Created
May 29, 2020 05:49
-
-
Save cs224/5fee6b3e2fd8685f12a54053431d06ea to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
{ | |
"cells": [ | |
{ | |
"cell_type": "code", | |
"execution_count": 1, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
"cs224 \n", | |
"last updated: 2020-05-29 \n", | |
"\n", | |
"CPython 3.6.10\n", | |
"IPython 7.13.0\n", | |
"\n", | |
"numpy 1.18.1\n", | |
"xarray 0.15.0\n", | |
"scipy 1.4.1\n", | |
"pandas 1.0.2\n", | |
"sklearn 0.22.1\n", | |
"matplotlib 3.1.3\n", | |
"seaborn 0.10.0\n", | |
"pymc3 3.8\n", | |
"lifelines 0.24.2\n", | |
"rpy2 3.2.6\n" | |
] | |
} | |
], | |
"source": [ | |
"%load_ext watermark\n", | |
"%watermark -a 'cs224' -u -d -v -p numpy,xarray,scipy,pandas,sklearn,matplotlib,seaborn,pymc3,lifelines,rpy2" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 2, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"%matplotlib inline\n", | |
"import numpy as np, scipy, scipy.stats as stats, scipy.special, scipy.misc, pandas as pd, matplotlib.pyplot as plt, seaborn as sns, xarray as xr\n", | |
"import matplotlib as mpl\n", | |
"\n", | |
"import pymc3 as pm\n", | |
"\n", | |
"import theano as thno\n", | |
"import theano.tensor as T\n", | |
"\n", | |
"import datetime, time, math\n", | |
"from dateutil import relativedelta\n", | |
"\n", | |
"from collections import OrderedDict\n", | |
"\n", | |
"SEED = 41\n", | |
"np.random.seed(SEED)\n", | |
"\n", | |
"pd.set_option('display.max_columns', 500)\n", | |
"pd.set_option('display.width', 1000)\n", | |
"# pd.set_option('display.float_format', lambda x: '%.2f' % x)\n", | |
"np.set_printoptions(edgeitems=10)\n", | |
"np.set_printoptions(linewidth=1000)\n", | |
"np.set_printoptions(suppress=True)\n", | |
"np.core.arrayprint._line_width = 180\n", | |
"\n", | |
"sns.set()" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 3, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/html": [ | |
"<style>.container { width:70% !important; }</style>" | |
], | |
"text/plain": [ | |
"<IPython.core.display.HTML object>" | |
] | |
}, | |
"metadata": {}, | |
"output_type": "display_data" | |
} | |
], | |
"source": [ | |
"from IPython.display import display, HTML\n", | |
"\n", | |
"from IPython.display import display_html\n", | |
"def display_side_by_side(*args):\n", | |
" html_str=''\n", | |
" for df in args:\n", | |
" if type(df) == np.ndarray:\n", | |
" df = pd.DataFrame(df)\n", | |
" html_str+=df.to_html()\n", | |
" html_str = html_str.replace('table','table style=\"display:inline\"')\n", | |
" # print(html_str)\n", | |
" display_html(html_str,raw=True)\n", | |
"\n", | |
"CSS = \"\"\"\n", | |
".output {\n", | |
" flex-direction: row;\n", | |
"}\n", | |
"\"\"\"\n", | |
"\n", | |
"def display_graphs_side_by_side(*args):\n", | |
" html_str='<table><tr>'\n", | |
" for g in args:\n", | |
" html_str += '<td>'\n", | |
" html_str += g._repr_svg_()\n", | |
" html_str += '</td>'\n", | |
" html_str += '</tr></table>'\n", | |
" display_html(html_str,raw=True)\n", | |
" \n", | |
"\n", | |
"display(HTML(\"<style>.container { width:70% !important; }</style>\"))" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 4, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
"3.2.6\n" | |
] | |
} | |
], | |
"source": [ | |
"import rpy2\n", | |
"print(rpy2.__version__)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 5, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
"$version.string\n", | |
"[1] \"R version 3.6.1 (2019-07-05)\"\n", | |
"\n", | |
"\n" | |
] | |
} | |
], | |
"source": [ | |
"import rpy2.robjects.packages as rpackages\n", | |
"baseR = rpackages.importr('base')\n", | |
"print(baseR.R_Version().rx('version.string'))" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 6, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"name": "stderr", | |
"output_type": "stream", | |
"text": [ | |
"/home/local/cs/local/install/anaconda3-5.3.1-Linux-x86_64/envs/py36ds/lib/python3.6/site-packages/rpy2/robjects/pandas2ri.py:14: FutureWarning: pandas.core.index is deprecated and will be removed in a future version. The public classes are available in the top-level namespace.\n", | |
" from pandas.core.index import Index as PandasIndex\n", | |
"/home/local/cs/local/install/anaconda3-5.3.1-Linux-x86_64/envs/py36ds/lib/python3.6/site-packages/rpy2/robjects/pandas2ri.py:34: UserWarning: pandas >= 1.0 is not supported.\n", | |
" warnings.warn('pandas >= 1.0 is not supported.')\n", | |
"/home/local/cs/local/install/anaconda3-5.3.1-Linux-x86_64/envs/py36ds/lib/python3.6/site-packages/rpy2/robjects/lib/ggplot2.py:72: UserWarning: This was designed againt ggplot2 version 3.2.1 but you have 3.3.0\n", | |
" 'have %s' % (TARGET_VERSION, ggplot2.__version__))\n" | |
] | |
} | |
], | |
"source": [ | |
"import IPython.display\n", | |
"import rpy2, rpy2.robjects, rpy2.robjects.pandas2ri, rpy2.rinterface, rpy2.robjects.packages, rpy2.interactive, rpy2.robjects.lib.ggplot2, rpy2.robjects.lib.grdevices\n", | |
"# rpy2.robjects.pandas2ri.activate()\n", | |
"\n", | |
"from rpy2.robjects.packages import importr\n", | |
"# import R's \"base\" package\n", | |
"base = importr('base')\n", | |
"\n", | |
"# import rpy2's package module\n", | |
"import rpy2.robjects.packages as rpackages\n", | |
"\n", | |
"# import R's utility package\n", | |
"utils = rpackages.importr('utils')\n", | |
"\n", | |
"# select a mirror for R packages\n", | |
"utils.chooseCRANmirror(ind=1) # select the first mirror in the list\n", | |
"\n", | |
"# R package names\n", | |
"packnames = ['lmtest']\n", | |
"\n", | |
"# R vector of strings\n", | |
"from rpy2.robjects.vectors import StrVector" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 7, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"grdevices = rpy2.robjects.packages.importr('grDevices')\n", | |
"# Selectively install what needs to be install.\n", | |
"# We are fancy, just because we can.\n", | |
"names_to_install = [x for x in packnames if not rpackages.isinstalled(x)]\n", | |
"if len(names_to_install) > 0:\n", | |
" utils.install_packages(StrVector(names_to_install))" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 8, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"# lexis_grid = rpy2.robjects.r['lexis.grid']\n", | |
"# lexis_lifeline = rpy2.robjects.r['lexis.lifeline']" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 9, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"# def plot_lexis(mortality_analysis_instance):\n", | |
"# mylexis = lexis_grid(year_start = 2020, year_end = 2021, age_start = 0, age_end = 1) # lwd = 0.1\n", | |
"\n", | |
"# alpha = 1.0\n", | |
"# ix_present = ~mortality_analysis_instance.df_lifelines_individual.observed_death\n", | |
"# ix_lost = mortality_analysis_instance.df_lifelines_individual.observed_death\n", | |
"# mylexis = lexis_lifeline(lg = mylexis , entry = mortality_analysis_instance.df_lifelines_individual['start_date'][ix_present], exit = mortality_analysis_instance.df_lifelines_individual['end_date'][ix_present], colour = \"orange\", alpha = alpha, lwd = 0.4)\n", | |
"# mylexis = lexis_lifeline(lg = mylexis , entry = mortality_analysis_instance.df_lifelines_individual['start_date'][ix_lost] , exit = mortality_analysis_instance.df_lifelines_individual['end_date'][ix_lost] , colour = \"blue\" , alpha = alpha, lwd = 0.4, lineends = True)\n", | |
"\n", | |
"# with rpy2.robjects.lib.grdevices.render_to_bytesio(grdevices.png, width=1.5*1024, height=1.5*896, res=90) as img:\n", | |
"# rpy2.robjects.r.print(mylexis) \n", | |
"# IPython.display.display(IPython.display.Image(data=img.getvalue(), format='png', embed=True))" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 10, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"rlm = rpy2.robjects.r['lm']" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 11, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/html": [ | |
"<div>\n", | |
"<style scoped>\n", | |
" .dataframe tbody tr th:only-of-type {\n", | |
" vertical-align: middle;\n", | |
" }\n", | |
"\n", | |
" .dataframe tbody tr th {\n", | |
" vertical-align: top;\n", | |
" }\n", | |
"\n", | |
" .dataframe thead th {\n", | |
" text-align: right;\n", | |
" }\n", | |
"</style>\n", | |
"<table border=\"1\" class=\"dataframe\">\n", | |
" <thead>\n", | |
" <tr style=\"text-align: right;\">\n", | |
" <th></th>\n", | |
" <th>month</th>\n", | |
" <th>temperature</th>\n", | |
" </tr>\n", | |
" </thead>\n", | |
" <tbody>\n", | |
" <tr>\n", | |
" <th>355</th>\n", | |
" <td>356</td>\n", | |
" <td>23.09</td>\n", | |
" </tr>\n", | |
" <tr>\n", | |
" <th>356</th>\n", | |
" <td>357</td>\n", | |
" <td>22.39</td>\n", | |
" </tr>\n", | |
" <tr>\n", | |
" <th>357</th>\n", | |
" <td>358</td>\n", | |
" <td>17.05</td>\n", | |
" </tr>\n", | |
" <tr>\n", | |
" <th>358</th>\n", | |
" <td>359</td>\n", | |
" <td>12.28</td>\n", | |
" </tr>\n", | |
" <tr>\n", | |
" <th>359</th>\n", | |
" <td>360</td>\n", | |
" <td>7.19</td>\n", | |
" </tr>\n", | |
" </tbody>\n", | |
"</table>\n", | |
"</div>" | |
], | |
"text/plain": [ | |
" month temperature\n", | |
"355 356 23.09\n", | |
"356 357 22.39\n", | |
"357 358 17.05\n", | |
"358 359 12.28\n", | |
"359 360 7.19" | |
] | |
}, | |
"execution_count": 11, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"df = pd.read_csv('ex5task2.csv', names=['month', 'temperature'])\n", | |
"df.tail()" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 12, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/html": [ | |
"<div>\n", | |
"<style scoped>\n", | |
" .dataframe tbody tr th:only-of-type {\n", | |
" vertical-align: middle;\n", | |
" }\n", | |
"\n", | |
" .dataframe tbody tr th {\n", | |
" vertical-align: top;\n", | |
" }\n", | |
"\n", | |
" .dataframe thead th {\n", | |
" text-align: right;\n", | |
" }\n", | |
"</style>\n", | |
"<table border=\"1\" class=\"dataframe\">\n", | |
" <thead>\n", | |
" <tr style=\"text-align: right;\">\n", | |
" <th></th>\n", | |
" <th>month</th>\n", | |
" <th>temperature</th>\n", | |
" <th>x2</th>\n", | |
" </tr>\n", | |
" </thead>\n", | |
" <tbody>\n", | |
" <tr>\n", | |
" <th>355</th>\n", | |
" <td>356</td>\n", | |
" <td>23.09</td>\n", | |
" <td>-0.50</td>\n", | |
" </tr>\n", | |
" <tr>\n", | |
" <th>356</th>\n", | |
" <td>357</td>\n", | |
" <td>22.39</td>\n", | |
" <td>-0.00</td>\n", | |
" </tr>\n", | |
" <tr>\n", | |
" <th>357</th>\n", | |
" <td>358</td>\n", | |
" <td>17.05</td>\n", | |
" <td>0.50</td>\n", | |
" </tr>\n", | |
" <tr>\n", | |
" <th>358</th>\n", | |
" <td>359</td>\n", | |
" <td>12.28</td>\n", | |
" <td>0.87</td>\n", | |
" </tr>\n", | |
" <tr>\n", | |
" <th>359</th>\n", | |
" <td>360</td>\n", | |
" <td>7.19</td>\n", | |
" <td>1.00</td>\n", | |
" </tr>\n", | |
" </tbody>\n", | |
"</table>\n", | |
"</div>" | |
], | |
"text/plain": [ | |
" month temperature x2\n", | |
"355 356 23.09 -0.50\n", | |
"356 357 22.39 -0.00\n", | |
"357 358 17.05 0.50\n", | |
"358 359 12.28 0.87\n", | |
"359 360 7.19 1.00" | |
] | |
}, | |
"execution_count": 12, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"df['x2'] = np.cos(2*np.pi*df['month']/12)\n", | |
"df.tail().round(2)" | |
] | |
}, | |
{ | |
"cell_type": "markdown", | |
"metadata": {}, | |
"source": [ | |
"# r-statistics.co/Linear-Regression\n", | |
"\n", | |
"* [r-statistics.co/Linear-Regression](http://r-statistics.co/Linear-Regression.html)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 13, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"import rpy2.robjects as ro" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 14, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/html": [ | |
"\n", | |
" <span>R/rpy2 DataFrame (360 x 3)</span>\n", | |
" <table>\n", | |
" <thead>\n", | |
" <tr>\n", | |
" \n", | |
" <th>month</th>\n", | |
" \n", | |
" <th>temperature</th>\n", | |
" \n", | |
" <th>x2</th>\n", | |
" \n", | |
" </tr>\n", | |
" </thead>\n", | |
" <tbody>\n", | |
" \n", | |
" <tr>\n", | |
" \n", | |
" <td>\n", | |
" ...\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" ...\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" ...\n", | |
" </td>\n", | |
" \n", | |
" </tr>\n", | |
" \n", | |
" </tbody>\n", | |
" </table>\n", | |
" " | |
], | |
"text/plain": [ | |
"R object with classes: ('data.frame',) mapped to:\n", | |
"[IntSexpVector, FloatSexpVector, FloatSexpVector]\n", | |
" month: <class 'rpy2.rinterface.IntSexpVector'>\n", | |
" <rpy2.rinterface.IntSexpVector object at 0x7f3fcc1b22c8> [RTYPES.INTSXP]\n", | |
" temperature: <class 'rpy2.rinterface.FloatSexpVector'>\n", | |
" <rpy2.rinterface.FloatSexpVector object at 0x7f3fcc1b14c8> [RTYPES.REALSXP]\n", | |
" x2: <class 'rpy2.rinterface.FloatSexpVector'>\n", | |
" <rpy2.rinterface.FloatSexpVector object at 0x7f3fc6b0d648> [RTYPES.REALSXP]" | |
] | |
}, | |
"execution_count": 14, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"with rpy2.robjects.conversion.localconverter(ro.default_converter + rpy2.robjects.pandas2ri.converter):\n", | |
" rdf = ro.conversion.py2rpy(df)\n", | |
"\n", | |
"rdf" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 15, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"rformula = ro.r('temperature ~ month + x2')" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 16, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"rmodel1 = rlm(rformula, rdf)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 17, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"rprint = rpy2.robjects.r['print']\n", | |
"rsummary = rpy2.robjects.r['summary']" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 18, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
"\n", | |
"Call:\n", | |
"(function (formula, data, subset, weights, na.action, method = \"qr\", \n", | |
" model = TRUE, x = FALSE, y = FALSE, qr = TRUE, singular.ok = TRUE, \n", | |
" contrasts = NULL, offset, ...) \n", | |
"{\n", | |
" ret.x <- x\n", | |
" ret.y <- y\n", | |
" cl <- match.call()\n", | |
" mf <- match.call(expand.dots = FALSE)\n", | |
" m <- match(c(\"formula\", \"data\", \"subset\", \"weights\", \"na.action\", \n", | |
" \"offset\"), names(mf), 0L)\n", | |
" mf <- mf[c(1L, m)]\n", | |
" mf$drop.unused.levels <- TRUE\n", | |
" mf[[1L]] <- quote(stats::model.frame)\n", | |
" mf <- eval(mf, parent.frame())\n", | |
" if (method == \"model.frame\") \n", | |
" return(mf)\n", | |
" else if (method != \"qr\") \n", | |
" warning(gettextf(\"method = '%s' is not supported. Using 'qr'\", \n", | |
" method), domain = NA)\n", | |
" mt <- attr(mf, \"terms\")\n", | |
" y <- model.response(mf, \"numeric\")\n", | |
" w <- as.vector(model.weights(mf))\n", | |
" if (!is.null(w) && !is.numeric(w)) \n", | |
" stop(\"'weights' must be a numeric vector\")\n", | |
" offset <- model.offset(mf)\n", | |
" mlm <- is.matrix(y)\n", | |
" ny <- if (mlm) \n", | |
" nrow(y)\n", | |
" else length(y)\n", | |
" if (!is.null(offset)) {\n", | |
" if (!mlm) \n", | |
" offset <- as.vector(offset)\n", | |
" if (NROW(offset) != ny) \n", | |
" stop(gettextf(\"number of offsets is %d, should equal %d (number of observations)\", \n", | |
" NROW(offset), ny), domain = NA)\n", | |
" }\n", | |
" if (is.empty.model(mt)) {\n", | |
" x <- NULL\n", | |
" z <- list(coefficients = if (mlm) matrix(NA_real_, 0, \n", | |
" ncol(y)) else numeric(), residuals = y, fitted.values = 0 * \n", | |
" y, weights = w, rank = 0L, df.residual = if (!is.null(w)) sum(w != \n", | |
" 0) else ny)\n", | |
" if (!is.null(offset)) {\n", | |
" z$fitted.values <- offset\n", | |
" z$residuals <- y - offset\n", | |
" }\n", | |
" }\n", | |
" else {\n", | |
" x <- model.matrix(mt, mf, contrasts)\n", | |
" z <- if (is.null(w)) \n", | |
" lm.fit(x, y, offset = offset, singular.ok = singular.ok, \n", | |
" ...)\n", | |
" else lm.wfit(x, y, w, offset = offset, singular.ok = singular.ok, \n", | |
" ...)\n", | |
" }\n", | |
" class(z) <- c(if (mlm) \"mlm\", \"lm\")\n", | |
" z$na.action <- attr(mf, \"na.action\")\n", | |
" z$offset <- offset\n", | |
" z$contrasts <- attr(x, \"contrasts\")\n", | |
" z$xlevels <- .getXlevels(mt, mf)\n", | |
" z$call <- cl\n", | |
" z$terms <- mt\n", | |
" if (model) \n", | |
" z$model <- mf\n", | |
" if (ret.x) \n", | |
" z$x <- x\n", | |
" if (ret.y) \n", | |
" z$y <- y\n", | |
" if (!qr) \n", | |
" z$qr <- NULL\n", | |
" z\n", | |
"})(formula = temperature ~ month + x2, data = structure(list(\n", | |
" month = 1:360, temperature = c(3.23, 4.48, 9.04, 8.64, 16.79, \n", | |
" 17.34, 20.21, 19.17, 15.94, 11.47, 3.27, 3.56, 3.44, 7.3, \n", | |
" 8.52, 9.42, 16.64, 16.94, 19.76, 20.87, 13.75, 11.36, 5.9, \n", | |
" 1.84, 2.82, -0.55, 8.49, 9.33, 12.16, 15.98, 22.25, 21.84, \n", | |
" 17.67, 9.51, 4.97, 1.33, 2, 3.48, 7.17, 10.29, 17.11, 18.67, \n", | |
" 20.61, 21.55, 15.43, 8.22, 7.21, 2.7, 4.51, 0.47, 5.86, 13.09, \n", | |
" 16.69, 18.48, 18.89, 19.44, 14.48, 9.3, 2, 5.26, 4.4, 2.67, \n", | |
" 9.08, 9.97, 15.04, 18.92, 23.65, 19.85, 15.38, 9.37, 8.91, \n", | |
" 5.49, 2.08, 6.59, 5.34, 10.98, 14.59, 16.52, 22.27, 20.18, \n", | |
" 13.92, 13.05, 4.23, 0.97, -0.2, 0.77, 4.22, 10.68, 13.19, \n", | |
" 18, 18.36, 19.05, 13.03, 10.44, 5.99, -0.27, -2.95, 5.73, \n", | |
" 8.79, 9.1, 14.81, 17.65, 18.95, 21.84, 16.02, 9.33, 5.35, \n", | |
" 3.91, 3.47, 4.53, 7.74, 10.53, 16.3, 18.99, 18.72, 19.85, \n", | |
" 15.17, 10.76, 3.35, 2.6, 4.05, 2.16, 7.38, 11.39, 16.29, \n", | |
" 17.59, 21.58, 19.79, 19.13, 10.4, 4.7, 3.41, 2.83, 5.42, \n", | |
" 7.57, 11.94, 16.6, 19.71, 17.21, 20.38, 16.02, 11.41, 7.47, \n", | |
" 4.75, 2.78, 4.94, 7.46, 9.06, 17.35, 16.69, 21.14, 21.1, \n", | |
" 13.34, 13.86, 4.27, 1.56, 0.75, 6.92, 7.75, 10.61, 14.81, \n", | |
" 19.85, 19.42, 20.26, 14.8, 10.82, 8.17, 4.27, 0.82, 0.61, \n", | |
" 8.6, 11.06, 16.21, 22.64, 21.68, 24.1, 16.25, 8.35, 7.13, \n", | |
" 2.9, 2.05, 4.21, 5.79, 11.8, 13.44, 17.96, 19.6, 20.57, 16.49, \n", | |
" 11.79, 5.54, 1.69, 3.15, 0.76, 6.4, 11.42, 15.07, 19.52, \n", | |
" 20.64, 18.31, 17.54, 12.41, 5.28, 2.14, -0.83, 1.83, 4.53, \n", | |
" 10.78, 15.79, 19.23, 24.43, 16.98, 18.38, 13.45, 8.32, 4.87, \n", | |
" 6.35, 6, 7.66, 14.45, 16.45, 19.27, 19.29, 18.89, 14.07, \n", | |
" 10.02, 4.84, 2.19, 4.98, 4.66, 6.46, 9.6, 17.82, 19.36, 20.23, \n", | |
" 19.26, 13.91, 10.46, 6.3, 1.95, -2.03, 1.99, 6.07, 13.67, \n", | |
" 16.11, 17.68, 19.84, 21.03, 16.6, 10.32, 8.64, 2.46, -1.06, \n", | |
" 2.36, 6.31, 11.18, 12.54, 18.93, 21.83, 18.36, 14.16, 9.28, \n", | |
" 7.06, -1.21, 2.58, 3.62, 7.92, 14.15, 16.6, 18.91, 18.02, \n", | |
" 19.77, 17.18, 10.71, 5.2, 5.4, 3.83, -1.04, 8.97, 10.25, \n", | |
" 16.69, 18.03, 19.72, 21.18, 15.4, 9.85, 6.57, 3.89, 2.24, \n", | |
" 0.99, 3.44, 10.54, 12.9, 17.79, 22.36, 19.84, 15.62, 12.13, \n", | |
" 5.82, 4.47, 4.61, 5.81, 8.79, 13.28, 14.43, 18.85, 20.85, \n", | |
" 17.76, 16.53, 13.37, 7.63, 4.31, 3.09, 2.04, 7.01, 10.97, \n", | |
" 15.35, 18.76, 22.83, 21.99, 15.04, 10.08, 8.42, 7.37, 3.96, \n", | |
" 5.34, 6.04, 9.84, 15.15, 18.44, 21.08, 20.25, 18.63, 9.81, \n", | |
" 5.56, 2.29, 0.55, -1.19, 5.4, 9.45, 9.98, 16.22, 20.78, 20.96, \n", | |
" 19.99, 14.43, 11.98, 6.17, 4.04, 6.6, 0.46, 5.22, 14.19, \n", | |
" 17.45, 20.28, 23.09, 22.39, 17.05, 12.28, 7.19), x2 = c(0.866025403784439, \n", | |
" 0.5, 6.12323399573677e-17, -0.5, -0.866025403784439, -1, \n", | |
" -0.866025403784439, -0.5, -1.83697019872103e-16, 0.5, 0.866025403784438, \n", | |
" 1, 0.866025403784439, 0.500000000000001, 1.19434011948696e-15, \n", | |
" -0.499999999999999, -0.866025403784439, -1, -0.866025403784439, \n", | |
" -0.5, -4.28626379701574e-16, 0.499999999999999, 0.866025403784438, \n", | |
" 1, 0.866025403784438, 0.5, 5.51091059616309e-16, -0.499999999999999, \n", | |
" -0.866025403784439, -1, -0.866025403784439, -0.500000000000002, \n", | |
" 1.10280109986921e-15, 0.5, 0.866025403784439, 1, 0.866025403784439, \n", | |
" 0.500000000000002, 2.57237725884603e-15, -0.5, -0.866025403784437, \n", | |
" -1, -0.866025403784439, -0.500000000000002, 8.57871740039736e-16, \n", | |
" 0.499999999999997, 0.866025403784438, 1, 0.866025403784439, \n", | |
" 0.499999999999999, 2.8173066186755e-15, -0.5, -0.866025403784438, \n", | |
" -1, -0.866025403784438, -0.500000000000002, -2.93977129859024e-15, \n", | |
" 0.5, 0.866025403784438, 1, 0.86602540378444, 0.500000000000002, \n", | |
" -4.9047770029553e-16, -0.499999999999997, -0.866025403784438, \n", | |
" -1, -0.86602540378444, -0.499999999999999, -3.18470065841971e-15, \n", | |
" 0.5, 0.86602540378444, 1, 0.866025403784441, 0.500000000000002, \n", | |
" -2.45548340466059e-16, -0.499999999999997, -0.866025403784438, \n", | |
" -1, -0.86602540378444, -0.499999999999999, -3.42963001824918e-15, \n", | |
" 0.499999999999993, 0.86602540378444, 1, 0.866025403784438, \n", | |
" 0.500000000000003, 7.10480837696441e-15, -0.499999999999996, \n", | |
" -0.866025403784438, -1, -0.86602540378444, -0.500000000000006, \n", | |
" -3.67455937807865e-15, 0.499999999999999, 0.86602540378444, \n", | |
" 1, 0.866025403784442, 0.500000000000003, 2.44310379192882e-16, \n", | |
" -0.500000000000002, -0.866025403784438, -1, -0.86602540378444, \n", | |
" -0.5, 3.18593861969288e-15, 0.499999999999999, 0.866025403784436, \n", | |
" 1, 0.866025403784442, 0.499999999999997, 4.89239739022353e-16, \n", | |
" -0.499999999999996, -0.866025403784438, -1, -0.866025403784437, \n", | |
" -0.5, -4.16441809773759e-15, 0.499999999999999, 0.866025403784436, \n", | |
" 1, 0.866025403784438, 0.500000000000003, 7.83959645645283e-15, \n", | |
" -0.499999999999996, -0.866025403784438, -1, -0.866025403784437, \n", | |
" -0.500000000000006, -4.40934745756706e-15, 0.499999999999999, \n", | |
" 0.866025403784439, 1, 0.866025403784442, 0.500000000000003, \n", | |
" 9.79098458681294e-16, -0.500000000000002, -0.866025403784441, \n", | |
" -1, -0.86602540378444, -0.5, -1.17597041749975e-14, 0.500000000000005, \n", | |
" 0.866025403784436, 1, 0.866025403784439, 0.50000000000001, \n", | |
" -5.88139953909024e-15, -0.499999999999995, -0.866025403784437, \n", | |
" -1, -0.866025403784444, -0.500000000000007, -4.899206177226e-15, \n", | |
" 0.499999999999998, 0.866025403784439, 1, 0.866025403784442, \n", | |
" 0.500000000000004, 1.46895717834024e-15, -0.500000000000001, \n", | |
" -0.866025403784434, -1, -0.866025403784441, -0.500000000000013, \n", | |
" -1.22495628946565e-14, 0.500000000000004, 0.866025403784443, \n", | |
" 1, 0.866025403784439, 0.499999999999998, 8.81931389577071e-15, \n", | |
" -0.499999999999995, -0.866025403784437, -1, -0.866025403784444, \n", | |
" -0.500000000000007, 8.82178981831706e-15, 0.499999999999998, \n", | |
" 0.866025403784439, 1, 0.866025403784443, 0.500000000000004, \n", | |
" 1.95881589799918e-15, -0.499999999999989, -0.866025403784433, \n", | |
" -1, -0.866025403784434, -0.500000000000001, 1.47143310088659e-15, \n", | |
" 0.500000000000004, 0.866025403784435, 1, 0.866025403784439, \n", | |
" 0.500000000000011, 9.30917261542965e-15, -0.499999999999995, \n", | |
" -0.86602540378443, -1, -0.866025403784437, -0.499999999999995, \n", | |
" -5.87892361654388e-15, 0.499999999999997, 0.866025403784439, \n", | |
" 1, 0.866025403784443, 0.500000000000005, 1.66595293328601e-14, \n", | |
" -0.5, -0.86602540378444, -1, -0.866025403784441, -0.500000000000002, \n", | |
" 9.81574381227648e-16, 0.499999999999991, 0.866025403784435, \n", | |
" 1, 0.866025403784446, 0.500000000000011, -4.41182338011341e-15, \n", | |
" -0.500000000000006, -0.866025403784437, -1, -0.866025403784438, \n", | |
" -0.500000000000008, -6.36878233620282e-15, 0.499999999999997, \n", | |
" 0.866025403784431, 1, 0.866025403784443, 0.499999999999993, \n", | |
" 2.93853333731706e-15, -0.5, -0.86602540378444, -1, -0.866025403784441, \n", | |
" -0.500000000000002, -1.37191390536333e-14, 0.499999999999991, \n", | |
" 0.866025403784435, 1, 0.86602540378444, 0.499999999999999, \n", | |
" -3.92196466045447e-15, -0.499999999999994, -0.866025403784436, \n", | |
" -1, -0.866025403784438, -0.500000000000008, -2.10694957710638e-14, \n", | |
" 0.499999999999997, 0.866025403784445, 1, 0.866025403784443, \n", | |
" 0.499999999999993, 3.428392056976e-15, -0.499999999999987, \n", | |
" -0.86602540378444, -1, -0.866025403784449, -0.500000000000003, \n", | |
" -1.42089977732922e-14, 0.500000000000003, 0.866025403784434, \n", | |
" 1, 0.86602540378444, 0.500000000000012, -3.43210594079553e-15, \n", | |
" -0.499999999999993, -0.866025403784429, -1, -0.866025403784445, \n", | |
" -0.499999999999997, -7.34849977552071e-15, 0.500000000000008, \n", | |
" 0.866025403784438, 1, 0.866025403784436, 0.500000000000006, \n", | |
" 1.81291054918369e-14, -0.499999999999999, -0.866025403784432, \n", | |
" -1, -0.866025403784442, -0.499999999999991, -4.88001777749176e-16, \n", | |
" 0.49999999999999, 0.866025403784441, 1, 0.866025403784447, \n", | |
" 0.5, 1.12686074940654e-14, -0.49999999999998, -0.866025403784436, \n", | |
" -1, -0.866025403784438, -0.500000000000009, 6.37249622002235e-15, \n", | |
" 0.499999999999996, 0.866025403784431, 1, 0.866025403784444, \n", | |
