{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Graphical Data Analysis\n", "\n", "When you get a set of measurements, ask yourself:\n", "- What do you want to learn from this data?\n", "- What is your hypothesis, and what would it look like if the data supports or does not support your hypothesis?\n", "- **Plot your data!** (and always label your plots clearly)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Reporting of Numbers\n", "\n", "- Keep track of **units**, and always report units with your numbers!\n", " - Make sure to check metadata about how the measurements were made\n", "- Significant figures\n", " - From our snow depth example last week:\n", " - Should I report a snow depth value of 20.3521 cm?\n", " - Should I report a snow depth value of 2035 mm?\n", " - Should I report a snow depth value of 20.0000 cm?\n", " - Consider the certainty with which you know a value. Don't include any more precision beyond that\n", " - Note: Rounding errors - Allow the computer to include full precision for intermediate calculations, round to significant figures for the final result of the computation that you report in the answer\n", " \n", "---" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "To start, we will import some python packages:" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "# numpy has a lot of math and statistics functions we'll need to use\n", "import numpy as np\n", "\n", " # pandas gives us a way to work with and plot tabular datasets easily (called \"dataframes\")\n", "import pandas as pd\n", "\n", "# we'll use matplotlib for plotting here (it works behind the scenes in pandas)\n", "import matplotlib.pyplot as plt \n", "\n", "# tell jupyter to make out plots \"inline\" in the notbeook\n", "%matplotlib inline " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Why are you plotting?\n", "\n", "**You have an application in mind with your data.** This application should inform your choice of analysis technique, what you want to plot and visualize." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Open our [file](https://mountain-hydrology-research-group.github.io/data-analysis/_downloads/55754612f1cf8f2d340fa84ba0f399b4/my_data.csv) using the pandas [read_csv function](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.read_csv.html)." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "# Use pandas.read_csv() function to open this file.\n", "# This stores the data in a \"Data Frame\"\n", "my_data = pd.read_csv('my_data.csv')" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
timetair_maxtair_mincumulative_precip
01920-12-3120.455167-9.901765102.502512
11921-12-3120.119887-10.36425497.108113
21922-12-3119.872675-10.31318197.166797
31923-12-3120.449070-11.35963997.902843
41924-12-3120.449110-10.04653999.329978
\n", "
" ], "text/plain": [ " time tair_max tair_min cumulative_precip\n", "0 1920-12-31 20.455167 -9.901765 102.502512\n", "1 1921-12-31 20.119887 -10.364254 97.108113\n", "2 1922-12-31 19.872675 -10.313181 97.166797\n", "3 1923-12-31 20.449070 -11.359639 97.902843\n", "4 1924-12-31 20.449110 -10.046539 99.329978" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# look at the first few rows of data with the .head() method\n", "my_data.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Scatterplots\n", "\n", "- If we're looking for relationships btween variables within our data, try making [scatterplots](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.plot.scatter.html). \n", "- Later this quarter we'll get into statistical tests for correllation where we'll use scatterplots to visualize our data. \n", "- **Remember that correlation =/= causation!**" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "my_data.plot.scatter(x='tair_max', y='tair_min')\n", "plt.title('Maximum vs Minimum\\nAir Temperature');" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Timeseries plots\n", "\n", "- If we are interested in how some random variable changes over time.\n", "- Similarly, if we have a spatial dimension and are interested in how a variable change along some length we could make a spatial plot." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYwAAAEWCAYAAAB1xKBvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nOx9aZgcV3X2e3qfXmY0Gu3ryMa2vGEDsnGwAQsDMQ6LE0KCE/gMST74AknsQEKMHwiGAHESQgIBAg4EA5HDZrywhF3GMUY2kpEtGUm2JI9k7TOjWXut7j7fj3tv1a3qqq7qmenpmVG9zzPP9FZVt6ur7rnvec9CzIwQIUKECBHCD5FODyBEiBAhQswPhAYjRIgQIUIEQmgwQoQIESJEIIQGI0SIECFCBEJoMEKECBEiRCCEBiNEiBAhQgRCaDDmIYhokojO6vQ45hOIiInoOVPc9lYi+vxMjynE9EBELyeixzs9jmYgoqS8X1d1eiwzgdBgzDCIaICIKkS0xPH6Tjlp9U/3GMycZeaD091Pu0BEt8nvenmnx9IKiOhOIqo6b25m/igz/0nAffyhnCAmiahIRHXt+WR7Rj47IKJriWj/LB7vp9q5M4iorD3/V2b+MTNfMlvjmQqYuSzv12OdHstMIDQY7cEzAG5QT4joYgBdnRvO7IGICMCbAZwGcGOHhxMYRJQB8HoAYwD+sIXtYvpzZt4iJ4gsgFcBOKaey9fmJIgoQkRtnQ+c58oPzPwy7bzdDeDvtHN5c3tGOXNo9fvOB4QGoz34CoD/oz2/EcCX9Q8Q0W8R0a+IaJyIniWi27T3fp+IDhJRt3z+KiI6QURL5XPTvSJXxZ8hov+RK6+fE9EKIvpXIhohor1E9Dxt3zbXjNz+w/Lx1UR0hIjeQ0SniOg4EV1PRNcR0VNEdJqIbvX57i8GsArATQDeSEQJ7VhvIaKHiOhjcmzPENGrtPcfIKK/k99hgoh+qJiaGpvjHA4Q0cvl48uJ6BdENCrH/Sn92AHwegCjAD4Eh6GTjOm/5ON+eQ7/mIgOA/hpC8dQ+1tLRPcR0ZD8nf+f9t7tRLSFiL4mf8+dRLSBiD4gPz9ARJu1z2+T52wHEY0R0d1E1KO9/2IiekSel8eI6ErHth8iokcAFACsIqK3y2tmgoj2E9Efyc/2AbgHwFnaKr+PiL5KRO/T9mljIfK6/SsiehLAuN/3b/E8uh3rXUT0pBzfvxPRSiL6kbzPvq/uqQDn5v/Kcz0hx/gG7b23E9E+eT98l4hWy9dT8tr4UyI6AGC39toa+ZkuEvfms3K8/0ZESfneCjnGUSIaJqKWr622g5nDvxn8AzAA4OUA9gE4H0AUwLMA1gNgAP3yc1cDuBjCaD8XwEkA12v72QLgTgB9AI4BeLX2HgN4jnx8J4AhAC8AkIKYwJ6BMFhRAB8GsNVtW237D2tjqgL4WwBxAP8XwCCAuwDkAFwIoATgrCbf/wsAvi63HwbwO9p7bwFgyP1GAfyp/G4k338AwAEA50IwsgcA3K6N7YjbuZaPXwDgCgAxAP0A9gC42et7u4z7JwD+EcByeQ6er713G4D/ko/75b6+DCADoKvJPt3GHAWwC8DfAEjI73oYwEvl+7dDTN6b5Xf5mvw9/0o+/3MAe7T9bQNwCMBGAFkA3wbweW2swxDXYwTAdfL37NW2PQjgPPl7xQC8FsAGACS3KwK4UH7+WgD7Hd/nqwDepz23fQbACQC/hFhEdPl9/ybn0nacJsf6XwBLAKwDMALgUYj7rAvAQwD+xu/cyL9RAGfLz64GcL58/EZ5bZ0rz5l5f0HcfwzguwAWyWOq19bIz3wWwDfl+z0AfgDgA/K9fwHwCfk7JAC8pNPzmfMvZBjtg2IZrwCwF8BR/U1mfoCZdzFznZmfAPDfAF6qfeSdAF4GMWl+m5m/0+RY9zDzDmYuQawCS8z8ZWauQUw4z2uyrRMGgI8wswFxky4B8AlmnmDmJwE8CWHgGkBEaQBvAHCX3P6baHRLHWLm/5Bj+xKAlRCTtMIXmfkpZi5CGJ5Lgwxafv9tzFxl5gEAn4P9fHqCiNZBTNB3MfNJCOPh5067jZnzcpyt4CoAKWb+B2auMPNTAL4IMREp/ISZtzJzFeIcdgP4Z/n8qwA2EpHu4vwiM+9l5kkAH4DlDr0RwLdY+PrrzPw9AL8G8Ept288z8z5mNuS5u5+Zn2GBHwP4mRzzdPAvzHxMnqsg3386+FdmHmLmwwAeBvBzeZ8VAdwH614Icm4uIqIUMx9l5j3ytbdDLLCektf4BwFcRUT6NfwRZh51XhskXFR/BOAm+f4YxAJBfXcDwrCuk+fmwRk6JzOG0GC0D18B8AcQq+ovO98kohcS0VYiGiSiMQD/D2JyBgAw8yiAbwC4CMA/+xzrpPa46PK8Fd/5sJzM1bZu+/fa329DrM6/J59vAfAqkq40iRPqATMX5MOs2/sQK+1AYyeic4noO5LmjwP4KLTz6YM3Q6zad2rj/gMiijfZ5tmA+3ZiPYB+6XYYJaJRAO8CsEL7jPN8D7JcgsL6TTIeYzkEIC3dUusBvMlxrE0Qk5Lr9yCi1xLRo9LdMgqxaAl6Hr2gHyPI958Ogt4LnueGmUcgdKy/AHCCiO4ny427HsBntW0GIa75NdpxvK6NVRCs5Elt+3sBLJPvfwSCcW+V7sB3tfzt24zQYLQJzHwIwpVwHYBvuXzkLgD3A1jLzD0QVJXUm0R0KcRq5L8BfHIGh1YAkNaez9SNCohVWxbAYSI6AWHw4tACAKaBPLRxE1EUgG6I/h2CyZ3DzN0AboV2Pn3wfyB88yfkuD8OMUm+qsk2Uy3z/CyAvcy8SPvLMfNvT3F/ALBWe7wOQEGuXp+FYBD6sTLM/C/a583vQUL4/waAvwOwjJkXQbg4yflZDbbfBe7Xk75dO77/VND03DDzd5n5GohJ/jDE9aW2e4tjuy5m3qHt2+vaOA5hXM7Wtu1h5j55zDFmvomZ10Noau/TdZW5gNBgtBd/DOBlzJx3eS8H4DQzl0iEn/6BeoOIUgD+C2LSeyuA1UT0jhka006I1XOUiK5FQLeNH6Twdw2AV0O4kS4FcAmAf8DMREs9BSBFIlggDuB9AJLa+zkIUXWSiDZC6CNBxv0bAM4GcLk27osgDHo7orwekse9WQqiMSJ6LhE9fxr7fItkWFkIveVr8vUvAXgDEV0jf+8u+dhrkdAFYeBPAagT0WshdBiFkwCWyeMo7ATwaiJaJK+BP/cZazu+/1TgeW6IaLW8ztIAygAmASjW/VmIifw8+T16iej1QQ4oXVj/CeATRLSEBNYS0Svkvl5LIsCBIKL1atpx5wRCg9FGMPMBZt7u8fY7AHyIiCYgROava+/9PYRY+u/MXAbwJgAfJqJzZmBYNwF4DYSo94cQlHgm8GYAO5n5h8x8Qv1BsKPnEtFF09m5XDG/A8DnIfSgPAA9auqvIIzuBID/gDVp+uFGAPdJP7c+7k9ATISLpzNuJ+SkcR2AF0G4jwYhVq/TCbn9CgQTPQqgDuDd8lgHIVaqH4QIjDgE8fu73vfMPARxHr8NIQhfD8u9CACPQ7DiQ9KlshhiAtwPsQr/jhyHJ9r0/VuGz7mJAngvhHt0GMBlkIaQmf8bwKcAfEu6PndC6JRBcTOE22k7hFH4PgDl7jofQrOcAPAggI8x87apfsd2QEWnhAgRYh6CiLYB+BQz/1enxxJi4SNkGCFChAgRIhBCgxEiRIgQIQIhdEmFCBEiRIhACBlGiBAhQoQIhAVXHEvHkiVLuL+/v9PDCBEiRIh5gx07dgwx81K39xa0wejv78f27V5RrSFChAgRwgkiOuT1XuiSChEiRIgQgRAajBAhQoQIEQihwQgRIkSIEIGwoDWMECFCLHwYhoEjR46gVCp1eijzCqlUCmvWrEE83qwosx2hwQgRIsS8xpEjR5DL5dDf3w9Rty+EH5gZw8PDOHLkCDZs2BB4u9AlFSJEiHmNUqmEvr6+0Fi0ACJCX19fy6wsNBghQoSY9wiNReuYyjkLDUaIEG3ERMnAfTuP+n8wRIh5gNBghAjRRnz3ieO46as7cWIsFGRDzH+EBiNEiDZivGQAAErGnGqcFmIGMTo6is985jO+n3vRi140C6NpL0KDESJEG5EvC0NRrdc7PJIQ7UJQg/Hwww83vFarza+FRMthtbJRfImZ59c3DRGiAyhUqgCASjVsIzAb+OC3n8Svj43P6D4vWNWND7zmQs/3b7nlFhw4cACXXnopNm/ejCeeeAIjIyMwDAMf/vCH8brXvQ4AkM1mMTk5iQceeAAf/OAHsXLlSuzcuRO//vWvG/Y5MDCAa6+9FldddRW2bduGSy65BG9961vxgQ98AKdOncKWLVtw+eWX49FHH8XNN9+MYrGIrq4ufPGLX8R5552Hj3/849i9ezf+8z//E7t27cINN9yARx99FOl0elrnwtdgEFEEwBsh+j9fBtEUPUlEgxD9fu9g5qenNYoQIRYo8pWQYSx03H777di9ezd27tyJarWKQqGA7u5uDA0N4YorrsBrX/vahoikRx99FLt3726aA7F//3584xvfwB133IHLLrsMd911Fx566CHcf//9+OhHP4p7770XGzduxIMPPohYLIYf//jHuPXWW3H33Xfj5ptvxtVXX4177rkHH/nIR/C5z31u2sYCCMYwtgL4MURT9N3MXAcA2QB+M4DbieiesKdwiBCNyJcFwzBqocGYDTRjArMBZsatt96KBx98EJFIBEePHsXJkyexYsUK2+cuv/xy34S5DRs24OKLLwYAXHjhhbjmmmtARLj44osxMDAAABgbG8ONN96Ip59+GkQEwxCaWSQSwZ133onnPve5ePvb344rr7xyRr5fEIPxcmY2nC8y82kAdwO4m4iC55aHCHEGQWkYRi10SZ0J2LJlCwYHB7Fjxw7E43H09/e7JsdlMhnffSWTSfNxJBIxn0ciEVSrYiHy/ve/H5s3b8Y999yDgYEBXH311eY2Tz/9NLLZLI4dOzbNb2UhiOh9CxG9u9kH3AxKiBAhLA0jZBgLF7lcDhMTEwDEin/ZsmWIx+PYunUrDh3ybC0xIxgbG8Pq1asBAHfeeaft9ZtuugkPPvgghoeH8c1vfnNGjhfEYLwZwL87XySiPyGi987IKEKEWKAwNYyQYSxY9PX14corr8RFF12EnTt3Yvv27di0aRO2bNmCjRs3tvXY73nPe/De974XV155pS3i6i//8i/xjne8A+eeey6+8IUv4JZbbsGpU6emfTxibn4hE9HjzHyJy+tJAI8xc2edhk2wadMmDjvuhegkXv7xn2H/qUnc8eYX4JUXrvDfIETL2LNnD84///xOD2Newu3cEdEOZt7k9vkgDKNIRCudLzJzGUB1SqMMEeIMQcEUvUOGEWL+I4jB+GcA9xHRev1FIloGILBjlojWEtFWItpDRE8S0U3y9TfI53UicrVq8nPXEtE+ItpPRLcEPW6IEJ1EGFYbohmGh4dx6aWXNvwNDw93emiu8I2SYuZvEFEawA4i2gZgJ4SheQOA21o4VhXAu5n5MSLKyf39CMBuAL8D4HNeGxJRFMCnAbwCwBEAvySi+5m5MeMlRIg5AmY2w2or1dBgtBPMPC8r1vb19WHnzp0dObafHOGGQKVBmPlLADYA+DqAOIASgBuYeUsLgzvOzI/JxxMA9gBYzcx7mHmfz+aXA9jPzAeZuQLgqwBeF/TYIUJ0ApVaHdW6uCnV/xAzj1QqheHh4SlNgGcqVAOlVCrV0naBSoNIhlFn5i9PZXAu++sH8DwAjwTcZDWAZ7XnRwC80GPfbwPwNgBYt27dlMcYIsR0UShbUSthWG37sGbNGhw5cgSDg4OdHsq8gmrR2gqClAb5CwC/AYCJ6FFm/tcpjk/tLwuR8HczMwct+uLGNV2XE8x8B4A7ABElNaVBhggxA8hXrJiQUPRuH+LxeEttRkNMHUEYxo0ANkFM2tsBTNlgyIzwuwFsYeZvtbDpEQBrtedrAMxc+mKIEG1APmQYIRYYghiMfwSgJvfpGAsC8AUAe5j54y1u/ksA5xDRBgBHIYoh/sFUxxIixGxAZxjV0GCEWAAIEiX1NQBfm4FjXQmRNb6LiFRYwK0AkgD+DcBSAN8lop3M/JtEtArA55n5OmauEtGfAfgBgCiA/2TmJ2dgTCFCtA26hlEJXVIhFgCCaBjEPuEHQT7DzA/BXYsAgHtcPn8MwHXa8+9BlFMPEWJeYLIcMowQCwtBwmq3EtGfE5Et5IiIEkT0MiL6EoTOESJECA0Fm+gdGowQ8x9BDMa1AGoA/puIjhHRr4noIICnAdwA4F+Y+c42jjFEiDmJhw8M4c6fP+P5vsryBsIoqRALA0E0jBKAzwD4jIxyWgKgyMyj7R5ciBBzGd967Cge2HcKb7nSPaRT1ZFKxSMhwwixINBST2/Z9+J4m8YSIsS8Qq3OKBnehiBfroIIyKXiYXnzEAsCgUqDhAgRohFGrY6SUfN8P1+pIR2PIhGNwAiLD4ZYAAgNRogQU0StzqjW2dPdVKhUkU7GkIhFQg0jxIJAYINBAm8ior+Vz9cR0eXtG1qIEHMbygh4sYx8uYZsMoZYhMKw2hALAq0wjM9A1JS6QT6fgCg5HiLEGYmadDN56Rj5chXpRBTxaCh6h1gYaEX0fiEzP5+IfgUAzDxCRIk2jStEiDkPVbLck2FUqsgkYihXa6FLKsSCQCsMw5CNjBgAiGgpWui4FyLEQkPVxyVVqNSQSYYMI8TCQSsG45MQJTyWEdFHADwE4KNtGVWIEPMAqu1q0cNgTJaF6B2LUhhWG2JBIGgDJQLwIIAdAK6BqAl1PTPvaePYQoSY07BcUh5RUuUaMokoxqMRTBhV18+ECDGfEMhgMDMT0b3M/AIAe9s8phAh5gUUa/BiGPlKFelEDPFoxGQjIULMZ7TiktpGRJe1bSQhQgTEnuPj+OtvPI5ah/tkNxO9mRmFigirjUcJRjV0Sc1n/PrYOH6y52Snh9FxtGIwNgP4BREdIKIniGgXET3RroGFCOGFn+8fwjd2HMHwZLmj41C5FW4Go1yto1ZnpJNRxMJM73mPzz14ADd/bSd8ujgseLQSVvuqto0iRIgWUK6KyVevBtsJ1JowjLwsPJhJxERpkDBKal4jX65iolTFoeEC+pdkOj2cjiGwwWDmQ+0cSIgQQWEajHJnhWSjSeJeQRqzdCIqM73P7JXpfIf6PXcdHQsNRhCokiBOMPOHZm44IUL4o1wVN2+h0wyjieituu1lkzHhkpqjDOPh/UP40Hd+jfv+7EokY9FOD2fOQl1ru4+O4TWXrOrwaDqHVjSMvPZXg3BR9bdhTCFCNEXZUC6pzjKMZqK36raXTsaQiNKczfTefWwMe09MYLRgdHoocxrq99x1dKzDI+ksWnFJ/bP+nIg+BuD+GR9RiBA+qMjVeqHcWYahDIYbw8jLsWUSQvSeq8UHixVv4T6EBd0lxcwQqWlnHqZT3jwN4KyZGkiIEEExZxiGNAJlVw1DMgyZhzFXGYYyds0aQYUAipUaUvGIKXyfqWilvPkuGU77BBE9CWAfgE+0b2ghQrhDaRidFr2buaQmJcMw8zDq9TkZkqnGrs5pCHcUKjU8f10vgDPbLdUKw9gD4DXy75UAVgNY245BhQjRDBUZJdVp0buZS8rSMETxQWZ0PNHQDWqcIcPwRr3OKBo1XLp2ERLRCHaHBiMQnsPMh+TfUdnfO8zNCDHrmCthtc0S9ywNQxQfBCwDM5dQlIYiZBjeUAuCnq44Nq7MhQyjGYjoT4loF4DzNJfUE0T0DIAw03uaGJ4s4y+/ttMMwwzhj7kQVluvM9T8X/TQMCIEpOIRJKLiNqvMQeG7WAk1DD/oOTUXre7Bbil8n4kIwjDugnBD3Q/LJfUaAC9g5je1cWxnBLYfGsE9vzqKJ54d7fRQ5g3mAsPQ2YK7hiGaJxERYhHJMOag8F0yRe+QYXhBue26EjFcvLoH46UqDp8+M4Vv37BaZh4DMAarNWuIJhgrGBjKl3H20mygz6uLcShfaeewFhTmgoah6xFlNw2jXEM6KRLh4jGxLpuLyXvq+lNGOEQj1HWWSUSxdnEOAPDEkTGs7zvzMr5bCqslol4iupyIXqL+2jWw+Yq//589eNPnHwn8eXUxnu5wIb35BKuWVOcYhl5M0DUPQ7ZnBYB4ZO4aDOVOCxmGN9Q92pWI4tzluTNa+G4lrPZPIJoo/QDAB+X/29ozrPmLRwdOY2iyHNjHqZLPhkOGERjTCaut1up46xcfxbaDw9MaQ62mu6QaDUG+XNUYhnBJzcVcDCusdu4Zs7mCoqlhxJCIRc5o4bsVhnETgMsAHGLmzQCeB2CwLaOapxgtVHBwMA+jxoFvQLV6GZoMDUZQVEwNo/VV8elCBVv3DWLLI4enNQZ/hlEzGUZMMoy5mO1thdWGDMMLeTMJUywALlzVjb0nJjo5pI6hFYNRYuYSABBRkpn3AjivPcOan/iVJlyPl4LV5lE37Ol86JIKirKpYbTOMJSRefCpwWlN4ErD6IpHPWtJZZLSJTUPoqTcdJgQAkUtSgoAlmSTGC1UzshIqVYMxhEiWgTgXgA/IqL7ABxrz7DmJ351aMR8PFkKNpkphjEcMozAsEqDtD7JKTfWWNHAzmlEpqmIp2wq5m4wyjVzgolHOx8ldf/jx/Cur+1seL1k5mHMPWM2V1DQXFIAkEvFUOfOJ452AoEMBolKW3/BzKPMfBuA9wP4AoDr2zi2eQedYUwENBh5k2GEBiMorOKDrTMMPd/lp3tPTXkMKqw2l4zBqHFDFvdkuYqsZBixaOdF75/tG8T3dh+3vVat1c1z