up4tdP2eCq5iYYacc9bw9vwZTjLUfSuZZwluCDkyFpQwZH8r25dmERcepGoSieEBXl5fvoH15r0IbrjocbU6EW6SUi4HFR2x7pNX937t7LqWt9B0lVBXnIaWzOcV3E1e9A2dpPcYJ4R4vR2+vB9HEnpdH3dat2DIzsaVnaD8zM3EWF+Nz+kNULqshZO8XHBpzL9X7K6mozqds5Qp8R43GEBLSq2VRlBNVVwFihJRyNIAQ4t/ABs8XSXHHzp+yMXmXYaviqABhL9DWbjZGeG7+Q5Mjm7Z6QtrtVK1cSflnn1GzajU0riBmCA/HFBeH9fTpmOLicdZFYS/UEzV4OrHpqWS4EoDd7F3wBwJrG7BMnUroggV4j+pqJLaiKJ3pKkA0Nyc1Nhl5uDiKO8rya8jeV0ZodD11pYajFsyxN+YralrL2ZMefvjhHp/DlplJ+ef/pfzLL3AWFWMIDyfk1lvwnTULU3w8Oh+fNvtX/5JH+ZeHCL11AadLyXsPfQ+1YLj5d4S69JS+8y7pl1yC75w5hP7+//BSi80oyjHpKkCMFUI0TWYTgHfjYwFIKeXxneTnOLXzpxx0BoFOlBEYFYNO3zYtlj2/BmHWo/f36qcSuqd6zRpK3nyT2nXrQafDOn06AZdeivW0aQhDx/+apiitb8WWU43f6BBGTR/Gpq9MFNlcTLzzDgKvuYbS//yHkrffoWrZMgIuvpiQ22/DGO75JjdF+TXpNEBIKVVCvgHGVu9g/7o8hpwcRvqWLCKThh21jz2vBmOEpU+Gnx5LH0T9/v0UPvMsNWvWYIiMJOT/7iRg3jyMERFuHW+MsIBOYM+phtEhpJyTwJZvQ8ncdQAAva8vof/3fwRedRXFr75G2SefUPHVVwRdew0ht9+Oztszqc8V5dfGczkYFI84sKEAW72TYVNCqCwqJCQmrs3z0iWx59f0SfMSgF6vR6937zrCXlBA7oMPkXbhRdTt2kXY/X9k8HdLCb3tNreDA6Dllwr3wZajJe2z+HsRlpBIbXkuxVkt2VsMISFE/OkhBi9ZjN+csyj591tk/vYGnOXl3XqPinKiUgHiOCKlZNdP2YTEWtEbygEIjotvs4+zvAHZ4OyzAPGnP/2JP/2p3QnwLWWqrqHwH//g8FlzqPz6a4Lmz2fI998RPH8+OpPpmF7XGG3FnlONbOzIHjZlNGBj9WdbjtrXFBND1F//SvQLL1C/axcZ116LPT//mF5XUU4kKkAcR/IOVVCSU8Po02Mozc4EOKoG0dxB3QcjmLoiXS7KP/+cw2edRckrr+J7xhkMWrKY8D/eh96/ZzmiTNFWXLUOnBUNAEQmaVlsM3cdoCCtst1j/M6aTewbb2DPzSP9qqtoSFVzJxSlMypAHEd2/pSNl4+BpAnhFGdnYjCa8A9v2zRjz68BAcbwvgkQHeViqtuxg/TLryDvTw9jio8n4dNPiH7+b5hiYnrldY3RWke1PVtbozokLh6EQKcrYcfKrA6Ps0yeRPx77yIbbGRcfTV1O3f2SnkU5ddIBYjjRE1FA6lbihg2NRKjSU9JdiaB0THodEePYDIEmdF59c34giP7IBwlJeQ+9BDpl12OIz+fqGefIf6D9/EeM6ZXX9cUaQGdNpIJwGT2JiA8Am9rJalbi7A1rjrXHvOIESR8+AE6i4WM6+dTvWZNr5ZNUX4tVIA4TuxZnYvLJRl1mjbnoSQr86jmJWgZwdRXmvogpMNB6bvvcnjO2VR8tYigG37LoCVL8D//fI+MphJGPcYwS3OAAAiNS8RpK8Jhc5G2rfO08Kb4eOI//ABTbCxZt9xK5ZIlvV5GRTneqQBxHHA6Xez+OYe4kUEEhPnQUFtDVUkRwUcECJfNiaOkrs86qAEKCwvJWrSItIsuouDJp/AeM4ZBi74i/N570Vs9W44jO6pD4xOpKi3AN0jH/l+67oQ2hoUR/967eI8dQ86Cu6letcqj5VWU440KEMeBtG3F1FTYGDVda78vaeygDo5tO4LJUVALsm86qKWU1Kxdyysvv8xbW7bgarAR868XiX3zDbwGDer6BL3AFG3FVWPHWWkDICQ+AaQkaoiL7H1lVJc1dHkOvZ8fcW+8gVdyMrn33Is9p7OM9IpyYlEB4jiw66dsfIPMxI8KBqA4q3EE0xEBoq9SbNRu2Urm9fPJ/O0NWKuqMTudDP72G3xnzuzbtSGO6KgOjdNSalgDqpESDmx0byirztubmH/+A+l0kv37u3A1dB1YFOVEoALEAFeWX0POgXJGTY9Gp9O+fEuyMzF4eeEfGtZmX3t+DcKkRx9obu9UPVa3azeZN99MxlVX0XD4MOEPPsgfnnyCP/7lLwij0SOv2RljpAUEzRPm/EPD8LJYqCzOIDzRjwNuNDM1McXHE/XXp6nftYuCJ5/yVJEV5biiAsQAt29dPkInGDYlsnlbSXYmwdGxiCMyqdryajBG+CB0vXcV76yqouzjj0m75FLSL7mEum3bCV2wgCHLvifoumvJyM0lLa1/5hPoTHoMYT5ayg1A6HREJg0j78A+hk6KoCSnhuLsKrfP53vmmQTf+DvKP/mE8i//56FSK8rxw7PLjSk94nJJDmzIJ25kED5+LTOOS7IyiBs9rs2+UmopNnzG9HwtBCkldZs2Uf75f6n87jtkfT1eSUmEP3A//hddhN6vJUfju+++C/T+ehDuMkVbqT9Y1vw4cshQ1m3fQvxIX3R6wf71+YRc4v6yq6G//z11O3aS/+ijmIcPwzzs6FxXinKiUAFiAMs5oHW0Tr14SPO2+upqqstKjxrB5Ky0Iescx9xBLR0O6nftonrNGiq//gZbejo6iwX/Cy4g4JKLMY8e3W7/grEfmpbavH60ldothTgrG9D7eRGVNBSkpCwvjfhRwRzYWMCUeUOam+e6IgwGov/2HGnzLib7/35P4ueftQmIinIiUQFiANu/Lh+Tt4HEsS21guLsDKDnHdRSSmxp6dSsXUvNunXU/vILrmqtqcY7ZTyRN9+M31mzj1qL4UgPPvig2+/HE0yNHdW27Gq8R3gRkTQUgLyD+xg66QzStheTva+UuBHBbp/TEBJC9AsvkHHddeT+8X5iXvrXUc15inIiUAFigLLVOzi8tZDkSREYjC0zlUsaRzAdWYOw53eeg8nV0EDDvn3Ubd9B3Y4d1G7ahKMxYZ0xOhq/s8/GMnUKPpMmYQgKcruc27dvB2Ds2LHuv7lG5fXlpFemU2mrpKKhgkpbpXZr0H4CBJmDCDIHEWgOJMgcRLA5mEBzIOE+4eh1eoyR1saO6mq8RwRjtlgJio4l9+A+JlxwGV4+Bvavz+9WgADwOfkkwu+7j4Inn6TkjTcJufmmbr8/RTneeTRACCHmAP8A9MCbUsqnj3h+AfA7wAEUAb+VUmZ4skzHi9Rt2ozgYZPa5loqyc7EaPbGLyS0zXZ7Xg36AC90ZoPWH5GdTd2WLVpA2LmT+n37wK4tEGgIDcX75JOx3HILlqlTMMUdPSPbXf/73/+ArgNEpa2SvSV72V2ym93Fu9ldspuc6vbnHFiMFvxN/kgkpfWlNDiPHnZqMVoYFzaOlPAUZgYOoyGnJUFfVPIwDm1cj84gGDI+jP2/5GOrd2Ayd+/fPfDaa6jbtpWif/4T67RTMY8Y0a3jFeV457EAIYTQAy8Bs4BsYKMQYpGUck+r3bYCKVLKWiHErcAzwOWeKtPxZN+6fPxCvYkY3DbraUl2BsExbUcwSSmxZZaBq5qce+6lduNGHAUFAOh8fDCPHk3w/Osxjx6N99ixvbqympdX+6vW2Z12NhZsZEXmCtblrSOjsiXuR1ujGRk8ksuGXsaQgCEEeAXg7+WPn8kPX5MvBl3Lv6WUkjpHHSX1JZTWl1JaV0pxfTH7SvaxuWAz/8j5B0bH9Yw7NJT7v/s748PHkxwVQH11FWV5uQydFMHuVbmkbiti2OTI9oraISEEEY88Qs0vG8hb+CgJH3+EcHPtC0X5NfBkDWIicEhKmQoghPgYmAs0Bwgp5YpW+68HrvFgeY4bVaX15BwoY+J5iUd1DBdnZZJ4Ugqu2lqqlq+g6ocfqN28De8pC7Ed/BFn8XosEybgM2EC3iePx2vIYI9+qd1///3N96tt1azOWc3yrOWszl5Nlb0Ks97MpMhJzB08l5HBIxkRPIIAc4Db5xdC4GP0wcfoQ6xv7FHPl9aXkv79doLXGhDVLl4veB2/Sj0XEsXXq97nkovuwC/EzP71+d0OEAD6gADC77+f3HvvpezDjwi6Vv2LKicOTwaIaKB13uVsYFIn+98AtJsxTQhxE3ATQFwPmkOOFwc25IOE5Iltm5dqS0qorSjHuGU7B14/FVlXhyE0FJ8ps0GnJ+T2a/E786k+nc28ctVK9pXvY5t5G7/k/4LD5SDIHMTM+JnMiJ3B5KjJeBs8t8RnkDkIy6iTKVq7g3+NfJ6GBB2LDn1F3rrPWL1pMS87vuDiqFuo3BlPVVk9vscwidDvvHOp+N//KHrhBXxnzezW6neKcjwbEJ3UQohrgBRgenvPSylfB14HSElJkX1YtD4npWTfunwih/jjH+qNlJLaDRup/OZr0n5aAZEBeB08jP/cC/A/91y8x4+ndmsRZZ8dwGfs0TUOT3BJF5sLNrPo8CL0y7XaSfqodK4dfi0z4mYwJmQMel3fNcUYo7UZ1facagKGxXHdqOv5bPhBAsoKCI8PZ+neT7hY3sfCd/7GWRemMCt+Fjrh/qgkIQQRCx8h9fwLKHjiCWJefNGD70ZRBg5PBogcoHWbQEzjtjaEEDOBh4DpUsoTPglOYXoV5QW1jJs1lOqffqLoXy9Rv3MnOh8fbJNOgtJ8xnz8CX4RLc0l9rwaMOgwBHvuSh0gqzKLRamL+Prw1+RU52AxWphpmolZmFkyb0mf1lxa03kZMIR4t0n9HZU0lKwvd/BIyj9pmOjgw9y1hGUncc/KexgePJz/O/n/OCXqFLfLbIqLI+T22yl6/nmqfvwR3zPP9NTbUZQBw5MBYiOQJIRIRAsMVwBXtd5BCHES8BowR0pZ6MGyHDf2rc9DrwfjP/9I1o7NGKOjifjLn/E/7zyyPnwb06oV+Lazilxvp9hoUueoY1nGMr48+CWbCjYhEEyOnMydJ93JGXFneLT5qDuMUVZs6RXNjyOThyGli4LDB4kdOYZTTh/JTx8d4NHkp3kj70Vu/eFWxoeP566T72Jc2Di3XiP4N/Op/Ppr8v/yOD6TJns8nbmi9DePBQgppUMIcQfwHdow17eklLuFEH8GNkkpFwHPAlbgs8YruUwp5QWeKtNAJqWkcsVP7F9RS3DRLnSV+YQ9/hf8585tToRXkp1BcGzcUVe99vwazMPcn7vgTll2Fu/ky0NfsiRtCTX2GuJ84/i/k/6P8wefT4SlJUCtWKGNM5gxY0avvf6xMEVbqdtehLPaht5qInKINmEu9+B+YkeOYUhKOKs+OUhk3lC+vvBrPj/4Oa9tf41rl1zL9Jjp3HnSnQwNGtrpawijkYg/P0bGlVdR9M9/ENHPkwQVxdM82gchpVwMLD5i2yOt7s/05OsfLxoOHiTvTw+TmQP2UTcx8qyhDL7hD0dlSC3JymTIhMlttjmrbLiq7b2yBkR+TT5L05by1eGvOFR+CG+DN7PiZ3HRkIsYHz6+3eaYn3/+Gej/ANGc+junGv3QILx9/QiMjCbv4D4AzBYjsSODOLipgCkXDebKYVcyd/BcPtz3IW/teotLv76UK4ddyZ0n3YnVZO3wdXxOOomAKy6n7P0P8D//ArxHj+qT96co/WFAdFKfqKTTSenbb1P0wj/QWa2UzXwEnwYjw248H6Fv24laW1FOXVUlwTFHpNjI79kaEGX1ZSzLWMbitMVsKdiCRDImZAwLpyxkTsKcTr8sASyWgdHM0pxyI6ca81CtNhWZNJT07VuQUiKEICklnIydJeSlVhA1JAAfow+/G/07Lk2+lJe2vcRH+z7ih8wfeHDig5wRd0aH/RNhCxZQ9eOP5C18hMRPP0UY1MdI+XVS/9n9xJaRQe4DD1K3ZQvWmWcScN/D/PjXPYyZEYFOf/QIm6ZFgoJjj0ixkdd5io32VNuqWZG1giVpS1iXuw6HdJDon8it427lnMRziPeL7/okje655x639/UkndmAIdjctqM6eRh7fl5ORWEBAeERJI4NwWDUcXBjAVFDApr38/fy58FJD3LeoPN4bN1j3LXyLk6PPZ0HJz5IpPXouRN6Pz8iHnqInLv+QOl77xP8m/l98A4Vpe+pANHHpJSUf/wxBc88izAYiPrr0/hdcAE7V+bgcso26z60VtKUpO/IHEx5Nej8TOgtHWdVlVKyv2w/q3NWsyZnDdsKt+GQDiIsEVw74lrOGXQOQwOHHtMopMWLtRbEc845p9vH9jZjtBVbZsv6D5FJWqruvIP7CAiPwGQ2kDAmhMNbCpl2WdJRgXhM6Bg+Pu9jPtjzAS9vf5m5X83ljnF3cNXwq9rM7gbwPessLNNPo/jFF/E/71wMoW1TnyjKr4EKEH3Inp9P3oMPUbN2LZZTTiHyicebJ13tX59HSKyV4Oj2m3SKszLwsliwBLbtjNZGMLWtPUgpKagtYGvhVlbnrGZt7lqK64oBGBY0jOtGXsf0mOmMCxvXrfkA7dm4cSMwMAKEKdqXuh3FOGvs6C1GQmLjMXqZyT2wj+Gnng5A0oRwDm0uJHtfGXEjj07gZ9QZmT9qPrMSZvH4+sd5dtOzfJv2LY+f8jhJgUnN+wkhiHjgAQ6ffwGFL7xA1BNP9NXbVJQ+owJEH6ndupXs227H1dBAxKMLCbj88uYr9pKcagozqjjlkiEdHl+SnUlwTHybq3zpdGEvrEWXaGJF5gp2l+xmT8kedpfsprS+FNCaT6ZGTuWU6FOYGjWVUJ/evdL19XV/MR5Pa9NRnRyITq8nYnASeQf3N+8TPzIYk7eBgxsL2g0QTaKt0bx85st8l/EdT/3yFJd/czl3nHQH14+4vnkSoCkhgaBrrqH07bcJvPIqvEeN9OwbVJQ+pgJEH6j87nty77sPQ0Q48a9+gFdiYpvn967JQ6cXDJ3cfgoHp8tJUVYGIeNG8PXhr8muzia7Khtbfg2/d17MU+l/Y2XZJnRCxyD/QZwafSojgkcwJmQMI4JHeHRW84IFCzx27u4yRWk1KVtWFebkQECbD7Hp6y+w2xowmrzQG3UMOimU1C2FTLc726RSP5IQgjkJc5gQPoHH1z/O3zf/neWZy3ni1Cea+2lCbruViq++ouDJJ4n/4P1+myyoKJ6gAoSHlb7zDgVP/xXvMWOIeeXl5rUWXNJFZUMlRdXF7FqXhXeSk/9mfkpJfQkldSVa9tK6UkrqS6gpK2NeTThflH3HvtWfIxCEGq2cV6ENeT0j0MwN/lNI1lvwQQfltVC2EVK3gJcfmP2PuAWATyBYw8HUs1FIX3zxBQDz5s3r0Xl6g87HiDHCQkNaqwlzScNwOZ0UpB4iZph2hZ+cEs6+tXlk7Cph8ElhXZ432DuY509/nsVpi3nilye4ZNEl3DX+Lq4cdiV6X19C7/o9+Y8spGrJEvwGQFObovQWFSA8xGarJ/3xR3F++hVlk4fxze9Gk7H1EQpqCyipK6GsvgyHdDC4+CRm1c3nE/Ea2Rv3oxd6gsyBBOt9CJYwqMFBUIl2lTvfXsiI7DKiHA68JJTZz6CWai7M+haRAwiddkNoP10OaKjUfnbE5Au+4WCNaPUzAnwjwS9S++kb0WEg2blzJzAwAgSA12B/qn/JRzpcCINOW4IUyDuwrzlARA8NwNvXyMGNBW4FCNBqE+cOOpcJERNYuHYhT294muWZy/nLKX8h8uKLKfvwIwqeew7rGWegM3c/IaCiDEQqQPQCh8vBvlJtfYLNBZtJK9jHJR9lM/GA5JsJgvemH8SckUOsbyyRlkhGBI9oXh3NtigCpz/8c8pMQopi8c/dji5jJzht2smt4Wyxj2IFNk4/7UYsEYmNX+BRNPynFK8wC+L6zI4LJyXYa6G+otWtEmqLoSofqgtafuZs0X7aa48+j5ef9rrWcLCEgCUULKEEeQM6A2Su12omTbUUozf0Q3OL1+AAqtfkYsusxGtQAD7+AfiHR7Tph9DpdQwZH86eNbnY6hyYvN3/GIT5hPHymS/z5aEv+euGvzJv0Tz+OOGPzH7gATKvv56St94i9LbbPPHWFKXPqQBxDOxOO1sLt7K5cDNbCrawvWg7dY46AEaIaO75qIrQdEnhzRcw8+or+I1vLMHm4Lbt07WlVG5YynvpOib6fkrS4o/A6AOR42DSzRCdAtHjwT+G4jf+hTljHT6z7mv+0nVW2nCU5GGZFNV5YYXQrv5NFvDrYl/QAkpDlRY0qnIbf+ZpPytzoaYY8ndBTSHUV3Bn03FvPdP2PDpDS7Dw8gWjRQsaRm+tLEZv7f0azGDwAr1JuzXdP2qbseVxm/3Mbe57JfqDgPrDFeji/ai1OQlMSCJr7y525ZSjEzr0OoH3EF+cK11sXJND/PhQTHodvmYjZqOuy34EIQTzkuYxKXISD695mEfWPsKPMdP5w8zTKXnjTQLmzVMpwZVfBRUgumFf6T6+OvQV36Z+S1lDGQJBcmAyFw65kJPDT+YkwyCqb16APa+e6BdfZMTMIzKJVBXA3kXaLX0NeysvAy5h2NQoOHkpxEwA/dF/kpKsTEJi2uZgakgrB8BrkP9R+/eIEGD2026hyZ3v67Dx0YfvgcPGlaclHVFLaayp1FdozVz2Oqgv14KNvVZ7bK/T7nfWBNZNEh16/kbN8n1UrX6aBmnEXBZMfWUg+S+ehzSaqMWLGpcXQvcbtn71E2tWrKIOL2ow0yC8ESYfhMmKzmzFYLZitvjh7x9IYEAgEYFWwv3NRPqbCbdG8ubsN/lw74e8sOUF7h5p4smf7BT+7Xmin32m68IqygCnAkQXyuvL+TbtW/536H/sK92HUWdkRuwMzh10LikRKfiZ/ABwlJWRef187Dk5xL3xOj4TJmgnkBLSV8HGN2Hft9qXYUgyrlP+wL7vpxE3IgDfix7p8PWllBRnZTSP42/SkFqB8NJjjOw8FYZHGUwcSMvW7g+59djP43KBswEcDeC0g7OBkvIq0gtLySutJK+0ioKyKorKq6ioqcEk7Rhx4oUNL2HH3yQJ9nIRaHLhb3QRVdWAX+VwqkNPxUtXS5xPPfsKJYFGM0lBlegdRegdtWyyrWRH1TncqV+Kt66yVXmA+sbbEeqlkRrM1EozhzBj0/sw2mTlcZ8wXvUr4csJTuZ9/TW6YU4ix41uGRRg9gfvgJb7fbhehqIcKxUgOrCreBdv7XqLFVkrcLgcDA8azgMTH+CcxHOOWjLTWVlJ1u9uxJaeTuyrr2jBob4Ctn8MG/8NxfvBOxAm3wrjroGwYWTtKqG6cjunntp5s091aQm2ulpCYtumv2hIq8AU74fQ9++wypCQkB4dL6Ukq6yeXbkV7M6tYHduJbtyKimublkaxGLyIzE0kkGJVsaFWkgMsRAT6EOYrxehvl6YjxiqWn+gjOK3dpE4+wXMyYE4HQ6W/+ZyKofMxe/6G5v3G5ZdxfbHN3L49FWMmuwHthrtZq9puW+rAVs1sqGaupoqaqvLqa+pxF5bib6+GnN9FdiqSC4r4F+ltfxvmKBshxfp7y/mYNH7nFbfTpQBLVA09uO07tPBEgL+MS03c0C/9OUoCqgAcZT0inT+ufWfLMtYhr+XP1cMvYILh1zYYSpoZ3UNWTfdTP2BA8T+60UsgwPg69/Djk+15pPoFLjwVRh5odbu3mjPmly8fY0kjOn8CzZn324Awge1TKJzVttwFNbhc3J4j99vT91+++3d2r+81sbWrHK2ZpSxJbOc7dnlVNVrTUx6nSApzMr05FBGRvkxLNKXwaFWwny9ujW/wJTgB3pBw+FyzMmB6A0GwgcNIe/Avjb7BUdbCYzw4cCWEkadkQA+HadMF4BP460jFXV2Tiup4ZB4leH/fod/FYfwaNIoQnLG4utw4E8NfqKWOO8GEoz1RFNNaG0l1up9GOvXIGpLgSMWTDT5tgSLgDgISdaa/kKGan1KKngoHqQCRKOi2iJe2f4KXxz8ApPexC1jb+H6Edd3ms3UVVdH9i23ULdzJ9EP34E191VYsRgM3jD6EphwA0SddNRxtZU20rcXM+aMGPSGzlNdZOzchpfFQvjglgDRNM7fK7GX+x+Owdtvvw3A/Pnzj3rO5ZIcKqpmU3oZWzK1W2qRllxQJ2BYhB/nj41iVJQ/o6L9SA73Pao2cCx0Jj2mWF/qD5fT9BuKTBrK1iWLcNjtGBrTqAshSJoQzoav06gqrcc3qGfDU/29jYyJCWD03feRtmEzN65O5bbk/eSPyOPcEQsIExM5WFjD5oIqPsirJC23BldjPLB6GRgR7kNKmJOT/WsZ6l1OJCUYqnKgIku7ZW/QaqZNTL4QkgShwyBsOESfrA1y8OrHZkflV+WEDxCVtkr+s+s/vL/nfRzSwWVDL+OmMTcR4t35lb2roYHs2++gdstmos4Px2/fH7XmgNMfgIk3dXo1un99Pi6XZPgpnTcvSSlJ37GVuFFj0bVqs25IrUAYdZhi+v+LICMjo/l+vd3JzpwKNqaXsim9jM0ZZVTU2QEIspg4OS6Ai0+O4eS4QMbE+GPx8ty/n9fgAKqWZ+Kqd6AzG4hqnFFdkHqI6KHDm/drChCHNhVy0uy4Ts7oPqHTEfnIQhouu4x/55zHo1EZPLf1Yc6IPYOHpjxEmI+W06nO5uRAQRV78irZ23h7d2ctLzdIwB+TPpDkiHGMivJn5Fh/RkX6MtyvAXP5ISjaB8UHoGg/HF4O2z9senEIHQ4x47VRcNEpWgBpZ/CDonTlhP2vkVLybdq3PL3haSoaKjgn8RzuOOkOYn1juz62oYGcG66iZtMeIieV4R/qgCl/gZTfaEM6u3jdPWtyiRzsT1AXaziU5mZTXVJM/EWXt9lua+5/6FmivZ4qqW7A7B9EfYOTi19Zy87sCmxOFwCDQy2cPSqClIQgUuIDiQ/26dM0FObB/lT9qAVT7xHBRA8bidDpSNu6sU2ACAjzISzel4ObCnotQAB4jx5FwGWXUf7557x1xSd8mrCBl7a9xIX/u5B7JtzDRUMuwtukZ2xsAGNjA5qPc7kkGaW17Mqp0PplcipZujufjzdmAS3NcKOjJzI6ZhajRvkzItIPs60McjZrt+xNsGcRbHlXO6mXHyRMg8EzYNDpEDxENU0pbjkhA0RxXTGPr3+cHzN/ZGzoWB6a9BDDg4d3faCUyAPfk3vvH6k+0EDENB0BNz2udTwb3WueyDtcQXlBLSefNazLfTN2bAMgYWxLM5Wzxo49vxa/WX2bXrqpuWhzRlnzLa24BhiEUS8Y5SuZf0oCKfGBjI8PJNjq1aflO5Ipzg8MOhoOl+M9IhgfP39iR4zmwPo1nHL5tW2CVdKEcNZ8foiSnOoOs+kei9C7fk/Vd99R9PhTzH//Pc6IO4OFaxeycO1CFh1exIOTHiQ5sO1QYp1OkBiidcSfP1arYUopySmvY1dOJbtyKtiRU8GP+wr5bLM2gqwlaEQwJuYqRk2/leERvpirMrRgkbEGUlfA/m+1F/GL0QLFoNNhyJmd1naVE9sJFyC+S/+OJ9Y/QbW9mgXjF3DdiOvcS2aX9jPyx79QsGgflQeshF11JoEP/V2bwNUNe1fnYjTrGXxy1ykeMnZsISA8Ev+wlklXtvTG/ofenv/QipSSvIp6tmeVsz27gu1Z5ezMqaC6QetM1pqLArl8Qiz1u37Ax6TnphsHVg4iYdDhleBHw+GWNvvkyafww5svU5KVQUhcQvP2YVMi+eXrNLZ+n8nM34zotTIYAgMJvXsB+Q8/QuWiRcTPnctbZ73FFwe/4B9b/sFlX1/GFcOu4LZxtzUPl273vQhBTKAPMYE+zBml/S9IKcmtqGdndgU7c8rZmVN5VNBIDvdldPQwRkdPYtTYRxnhXYpXxk+QuhL2fQPb3gehh4RTYPgFMOw8Lb2KojTyaIAQQswB/gHogTellE8f8fxpwAvAGOAKKeXnnipLeX05T/zyBEvTlzIyeCRPnPoEgwMGd31g5i+w/C+QvorS1CjKDlgJuu4agh54sNvV9IY6B4e2FJI8SVu8pjNOh4OsPbsYMe30tudIrQCDDlNs76TZdrok6SU17M2rZF9eFXvzKtmRU0FRlTbM1KgXjIj046KTohkT48/4+EASQyzNV+CPrXyfis5eoB95Dfan8rsMnNU29FYTQyZM4Yd/v8KBX9a0CRBmi5GRp0Sxc2U2k+YO6nFndWsBF19M+WefU/CslqdJ7+vLJcmXMDNuJi9ufZEP937IkrQlLBi/gPMHn+/2+hxCCKIDvIkO8G4TNLSaRgU7cyrYkV3Bsj0FfLqpdU0jiVHRKYyZ9igTvDIYUroK4/6vYfE92i1mIoy4AIafD4EJnZRAORF4LEAIIfTAS8AsIBvYKIRYJKXc02q3TGA+4NF1K1dmreTRtY9SYavgjnF38NvRv8Wo6+LKP3cbLH8cDi0DSyjlvtdTuGEZfuecQ9j9DxxTe/rBjQU4bC5GTO065UXewX3Y6+uIH9N2FFRDWgVecb6ILkY/HcnhdJFTXkdqUQ2Hi6o5WFDNvvxK9hdUUW/X+g30OsHgUAunDglhXGPb+PBIX7wMHdewoqOju1WOvuQ1OADIoCG1Ap8xoVgCAokZPpID69cw9dKr2+w7dmYsO1dms/2HLE69LKnd8x0LodMR8cgjpF96KUUvvkjEgw8CEGAO4OEpDzMveR5Prn+SP635E58f+JwHJz3oXnNne6/Vpqah1QRaB41dOZXszKlgxb5CPm+saehECoNCT+fMxFJmiQ0MK/8J6/d/gu//pHVwj7sSRs5TzVAnKE/WICYCh6SUqQBCiI+BuUBzgJBSpjc+5/JEAewuO3/f/Hfe2/MeyYHJvDbrtQ7nMzQrPqgFhj3/00YlzXyU6oaR5P3fAnymTCby6acQuu53Dksp2b0qh+BoC2EJXV/9Z+zYihA6YkeOad7mqnNgz6vB94z2O1Pr7U6yy+rIKa8ju6yWzNJaUotqSCuuIaOkBruzZYx9oI+R4ZF+XD0pnmERvgyP9GNImLXbw0x/97vfdWv/vmSK9kV46ZsDBEDSxFNY8fZrlORkERzdMiDBN8hM0oRwdq/JJeXcBMydLOHaXd6jRhJwxeWUvf8BAfPmYR7W0v80Mngk753zHl8d+ooXtrzAFd9ewdzBc7lt3G1EWHqez6mjoJFfqTVP7c6tZHduBV/lOHit8lTgVGJFAVdatzK3aBXR396Na8n9NAw+C3PKNYghM7vdrKocvzwZIKKBrFaPs4FJx3IiIcRNwE0AcXHujTQpqCngnp/uYVvRNq4cdiX3pNyDSW/q+ICKbFj5NGz7UEsAd9q9MPVO6g5kkH39fLyGJhPz4ovoTJ2coxMZu0oozqpmxjXD3Kp9ZOzYRsSQJMwWrdO03u6kcGchOgmbpYPDP6eSX1lPfkU92eV15JTVUlxta3MOo14QH2xhUIiFmcPDGRRiYVDjTOQgi6lXRhW9+uqrANxyyy09PldvE3qBV6I/DYfLm7clTZrCirdf4+Avawme13Z02Emz49j/Sz67fsom5ZxEelPY739P1ZKl5P/5L0ctLKQTOi5Kuogz4s7gtR2v8fG+j1mctphrhl/DDaNvwNfUu6v2CSGI9Pcm0t+b2SNbglBxdUNzwNidM47PcufhU7Wbi/WrmHvgJ7wPfkOFLoC9oXOoHH4V0cnjGBJm7bSGqRzfjotOainl68DrACkpKbKL3Vmft54//vxH6hx1PHPaM5ydeHbHO9eUwOrnYcMbgISJN8K0u8EaRkNaGlk334IhOJi4115Dbz22ES5SSjZ+m45vkPmoVeNcLklRdQN5FfUUVNZTWFlPQVEZHDpAweBpvPvCzxRU1lNWa+dWvLgUE7cs34cNMBt1RPpr7dAzh4cTHeBNTJA30QE+xAR6E+5nRq/z7HDGgoICj56/p7wG+1O/rxRnRQN6fy98g0KISh7OgV/WMPmIABEcbSV+dDA7VmQzbmYcBlPvffHpAwIIu+du8v70MBVffUXAhRcetY+/lz/3TbiPq4ZdxYtbX+Tfu/7Nfw/+l5vG3MTlQy/v/AKnF4RYvZieHMr05JYRctUN09iffxVLsktwHVjGkLyvScn/DFPBx6xfPpx7XbM4HHQ6SVFBDIv0Y1iEL8Mi/Aj3697sd2Vg8mSAyAFaTyqIadzmMS7p4o0db/DStpcY5D+I/5z1HwYFDGp/5/oKWPcyrHtJy70z9ko4/X4tnQFgLywk63da3p64N9/AEHpsw0odThdbN+RRmF6JeUoIzyzbT155PXkVdeSWa0HB4Wob85JqU5mDJN8SS2yQD+PjA4nwMzNzUylOk55vrjqJcD8zfmZDv38I4+Pju96pH2n9EFCfWoGlcXGg5MmnsPLdNynLzyUwom1/0Mmz4/jyb1vZuzaP0afH9GpZ/OfNo+yzzyh89jl8zzgDvV/7I5difGP462l/5fqR1/P3zX/nmY3P8MHeD7jzpDs5O/Fstzuye4PVy8D4+CDGxwfBKUnAbTgqCyhe+x9G73iPf9b+k4rq9/jy4Bm8se00ctA+J/7eRoaG+5IcYWVohB9Dw30ZGu6Lv49qnjqeCCm7vCA/thMLYQAOAGeiBYaNwFVSyt3t7Ps28I07o5hSUlLkpk2bjtpeXl/OA6sfYHXOas4ddC6PTH4EH2M7mXNsNbDhdVjzD6gr04b3zXgIwlrahZ3l5WRcex22nBzi33kb79Gju3y/ZTU29uVXsT+/ktTiGtJLasksqSG7tI7LK01YXPCmXwMGg44IfzMR/mai/M1EBnhrP/29ifA3E+bnxbaP3mT/2p+47c2P0Bu0GO5qcJD72Dp8p8fif1ZCl+VRNNIlyXt8PebhwQRdqs05qCwu5I3bf8u0q+Yzce4lbfeXkv8+s5m6KhtXPzYZXS9PRqzbvZv0Sy/D/8ILiXryCbeOWZuzluc3P8/+sv0MCRjCLWNvYVb8rD4NFO1yueDwj1pCyoPfIYGy6BlsiriclbbhHCioZn9BVXOuLYBQXy+Sw60khfmS1PgzOdxKgI9na0cKoKUU6xaP1SCklA4hxB3Ad2jDXN+SUu4WQvwZ2CSlXCSEmAB8CQQC5wshHpNSjuzua+0u2c2CFQsoqivi4ckPc2nypUdfWTsaYPPb8PNz2mI3Q2bBGX+CqHFtdnPV1JB5881aZtbXXzsqODhdkkOF1ezMqWB/fmVjUKiisKol+6iv2UBCsIWR0f6cHxGIeU0pcbOiWTsznjBfL3RdNPtk7txK7MgxzcEBwJZRBS7Pzn84Fi+99BLQ/aR9fUXoGvshUsubt/mFhBExOIkD69ccFSCEEJx8VjxLXt3J4a1FJKX0bkJE75EjCb7xRkpeew3rtFPxO7uT5s9GU6OnMjlqMt+lf8cr21/hnp/uISkwidvG3sYZcWf0X6DQ6SBplnYrz0Rsfpugze8wO/tHZoeNgMm3IkddQl6tYH9BFQfyqzhYWM3Bgio+3ZRFrc3ZfKoQq4nBoVaGhFlbfoZZifI393st+UTm0T4IKeViYPER2x5pdX8jWtPTMfvy4Jc8vv5xgr2DeffsdxkVMqrtDk47bP8IfnpGS3gWfwpc9i7ETznqXK6GBrLuuIP6XbuJfuHvWKZMobi6gW2Z5WzNKmNbVjnbs1omjJkMOpLCrJyaFMKwCF+GRmhtsE3ZR6WUfPncFqoCvTjngiT0xq4/yOX5eVQUFjD+3AvbbG9IrQCdwBTf8YSq/lBcXNzfReiS15AA6naX4Citx9A4xyFp0ims+vBtKgoL8A9rGwQSx4QQGOHDlu8yGDI+rNe/oELvuJ2a9evIe2Qh3mPGYHRjqLBO6Dg78Wxmx89mafpSXt3+Kn9Y+QeGBQ3j1rG3MiN2Rv9+kQbEwZmPwGn3wa7/wvqXYdGdiB8eJSrlBqIm3MCMoS3zjlwuSW5FXXPAOFRYzeGiGr7ZkdecvwvAx6RvHFhhbR5kMTjUSmKIxaO5vBTNcfsbtjltPLXhKT4/8DmTIyfzzGnPEGgObNnBadfWY/j5WSjP0LKqnv8PGHxGuxPcpMNBzt13U7tuPbm33sdb5WGsf2Y5WaXaUqJ6nWB4pC8XnhTFSbGBjI31JyHYgqGTJoic/WXkHa7gtCuS3QoOABk7twIQP+bkNtsb0iowxVjR9WLHaW9ITu5i1bkBoKkfouFwOYYgbZBAcmOAOLhhLSnnXdRmf6ETjJsVx4r39pG9t4zYEb07B0AYjUQ/9xxpF15Ezr33Ef/uOwiDex9FvU7PuYPO5ayEs1iStoRXt7/K71f8nuFBw7l5zM3MiJvRv01PRjOcdDWMu0pbKGv9K9pncPXfYdTFMOV2iByDTtcy/HbG0JasAlJKSmpsjQGjujlwbM0s45sdubRuEY/wMzePymt9iw3ywdjPecp+LY7LAJFfk8+ClQvYWbyTG0bdwJ0n3dmSLuPIwBA5Ds7+KyTPaTcwlNXYWH+oCNdf/8KgLT/xyui5LMoLw7+8gMmDgrh2cjzjYgMZHe2Pdze/nDd+m47F38TwU9xPX5C+fSu+IaEERrZ0nrpsTmzZVfieOvAmpV155ZX9XYQuGUK90fkaqT9cjmWCFiACIiIJSxjMgV/WHBUgAIZOjGDDolS2fJ/R6wECwBQbS8SjC8m99z6KX3mV0Dvv6NbxBp2B8wefz9mJZ/NN6je8seMN7lp5F0MChnDTmJuYHT/bvRQyniIEJJ6m3UoOwy+vwdb3YcfHkDgdpt4JQ2Ye9ZkUQhBi9SLE6sXkQcFtnqu3O0kvqSG1qIbUomptjk/J0bUOvU4QG+hNQqugkRCs/YwK8Pb4yL5fk+MuQNTYa7j8m8upd9Tz99P/zsz4xnWf2w0Mz0DyWW3+CaWU7M6tZPm+QpbvK2R7Vhk37/iKuamrWX3qRYy+7nfcNDiYEZF+XfYVdCZnfxm5B8s59bIkDG5OPnM5nWTt3kHSpFPaNBfYMivBKTENsP4HgBdeeAGAu+66q1/L0RkhBF6DA2g4XI6Usvl3mzz5FFZ//C5VJcX4BrdN76436hhzZizrvjhMYUYlYR5o2vM//3xqVq+m+JVXsEydgs/48d0+h0Fn4MIhF3LeoPP4Lv073tjxBvf9fB8v+b3E70b/jnMHndt11gBPCx4M5zwDMx7Q+gF/eQ0+uERLQz7ldhh9mVvJLs1GPcMi/BgWcfTfoqzGpg0OKdYmhjbdNqSVtunrMOl1xAX7NAYMHxJDrCSE+JAYYiHc19yjz/yv0XEXIDIqM0jySuI/c/7DIP9BYKuFbR/A2n9CeabWlHTOs5A0uzkwVDc4WH2wiOX7Clm5v4jCqgaEgDExAfyzYRNDUlfjf911/O6B+3utHXfjt2n4+JkY2cWSoq3lHz5AQ21Nm+yt0Nj/IMBrgPU/AFRUDNRMTG2ZBwdQt60IR1EdxjBtdFvSJC1AHNywlpPPvuCoY0ZNi2bzkgy2fp/JWTeOOur53hD+8CPUbt1Gzr33MujLL9H7H9tFgEFn4NxB53J24tksz1zO6zte5+E1D/Pq9lf57ajfMnfIXLz0/ZthF+9AOPUPMPl22P0FrP0XLLoTfvyLtoZKym/BEtz1edoRaDEx3mJifHxgm+1SSgqrGkhrCh4lNaQV1ZBeUsPPB4uwOVqSOJiNOhKCLa2arbS+jkEhFgItJ+YoK48Nc/WU8KHhMnVXKhaHDTa+CetfhdpiiJ2kTXBrDAxFVQ0s21PA93vyWXuoBJvTha/ZwGnJocwYGsb05BDkG69Q8vrr+M+bR+QTj/dacMg9WMaXf9vKqZcmMfbMrteXaLLu849Y+/mH3PbGB3j7tgSDwte2I20uwu88enW6/vbFF18AMG/evH4uSeccFQ3k/3UD1qnRBJzXMjfmnXtux8ti5YrH/truceu/OszmJRmcffNoBp3kmRTrdTt2kH7V1fjOnEn035/vlf9DKSWrclbx2vbX2FG8gxDvEK4bcR2XDb0Mi7HzdUj6jJSQ9pMWKA4t0zIYjL0CJt8GoV2kxOkFLpckr7K+udbRuvaRWVrbZn5SgI+xsZNcG2U1OFS7Hx98XPV3dPsf67gLECknjZWb/no+bPoP2Kq1gHDqHyB+KhklNXy3O5/vdxewObMMKSEuyIfZI8KZOSKc8fGBGPU6pMNB3qOPUvH5fwm49FIiHl2I0Pdee+1XL2ylJLeGax+fgrEb/RYfL7wPh83GNU+90LxN2l3kPLYW6+SoNl9sSveVfnaA2u1FRP5xAnpf7Ypw7Wcfsu6/H3HLq+9iCQg86hin3cUXf9tCWV4Nl9yf0uUiT8eq+PU3KHr+eSKfeJyAiy/utfNKKdmYv5E3dr7B+rz1+Jp8uXLYlVwz/Jq2gzr6W+E+beTT9o/B2aB9rqfcrvVX9MPoLLvTRXZZHalF1aQV13C4qIa0Yq3DvKjVkHaDThAX7MOQxqG5TXM7Bodau91n2QdOgAARbZSbbrLCqIuRU/+P/SKBpbvyWborn335VQCMjPJj9ogIzhoVztBw3zZXZK76enIW3E318uWE3HYrIXfe2avDA/MOlfPFc1uYevEQTprl/gplDbW1vHTDFUy44GKmXXl98/aajfmU/fcgIb8ZiXnowMuo+fzzzwOwYMGCfi5J1xzFdeQ/v6lNLaI4K4N37rmdM2+4jXGz21/Torqsnk+f3IiXj5FL7k/By7v3W2aly0Xmb2+gbvt2Ej76sE1Cv96yq3gX/975b37I/AFvgzcXJ13M9SOv75WkgL2mphg2vaVNZq0pgvBRMOkWbY13o3d/lw6Aynp7c0f54aJqDhfWcKiomvTimuZahxAQE+jNkFAryRHaLPLkcN9jSojZi06AAJEULt//ahlfZZpYuiuftOIahIAJ8UGcNSqC2SPCiQ1qZwY14KysJOu226jbvIXwhx4i6Jqr293vWDnsTm3eQ2k91z4+FaOX+/8Ihzb9wlfP/oVLH36SuFFaBldXg5P85zZhCPAi9LaxA3LC0GOPPQbAwoUL+7kk7in9dD91O4uJuE+rRUgpeXvBrViDgrj04Sc7PC73YBlf/X0bcaOCOeeW0QgPdGbaCwpJv/xyZH09cW//xyNBAiC1PJV/7/o3i1O1KUpzEucwf+T8rjMd9yV7Pez6XEuFU7hH6784+TpIuQECB2Z6F7vTRUZJDQcLqrX5HY1zPA4XVTdnUtYJSAixaKlHGvNWDY/0JTbQpy86yH/9AcIaM1SGXPM8ep1gyqBg5oyKYPbIcMJ8Ox8FYS8oJOvGG2lISyP6r0/jd07vroAmpWT5e/vYtzaPs28ZzaBx3Wuv/vGtV9i18gdu//fHGIzaqJOKZRlU/ZhJ6K1jB2QHNcDixdqXzDm9/Pv0lPZqEWs+eY9fvvyMG/75epvV+460Y0U2qz45wMTzE5lwbu9me21iy8gg4/r5Hg8SALnVuby35z3+e/C/1DnqmBo1lfkj5zM5cvLAuRiREtJXazWKfd+CdMHQs7WkmoNmHBdrazcFjn352mzy/QVVHCioJr2kpnleh8Wkb052ODzSj+GN93t5MuCvP0CEJA6XL3/2PbOGh7s9sqAhLY2s392Is6yMmH+9iGXq1F4v166fsvnpowOknJPApAu611cgpeQ/f7iFgPAI5j2gXZE7KxvIf3YT5mFBBF99bAvIKO07shZRWVzIO/fcTmj8IC5b+CS6DuYPSCn58Z297F+fz7m3jSFhTEi7+/WULTOTjOuu75MgAVDRUMFnBz7jg70fUFxXzLCgYcwfOZ/ZCbP7f4hsaxXZWt/j5re1gSnBSTB+vtaxbfHM38KT6mxODhRoqzjuzatkb752vyl3lRCQGGJhRKQfI6P8GRHlx8goP0KOfb33X3+A6ChZX0cql35H/sKFoNcT+9preI/u/eGKuYfK+er5rcSOCOLc28Z0u/lhx49LWfb6v5h10x2MOXMO0Nihuq2QiAXjMQQPjLbX9jz33HMA3HOPRxcF7FX24joK/rYJ66nRBJyrBfPdP/3I0pf/3m4Cv9YcNidfPLeFisJaLn1gAgHh7Tdn9pQtM1OrSdTWakFiuOcvEmxOG9+kfsPbu98mrSKNMO8wLh16KZckX0KI9wD6AnY0wO7/wcY3IHsj6IxareKka2HImdCfEwR7qGmt8T25lexpWpsjt5Kc8rrmfcL9vBgV5c/IaH9GRfkxMtrf3ZxVKkA0cVZUkP/4E1R+/TXmkSOJ/ttzmBISer081WUNfPrURkxeei59IAWvbqYzLsnO5P0H/kDU0OFc8uCfETodttxqCl/c2uYLbKA63vogmpR+sp+6XW37Ir7++1Mc3rSBq598nrCEjn/vlSV1fPbUJrytWqd1V+uLH6s2QeI/b2EeMcIjr3Mkl3SxKnsVH+37iDW5azDoDJyVcBZXDruSMSFjBk7zE0DhXm2G9vaPoLYEfKO0NB8nXQ1BA/uz0x0VtXZ251U0Bg0tcBwqrKZpJG6gj5FR0f6MjPJnVLQfo6L8iQs6ql9DBQiA6tVryHvoIRzFxYTceishN9+EMPZ+Vdlpd/Hl81soya3hkj+OJziqewsK2W0NfPjgAmoqyrnumRexBgYhpaT4zZ3Y82qIuCcF3QDPn79ixQoAZsyY0c8l6R57US0Fz2/GOi2agHO0L5K6qkreufcOzBYr1zz1AoZOVg/M3l/Gon9sIzo5gFm/HYmPn2cmUtmysrTmpj4OEk3SK9L5eP/H/O/Q/6ix1zAieARXDruS2fGz20+n318cNjiwFLa+B4d+0PoqYibAyItgxFzw7921PQaCOpuTvfmV7G613vjBwqrmDnFfLwMjovwYFa0FjYtOijmxA4SrtpaCZ5+l/KOPMQ0eTNTTT3ukSanJig/2sWdVLnNuGsXgk8O6PuAIP7z5MtuXLWbeA4+ROE5Ls1C3t4SSd/bgf/4gfE8ZeLmXfk2aaxF/nIDeqn3Bp23bzBdPLeTkc+Yy4/obOz1+79pcfvrwAEYvPdOvGsqQ8d3/H3BHU5Bw1dQQ8dCD+F1wQZ9fxdfYa/jm8Dd8tO8jDlccxsfgw6z4WVww+AJSIlL6f22K1ipyYOensPtLyNuubYud1BIs/NzPbnC8aXA4OVhQza6cCnblaoFjb14lDQ4X6U+fe+IGiJpfNpD38MPYs7IIuv56Qu/6PTpz1/ldjtXuVTms/GA/J8+JZ8qFg7s+4AgHf1nLouefJOX8eUy/5rcASKeLghe2gITwu05GGAbQh64DzzzzDAD33XdfP5ek+1pqETEEtFqD+se3XmHbd99yyUOPEz9mXKfnKM2r4ce391CYUcWQlDBOuyIZb2vv1yZs2dnk3n0Pddu34zNpEhELF+I1yDMjqTojpWRTwSa+Pvw132d8T429hkhLJOcNOo/zB59Pon/fl6lTJYe1QLH7f1CwU9sWO0mbiDdkJkSM0da1+BVzOF0cLqphaITviRUgpNNJ1fLllL71H+q2bsUYHU3kU09imTjRo2VI31HMktd2EjM0kHPvGNvt8cuVRYW8+8c7CYyI4oo/P4PeoDUjVa/LpfyrwwRfOwLvkceWk6avHa99EE3aq0XYG+p5//67sNXXcd2z/8Lb6tvpOVxOF1u+y2Tjt2l4WYycftXQbg9zdod0uSj/9DMKn38eWVdH8I03EnzzTei8+ifHUp2jjhWZK1iUuoh1uetwSRcjg0dyRtwZTI+ZTnJg8sDqryg+qAWKvYsgf4e2zRIKg8/UgsXgGcflaKhuODECxIZVq6j43/8oeftt7BmZGKOjCZo/n4CL56Hz8Vy7aEVRHWs+P0ja9mICIy3Mu+dkzJbu9RG4nE4+eewBijPTuPbpfxIQoaUCd9U7yH92I4YwC6E3jR5YH6xOrF27FoCpHhg63Bc6qkUUpB7iwz/dTdLEqZz7+/vc+nsUZ1fxw9t7KcmuZuikCE69NAmztff7kBzFxRT89Rkqv/4aY3wcEY88gvWUU3r9dbqjqLaIxWmLWZK2hN0l2qrC4T7hTI+ZzvTY6UyMmIjZ4LkafbdVF8Lh5Vp/xaEfoa4UEBA5RqthxE7S+jAC4o6LuRZu+vUHiHFxcfLT6BicZWWYR48m+Ibf4jtzptsLrhwLe4OTzUvT2bYsC6EXpJwdz7gz49xeBKi1NZ+8x/ovPuGcO+9h+KmnN2+vWJJG1U/ZhN0xDlNM51esSu8q/XgfdbtLCJg7GJ+TwhF67XO0/otPWPPJe5xzx90Mn+ZeJ7zT4WLT4nQ2L81ApxMkjg1h6OQI4kYE9fr61jXr1pH/6GPYMjLwnTUT/wsvxHLqqf1Wo2hSXFfMquxV/JT9E2tz11LnqMOsNzM+YjzjQscxNnQso0NGYzV1b1CHx7ickLcNDi3XkgfmbAF7jfacNVwLFLGTIPpkCB1+zBlnB4Bff4AYZfaWS37zG4J/+xu8U1I8eqUtpeTQpkLWfnGI6rIGkieGM+WiIVgDj+0DmLlrO589/idGnnYmc267CyklDQfLqVqdQ8OBMnxOCiPo8gGU7sANTz/9NAD3339/P5fk2DnK6yl5fy/27GoMod74zY7He1QI0uXik0fvJ//wAYakTGbs7HOIHeneMM+SnGp2r87l4IYC6mvs+PiZSJ4UwbDJEQRH994Xo6uhgZI33qTsvfdwVlSg8/HBOmMGvmfNxjptGjrv/p1DY3Pa2FSwiZ+zf2Z97npSK1KRSASCwQGDGRs6ljGhYxgRPII437iBMTLK6dDSe2T9os2zyNoAZWktz1tCIWy4FizCGm9Bg7XmqYFd2xhYAUIIMQf4B6AH3pRSPn3E817Au8B4oAS4XEqZ3tk5x48eLTfv3OmZAqNNhCrOrqYos4qDmwrIO1RBaJwv0y5LInJIQLfOJaWkKCON1M0bOLxlA/mHDhAYFcPVjz2HfU8V1WtycBTVofM1Yp0chfXUaHTdyN80EBzvfRBNpJTU7y6h4vt0HIV1GKOt+J+VgDMMNn79BbtX/kB9dRWBkdGMmTmHkafP7LJvArQaRcbOEvatzyNjZwkulyQ0zpeYoYEEx1gJibESEOGDvoe1C2m3U7NhA1VLv6Pqhx9wlpUhvL2xTp+O5ZSpeA0ahCkhAX1QUL82X1bZqthZvJPtRdvZUbSDHUU7qLRVNj8fYYkg3i+eBL8E7eafQJQlimDvYPxMfv1X9upCrd+icJ8296Jor3a/qaYBYPCGgFitWcq/8WdAnFYLsYRogcU7sD8n8g2cACGE0AMHgFlANrARuFJKuafVPrcBY6SUtwghrgAuklJe3tl5uzuTujUpJU6HC3uDE3u9E3uDk4Zae3NAKMqsojSvFtk4+8Tib2LCeYkMPyWqy45op8OBw2bDYWugMD2Vw5s3kLplAzUlpRiFicjEocQPG0e87whsOyqQdQ6M0Vasp0bjMzrkuBix1J7t27VhhGPHju3nkvQO6ZLUbi2kclkGzvIGTIn++E6LRvoIMvZuY+eaZWQf3IXOaGDolGkMSZmM2dcXs8WKl8WKt9UXg5dXu19kdVU2Dmws4MCGAoqzq3A5GhO46QWBkRZCYqwERVnw8TPh5WPEy8eAl48Bc+N9g5vpo6XDQe2mTVQuXUrVsh9wlpQ0P6fz88OUkIApIR6vxEQMoaHorL7ofK3ofX3RWX3R+1rR+foiOngfvcklXaRXpnOw7CDpFelkVGaQXplOekU6VfaqNvuadCZCvEMI8Q4h2DuYEO8Q/Ex+WE1WLEYLVmPbnya9CS+9Fya9SbvptMcGnaF33pfLBRWZWqAoS4PyLO1xeaZ2v6706GOEDryDtGBhCQEvP/DybbmZGx+bfLVV9gyNN6N3y0+9qfFmBJ2h8adR+9n5+xpQAWIK8KiU8qzGxw8ASCmfarXPd437rBNCGIB8IFR2UqhREUPkp1c/3+HrdvpuOntSNP72hLZEpWh60NnBUgs6SNlmKUud0GPUeWHQmdBxxBe/AO9RIVhPicIU349XREqnpMNFzYZ8Kpdn4qq2H/W8SzhpcNThcNlp+t9o/rcVojndSvt/XwFI7ajm/yGaH3fKjX+X9nZpPu1x1qSs9J6Rz8/t9peNJ5ccjQayWj3OBiZ1tI+U0iGEqACCgeLWOwkhbgJuAhgWlkiNqO/0hY/6btdOgqAxwDbebwoGOr3oOH+SOPKhaN6u0+kRej06nQ6h06HT6xE6HUZvM9awEPQ+RoSXHp1Jj/AyILz0mGKsGAIH0GiOHnrySS1F9oMPPtjPJeldwqDDOjUKn5Rw7NlVuOqcuOoduOodyDoHrnonjpoGGqqqcdoduBwOnI62P6V0NQYN2fwNLVvdb4+UEumi8aKj8bFseaydg7Zf9J2f8og31ubVtAObT9xylqOu0QZsYGn/zctOHvXSq/axbrxiLxbuuFiTWkr5OvA6aE1ME567rJ9LpDSx24++uv410Zn0eA0K6O9iKEq/8GSAyAFaL8gc07itvX2yG5uY/NE6q5XjxHXXXdffRVAUxUM8GSA2AklCiES0QHAFcNUR+ywCrgfWAZcAyzvrf1AGnsTEAZZaQVGUXuOxANHYp3AH8B3aMNe3pJS7hRB/BjZJKRcB/wbeE0IcAkrRgohyHHniiScAeOihh/q5JIqi9DaP9kFIKRcDi4/Y9kir+/XApZ4sg+JZDoejv4ugKIqHHBed1MrAddttt/V3ERRF8RAVIJQeCQ3t/ayliqIMDMfn9F1lwHj88cd5/PHH+7sYiqJ4gKpBKD3idDr7uwiKonjIcZfNVQhRBezv73IMECEcMev8BKZ+Fy3U76KF+l20MEspu7UG8/FYg9gvpUzp70IMBEKITep3oVG/ixbqd9FC/S5aCCG6neVU9UEoiqIo7VIBQlEURWnX8RggXu/vAgwg6nfRQv0uWqjfRQv1u2jR7d/FcddJrSiKovSN47EGoSiKovQBFSAURVGUdh1XAUIIMUcIsV8IcUgIcX9/l6e/CCFihRArhBB7hBC7hRC/7+8y9SchhF4IsVUI8U1/l6W/CSEChBCfCyH2CSH2Ni79e8IRQvyh8bOxSwjxkRDi17OMoxuEEG8JIQqFELtabQsSQiwTQhxs/BnY1XmOmwAhhNADLwFnAyOAK4UQI/q3VP3GAdwtpRwBTAZuP4F/FwC/B/b2dyEGiH8AS6WUw4CxnIC/FyFENPB/QErjxDA9J95SAm8Dc47Ydj/wo5QyCfix8XGnjpsAAUwEDkkpU6WUNuBjYG4/l6lfSCnzpJRbGu9XoX0JRPdvqfqHECIGOBd4s7/L0t+EEP7AaWjrrCCltEkpy/u1UP3HAHg3rlTpA+T2c3n6lJTyZ7Q1dlqbC7zTeP8d4MKuznM8BYhoIKvV42xO0C/F1oQQCcBJwC/9XJT+8gJwH+Dq53IMBIlAEfCfxia3N4UQlv4uVF+TUuYAzwGZQB5QIaX8vn9LNSCESynzGu/nA+FdHXA8BQjlCEIIK/Bf4C4pZWV/l6evCSHOAwqllJv7uywDhAE4GXhFSnkSUIMbzQi/No1t63PRAmYUYBFCXNO/pRpYGpd27nKOw/EUIHKA2FaPYxq3nZCEEEa04PCBlPKL/i5PPzkFuEAIkY7W5HiGEOL9/i1Sv8oGsqWUTbXJz9ECxolmJpAmpSySUtqBL4Cp/VymgaBACBEJ0PizsKsDjqcAsRFIEkIkCiFMaJ1Oi/q5TP1CCCHQ2pn3Simf7+/y9Bcp5QNSyhgpZQLa/8NyKeUJe6UopcwHsoQQQxs3nQns6cci9ZdMYLIQwqfxs3ImJ2BnfTsWAdc33r8e+KqrA46bbK5SSocQ4g7gO7RRCW9JKXf3c7H6yynAtcBOIcS2xm0PNq4BrpzY7gQ+aLyISgV+08/l6XNSyl+EEJ8DW9BG/G3lBEu5IYT4CDgdCBFCZAMLgaeBT4UQNwAZwGVdnkel2lAURVHaczw1MSmKoih9SAUIRVEUpV0qQCiKoijtUgFCURRFaZcKEIqiKEq7VIBQlA40Zka9rfH+6d3NFiuEmC+EiPJM6RTF81SAUJSOBQC39eD4+WipHhTluKTmQShKB4QQTRmD9wN2tNxGxcAoYDNwjZRSCiHGA88D1sbn56NNZnwbLR1MHTAFuBc4H/AG1gI3S/UBVAYwFSAUpQONmXK/kVKOEkKcjpaaYCRa6ug1aF/4vwA/AXOllEVCiMuBs6SUvxVCrATukVJuajxfkJSytPH+e8CnUsqv+/ZdKYr7jptUG4oyAGyQUmYDNKY4SQDK0WoUy7S0P+jRUky3Z4YQ4j609QmCgN2AChDKgKUChKK4r6HVfSfa50cAu6WUnS7t2bjk5ctoq5xlCSEeBU6oZTCV44/qpFaUjlUBvl3ssx8IbVr7WQhhFEKMbOf4pmBQ3LiOxyW9XVhF6W2qBqEoHZBSlggh1jQu/F4HFLSzj00IcQnwz8YlPw1oq9ztRuukflUI0dRJ/QawC201r4198iYUpQdUJ7WiKIrSLtXEpCiKorRLBQhFURSlXSpAKIqiKO1SAUJRFEVplwoQiqIoSrtUgFAURVHapQKEoiiK0q7/B8JbnFK1JrNNAAAAAElFTkSuQmCC\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "distance = pyabc.PNormDistance(p=2)\n", "\n", "abc = pyabc.ABCSMC(model, prior, distance)\n", "abc.new(db_path, observation)\n", "h_uni = abc.run(max_nr_populations=7)\n", "\n", "plot_history(h_uni)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Next, we consider again a Euclidean distance with adaptive weights." ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "ABC.Distance DEBUG: Fit scale ixs: \n", "ABC.Sampler INFO: Parallelize sampling on 4 processes.\n", "ABC.History INFO: Start \n", "ABC INFO: Calibration sample t = -1.\n", "ABC.Distance DEBUG: Scale weights[0] = {'s1': 4.0250e-01, 's2': 1.4861e-01, 's3': 1.4870e+01}\n", "ABC.Population INFO: Recording also rejected particles: True\n", "ABC INFO: t: 0, eps: 4.47989200e+01.\n", "ABC INFO: Accepted: 100 / 198 = 5.0505e-01, ESS: 1.0000e+02.\n", "ABC.Distance DEBUG: Scale weights[1] = {'s1': 3.6312e-01, 's2': 1.5042e-01, 's3': 1.4881e+01}\n", "ABC INFO: t: 1, eps: 4.38143932e+01.\n", "ABC INFO: Accepted: 100 / 344 = 2.9070e-01, ESS: 9.7743e+01.\n", "ABC.Distance DEBUG: Scale weights[2] = {'s1': 4.3046e-01, 's2': 1.4578e-01, 's3': 1.5332e+01}\n", "ABC INFO: t: 2, eps: 4.44628941e+01.\n", "ABC INFO: Accepted: 100 / 557 = 1.7953e-01, ESS: 9.5587e+01.\n", "ABC.Distance DEBUG: Scale weights[3] = {'s1': 4.5866e-01, 's2': 1.4819e-01, 's3': 1.4176e+01}\n", "ABC INFO: t: 3, eps: 4.04222910e+01.\n", "ABC INFO: Accepted: 100 / 1450 = 6.8966e-02, ESS: 9.7915e+01.\n", "ABC.Distance DEBUG: Scale weights[4] = {'s1': 4.3275e-01, 's2': 1.4706e-01, 's3': 1.3978e+01}\n", "ABC INFO: t: 4, eps: 3.94279313e+01.\n", "ABC INFO: Accepted: 100 / 2733 = 3.6590e-02, ESS: 9.3574e+01.\n", "ABC.Distance DEBUG: Scale weights[5] = {'s1': 4.6633e-01, 's2': 1.5571e-01, 's3': 1.5107e+01}\n", "ABC INFO: t: 5, eps: 4.20907747e+01.\n", "ABC INFO: Accepted: 100 / 6682 = 1.4966e-02, ESS: 9.6457e+01.\n", "ABC.Distance DEBUG: Scale weights[6] = {'s1': 4.9762e-01, 's2': 1.5082e-01, 's3': 1.5126e+01}\n", "ABC INFO: t: 6, eps: 4.17007190e+01.\n", "ABC INFO: Accepted: 100 / 16089 = 6.2154e-03, ESS: 8.6932e+01.\n", "ABC.Distance DEBUG: Scale weights[7] = {'s1': 4.6989e-01, 's2': 1.4659e-01, 's3': 1.4655e+01}\n", "ABC INFO: Stop: Maximum number of generations.\n", "ABC.History INFO: Done \n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "distance_adaptive = pyabc.AdaptivePNormDistance(\n", " p=2,\n", " scale_function=pyabc.distance.mad,\n", ")\n", "\n", "abc = pyabc.ABCSMC(\n", " model,\n", " prior,\n", " distance_adaptive,\n", " acceptor=pyabc.UniformAcceptor(use_complete_history=True),\n", ")\n", "abc.new(db_path, observation)\n", "h_ada = abc.run(max_nr_populations=7)\n", "\n", "plot_history(h_ada)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "These results are as expected: The adaptive weights make the situation much worse, as the impact of the low-variance outlier statistic is increased. Our solution is to firstly use outlier-robust metrics such as a Manhattan distance (p=1), instead of the Euclidean distance used above, which emphasizes large errors." ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "ABC.Distance DEBUG: Fit scale ixs: \n", "ABC.Sampler INFO: Parallelize sampling on 4 processes.\n", "ABC.History INFO: Start \n", "ABC INFO: Calibration sample t = -1.\n", "ABC.Distance DEBUG: Scale weights[0] = {'s1': 3.7101e-01, 's2': 1.7356e-01, 's3': 1.2818e+01}\n", "ABC.Population INFO: Recording also rejected particles: True\n", "ABC INFO: t: 0, eps: 4.06814929e+01.\n", "ABC INFO: Accepted: 100 / 210 = 4.7619e-01, ESS: 1.0000e+02.\n", "ABC.Distance DEBUG: Scale weights[1] = {'s1': 3.6031e-01, 's2': 1.5561e-01, 's3': 1.4466e+01}\n", "ABC INFO: t: 1, eps: 4.41407080e+01.\n", "ABC INFO: Accepted: 100 / 448 = 2.2321e-01, ESS: 8.7540e+01.\n", "ABC.Distance DEBUG: Scale weights[2] = {'s1': 4.3109e-01, 's2': 1.4271e-01, 's3': 1.4640e+01}\n", "ABC INFO: t: 2, eps: 4.42378486e+01.\n", "ABC INFO: Accepted: 100 / 648 = 1.5432e-01, ESS: 9.5623e+01.\n", "ABC.Distance DEBUG: Scale weights[3] = {'s1': 4.9523e-01, 's2': 1.5261e-01, 's3': 1.4854e+01}\n", "ABC INFO: t: 3, eps: 4.42414461e+01.\n", "ABC INFO: Accepted: 100 / 923 = 1.0834e-01, ESS: 9.1566e+01.\n", "ABC.Distance DEBUG: Scale weights[4] = {'s1': 7.5066e-01, 's2': 1.5755e-01, 's3': 1.5398e+01}\n", "ABC INFO: t: 4, eps: 4.53395178e+01.\n", "ABC INFO: Accepted: 100 / 1773 = 5.6402e-02, ESS: 9.4182e+01.\n", "ABC.Distance DEBUG: Scale weights[5] = {'s1': 8.9670e-01, 's2': 1.4815e-01, 's3': 1.5066e+01}\n", "ABC INFO: t: 5, eps: 4.39434732e+01.\n", "ABC INFO: Accepted: 100 / 2823 = 3.5423e-02, ESS: 7.6064e+01.\n", "ABC.Distance DEBUG: Scale weights[6] = {'s1': 1.3076e+00, 's2': 1.5372e-01, 's3': 1.4739e+01}\n", "ABC INFO: t: 6, eps: 4.28698245e+01.\n", "ABC INFO: Accepted: 100 / 5050 = 1.9802e-02, ESS: 9.2703e+01.\n", "ABC.Distance DEBUG: Scale weights[7] = {'s1': 1.2742e+00, 's2': 1.4557e-01, 's3': 1.4966e+01}\n", "ABC INFO: Stop: Maximum number of generations.\n", "ABC.History INFO: Done \n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "scale_log_file_l1 = tempfile.mkstemp(suffix=\".json\")[1]\n", "\n", "distance_adaptive = pyabc.AdaptivePNormDistance(\n", " p=1, # new, previously p=2\n", " scale_function=pyabc.distance.mad,\n", " scale_log_file=scale_log_file_l1,\n", ")\n", "\n", "abc = pyabc.ABCSMC(\n", " model,\n", " prior,\n", " distance_adaptive,\n", " acceptor=pyabc.UniformAcceptor(use_complete_history=True),\n", ")\n", "abc.new(db_path, observation)\n", "h_l1 = abc.run(max_nr_populations=7)\n", "\n", "plot_history(h_l1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This visibly improves the estimates again. However, the outlier statistic still seems to have a considerable impact, as the estimates are considerably worse than without s3. This can be tackled by actively identifying and down-weighting outlier data points, by using a different weighting scheme than MAD. We propose to use the PCMAD (Perhaps-Combined MAD), which either uses MAD, or, if the fraction of severe outliers detected is not too high (precisely not more than 1/3, see the publication and the API for details), uses MAD + MADO, where MADO is the median absolute deviation to the observed value." ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "ABC.Distance DEBUG: Fit scale ixs: \n", "ABC.Sampler INFO: Parallelize sampling on 4 processes.\n", "ABC.History INFO: Start \n", "ABC INFO: Calibration sample t = -1.\n", "ABC.Distance INFO: Features ['s3'] (ixs=[2]) have a high bias.\n", "ABC.Distance DEBUG: Scale weights[0] = {'s1': 2.0023e-01, 's2': 6.4406e-02, 's3': 3.2667e-01}\n", "ABC.Population INFO: Recording also rejected particles: True\n", "ABC INFO: t: 0, eps: 2.04345057e+00.\n", "ABC INFO: Accepted: 100 / 214 = 4.6729e-01, ESS: 1.0000e+02.\n", "ABC.Distance INFO: Features ['s3'] (ixs=[2]) have a high bias.\n", "ABC.Distance DEBUG: Scale weights[1] = {'s1': 2.0080e-01, 's2': 6.7454e-02, 's3': 3.2413e-01}\n", "ABC INFO: t: 1, eps: 1.64061092e+00.\n", "ABC INFO: Accepted: 100 / 270 = 3.7037e-01, ESS: 9.1801e+01.\n", "ABC.Distance INFO: Features ['s3'] (ixs=[2]) have a high bias.\n", "ABC.Distance DEBUG: Scale weights[2] = {'s1': 3.3611e-01, 's2': 7.4539e-02, 's3': 3.2369e-01}\n", "ABC INFO: t: 2, eps: 1.63893421e+00.\n", "ABC INFO: Accepted: 100 / 348 = 2.8736e-01, ESS: 9.6167e+01.\n", "ABC.Distance INFO: Features ['s3'] (ixs=[2]) have a high bias.\n", "ABC.Distance DEBUG: Scale weights[3] = {'s1': 5.3300e-01, 's2': 7.4961e-02, 's3': 3.2599e-01}\n", "ABC INFO: t: 3, eps: 1.54996332e+00.\n", "ABC INFO: Accepted: 100 / 394 = 2.5381e-01, ESS: 9.8993e+01.\n", "ABC.Distance INFO: Features ['s3'] (ixs=[2]) have a high bias.\n", "ABC.Distance DEBUG: Scale weights[4] = {'s1': 7.4951e-01, 's2': 7.9142e-02, 's3': 3.2542e-01}\n", "ABC INFO: t: 4, eps: 1.48381284e+00.\n", "ABC INFO: Accepted: 100 / 482 = 2.0747e-01, ESS: 8.8783e+01.\n", "ABC.Distance INFO: Features ['s3'] (ixs=[2]) have a high bias.\n", "ABC.Distance DEBUG: Scale weights[5] = {'s1': 1.3840e+00, 's2': 7.5257e-02, 's3': 3.2620e-01}\n", "ABC INFO: t: 5, eps: 1.43561651e+00.\n", "ABC INFO: Accepted: 100 / 733 = 1.3643e-01, ESS: 9.7654e+01.\n", "ABC.Distance INFO: Features ['s3'] (ixs=[2]) have a high bias.\n", "ABC.Distance DEBUG: Scale weights[6] = {'s1': 1.8054e+00, 's2': 7.5495e-02, 's3': 3.2604e-01}\n", "ABC INFO: t: 6, eps: 1.33302716e+00.\n", "ABC INFO: Accepted: 100 / 874 = 1.1442e-01, ESS: 8.7795e+01.\n", "ABC.Distance INFO: Features ['s3'] (ixs=[2]) have a high bias.\n", "ABC.Distance DEBUG: Scale weights[7] = {'s1': 3.3984e+00, 's2': 7.4986e-02, 's3': 3.2583e-01}\n", "ABC INFO: Stop: Maximum number of generations.\n", "ABC.History INFO: Done \n" ] }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEGCAYAAAB/+QKOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8/fFQqAAAACXBIWXMAAAsTAAALEwEAmpwYAABN5ElEQVR4nO3dd3xb1fn48c/RsLb3iB3H2YEsCBBWIBBKwip7tKxQCi3lW2hLw2goBX5hlVHohPLl28Eqo2VTKJuUQBgJEDII2ctOvOJtbd3z++PKjh0vObYi23ner5ci+ereqyPF1nPPeo7SWiOEEELszpLqAgghhBiYJEAIIYTolAQIIYQQnZIAIYQQolMSIIQQQnTKluoC9NZJJ52k33jjjVQXQwghBhvV2wMGXQ2iuro61UUQbbz++uu8/vrrqS6GECIJBl2AEAPLkiVLWLJkSaqLIYRIgkHXxCQGFp/Pl+oiCCGSRA22mdTTp0/XS5cuTXUxhBBisOl1H4TUIESfvPDCCwCcffbZKS6JGKoikQilpaUEg8FUF2VQcDqdFBcXY7fb+3wuCRCiT1asWAFIgBDJU1pais/nY9SoUSjV64vgfYrWmp07d1JaWsro0aP7fD7ppBZ9kpGRQUZGRqqLIYawYDBITk6OBIcEKKXIycnpt9pW0moQSqkRwONAAaCBR7TWv99tn1nAy8Cm+KYXtNa3JatMov9dc801qS6C2AdIcEhcf35WyWxiigLXaq2/UEr5gM+VUm9rrb/ebb9FWutTk1gOkURPP/00AGdMno1jbCZWT9/bPYUQA0PSmpi01ju01l/EHzcCq4HhyXo9kRpr165l3Zq11Dz1Df4vKlNdHCGSwmq1Mm3aNKZMmcJ5552H3+9vt33y5MkceOCB3H///RiGAcDChQvJyMhg2rRpTJs2jdmzZ3c478KFC1m8eHGvyvLYY48xfvx4xo8fz2OPPdb3N9eNvdJJrZQaBRwEfNrJ00cqpb4CtgPXaa1XdXL8FcAVACUlJUksqeit3NxcHBErlIMRjKa6OEIkhcvlYtmyZQBcdNFFPPzww8ybN6/d9srKSi688EIaGhpYsGABADNnzuTf//53l+dduHAhXq+XGTNmJFSOmpoaFixYwNKlS1FKccghh3D66aeTlZXVp/fXlaR3UiulvMDzwDVa64bdnv4CGKm1PhD4I/BSZ+fQWj+itZ6utZ6el5eX1PKK3rnqqquYe/aFAGgJEGIfMHPmTNavX99he35+Po888gh/+tOfSGR+2ebNm3n44Yf57W9/y7Rp01i0aFGPx7z55pvMmTOH7OxssrKymDNnDsnMTZfUGoRSyo4ZHP6htX5h9+fbBgyt9etKqYeUUrlaa0m4NEg8+uijpDc5OIxijFAs1cURQ9yCV1fx9fbdrzP7ZlJROreeNjmhfaPRKP/5z3846aSTOn1+zJgxxGIxKivN5tZFixYxbdo0AM477zxuuumm1n1HjRrFlVdeidfr5brrrgPgH//4B/fdd1+H844bN47nnnuOsrIyRowY0bq9uLiYsrKyhMq+J5I5ikkBfwVWa60f6GKfYUCF1lorpQ7DrNHsTFaZRP/bsmULTm3nMIrREiDEEBUIBFq/6GfOnMnll1+e0HE9NTHt7qKLLuKiiy7akyImRTJrEEcBc4EVSqll8W2/BEoAtNYPA+cC/6OUigIB4Hw92HJ/7OMKCgrIb/RCUPogRPIleqXf39r2NXRn48aNWK1W8vPzWb16da9fp6caxPDhw1m4cGHr9tLSUmbNmtXr10lU0gKE1vpDesj9obX+E/CnZJVBJN+VV15Jw7tbaXh7i9QgxD6tqqqKK6+8kquvvjrhuQg+n4+Ghl1NZj3VIE488UR++ctfUltbC8Bbb73Fr3/9674VvBuSakP0yV/+8hdG7ExnMoXSByH2OS1NT5FIBJvNxty5c5k3b17Cx5922mmce+65vPzyy/zxj39k5syZ3e6fnZ3NzTffzKGHHgrALbfcQnZ2dp/eQ3ckQIg+KSsro8moYTKF6KAECDE0NTU1dbo9Fuv6d37WrFk9Nv9MmDCB5cuX96osl112GZdddlmvjtlTEiBEnwwfPpxx5VkQAiMkfRBCDCWSrE/0yQ9+8AMm5Y0DQIdiCY3/FkIMDhIgRJ88/PDDrCvbaP6gQUeM1BZICNFvpIlJ9ElFRQWGCjCSXACzHyLNmuJSCSH6g9QgRJ+MHDmSGdEJWHxmFlfphxBi6JAAIfrke3MvITeWji3TCSAjmYQYQiRAiD555MH/pcJSjzXTASBzIcSQNJDSfZ900klkZmZy6qnJX0ZHAoTok6qaapbaNmDNitcgpIlJDEEtqTZWrlxJWloaDz/8cLvtq1at4u233+Y///lPa6pvMHMxLVu2jGXLlvHOO+90OO+eBIjrr7+eJ554om9vKEESIESfTC3ej6Mi+2HLitcgpIlJDHGpTPcNcPzxx+Pz+Xpd7j0ho5hEn5xw9Gx2rl/V2sQk+ZhEUv1nPpSv6N9zDpsKJ9+d0K6pTve9t0mAEH3ywksvMMGST368k1r6IMRQJOm+hdgD24NVNNmaOdhnB6uSPgiRXAle6fe3gZLue2+TACH65LDcKeSVObC4bFicVumDEPusvZHue2+TTmrRJweOmkyG3YOyWlAOm/RBiH1KS9PT5MmTmT17NieccAK33nprwsefdtppvPjii73qpJ45cybnnXce7777LsXFxbz55pt7WvweSQ1C9MkHny8m03AxHLA4rLKqnBiSBlK670QDSX+QACH6ZJNRjtVq4RhAOaxSgxBiCJEAIfrkGOfU1oVlLU4bscZwagskhOg3EiBEnxQ6c7HluIB4DaJaahBCDBXSSS365JvqTXy0aSkgfRBCDDVSgxB98o21FL8RAkA5pQ9CiKFEAoTYYzpq8K3wFBrGm50QFocNHTHQMY2yJjYOXAgxcEkTk9hjRiCKGwf7T5oImH0QIBldxdAzUNJ9L1u2jCOPPJLJkydzwAEH8Oyzz/bPG+yCBAixx4xAlK2Wav757ksAWJxmgJB8TGKoGSjpvt1uN48//jirVq3ijTfe4JprrqGurq7P768rEiDEHjMCUVbYtlIRrQXa1iAkQIihK5XpvidMmMD48eMBKCoqIj8/n6qqqt6/iQRJH4TYY0YgyvHhKdQcbab6tjjMXyepQYhkueeze/im5pt+Pef+2fvzi8N+kdC+Aynd92effUY4HGbs2LEJlX1PSIAQe8wIRHGSxkGHHwyYo5gAtAx1FUPMQEv3vWPHDubOnctjjz2GxZK8hiAJEGKPaX+EzZZKnv3Lg1x74/VYHNIHIZIr0Sv9/jaQ0n03NDTw7W9/mzvvvJMjjjii16/RGxIgxB4zAlFW2UppCpkjOlS8iUlLym+xD9ob6b7D4TBnnXUWl1xyCeeee26fy9yTpNVNlFIjlFLvK6W+VkqtUkr9rJN9lFLqD0qp9Uqp5Uqpg5NVHtH/jECU2fpAzjzrTDYv/5JIfMKc1CDEvmJvp/v+5z//yQcffMCjjz7aOnw2kZrNnkpmDSIKXKu1/kIp5QM+V0q9rbX+us0+JwPj47fDgT/H78UgYASiuD1u8ouH88iPL+W4S35IPtkyD0IMOQMl3ffFF1/MxRdfnPD+fZW0AKG13gHsiD9uVEqtBoYDbQPEGcDj2hwT9olSKlMpVRg/VgxwRiDKsuZ1fPmHV3EA/sZ6VFqerConxBCxV+ZBKKVGAQcBn+721HBgW5ufS+Pbdj/+CqXUUqXU0mSO+RW9YwSirLeUE4n/GgWbmmRNCCGGkKQHCKWUF3geuEZr3dDT/p3RWj+itZ6utZ6el5fXvwUUe8wIRDm1aCZjLOYaEMHmJnNdamliEmJISGqAUErZMYPDP7TWL3SySxkwos3PxfFtYhAwAlEycjJp3mH+l4WapQYhxFCSzFFMCvgrsFpr/UAXu70CXBIfzXQEUC/9D4OHDkT5z/L32OnwAS01CJv0QQgxRCRzFNNRwFxghVJqWXzbL4ESAK31w8DrwCnAesAPfD+J5RH9SEcMdMRgh6OOaEY23lBjaw3CaJJlR4UYCpI5iulDWlcr7nIfDVyVrDKI5DECZj/DsVnjWPjx44w4+hg2f/UFFoeViNQgxBBjtVqZOnUq0WiUiRMn8thjj+F2u1u3RyIRbDYbl1xyCT//+c+xWCwsXLiQM844g9GjRwOQm5vbIaPrwoULSUtLY8aMGQmVY8uWLZx11lkYhkEkEuEnP/kJV155Zb+/3xYyk1rsESMQASDQvJOcwkJ8uXkEW/ogwhIgxNDSNtXGRRddxMMPP8y8efPaba+srOTCCy+koaGhNeV3T7mYFi5ciNfrTThAFBYW8vHHH+NwOGhqamLKlCmcfvrpFBUV9en9dUXSfYs90lKDWF67ne0OH06PF20YaKvGCMYSSncsxGCUynTfaWlpOBxm9uRQKNS6OFGySA1C7JGWANHgthBTHhweLwAxFQVDQ9QAuzWVRRRDUPlddxFa3b/pvh0T92fYL3+Z0L4DId33tm3b+Pa3v8369eu57777klZ7AAkQYg+1BIicDes4+cbrCTabqQii2mx6MoIxrBIgxBAxkNJ9jxgxguXLl7N9+3bOPPNMzj33XAoKChJ+jd6QACH2SEuACBtB8kePpXLTBgCiOozCTNhn9aWwgGJISvRKv78NpHTfLYqKipgyZQqLFi1KWmZXCRBijxj+eBNTUTG/+e3v+OFFFwAQjoVwIMuOin3P3kj3XVpaSk5ODi6Xi9raWj788EN+/vOf97nsXZFOarFHdCBKRIeJetKJxWI4430Q4VgAAENWlRP7gL2d7nv16tUcfvjhHHjggRx77LFcd911TJ06tS9voVtSgxB7JNIYIBwNcNIRUzj8rO8Q8jcDEAr78eGQGoQYUgZKuu85c+b0av++khqE2CPBmkbCRpCCMeMASHO6UMpCMGwGClk0SIjBTwKE2CORhgBhI8gTL73K7bffjrJYcHg8BIKNALJokBBDgDQxiT1iBCIYNgMDID5Zx+nxEgiaHW6SsE+IwU8ChNgzEbB5ne065BweL4HmRrAo6YMQYgiQJibRa+GAHztpOLK87bY7vd5dGV1lFJMQg54ECNFrlRs2YlU23HlZ3Hbbbdx2222AWYMINjdhkUWDhBgSJECIXqtetxEAX2EeWuvWxGROj6fNsqMSIMTQYbVamTZtGlOmTOG8887D7/e32z558mQOPPBA7r///tYEegsXLiQjI4Np06Yxbdo0Zs+e3eG8CxcuZPHixb0uT0NDA8XFxVx99dV9e2M9kD4I0Ws1W0rJJQtXTka7Pginp6WJySY1CDGkDJR03y1uvvlmjjnmmD16L70hNQjRaw2lFQBYXO2vLxweL0YsBnaZSS2GrlSm+wb4/PPPqaio4IQTTuh12XtLahCiVyKhIIGdDVAAFrettf/hlltuwek1O60NqwapQYgkWPTPtVRv63xW857KHeFl5ncmJLRvqtN9G4bBtddey5NPPtlhdbpkkAAheqVqy2bSLOaCJRaXrd2VUks+JsNiSIAQQ8pASff90EMPccopp1BcXJzwOftCAoTolcpNG0izOAEzQOw+DwLMRYMsMlFOJEGiV/r9baCk+/74449ZtGgRDz30EE1NTYTDYbxeL3fffXevXysREiBEr1Rs2oDblQGAcrb/9WmpQUR1GHtYow2NsiSW9liIwW5vpPv+xz/+0fr40UcfZenSpUkLDiCd1KKXKjatJz0jD+W0oiyq3TyIlj6IqBEGQIelFiGGtr2d7ntvkxqESFg0EmHnti34DjgJizJ/ddr2QTha14QI4sKOEYphccqvmBj8Bkq677YuvfRSLr300j06NlHy1ysStnPbFoxYDKfDh8VmB2jfB+Fyg1KEowHAhw5GIcORotIKIfpKmphEwio2mWO/0yzODnMgAJTFgtPtIRgxZ5nKbGohBjcJECJhFRvX43B7sERVa4Bo2wcB4PB6CQbN6riWkUxCDGrSxCQSVrlpA/mjx2IEoq0BYvcZo06Pl2DIXDTIkEWDhBjUJECIhNWV72C/GcdibIqi4gFi9xEbDo8Xf8Actif5mIQY3KSJSSQkFo0QbG7Cm54FMd1pHwSYNYhmv6wqJ8RQIAFCJMTfUA+Ax5UJ7ErUt2DBgtbMlWAGCH9zLSA1CDF0DKR03y2vOW3aNE4//fS+v7luSBOTSIi/3gwQLqcPMLqsQTi8XgJNjagCi/RBiCFjIKX7TjTtR39IWg1CKfU3pVSlUmplF8/PUkrVK6WWxW+3JKssou8C9XUAONPMyXCWNn0Qu68JYcSiqDSL1CDEkJTqdN97UzJrEI8CfwIe72afRVrrU5NYBtFPWpqYHFYXIRq67YMAIM0ifRCi373/6CNUbtnYr+fMHzmG4y69IqF9U53uGyAYDDJ9+nRsNhvz58/nzDPPTPSt9lrSAoTW+gOl1KhknV/sXf54DSLN4iRE+z4I2DWaqSXdhrZpqUGIIWOgpPsG2LJlC8OHD2fjxo1861vfYurUqYwdOzbh1+iNHgOEUsoKrNJa75+E1z9SKfUVsB24Tmu9qosyXAFcAVBSUpKEYoie+BvqsdpsWGJmq2RPNQht0bKqnOh3iV7p97eBku4bYPjw4YBZW5k1axZffvll6gKE1jqmlFqjlCrRWm/tx9f+AhiptW5SSp0CvASM76IMjwCPAEyfPr3nxj3R7/z19bgyMjECUVC7Un3vPg+iJaNrzBKTGoTYp+yNdN+1tbW43W4cDgfV1dV89NFH3HDDDX0ue1cSbWLKAlYppT4Dmls2aq33eIyV1rqhzePXlVIPKaVytdbVe3pOkTz+hjrc6RkYgSjKYetynYe2iwZJH4QY6lqaniKRCDabjblz5zJv3ryEjz/ttNM499xzefnll/njH//IzJkzu91/9erV/OhHP8JisWAYBvPnz2fSpEl9fRtdSjRA3NzfL6yUGgZUaK21UuowzBFVO/v7dUT/8NfX48nIRAeiWNy7fm1274PYtWhQBC3DXMUQMVDSfc+YMYMVK1YkvH9fJRQgtNb/VUoVAIfGN32mta7s7hil1NPALCBXKVUK3ArY4+d7GDgX+B+lVBQIAOfrRMaGiZTwN9SRO6KkXR6mzjjcZsrviBHCCLr2YgmFEP0toQChlPoOcB+wEFDAH5VS12utn+vqGK31Bd2dU2v9J8xhsGKA01oTqK/HlZ6B0dA+QOzeB6EsFhxuN+FoEGIaHTVQNpmwL8RglGgT003AoS21BqVUHvAO0GWAEENHOBAgGgnjzsjEqIhi72ERIKfHG180yFwTwioBQohBKdG/XMtuTUo7e3GsGOT8DXUArZ3UbWsQu+diArOjOhgxxzJoGeoqxKCVaA3iDaXUm8DT8Z+/C7yenCKJgaYlD5MZIILd9kFAfE2IoBkgZFU5IQavRDupr1dKnQMcFd/0iNb6xeQVSwwkLTUIlzudaCzQbhTT7n0QEA8QtQ2QJqvKCTGYJdxMpLV+Xms9L36T4LAPCbRkcnX4AFoXC+qKw+ulOb5okGR0FUPBQEr3vXXrVk444QQmTpzIpEmT2Lx5c5/fX1e6/UtXSn2otT5aKdUItB2CqgCttU5PWsnEgNGah8nqBujQBwF0yOjq99dBhqwJIYaGgZTu+5JLLuGmm25izpw5NDU1YbEkrzu42zNrrY+O3/u01ultbj4JDvsOf0M9DrcHS8S8RuipD8Lh8RKKmFdY0gchhppUpvv++uuviUajzJkzBwCv14vb7e79m0hQqpP1iUHAX19nDnENmM1FFpe99bmu+iCiRhiQPgjRv+pe3UB4e3PPO/ZCWpGHzNMSS3aX6nTfa9euJTMzk7PPPptNmzYxe/Zs7r77bqxWa2/ecsJSmaxPDBL+hnrcGRltAkQPo5i8XqI6AkgfhBgaBkq672g0yqJFi/jyyy8pKSnhu9/9Lo8++mjC5emtlCXrE4OHv76O7KLiTgNEZ30QLQn7sCvpgxD9KtEr/f42UNJ9FxcXM23aNMaMGQPAmWeeySeffJLyANHvyfrE4OFvqKd44uRdqb4d3VdnnW0WDZKMrmJfsTfSfR966KHU1dVRVVVFXl4e7733HtOnT+9z2bvSm2R9I4HxWut3lFJuIDmNXmJAMWIxAo0NuNIzW2dRt0313VUfBIBhNSSjqxjS9na6b6vVym9+8xuOP/54tNYccsgh/PCHP+zr2+hSosn6foi5ols2MBYYDjwMHJ+0kokBIdDYAFqbfRC10R7nQIA5DwIgpgwZxSSGhIGS7htgzpw5vT5mTyU6gPYqzFnUDQBa63VAfrIKJQYOf0NLmo1MDH/HVN+d5mKKD7uLEZFRTEIMYokGiJDWOtzyg1LKRvuJc2KIapkk587IMBcLSqAGYbFYcbg9RHVEahBCDGKJdlL/Vyn1S8CllJoD/Bh4NXnFEgNF2xpEJFCBPat9qu/O+iDAHMkUMULSByH6hdY64Y7ffV1/rruWaA1iPlAFrAB+BLyutb6p+0PEUBBoU4PoaTW5tpweL+FYUEYxiT5zOp3s3LmzX7/4hiqtNTt37sTpdPbL+RKtQfxEa/174P9aNiilfhbfJoYwf0M9FqvZZGQEIljc9nbPdzYPAsDp9RAO+NE6Jld/ok+Ki4spLS2lqqoq1UUZFJxOJ8XFxf1yrkQDxPeA3YPBpZ1sE0OMv74OV3oGhDUYtEv1DXT5xe/weAnWN4MNdNjoce6EEF2x2+2MHj061cXYJ/WUzfUC4EJgtFLqlTZPpQM1ySyYGBj8DfXmQkF+M3XG7jWIW265pdPjnB4vwVCjGSBCUZAAIcSg01MNYjGwA8gF7m+zvRHYOwNxRUq1Jurzx9NsuBOrdDo8XhqD28EDRjCGVXL/CjHodPvXrrXeAmxRSs0GAlprQyk1Adgfs8NaDHH+hnoyCwq7rEF02QfRJuW35GMSYnBKdBTTB4BTKTUceAuYCzyarEKJgcNfV2eOYOqiBqGU6rQfwun1Eomn/DaCMtRViMEo0U5qpbX2K6UuBx7SWt+rlFqWxHKJASASDBIJBc08TL3sg3B4vER1CJAahBCDVaI1CKWUOhK4CHgtvk16HYe41klyGRnEWmoQzsTnQbTWICRACDEoJRogrgFuBF7UWq9SSo0B3k9aqcSA4G+oA1ryMEVQThvK2r45qbNcTNA+QGhpYhJiUEo43Tdmug2vUsqrtd4I/DS5RROp5q/fVYMw1kWxeDr+unQ5D8K7a9lRqUEIMTglmu57KvA4ZrpvpZSqAi7RWq9KZuFEarXUIDwZWYT8lZ2m2ehuHoRBDK209EEIMUgl2sT0v8A8rfVIrXUJcC1t0m6IoamlBuFKT8fwd0yz0R2HxwOYiwZJDUKIwSnRUUwerXVrn4PWeqFSypOkMokBItBQh93pwu5wYvij2HNdHfZZsGABaBjPiRx22hjGHWIuE2KxWElzuYmpmPRBCDFIJVqD2KiUulkpNSp++xWwsbsDlFJ/U0pVKqVWdvG8Ukr9QSm1Xim1XCl1cG8LL5LLX1+POyMDoMsahMKKJeqmttzPR8+tIxYxWp9zer3EkDUhhBisEg0QlwF5wAvA85ipNy7r4ZhHgZO6ef5kYHz8dgXw5wTLIvaSljxMOqbRwViHSXKVWxooqj+GEmMGx128P021IVYuKmt93uHxEtFhSfktxCDVU7I+J3AlMA4ztca1WutIIifWWn+glBrVzS5nAI9rM8n7J0qpTKVUodZ6R2JFF8nmr68jPS8fI9BxklzFpgZe+cMyHG4bZ/78INJzXaxdUs7n/9nMpKOKsDus5lDXQAgdlgAhxGDUUw3iMWA6ZnA4GbivH197OLCtzc+l8W0dKKWuUEotVUotlZzwe4+/vi6eybV9mo3yjfW88vsvcXpsbLK/zW8fvBeAI84YS6AxwvL3zf9Wp8dLJBaUPgghBqmeAsQkrfXFWuv/Bc4FjtkLZepAa/2I1nq61np6Xl5eKoqwz9GGYTYxZbRPs7FjfR2v/GEZLl8aZ117MBarBYvF/DUaNiaDUVNz+PKtrYT8ERweL6FIQPoghBikehrF1NqcpLWO9vOqYGXAiDY/F8e3iQEg2NyENox2NYid1QH+/ew6vJkOzrjmILxZDm6++eZ2xx12+hj+eecSvnx7K06vl2C4XvoghBikeqpBHKiUaojfGoEDWh4rpRr6+NqvAJfERzMdAdRL/8PA0ToHos1aEIte3oA308GZ88zg0Jm8ET7GHZLPV++VYrE6CUf9EDXQMaPT/YUQA1dP60HscUI+pdTTwCwgVylVCtwK2OPnfRh4HTgFWA/4ge/v6WuJ/rdrFnUmRo1ZkayvjzDztLF4MnYFh87WgzjstNFs+KKSys1h7C35mEIxlDvRQXNCiIEg0Ylyvaa1vqCH5zVwVbJeX/RNax6m9AyM0ihaQRQoHJfRbj+rteM1RNYwD/sdWcjqD9ZSpFvWhIj1aia2ECL1khYgxODWmsk1I5NQoIqYReHy2ckscLfb71e/+lWnxx/67VGsXvQJkZi5JoR0VAsx+EidX3TKX18PSuH0+TD8UUIxTeG4zC6zt+4uPcfFmIOKd6X8DslQVyEGGwkQolOBhjpcvnQsFiuR+hDBqEHRuMwO+3W1HgTAgbPHETGCAK0d3UKIwUMChOiUv95MswEQrg8R1h37H8Dsg+isHwIga1g2gVgzAI1lTckrrBAiKaQPQnTK31CHOyMTMK/+Y0qRW+ztsF9XfRBgzqQOxQNEzYY68ueMTEpZhRDJITUI0amWNBsAKmJgz0jDYu3461JVVUVX6U8sVis2l4OQEcZf7k9qeYUQ/U8ChOiUmeo7k0BdCCvgzuu4FgTAQw89xEMPPdTleRweLxEVJtYYJuRPKM+jEGKAkAAhOohGIoT8zbjTMyj/pgYAX1HH5iUAm82GzdZ1S6XT4yViCeJUsG11bVLKK4RIDumDEB0EGuKT5DIzqV5XRyGQOaLzAHHTTTd1ey6nx0s4FCDdqtiyamfrinNCiIFPahCiA399HQDu9Exqt5gpt2zpnede2rRpExs3bqTx3XeJ7tzZ4XmHx4s/0ohDKbauqsacQC+EGAykBiE68MdrEGluH/7KKnBbu0yT8fhjjwHwnWeexVZYyIiH/4xzv/1an3d6vfi31mFxQ7QhQnVpE3kjfMl/E0KIPpMahOigpQbhb7Biw7zi3325UYBodTW+piYyamvJ/t73IBZj8wUX0vje+637ODxeGgNm34NDwdZVHWsZQoiBSQKE6KClBlFboUmzmKk1rLsFiODatWz+znc55Z13ueyUb1Nw43xG/etfOEaPpvSqq9j5t7+jtcbp8dIcqgOgoMDNlpUSIIQYLCRAiA789XXY7GlUbAqQ7ktD2S0o+67Z0k2LFrHlggvRkQiB2xawaVgBAPaCfEY++QS+E06g8t572XHzzTicToIxcxZ1YZGH8o0NMtxViEFCAoToINBQjysjg4rNjWT47Fhcu2oPNU/+g20/uhJ7SQmj/vVPXvnyS1566aXW5y0uF8N/+wA5/3Ml9c89T9OTT7Wm28jJdqANLcNdhRgkJECIDvz1ddidPqKhGG7Hrg7qinvupeKOO/DOmsWoJ5/APmwYDocDh6P9CCdlsZD/s59RdN+96E2biekI2gZumwWH28YW6YcQYlCQUUyig+b6OpQyZ06nWRWWNCuNCxdS8/e/k3nB+Qz71a9Q8QR98+fP7/I8GaedxkirYsljDxOLNmE0RRgxMZutq3aitU44dbgQIjWkBiE68DfUE404SM9zocIxVJqifMFtOMaPY9iNN7YGB4DFixezePHiLs+Vf8yxAAQbq4jVBymZnIO/Pkx1qWR3FWKgkwAh2tFaE6ivI9hsp2hchrlY0PqviZaXU3j77ai0tHb7v/3227z99ttdns/p8ZKW5iAYbiBSXkvJ5GxAhrsKMRhIgBDthAN+YtEosaiTwrEZGM0Rgl9/RdbFF+OaNq3D/i6XC5er80R+AEop0ocVErRGMAIat89O7givDHcVYhCQPgjRTsskOaVcFBS5CGiwOC3kX/OzTve/4YYbejxnRv4w/KEYyppG4zv/ZeTkEr54ayshfwRHFzO0hRCpJzUI0Y6/3pwk5/BmEH3pOQDSv30CFo+n0/3ff/993n///U6fa5GRX0B9yBzaWvvMi5RMyZHhrkIMAhIgRDv+hjoA8vK91L3wCgDugyZ3uf8HH3zABx980O05M/ILaA6a5w2t2UJmaLsMdxViEJAAIdqp2VENgGftF1jTc4HO8zC18Hg8eLqoXbTIyC9onSxnSc+l7okn2g13FUIMTBIgRDuVmysAyFj9CZnnng/QZSZXgOuuu47rrruu23Nm5BW0pttwTz+ahtdeo7gkTYa7CjHASYAQ7dRs3QE4KDx4DI79pgLd1yBef/11Xn/99W7PmZ5fQNgIopXGOfkgdDSKb9U7gAx3FWIgkwAhWmmtaaioxoaNoltvQQeiAFhcXdcglixZwpIlS7o9b5rThSs9g4g1AsqFd9Ysgv96ktxijwx3FWIAkwAhWm1/6W0iRhC314d9+HAMfxTlsKKsXafE8Pl8+Hw9LwCUkV9AyPATawyT/b3vEautpcBaRfnGBoLNkt1ViIFI5kEIAIzmZtY89hbaWUPhlBnmNn8Ei6f7eQrz5s1L6PwZeQX4SxuINYZxH34YjokTSf/sRXTBRWxcVsWko4r6/B6EEP1LahACgKoHH6Lclg86yLDxowGI+aPd9j8AvPDCC7zwwgs9nj8jv4DGQA2xhjBKKbK/dwnO1YtJ98Hazyr65T0IIfpXUgOEUuokpdQapdR6pVSHtJ9KqUuVUlVKqWXx2w+SWR7RueCatVQ//iR1mebCP9lFxQAYgWi7tSA6s2LFClasWNHja2TkDyMQbUQHouioQcYpp2DLy2VY/QrK1tbSXBfq+xsRQvSrpAUIpZQVeBA4GZgEXKCUmtTJrs9qrafFb39JVnlE57RhUL5gAY3DJhOJNQCQVWg29xj+SLdDXAEyMjLIyMjo8XXS8/IJRs0hrbHGMCotjewLLyRryQugYd1SqUUIMdAkswZxGLBea71Rax0GngHOSOLriT1Q/+JLBL74gsDsi8CoxWK1kZE/DACjuecmpmuuuYZrrrmmx9cxJ8vtChAAGWefgydUTabDz7olEiCEGGiSGSCGA9va/Fwa37a7c5RSy5VSzymlRiSxPGI30dpaKu+7D9dBB1EZzSPN1UhmwTAsViva0OhgtMcaxNNPP83TTz/d42v5cvMJGn4AjHiAsBfk45l5NPlbPqRySyN1Ff6+vykhRL9JdSf1q8AorfUBwNvAY53tpJS6Qim1VCm1tKqqaq8WcCireuC3xBobSb/+V+aMZl1HVpEZw42WORA91CDWrl3L2rVre3wtm92OxWcGm5YaBEDm2eeQu+m/gDQzCTHQJDNAlAFtawTF8W2ttNY7tdYtvZN/AQ7p7ERa60e01tO11tPz8vKSUth9TWDZMur+9S+y586lMpyF1gbBpmqyCuMBwm/OTbD2UIPIzc0lNzc3odd05mSg0cQadgUI33Gz8Lgg11LN2s8qJDeTEANIMgPEEmC8Umq0UioNOB94pe0OSqnCNj+eDqxOYnlEnI5G2fH/FmArKCD36qvZuqoGhzuAEY3sGsHkT6wGcdVVV3HVVVcl9LoZBfmEjABG466JcSotjfTTTiN37TvUVfip3ia5mYQYKJIWILTWUeBq4E3ML/5/aq1XKaVuU0qdHt/tp0qpVUqpr4CfApcmqzxil5onniT0zTcU3HgjFrebbatryC40r+pbm5jiNYie+iAeffRRHn300YReNyO/gECkkWh9sN32zLPPIq/8cyxKs1Y6q4UYMJI6k1pr/Trw+m7bbmnz+EbgxmSWQbQXXLuWqt/+Fu9xx+E78QSqtjYSbIrgHhsAILu1iSmxGsSWLVsSfu2M/GE0x7YRqWvfGe2cOBHvhJHkBjezbomTGWeNRVm6Tu8hhNg7Ut1JLfYiIxxm+/U3YPH5yL/t/9EQbmDFF5sA2Olfh9Xl4M3K93l1w6ts3LEegCZroNt+gYKCAgoKChJ6fTPtdzOxxo65lzLPOpu8De/TXBdi+/q6Xr83IUT/k1xM+5Dt999HaM0anr18LC+8PhuN5oyVP8XmSWPjhg+xOyz8dbFZwbuk8jS+w4kc89Is7FY7ua5ccl255LnzyHPlUeApIM+VxwGnH0C+K5/6UD3paeko1fWVf3rLXIiggTZ0u1pC+qnfJu83v8OqYqxdUsHwCVlJ/zyEEN2TADHEGdrg84rP+fjf/8fxjy3i3YMUy8bb+EHJD8hU2TR+UkjOkVD1UQF5+43nqrN/QMyIEfr3dnRzlOsPu56qQBXV/mqqAlVsadjCkvIlNITNWdfTK6YDsLRgKW6bmyJvEYWewnb3Rd4i9s/eH292NiEjgNIKozmC1ZfWWk5bVhYZs44mr3wFGz53cMx3J2C1SQVXiFSSADFE1YfqeeabZ3hp/UvUVm3j/r8ZNBf4OO6eB/nJiOkopdjwZSVv6JXMOGQi/3qtgVGjJjHCZ45M3kmAiK+ZuZPmdnr+YDRIlb+Kv/72rwDMmj6L8uZytjdtZ0fzDpZXL6c+VN+6f5oljal5UznZYQ5ca65tIN3Xfnhs5jlnkz//D5RnTWPb1zWMOiCx4bNCiOSQADHEaK15e8vb3PXpXewM7uTwYYez4K1MfM0rGPXXv+Iqmdq679ZVNdidVuxOc2hpdtGuie49Jepz2pyMSB9B8XBzWOz3Jn+vwz7NkWZ2NO1gS8MWvqj8gs8rPmcbOwCY9+pPCYxWHD7scM6dcC5F3iI8Rx1Fvu1W0gixdkmFBAghUkwCxBBS6a/kzk/u5L1t7zExeyJ/nv1nihZvYPt/ryfvZz/FNXVXcNBas/XrnRTvl0V9+XYAsuJzIACzCSjD0eNr/uAHXSfg9dg9jMsax7iscRw/8ngA3t74J9gOpxaczPO8xd9W/o2/rvwrs0tmM3fSXIrOOI28D5awaZmTcHA/0pzyKypEqshf3xBgaIPn1z3PA0sfIGJE+PkhP+eSSZegyyvZeNttuA46iJwf/rDdMXUVfppqQhxy0ihqd6wCpcgctmveouGPYi/09PjaDz/8MABXXnllQmVNH54P2+HY9BmcecJcypvLeeqbp3hu7XO8teUtZuVM4MLyEGWFR7N5eTUTDhvWi09CCNGfpBdwkNvSsIXL37yc2z6+jUk5k3jh9Be4bMplWLVi+y/mQyxG0b33oGztrwW2rqoBoGRSNjXbS0nPzcOetqvGkEiqb4CKigoqKhKf3JYxrIBwLEig0uzkHuYZxrxD5vHOue/wy8N/ycb0IDt8G7GFa3n/vc+JGLIcqRCpIgFikNJa8+w3z3LOK+ewpmYNC2Ys4C8n/IWS9BIAah59FP+SJRTcdBNpIzomyd369U4yC9yk57qo3VHWmoMJQEcMdMTocZIcwMiRIxk5cmTC5c7IH0Yg1kSktrnddrfdzQX7X8CrZ71K0XkXMnzHZ0Q2O/j+sz/iq6qvEj6/EKL/SIAYhBrCDVz732u549M7mD5sOi+f+TJnjz+7dQ5CcPVqKn/3e3xz5pBx9lkdjo+GY5StraNkUjZaa2q2l7XmYAIwAoml2QC49NJLufTSSxMue0Z+AcFYU6eT5QAsysKhF/2cETs/wYJB4dopzH19Lnd8cgeN4caEX0cI0XcSIAaZFVUr+M6r3+H9re8z75B5PHT8Q+S5d2W4NYJByq6/HltmJsNuW9DpxLXt6+uIRQxKJufQXFdLJBhoXUUOEk+zAfDggw/y4IMPJlx+V3oGIR1ABYwu97F6PeR9awbF5R8zumIaFxd/n3+t/RdnvHQGb25+UzK+CrGXSIAYJLTWPLbqMS75zyVorXn05Ef5/pTvY1Ht/wsrH3iA8PoNFN51F7aszmcjb/6qGqvNQtGETGq3lwK7jWBqTdTXJkCEmqBmI1R8DRWroHwl7FhOdXU11dXVsP1LqFoD9WUQrAcj1ulrK6UwHGCNWLv9os+5/DJKNr2BMgwOKz+Zp779FLmuXK7773Vc9e5VlDWVdXmsEKJ/yCimQaA2WMuvPvoVH5R+wPElx7NgxgIyHB3XgW768CNqH3+CrLlz8c48utNzRUIx1nxazpiD8rCnWanZbn7RZmd6YMP7UL4cY10IOBLLmz+B11ZAUyVEOl/tbSpzzAePPNDxSbsHHF5w+MCVDZ5ccGdjoQQLVvRnT6OycsFXAN5h5vMWKwCO8ePJP+lYCtd8zOrFR3PIyUfy1Lef4ulvnuaPX/6Rc145h1uOuIVTxpzS+w9UCJEQCRAD3IqqFfx84c+pCdZw42E3csH+F3TabBStrWXHjTeSNm4s+dfO6/J8az8rJxyMMXXUFnj/ZWrfX4LNYuD766EQP23McjZwJBaHhmGHgicfvPGb3QUoUBZQirNbHnOZGUTCTWZtI9QYf9xo3vw7oW4bbP8Se2gmuGcQ/fedpFlLdxVOWeOvUwC+QnInZzDy7Q/ZPuwIvnxxKcecP5G5Ey/m+JLjmb9oPr9Y9As+3vExNx52I267u18/dyGEBIgB7cV1L3L7J7eT787nyVOeZFLOpE7301pTfsutROvqGP3I/2JxOtvvEAnC1o/R699j5Rv7k2MLMuyDeWCxUFs7nSyfB3XC7VB4AAw7AOOzZnhjM5ZLHoc0a7dl/N3vfgfANddck/D7cjz5b1gJwZOfIG1EFBrLoanCvG8sh6ZyqN9GWsMnFA6PMWzHx3y95DAO2XwIHleEoswS/pY9hofTJ/PI+pdZtuMz7jv2AfbPm5xwGYQQPZMAMQBFYhHuWXIPz655liMLj+TeY+4l05nZ5f71L75E49tvk3/dtTgnTgStoXotrH0TNr4PWxZDNEhFdBLVgeM49tCtqBPehoIp1Fz/M/JHjYWjftp6PsNfDzaFsvfcRVVfX9/jPrtzF2bDyhhNjTbSRxzc7b65F29i9NmXUz5sBl/m/Yajxy+Fui3YqtZwdc0GDnfYmZ8X4cLXvsu1ERcXZh+EGn4wlBwOBVPBKr/iQuwp+esZYKoD1cxbOI8vK7/k+5O/z08P/ik2S9f/TeFt26i44w7ch04n+7j94M2bYM3rZocyQN7+MP0yGHMcKxcVYG+qY8JFF4PTRiwaob6ygv1nHNPunC2T5LpL3d1iapv0HYnyFecTYQeByp6Di234aIaffzoFCz9lpf1IDr70Ntzp8SywkQCHVq3hue1LuWX9s9ytylm88yNuX/Uvsg3D7AMZcSiMOAJKjoDi6WZ/iBAiIRIgBpCvqr5i3vvzaIw0ct8x93HS6JO63V8317D9Jz8EI0TRqA9RT7wC1jQYfQwceRVMOAkyzNFJgaYw6778iMlHFbXmN6orL0cbRrtJcmAOc+0uUV9bZ599dq/fZ0bxMMqNrURqwgntn3P5ZYx58QLKhx3Osre3MuOcceYTdhcUTSOraBp/OORynvrmKe5fej/nTjiAu0tO5bDaCtj6Cfz3HkCbfSXFh8L4OTBuDgw7ACwykE+IrkiAGCCeX/s8d356J/nufJ6Y/QT7Ze/XcSetoXw5rH8H1r3DzjdWEvjGS9ExYewHnwj7nQxjv9XpVfLqj3ZgRDWTj90VDGp3mCOYsop2DxCJpdkAeOABc/TSvHldd4zvzunxEtJ+Yo1dz4Voy5qezoi5Z1Lw+lJWvG/hoBNLcHnT2u2jlOKiiRdxSMEhXP/f6/nBuif44QE/5H9OvhtbuBlKl8CWj83P7r07zJu3AMbNNm9jjwOXLFIkRFsSIFIsakS5b8l9PPXNU8womsG9x9zbfghrfRls/Rg2vGd+uTWZeY8CTKRqpY/0WYeR8eBfWtvaDUNT1xympjlMfSBMcyhGcyjClne2Yi1w8vz6CppXbScUNVArvsAG/N9XTRhfr0JrjaHhvIomGtw2/v7GN6RZLaTZLK33LrsVr9OGx2HD67DR2GjObq73R/A6bVgTXEs6ao1AIPGr9+yLL2bMMxdSkX8oy97ZxpFnju10v/2z9+fZU5/lrk/v4pHlj7CkfAl3z7ybopZAcPzN5rDd9e/Curfgm9dg2T/MEVSjjoaJp5k3nyQJFEINtlmp06dP10uXLk11MfpFfaie6/57HZ/s+IRLJl3CvGk/wVq1GrZ+Cts+hW2fQYM5DFQ7M2guPoby/JmsSzuA/Ft+hQ6FeOLyu9hh2Knxm0Ghzh/G2O2/dHTEwrnNDl5xh1mTZk5gS7NamFX5HiXNm3l2/A9QCiwWhUUpHmt28LElyn0qSCTW/e/HoWozAEv0KJQCn8NGhttOhsu8ZbrSSHfZyXS3/Gze5z+zGHckDd+Nx+Nz2vE6eg4uNU89xbvPlVFbdAiX3DMTp6f7Ws5rG1/j9k9ux6Is3DbjNmaPnN1xp1gUyj6HtW/A6ldh5zo0imjRdJrGnEJ18Qk0uIoIhA0CkRjBSKz1PhiJEYoYxLTGMDRRQ7c+jhmgFKTZLNitFhw2C3arwh4PtF6HjXRX28/JTrrLjt0qTV4iaRK7emt7gASIFNCajds/5Scf3sj2YA23eCZwel0DqvwrLNEAAA32fNY4JvOFMZ4PAqP5JFBMDHPI6dXLnuPkzZ9y+6wfUz1hKjmeNLLb3LLcaeR408hwmV+865/fRNMOP6fedAhedxouuxWrRfH0LTeglOL8Bfe0KZqm7Fcf4Tt6OBknj8YwNBHDIBw1b/5wjOZwlKZglKaQeWsORWkMRmkIRKjv5tY22NwTrOVQewHfsoTMb1LAk2Y1g4XThs9pw5Nmw2m34LBbcdqsuJTBnD/fxcoJ/0N0fy/G5EwsSpnBTZnNTEqZtahwTBONGdSEt7O44ffUxjZSYjueUep8wlEbgbD5Re8Pm1/0/nAUfyhKYWQLJ6rPOMm6hMmWLQCsNEbxWuwIXjMOZ6su6PK/1WZRWCwKq1JYLQqtNZGYJhxLrCmt5TPIT3cyLN3JsIz4Lf64MMNJSbabTHdazycSoqNeBwhpYkqWcDPUl5qTw+q3oetLCVVvJlq1nqXBLdyY4yFNa/5WUcXoUDWfG8NZZRzD58YEPjcmUBvLo9jlpjjHxahMFzMyXQxLdzLim6X4XvoE36Xf55n5V/dYjIbqAJ+ur2f6yaMYltV+MlntjjLGTT+83TYdjkFMt/ZBWCwKh8WKw2YGp5zdzv+b3/wGgOuuu67bcmit8Ydj1Aci1Pkj1D+7iLQKJ7/+Vj7NLh+NQTPINIUi8XvzVtNsEIyaV+qhaIydI6ZxXOUXlOsDeaysmjpr1xc4dqvCZrFgs16BNecNtqa/S2nsS7IC3yXHMg2X3Uq2xwyYLrsVj8OGxzEad9psPndY2RbZwaiqdynZ/ha/2PkMv+AZgnkHEBx/GsakM3DkjSXNZsGqzMDQ3XtvCRSRqEE4ZtAYjFIfiLQLqnX+CHWBMJWNIcrrg3y2qYbKxo61uAyXnZE5bkbmeBiV46Yk283oXA9j87xkeSR4iP4jAaK3YhGzDbuxHBp3mLemCmjcgW4oJ1q/HRq2Yw/XtT9MW6gim7+nZ/Jcrg9fyMv42lP5c+bBZOUWUZLroSTbzeVZLm7NcpPrTeswzDRaVcXGq+/FNnEiRfOuSai4qz7cjgImHV3UbnuwqYlAQ327HEwARnPiifoAmpube94J8+re/AK2UZTpYnNJDlREmJmlKD50TELnANCxb7H6nLnUxiZyk9fHWTcdDQo0YGiN1mC1KGwWtdvn922WlC/h9k9uZ1P9n5lWcjzzD5vPME93fQ2jgCOBX0HdVvj6ZZyrXsK5+E5YfCcUHQSTzoRJp0N21+9BKUWaTZFms0B8yY2C9MTer2FodjaHqWgIUlYXYFuNn807m9my089X2+p4bfn2dk2KOZ40xuZ7GZfvZWyeeT8+30thhjOhYctCtCUBoq1QEzSUmVf+DWVmB3HjjjbBoBzdXIWi/RWdgYUalckOI5NyI5NKfQhlOpdylUfUNxx7dgmu3GzWGk+xuum/HFEwi3uPvYssV+Jj8rXWbP/VrzD8fobfdy+WtJ6vFGMRg9UfbWfUAbn4stvPrq5pSdK3+xDXQO8CxDHHHNPzTp1wF2YRppLm7Tt7dZyyWhn3h7up+ukDrLSdzpePL+bQyzrPO7W7Q4cdyvOnPc9jXz/G/371v5z+0un8+MAfc9Gki7Bbehi1lVkCM35i3mq3wNcvw6oX4Z1bzVvBVDNQTDwd8vfv1XvqjsWiyPM5yPM5mDK8Y/6tcNSgrC7A5upm1lc2mbeqJl5bvoP6wK6U6j6HjXEFXibk+xhf4GVCgY8JBT4K0h0SOESX9q0AEfabV4K1m6B2865bfal5C9a1212jCDqyqbfmUkkWpdGD2Gh4KYtlUqkzKdfZ1FmycWcVMDIv3azy55rV/mNyPBRmOLFZLSzevphbPrqF6kA1V027iisOuKJDFtae1D79NM3//YCCm27CMW5cQsds+LKSQGOEKccO7/BcyxDX7E6GuEJia0EAHHfccQnttztfcT47qUxostzu0kpKOPLh+VT84t8s/biYopz3GH7GtxI61m6184OpP+CkUSdx92d3c//n9/Pyhpe5+YibObig+1ndrbJGmjPPj/qp+fu0+lX4+hV4/07zlrufGSz2PxUKD2ztY0mGNJuF0bkeRud6OG7//NbtWps1j/WVTayrbGJdRSNrKxp5Z3UFzy7d1rqfz2ljfL4ZMMYX+FofS+AQMBQDRCwKdVtg53oz3UT1Wqheb84sbipvt6u2ewh4R1CXNozKzP3ZEs1ibSCd5Y0+NkWyqNBZRII27FbFiGw3o/M9jMzxMCnXzSm5HkbleCjKdHU5+sYf8XPPkt/yzJpnGJ0xmiePe5IpuVN6/ZZCGzZQec+9eI4+mqyLL0r4uJUflJGR52LE/tkdnqvdUYbFaiUjv30TS6epvrtx7733AnDDDTckXC6AtGxzvetwTWJNVB2OHzaME287nWfu/Jz3n9nAqaqZzNNPS/j4Yl8xf/zWH3lv23vc/dndfO+N7zG9YDrnTjiX2SNn47A6ej4JmDWLI68ybw074Jt/m7WLRffDB/eBrxAmnGhOWhx9LKTtnaSCSilyvQ5yvQ6OGNO+52hnU4i1FU2sqzSDxtqKJt5cVc4zS9oHjnH5Xsblec0mq3hz1Yhsd8JDmcXgN3gDhBEzr/4rvzbXKKj82lyPoGYDxHbN0DXcuQR8o9mZPYPSnALWR3JZ4c9iSV06mxtd0Gj+slstiuIsF6NyPIzdz8PsXLM2MDrHQ1GmWRPoja+qvuKmD29iS8MWLp54MT87+Gc4bc6eD9xNtKqKsuuux+JyUXjXnQlf1ZVvrGfH+nqOOnccqpM/6JrtpWTkD8O621rVuxYLSqwGEQgEEtpvdxa3HY2B0ZjYbOrOZIzMZ+YF+/P+sw4++8O/mN7USPaFFyZ8vFKK40uO58jCI3lmzTM8t/Y55i+aT+ZnmZw+9nTOmXAOYzIS7x8hvRAO+6F5a642c2GtfQNWPAefPwo2pxkkJpxozsnISnyp1v6U43VwpNfBkWPbB47qphDr2gSO9ZVNvL+min99vivjbprVrLGMyfO01lzG5JkXS9mejv1mYnAbfAGibiv877FmMIgPCdUowukl1LnHUFZ0GOtihSwP5PFJQzYbahxQs+vwogwno3I9HDXSw8XxWsDoPA8jstxmJ2IfhWNh/vzVn/nbyr9R4C7gryf8lcMKD9ujczW89Rblt9yKEQgw/Pe/w56f3/NBwLolFbz35De409PY/8jCTvep3V7WbhW5Fq0BIsFUG3PmzElov90piyJqiUI3K8slYuKsUWxcUcdGziLnvjswGhrJ+dEVvfqictvdXDblMi6dfCmflX/Gv9b8i6dWP8XjXz/O9ILpnDX+LA4pOIQiT1Hi5/XkwkEXmbdoyEyYuPYNWPMfWPemuU9mCYw6xkyNMnompHf8/9ibWmocuweOen+E9VVNbIj3b6yvbGJNeSNvf11BtE0PebrTxug8LyOz3YzMcTMi2x1/7CHf5+h2pJcYmAbdPIhDih36hWuPYaNlJCsjw/nUP4zP/fkE2HV1nu9ztF79j8r1MDrXzahcDyOzPbh6SF+9p0obS3l+3fO8sO4FaoI1nDXuLG449Aa8ad5enyvW2EjFnXdR/9JLOCdPpujee3CM7XzmcLvjogaLn1/P8vdLGTYmgxN/OAVvVsemEm0Y/P6Sc5h24qnMmnt5u+fqXt1A89IKhi+Y0ety99aGBe9Qs7OMafd9F7uj97WrFk21IZ6+7RN84SoOePcW0uccT/b3L8N10LQ9vqKtDlTz8vqXeX7d82xrNJtesp3ZTMmdwpTcKUzNncrU3KmdLtzUrZZMuxsXwqYPYPOHu/q+sseagaL4MHOEVO6EAZ2NNhIzKK0NsKm6iU3V/vi9OcJqe12g3egqh83CiGw3wzNdDM9yMTzTRXH8fniWi3yfU5qukm/oT5RzFI7Xwy/9HUWZLkbmuCnJ9phjwrPdlOS4GZXjwePYO39UUSPKotJF/HPtP/mo7COUUhxbfCwXTbyIwwsP7/kEnWj+7DO2z59PtKKS3B/9iNz/uRJl77m5p6k2xJv/t5LyjfUc+K0RHHnOWKxdNIvVV1bwl59czpwrruaA49snBKx5dg2hTfUUzk+s1nP33XcDMH/+/IT2b2vr7z+ifvMOPg69xpHnXsDkY4/HYt2zAL568Q7ee3w1BxWWk/Pa7zEaGnBOmUL23IvxnXxyQqO+OmNog29qvmFl9UqWVy1nZfVKNtZvRMdHshV5ihjhG0Gxr7jd/QjfCHxpCYxSMwyoWAGbFsHmRbD5Iwib6Uuwu82EgkXTzIBROM0cTmsb+HMdIjGDstoAW2v8bKnxs63Gz9adfsrqApTVBahpbt+0aLMo8n2O+IRAFwXp5sTAggwnBfFRXLk+Bz6HTZqx9tzAChBKqZOA3wNW4C9a67t3e94BPA4cAuwEvqu13tzdOadOO1h/vnRpvzQH7Ym6YB3r69azpGIJz699ngp/BfmufM6ZcA5njz+7h3H1XTNCIap+/wdq/v537CUjGH7PPbimTUvo2NJvanjrr6uIhg2Om7s/46d3PdsXYPOyz3n+17fynVt/zYhJU82JXDuaCa6uoemT7VjTHRT85KCEXnvBggUA3HrrrQnt31btC+toWl7BB5EXKF+/lqzC4cz4zkXsd8TRqF5mWdVa89qDyylbW8sZP55E2hfvUvvkk4Q3bsSam0vW+eeTdf53seXm9rqcu2sKN/H1zq9ZXr2ctbVrKWssY1vjNmpDte32y3BkUOQpothXTJGniOG+4Qz3mrdhnmF47J6OJzdi5gCL7cvMdb63f2kmaGxZ8lVZzKap7LGQM3bXfdZoM3+Uo/c11lTwh6OU1QYorQtQVmsGjYr6IOUN8Vt9EH+447rmDpvFDBZeM2jkeNLIdKeR5baTFc8ikO2xk+lOI91pJ91la53kKQZQgFBKWYG1wBygFFgCXKC1/rrNPj8GDtBaX6mUOh84S2v93e7Om6xUG1prAtEA/qifQMS8bwg3sKl+E+vr1rOhbgMb6jawM7hr3P5RRUdx3n7ncWzxsd2u2dDudcJhwqWlhDdvIbx5c+sttH49sZoaMi84n4Lrr8fi7ny0SzQSI+SPtt5Kv6lhyb83kTnMw8k/mkLWsI5fOlprwoEAwaZGgs1NrF70Hl++9irfn/8gemuI4OoaYnUhUJA2wofvmGJcUxL7Iv3qq68AOPDAAxPav636t7fQ+N5WCq49hC3ffMVHL/6D6tLN5I0czdHnX8Log6b36mqxqTbEM7d/SsgfJc1lI6/ES5a1AcfKD0j77E2csUac48aRVlJCWkkJ9pIRpJWMJG1kCbb8/F4HpQ6vH26itKmUbY3b2Na4jbLGMsqayyhrLGN703bCRvurZo/dQ54rjwJ3AXnuPPLceeS78slwZOC1e/Gmec17mwtvYwXeqg3YajaiajaYgzF2btxV22hh9+xatrXl3pNrZvhNi68P7kjftVa43QVWB9gcZqr4lvsUX6VrrWkMRSmvD1LZEKKqKUh1Y5iqphDVjSGqmkJUNYbi+cci3aYzSbNZSHfaSHfa8Tlt+Jx23GnmzHlXmhVPmhV3mg13mhV3mhWH3YrDZsFhs+KwW3DG71sSVtosZk4t86awxe8typygae0wSXPAGFAB4kjg/2mtT4z/fCOA1vrXbfZ5M77Px0opG1AO5OluCjW5YJz+50X3J6XMKadUl/+DuvWfTg6zmBOqOvxRa41hGGij4x+P2+bDbnGg7BYc47NwTczGuX82Vt/ea75oXlJO7fPr2m3TFk0kFiIcDUJPF36d/RHGZ1NrrUGbTf677dCXIgsxaE1+4IwBlYtpOLCtzc+lwO4N8637aK2jSql6zHQ/1W13UkpdAVwBsH/eKBqjDXtUoMQ/HdV+f9W7o7s/dfzqou0tscPix+1KSqeUMoOD1ULHyrhZYovNhs3mwGKzYY3fLDYbjiwvvgMLcY7LRNn3vAp+1113AfDLX/6y18e6D8rH4rZjBKIYoSg6FEOHYsSCEWq3lOKv7f0kug60xohpYjGNETPMoNHmuZab1p1Gk71Md4hfupNHvTjbnu+R6o9CDAgDd4hEG1rrR4BHwGxiOuIPc1NcItEiEon0vFMXlM2Ca/Lu6f9M2UzY4/MKIfpHMgNEGTCizc/F8W2d7VMab2LKwOysFoPEJZdckuoiCCGSJJkBYgkwXik1GjMQnA/sPs31FeB7wMfAucB73fU/iIFn9OjRqS6CECJJkhYg4n0KVwNvYnY3/k1rvUopdRuwVGv9CvBX4Aml1HrM+c7nJ6s8IjnuvPNOAG666aYUl0QI0d+S2gehtX4deH23bbe0eRwEzktmGURyRaPRVBdBCJEkg6KTWgxcP/7xj1NdBCFEkkiAEH2Sl5eX6iIIIZIkNfkqxJBxxx13cMcdd6S6GEKIJJAahOiTWKyzKXpCiKFg0GVzVUo1AmtSXY4BIpfdZp3vw+Sz2EU+i13ks9jFqbXu1ZKWg7EGsUZrPT3VhRgIlFJL5bMwyWexi3wWu8hnsYtSqtdZTqUPQgghRKckQAghhOjUYAwQj6S6AAOIfBa7yGexi3wWu8hnsUuvP4tB10kthBBi7xiMNQghhBB7gQQIIYQQnRpUAUIpdZJSao1Sar1San6qy5MqSqkRSqn3lVJfK6VWKaV+luoypZJSyqqU+lIp9e9UlyXVlFKZSqnnlFLfKKVWx5f+3ScppX4e//tYqZR6WinlTHWZ9hal1N+UUpVKqZVttmUrpd5WSq2L32f1dJ5BEyCUUlbgQeBkYBJwgVJqUmpLlTJR4Fqt9STgCOCqffizAPgZsDrVhRggfg+8obXeHziQffRzUUoNB34KTI9PDrOyby0n8Chw0m7b5gPvaq3HA+/Gf+7WoAkQwGHAeq31Rq11GHgGOCPFZUoJrfUOrfUX8ceNmF8Cw1NbqtRQShUD3wb+kuqypJpSKgM4BnOdFbTWYa11XUoLlVo2wBVfrdINbE9xefYarfUHmGvstHUG8Fj88WPAmT2dZzAFiOHAtjY/l7KPfim2pZQaBRwEfJrioqTK74AbACPF5RgIRgNVwN/jTW5/UUp5Ul2oVNBalwG/AbYCO4B6rfVbqS1VyhVorXfEH5cDBT0dMJgChNiNUsoLPA9co7VuSHV59jal1KlApdb681SXZYCwAQcDf9ZaHwQ0k0AzwlAUb18/AzNoFgEepdTFqS3VwBFf2rnHOQ6DKUCUASPa/Fwc37ZPUkrZMYPDP7TWL6S6PClyFHC6UmozZpPjt5RST6a2SClVCpRqrVtqk89hBox90Wxgk9a6SmsdAV4AZqS4TKlWoZQqBIjfV/Z0wGAKEEuA8Uqp0UqpNMwOp1dSXKaUUEopzHbm1VrrB1JdnlTRWt+otS7WWo/C/H14T2u9z14laq3LgW1Kqf3im44Hvk5hkVJpK3CEUsod/3s5nn20w76NV4DvxR9/D3i5pwMGTTZXrXVUKXU18CbmiIS/aa1XpbhYqXIUMBdYoZRaFt/2y/ga4GLf9hPgH/GLqI3A91NcnpTQWn+qlHoO+AJz1N+X7ENpN5RSTwOzgFylVClwK3A38E+l1OXAFuA7PZ5HUm0IIYTozGBqYhJCCLEXSYAQQgjRKQkQQgghOiUBQgghRKckQAghhOiUBAghuhDPjPrj+ONZvc0Wq5S6VClVlJzSCZF8EiCE6Fom8OM+HH8pZpoHIQYlmQchRBeUUi0Zg9cAEczcRtXAFOBz4GKttVZKHQI8AHjjz1+KOZnxUcx0MAHgSOB64DTABSwGfqTlD1AMYBIghOhCPFPuv7XWU5RSszBTE0zGTBv9EeYX/qfAf4EztNZVSqnvAidqrS9TSi0ErtNaL42fL1trXRN//ATwT631q3v3XQmRuEGTakOIAeAzrXUpQDzFySigDrNG8baZ8gcrZnrpzhynlLoBc22CbGAVIAFCDFgSIIRIXKjN4xjm348CVmmtu13aM77c5UOYK5xtU0r9P2CfWQJTDE7SSS1E1xoBXw/7rAHyWtZ+VkrZlVKTOzm+JRhUx9fxOLe/CytEf5MahBBd0FrvVEp9FF/4PQBUdLJPWCl1LvCH+JKfNsxV7lZhdlI/rJRq6aT+P2Al5mpeS/bKmxCiD6STWgghRKekiUkIIUSnJEAIIYTolAQIIYQQnZIAIYQQolMSIIQQQnRKAoQQQohOSYAQQgjRqf8PmtElabrVmLUAAAAASUVORK5CYII=\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "scale_log_file_pcmad = tempfile.mkstemp(suffix=\".json\")[1]\n", "\n", "distance_adaptive = pyabc.AdaptivePNormDistance(\n", " p=1,\n", " scale_function=pyabc.distance.pcmad, # new, previously mad\n", " scale_log_file=scale_log_file_pcmad,\n", ")\n", "\n", "abc = pyabc.ABCSMC(\n", " model,\n", " prior,\n", " distance_adaptive,\n", " acceptor=pyabc.UniformAcceptor(use_complete_history=True),\n", ")\n", "abc.new(db_path, observation)\n", "h_pcmad = abc.run(max_nr_populations=7)\n", "\n", "plot_history(h_pcmad)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This visibly improves the obtained estimates further. This is due to the respective weights assigned: PCMAD correctly identifies the outlier and assigns a low weight." ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "pyabc.visualization.plot_distance_weights(\n", " [scale_log_file_l1, scale_log_file_pcmad],\n", " labels=[\"L1+MAD\", \"L1+PCMAD\"],\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This result is further obtained with less samples." ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "pyabc.visualization.plot_sample_numbers(\n", " [h_uni, h_ada, h_l1, h_pcmad],\n", " [\"Uniform+L2\", \"Ada.+L2\", \"Ada.+L1\", \"Ada.+L1+PCMAD\"],\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Conclusion\n", "\n", "In general, we recommend the consistent use of an L1 norm over L2, unless squared residuals are needed. This makes the analysis robust to outliers, and further has also shown advantageous on high-dimensional data, where highly deviant points are likely to exist. Further, PCMAD has proven stable and advantageous on multiple test problems, on both outlier-free and outlier-corrupted data. On highly flexible models, it may be unable to reliably detect errors, such that it resorts to MAD. In the implementation in pyABC, PCMAD informs about potential outliers early-on via the logging modules." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Problems\n", "\n", "The scale-normalized outlier-robust distance functions presented above clearly still have several shortcomings. In particular, they do not account for how \"informative\" data points are. This can e.g. be problematic if the data contain large quantities of uninformative data points with background noise that would be enlarged by the here-employed scale normalization. Another problem occurs when many data points are informative of one parameter, but only few of another, in which case one could argue that the impact of the many and the few should be balanced better in order to extract information on both parameters. To some degree, these problems can be tackled by additional `fixed_weights`, which pyABC allows to define manually, but this may be cumbersome, and it may be preferable to a-priori define a set of similarly informative summary statistics, e.g. using the [regression-based summary statistics](https://rss.onlinelibrary.wiley.com/doi/pdfdirect/10.1111/j.1467-9868.2011.01010.x) by Fearnhead and Prangle, which are also implemented in pyABC.\n", "\n", "For an overview of approaches accounting for both scale and informativeness of data, see the notebook on \"Informative distances and summary statistics\"." ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.9.7" } }, "nbformat": 4, "nbformat_minor": 4 }