" 0.500000000000019, 4.40810949629388e-15, -0.499999999999986, \n", | |
" -0.866025403784439, -1, -0.866025403784435, -0.500000000000003, \n", | |
" -1.51887152126101e-14, 0.500000000000002, 0.866025403784434, \n", | |
" 1, 0.86602540378444, 0.500000000000013, -2.45238850147765e-15, \n", | |
" -0.499999999999992, -0.866025403784443, -1, -0.866025403784446, \n", | |
" -0.499999999999997, -8.32821721483859e-15, 0.499999999999983, \n", | |
" 0.866025403784437, 1, 0.866025403784451, 0.500000000000007, \n", | |
" -9.31288649924918e-15, -0.499999999999974, -0.866025403784432, \n", | |
" -1, -0.866025403784442, -0.499999999999992, -1.46771921706706e-15, \n", | |
" 0.500000000000014, 0.866025403784427, 1, 0.866025403784448, \n", | |
" 0.500000000000001, 1.22483249333833e-14, -0.500000000000004, \n", | |
" -0.866025403784421, -1, -0.866025403784439, -0.50000000000001, \n", | |
" 5.39277878070447e-15, 0.499999999999995, 0.866025403784444, \n", | |
" 1, 0.866025403784444, 0.50000000000002, 5.38782693561177e-15, \n", | |
" -0.499999999999986, -0.866025403784439, -1, -0.86602540378445, \n", | |
" -0.500000000000004, -1.6168432651928e-14, 0.500000000000001, \n", | |
" 0.866025403784433, 1)), class = \"data.frame\", row.names = c(\"0\", \n", | |
"\"1\", \"2\", \"3\", \"4\", \"5\", \"6\", \"7\", \"8\", \"9\", \"10\", \"11\", \"12\", \n", | |
"\"13\", \"14\", \"15\", \"16\", \"17\", \"18\", \"19\", \"20\", \"21\", \"22\", \"23\", \n", | |
"\"24\", \"25\", \"26\", \"27\", \"28\", \"29\", \"30\", \"31\", \"32\", \"33\", \"34\", \n", | |
"\"35\", \"36\", \"37\", \"38\", \"39\", \"40\", \"41\", \"42\", \"43\", \"44\", \"45\", \n", | |
"\"46\", \"47\", \"48\", \"49\", \"50\", \"51\", \"52\", \"53\", \"54\", \"55\", \"56\", \n", | |
"\"57\", \"58\", \"59\", \"60\", \"61\", \"62\", \"63\", \"64\", \"65\", \"66\", \"67\", \n", | |
"\"68\", \"69\", \"70\", \"71\", \"72\", \"73\", \"74\", \"75\", \"76\", \"77\", \"78\", \n", | |
"\"79\", \"80\", \"81\", \"82\", \"83\", \"84\", \"85\", \"86\", \"87\", \"88\", \"89\", \n", | |
"\"90\", \"91\", \"92\", \"93\", \"94\", \"95\", \"96\", \"97\", \"98\", \"99\", \"100\", \n", | |
"\"101\", \"102\", \"103\", \"104\", \"105\", \"106\", \"107\", \"108\", \"109\", \n", | |
"\"110\", \"111\", \"112\", \"113\", \"114\", \"115\", \"116\", \"117\", \"118\", \n", | |
"\"119\", \"120\", \"121\", \"122\", \"123\", \"124\", \"125\", \"126\", \"127\", \n", | |
"\"128\", \"129\", \"130\", \"131\", \"132\", \"133\", \"134\", \"135\", \"136\", \n", | |
"\"137\", \"138\", \"139\", \"140\", \"141\", \"142\", \"143\", \"144\", \"145\", \n", | |
"\"146\", \"147\", \"148\", \"149\", \"150\", \"151\", \"152\", \"153\", \"154\", \n", | |
"\"155\", \"156\", \"157\", \"158\", \"159\", \"160\", \"161\", \"162\", \"163\", \n", | |
"\"164\", \"165\", \"166\", \"167\", \"168\", \"169\", \"170\", \"171\", \"172\", \n", | |
"\"173\", \"174\", \"175\", \"176\", \"177\", \"178\", \"179\", \"180\", \"181\", \n", | |
"\"182\", \"183\", \"184\", \"185\", \"186\", \"187\", \"188\", \"189\", \"190\", \n", | |
"\"191\", \"192\", \"193\", \"194\", \"195\", \"196\", \"197\", \"198\", \"199\", \n", | |
"\"200\", \"201\", \"202\", \"203\", \"204\", \"205\", \"206\", \"207\", \"208\", \n", | |
"\"209\", \"210\", \"211\", \"212\", \"213\", \"214\", \"215\", \"216\", \"217\", \n", | |
"\"218\", \"219\", \"220\", \"221\", \"222\", \"223\", \"224\", \"225\", \"226\", \n", | |
"\"227\", \"228\", \"229\", \"230\", \"231\", \"232\", \"233\", \"234\", \"235\", \n", | |
"\"236\", \"237\", \"238\", \"239\", \"240\", \"241\", \"242\", \"243\", \"244\", \n", | |
"\"245\", \"246\", \"247\", \"248\", \"249\", \"250\", \"251\", \"252\", \"253\", \n", | |
"\"254\", \"255\", \"256\", \"257\", \"258\", \"259\", \"260\", \"261\", \"262\", \n", | |
"\"263\", \"264\", \"265\", \"266\", \"267\", \"268\", \"269\", \"270\", \"271\", \n", | |
"\"272\", \"273\", \"274\", \"275\", \"276\", \"277\", \"278\", \"279\", \"280\", \n", | |
"\"281\", \"282\", \"283\", \"284\", \"285\", \"286\", \"287\", \"288\", \"289\", \n", | |
"\"290\", \"291\", \"292\", \"293\", \"294\", \"295\", \"296\", \"297\", \"298\", \n", | |
"\"299\", \"300\", \"301\", \"302\", \"303\", \"304\", \"305\", \"306\", \"307\", \n", | |
"\"308\", \"309\", \"310\", \"311\", \"312\", \"313\", \"314\", \"315\", \"316\", \n", | |
"\"317\", \"318\", \"319\", \"320\", \"321\", \"322\", \"323\", \"324\", \"325\", \n", | |
"\"326\", \"327\", \"328\", \"329\", \"330\", \"331\", \"332\", \"333\", \"334\", \n", | |
"\"335\", \"336\", \"337\", \"338\", \"339\", \"340\", \"341\", \"342\", \"343\", \n", | |
"\"344\", \"345\", \"346\", \"347\", \"348\", \"349\", \"350\", \"351\", \"352\", \n", | |
"\"353\", \"354\", \"355\", \"356\", \"357\", \"358\", \"359\")))\n", | |
"\n", | |
"Coefficients:\n", | |
"(Intercept) month x2 \n", | |
" 10.702370 0.002824 -7.786434 \n", | |
"\n", | |
"\n" | |
] | |
} | |
], | |
"source": [ | |
"print(rmodel1)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 19, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/html": [ | |
"\n", | |
" <span>ListVector with 1 elements.</span>\n", | |
" <table>\n", | |
" <tbody>\n", | |
" \n", | |
" <tr>\n", | |
" <th>\n", | |
" coefficients\n", | |
" </th>\n", | |
" <td>\n", | |
" <rpy2.rinterface.FloatSexpVector object at 0x7f3fc6b14d88> [RTYPES.REALSXP]\n", | |
" </td>\n", | |
" </tr>\n", | |
" \n", | |
" </tbody>\n", | |
" </table>\n", | |
" " | |
], | |
"text/plain": [ | |
"R object with classes: ('list',) mapped to:\n", | |
"[FloatSexpVector]\n", | |
" coefficients: <class 'rpy2.rinterface.FloatSexpVector'>\n", | |
" <rpy2.rinterface.FloatSexpVector object at 0x7f3fc6b14d48> [RTYPES.REALSXP]" | |
] | |
}, | |
"execution_count": 19, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"rmodel1.rx('coefficients')" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 20, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
"$coefficients\n", | |
" (Intercept) month x2 \n", | |
"10.702370255 0.002823587 -7.786433525 \n", | |
"\n", | |
"\n" | |
] | |
} | |
], | |
"source": [ | |
"print(rmodel1.rx('coefficients'))" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 21, | |
"metadata": { | |
"scrolled": false | |
}, | |
"outputs": [ | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
" (Intercept) month x2 \n", | |
"10.702370255 0.002823587 -7.786433525 \n", | |
"\n" | |
] | |
} | |
], | |
"source": [ | |
"rcoeff = rmodel1.rx('coefficients')[0]\n", | |
"print(rcoeff)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 22, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"['(Intercept)', 'month', 'x2']" | |
] | |
}, | |
"execution_count": 22, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"names = rcoeff.names\n", | |
"list(names)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 23, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"[10.702370255288873, 0.0028235873822100156, -7.786433524505263]" | |
] | |
}, | |
"execution_count": 23, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"list(rcoeff)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 24, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/html": [ | |
"<div>\n", | |
"<style scoped>\n", | |
" .dataframe tbody tr th:only-of-type {\n", | |
" vertical-align: middle;\n", | |
" }\n", | |
"\n", | |
" .dataframe tbody tr th {\n", | |
" vertical-align: top;\n", | |
" }\n", | |
"\n", | |
" .dataframe thead th {\n", | |
" text-align: right;\n", | |
" }\n", | |
"</style>\n", | |
"<table border=\"1\" class=\"dataframe\">\n", | |
" <thead>\n", | |
" <tr style=\"text-align: right;\">\n", | |
" <th></th>\n", | |
" <th>(Intercept)</th>\n", | |
" <th>month</th>\n", | |
" <th>x2</th>\n", | |
" </tr>\n", | |
" </thead>\n", | |
" <tbody>\n", | |
" <tr>\n", | |
" <th>0</th>\n", | |
" <td>10.70237</td>\n", | |
" <td>0.002824</td>\n", | |
" <td>-7.786434</td>\n", | |
" </tr>\n", | |
" </tbody>\n", | |
"</table>\n", | |
"</div>" | |
], | |
"text/plain": [ | |
" (Intercept) month x2\n", | |
"0 10.70237 0.002824 -7.786434" | |
] | |
}, | |
"execution_count": 24, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"coeffdf = pd.DataFrame(columns=list(names))\n", | |
"coeffdf.loc[0] = list(rcoeff)\n", | |
"coeffdf" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 25, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/html": [ | |
"\n", | |
" <span>ListVector with 11 elements.</span>\n", | |
" <table>\n", | |
" <tbody>\n", | |
" \n", | |
" <tr>\n", | |
" <th>\n", | |
" call\n", | |
" </th>\n", | |
" <td>\n", | |
" <rpy2.rinterface.LangSexpVector object at 0x7f3fc64b8948> [RTYPES.LANGSXP]\n", | |
" </td>\n", | |
" </tr>\n", | |
" \n", | |
" <tr>\n", | |
" <th>\n", | |
" terms\n", | |
" </th>\n", | |
" <td>\n", | |
" temperature ~ month + x2\n", | |
"attr(,\"variables\")\n", | |
"list(temperature, month, x2)\n", | |
"attr(,\"factors\")\n", | |
" month x2\n", | |
"temperature 0 0\n", | |
"month 1 0\n", | |
"x2 0 1\n", | |
"attr(,\"term.labels\")\n", | |
"[1] \"month\" \"x2\" \n", | |
"attr(,\"order\")\n", | |
"[1] 1 1\n", | |
"attr(,\"intercept\")\n", | |
"[1] 1\n", | |
"attr(,\"response\")\n", | |
"[1] 1\n", | |
"attr(,\".Environment\")\n", | |
"<environment: R_GlobalEnv>\n", | |
"attr(,\"predvars\")\n", | |
"list(temperature, month, x2)\n", | |
"attr(,\"dataClasses\")\n", | |
"temperature month x2 \n", | |
" \"numeric\" \"numeric\" \"numeric\" \n", | |
"\n", | |
" </td>\n", | |
" </tr>\n", | |
" \n", | |
" <tr>\n", | |
" <th>\n", | |
" residuals\n", | |
" </th>\n", | |
" <td>\n", | |
" \n", | |
" <span>FloatVector with 360 elements.</span>\n", | |
" <table>\n", | |
" <tbody>\n", | |
" <tr>\n", | |
" \n", | |
" <td>\n", | |
" -0.731945\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" -2.334801\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" -1.670841\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" ...\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" 9.230002\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" 7.307211\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" 3.257572\n", | |
" </td>\n", | |
" \n", | |
" </tr>\n", | |
" </tbody>\n", | |
" </table>\n", | |
" \n", | |
" </td>\n", | |
" </tr>\n", | |
" \n", | |
" <tr>\n", | |
" <th>\n", | |
" ...\n", | |
" </th>\n", | |
" <td>\n", | |
" ...\n", | |
" </td>\n", | |
" </tr>\n", | |
" \n", | |
" <tr>\n", | |
" <th>\n", | |
" adj.r.squared\n", | |
" </th>\n", | |
" <td>\n", | |
" \n", | |
" <span>FloatVector with 1 elements.</span>\n", | |
" <table>\n", | |
" <tbody>\n", | |
" <tr>\n", | |
" \n", | |
" <td>\n", | |
" 0.661862\n", | |
" </td>\n", | |
" \n", | |
" </tr>\n", | |
" </tbody>\n", | |
" </table>\n", | |
" \n", | |
" </td>\n", | |
" </tr>\n", | |
" \n", | |
" <tr>\n", | |
" <th>\n", | |
" fstatistic\n", | |
" </th>\n", | |
" <td>\n", | |
" \n", | |
" <span>FloatVector with 3 elements.</span>\n", | |
" <table>\n", | |
" <tbody>\n", | |
" <tr>\n", | |
" \n", | |
" <td>\n", | |
" 352.348000\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" 2.000000\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" 357.000000\n", | |
" </td>\n", | |
" \n", | |
" </tr>\n", | |
" </tbody>\n", | |
" </table>\n", | |
" \n", | |
" </td>\n", | |
" </tr>\n", | |
" \n", | |
" <tr>\n", | |
" <th>\n", | |
" cov.unscaled\n", | |
" </th>\n", | |
" <td>\n", | |
" \n", | |
" <span>FloatMatrix with 9 elements.</span>\n", | |
" <table>\n", | |
" <tbody>\n", | |
" <tr>\n", | |
" \n", | |
" <td>\n", | |
" 0.011158\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" -0.000046\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" 0.000046\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" ...\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" 0.000046\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" -0.000000\n", | |
" </td>\n", | |
" \n", | |
" <td>\n", | |
" 0.005556\n", | |
" </td>\n", | |
" \n", | |
" </tr>\n", | |
" </tbody>\n", | |
" </table>\n", | |
" \n", | |
" </td>\n", | |
" </tr>\n", | |
" \n", | |
" </tbody>\n", | |
" </table>\n", | |
" " | |
], | |
"text/plain": [ | |
"R object with classes: ('summary.lm',) mapped to:\n", | |
"[LangSexpV..., LangSexpV..., FloatSexp..., FloatSexp..., ..., FloatSexp..., FloatSexp..., FloatSexp..., FloatSexp...]\n", | |
" call: <class 'rpy2.rinterface.LangSexpVector'>\n", | |
" <rpy2.rinterface.LangSexpVector object at 0x7f3fc6b033c8> [RTYPES.LANGSXP]\n", | |
" terms: <class 'rpy2.robjects.Formula'>\n", | |
" R object with classes: ('terms', 'formula') mapped to:\n", | |
" residuals: <class 'rpy2.robjects.vectors.FloatVector'>\n", | |
" R object with classes: ('numeric',) mapped to:\n", | |
"[-0.731945, -2.334801, -1.670841, -5.966881, ..., 10.679609, 9.230002, 7.307211, 3.257572]\n", | |
" coefficients: <class 'rpy2.robjects.vectors.FloatMatrix'>\n", | |
" R object with classes: ('matrix',) mapped to:\n", | |
"[10.702370, 0.002824, -7.786434, 0.416126, ..., -26.517467, 0.000000, 0.158453, 0.000000]\n", | |
"...\n", | |
" sigma: <class 'rpy2.robjects.vectors.FloatVector'>\n", | |
" R object with classes: ('numeric',) mapped to:\n", | |
"[0.663746]\n", | |
" df: <class 'rpy2.robjects.vectors.FloatVector'>\n", | |
" R object with classes: ('numeric',) mapped to:\n", | |
"[0.661862]\n", | |
" r.squared: <class 'rpy2.robjects.vectors.FloatVector'>\n", | |
" R object with classes: ('numeric',) mapped to:\n", | |
"[352.348000, 2.000000, 357.000000]\n", | |
" adj.r.squared: <class 'rpy2.robjects.vectors.FloatMatrix'>\n", | |
" R object with classes: ('matrix',) mapped to:\n", | |
"[0.011158, -0.000046, 0.000046, -0.000046, ..., -0.000000, 0.000046, -0.000000, 0.005556]" | |
] | |
}, | |
"execution_count": 25, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"rsummary(rmodel1)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 26, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"%load_ext rpy2.ipython" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 27, | |
"metadata": { | |
"scrolled": true | |
}, | |
"outputs": [ | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
"\n", | |
"Call:\n", | |
"(function (formula, data, subset, weights, na.action, method = \"qr\", \n", | |
" model = TRUE, x = FALSE, y = FALSE, qr = TRUE, singular.ok = TRUE, \n", | |
" contrasts = NULL, offset, ...) \n", | |
"{\n", | |
" ret.x <- x\n", | |
" ret.y <- y\n", | |
" cl <- match.call()\n", | |
" mf <- match.call(expand.dots = FALSE)\n", | |
" m <- match(c(\"formula\", \"data\", \"subset\", \"weights\", \"na.action\", \n", | |
" \"offset\"), names(mf), 0L)\n", | |
" mf <- mf[c(1L, m)]\n", | |
" mf$drop.unused.levels <- TRUE\n", | |
" mf[[1L]] <- quote(stats::model.frame)\n", | |
" mf <- eval(mf, parent.frame())\n", | |
" if (method == \"model.frame\") \n", | |
" return(mf)\n", | |
" else if (method != \"qr\") \n", | |
" warning(gettextf(\"method = '%s' is not supported. Using 'qr'\", \n", | |
" method), domain = NA)\n", | |
" mt <- attr(mf, \"terms\")\n", | |
" y <- model.response(mf, \"numeric\")\n", | |
" w <- as.vector(model.weights(mf))\n", | |
" if (!is.null(w) && !is.numeric(w)) \n", | |
" stop(\"'weights' must be a numeric vector\")\n", | |
" offset <- model.offset(mf)\n", | |
" mlm <- is.matrix(y)\n", | |
" ny <- if (mlm) \n", | |
" nrow(y)\n", | |
" else length(y)\n", | |
" if (!is.null(offset)) {\n", | |
" if (!mlm) \n", | |
" offset <- as.vector(offset)\n", | |
" if (NROW(offset) != ny) \n", | |
" stop(gettextf(\"number of offsets is %d, should equal %d (number of observations)\", \n", | |
" NROW(offset), ny), domain = NA)\n", | |
" }\n", | |
" if (is.empty.model(mt)) {\n", | |
" x <- NULL\n", | |
" z <- list(coefficients = if (mlm) matrix(NA_real_, 0, \n", | |
" ncol(y)) else numeric(), residuals = y, fitted.values = 0 * \n", | |
" y, weights = w, rank = 0L, df.residual = if (!is.null(w)) sum(w != \n", | |
" 0) else ny)\n", | |
" if (!is.null(offset)) {\n", | |
" z$fitted.values <- offset\n", | |
" z$residuals <- y - offset\n", | |
" }\n", | |
" }\n", | |
" else {\n", | |
" x <- model.matrix(mt, mf, contrasts)\n", | |
" z <- if (is.null(w)) \n", | |
" lm.fit(x, y, offset = offset, singular.ok = singular.ok, \n", | |
" ...)\n", | |
" else lm.wfit(x, y, w, offset = offset, singular.ok = singular.ok, \n", | |
" ...)\n", | |
" }\n", | |
" class(z) <- c(if (mlm) \"mlm\", \"lm\")\n", | |
" z$na.action <- attr(mf, \"na.action\")\n", | |
" z$offset <- offset\n", | |
" z$contrasts <- attr(x, \"contrasts\")\n", | |
" z$xlevels <- .getXlevels(mt, mf)\n", | |
" z$call <- cl\n", | |
" z$terms <- mt\n", | |
" if (model) \n", | |
" z$model <- mf\n", | |
" if (ret.x) \n", | |
" z$x <- x\n", | |
" if (ret.y) \n", | |
" z$y <- y\n", | |
" if (!qr) \n", | |
" z$qr <- NULL\n", | |
" z\n", | |
"})(formula = temperature ~ month + x2, data = structure(list(\n", | |
" month = 1:360, temperature = c(3.23, 4.48, 9.04, 8.64, 16.79, \n", | |
" 17.34, 20.21, 19.17, 15.94, 11.47, 3.27, 3.56, 3.44, 7.3, \n", | |
" 8.52, 9.42, 16.64, 16.94, 19.76, 20.87, 13.75, 11.36, 5.9, \n", | |
" 1.84, 2.82, -0.55, 8.49, 9.33, 12.16, 15.98, 22.25, 21.84, \n", | |
" 17.67, 9.51, 4.97, 1.33, 2, 3.48, 7.17, 10.29, 17.11, 18.67, \n", | |
" 20.61, 21.55, 15.43, 8.22, 7.21, 2.7, 4.51, 0.47, 5.86, 13.09, \n", | |
" 16.69, 18.48, 18.89, 19.44, 14.48, 9.3, 2, 5.26, 4.4, 2.67, \n", | |
" 9.08, 9.97, 15.04, 18.92, 23.65, 19.85, 15.38, 9.37, 8.91, \n", | |
" 5.49, 2.08, 6.59, 5.34, 10.98, 14.59, 16.52, 22.27, 20.18, \n", | |
" 13.92, 13.05, 4.23, 0.97, -0.2, 0.77, 4.22, 10.68, 13.19, \n", | |
" 18, 18.36, 19.05, 13.03, 10.44, 5.99, -0.27, -2.95, 5.73, \n", | |
" 8.79, 9.1, 14.81, 17.65, 18.95, 21.84, 16.02, 9.33, 5.35, \n", | |
" 3.91, 3.47, 4.53, 7.74, 10.53, 16.3, 18.99, 18.72, 19.85, \n", | |
" 15.17, 10.76, 3.35, 2.6, 4.05, 2.16, 7.38, 11.39, 16.29, \n", | |
" 17.59, 21.58, 19.79, 19.13, 10.4, 4.7, 3.41, 2.83, 5.42, \n", | |
" 7.57, 11.94, 16.6, 19.71, 17.21, 20.38, 16.02, 11.41, 7.47, \n", | |
" 4.75, 2.78, 4.94, 7.46, 9.06, 17.35, 16.69, 21.14, 21.1, \n", | |
" 13.34, 13.86, 4.27, 1.56, 0.75, 6.92, 7.75, 10.61, 14.81, \n", | |
" 19.85, 19.42, 20.26, 14.8, 10.82, 8.17, 4.27, 0.82, 0.61, \n", | |
" 8.6, 11.06, 16.21, 22.64, 21.68, 24.1, 16.25, 8.35, 7.13, \n", | |
" 2.9, 2.05, 4.21, 5.79, 11.8, 13.44, 17.96, 19.6, 20.57, 16.49, \n", | |
" 11.79, 5.54, 1.69, 3.15, 0.76, 6.4, 11.42, 15.07, 19.52, \n", | |
" 20.64, 18.31, 17.54, 12.41, 5.28, 2.14, -0.83, 1.83, 4.53, \n", | |
" 10.78, 15.79, 19.23, 24.43, 16.98, 18.38, 13.45, 8.32, 4.87, \n", | |
" 6.35, 6, 7.66, 14.45, 16.45, 19.27, 19.29, 18.89, 14.07, \n", | |
" 10.02, 4.84, 2.19, 4.98, 4.66, 6.46, 9.6, 17.82, 19.36, 20.23, \n", | |
" 19.26, 13.91, 10.46, 6.3, 1.95, -2.03, 1.99, 6.07, 13.67, \n", | |
" 16.11, 17.68, 19.84, 21.03, 16.6, 10.32, 8.64, 2.46, -1.06, \n", | |
" 2.36, 6.31, 11.18, 12.54, 18.93, 21.83, 18.36, 14.16, 9.28, \n", | |
" 7.06, -1.21, 2.58, 3.62, 7.92, 14.15, 16.6, 18.91, 18.02, \n", | |
" 19.77, 17.18, 10.71, 5.2, 5.4, 3.83, -1.04, 8.97, 10.25, \n", | |
" 16.69, 18.03, 19.72, 21.18, 15.4, 9.85, 6.57, 3.89, 2.24, \n", | |
" 0.99, 3.44, 10.54, 12.9, 17.79, 22.36, 19.84, 15.62, 12.13, \n", | |
" 5.82, 4.47, 4.61, 5.81, 8.79, 13.28, 14.43, 18.85, 20.85, \n", | |
" 17.76, 16.53, 13.37, 7.63, 4.31, 3.09, 2.04, 7.01, 10.97, \n", | |
" 15.35, 18.76, 22.83, 21.99, 15.04, 10.08, 8.42, 7.37, 3.96, \n", | |
" 5.34, 6.04, 9.84, 15.15, 18.44, 21.08, 20.25, 18.63, 9.81, \n", | |
" 5.56, 2.29, 0.55, -1.19, 5.4, 9.45, 9.98, 16.22, 20.78, 20.96, \n", | |
" 19.99, 14.43, 11.98, 6.17, 4.04, 6.6, 0.46, 5.22, 14.19, \n", | |
" 17.45, 20.28, 23.09, 22.39, 17.05, 12.28, 7.19), x2 = c(0.866025403784439, \n", | |
" 0.5, 6.12323399573677e-17, -0.5, -0.866025403784439, -1, \n", | |
" -0.866025403784439, -0.5, -1.83697019872103e-16, 0.5, 0.866025403784438, \n", | |
" 1, 0.866025403784439, 0.500000000000001, 1.19434011948696e-15, \n", | |
" -0.499999999999999, -0.866025403784439, -1, -0.866025403784439, \n", | |
" -0.5, -4.28626379701574e-16, 0.499999999999999, 0.866025403784438, \n", | |
" 1, 0.866025403784438, 0.5, 5.51091059616309e-16, -0.499999999999999, \n", | |
" -0.866025403784439, -1, -0.866025403784439, -0.500000000000002, \n", | |
" 1.10280109986921e-15, 0.5, 0.866025403784439, 1, 0.866025403784439, \n", | |
" 0.500000000000002, 2.57237725884603e-15, -0.5, -0.866025403784437, \n", | |
" -1, -0.866025403784439, -0.500000000000002, 8.57871740039736e-16, \n", | |
" 0.499999999999997, 0.866025403784438, 1, 0.866025403784439, \n", | |
" 0.499999999999999, 2.8173066186755e-15, -0.5, -0.866025403784438, \n", | |
" -1, -0.866025403784438, -0.500000000000002, -2.93977129859024e-15, \n", | |
" 0.5, 0.866025403784438, 1, 0.86602540378444, 0.500000000000002, \n", | |
" -4.9047770029553e-16, -0.499999999999997, -0.866025403784438, \n", | |
" -1, -0.86602540378444, -0.499999999999999, -3.18470065841971e-15, \n", | |
" 0.5, 0.86602540378444, 1, 0.866025403784441, 0.500000000000002, \n", | |
" -2.45548340466059e-16, -0.499999999999997, -0.866025403784438, \n", | |
" -1, -0.86602540378444, -0.499999999999999, -3.42963001824918e-15, \n", | |
" 0.499999999999993, 0.86602540378444, 1, 0.866025403784438, \n", | |
" 0.500000000000003, 7.10480837696441e-15, -0.499999999999996, \n", | |
" -0.866025403784438, -1, -0.86602540378444, -0.500000000000006, \n", | |
" -3.67455937807865e-15, 0.499999999999999, 0.86602540378444, \n", | |
" 1, 0.866025403784442, 0.500000000000003, 2.44310379192882e-16, \n", | |
" -0.500000000000002, -0.866025403784438, -1, -0.86602540378444, \n", | |
" -0.5, 3.18593861969288e-15, 0.499999999999999, 0.866025403784436, \n", | |
" 1, 0.866025403784442, 0.499999999999997, 4.89239739022353e-16, \n", | |
" -0.499999999999996, -0.866025403784438, -1, -0.866025403784437, \n", | |
" -0.5, -4.16441809773759e-15, 0.499999999999999, 0.866025403784436, \n", | |
" 1, 0.866025403784438, 0.500000000000003, 7.83959645645283e-15, \n", | |
" -0.499999999999996, -0.866025403784438, -1, -0.866025403784437, \n", | |
" -0.500000000000006, -4.40934745756706e-15, 0.499999999999999, \n", | |
" 0.866025403784439, 1, 0.866025403784442, 0.500000000000003, \n", | |
" 9.79098458681294e-16, -0.500000000000002, -0.866025403784441, \n", | |
" -1, -0.86602540378444, -0.5, -1.17597041749975e-14, 0.500000000000005, \n", | |
" 0.866025403784436, 1, 0.866025403784439, 0.50000000000001, \n", | |
" -5.88139953909024e-15, -0.499999999999995, -0.866025403784437, \n", | |
" -1, -0.866025403784444, -0.500000000000007, -4.899206177226e-15, \n", | |
" 0.499999999999998, 0.866025403784439, 1, 0.866025403784442, \n", | |
" 0.500000000000004, 1.46895717834024e-15, -0.500000000000001, \n", | |
" -0.866025403784434, -1, -0.866025403784441, -0.500000000000013, \n", | |
" -1.22495628946565e-14, 0.500000000000004, 0.866025403784443, \n", | |
" 1, 0.866025403784439, 0.499999999999998, 8.81931389577071e-15, \n", | |
" -0.499999999999995, -0.866025403784437, -1, -0.866025403784444, \n", | |
" -0.500000000000007, 8.82178981831706e-15, 0.499999999999998, \n", | |
" 0.866025403784439, 1, 0.866025403784443, 0.500000000000004, \n", | |
" 1.95881589799918e-15, -0.499999999999989, -0.866025403784433, \n", | |
" -1, -0.866025403784434, -0.500000000000001, 1.47143310088659e-15, \n", | |
" 0.500000000000004, 0.866025403784435, 1, 0.866025403784439, \n", | |
" 0.500000000000011, 9.30917261542965e-15, -0.499999999999995, \n", | |
" -0.86602540378443, -1, -0.866025403784437, -0.499999999999995, \n", | |
" -5.87892361654388e-15, 0.499999999999997, 0.866025403784439, \n", | |
" 1, 0.866025403784443, 0.500000000000005, 1.66595293328601e-14, \n", | |
" -0.5, -0.86602540378444, -1, -0.866025403784441, -0.500000000000002, \n", | |
" 9.81574381227648e-16, 0.499999999999991, 0.866025403784435, \n", | |
" 1, 0.866025403784446, 0.500000000000011, -4.41182338011341e-15, \n", | |
" -0.500000000000006, -0.866025403784437, -1, -0.866025403784438, \n", | |
" -0.500000000000008, -6.36878233620282e-15, 0.499999999999997, \n", | |
" 0.866025403784431, 1, 0.866025403784443, 0.499999999999993, \n", | |
" 2.93853333731706e-15, -0.5, -0.86602540378444, -1, -0.866025403784441, \n", | |
" -0.500000000000002, -1.37191390536333e-14, 0.499999999999991, \n", | |
" 0.866025403784435, 1, 0.86602540378444, 0.499999999999999, \n", | |
" -3.92196466045447e-15, -0.499999999999994, -0.866025403784436, \n", | |
" -1, -0.866025403784438, -0.500000000000008, -2.10694957710638e-14, \n", | |
" 0.499999999999997, 0.866025403784445, 1, 0.866025403784443, \n", | |
" 0.499999999999993, 3.428392056976e-15, -0.499999999999987, \n", | |
" -0.86602540378444, -1, -0.866025403784449, -0.500000000000003, \n", | |
" -1.42089977732922e-14, 0.500000000000003, 0.866025403784434, \n", | |
" 1, 0.86602540378444, 0.500000000000012, -3.43210594079553e-15, \n", | |
" -0.499999999999993, -0.866025403784429, -1, -0.866025403784445, \n", | |
" -0.499999999999997, -7.34849977552071e-15, 0.500000000000008, \n", | |
" 0.866025403784438, 1, 0.866025403784436, 0.500000000000006, \n", | |