2U6X1O6jYxgYyrdt/+2GFVYrFgDZZBxA8Ht8ISGQwWDBve7Vnv+Mme9n5nCWk6jXGTsPj+Lc5SKcNujFVDQ1jNAlFQTVmmh9GiGgYNRQbzF7WjGM7lQMW/dNXYJT7qxsShgF54RbqNTMFaliGJ0UvUcKFZSMus0Np2sv7Uzce8eWx/BPP9zXtv23GwWHSyonf/PJ8plXEr4Vl9Q2IrpsqgciorVEtJWI9hDRk0R0k3x9MRH9iIielv97PbYfkAUQdxLR9qmOo13YPziJiXIVLzlnKYDgF5Nyq4yXqqaYG8IbynXSm06AGSi1WNJCMYzfvHAF9hwfx4mx0pTGoRiGYhH65MvMIqxWRUlJhlHtYF9vxWB1N54+5naVBpkoGTh8ujCvV+OFSg2JaMT8HdUiYXwef6epohWDsRnCaByQpUF2EVErpUGqAN7NzOcDuALAO4noAgC3APgJM58D4CfyuecYmPlSZt7UwnFnBY9J/eIl5wqDEfRi0t0qI4WQsPlBGdXeTAIAWi6pokTvV8uuaQ/sm5pbytQwko0Mo2jUwAyNYXTeJaWuLT1QoFSxxtMuhvHUSRFNNJ9F9WKlarqjAMFOgeA65UJC4AZKmGahQWY+DuC4fDxBRHsgKt6+DsDV8mNfAvAAgL+ZzrE6gccOj2BROo6LV/cACO6SKlRqIAKYhVtqeXeqncOc91AMY7E0GIVyDcgF3165pJ6/bhFW9aSwdd8pvPHydS2PQ7EFN5dU3ixtLiYZVRqkky4pk2FoBnY2GIYKPy3NY/Ys3IuWwQg1jGA4DODFAG5k5kMAGMDyqRyUiPoh8jgeAbBcGhNlVJZ5bMYAfkhEO4jobU32/TYi2k5E2wcHZy9N5FeHR/G8tYtM/+ZEC2G1K6WRCCOl/KEmtt60uGlbzfZWjCSTiOHqjcvw0NNDU3IF6qI3YF+h682TACDR4dIgRq1uTm567oqNbbSJYew9Pv8ZhtNghBpGMHwGwG/Aqik1AeDTrR6QiLIA7gZwMzOPt7Dplcz8fAim806vsiTMfAczb2LmTUuXLm11eFPCWNHA06cm8fx1vYhFI0gnoi2F1a5ZnAYQRkoFQcXJMFoMbZwsV5FORBGJEF523jLkKzVsHzjd8jiUSyrjomGYpc0dDKNTYbW6q9ONYSSikZa1oKDYpxjGvDYYVdP4AxarDBlGc7yQmd8JoAQAzDwCINHKwYgoDmEstjDzt+TLJ4lopXx/JQBXpzIzH5P/TwG4B8DlrRy7VbSSlPO4DKd9nuzIlUvFWnJJre0VBiOMlPKHLnoDrZcHyZethLoXPacPEQK2PdO6wVBhtK4uKQfD6HTi3kjeWgnrorca86J03LXN7HTBzNh7QqwJ53OeR6FSs2kY2URoMILAIKIohGsIRLQUQOCrQOZyfAHAHmb+uPbW/QBulI9vBHCfy7YZIsqpxxAd/3a3MPbAqNcZV3z0J/iXHz3V9HP7T03gK9sO4SvbDuGuRw6DCLhkrdAvsskYJgLQVWZGoVLFip4kYhEKGUYAWC6pqTMM5UZKJ2LIJmMYL7buWlClQdxcUsqIOTO9O8Uw9OvKxjAqlvFtB8M4PlbCeKkqGMw8ZhhFw+6SikRI3ONnoMFoRfT+JMTKfjkRfQTA7wJ4XwvbXwngzQB2EZFKOb0VwO0Avk5Efwyhk7wBAIhoFYDPM/N1EFrJPcLmIAbgLmb+fgvHDoyIdB8c8wm3vO3+X+Oh/UPm8xes70UuJfzquVQ80MVUrtZRZzGxLM4kQg0jANRKWEVJTYdhAOK3mkrzqlrNzjB0l5RZDjtpz/TulIahGwz9uyoNoycdx9GR4owfV7mjLljVbT6ej8iXq6YXQEF4Ec48DSOwwWDmLUS0A8A18qXrmXlPC9s/BIA83r7G+YJ0QV0nHx8EcEnQY00Xy3tSODne3GCczlfw4nOW4OO/dykA0b5RIahLyqqzH0NfNonhsJ6UDcVKDeMlwxY5Vq4pDUOK3lMIq1UTOSAm9amER5pRUjJixh4lZQnrQOczvU9rGoZd6LYCCA4OTs74cfdId9Qla3rw+JFRMDPkom9eoehwSQHCi3Amdslspbx5CmICfzmAlwG4Vr624LCiO4njPgxjvGRgaTaJpTnxpyJhgOCrDzWxdCWiWJJNhCXOHfiH7+/F7372YdtrJsNQGsYUXFJZjWFkk7Ep9dVwJu65GQzlxkiYBqNDoreNYTQm7vWmE22Jktp3YgKrejqNhuwAACAASURBVFJY1p0C89ws7x4EBYdLCmhNp1xIaEXD+DKACyFcU58CcD6Ar7RjUJ3Gyp4unPQxGBOlKro1VqEjlwzm5tAZRuiSasTDB4ZwatzOupSGkUvFEItQyxVr8xW7SyozRV+00iNyLqL3pEPDiJnFBzvnksqlYkgnorZEUXX9LUon2pKHse/EBDau7EZSLqbaFYnVbuhlXhSyqTgmzkCG0YqGcR4z626hrUT0+EwPaC5geXcKE+Vqg79boV5nTJQMc7JwIrhLylqJ9mWSGA6jpEyMFQw8dVK4SWp1RlRqSyqsNhmLIp2IttwTo1HDiE2pPIhiGGZYrZY1PVIw0BWPIhV3Ju51KEqqUMHiTAKFSs3GpopGDYmYCANXBRTVeZ4uKtU69p+axOaNy5CU56Fk1NCdcl9kzVVUa3VUqnVXhnFk5Mzr690Kw/gVEV2hnhDRCwH8fOaH1Hms6EkCAE546Bj5ShV1hufFn0vFUajUfFeUelGzvmwC+UptXkeTzCQee9YqFW/PSFYGI4JMMtYyw3C6pDKJqfmi1W+bikcQj5Jt9TxSqJiJhQBARIhHCUaLhRJnCqfzFfSmE8g4DGypUkNXPGoygJlkGQeHJlGtMzauyCGl9t/GAoftQsGwFx5UyCVjZ2RpkJbyMAA8LIsADgD4BYCXTqGm1JzHiu4uAPBceao6Ud1d7gxDRc74rX71Ovt9Muon1DEEdgxoBqPiZjAkw2hBw6jW6igZdVOMBsRvNSWDISf/aISQitmbKI0WDCxK21OUYpEIjA7lIiiGkUnGGhL30gmLCc2kjqGiojau6Nb2P/8WQ+rac4reZ6qG0YpL6tq2jWKOYUWP0PI9DYaM2/dmGKqapYGetDcFN11SySj6soLVnJ6sYPWirqkNfAFhxyF3g2G6pOKSYbQw2Zs1nlIO0btcbTmCRzGMWCSCVMJuMNQErSMeJdPIzDZG8gbOW96NyVLVEVZbk64zqTHM4IS+5/gE4lHCWUszOHxauG7mUvJevc4ggu9v7ox4U8il4igawougouBmGzsOnUadgcv6F8/aMQN/U2Y+1OyvnYOcbayQYZxeLinTYHiI3t0BSwc4RW8AGApDa2HU6tj57CiWdwsjWjCs86jcJglZgqUVDWNSGuhsUi8kJ7qnufXlbgY1+ceihFQ8YludC4Zhvzbi0UhH8zAWZ+LIJKO2RMeSUUMqHkUyJs7HTE7o+06M4+ylWcSjEUv0nkMM47pP/i/uePCg7+cKHgxDuTU7GVr7d9/Zg7//XuDMhhlBK2G1m4joHiJ6bIrlzecNuhJRdKdinrkYpkvKg2FY1Sybh9Y6w2qBsAAhIArWFY0arnqOqAVWcLik4lFCJELIJFoLiXVmYOuPm/mj3/31x3HrPbtsr5kGI0LoikdtLEhoGE6G0RmDUazUUDRq6G3qkpr5Cf2pk5M4b4UoI9wOl9d0MTCcx4EAuSdFLw1jDtSTOjJSxGhhdpMHW3FJbQHw1wB2oYWSIPMVK3u6PHMxlCHw0jCsapbNLya9uXw0Il1SIcPA9kOittNLzl2Cux870uCSUitiIXq3wDBcDIZ545ernmWSf3FgCKt77W7CmmkwIkjFo6boXaszxoqGTfQGBBPpRGkQVXhwcTrRYGCLlRoyyZh5PmfKYNTrjBPjJayR56wdBmk6YGaUq/VA7NTZbU+h0wajZNQwNFluqebdTKAVgzHIzPe3bSRzDM2yvYNqGH4XU17r5BWLEBKxSMgwIPSLVT0pnLVEtLu1i94108WRSUZbyvRWn3VGSenvOVEyajg2VsJSR58Sw9Qw7KL3eNEAMxpE70Q00pHig6osiMUw7CVM+rJJJOMqSmpmxjecr6BWZzNDXzGMuaJhVGp1MAdzJxXMJMxGDQPonEtK6atjRWNWM+hbUWs+QESfJ6IbiOh31F/bRtZhrOhO+kZJZT3yMLIBe2IUK1WkpT+diLAkk8BQaDCw49AInr++F10JcXkW9LBao25m1acTdoZRqdbxX9sOmat/J9wETPVbebmkDg0LwdYZ4aT6ikcihFQiiqJ0t6gyHL2ZucEwlMEQUVJR5CtVc1VaMmq2fJGZYgCnJsR9sywnWPNc0zCU4WoludbJMNSio1P1pI6Nitpf1Tq3XIBzOmiFYbwVwEYAcVguKQbwLc8t5jFW9HRhcLIMo1Y3q40qjBcNpBPRhtcVFPPwywTNV2pIx60LcXE2cca7pI6NFnF8rIRN63vRlVBJcdZ5rNTqFsNIWBMgEeGne0/hfffuxsYVOWxyiRyZNDvh2aOkxHvuv9UzQ3kAjUl3Ro0Ri4hxpGIRnJKT4ag0GE6G0SkNQ7mketOCYTAr7SJmahhWHsbMjE9l5y9zMIy5kumtDFcQdqoWK25htUDnXFJHRq1ikWNFwzXBuB1o5SiXMPPFbRvJHMMKWf9mcKKMVY4w1/GS0TRjNRkTyVz+UVJVpLUfui+TPOPzMLbLcNoXrF9sGlOb6G1YGkZaToAlo46uRNTMvB3zKFduid72KCnA22AcGnY3GLV63Sz50ZWImuKo6j3hFL1j0UhHEvfsDMP6roqdpdrIMEyXVMxd9N51ZAxjRQNXnbNkRo4bFCqBMAjDKFbcXVJZTfvqBI45DIZzjmoXWnFJbSOiC9o2kjmGZtne48Wqp+ANCPeSKHHenK46Wz/2ZcN6Uo8dGkFXPIrzV+bMVV3BoWEkNIYBWDf+EVmie9zjvKvP6a5ENYl6rTYHpMFwtnE1alYZDV3D0EVmHYkodSRxbyRfQYRENWV1vgqSaZUMUYU1OcOZ2Cclw1gqc4uSHqL3J3/6NG779pMzcsxWoJhUEIahNJ+uuJ1hmF6EDrukAO8FUjvQisG4CsBOItq30MNqAWt15FaE0I9hALL8sR/DKDsMRiaB4fzsRz7MJRwZKWJ9XxoxGb8fIftEo7uk1KpPJUAqhjFedD/vk+Uq4lEyGQpgj5Jyg3JJVRz6Q63OpkuyK2GF1aowx0VODSMSMUuizyZOFypYlE4gGiEbwzBqdRg1tmkYQUqDMDO+9svDnkYZAE6Ol7A4kzANu5fLKy/rtc021PUUiGEYNaTikYYaW8mYCFTpVHmQo6NF87zOpsEIM709sLJHUDy30NrxkoFlueaV3YOUDigYVSzX9tOXTaJk1FGQ4Y5nIsZLhtlbhEjkODS4pOJWlBRgrQJNhtHEJeU8r8mYmAw8GcaQFL0dLqmqVqgvGY+gJCfDkUIFsQiZnfgU4rEICsXZ9+GP5K0Q34xpYK2aZa2WBjk6WsTf3L0LRo3xpivWu37m1ETZFLwB8TsmYxGUHQwjX6l1xGAow2XUWEbdRT0/6+znrSC8CJ0rD3JstITzVuTwhHTrzRZaYRiHAbwYwI0ys5shOuEtSPSm40jEIq6htROlqpnN7YVABqNsb8yi6kmdya1ax4uGrRlVlyMSqlytm/0l1ORfkMK3MhhebGGyXG0o8UBEnmywWKnhxHgJRC4Go1ZHXHNJVap11OuMEZnl7QxzjEeoI+XNRZa3uK4sA1s1GVGqxeKDytA0azB2arxkCt4KqXi0wSVVKFdRqNRmnVHrhitIvTdnhJTCVOuQTRfMjKOjRVywshuA9wKpHWjFYHwGwG8AuEE+nwDw6Rkf0RwBEWF5d9JDwzDMOGwvZJP+9fILlZptAuuT2d5DZ3CZ8/GiYSu5kk5EbVFS+opQrfzylRrGioZ58zZjGFkX5ia6pzVOHIdOC3fU2t60i+jNiGqiNyCigEalC8iJTkZJKQFed0kpkb4rLqL9ohEKxDDUZ5qVhHcyDAAN5VMAcf1X6zzr+Sm6a6xVt7GOXNJfp2wHhvMVVKp1nLciB6K5q2G8kJnfCaAEAMw8AqDxzlhAWNnd1XBjMDPGS81Fb0DUk/ItDVKp2hjG4oy4yUYKZy7DGCva9aF0ItqQa+F0SRXKVZNdAN6it7M9q4IwGI3bDEj94tzlWRg1tq2EjTojroXVAmIydZY2V+hkHobFMCxG5ix5kYxFAkVJqcn95IT7oqZeZ5yaKJt1wBRS8WgDg1HaU6s9TaYLfRx+DKFg1MzwbieyHXJJqf7ra3rT6E7F56zBMIgoCuGKAhEtxQIvEbK8J9XAMAqVGmp19hW9/VxSzCxLM7gIsGdg2WRAuHnylZrDJRVt6IdhuqQ0hqEE7654tKno7aYNiYzxxknrGalfPGeZqImkr4Rr9bqpYSijXzRqrqXNAcEwZnslzczCgEmDkU0ohlGzXFJy7GJC9x+fihY75eGScmZ5KwiD5BC9K8HzIWYS+jj8apEVK1VbrpSO7g4ZDBUhtWpRCj1dc9dgfBLAPQCWEdFHADwE4KNtGdUcgcr21leW46XmlWoVcinRptXLP1up1VGts01QyyXnhsHYfXQMb/7CI7OemTvh0mfEWdivrDEMtTrOawxj48qcJ7PzdEl5tNscGMpjSTaBxTLiSe9JbdTYLGut5zGIZkWN10a8AwxDREOxGeKb1hiZ2edBjj0VkGEoo+KlYTizvBX0eluA0ISU8QmaqVyt1c3EyOmgJYZRcWelgGKmHWAY0mCsXtQ16wbDNxSHiGLMXGXmLUS0A8A1AAjA9cw8u7V1ZxnLu1MoV+sYK1qrRrV69Q2rTcVQq7OZVeuEioXX/aPZgEULZwrfeeIYvv34MXzuzZtsrz92eAT/+/QQ9p+axEWre2ZlLIDli+1xaBh6RU5n8UFArBJPjZeRS8WwpjeN3UfHXPfvrWFEbXHtCgPDefT3ZczwWaNaB+Q8WKuz2XpVGYxiRTCM3kwjw4h1QMMwkwjleOLRCBKxCCYrdg0DAJItMoyRgmGWR9fhzPJWcDaZ0o1Es1X+ZLmKu3ccwf8+PYRtB4dRqdbxyK3XuJ7joNAZhq+GUfF2SQXJtWoHjo4WkU5E0dMVn5MM41H1gJn3MvOnmflTC91YAFZore6WGvepVKvg515ya/3YFY8iGqFZuwh/vn8IP917quF1NSk8e3p2exab5zZlj5Kyu6Rqtvj+CAnje2SkIH26MU/Re8LDJeUVJTUwnMd63WBoE75Rs1xSatIcKVRQqdUbsrwBUXxwtg3GsCwzs1jLCckkoiiUa1PWMPTV+aCLjuHFMJIO0VtvrVtoomF85ReH8IH7n8RTJydw/socKrX6tINC9O/g5w4rNHFJqSip2Y7yOjZaxOpFXSCiOWkwZqcM4hyEyvbWczH8KtUq5HwyQd2qYDYL8WwHRvIGjBo3FOtTK83DDoNRrzO27juFeptKXJgMQ3PppDWXVL3OMGpshoESiWS0fEW4pNb0diGXimO8ZDTcxMzsyTCcfSIAMVGcHC9jw5K0aaDsGgYjHlVhteL946PiOnEVvSOz33FPryOloL6rHlYLCIZRaoFhAO5uKTPL22kwYnYGozOMZox6/6lJLMsl8eB7NuNPrz4bAFpqy+sGPaM9iEvKWUdKIZeKyVyO2V0IHBstmaVAurvisxpWGyQ7bCkRvcvrTWb++AyOZ07BLdvb8rP7GAwfPcLstufwj2aTsVmrT6Mqq1aqddtNoeLUnx2xG4yHDwzjrV/8Jf7+dy7GDZevm/HxuLn7uhJRczWqJmw90SqTiJkaxhVn9aG7y7qJdXdJyaijznBlGLlkDJMVe5tWlbDXvyRjag/6ZFnVSoOoc6cWFq6id2x2GEalWkedGal4FKelS0pvF6v87kVHUb2US2Kd1/4VlHHQobK8nclwqbh9/wVbmXXv631gOI8NSzJirPGY7+eDoFStIUJA3afEuVtgig79Hne65tqJY6NF01WsGMZslTgPwjCiALIAch5/CxbLcikQubukcgES9wBvg6H8tuom0LebLdF7VDMYOso1xTDsfv19JycAAP/xvwfbwjLcNAw9SkqtDBXDAISQe3yshMlyFWt6u0xj41x1mXWkXG5+VcVVX/WqooM2DUMTrat1q4qxmiyOj4nz5eaSikeoITTXCz/Zc3LKkUN//c3H8dJ/2orHDo9gROuFoaDClBtE74AMQ19Nu+UoueVgmPs33HWLZoxhYMgyGFYY9fQZRlc8iq54834qboEpOvy8CO1AsVLDcL6C1YvEYranKw6jxi23GJ4qgjCM48z8obaPZA4iEYugL5M0XQ2ANRH5Gwx1MblfkEUPhpFLzaJLSorJwqdrTdJqYnZqGAdlS8uDg3n8eM9JvPLCFTM6Hjd9KB2PwqgxjFrd6uetGYxMIoa9J4QhW9ObNlnIeMmwCa9u7VkV9GAD9f4zymAsyZiCuM4Q9NIgXXE7w3CPkoqY2ylXlhuGJsv44y9tx9tfehbe+6rzPT/nheOjJZwcL+ONn9uGc5ZnG8qUZJJiQdIgek+BYbiF1rpleQMyca/qpWG4X+/jJQPD+Qr6pcGwEjWnzzCS8SgiRK4JmwpOo+pEJ8Lgj8lFieoAqRZXY0XD07DNJEINwwfrFneZGb+AaJ6Uikea1p8B9DatHiGeTRqzzEaUFDObK1CnD1Y9PzpStOkbBwYn8dw1PVjT24U7Hjw442MaKxqIR8l2g+oVa9W4bAwjETXFV6FhxOS+7OfQrT2rgluJcxFSm0Q2GUPcRcOoav0wVJivYhhuLikVgusXWqsmn289dnRKpUTylSou71+MK87uw5PHxtGbSdhcFZlETCTuVUTnwogm3AeKkpJjWt6ddNUwPBlGLOpZksOLYajEyf4+B8OYAQ0jGYsgl2rUrnR4NU9S8CuN3w6YORg9jQZjNhDEYFzT9lHMYfQvyZj+bECWrvARvAG9654Xw/Bu/TgbFHeyXDVFWGcGrnpeqdVtk8LBwTzOXZ7DH1+1AdsPjWCH7F0xU1DnVp/g1PkpGZrB0AyKbgDWysxXoNFN4NaeVcG88Uu6wShgw5K0OJ4eVitRrdfNsFonw1jkkYcBwDd5T628ByfKeGDfYNPPum9fw4qeFL74lsvw1795Ht7yon7b+6pNa9Gwi7mBo6TkZ9b0phs0DK8sb8BeoBGwt931mrRVpeANToYxzQla6VuZZLTpZK9+i7TLNQN0xiWlsryV6G0ajMIcMRjMfHo2BjJXsaEvgxPjJfMCHy8ZvoI3ILJqiax2rk7kXfIwgNkraKbnNjhXlrrbQUVKjZcMnJoo46ylGfzeprXo6YrjjgcPuO778HBhSt9hzFF4ELDOj2AY0iUVjTS8n0vG0N0VQ490ZznPu3JjeEVJAfaJ6PDpAtYtFhOVYhh2DYPNBkpKw5goVZFLxVw7MZouKR+DoU+kX9v+bNPPukFUVxXh2e/c/By8c/NzbO+rNq1FR7fHoAyjXBMtcle49LxXWd5ulZxTsShqdTbdeur3SEQjniL2M0N5EAHr+4ThVr91cZoMo2QIdpVJNL/XTIYxl1xSo0UQASt6LA0DmFsMAySwtt2DmYtQ/lPVSGe86F+pFhC9nrMJbz2i4MUwkrMjeusVcd1cUqrZjtIxDg6K73/20iwyyRjefMV6/PDXJ81VoEKlWse1n3gQmz/2AL712JGWYtTHS1XkHAZDTcaFStU0ZMoFBFjlQVb3irh0b9FbaUbeDENFpxm1Ok5OlEw/sZrsKzVrohIuKTLfV3qGm+Ct78PwcUmpSeoF63vx072nzLyGoBDF8ryvTzOs1qiZZUEAVRwwmIaRjEawPNdoMKxOe+6iN2BVu1Xfc0k24VlLamAoj1U9Xea28WgEiWhk+mG11TqS8SiyLuHUOvxcUp0wGEdHS1ieS5nX05w0GCzu+nvbPJY5CUWHlT91IiDDAFRxMo88jEoNsQjZBFxAXITlar0hcmmmoRc4dHZaK1fr6F+SQYR0gyEE77OXZgEA1z9vFZiBnc/a3VJjRUP0W6jU8K6vP47f/ewvzDpPfmjGMIoeGoYyAGt6xSpUuQmcBQiV4W7mklKThygHAzMSxXQnVa3Jvla3SoMAllvKTfAGYLIRv9BaJUbf+KJ+1OqMex472vTzOpgZeckwvJBNirDjsaJh04qSMhPbz8Cr0izLu5PIV2q2FbpXljcgDJLaHhDnOhYh9KQT3gxjuIB+6RZUSCej0w+rlQwj66NhKCbj5ZLKdEjDUAsZYI4aDIltRHRZ20YyR6EYhoqaGS9VfUubKzQLkfWqsz9bQprdJWVfsVWqooHTyp4u0yV1YHAS0Qhh3WJxA/c6SqUoqIn6766/CP/4u8/FE0dGcefPBwKNaaJoNLC3tI/orYTQNfImSsXd+6m79fNWcJZkUXWpVi8S3zXhkekd07qwqQnRTfD22ocb1CR10apuvGB9L76+/dnALK1cFbkmaY+8AcA6n0OTFZvBSMUjqDN8kwsrsvijconoLMMryxuwcmd0hpFORJH1KPzIzHhmcNIUvBVE3s30GYbQMJq7pPKmF8D9fMajEXTFo7OqYQxNlrEka11juZR0fc9Bg7EZwC+I6MCZ0KJVIZuMYUk2aTKMcZdJzQuqAKEbChWPMhXSGLU7tFZ3STXkYVRFFMm6xWk8KyfPg4N5rF9sZT3nPFw/erb2721aiyXZZODVjxvD0CvBKsFVj1BT7hdlMJRbyisPw9lACWg00ioSxemSMhyZ3rGobjCCMQy/Cdlyg8Tw+5vW4sBgHo8dDhZcYCaD+rikADHxdCXsGgbQ2HfbiXJVaBhKp9CTWr2yvAG9r7cqOCiu/7SM2nJipGBgvFQ1Gb5COjF9hqGub7+IxIJPWC0w+02UStWabTyRiLjeR+egwXgVgLMBvAzAawC8Wv4PBCJaS0RbiWgPET1JRDfJ1xcT0Y+I6Gn5v9dj+2tlP/H9RHRLC+OeNjYsSWNgqCB7YQR3SeWauKTyHiUHlF+0Wc/kmYBe9bNBwzAsg6EzjLOWWjdvIiZWV05j4Cydokp3+MHr3Kqbo1ipaZnebgzDcl10d8UbRe+ycNVEIo1R4s7+zKoa6Eq5ik7EvBiGNQ414XoxDPVZP1djwUzojOK6565ELhXDP3x/X6BEScWivEpZAJYxOZ2vOFxS7n23najIBlZKpzg5oRuMEnrTcdeQc6dBUte/16TtjJBScPZHmQrKmuhdMuqegQhFHw0DEPerV2BLO1Ay6g2/72zWkwpsMJj5kNtfC8eqAng3M58P4AoA7ySiCwDcAuAnzHwOgJ/I5zbIPhyfhjBaFwC4QW47K+jvy+CZ4TxKRh1Gzb8XhkK2iYBddHTbU8jNkktqpFmUVE1UhF27uAuDE2VMlqsYGCqY+oVCT1e8wbCpm0cxhYxHNzsnikYNRo1dNAyrD7XSWpyJe4DFMAB5EztuoLwHowO0mlTynB8dKWJJNmkTWwH7ZK9XqwV0DcPDJRULxjDMZDE5mb7/ty7Ao8+cxle2+d9qSv9ozjCi5vhtYbUBGUZFMgyzbI4WWitCat173atzqdyfBdku18sAmDkYDQbDnZG0AuWSUq5ILxeXVb7H+3zmZrH2GwCUKo09yOeUwSCih+T/CSIa1/4miGg86IGY+TgzPyYfTwDYA2A1gNcB+JL82JcAXO+y+eUA9jPzQWauAPiq3G5W0L8kg8GJspll6VepViHn0WcBEKtBd4YxSy6pQsWMhHLLw0jGIlgr9YpfHBhGpVZvMBjdXTFvhiHPUTYZxWQAtuRVNt5K3KtqGoZ13l5+wXL87asvMPsbq304md1kueYqeCvoNbyOjdmFxYQZJWUPq41GGzWM3oyHSyoSTMMoGPZgiDdsWoOXnrsUt//PXrNciReUwWumYejnwI1h+LVpVe6cTDKGXDJm1zDGS67uKMAq0KiMfl5qGG6FHwERlRghkVujw6vZlUKxUsM7tuzAP/1gL3YfHXPVf0zRW56nSQ8DVKhUQWRntE7MdonzUrWxpPycMhjMfJX8n2Pmbu0vx8zdftu7gYj6ATwPwCMAljPzcXmM4wCWuWyyGoAelH5Evua277cR0XYi2j442HrikxsULd51RPRZCMowmpXaLho1c8LWYSb8eWSIzxRGCxUsly6XBg3DEJEwSuB+YJ8oga67pADJMByi95jDJZVNBhMp3epIAfYoqUpVaRjWZdvTFccfXbXB5mrq7mp0E+TLVc8ictY4LYahIqQAIB5rjHCqai1aAV3D8Aur9Re99YUEEeH211+MWITwnm8+0dQ1ZbpQmvjc9ZBbNw3DuXhwQjEMAFjWnTQjo4DmDMNkMHL/oqifxTCcE/szQ3ms6U03RBH6MYxdR8fwvV0n8OmtB/Dqf3sIL/2nBxr6o+iiN+CdCFiQuSrNivrNZhOlmqzWnIrbz8mcMhheIKIriejTU9guC+BuADczc1CG4vaLud45zHwHM29i5k1Lly5tdXiuUJEaTyiDEVDDWJpLolytmyU4dAiferCs43ZgJG9ghby53VxSiahuMIThbWAYKTeXlIFELGJOQH6RKPp2QCN7i0dF1FNRy/R2TiJOeIneTRmG1tvgqOw3oI8BsDK9mUVJ+GjETfT2MhjSJeWTh1F0iZ5b2dOF97/6AjzyzGl8c8cRz23zAVwoXgzD0hh8NIyaZTCWd1stjI+NFnFyvGRzDepIOURvFf6bScZQrXNDBvwzQ/kGdxQgFhBNixVKFnbPO16ED19/EQ6fLmDbwWHbZxSD9guLbdY8SWEqxUKnmqmu3IVOhjGbJc5bMhhEdCkR/SMRDQD4MIC9LW4fhzAWW5j5W/Llk0S0Ur6/EkBjRx/BKPTEwTUAjrVy7OlAxYI/cWQUAAJHSW1whOTq8AqrNZOB2q5hVCyD4czDMERi0+JMAulEFEdHi+hNxxu6nLmtbERio2VQg67AVGkDJ8MAxA3iFVbrBiFENpYGaWYwlNYynK+gXK3bDEYsQiCy2IHSIWIuBsOtLAgQnGEUPDo0vmHTGizLJfHLAe/CC6Zg3kSk1d1V7qK3T5SUDIgAhMFQLqk7Hx4AEeH1z1/jul3KGVZbli4ps82udVxmxsBQHme5GoxY00zvQ8N5RCOEi1b34I2XrW3Yt1qlJ2NREfBnDgAAIABJREFUSy9sUr6nGSsFVK5V8Hv16GgRl3zwh1Mqq2MajJg7w5iNRk5BNIxziehviWgPgE9BuIaImTcz878FPRAJXvcFAHscPTTuB3CjfHwjgPtcNv8lgHOIaAMRJQC8UW43K0gnYljencTuY4JhBM3DMA3GYHCDkYy55xHMNEYKFfRlE4hFyFPDILLyLpzsAnBf2YyXDLM8B2C5evwuZrduewrpRFS6pESXu5hL6Q3buFJxlAx78qNeidYNQrw0Gmr1AMItFI9GzLLvqiCjPXFPaRgeUVJm4p4fw3DvrUBEWNmTwkmXLncKQcJqbQzDzSUViGGIzy7vTuHUeBljRQN3PXIY11280tS9nLBcXjrDiJlJcfqqe3CyjHylhv6+xn2p0iZe19PAcAFrersQj0YQi0aQikdsUXrqWk/FI74uqXyl1jSkFrBC551NyLxwbLSIap0DJ7PqULW43DSM2SpxHoRh7IUoQPgaZr5KGompjOxKAG8G8DIi2in/rgNwO4BXENHTAF4hn4OIVhHR9wCAmasA/gzADyDE8q8z85NTGMOUsb4vY9LpoKL32sVpRCPUUD4DkDV/XCawIF33/ukHe7Hz2dGAI29EsVJDyaijN5NAMhZxNAYSyV9K6F3bzGCkhFCs+9XHi/bQWOVy8AvX9NIwAOm3NmqmIfODOr4uRuZ9DIYSU486cjAUEtEIDJnprViCG8PwysMImrjntZAARAa1W0lxhSCidzIWaWj8pF4HAkZJRRXDSKJSq+OzPzuAyXIVb3vxWU2Pq/avNybKaFFwCnrzKifSCdG7xMt1dni4gPVasp+T4Za0nip+SbKTpeasFAD65AJBr5zQDOp7+hlmN3i5pGYz2zvIzPd6iBX9ViL6PkSEUsslz5n5oSbbNVTEZeZjAK7Tnn8PwPdaPe5MYUNfBo8+I9wBQUXveDSCtb1dDS6pSlWE57qJ3kDzhL+SUcOntx5A2ajj0rWLWvgGFvTWnUlH0TmrIqy4wRXDcAregJiYmYX7TF2040XDloug35TNupIp8dytz0iXbNNa1gTXZujWChD2ZZPm8ZtHSYlzrpL21iyyr27jUTIne4thWJfzkmwSvem454rULG9e9zcYXr1Wlncnsb2pS8pf9CYipBNRTJSqrhqGn2EvV2vmtaEE7i889Ax+46w+XLymx3M7XSPRGxMp46azgAGPHAzACgvOVxqjDJkZA8N5PG+ddV+kEzFbvw2LYUR9DcZo0bC5Jt2wRF5fIgPbPUJMh6pSPRU2YBmMRpcUIAzGyp7m450ugkRJ3cPMvw9gI4AHAPwlgOVE9O9E9Mq2jm4OQa12dEE3CDYsyTS4pKxYe/eJQeRvuK8W1GR/2kVIDwrLYMRF4xzNJeUMXV0rV9peLinAnu09XqraGIazTpMXxooGssmYq7spnYiiaIjig0EYRi5pH1e1VkfJqPu4aqJmb/BMIuoqviuDodxKOsP4kxdvwP1/dpVnRI0SvY2qv+jtZXSW51IYKRieOkOhUkMiFvF12anfxFkaBAiY6a0xDEAsgN72Em92AdgZRkGr1GwyDE1nGBjOIxYh18lajdmt695IwcBEqWpjGM48ILNrYwCX1Fih4qlJKagyHYNNXIU6lJ4SpNCjEyY78mIYs1DivJXEvTwzb2HmV0OIzjvhkmS3UKF6IwRlFwr9SzIYGM7bfK4FQ5Wp8GjM0kRIG5F9moenYTBUHanedAKJWMTBMOxd7S7f0If1fWlc4sJm3KjweNGuYQQt0DZe8i650pWwRG+/xlWA7pISx7Sih5qXeGAGnj41YVa+1ZGIRcxIHjcNI52IefrvAU309mEYRcPbJaVW9KdcemkDVmlzP6jP2F1SwRiGbrTVeM5ZlsXV5zWPSIzI3JJStWayiUwiZv4m+vUxOFEW+pqL4VPXk7qHdAyYbXWt3yGTsLdhLZmh2VEkYqL6rVdi6WjRwCKfiEiVdzI0GcxgFKShmArDUKVxnAuK2XRJTSmslplPM/PnmPllMz2guQrFMILqFwpnLcmgUKnhlLYCUasMr2iW7mYGYwYYxmmt17NozWlNEhVHJNIFq7rxs7/e7JqQ5SwlziyqoDqjpADvbFqFsaJ3yRXLJVVr0SUlxmX18/avsbTvxKTryjYRjZjMQjGNqEuZES84Q3O90CyUc5lc0XuVPM+X3asHODFVhsHMZuIeIAzGpvW9eM+1G5vmKiioa810ndk0DOt6b+beSbtEVSmoxEYnw9D3ra519X2FdtV4r5WrNRQqNX+GoQzGREANQx7LL3zZDaXq/NAwQgBYL5vpTIVhAKJ4n1qRFX2iWZqFoqrJfjoGQ9WRWiTr/ugx8G7Z1F4wdQs5MReNGqp1trukfFrVKjjFch0quSuoS8ppyFRp9mYMQE2iQ5PlBsEbEBO+ShxUDKNZb24nghYfLDZhCW7lOGzbGkEZhjQYLgyj2USmDKYy2vFoBN/80xf5Hk9BNGmq2aK5LA3DMgBDkxXPjHGTYbgk7w0MFUAErF1s/X7ZZMwWkVRyFLD0Kh5oFdF0j3pTyCVjSMQiGAzKMCpTd0kVK3ZjpzDnGcaZiK5EFCt7Up6CpBc2OBowAf5lk5tVwBydEQ3Dckk1aBgu9Zq8YK7kpWDtVt7DLL8QgGG4RUgBQutRiXuBNAxHAcddMtP3olXeoqzOPla5MIx4jMwJUwnX0Ujw28cqL+I9ITMzCkYTDaO7saS4jnzZ252lI+PCMKIRQjzaGGKto1y1T7atQjRpqpurbLuGMX2Gcfh0Aat6uhzVjO21qspVB8Pw6Lqn9AA/lxQRYWk2iaGAGoYydFPTMFQehv38z2aJ85BhtICbX36OZyavF1b1dCERi9hCa/0as6j6NMzcQPVP563VvLOMRFCczleQS4pWogmnS6rWWH7DC92OlY2aoHscYbWAv+g9Uap6sjeVh1E2gkVJZRIxRMjSMHYdGcP6vjR6mrgXdIPh5pLSRW/FEuItuKSUQN4s07tcrYPZ21XZm44jHiVvhlFp3m1PQRlx53FSsWhThlEJmGnvhZRs0qTYRDoRM42Weo2ZMTxZaWIwmjCM4bzZzlXBWWnAyaC9uu6pcuF+LilAuKVaZRhTipLycEmpEudzwiVFRBNwL8NBEM34plRPaj7i9y9b1/I2kQihvy9ttjgFoIl+HgxDdkVTNW906PHew/ky1iS83SxeGC1UsEgWyUvGIja9pGzYNYxmyMqJWRkKs46Um+jtk4g43oxhxKOy+GDNs3y4jkiEkNPKg+w6OuYq2tu+i8Yc3cpbJKJWvoqa9FvRMNRnm+Vh+LUEJSIsy3nnYuQrVbMkezOkXRgGICKHSk0YhmJHUzYYMoTbbE+cFOXm04moyTDGi1VUanVbkyAd6p5xq3B7aLiA37xwhe01PXGUiDSXlNIwYq45FHpgiB+WZpOBE/Gm45IqGe4uKWD26kkFCat1Fh2cdvHBMw0bZKSUwq6jY4hFyLWVJWC5VNyosn5xT9UtNVIwsFjeCMlY1Ja4F7ReE2BNzCbDcBQeBCydplmUVK3OmChXPQMKuhJR1FkwhiCGDLAKEI7kKzgyUsTFq73dUYC/SyoRc2EYPuGrOojIJpy7Qa06m7mVlncnbT0odASpfQRoorfjOMlYtGlCWSuLCTckY6JvuDMjXe+ZolbqXhpG2kPDGCsaOJ2vNGSHp5Pi2lHXteWSaq5hKNev1yJGx9JcAkOTQRP3VB7GzCXuAXPIYOggol4iupyIXqL+2jWwhYT+JRkcGs6jVmfU64xv7zyGl5y71PNibNZcfqRgmKvVqRuMirlST8ab52H4oUcrD2IVELS+V1SuIJu5pCZcXFk61AQ6UqgEXt3mkmJcSr94bkCDEYuQ2U1ORzxqhdVWpxAlBQjhuxnDKJq1oLwnfVG/yTus1ou16ljT24XedLzBFz4bDKNk1Boy0kXoqziuCk/1ckmZLiyHhnF4WKzw1ztaujqT85wMI5twd0mNteKSyiZxOl8OVB5kOgyjbHi7i+ecwSCiPwHwIER5jg/K/7e1Z1gLC2ctycCoMY6OFLH90AiOjZXwuktXeX4+m/TuiTGSr5jZ19MxGItlSYNE1D0PI+lCe92g98SwRG/7hOdXsdZZEt0JZTDGikZgQ9bdJUKTlcG40MdgKNfZykUpV0MQj5KZdFd1yfQOgng04tndDQiWqa0X/GvYvhxM0/qDy9dh619d3dB9MOXDMEwNowVmZdu/Er0d31MvWe5nMKIRMl2UOswcjCUOhuFIDLQqGVjVlN0E9JFCBdEI+ZYGAQQbqnOw+3FaLikZ9OEWwtybCc5ypoNWfvmbAFwG4BAzb4boZzEzDScWODYsEVnSzwzncd/Oo+iKR/Hy85d7fl5dpG49MU7nKzhblumYssHIG+bKSTAMF5dUwEmhR2uHamkY9ok/52MwTEPjwTAUBa9zC4ZMll7ffVQK3j6uBVX0cZVHaQWb6G1merc2ccajBKPJKtRPwwBELsZEqdowYTKz6CoYwCUVi0ZctaCUg206YS0mphYllYxHzcQ9PSNdb4qkoo28NAzAvcS5aiW8zhE6nXUkBlqRXpJhSJeUs8/IaEEk7QXJL9HLg/jBFL2n0Ga2ZDQ2T1LY0JfGkZGCb7Xh6aKVK77EzCUAIKIkM+8FcF57hrWwoFY9T52YwHd3HccrLljevHJqE5fUaKGCdYsziEVoStnelWodk+WqKeY5NQwzca+FiVnXMNKJaINv36urmkKzwoOAvelPUEOmKuk+cWTMV78AhMbQnYrbeoPrSOguKRlWOxWG0SxxT00iqWYaRs4927tcFUUjmxUe9ENSRjF5odXFROP+ZeJe2d48zM4wxMq+mdicTkYbJtyBoTyW5ZINUWJmlJ4ZzmrXYZRBKTi+92jRaBpVp0MZjCDlQcyw2ilM7KUmIddnL8uizkL4byda+eWPENEiAPcC+BER3YdZ7Ekxn7E0m0Q2GcOWRw5htGA0dUcBmujtMBjlqghJXJwRvSlOT4GCjhatLG8AvrWk/KA3KxLlPRpvskwy2pxheDRPUtBX3EENWS4Vw6mJMo6O+gveCh/7vUvwzs1nu75nE71dakkFQSxKTRP3gjAMr1yMIO4sP6QcbNOJaYfVaol7+sSuJ6oOTZaxOJNocJfpyLjoDoeGC2ajMx3qOHmNYehuHa+w77GCf1kQhVbKg1gMYyqid901Qgqwar3tPzXZ8n5bQeA8DGb+bfnwNiLaCqAHwPfbMqoFBiLChiUZ7Do6hkXpOF58TvO6O15VNM1Qv0wCfZkETgcsqaxD1aJSZbjFRGh1kHNSdj/0pK2ue+NF90inbDJulg13gx/DcMtI9kN3Km5OzkENxubz3LoDC8S1CCergVKrLqlI08Q9M0oq3kz0FpOTsy+GJSRPPbUqKMOYapSUyvNw1rzSk+uCVH11JuMBQsN46bmN95WzNI3eAEp/f6JUxXIt5nO0WHENfnCDcp8FMhjKcE0l07uJS0pVkz7QZoMx1VpSP2Pm+5m5/SrLAoEqEXLdxSt9V2hmX29HxVqlWSxOJ7A4k5iShqGXNgesCVitHs1M76Cun1QMJaOOcrXmma2d9ajXo+AWjqvDxjACh9Va+/ITvIMgruVhuJU3D7SPSHPR24qSaqZhKJeUnWEoYxNEw/BCUIYxZYMRj5iJe7ph012Wg5OVpvqF+XlNwylUqjg1UfZs6QrYGYY+6XpVUx5tgWFkkzEkYxFfl5TK5AemXt7cSz9KJ2JYvagLBwbnCMMgor91e52ZPzRzw1m4UCVCXndJc3cUYFXSdLZpHTFrQAmD8eQxe0v099+7G2ctzeCtV27w3Pdog8GQJSuqdXQlomY/72YuAR1WH4wqxktWn3AdQaKkYjL81g1urUT9oCK1+gMI3kEQj1GDhtFqWK1eXsQNQVxS3akYUvFIg0vKZBhTyPxXUGGvXpiuSyoZi6JaZ4wXDYeGIRgDM2NoooyzXSZ+HV3xqE3DOWSG1DbqT1mHhiHaD1vjb+aSCqphEBGW5pK+UUolQ2Tyq+9r1Oot5fKUjXpDe1YdZy3N4IBLd8+ZRCu/fF77qwF4FYD+NoxpQeJ3n78Gf/XKc3FZ/+JAn8+5dN1T7qTFGWEwhjUKzMy451dH8b1dx5vuV5UW6c1YUVKAFT3ipOx+0MuDjJfcCwj69fVW23lFpLh1hgs6rotmgF0AqlptHcxsahjxFl1SsUgkUKZ3s7agROSai1EMYGz8IBLrmiTutahvOaH876OFik3D0LsyDk2WzQqwXsgkY7by5qrplVtJF6dBKFVrtvG7uX+NWh0T5SoWdQUvA7Qkm/R1SSnBW4W0txpaW6p6u6QAoWMcGJxsiPiaSbSiYfyz/pyIPoZZ7Ks937GuL40/e9k5gT+fcylxflprfLQ4k8B4qWquUgYny5gsV/HMUPMoCadLSrmerEzYYCXEFcwmSiVDaBguxRmzyRgqVdFj223fY8VqUxagTy5BJysVOBBUv/BDIhoBs3BHmQyjRZdUItrcYBQNIcj6sbvlucZcDL0+01ShRGkvqGq90xG9AdHL5RItmkuxjcGJMspV77IgCqKUiDVOZTxXuJRFSchwaXV+yg7h2K1fy3gLSXsKS3NJPHu6+b2nFgR9mQSOjBRRMuoIKJMAUGG13uf+7GVZFCo1nBgvuVYrmAlMp1ptGkDzNlshpgy3kgWjecsl5ewlrLr6DU2WPbv1AWJ11xWPmjev8omqiSJoCXEFpTuMFQTDcJv4/QoQjhYqns2TAPuKO+hk9ZxlWfT3pbF5o7eQ3QriMdWT+/+3d+ZRbtX3of98tc+MxjMe2+MFL4ONDQGCDRizhBAIBbJASEpyEkJaCnmlSWmSNs1CXtOU9L2816ZJ3yPNfnKynebAa/LCC9loCCUlJEDAhMXs4AW8Ys/YY1uaGWmk3/vj3t/VlXQl3attJOv3OUdnNFdXur+fdO/9/r67qqv4INhRUlVNUv7Kk4/Oixf1V9HvhQbDaqNh22ziPcaGnd72ze7IdInT2z4/tGmpltO71Iex9/A0IlY0YsX9a2gY7nMzSOFBjT8Nw7rG6tYwsuW15dycYEdKtdKPESTT+wkRedx+PAk8C3yhZSPrcZIeJqmJdIakXX9/ZMC6OLTj212rqlIstlKKR1465ETaQOHid9faCZKYpbvr7To0hVLeyXe1eifvOjRVdUUUDokzTr83q9HBBL/66EWsWzzoa/9aaFtzZjZfV/FB/RnVS4PkfWkIOtu7qItjjR4rfnD8WRXG2IywWk2RScp+rs9hP1FS09m8E3ywb3Kahcl4xda07hLmpRpG4dws3Lx1NGIQ39eiZIzxVKZqUEPKMUlZ8wsuMHJl5VzcrBltfaRUkF/+cuAK+3EpsEwp9S8tGZXBqrZaoikcSmcd34NepehcjK2u8unuUupubv/9LjbvOMifva6Qa1AuMHKBErO0gNh50LIje0U66aivlEdJ6nxesXNiqixDtxTtx/Cbh9FsYrb5KZPLu0qD1JHpXbX44Kyv0h6L58VJZ3JFAlivkOspd6/RN/RKfoxMLo9I8PwTjVvYFzm9ba1IZ2v7ERhQ0Kr2Hp72DLZwjhUPF5UGcWsYiWiIkBRrGJNTBU3eL4sG4yhF1VB37WdaYJvcgkZK1TJJLUrGGUxEeKGFGkaQ5che4CosR3cELAeciZJqDV7lNCZSGcf3oE86ne29/UCK44b72HVoymlV6WYyneV//OxpNqwY5p0bVzjbnV7O9k0iM5sPdFPWAuJlu7yzVx5GtRLnew9Pk8nlWekR4eKmPxrmEFli4fpviI3gtFjN5Z1VZODEPR9Obz8mKXfnvUH7+/cTYVULZ/GQzYHH6lo3sPJTLsPz86PlZigoaBj6vF04WMuHYe0/lckxmIiy7/B0xQx9KDZhTdt+Io2IlEXxHfLZPMmNUx7kSOX8jXKTVLDkvenZfNUqACJiOb5faV2kVBCB8SNgEtgM+OsWYqgbTx9GuiAw9F9tktp2IMWrls5jNp/3dHx/7hfPMpHK8O3rNhU5VWOOhuFegfkXGIlomHgkxE57deipYZTU83HjhESO1AilnGsNI+ISGPXmYURCNTO9q0VIaUYHC7kYJ4wmnfdaDt76vx+tYXzuF8/SH4uQjEf4wMUnFOXq1FsWBIo7xbkF24B9fuwYt1qsjtRY2Q+UtHXdd3iajWPzK+/vygz36jEzGC8OMNEdKQP5MOzIrmqNlEqjpIJoGPm8IjObr2qSAst3d+9zrSvxF0RgLFdKvaFlIzEUoaOk3F33JtIZVtuOLZ2pPZHKkM8rto+nufDEUY5MZ4v8GWB1nPvXB3dw7bljZWGm7jwM0BdUsJvCvL4oL2uTlKcPw9rm3VbTGqtXDL0bvaqs1+HaKMUaRp2Z3iEpqttVylQmVzNCCNzZ3oVIKb8O82qcuHiQkYEYP39iLzmlSGdyvHbtQs5evQCwzo1YnSG1UNz4x8uH8dJEmpH+WE1Tn7vcx3Q2x8G0d/6P8/nxsJNUV6phACweSjihuQCT6QwilRNJvVjkaBjVBEYhSkqPxS+lfTwqsWZRkh9s3lmxTE+jBDnjfysir276CAyeJONRcnlVpLa6q8xaFUejTKQy7J6cIjObZ2zBgNWsqcSHccvdz7FgIMaHL11XdpxEtNiHkSmx8fphqC/qaDreUVJawyiP3nppIk0kJDU7xWkNo16Ha6MUnN6KXN6y5dfj9NYhuV6kM/58GKMuk5QmNZNryOEN8OrlQzzyt5fwxKcv47YbzgGKC2DqOkz14j6vvHwY6Uyupv/Ceq9uopRzEvgqNSOD4sRRLw1j7WiS512O4kN2xQK/yatQ0DCqRUppLaeeKCmtjdRazK1pcYmQIL/++cAjIvKsHSn1hIg83pJRGQrlQeybrK4y61bXdXmQ7bYJamxhP2MLBxhPZRyHeS6veHDrBJeessRzxeH4MFxO76A3BXdIrLdJqjwSRbNjPM1x8/tqriq1qabepLFGiUUKTu9sXtXl+K0VVjudzdNXpY6UJhm3zEXuXAy/DnO/eEW2BQ25LqVIw/DwYUBt/wUUFg+pzCx77e+gqobhqobrdX6vHR3kwNEZDtqLniBlQQrHCNMXDVctD+I4veuIkqrWbc/NmlEdWtsaP0aQJckbWzICgyfzXCXORwcLJT2GBwoX1IKBGOOpGbYdsFYTqxcmnd4S2w+kOG35ME/vOcyRmVnOPt47w9zLhxF0Fe82QyU98imq5WG8NJGuGSEFBZt3J5ikcnkV2BylP6Na8cEgZqUlQwn2HCoIjFRJyfBG0c70IyUCoxENz32zc4/V7bfxpWHYGslUJudoQF5Je4X9rSZJSlkae+k5dMJiu9Lr/qOcNTBilzb3HyEFlsN54WCsuoaRsSIQ9TUSpCfGtE8NY+VIP9GwtCwXI8ivvxd4DXANcK3rYWgByZLIIu2Ic2sY8/stDWPbgTR90TCL58WdmlU6tPbBbRMAbKogMAqRMW6TVLCbgjZDDSYiFbrVhYhHQp4CY8d4uqb/AubeJKWdvdnZPNlcvi4NI1ozcc9flBRYbVZ3HioEN0yVlAxvFK8S+0EDIkpxCwy3NhRy1RELYpJKzcw6RRgXV9EwknGrRpp2kpfmGa21V+XP77NuspPpTGANQ4+9Wj2pKdvkqG/6Qfp6a9N0Lad3NBxi1YKBlpU5D/Lr/wi4EpiluK6UoQU4Kzz7gtU+gvmuyI0FyRgTqSzbDhxlbOEAIuLcfLWZ6nfbxlk50s/SCp3knAiYnCtxL6DZR5uhqjnZkvFIWTHFyXSWyalsd2gYrqS2XF4FjpCC6ol7ObuWkl+z0or5/bw8UXDUpprg9HYTj4SIhKTI79SohlGch+Hd6MiPwCjkYeTYOzlNXzRctVKAFqTa5FR6Di0b6qM/Fub5V44Alg8jSISUZlEyXtUklcpYWqC+6QcySc36M0mB5cdolYZhoqQ6lIIN2bpgnRpQA8U+jIPpDFsPpDhlmVXMPxENs2wowY7xFPm84nfbJri4SjvYWImGMZMNVksKChpGpRar4N11r9BWs3pILbijpObIh+GYpBTZnCJch0kqErbCat2RbxqnF0YADcMp+JiIWtpJA70wShERK7S7xOndaK0qTWkJk4FYmP1Ub83q7OuqQLv38DSL58Wr5oboa0mbi0pvuqGQcMJo0lmV1+PDAMvxvXnHwYqvT2WsnuuhkBCLhOryYfgJKz9vzULikbDnedYoJkqqQ9Ghk8/utU5iLTBGigRGnFxesWM87ZiiwOq9sW08xQv7j3Iwna3ovwAr0ifiapxUj9lBJ+tVW+UlPQTGDp8htQDLhhIsGIgRrWNl3wzcpUFy+Xxd49DZ4l7Z3mmnF4a/G/IKWyvbaWsZ6cxsU30YUK4VZnKNaRjhkDjfW6mGoQVRrUq1gJ08COmZHPsOT1c1R0FBOOlryOv8PmE0yfP7jpLLK6smWkAfBlja0US6cnmQVGbWEXZ9NUrJl6IXdH40jGvPG+MLV5/edGEBwaOkNpsoqfawIBln09gIP33C6oJ70Ck86DJJuYSHuz3lmB1aq/0XZx+/oOqxrDatebsSq6orrNb914tkvLz6rk7a82OSuuacVfzHX1/YkovAD1HnZm/lYQQNqYVCKRGv0Nppu2Wnn8Q9sExSUMiwT8/kmholBeX1zBpN3IOCDb50ntqRXamAoBsRsSOfcuw7PFPV4W19tnWTHrf9C1610taODrL38DS77Zpo9WgYujzIeIXGZu7EzEQ0FChxTwsXv+dHqzBRUh3MFeuX8rc/epJn9x7hYNpqOuO+mbvNU7pFI1hNgw6ms9z11D6WzEuwYqR6qeN4NOyUH7f+DxpW68ckFS7Lgn1pPM3CZNy5oKsRDYcY6p8b/4U+PhRqSdXn9NaOcwUlC1jd3yGISQqsGl5KKWv12kSnN1iOb3dY7UzAsjFexKMRm+4YAAAczElEQVRWE6XSHAdHw/AhMKz9rS6OtepIQcEkpW/kXk2ItOP74R3WIqseH8ZSexx7Jr21nqlMzun/3WdXBvZLEB9GK/H96yuldrgfwHLgY60bmuGNr15KSODHj+3mYCpTVgytooZhP//18/vZdPxIzVW5pWHkHLNU0FWkFhRVnd6JaFmm946JFCtrCLNOQZsxrNIg+cCFB8GlpXhoGE7zJJ8CY7g/SjIe4eWJNDOzefKqscKDXpQ2vmqGhhGPhBxtovRYUKiRVouBeMRJWK1pkrK/Fx044qlh2KG1D223fBB1CYxhaxx7J73716dciZmJaDighqFNUnO3aIKA/TBEZIOIfFZEtgP/HXgmwHu/KSKviMgW17b1InK/bd76sYjMq/De7fY+j4rIw0HG3M0sTMY5b81CfvL4bibSmSL/BRT8GYOJSNFr2p+hFJy9unaHv5htkqpXwyg4vav5MMJlJqmXJ6ZYtaC2w7sTiLrCamdzDWoYHjZup2OezxWkiFihtQfTrtLmTRYYiWi5SarBKLVENOTpONfnsN9aWH3RMFvt5LRaAiNZYpLy0jCWz+8nHgnx8HZLwxgK0G1PoyMR90xOe74+lSnkytRqh1uKk4cxR0Efmpq/joisE5FPicjTwBeBlwFRSl0UsLz5t4HSKKtvADcppV4N3A58tMr7L1JKbVBKbQxwzK7nivVL2T6eZvP2g2WrHi0kVtshtZoVI/3of6s5vDXxSIiZbL7uFpx+wmrdBeDAirjZPVm7rHmn0IywWi1kvHIxCtVm/ZuVls/vZ+fBqUI/7yZGSYF1oz1clofR2A0rEQ17mt1uuGA1t7xrg+/PGYiH2WXXf1oyVKscuh1Wm66sYYRDVqXX5+xcjHo0jPn9UeKRUEWBkZqZdcYS1Ok95TPTu9X4EefPABcDVyilzreFRLBC7oBS6l5gomTzicC99vO7sMqnG1xcdsoSomHhyMxsmYaRiIZJxiOMLRwo275sqI8FAzHW2MUKqxGPWIlNM3W24Fw+v4/3X7iGS0+pHL6bTESYyuacpjeW7d1fhFQnEHVFOGXz9YXVuv0gpRSipPzfEFaM9PHyRLoppc29sHwYzcvDgMoCY/WiJK9du8j357gFq28No0IehkabpaA+p7eIVROtooaRLSRmJqLV+6eXovedqzwkjZ8lyVXAu4B7RORO4DagWaEqW4C3YCUFvgNYUWE/BfxCRBTwNaXU1yt9oIjcANwAsHLlyiYNc+4Y7o9xwdpF3P3MK05JczefeduprB0t7yp3+fqlREP+ehcUfBj1nZShkPDxN5xUdR93baKhvigvBYiQ6gSioZKw2gZMUl4axnTAPAywNIxUJudUWm220zsZjzCdLWS2NxpWC/Cec5pzTbr9IJX6T2h0k6TxCnkYGu34hmDd9twsGUp4+jAys3myOeX8vn2xYD4MnR8VpCBiK6j56yulbldKvRM4CfgV8FfAYhH5iohc2uDxrwduFJHNwCBQKa/+NUqpM7AitW4UkQuqjPfrSqmNSqmNixb5X7F0MpevXwrgKTCu3HAcJy8rd/184o2v4iOXnejr82NlJqnmr2JK60k5SXtdomGE7ByCbC5vJ+4Fv3D1TchdNFDjOL0DmBxW2JFSz+6zMpSbrWG4+10369x42+nLedvpyxseWyGqKlZTiOkw3IkaGsYJ9sJrMB6pK6gBLD/G7kPlv++UE9RgjTsRCe7D8PK9tJsgUVIppdT3lFKXY0VIPQrc1MjBlVLPKKUuVUqdCdwKvFhhv93231ewfB2bGjlut3HJyUs4ackgG1YOt+TzdR6GTg5qRb2mUoGxYzxNfyzsK+6+U9ClPXJ5VVejog0rhwmHhAe3jZe9FjRKCgrJe8/u1QKjyRqGqwCmNqPNtUlEo4VjLe1CMxCPON9xRQ3DNkkN1eG/0CwZsvqt50saZemOf47TOxZUYJSXZZ8L6vr1lVITSqmvKaVe38jBRWTU/hsCPgl81WOfAREZ1M+x+olvKd3vWCYZj3DnX17A69a1RmOKR+w8jFx9Tm8/DMZ1uXatYaRYOdI/Z4l49WAJDMVsLl+XhpGMRzht+RD3v1guMKYyOUIS7IasczEcgeERrtoIgy4zYqaF2mc9aOFYK2lP4zZhVZrDKrvSq5cm75dlQwlm84oDqeKco9IFgaVhBMvDaHbYdD207dcXkVuB+4ETRWSniLwXuFpEnsNyrO8GvmXvu0xEfma/dTFwn4g8BvwO+KlS6s52jbsXiEdtH4auV9NiDWM6m+ORlw6xdnG576WTiYYtTWw2r+ouUXLO6gU8vnOyrExK2q42G0SADiaiDPdHecEuNNd0k1SiIDC0SWquqgWXolfqtRzezv6uCLJK53ckHGLt6KCvelaVWKJDa0vMUmlHw7CjpGJWprdSlasXu7FMUnMvMJqrw1ZBKXV1hZdu8dh3N/Am+/lWYH0Lh9bzOCaplvow7KY3M7Pc8dhuJlIZrj6rUoxDZxKzfRj1lgYBOHf1Ar7yqxfZvOMgF7g0xnobIK2Y388TuyaB5pukdMXko9MFDaNTBIYOIa6V5a3RN+pISKr6J25514aG+qLrzpF7JqdZ7zq9SyPZEpEwubxVyFI356qGZZKa++9+7kdgmHNiZQKjFSapQrn2b/1mOyctGeTcNdVrXHUasUiooUxvgDNXzScSEu7fWmyWmgrQC8ONNktB65zeR2ZmnZDruaoWXIrWMGrlYDj7x/2Vx1+7eLAsTD0IWmCURkppDUMLOr040CU/ajGdzXnmj7Qb32e9iGwUkdtF5BFTfPDYwvFh1Jnp7Qd9wd799Cs8vecw171mrKv8F1BwetdbSwos08j6FcM8UCIw3IXpgqAd37FwqKGVsRfuJkqOhtHkY9SLvuFW6+XtRpukWu04HhmIEQuXJ++Vahj65j/ts+vetEcv8rkgiA77PaxM7CcA/94aQ8fTaC0pP+gL9s4n9zIyEOPKDcc1/RitJhoOkZlVdmmQ+r+jc1aP8NX/3MrRmVlnFT+Vrc+pqUNrm+3whuKeLJ1mkjphNMm8RISTlvjzg+nzr9VOexGx2ufWEBh6ceDX8T2TzZHwUfq91QT59vYrpe5QSm0rKUJo6HLikTDZnHISiVqhYcQjIcdRfM3ZKztitRSUaCRkV6utr0Wr5tzVC8nllVO3CIK1Z3Wz3C5z3uykPbBubiKWhtFK/1Y9nLJsiMdvvqxiJ8lSBkpW9q1k6VCCvaUCQ5dv0XkYTptWfxrGVDbXEddMkF//70TkGyJytYj8oX60bGSGtqFXjbo4YCs0DBFhIB4hEhLec86qpn9+O4iFhexs/bWkNGeumk80XOzHsExSwW/6unR9K0IuRcSpJ9VpGkZQ2qVhgCUwdpf6MLKVNAz/PoxOcHoHOUOvw8r2jlIwSSngh80elKG9xB2BkbU68LXITr1qpJ+TlszzHQrZaeiM+Gyd1Wo1fbEwG1YM88DWgoYxVWdP7oKG0ZrV56Bd4rzTwmqDojWwdmgYS4b62Hd4D3lX34/0THGejRYYfjWMTkncCyIw1ttVZQ3HGNoEdXhqtqUrsO+/7zzmuBROQ0TDIY5Oz9oaRmPf0zmrF/DlX73Ikeksg4loUWG6ICSiYRYm4y1L6tJ9vTstSioojtO7DQJv2XCCbE4xnso4DZNK82zidWkYc//dB/n2HhCRk1s2EsOcoW8CR6azLRUYsUioZdpLO4iGQ2RyyinG1wjnrbH8GD95fA9gm6TqvOlvOn4+61qUBKmbKHVapndQnLDadmgYTue9glkqXaJBBjFJKaWY6cIoqfOBPxGRrcAMVsVapZQ6rSUjM7QNbWY4PD3btSvIdhBz1ZKqN3FPc87qEc5YOcznf/Ecl5+2lKk6w2oBvnzNmQ2NpRrJRJTJqaxTNqbbTVLt0DDcjZROs+sslgY1aH+EnygpbQ7sNh9GafMjwzGC24fRrTeEdhANCxm7NEijmpKI8KkrTuGtX/oNt/zyeWbzqumJd81gMB5h18G0U5iyezWM9vkwCq1aC5FSloZRuN1qbdKPD6NTuu1BMIFxbYXtf9+MgRjmjrgrSioZb1u1mK4jGg45F3ijJimADSuG+cPTj+Nbv90OFEpfdxKOSarLNYxkG6OkRvqt5L3dRSapEg0j4t8kVejnPfcCI8i3l3I9cli9KcZaMCZDm9FmqMNT2ZbkYBwrxCIhp69BI2G1bj72hpOcMOZO1DC007vTMr2DohMb22HWCYWExUPxIg0jlckVtdCtS8PogGvT95JGKfV59/8i8jngjqaPyNB29Koxlcl17Q2hHTRbwwCrPPf7L1zDP9/1XGcKjHiEVCbHVDbX0pDrVlPQMNrzHS8d6ivK9p7KzLLUFU6uNR0/Pgxdb6oTNIxGdOB+YHWzBmKYO9xqunF6VyYWCTk9yRspDVLKDRdYl9EFAXpatwtdT+pgKtPViwl3L+12sHQowSMvHXT+T2dyReVbRMTu6x3EJDX3379vgSEiT2Al6gGEgUUY/8UxgftENCapyrh7YDTLJAXWyvGDF69t2uc1E70yP3A009XnxkAswsJk3El0bDVLhhLsm5xxkve8Sr8kov667mkzaLc5vS93PZ8F9imlZivtbOge3FpFN68iW427GmwzNYxORjdRmkjNdPW5EQoJ9338orbNYdlQH5lc3kneS2dmy+p99UXDjjDI5vK8/Su/5epNK3nXppVF+2mTVCeUNw/iwzCFBo9R3JEvnXBSdiru76lZPoxORzdRGk9lujZCStNOH8CJdhXdu57axzvPWsF0Nl+WmJmIhh2f2I7xNI/tnOTJ3VsYWzjAOasLvWJmOsjpbfphGEp8GHN/UnYq7tVpM01SnYw2SY0fzZhzIwBnHz/C6SuH+eJ/PM/kVBYoj4KzTFKWf2LbgRRg5Yvc+L1H2HWoEJKr96k3sbOZBDkDvofVc/sq4AosE9UVrRiUob0UmaTMTaEibpNUo5ne3cKgq693rANs6N2CiPDhS9axe3Kab963DShvoet2em+1+7J/+7qzyMzmueG7DzvmqkJY7dx//6YfhqHYJGUERkXcAqPZ3e06FXcip1lMBOP8ExayaWyEb9y3FSjXMPpcTu9tB1IsGIhx+sr53HL1Bp7ac5iv3fsi0L0Cw/TDOEYJh8SJADJhtZVxR0n1ioahnd5gFhNBERH+6pJ1jkmpXMMo+DC27k+xepHVS/z1Jy1m/fJhfvuC1S9luoNqSQUZwXXABqyaUldQMEsZjgG0fd6sIivj/m6iPeLDcEf2GIERnHPXLOC8NZYDu5qGsfVAiuMXDjivnTU2n0d3HmJmNtdRtaSCnAHrlVIblVLXKqWusx/Xt2xkhraio6PMTaEysSIfRm98T+GQFNqbmnOjLj72hpNYMi9RJBCg4PQ+PJ3lwNEZVi9KOq9tHBshM5vniZ2TTGfzxMIhpxnTXGL6YRiAws3A3BQqU+TD6ICLt11os5TRPutjw4phHvivF7NipDhpUDu9t+63IqRWuwTKxlXzAXho+0Gms7mOSZoMMorzgUdF5FkTVnvs4QiMDnCsdSrRSO9FSUHB8d3NiXudSJ/tw9h2wIqQ0j4MgAXJOGsWDfDw9omO6bYHph+GwUavHuPmplCRVpUG6XSSdvKeCYhoLro0yNb9KcIhYeVIscnqrLERfr5lL69bt6gjHN5gMr0NNvpm0CmqbycSL8r07p3vaTBuTFKtoC8WJq/gmb1HWDG/r+z73Tg2wm0PvcyWXZMd4fAGHyYpEbnP/ntERA67HkdE5HDrh2hoB8aHUZteTNwDl0nKnBtNRV9rT+0+XOYQBytSCqwIqnr7vTebmmeAUup8+++gUmqe6zGolJrX+iEa2oHWLMxNoTK9mLgHBae3WUw0Fy0Edh2aKoqQ0qwc6Wd0MA50RkgtBKslVRYhJSIXNnU0hjlDOzSNnboyRsMwAqOZuIWAl4YhIpw1NgJ0jqk4yCj+TUQ+LhZ9IvIvwP9s1cAM7cXxYZibQkViRRpG7wiMQRNW2xLcZiZ3hJSbjbZZqlOipIKcAWcDK4DfAg8Bu4HXtGJQhvZjTFK1iUZ6rzQItL+9aa/gjnxa42GSAhwNoxsFRhaYAvqABLBNKVW7Ia2hKyg4vTvjxOxEYj3qw9A9McxiorloITAQCzu+ilJOWjLIYDxSVARyLgkyioeAHwFnAQuAr4nI25VSb2/JyAxtJWaipGrSs4l7xundErTAOH7RACLe51MkHOLb129i8TxvgdJugpwB71VKfUoplVVK7VVKXYklQHwhIt8UkVdEZItr23oRud/OGv+xiHhGXYnIG+wM8xdE5KYAYzb4RGsWZhVZmSINowfzMIzAaC66IdLqhd7mKM2Zq+a3rRd5LXyfAUqph0VkvohsEpELROQCIEgy37cpzxb/BnCTUurVwO3AR0vfJCJh4EvAG4GTgatNTavmY/IwalMUJdVDTm+nllQPmeHagaNheERIdSpBwmr/C3Av8O/Ap+2/N/t9v1LqXmCiZPOJ9mcC3IXVza+UTcALSqmtSqkMcBtwpd/jGvxRyPQ2PoxKhEOCtkT1Sk9vgLWjSc5YOcwpy4bmeijHFEvmJTht+RCvO3HRXA/FN0GWDB/C8l/sUEpdBJwO7G/w+FuAt9jP34EVhVXKccDLrv932ts8EZEbRORhEXl4//5Gh9c7nLZiiE1jIx3RN7iT0Sa7XhIYw/0xfvjnr2Hlgs4wixwr9MXC3PEX53PGyvlzPRTfBBEY00qpaQARiSulnsHSEBrheuBGEdkMDAIZj328rkxV6QOVUl+3+3ZsXLSoeyT3XHPRiaP82/vO7Slnbj1os5T5ngy9SJAoqZ0iMgz8P+AuETmIlYtRN7bQuRRARNYBb/Y6LsWax/JGj2sw1EssHCISkopRLQbDsUyQarVvs5/eLCL3APOAOxs5uIiMKqVeEZEQ8Engqx67PQSsFZHjgV3Au4B3N3Jcg6FeouFQT5U2Nxjc1BQYInJHpZeAP6Xgg6j1ObcCFwILRWQn8HdAUkRutHf5IfAte99lwDeUUm9SSs2KyF9gOdnDwDeVUk/6OabB0GyiEemp0uYGgxs/Gsa5WE7nW4EH8fYp1EQpdXWFl27x2Hc38CbX/z8DflbPcQ2GZhIzGoahh/EjMJYAlwBXY5mCfgrcalb5hl4kavswDIZexE8/jJxS6k6l1LXAOcALwK9E5AMtH53B0GHEIiETIWXoWXw5vUUkjhXBdDUwBnwBy+dgMPQUloZhfBiG3sSP0/s7wKnAz4FPK6W21HiLwXDMEg2L8WEYehY/GsYfASlgHfBBV/y5AMq0aTX0ErFI2PgwDD1LTYGhlDL6t8FgEwubsFpD79IZXTkMhi7h2vPGOJjOzvUwDIY5wQgMgyEAr11r6pMZehejWxsMBoPBF0ZgGAwGg8EXRmAYDAaDwRdGYBgMBoPBF0ZgGAwGg8EXRmAYDAaDwRdGYBgMBoPBF0ZgGAwGg8EXopSa6zG0DBHZD+yo8+0LgQNNHE430Itzht6cdy/OGXpz3kHnvEop5ZmhekwLjEYQkYeVUhvnehztpBfnDL05716cM/TmvJs5Z2OSMhgMBoMvjMAwGAwGgy+MwKjM1+d6AHNAL84ZenPevThn6M15N23OxodhMBgMBl8YDcNgMBgMvjACw2AwGAy+6BqBISLfFJFXRGSLa9t6EblfRJ4QkR+LyDx7+yUistnevllEXu96z5n29hdE5AvialJecrzPiMjLInK0ZPuHReQpEXlcRO4WkVUV3v8++ziPish9InKy67U7ReSQiPykWXN2vb5SRI6KyEe6cc71zFtETrNfe9J+PdFt8w54fkdF5Dv29qdF5BOu93TTnFeIyD32HJ4UkQ/Z20dE5C4Red7+O9/1nk/Yc3tWRC7rxnl3NUqprngAFwBnAFtc2x4CXmc/vx74b/bz04Fl9vNTgV2u9/wOOBcQ4OfAGysc7xxgKXC0ZPtFQL/9/P3A/6nw/nmu528B7nT9fzFwBfCTZs3Z9fr/Bb4PfKQb51zHbx0BHgfW2/8vAMLdNu+Ac343cJv9vB/YDox14ZyXAmfYzweB54CTgc8CN9nbbwL+0X5+MvAYEAeOB17sxt+6mx9do2Eope4FJko2nwjcaz+/C7jK3vf3Sqnd9vYngYSIxEVkKdYPfr+yfuHvAm+tcLwHlFJ7PLbfo5RK2/8+ACyv8P7Drn8HAOV67W7gSMXJFvbzPWcAEXkrsBVrznpbV