" 1.81291054918369e-14, -0.499999999999999, -0.866025403784432, \n", | |
" -1, -0.866025403784442, -0.499999999999991, -4.88001777749176e-16, \n", | |
" 0.49999999999999, 0.866025403784441, 1, 0.866025403784447, \n", | |
" 0.5, 1.12686074940654e-14, -0.49999999999998, -0.866025403784436, \n", | |
" -1, -0.866025403784438, -0.500000000000009, 6.37249622002235e-15, \n", | |
" 0.499999999999996, 0.866025403784431, 1, 0.866025403784444, \n", | |
" 0.500000000000019, 4.40810949629388e-15, -0.499999999999986, \n", | |
" -0.866025403784439, -1, -0.866025403784435, -0.500000000000003, \n", | |
" -1.51887152126101e-14, 0.500000000000002, 0.866025403784434, \n", | |
" 1, 0.86602540378444, 0.500000000000013, -2.45238850147765e-15, \n", | |
" -0.499999999999992, -0.866025403784443, -1, -0.866025403784446, \n", | |
" -0.499999999999997, -8.32821721483859e-15, 0.499999999999983, \n", | |
" 0.866025403784437, 1, 0.866025403784451, 0.500000000000007, \n", | |
" -9.31288649924918e-15, -0.499999999999974, -0.866025403784432, \n", | |
" -1, -0.866025403784442, -0.499999999999992, -1.46771921706706e-15, \n", | |
" 0.500000000000014, 0.866025403784427, 1, 0.866025403784448, \n", | |
" 0.500000000000001, 1.22483249333833e-14, -0.500000000000004, \n", | |
" -0.866025403784421, -1, -0.866025403784439, -0.50000000000001, \n", | |
" 5.39277878070447e-15, 0.499999999999995, 0.866025403784444, \n", | |
" 1, 0.866025403784444, 0.50000000000002, 5.38782693561177e-15, \n", | |
" -0.499999999999986, -0.866025403784439, -1, -0.86602540378445, \n", | |
" -0.500000000000004, -1.6168432651928e-14, 0.500000000000001, \n", | |
" 0.866025403784433, 1)), class = \"data.frame\", row.names = c(\"0\", \n", | |
"\"1\", \"2\", \"3\", \"4\", \"5\", \"6\", \"7\", \"8\", \"9\", \"10\", \"11\", \"12\", \n", | |
"\"13\", \"14\", \"15\", \"16\", \"17\", \"18\", \"19\", \"20\", \"21\", \"22\", \"23\", \n", | |
"\"24\", \"25\", \"26\", \"27\", \"28\", \"29\", \"30\", \"31\", \"32\", \"33\", \"34\", \n", | |
"\"35\", \"36\", \"37\", \"38\", \"39\", \"40\", \"41\", \"42\", \"43\", \"44\", \"45\", \n", | |
"\"46\", \"47\", \"48\", \"49\", \"50\", \"51\", \"52\", \"53\", \"54\", \"55\", \"56\", \n", | |
"\"57\", \"58\", \"59\", \"60\", \"61\", \"62\", \"63\", \"64\", \"65\", \"66\", \"67\", \n", | |
"\"68\", \"69\", \"70\", \"71\", \"72\", \"73\", \"74\", \"75\", \"76\", \"77\", \"78\", \n", | |
"\"79\", \"80\", \"81\", \"82\", \"83\", \"84\", \"85\", \"86\", \"87\", \"88\", \"89\", \n", | |
"\"90\", \"91\", \"92\", \"93\", \"94\", \"95\", \"96\", \"97\", \"98\", \"99\", \"100\", \n", | |
"\"101\", \"102\", \"103\", \"104\", \"105\", \"106\", \"107\", \"108\", \"109\", \n", | |
"\"110\", \"111\", \"112\", \"113\", \"114\", \"115\", \"116\", \"117\", \"118\", \n", | |
"\"119\", \"120\", \"121\", \"122\", \"123\", \"124\", \"125\", \"126\", \"127\", \n", | |
"\"128\", \"129\", \"130\", \"131\", \"132\", \"133\", \"134\", \"135\", \"136\", \n", | |
"\"137\", \"138\", \"139\", \"140\", \"141\", \"142\", \"143\", \"144\", \"145\", \n", | |
"\"146\", \"147\", \"148\", \"149\", \"150\", \"151\", \"152\", \"153\", \"154\", \n", | |
"\"155\", \"156\", \"157\", \"158\", \"159\", \"160\", \"161\", \"162\", \"163\", \n", | |
"\"164\", \"165\", \"166\", \"167\", \"168\", \"169\", \"170\", \"171\", \"172\", \n", | |
"\"173\", \"174\", \"175\", \"176\", \"177\", \"178\", \"179\", \"180\", \"181\", \n", | |
"\"182\", \"183\", \"184\", \"185\", \"186\", \"187\", \"188\", \"189\", \"190\", \n", | |
"\"191\", \"192\", \"193\", \"194\", \"195\", \"196\", \"197\", \"198\", \"199\", \n", | |
"\"200\", \"201\", \"202\", \"203\", \"204\", \"205\", \"206\", \"207\", \"208\", \n", | |
"\"209\", \"210\", \"211\", \"212\", \"213\", \"214\", \"215\", \"216\", \"217\", \n", | |
"\"218\", \"219\", \"220\", \"221\", \"222\", \"223\", \"224\", \"225\", \"226\", \n", | |
"\"227\", \"228\", \"229\", \"230\", \"231\", \"232\", \"233\", \"234\", \"235\", \n", | |
"\"236\", \"237\", \"238\", \"239\", \"240\", \"241\", \"242\", \"243\", \"244\", \n", | |
"\"245\", \"246\", \"247\", \"248\", \"249\", \"250\", \"251\", \"252\", \"253\", \n", | |
"\"254\", \"255\", \"256\", \"257\", \"258\", \"259\", \"260\", \"261\", \"262\", \n", | |
"\"263\", \"264\", \"265\", \"266\", \"267\", \"268\", \"269\", \"270\", \"271\", \n", | |
"\"272\", \"273\", \"274\", \"275\", \"276\", \"277\", \"278\", \"279\", \"280\", \n", | |
"\"281\", \"282\", \"283\", \"284\", \"285\", \"286\", \"287\", \"288\", \"289\", \n", | |
"\"290\", \"291\", \"292\", \"293\", \"294\", \"295\", \"296\", \"297\", \"298\", \n", | |
"\"299\", \"300\", \"301\", \"302\", \"303\", \"304\", \"305\", \"306\", \"307\", \n", | |
"\"308\", \"309\", \"310\", \"311\", \"312\", \"313\", \"314\", \"315\", \"316\", \n", | |
"\"317\", \"318\", \"319\", \"320\", \"321\", \"322\", \"323\", \"324\", \"325\", \n", | |
"\"326\", \"327\", \"328\", \"329\", \"330\", \"331\", \"332\", \"333\", \"334\", \n", | |
"\"335\", \"336\", \"337\", \"338\", \"339\", \"340\", \"341\", \"342\", \"343\", \n", | |
"\"344\", \"345\", \"346\", \"347\", \"348\", \"349\", \"350\", \"351\", \"352\", \n" | |
] | |
}, | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
"\"353\", \"354\", \"355\", \"356\", \"357\", \"358\", \"359\")))\n", | |
"\n", | |
"Residuals:\n", | |
" Min 1Q Median 3Q Max \n", | |
"-11.2334 -2.7756 -0.2854 3.0335 10.6796 \n", | |
"\n", | |
"Coefficients:\n", | |
" Estimate Std. Error t value Pr(>|t|) \n", | |
"(Intercept) 10.702370 0.416126 25.719 <2e-16 ***\n", | |
"month 0.002824 0.001998 1.413 0.158 \n", | |
"x2 -7.786434 0.293634 -26.517 <2e-16 ***\n", | |
"---\n", | |
"Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1\n", | |
"\n", | |
"Residual standard error: 3.939 on 357 degrees of freedom\n", | |
"Multiple R-squared: 0.6637,\tAdjusted R-squared: 0.6619 \n", | |
"F-statistic: 352.3 on 2 and 357 DF, p-value: < 2.2e-16\n", | |
"\n" | |
] | |
} | |
], | |
"source": [ | |
"%%R -i rmodel1\n", | |
"\n", | |
"summary(rmodel1)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 28, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"[<matplotlib.lines.Line2D at 0x7f3fc3bf32b0>]" | |
] | |
}, | |
"execution_count": 28, | |
"metadata": {}, | |
"output_type": "execute_result" | |
}, | |
{ | |
"data": { | |
"image/png": "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\n", | |
"text/plain": [ | |
"<Figure size 2560x640 with 1 Axes>" | |
] | |
}, | |
"metadata": {}, | |
"output_type": "display_data" | |
} | |
], | |
"source": [ | |
"fig = plt.figure(figsize=(32,8), dpi=80, facecolor='w', edgecolor='k')\n", | |
"ax = plt.subplot(1,1,1)\n", | |
"ax.plot(df['month'], coeffdf['(Intercept)'].values[0] + coeffdf['month'].values[0] * df['month'] + coeffdf['x2'].values[0] * df['x2'])\n", | |
"ax.plot(df['month'], df['temperature'], 'or-')" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 29, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"rmodel2 = rlm(ro.r('temperature ~ x2'), rdf)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 30, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
"\n", | |
"Call:\n", | |
"(function (formula, data, subset, weights, na.action, method = \"qr\", \n", | |
" model = TRUE, x = FALSE, y = FALSE, qr = TRUE, singular.ok = TRUE, \n", | |
" contrasts = NULL, offset, ...) \n", | |
"{\n", | |
" ret.x <- x\n", | |
" ret.y <- y\n", | |
" cl <- match.call()\n", | |
" mf <- match.call(expand.dots = FALSE)\n", | |
" m <- match(c(\"formula\", \"data\", \"subset\", \"weights\", \"na.action\", \n", | |
" \"offset\"), names(mf), 0L)\n", | |
" mf <- mf[c(1L, m)]\n", | |
" mf$drop.unused.levels <- TRUE\n", | |
" mf[[1L]] <- quote(stats::model.frame)\n", | |
" mf <- eval(mf, parent.frame())\n", | |
" if (method == \"model.frame\") \n", | |
" return(mf)\n", | |
" else if (method != \"qr\") \n", | |
" warning(gettextf(\"method = '%s' is not supported. Using 'qr'\", \n", | |
" method), domain = NA)\n", | |
" mt <- attr(mf, \"terms\")\n", | |
" y <- model.response(mf, \"numeric\")\n", | |
" w <- as.vector(model.weights(mf))\n", | |
" if (!is.null(w) && !is.numeric(w)) \n", | |
" stop(\"'weights' must be a numeric vector\")\n", | |
" offset <- model.offset(mf)\n", | |
" mlm <- is.matrix(y)\n", | |
" ny <- if (mlm) \n", | |
" nrow(y)\n", | |
" else length(y)\n", | |
" if (!is.null(offset)) {\n", | |
" if (!mlm) \n", | |
" offset <- as.vector(offset)\n", | |
" if (NROW(offset) != ny) \n", | |
" stop(gettextf(\"number of offsets is %d, should equal %d (number of observations)\", \n", | |
" NROW(offset), ny), domain = NA)\n", | |
" }\n", | |
" if (is.empty.model(mt)) {\n", | |
" x <- NULL\n", | |
" z <- list(coefficients = if (mlm) matrix(NA_real_, 0, \n", | |
" ncol(y)) else numeric(), residuals = y, fitted.values = 0 * \n", | |
" y, weights = w, rank = 0L, df.residual = if (!is.null(w)) sum(w != \n", | |
" 0) else ny)\n", | |
" if (!is.null(offset)) {\n", | |
" z$fitted.values <- offset\n", | |
" z$residuals <- y - offset\n", | |
" }\n", | |
" }\n", | |
" else {\n", | |
" x <- model.matrix(mt, mf, contrasts)\n", | |
" z <- if (is.null(w)) \n", | |
" lm.fit(x, y, offset = offset, singular.ok = singular.ok, \n", | |
" ...)\n", | |
" else lm.wfit(x, y, w, offset = offset, singular.ok = singular.ok, \n", | |
" ...)\n", | |
" }\n", | |
" class(z) <- c(if (mlm) \"mlm\", \"lm\")\n", | |
" z$na.action <- attr(mf, \"na.action\")\n", | |
" z$offset <- offset\n", | |
" z$contrasts <- attr(x, \"contrasts\")\n", | |
" z$xlevels <- .getXlevels(mt, mf)\n", | |
" z$call <- cl\n", | |
" z$terms <- mt\n", | |
" if (model) \n", | |
" z$model <- mf\n", | |
" if (ret.x) \n", | |
" z$x <- x\n", | |
" if (ret.y) \n", | |
" z$y <- y\n", | |
" if (!qr) \n", | |
" z$qr <- NULL\n", | |
" z\n", | |
"})(formula = temperature ~ x2, data = structure(list(month = 1:360, \n", | |
" temperature = c(3.23, 4.48, 9.04, 8.64, 16.79, 17.34, 20.21, \n", | |
" 19.17, 15.94, 11.47, 3.27, 3.56, 3.44, 7.3, 8.52, 9.42, 16.64, \n", | |
" 16.94, 19.76, 20.87, 13.75, 11.36, 5.9, 1.84, 2.82, -0.55, \n", | |
" 8.49, 9.33, 12.16, 15.98, 22.25, 21.84, 17.67, 9.51, 4.97, \n", | |
" 1.33, 2, 3.48, 7.17, 10.29, 17.11, 18.67, 20.61, 21.55, 15.43, \n", | |
" 8.22, 7.21, 2.7, 4.51, 0.47, 5.86, 13.09, 16.69, 18.48, 18.89, \n", | |
" 19.44, 14.48, 9.3, 2, 5.26, 4.4, 2.67, 9.08, 9.97, 15.04, \n", | |
" 18.92, 23.65, 19.85, 15.38, 9.37, 8.91, 5.49, 2.08, 6.59, \n", | |
" 5.34, 10.98, 14.59, 16.52, 22.27, 20.18, 13.92, 13.05, 4.23, \n", | |
" 0.97, -0.2, 0.77, 4.22, 10.68, 13.19, 18, 18.36, 19.05, 13.03, \n", | |
" 10.44, 5.99, -0.27, -2.95, 5.73, 8.79, 9.1, 14.81, 17.65, \n", | |
" 18.95, 21.84, 16.02, 9.33, 5.35, 3.91, 3.47, 4.53, 7.74, \n", | |
" 10.53, 16.3, 18.99, 18.72, 19.85, 15.17, 10.76, 3.35, 2.6, \n", | |
" 4.05, 2.16, 7.38, 11.39, 16.29, 17.59, 21.58, 19.79, 19.13, \n", | |
" 10.4, 4.7, 3.41, 2.83, 5.42, 7.57, 11.94, 16.6, 19.71, 17.21, \n", | |
" 20.38, 16.02, 11.41, 7.47, 4.75, 2.78, 4.94, 7.46, 9.06, \n", | |
" 17.35, 16.69, 21.14, 21.1, 13.34, 13.86, 4.27, 1.56, 0.75, \n", | |
" 6.92, 7.75, 10.61, 14.81, 19.85, 19.42, 20.26, 14.8, 10.82, \n", | |
" 8.17, 4.27, 0.82, 0.61, 8.6, 11.06, 16.21, 22.64, 21.68, \n", | |
" 24.1, 16.25, 8.35, 7.13, 2.9, 2.05, 4.21, 5.79, 11.8, 13.44, \n", | |
" 17.96, 19.6, 20.57, 16.49, 11.79, 5.54, 1.69, 3.15, 0.76, \n", | |
" 6.4, 11.42, 15.07, 19.52, 20.64, 18.31, 17.54, 12.41, 5.28, \n", | |
" 2.14, -0.83, 1.83, 4.53, 10.78, 15.79, 19.23, 24.43, 16.98, \n", | |
" 18.38, 13.45, 8.32, 4.87, 6.35, 6, 7.66, 14.45, 16.45, 19.27, \n", | |
" 19.29, 18.89, 14.07, 10.02, 4.84, 2.19, 4.98, 4.66, 6.46, \n", | |
" 9.6, 17.82, 19.36, 20.23, 19.26, 13.91, 10.46, 6.3, 1.95, \n", | |
" -2.03, 1.99, 6.07, 13.67, 16.11, 17.68, 19.84, 21.03, 16.6, \n", | |
" 10.32, 8.64, 2.46, -1.06, 2.36, 6.31, 11.18, 12.54, 18.93, \n", | |
" 21.83, 18.36, 14.16, 9.28, 7.06, -1.21, 2.58, 3.62, 7.92, \n", | |
" 14.15, 16.6, 18.91, 18.02, 19.77, 17.18, 10.71, 5.2, 5.4, \n", | |
" 3.83, -1.04, 8.97, 10.25, 16.69, 18.03, 19.72, 21.18, 15.4, \n", | |
" 9.85, 6.57, 3.89, 2.24, 0.99, 3.44, 10.54, 12.9, 17.79, 22.36, \n", | |
" 19.84, 15.62, 12.13, 5.82, 4.47, 4.61, 5.81, 8.79, 13.28, \n", | |
" 14.43, 18.85, 20.85, 17.76, 16.53, 13.37, 7.63, 4.31, 3.09, \n", | |
" 2.04, 7.01, 10.97, 15.35, 18.76, 22.83, 21.99, 15.04, 10.08, \n", | |
" 8.42, 7.37, 3.96, 5.34, 6.04, 9.84, 15.15, 18.44, 21.08, \n", | |
" 20.25, 18.63, 9.81, 5.56, 2.29, 0.55, -1.19, 5.4, 9.45, 9.98, \n", | |
" 16.22, 20.78, 20.96, 19.99, 14.43, 11.98, 6.17, 4.04, 6.6, \n", | |
" 0.46, 5.22, 14.19, 17.45, 20.28, 23.09, 22.39, 17.05, 12.28, \n", | |
" 7.19), x2 = c(0.866025403784439, 0.5, 6.12323399573677e-17, \n", | |
" -0.5, -0.866025403784439, -1, -0.866025403784439, -0.5, -1.83697019872103e-16, \n", | |
" 0.5, 0.866025403784438, 1, 0.866025403784439, 0.500000000000001, \n", | |
" 1.19434011948696e-15, -0.499999999999999, -0.866025403784439, \n", | |
" -1, -0.866025403784439, -0.5, -4.28626379701574e-16, 0.499999999999999, \n", | |
" 0.866025403784438, 1, 0.866025403784438, 0.5, 5.51091059616309e-16, \n", | |
" -0.499999999999999, -0.866025403784439, -1, -0.866025403784439, \n", | |
" -0.500000000000002, 1.10280109986921e-15, 0.5, 0.866025403784439, \n", | |
" 1, 0.866025403784439, 0.500000000000002, 2.57237725884603e-15, \n", | |
" -0.5, -0.866025403784437, -1, -0.866025403784439, -0.500000000000002, \n", | |
" 8.57871740039736e-16, 0.499999999999997, 0.866025403784438, \n", | |
" 1, 0.866025403784439, 0.499999999999999, 2.8173066186755e-15, \n", | |
" -0.5, -0.866025403784438, -1, -0.866025403784438, -0.500000000000002, \n", | |
" -2.93977129859024e-15, 0.5, 0.866025403784438, 1, 0.86602540378444, \n", | |
" 0.500000000000002, -4.9047770029553e-16, -0.499999999999997, \n", | |
" -0.866025403784438, -1, -0.86602540378444, -0.499999999999999, \n", | |
" -3.18470065841971e-15, 0.5, 0.86602540378444, 1, 0.866025403784441, \n", | |
" 0.500000000000002, -2.45548340466059e-16, -0.499999999999997, \n", | |
" -0.866025403784438, -1, -0.86602540378444, -0.499999999999999, \n", | |
" -3.42963001824918e-15, 0.499999999999993, 0.86602540378444, \n", | |
" 1, 0.866025403784438, 0.500000000000003, 7.10480837696441e-15, \n", | |
" -0.499999999999996, -0.866025403784438, -1, -0.86602540378444, \n", | |
" -0.500000000000006, -3.67455937807865e-15, 0.499999999999999, \n", | |
" 0.86602540378444, 1, 0.866025403784442, 0.500000000000003, \n", | |
" 2.44310379192882e-16, -0.500000000000002, -0.866025403784438, \n", | |
" -1, -0.86602540378444, -0.5, 3.18593861969288e-15, 0.499999999999999, \n", | |
" 0.866025403784436, 1, 0.866025403784442, 0.499999999999997, \n", | |
" 4.89239739022353e-16, -0.499999999999996, -0.866025403784438, \n", | |
" -1, -0.866025403784437, -0.5, -4.16441809773759e-15, 0.499999999999999, \n", | |
" 0.866025403784436, 1, 0.866025403784438, 0.500000000000003, \n", | |
" 7.83959645645283e-15, -0.499999999999996, -0.866025403784438, \n", | |
" -1, -0.866025403784437, -0.500000000000006, -4.40934745756706e-15, \n", | |
" 0.499999999999999, 0.866025403784439, 1, 0.866025403784442, \n", | |
" 0.500000000000003, 9.79098458681294e-16, -0.500000000000002, \n", | |
" -0.866025403784441, -1, -0.86602540378444, -0.5, -1.17597041749975e-14, \n", | |
" 0.500000000000005, 0.866025403784436, 1, 0.866025403784439, \n", | |
" 0.50000000000001, -5.88139953909024e-15, -0.499999999999995, \n", | |
" -0.866025403784437, -1, -0.866025403784444, -0.500000000000007, \n", | |
" -4.899206177226e-15, 0.499999999999998, 0.866025403784439, \n", | |
" 1, 0.866025403784442, 0.500000000000004, 1.46895717834024e-15, \n", | |
" -0.500000000000001, -0.866025403784434, -1, -0.866025403784441, \n", | |
" -0.500000000000013, -1.22495628946565e-14, 0.500000000000004, \n", | |
" 0.866025403784443, 1, 0.866025403784439, 0.499999999999998, \n", | |
" 8.81931389577071e-15, -0.499999999999995, -0.866025403784437, \n", | |
" -1, -0.866025403784444, -0.500000000000007, 8.82178981831706e-15, \n", | |
" 0.499999999999998, 0.866025403784439, 1, 0.866025403784443, \n", | |
" 0.500000000000004, 1.95881589799918e-15, -0.499999999999989, \n", | |
" -0.866025403784433, -1, -0.866025403784434, -0.500000000000001, \n", | |
" 1.47143310088659e-15, 0.500000000000004, 0.866025403784435, \n", | |
" 1, 0.866025403784439, 0.500000000000011, 9.30917261542965e-15, \n", | |
" -0.499999999999995, -0.86602540378443, -1, -0.866025403784437, \n", | |
" -0.499999999999995, -5.87892361654388e-15, 0.499999999999997, \n", | |
" 0.866025403784439, 1, 0.866025403784443, 0.500000000000005, \n", | |
" 1.66595293328601e-14, -0.5, -0.86602540378444, -1, -0.866025403784441, \n", | |
" -0.500000000000002, 9.81574381227648e-16, 0.499999999999991, \n", | |
" 0.866025403784435, 1, 0.866025403784446, 0.500000000000011, \n", | |
" -4.41182338011341e-15, -0.500000000000006, -0.866025403784437, \n", | |
" -1, -0.866025403784438, -0.500000000000008, -6.36878233620282e-15, \n", | |
" 0.499999999999997, 0.866025403784431, 1, 0.866025403784443, \n", | |
" 0.499999999999993, 2.93853333731706e-15, -0.5, -0.86602540378444, \n", | |
" -1, -0.866025403784441, -0.500000000000002, -1.37191390536333e-14, \n", | |
" 0.499999999999991, 0.866025403784435, 1, 0.86602540378444, \n", | |
" 0.499999999999999, -3.92196466045447e-15, -0.499999999999994, \n", | |
" -0.866025403784436, -1, -0.866025403784438, -0.500000000000008, \n", | |
" -2.10694957710638e-14, 0.499999999999997, 0.866025403784445, \n", | |
" 1, 0.866025403784443, 0.499999999999993, 3.428392056976e-15, \n", | |
" -0.499999999999987, -0.86602540378444, -1, -0.866025403784449, \n", | |
" -0.500000000000003, -1.42089977732922e-14, 0.500000000000003, \n", | |
" 0.866025403784434, 1, 0.86602540378444, 0.500000000000012, \n", | |
" -3.43210594079553e-15, -0.499999999999993, -0.866025403784429, \n", | |
" -1, -0.866025403784445, -0.499999999999997, -7.34849977552071e-15, \n", | |
" 0.500000000000008, 0.866025403784438, 1, 0.866025403784436, \n", | |
" 0.500000000000006, 1.81291054918369e-14, -0.499999999999999, \n", | |
" -0.866025403784432, -1, -0.866025403784442, -0.499999999999991, \n", | |
" -4.88001777749176e-16, 0.49999999999999, 0.866025403784441, \n", | |
" 1, 0.866025403784447, 0.5, 1.12686074940654e-14, -0.49999999999998, \n", | |
" -0.866025403784436, -1, -0.866025403784438, -0.500000000000009, \n", | |
" 6.37249622002235e-15, 0.499999999999996, 0.866025403784431, \n", | |
" 1, 0.866025403784444, 0.500000000000019, 4.40810949629388e-15, \n", | |
" -0.499999999999986, -0.866025403784439, -1, -0.866025403784435, \n", | |
" -0.500000000000003, -1.51887152126101e-14, 0.500000000000002, \n", | |
" 0.866025403784434, 1, 0.86602540378444, 0.500000000000013, \n", | |
" -2.45238850147765e-15, -0.499999999999992, -0.866025403784443, \n", | |
" -1, -0.866025403784446, -0.499999999999997, -8.32821721483859e-15, \n", | |
" 0.499999999999983, 0.866025403784437, 1, 0.866025403784451, \n", | |
" 0.500000000000007, -9.31288649924918e-15, -0.499999999999974, \n", | |
" -0.866025403784432, -1, -0.866025403784442, -0.499999999999992, \n", | |
" -1.46771921706706e-15, 0.500000000000014, 0.866025403784427, \n", | |
" 1, 0.866025403784448, 0.500000000000001, 1.22483249333833e-14, \n", | |
" -0.500000000000004, -0.866025403784421, -1, -0.866025403784439, \n", | |
" -0.50000000000001, 5.39277878070447e-15, 0.499999999999995, \n", | |
" 0.866025403784444, 1, 0.866025403784444, 0.50000000000002, \n", | |
" 5.38782693561177e-15, -0.499999999999986, -0.866025403784439, \n", | |
" -1, -0.86602540378445, -0.500000000000004, -1.6168432651928e-14, \n", | |
" 0.500000000000001, 0.866025403784433, 1)), class = \"data.frame\", row.names = c(\"0\", \n", | |
"\"1\", \"2\", \"3\", \"4\", \"5\", \"6\", \"7\", \"8\", \"9\", \"10\", \"11\", \"12\", \n", | |
"\"13\", \"14\", \"15\", \"16\", \"17\", \"18\", \"19\", \"20\", \"21\", \"22\", \"23\", \n", | |
"\"24\", \"25\", \"26\", \"27\", \"28\", \"29\", \"30\", \"31\", \"32\", \"33\", \"34\", \n", | |
"\"35\", \"36\", \"37\", \"38\", \"39\", \"40\", \"41\", \"42\", \"43\", \"44\", \"45\", \n", | |
"\"46\", \"47\", \"48\", \"49\", \"50\", \"51\", \"52\", \"53\", \"54\", \"55\", \"56\", \n", | |
"\"57\", \"58\", \"59\", \"60\", \"61\", \"62\", \"63\", \"64\", \"65\", \"66\", \"67\", \n", | |
"\"68\", \"69\", \"70\", \"71\", \"72\", \"73\", \"74\", \"75\", \"76\", \"77\", \"78\", \n", | |
"\"79\", \"80\", \"81\", \"82\", \"83\", \"84\", \"85\", \"86\", \"87\", \"88\", \"89\", \n", | |
"\"90\", \"91\", \"92\", \"93\", \"94\", \"95\", \"96\", \"97\", \"98\", \"99\", \"100\", \n", | |
"\"101\", \"102\", \"103\", \"104\", \"105\", \"106\", \"107\", \"108\", \"109\", \n", | |
"\"110\", \"111\", \"112\", \"113\", \"114\", \"115\", \"116\", \"117\", \"118\", \n", | |
"\"119\", \"120\", \"121\", \"122\", \"123\", \"124\", \"125\", \"126\", \"127\", \n", | |
"\"128\", \"129\", \"130\", \"131\", \"132\", \"133\", \"134\", \"135\", \"136\", \n", | |
"\"137\", \"138\", \"139\", \"140\", \"141\", \"142\", \"143\", \"144\", \"145\", \n", | |
"\"146\", \"147\", \"148\", \"149\", \"150\", \"151\", \"152\", \"153\", \"154\", \n", | |
"\"155\", \"156\", \"157\", \"158\", \"159\", \"160\", \"161\", \"162\", \"163\", \n", | |
"\"164\", \"165\", \"166\", \"167\", \"168\", \"169\", \"170\", \"171\", \"172\", \n", | |
"\"173\", \"174\", \"175\", \"176\", \"177\", \"178\", \"179\", \"180\", \"181\", \n", | |
"\"182\", \"183\", \"184\", \"185\", \"186\", \"187\", \"188\", \"189\", \"190\", \n", | |
"\"191\", \"192\", \"193\", \"194\", \"195\", \"196\", \"197\", \"198\", \"199\", \n", | |
"\"200\", \"201\", \"202\", \"203\", \"204\", \"205\", \"206\", \"207\", \"208\", \n", | |
"\"209\", \"210\", \"211\", \"212\", \"213\", \"214\", \"215\", \"216\", \"217\", \n", | |
"\"218\", \"219\", \"220\", \"221\", \"222\", \"223\", \"224\", \"225\", \"226\", \n", | |
"\"227\", \"228\", \"229\", \"230\", \"231\", \"232\", \"233\", \"234\", \"235\", \n", | |
"\"236\", \"237\", \"238\", \"239\", \"240\", \"241\", \"242\", \"243\", \"244\", \n", | |
"\"245\", \"246\", \"247\", \"248\", \"249\", \"250\", \"251\", \"252\", \"253\", \n", | |
"\"254\", \"255\", \"256\", \"257\", \"258\", \"259\", \"260\", \"261\", \"262\", \n", | |
"\"263\", \"264\", \"265\", \"266\", \"267\", \"268\", \"269\", \"270\", \"271\", \n", | |
"\"272\", \"273\", \"274\", \"275\", \"276\", \"277\", \"278\", \"279\", \"280\", \n", | |
"\"281\", \"282\", \"283\", \"284\", \"285\", \"286\", \"287\", \"288\", \"289\", \n", | |
"\"290\", \"291\", \"292\", \"293\", \"294\", \"295\", \"296\", \"297\", \"298\", \n", | |
"\"299\", \"300\", \"301\", \"302\", \"303\", \"304\", \"305\", \"306\", \"307\", \n", | |
"\"308\", \"309\", \"310\", \"311\", \"312\", \"313\", \"314\", \"315\", \"316\", \n", | |
"\"317\", \"318\", \"319\", \"320\", \"321\", \"322\", \"323\", \"324\", \"325\", \n", | |
"\"326\", \"327\", \"328\", \"329\", \"330\", \"331\", \"332\", \"333\", \"334\", \n", | |
"\"335\", \"336\", \"337\", \"338\", \"339\", \"340\", \"341\", \"342\", \"343\", \n", | |
"\"344\", \"345\", \"346\", \"347\", \"348\", \"349\", \"350\", \"351\", \"352\", \n", | |
"\"353\", \"354\", \"355\", \"356\", \"357\", \"358\", \"359\")))\n", | |
"\n", | |
"Residuals:\n", | |
" Min 1Q Median 3Q Max \n", | |
"-10.7520 -2.7934 -0.2384 3.0898 11.1780 \n", | |
"\n", | |
"Coefficients:\n", | |
" Estimate Std. Error t value Pr(>|t|) \n", | |
"(Intercept) 11.2120 0.2079 53.93 <2e-16 ***\n", | |
"x2 -7.7836 0.2940 -26.47 <2e-16 ***\n", | |
"---\n", | |
"Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1\n", | |
"\n", | |
"Residual standard error: 3.945 on 358 degrees of freedom\n", | |
"Multiple R-squared: 0.6619,\tAdjusted R-squared: 0.6609 \n", | |
"F-statistic: 700.7 on 1 and 358 DF, p-value: < 2.2e-16\n", | |
"\n", | |
"\n" | |
] | |
} | |
], | |
"source": [ | |
"print(rsummary(rmodel2))" | |
] | |
}, | |
{ | |
"cell_type": "markdown", | |
"metadata": {}, | |
"source": [ | |
"# lrtest\n", | |
"\n", | |
"* [lrtest](https://www.rdocumentation.org/packages/lmtest/versions/0.9-37/topics/lrtest)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 31, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"rlmtest = importr('lmtest')" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 32, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"rlrtest = ro.r['lrtest']" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 33, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
"Likelihood ratio test\n", | |
"\n", | |
"Model 1: temperature ~ month + x2\n", | |
"Model 2: temperature ~ x2\n", | |
" #Df LogLik Df Chisq Pr(>Chisq)\n", | |
"1 4 -1002.9 \n", | |
"2 3 -1003.9 -1 2.0085 0.1564\n", | |
"\n" | |
] | |
} | |
], | |
"source": [ | |
"rtest_result = rlrtest(rmodel1, rmodel2)\n", | |
"print(rtest_result)" | |
] | |
}, | |
{ | |
"cell_type": "markdown", | |
"metadata": {}, | |
"source": [ | |
"# By Hand" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 34, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"import sklearn.linear_model" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 35, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"(10.702370255288873, array([ 0.00282359, -7.78643352]))" | |
] | |
}, | |
"execution_count": 35, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"lr = sklearn.linear_model.LinearRegression()\n", | |
"X, y = df[['month', 'x2']].values, df['temperature'].values\n", | |
"lr.fit(X, y)\n", | |
"# lr.coef_.squeeze()\n", | |
"lr.intercept_, lr.coef_" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 36, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/html": [ | |
"<div>\n", | |
"<style scoped>\n", | |
" .dataframe tbody tr th:only-of-type {\n", | |
" vertical-align: middle;\n", | |
" }\n", | |
"\n", | |
" .dataframe tbody tr th {\n", | |
" vertical-align: top;\n", | |
" }\n", | |
"\n", | |
" .dataframe thead th {\n", | |
" text-align: right;\n", | |
" }\n", | |
"</style>\n", | |
"<table border=\"1\" class=\"dataframe\">\n", | |
" <thead>\n", | |
" <tr style=\"text-align: right;\">\n", | |
" <th></th>\n", | |
" <th>(Intercept)</th>\n", | |
" <th>month</th>\n", | |
" <th>x2</th>\n", | |
" </tr>\n", | |
" </thead>\n", | |
" <tbody>\n", | |
" <tr>\n", | |
" <th>0</th>\n", | |
" <td>10.70237</td>\n", | |
" <td>0.002824</td>\n", | |
" <td>-7.786434</td>\n", | |
" </tr>\n", | |
" </tbody>\n", | |
"</table>\n", | |
"</div>" | |
], | |
"text/plain": [ | |
" (Intercept) month x2\n", | |
"0 10.70237 0.002824 -7.786434" | |
] | |
}, | |
"execution_count": 36, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"# for comparison:\n", | |
"coeffdf" | |
] | |
}, | |
{ | |
"cell_type": "markdown", | |
"metadata": {}, | |
"source": [ | |
"* [statsmodels-linear-regression](https://datatofish.com/statsmodels-linear-regression/)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 37, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"import statsmodels.api" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 38, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"X, y = df[['month', 'x2']].values, df['temperature'].values\n", | |
"X = statsmodels.api.add_constant(X) # adding a constant" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 39, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"array([[ 1. , 1. , 0.8660254],\n", | |
" [ 1. , 2. , 0.5 ],\n", | |
" [ 1. , 3. , 0. ],\n", | |
" [ 1. , 4. , -0.5 ],\n", | |
" [ 1. , 5. , -0.8660254]])" | |
] | |
}, | |
"execution_count": 39, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"X[:5,:]" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 40, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"X, y = df[['month', 'x2']].values, df['temperature'].values\n", | |
"X = statsmodels.api.add_constant(X) # adding a constant\n", | |
"smmodel = statsmodels.api.OLS(y, X).fit()" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 41, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
" OLS Regression Results \n", | |
"==============================================================================\n", | |
"Dep. Variable: y R-squared: 0.664\n", | |
"Model: OLS Adj. R-squared: 0.662\n", | |
"Method: Least Squares F-statistic: 352.3\n", | |
"Date: Fr, 29 Mai 2020 Prob (F-statistic): 3.24e-85\n", | |
"Time: 07:45:47 Log-Likelihood: -1002.9\n", | |
"No. Observations: 360 AIC: 2012.\n", | |
"Df Residuals: 357 BIC: 2023.\n", | |
"Df Model: 2 \n", | |
"Covariance Type: nonrobust \n", | |
"==============================================================================\n", | |
" coef std err t P>|t| [0.025 0.975]\n", | |
"------------------------------------------------------------------------------\n", | |
"const 10.7024 0.416 25.719 0.000 9.884 11.521\n", | |
"x1 0.0028 0.002 1.413 0.158 -0.001 0.007\n", | |
"x2 -7.7864 0.294 -26.517 0.000 -8.364 -7.209\n", | |
"==============================================================================\n", | |
"Omnibus: 6.982 Durbin-Watson: 0.569\n", | |
"Prob(Omnibus): 0.030 Jarque-Bera (JB): 4.158\n", | |
"Skew: 0.003 Prob(JB): 0.125\n", | |
"Kurtosis: 2.474 Cond. No. 417.\n", | |
"==============================================================================\n", | |
"\n", | |
"Warnings:\n", | |
"[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n" | |
] | |
} | |
], | |
"source": [ | |
"print(smmodel.summary())" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 42, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
" Results: Ordinary least squares\n", | |
"==================================================================\n", | |
"Model: OLS Adj. R-squared: 0.662 \n", | |
"Dependent Variable: y AIC: 2011.7680\n", | |
"Date: 2020-05-29 07:45 BIC: 2023.4264\n", | |
"No. Observations: 360 Log-Likelihood: -1002.9 \n", | |
"Df Model: 2 F-statistic: 352.3 \n", | |
"Df Residuals: 357 Prob (F-statistic): 3.24e-85 \n", | |
"R-squared: 0.664 Scale: 15.519 \n", | |
"--------------------------------------------------------------------\n", | |
" Coef. Std.Err. t P>|t| [0.025 0.975]\n", | |
"--------------------------------------------------------------------\n", | |
"const 10.7024 0.4161 25.7191 0.0000 9.8840 11.5207\n", | |
"x1 0.0028 0.0020 1.4133 0.1585 -0.0011 0.0068\n", | |
"x2 -7.7864 0.2936 -26.5175 0.0000 -8.3639 -7.2090\n", | |
"------------------------------------------------------------------\n", | |
"Omnibus: 6.982 Durbin-Watson: 0.569\n", | |
"Prob(Omnibus): 0.030 Jarque-Bera (JB): 4.158\n", | |
"Skew: 0.003 Prob(JB): 0.125\n", | |
"Kurtosis: 2.474 Condition No.: 417 \n", | |
"==================================================================\n", | |
"\n" | |
] | |
} | |
], | |
"source": [ | |
"print(smmodel.summary2())" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 43, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"array([10.70237026, 0.00282359, -7.78643352])" | |
] | |
}, | |
"execution_count": 43, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"smmodel.params" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 44, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
" OLS Regression Results \n", | |
"==============================================================================\n", | |
"Dep. Variable: y R-squared: 0.662\n", | |
"Model: OLS Adj. R-squared: 0.661\n", | |
"Method: Least Squares F-statistic: 700.7\n", | |
"Date: Fr, 29 Mai 2020 Prob (F-statistic): 2.64e-86\n", | |
"Time: 07:45:47 Log-Likelihood: -1003.9\n", | |
"No. Observations: 360 AIC: 2012.\n", | |
"Df Residuals: 358 BIC: 2020.\n", | |
"Df Model: 1 \n", | |
"Covariance Type: nonrobust \n", | |
"==============================================================================\n", | |
" coef std err t P>|t| [0.025 0.975]\n", | |
"------------------------------------------------------------------------------\n", | |
"const 11.2120 0.208 53.926 0.000 10.803 11.621\n", | |
"x1 -7.7836 0.294 -26.472 0.000 -8.362 -7.205\n", | |
"==============================================================================\n", | |
"Omnibus: 7.806 Durbin-Watson: 0.566\n", | |
"Prob(Omnibus): 0.020 Jarque-Bera (JB): 4.591\n", | |
"Skew: 0.055 Prob(JB): 0.101\n", | |
"Kurtosis: 2.458 Cond. No. 1.41\n", | |
"==============================================================================\n", | |
"\n", | |
"Warnings:\n", | |
"[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n" | |
] | |
} | |
], | |
"source": [ | |
"X, y = df[['x2']].values, df['temperature'].values\n", | |
"X = statsmodels.api.add_constant(X)\n", | |
"smmodel2 = statsmodels.api.OLS(y, X).fit()\n", | |
"print(smmodel2.summary())" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 45, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
"Likelihood ratio test\n", | |
"\n", | |
"Model 1: temperature ~ month + x2\n", | |
"Model 2: temperature ~ x2\n", | |
" #Df LogLik Df Chisq Pr(>Chisq)\n", | |
"1 4 -1002.9 \n", | |
"2 3 -1003.9 -1 2.0085 0.1564\n", | |
"\n" | |
] | |
} | |
], | |
"source": [ | |
"print(rtest_result)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 46, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"2.0" | |
] | |
}, | |
"execution_count": 46, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"lmda = 2 * (-1002.9 - (-1003.9))\n", | |
"lmda" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 47, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"x = np.linspace(0,10,100)\n", | |
"# x" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 48, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"[<matplotlib.lines.Line2D at 0x7f3f88dfe128>]" | |
] | |
}, | |
"execution_count": 48, | |
"metadata": {}, | |
"output_type": "execute_result" | |
}, | |
{ | |
"data": { | |
"image/png": "iVBORw0KGgoAAAANSUhEUgAAB+4AAAIKCAYAAAANn2YcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAMTQAADE0B0s6tTgAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nOzdaZBc530f6l/37CswwGAnAZAACe4UKdIiJdEyqdW2IouyQyflxNeyHEtWqmTrQ1yplFOqUrlSub4fbpzIsn0V+yqJXb6JS7JsyxYlWSsVS6JISuIKggQXEAsBDjAAZp/p5X7onpmeAUCCJIBuAM9T1T7be97zPy3gxXh+fM8pVKvVagAAAAAAAACApig2uwAAAAAAAAAAuJgJ7gEAAAAAAACgiQT3AAAAAAAAANBEgnsAAAAAAAAAaCLBPQAAAAAAAAA0keAeAAAAAAAAAJpIcA8AAAAAAAAATdTe7AJezvHjUymXK80u44I2NNSX0dGJZpcBwEkYowFalzEaoDUZnwFalzEaoHUZo8++trZiBgd7XrZNSwf35XIlpZLg/mwpFGrLcrmSarW5tQCwlDEaoHUZowFak/EZoHUZowFalzG6dXhUPgAAAAAAAAA0keAeAAAAAAAAAJpIcA8AAAAAAAAATSS4BwAAAAAAAIAmEtwDAAAAAAAAQBMJ7gEAAAAAAACgiQT3AAAAAAAAANBEgnsAAAAAAAAAaCLBPQAAAAAAAAA0keAeAAAAAAAAAJpIcA8AAAAAAAAATSS4BwAAAAAAAIAmEtwDAAAAAAAAQBMJ7gEAAAAAAACgiQT3AAAAAAAAANBEpxXc/9mf/Vk+8IEP5LrrrsvHP/7xl217//33573vfW9uvPHG3HPPPXnqqafOSKEAAAAAAAAAcCE6reB+7dq1+ehHP5p77rnnZduNjo7mox/9aH791389P/jBD/L2t789H/3oR1Mqlc5IsQAAAAAAAABwoTmt4P5d73pX3vGOd2RoaOhl2331q1/N1q1b8773vS+dnZ35tV/7tUxMTOQHP/jBGSkWAAAAAAAAAC407Weys127duWqq65a2G5ra8sVV1yRXbt25fbbb39NfRYKZ6o6lpv/bn3HAK3HGA3QuozRAK3J+AzQuozRnEnVajXVJKkm1VRTrc7vr+2sVvPqji+s1xpVF9qeeK35dvPHT9ZfGs8/RZ+L9/Iq2iy5l/krpaGmpec09rPkHhv3L56y5LtavO7iSae8x5PUkobrvtI1Fo401Nd4nVN95yeraUld9Y3qsv1ZVsOSe1048cRrLu1j2TknuW5jfYv9Lb3w0v4azll2XuPx5efV7+hV95dlx3/mrZdn01B3aK4zGtxPTk5mxYoVS/YNDg5mYmLiNfU3NNR3JsriFaxePdDsEgA4BWM0QOsyRgO0JuMzcDGrVuvh10IIWE2lmlQry7Yb2lUaz6nmhO2TtVnoq6Hf+XNzsj5Szd7D07X9y+usLtaVnFjf8vtZWuviOakuXc4HhC/bz8na5CTXbrjn5W0W7rdWSu07yMv02diu3u/i+tJrzvez2Ge9/sqJwfXJal9yTsN9LITElaVBamXhvMU+lwSq9WvMf88LwWbjek7yfZ3Qz9L2lerL97N0P3A2bFgzkBt+9ppml3HRO6PBfW9vb8bHx5fsGxsbS1/fawvgR0cnUi5XzkRpnEShUPt/Zg8fHvMPHkCLMUYDtC5jNEBrMj7Da9MYylYqS9cXQ9xasFapVlOtNKw3hMCN7Zef2xgGNx6vpppKZVkNDddoPL68lqUBdGP9J7Zffv3qKdosuc+ThMon364uBK8n9vtKYfjSYPdUgfXC/rxcfw2hMOe1Qv3/FFJIoTD/lIL6epJCobDkCQaFQiHzDzKYP1aoHyw0tFloX18pzvdRu1iKJ2kzX0tbcfFYrZaGvuvnL687qfd5imPLa2y8xyzr96TH589dVuv8d3SqeznxeOO9NXyPp2i//PjJ72XpvS797pbd45J+l7cpLPTf2O/y72XpvS62WXruifWccP1l39Xi+YUTz1tyzsnvd/FP5im+34VzF/efrG3jdU5VU8O3c8o+Tnavi/299vYLV25su9DmZN/xsmpPcV8nnLeshhOud9L+FitMTlJ3fbtYKGTjhpV+jj7L2tqKrzhp/YwG91deeWX+8i//cmG7Uqlk165d+fCHP/ya+/QH5Oxb/C/eAGg1xmiA1mWMBmhNxucL22LoW025Ugt15/ctLOfXq0m5UguBywv7Gts0hsSL/VUb21VPcY2FYPvEffPXqS4/rzEYr1RTbjjv5MH3Sfps3G6stZqF+2g8b8l6lt+foPfVKBRqwUZjUFoo1sLPxYC1vl1o3C6ceG59u72YJMUUiw19nrSP+Roajy3t98TjDcuF+hr7W9Y2y85ZVmtj/0nDdZdfM8u+o4ZrDgx0Z3xiZuE+F5bz1zuhhqWBdLGQLAlsT3Zulp4/X2tOOLY0aDx5m4YgtvF+srz/ZccbwsdXusbJ9gOca/NDj5+jm++0gvtSqZRyuZxSqZRKpZKZmZkUi8V0dHQsaffOd74zv/d7v5cvfvGLede73pXPfvaz6evry6233npWigcAAACAC9180FouLwbQ5Xr4XC4vBsDlSjXlcmXJ8fn9Jyyrp9i/7Hj1JO0rjaHzSbYX2lVrE3satxcC9Ibl4vGloXp5+b7KhR0yzwekxWKhvqyFfsX5cHh+/3wQutCuFgAX24uL5xVO1raxj4bQt36sWA+Dl5xfrLdLIYVilp5bbOhj2bVOur/h+oUsnn9iuN3YdvF45r+f5efM95PG7WXXPVn/hcbrn3y7MWzn9SkUkuHhgYyMmM0JAKdyWsH9H/7hH+ZTn/rUwva9996bu+++O//xP/7H3HTTTfnMZz6TW265JUNDQ/mDP/iDfPKTn8y/+3f/Ljt27MinP/3ptLef0Yn9AAAAALBEtVrNXKmSmdlySuVKSvOhdj04ng+9y5VKw/orb1fqYXi5Uq33WVkSoi+2qyyE6OVyQ5A+f6yytH25PB/GVxaD8XLDNRv6qZxnKdd84NtWXAyM24q1gLitIWwuFmrb7R3FxX3FpG1hfWnb4rL+5kPnhT4b+jj1ecWlAXb9/BND88Xw/MT9OaFd4YR9i+cVTnZeY0Be3wcAwMWtUK227k/+o6MTKZW84/5s8V85ArQuYzRA6zJGAxey+cePzwfepXK1IQSvBdOl+v7yCftrbefblJe1KdVD6VJ5PkhfFq4vO77YbrG/hX0NYXpjCN5q2oqFtLXVguG2YnEhJJ7/FOv7F9bbCmkvLgbN8wF2W1txIZye73Pp8cKS48Vl11tYFpZtL1tvKyw7drL2r9DG7GTgZPwMDdC6jNHnRnv7OX7HPQAAAACnNj/Deq5UD8Trn7lyNaVSpRaKl2qBdqnUEJzXQ+q5UqV2fmOIvqSvZesNgXdj2F4L2E8M5kvl5v2mrpDUA+hi2ufD7npg3dZWTHuxkM6O9lrQXQ+4l7YppLenM6W58tL9C8F2Q7/FYpaH6ovri9u1R5AvC92XtVseuDcG8QAAAKdLcA8AAABcsKrVWjA9V6pmrlzJXKmcUrkWgNf210LwUqmyuG8+0F7eplwLyefKi20X2tdD9rnyfLC+eH5joF6unLtgvK1YSHtbPayuL9vrgXRXR1vau9sXAvFau3ogXt/XVt/uaAzP66H1Qvv50Lq+vnC94uJ150PuxmON5yzMIK+/e/q1MlMIAAA4nwnuAQAAgLNm/r3jc+VKZucqi6F3PSifK5VPOLbwaWg7Wyov2bckeD9ZCN8wq/1sagyk29uLtZC7rZiOtkJ6OjvS3l4PxuvHFtq2FRvaFxqONRx/mfWFQH1heeJx78wGAAA4fwjuAQAA4CIxH6LP1sPu2VI5c3Pz2+XMzNWWC8fnygvtF4LzhXMrmZurh+6N++v75vs9W8F5x3wQ3rhsb0tvd0c62gq14+1taV9YrwXajcsl57cV095eSEdbWzraFwPwxnbt8wH7fB9tRY9DBwAA4IwQ3AMAAECTlSu1Geezc+XMzNWC8/nt2VK5tr6wXL6/vr1w3tLl3LL1M2k+2O6sB+Md7cX0dXcsrC8ea1u2XVwI1ueD8cZzOtqK6ehYvr9tYbu97fU9Uh0AAABajeAeAAAAXkGpPB+qVzIzV87MbD1grwftM/VjC8H7shB+ZnYxaJ9pCNnn95fKZ+aF3PPBeGdHW329Lf09HQvheWfHYpDe2V4LxzsX9rctCdrn23Z2tC0E6Z0NAXxHe9Gj2AEAAOAMEdwDAABwQahUq5mbq2R6rpyZ2VItZK8H7NOz5czMLe6bni1ldnnb+aB99sQwvlx5fcF6sVBIV2ctBO+qh+L9PR3p7KgF6F2dbYsBekdbuur75493dBTr5y2G7Athe/1YR4cgHQAAAM5XgnsAAADOuWq1mtlSpRaoz5YyPVte+NSC9lL9WLlhWcr0XHnp/rmlAf1rVUjS2VkLzLs6iunqaFsI1rs62tLVWdu/sN2wv7PevvHYfKg+f057W/HMfXkAAADABUdwDwAAwGkplSv1cL0etM8srk81hu8zpaXt6utT9fbzM+Crr3ESe1dHW7rrIXtPZ3tW9LeleyFcry27GwP3k2x3dSw9p6O96J3pAAAAQNMI7gEAAC5wc6VKpmZKtXB9prx0fbaUqXrQPjlTyvTMYsA+VT8+PVPK1Gw5c6XKq752sVBId2dberra0t3ZntWD3bXQvbO9vqwF6I375kP52npDu87arHaPgwcAAAAuNIJ7AACAFlWtVjM7V8nkTC1cn/8sbpdfdv/kTCnTs6WUyq9uantbsZCervZ64N6elX2d6e5qXxKkzx9bvq+7s30hpO/uNJMdAAAA4HQI7gEAAM6SSrWamdlyJqdrIfrk9Fx9WQ/ZF/YvBu+1Y3P1feVUXsXz5IuFQnq6aoF6b1d7hga60tNVD9K72tPTEKr3dNUeM99dP97TcLy9TdgOAAAAcC4J7gEAAF5GuVLJxHQtUJ+YnsvUdCnFPcdy8KWxjE/Xwvj545PTcw0hfe1x9KebuxeS9Ha310L37toj5Wuh++KntyFk7z3Jsc4OgTsAAADA+UhwDwAAXPCq1WqmZ8uZmKqF7OPTcwvrteVcJqZqwXwtoK/Nep+YLmVmtnxa12gM3vu6O7JmZU8tXO+uhepL1heWHQvbXZ3e3Q4AAABwsRLcAwAA5435AH58ai7jU7XwfbwhdF9c1kP5hvVy5ZWnvhcKSV93R3q72zPY25kNq3rT292Rvu72hWVfT3vWrxlMeW4uvV21tn3dtUfOC94BAAAAeC0E9wAAQFNUKtVMzpRqIfzk3EIY3/iZOMm+0wng29sK6evuSH9PR9av6k1fT0f6ujvS19NeX86H8B3p715c7+5se8VHzRcKyfDwQEZGxk77MfgAAAAA8HIE9wAAwBkxM1vO2ORsxqbmMjY5l/Gp2fpyrrZ/ci5jDSH9xNRcXin3LiS1kL2nI8Mru7N1/UD6e2uBfH99f39DEN9fD+i96x0AAACA84ngHgAAOMH8I+mPT85mbGIuxydna+uTtRB+fD6cn5zL2NRsxifnMluqvGyfxUIh/T3tGejtzKbhvqUBfH12fOO+/p7a+9+LRQE8AAAAABc2wT0AAFwk5kqVjDUE8Mcn6svJ2YxNzOb4/PrkbI5PzKVUfvkgvqujLf09HRns7czG4b4M9HRmoLej/ulMf8/S9d5u74AHAAAAgJMR3AMAwHlsdq6c4xOzOTY5m+Pj9eXEbI5N1Jbz62OTs5maKb9sXx3txQz21oL4S9b0Z6C+PtDbmcG+2np/b8dCQN/Z0XaO7hIAAAAALmyCewAAaDGlciXHxmuB+7GJmZMG8ccnajPnXy6MLyTp7+3IYF9ntq4fzGBf50IY37g+0NeZwd6OdHW0eS88AAAAADSB4B4AAM6Rmdlyjk7M5Nj4bI6O15f17WPjMzk6MZtj47X3x59KIUlfT0dW9NfC+BV9tRB++XI+mG8rFs/dDQIAAAAAr4ngHgAAXqeZ2XKOjs9kdGwmo+Mzi6H8QjhfC+anZ089O76roy0r+juzcXVvVvR3ZUV/Z1b2dy0G8b2dWdEvjAcAAACAC5HgHgAATqFcqeT4xNxiKD9WC+WPLgT0sxkdm8nUTOmUffR1t2dFf1cu2zCYlf2dWdHflZV99WV9e0VfZ3q6/GgOAAAAABcrvx0EAOCiNDtXzuj4TI4cn8no2HRtOR/K1wP6YxOzqVZPfn5XZ1uG+ruyZV1/hga6srK/KysHujJUX9bC+c50tLed2xsDAAAAAM47gnsAAC44Jw3lx2Zy5Ph0bTk2c8r3yBcLhazo78yqwe5s27iiHsjXHls/NNC1ENKbIQ8AAAAAnCl+2wgAwHmlXKnk6NhsDh+fzuHj0zlyvBbMn04o31YsZGigKxtX92ZosDurBubD+O6sGuzKqoGuDPR2plgsnOO7AgAAAAAuZoJ7AABayvRsKYePz+TwsVoovxDQH6stR8dmUznJ8+tPK5Tv60yxIJQHAAAAAFqL4B4AgHOmWq1mfGouI8emc/jYdG1ZnzV/uL4+MV066bn9PR1ZPdidzesGsnpFd1YP1j8ruoXyAAAAAMB5TXAPAMAZNTldysixqYwcm87I0fry2HRequ+bmS2fcM78bPlNa/rrYXzXsmC+O12dbU24GwAAAACAs09wDwDAqzIzW87Isam8tCyYHzk2lZGj05mcOXHGfFuxkNWD3dm2cTDDK7ozvKJnYblqsCsr+7u8Vx4AAAAAuGgJ7gEAWKJareb4xGxeOjqdQ0cnc2h0Ki8dncqho1N56eh0jk/MnnBOoZCsGujOpWv7M7xyMZhfs7K2FMwDAAAAAJya4B4A4CJUKldy+Ph0XhqdD+SnFgL6l45OZ2buxMfZD/R2ZO3KnlyzZSjD9UB+zYruDK/sydBAV9rbik24EwAAAACA85/gHgDgAjVXquTQ0akcOjKZg6NTOTi6OHv+8PHpVKtL2xcLhawa7Mq2TYNZu7Ina4Z6asv6p6fLj44AAAAAAGeD374CAJzHypVKRo5N5+CRyRw8UgvnD9aD+pOF810dbVmzsic3XbEma1Z2LwnoVw12mzUPAAAAANAEgnsAgBZXqVZz5Ph0QzC/GNCPHJtOubI0ne9sL2btUG9uvnJN1g31Zt2qntpyqCeDfZ0pFLxrHgAAAACglQjuAQBaxPRsKS8emcyBw7XPi4cncqA+k75Urixp295WyJqVPbn+8tW1YH5V70I4v3KgK0XhPAAAAADAeUNwDwBwDlWr1YyOzeTAkcm8eHgyBw5PLIT1o2MzS9oWkgyv7M7VW4aybqgeztdnz68e7E6xKJwHAAAAALgQCO4BAM6CuVIlB0cXw/kD9XD+xSOTmZktL2nb1dGW9at7s2Pzyqxf1ZsNq/uyoR7Sd7S3NekOAAAAAAA4VwT3AACvw1ypkhePTGbfyHj2j0xm/8hE9o1M5NDoZKpLXz2foYGuXL5hMBtW18L59at7s2FVb4YGurx3HgAAAADgIia4BwA4DXOlSg4emcy+ejB/YCGgn0qlIaEvFgpZO9STm69Yk/Wre7OxHtCvX9Wbni4/egEAAAAAcCK/PQYAaFAq12bQ7x+ZyL6XJrJ/ZCL7D0/k4JGlAX2hkKwd6s0brhjOxuHebBzuy6bh/qxf1ZuO9mIT7wAAAAAAgPON4B4AuChVq9UcHZ/NC4fGs/el8ew9NJ4XXhrPi4cnU64sC+hX9uTG7avr4XxfNg73ZcPqXu+fBwAAAADgjBDcAwAXvJm5cvaPTNRC+npQ/8Kh8UxMl5a0G17RnesvX51Na/oWQvr1q3rT2SGgBwAAAADg7BHcAwAXjEq1msPHphdmz9eWEzl0ZDLVhnZdnW25dE1/Llnbn0vX9OWStf3ZNNyf3m4/GgEAAAAAcO757TQAcF6aK1Wyf2Qizx8cy/MHx/LCwdpM+unZ8kKbQpK1q3rzxh1r6iF9LaxfvaI7xUKhecUDAAAAAEADwT0A0PKmZ0t54dB4nn9xLHsOjmfPwbHsG5lY8i76vu72bF0/kEvmZ9Kv7c/G4b50ecw9AAAAAAAtTnAPALSUscnZhXC+Npt+/IRH3Q8NdOW6y1Zly/qBbF43kM3r+rN6sDsFs+gBAAAAADgPCe4BgKYZHZvJ8y/WAvr5oP7I8ZklbdYO9eSWq9Zm87r+bFlXC+oH+zqbVDEAAAAAAJx5gnsA4JwYn5rLcy8ez7MHxvLcgeN59sDxHB2fXTheLBSycbgvV183lM3rBrJl/UAuXdufni4/rgAAAAAAcGHzm3AA4IybmS3n+YNjebYe0D93YCyHjk4tHC8WCrlkTV9u2DacrRsGsmXdQC5Z05eOdu+jBwAAAADg4iO4BwBel1K5khcOjddn0Y/l2RePZ//IRKoNL6Vfv6o3t1+7Lls3DOayDYPZvLY/nR1CegAAAAAASAT3AMCrUK1Wc2h0Kk/vO1afTT+WFw6NpVReTOlXDXbl5ivWZOuGgVy2YTBb1w+kt7ujiVUDAAAAAEBrE9wDAKc0M1vOsweOZ/f+Y3l677Hs3n8841NzC8f7ezpy9ZZVuWzDwMJs+hV9nU2sGAAAAAAAzj+CewAgSW02/UtHp7J73/E8vf9Ydu87lr2HJlKpP/O+rVjI5nX9ue2addm2aUUu3ziY4RXdKRQKTa4cAAAAAADOb4J7ALhIzcyV89yB43l637Hs3nc8z+w/luOTi7PpB/s6c+P21dm+aUW2bVqRresHvJceAAAAAADOAsE9AFwkjhyfzq69R7N7b21G/d5D4ylXarPpi4VCLl3Xn1uvWpdtmwazbdMKs+kBAAAAAOAcEdwDwAWoWq3m4OhUdr1wdOEzcmx64fhAb0euv3x1tm0azPZNK7J1/WC6Os2mBwAAAACAZhDcA8AFoFKpZu9L44tB/d5jOT4xu3B83are3HHDhlx56cpcccmKrFnZYzY9AAAAAAC0CME9AJyHSuVKnjswll17a0H9U3uPZWqmlCQpJLl0bX9uvWptdtSD+hX9Xc0tGAAAAAAAOCXBPQCcB2Zmy3l6/7E8VZ9Rv3v/8cyVKkmStmIhl20YzJWXrsyVl67I9k0r0tvd0eSKAQAAAACA0yW4B4AWNFcq55GnR/K9h/fl8edG8+yB4ylXqkmSro62XHHJilpQf8nKXL5xMJ0d3k8PAAAAAADnK8E9ALSAcqWS514cy87nR/PE86N5au+xhRn1PV1tue6yVdmxeShXXroym9f1p72t2OSKAQAAAACAM0VwDwBNUKlWs++liTzx/Gh2Pj+aJ18YzdRMOUnS0V7MFZesyC3XrM/m4d5sWT+QtqKgHgAAAAAALlSCewA4B6rVag4dncoTz9Vm1O/cM5qxybkk9XfUbxzM1ZuHcvWWoWzbVHv0/fDwQEZGxlKtNrl4AAAAAADgrBLcA8BZcnR8Jo8/d6QW1u8ZzZHjM0mSQpJL1/XnLddtyFVbhnLFJSvS0+WfZAAAAAAAuFhJCQDgDCmVK3lq77E8+uzhPPrMkbxwaHzh2IbVvbnz5k25ZstQdmweSn9PRxMrBQAAAAAAWongHgBeh0NHp/LoM7Wg/ok9o5mZrb2nfqC3I7dduy7XXbYqV29ZlaGBriZXCgAAAAAAtCrBPQC8CjOz5ezcM5pHnzmSR549nEOjU0mSYqGQ7ZsGc+3lq3P95auyed1AioVCk6sFAAAAAADOB4J7AHgZ1Wo1+16ayCP1x98/tfdoSuVqkmT1YFfe9oaNC7Pqe7v9swoAAAAAALx6EgYAWGZiei6PPXskjz5zJI8+ezhHx2eTJB3txVy1eSjXXb461122KhtW96ZgVj0AAAAAAPA6Ce4BIMnB0cn8+KmR/Ojpkex64Vgq1dqs+g2re3PrVety/eWrcuWlK9PZ0dbkSgEAAAAAgAuN4B6Ai1KlUs3u/cfyo3pYf+DwZJLarPrrL1+VG7cP57rLV2V4RU+TKwUAAAAAAC50gnsALhpTM6U89uyR/Pjpkfx49+GMT80lSVb0deYnb9yQG7cP55qtq9JlVj0AAAAAAHAOCe4BuKAdOT6dHz09kh89NZKde0ZTKtcegX/Jmv781E2b8obtw9m6YSBF76oHAAAAAACaRHAPwAWlUq3m+RfH8qOnRvLjp0ey59B4kqStWMhVW4byhu3DuXH7ao/ABwAAAAAAWobgHoDzXrlSyc49R/PgzkP54dMjOTY+myTp7+nI7deuz01XDOfay1alp8s/ewAAAAAAQOuRYABwXiqVK3ni+dE8sPNQfvjUyML76tet6s3tb1qfN2wfzrZNg2krFptcKQAAAAAAwMs77eD++PHj+ff//t/n29/+dvr7+/ORj3wkv/RLv3TStn//93+fT33qUzlw4EDWrl2b3/iN38j73//+M1Y0ABenuVI5jz07mgefrIX1kzOlJLX31b/jlkvyxh1rs2m4r8lVAgAAAAAAvDqnHdx/8pOfTLlczn333Zc9e/bkgx/8YLZt25bbbrttSbsDBw7kt3/7t/Of//N/zp133pkHH3wwH/rQh3Lddddl+/btZ/wGALiwzc6V88gzR/Lgk4fyo6dHMj1bTpJsWTeQW65akzfuWJv1q3qbXCUAAAAAAMBrd1rB/eTkZO6999584QtfSH9/f6655prcfffd+dznPndCcL9///4MDAzkrrvuSpLccsst2bx5c55++mnBPQCnZWa2nIefOZwHdh7Kw7sPZ2auFtZfvnEwt+xYmzfuWJM1K3uaXCUAAAAAAMCZcVrB/XPPPZckS4L3q666Kp/97GdPaHvjjTdm69at+epXv5q3v/3tuf/++zMyMpKbb775NRVYKLym0zgN89+t7xhoBVMzpfz46ZE88ORLeWT34cyWKikk2X7JioWwfvWK7maXec4YowFalzEaoDUZnwFalzEaoHUZo1vHac+47+tb+s7gwcHBTExMnNhhe3s+8IEP5Ld/+7czMzOTYrGY3/3d383atWtfdXFDQ95TfC6sXj3Q7BKAi9T0bCk/eOxgvvXDvXnoyUOZK1VSLCTXXj6ct9ywIbddvyGrV1zcM+uN0QCtyxgN0JqMzwCtyyHjjnYAACAASURBVBgN0LqM0c13WsF9b2/vCSH92NjYCWF+ktx33335vd/7vfzpn/5pbrzxxuzevTsf/vCHs3LlyvzUT/3UqypudHQi5XLlVZ3D6SsUan8JDx8eS7Xa7GqAi0W5UskTz43mu48dzEO7Xsr0bDmFQnL1lqHcctXa3Hzlmqzo60ySVOdKGRkZa3LFzWGMBmhdxmiA1mR8BmhdxmiA1mWMPjfa2oqvOGn9tIL7rVu3Jkl2796dbdu2JUl27tyZK6644oS2u3btys0335ybbropSXLFFVfkbW97W7797W+/6uA+iT8g50C16nsGzq5qtZrnXhzLdx97Mfc/cSjHJ2aTJJdtGMxt167LT1y9biGsr7VvVqWtxxgN0LqM0QCtyfgM0LqM0QCtyxjdfKc94/7d7353fv/3fz//4T/8h+zduzef//zn85/+0386oe0NN9yQP/7jP87DDz+cG264Ic8880y+9a1v5SMf+cgZLx6A1nZwdDLfe+xgvvfYizk4OpUkWTfUkztvuiy3XbMu61b1NrlCAAAAAACA5jut4D5JPvGJT+R3fud3cscdd6Svry8f+9jHcvvttydJbrrppnzmM5/JLbfckltvvTUf//jH82/+zb/JoUOHsmLFirzvfe/LL/zCL5y1mwCgdRybmM39TxzM9x47mGcPHE+SDPZ15h23XJLbr12fresHUigUmlwlAAAAAABA6yhUq6370IPR0YmUSt5xf7YUCsnw8EBGRryzAnh9pmZK+eFTL+V7jx3M48+NplKtpquzLW+8ck1uu3Zdrt4ylLZisdllnleM0QCtyxgN0JqMzwCtyxgN0LqM0edGe/sZesc9ACxXrlTy6DNH8t3HXsyPnhrJbKmStmIhN2xbnduuXZcbtw+nq6Ot2WUCAAAAAAC0PME9AK/Ki0cmc9/D+/OPj7yYYxOzSZLtl6zI7desyy1Xrc1Ab2eTKwQAAAAAADi/CO4BeEUzs+U88OSh3Pfj/dm191iSZHhFd95/x2V587XrM7yyp8kVAgAAAAAAnL8E9wCcVLVazbMHxnLfw/vz/ccPZnq2nPa2Ym67Zl3uuGFDdmwZSrFQaHaZAAAAAAAA5z3BPQBLjE3O5ruPHcx9D+/PvpcmkiSb1/Xnjhs25rZr16Wvu6PJFQIAAAAAAFxYBPcApFKp5vHnjuTbDx/ID3e9lHKlmt6u9tx186bcccPGbFk/0OwSAQAAAAAALliCe4CL2MjRqXznkQP5ziMHcuT4TJLk6i1DueOGDbn5yjXp7GhrcoUAAAAAAAAXPsE9wEVmrlTJQ7teyn0P788Tz42mmmRooCvvffPWvPWGDVm7sqfZJQIAAAAAAFxUBPcAF4kjx6fzjR/uy7d/vD9jk3NpKxZy8441ueOGjbnuslUpFgvNLhEAAAAAAOCiJLgHuIBVq9Xs3HM0X39wb3741Egq1WqGV3TnPXduzluu25DBvs5mlwgAAAAAAHDRE9wDXICmZ0v57qMv5msP7cv+kYkkyXWXr8rbb74k11++2ux6AAAAAACAFiK4B7iAHDg8kW88tC//+9EDmZopp6erPe+85dLcdfOmrFvV2+zyAAAAAAAAOAnBPcB5rlKp5uHdh/O1h/bmsWePJEk2renL22++JLdduy7dnYZ6AAAAAACAVibNAThPjU/N5b6H9+cbD+3LyLHpFAuF3LJjTd7+xkty5aUrUyh4HD4AAAAAAMD5QHAPcJ55/sWxfO2hvfn+4wczV6pksLcj/+TNW/O2N2zMqsHuZpcHAAAAAADAqyS4BzgPlCuVPPjkS/mHB/bm6X3HkiTbNg3m7TdfkjfuWJuO9mKTKwQAAAAAAOC1EtwDtLCZuXK+8/CBfPn+PRk5Np32tmLeev2G3PXGTdm6frDZ5QEAAAAAAHAGCO4BWtDY5Gy+/tC+fO3BvRmfmkt/T0d+7q2X5c6bN2Wwt7PZ5QEAAAAAAHAGCe4BWshLR6fylftfyH0P789sqZLhFd35ubdelrfesCFdHW3NLg8AAAAAAICzQHAP0AKef3EsX/r+8/nBzkOpVpMt6wby07dtzht3rElb0fvrAQAAAAAALmSCe4AmqVarefy50Xzp+8/n8edGkyTXXrYqP/2mzbl6y1AKhUKTKwQAAAAAAOBcENwDnGPlSiU/2Hko935vT/YcGk+xUMht16zLe960OZvXDTS7PAAAAAAAAM4xwT3AOTIzW859D+/PV37wQkaOTaezo5h3vPGSvOvWSzO8sqfZ5QEAAAAAANAkgnuAs+z45Gy+/uDefO3BvZmYLqW/pyPvv+Oy3HXzJenv6Wh2eQAAAAAAADSZ4B7gLDk+OZt7v78nX39ob2bnKlmzsjsf+MnL85brN6Szo63Z5QEAAAAAANAiBPcAZ9j41Fzu/f6efO3BvZmZK+eSNX1575u35pYda1MsFppdHgAAAAAAAC1GcA9whoxPzeXL9+/JPzy4NzOz5Wwa7svPvfWy3LxjTYoFgT0AAAAAAAAnJ7gHeJ0mpufylftfyFcfeCHTs+VsWN2bn3vrZbnlqrUCewAAAAAAAF6R4B7gNZqcLuWrD7yQr/zghUzNlLJ+VW/e95at+Ymr13kkPgAAAAAAAKdNcA/wKk3NlPIPD7yQL9//QiZnSlk71JN/8c4r86ZrBPYAAAAAAAC8eoJ7gNM0NVPK1x7cmy/fvycT06WsWdmdf/6OK3LbtevSViw2uzwAAAAAAADOU4J7gFcwPVvK1x/al3u/vyfjU3MZXtGde+7cntuvW5/2NoE9AAAAAAAAr4/gHuAUZubK+cZD+/Kl7z+fscm5rB7syi/81FV5s8AeAAAAAACAM0hwD7BMuVLJfT8+kC9859kcn5jN0EBXfvndO/LWGzYI7AEAAAAAADjjBPcAddVqNY88czj/6xu7s39kIiv6OvNL77wyP3njxnS0C+wBAAAAAAA4OwT3AEn2HBzL//z603ni+dF0thfzvrdszXvetDndnYZJAAAAAAAAzi6JFHBRGx2byee/vTv/+MiLSZK3Xr8hd//k5Rka6GpyZQAAAAAAAFwsBPfARWlqppR7v78nX75/T2ZLlVyzdSj33Lk9m9cNNLs0AAAAAAAALjKCe+CiUq5U8p2HD+Sv7ns2xydms3G4L/fcuT3XX74qhUKh2eUBAAAAAABwERLcAxeFarWaR545kr/8xtPZNzKRwb7O/PJ7duSOGzakrVhsdnkAAAAAAABcxAT3wAVvz8Gx/OU3ns5jz42ms72Y9755a376TZvT02UIBAAAAAAAoPmkVsAFa3RsJn/17Wfyvx85kCR5y/Xrc/cdl2fVYHeTKwMAAAAAAIBFgnvggjM9W8q939+Te+/fk9m5Sq7eMpR77tyeLesHml0aAAAAAAAAnEBwD1wwqtVqvvfYwfyvbzydYxOz2bC6N/fcuT03bFudQqHQ7PIAAAAAAADgpAT3wAXhwOGJ/I8vP5mde46mv6cj//LdO/KTN25IW7HY7NIAAAAAAADgZQnugfPa7Fw5X/zuc/nS9/akXKnmbW/YmJ9/27b093Q0uzQAAAAAAAA4LYJ74Lz18O7D+fOvPpmXjk7nkjX9+eX37Mj2TSuaXRYAAAAAAAC8KoJ74LwzOjaTv/iHXXngyZfS1dGWX7xre95xyyUeiw8AAAAAAMB5SXAPnDfKlUq+9sDe/NV3ns3MbDlvvHJN/vk7rsiqwe5mlwYAAAAAAACvmeAeOC/s3ncs//3LT+aFQ+MZXtGdj7zv2ty4fbjZZQEAAAAAAMDrJrgHWtrE9Fw+983d+daP9qdYLORnb9+S9755a7o62ppdGgAAAAAAAJwRgnugJVWr1Xz3sRfzP7/+dMYm57Lj0pX5l+/ekY3Dfc0uDQAAAAAAAM4owT3Qcg4cnsj/+PKT2bnnaAZ6O/Khn706b75ufQqFQrNLAwAAAAAAgDNOcA+0jNm5cr743efype/tSblSzdvesDE//7Zt6e/paHZpAAAAAAAAcNYI7oGW8NizR/Lf7t2ZkWPTuXRtf3753TuybdOKZpcFAAAAAAAAZ53gHmiqmdly/tc3ns43frgvXR1t+Wd3bc/bb7kkbcVis0sDAAAAAACAc0JwDzTNU3uP5k+++EQOHZ3KjktX5kM/e3WGV/Y0uywAAAAAAAA4pwT3wDk3V6rkC995Jvd+f0/a24r5Z2+/Iu+45ZIUC4VmlwYAAAAAAADnnOAeOKf2HBzLf/3i49n70kQu2zCQD/3sNdk43NfssgAAAAAAAKBpBPfAOVGuVPL339uTv/nOs0mS999xWX729i3eZQ8AAAAAAMBFT3APnHUHDk/kT/7uiTyz/3g2DvflX733mmxZP9DssgAAAAAAAKAlCO6Bs6ZSreZrD+7N5765O3OlSt7zE5tz909elo72tmaXBgAAAAAAAC1DcA+cFYePTedP//6JPPH8aIZXdOfX3ntNrrx0ZbPLAgAAAAAAgJYjuAfOqGq1mv/9yIv5i6/tytRMOW97w8bcc+f29HQZbgAAAAAAAOBkJGnAGXNsYjb/7Us786OnR7KivzMfft91uWHb6maXBQAAAAAAAC1NcA+cEQ/sPJT//uUnMz41lzddsy6/9M4r09/T0eyyAAAAAAAAoOUJ7oHXZXJ6Ln/+1V357mMH09/Tkd94/3W59aq1zS4LAAAAAAAAzhuCe+A1e2b/8fzhFx7J4eMzuXHb6vzKT1+VFf1dzS4LAAAAAAAAziuCe+BVq1ar+fpD+/L/fe2pFIuF/B/v2ZGfvHFjCoVCs0sDAAAAAACA847gHnhVpmdL+eyXdub+Jw5l7cqefPTu67J53UCzywIAAAAAAIDzluAeOG37Riby6b96JAcOT+bmK9fkV3/m6vR2G0YAAAAAAADg9ZC4Aaflu4+9mP92786UStXcc+f2vPsnLvVofAAAAAAAADgDBPfAy5orlfMXX3s63/zhvqzs78xH7rkuV166stllAQAAAAAAwAVDcA+c0sjRqfzBFx7N8y+O5eotQ/nw+67NYF9ns8sCAAAAAACAC4rgHjipHz09kj