83Z3jfIvC8FHldKPWa/d1wpleu2eQecswIGRCQC9AEZ4HAXznmPUuoR+/kR4GngOOBK4Dv2bt9xzeFKLEE5o5TaBrwAbOq2eXczXSMwKrAFS8oDvANY4bHPVcDvlVIzWCfjTtdrO+1t9fJerNWMJyJyo4i8iLVi+mADx3HjOWcRGQA+Dny6ZP9jYc5Q+bdeBygR+XcReUREPmZvPxbmXWnOPwBSwB7gJeBzSqkJunjOIjKGZRl4EFisb+r231F7t+OAl11v0/Pr2nl3G90uMK4HbhSRzVgqbcb9ooicAvwj8Gd6k8dn1BVXLCLvATYC/1RpH6XUl5RSa7Bu5J+s5zgeVJrzp4H/pZQ6WrL/sTBnqDzvCHA+cI39920icjHHxrwrzXkTkAOWYZlm/lpEVtOlcxaRJJYp9S9LVvBlu3oNo8r2esYyl+d4xxOZ6wE0glLqGSyTBCKyDnizfk1ElgO3A3+slHrR3ryTYlVzObBbRMLAZnvbHUqpT1U7roj8AfA3WPblGXvbZ/TxlVIbSt5yG/CVwBP0oMqczwbeLiKfBYaBvIhMY12IXT1n+/MrzXsn8J9KqQP2az/D8gX8K10+7ypzfjeW7TwLvCIiv8G6yf2aLpuziESxztHvKaV+aG/eJyJLlVJ7bHPTK/b2nRRbEZYDuzkGruuuoZkOkVY/gDGKnYKj9t8Qlt3yevv/YSzn2FUen/EQluNLO8feVOOYpc6x07GcbWtrvG+t6/kVwMMlr1+IPwewrzmXvOdmip3eXTXngL/1fOARLOdvBPgl8OZunHeAOX8c+JY9rwHgKeC0bpuzPcbvAv+7ZPs/Uez0/qz9/BSKnd5bKTi9u2be3fyY8wH4HijcimWzzWKtKN4LfAgrsuI54B8oZK5/EsvG+6jroS++jVi24ReBL+r3eBzvs/Zx8vbfm+3tvwT2uT73jgrvvwXL+fwocA9wiuu1XwP7gSn7sy9rdM4l77uZYoHRNXOuZ97Ae+zjbsG+uXTbvIPMGUhiRcI9iSUsPtqlcz4fy3T0uOt4b8KKdLsbeN7+O+J6z9/Yc3sWVyRUN827mx+mNIjBYDAYfNHtTm+DwWAwtAkjMAwGg8HgCyMwDAaDweALIzAMBoPB4AsjMAwGg8HgCyMwDIYmISLDIvLn9vNlIvKDuR6TwdBMTFitwdAk7HpIP1FKnTrHQzEYWkJXlwYxGDqMfwDWiMijWElnr1JKnSoif4JVPTWMVW7/80AM+CNgBisreUJE1gBfAhYBaeBPlVUexGDoCIxJymBoHjcBLyqr5tBHS147FasG1CbgM0BaKXU6cD/wx/Y+Xwc+oJQ6E/gI8OW2jNpg8InRMAyG9nCPsno+HBGRSeDH9vYngNPsiq3nAd93NYuLt3+YBkNljMAwGNrDjOt53vV/Hus6DAGHVHlFVIOhYzAmKYOheRzB6lsRGGX1gdgmIu8AEIv1zRycwdAoRmAYDE1CKTUO/EZEtlClAU8VrgHeKyKPYVVEvbKZ4zMYGsWE1RoMBoPBF0bDMBgMBoMvjMAwGAwGgy+MwDAYDAaDL4zAMBgMBoMvjMAwGAwGgy+MwDAYDAaDL4zAMBgMBoMv/j8O+0Ai7/eUOwAAAABJRU5ErkJggg==\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "my_data.plot(x='time', y='tair_max')\n", "plt.ylabel('Maximum Annual Air Temperature ($\\degree C$)')\n", "plt.title('Maximum Annual Air Temperature Timeseries');" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Histogram plots\n", "\n", "- We are probably interested in what kind of distribution our data has.\n", "- Make a histogram plot to quickly inspect (Note: Careful with the choice of [number or width of bins](https://en.wikipedia.org/wiki/Histogram#Number_of_bins_and_width))\n", "- See documentation for making [histograms with pandas](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.Series.plot.hist.html#pandas.Series.plot.hist), and [histograms with matplotlib](https://matplotlib.org/3.1.3/gallery/statistics/hist.html)" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "my_data['tair_max'].plot.hist(bins=10)\n", "plt.xlabel('Maximum Annual Air Temperature ($\\degree C$)')\n", "plt.title('Maximum Annual Air Temperature Histogram');" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Boxplots\n", "\n", "- A [boxplot](https://en.wikipedia.org/wiki/Box_plot) (sometimes called \"box-and-whisker\" plots) can also help visualize a distribution, especially when we want to compare multiple data sets side by side.\n", "- The box usually represents the interquartile range (IQR) (between the 25th and 75th percentiles)\n", "- Symbols (lines, circles, etc) within the box can represent the sample mean and/or median\n", "- Vertical line \"whiskers\" can represent the full range (minimum to maximum) or another percentile range (such as 2nd and 98th percentiles)\n", "- Data points beyond the \"whiskers\" are \"outliers\"\n", "- What each symbol represents can vary, so be sure to check documentation to be sure! See documentation for making [boxplots with pandas](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.boxplot.html), and [boxplots with matplotlib](https://matplotlib.org/3.2.1/api/_as_gen/matplotlib.pyplot.boxplot.html)." ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "my_data.boxplot(column=['tair_min','tair_max'], grid=False)\n", "plt.ylabel('Air Temperature ($\\degree C$)')\n", "plt.title('Min/Max Annual Air Temperature Boxplots');" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Let's look at a different set of data:" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "from scipy.stats import linregress\n", "\n", "# Plot the same set of points three different ways to show how plots can be manipulated to trick us!\n", "fig, [plotA, plotB, plotC] = plt.subplots(ncols=3, nrows=1, figsize=(15,5), tight_layout=True)\n", "\n", "# The underlying data is a linear relationship, but with a lot of random noise added\n", "# There is a trend in the data, but it is hard to detect\n", "x = np.linspace(0,20,21)\n", "y = x + 15*np.random.randn(21)\n", "\n", "# Be careful! Depending only on the axes limits we choose, we can make the data look very different\n", "plotA.scatter(x,y)\n", "# Adding a regression line can sometimes be misleading (suggesting there's a trend even if there isn't)\n", "m, b, _, _, _ = linregress(x, y)\n", "# Just because I've plotted a linear regression here, doesn't mean that it's statistically significant!\n", "plotA.plot(x, m*x + b, color='red')\n", "plotA.set_xlim((-1,21)); plotA.set_ylim((-50,50))\n", "plotA.set_xlabel('X'); plotA.set_ylabel('Y')\n", "plotA.set_title('A', fontsize=25, fontweight='bold')\n", "\n", "# We can make the data look a lot different by just changing the axes limites\n", "# This can be misleading, be careful!\n", "plotB.scatter(x,y)\n", "plotB.set_xlim((-1,21)); plotB.set_ylim((-150,150))\n", "plotB.set_xlabel('X'); plotB.set_ylabel('Y')\n", "plotB.set_title('B', fontsize=25, fontweight='bold')\n", "\n", "# We can make the data look a lot different by just changing the axes limites\n", "# This can be misleading, be careful!\n", "plotC.scatter(x,y)\n", "plotC.set_xlim((-50,50)); plotC.set_ylim((-50,50))\n", "plotC.set_xlabel('X'); plotC.set_ylabel('Y')\n", "plotC.set_title('C', fontsize=25, fontweight='bold')\n", "\n", "fig.suptitle('Is there a trend in any of these plots?', fontsize=20, fontweight='bold', y=1.05);" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Ethics in graphical analysis\n", "\n", "**Be careful!**\n", "- Others could try and manipulate plots and statistics to convince us of something\n", "- We can end up tricking outselves with \"wishful thinking\" and \"confirmation bias\" if we are not careful\n", "- This is why we have statistical tests, they're our attempt to find objective measures of \"is this a true trend\"\n", "- **Don't draw a trendline through data when there isn't a statisticaly significant trend!**" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.10.10" } }, "nbformat": 4, "nbformat_minor": 4 }