/54uOZmC7lvW/emve/9bIUix6NDwAAAAAAAGea4B5Yolyp5Av3PZu/++7z6etuz2/90xtzw7bVzS4LAAAAAAAALliCe2DBsfGZ/PHfPJade47msg0D+Y33X5fhFT3NLgsAAAAAAAAuaIJ7IEny5J7R/NHfPJZj47O56+ZN+cW7rkhHe7HZZQEAAAAAAMAF77RTuePHj+c3f/M3c9NNN+WOO+7In//5n5+y7czMTH73d383t99+e26++eZ84AMfyPj4+BkpGDizqtVqvvT95/N//cWPMj1Tzq+/75r8i3ftENoDAAAAAADAOXLaM+4/+clPplwu57777suePXvywQ9+MNu2bcttt912QttPfOITmZyczN/+7d9m1apV2bVrVzo7O89o4cDrNzk9lz/5uyfyw6dGsmF1b/713ddn43Bfs8sCAAAAAACAi8ppBfeTk5O5995784UvfCH9/f255pprcvfdd+dzn/vcCcH9s88+m6985Sv55je/mcHBwSTJVVdddeYrB16X518cy6e/8EheOjqd265Zl19+z450d3p7BgAAAAAAAJxrp5XSPffcc0mS7du3L+y76qqr8tnPfvaEtj/+8Y+zadOmfOpTn8pf//VfZ9WqVfmVX/mV/OIv/uJrKrBQeE2ncRrmv1vf8cXnBzsP5TN/+3iq1Wr+5buvzJ03bUrBHwRoKcZogNZljAZoTcZngNZljAZoXcbo1nHaM+77+pY+PntwcDATExMntH3xxReza9euvPOd78x9992XnTt35ld/9VezZcuWkz5W/+UMDXlk97mwevVAs0vgHKlWq/mrb+7O//vFxzI00JXf+dU35crNQ80uC3gZxmiA1mWMBmhNxmeA1mWMBmhdxujmO63gvre394SQfmxs7IQwP0m6u7vT0dGRj370o2lvb88NN9yQ97znPfnWt771qoP70dGJlMuVV3UOp69QqP0lPHx4LNVqs6vhbCtXKvnzrzyVb/xwXzYN9+W37rkhq3rbMzIy1uzSgJMwRgO0LmM0QGsyPgO0LmM0QOsyRp8bbW3FV5y0flrB/datW5Mku3fvzrZt25IkO3fuzBVXXHFC2x07drzKMl+ePyBnX7Xqe77QTc2U8sd/81ge3n04V28Zyr+++7r0dnf43x3OA8ZogNZljAZoTcZngNZljAZoXcbo5iueTqPe3t68+93vzu///u9nfHw8O3fuzOc///l84AMfOKHtrbfemk2bNuWP/uiPUiqV8vjjj+fLX/5y7rrrrjNePPDKRsdm8n/++UN5ePfhvPX6Dfn4PTemt7uj2WUBAAAAAAAAdacV3CfJJz7xiSTJHXfckV/7tV/Lxz72sdx+++1JkptuuikPPPBAkqS9vT2f/vSn84//+I+55ZZb8lu/9Vv5t//23+bWW289C+UDL+eFQ+P53f/+QPYcGs/dd1yWD/7MVWlvO+2/9gAAAAAAAMA5UKhWW/ehB6OjEymVvOP+bCkUkuHhgYyMeGfFhejRZw7n0194NKVyJR/8matz+7Xrm10S8CoYowFalzEaoDUZnwFalzEaoHUZo8+N9vYz9I574PzyzR/ty599eVd6utrym7/whuzYPNTskgAAAAAAAIBTENzDBaRSrebz33omf/+95zO8ojsfv+fGbFj98v/1DgAAAAAAANBcgnu4QMyVyvmTv3si9z9xKJdvHMzHfv6GDPZ1NrssAAAAAAAA4BUI7uECMDY5m//y+Ufy9N5jeeOONflX770mnR1tzS4LAAAAAAAAOA2CezjPHTwymf/7L3+cQ6NTec9PbM4v3LktxUKh2WUBAAAAAAAAp0lwD+exp/YezX/53COZmJ7Lv3jXlbnr5kuaXRIAAAAAAADwKgnu4Tx1/xMH81+/+ETaioV87OdvyI3bh5tdEgAAAAAAAPAaCO7hPFOtVvP333s+n/vWM1nR35nf+oUbs2X9QLPLAgAAAAAAAF4jwT2cR8qVSv7Hl3fl2z/en0vW9OW3/umNWTXY3eyyAAAAAAAAgNdBcA/niVK5kv/nbx/PAzsP5drLVuWj778uPV3+CgMAAAAAAMD5TuoH54FSuZI//pvH8uCTL+WWHWvy6++7Nu1txWaXBQAAAAAAAJwBgntocaVyJX/014/loV0v5dar1uZf/ZNrhPYAAAAAAABwARHcQwsrlSv5wy88mh8+NZKfuLoW2rcVhfYAAAAAAABwIRHcQ4uaK9VC+x89PZLbrlmXD733aqE9AAAAAAAAXIAE99CC5kqVfPqvHsmPdx/O7deuy4d+9poUi4VmlwUAAAAAAACcBYJ7aDFzpXL+4K8ezcO7D+fN163Pr/7M1UJ7AAAAAAAAuIAJ7qGFzJXK+S+ffySPPnMkb7l+ff7/9u49yO66sPv452Rz3d1ssskmkBtESUKMCmwAA5QgBBAvFAU1Vqv1hqDUEjvTTmeKM7TY8dJOHwVb+2CxgyNoB5o8XLRyU4uxyv0mlxgMhJBsEthkySa7Ya/n+QNISRNgk+zu72T39Zphxpz8Vj6su1+Q9/7O79PvEe0BAAAAAABgqBPuoUJ0dr0U7R97emsWHzUtn3zP/IwoifYAAAAAAAAw1An3UAE6unry7eWP5PG1LTnl6On5k3cfKdoDAAAAAADAMCHcQ8E6unpyxX88kieeacmpx0zPx88S7QEAAAAAAGA4Ee6hQB2dPbn8Px7OqnUv5LTGGfnjd80T7QEAAAAAAGCYEe6hIB2dPfnW9Q/nd8++kNMXzszHzpybkmgPAAAAAAAAw45wDwV4sbM737ru4axevy1nHDszHz1DtAcAAAAAAIDhSriHQbazozvfuv7hPLl+W848blb+6PQ5oj0AAAAAAAAMY8I9DKKdHd355nUP5/cbtuWsd8zK0tNEewAAAAAAABjuhHsYJO0vdueb1z2UNU2tefeiw/LhU48Q7QEAAAAAAADhHgZD+4vd+T/XPZSnmlrz3hMOzwff+WbRHgAAAAAAAEgi3MOA6+ruzT+teCRPNbXmfScenvNOEe0BAAAAAACA/zGi6AEwlJXL5Vz90yeyat0LOX3hTNEeAAAAAAAA2INwDwPoxl89nd88tjnHzGnIR8+YK9oDAAAAAAAAexDuYYD86pGNuem/1+bwQ8fnwnPemhEjRHsAAAAAAABgT8I9DIDH127N929Zlcl1Y/OlDx2VMaOrip4EAAAAAAAAVCjhHvrZ+ud35J//328zelRVvrT06EyoHVP0JAAAAAAAAKCCCffQj17Y0ZHLr384nV29+eK5b8uMhpqiJwEAAAAAAAAVTriHfvJiZ3cuv/6RbGntyKfeMz9vmT2p6EkAAAAAAADAQUC4h37Q09ub/3vjY3lm8/a8/+Q35Q/ePq3oSQAAAAAAAMBBQriHA1Qul/PDO57MI2u25KS3HZpz/mB20ZMAAAAAAACAg4hwDwfotnufzS8e2JD5h03Mp94zP6VSqehJAAAAAAAAwEFEuIcDcN+q53Ldz3+f6Q01+eJ5b8/IKt9SAAAAAAAAwL5RGWE//X7Dtvzrjx/P+JrR+dKHjkr12FFFTwIAAAAAAAAOQsI97IfnWtpzxX88klIpWfaho9IwcVzRkwAAAAAAAICDlHAP+2jHzq588/pH0vZiVy48561507S6oicBAAAAAAAABzHhHvZBV3dPvr38kWze2p6PnTEvjXOnFD0JAAAAAAAAOMgJ99BHveVyvveTJ/Lk+m151/GzcvqxM4ueBAAAAAAAAAwBwj300f/75VO554nnsnDelCw9bU7RcwAAAAAAAIAhQriHPrjzoQ35yW+eyZum1eVzf7ggI0aUip4EAAAAAAAADBHCPbyBR5/akh/cujoNE8bm4g8dlTGjqoqeBAAAAAAAAAwhwj28jmef25Hv3PBoxo6uyp8vPToTakYXPQkAAAAAAAAYYoR7eA3tL3bn28sfSVd3b/7sg2/PtMk1RU8CAAAAAAAAhiDhHvaiXC7nB7f9Ls3bXswfnT43Rx5WX/QkAAAAAAAAYIgS7mEv/vu3m3L345tzzJyGLFk4o+g5AAAAAAAAwBAm3MP/snFLW669fXXqx4/JZ973lpRKpaInAQAAAAAAAEOYcA+v0tXdmytveiydXT353NkLUjtuVNGTAAAAAAAAgCFOuIdXWX7nmqzbvCPvO2l25h/uufYAAAAAAADAwBPu4WWPrGnObfc+myNm1OX9J88ueg4AAAAAAAAwTAj3kOSFHR353k+eyLgxI3PhH741VSN8awAAAAAAAACDQ51k2Ostl3PVjx/P9vaufOo989MwcVzRkwAAAAAAAIBhRLhn2Lv17nV5fG1LTjl6eo6fP7XoOQAAAAAAAMAwI9wzrD3V1JoVv3wq0yZX56NnzC16DgAAAAAAADAMCfcMWzs7unPlTY+mVCrlwnPemjGjqoqeBAAAAAAAAAxDwj3DUrlczg9u/V2ef+HFfGTJnBx2yPiiJwEAAAAAAADDlHDPsPTrRzflrsc355g5DVmycEbRcwAAAAAAAIBhTLhn2Nm0tT3X3LY6E2tH59PvnZ9SqVT0JAAAAAAAAGAYE+4ZVrp7enPljY+ls6snn/vDt2Z89eiiJwEAAAAAAADDnHDPsLL8zjV5ZvP2vO+kw/OWw+uLngMAAAAAAAAg3DN8/PapLbn1nmdzxIy6nPMHbyp6DgAAAAAAAEAS4Z5hYtuOjnzvx49n3JiqXPiHb83IKl/6AAAAAAAAQGVQLxnyesvlXPWTJ9La3pVPvnt+GiaOK3oSAAAAAAAAwC7CPUPebfc8m8ee3prFR03LO95ySNFzAAAAAAAAAHYj3DOkPb2xNcvvXJNDJ1XnY2fMK3oOAAAAAAAAwB6Ee4asnR3dufLGx1IqJZ9//1szZnRV0ZMAAAAAAAAA9iDcM2Rdc9vqPPfCznz4tDk57JDxRc8BAAAAAAAA2CvhniHp149uzG8e25SjjpicM46dWfQcAAAAAAAAgNck3DPkbNvRkWtvX50JtaPzmfe9JaVSqehJAAAAAAAAAK9JuGfIWX7nU9nZ0ZOPnzkvddWji54DAAAAAAAA8LqEe4aUpze25le/3Zi3HF6fhfOmFD0HAAAAAAAA4A0J9wwZveVyfnj76owolfKxM+Z6i3wAAAAAAADgoCDcM2Tc9dimrGlqzWkLZ2TGlNqi5wAAAAAAAAD0SZ/DfWtra5YtW5bGxsYsXrw411577Rt+zIoVK3LkkUfmRz/60QGNhDeys6M71//XmtSOG5UPLH5T0XMAAAAAAAAA+mxkXy+87LLL0tPTk5UrV2bdunX59Kc/nSOOOCInnHDCXq9vaWnJd7/73cybN6/fxsJr+clvnsm2HZ35xFlHpmbsqKLnAAAAAAAAAPRZn8J9e3t7brnlltxwww2pra3NggULcu6552b58uWvGe6/8Y1v5LOf/WxuuummAxroMeUD55XP7cH+Od68tT233bsus6bW5tRjph/0fz0AydA5owGGImc0QGVyPgNULmc0QOVyRleOPoX7tWvXJknmzJmz67X58+fn6quv3uv1d999d9auXZuvfe1rBxTu6+tr9vtj6bvJk8cXPeGA/N+bHk93TzkXfejoTJ1aV/QcgH51sJ/RAEOZMxqgMjmfASqXMxqgcjmji9fnO+5ranaP6HV1dWlra9vj2s7Ozlx22WX5h3/4h5QO8EczWlra0tPTe0D/Hby2Uumlb8ItW7anXC56zf757VNbcvdjm3L8W6bm0Alj0ty8vehJAP1iKJzRAEOVMxqgMjmfASqXMxqgcjmjB0dV1Yg3vGm9T+G+urp6j0i/ffv2PWJ+knz3u9/NiSeemAULFuzD1NfmC2TglcsH5+e5u6c3P7rjyYweOSJLT51zUP41ALyRg/WMBhgOnNEAlcn5DFC5nNEAlcsZyhkGEAAAHY1JREFUXbw+hfvZs2cnSdasWZMjjjgiSbJq1arMnTt3j2vvvvvuPPHEE7n55puTJDt27Mijjz6aBx98MH//93/fT7Mh+fn967NxS3vef/KbMnnC2KLnAAAAAAAAAOyXPt9xf9ZZZ+Xyyy/PV7/61axfvz4rVqzIt771rT2uvfzyy9PZ2bnr18uWLcvpp5+eD3/4w/23mmGvta0zN/7305lcNybvXnRY0XMAAAAAAAAA9tuIvl546aWXJkkWL16c888/PxdffHFOPPHEJEljY2Puu+++JMmkSZNy6KGH7vpj9OjRGT9+fOrr6wdgPsPVil+uyc6OnnxkydyMGVVV9BwAAAAAAACA/danO+6TpK6uLldcccVef+/BBx98zY/7wQ9+sO+r4HWs3dSalQ9vzPzDJubYI6cUPQcAAAAAAADggPT5jnuoBOVyOT+848mklHz0jHkplUpFTwIAAAAAAAA4IMI9B5W7H9+c36/fllMbZ2TW1Nqi5wAAAAAAAAAcMOGeg0ZHZ0+u/681qRk7MucufnPRcwAAAAAAAAD6hXDPQeMnd61Ny/aOfGDxm1M7blTRcwAAAAAAAAD6hXDPQeG5F3bmlrufzcwpNTm1cXrRcwAAAAAAAAD6jXDPQeG6n/8+3T29+egZ81I1wpctAAAAAAAAMHQooFS8x9ZuzQOrn8+xR07JWw6vL3oOAAAAAAAAQL8S7qlo3T29+dEdT2bUyBH5yGlzip4DAAAAAAAA0O+EeyraLx7ckKbmtrz7HYelYeK4oucAAAAAAAAA9DvhnorV2t6ZG1c+nUl1Y/LeEw8veg4AAAAAAADAgBDuqVg3/PKptHd0Z+lpczJmVFXRcwAAAAAAAAAGhHBPRVq3eXvufKgp82ZNzPHzpxY9BwAAAAAAAGDACPdUnHK5nB/evjopJR87Y25KpVLRkwAAAAAAAAAGjHBPxbl31XNZvX5b3nn09Bx2yPii5wAAAAAAAAAMKOGeitLR1ZPrfvH7VI8ZmXNPeXPRcwAAAAAAAAAGnHBPRfnpXc9ka2tH3r/4TRlfPbroOQAAAAAAAAADTrinYrRs78hP716X6Q01Oa1xRtFzAAAAAAAAAAaFcE/F+K8HN6SruzfnnfLmjKzypQkAAAAAAAAMD+ooFaGruzd3PrQhDRPG5pg5DUXPAQAAAAAAABg0wj0V4b5Vz6W1vSunLZyRESNKRc8BAAAAAAAAGDTCPRXhZw+sz6iRI7L4qOlFTwEAAAAAAAAYVMI9hXt6Y2ueamrNCQsOSe24UUXPAQAAAAAAABhUwj2Fu+O+9UmS04+dWfASAAAAAAAAgMEn3FOo1rbO3Ltqc+bNnJDDDhlf9BwAAAAAAACAQSfcU6g7H25Kd085S9xtDwAAAAAAAAxTwj2F6e7pzX89uCH148dk4bwpRc8BAAAAAAAAKIRwT2EeerI5Lds7cuox0zOyypciAAAAAAAAMDyppRTmjvvXZ2RVKe88ZkbRUwAAAAAAAAAKI9xTiGef25HVz76Q4+dPTV3N6KLnAAAAAAAAABRGuKcQP7v/2STJ6cfOKngJAAAAAAAAQLGEewbdjp1dueuxzXnTtLq8eXpd0XMAAAAAAAAACiXcM+h+9cjGdHb35vRjPdseAAAAAAAAQLhnUPX2lvPzB9anrnpUjp9/SNFzAAAAAAAAAAon3DOoHlmzJc3bXswpx8zIqJG+/AAAAAAAAACUUwbVz+5/NiNKpZx6zPSipwAAAAAAAABUBOGeQbNxS1seW9uShUdOyaS6sUXPAQAAAAAAAKgIwj2D5uf3b0iSnHHszIKXAAAAAAAAAFQO4Z5BsbOjO796dGNmTa3N3JkTip4DAAAAAAAAUDGEewbFf/92Yzo6e3L6sTNTKpWKngMAAAAAAABQMYR7BlxvuZyfPbAhNWNHZtGCQ4qeAwAAAAAAAFBRhHsG3ONrt2bz1vYsPnp6xoyqKnoOAAAAAAAAQEUR7hlwP7tvfUpJTmucUfQUAAAAAAAAgIoj3DOgnnthZx5ZsyVHz2nIlInjip4DAAAAAAAAUHGEewbULx5Yn3KS04+bWfQUAAAAAAAAgIok3DNgOjp7svLhjZk2uToLDq8veg4AAAAAAABARRLuGTC/eXxT2ju6s2ThzJRKpaLnAAAAAAAAAFQk4Z4BUS6X87P712fs6Kqc9LZDi54DAAAAAAAAULGEewbE6mdfyIbn23Ly26dl3JiRRc8BAAAAAAAAqFjCPQPijvvXJ0mWHDuz4CUAAAAAAAAAlU24p99tbX0xD65uztveNCmHTqoueg4AAAAAAABARRPu6Xe/eHBDesvlnO5uewAAAAAAAIA3JNzTr7q6e3LnQ02ZMnFs3n7E5KLnAAAAAAAAAFQ84Z5+dc8Tz2XHzq4sWTgzI0qloucAAAAAAAAAVDzhnn5TLpdzx/3rM3rUiCw+alrRcwAAAAAAAAAOCsI9/eapptY8s2l7TnrroakeO6roOQAAAAAAAAAHBeGefvOz+9cnSZYsnFnwEgAAAAAAAICDh3BPv9i2oyP3rnou8w+bmJlTa4ueAwAAAAAAAHDQEO7pF3c+1JSe3nJOP9bd9gAAAAAAAAD7QrjngHX39OYXD23IpLoxOWZuQ9FzAAAAAAAAAA4qwj0H7JE1W7JtR2dOa5yRqhG+pAAAAAAAAAD2hcrKAVu7qTVJcvQR7rYHAAAAAAAA2FfCPQesqbk9I0qlHDKpuugpAAAAAAAAAAcd4Z4DtqG5LVPrx2XUSF9OAAAAAAAAAPtKaeWAdHX35rmW9kxvqCl6CgAAAAAAAMBBSbjngGza2p5yOcI9AAAAAAAAwH4S7jkgTc1tSZLpDZ5vDwAAAAAAALA/hHsOyIaXw/2MhtqClwAAAAAAAAAcnIR7DsjG5raUSsmhk8YVPQUAAAAAAADgoCTcc0A2NLdlan11Ro2sKnoKAAAAAAAAwEFJuGe/dXX35rmWnZk+2fPtAQAAAAAAAPaXcM9+27y1Pb3lcmZMqSl6CgAAAAAAAMBBS7hnvzVtaUuSTJ8s3AMAAAAAAADsL+Ge/bbh+ZfDfYNwDwAAAAAAALC/hHv2W9OWtpRKyTTPuAcAAAAAAADYb8I9+62puS1TJ47LqJFVRU8BAAAAAAAAOGgJ9+yX7p7ebN6609vkAwAAAAAAABwg4Z79smlre3rLZeEeAAAAAAAA4AAJ9+yXpua2JBHuAQAAAAAAAA5Qn8N9a2trli1blsbGxixevDjXXnvtXq976KGH8tnPfjaLFi3KokWLcsEFF2Tt2rX9tZcK8Uq4nyHcAwAAAAAAAByQPof7yy67LD09PVm5cmWuvPLKXHHFFbnrrrv2uG7btm354Ac/mDvuuCMrV67M3Llz84UvfKFfR1O8pua2lErJoZOqi54CAAAAAAAAcFAb2ZeL2tvbc8stt+SGG25IbW1tFixYkHPPPTfLly/PCSecsNu173znO3f79Wc+85lcddVVaWlpSX19/T4PLJX2+UPoo1c+t/vzOW5qbsuUieMyZnRV/44CIMmBndEADCxnNEBlcj4DVC5nNEDlckZXjj6F+1fe6n7OnDm7Xps/f36uvvrqN/zYe+65J1OmTNmvaF9f723YB8PkyeP36fqu7t5sbtmZ495ySBoa9u1jAdg3+3pGAzB4nNEAlcn5DFC5nNEAlcsZXbw+33FfU7N7RK+rq0tbW9vrftz69evzd3/3d7nkkkv2a1xLS1t6enr362N5Y6XSS9+EW7ZsT7nc94/b8PyO9PSW01A3Js3N2wduIMAwtr9nNAADzxkNUJmczwCVyxkNULmc0YOjqmrEG9603qdwX11dvUek3759+x4x/9U2bdqUT33qU/nc5z6X9773vX350+yVL5CBVy7v2+d5/fMvfS1Mn1zjfx+AAbavZzQAg8cZDVCZnM8AlcsZDVC5nNHFG9GXi2bPnp0kWbNmza7XVq1alblz5+71+s2bN+dP/uRPsnTp0nzqU5864JFUlqbml8N9g0cZAAAAAAAAAByoPoX76urqnHXWWbn88suzY8eOrFq1KitWrMh55523x7WbN2/OJz7xiZxzzjm54IIL+n0wxWtqbkspyaGTq4ueAgAAAAAAAHDQ61O4T5JLL700SbJ48eKcf/75ufjii3PiiScmSRobG3PfffclSa6//vo888wz+d73vpfGxsZdfzQ1NQ3AfIrQtKU9UyaOy5hRVUVPAQAAAAAAADjo9ekZ90lSV1eXK664Yq+/9+CDD+76z1/84hfzxS9+8cCXUZG6e3qzeWt73v7myUVPAQAAAAAAABgS+nzHPSTJ5pad6ekte749AAAAAAAAQD8R7tknTc1tSZLpDZ5vDwAAAAAAANAfhHv2ySvhfkZDbcFLAAAAAAAAAIYG4Z59sqG5LaUkh052xz0AAAAAAABAfxDu2Scbm9vSMHFsxoyqKnoKAAAAAAAAwJAg3NNn3T292bS1PdMn1xQ9BQAAAAAAAGDIEO7ps+dadqant5zpU4R7AAAAAAAAgP4i3NNnTc1tSZIZDcI9AAAAAAAAQH8R7umzV8L9dOEeAAAAAAAAoN8I9/TZhua2lJJM84x7AAAAAAAAgH4j3NNnTVvaMnnC2IwZVVX0FAAAAAAAAIAhQ7inT7p7erNpS7vn2wMAAAAAAAD0M+GePnn+hZ3p6S17vj0AAAAAAABAPxPu6ZMNz7cliXAPAAAAAAAA0M+Ee/qkaYtwDwAAAAAAADAQhHv6pKn55XA/WbgHAAAAAAAA6E/CPX3S1NyWhgljM2Z0VdFTAAAAAAAAAIYU4Z431NPbm01b271NPgAAAAAAAMAAEO55Q8+17Ex3T1m4BwAAAAAAABgAwj1v6JXn288Q7gEAAAAAAAD6nXDPG3ol3LvjHgAAAAAAAKD/Cfe8oQ0vh/tpk6sLXgIAAAAAAAAw9Aj3vKGm5vZMrhubsaNHFj0FAAAAAAAAYMgR7nldPb292bS1LTOmeJt8AAAAAAAAgIEg3PO6nn/hxXT3lDN9snAPAAAAAAAAMBCEe17Xhudfer799AbhHgAAAAAAAGAgCPe8rqYtwj0AAAAAAADAQBLueV1NzS+F+2mTqwteAgAAAAAAADA0Cfe8rqbmtkyuG5NxY0YWPQUAAAAAAABgSBLueU29veVs3NKe6Q21RU8BAAAAAAAAGLKEe17T8y/sTHdPb6Y3eJt8AAAAAAAAgIEi3POaNrz8fPvpDTUFLwEAAAAAAAAYuoR7XlOTcA8AAAAAAAAw4IR7XtOucD9ZuAcAAAAAAAAYKMI9r6mpuS2T6sZk3JiRRU8BAAAAAAAAGLKEe/aqt7ecjVvbvU0+AAAAAAAAwAAT7tmr57ftTFd3r7fJBwAAAAAAABhgwj171fT8S8+3n+GOewAAAAAAAIABJdyzV01bXgr33iofAAAAAAAAYGAJ9+zVhmbhHgAAAAAAAGAwCPfsVVNzW+rHj8m4MSOLngIAAAAAAAAwpAn37KG3t5yNW9o93x4AAAAAAABgEAj37KF52850dfd6m3wAAAAAAACAQSDcswfPtwcAAAAAAAAYPMI9e2gS7gEAAAAAAAAGjXDPHnaF+8nCPQAAAAAAAMBAE+7Zw4bmttSPH5PqsSOLngIAAAAAAAAw5An37Ka3t5yNW9q9TT4AAAAAAADAIBHu2U3ztp3p6u71NvkAAAAAAAAAg0S4ZzdNze1JkhlThHsAAAAAAACAwSDcs5sNzTuSxB33AAAAAAAAAINEuGc3r9xxP72huuAlAAAAAAAAAMODcM9umprbMrF2dKrHjip6CgAAAAAAAMCwINyzS2+5nI1b2jKjwdvkAwAAAAAAAAwW4Z5dmre9mM7u3kwT7gEAAAAAAAAGjXDPLk3NbUnijnsAAAAAAACAQSTcs8v/hPvagpcAAAAAAAAADB/CPbu8Eu6nN1QXvAQAAAAAAABg+BDu2WVDc1sm1o5O9dhRRU8BAAAAAAAAGDaEe5IkveVyNm5py3TPtwcAAAAAAAAYVMI9SZIt215MZ1evcA8AAAAAAAAwyIR7krz6+fbCPQAAAAAAAMBgEu5J8j/hfoZwDwAAAAAAADCohHuSuOMeAAAAAAAAoCjCPUmSDc1tmVA7OjVjRxU9BQAAAAAAAGBYEe5Jb7mcjVvaM32yu+0BAAAAAAAABptwT7ZuezEdXT2ebw8AAAAAAABQAOGebPB8ewAAAAAAAIDCCPekSbgHAAAAAAAAKIxwj3APAAAAAAAAUCDhnmxobsuEmtGpHTeq6CkAAAAAAAAAw45wP8yVy+U0Nbe72x4AAAAAAACgIML9MPd8y850dPUI9wAAAAAAAAAFEe6HuXWbtyfxfHsAAAAAAACAogj3w9y6TS+F+xnCPQAAAAAAAEAhhPth7ll33AMAAAAAAAAUSrgf5tZtbk1dzejUjhtV9BQAAAAAAACAYUm4H8bK5XKe3bw90xuqi54CAAAAAAAAMGz1Ody3trZm2bJlaWxszOLFi3Pttde+5rX33HNPzj777Bx99NFZunRpnnzyyX4ZS//a2tqRnR09mdFQW/QUAAAAAAAAgGGrz+H+sssuS09PT1auXJkrr7wyV1xxRe666649rmtpaclFF12UCy64IPfee29OP/30XHTRRenu7u7X4Ry4pua2JHHHPQAAAAAAAECBRvblovb29txyyy254YYbUltbmwULFuTcc8/N8uXLc8IJJ+x27e23357Zs2fnnHPOSZKcf/75+f73v5977703J5544j4PLJX2+UPoow1bXgr3M6fU+DwDVJhXzmXnM0DlcUYDVCbnM0DlckYDVC5ndOXoU7hfu3ZtkmTOnDm7Xps/f36uvvrqPa5dvXp15s+fv+vXVVVVmTt3blavXr3P4b6+vmafrmff9JRLGVlVytuPPDTjq0cXPQeAvZg8eXzREwB4Dc5ogMrkfAaoXM5ogMrljC5en++4r6nZPaLX1dWlra1tr9dOmDChT9e+kZaWtvT09O7zx9E3ZyycllMaZ6RzZ0ea2zuKngPAq5RKL/2D0pYt21MuF70GgFdzRgNUJuczQOVyRgNULmf04KiqGvGGN633KdxXV1fvEd63b9++R8x/5dodO3b06dq+8AUycMaMGpkZDePT3OwbEaBSlcv+XghQqZzRAJXJ+QxQuZzRAJXLGV28EX25aPbs2UmSNWvW7Hpt1apVmTt37h7Xzps3L6tWrdr1697e3qxevTrz5s07wKkAAAAAAAAAMPT0KdxXV1fnrLPOyuWXX54dO3Zk1apVWbFiRc4777w9rj3zzDPz9NNP58c//nE6Oztz1VVXpaamJscff3y/jwcAAAAAAACAg12fwn2SXHrppUmSxYsX5/zzz8/FF1+cE088MUnS2NiY++67L0lSX1+ff/7nf86//Mu/5Ljjjsvtt9+e73znOxk5sk/vyg8AAAAAAAAAw0qpXK7cpxW0tLSlu7u36BlDVqmUNHjGPUBFckYDVC5nNEBlcj4DVC5nNEDlckYPjpEjR6S+vuZ1r+nzHfcAAAAAAAAAQP8T7gEAAAAAAACgQMI9AAAAAAAAABRIuAcAAAAAAACAAgn3AAAAAAAAAFAg4R4AAAAAAAAACiTcAwAAAAAAAECBhHsAAAAAAAAAKJBwDwAAAAAAAAAFEu4BAAAAAAAAoEDCPQAAAAAAAAAUSLgHAAAAAAAAgAIJ9wAAAAAAAABQIOEeAAAAAAAAAAo0sugBr6eqys8VDAafZ4DK5YwGqFzOaIDK5HwGqFzOaIDK5YweWH35/JbK5XJ5ELYAAAAAAAAAAHvhRycAAAAAAAAAoEDCPQAAAAAAAAAUSLgHAAAAAAAAgAIJ9wAAAAAAAABQIOEeAAAAAAAAAAok3AMAAAAAAABAgYR7AAAAAAAAACiQcA8AAAAAAAAABRLuAQAAAAAAAKBAwj0AAAAAAAAAFEi4BwAAAAAAAIACCfcAAAAAAAAAUCDhfphqbW3NsmXL0tjYmMWLF+faa68tehIASTo7O3PJJZdkyZIlaWxszPve977cfPPNRc8C4FVaWlqyaNGiLF26tOgpALzKrbfemrPPPjvHHHNMTjvttNx2221FTwIgyfr163PBBRfkHe94R0466aT89V//ddrb24ueBTDsXHPNNTnvvPPytre9LX/+53++2++tXr06S5cuzdFHH52zzz479913X0Erh7eRRQ+gGJdddll6enqycuXKrFu3Lp/+9KdzxBFH5IQTTih6GsCw1t3dnalTp+b73/9+ZsyYkQceeCAXXnhhZs6cmcbGxqLnAZDkG9/4RubNm5eOjo6ipwDwst/85jf56le/mn/8x3/MwoUL09LSIgoBVIhLL700kyZNyi9/+cvs3LkzF110Ub7zne/kL/7iL4qeBjCsTJ06NRdddFF+/etfp6WlZdfrXV1d+cIXvpCPfOQjueaaa/LTn/40F110UW6//fZMmDChwMXDjzvuh6H29vbccsst+dKXvpTa2tosWLAg5557bpYvX170NIBhr7q6OsuWLcusWbMyYsSIHHfccVm4cGEefPDBoqcBkOTuu+/OunXr8oEPfKDoKQC8yhVXXJE//dM/zXHHHZcRI0Zk8uTJmTVrVtGzAEjy7LPP5uyzz87YsWNTX1+fM888M08++WTRswCGnXe9610544wzUl9fv9vr99xzT1588cWcf/75GT16dN7//vdn5syZ3sGqAML9MLR27dokyZw5c3a9Nn/+fP+wBFCB2tvb8+ijj2bu3LlFTwEY9jo7O/OVr3wll156aUqlUtFzAHhZT09Pfvvb36alpSVnnnlmTj755PzVX/1Vtm3bVvQ0AJJ88pOfzM0335z29vZs2bIlt912W0455ZSiZwHwsieffDLz5s3LiBH/k411w2II98NQe3t7ampqdnutrq4ubW1tBS0CYG/K5XIuueSSHHXUUTn55JOLngMw7F155ZU5+eSTc+SRRxY9BYBXaW5uTldXV2699dZcc801+c///M9s3bo1X/3qV4ueBkCSRYsW5amnnsqxxx6bk046KRMnTszSpUuLngXAy9ra2jJ+/PjdXtMNiyHcD0PV1dV7fLNt3759j5gPQHHK5XIuvfTSbNy4Md/85jfd2QlQsLVr1+bGG2/Mn/3ZnxU9BYD/Zdy4cUmST3ziEznkkENSV1eXz3/+87nzzjsLXgZAT09PPvvZz+bUU0/NQw89lPvvvz/19fX5y7/8y6KnAfCympqa7NixY7fXdMNiCPfD0OzZs5Mka9as2fXaqlWrvA0zQIUol8v527/92zz++OO56qqrUl1dXfQkgGHvgQceyObNm7NkyZIsWrQoX/nKV/LYY49l0aJFe/yfWwAGV11dXaZNm+aHXQEq0LZt27Jp06Z8/OMfz5gxY1JbW5uPfvSjfrgKoILMnTs3q1evTm9v767XnnjiCd2wAML9MFRdXZ2zzjorl19+eXbs2JFVq1ZlxYoVOe+884qeBkCSyy67LA8//HC+973vpba2tug5ACR5z3vek9tvvz033nhjbrzxxixbtizz5s3LjTfe6CfQASrAhz70ofzgBz9Ic3NzduzYkX/913/NkiVLip4FMOxNmjQps2bNyg9/+MN0dnamvb091113ncdPARSgu7s7HR0d6e7uTm9vbzo6OtLV1ZV3vOMdGT16dP7t3/4tnZ2dufnmm7N+/fqceeaZRU8edkrlcrlc9AgGX2tra7785S9n5cqVqampyRe+8IX88R//cdGzAIa9DRs2ZMmSJRk9enRGjhy56/ULL7wwn//85wtcBsCrrVixIv/+7/+e6667rugpAOSlfwn59a9/PTfddFOqqqpy6qmn5pJLLvGDsAAVYNWqVfna176WJ554IqVSKUcffXS+/OUv57DDDit6GsCw8u1vfzv/9E//tNtr5557br7+9a/nd7/7Xb785S/nd7/7XWbNmpW/+Zu/yfHHH1/Q0uFLuAcAAAAAAACAAnmrfAAAAAAAAAAokHAPAAAAAAAAAAUS7gEAAAAAAACgQMI9AAAAAAAAABRIuAcAAAAAAACAAgn3AAAAAAAAAFAg4R4AAAAAAAAACiTcAwAAAAAAAECBhHsAAAAAAAAAKND/B0Dh7/dU5dy/AAAAAElFTkSuQmCC\n", | |
"text/plain": [ | |
"<Figure size 2560x640 with 1 Axes>" | |
] | |
}, | |
"metadata": {}, | |
"output_type": "display_data" | |
} | |
], | |
"source": [ | |
"fig = plt.figure(figsize=(32,8), dpi=80, facecolor='w', edgecolor='k')\n", | |
"ax = plt.subplot(1,1,1)\n", | |
"ax.plot(x, scipy.stats.chi2.cdf(x, 1))" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 49, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"0.15729920705028488" | |
] | |
}, | |
"execution_count": 49, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"1-scipy.stats.chi2.cdf(lmda, 1)" | |
] | |
}, | |
{ | |
"cell_type": "markdown", | |
"metadata": {}, | |
"source": [ | |
"# Even more by Hand" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 50, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"(10.702370255288873, array([ 0.00282359, -7.78643352]))" | |
] | |
}, | |
"execution_count": 50, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"lr = sklearn.linear_model.LinearRegression()\n", | |
"X, y = df[['month', 'x2']].values, df['temperature'].values\n", | |
"lr.fit(X, y)\n", | |
"# lr.coef_.squeeze()\n", | |
"lr.intercept_, lr.coef_" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 51, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"0 3.961945\n", | |
"1 6.814801\n", | |
"2 10.710841\n", | |
"3 14.606881\n", | |
"4 17.459737\n", | |
" ... \n", | |
"355 15.600784\n", | |
"356 11.710391\n", | |
"357 7.819998\n", | |
"358 4.972789\n", | |
"359 3.932428\n", | |
"Length: 360, dtype: float64" | |
] | |
}, | |
"execution_count": 51, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"y_ = lr.intercept_ + lr.coef_[0] * df['month'] + lr.coef_[1] * df['x2']\n", | |
"y_" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 52, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"0 -0.731945\n", | |
"1 -2.334801\n", | |
"2 -1.670841\n", | |
"3 -5.966881\n", | |
"4 -0.669737\n", | |
" ... \n", | |
"355 7.489216\n", | |
"356 10.679609\n", | |
"357 9.230002\n", | |
"358 7.307211\n", | |
"359 3.257572\n", | |
"Length: 360, dtype: float64" | |
] | |
}, | |
"execution_count": 52, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"(y-y_)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 53, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"(1.3273333049962983e-15, 3.9229754827902212)" | |
] | |
}, | |
"execution_count": 53, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"mu, sigma = scipy.stats.norm.fit((y-y_))\n", | |
"mu, sigma" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 54, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"<matplotlib.axes._subplots.AxesSubplot at 0x7f3f88dbaf28>" | |
] | |
}, | |
"execution_count": 54, | |
"metadata": {}, | |
"output_type": "execute_result" | |
}, | |
{ | |
"data": { | |
"image/png": "iVBORw0KGgoAAAANSUhEUgAAB/UAAAIKCAYAAADS0RwDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAMTQAADE0B0s6tTgAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nOzdeXxU9b3/8fc5s2Um+wIhCWHLQsK+rwoqi7KpoKBotdpqW/Xeelu9bX/W9rbaamtrq9UuVtu6VhFBQFSUxQWxshN2AmEPa0IgZJ3t/P6IYi2gYcuZSV7Px8NHyMyZyXti5vN4zHmf8z2GZVmWAAAAAAAAAAAAAABAxDHtDgAAAAAAAAAAAAAAAE6NUh8AAAAAAAAAAAAAgAhFqQ8AAAAAAAAAAAAAQISi1AcAAAAAAAAAAAAAIEJR6gMAAAAAAAAAAAAAEKEo9QEAAAAAAAAAAAAAiFCU+gAAAAAAAAAAAAAARCin3QHORWVlrUKhsN0xWpzk5FhVVFTbHQMAvhLzCkA0YFYBiBbMKwDRgnkFIFowrwBEC+bVhedwmEpI8J72/qgu9UOhsIJBSv2mZBgNX0OhsCzL3iwA8GWYVwCiAbMKQLRgXgGIFswrANGCeQUgWjCvIgPL7wMAAAAAAAAAAAAAEKEo9QEAAAAAAAAAAAAAiFCU+gAAAAAAAAAAAAAARChKfQAAAAAAAAAAAAAAIhSlPgAAAAAAAAAAAAAAEYpSHwAAAAAAAAAAAACACEWpDwAAAAAAAAAAAABAhKLUBwAAAAAAAAAAAAAgQlHqAwAAAAAAAAAAAAAQoSj1AQAAAAAAAAAAAACIUJT6AAAAAAAAAAAAAABEKEp9AAAAAAAAAAAAAAAiFKU+AAAAAAAAAAAAAAARilIfAAAAAAAAAAAAAIAIRakPAAAAAAAAAAAAAECEotQHAAAAAAAAAAAAACBCUeoDAAAAAAAAAAAAABChKPUBAAAAAAAAAAAAAIhQlPoAAAAAAAAAAAAAAEQop90BAAAAAABA06qtD+rAkZqG/8prVFUX+NLtDUkxMS7V1QVkfcl2Loep9GSv2qT4lJ7iU3K8R4ZhnNfsAAAAAAC0NJT6AAAAAAA0Q8FQWIeP1urgkdrPC/wjNTp4pEbHqv1NksHjcig9paHk/6zo/+zfXg+7JAAAAAAAaAw+QQMAAAAAEKUsy9Kxar8OHqnR/k8L+wPlNTpQUauyo7UKhb94Xr3P41SbVJ+6dEhRm9TPC/bEWHfD6finYRhSakqcyo9UyfqSU/Xr/aGGDP9xEMGyTYdO2jYx1n1S0d8m1ae0xBg5HVwtEAAAAACAz1DqAwAAAAAQJSzL0v7yGhWVlGnttnLtPnRctfWhL2zjMA21TvaqR07qFwvzFJ/ifa6zWg7fMKTEOI8Cdf4vLfXlk1oledWtU+oXbq4PNJT9BytqdaC8Wgc+XT1g96Eqbdlz9JT5u3ZMUc/cNHXOTqLkBwAAAAC0aJT6AAAAAABEsGAorC17jqpoW5mKtpXp8NE6SZLbZapTRsIXl7b/9Ex3hxlZJbjH5VC79Hi1S4//wu2WZel4TeALZ/YfKK/R7kPHtWDFXi1YsVcxboe6fVrwd89JVYLPbdOrAAAAAADAHo0u9SsrK/WTn/xEH374oeLi4vSd73xHN95440nb+f1+3XvvvVq/fr1KS0v19NNPa9iwYV/Y5sUXX9RTTz2lqqoqDR8+XL/4xS8UFxd37q8GAAAAAIBmoLLGr3Ul5SraVqb1O46ozt9wNn5KgkeX9slSr9w0FbRLksvpsDnpuTEMQwmxbiXEupWfnXTidsuyVHq4Wmu2lamopEwrtxzWii2HZUjqlJWgXrlp6pmTpqxWsWe18gAAAAAAANGk0aX+Aw88oFAopMWLF2v37t269dZblZOTo0GDBp20bZ8+fXTzzTfrnnvuOem+JUuW6Mknn9Q//vEPZWdn6wc/+IEefPBB/frXvz63VwIAAAAAQJSyLEulZdWfno1frpLSY7KkEyV2z5w09cxNU9sWUmIbhqG2rePUtnWcxg/poMpqv9Zt//wgh5LSSs34YLtSE2LUMzdVPZvJQQ4AAAAAAJxKo0r9mpoazZs3T7NmzVJcXJy6dOmiiRMnasaMGSeV+m63W7fccoskyeE4+cP0zJkzNWnSJBUWFkqS7r77bl177bX62c9+Jq/Xe44vBwAAAACA6BAIhrVld4WKtpWrqKRMZccaltX3uB3q07mVeuWmqXunVCXEstx8QqxbQ7tnaGj3jM8vR7C1TGu2lWnRqlItWlUqj8uhLh2S1Ss3TT1yUpUY57E7NgAAAAAA50WjSv2dO3dKknJzc0/cVlBQoGefffaMf+DWrVs1fPjwE9/n5+crHA5r165dKigoOOPnawEnKESUz37f/N4BRDrmFYBowKwCWh5/IKTlmw9pVXGZNuw4ovpAw7L6rZJiNLJfW/XKTVN+dpJcTtPmpF8USfPK5TTVrWOKunVM0Q2j8rSvvEZF28q05tOSf/XWMklSp4wE9cxN1ZDubZSWyEkEQEsRSfMKAL4M8wpAtGBeRYZGn6kfGxv7hdsSEhJUXV19xj+wpqZG8fHxJ743DENxcXGqqqo64+dKTo796o1wQaSmxn/1RgAQAZhXAKIBswpo/ioq6/Tmxzv09sc7VVntl2lIBR1SNKBLG/Xvkq7s9PioWFY/EudVq1YJ6lnQRjdLqqz2a9Xmg1q28aBWbT6o1xdXavaSnRrcPUNXD8tRQYcUu+MCaCKROK8A4FSYVwCiBfPKXo0q9X0+30kF/vHjx08q+hv7XP9Z4FdVVSkuLu6Mn6uiolqhUPiMH4ezZxgNb9ry8uOyLLvTAMDpMa8ARANmFdD87T54XO8u36OlGw8qGLLUKilGU0fkaXC3dMX7Pl9Wv7z8zA90b0rRNK+6tU9St/ZJunl0njbsOKL5K/ZqSdE+LSnap06ZCRrdP1v9ClrJYUbWaggAzo9omlcAWjbmFYBowbxqGg6H+aUntDeq1O/QoYMkqaSkRDk5OZKkzZs3Ky8v74wD5eXlafPmzZowYYIkqbi4WKZpqn379mf8XJL447GJZfG7BxAdmFcAogGzCmhewpaltdvK9e7y3dq8+6gkKa9tokb3z1bvvFYyzYYz8qPxfR9N88phmuqRk6YeOWkqPVyl+Sv26OP1B/WX2RuU8p5HI/q21fCemfLFuOyOCuACiKZ5BaBlY14BiBbMK3s1+kz9yy+/XI8//rgeeugh7d27VzNnztRjjz12yu39fr8sy5JlWQoGg6qvr5fL5ZJpmpo0aZLuvfdeTZgwQW3bttXjjz+usWPHyuvl+nYAAAAAgOhV7w/po3X7tWDFHh2sqJXDNDSoS7pG9c9Wx4wEu+O1aFmt4nTLmEJNGp6j91eXatGqUk1/r0RzPtqpi7pnaGT/tkpP9tkdEwAAAACAUzIsq3HHVFRWVur+++/X4sWLFRsbqzvuuEM33nijJKl37956+umn1a9fP0nSZZddptLS0i88/vnnn9fAgQMlSS+88IKeeuopVVdXa9iwYfrlL3951svvB4Msv9+UDENKS4tXWRlLbACIbMwrANGAWQU0D0cq67Rw5V59sGafauqDio1xanivLI3o21bJ8R67450XzW1eBYJhLdt0UO8u36M9h6pkSOqZm6bR/bPVuV2SDMOwOyKAs9Tc5hWA5ot5BSBaMK+ahtP55cvvN7rUj0SU+k2PNy6AaMG8AhANmFVAdNu+r1LvLt+tFZsPK2xZSk/xaXS/thrSLUMet8PueOdVc51XlmVp8+6jmr98j4q2lcmS1K51nEb1z9bALulyOky7IwI4Q811XgFofphXAKIF86ppfFWp36jl9wEAAAAAgBQKh7W6uEzvLt+jbaXHJEmF7ZM1un+2uuekyuQM76hiGIYK2yersH2yDh6p0fwVe/TRuv3625ub9Nr7JbqsT5Yu6Z2leJ/b7qgAAAAAgBaMUh8AAAAAgK/gD4T0/upSzV+xV+WVdXI6DF3UPUOj+mcru/WZX04OkSc9xaevje6sicM66cM1+7Rg5V69vniH5v5rlwZ3baOxg9qpdbLP7pgAAAAAgBaIUh8AAAAAgNOwLEsrthzWq4u2qbyyTvE+l64c2kGX9mmrxFjO3m6OYmNcGjOovUb1z9bKLYf17vI9+rBon5as26/R/bM1fkgHeT3sTgEAAAAANB0+hQIAAAAAcAq7DhzXywuKVbz3mLwep667LFeX9cmSy+mwOxqagNNhamCXdA0obK0tu4/qlUVb9fbS3Vqy/oCuGdZJQ3tkcLkFAAAAAECToNQHAAAAAODfVFb7NfPDEi0u2i9JuqRXpq4e1kkJXFe9RTIMQwXtk/XTr/fXR+v2a8YHJfrH25u1aFWppo7MU352kt0RAQAAAADNHKU+AAAAAACSgqGwFqzYqzc+3qHa+pAK2iXp+hF5apceb3c0RADTNDSsZ6b6dW6tuR/v1PwVe/Srl1ZpQGFrTb4kV6mJMXZHBAAAAAA0U5T6AAAAAIAWzbIsFZWUa9rCrTpYUau0xBjdOqZQfTu3ksHy6vgPvhinplyWq+G9MjVt0TYt23RIq7eWaczAdhozsL08bi7PAAAAAAA4vyj1AQAAAAAtVmlZtV5ZuFUbdhyRx+XQpGGddPmAbLmcFLP4cukpPn332h5av6NcryzcpjlLdmrx2v2afGmOBhamc0AIAAAAAOC8odQHAAAAALQ4VbUBzf5oh95bVaqwZWlItza6ZniOkuM9dkdDlOnWMVU/uzVZ768u1eyPduivczZq0cpSTR2Zp44ZCXbHAwAAAAA0A5T6AAAAAIAWIxQO6/3V+zRr8XZV1wWVk5mgqSPz1SmT8hVnz+kwNbJftgZ1baPZi3fovdWlevC5FRraveFgkaQ4DhYBAAAAAJw9Sn0AAAAAQIuwYecRvbJgq0rLqpUU59btE7poYJd0mSyTjvMkzuvSjaPzNbx3pl5ZuFVL1h3Qii2HNX5we43uz2UdAAAAAABnh1IfAAAAANCsHayo0auLtmn11jK5nKbGD+mgsYPaKcbNR2JcGG1bxeme63ppzbYyTVu4TTM+2K4Pi/ZpyqV56pOfJoMDSQAAAAAAZ4A9GAAAAACAZikctvTWJ7s0Z8kOBUOW+hW01pRLcpSW5LU7GloAwzDUO6+VunVM1YKVe/TGkp364+vr1K1Tim4dU6jkeJbkBwAAAAA0DqU+AAAAAKDZOXy0Vk/P3ahte48pMy1WN43OV+d2yXbHQgvkcpoaM7C9hnTL0GvvbdOS9Qf0078t1devKFC/gtZ2xwMAAAAARAFKfQAAAABAs2FZlj5ef0AvzS9WnT+kkf3a6trhOXK7uJY57JUY69Y3x3dR7/xWevbtzfrTrPUa2r2NbhiZL6+H3TMAAAAAgNPjUyMAAAAAoFmoqg3ouXmbtXLLYSXFuXXXxO7q2jHF7ljAF/TJb6VOmQn6+1ubtGTdAW3ZfVS3je+i/Owku6MBAAAAACIUpT4AAAAAIOqt31Guv725Sceq/OrXuZVuvqJAcV6X3bGAU0qK8+h7k3vqvdWlmrZom379z1UaO6i9rrqoo5wO0+54AAAAAIAIQ6kPAAAAAIha/kBIr71fogUr9yrG7dA3xxVqSLc2MgzD7mjAlzIMQ5f1aavC9sn665yNevNfu7R+xxF9a0IXZaTG2h0PAAAAABBBKPUBAAAAAFFp14HjenruRu0rq1Z+20TdNr6L0pK8dscCzkhGaqx+fHNfzf5oh976ZJd+/o/lmnJZri7tncXBKQAAAAAASZT6AAAAAIAoEw5bmrdst17/cLsk6ZrhnTRmYHuZJgUoopPTYeqa4Tnq3ilVz8zdqBffLVbRtnJ9Y2yBEuM8dscDAAAAANiMC7UBAAAAAKJG2bFaPfLyar32folaJ3t1/839NG5wBwp9NAv52Un6+TcGaGi3Nlq3vVw/+dsyrSo+bHcsAAAAAIDNOFMfAAAAABDxLMvSJxsO6sX5W1RbH9KIvm01+ZIcuV0Ou6MB55XX49Q3x3dRz9w0PTdvs56cuU4X98jQ1JF5inGzGwcAAAAAWiI+DQIAAAAAIlp1XUAvvLNFyzYdUmKsW3dc1U3dOqXaHQu4oPoVtFZOVqL+/uZGLV67X1t2H9VtE7ooNyvR7mgAAAAAgCZGqQ8AAAAAiFgbdx7R397cpIrj9eqb30o3X9FZ8T633bGAJpEc79H3ruulhSv3avp7JXr4xZWaMKSDxg/pIKeDKyoCAAAAQEtBqQ8AAAAAiDiBYEgzPtiud5fvkcft0K1jC3RR9wwZhmF3NKBJmYahUf2y1aV9sp5+Y6PmLNmpddvLdfuErmqT4rM7HgAAAACgCXBYNwAAAAAgohyprNPDL67Su8v3KDcrUT//xgBd3COTQh8tWlarOP345n4aM6iddu4/rp8/u1yriw/bHQsAAAAA0AQo9QEAAAAAEWPb3mN64LkV2nnguMYPaa8f3thbrZO8dscCIoLLaWryJbm6d2pvuRymnpi5TnOW7JBlWXZHAwAAAABcQJT6AAAAAICIsLhonx55eZXq/EHdeXU3TRqWI4fJx1bgPxW2T9ZPb+mn7NZxmrV4h/48a73q/SG7YwEAAAAALhD2jgAAAAAAbBUMhfXS/GL94+3NSoz16L6v9VW/gtZ2xwIiWlqi98R7ZcWWw/rlCytVdrTW7lgAAAAAgAuAUh8AAAAAYJuq2oB+/2qRFq7cq4J2SfrJLf3ULj3etjxWOCwrUC+rrkrh6gqFKw8pfHS/rEC9bZmA0/G4Hbrjqq6aNKyTSg9X6YHnVmjzrgq7YwEAAAAAzjOn3QEAAAAAAC3T3kNV+sOMtSo7VqfL+mTp+hF5cjrO77HnVtCv0OEdCu3fotDBbbL8NVIoIIUCskLBhn8HA7JCASkUlKzTLWFuyIhPkyOlrczkLJkpbWWmZMlMzJDh4KM17GMYhsYP6aC2reL01zc26NFpazR1ZJ4u7Z0lwzDsjgcAAAAAOA/Y8wAAAAAAaHIrtxzWM3M3KhgK6+tXdNbwXlnn5Xktf61Ch0oaSvz9WxQ6tF0KBxvuNJ0yYuIkh0uGwynDFSPFxMtwuCSH89OvLsnpkmF++tXhkgyz4Yz9ilIFd6+Vdq3+/AcaDplJ6Q1Ff3JD0e9IbisjobUMk8Xx0HR65aXpxzf30xOvrdWL7xZrz6Eq3Tgq/7wfKAMAAAAAaHqU+gAAAACAJhO2LL2xZKdmf7RD8T6Xvjelp/Kzk876+ay6KoUObFXwQEOJHy7bJVnhhjtdMXJkFcrRprMcGZ3laNWhoaQ/B1YoqPCxAwof2atwRanCR/YqVFGq4PYVkpZ/vqHDJTMpU87s7nLmDpIjpe05/VygMbLSYnX/1/vpqdnr9cGafSotq9ZdE7srMdZtdzQAAAAAwDmg1AcAAAAANIk6f1DPzN2kVcWH1T49Xv81qbtSE2PO6DmscFDBXUUKlW5U6MAWhY/s/fxOT6yc7XudKPHN1GwZpuO8vgbD4ZQjpe1JJb0VrFe4Yr/CFXsV+qzwL9sl/5q58q+ZKzO5rZy5A+XKGSgzofV5zQT8uzivS/8zpaemv1eid5fv0YPPLdd/T+qh9m3i7Y4GAAAAADhLlPoAAAAAgAvu0NFaPTFjrUoPV2tAYWvdOrZQHlfjC3fLX6PApg/kXz9fVvURSZLhS5IzZ2DDWfhtOstMzpBh2LPUuOH0yNGqgxytOuiztQCscFihA1sU3LZUgR3L5V8+Q/7lM2S27iRXziA5cwbI9J39KgXA6ThMU9ePyFN26zg9N2+LHn5xpW4dW6iBXdLtjgYAAAAAOAuU+gAAAACAC2rTziP606z1qqkL6prhnTR2UHsZhtGox4aPl8m/fr4Cmz+QAnUyvIly95skV87AhuvWN/J57GCYppyZhXJmFsoz9GsKla5XYNtSBXeuUv2hf6r+k5flyCiQM3eQXB37yfDE2h0ZzczQ7hlqk+rTkzPX6ak5G7TnUJUmDesk04zc9w0AAAAA4GSU+gAAAACAC8KyLC1cuVevLNwmj9vUd6/toZ65aY16bOjQdvnXzlNwxwrJCstMbit3j8vlzB0kw+H66ieIMIbDKWe7XnK26yUrWK/g7iIFty1VcHeRQvs2qf6j5+Vo212u3IFytu8tw3VmlyUATicnM1E//Xp//fH1dXrrk13ae7hK35rQVb4YdgkBAAAAQLTgExwAAAAA4LwLBMN68d0tWrx2v9JTfPruNd2VkfrlZ6Jb4bCCu1crsPYdhQ4US5IcbbvJ3eMKObK6RvRZ+WfCcHrk6jRArk4DZPlrFNyxUoGSpQrtWavQ7jWS0y1X54vl7jlWZlyq3XHRDCTHe/TDG3rr+XlbtGT9Af3i+RX67rU91CbFZ3c0AAAAAEAjUOoDAAAAAM6rY1X1evL1dSoprVS3Tin6zpVd5Ys5/dn1VqBegeLF8q+bL6vyoGQ65cy/WO4el8uR0rYJkzc9w+2Tq/PFcnW+WOHaSgW3L1Ngy2IFNixUYNP7cuVfJHevcTITWtsdFVHO5XToG+MKlZ0er2mLturB51boO1d1VfdOHDgCAAAAAJGOUh8AAAAAcN7sPFCpJ2asU8Xxel0xsJ2uHZ5z2ut3h2uOKrB+gfyb3pPqqyVPrNy9J8jVdYRMX1ITJ7ef6U2Qu+tIubqMUGjvOtWvmqPA5g8U2LJYztzBcvceJ0dSpt0xEcUMw9Do/tnKSovVX2av12PTi3TtJTm6YkC7ZrMSBgAAAAA0R5T6AAAAAIDzYv32cj35+jpZlnT7hC4a3LXNKbezwkH5i+bJv2q2FArISEiXu98kufIvkuHyNHHqyGMYhpzZPeRo212h/ZvlXzVHwa1LFNz6sZyd+svde4Icqdl2x0QU69oxRfd/vZ+emLFO098r0ZFj9Zo6Kk8mxT4AAAAARCRKfQAAAADAOftk4wH9be4mxbgduvvansptm3jK7UKHSlT34T8UPrJXZmIbuQdOlrNdbxmm2cSJI59hGHJmFsqZWajQga2qX/2GgtuXKbh9mZztezeU+6072R0TUSo92acf39RXT8xYq4Wr9up4rV+3je8ip4P3IgAAAABEGkp9AAAAAMA5WbBij15esFVJ8R59f0pPZbWKO2kby1+r+uUzFNiwUDJNuftcJXevcTKcbhsSRx9Hmzz5xnxfocM75V/9hoI7Vyq4a7UcbbvJ3ecqOdvk2R0RUcjrcep7U3rqr3M2atmmQ6quC+quid0U42Z3EQAAAABEEj6lAQAAAADOimVZmrV4h974eKfapPh0z3W9lJoYc9J2wZ2rVbfkBVnVR+RIz5Nn2C1yJGfZkDj6OVp1kHf0fyt0ZK/8q+cquH2paveulyOjQO4+V8qRWci10XFGXE6H7ri6m154d4s+WLNPv3l5jf5ncg/F+zjgBgAAAAAiBaU+AAAAAOCMhcOWXnx3i95fs08dM+L1P5N7nlQChqsrVP/xSwruWCG5vPJc9HW5CofLMFje+1w5UtrKO+I7Cve9WvVr5iq49WPVvvmIHFld5RlyAwdN4IyYpqGbL++seJ9bcz/eqYdfXHXag3QAAAAAAE2PUh8AAAAAcEYCwbD++sYGrdxyWF07JOuuSd2/sFy3ZYUV2PS+6pdOlwK1cnbsJ8+QG2XGJtuYunkyk9rIe8ltCve5SvWrZilYvEQ1r/1Erq4j5Ol7tQxPrN0RESUMw9CkYZ0U73Pp5QVb9dCLK/X963opK42/IQAAAACwG6U+AAAAAKDRauuDenLmOm3aVaEBha31zXFd5HJ+fuZ96Eip6hb/Q+GD22TEpijm0m/J2aG3jYlbBjOhlbyX3K5QlxGq+/glBdbPV3Drv+TuP0mugktkmKyOgMYZ1S9b8V6X/vbmJv3qxZW6e3JP5WYl2h0LAAAAAFo0Sn0AAAAAQKNUVvv1+1eLtOvgcV3WJ0s3jMyXaTZcv90K+uVfM1f+NW9K4bBc3UbJ02+SDLfX5tQti6N1J/mu+rGC2z5R/dJXVf/R8wpsfE+eITfImVlodzxEiUFd2yjW69IfX1+n376yWnde3V09clLtjgUAAAAALRaH6gMAAAAAvtLho7V66MWV2nXwuK6+qKNuHPV5oR/cv0U1M34q/6o5MpMy5bv6J4oZciOFvk0Mw5Qrb4hir/uV3L3GK3xsv2rn/lq1859U+Phhu+MhSnTvlKr/ndpbLoepJ2as1b82HLA7EgAAAAC0WJypDwAAAAD4UnsPVenRV9eossqvm0bn69I+bSVJlhWWf9Uc+VfOlhxOuQdMlrvH5TJMPmpGAsMVI8+Aa+UqGKb6T6YpuGOFgrvXyN1jjNy9xstweeyOiAiXk5mo//e1vnp02ho9/cZGVdUENKp/tt2xAAAAAKDFYU8LAAAAAOC0ivcc1R9eW6v6QEjfubqb+he0liRZ9dWqXfSUQnvWykxtJ+/IO2UmtrE5LU7FTGgt7+j/VrB0o+o//qf8q99QoPgjeQZOkTNnkAzDsDsiIlhmWqzu+1pf/e7VNXp54VZV1vg1aVgn/m4AAAAAoAmx/D4AAAAA4JTWbC3To9PWKGRZ+p8pPU8U+qGyXaqe+TOF9qyVM3+ofFfdT6EfBZxZXeS75ufyDL1JVtCvukVPqWbOLxU6vNPuaIhwqYkx+tGNfdQxI0Fv/muXnpu3WaFw2O5YAAAAANBiUOoDAAAAAE6yZN1+PTlznTwuh34wtbe6dkiRJAWKl6hm9i9kVR+R56KbFTP8NhlOt81p0ViG6ZC76wjFXfdrubqOUPjQdtW8/nPVLX5OVn213fEQweJ9bv3v1F7q2jFFHxbt159nbVAgGLI7FgAAAAC0CJT6AAAAAIAveHvpLv3tzU1Kjvfovpv6qmNGgqxQUHUfPa+695+WERMn35X3yd3lMpbgjlJGTJxiht4k3zUPyJGRr8Cm91T96n0KlCyVZVl2x0OEinE7dfe1PTSgsLVWFR/W76YVqaYuaHcsAAAAAGj2KPUBAAAAAJIky7L06qJtmv5eibLSYnXfTX3VJsWncHWFat54WIGNi+TILJRv0s/laJ1jd1ycB46UtvKO/5Fihn9TVjiouoV/Vu283ylcecgbfAUAACAASURBVNjuaIhQToepb13ZVSP6tNWWPUf1yD9X6Vi13+5YAAAAANCsUeoDAAAAABQKh/X3Nzdp3rLdys1K1A9v7KPkeI+C+zapZub/KXyoRO6eY+Ude69Mb4LdcXEeGYYhV+eLFTvlYTnzhyq0Z52qp/9Y9WvekhXmLGyczDQM3TAqTxMv7qjdh6r08Asrdehord2xAAAAAKDZotQHAAAAgBYuEAzrjzPXa8n6A+qRk6p7ru+l2Bin/GvfVu2bv5EV9Ctm5F3yDJwiw3TYHRcXiOlNkPeS2+Ud/0MZccnyL3tVNTN/ptDBbXZHQwQyDEMThnbUzVd01uFjtXrohZXae7jK7lgAAAAA0CxR6gMAAABACxYIhvXH19dpzbYyDeqarv+a1F1uy6+6hX9S/SfTZCamyzfxp3J16m93VDQRZ2ahYq95UO4+Vyl8dL9qZv9StYufV6iu2u5oiECX9MrSHVd1U3VtQL95eTXFPgAAAABcAJT6AAAAANBCfVbory0p19BubXTbuC4yjh9QzawHFdy+XM6O/eS7+qdyJGXaHRVNzHC65ek3Ub5rH5QjI1+BjYu09y/fVaBkmSzLsjseIky/gta6c2I31dQFKfYBAAAA4AKg1AcAAACAFug/C/1bxxYqtGulal5/QOFj++UZeJ1iRt4lw+21Oyps5EjKlHf8jxQz/JuywkHVLviTauf9XuHjh+2OhgjTO68VxT4AAAAAXCCU+gAAAADQwnyh0O/eRreMKVBw3duqm/+kDIdL3nE/kLvnGBmGYXdURADDMOQuuFjZ3/6DXPlDFdqzVtWv/lj+ordkhYN2x0ME6Z3XSndN7E6xDwAAAADnGaU+AAAAALQgJxX6l3dWYOnLql/6qszkLPkm/UzOzEK7YyICOWIT5b30dnnH/UBGXLLql76qmpk/V+jgNrujIYL0yks7Uew/8k+KfQAAAAA4Hyj1AQAAAKCFCARDXyz0R+XK/96fFVg/X46MzvJdeZ/MuFS7YyLCObO6KPaaB+Xuc5XCR/epZvYvVffR87L8NXZHQ4T4rNivrafYBwAAAIDzgVIfAAAAAFqAhkJ/vdaWlOui7hm65bJ2qp/3WwV3rJCz0wB5x94rwxNrd0xECcPplqffRPmueVCOjHwFNi5S9av3KbB9uSzLsjseIsBJxf4hin0AAAAAOFuU+gAAAADQzP1noX/zsFaqm/uQQgeK5ep+uWJGfEeGw2V3TEQhR3KmvON/qJhh35AVCqhuwR9V+85jCh8/bHc0RIBeeWm6a9Knxf7LFPsAAAAAcLYo9QEAAACgGQsEQ3py5r8V+gNjVTf7FwpX7JNn0PWKGTxVhsFHQ5w9wzDlKhim2CkPy5k3RKHdRaqe/mP5i96WFQ7aHQ8265VLsQ8AAAAA54o9NwAAAADQTH1W6K/b3lDo39TLUu0bD8uqq1LMZd+Ru8cVdkdEM2J6E+S99FvyjvuBDF+y6pdOU83rP1foUInd0WCzz4r9Oj/FPgAAAACcDUp9AAAAAGiG/rPQ/1rnY6p7+1FJknfsPXLlDrI5IZorZ1YXxV77oNx9rlS4Yp9qZv1CdR+9IMtfY3c02KhXbprunPh5sb+HYh8AAAAAGo1SHwAAAACamUAwpCdmrjtR6N/YdpfqF/1FhjdBvqvukzOz0O6IaOYMp1uefpPku+YBOdrkKbBxoapfvU+B7ctlWZbd8WCTfy/2f0OxDwAAAACNRqkPAAAAAM3IZ4X++u1HdFH3NpqavFr+pa/ITM6S76r75UjJtjsiWhBHcpa8E34kz7BbZYUCqlvwR9W+85jClYfsjgab9MpN010U+wAAAABwRij1AQAAAKCZ+PdCf1j3Vrre856C69+VI6OzfFfeJzMu1e6IaIEMw5S7YLhipzwsZ+5ghXYXqXr6fapfMVNWsN7ueLBBT4p9AAAAADgjlPoAAAAA0Az8e6F/abdkTQ7PVWj7cjk79pN3zD0yPLF2R0QLZ3oT5L3s2/KO/6HMhDbyr5qj6mn/T4Hty1iSvwWi2AcAAACAxqPUBwAAAIAoFwiG9MSMhkJ/VNd4Tax9TeH9W+TqNkoxI++U4XTbHRE4wZlZKN81P5dnyNdkBepUt+BPqn3zEYWO7LU7GppYz9w0/dckin0AAAAA+CqU+gAAAAAQxU4U+juO6PIuPk2oni6rolSegVPkGXyDDIOPfYg8humQu9tIxV73K7kKLlFo32bVzPip6pa8KKu+2u54aEI9cij2AQAAAOCrsHcHAAAAAKJUIBg+UeiPKfRo3PFXZVUelufiW+TuOVaGYdgdEfhSpjdBMcNukW/i/8ls3UmBDQtUPe1H8m96X1Y4bHc8NBGKfQAAAAD4cpT6AAAAABCFQuGwnpqzQet3HNG4QqfGVE6TVXNUMZfeLnfhJXbHA86Io1UH+a78sWIuuV0yTNUvflY1sx5Q6OA2u6Ohifx7sf/otDU6WFFjdyQAAAAAiBiU+gAAAAAQZcKWpWff2qxVxYc1Ji+s0UenyaqvUczIu+TKG2J3POCsGIYhV/7QhiX5e4xR+Mge1cz+hWrfe1rhmqN2x0MT6JGTpm9f2U3Ha/z67ctrdKSyzu5IAAAAABARKPUBAAAAIIpYlqVXFmzVkvUHNLpdra6ofE0KBuS9/G65Ova1Ox5wzgy3VzGDrlPstb+QI7u7gluXNCzJX/S2rFDA7ni4wPp2bqVvjC1UeWWdHp22RpU1frsjAQAAAIDtKPUBAAAAIIrM/miHFqzcqxEZxzSudrYkyTv2Hjmzu9ucDDi/zKQMea/4vryX3y0jJl71S6c1lPubP5AVDtkdDxfQ0O4ZmjoiT/vLa/T7aUWqqQvaHQkAAAAAbEWpDwAAAABR4t1luzVnyU6NaH1YVwbelJwu+cb9QM6MznZHAy4IwzDkbN9bsZN/Kc+g66RAveo//Ieqp9+nwLZPZFlhuyPiAhnVP1tXXdRRuw4e1x9eK1J9gAM5AAAAALRclPoAAAAAEAUWF+3TK4u2aWTKXl0ZekeGxyffhB/J0bqT3dGAC85wuuXuMUaxU38jd9+JsmoqVbfoL6qZ8X8K7lwty7LsjogL4MqhHTSqX7aK9x7Tn2etVzDEQRwAAAAAWqZGl/qVlZW6++671bt3b1188cV66aWXTrvtsmXLNH78ePXs2VNTpkzR1q1bT9xnWZYee+wxDRs2TH379tX111+vtWvXnturAAAAAIBmbMXmQ3p23maNStqh8XpPhi9Jvgn3yZGSbXc0oEkZbq88fa9S3NTfyN1zrMLHDqr23cdVM/tBBUs32h0P55lhGLpuRK4u6p6htSXlembuRoXDHMABAAAAoOVpdKn/wAMPKBQKafHixXrqqaf0hz/8QZ988slJ21VUVOjOO+/Ut771LS1fvlwjRozQnXfeqWCw4fpnb775pl577TU9//zzWr58uUaNGqU777yTo+oBAAAA4BTW7yjXU3M26Ir4LRpvLpaZ0Eq+K++TmdTG7miAbYyYOHkGTlHs1Efk6jpC4bJdqn3zEdXM/bVCB7fZHQ/nkWkY+vqYzuqb30rLNh3SC+9uYR8SAAAAgBbH2ZiNampqNG/ePM2aNUtxcXHq0qWLJk6cqBkzZmjQoEFf2Hb+/Pnq0KGDrrzySknSbbfdpueee07Lly/X4MGDtWfPHvXt21cdOnSQJE2aNEmPPPKIKioqlJKScsYvwDDO+CE4B5/9vvm9A4h0zCsA0YBZha+yde8xPTlzrcb61mqkc7XM5Cz5xv2vzNgku6OhhYnUeeWITZL3opvk6XmF6lfOUaD4I9XM/oWc7XrK03+SHGnt7Y6I88DpMPXtq7rq8elF+mDNPsXGODX50ly7YyFCReq8AoD/xLwCEC2YV5GhUaX+zp07JUm5uZ9/YCooKNCzzz570rbFxcUqKCg48b3D4VBeXp6Ki4s1ePBgjR8/XvPmzVNJSYnat2+v6dOnq0ePHmdV6Ccnx57xY3B+pKbG2x0BABqFeQUgGjCrcCo79h3T49PXaLxnpYa718vdppMypv5EDl+C3dHQgkXsvEqLlzr+j/zlk1Xx4TRVb1yi4O4ixRYOUfLw6+VOzbI7Ic6Dn31riH7y1Md665PdSkuJ1eQR+XZHQgSL2HkFAP+BeQUgWjCv7NXoM/VjY79YoCckJKi6uvqU2yYmJp522/T0dPXv31/jxo2TaZpKTk7WM888c1bhKyqqFQqFz+qxODuG0fCmLS8/Lla7AxDJmFcAogGzCqdzoLxGv3pxhcY7lmiIu1iONnnyXPE9VdQYUs1xu+OhBYqeeZUgx8W3K7ZwtOqXz1T1po9VvflfcnboI3f30XK0yZfB6SVR7b8mdtOv/7laz7+1SVYopMv6tLU7EiJM9MwrAC0d8wpAtGBeNQ2Hw/zSE9obVer7fL6TCvzjx4+fVPR/tm1VVdVpt33iiSe0evVqLVy4UG3atNH777+vb3zjG5o1a5bS09MbE+cL+OOxh2XxuwcQHZhXAKIBswr/7khlnR59ZaXGGR9ooKdEjqyu8o7+ruTy8HcC20XLvDJT28t7xfcUOrhN9avmKLhjpYI7VspMbS9399Fy5gyQ4XDZHRNnwRfj0vev66WHX1ypF98pltft1KCubeyOhQgULfMKAJhXAKIF88peZmM26tChgySppKTkxG2bN29WXl7eSdvm5+dr8+bNJ74Ph8MqLi5Wfn7DkmjFxcUaO3assrKy5HA4NGLECKWkpGj16tXn8joAAAAAIOpVVvv16CurNDa0sKHQb9dT3svvluHy2B0NiEqO9Fz5xnxfsVMelqvLCIWPHVDd+0+r+p/3qH7lLIVrjtkdEWchMdate6/vpaR4j56Zu0lrtpbZHQkAAAAALqhGlfo+n0+XX365Hn/8cVVVVWnz5s2aOXOmJk2adNK2o0aN0o4dOzR37lz5/X4988wzio2NVf/+/SVJPXr00Lx583Tw4EFZlqUPPvhAe/bsOeUBAgAAAADQUtTUBfXYtJUa45+nvp6dcnboK++o/5bhdNsdDYh6ZlKGYi66SXE3/k6eQddJDpf8K2ep+p/3qPb9pxUq22V3RJyhtESv7r2+l3wxTv1p1npt3lVhdyQAAAAAuGAMy2rcQgmVlZW6//77tXjxYsXGxuqOO+7QjTfeKEnq3bu3nn76afXr10+StHTpUj3wwAPas2ePOnfurIceeuhEae/3+/XII4/onXfeUXV1tTIzM/Xtb39bEyZMOOPwFRXVCgbDZ/w4nD3DkNLS4lVWxnUzAEQ25hWAaMCswmfqAyE9/soKDTs2R13dpXLmDFTMpbfLMBt1xTTggmtu88oKhxTctVqBde8qdKBYkuTI6CxXt9Fytu8tw2zUORCIALsOHNcjL69S2JJ+MLW3OmYk2B0JNmtu8wpA88W8AhAtmFdNw+k0lZwce9r7G13qRyJK/abHGxdAtGBeAYgGzCpIUjAU1h9fW6nBZTNV6NovZ95QxQz/JqUiIkpznlehsp3yr5uvYMlSKRyUEZ8md9eRcnW+WIbn9DtUEDmK9xzVo9PWyO009aOv9VVWGv/fWrLmPK8ANC/MKwDRgnnVNL6q1GcvEQAAAADYJBy29I/Za3Rx2WsNhX7BcMVcQqEPNCVHWgd5L71dsTf8Vu4+V0lBv+o/eUVVL31ftYv+ouCuNbJCQbtj4kvkZyfprondVecP6dFXVuvw0Vq7IwEAAADAecWeIgAAAACwgWVZ+ufbRRqw/2XluQ7K2WWkYi6+RYbBxzTADqYvSZ5+ExV7w6OKueQ2manZCm77RLXvPKaqF76rug/+ruDeDbLCrBgYiXrkpOr2CV10rMqv376yWker6u2OBAAAAADnDRdoBAAAAIAmZlmWXl+wXr12vaD2rnI5ul2hmMHXyTAMu6MBLZ7hcMmVf5Fc+RcpfPywAiXLFCxZqsCWDxXY8qEMb4KcnfrLmTNIjvQcDsSJIAMK01VbH9Rz87bo0Wlr9MMb+ijO67I7FgAAAACcM0p9AAAAAGhiCz7epMKtf1dbZ4XMnhPkHTCJQh+IQGZ8K3l6jZOn1ziFju5TcNtSBUqWKrBhoQIbFsqIS5Wz0wC5cgfKTG3P+zgCDO+VpZr6oKa/V6LHpxfp3qm95XE57I4FAAAAAOeEUh8AAAAAmtCyVVvVvuivynAeldHrasUOuNruSAAawZGUKUe/iXL3vVrh8t0NZ++XLFVg7dsKrH1bRmK6XDkD5ew0UGZyJgW/jcYMbK+qmoDeXrpbT83eoLsmdZPDZEUFAAAAANGLUh8AAAAAmsjmTduVsvQJpTsrFep1jZIGTLA7EoAzZBiGHGnt5UhrL/eAyQofKlFg2ycKbl8m/6o58q+aIyM2RY7MQjmzCuXI7CIzLsXu2C3ONZfk6GhVvf614aBefLdYN1/emQMtAAAAAEQtSn0AAAAAaAJ7tu9SzAe/V6rjuGp7TlbrAePsjgTgHBmGIUd6rhzpubIG36DQ/s0K7lypUOkmBbcuUXDrkobtEtPlzOwiR1ahHBkFMr0JNidv/kzD0K1jC3Ws2q8P1uxTSrxHE4Z2tDsWAAAAAJwVSn0AAAAAuMDK9u6W5v9GyUaVjnadouyBY+2OBOA8M0xTzqwucmZ1kSSFa44qtG+TQqUbFSzdqMCm9xTY9J4kyUzNliOzi5yZhXJkdJbh9toZvdlyOkzdNbG7fv3SKr2+eIeS4j26uEem3bEAAAAA4IxR6gMAAADABXT8wG753/qVElSrgwXXKX/oGLsjAWgCpi9JZu5guXIHS5LClYcV3LdRodJNCu3bqMC6dxRY945kmDJbdZQzs1Dmp8v6G/FpMgyuAX8+eD1OfW9KT/3yhZV67u0tSox1q0dOmt2xAAAAAOCMUOoDAAAAwAVSe2CHauf8Wj6rXiUdJ6vPcAp9oKUyE1rJnTBcKhguy7IUrtin0L6NCu3bpOC+zfIfKvl8Y5dXjtRsmant5EhtJzOtnczkLBkOl30vIIolxnn0vSk99dALK/WnWev1wxv6qGMGl0AAAAAAED0o9QEAAADgAvAf2KrqN34rpxXQmszJumQ0S+4DaGAYhhwpWXKkZEndRskKhxWu2Ktw+W6FynYrXL5LofLdCh0oVuDEgxwykzM+L/o//WrExNn5UqJGRmqs7p7cU799ebUem16k+27qq/Rkn92xAAAAAKBRKPUBAAAA4DwLlG5S9Vu/k8KWFqdeqwnjOUMfwOkZpinHpyW9K7/hNsuyZFWVKVS+W+Gy3Q2Ff/luBbd+rODWjz9/bGyKzMR0mQmtZCSky0xofeI/w+216RVFptysRH37qq56cuY6/X5aQ7GfEOu2OxYAAAAAfCVKfQAAAAA4j4K7i1T9zhMKhAy9EzdRUydeIdMw7I4FIMoYhiEjvpXM+FZSh74nbrfqqhQ6skfhsoaSP3xkj0KHdyi0b9PJzxETLyMxXWZ8q0+L/0/L/oTWDfe1wNnUO6+VbhrdWc+/s0WPTS/SD27orRg3u8cAAAAARDY+tQAAAADAeRLYvlw1C/+i2pBTr7sm6NbJl8vpMO2OBaAZMWLi5MwslDILT9xmWZasuuOyKg8pXHlI4WMHG75WHpJ17KCCB7ed/ESuGJlxKTLiUmXGpsiIS5EZl9pw5n9ciozYFBnO5nkW+yW9s1RxvF5vfLxTf561Qf99TXdmNQAAAICIRqkPAAAAAOdBoHiJat9/RlXhGL0UHqdvXTdKXg8fuQBceIZhyPAmSN4EOdJzT7rf8teeKPnDlQc/Lf8PK1x9ROF9mxQKBU/9vDHxJ5f9/3YQgOFLkmE6LvTLuyCuvrijKqrq9dHa/Xpu3mZ9Y2xhi1y5AAAAAEB0YA8TAAAAAJwj/8ZFqv/oeR0Nx+pv9WP0na9dqqQ4j92xAECSZLi9cqS1lyOt/Un3nTjLv/qIwlXlsqqOKFx1RFZVucLVR2RVHVGwfLdkWad4YrOh2D9V8f/Z2f4Rusy/YRi6+fLOqqz2a8m6A0qOj9GkYZ3sjgUAAAAAp0SpDwAAAADnwF/0luqXvqqycIL+Uj1at19/sTJSY+2OBQCN8oWz/NM6nHIbKxySVXP008K//NMDAP6t+D/dEv+S5HCdKP3NhNYyE9NlJKTLTEiXmdDK1iX+nQ5Td1zVTY+8vEpzP96p5HiPLu2dZVseAAAAADgdSn0AAAAAOAuWZcm/cpb8q2brkJWiJypH6OtXD1BuVqLd0QDgvDJMh4y41P/P3n3HSVXf+x9/nzN9dmF32aUtSO9llyaKBRFQQEGplmhM0VjQADHJTfklMdHEm8RrFLBrmjcmRnqRKjYsSJWlLb23pWxhd/qc8/uDhHu9qCy4u2eWfT0fD/+QPfP9vpZHcpw5n5kzUnq2XGr/ucfYidjZw/7/9SaA5NGdSh7c/H9XlpGWJTPjX0P+jEanB/4Z/x74V/8dT3xelyaOzdfjf1ujvy3Zqsw0r3p2aFjt+wIAAADA+WCoDwAAAADnybZtRVe8rviGxTpiNNLk4gEae30+gyAAdZbh9srIaCIzo8nn/ty2bdnhUlmlR2WXFckqPSqr7Kis0iIlj+1W8tCWs9dMy5KZ0URmdgu5slvIzGkhM7OpDLNqL2fVT/Pq4Vvy9Zv/XqMX5m7SD2/rqXbNeYMWAAAAgNTBUB8AAAAAzoNtWYp+8Krihe/qiCtXTx3vr8H9OmgAt2wGgC9kGIaMYKbMYKbUtONnfmbbtuzIqX8N/I/+a+B/evD/74F//N8Hu9wys5rLldNCZnbL08P+7EtkePxfqa9RVlCTxuXr939fp8nT1+unX+/NV6kAAAAASBkM9QEAAACgkmwrqci7ryix42Md8bXSk4f7qW/3Fhp1dWun0wCg1jIMQ0agvsxAfanJZ2/vb9uW7LIiJY/vk3Vin5In9so6vk/x43v+9woyMhqf+TS/K7ulzJyWp9c7D62b1tcDI7tpyvQC/eGf6/X/7uqtzPTq/woAAAAAADgXhvoAAAAAUAl2IqbI2y8osWetjqV30BP7LlXnNo1019COMgzD6TwAuCgZhvk/t/Vv2/fMn1uhktND/uP7ZJ3Yq+SJfUrsWintWnnmGDOrmVy5neTK7Sx3004y/Onn3C+vbba+OayT/rRgi556Y71+fEcvBXxcPgMAAADgLF6VAAAAAMA52LGQwosnK3l4q042yNPjO/LUokmGHhjZVW6X6XQeANQ55r9u5e++JO/Mn9mxsJInD8g6vlfJop2nb9u/aZnim5ZJMmRmX3J6wJ/bWa6mHWR4g5+79lV5TVVcHtWs93fpmZkb9L1b8jnXAwAAAHAUQ30AAAAA+BJWqEThhU/KOrFfpc2u1GMb2ygn8/R3L/u9vKQCgFRheANyN2n/r1v4D5Zt27JLjypxaIuS//onvmGx4hsWS4YhM6e13P/6JL+rSQcZnv+51f7wfi1VfCqqd9cd1J8WbNF3hnfhriwAAAAAHMMVKAAAAAD4AlZZkUJvPiH71DGFOt2oX6/KUdDv1vduzVf9NK/TeQCAL2EYhozMJvJmNpG6XCvbtmUVHzoz4E8cLlTs2C5p/QLJdMnVsI1czbrI3aaPzKzmuvO6Diotj2rFpqNqmBHQqP5tnP6VAAAAANRRDPUBAAAA4HMkj+9VeOGTsiOnlOhzh377YUCWldCEMXlqnPX5t2wGAKQuwzDkatBMrgbNpG6DZduWrBP7Tw/4D21R8vA2JY9uV2ztHJkZTeRu21f3XNVbT5RFNO+jPWqYGdBVeU2d/jUAAAAA1EEM9QEAAADg/0gc2qLw4smSlZB5zQP6w3JbpRUVemBkN7VrnuF0HgCgChiGKVdOS7lyWsqbN1S2lVTyyDYldq1SYvdqxdbOldbO1aT6jbU8M1dLlpQou9416tw62+l0AAAAAHUMQ30AAAAA+F/iu1Yp8vaLkssj75Dv65kPIjp4rFhjB7TVpZ0aOZ0HAKgmhumSO7ez3LmdZV9xp5JHtp4Z8Pc316l//XUqWvyejne5QlndrpCZ1VyGYTidDQAAAKAOYKgPAAAAAP8S2/yOoh+8KiNQT/5h39ffVoW0aU+x+ufnathlLZzOAwDUEMM0zxrwF63/QIG9a+QrXKhQ4cIzt+h3t7mUAT8AAACAasVQHwAAAECdZ9u2YmvnKrZmloz6jRS84QdauCmk99cfVtfWDXTn9R0Y1gBAHfXvAX9ubmct/3SYlr/1vvpnHlJ+dN/p/3asnSszK1eezgPl6XClDG/A6WQAAAAAFxmG+gAAAADqNNuyFP3oNcU3L5OZ3VKBYQ9r9d6IZry3S80bpmn8yG5yu0ynMwEAKeDqHs1VVHqZ/vTxXvVuf53uHeRRcvdqxXesUPSjvym6aro87a+Up+sgubJync4FAAAAcJFgqA8AAACgzrKTcUXeeVmJXSvlyu2swPUTtKMoqlfmb1FGuleTxuUr4ONlEwDgf4zq30bHSsJauaVIM7Mu0a0D75Lv8lsV37FC8U1vKb55meKbl8mV21meroPlbtlDhulyOhsAAABALcbVKQAAAAB1kh0LK7x0qpIHN8vduo/8A+9TUVlcU2dskMs0NGlsvhrU9zudCQBIMaZh6O4bO+vkqagWr9yvRpkBXdurubydrpGnY38lj+5QfNNbSuxareShLTLSGsjTZaA8nfrLDNR3Oh8AAABALcQ9JAEAAADUOVa4TKH5v1Py4GZ5ugyUf9B4VcSkp99Yr4pIXPfd3FUtm9RzOhMAkKI8bpe+O7q7GmUF9Lel21Sw87gkyTAMuZu0V2DQA0q740l5e4+SbEuxVdNV8drDCr/zspJFuxyuBwAAAFDbMNQHAAAAUKdYpUcUmvsbWcf3yNt7pHxXfl0JS3pmRoGOFof1tcEd1KNdjtOZAIAUVy/o1ffG5Svoc+v5OZu07+ipz/zcDGbK1/tmpX3tv+QfNF6uRm2U2P6hQrMfVcWsRxXf9qHsRMyhUJAo6wAAIABJREFUegAAAAC1CUN9AAAAAHVG4vBWVcx+THZZkXxX3SVf75GSpD8v2KJtB0o1uE9zDerd3OFKAEBt0bhBUN8dk6dk0tLT09brZFnkrGMM0y1P274K3vRTBcc8Jk+nAbKKDyjy7suq+McPFNuwhOE+AAAAgC/FUB8AAABAnRDf/pHCbz4hWUkFhnxP3i4DJUmzlu/Wis1H1bN9jm4b2N7hSgBAbdPhkkx9+8bOKimPafL0AoWjiS881pV9ifz9v6n0O56S7/LbJcNU9OO/q+KfP1Js89uyk1/8WAAAAAB1F0N9AAAAABc127YVXTNHkXdekhGor+BNP5W7RZ4kaXnBIc3/aI9aNamne0d0lWkaDtcCAGqjy7s00aj+bbS/qFzPz9mopGV96fGGL03evCFKu+338vW7XbKSin7wqire+LHiW5fLtpI1VA4AAACgNmCoDwAAAOCiZSfjirz7smJrZsnMbqngyJ/Lld1CkrRlz0m9umirsuv7NHFsnnxel8O1AIDabHi/lroqr6k27jqp15Zul23b53yM4fbK232I0m57Qt6+42THwoq890dVTPup4js+ln2ONwcAAAAAqBvcTgcAAAAAQHWwI+UKL52q5OGtcrXoocCg+2V4/JKkg8cr9MysjfJ6TE0al6+MdJ/DtQCA2s4wDN01pKNOlEb07rqDapQZ0NDLWlTusR6ffD1ulLfLtYptWKJYwWJF3n5R5rr58vYZJXer3jIM7iYDAAAA1FV8Uh8AAADARccqPaqKOb9W8vBWebpdp8D1E84M9EsrYpo8bb1i8aTGj+quZg3THa4FAFws3C5TD47qptycNL3xzg6tLiw6r8cb3qB8vUcq/fYn5O1xo6xTxxRZ+oxCs36pxL5PK/XpfwAAAAAXH4b6AAAAAC4qiSPbFZr9mOyyo/Jdcaf8V9whwzz90icaT2rK9PU6XhrRXUM6qmurBg7XAgAuNkG/R5PG5al+mlcvz9+snYdKz3sNw58uX99xSrvtCXm6D5FVfFDhRU8rNOfXShzYxHAfAAAAqGMY6gMAAAC4aMR3rFB4/u9kJ+MKDJkob7fBZ35mWbZenrdZuw+f0o39Wurq/FwHSwEAF7OcjIAmjs2TIWnK9AIdKwlf0DpmMEP+frefHu53GSjr+B6FFzyh8Ju/V/LkwaqNBgAAAJCyGOoDAAAAqPVs21Z07VxF3n5Bhj9dwZt+KneLHp855o13dmjttmPq27mRRvVv41ApAKCuaN20vu69qavKQ3E9PW29KiLxC17LTMuS/6q7lHbrb+XucLWShwoVmvELRT95Q3Y8WoXVAAAAAFIRQ30AAAAAtZqdTCjy3iuKrZ4pM7uFgiN/IVdOy88c8/baA1qyar/aNc/Q3Td2lmkYDtUCAOqSXh0a6taB7XT4REjPztygRNL6SuuZ9RoqMOBuBUf+XGZ2c8XWL1DFGz9RfM8abskPAAAAXMQY6gMAAACotexIucIL/kuJbR/K1SJfwZt+KjO9wWeO2bjrhP6+dLsaZQb03dHd5XG7HKoFANRF1116iQb2aqbCfSV6dfHWKhm+uxq1UXDkI/JdcafsWFiRJVMVXvy0rLJjVVAMAAAAINW4nQ4AAAAAgAuRLDmk8OIpskuPyNN1sHz9vibD/Oz7lg8eK9fzczbK73Vp4rg81Qt6HaoFANRVhmHo9sHtdawkog8KDqtpg6CGXd7y3A8817qmKW+3wXK36aPoiteV2LFCFQc3y9tzhLz5w2S4PFVQDwAAACAV8El9AAAAALVOYu+nCs16THZZkXxX3CH/lXeeNdAvq4hp8vQCxeKWHhzVTU2z0xyqBQDUdS7T1P03d1WznDRNf3en1mytuk/Um8FMBQber8CN/yEzPVux1TMVmv5zJQ5urrI9AAAAADiLoT4AAACAWsO2bUXXzlV48WQZpkuBG34gb7frzjounkhq6swCHS+N6OtDOqpzqwafsxoAADUn4HNr4tg81Qt69PK8TdpzpKxK13c366Lg2MfkvXSMrPITCr/5e4WXvSArVFKl+wAAAACoeQz1AQAAANQKdjyiyFvPKrZ6pswGzRUc9YjczbqcfZxt608LCrXzYJmG9m2h/vm5DtQCAHC2nMyAHhqTJ8uWJk8v0MmySJWub7g88vUcobRxj8vVIl+JnStU8c+fKLZxqWwrWaV7AQAAAKg5DPUBAAAApDyrrEihOb9WYvdqudv0VfDmn8ms3/Bzj5374R59svmoerbP0dgBbWu4FACAL9euWYbuvrGzSstjmjK9QJFYosr3MOs3VGDIJPmvnyDDF1T0o9cUmvWokkW7qnwvAAAAANWPoT4AAACAlJY4sEkVs34l6+RBefuOlX/QAzI8vs89dsXmI5rzwW61aJyue0d0lWkaNVwLAMC5XdalsUZe1Vr7isr10tzNsiy7yvcwDEOeVr2UNu5xefNvkHXygEJzHlN09UzZVtW/kQAAAABA9WGoDwAAACAl2batWMEihRf+l2RbCgydJF+P4TKMzx/U7zhQqj+9WajMdK8mjs2Xz+uq4WIAACpvxJWtdHnXxvp0x3FNf3dnte1jeHzyXXaLgmMeldngEsXWzlVozuOySo9W254AAAAAqhZDfQAAAAApx07EFHnnJUVXvC4zo4nSRj4id4v8Lzz+WElYU2cWyDSliWPzlVXv8z/JDwBAqjAMQ98a1kntmmVo0cp9en/9oWrdz9WgmYIjfy5P3jBZx3arYsYvFCt8T7Zd9XcJAAAAAFC1GOoDAAAASClW+QmF5j6uxI6P5WrRQ8GRv5CZ2eQLjw9FEpo8vUDlobjuHdFVLZvUq8FaAAAunMft0kNjuisnw6//XrxVm/ecrNb9DJdH/stvVWD4f8jwpSn6/p8VWfqM7Eh5te4LAAAA4KthqA8AAAAgZSQOb1Vo1q9kHd8jb6+bFBgyQYY38IXHJy1Lz8/ZqEPHKzT22rbq1aFhDdYCAPDV1Q96NXFcvrweU8/N2qjDJyqqfU93bmeljX1M7jZ9ldizRhXTf6bEgY3Vvi8AAACAC8NQHwAAAEBKiG1+W+H5v5cdj8p/3UPy9Rktw/jylyz/eGu7Nu0+qavzmmpo3xY1VAoAQNVqlpOm8SO7KxJLavK0Ap0Kxap9T8OXJv+gB+Qf8B3Z8YjCC/5LkY//ITtR/XsDAAAAOD8M9QEAAAA4yk7EFHn/z4p+8KqM9Aanv++3dZ9zPu6t1fv19tqD6tQiU18f0lGGYdRALQAA1aNr6wa64/oOKioJ69mZGxRPWNW+p2EY8nS4UmljHpXZuJ3iGxYrNPtRJU8eqPa9AQAAAFQeQ30AAAAAjrHKihSa8xvFC9+Tq1lXpY16RK4Gzc/5uIKdx/WPZdvVuEFQD47uLreLlzYAgNrv2p7NdF2fS7TtQKn+uqhQtm3XyL5m/UYKjviJvL1HySo+pNCsXyq2YYlsu/rfWAAAAADg3NxOBwAAAACom+K71yjy3itSLCJvr5vl7XWzDPPcw/kDReV6Yc4mBX1uTRqbpzS/pwZqAQCoGbcObKejxSF9tPGImjQIavgVrWpkX8N0ydf7Zrmbd1X4nZcU/fjvSuwvkH/APTKDmTXSAAAAAODz8XEWAAAAADXKthKKfPwPRZZOlWG6Fbjh+/L1GVWpgX5peVSTp69XPGHpodHd1bhBsAaKAQCoOaZp6L6buqp5w3TNfH+XVhUW1ej+rsbtlDb6V/J0vFrJAxsVmvYzJfasq9EGAAAAAJ/FUB8AAABAjbHKTyg077eKb1gsV+P2Co7+ldzNu1XqsbF4UlNnbtCJsqi+MbSTOrbIquZaAACcEfC5NXFsnuqnefXK/M3afbisRvc3vAH5r7lb/sEPypat8JLJiq6cLtvidvwAAACAExjqAwAAAKgRif0FCs14RNbRHfLkDVVgxI9kpjeo1GMt29Yf39yiXYfKdMPlLXVVXtNqrgUAwFnZGX5NGJMnSZoyvUAnSiM13uBpc6nSxjwms2FrxT6dr/Dip2VHK2q8AwAAAKjrGOoDAAAAqFa2ZSm6aobCC5+SbSflv36C/JffJsN0V3qNOct3a1VhkXp3bKjR17SpxloAAFJHm9z6+s7wLiqtiGny9AKFo4kabzDTGyg44idyd7hKyf0Fqpj9qJLFB2u8AwAAAKjLGOoDAAAAqDZWqEThBU8otm6ezJyWp7+jt1Wv81rj441HNO+jPWrVpJ7uGd5FpmFUUy0AAKmnT6dGGt2/jQ4cK9eLczfJsuwabzDcXvmvuVu+K+6UXXZModmPKb57TY13AAAAAHUVQ30AAAAA1SJxqFChGY8oeWiLPF0GKnjTT2XWb3Rea2w/UKI/L9yirHo+TRibJ5/HVU21AACkrhv7tdSV3ZqoYOcJ/fPtHY40GIYhb7fBCtz4HzJcHkWWTlV01QzZtuVIDwAAAFCXMNQHAAAAUKVs21J03XyF3/yd7HhE/oH3y3/VXTLc3vNap6gkrKkzNshlmpo4Nk+Z6b5qKgYAILUZhqG7hnZSh+YZWrp6v95Z59zt7925nRQc/UuZOa0UWzdP4cWTZcdCjvUAAAAAdQFDfQAAAABVxo6UK7x4smKrpsvMzFVw9CPytLv8vNcJRRKaPG29KsJx3XdTV7VoXK8aagEAqD08blMPjclTo8yAXluyTZv2nHSsxUzPVvCmn8rd/kol961XxaxHlSw+5FgPAAAAcLFjqA8AAACgSiSLdqpi5iNK7lsvd/srFRz5C7kyc89/HcvSC3M26vCJkMZd20492udUQy0AALVPesCjiePy5PO69NysjTp8osKxFsPtlX/APfL1+5rssiKFZj+q+J61jvUAAAAAFzOG+gAAAAC+Etu2Fdu4VKG5j8sOl8nX/1vyD7hHhufCbpf/+ls7tHH3SfXPb6ohfS+p4loAAGq3ptlpGj+ym6KxpCZPK1B5OO5Yi2EY8na/XoEbfyjD5VFkyRRFV8+SbVuONQEAAAAXI4b6AAAAAC6YHQsrsuw5RT96TUZ6joIjfy5vp2tkGMYFrbdszQEtW3tAnVpk6s7rO17wOgAAXMy6tm6gO67voKKSsJ6ZuUGJpLNDdHduZwVHPSIzu6Via+cosmSq7FjY0SYAAADgYsJQHwAAAMAFSZ7Yp4qZv1Ri1yq5W/dR2uhfypXd4oLX27jrhP7x1nY1zgpo/Kjucrt4uQIAwBe5tmczDe7dXNv2l+jVxVtl27ajPWa9HAVv/n9yt+unxN51Cs36lZIlhxxtAgAAAC4WXCUDAAAAcF5s21as8D2FZj8mu/y4fFfcIf/gB2V4Axe85sHjFXp+zkYFfC5NHJev9ICnCosBALg43Tqonbq3ydYHBYe1aOU+p3NkuL3yX3uvfJffLqusSKHZjylxaIvTWQAAAECtx1AfAAAAQKXZ8agi772i6Pt/lhGor+BNP5W323Vf6Tb5ZaGYJk9br1jc0viR3dSkQbAKiwEAuHi5TFP339xVzXLSNP2dnVq77ZjTSTIMQ968IQoMe1iybYUXPKn4zpVOZwEAAAC1GkN9AAAAAJWSLDmk0OxHldj2oVyX5Clt9K/katT2K60ZT1h6ZuYGHS+N6OtDOqpzqwZVVAsAQN0Q8Lk1cWye0oMevTRvk/YeOeV0kiTJ3bybgiN+IsOfrsiy5xTbsNjpJAAAAKDWYqgPAAAA4JziO1YoNPNXskoOydt3rAJDJ8nwp3+lNW3b1l8XFWrHgVJdf+kl6p+fW0W1AADULTmZAX13dJ4sy9aUGQUqPhV1OkmS5MppqeDNP5OZ2VTRj/+hyIrXZduW01kAAABArcNQHwAAAMAXspNxRT54VZG3X5Dh8Stw44/k6zFchvHVX0osWLFXH208ovy22brl2nZVUAsAQN3VrnmGvnVDZxWfimrqjAJF40mnkyRJZr0cBW/6f3I1bq94wSJF3n5RdjLudBYAAABQqzDUBwAAAPC5rLIiheb8RvHNb8uV21nBMY/KndupStZeXVikGe/tUvOG6br3pq4yTaNK1gUAoC7r17WJhl/RSnuOnNIf52+WZdtOJ0mSDH+6Ajf+UO5WvZXY+YnCC56UHQs5nQUAAADUGgz1AQAAAJwlvmetKmY+Iuv4Hnl7jlDghh/KDGZUydp7jpTplfmbVT/Nq4lj8xTwuatkXQAAII28urX6dGqk1VuPafby3U7nnGG4vfIPflCeLoOUPFyo0NzHZVUUO50FAAAA1AoM9QEAAACcYVsJRVa8rsiSKTIMlwLDHpbv0jEyzKp56VB8Kqop0wtk2dJ3x3RXdoa/StYFAACnmYahu2/srFZN6mn+R3v08aYjTiedYZimfFfeKW/fsbJOHlBo9mNKnjzodBYAAACQ8ip9Za6srEwTJ05Uz549dfXVV+u11177wmNXrlyp4cOHKz8/X7fccou2b9/+mZ8fPHhQ48ePV69evdS3b1/96Ec/uvDfAAAAAECVsMpPKjzvd4oXLJLZuJ2CY34l9yV5VbZ+NJbU5OnrVVIe0z3DO6ttbtV88h8AAHyWz+PShLF5yqrn058XbNH2AyVOJ51hGIZ8PYbLP+A7skOlCs39jRKHtzqdBQAAAKS0Sg/1H330USWTSS1fvlwvvviipkyZohUrVpx1XHFxscaPH697771Xq1at0qBBgzR+/HglEglJUjwe17e//W316NFDy5cv1wcffKC77rqr6n4jAAAAAOctcWCjQjMfUfLodnm6D1FwxI9lpmdX2fqWbeuleZu072i5Rl7VWn07N66ytQEAwNky032aODZPLtPUMzM36HhJ2Omkz/B0uFKBYd+TbEvhBU8ovmuV00kAAABAyqrUUD8UCmnRokWaNGmS0tPT1aVLF40aNUozZsw469ilS5eqVatWuummm+T1enXPPfeooqJCq1adfmI+e/ZsZWVl6d5771VaWpq8Xq+6du1atb8VAAAAgEqxLUvR1bMUXvCk7GRC/uu+K3+/22WYVfs99zPf26V124/rsi6NNeLKVlW6NgAA+HwtGtfTvSO6qDwU1+TpBQpHE04nfYa7eTcFR/xEhjdNkbeeU2zjUqeTAAAAgJRUqSt1e/bskSS1a9fuzJ916tRJf/nLX846dtu2berUqdOZf3e5XGrfvr22bdumfv36ad26dWrevLnuuecebdiwQa1atdIPf/hD9enT54J+AcO4oIfhAv3775u/dwCpjvMVgNrA6XOVFS5TeNmLSh7cJDOnpYKDH5SZ0ajK9/mg4LAWrNirtrn1dfeNnWSanJyB2sbp8xWAC9erY0ONu7at3nhnp16Ys0kTx3WXy6z0zTurnbthS6WN/JlCC55U9KPXZFeclO+ycTKMC2vkfAWgtuB8BaC24HyVGio11A+FQkpLS/vMn9WvX18VFRWfe2xGRsYXHnvkyBF98skneu6553TFFVdo9uzZGj9+vJYsWaLMzMzzis/KSjv3QagW2dn1nE4AgErhfAWgNnDiXBXet1lFs/6gZHmx6vW8XtnXf0um21vl+2zceVx/XVSohlkBPXJvP2XV81f5HgBqDs+tgNrpzhu7qrgirqUr92nOR/t078juTid9Vk49Jb/9nzryxm8VXb9Q3mSFckY8JMN0XfCSnK8A1BacrwDUFpyvnFWpoX4wGDxrgH/q1KmzBv3/Pra8vPwLj/X7/erVq5euueYaSdK4ceP08ssv69NPP9WAAQPOK764uELJpHVej8FXYxin/0974sQp2bbTNQDwxThfAagNnDhX2bal2PpFiq6cLrnc8l97r4wOV+hkSVRStEr3KioO6Td/XSO3y9R3R3dXMhrX8Wi8SvcAUDN4bgXUfrcMaKN9R8o0b/kuZQbdGtirudNJ/4ch75CHlVz2gso3vq9IRUiBQffLcJ3fVwJxvgJQW3C+AlBbcL6qGS6X+aUfaK/Us+JWrVpJknbu3Km2bdtKkgoLC9W+ffuzju3QoYOmTZt25t8ty9K2bdt03333SZI6duyoNWvWVPoXOBf+x+MM2+bvHkDtwPkKQG1QU+cqO1Ku8LuvKLnvU5mZufJf96BcWc2qZe9QJK6npxWoIhLXhDF5at4wnfMxcBHguRVQe7lMUw+O6q5f/3W1XluyXY0yg+rauoHTWZ/l9sl/3YOKvP2SErtWKrRkqgKDH5RxAXcT4nwFoLbgfAWgtuB85axKfTlVMBjUkCFDNHnyZJWXl6uwsFAzZ87U6NGjzzr2uuuu0+7duzV//nzFYjG98sorSktL06WXXipJGjlypAoKCvTBBx/IsizNmjVLZWVl6tmzZ9X+ZgAAAADOSBbtUsXMR5Tc96nc7fopOOoXcmU1q5a9EklLz8/eqMMnQrp1YHvlt8upln0AAMD5SQ94NHFcnvxel56bvVGHjp/91ZpOM0y3/APvl7v9FUruW6/wkimyE1V7NyEAAACgtqnUUF+SHnnkEUnS1VdfrXvuuUcTJkxQv379JEk9e/bU6tWrJUlZWVl69tln9fzzz6tPnz5aunSpnnvuObndp28K0LJlSz355JN67LHH1Lt3b/3973/XCy+8oIyMjKr+3QAAAIA6z7ZtxTa9pdDcx2WHS+W7+punb7nvqZ7vtrdtW39/a7s27SnWNT1ydV2fVLu1LwAAdVvT7DSNH9VN0VhSk6ev16lQzOmksximKf8198jTqb+SBzYqvOhp2fGI01kAAACAYwzbrr03SigurlAiYTmdUacYhpSTU0/Hj/O9GQBSG+crALVBdZ+r7HhUkff/pMTOT2TUb6TA4AflymlZ9Rv9L0tW7dfry7arS6ssTRqXL7er0u8jBpDCeG4FXHze/fSgXl20Ve2bZ+gHt/WUx516/822bUvRD19TfPMyuRq3V2DYwzK8gS99DOcrALUF5ysAtQXnq5rhdpvKykr7wp+n3rN1AAAAAF+ZVXpUodmPKbHzE7lb9Vba6F9W+0D/0x3H9c9l29U0O6jxI7sx0AcAIIUN6NFM1196ibYfKNVfFm5RKn7uxzBM+a68U57uQ5Q8ul2hN5+QHU29rwwAAAAAqhtX2QAAAICLTGJfgSpm/UpW8UF5Lx0r/3UPyfAGq3XPfUdP6cU5m5QW8Gji2DwF/Z5q3Q8AAHx1t1zbTj3a5ejjTUc1/6M9Tud8LsMw5Lv8Nnl7DJd1bJdC838vO1LudBYAAABQoxjqAwAAABcJ27YUXTtX4UVPSYahwLCH5es5XIZhVOu+JeVRTZ5eoKRl6aHR3dUoq3rfQAAAAKqGaRq696YuatEoXbOW79bKLUedTvpchmHI13esvH1GyTqxV6H5v5UVKnU6CwAAAKgxDPUBAACAi4AdCyuyZKpiq2fKbNBcaaMekfuS7tW+bzSe1JTpBSo+FdW3hnVWh0syq31PAABQdfxetyaMzVNGulevzN+inQdTd1ju63WzvH1vkXXygMLzfyurotjpJAAAAKBGMNQHAAAAarlkySGFZv1Kib3r5G57uYI3/0xm/UbVvq9l23pl3mbtOXJKw69opX7dmlT7ngAAoOo1qO/XxLF5Mk1p6owCHS8JO530hXw9bpDvijtklRxWaN5vZZWfcDoJAAAAqHYM9QEAAIBaLL5njUKzHpVVViRfv9vlH3ifDI+vRvae+d4urdl2TJd2aqSRV7eukT0BAED1aNWkvr4zvKtOheKaPKNA4WjC6aQv5O12nXxXfUN22VGF5v2nrLJjTicBAAAA1YqhPgAAAFAL2Zal6KoZiiyZKsPlUeDGH8rbfYgMw6iR/ZcXHNKCFXvVJre+7r6xs8wa2hcAAFSf3h0bauyAtjp4rELPz9mopGU5nfSFvF2ulf+au2WfOnF6sF96xOkkAAAAoNow1AcAAABqGTtaofDipxRbN09mw9YKjv6l3Lmda2z/rfuK9eqircqu79N3x+TJ63HV2N4AAKB6Db2sha7Ka6qNu07q9bd2OJ3zpTwdr5Z/4L2yQyUKzf1PJYsPOp0EAAAAVAuG+gAAAEAtkjyxXxUzf6nk/g3ydLxawRE/kZmeXWP7Hz0Z0jMzN8jjNjVxbL4y0rw1tjcAAKh+hmHoriEd1alFppatPaBlaw44nfSlPO36yT/oAdmRcoXn/07J4kNOJwEAAABVjqE+AAAAUEvEd6xQaM5jsitOynfVXfL1/7YMd80N1cvDcT09bb1C0YTuv7mbmjdKr7G9AQBAzXG7TI0f1V2NGwT197e2qWDnCaeTvpSnzaXyXzdedqRcofm/V7yYW/EDAADg4sJQHwAAAEhxtmUpsuKfirz9ggxvUMHhP5a3y0AZNfg99omkpedmbdDR4rBuH9ReeW1r7u4AAACg5qUHPJo0Nk9Bn1svzNmoA0XlTid9KU+r3vIPvE92uFSHX/ulrFOp/UYEAAAA4Hww1AcAAABSmB2PKLJ0quIFC2U2bqfg6F/K1aR9zTbYtl5dtFWF+0o0qFdzDe5zSY3uDwAAnNG4QVAPje6ueMLS5OnrVVoedTrpS3naXib/NXcrUXpMofm/lxUqcToJAAAAqBIM9QEAAIAUZZWfVGju40rsXSd3u34KDv+RzGBmjXcs/GSfPthwWN3bZOu2we1qfH8AAOCcji2y9I2hnXSiLKqpMzcoFk86nfSlvB2vUs7Qe2WVHVX4zd/LCpc5nQQAAAB8ZQz1AQAAgBSUPL5HodmPyjqxT94+o+S/9l4ZLk+Nd6zZWqTp7+5Us4Zpuv/mrnKZvIQAAKCuuSqvqW64vKV2HSrTH9/cIsu2nU76UvV7D5Gv3+2yig8pvOAJ2dEKp5MAAACAr4QrcgAAAECKie9Zq9Dcx2VHy+UfeL98vW6WYRg13rH7cJlenrdZ9YMeTRybp4DPXeMNAAAgNYy+po16d2yoVYVFmr18t9M55+TLGyLvpWNkndiv0IInZcfCTicBAAAAF4yhPgAAAJAibNtWrGChIkumynD7FBz+Y3naXe5Iy8myiKbMKJBlS98dk6ecjIBDW8BZAAAgAElEQVQjHQAAIDWYhqF7hndRqyb1NP+jPfpo42Gnk87J13OEvD1HyDq2S+FFT8mOR51OAgAAAC4IQ30AAAAgBdhWQtHlf1F0xT9lZjVVcOQv5GrszPfXR2IJTZleoNLymO4Z3lltm2U40gEAAFKLz+PShLF5alDfp78sLNS2/SVOJ52Tt89oeboPUfLINoWXTJadiDmdBAAAAJw3hvoAAACAw+xohcIL/6B44XtyNe+m4M0/k1m/oSMtlmXrpbmbta+oXCOvbq2+nRs70gEAAFJTZrpPE8bkyeUy9czMDSoqDjmd9KUMw5Dv8tvk6TJQyYObFV76jOxkwuksAAAA4Lww1AcAAAAcFC8+oorZv1by4GZ5Ol+rwNDvyfAGHet5450d+nTHcfXr2lgjrmjlWAcAAEhdLRrX0303dVVFJK7J0wtUEYk7nfSlDMOQ78o75e5wlZL7CxRZ9rxsK+l0FgAAAFBpDPUBAAAAhySObNfBv/xEVskR+frdLt9Vd8kwXY71vLP2gJas2q92zTP0zWGdZRiGYy0AACC19WiXo1sHttfhEyE9O3ODEknL6aQvZRim/P2/LXfby5TYs0aRd1+WbaV2MwAAAPBvDPUBAAAAB8S3f6TQvN/JjscUGDJB3u5DHB2iF+w8rr8t3aZGmQE9NLq7PG5eKgAAgC93XZ/murZXMxXuK9FfFxbKtm2nk76UYZryX/sduVv1UmLHCkWX/1m2zWAfAAAAqY8rdQAAAEANsm1b0dWzFHnnJRmB+sq969fytOrpaNO+o6f0/JxNCvrcmnRLvuoHvY72AACA2sEwDH1tcHvltc3WhxuPaP5He5xOOifDdMs/6AG5Lumu+Nblin74Wsq/GQEAAABgqA8AAADUEDuZUOSdlxRbO0dmTiuljfq5fE1aO9p0siyiydMLlExa+u6YPDVpEHS0BwAA1C4u09T9N3dVi0bpmrV8t1ZsOuJ00jkZLo8C131XrtzOim9epugnbzDYBwAAQEpjqA8AAADUADseVXjJZCV2fCx3y54KjviJzLQsR5vC0YQmTy9Q8amovn1DZ3W4JNPRHgAAUDv5vW5NHJevrHo+/WnBFm3bX+J00jkZbq8CQybKbNxO8YKFiq1f4HQSAAAA8IUY6gMAAADVzI6UK7TgCSX3b5CnY3/5r3tIhsfnaFPSsvTCnE3aX1SuUVe31uVdmzjaAwAAaresej5NHJsnl8vU1BkFOnIy5HTSORkev4JDvyezQXPFVk5TrPA9p5MAAACAz8VQHwAAAKhGVkWxQvN+K+voDnnzb5Cv/7dkmC5Hm2zb1t+XbteGXSd0ZfcmGn5FK0d7AADAxaFF43p64OZuCkeTevqN9SoLxZxOOifDl6bAsO/LqJej6PK/KL5njdNJAAAAwFkY6gMAAADVxCo9qtDcx2UVH5Dvslvku+wWGYbhdJYWr9yvd9YdVOeWWfrG0E4p0QQAAC4OeW2zdcf1HVRUEtYzMzYonkg6nXROZlqWgjf8QIa/niLLnlfiUKHTSQAAAMBnMNQHAAAAqkHy+F6F5v5Gdvlx+ft/W978G5xOkiStLizStHd2qGl2UA+O6ia3i5cEAACgal3bs5mG9m2hHQdL9cr8LbJs2+mkczIzmigw7GHJdCu8eLKSx/c6nQQAAACcwRU8AAAAoIolDm9VaN5vZUdD8g9+SJ5O/Z1OkiTtPFSql+dvVr2gR98bl6+g3+N0EgAAuEiNvbatendsqFWFRZr53i6ncyrFldNKgSETpWRc4YVPyiorcjoJAAAAkMRQHwAAAKhSib2fKrzgvyTZCtzwfXla93Y6SZJ0rCSsKdMLZEiaMDZfOZkBp5MAAMBFzDQMfWd4F7XJra8FK/bqvU8POp1UKe7czvIPul925JRCbz4hK1TidBIAAADAUB8AAACoKvFtHyq8ZIoMj1/B4T+WO7ez00mSpIpIXE9PW6/yUFzfGdFVbXLrO50EAADqAK/HpQlj8pST4dd/L96mTbtPOp1UKZ7WfeS76huyTx1TeOGTsqMVTicBAACgjmOoDwAAAFSB2IbFirz7soy0LAVv+qlcDVs5nSRJSiQtPTtzgw6fCOmWge3Uu2NDp5MAAEAdUj/Nq0nj8uX3uvTc7A06cKzc6aRK8XYeIO+lY2Sd2K/wkimyEzGnkwAAAFCHMdQHAAAAvgLbthVdNUPRj/8hMzNXwZt+KjOzqdNZkk63/XVhoQr3lejaXs10/aWXOJ0EAADqoNycND00urticUtPT1uvkvKo00mV4u0xXJ5u1yt5eKsiy56XbSWdTgIAAEAdxVAfAAAAuEC2ZSn6wV8VWzdPZsM2pwf66dlOZ50x76M9+nDjEeW1zdbXBreXYRhOJwEAgDqqU8ssfXNYJ50si2ry9AJFY6k/IDcMQ75+t8ndrp8Se9cp8v5fZNu201kAAACogxjqAwAAABfATiYUefsFxbe8K1ezrgoO/w8Z/nSns874eNMRzV6+Wy0apev+m7vKZfLUHwAAOOvK7k1105WttPfIKb04d5MsK/UH5IZhyj/gbrkuyVNi23LFVk5zOgkAAAB1EFf2AAAAgPNkJ2IKL5msxK6Vcrfuo8DQSTI8fqezzti6r1h/XrBFWfV8mjguX36v2+kkAAAASdLNV7VWv66N9emO43p92XancyrFMN0KDH5QZuN2iq1foFjBQqeTAAAAUMcw1AcAAADOw+mB/hQl92+Qp2N/+QeNl+HyOJ11xuETFXpm5ga5XKYmjs1TVj2f00kAAABnGIahbw7rrA6XZOqtNQe0dPV+p5MqxfD4FBwySWZWM0VX/FPxbR84nQQAAIA6hKE+AAAAUElnBvoHNsrTqb98/b8pI4Vua18WiunpaesVjiY1fmQ3tWhcz+kkAACAs3jcph4a3V1NGgT1+lvbtW77MaeTKsXwpytwww9kpGcr8t6flNj7qdNJAAAAqCNS5wokAAAAkMLOGuhf/U0ZRuo8nY7Fk5o6o0DHSiK68/oO6t4m2+kkAACAL5Qe8GjSuDylBTx6ce4m7TlS5nRSpZhpWQre8EMZvjSF33pOyaM7nE4CAABAHZA6VyEBAACAFGUnYgovnZqyA33LtvXHN7do58EyDb2shQb0bOZ0EgAAwDk1ygpqwtg8WZY0eVqBjpeGnU6qFDOziQJDvycZUnjxZFmlR5xOAgAAwEUuda5EAgAAACnozEB//wZ5OqbeQF+Spr+7U6sKi9SnY0ONHdDW6RwAAIBKa9csQ98Z0UWlFTE9Pa1AFZG400mV4mrURoFB42VHyxVa8KSscO240wAAAABqp9S6GgkAAACkkLMG+v1Tb6C/dPV+Lfpkn9o1y9A9w7vINAynkwAAAM7LpZ0a6ZZr2+nQ8QpNnbFB8UTS6aRKcbfsId9V35B96pjCi56SHY86nQQAAICLVGpdkQQAAABSRG0Y6K8uLNLrb21Xkwanb13r9bicTgIAALggQ/peosF9mmvb/hK9Mn+LLNt2OqlSvJ0HyNtzhKxjuxVe9pxsq3a8IQEAAAC1S2pdlQQAAABSwOmB/jP/GuhfnZID/W37S/TSvM2qn+bVw7fkKz3gcToJAADgghmGodsGtVefjg21qrBIb7y9w+mkSvP2GS13hyuV3Lde0Q9elV1L3pAAAACA2iO1rkwCAAAADvufgX7Bvwb630q5gf7pW9MWyOUyNGlcvnIyA04nAQAAfGWmYeg7I7qoQ/MMLVm1X0tW7nM6qVIMw5C//7fkat5N8cL3FFs3z+kkAAAAXGRS6+okAAAA4KD/PdB3d0jNgX7xqaieeuNTRWJJPTiqm1o2qed0EgAAQJXxuF16aEyemmYH9frbO7Ryy1GnkyrFMN0KDH5QZnYLxVbPVHzrcqeTAAAAcBFJrSuUAAAAgEP+70Dff03qDfTD0YSenrZeJ8qi+uawTurWOtvpJAAAgCqXHvDo4Vt6KCPdq1fmb1bh3mKnkyrF8AYUGPawjPRsRd7/ixIHNjqdBAAAgItEal2lBAAAABxQGwb6iaSlZ2dt0P6ico3q30ZXdm/qdBIAAEC1yc7w63vj8uV2mZo6c4MOHCt3OqlSzGCmAsO+L3n9p59fHt/rdBIAAAAuAql1pRIAAACoYbVhoG/btv68YIs27ynWgJ7NNLxfS6eTAAAAql2LxvX00OjuisWTeuqN9TpZFnE6qVJcWbkKDJkoWQmFF/5B1qljTicBAACglkutq5UAAABADbKthMJvPZvSA31JmvHeLn286ah6tMvRndd1kGEYTicBAADUiC6tGujbN3ZW8amonp62XqFIwumkSnE36SD/tffJDpcpvPAPsiO1404DAAAASE2pd8USAAAAqAG2bSny7h+V3Lde7vZXpOxA/+21B7RgxV61za2v+27uKtNkoA8AAOqWfl2baOyAtjpwrELPzCxQPGE5nVQpnjaXytfvdlklhxVeMkV2IuZ0EgAAAGqp1LtqCQAAAFQz27YV/eg1JXZ8LHfLnvJfc3dKDvTXbjum15ZsU+OsgCaMzZPP43I6CQAAwBHDLmuhgb2aqXBfif60YIss23Y6qVK83a+Xp/sQJY9sU+Sdl2TbteMNCQAAAEgtqXflEgAAAKhmsTWzFd+0TK6mneQf9IAMM/WG5TsOlOrFuZtUL+jR927toXpBr9NJAAAAjjEMQ18b3EG9OjTUJ5uPavo7O51OqjTf5bfK3aavErtXK/rxP2TXkjckAAAAIHUw1AcAAECdEtu4VLG1c2TmtFJgyEQZ7tQblh8+UaHJ09fLNAxNHJevRpkBp5MAAAAcZ5qG7h3RRe2aZWjRyn1aunq/00mVYhim/APukatpR8U3LlV8wyKnkwAAAFDLMNQHAABAnRHf9qGiH70mM6OJAsMeluFNvWF5aXlUT72xXuFoUg+M7KbWTes7nQQAAJAyvB6XJozNU5MGQb3+1natLixyOqlSDLdXgesnyMzKVXTFPxXftcrpJAAAANQiDPUBAABQJyT2rFPkvT/KSGugwI0/lBlIvWF5OJrQ09MKdLw0om8M7ai8ttlOJwEAAKSc9IBHD9+Sr/ppXr00b7O27S9xOqlSDF+aAkMflhHIUOSdl5Q8usPpJAAAANQSDPUBAABw0Usc2qLwsmdl+NIUvPGHMtNTb1ieSFp6fvZG7T16SiOvaq2r83OdTgIAAEhZOZkBTRqXL5fL0NQZBTp0vMLppEox6+UoMHSSZBgKL54sq6x23GkAAAAAzmKoDwAAgIta8tgehRdPlky3Ajd8X2ZmU6eTzmLbtv66sFAbd59U//xcjbiyldNJAAAAKa9lk3p6cFQ3RWJJPfXGpyo+FXU6qVJcDVsrMOgB2dFyhRY+KTtS7nQSAAAAUhxDfQAAAFy0rJLDCi98UrKSCgyZJFdOK6eTPtes93frw41HlNc2W18f0kGGYTidBAAAUCt0a52tbw7rpBNlUT31xnqFowmnkyrF3bKnfFfcIbv0qMJLpshOxJxOAgAAQApjqA8AAICLklV+QqE3n5AdrVBg8Hi5czs5nfS5Fn68R/M+2qPWTevpgZu7yWXyFB0AAOB8XNm9qUb1b6MDx8r1zMwNiicsp5Mqxdt1sDzdhyh5ZJsi7/1Rtl07ugEAAFDzuGIIAACAi44VLlP4zSdkV5yUf8A9crfs6XTS51q3/ZhemLFejTIDmjg2Xz6vy+kkAACAWml4v5Ya0CNXm/cUa8o/18mybaeTKsV3+a1yt+qtxM5PFFs10+kcAADw/9m77/is6rv/4+9z7etKSAgJe0MIYSZhCQoKIsuNAxTUqlXrqiBqra2t3t6tvesAoaK12opbHKDIBkVAkb2REPbeBMi49nV+f3iX301xREhyrit5PR8P/iCcnM87MXwfx/PmfA8Qpyj1AQAAUKWYIb/8M0YrduKA3OcPl7PV+VZH+l6bdhXqpckbVCPJpVFDc5SS5LI6EgAAQMIyDEPD+2cpr1WGvly5R+/N3SwzAYp9w7DJc/FdstVpodDqqQpt/NLqSAAAAIhDlPoAAACoMsxISP5ZYxU7skOuzlfL1b6f1ZG+184DRRr38Vo57IaevLOH6tbyWR0JAAAg4dltNt19VTu1a5Guucv3aOqiHVZHKhPD4ZZ3wEgZNWor+NWbiuxea3UkAAAAxBlKfQAAAFQJZiyqwOcvK7o/X852l8jV6SqrI32vg8dKNeaD1QpHTI24rqMyG9W0OhIAAECV4XLa9Yfbz1OTusmavHC75q3cY3WkMrF5U+QbNEpyeeWf+5KiR3dZHQkAAABxhFIfAAAACc80TQUWvK7IzlVyZPaQ+/xhMgzD6lhnKCwK6rn3V6vIH9Y9V7VTdtM0qyMBAABUOUlep0YNzVWdNK/enl2gpRsPWh2pTGw168vb/wEpGpF/5hjFio9ZHQkAAABxglIfAAAACS+04hNFCr6SvXFHeXr/UoYRf5e5xf6wRk9craMnA7p1ULbysmpbHQkAAKDKSk1y6eGhuUpNdunVz77V+u1HrY5UJo76reXp/UuZJYXyzxwjM+S3OhIAAADiQPzd7QQAAAB+hvCmhQqt/FS2jGbyXnKvDJvD6khnCIaiGvvhGu09UqIhfTLVq2MDqyMBAABUeRk1vRo1NFcel10vTlqnrXtPWB2pTJyZPeTqeq1ix3bLP3e8zFjE6kgAAACwGKU+AAAAElZkz3oFFkyQkZwu78CRMpweqyOdIRKNafzkddq676QGdW+igec1sToSAABAtdGodrJGXJ8jSXrhwzXae7jY4kRl48q9XM7sCxXds17Br96UaZpWRwIAAICFKPUBAACQkKJHd8k/50XJ6ZZ30EOy+WpaHekMMdPUa1O/1frtx9SrY31dd1FLqyMBAABUO5kNU3X/4A4KhKJ6fuJqHTke/1vaG4Yhd89bZG/UXuH8BQqtmWZ1JAAAAFiIUh8AAAAJJ1Z8TP6ZY6RoRN4BI2RPi7/t7E3T1LtzCrR04yF1zqqtWwa2lmEYVscCAAColtq3SNcdl7fVieKQnp+4WidLQlZH+kmGzSHvJffJVquRQks/UnjLYqsjAQAAwCKU+gAAAEgoZqhU/pmjZZYUytP7Djnqt7Y60vf69Kvt+mLlXrVpmqa7rmwru41LbwAAACud17aubuqfpYOFfo3+YLX8wfh/V73h8so7cJQMX00FvnxNkf2brI4EAAAAC3BnEQAAAAnDjEbknzNesWN75Op2vZyZ3a2O9L3mLt+tKV/vULN6NXT/NR3kdNitjgQAAABJfTo10tU9m2vXwWKN+2itwpGo1ZF+ki25lrwDH5TsDvlnj1Ps+H6rIwEAAKCSUeoDAAAgIZimqcDC1xXdu0HONn3kyrnU6kjf65sNB/Tu3M2qV8unkUNy5HU7rI4EAACA/+OKC5qpb+dG2rT7uP7+6QZFYzGrI/0ke0ZTefveK4X8Kp0xWjH/SasjAQAAoBJR6gMAACAhhFZ8okjB17I3yZH7gpvi8v30a7ce0b+mbVRaDbceGpqrFJ/L6kgAAAD4D4Zh6MZLWql727patfmIJszIl2maVsf6SY4mHeXueYvMosPyzxorMxKyOhIAAAAqCaU+AAAA4l5400KFVn4qW0YzefveI8MWf9vZb95zXC9NXi+v26GHhuYqPdVjdSQAAAD8AJth6PbL2qhDi3R9ve6APpy31epIZeJq01uunEsVO7RVgXn/kGnG/y4DAAAAOHeU+gAAAIhrkT3rFVgwQUZyurwDR8pwxl9ZvvtQscZ+uFaGYWjk9TlqkJFkdSQAAAD8BIfdpnsHt1dmo1TNXLpLMxbvtDpSmbi6XSdHi26KbF+u4JIPrI4DAACASkCpDwAAgLgVPbpL/jkvSk63vIMeks1X0+pIZzh03K/RE1crGI7q/ms7qEWDFKsjAQAAoIzcTrtGXNdRjWon6cMvt2rBmn1WR/pJhmGTp/cdstdtpfDamQptmGt1JAAAAFQwSn0AAADEpVjxMflnjpGiEXkHjJA9rYHVkc5wojio599fpZMlIf3qynZq16yW1ZEAAADwMyV5nBo1NFcZqR69MTNfKzYdsjrSTzIcLnkGPCAjta6Ci95RZOdqqyMBAACgAlHqAwAAIO6YoVL5Z46WWVIoT+875Kjf2upIZygNhPX8xDU6fDygmwe2VpfsOlZHAgAAwFmqmezWwzfkqobPpVembNDGHcesjvSTbJ4a8g0cJcOVJP/nLyl6eIfVkQAAAFBBKPUBAAAQV8xoRP454xU7tkeubtfJmdnd6khnCIajGvvRWu05XKxrL2qh3rkNrY4EAACAc1QnzadRQ3LkdNg1btI6bd9/0upIP8mWWlfeASMkMyb/zDGKFR2xOhIAAAAqAKU+AAAA4oZpmgosfEPRvRvkbNNbrpzLrI50hnAkppcmr9fmPSfUv2tjXdq9qdWRAAAAUE6a1K2hEdd1VCxmaswHa7T3cLHVkX6SvV4refrcJdN/Qv6ZY2QGS6yOBAAAgHJW5lL/5MmTGjFihPLy8tSrVy+98847P3js0qVLdfnllysnJ0dDhgzR5s2bv/e43/72t2rdurW2bt3685MDAACgygmvnalIwULZG3eU+4KbZRiG1ZFOE4nG9PdP12vdtqPq2aG+hlycGXcZAQAAcG6yGtfUfYPbyx+M6Nn3V+vAsVKrI/0kZ4tucp83VLHCvfLPeVFmNGJ1JAAAAJSjMpf6Tz31lKLRqBYuXKhXXnlF48aN0+LFi884rrCwUPfee6/uuusuLVu2TH379tW9996rSOT0C8klS5Zo79695/4VAAAAoEqI7Fyl4JIPZEtrIG/fe2TY7FZHOk0sZuq1qd9q1eYj6t62rm4dlC0bhT4AAECV1LFlhu6+qr2KS8N69r1VOnTcb3Wkn+TsOFDOthcrum+jAgtfl2maVkcCAABAOXGU5aDS0lLNnDlTn3zyiZKTk9W2bVsNHjxYH3/8sbp3P/0dp3PmzFGzZs105ZVXSpLuuOMOvfHGG1q2bJl69OghSQqFQvrTn/6kMWPG6LLLzm1LVe6jVq5/f7/5vgOId6xXQGKJHt0t/xevyPAkyzdwpGxur9WRThMzTf1r+kYt3XhInVvX1h1XtJHddu4LDGsVgETBegUgUZTnetUlu7buvKKN/jHlWz333ir99qZOSk/xnPuJK4hhGPJcMFxm8RFFCr5WOKWO3J2vsjoWgB/A9RWARMF6FR/KVOrv2LFDkpSZmXnqY9nZ2ZowYcIZxxYUFCg7O/vU7+12u1q1aqWCgoJTpf4rr7yiCy+88LTznY20tKRz+nycvfT0GlZHAIAyYb0C4l+k+Lj2zR4rRSOqd8Pv5W3S0upIpzFNU+M/WqNF6w+oS5u6+t2t3eR0lHnDqzJhrQKQKFivACSK8lqvLr+ohjxet8ZOXKXnJ67R/9zXU7XiuNiXpNjQ32jfm39QcPlkpdRvpBode1sdCcCP4PoKQKJgvbJWmZ/UT0o6vUBPSUlRSUnJ9x6bmpr6g8du375dU6dO1eTJk8828ymFhSWKRmPnfB6UnWF895f26NEisYMXgHjGegUkBjMSUunUZxQ9eUSei25Xia+xSo4UWR3rFNM09e7czZq7fI/aNUvTXZdn68TxM6+BzxZrFYBEwXoFIFFUxHqV2yJNtwxorTdnbdJjLy7Uo8M7KSXJVT4nryCufg8o/Ml/6/C0l1Ri+uRo2MbqSAD+A9dXABIF61XlsNttP/pAe5lKfZ/Pd0aBX1RUdEbR/+9ji4uLf/DYJ554Qg8//LB8Pl9ZRv8kfnisYZp87wEkBtYrIH6ZpqnAggmKHtzy3fs/W18YV39fTdPUR/O3au7yPcpqXFP3X9tRDru9QjKyVgFIFKxXABJFea9XvfMaKhSJ6f3PN+vZ91brN8PylOx1lt+Acmb40uQd+KBKP31apbPHyXfV47KnNbQ6FoDvwfUVgETBemWtMu0b2qxZM0nS1q1bT30sPz9frVq1OuPYrKws5efnn/p9LBZTQUGBsrKyJElLlizR448/rvPOO0/nnXeeJOmGG27Q+++/f9ZfBAAAABJPaM00RTYvkr1Jjtzdhlgd5wxTvt6hGYt3qWWDFI24rqPcTrvVkQAAAGCh/l0b69qLWmjP4WKNnrhapYGI1ZF+lL1WY3n73S+FQ/LPGK1Y6XGrIwEAAOAslanU9/l8GjBggMaOHavi4mLl5+dr0qRJuuaaa844tl+/fqe22A+FQnrttdeUlJSkrl27SpLmz5+vTz/99NQvSRo/fryuuOKKcvyyAAAAEM/CO1YotPQj2dIayXvx3TJs5fuO+nM1ffFOffrVdjWtW0MPDsmR112mDa4AAABQxV3Wo5muvKCZdhwo0gsfrlEgFN/FvqNRO3l6/UJm8VH5Z42VGQ5aHQkAAABnocx3T5944glJUq9evXTHHXfogQceUI8ePSRJeXl5Wr58uSQpLS1N48eP18svv6wuXbpozpw5eumll+RwfHcjtF69eqf9kqT09PTv3cofAAAAVU/0yE4FvviHDE8NeQeOkOHyWh3pNHOW7dZHX25Vo9pJeuiGXPk88butKgAAACrfVT2ba+B5TbRl7wmN+2itguGo1ZF+lDP7QrnyrlDs8HYFvvi7zFjM6kgAAAD4mQzTTNy3HxQWligS4SK0MhmGlJFRQ0eOFPHeDABxjfUKiE+x0uMqnfyUTP9JeS9/VI56Z77OyUpfrtqrN2dtUv10n34zrJNSk1wVOo+1CkCiYL0CkCgqa70yTVPvzt2sz1fsUbvmtfTAtR3kdMTv65pM01Rg3j8U2fKNnO37yXP+cKsjAdUe11cAEgXrVeVwOGxKS/vhh+Dja59TAAAAVFlmJCT/7HEyS47Jc+FtcVfof71uv96ctUl1anr18A15FV7oAwAAIHEZhqEbL2mlC3Pqa8P2Y3r5kw2KROP34SPDMOS56HbZ67dWeP0chdbNsjoSAAAAfs6fXiUAACAASURBVAZKfQAAAFQ40zQVWPAvxQ5tkyv3MjmzLrA60mmWbjyof03fqPQUtx6+MVdpNdxWRwIAAECcsxmGbhmQrR7t6mn1liP6x5QNisbx1vaG3Slv/wdkq1lfwW/eV3j7CqsjAQAAoIwo9QEAAFDhQqs+U2TLYjmadZKr67VWxznNik2H9Y8p3yo1yaVHbsxTRqrX6kgAAABIEDabodsvy1aX7Dpavumw/jlto2Kx+N2X1nAnyTtwlAxvDQW+eEXRQ1utjgQAAIAyoNQHAABAhQpvW6bQ8kmypTeWp89dMoz4uQRdu/WI/v7peiV7HXrkxjzVSfNZHQkAAAAJxm6z6a4r2io3M0OLNxzUGzPzFYvjF87aUmrLO2CkJMk/8wXFTh6yOBEAAAB+SvzcUQUAAECVEz2yQ4F5r8rwpsg7YKQMp8fqSKd8u+OYXpy0Xh6XXQ/fkKf66UlWRwIAAECCcthtuufq9mrfopYWrt2vd+cUyIzjYt9ep4U8fX8lM1As/4zRMgPFVkcCAADAj6DUBwAAQIWIlR6Xf9ZYSbHv3t2ZnG51pFMKdh/XuI/Xyumw6eEb8tSoTrLVkQAAAJDgnA6b7h/cQdlNauqLlXv1wbwtcV3sO5t1lrvHjYqdOCD/nL/JjIatjgQAAIAfQKkPAACAcmdGQvLPGiuzpFCeC2+XvW6m1ZFO2brvhMZ8uEaGYWjUkBw1rVfD6kgAAACoIlxOux64rqMyG6Vq1tLdmrxwu9WRfpSrQ3852/dTdP8mBeb/M67/EQIAAEB1RqkPAACAcmWapgJfvqbY4e1y5V0hZ6vzrY50ys4DRRozcY3MmKmR13VUy4apVkcCAABAFeNxOTTyuhw1q1dDUxft0NRFO6yO9KPc3W+Uo2meIlsWK7R8ktVxAAAA8D0o9QEAAFCuQiunKLJtqRzNOsvVZbDVcU7ZeaBIz72/SqFITL++rqNaN0mzOhIAAACqKJ/HoVFDc9W4TrImLdgW18W+YbPJc/HdstVurtCqzxTOX2B1JAAAAPwHSn0AAACUm/DWpQqtmCxbelN5+twlw4iPy81t+07q2fdWKRiO6dfXdlC7ZrWsjgQAAIAqLtnr1MM3/P9i/9Ovtsft9vaG0y3vgJEyamQosHCCInvWWx0JAAAA/0d83GUFAABAwose2qbAl6/K8KbKO2CEDKfb6kiSpC17T+j5iasUicY04vqO6tAi3epIAAAAqCZq+Fx65MY8Na1bQ59+tV2TF26L22Lf5kuVd+AoyemRf86Lih7bbXUkAAAA/C9KfQAAAJyzWEmh/LPHSZK8A0bIlhwfT8IX7D6u5yeuViwmPTgkhyf0AQAAUOmSvU49cmOumtdP0dRFO/XRl1vjtti3pzWQt/+vpWhY/hmjFSs+anUkAAAAiFIfAAAA58iMBOWf9YLM0uPy9L5D9jotrI4kSdq4s1CjP1gt6btCv3WTNIsTAQAAoLryeZx6aGiuWjZM0Ywlu/T+51vitth3NGgjT+87ZJYUyj9jtMxgidWRAAAAqj1KfQAAAJw104wpMO9VxY7slKvTVXK2PM/qSJKkDduP6YUP18huM/Tw0FxlNa5pdSQAAABUcz6PQ6OG5CqrUarmLN+td+YUKBanxb4zs4dc3YYoVrhX/tnjZEbDVkcCAACo1ij1AQAAcNZCKz5VZPtyOVp0lavzVVbHkSSt3XpUYz9aK6fdpodvyFPLhqlWRwIAAAAkSV63Qw8OyVV2k5r6YuVevTVrU9wW+66cQXK2u0TR/ZsUmPeqTDNmdSQAAIBqi1IfAAAAZyW8ZbFCKz+VLaOZPL3vkGFYf2m5evMRvThprdxOmx65MU/N66dYHQkAAAA4jdtl14jrc9SuWZrmr96nCdPzFYvFX7FvGIbcPYbJ0byLItuWKrh4otWRAAAAqi3r77wCAAAg4UQPbVVg/msyfDXlHTBChsNtdSSt2HRI4yevk9ft0G+GdVLTejWsjgQAAAB8L7fTrgeu66iOLdP11br9+ue0bxWNxd+T8IbNJk+fu2Svl6XwulkKrZ1pdSQAAIBqiVIfAAAAP0us+Kj8s8ZJssk7YIRsSWlWR9LSjQf18icblOR16jfDOqlxnWSrIwEAAAA/yumw677BHZSbmaFvNhzUq5/FabHvcH133Z/WQMHF7yu8ZbHVkQAAAKodSn0AAACUmRkOyj9rrEz/CXn63CF77eZWR9I3Gw7olSkblJLk1KPD8tQwI8nqSAAAAECZOB023Tu4vTpn1dbSjYf09083KBKNw2LfnSTvoIdk+Goq8OVriuzbaHUkAACAaoVSHwAAAGVimjEF5v1DsaO75OoyWM4W3ayOpK/X7ddrn32rtBpuPTq8k+qnU+gDAAAgsTjsNv3qqnbq1qaOVmw6rJc/Wa9wJP6KfVtyuryDHpLsTvlnjVP02G6rIwEAAFQblPoAAAAok9CySYrsWCFHy/PkyrvS6jhasGaf/jVto2qlePTosE6qm+azOhIAAABwVhx2m+68oq16tKurVZuPaPzkdQpHolbHOoM9vbG8/X8tRUPyzxitWPFRqyMBAABUC5T6AAAA+EnhzYsUWj1Vttot5LnolzIMw9I881bu0YQZ+cqo6dGjw/NUu6bX0jwAAADAubLbbPrlZW11QYd6Wrv1qMZ9vE6hcPwV+46GbeXpfafMkkL5Z4yWGSyxOhIAAECVR6kPAACAHxU9uEWB+f+SkVRL3gEPyHC4LM0zZ/luvTW7QHXTvHp0WCdlpFLoAwAAoGqw2QzddmkbXZjTQBu2H9PYj9YqGIq/Yt+Z2V3u7kMVK9wr/+xxMiMhqyMBAABUaZT6AAAA+EGxoiPyzx4n2WzyDhghm6+mpXlmLtml9+ZuVv10n34zrJNqpXgszQMAAACUN5th6JaBrXVxp4bauLNQYz5cI38wYnWsMzg7DJSzfX9F929S4MtXZZoxqyMBAABUWZT6AAAA+F5mOCD/rLEy/Sfl6XOX7BlNLc0z7Zsd+mDeFjXMSNJvhnVSWg23pXkAAACAimIzDA3vl6V+XRqrYPdxjfkg/op9wzDk7nGDHC26KrJtmYLfvCfTNK2OBQAAUCVR6gMAAOAMphlT4ItXFDu2W66u18rZvIuFWUxN+Wq7Pp6/TY1qJ+uRYXlKTbL2FQAAAABARTMMQzf0zdSg85poy94Ten7iapUEwlbHOo1h2OTpfafs9VsrvH6OwutmWh0JAACgSqLUBwAAwBlCSz9SZOcqOTJ7yJV7uWU5Yqap9z7frE++2q6mdWvoN8PylOKj0AcAAED1YBiGruvdUpef31Tb9p3UX99ZqcKioNWxTmM4XPL2f0C2tIYKLp6o8JZvrI4EAABQ5VDqAwAA4DThgq8UWjNdtjot5bnwNhmGYUmOSDSmf079VnOX71FW45p65MY8JXudlmQBAAAArGIYhq65sKWu79NSew6X6C9vr9DBwlKrY53GcCfJO2iUjKQ0Bea9psjudVZHAgAAqFIo9QEAAHBK5ECBAgtel5FUS97+D8hwWPNUfDAc1YuT1umbDQeVm5mhUUNy5PM4LMkCAAAAxINB5zXVbZdm6+jJgP7y1grtPFBkdaTT2JLT5R30sOTyyD/nb4oe3GJ1JAAAgCqDUh8AAACSpFjRYQVm/02yOeQdOFI2X6olOYr9YT3//mqt3XpUPTvU133XtJfLabckCwAAABBPenVsoPsHd1BpMKpn3lup/J2FVkc6jb1WQ/kGPihJKp05RtHCvRYnAgAAqBoo9QEAACAz5Jd/5liZgWJ5L75b9vQmluQoLArqr++s1Ja9JzTwvCa67dJs2W1csgIAAAD/lpdVWw8NzZEkjf5gjVYWHLY40ensdTPl7Xe/FArIP/05xYqOWB0JAAAg4XGHFAAAoJozYzH5v/i7YoV75Op2vRzN8izJceBYqZ5+a4X2HinR9X1aakifTBmGYUkWAAAAIJ61bpKmR4d1ks/j0PjJ67RgzT6rI53G0bijPH3ulFlyXKXTn1PMf9LqSAAAAAmNUh8AAKCaCy79QNFda+TIukCunEGWZNhx4KT+8vYKFRYFdfulbTTovKaW5AAAAAASRZO6NfS7mzopPcWjCTPyNX3xTpmmaXWsU5yZ3eW+YLjMEwfknzFaZshvdSQAAICERakPAABQjYXzFyi8dqbsdVvJ0+tWS56M37izUM+8u0qBUFT3XdNePTvWr/QMAAAAQCKqk+bT727urEa1k/XRl1v1wbwtisVRse9qd4lcna5S7MgO+WePkxkJWR0JAAAgIVHqAwAAVFOR/ZsU+OoNGcnp8vT/tQy7s9IzrNh0SGM+WC3DkEYNyVFeq9qVngEAAABIZDWT3frt8Dy1apSqWUt36/VpGxWJxqyOdYqr89Vytu2r6L6NCnzxisxY/GQDAABIFJT6AAAA1VDs5CEFZv9NsjvlHThSNm9KpWeYv3qvXvpkvXwepx4d1kmtm6RVegYAAACgKvB5nBo1NFc5LdP19foDGj9pnYLhqNWxJEmGYch9wXA5WnRTZMcKBb96I65eEwAAAJAIKPUBAACqGTNUKv+sF2QGS+S9+G7ZazWu3PmmqamLduiNmZuUkerR727qpCZ1a1RqBgAAAKCqcTvtuu+aDjq/fT2t2XpUoyeuVmkgbHUsSZJh2OTpc5fsjdornD9foWUfWx0JAAAgoVDqAwAAVCNmLCr/5y8rVrhP7u5D5GiaW6nzY6ap9z/fokkLtqlR7WQ9dlNn1UnzVWoGAAAAoKpy2G26/bI2GtCtsTbvOaH/eWeljhcHrY4lSTLsDnn73S9bnRYKrZ6q0LpZVkcCAABIGJT6AAAA1Uhw8URFd6+Ts3UvOTsMrNTZkWhM/5z6reYs362sRqn67fA81Ux2V2oGAAAAoKqzGYaG9MnU9b1bas/hEj391godLCy1OpYkyXB65Bs4SraaDRT85j2FC762OhIAAEBCoNQHAACoJkIbv1R4/WzZ67eWu+cvZBhGpc0OhqN6cdI6fbPhoHIzMzRqaK58HmelzQcAAACqE8MwNKh7U902KFtHTwb0l7dXatfBIqtjSZIMT7K8lz4sIzldgfn/VGTXaqsjAQAAxD1KfQAAgGogsnudgl+9KaNGbXn63S/D7qi02SWBsJ5/f7XWbj2qCzrU033XtJfLaa+0+QAAAEB11Sunge4b3EGlgYj++u5KbdpVaHUkSZItuZZ8lz4sw50k/5zxihwosDoSAABAXKPUBwAAqOKiR3fLP3e85PLKO+hB2Tw1Km12YVFQ//POSm3Ze0IDuzXR7Ze2kd3GJSgAAABQWTpl1daoITkyTen5iWu0suCw1ZEkSbaa9eUdNEqy2eWfOUbRo7utjgQAABC3uKMKAABQhcVKCuWfOUaKhuXt92vZazaotNn7jpToL2+v0N7DJbq+T0sNuTizUrf8BwAAAPCd7KZpenRYJ/ncdo2fvE5frt5rdSRJkr12c3n7PyBFwvJPf1ax4wesjgQAABCXKPUBAACqKDMckH/mCzJLjslz0S/laJBdabM37DimP7+1QsdOBnXbpdkadF7TSpsNAAAA4ExN69XQYzd3VkaqR2/O3KQPvtiiWMy0OpYcDdvK0/cemYFilU57RrGi+NhJAAAAIJ5Q6gMAAFRBZiwq/+cvK3Z0p1ydB8vZ6vxKmz1/9V6NmbhGkvTg0Bz16lh5uwMAAAAA+GF103z6/S1dlNkoVTOX7tL4yesUDEWtjiVn887y9LlTZkmhSqc+o1hJodWRAAAA4gqlPgAAQBVjmqaCi95VdNcaObIukKvTlZUyN2aa+mDeFr0xc5Nqpbj1+5s7q12zWpUyGwAAAEDZpPhceuSGXHVvW1erNh/R/7yzUoVFQatjyZnZQ+4Lb5VZdFj+qX9VrPSE1ZEAAADiBqU+AABAFRNeN1vhbz+XvUEbeXrdVinvsQ+Goho/aZ1mLtmllg1T9PgtXdQgI6nC5wIAAAD4+ZwOu+68oq2u6tlcOw8W6U9vLteug0VWx5Ir+yK5zx+u2IkD8k9/Vmag2OpIAAAAcYFSHwAAoAoJb1+h4OL3ZavZQN5+98uwOyp8ZmFRUP/z7kqt2nxE3drU0W9uzFNKkqvC5wIAAAA4e4Zh6KqezXXnFW1VVBrSX95eqdVbjlgdS672/eTqdp1ix/aodMbzMkN+qyMBAABYjlIfAACgioge2qbAF6/I8NaQd9CDMtwV/6T8rv99qmfngSJdeUEz/erKdnI67BU+FwAAAED56NGunh6+IU9Oh01/+3it5izbLdM0Lc3kzr1crk5XKnZ4u/wzx8gMW/96AAAAACtR6gMAAFQBsZOH5Z/1giTJO2CkbDVqV/jMNVuO6C/vrFRRaUh3Xt5WV/dqUSlb/QMAAAAoX1mNa+rxWzqrbppP732+WW/PKVA0FrM0k6vzYDk7DFD0QIH8s8fJjIQszQMAAGAlSn0AAIAEZwZLvnt6xV8kz8W/kr1OiwqfOXf5bo37eK2cdpseviFPPdrXq/CZAAAAACpOnTSffn9LZ7VpmqZ5K/dq7Edr5Q9GLMtjGIbc3W+Qs00fRfdukH/ueJlR6/IAAABYiVIfAAAggZnRiPxzXlTs+L7vbng171yh86KxmN6ZXaB3524+ddMvq3HNCp0JAAAAoHIkeZx6cEiOenWsr/Xbjunpt1foyAnr3mlvGIbcPW+Wo9UFiu5ao8C8V2TGopblAQAAsAqlPgAAQIIyTVOBha8rum+jnG37ytmhf4XO8wcjGvfROn2+co+ym9TU72/+bntOAAAAAFWHw27TrYOydX3vltp7uER/enOFtu07aVkew7DJc9HtcrToqsi2ZQrM/6dM09pXAwAAAFQ2Sn0AAIAEFVo1RZGCr2VvkiP3+cMq9H32R08E9Je3V2jdtqPq2bG+Rg3NVbLXWWHzAAAAAFjHMAwN6t5U9w1ur0Awor++u1LL8w9Zl8dml6fPr2RvkqPI5kUKfvWWTNO0LA8AAEBlo9QHAABIQOHNixRaPlm29Kby9r1Hhs1eYbO27z+p/35zufYcLtF1vVvqtkHZcti5jAQAAACqus6t6+jR4Z3kczv00ifrNe2bHZaV6YbdIe8l98nesJ3CG+cpuPh9in0AAFBtcDcWAAAgwUT25Ssw/58ykmrJO3CkDKenwmYtzz+kv76zUv5gRPde3V6Xdm9aoTsCAAAAAIgvzeun6PFbuqhR7SR9PH+bXp+er0jUmu3vDYdL3v4PyF4vS+F1sxRaPsmSHAAAAJWNUh8AACCBRI/ukn/WWMnulHfQg7IlpVXIHNM0NX3xTr30yXp53A49OqyTumTXqZBZAAAAAOJbeqpHj93UWR1apOurdfs1euJqFfvDlmQxnG55Bz4oW+3mCq36TMFVn1mSAwAAoDJR6gMAACSI2MlD8k9/ToqG5R0wQvZajStkTiQa04QZ+froy61qVDtJj9/SWS0apFTILAAAAACJwet26IHrOqhv50bK33Vcf35rhQ4eK7Uki+HyyjfoIdnSGyu07GMFV06xJAcAAEBlodQHAABIALHS4yqd9qzMQJE8fe+Wo0GbCplTWBTUM++u0sK1+9W+RS09dlNnZaR6K2QWAAAAgMRit9k0vF+Whl3SSocKS/XfbyzXmi1HLMlieJLlu+xR2dKbKLR8koIrPrEkBwAAQGWg1AcAAIhzZqhU/hnPyyw6LHfPX8jZvEuFzNm0q1D/NWGZtuw9oYHdmmjEdR3ldTsqZBYAAACAxHVJl8Z6cEiODEMa+9FafbJwm2KmWek5viv2fyNbRlOFVnyi4PLJMi3IAQAAUNEo9QEAAOKYGQnJP2usYkd3y9X1Wrna9C7/Gaap2ct269n3VisYiuqeq9tryMWZstu4VAQAAADw/do3T9cTt3ZVk7rJmvL1Do39cK2K/eFKz3Gq2K/dXKGVnyq0fBLFPgAAqHK4UwsAABCnzFhUgc9fVnT/Jjnb95Mr9/JynxEMRfXKlA16//PNqp3m1eO/6KKu2XXKfQ4AAACAqiejple/u6mzLuhQT+u2HdVTE5Zp18GiSs9huJPku/Rh2Wq3UGjVZwot+4hiHwAAVCmU+gAAAHHINE0FF05QZOcqOTJ7yN3jRhmGUa4zDh4r1Z/eWq6lGw+pU1Zt/fEXXdQwI6lcZwAAAACo2lxOu26/tI1uHtBahUVB/fmtFVq0fn+l5zDcSfJd9rBsdVoqtHqagks+oNgHAABVBqU+AABAHAot/VDhTQtlb9xRnt6/lGGU72Xb6s1H9NQby7TvSImuvaiF7hvcXl63o1xnAAAAAKgeDMNQn7yG+u3wTkr2OvXa1I16e/YmRaKxys3h8n33xH7dTIXXzlBw8fsU+wAAoEqg1AcAAIgzobUzFFozXba6mfJecp8MW/mV7bGYqUkLtmncx2tlt9k0amiuLuvRrNx3AQAAAABQ/bRsmKo/3tpV2U1q6ouVe/XMu6tUWBSs1AyGyyvfoIdkr5el8LpZCn7zLsU+AABIeJT6AAAAcSRc8LWCiyfKltZAvgEjZTjd5XbuYn9YL3y4RlMX7VDTejX0x1u7qF2zWuV2fgAAAABITXLpoRtyNaBbY23Ze0L/NWGZNu0qrNQMhssr76BRstdvrfD6OQoueodiHwAAJDRKfQAAgDgR2blagfn/lJGcLu+gh2V4ksvt3DsPFOmpCcu0fvsx9epYX7+7qZMyUr3ldn4AAAAA+De7zaahF7fS3Ve1UzAU1bPvrdbsZbsrtVg3nB55B46SvX62whvmKvj1WzLNyn0dAAAAQHnhxakAAABxIHKgQP6542W4k757B2Ry+T1B/9Xa/Xpr9iaZpqlfDGyti3Ibltu5AQAAAOCHdGtTVw0zkvTi5PV6//PN2rbvhG4b1EZul71S5htOt7yDHpR/5gsKf/uFZMbk7nmLDINn3QAAQGLh6gUAAMBi0WO75Z/5gmSzyztolGw165fLecORmN6ctUn/mr5RNXxO/XZ4Zwp9AAAAAJWqYe1k/eGWLsprlaGlGw/pT28t18FjpZU233C45R04UvaG7RTe+KWCCyfwxD4AAEg4lPoAAAAWip08LP/056VISN5+v5a9dvNyOe+xkwH99d2V+nLVXrVpmqY/3tpVLRqklMu5AQAAAODn8Hkcuu+aDrr2ohbad6RET72xTKs3H6m0+YbDLe+AEbI3aq9w/gIFF7xOsQ8AABIKpT4AAIBFYv6TKp3+nMzSE/JcfJccjdqVy3nzdxbqqQnLtG3fSQ06r4lGDc1Ris9VLucGAAAAgLNhMwxd1qOZRg3Jld1m07iP12rSgm2KxcxKmW84XPL2f0D2xh0V3rRQgS9fkxmLVMpsAACAc0WpDwAAYIFYoEj+ac/IPHlQ7p43y9mi2zmf0zRNzVyyS8+9v1rBSEz3Xt1e1/fJlN3GJR8AAACA+NCueS398dYualqvhqYu2qEXPlyjYn+4UmZ/V+z/WvYmuYpsXiT/7BdlRkKVMhsAAOBccIcXAACgkpmBYvmnPqPYsT1ydbterrYXn/M5i0pDenHSOn0wb4vq1vLqD7d0UZfsOuWQFgAAAADKV0aqV7+7qZN6dqyv9duP6b9eX6qC3ccrZbZhd8rb/345Wp2v6K7V8s94XmaotFJmAwAAnC1KfQAAgEpkBopVOu0ZxY7tlqvbdXLnXnbO59y445ie+NdSrdp8RN3a1NHjt3RRg4ykckgLAAAAABXD6bDrtkHZunVQtopKw/rruyv1ycJtisYq/l33hs0hT+875GzfT9H9m1T62V8V85+s8LkAAABny2F1AAAAgOriu0L/WcWO7pKr67Vy515+TueLRGOavHCbZi7eJZfLrl9e1kbnt68nwzDKKTEAAAAAVBzDMHRhTgNlNkzVP6Zs0JSvd2jDjmO664p2ql3TW8GzbXL3GCbDk6zQ8skqnfJn+S59RLYaGRU6FwAA4GzwpD4AAEAl+P+F/k65ulwjd94V53S+g8dK9fRbKzRj8S41q19DT97WVRd0qE+hDwAAACDhNMhI0u9v6aL+XRtr696TevL1pVq84UCFzzUMQ+5OV8l9wc0yTxxS6ZQ/K1q4t8LnAgAA/FyU+gAAABXMDJaodPpz/1voD5a705Vnfy7T1MK1+/Tk68u080CRLuvRVI/d1Fl103zlmBgAAAAAKpfTYdMNfVtp1JAcOR12/eOzb/XqZ9/KH4xU+GxXu77yXPwrmaUnVTrlaUUPbavwmQAAAD8HpT4AAEAFOlXoH9khV+er5e501VmfqyQQ1t8/3aDXp+fL53HokRvzdO1FLeWwc0kHAAAAoGpo3yJdT93eTR1bpuubDQf05OtLtXXfiQqf68zsLu+AEVIkrNKpf1Vkz4YKnwkAAFBW3AEGAACoIKcK/cPb5ep0ldydrz7rcxXsPq4n/7VUy/IPqXNWbf3X7d2U3TStHNMCAAAAQHxISXJpxHUdNbxflgqLQvrLWyv12aIdisXMCp3raNJR3ssekWx2+WeOUXj78gqdBwAAUFaU+gAAABXADJWqdPrz3xX6eVfIdZaFfjQW0+QF2/TXd1eqyB/WLwa21r2D2yvZ6yznxAAAAAAQPwzDUN/OjfTHX3RR/XSfJi/YpmfeW6VjJwMVOtdRr5V8Vzwmw+1TYO54hfLnV+g8AACAsqDUBwAAKGdmyP+/T+hv+67Q73KNDMP42ec5fNyv/3nnuydSGtdJ1hO3dtVFuQ3P6lwAAAAAkIga1UnWH37RRX07NVLB7uP64z+Xann+oQqdaU9vLN9Vj8tIzlBwwesKrp5eofMAAAB+SplL/ZMnT2rEiBHKy8tTr1699M477/zgsUuXLtXll1+unJwcDRkyRJs3bz71Z5MnT9a1116rzp07q2fPnnriiSdUUlJybl8FAABAnDhV6B/aJlfu5Wdd6C/+97sj957UgG6N9fubu6h+elIFJAYAAACA+OZy2jW8f5YeIJCotQAAIABJREFUuK6jbDZDL32yXq9P36hgKFphM20pdeS78neypTVSaOkHCi75QKZZsdv/AwAA/JAyl/pPPfWUotGoFi5cqFdeeUXjxo3T4sWLzziusLBQ9957r+666y4tW7ZMffv21b333qtIJCJJ8vv9evTRR7Vo0SJNmTJFO3fu1DPPPFN+XxEAAIBFzJBfpTOeV+zQVrlyLpWr67U/u9D3ByN69bMN+sdn38rlsGvU0BwNvbiVnA42WAIAAABQveVmZuipX3ZTu+a1tHDtfj05YZl2HDhZYfNsSWnyXfmYbHUzFVozXcGFr8uMxSpsHgAAwA9xlOWg0tJSzZw5U5988omSk5PVtm1bDR48WB9//LG6d+9+2rFz5sxRs2bNdOWVV0qS7rjjDr3xxhtatmyZevTooWHDhp061u12a8iQIXr55ZfP+gtg99nK9e/vN993APGO9QqVzQz55Z85WrGDW+TKGST3edf/7EJ/694TemXKBh0+HlBOZrpuv6yNUnyuCkqMeMBaBSBRsF4BSBSsV1VfWg23Rg3N0Zxlu/XhvK3685srdG3vlhrQrbFsFfAf3vAkKemyR1Q6+28K5y+QGSiS9+K7ZTjd5T4L1QvrFYBEwXoVH8pU6u/YsUOSlJmZeepj2dnZmjBhwhnHFhQUKDs7+9Tv7Xa7WrVqpYKCAvXo0eOM45ctW6ZWrVr9zNjfSUtjC1qrpKfXsDoCAJQJ6xUqQyzo14GJ4xQ9sFmp512pWn1v+VmFfjRm6qMvCvTurE2y2wzdPbiDLr2g+Vlt24/ExFoFIFGwXgFIFKxXVd/wS9upe8eGevbtFfrgiy0q2H1CI2/MU3qqtwKm1ZB50+M6/Nl4FW9YqOD0Z1RvyG/lqFGrAmahumG9ApAoWK+sVeYn9ZOSTi/QU1JSVFJS8r3HpqamlunYefPmadq0afrwww9/TuZTCgtLFI2y3VFlMozv/tIePVokXiEFIJ6xXqGyxALF8s8YreihbXJ1HKBYzmAdPVpc5s/ff7RE/5qWry17T6hh7STdfVU7Naqd/LPOgcTFWgUgUbBeAUgUrFfVS6rHrj/c0lnvfb5Z81fv033PfKFh/bLUo13dCvlH0sYFt8vlrqXQyk+1+5+PyjdopOzpTcp9DqoH1isAiYL1qnLY7bYffaC9TKW+z+c7o5QvKio6o+j/97HFxcU/eew333yjxx57TC+99JKaNm1alhjfix8ea5gm33sAiYH1ChUpVnxU/unPKXZ8v1y5l8nV9TpJRpl+5iLRmGYu2aUpX29XNGqqX5fGuvaiFnI57fzMVkOsVQASBesVgETBelV9uJx2/WJgtto3T9dbs/L16mffasm3B3XLgNaqleIp52mG3F0Gy5ZSR4EF/1LJp0/L2/ceOZrklPMcVCesVwASBeuVtWxlOahZs2aSpK1bt576WH5+/vdum5+VlaX8/PxTv4/FYiooKFBWVtapjy1evFgjR47UCy+8oC5dupxtdgAAAMtEj+1V6ad/Uuz4frl7DJO72/VlfhJk54Ei/emN5Zq0YJtq1/TqsZs668ZLWsnltFdwagAAAAComjq3rq0/3dldPdrV09qtR/X4a0s0b9VexSqgfXBmXSDvZb+RbHb5Z72g0Ia55T4DAADg/ypTqe/z+TRgwACNHTtWxcXFys/P16RJk3TNNdeccWy/fv20fft2TZ06VaFQSK+99pqSkpLUtWtXSdKSJUv0wAMP6LnnnlP37t3L96sBAACoBJEDm1U65c8y/SflufhuuTr0L9PnhSNRfTx/q/77jeXac7hEl5/fVE/e1lWZjVJ/+pMBAAAAAD8q2evUnVe01cjrO8rrduitWZv07LurdLCwtNxnOeq3VtLVf5CRUkfBr99WYNE7MmO8KhYAAFQMwzTL9k8VT548qccff1wLFy5UUlKS7rnnHg0fPlySlJeXp1dfffXUU/dLlizRU089pd27d6t169Z6+umnTz3Vf/PNN2vFihVyu92nzt2gQQNNmzbtZ4cvLCxRJMKFUmUyDCkjo4aOHOG9GQDiG+sVKkpk5yr5574kGTZ5+/9ajkbty/R5m/cc1+vT83XgWKma1E3W7Ze2UZO6NSo4LeIdaxWARMF6BSBRsF7h3/zBiD76cqvmrdorp8Omwb1aqF/XRrLbyvScW5mZgWL55/xN0f2bZG+SI2/fe2Q4y3vbf1RFrFcAEgXrVeVwOGxKS0v6wT8vc6kfjyj1Kx9/cQEkCtYrVITwpoUKLHhdhjtJ3oEPyl6nxU9+TiAU0cfzt+mLFXtkt9t0Vc9mGtCtiRz28r2RhMTEWgUgUbBeAUgUrFf4T5t2Fer1Gfk6VOhX8/o1dNugNmpUJ7lcZ5jRsAILXldk8yLZ0pvKO3CkbElp5ToDVQ/rFYBEwXpVOX6q1Lc/+eSTT1ZenPIVCIQVi/HTU5kMQ/L53CotDVkdBQB+FOsVypNpmgqtmabgondkJNeS7/Lfyp7e+Cc/b8P2YxrzwRpt2HFMmY1S9eCQHHXKqiObzaiE1EgErFUAEgXrFYBEwXqF/5SR6tWFOQ0UiZlasemwFqzZp1jMVGaj1HL7fzPDZpejWSfJZlNk62JFti6RvUEb2Xw1y+X8qJpYrwAkCtarymGzGfJ6XT/4545KzAIAAJBwTDOm4DfvK7x+tmy1Gsk76KGffOKiJBDWxM+36Kt1++V22jW8X5b6dGoom0GZDwAAAACVzeW0a0ifTHXNrqPXp2/UlK93aEXBYd02qI1aNEgplxmGYcjd6SrZUuoq8OVrKp3ytLx975ajaV65nB8AAFRvlPoAAAA/wIxGFJj/miJbFsteL0veASNkuH94CyRJWrHpsN6evUknSkJq1yxNvxiYrYya3kpKDAAAAAD4Ic3rp+iPt3bV9G926rNFO/Tnt5arf9fGurpXC7md9nKZ4czsLiM5XYHZ4+SfPU7uHsPkat+vXM4NAACqL0p9AACA72GGA/LPeVHRPevlaNZJnovvluH44e2PTpSE9M6cAi3PPySf26HbL22jCzrUk8HT+QAAAAAQNxx2m67s2VydWtfW69PzNWvpbq0qOKJbB2Uru+mP78pW5hn1Wsl39R/knzFawUXvKHbigNw9hsmwlc8/HAAAANUPpT4AAMB/iPlPyj9zjGKHt8uZfZHcPW/5wZsvpmnqmw0H9N7czSoJRNQpq7Zu6p+lmsnuSk4NAAAAACirRrWT9fubO2vO8t2avGCbnnlvlXrnNtB1vTPl85z7bXNbSp3viv05Lyq84XPFjh+Qp+/dsnlqlEN6AABQ3VDqAwAA/B+xosMqnf68zBMH5Op0pVydB//g0/ZHTwT05qxNWrftqFJ8Tt17dXt1ya5TyYkBAAAAAGfDZjM0oFsT5bbK0Bsz8vXl6n1as/WobhnQWjmZGed8fsOdJO+ghxT8+m2F879U6aQn5b3kPtnrtCiH9AAAoDqh1AcAAPhf0aO75J8xWmbpCbnPv0mu9pd873HhSFSzlu7W1G92KBSO6fz29XRD31ZK9jorNzAAAAAA4JzVTfPp4RvztGDNPn3wxRaN/Wit8lplaGjfVqpT03tO5zbsDnkuvFX2ui0V+OoNlU55Wu4LbpIz+yJe1wYAAMqMUh8AAEBSeOtSBea/JsWi8vS9R86W3c44xjRNrd5yRO9/vlmHjwdUN82rYf2y1KFFugWJAQAAAADlxWYY6p3bUB1bpGviF1u0LP+Q1m07poHnNdZl3ZvJ7fr+V7KVlbN1L9nSG8s/50UFF05Q9OBW/b/27jy6qvu+9/57T2fS0TxLCAECMRobGzN4AE/EzmTHztAmjnOT56b3Nr69SbrSleaPpE8et+lqmtV1m1v/cdPmxsmK07RuQpvYtZ3YiW3wgBkMmFlikhAgIQkh6Rydce/9/HGEQAYM2KCjgz6vtfba++xJv21zvto6n/37ndBtj2DYgSt0BSIiInItU6gvIiIiU5rveaQ3/YL09mcxwqWE1/wJVt2cc/Y73h/n5y+2s/PQSYIBi0/e2cKapU3YlpmHVouIiIiIiMjVUFES4ksfW8SdHQP884ttPPN6B6/t6OZTd85m2fya99W73qqaQdGD3ybx0j+SbVvPSH8n4TX/A7NEX+MmIiIi787wfd/PdyPeq4GBONmsl+9mTCmGAVVVxfT1DVO4/3JEZCpQvZJL4SdjJH7/f3C7dmLWzCK85n9iFpWP2yeRyvLr1w7x4uYuXM/nlkV1fOKOFsqiwTy1Wq4lqlUiUihUr0SkUKheyZXkeh4vbz3Gf6w/SDyZpXVaKZ9Z08r02uL3dV7f90i/9WvSW34FwQjhO/8b9vTrr1CrpVCoXolIoVC9mhi2bVJeXnTh7RPYFhEREZFJwz15hMRv/jf+cC/OvFUEb30Ew3LGtnu+z+s7uvnFKwcYiqdprivm4TWtzG4szWOrRUREREREZKJYpsndN01j2fwa/n39IV7ZepT/78ebuGNJIw/ePoto2Ln4Sc7DMEyCN30Mq3oWiZd+QOL5vydw4/0EbnwAw9RocCIiInIuhfoiIiIy5WQObiT58v8FN0vwts/hzL9z3BCKB48N8bMX2jh0fIjiiMPnPziP2xbXY76PYRZFRERERESkMBVHAnzu3rmsvr6Bf36xjZfeOsrG3T08tGoWq29oxDTf29+K9vTFueH4X3ic9Fu/wu09SPjO/44Ril7hKxAREZFCp+H35bJoiA0RKRSqV3I+vueR3vxL0tv+EyNcQmjNn2DXtY5tH4yn+eXLB3h1x3FMw+CepdO4/9YZRELvrfeFyMWoVolIoVC9EpFCoXolV5vv+7y5p4d/e+kAA8MpmmqifOaeOcydXn7xgy90zmya1Gs/JbNvPUa0kvCa/4lVPePKNVomJdUrESkUqlcTQ8Pvi4iIiAB+Kk7i9/8H98gOzOpZhNf8CWa0AoCs6/G7LV38+rVDJFIuC2aU8+l7WmmsuvBNlIiIiIiIiEw9hmGwYkEdN8yu4j/f6OA3Gzv57j9vZdn8Gj5152wqSkKXf047QGj1f8WsnU3q1Z8y8uu/InjrIwTmrb4KVyAiIiKFSKG+iIiIXPPck10kfvu/8YdO4My9neCtj2DYAQB2Hurn5y+2c7x/hKrSEP/1wwtYMqdq3HD8IiIiIiIiImcLBWw+vrqF2xfX8y+/28/GPSfYtr+PD69o5r7l03Fs67LPGZi3GqtyOokXHie17gm8ngMEb/3s2N+vIiIiMnUp1BcREZFrWubgJpIv/xDcLMFbH8FZcBeGYXDiVIJ//V07W9v7CNgmD94+k3uXTSfgXP4HLyIiIiIiIjI11ZRH+PInFrPjYO6B8X9ff4j1bx/n03fP4Yb38MC4VT2TyEPfJvn7H5DZtw639yChu/47VkXTVboCERERKQSG7xfutx8MDMTJZr18N2NK0fdmiEihUL0S3/NIb15LetszGOESQvf8D+z6uYwkMzz3Zie/2XiErOuxbH4Nn7xjNpWllz9Eosj7pVolIoVC9UpECoXqleRT1vV4cXPuq92SaZeFMyv45B0tTK8tvuxz+Z5HetvTpLf8CgyT4LJP4Fz3AQzDvAotl3xQvRKRQqF6NTFs26S8/MJfB6ue+iIiInLN8VNxEr//Ae6RtzGrZxJe8ydkgmX8dkMHz23oIJ7MMq26iIfXtDJ3enm+mysiIiIiIiLXANsyuW/5dFYurOUXrxzgtR3d7Dp0kuULavnYbTOprYhc8rkM0yR44wPY0xaReOkfSW34F7Kd2wnd8UXMaOVVvAoRERGZjNRTXy6LnsYRkUKhejV1ZY/uJvnyD/HjJ7Fbb8Ne+QjrdvbyzOuHGYynqSwJ8bHbZ7JyYR2meXnDIIpcaapVIlIoVK9EpFCoXslk0tE9zL+vP8jbB/oxDYPbFtdz/60zqCi5vJHi/EyK1Iafk9nzMgQihG77HM7sFVen0TJhVK9EpFCoXk2Mi/XUV6gvl0VvXBEpFKpXU4+fTZPa9EsyO34DdpDAik+zOdvKr147TN9gkpKiAB+9ZQarrm/AsTVcoUwOqlUiUihUr0SkUKheyWTUduQUa185QFvXILZlcteNjXxoZTMlkcBlnSfbsY3kuh/hJ4awZ68gdOsjGMELf/gvk5vqlYgUCtWriaFQX64ovXFFpFCoXk0tbl8HyZd+gDdwDLN2Nvunf5ynNg9yvH+EopDNB1c0c/eN0wgGrHw3VWQc1SoRKRSqVyJSKFSvZLLyfZ9dh07yy1cO0tEzTDBg8YGlTdy7bDqR0KV/S66XGCK17gmyHVsxiioI3flH2A3zr2LL5WpRvRKRQqF6NTEuFupf+t2CiIiIyCTjex7p7c+S3vLv4MNAy4f4cWczh/ccJ+hYfOSWGdy3rIlIyMl3U0VERERERGQKMwyDRbMqWTizgi37evn39Qd5+vXD/P6tLj60opm7bppG0Ln4g+hmuITQB75MZu8rpN74OYln/hZn8b0Eb/44hqW/fUVERK5VCvVFRESkIHlDJ0i+9E+4Pe1konWszd7B65sC2NYIa5Y28eGVzZQUXd5QhiIiIiIiIiJXk2EYLJ1Xw5LWKjbs6uE/1h/i314+wG83HeEjt8xg9Q0N2Na7f2WcYRgE5t+B3TCfxEs/IPP287hduwjd9d+wKpom6EpERERkImn4fbksGmJDRAqF6tW1y/d9MvvWkXr9nyGbYmdwCT8+Ph/XcLhtcR0fvWUmlaWhfDdT5JKoVolIoVC9EpFCoXolhSaT9Vi3/RhPv36YoXiaqtIQD9w2k5UL6zBN46LH+55LeuszpN/6FRgmwWWfwLnuAxjGuz8YIPmneiUihUL1amJo+H0RERG5ZngjgyTXPYHbuY24WcyPh1bRlq1n2fwaPnb7LOoqIvluooiIiIiIiMglc2yTu2+axm3X1fO7t7p4bkMH//c/9/Dshg4evH0WN82txjAuHO4bpkXwpgewmxaReOkfSW34F7Kd2wnd8UXMaOUEXomIiIhcTeqpL5dFT+OISKFQvbr2ZA6/ReKVH2GkYmxOzeQXI8tpnVXPg6tmMb22ON/NE3lPVKtEpFCoXolIoVC9kkI3kszw/MYjvLDpCKmMS3NdMR9fNYuFMyveNdwH8DMpUht+TmbPy2AHCS59EGfRGgzTmpjGy2VRvRKRQqF6NTEu1lNfob5cFr1xRaRQqF5dO/x0goGXf4pz+HVGvABPjSxnpHYJH1/dwuxppflunsj7ololIoVC9UpECoXqlVwrhuJp/vONDl7a2kXW9ZnVUMIHlzezpLUK8yLhfrZrJ8n1P8Ef7sWsbCJ0++exalomqOVyqVSvRKRQqF5NDIX6ckXpjSsihUL16trQuX0j9qafUewNsidTz4bi+7jr9kUsnHHxHgoihUC1SkQKheqViBQK1Su51vQPJvnPNw7z6o5usq5HXUWE+5ZPZ+XCOhzbvOBxfjZNeuvTpLc/C56HM/8Ogss+gRG8cFggE0v1SkQKherVxFCoL1eU3rgiUihUrwqX7/vs2rWfzJv/ymx3P2nfYkvRamatvp/ZTWX5bp7IFaVaJSKFQvVKRAqF6pVcqwZjKV7Y3MVLW4+SSGUpiwZYc3MTd9zQSDhoX/A4d+AYqVd/gnt8H0a4hODKz2C3LNeD8pOA6pWIFArVq4mhUF+uKL1xRaRQqF4VnqzrsXFnFwMbn2GFt4WA4dIVaqV09WdpaJ6e7+aJXBWqVSJSKFSvRKRQqF7JtS6RyvLKtmP8dlMnp2JpwkGLO5Y0smZpE2XR4HmP8X2fbNurpDb8K34qhtW4kNBtj2CW1k1w6+VsqlciUihUryaGQn25ovTGFZFCoXpVOJLpLOu2H+fQ5ldZw+tUW8PEnApCtzxM+dyb8t08katKtUpECoXqlYgUCtUrmSoyWY8Nu7p57s1Ouk+OYFsGtyyq577l06mriJz3GC85TPrNp8jsWw+WTeCGjxK44UMYljPBrRdQvRKRwqF6NTEU6ssVpTeuiBQK1avJb2gkze82d7Ft624+aL3BwsBRXNPBvuF+ipZ8EMO68PCBItcK1SoRKRSqVyJSKFSvZKrxfJ9t7X08t6GDA8eGMIAb51bzoRXNzKwvOe8x2eP7SK3/Cd6pYxildYRu/y/YDfMntuGieiUiBUP1amJcLNTXp+UiIiIyoU6cSvCbjZ1sfLuT1c7bfDW0C9vwMGctp2jlH2IWlee7iSIiIiIiIiIFwTQMbmytZsmcKtq7Bnl2Qwdb9vWyZV8v86aX8cEVzSyaWYFhGGPH2PVzsT7+GOm3nyf91q9IPPNd7Dm3EFzxh5jh8z8IICIiIvmlUF9EREQmREf3MM+92cGmvT1cb3fwjZK3KCGGWdFE8NbPYtfPzXcTRURERERERAqSYRi0NpXR2lRGV2+M5zZ0snFPD3uf2k5TTZQPLp/OzfNrsEwzt79lE1zyEZyWZSRf+ynZ9tfJdm4neNODOPPv0Oh5IiIik4yG35fLoiE2RKRQqF5NDp7n8/bBfn63pYtdh05Sa57iv1RupdE9AoEIwaUP4Sy4E8O08t1UkbxQrRKRQqF6JSKFQvVK5Iz+wSS/3XSEdduPkcq4VJWGuOemadxyXT3RsDO2n+/7ZA9tIvX6P+OPnMIoqSF488exZ92MYZh5vIJrm+qViBQK1auJcbHh9xXqy2XRG1dECoXqVX4NxlKse/s467YdpX8oRcRM87mGduYltmL4Hs7cVQSWfULD+smUp1olIoVC9UpECoXqlci5YokMv3+rixc3dxFLZHBsk2XzarjjxkZm1ZeMDc3vZ1Kkd/6W9LZnIZPArJ5JcNknsRsX5PkKrk2qVyJSKFSvJoZCfbmi9MYVkUKhejXxfN9nb+cpXt56lLfaenE9n9qozx/WdzBzaCNGJolZPZPQrY9g1czKd3NFJgXVKhEpFKpXIlIoVK9ELiyTddm8t5eXth5l/9FBAKbXRrljSSMrFtQSCuSG3PeSw6TfeprM7t+B52I1XUdw2SexKqfns/nXHNUrESkUqlcTQ6G+XFF644pIoVC9mjjxZIbXdnTz8tajdJ8cAeD66WHurzhA1fHXIJPAKK0luOR+7DkrNXSfyFlUq0SkUKheiUihUL0SuTRHTsR4eetRXt/VTSrtEgpYrFxUx51LGplWHQXAG+oltXkt2f1vAAb2nJUElz6IWVyd38ZfI1SvRKRQqF5NDIX6ckXpjSsihUL16uryfZ9Dx4d5aWsXG/ecIJP1KArZrF5Qxp3hPQT2vzw+zJ+9AsO08t1skUlHtUpECoXqlYgUCtUrkcuTSGV5c3cPL209ypETMQBmTyvlziWNLJ1bg2ObuH0dpDb+G27XTjBtnIV3E1zyUYxQNM+tL2yqVyJSKFSvJoZCfbmi9MYVkUKhenV1pNIuG3Z389LWo3T25P7Yb2ks4e6F5SzObMPd/eKZMP/GB7BblivMF3kXqlUiUihUr0SkUKheibw3vu9z4NgQL289ysY9J8i6HtGww22L67njhgZqyiNku3aR2vgUXl8HBMIEbvgwgUVrMOxgvptfkFSvRKRQqF5NDIX6ckXpjSsihUL16srq6s0Ny/fGrm4SKZdgwGLlwjruXFBKTfdrpHe+AJkkRmkdwRvvV5gvcolUq0SkUKheiUihUL0Sef9iiQyv7TjOy1uP0jOQAGDhzAruXNLI4pZy/EObSW36Jf5wL0ZROYGbPobTeps+B7hMqlciUihUrybGxUJ9ewLbIiIiIgUkk3XZsq+Xl7Yepb1rEIBp1UXcuaSR5bOjWPt+R/rFF0iPC/NXYJhmnlsuIiIiIiIiIu9VNOxw77LprLm5iT0dA7y89Shb2/rYdegk5cVBVl3fwO33/b9Ej75B+q1fk1r3BOmtzxBYfB/O3Nsx7EC+L0FEROSao576cln0NI6IFArVq/fG83z2dQ7wxu4etuw7QSLlYlsmN8+r4c4ljcysgOyuF8/TM19hvsh7oVolIoVC9UpECoXqlcjVMTCcYv32Y7yy/RgDwykMYO70Mm6dW8b12e2w93f4qRhGqBhn0RoCC+/GCF64t6GoXolI4VC9mhjqqS8iIiLvyvd9jpyIsWFXD2/u6WFgOAXA9NooKxfWccvCWiLDnaR3/wsjBzeBl1WYLyIiIiIiIjKFlBcHuf+2mXz4lmbe3t/P6zu72X6gj72dp7Ctcpa2/D/c3dRBTferpDevJb39WZx5qwlcdy9mtCLfzRcRESl4CvVFRESmqL7BBG/u7mHDrh6O9sUBqCoN8ZFbmlmxoI76MpvM/g1knv8hI/2dAFj1c3EW3I09c6nCfBEREREREZEpxjJNlrRWs6S1mngyw5Z9vWzY1c2GtlNsoJTi0Ed5oKmPG1JbYMdvyOx6EXv2LQSu/yBWeUO+my8iIlKwFOqLiIhMIbFEhs17T7BhVzdtXYMAFIVs7lzSyIqFtcxuLMUf7CG9+1fE2l6F9Ag4IZwFd+EsuAurYlqer0BEREREREREJoOikMOq6xtYdX0DJ4eSvLm7hzd29fBkewVPcg/LS09wX3gPFW3rybatx25eQuCGD2PVzs5300VERAqOQn0REZFrXDrjsv1APxt2dfP2gX5cz8exTZbNr2HFgjoWzarAMiDbuY3Es7/DPboLALO8Aefmh3Dm3IoRCOf5KkRERERERERksqooCfHBFc18cEUzXSdivLG7mzd3h3jzaC0z7F4+UrKXOR1byXZsxaprzYX7TYsxDCPfTRcRESkICvVFRESuQZ7ns7dzgA27etjSdoJEysUwYEFzOSsW1nFjazXhoI2XGCLz9rMk97yEH+sHw8SeuRRn4d1Y9fP0x7WIiIiIiIiIXJZpNVE+WTObj69uof3IKd7Y1cOP9tZTnLmOu8O7WNq9H/f5/wVljQQXfwCnZTmGE8p3s0VERCY1hfoiIiLXCNfzaDsyyNa2XjbvO8GpWBqA5rpiVi6oZdmCWsqiQXzPwz16v/f4AAAgAElEQVS2h0Tbq2QPbgIvixEpI3DjAzjz78AsKs/zlYiIiIiIiIhIoTMNg7nTy5k7vZyH17Sy42A/b+yaw28OdHK7s4uVfhuse4KRV3+G1bKCyKK7sKpn5LvZIiIik5JCfRERkQKWTGfZefAkW9t7eftAP/FkFoCq0hAfuWUGKxfWUl9ZhO97uD0HSG7fQPbgJvzEEABW/VychXdjz7gRw9RtgYiIiIiIiIhceY5tcmNrNTe2VjOSnMfmfUt4cmcHJSe2szLYRlP7Okba1xELNxCcv5ryxav0VYAiIiJn0af3IiIiBWYwlmLr/j62tfex+/AAWdcDoLm2mDVLq7hhThVNNVEAvP4OkhueIXtwY254fcAsrcNZcBdOy3LMsvq8XYeIiIiIiIiITD2RkMOq6xtYdX0DQ/Gb2L6/j617dlHdt4kl/iGct37Oybf+jd7SRUQX303D3IWYppnvZouIiOSVQn0REZECcLw/zlttvWxr7+PgsSF8wDIN5k4vY8mcam6YXUVlae7759yBY6S3vEjmwJv4g90AGNFKAtd/CLtlOWbldAzDyOPViIiIiIiIiIhASVGA269vgOsbSKXvYvf+4wzsfJW6k1toGtwG67dxYF0FJypvonzxHbS21GNbCvhFRGTqUagvIiIyCXmez4Fjg2xt72Nrex89J0cACAUsls6rYcmcKha3VBIJObn9h3pJbXuR7IE38fqPAGCES3AW3oMzewVmTYuCfBERERERERGZtIIBiyULpsGCP8T1PsXh3bsZ2fkSdYM7qDv5AumXfs+rL8ykv/ZmmhZcz3UtVYSDijhERGRq0G88ERGRSSKdcdl9eICt7b1s39/H0EgGgLJogDuWNLJkThXzppfj2Ca+7+P1d5Las4Nsx1a8EwdyJwkW4cxbhd2yAqt+HoaGpxMRERERERGRAmOZJi2LFsGiRXiZJL3bXyWz9xVuGtkPJ/fT98rT/ObFGQxVXUfT3AXc0FpDeXEw380WERG5ahTqi4iI5Inn+XSeGGb34QF2HTpJe9cgWdcDoLGqiNuvb+DG1mqa64oxDQM/FSfbuZlE5w7crh34I6dyJ3JC2LNX4rQsx5q2CMPSr3cRERERERERuTaYTojapffA0ntwTx4ltuMlSg5u5G5rJ8R20rcxykuvzuBoZB6VM+awcGYlrU1l6sUvIiLXFP1WExERmUAnTiXYffgkuw8PsOfwSeLJLACObTK3qZSFMytZ0lpFbXlkrDd+ZtsruEd24PbsBz8X+ptlDdgty7GbFmPVt2JYTj4vS0RERERERETkqrMqGild/Vn8VZ/BO3GQRNsGSg9sZI21E/yd9O4vZtvuZtZmZxKqaWbhzEoWzKhgRn0xtqXRDEVEpHAp1BcREbmKYokMezoG2H34JLsOnaRvMAmAATTXFbN6RgULZ5Qze1opjm3leuN37SSx/W3cIzvwE4O5E9lB7OnXYzUtxm66DrO4Kn8XJSIiIiIiIiKSR4ZhYtXOJlo7G/+2z+D2HCB7cCOV+3MB/xp20hsvYeuWZn78RjMnrSrmTa9gwYxyFsyooL4ygmEY+b4MERGRS6ZQX0RE5ApKZ1zajw6O9cbv7B7GH91WUxbmjhsaWDCjgnnN5UTDDn56BLdnP+7Wlxg5thf3xH7wc0eY5Y3Yc1bmeuPXtWpYfRERERERERGRdzAME7tuDnbdHIIrP43bs5/sgY1UH9rEB8wdfCC8gwGjjE1Hm3j1UBM/dysoKw6zoDkX8M+fUU5ZNJjvyxAREXlXSgdERETeh6zrceREbKw3fnvXIJlsboj8aNhh6byasafAq8vCePEB3O423Ld+Q7y7He/kkbEQHyeE3bzkTG/8aGUer0xEREREREREpLDkAv5W7LpWgrd8Bre7nezBjVQc3MwH/FzAnzbD7Pcaeautln/Z1cCwH6axqogFM3I9+edMKyUS0tcciojI5KJQX0RE5DLEkxkOHB1k/9FB9ncNcvD4EOlMLsR3bJPWaaWjfwRWMK0mAoPduN3tuFvbiHW34w/3jp3LCJdiz7gJq34uVt0czIomDNPK16WJiIiIiIiIiFwzDMPErp+LXT8Xf+XDuD3tuEd2YHbtYEHffhZE9wNw0q5hZ6qebdvq+N3manxMGqqKmD2tlNmNpcyeVkpNWVjD9YuISF4p1BcREbkA3/fpGUiwv2uQ/UdPsf/oEMf64mPbg45FS0PuD7y508uYXe1gDnbhntiFt72dRHc7fio2tr9ZWoc9dxVWfWtuOP3iav1BKCIiIiIiIiJylRnmmYA/uOwTeCODuF07yXbtoPLITlZZ21lVsp2sGeSYM53t8Vq2vF3DK9uiAJREHFpGA/7ZjaXMqCvGsdUxQ0REJo5CfRERkVHpjMvh7uGxXvj7jw4SS2TGtleWBFm+oJbZjaXMqTSoM/rwT3bi9b2G+0YHqbN64WNYmFXN2HW35kL82jmY4ZI8XJWIiIiIiIiIiJzNjJRitt6K03orvufh9R0mO9qLf/qJ/Uy32/loGaTCNXQ5M9gxUsWmAzG2tvcBYFsGzXXFzGksGwv7S4sCeb4qERG5linUFxGRKetULMWBo4O0dw1y4Oggh7uHcb3c99ubhsH02igrFtQwr8pnZvAUkZHjuH0b8fZ04I+cInX6RIaBWVaPNXslVlUzZlUzVvUsDCeYt2sTEREREREREZGLM0wTq2YWVs0sgjc9gJ+MkT26m+yRHRhdO2gZ2kgL8LEyyBTV0huYRluqmjf7Ejx/dBDIjcJYUxampbGUOaO9+RuqijBNjdAoIiJXhkJ9ERG55vm+T/9gko6eGB09w3T2DNPRM8xgLD22T1HI5sbmMIsqUjSHY1RyCmPwGN6xDvxDuSH00wCmhVk+DbtpMWbVdKyqGZgVTQrwRURERERERESuAUYoitOyDKdlGb7v4w0cxe1uwz3ehtHdRsPAFhqAOyLgVZZxKjydQ24tW06VsmHXCG/s6gYg4Jg01URpri1mem0xzbXFNFYXYVtmfi9QREQKkkJ9ERG5pni+T8/JETp7hukd6mTv4X46uoeJJ7Oje/hUWAkWV6SZXR2nMThMuTeAHeuGwWEYHD0PgBXArGzCrlqW631f1YxZ3ohhOXm6OhERERERERERmSiGYWBVTMOqmAYL7gLAG+7Lhfzdbbjd7VScfJsK4CYD/LoII9Fmuow69oxUsrXf48DRobHzWaZBY3URzbXFzKgrZvHcWoqDJgHbytMViohIoTB83/fz3Yj3amAgTjbr5bsZU4phQFVVMX19wxTuvxwRuVZkXY9jfXE6R3vgd/QMc+REjFTaJUCGKmuYukCcluI0TaFhqvwBwqlezGxy/IkCYcyyBqzyBsyyBszyesyyRoziSgxDT0+LyNWjeysRKRSqVyJSKFSvRGSi+ckYbk87bnc72e42vN5D4Lm5jYaBV1LPcKiebr+KtkQp2/vD9MbOFCjDgPrKIqbX5nr153r2R4mE1KlERCYH3V9NDNs2KS8vuvD2CWyLiIjIe+L7PqdiaY71xTnWH+dob5zO7iEG+/soY4gqc5gqa5jbnDj1JXHKjWGCbvzMCbJADIxwCWb1DMzyBsyyeszyRsyyeoxIGYah7zgTEREREREREZHLY4Si2M1LsJuXEAT8bBr3xEHcnna83sO4fYcp7dlCKTAX+GjAwJ9RRzzSwMlAHXsGo2w7abJhV5wNu3rGzltdFqK5tphpNVEaKotoqCqipjys4ftFRKYohfoiIjJpeL7PyaEkx/pGONYbo6+3j5H+HrJDfUTcYSrMOJXWMCvNYT5qxQiUZM85hxEswSipxyqppqh2GkmnDKO4BqusHiMUzcNViYiIiIiIiIjIVGHYAeyGedgN88bWeclhvL4O3L7Do0F/B9HuLUSB6cC9lgEzahkpaqCHag6nitl10mPLvgSb9/WOnccyDWrKwzRUFY0F/Q1VRdRVhHE0hL+IyDVNob6IiEw4z/Pp7R+k99hRBk90kzjZizfcj5UaoJQYZeYIN5txAsboUGWhM8f6hglFFdilczFLajBLajBKajBLqjGLqzECYSA3JFCFhgQSEREREREREZE8M0PFmNMWYU9bdGZlOk40fYKTB/bgjgb9keNvMROYCdwJUBsmE61j2KnihF9OR7KYvUMub7XF2eKfGXXSMKCmLEz9WNAfoaGqiPqKIoIBhf0iItcChfoiInJF+Z6LnxjCiw8QP9nHcH8vyVP9ZGMDGIkB7PQwRV6MIjNF8zsPdiBrBMiEyvCijVBWTaC0CjNaiRGtHJ1XYJj6Y0RERERERERERAqXESwi3LiYYPHMsQ4pfiqO29+JN3AUb+AY3sBRjJNHqUgdogKYB9xrALURstE6hgNVnPDK6UwVs3fQZ8fBBNv29437OVWlodHe/BGqy8LUlIepKQtTWRrSUP4iIgVEob6IiFwSP5PCTwzmAvvEIH5iGG/kFImBftLDJ/Hjp7BSgwTcOCa5v0RMoHR0Akj7FjGKGHYqORUuxyquIlJZQ1lNHaGKGsyiCghE9P32IiIiIiIiIiIy5RjBIuyG+dAwf2yd7/u5z+POCvq9gaMwcJTykwcpB+YCawCqg7iRauJOBScp4Xi6iEOxCG0dp3j7QBAY37u/siQ0FvJXn56PBv+hgOIjEZHJRFVZRGSK8n0f0iOjIf1QLrAfGZ0nhvETg3iJIbyRXJBvuOnznscGDN9gyAsz6EcY8ipIOcX4oVKsaDnB0kqiFdWU19RSWV1OZdCZ2AsVEREREREREREpUIZhYERKMSOl0LhgbH0u7B8cF/R7gz0YQycoGdxBCT4zgJUAJeDbQTLhKmJ2OQOU0J0ppmMkzP5jQXYfDnB24A9QEnHOCfprynO9/YsjDqY65YiITCiF+iIi1xDf8/BTsXHB/OmgPhfcj5/wshc8l4dBzAsx5IUY9isZ9sIMeyGG/TBJI4xZVEqouJxwWRVlVRXUlBdRUx5mfomG7hIREREREREREbmacmF/GWakbFzYD+Bn03jDvXiDPfhDPXiDPXhDJzAHewgM7KICnxbgVoAo+JaDGyon4ZQyRDH92QjHU2E6TwXYfSzEkB/G58znfbZlUF4cpKI4REVJiIqSYG5enJtXlgQJB22NxikicgUp1BcRmcT8bBo/OZybEsNnlpOxc18nh/FTMca+hOs8XMMiYUSI+WFOZes5lQ3mgnovzJCfm8f8EAkjQrC4ZPSmPDQ2FFfT6FO5pUUB3ZSLiIiIiIiIiIhMQoYdwCpvxCpvPGfb6cDfHzyBN9SNN3gCL9aPNdyHHeugOJumEVgMEMxNvmGSCZYxYpUySJR+t4ieVJjj/QEOHg3ylhchRWDczwkGrLGQv6I4SGVJiPJ3hP9Bx5qI/xwiItcEhfoiIhPE9z38VPysgD52bmCfGh/Wkz3/kPdj58Qga4dJmRESlBKzahh0Q5zMBOhPObne9aNh/ZAXIoWDaZiUFQdGn6TNPVHbWBLkupIzr4sjjkJ7ERERERERERGRa8zpwJ/zBf6+n/tcMtaPN9yHH+vDG+7HHw39A8PdlGUSNJ8+IDQ6AZ4VJO0UM2JGGSLKQDbEiXSI490OHZ1htnsRhv3QuB7/4aBFSVGQ0qLAmSkaoLQoODrPTcWRAKapzypFZGpTqC8i8h74bjYXwCfjo/PYmXkyBqnYaIAfu+Re9ACeFcC1I6StCMlgGfFgiGE3yGA2wMm0Q1/SZjAbIOaFiPtBRvzAuBvhgGNSVhSkpCR3w1sVDTL3rLC+oiR3Q2yZGh5fREREREREREREzjAMAyNcAuESrOqZ593HT8XxYv34w/14IwP4sZO5efwUdnyAULybikySGacPCI9OjPb4t6MkzCJiRBjyQgxkQ/QPBOjtDnDQDzHkhRnywqRxzmoXFEcC7wj/cw8DlBQFKI44RMMOxZEA0bCNY2sEABG59ijUF5Epzfc9SCfGB/DvDOlPbzvrNZnkxc+NgedEyFgRUlYFiUgDcULE3CCDbpBTGYf+lEN/0ibmB4l5ITLnKcumYVBS5OSeUK0M0HCem9fTT66GAirrIiIiIiIiIiIicnUYwSKsYBFUTr/gPn46MRb0+/EBvPhJ/PgAfnwAa+QUwZFBShMnaPTc3AGB0eksrumQtqKMmBFifoRBN5T7PLXbYdAN0OuFiPkhYl6QuB/E5UyQHwxYREMO0YhDcTg3j4ZHl8MO0UjgzOvRbbalTlAiMrkp/RGRguf7PmSSufA9PYKfGsFPxyF1ennkrG3xXCh/OqxPxcH3LvozXCtA1oqQNsOknBISTogRP0jMDTDsBhjMOpxKO/SnbOJegLgfJOkH8Dl3WCjLNHI3j2GHaIVDTcShZewGMvc0aUnR6DBTRQGiEQdTQ+GLiIiIiIiIiIhIATACYaxAGMoaLriP7/uQiuMlBvETQ/gjg7kpMYiXGMQaGcRJDBIZGaQyeSw3AqrBWK//d8qYQdJmhIQRIu6Hc6OfJgMMDDkMZR16/CCHRkc+jXtBRvwg2bMeBAgFLIpCDpGQTSRonzU/37rR9aPLoYClrzIVkatOob6ITAq+mxkL3s+E8fFxoTzps0P6M0E96ZGLDmt/moeZC+bNEElKGLFqiPtBYq7DUDbAqbTDsJsL5eN+kJHRgP7sJz3PFgpYozdzDpEii8pIgOawMzbk09h0+qnQcIBwUDd5IiIiIiIiIiIiMnUZhgGhKFYoCuWN77qv73m5rzc9e0qMX7aSwwSTw0STMaoSJ8B3wQRCFz6vazikzRBJI9eBa8QLEMsEGE44DGUcRnyHPj9A0g+Q8AMkPIeEHyDpO6SxYbRDl2FwVujvEA5ahAK5sD832YTOt+70cjC3HA7YOLZGDBCR81OoLyLvie97kE3jZ5KQTuJnkviZRK7HfCaJn06eWR597aUTo1NuHaOTkU1getlL/tkpAiQJkvADjHglxL1K4qM3VCNekITvkPBz2xO+w4gfJOEFGPEDZLAYu9kiNxTT6acwwyVnnrSsCo5/2vLME5hn1oeDlr6bXkREREREREREROQqMkwTI1IKkdJL2t/3/VwHsWQsF/qnYvjJ+NjIrX4qN4qrlYrjJONEUnHKUwOQGgHDB4fc9C48TLJmkLQRJEWABLnQP55yGBmxibsWCc8h6TvEfIekb5Pyc6/Pnqexx432apnGuNA/4FgEHZOAc2Y5OLZsETj92rYIBkb3HV0O2Gf2dWwTxzaxTEMdzkQK1CWH+kNDQ3zrW99i3bp1RKNR/viP/5iHH374vPtu3LiRxx57jCNHjjB37ly+853vMGfOnLHtTz75JD/4wQ+IxWKsXr2av/qrvyIajb7/qxGRMb7n4WfTeNkU2VSabCaJm07jplK4mRRuJo2XSeNlUniZNL6bxs+k8bNpyKYgm8LIpjCySUw3heWmML0UlpvG8tLYfvo8A8tfXMY3x25a0r5N0g+R8EtygfzpJx79AAkvMG7diB8gbQTx7BCOY4/dqJzvqceygEXdO9aHz/M0ZMCxNKy9iIiIiIiIiIiIyDXEMAwIFmEEi6C09pKP830P0olc8J+M4acTuVFjR0eL9dOJ0ZFkE5AewUmPEBpbfxLcBJh+boSAizwUMPYzAdcMkDECZLHJYJP2bdK+RcqzSSUsUnGLpGeRdE0y2KR8m7hvM+BbpLHJjO5/Zjn3+vS5spicPapAwD4T8p+eAraJY5k4joVjmQSc0de2iXPW/rZl4Fgmtm1iW7nXuXlucizjrG3jt58+3rb0cIHIe3HJof5jjz2G67qsX7+ezs5OvvCFL9DS0sKKFSvG7TcwMMCjjz7KX/zFX3DffffxxBNP8Oijj/Lcc89h2zavvfYajz/+OE888QRNTU18/etf5y//8i/57ne/e8UvTqY23/fxfB/fB88bnfv+6PrRZS+37Ps+npfb3xvd/8xrH8/jzOuz556H53r4XhbfPT25eKdfe25uns0t443O3TOvDc/F81wMLwuei+G54Lm54YE8F8N3Mb0sppfB9LNYfm5u+1ksP4tFbtkmi42LMzq3jXO/J97ksu4nAMj6JinfHntyMOWHSVI8uuyQ8u3cDYcRwLWCubkdxDOD+FYA3w7i22FwguCEcBwHx7YIBkafGBx9mjDsWJSd7wnDs55EtC31ihcRERERERERERGRK88wzDMPA5TUXPbxvu9BJnVmVNuzR7g973JuJFs7nSCQSYJ7ptPdWOc7srkP9d8Hn9xXDWQNZ+zBgSwWGd/C9U0yrkk2Y5LxTTKeQdo3yXomWUxc38rNMRk5veybZLFG5yZZ38Jl/PLpfbJ+7tisb+Jh4mHkJt/AMK3cKAyWhWmaWJaJbZpYloFlGlhm7iGA3OvcgwCnHwiwRh8OMA1jbH/TNLCM0bl5Zn5m2Txn2+njzdHjTIPRufGO+UXWn143ut4wwDRG56aBgYFp5h44ye1zZlnkUl1SqD8yMsLzzz/Pf/zHfxCNRlmwYAEPPvggv/zlL88J9V944QVmzJjB/fffD8AXv/hFfvKTn7Bp0yZWrlzJ2rVreeihh5g/fz4AX/nKV/jEJz7Bt7/9bcLh8GU13lLAd1W5/Ufw+g6Pfld5LhTf33WKtzMenueCn/tlcOa7zP3RCYzR/cfW+2dtP2f9+GPHnW/cvuduN0aPMkaPN0fbaoxO5un1hj/6ey+37sx2xu1rGLlzmvijg954GAZj63K/bnwsA0w8rPME55fNAKzR6RJ5vkUWB9fI/fL1sHCxcA2LpGHjGja+YeXmpoVv2PiWjW/YYNlgOmBZYDkYpyfbxrAcTNvBsIMYTgDTCWI7DqZpELVMyk4/VWfnfrk6du4XH++pz77I1XX6fsi2zbPqkYjI5KJaJSKFQvVKRAqF6pWIFArVq2uRCY4NFF25U/perpOgm8mF/m4m9zqbyX09rpuBsXVpcDNj63w3A9nMWftkcsed3jbawfB0B0Q8l1zi4o1OEyuXwJxJcDzfGPcQgH/WsucbeNnRdZmztp117Om06vS5T6dHnD33z061Tq83cAH37ON8zjruzL7jf8bpKMwY/a83frv3zv3GcpXR+ehDAGCM5VIYY2MsnNnXgAFK6KI+93L04QDjrHOYY+c6c6Q5etLczBg7tzF6rGGcWYdx+lmS3Hlsy+TBu+YQUL26qi6We19SqH/48GEAZs+ePbZu3rx5/PjHPz5n37a2NubNm3dWAyzmzJlDW1sbK1eupL29ndWrV49tb21txfM8Ojo6xh13KUpKLu8hALlM5fNg9vj/J5c+UI2IyORQVnYFb6JFRK4S1SoRKRSqVyJSKFSvRKRQqF6JiMiluKSu7iMjIxQVjf/FUlJSQjweP+++xcXFF9z3ndsNwyAajRKLxS678SIiIiIiIiIiIiIiIiIiIteySwr1I5HIOQH+8PDwOUH/6X3fGdCfve/5tsdiMaLR6GU1XERERERERERERERERERE5Fp3SaH+jBkzADhw4MDYur179zJnzpxz9m1tbWXv3r1jrz3Po62tjdbWVgDmzJkzbntbWxumadLc3PyeLkBERERERERERERERERERORadck99e+9916+//3vE4vF2Lt3L2vXruWhhx46Z981a9Zw6NAhnnnmGdLpND/84Q8pKiri5ptvBuChhx5i7dq17N27l1gsxve//30+9KEPEQ6Hr+yViYiIiIiIiIiIiIiIiIiIFDjD933/UnYcGhrim9/8JuvXr6eoqIgvfelLPPzwwwAsWbKEf/qnf2Lp0qUAvPnmmzz22GMcOXKEuXPn8td//dfjevX/9Kc/5Qc/+AHxeJxVq1bxne98R8Pvi4iIiIiIiIiIiIiIiIiIvMMlh/oiIiIiIiIiIiIiIiIiIiIysS5p+H0RERERERERERERERERERGZeAr1RUREREREREREREREREREJimF+iIiIiIiIiIiIiIiIiIiIpOUQn0REREREREREREREREREZFJSqG+iIiIiIiIiIiIiIiIiIjIJKVQX0REREREREREREREREREZJJSqC+XZMOGDTzyyCPcdNNN3Hrrreds/8Y3vsGiRYtYsmTJ2HTs2LE8tFREprqL1auhoSG+8pWvsGTJEm6//XZ+9rOf5aGVIiLnWrt2LfPnzx93P/XrX/86380SEdH9k4gUBH02JSKT1ZNPPslDDz3EokWL+NM//dNx29ra2vjUpz7F9ddfz0c+8hE2b96cp1aKiLx7vbrrrrtYvHjx2H3Whz/84Ty1cuqy890AKQyRSISPf/zj3H///fz93//9eff5/Oc/z5/92Z9NcMtERMa7WL167LHHcF2X9evX09nZyRe+8AVaWlpYsWJFHlorIjLeddddx1NPPZXvZoiIjKP7JxEpFPpsSkQmo5qaGh599FFef/11BgYGxtZnMhm+9KUv8Qd/8Ac8+eSTPPfcczz66KO88MILlJaW5rHFIjJVXahenfb444+zatWqPLRMQD315RItXryYj33sY0yfPj3fTREReVfvVq9GRkZ4/vnn+epXv0o0GmXBggU8+OCD/PKXv8xDS0VEREQmP90/iYiIiLw/H/jAB7jnnnsoLy8ft37jxo0kk0m++MUvEggEeOCBB5g2bRq//e1v89RSEZnqLlSvZHJQqC9XzFNPPcWyZcu4//77+cUvfpHv5oiInOPw4cMAzJ49e2zdvHnzaG9vz1OLRETG27dvHytWrGDNmjV873vfI5lM5rtJIjLF6f5JRAqJPpsSkULS3t5Oa2srpnkmptF9lohMZt/4xjdYsWIFjzzyCFu2bMl3c6YcDb8vuK6L7/vn3WYYBpZlXfQcjzzyCF//+tcpLS1l8+bNfPnLX6a4uJh77733SjdXRKaw91uvRkZGKCoqGreupKSEeDx+xdooInI+l1K/br75Zp555hkaGxs5fPgwf/7nf873vvc9vvWtb01wa0VEztD9k4gUCn02JSKFJh6PU1xcPDkE/P4AAANoSURBVG5dSUkJw8PDeWqRiMiF/e3f/i2LFi0CYO3atfzRH/0RTz/9NI2NjXlu2dShUF/4/Oc/z8aNG8+7raqqitdee+2i51i4cOHY8vLly3n44Yd5/vnn9YeTiFxR77deRSKRcz6AHh4ePueDahGRK+1S6ldTU9PYulmzZvG1r32Nr33tawr1RSSvdP8kIoVCn02JSKEpKioiFouNW6f7LBGZrJYuXTq2/JnPfIZnn32WdevW8elPfzqPrZpaFOoLP/3pT6/4OU3TvGBvNBGR9+r91qsZM2YAcODAAVpaWgDYu3cvc+bMeb9NExF5V++lful+SkQmA90/iUih0r2UiEx2c+bM4Yc//CGe540Nwb9nzx4FZCJSEAzD0L3WBDMvvosIeJ5HKpUik8kAkEqlSKfTY9ufe+45YrEYnuexefNmnnzySdasWZOv5orIFPZu9SoSiXDvvffy/e9/n1gsxt69e1m7di0PPfRQPpssIgLAK6+8Qm9vLwBHjhzh7/7u77jnnnvy3CoRmep0/yQihUKfTYnIZJXNZkmlUmSz2XGfWy1btoxAIMCPfvQj0uk0Tz/9NF1dXapdIpI3F6pXx44dY/PmzaTTadLpNE899RQ7d+7ktttuy3eTpxTD12MUcgnefPNNPve5z41b19jYyO9//3sAHn74Yfbt24frujQ0NPDZz35WTxSKSF5crF4NDQ3xzW9+k/Xr11NUVMSXvvQlHn744Xw0VURknO9+97v8+te/Jh6PU15ezr333suXv/xlIpFIvpsmIlOc7p9EpBDosykRmaz+4R/+gccff3zcugcffJC/+Zu/Yd++fXzzm99k3759NDU18e1vf5ubb745Ty0VkanuQvXqi1/8Il/72tfo7OzEcRxaWlr46le/yvLly/PU0qlJob6IiIiIiIiIiIiIiIiIiMgkpeH3RUREREREREREREREREREJimF+iIiIiIiIiIiIiIiIiIiIpOUQn0REREREREREREREREREZFJSqG+iIiIiIiIiIiIiIiIiIjIJKVQX0REREREREREREREREREZJJSqC8iIiIiIiIiIiIiIiIiIjJJKdQXERERERERERERERERERGZpBTqi4iIiIiIiIiIiIiIiIiITFIK9UVERERERERERERERERERCap/x9k4acKZKgeVQAAAABJRU5ErkJggg==\n", | |
"text/plain": [ | |
"<Figure size 2560x640 with 1 Axes>" | |
] | |
}, | |
"metadata": {}, | |
"output_type": "display_data" | |
} | |
], | |
"source": [ | |
"fig = plt.figure(figsize=(32,8), dpi=80, facecolor='w', edgecolor='k')\n", | |
"ax = plt.subplot(1,1,1)\n", | |
"x = np.linspace(-15,15,100)\n", | |
"ax.plot(x, scipy.stats.norm(loc=mu, scale=sigma).pdf(x))\n", | |
"sns.kdeplot((y-y_), bw='scott')\n", | |
"#sns.kdeplot((y-y_), bw=1.0)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 55, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"-1002.884022289862" | |
] | |
}, | |
"execution_count": 55, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"model1_ll = np.log(scipy.stats.norm(loc=mu, scale=sigma).pdf(y-y_)).sum()\n", | |
"model1_ll" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 56, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"-1003.8882496353907" | |
] | |
}, | |
"execution_count": 56, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"lr = sklearn.linear_model.LinearRegression()\n", | |
"X, y = df[['x2']].values, df['temperature'].values\n", | |
"lr.fit(X, y)\n", | |
"# lr.coef_.squeeze()\n", | |
"lr.intercept_, lr.coef_\n", | |
"y_ = lr.intercept_ + lr.coef_[0] * df['x2']\n", | |
"mu, sigma = scipy.stats.norm.fit((y-y_))\n", | |
"model2_ll = np.log(scipy.stats.norm(loc=mu, scale=sigma).pdf(y-y_)).sum()\n", | |
"model2_ll" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 57, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"0.15642458027967598" | |
] | |
}, | |
"execution_count": 57, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"lmda = 2 * (model1_ll - model2_ll)\n", | |
"1-scipy.stats.chi2.cdf(lmda, 1)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 58, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
"Likelihood ratio test\n", | |
"\n", | |
"Model 1: temperature ~ month + x2\n", | |
"Model 2: temperature ~ x2\n", | |
" #Df LogLik Df Chisq Pr(>Chisq)\n", | |
"1 4 -1002.9 \n", | |
"2 3 -1003.9 -1 2.0085 0.1564\n", | |
"\n" | |
] | |
} | |
], | |
"source": [ | |
"print(rtest_result)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": null, | |
"metadata": {}, | |
"outputs": [], | |
"source": [] | |
} | |
], | |
"metadata": { | |
"kernelspec": { | |
"display_name": "Python 3", | |
"language": "python", | |
"name": "python3" | |
}, | |
"language_info": { | |
"codemirror_mode": { | |
"name": "ipython", | |
"version": 3 | |
}, | |
"file_extension": ".py", | |
"mimetype": "text/x-python", | |
"name": "python", | |
"nbconvert_exporter": "python", | |
"pygments_lexer": "ipython3", | |
"version": "3.6.10" | |
} | |
}, | |
"nbformat": 4, | |
"nbformat_minor": 2 | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment