diff --git a/wk7/week7.tex b/wk7/week7.tex
index 3ddb765..a293c64 100644
--- a/wk7/week7.tex
+++ b/wk7/week7.tex
@@ -26,6 +26,76 @@
 
 	\section{Results} \label{sec:results}
 
+	\subsection{The advantage of height}
+
+	\begin{table}[ht]
+		\centering
+		\label{tab:chi-height}
+		\begin{tabular}{|l|r|r|r|r|}
+			\hline
+												& \textbf{M: 168 - 188} & \textbf{M: 189 - 210} & \textbf{F: 155 - 171} & \textbf{F: 172 - 189} \\ \hline
+		\textbf{1 - 99}    & 67 / 73               & 32 / 26               & 38 / 42               & 60 / 55               \\
+		\textbf{100 - 199} & 69 / 72               & 30 / 26               & 31 / 27               & 32 / 36               \\
+		\textbf{200 - 299} & 75 / 68               & 17 / 25               & 18 / 17               & 22 / 23               \\
+		\textbf{300 - 399} & 61 / 60               & 21 / 23               & 11 /12                & 17 / 16               \\
+		\textbf{400 - 499} & 59 / 60               & 22 / 22               & 7 / 6
+		& 7 / 8 \\
+		\hline
+		\end{tabular}
+		\caption{Observed / Expected values used for the $\chi^2$-test. The groups are divided by their rank (vertical) and, per gender, their height (horizontal).}
+	\end{table}
+
+	$$
+		\chi^2 \approx 7.697606186049128
+	$$
+
+	$$
+		df = (5-1)(4-1) = 12
+	$$
+
+	$$
+		\chi^2(7.69\dots,12) \approx 0.8082925814979871
+	$$
+
+
+	\textbf {t-test men:} T score: 1.711723, P score: 0.043815
+
+	\textbf {t-test women:} T score: 1.860241, P score: 0.032030
+
+
+	\subsection{The advantage of left-handedness}
+
+	\begin{table}[ht]
+		\centering
+		\label{tab:chi-hand}
+		\begin{tabular}{|l|l|l|l|l|l|}
+		\hline
+								& \textbf{1 - 99} & \textbf{100 - 199} & \textbf{200 - 299} & \textbf{300 - 399} & \textbf{400 - 499} \\
+		\hline
+		\textbf{L} & 22 / 21         & 23 / 18            & 17 / 15            & 6  / 12            & 8 / 10             \\
+		\textbf{R} & 174 / 177       & 139 / 144          & 117 / 119          &
+		105 / 98           & 88 / 86 \\
+		\hline
+		\end{tabular}
+		\caption{Observed / Expected values used for the $\chi^2$-test. The groups are divided by which hand they use (vertical) and their rank (horizontal).}
+	\end{table}
+
+
+	$$
+		\chi^2 \approx 6.467312944404331
+	$$
+
+	$$
+		df = (2-1)(5-1) = 4
+	$$
+
+	$$
+		\chi^2(6.46\dots,4) \approx 0.1668616190847413
+	$$
+
+	\textbf {t-test:} T score: 0.451694, P score: 0.325815
+
+
 	\section{Discussion} \label{sec:discussion}
 
 \end{document}
diff --git a/wk7/wk7.ipynb b/wk7/wk7.ipynb
index 4caf0b3..64bdd23 100644
--- a/wk7/wk7.ipynb
+++ b/wk7/wk7.ipynb
@@ -16,7 +16,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Using matplotlib backend: Qt5Agg\n",
+      "Using matplotlib backend: MacOSX\n",
       "Populating the interactive namespace from numpy and matplotlib\n"
      ]
     }
@@ -43,106 +43,106 @@
      "data": {
       "text/html": [
        "<style  type=\"text/css\" >\n",
-       "    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row0_col0 {\n",
+       "    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row0_col0 {\n",
        "            background-color:  #fc7f00;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row0_col1 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row0_col1 {\n",
        "            background-color:  #ffcc09;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row0_col2 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row0_col2 {\n",
        "            background-color:  #faec2c;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row0_col3 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row0_col3 {\n",
        "            background-color:  #ffd20d;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row0_col4 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row0_col4 {\n",
        "            background-color:  #ffb300;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row0_col5 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row0_col5 {\n",
        "            background-color:  #ffc404;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row0_col6 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row0_col6 {\n",
        "            background-color:  #ffbc00;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row1_col0 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row1_col0 {\n",
        "            background-color:  #ffcc09;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row1_col1 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row1_col1 {\n",
        "            background-color:  #fc7f00;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row1_col2 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row1_col2 {\n",
        "            background-color:  #ffb600;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row1_col3 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row1_col3 {\n",
        "            background-color:  #ffc908;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row1_col4 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row1_col4 {\n",
        "            background-color:  #e4ff7a;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row1_col5 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row1_col5 {\n",
        "            background-color:  #ffb400;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row1_col6 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row1_col6 {\n",
        "            background-color:  #ffbf01;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row2_col0 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row2_col0 {\n",
        "            background-color:  #faec2c;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row2_col1 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row2_col1 {\n",
        "            background-color:  #ffb600;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row2_col2 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row2_col2 {\n",
        "            background-color:  #fc7f00;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row2_col3 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row2_col3 {\n",
        "            background-color:  #ffbd00;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row2_col4 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row2_col4 {\n",
        "            background-color:  #ffc907;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row2_col5 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row2_col5 {\n",
        "            background-color:  #ffb400;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row2_col6 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row2_col6 {\n",
        "            background-color:  #ffbf01;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row3_col0 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row3_col0 {\n",
        "            background-color:  #ffd20d;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row3_col1 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row3_col1 {\n",
        "            background-color:  #ffc908;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row3_col2 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row3_col2 {\n",
        "            background-color:  #ffbd00;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row3_col3 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row3_col3 {\n",
        "            background-color:  #fc7f00;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row3_col4 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row3_col4 {\n",
        "            background-color:  #ffb500;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row3_col5 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row3_col5 {\n",
        "            background-color:  #ffc908;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row3_col6 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row3_col6 {\n",
        "            background-color:  #ffc304;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row4_col0 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row4_col0 {\n",
        "            background-color:  #ffb300;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row4_col1 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row4_col1 {\n",
        "            background-color:  #e4ff7a;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row4_col2 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row4_col2 {\n",
        "            background-color:  #ffc907;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row4_col3 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row4_col3 {\n",
        "            background-color:  #ffb500;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row4_col4 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row4_col4 {\n",
        "            background-color:  #fc7f00;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row4_col5 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row4_col5 {\n",
        "            background-color:  #ffcc09;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row4_col6 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row4_col6 {\n",
        "            background-color:  #ffbb00;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row5_col0 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row5_col0 {\n",
        "            background-color:  #ffc404;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row5_col1 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row5_col1 {\n",
        "            background-color:  #ffb400;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row5_col2 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row5_col2 {\n",
        "            background-color:  #ffb400;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row5_col3 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row5_col3 {\n",
        "            background-color:  #ffc908;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row5_col4 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row5_col4 {\n",
        "            background-color:  #ffcc09;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row5_col5 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row5_col5 {\n",
        "            background-color:  #fc7f00;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row5_col6 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row5_col6 {\n",
        "            background-color:  #fd8f00;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row6_col0 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row6_col0 {\n",
        "            background-color:  #ffbc00;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row6_col1 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row6_col1 {\n",
        "            background-color:  #ffbf01;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row6_col2 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row6_col2 {\n",
        "            background-color:  #ffbf01;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row6_col3 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row6_col3 {\n",
        "            background-color:  #ffc304;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row6_col4 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row6_col4 {\n",
        "            background-color:  #ffbb00;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row6_col5 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row6_col5 {\n",
        "            background-color:  #fd8f00;\n",
-       "        }    #T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row6_col6 {\n",
+       "        }    #T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row6_col6 {\n",
        "            background-color:  #fc7f00;\n",
        "        }</style>  \n",
-       "<table id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6\" > \n",
+       "<table id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2\" > \n",
        "<thead>    <tr> \n",
        "        <th class=\"blank level0\" ></th> \n",
        "        <th class=\"col_heading level0 col0\" >ranking</th> \n",
@@ -154,73 +154,73 @@
        "        <th class=\"col_heading level0 col6\" >height</th> \n",
        "    </tr></thead> \n",
        "<tbody>    <tr> \n",
-       "        <th id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6level0_row0\" class=\"row_heading level0 row0\" >ranking</th> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row0_col0\" class=\"data row0 col0\" >1</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row0_col1\" class=\"data row0 col1\" >-0.165935</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row0_col2\" class=\"data row0 col2\" >-0.586707</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row0_col3\" class=\"data row0 col3\" >-0.244073</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row0_col4\" class=\"data row0 col4\" >0.17403</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row0_col5\" class=\"data row0 col5\" >-0.0826093</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row0_col6\" class=\"data row0 col6\" >0.0196139</td> \n",
+       "        <th id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2level0_row0\" class=\"row_heading level0 row0\" >ranking</th> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row0_col0\" class=\"data row0 col0\" >1</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row0_col1\" class=\"data row0 col1\" >-0.165935</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row0_col2\" class=\"data row0 col2\" >-0.586707</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row0_col3\" class=\"data row0 col3\" >-0.244073</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row0_col4\" class=\"data row0 col4\" >0.17403</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row0_col5\" class=\"data row0 col5\" >-0.0826093</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row0_col6\" class=\"data row0 col6\" >0.0196139</td> \n",
        "    </tr>    <tr> \n",
-       "        <th id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6level0_row1\" class=\"row_heading level0 row1\" >age</th> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row1_col0\" class=\"data row1 col0\" >-0.165935</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row1_col1\" class=\"data row1 col1\" >1</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row1_col2\" class=\"data row1 col2\" >0.121731</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row1_col3\" class=\"data row1 col3\" >-0.140033</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row1_col4\" class=\"data row1 col4\" >-0.994296</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row1_col5\" class=\"data row1 col5\" >0.157223</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row1_col6\" class=\"data row1 col6\" >-0.0282972</td> \n",
+       "        <th id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2level0_row1\" class=\"row_heading level0 row1\" >age</th> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row1_col0\" class=\"data row1 col0\" >-0.165935</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row1_col1\" class=\"data row1 col1\" >1</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row1_col2\" class=\"data row1 col2\" >0.121731</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row1_col3\" class=\"data row1 col3\" >-0.140033</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row1_col4\" class=\"data row1 col4\" >-0.994296</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row1_col5\" class=\"data row1 col5\" >0.157223</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row1_col6\" class=\"data row1 col6\" >-0.0282972</td> \n",
        "    </tr>    <tr> \n",
-       "        <th id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6level0_row2\" class=\"row_heading level0 row2\" >points</th> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row2_col0\" class=\"data row2 col0\" >-0.586707</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row2_col1\" class=\"data row2 col1\" >0.121731</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row2_col2\" class=\"data row2 col2\" >1</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row2_col3\" class=\"data row2 col3\" >-0.004905</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row2_col4\" class=\"data row2 col4\" >-0.129971</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row2_col5\" class=\"data row2 col5\" >0.159385</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row2_col6\" class=\"data row2 col6\" >-0.0153843</td> \n",
+       "        <th id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2level0_row2\" class=\"row_heading level0 row2\" >points</th> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row2_col0\" class=\"data row2 col0\" >-0.586707</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row2_col1\" class=\"data row2 col1\" >0.121731</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row2_col2\" class=\"data row2 col2\" >1</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row2_col3\" class=\"data row2 col3\" >-0.004905</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row2_col4\" class=\"data row2 col4\" >-0.129971</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row2_col5\" class=\"data row2 col5\" >0.159385</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row2_col6\" class=\"data row2 col6\" >-0.0153843</td> \n",
        "    </tr>    <tr> \n",
-       "        <th id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6level0_row3\" class=\"row_heading level0 row3\" >tournplayed</th> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row3_col0\" class=\"data row3 col0\" >-0.244073</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row3_col1\" class=\"data row3 col1\" >-0.140033</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row3_col2\" class=\"data row3 col2\" >-0.004905</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row3_col3\" class=\"data row3 col3\" >1</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row3_col4\" class=\"data row3 col4\" >0.13293</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row3_col5\" class=\"data row3 col5\" >-0.139194</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row3_col6\" class=\"data row3 col6\" >-0.0712482</td> \n",
+       "        <th id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2level0_row3\" class=\"row_heading level0 row3\" >tournplayed</th> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row3_col0\" class=\"data row3 col0\" >-0.244073</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row3_col1\" class=\"data row3 col1\" >-0.140033</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row3_col2\" class=\"data row3 col2\" >-0.004905</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row3_col3\" class=\"data row3 col3\" >1</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row3_col4\" class=\"data row3 col4\" >0.13293</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row3_col5\" class=\"data row3 col5\" >-0.139194</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row3_col6\" class=\"data row3 col6\" >-0.0712482</td> \n",
        "    </tr>    <tr> \n",
-       "        <th id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6level0_row4\" class=\"row_heading level0 row4\" >born</th> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row4_col0\" class=\"data row4 col0\" >0.17403</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row4_col1\" class=\"data row4 col1\" >-0.994296</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row4_col2\" class=\"data row4 col2\" >-0.129971</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row4_col3\" class=\"data row4 col3\" >0.13293</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row4_col4\" class=\"data row4 col4\" >1</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row4_col5\" class=\"data row4 col5\" >-0.163677</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row4_col6\" class=\"data row4 col6\" >0.0333731</td> \n",
+       "        <th id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2level0_row4\" class=\"row_heading level0 row4\" >born</th> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row4_col0\" class=\"data row4 col0\" >0.17403</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row4_col1\" class=\"data row4 col1\" >-0.994296</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row4_col2\" class=\"data row4 col2\" >-0.129971</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row4_col3\" class=\"data row4 col3\" >0.13293</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row4_col4\" class=\"data row4 col4\" >1</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row4_col5\" class=\"data row4 col5\" >-0.163677</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row4_col6\" class=\"data row4 col6\" >0.0333731</td> \n",
        "    </tr>    <tr> \n",
-       "        <th id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6level0_row5\" class=\"row_heading level0 row5\" >weight</th> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row5_col0\" class=\"data row5 col0\" >-0.0826093</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row5_col1\" class=\"data row5 col1\" >0.157223</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row5_col2\" class=\"data row5 col2\" >0.159385</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row5_col3\" class=\"data row5 col3\" >-0.139194</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row5_col4\" class=\"data row5 col4\" >-0.163677</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row5_col5\" class=\"data row5 col5\" >1</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row5_col6\" class=\"data row5 col6\" >0.757689</td> \n",
+       "        <th id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2level0_row5\" class=\"row_heading level0 row5\" >weight</th> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row5_col0\" class=\"data row5 col0\" >-0.0826093</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row5_col1\" class=\"data row5 col1\" >0.157223</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row5_col2\" class=\"data row5 col2\" >0.159385</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row5_col3\" class=\"data row5 col3\" >-0.139194</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row5_col4\" class=\"data row5 col4\" >-0.163677</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row5_col5\" class=\"data row5 col5\" >1</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row5_col6\" class=\"data row5 col6\" >0.757689</td> \n",
        "    </tr>    <tr> \n",
-       "        <th id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6level0_row6\" class=\"row_heading level0 row6\" >height</th> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row6_col0\" class=\"data row6 col0\" >0.0196139</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row6_col1\" class=\"data row6 col1\" >-0.0282972</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row6_col2\" class=\"data row6 col2\" >-0.0153843</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row6_col3\" class=\"data row6 col3\" >-0.0712482</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row6_col4\" class=\"data row6 col4\" >0.0333731</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row6_col5\" class=\"data row6 col5\" >0.757689</td> \n",
-       "        <td id=\"T_2eeaf31a_4f3b_11e8_82d6_780cb8866ae6row6_col6\" class=\"data row6 col6\" >1</td> \n",
+       "        <th id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2level0_row6\" class=\"row_heading level0 row6\" >height</th> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row6_col0\" class=\"data row6 col0\" >0.0196139</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row6_col1\" class=\"data row6 col1\" >-0.0282972</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row6_col2\" class=\"data row6 col2\" >-0.0153843</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row6_col3\" class=\"data row6 col3\" >-0.0712482</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row6_col4\" class=\"data row6 col4\" >0.0333731</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row6_col5\" class=\"data row6 col5\" >0.757689</td> \n",
+       "        <td id=\"T_ed54bffa_4f56_11e8_b9a6_787b8ab7acb2row6_col6\" class=\"data row6 col6\" >1</td> \n",
        "    </tr></tbody> \n",
        "</table> "
       ],
       "text/plain": [
-       "<pandas.io.formats.style.Styler at 0x7f5347ae08d0>"
+       "<pandas.io.formats.style.Styler at 0x1a19349ef0>"
       ]
      },
      "execution_count": 2,
@@ -252,9 +252,9 @@
     {
      "data": {
       "text/plain": [
-       "[<matplotlib.lines.Line2D at 0x7f533f51c438>,\n",
-       " <matplotlib.lines.Line2D at 0x7f533f51c2b0>,\n",
-       " <matplotlib.lines.Line2D at 0x7f533f51c6d8>]"
+       "[<matplotlib.lines.Line2D at 0x1a19493828>,\n",
+       " <matplotlib.lines.Line2D at 0x1a194939e8>,\n",
+       " <matplotlib.lines.Line2D at 0x1a19493e48>]"
       ]
      },
      "execution_count": 3,
@@ -265,7 +265,7 @@
      "data": {
       "image/png": "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\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f5347ae0278>"
+       "<matplotlib.figure.Figure at 0x1a19349978>"
       ]
      },
      "metadata": {},
@@ -293,7 +293,7 @@
     {
      "data": {
       "text/plain": [
-       "<matplotlib.collections.PolyCollection at 0x7f533f09f6a0>"
+       "<matplotlib.collections.PolyCollection at 0x1a19443e48>"
       ]
      },
      "execution_count": 4,
@@ -304,7 +304,7 @@
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAEICAYAAABWJCMKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzsvXfcJFWV//8+tzo8YSLMAAMDEkQUWEAYgpgQEEkSZAYxAZIRMcKubvL3XXW/u1/AvJKEFRQDDCMSJQqKIjiYAZE4ZBhgYMITurvq/v6o0NX9VHdXuI3PzHPfvIbp6a4+dW/Vuefcqr7nU6K1xmKxWCxTD/X3boDFYrFY/j7YBGCxWCxTFJsALBaLZYpiE4DFYrFMUWwCsFgslimKTQAWi8UyRbEJwGIpiIgcKyJ3dvn8BhE55rVsk8WSBpsALFMGEXlcRPZte69r8DaB1voArfUl/dyHxZIHmwAsFotlimITgMUSICIbi8iVIrJcRB4TkU/EPrteRM6J/fvHInJx69flmyLyqoj8VUT2iX1wu4icELw+VkTuFJGzRWRFsJ8DXpMOWixtlP7eDbBYJgMiooBrgJ8CHwDmA7eIyINa6xuB44A/ich1wDxgV2DHmIndgcXAHOB9wBIR2UJr/XLC7nYHLgm2PQm4SEQ20VaXxfIaY68ALFONq0TklfAP8O3g/V2BuVrr/9Ba17TWjwIXAkcBaK2fA07BD9xfB47WWq+K2X0B+JrWuq61/jHwIHBQhzYs01pfqLV2A3vzgA0N99Ni6YlNAJapxmFa61nhH+BjwfuvAzZuSw7/TGtgvhZwgAe11u0/HD/dNoNfBmzcoQ3PhS+01iPBy2k5+2Ox5MYmAIvF50ngsXhy0FpP11ofGNvmy8ADwDwR+UDb9zcREYn9ezPgmT632WIphE0AFovPPcBKEfknERkUEUdEtheRXQFE5B3AR4Gjgz/fFJFNYt/fAPiEiJRFZBHwJuD617gPFksmbAKwWIDgfvx7gZ2Ax4AXge8AM0VkBnAp8HGt9dPB7Z+LgP+NzfrvBrYOvvdlYKHW+qXXuBsWSybELjywWCyWqYm9ArBYLJYpik0AFovFMkWxCcBisVimKDYBWCwWyxRlUktBzJkzR2+++eZ/72ZYLBbLWsW99977otZ6bq/tJnUC2HzzzVm6dGnm7z3+3JNcd/et7Lvz29lm060Kt2PVyGqu/OX1zJ87j713eitKFbtw8jyPm+69g+WvvMT73nYgw4NDhdt43+MPcvsf7+LQPfdj/txOBajpefHVl7nyl9ez01bbstsb30xrjVN2avUaV991EyLCe/d4N5VypZA9rTX3/PX3/OGR+1n4joNYf8bsQvYAnnzhGa6+6yb22vEtbLf5NoXtrRkd4co7r2eDWeuz3y7vNOI3t/7hTp5+6XkWvu1Apg0OF27jg08+wi2/+yUH77Evr9twfmF7L69cwZV3Xs92r9uGt2y7S2G/qTfqXHv3rdTdOofusR/VSrVwG5c++EeW/u1PvO9tB7DB7DmF7T3z4nNc9esbedv2u7LDltsWtjc6PsqSO29g1rSZ7L9gLxzHyWxDRJal2m4yLwNdsGCBTpsAXNfl+ntu4+wrzuOev/4BUYLWmn/Y4o2cuehUDt1zv8xB54+P3M/XlnyHH99+NUopRITpg8N86n0ncNz+RzFn5nqZ7D2/YjnfueGHfH3JRYzWxgB/UH943/fxicOOyxx0xmvjXHnn9Zx1+bk8+NSjgT3N27bflTMWnZw56Git+dV9v+Uriy/g+ntuw1EOIsJG683ljIWn8KF9Dmf6UDbFgseefYJvX3MpF1z3fUJfU0px0oEf5mOHHM3mG22ayd6qkdVcdutPOPuK83huxXK01riex0G7781njjiJPbdbkCnouK7LTffewdlXnM+v7luKUgIattl0S8488lSOeNuBmYPOXx77K9/4ycVcdttPouM/VB3kE4cfx4kHfDBz0Fn+yktcfNOP+dpVF7F6dASNxtMeR73jED51+PHssMWbMtmr1Wtc9esbOevyc/nL4w8iImhPs/ub3swZi07hgF3flSnoaK25+4Hf8ZUrL+Sau26OxsrcWetzxsKT+ci+RzBjeHqmNj65/BnOvfZ7nHvd93A9N3hXOGH/ozjt4KPZauPNM9lbMzrCD39+FWddcR5PLX8WANdzec+Cd/KZI07iHTvskclvPM/j1t/fydlXnM8df7rL7zOw1bzNOfPIU1j4joMYrA5mauNfn3iYb1x1MZfevBgRQYBqucLphx3HiQd+kHnrp5eLEpF7tdYLem63tieAZ158jvOv+z7/c/Ul1Oo1Vo2umbDN9MHh1EFndHyUy++4lrMuP5fHnnuS8Xot5oA+g9UBPM/jwN325jMLT+Kt2+3a0Xm01tzxp7s454oLuOV3vwQRxoLgH1JyHMqlMtvMTxd0Hn76Mb710+9y0c9+hEBin6cNDjNYGeCT7zueEw74ABvO7nw1+OqalVx602LOWXw+L65cwcj4KO1+MTww5AedvQ7lk4cfz45bdZ7pNNwG19/tJ+Pf/u2PeJ6m1qi1bFMpV1AIu75xJ85cdAoH7rZ316DjJ+ML+fEd16BEsWZspOVzEWGoOsicmev5QefdRzBzeEZHe8+vWM6F1/+Ab/zkYsZqYx39BuD4Az7Axw89tmvQGauNceUvr+esy8/jb089St2t03Db/KYygKc93r3LOzhj4cldg47Wmjvvu4evLLmQG5bejlKK0fFWv3GUQ6VcZquNNuPMhaew6O0HM1gd6NjG9mTcqc+VcoXTD/0oJx30oa5BZ9XIar5/6xLOvuI8nl/xIqPjY3jaa9lmeGAI13NZ+PaD+PQRJ7Lz1v/Q0Z7rutx47x2cfeX53PXA79BoxuttflMqo5TizVttz5lHnMx799iXktP5RsZfHvsrX//JRVx2209wRLE6wW+Gq4PMmjaTzy48iWP2W8Ts6bM62lv+yktc/LMf8bUl32H12AirO4w9tObY/Y7k9MOP4w3zt+xor1av8ZNf/YyzLj+X+5c9RN1t0HAbLdsMVKpordl7p7fy2YUn866d9uw5sVunE0A8+/7iT78BgbHaeE97lVIFpYRdt5kYdNqzb9KJbScKOjNm89mFJ3P0fgujoPPyyhV896Yr+MqVF/DqmlWsGRuZEFSTmD44jAaO3/8oPn7osbx+ky0A/1L4mrtu5qwrzuMPj9yH63nUG/We9gYqA6A1++78dj676CTeucNboqCz9ME/8tUlF7LkzhtQSjEyNtrTnqMcquUKW87bjDOPPJVFsZnO0y8+y/nXXsb/XP1d6o16YoDp1OdKucJphxzDyQd9mI3nbASkS8ZJhEHniLcdyKePOJFd3rAD4AfV2//4az8Z//5OlEh0JdaNcqmMoxRvfv32nLnoFA7eY1/KpTIADz31KN/66Xe5+MYfA6T2m+GBIWYNz+AzC0/i2FjQeWX1q1x6y5Wcs+QCXl71CmsSknESYdA5Zt+FnH7oR9lmvn/rM0zGZ11xHks7JOMkwqCz1457csaik1tuff7h4fv42pLvcPkd16DUxGScROg3m22wMWceeSrvf+ch0a3P515+gQtv+AHfuPq7jNfHM/mNoxw+dvBHOPWgo5k/dx7gJ+PFv7iOsy4/j4eeeYxavZ7Kb4aqg3ja45C37MdnjjgxuvWptebOv9zDOYsv4MaltyOk9BunhOM4iXchHnv2Cf7np9/lwht+gNawanR1qj5PGxxm+uAwnz7iRD76nvd3vAuxTiYArTVfXXIhZ/343I7ZNy3TB4eplMoctPu+/OXxB3ngieTsm5ahgUE8z+M9C96J63rc8vtfokQxMt47qCYRBp0dttyWN2yyBdf85hY8z0vtKO34QWeQmcMz2H/BXvzyz/fw1IvPMlYfx/O83gYSmDY4jNaag3bfh+WvvMRd99+LBsbrvZNxEgOVKmjYc7sFzJm5HtfdfWvqZJyEUorBSpX5c+bx1u1342e//TkrR1anTsZJTB8cxnEcDtp9Hx56+jH+9OgDqZNxEmHQ2ffNb0eVHG76/S+K+Y1TouSU2Hazrdl2/uu57u5bMyXjJKYNDjNtcJgDd30Xdz3wO5Y9/1TqZJxob2AIT2sO3G1vXh1bxS/+cg8ikmoSl0S1XEEQdt9mJ+bN3oBrf3MLkC4ZJ6FEMVgdYKPZc9lrx7dw072/YEWGZJxEeBfioN335fHnnuB3D/8Fz/Oo5fSbweoAWmsO2PVdnH3yv7HlvNe1fL5OJoBnX3qe1314j9yDbW0lnIVMVpSoCZf+k9GmSQRBY/CciCCOmLOoNeJito2GUSUHz2T7PI3SMrn9xvBYFoTTD/soXz/tP9r3kyoBrFV1AFpryl3u9+Wh6CqF1wL/56DJSz+O4WQ/L+bb14dQPbkPofUbA4QLAvKyViUAi8VisZjDJgCLxWKZotgEYLFYLFMUmwAsFotlimITgMVisUxRpnwCmMzLK0Mmexv70b7JvHwR+tBnDZi0qfXa4TdTsc+TiLUqAcwYns5gdcCICBb4a81NErdn0rZGG7OngmVoppajxdfrm7AojoKSwhN/nbgJwmW05o6hMpqgona5GlU0EWgNngavzXZBQjumliQ7ovAaLrga8TDTZw2e9gqL7oWEY0QZHCsmE0ClVGagXGXb170hf5uMteY1YNrgME//cCkXfOq/efNW2zFYGaCUUSkvdGCR1oIRJ3CaPAMmdJDWoKBbPstkL2iLE3NkT3u5g3Y8AHqBA2qtC/U5FKuKO7TKay8M9iWFhH32NdmQkkJKTubsErbBUc1g7WmvcCDr6Dc5gk7TN5rHUHvaTwQeQVBLETCCmbTyAFfjxPoW95usfW71Gy9oqW6e5xx9DtsSLwATTb4+B4VfuBoV65v2PCT4LyuhfyjVDNaebk7A8oxBwazfTB8cZvrQNE4/9KPcd9FtnPreozPbiNo22S5J4vQSg7t/2d/45lUXc+nNV6JEJgg9xQmdWCnVVfogCGvQo2JPiUSO0asQI9ymV/Vo6LC9ZpeOUrhe775Efe7RxrA6MexTL3vh/rsR2uzWZ1EKrei5X/BnjK72UIg/c+zYxvC8pOtLr2MTHuNefZbY/7ue4+hYp/CbYN/iBEmsPfgEQVDCbNnDc6JjmLLPPcdKeI57jpX0fhONFaXQ0qHPGr/fKapqo7HSq8+Bv6Ttiym/gSCp6O7nruSUKJd8iY8zjzyVw/Z8T1fByHVSCqITI2OjXH7HNZx1+bk83qZTItAzmHciyXmaDtA7aLWT5DzhoEzjKEl0Shp5pRSS+pd2oCfaC4NJ2B4B5fgSAHn63JJYXBd0tmSc2MaE75nwm/DcxveRp40tkxf/xwKUOHiui6OcXJo87UEnTNR5/BoSzjOG/MZx8DzX95uozwX8JtaWLMk4sY3RxK55DecfV8jzK1ZSvAn1to5+90I+cdhxvHGz16eyNaUSQJzfP/wXvrbkO3zv5itRSnIF1XaKOHJP24Y0ZaIg2GPWltlezoCQiAg45krhBUE3XGNtzDJrS9W+FFdBmW0S3Boz5IvhsTPV5774jWGKBv52kpJfofYFZ3mTuRvzpWPP5Mh3Hpz52QLrpBZQGt78+u255B+/RqlUMuLQEN4vd4wHf0c5xgJDONhMBP/QnvFBbFimRUczakPHMDh2Zv3G7A/GGoz6YnjsTPW5eSVmfqJkinAyZ0o0LvQbU/Y0moHKAP/ygdM5Zr9FmYN/Fta5BGCxWCyWdNgEYLFYLFMUmwAsFotlimITgMVisUxRbAKwWCyWKUqqBCAinxaR+0TkLyLyQxEZEJEtRORuEXlIRH4sIpVg22rw74eDzzeP2fl88P6DIvKe/nTJYrFYLGnomQBEZBPgE8ACrfX2gAMcBfw38FWt9dbACuD44CvHAyu01q8Hvhpsh4hsG3xvO2B/4NsiYkbsJcaqkdWcd833cF3X6KpD13ONPs5NwC9wmcR4wRp2U/ThAYB9ef6ryT67gSyBKQRyP4z9tSKsejdFWOpokslc/wQw3qhx2W1L+NOj9/d1P2lvAZWAQREpAUPAs8DewOLg80uAw4LXhwb/Jvh8H/Ej56HAj7TW41rrx4CHgd2Kd8Hnj4/cz0fP+jQbHrkTZ1zwxaicHPLpbUBT9yMK/DrU98l/5yz8rqZZ0JNXtwQmahg5MQ2cPDQFsELtk9b9FLEZ6txAsSAbtiW0EOm35EzQkfZS2GclLfvJSlxvCpo1oab8Jm4776SkXc8nroGThwnCadK6nyI2o7FC/v7G29L0l/xaWEnfLzz28H1HRPA8j7vuv5e3fOIQdjhxX75382JGx0dz2e1Gz5ZqrZ8GzgaewA/8rwL3Aq9orRvBZk8BmwSvNwGeDL7bCLZfP/5+wnciROQkEVkqIkuXL1/etW0jY6N898bL2f6Evdnzk4fyvVuWMDo+xppAEygq0PCapelpaAYYCY+B/3fweVxULE3QiTtK+4xVa41GZw46oRib21aEEkoPhH1O69yqrc+hvXihUJagEx9sLbMtraHhoV03EvBKEyTigmZuvRHYaC/ASW8vbBvECsDajp0/e8/uN5E4YNsss9VverdRxZJxkt9A83xl9Zv2sRGJvXnZklX7WImLDYb28/iNoyYqZ2paRQydFG1sEX9s62tc4A4yjJUwGUdica3+E46VtNO66BgGRYM6NuZGxsf48+N/5WPf/BfmLtyRj3/zX3jwyUdStTNVX3ptICKz8WfvWwAbA8PAAQmbNmVFkj/r9H7rG1pfoLVeoLVeMHfu3MQ2PbDsIU79+ufZYNGOnP4//8Z9y/7GyPhoz0vjsPwbJgaJePZtD6pJpAk6zVmbbvtOgr0UQSeuIKh7iEclJ6u2PsdmbWmqGdMEHRWf7Xezp/FF3RoeutE5QUeDQxNs63a8G+DFAninoNMtGXdopu83CQqtIe3J2E3hN92CTvt7afywW9BpT8bd/SamnNpB3TV+bNOMlbj2DnTym7Zk3KUyuf04dx0rHZJxUvtCDawkm0nJuNuRbE7s0vtNtyr+1aNrWDM2wgXXXcZOp+zHrqcdyOV3XEOtXuv4nTSkSXn7Ao9prZdrrevAEmBPYFZwSwhgPvBM8PopYFOA4POZwMvx9xO+k4rx2jh7fuJQdvnYAXznhh+yZmyE1aNrspiYMNMJac++We35WiBNp1GxUvOsNtuDTkgaR+nWxnaHbZ+1ZbUXvyqAQBMlulpKb1N7XhTc48EmPtvvpgCaaLMtWUXvp0jGSXQKTGmScbf2xYMOtPpN1vPSHnSi93sl4w6EfW73X0kRVLvZm+A3BTV0tNYtVwOCpErGiW3s4DfNz834jZPTb+pug7HaOEv/9ieOP+cM5i7cgWt/c0tGK03SJIAngD1EZCi4l78PcD/wc2BhsM0xwE+D11cH/yb4/Dbte8rVwFHBKqEtgK2Be7I09qWVK/j9I39htDZGw230/kIX2g+9CQ2d+IA1oYUS3boqbMkn6ZaECZthACt8DDV4QcBXHl1n+2npdOukKOE5mcx+E1JUi6gfYyXU4wEzfthUXTUjvNfeJlPaRmFiMaG9tHp0DStHVnPj0ttz2yj12kBrfbeILAZ+BzSA3wMXANcBPxKRLwXvXRR85SLgeyLyMP7M/6jAzn0icjl+8mgAp2mtMy9nMP0Ur34ofJpGRKH7sNrFFCIKsp/K7jb7sGbIJJP9nMDk920lgmtaNK4PvmgSUTLh7sPfk54JAEBr/QXgC21vP0rCKh6t9RiwqIOdLwNfzthGi8VisfQBWwlssVgsUxSbACwWi2WKYhOAxWKxTFFsArBYLJYpylqXAEzroEzmVRIh/dC7MYmpx1DGMb7CxvCiosl+TmDy+7bXh9Uwk/286D6MlSKsVQlgg9lzOHiPfamWKwxWBgrZamrdtGrfFKWYsk8r7VWJRdsYyT0Yshf1MvzLgFheXP8lZjqnMVAlB0qq9e8CTPQbA30Gg15j3redNntpJBhS0Qe/aZouZnPCMSygaeTbizrb+u+ciAjTBobZdO7GLHrHwfnbVagVrzElp8Tif7+AJy67h3/90CfZYNYcpg8OZ7IRlYbTplkSk0zI4jztEgE6+q9VuiEt8UHbrqvSlA/I5jyhRIBuq+z0tJdr8EWJRLVpJcV0WrIEnUhMK6b/0pQPyCFSphSUFOI4eFHlb/MYSkkhJSdTdunsN3pCsEjXxFifIeY1Of2GuN+0nucifgMT9X3cmN9kGStO27mMn+t2cbo0dPebfMkvlGGZcAwDVdesiaV9/0X9ZrA6QLVc4dC37Mf1/3kpyy67m3fssEemNsWRyXyZuGDBAr106dKOn3uex833/oJzFp/PL/98DxoYr49P2M4vr9epS86VUr60Q/B3EgKQstAmHCpI58vysG2p2xhs16nYJ/zcUSpV1WG4Xac+C36FZdriouYxl45VlOFnaW2G24VtaWsgqlTy+ywqlQRAdMyRRKmJcH9Zz0m37UUkUpVNM/IiQcIOW0e+mrmNyeclje8n2+u+/8znuMv2Wf0mGitJfkP2PodjpZOfhfsx6TfTBoeZNjDEp953AsftfxRzZ63f1aaI3Ku1XtBr32t1Aojz9IvPcv61l/E/V3+XeqPOqtE1PZ09DXEnyxpUkwi/G3fGIloo7c6TZgD1wg9StDhx2sHRrY1hItToYvaC74ooNB7iqMhuHloSS6BHlCUZJ7Yx9l2TfqOQ6KqmiF+3T4rM+Y1/FsJzVKTP0XmOJcFC5yRmL5zM+Icyn+e0T4qyJuPENsaSVaVcBmCvHffkswtPYp83vy31FdKUSwAhDbfB9XffxuH/54S+/Dhpkk4zEksGSobvYjYmt8+Af0U51bxmqvXZUYoDd9ub8z75X2w8Z6PM30+bANaq3wDSUHJKHLLnfjjK7MPGTNuD4j8s9RsTP3D2m8nexiIP0umE6oMvmsS0XpeP2fNs8ul+/aBSqnDgbvvkCv5ZmNwRyGKxWCx9wyYAi8VimaLYBGCxWCxTFJsALBaLZYpiE4DFYrFMUWwC+LsylRa2WSyWLLwWukbrVAJwXZfr776Vvc88knqjbmw5mlIqEqEzsXgs/lxQkxpEYG7ZYfiAd1PL5eJ2jC3BE4kqOE0QngtTS0uVNIugTOj8hBZczzV2nk37jaOahUwmaGpDaWPnRWiVLSlKU/7EnN+M1sb454v/iy//4Bs8v2K5EbtJrBOFYM+9/AIXXv8DvnHVxYzXxlk1uib6LGvZeJykir6sZeNxomrOmN2sZeOt7UuWuOgqmdCDpOMUlryb6nORytjEClOlQOUrrIuqORG023wIfS+pja5tTJAeKFIZm+RzYWFUHr9pSly0VhIXqQZO+k6RKvJufpPHXqdxm0XSJU4neZciVeRJPjdQGQCt2Xfnt/PZRSfxzh3ekmoCtc5XAmut+fkffs05i8/n1t//ChEYq03UAQpJG3SynMC0zhMvke9GWumKtNuFgbtnn3PqyXQiSyBJu22qtgmoQACuV5CI9FwQ3AQdoPb99rKXRW+qf37Tfd9pz7Npv2lOcrr7az/9ptfkoJe+T/t2vfwhrY5UGs2s0N7wwBAzh6fzmSNO4tj9FrHejNndtl83E8BLK1fwvz/7MV9dciErR1azOjbbT0uS85iY8cWdIq2jJNubGEw6zdrSEtfhad9PnplzUtApooWSlKCLzKZEKbSaqJcjwdxZexoy6tEnXU0W0ZtKuiqIrrYKzewn+k1ejZ8kvZxCYyXBR9IG30R7CWPPhJZTe9/SJuPkNib4TYGxMlQdxNMe793j3Xz6iBPZ4007T5TDXhcTwMo1q9j4qF3wtMfo+Fhh+0UcOYl+aPsUuUWUhAnBqpb2GTp2ibZNHM/wqkB7KHHwGo3C7TLtN3FM+VC7QFlRjPtNH3WwTJ0XEyJ+SfZMtm+gUuXfP/Jp/un9H2v5bJ3UAlo9ugattZHgD81f2U0N4n44dHh/2tSKgDAYmLIXzjhNo0SZOZ4aX+rZ1UaCP9C8pdCHxGfKhyLdeUOCiMb9xuCPunGUKGPnJeyrieAf2jM5afC0x8j4KE8tfya3jbUqAfSDyS4KBZO/jf1o31Tss2kmexut3/z9mfIJwGKxWKYqNgFYLBbLFMUmAIvFYpmi2ARgsVgsUxSbACwWi2WKslYlAMdxqDfqRm1O5jqIkNdCFKoILhqU2dUNofbSZGWynxOY/L5tanllq81J7jeG++woh3KpnPv7a1UC2HD2XL73uW+wwxZvYrA6UPg5veGSrFDMySmwnj1c0ywi0esi6+NDgbOwbSLF5LXCb0eCZ5GAVc42CkjJgZJCFKjgtSo5uRXz4m0L21tk2VynvhY7jhP9pogYXdQmkchukfXxob2435ig3W4R35aWsVLcXrvfxM9RPnsS2YvbLzICTfvNtMFhpg0Mcdohx/CPR56av12TeZbQTQvoz489wNeXXMQPfn4VShRrxkZ62uslyBUJg6Us1kij95FWiyRus730Pk5WiYleFZxZJSZUydfa6Va2H/XVA1LOeLod8yRhsK5t7CEAmFXQr9f2nYTBOpGmujtr1WgvAcDMftNDAymthk27vW5VtXmkSbodn7waV536lFViIo0AYJhU0vS35DiUnTLbbLolZx55Ku972wG+WFyS3XVRCiKJ1aNr+NHPf8r/u/xcnnnpOUZr4xMcNo/iX7dBn0drpNuAylNy3i3opBXf6tTGCYNKQBwnU3KENl2bRqPl8Qd51DG76dqkFd9qp1vQac740tfndktWeeQUuvlaXmXaXro2WatVuwXOPGq83Y6Tcb/JKbXSbfvw+LZrb3Wjmx7StIEhPK358D6H84nDj2e7zbfpaW/KJIA4Sx/8I19dciFL7ryBhuvScBuFdTzancaEFkoz6PgxsWh5eNjHyJkL6r9EfQxklk3oyfjSCR64+YXJWtoYtMmUVlK7n0xOvwmvJvvjN3nE2Frb15oATPS5eWUDoQh2Ec+ZMFYKtrHph4E8d8GxEr8LUS1V2Hyj+Zx55Km8/52HMDw4lN7OVEwAIa+sfpU5R+xg9AehvCqcr7VNk0hJmVU3cj3jD0Hrp6iYCaak3/RFFHFqPT+vXCrz2YUn8X+P/3zoNgw+AAAgAElEQVSu76+TYnBpmTVtprGnRIVIHwTP+mHTJKae6tRPJpm0ygT6o3czyf2mLyfFrM3JpsnTTkk5vG6D+X3fTypPEpFZIrJYRP4qIg+IyFtEZD0RuVlEHgr+nh1sKyLyDRF5WET+JCI7x+wcE2z/kIgc069OWSwWi6U3aacSXwd+prV+I7Aj8ADwOeBWrfXWwK3BvwEOALYO/pwEnAsgIusBXwB2B3YDvhAmDYvFYrG89vRMACIyA3gHcBGA1rqmtX4FOBS4JNjsEuCw4PWhwKXa5zfALBGZB7wHuFlr/bLWegVwM7C/0d5YLBaLJTVprgC2BJYD/ysivxeR74jIMLCh1vpZgODvDYLtNwGejH3/qeC9Tu+3ICInichSEVm6fPnyzB2yWCwWSzrSJIASsDNwrtb6zcAamrd7kkj6dUV3eb/1Da0v0Fov0FovmDt3bormWSwWiyUPaRLAU8BTWuu7g38vxk8Izwe3dgj+fiG2/aax788HnunyvnHWjI74a84NYtpev2yaZDIvrwyZxKshgf60bzIv3YZ++Y1Zm5P9GLqey7MvP9/3/fRMAFrr54AnRSQsP9sHuB+4GghX8hwD/DR4fTVwdLAaaA/g1eAW0Y3AfiIyO/jxd7/gPWP8+bEHOOGcM5i7aIfoPafgctBwKWS47trE829DG2GFZNEFae19LNrGcIlcmJ+KHsPIpiqu7xMStikMNkVtti8bLrqMWKDlWc5m/cYL9lGsz5FeUGi/aJ9DvwnGigm/CXV5wsK3wvZUq65PYb9pO69GxgpCvdHgv3/0bXY6ZT8uu3UJYzUzz0GfsK+UOiM7Ad8BKsCjwEfxk8flwGbAE8AirfXL4h/Rb+H/wDsCfFRrvTSwcxzwz4HZL2ut/7fbftMUgo3Vxlj8i+s46/LzeOiZx6jV64kFYFmrJruVZucqbw+1U5TCQ7fcEFMoPNc1VtKfu7w9LMF3HDy8lkX2UYVihiKfyB6C57otk7hIJiJDn7uV9OeR04h/L6nQKI+ESDfNnV5aM8nt66y5Ex7DrH3u5bdZ/aab/EW3c9Zr/4lSED00ipKIvLiL32Qde900vkz7zbTBYQCOe8/7+fihx7L1/C172lvnK4H/9tSjfPOqi/nujZeDCKtH1/S0l+ZkZxFnSmOvGVTDwN9hxqE1aBAd/rODwFVWQaoeQadlgPrT1q7VVWmCTpQgNWg3m+ZN4j67JOMkegWd3Mewm9/00NdptRcG9S5icBl1g3oFnfzCaN1FBNMG9mxicL1v+KTRu8ougNdDRDDFeWvZPo0YXIZEUS6VcZRixy235cwjT+GQt+zXUQp6nU0AV/3qZ/znD7/FXx77Kw3PzfV8gPYZclZHSbQZGxDgB/CW2X7aS80gETiicGNXBb3UHnvR7rzdZvtpiSfLlkTTcHPdsm0POlmScaK9NrG0PDPwOO1BJ+8MvKWNsaADZJ4tT7DXFnTyXK3GaRfZy5qMO9lE03Isi2jytJ/fLMk40V77GMkh4pdEdCwN2Js+OIyjHE46+EOcuehU5sxcr31fqRJAKdfe/068sOJFjvzSqYUfChO6Rnjww8FbRMQpHFw6XPDkCJ4/LctmSHyFT9fPAnieTrCfnTDghX32tBe0sdMCrd5EbdHBsfM8ivys3dI28vc1std2Xpv289lt+o3/t9vmP7naqFvPb/vrzPa8Np9OsJ8FrZPHShHRuHhb2v0yD+3nt+ikNvKTDv6Tl+hYGrC3Krjj8ZXFF+K6Lmef/O+57ExuUZE2Gm6DsmM2ZxnXBBE/+BsRqRGJ1DiNUlLgKGNCOg6SWvc/tc2CD/vpNyZ+1O03k13vxsQPphNtTnK/MdznhttgvF7L/f3J78UWi8Vi6Qs2AVgsFssUxSYAi8VimaLYBGCxWCxTFJsALBaLZYpiE4DFYrFMUdaqBOA/FFkzUKkasadEBUUuhpbLKRXp3RRGa/A0oiUoGjG0ZFMUNDyUB3i6mFpZ0EbXdY0+PlIQ489zhqDvJuyVFJ7SSMkx/aRCY8QLuEwQ2jFlzz/HZpcOh35jyhfDJZumlm6KSPAAeTPtExGGqoPMW2+D3ht3YK1KADOHZ/DIpb/mnz/wcebMXI/pgUZGVkInjhf2FHFsVSpBSaGcpo1wjXPmtc5aIx7gahSCTtCTyUO4JtyLFaOIxt9XlkQQBH1HS9RG8I9l2Nc8bQy/4ygVnZdw4OUN3H4BtsSKjQoIxwmokuPXUIhf56HR0bkXJ0efE3zEKRB0IlG3YGIDTR/PO4GIKpRjxZNF6gvC8yyxiZIJv1Exvwlt5w3ckd8EY08HQTtvv6M+t1XOQz5fHKgMUC1XOGj3fbjuy5fw+Q+cnqtdsBZKQYS4rstN997B2Vecz6/vX4rWumtBRHtJeydSl2krQZQi1DzodBQjEbVupfiB/INCUu27mxBVYl96SFxMkIWAiUVi2r8a0TFZik6kPdZ+X9JJFaTWnEmhvxLfrtexEUehpbdAWouMRaPRVQojlEKAztXOEv5fele2ppWkiEslpBGD62Uvq4RKer/pLdWRVtajqS+Ubqyk7nMPfwj1jDL5TY+xMn1wmIHKAJ983/GccMAH2HB25+elrLNaQEk8+cIznHftpXz7mu/huo2oTBqyC3/FSdLfUSUHL5jtZr2EnZBcYkE1T/s6iZTlUWCEtqAjED7HJ9QlytXnhKCTJUG0E2nRtLWleVmdTUAi8VgFs/28GjVR0PGIruDSJpwkkoJOqolFx/Ylj4kiukFJ+jtpg2piGxMmYnlVbjvZi7cbsslXdRpjReJNkjJttezf7n7b9rtyxqKT2W+Xd6a6splSCSCk3qhz7W9u4awrzuOu++8tLN4U4ojCFe3P+gsIdYWIiK+S6elcASHRJsGMw5C9cNCa6C+0zoRMnZeQImJscZRSeKIRpcz0WSm8un81YMxvYs+lMHEMiwTVJCK/wcwjXPrR55BJO1aCdk0bGOJTR5zAyQd9mPlzN85kY0omgDiVA7YoLBoXxymXCglgTcDTOCijP3aaxpRD95O8yp4dKZn9WczRguuaPceOcia135gO1EDPWziZ7U1y3x6sDHDWyf/KaYccm+v7aRPAWvUjsOW1ZZIucLFYpgSO9F/YziYAi8VimaLYBGCxWCxTFJsALBaLZYpiE4DFYrFMUWwCsFgslinKOpcAPM/jxt/ebnyZnIn1wq+FTZMYXV7ZJ4y30bA903o3AN4kXgIKxZ+dm4TJJaBQ/LnB/abWqHHTvXew/JWX+rqfdSYBvLDiRf7zB99k46N2ZtEXT24RUCsipBbpBsV0b4qIioW6JyIqWmdZSIeorY8i4uuW5LbYqk8S11rJS9RnpRL1b7IS2hBHRc9fLqJPI+EfEb84j+axzN3G8Jw6YkRUrOk3EoXCIn4jCX4DFPIc034TnWekKehXxF5cf8c/4YWfmxwfbU3BvOJ+42nNDff8nE0/uCuHf+F4fvnnu/uStNbqQjCtNb/40284Z/EF3HzvLxARRmtjE7ZLq23Svn3HghalQGUrTpGgNlJ7vpjaxH1KID2QrrCpl2ZR1jL8XvayVoy2lMrjV1FH+kKh0inZJCvCNkpJoR1pKq8G9pQneA03dYVnLy2bUPYjbZ+7nsNA70lCDaCU4y6tbk9av+mlZZNWVyitvfZ29iJKQF00kLJKnUR+oxRadKsf6qBtrkG/ifwg21jpdA5FhOHqIOvNmMVnF57M0e9eyKxpM3vZXHcrgVeseoVLbrqCcxZfwCtrVrJmbCTzgEpynrRiWU1jBAJqyZIO0WAShdtDIKy9DUn2Cmn8dPhOWmG5xDYmOHg0OLoJy4UE7XHorDUU7ctx8Jy2RNLBprgaXe88QJN0a7oiII7TUXsntYBg0L4W4b+ENmayF36nS9DJ7zetSqCt+5uoW9O7jSn8JoM8Q7eJXXQ8lMJDRzP+jrTocnUOxNl1lzpPCtImzzhD1UE87XH4W/fn00ecyII37Jh4FbNOJgDXdTn2rE+z+JfXoUQxMj6a23b7zKmopIAohQ6uCsLZS1wMLJfNtplBUe2Sds2SoiX78cAS6Mb5InK9BlsS8RlyaEtrpOygFc3ZfhZ7nka5gue6UVuL6gY1rwryC6fF2xhXWPXf6q1c2bONbT5d9Dw3E5IZexP8xsAd/vYkF01AcvphmKBNjZX2eFNUikIpxWClyrz1NuR7//R19th2l5bP10kpiOdXLOeKX1zLWG28UPCH5g9V7VrxedGeBw0P7brohuu/LvgDoG5rW9EfjcPA17Rrpn0IaIdmoM5zD1T8Wzpa+bYoKxhw0CXJHvxDe47CqwhUmuJuRX+U9YJzG56TQhMo8fuL4wf8dg3/3G1s95uC5zn0O1P2omOm/b6amIJGNpWAI3jxWz1ZCPzQU4AyN1bC72sTfoN/TtaMjfLwM49z2W0/yW1nrUoA4AthmaToj0ATCGewBjH1FKZ+oRyH8CEphQkGoCqXzNiDfAmkp0mDNk0duwlmJ7eakzJ9XiTmi0bs9eGcGHq6mCkmV2ssFovF8pphE4DFYrFMUWwCsFgslimKTQAWi8UyRbEJwGKxWKYoa10CMK0zMpnrIEImexuNt08Xq5+YaM/88dOeNmu3H22cqn5jym6//GYSsVYlgPVmzGLb172Boeqg8eWgJpZaxm0YsRcsGdPoguo+HewXXeYmIEFhlG/P3DH06i7iBrIZeQeip1ENDeOmn08L2vXA1SgtQfFQjjZGEhZES4dN+6EJr2n3PRNtlFjhpQl7oUaQ13ARD3ALJAKt/YLEmNuYGH+RLArFNJJCpg0OMzwwxD5vfltuG2tVAhioDHDvt2/gV1+7io/s+z4GqwMMDwxlstFJmCs8MVnXTseFoOJXJ2HJu79NNqKq0NhsQaMjJ8/qPKF4XSchrawDUJUcKCmccqmljCeU0ciLHxSax1B7GkcLNMIgmWJAaw0ND1XzYNwv2mppeyhGl7XPMRG7eCvCRJAp6GiNeDpKIPECv1a/yeiLod8Q95vOPt+LqB1t6/WjIJbxXMcF4uKz/yJ+E4r4xQv8tNb++Y0n6F7Ek3F4PuP7CY5B3mPotPVZe778StZ+O8phqDrAP2z+Rr59+pdZvviPHPbW/TPZiJNaCkJEHGAp8LTW+mAR2QL4EbAe8DvgI1rrmohUgUuBXYCXgPdrrR8PbHweOB5wgU9orW/sts9eYnCrRlZz2a0/4awrzuP5FcsZHR/reIsoS9l5Go2OnoJxCfZCbaBEe2G5fQq5h1AqoFs5eZptkvrTUYYgpnuUxmaa/mQRrItK8pUKqjxpLdTxAumNenpRrzT9CHWDSCHPMEEHKd6+NomBNH3upgvVvk2qYxj+v4vQWrRtVr/psn0WuZU0mkB5/EZEkmVKgtm+9rKL/nWT7Miiv5Qm3gwPDOFpjw+861A+efjx7LDltl1tGtcCEpHPAAuAGUECuBxYorX+kYicB/xRa32uiHwM2EFrfYqIHAUcrrV+v4hsC/wQ2A3YGLgFeIPWuqO4ea8EEKK15p6//p6vXHkhV//6JpTydYKyKhsm9DlRpa+ILki7Q+QV6ora0hZoo4DQJdl0tdeW2KTkz3jz9lli/4+usrqIjKVqY6ju6ChfvMvz9X7ytjFK0G0DMa/20gQl1FBmuoDe1EQfLKZh1amvecfKBL/JqB6bRLtoX1HtoGiMOA6edlGkn3R1tdfW96LxJv73UHWAeettyJlHnsoH9z6MaYPDaW2ZSwAiMh+4BPgy8BngvcByYCOtdUNE3gL8f1rr94jIjcHru0SkBDwHzAU+B6C1/r+BzWi7TvtNmwDihEqhnz3/iy36KiYoKtLVL1t9wxFf5G4y/5jYyBdcupFF4TKdPZN+Y1xppC8UFTtrsRX8barf/TqGJs+zEsWu2+zINz/+xY6Kn13bYlgM7mvAP9L8WWR94BWtdSP491PAJsHrTYAnAYLPXw22j95P+I4xZk+fxaeOOBHHcYwGLkc5RgO2f0/f7A/ZplEyyYN/oNJo3KxBW/5vBib9xrwelmlM+0143WjSXj90kkye52q5wrH7Hcmu2+zUV02nnqNHRA4GXtBa3xt/O2FT3eOzbt+J7+8kEVkqIkuXL1/eq3kWi8ViyUma6dNbgUNE5HH8H333xr8imBXc4gGYDzwTvH4K2BQg+Hwm8HL8/YTvRGitL9BaL9BaL5g7d27mDlksFoslHT0TgNb681rr+VrrzYGjgNu01h8Cfg4sDDY7Bvhp8Prq4N8En9+m/evBq4GjRKQarCDaGrjHWE8sFovFkolS70068k/Aj0TkS8DvgYuC9y8CviciD+PP/I8C0FrfF6wcuh9oAKd1WwFksVgslv6SKQForW8Hbg9eP4q/pLN9mzFgUYfvfxl/JZHFYrFY/s6sVZXAFovFYjHHOpcA7l/2N0752udouI3M5f6dEMD1XKPLsUKbxuxJU5LClL2wyGWyIrEqVBPEpQpM4Xrm2gf+WnPXc40tigzlHEz1OZTzMPrITMwusWx5KL0B+uE3Y7Vx/vvyb/Pj26+mVq8Zs9vO5B3dGRivjfPD265i51Pew4KPHchFP/shWuuoQjLviQ61c8LneIZVwXltRoNNJHLnuMZMHiL9l7aHTecN3FGf8atXvUbD1ywh30rsMAkrxNfKcT3/dc42NtsiaNdttpX8xzAS3RNAEYnbFfUbleA3eQKjxIK0Fg2q6ZOF/UbEf/h5QUG/yG8Cu14swOYbK2pCe4oE2tBnhGaBWqTjlddvpCnWCP6D2kNtojzE/Uajefy5JznxK2cyZ+EOfPa8/8Ojzy7LZbcbqaUg/h70qgR+5JnH+dZPv8tFN/wI0KwaXdNx20yaKW0l6EXttZeM99p3ap2dHvaaJerd5QLSylGoQPkznY5NoJeiA8G0pP06Ci0TZQ6SiKQLELxG5yuntH1p0RYiQVso1O4R5ctMpNFoSnPuMshMNLWFwja2BZagjaLDf6bzG8dxfJmQTn1GUvlXJFbXQ1soiyRLT00qmhW3acZTWo2tzH7Tw2ezaIWl8f9KqYwSxc5v+AfOXHQKB++xLyWn80+4xrWA/h4kJYB6o841d93MWVecxx8euQ/X86g36qlthmXg7SemiG5QkvMU0fjpFCTSiFBlsZdGhKqDwUAdsbUdLfo0DTd9Sa0EySVhYElwxrTnqzWmpdP5bNGDaRds64TWiBZfd6jtWGVJ7u309Jsk8bIO7UMHfXbdiX0Og6oKbabrs0Lw3M5+k0dvKkkHquk32Y9hp0CbZhKXaK9Dwsg7VjqJMkbJOMcxnD44TMkp8bFDjuaUgz/C/LkbT9zvupgA1oyOsNUxb2VkbKTrbD8tRYSqetnLKzAVJx7oTdiDWIAxoVsi4JRKfrCRILkW1OYJxefQQZ8bjcLaDNGMsdNsPy1tM+S47SLEz63jOLhpE1OHNobJKrJtsM+mdHTifTYxDlvGiqFxnVVRtxd5ReI6US1XAOFbp3+REw74YMtnaRNAkTqA15xX16xk5cgqRsfHjNgz/XSxlucBGDjRcfVME/Yg9luBiWGswa03/MHsekYCQ5hAlFJ49UaPrVPa9KfHeAKFNGXEz3Ie+EFWayPHMTq3juBKJ9WU9G2MZvkihGmqiL2wz3jmdJL8qyZ/5m9iHJoO/nGbpibJ4YIAU/bGgx+H//jI/bltrHU/Apt+MlY/hZZMMdnb2I+nlRm3afgYGj8neWfoHe1lf9jIa00/2jcV+1yEtS4BWCwWi8UMNgFYLBbLFMUmAIvFYpmi2ARgsVgsUxSbACwWi2WKstYlgIZrZmlgyGSugwgxvVwVJSafsGdU06gvNvuw8MJT+MfRFCbbqP2iOd1w/dembBrG5Jr4ps3JrTBvajl3nCIr5taqBDBv/Q359w9/mvVnzGb64HAhW6GOhyBNfZ8iB1I1tUvir/MStkTFlvMVEmYTkJIDJeVX8joOErzOS1yvxUSf41ovykSflSAlhTjNZ+gWWoYnoMoOlBVSdlADJRgoIRUndwB3gnOinFJxUTGtEQ9wNUoLosV/7eFXUWcN4kEiEVeDqyOdm6K5KtK8ETEzVmK+EvpNUSHIUDeofR95ieKNSEvsyctApUq1XGX/Xffi44d9NLedtaoSOKThNrjhnp9z9hXncc+Df8DzNLVGb8W8Xhoiecr6w5LzdlmE8LOschC9NESaWinpinJ66fek1dhpp1Of8pT1t0hIJNjMWjyTus8p2yil7ppFjgTSCAhurdFbskJAHKdjlWmmCtSUmkXh+xIKynULaFqjSGcvS4W6X4eX3Ke0GjstbegxVrL6TS85mMhvMlbl9xrLWexNHxymWq5w+mHHceKBH2Te+hsmbrdOSkEk8fhzT3LuNZdy/nXfx/O8RImILMJMId2cJ4sQXEi3oNNMTOmdv2sQEwKtm/TJR6KUEqhstn0l0i7J0Oduwl5ZBL1Cug7AIKiGDU9zFLsGHWkmkkx+E6hNiqvR9daEGiWmDPovHSclQeDHyznBaJeIaJd8SO03zf8nVUUX8psuydGY35BdHqXXxC5PYO+m81UpV1AIe7xpZ8448hT2X7AXTuyqtkMbp0YCCKk36vz01zdy1uXn8efH/kqtUcsndNZG6Dxx2eUi5dxNZyOaDRXV+Yna4yg/cBXtcxigPI14yVc3WYgnAhFyC3+FhEFHA+L40gdFdVai9igFTjqFxm6EVwWiAVdnTvDtxH1OeWQOqkm0B9qifhhvT/P2Rn7PafUbyRxUk4iPP40Bv2lrT1Gph3jCH6xWKTllTjn4w5z63qN53YbzU9uZcgkgzt+eepRtj9/L6I9MJjU8+mazZPgnHdeg+Eu/MN3nkjL6g6y4gZKpKbxAYnsSj1tLcUqOw3HvOYpvnf4lyqVy5u+nTQBr1Y/AaXnD/C1RqvslUlb68WSsyfy0LQC1FriH6SdPmV4xZPIpUZHNye43fWjfVNMAKztl3vz67XMF/yxMbk+yWCwWS9+wCcBisVimKDYBWCwWyxTFJgCLxWKZotgEYLFYLFOUdTIBeJ7Xh2Vy/Vh2N7mX8hl5bGSfMd5Cwwb7s1pzsp+Xyd4+Jn0TtdasGRvp+37WqQTw4qsv8/9+/G3mf2BBJAplYrmXkmaxSFGNEQAnWKIaVicWJeqjNre8TaT1IdtFiffTRJ9DG2FhXlEk+BNGbBPLDpVSeNr3Q8dAnx3xaxTCAseixI+bGFpoGS92M9HG8LhptBm/kWbVssmxkvQ6L0opxurjfP6i/2Lhf5zEr+9b2re6j7XqofBJaK351X2/5SuLL+D6e25DKdXy0Pi8VY5NmYDWitXQufMUcflBFVyloeInAc8N9uFmf2B0VKVcUmgFBBX9CgfPdTNXOUYSFwhuoykHoWMDOkv1bjQYlOBVBaoVALxxFxlXkWZOJp0f5bdBVxVUSqBANzSqAV49R59DjRwUXqMBDXxZiUqpWWGcsc9aa5Sj8JT226jBrWuoBYIbefxGa1zXi46ZL23SXUOpE5HfKIUOHhuv8es+tJu9wriT9IEOjltmvwn/L62KoXGJisz6UEEb4t+Ij++s8aGTvEQkVZEz3ohI9L26W2fJnTfws9/eztxZ63PGwpP5yL5HMGN4emq7Pfc7mSsKu1UCv7pmJZfetJhzFp/PiytXMDI+2tMhfOE2/3Wn2xtZBOFCyYRuJzuyV3L8gKAkWYjL81UbewWxcGYuotAO4HSwpzV4EMq9dDs2kQ6JBu1273NTV6lz0IkGR7XkB/6kNmoNDY2qabzxRtc+R4Gu7OBVBCqq4zGUukbXXL+dXQZgpLniAd0GqiNIxfFlJ7oEnajP5dh5bkf7qpqq7p/nNH7jSGsyTtw2RRBrEZijix8GmkDSlvy77TeNhk4aPa4sAnNp9Lg6TeISt02pMZQl8aRJVlmECYcHhnA9l4VvP4hPve8EdnnDDt32vW5KQfzuoT9zzuLzWXLnDSilGBkbzWy33Xnyzio62fVfiy/GVnL8QJ320jAIEuL6M6hQXMxRDq7n+onEIb2McyTy1aru2BLIewSYbn2ND1YRf9amB5R/hZO2jZ6GmouM+dIT0RVHMNtvJpKUtwDC5FIHr+FOaKtC8BIE77oioCplPNHRVUHkN0qhHd05GXfoc5SsiN82CV5rfCmOjLT7cujjjuPgBrP9VG0M2uCIwo1dTeZRe01qX/z7oaJu3pHXnNj5NvKo+sZpT1Z5BO1a7LUltkxqrwk4yqFarrDp3I0588hT+ODehzFYHWzZZp1MAC+tXMHG79+ZhucafbCCUU2eIAhSks6zrLQEWuza9XyNmiwBJgmtES3ohus7YY4Ak0igu6+ryu933jYGgVvGPb9tA07n2X5aPI3UNHq84ffdhN+UlP9MgFJwFVbkwTBBnxkLbu94ZvSXIp8OH/5T0G/yKI+mIa2seTpbxUQLJ9oL9XHN/mZs8hgOVKp85ogT+fJxn2vfR6oEsFb9BjBeG6fklKg16sZsmhdkE3MCZUrQSlCVkpmngomvnBmqU5rCmV7FdYxELSgLuqyiK57CKEFXgBGDfW54qGkV/1ZPUYI+0/Bg1NykRmuNlB0zvh1OajxtdKwUVeJsR6PN+Q3NoG8y+BdVM21nrDbOypHV+dtjrCUWi8ViWauwCcBisVimKDYBWCwWyxTFJgCLxWKZotgEYLFYLFOUnglARDYVkZ+LyAMicp+IfDJ4fz0RuVlEHgr+nh28LyLyDRF5WET+JCI7x2wdE2z/kIgc079uWSwWi6UXaa4AGsBntdZvAvYAThORbYHPAbdqrbcGbg3+DXAAsHXw5yTgXPATBvAFYHdgN+ALYdJIy0ClSsNtRFo6JtBBKb0xGhrG3eB5ugUWkGkdPZTdqzXA1UaUxUQJDJeQgZKZxx8KuCM1ZMT1lzIW7bPrr9t3R2rF7UFkD0cQE49nFGDAwRNDWjLBecZRMFTylxEXJZSyKFqj0CLkw5cAACAASURBVGLSr+gWg89gdoOiKCMokGoJt4xfL2OAfjw0MpSzMMVApcqMoWm5v9/zbGqtn9Va/y54vQp4ANgEOBS4JNjsEuCw4PWhwKXa5zfALBGZB7wHuFlr/bLWegVwM7B/lsauN2M2v/nmNXzgXYcyUKkyPDCU5esRoahUU2jKJ++JCb+lREUDWjWAcRdpZAzcWoPrIS5+0A+WDDuIX93qiR8wstgUUCUHSoI4yi/cqigYLqOGK7mCjm9Pocq+1o1ueMioh6xx/QSY5UHoWkPdQ8aD6tjgYfTKFRjz/GOZw54a8WBNA6lrkCB4lZTf9qxddgRnehVmVXGGKy3fzxXEgmQXnVPx96EGyzCtnC9BO4IMlmGgFBXkOSUHygpVztFnmoJufpW3+H/nPYY0xdhCu4VF2UoKGSgh1ZKvh6UEp+ofA1XNN8lp9tn/O4oXOcXtQnuRIGDgynn77SiHoeog2262Nd8+/T/51w99MpcdyFgJLCKbA78Atgee0FrPin22Qms9W0SuBf5La31n8P6twD8BewEDWusvBe//GzCqtT67bR8n4V85sNlmm+2ybNmyxLa8umYl379lCWcvPp/lr7yUUQuoc0lQljLyTPomvbSAtEZp1VPEraXUX7rY8xvoz4x6VEiGUgTS8CtmO24qoBzHl7hIo29SKeGVOkgkBFpFyvOlKFJpAYUCa50qovuhBVRVyGAZLa0KqRP6HOpCdfObDrIcHe15HkoL3ljdTxbJHfGPszSlMzr1N6xr1a4bTSw6IeAXDqbRAjKtIdWrWErAqZRxY7IcyZsF+j5I8yq6B2m1gNJUHaeRmMmifzQ8MITneSx658F86n0n8ObXb99t32YrgUVkGnAl8Cmt9cou2SvpA93l/dY3tL4AuAB8KYhOO5k5PIPTDj2Wjx1yDHfdfy9fufICrv3NrSglLWqgaUSj4oQnoSmF3OqwcR2P0FG6OUxoz2u4zQBSkuZlqsYf5J4G8bft1s7QkbTWfpBoeChH8ERHFZt+kPaioNTLUb1AMU4qDrokE4KOlJSvsRIbHN0CQ9h+XWsg9UAfqByIuIEv9xAI1YWKkd36HB5/z/X8JFn3kHJTAZWYGqgoFVyEdT/X4TFxSgrXC4Jyo+F/6AhqsIJXblUD7XYc3bbj0uI3oTCfF/qNX6nabcC72gMBTzRqqOwn6LqHrgUaRk5QIY5GB1o63frcbLtGlfzKcoXg1ZtVsxM0b3pMqCItH6XQwgRdqSSBtW6qpZHfxG4NtRzzmDBf2NNuSq3hdz00qlr2b794oOvNSU67oGNPv4mdXzdBEykp3nQbK17bNu3JQolisDrABrPmcOaRp/ChvQ83qgaaKgGISBk/+F+mtV4SvP28iMzTWj8b3OJ5IXj/KWDT2NfnA88E7+/V9v7t+ZsetY09t1vAntst4KWVK/jfn/2Yry65kGdffgGtm+X/WUvOw5PQLnwlStA5S+LD7ziu4NZd/1I6lvmzSPpCrE/hLaGqf5/bC4ZHFhlj30wz6MhQGV1rgOtfCmvde3C0419s+bNep65waw0oqUISAJHwXAO05/rBPxZss5bZh+3w8HzhviEFjkIrP6hmPYZR0An9ptEALwg05PObKEFXS+iSgNs6289qM/peqFnVCGa0Xu9k3M2ep7V/pRskgTCY5fKbMEEr/7hRcVIn445tFJCSoJUDY/6ELLSX2W9ik7940C4ab+LS9RWnzKFvfQ+fOeJEdn/TzkZ/OwhJswpIgIuAB7TWX4l9dDUQruQ5Bvhp7P2jg9VAewCvaq2fBW4E9hOR2cGPv/sF7xlj/RmzOePIU3jqh0ujH4pNaJeEGR2yO0oSruc7o2NIFyRKHF1uUWRF4wvH+fZN9NmNbtuY0H/xtBcFmayJsxOa4MdYMec3SoKH/xg4huF5cJx8KpftxPtoSikpTHz+a0NjRdESrIsQKq2a1ABrOY4m/MbzGChX+c/jP8fl/3Yee2y7S1+CP6S7Angr8BHgzyLyh+C9fwb+C7hcRI4HngAWBZ9dDxwIPAyMAB8F0Fq/LCJfBH4bbPcfWuuXjfSiDQl/rDJr1bC9PmC4iabVFfvBZG9jf7xmLfBFSyFEJPcilyz0TADBj7mdPG6fhO01cFoHWxcDF2dpoMVisVj6g60EtlgslimKTQAWi8UyRbEJwGKxWKYoNgFYLBbLFGWdTAAPPvkInqHHwoUYeSTjBJuGV68YNtePPpvGfBvNHsT++M3kPi/G/Zrs6/572pvEz0IHqDfq/O7hP1M3+PjbJNaZBFCr17j8jmvY9bQD2emU/aL3iwq9hTogqEBHR/JreEDzu+IoX5iNmEZIXpsqeGB82W+fKmjPERVUmyok+K8ISgXHrepE+ixF2ug/4zwQJys7zX0UwHEcX0emXEKUmDmG4GvVODE9nZxE33UU4pjxGxVUjkvFMeM3oUhj2e+zSDHPCceueILS+FXzOfV4QsIWmRp77e0puvw8PAee1vzgtp8wd+EO/NOFX2bZ808VstuJTFpArzULFizQS5cu7brNY88+wbevuZQLrvs+WmtWja6ZsE0WfZ8QX7YBX76gJE1VRTeQHag1MlW0turjEJOC0IgbiKmRvjglrAR1Bsu400ow6PiFVlrDmIsadX35iZQ2I4kLQNe8CbopWTRLou84ypcwmFlFzyj7CQp84bfVDfSqGkoJXg8dmQltKDvNYxj2edxFxnUkGpeqzyJogdJAGWfOAM7MAb/SW2vc0Treyhr1sVpPXaEJNgnaVoppFrkeqp7db1q0pOLqnp4vJEfN889ZWr8JNXlKDl6onBkew5qHqmm8uttVs6elv36n/SLEqoKqLzaI1tDwUGMab7Serc+hxlB7sZbgP+heie83Kfsc6SoheLX6BB2kNJo9Sdt3OkbRMc4yVgJbSXUtlXIFhbDbG3fijEWncOBue/sTlu5tTKUFtFYmgIbb4Pq7b+OsK85j6d/+iOdpao1aT3tJ2iRxOgaYJEIFyzqBONTEQRhq/4hSvhZOqYc9DxxPcBudhcLCoMr0CkwvQTd53oaHGvXwRmvRIGgnfN/R4ks1pHSHToNFiS9FoaplvFllGC517/NIA7WqgVdrdExW4axKtyfjJHuuRo1rvPHOgdZxHFztUZk9hDNnADXQuRzGa7i4q2rUV44hSPI5CTV0Sg5uN8G/sI11zxet8zr4DTG/Udo/x93suU0tpI5+E0wYQr2nrjLRrhcdw07JLxor1RJepYdve/6kREZ9ldhO57hbUJ3YAPH3TefvRP7kelDvHYh76YZFk7iUE8lmvDGXKKYPDlMpV/j4Icdy0kEfYuM5GyXve11MAFprvnTZ1/naku9Qb9QTZ/tpiWYZBGJlgFRU78HRTiD/7DT8ABpdwnkequLgxYXf0uJpcPEDuKigTD8QtJru+LrxWS41g6sCGXEDqWV/poHW6IbnzyRzEFd4RPnBUc0cwJtRgkrGZzbUXdRqF2/VeCSr7WmdLhkn4WmouciYF9P79pUznbmDlGYNRLcB0qC1xh2p466s0RivRbfGPO35AbpbAOyE6+HUmeA3TsnBdQjvdaW35wWKrjW3KaYW6vOUyN7G8Kpg3MNzPV8aRPx26arTnO1nsVf3cMY07ljdP89C5qvzFgSk7EtBq6AtnhdTAM0iIx4326YcXFQ6Igzy4W2xTrP9tAxUqmiteccOe/DN077INptu1d7+dS8BPPPic2zxkbdQM/nDSFn8AZw1wCQRDBiE4H68AXsifjCdFruFUoS6By+MtDxroDDTyzBY8ttY9AEknoaVNRhzu8/206L9ZFqqlijNGsAZKhezhz/THnthtT94u832s7RxzPX/7jbbz2LPDf6UMwbpTjQ8qLlQdvIlu3Y8DStq/uzcFOEYbkzuH8lNc/qhH+UbH/9iy3vG5aAnCyWnZDQBSLWENiWtIgJVc08rQ/wfdtXsAV8R0gRl5c/4DeZ9Z8YA7qCh9QRKYLCEg/IF5IoiAmWhMm+aMX0oVXZwBso06g0j9qIkn3O2mmivJEjZnOAZwcN0jKEEp+LgjhoM1m7wfABzFo2T5XZPWoqskFpnVgFZLBaLJRs2AVgsFssUxSYAi8VimaLYBGCxWCxTFJsALBaLZYqy1iUA45ogRq0RrZM2btMgpYGKUXsqKPs3hShFqVx8uWacamXAqL1SuWz2vGRd85/GpDO5h7dWGPdtYyv6+sRkW3S/Vi0DnTk8g5nDMygpp1ARGAAl8QurglNS+NGCFUVp1qBfDi8gox6NV0ZTVSAmISVFZdYgzswB/5moWhgbHcdt5FvkJkoxe73ZbDxvY8q7lGmM13n+wSd46fHncHMuZxxafzobbDWP6nr+o+tWrlzNiyteoV7Pt0x3cHCITTfdgg02mIeIsPLFl1n2twd5ZfmLueyVK1U233Ib3rDdTlQHhxgdW80jy+7j2ReeyC0WOH36TDaYsxGVahXP9Xjh6Wd4/qlnqNd6V6In4ZQdhmcOUwqqkRujddasWIM7nu+cKEcxNGuIgVlDiBK8msvql1Yxvno8lz2/kYIq+1W3AkHxYIGljCVBDVXxZvvLkdWqBt5LI34NTR4UyGAFXfFLW5RXwltTy13gGBJVZJt4fnBYuaw1oBBPo+tuoYxQLfsTuZ222i63jbWqEAx8GYhrf3MLZ11xHr976M94npe+LkCCdf8l8cuzY0c/j14QAjJcxpk5gHYkkn7wPwr0RVyN+8oYeqSe6mQ7Q2UG1p+Grir/ofGxL4WVp27NpTZeT+WUAwMDbLjRRsxeb71IlC2y54GrXVY9u4Jn//o4IytW9W5f2WHWpnNYb4sNcEpOyyxTANfT1Ot1lr/0MqtW907SIsL6czZg8822YnBoGNWmceI1XBqNOk8+9DDPLVtGo9b7XK83ZwPeuN3ObLjJZihRLW3UnounNc8+v4zHnnyA1SMre9orlcqsv95cZs9e3xd5ixVWifYf+D7y6iqeXvYEr768oqc9BKpDVYZmDqEcQYJqb/8jQWsP7WrWvLyG8dVjqfymPFRh+vrTcaqliX6j/WrbsVdHWfPKGrw0wVvwg75qykiERNo6WvDqE7V1OtmjWkIGSqBau6QCiQ1V13gvjsDqdGOFskINlYM2tsotRDIQoy6Mp5c46SQHkVUGImpH2UE7E7WLIgkWAgmWDDUg0weHKZfKnHbIsZx88IfYZM68iftdFyuB23n02WV8+6eXcMENP4AOQnCAP4MZKOOJnuDMSUggjtXxyJQVpZmD6EHl2+vhXaGDM+LirhybeFXgCJWZg5RnD/j6Lz0uY5X4RVLiwdjYOG7bgBYRZs2ezbyNN6ZSraRSefQaLo3xOs8+8DgrnngBz22dIQ/OHmbuVvMYXH8aToo2EgSdV1au4qUVr9BotM5oqwODzN/kdWw0bz5OW5BObqDG0x6vvLCcZQ/+jZUvv9zycalUZtMttuaN2+/MwOAQjlPqfntBg+u5jI6u4uFl9/H88icn+MW0adPZYM5GVAcG/T736rLr4bouzz35NC888yyNtishp+wwNGOI8mB5QpBOQiG4nkd9pM7IitW4tdZzIkoYmjnE4OxhlKO69zdmz6u5rHppFbU1CVcFsdl+L8G1aHcafzbrJmzrBLP9ksZRvgZT1zZqf5Ijr9bQLyeMFQUMlKGqJkzikoiSlQveSOergrSCcKk0i5T4oo+kjDexyaKuJ1+ZVkoVlBJ2fcOOviDc7ntTcjrfwJkSCSBkvDbOVb++kbMuP5f7lz1E3W3Q8Bq+8FU5naMk0ZL18ZChMs6sibP9tISzO2mA+8ooooWB9YdhwEkVEJJt+nov9fEGIooNN9iQ2XMmzlRT42k8z/v/2zvz4LiqK/9/znuvd7UkW5LlRd4XjGPAMDYEMITFEDazB0gcJkwywRRFJakMzGTmr99S86upImSZlUBCEookhBhstiSEsWGCgRjjhcXxglfhTZIlW9bS63vn98dryUJuSS2p3SbS/VS5bEntq/tO3z7n3nPv+V5aDxyhYefHhOJhqmfUYgedIeWUJed00uk0Tc0thMIxpk6ZSSwexxJ7SDlgL+uSSaep37GDxPEOZs89i0lTZmKJIAOoJOZtz3Xx1OPAod3UH9pJJBKmamwNlmX5UtuDpGtV0H60lQP76kmlE8QqYogj3dpOg2ovt/LTrNLR0o66HvGxcewCA0leFNTzSBzrpONYJ9hWTk9naJo8J1YFvuIpQRuJBFBhSJ+V7lVBOrcqSLk9ZvtD66MAeKDJrK+YO4D424B97KGF5XXJeDjWkJ63u71u1VJ/VRCPxBARll9/N/ff+NdMGz+5oLZGVQDoybb6nfzrqid49PVf+TPlIlyeIYBUhbFiwaJtMtmWRcCxi7bx59gO06fPKmimWggCJDrbUM8r2kZdJFpJWXnVkJxqPmxsxsRqEAae/RaEKvuP7EA9t2ibdY0NB2hvay3a4QXBTycUa9yIQsP+I0MXY+vdnggaECzbLp58SWfGTwsVCfFAj6WKk9snJ+UctPxnLtJlPYIwvWoS/3LPd7jpoqsJBgZ3cKPQAPDpPiYwBOZOmcV/fuP/4TiBojh/yN2NEggU/YRB9wUaRaDLwRTLcXUpIBbzlIZlD21G3RdiWcXto+Ty70VqDkBUi3tyTYo7btycAyyW41LV4jp/AKWop8zUy3PXwDDwcqKNxbypLRwM8dAd9/GFz90waOc/GEZcADAYDAZDYZgAYDAYDKMUEwAMBoNhlGICgMFgMIxSTAAwGAyGUYoJAAaDwTBKGXEBoKXtGI+sfBzXzRb16JjnnrhouxhIf5WEQ8BCui+cLhYKRX1mCwuryOpfxS5jEd+SRWtPxS/+Kl6DFO14M9D9fhTzmT3XK+5npRh3Gvdqr+j1T7lL3otFKpPmiT88w7ptm4rf1x6MiEIwVeVP2zfyvVU/4aX1q7Esi85UAtuxcNFhVfphQaQsRvWMSXhBobGpAVCy7tDExBzbQUQYN84XFDvW0kQqnQLVIQWEaCCMp8oN86/kijMX8/ah99l0+M+ICGl38MUzXZWqkWAZ0Uglyc5WmlsPYiFkvcELlAlC2AlRGY5z59nXE4mW88Ket2lLd5JyM0M6I98t0gUE7BBlwXIsy8ESGdLpczv3zLXRKmrC5ew+spMdRz7CwiLtDd6Ggq/vEw5EqCobT0tLA9vr38PzXDLZoYnGdVUDJ1MZPE8pi4YRa2iVxZCr3FUP27JJp9Ik2pJ0diawxR7SXcxWrn9WVvA6Un4VcHmgIKmUvuhyp+p6kMgiKc/XzNGhqQJ3V+4qeMksdhbcjDt0/9ClH2bbeHjd8g/2EAtQu+UoEDTrIlhEQ2HGj6nhwduWs+zym4lHywpta+RXAh/vbOOp11fx3ZWP0djaTCKVzF+MYUluFlFYsVTACeCqy7jJE6meUUe0/ITRVZXW48doamygraOtW5enP2zLrxCsiFcyrnYCFRWVn1idJBKdtDQ30nK0GduyyLr9O1pLhJAToqasiuUXfYmbzr6aslCs++ctiVZ+v+sNVm1fTdrNkMgmB3hisMVGgcpYNeXxcQScUPfPPM+lta2J5pZ6kukOtIDiprATxFNl8dTzWHbOjSwYP7f7mVWVLS17eW7nG6xv2I6FkBrA0Xa9e32ptlpiEw3ECAeiSIEzbkd8CY5ZlZOZVj6RcI9nTmXTbGvczoaDG+lMJ8h62QGf2bEcPPWoiU+gtnIKsVC8+2eeehw6so9tezbSePQAUkBAtRBc9VAP2hMJMr10YmzboiwaIRh0CgoE3SMu9z5YlnxiHLquR6ItQcfxTjSn5TQQXZINkvDQttTJekBlAayxYTxHfJ2rQrR71MPCwuulH4UquIqVUrxUpiDH3TVeLAQvm0d901Mk42vwdEkxDPjMudW7r9ulJ1dlC4hjF6wk2h2YELw+tIBi4Sie53HnpUv51i1f45wZ8/ptc0QHgE27PuT7LzzBb958GVtsOlKdBbdpOX607h2lLRGwLMLRMLWzpjB20nhsp/+Ky1QqxZHmRpqaGhGRkxy3YzsoUDtuAjU1tYRCofwN5XBdl6NHm2k+0kAmm0a9TzraSCCMpx5XnXEJX7vwLhZMmtdvmsv1PDYe3sKKrX9gS9NHAGR6OB0LQQXCgSgV8VrKIpUDOs9kqp3mo/s52noIS6yTnFg0ECYaiLDsnKXcOPcKxkTK+22vNdXBK/vWs2rXWhJuikSvGbKFnJAh7relE4ScMLFgHEtOXhX4s32oCVcwa8xUaqNj+7WhqnK4rYFNBzezs3k3lsgnbCj4s/CgE2J85VRqyidgW/2rrHck2thZ/x7b698HlHSvZ+6a7adSWRKpFF4BSpHhUDC3KrByz9xD6baH5k1XFBjomVPJNMnjSToTyW6dn+728B2g5YJ3POkrbg5EwMIaE8Ers/OuCroDuwcUMhv3FNIukvSdem8/diKQ9O1Uez00ZBUr62tN9ZbG6NYF6+p7Ifc3SJe/OTkLIT2mNOq6hSmq4k8mQ4Eg02vrePD2+7jjkhuIhiMn/+qRGACS6SQXPnQrOw7uJpXJDGmp2o0liAVYfha0qm4842bUEavs32Hlw/M8WluP0dh4mM6EH4zKYnFqx0+gsrJ/B5MPVSWR6KD5SCOtrUcJOUEqI+V8/cIvctuCa6kYwKnmo7Gjhd/u/B9e3PEaCTeFp+rP9svGEQwM/rIUz3M5dryBIy31ZLMpLBEWTprP3QtuYtGk+YPOe3vqsblpF8/t/CObm3aRVXfYdzTYYhMNxgk5YRzLnyXPrKhjesUkokN45mQ2yZ8btrHxwCY6swk8VapjtdSOmUJZqHzQ77Pnuexv3MO2PRs40noY13NRDzoSSdJDvKPBcWzKohECATu3JySI+HnvoeyJuVmXRFuS9uOdqOf5Ug8JD68tNTS9fSG3KojgOYD4uy6eO0RtfFXI5kTjcncodKVQhjx0PEUyoOkegd7yJ0tD1mCyBLFzkxHNrSKG00egLBzFU+Wph37ILRdf84mfjcgAcLC5gVnLLyWRHsblFr2YffEC4tVjcALFuRsnnU5j2TahUHFuoJoZn8SXzriGhZPPKsqmtuu53P/6D7HtcEEy0YXwhanncf74M6iJjS1Key/sfosff/gy6SHsOeQjaAdZOusqqiKVRdmQVVXW7t+M2AEcuzg3l63d/Cqbtv2poNl+IQQcm+qxFX46ogjjRlU5tKEeUsNzWj2xJsTwQkXcIE9ksDpcPLdIm+SeB8lcW8US3xMLLeImPsADS7/Cv93/f3v9nsICwF/UjWBAwfndQolVxIvm/AHC4UhRBc+CgSDzJ8wp2okm27KJhuJkhrN66sXkyolFc/4AZYGIL2ddRMaGK4p2GkdEKI9WnpSuGg5BJ1Q05w/FF/MTET9XXkxdu+L6QciteIrZXtGv6ZTin1wbDiU/Bioi14jIdhHZKSLfKfXvNxgMBoNPSQOAiNjAfwDXAvOAL4pI/9vZBoPBYDgllHoFcD6wU1V3q2oaeBq4qcR9MBgMBgOlDwCTgI97fL0/971uROReEXlXRN5tamoqaecMBoNhNFHqAJBvR+UTWyKq+piqLlTVhTU1NSXqlsFgMIw+Sh0A9gM9bzWuAw4OpgF3iBIMf8kUVUvmFHAqTjUUUycJinulIFC8uzdHMafiCPqn+Vg78Ok6AkTpA8B6YLaITBeRIHAX8EKh/7m2spo7L7mBcDBENHRy9dtgiATDhAMhapJh4oEoUWd45/ZtsQjZQSbFqpgYqyJsB4ftuKNOiKPJVjozhVc6F8I1k88lZAcID/MMuyMWActma8vHRb0P9ezqGcyunEjQcnBkePffRpwgtoDnJrFFsIcZCKxc4dKZlROIOkFCA1T9DoQgBC2bRTPP44zaGUQDYZxh3vkbD8UoD5ZxRvV0QnaQsDO8O2UjgTAhJ8iFf7WQeCRGWSQ28H/qh4ATIBwMMad8MnUVtcRC0WEFaEEoC0Wprapl3uTZhAMhgsN85rJwjGgownmzzyISDBMLR4fVXjgQIhwIMW/qHGoqxhIfpg0tyyIWijBn0nS+fMWtQ26n5IVgInId8APABp5Q1X/u67V9SUEca2/lyTXP8ciqx2lpP0ZHKlFw5C8LxygLR/nWjV/lq1fdQU1FFVnP5c39H/DUllfY1rIPT7Vg4bOIE8JTj89Pv4A7z7ySWWPqANje8jHPbF/DmvqNCELSLezMeMByEGB+9Qy+dOYSLpgwD9uy8FRJZD2S7tCktSwg4liEbF8qIO1mWde4kxf3buRwZytZzy1YtCtsBxCEK+s+w9WTz2JcpCKnDwRZTwutas/bR8eSXJW9cLC9mVW73+KlPetQIJEtrADQQgjYDnVl1dw15zIunXQWITtAxnPZ23aUba1NJF13UIJdjljYIsyuqGZmfCwRJ4CnytZjh1lzcDv72ppRClfqDFq+9tK5VXVcOmEOdbFKAN4/sJ0fvv4LfrXhd1hi0ZEuLPgH7YBfjT1lPg8tuYfr5l2CYzsc6TjKrzb/lh+vX0Eik6QjnSioPUGIBMKMiZRz32fv4LazrqYiHCeVTrHyrVd4eMWjbP14Jxk3O6B2VRddgeNrV9/JAzfew6yJ01BV3t69me+t/jkvf/BHRIREZmDtKvCdqqqyZO5n+bsl93DZnPMREfY17uc/f/8UP/rDL/HUoy3RUVB73TIL4+p46Obl3HHR9URCYdoTHfzy9ed5eMWjHGppJJFOFqQZBBCPxAg4AR5Y+hWWX7eMiVXj8TyPVze9wXdX/Ig3tqwHlFSmMP8QDUXw1OO2i6/lWzd/jYVzzsn7uhFZCdwbVWXtn9fzyKrH+f3G/8ESi0T65METDvgaPJfOv4AHb/46V55zcZ9VsAfamlix7TVW7XwDVOnM43QsEYJWgJpoJXfPv4arpi3qU1qgI5PglT3r+eXWVzmaaiOZza+AGXVCWGJxy+xLX+ThiAAACLZJREFUuGX2pYzvo7BKVcl4SmfWI1vAexe0hIhj4UjfUgD1bUf4Xf17vHl4B5ZAMs8H2kawLYu6WBVLp53HonEz+pypqipZ1ZN0wfrCFvrtX9rNsvbghzy943X2tTWS9fI777AdRFGumnwut866hBkV4/vsX3Oqk22tRzjYeRwB3Dy2tPALqapDUeZW1jA+Evc1o/LQkurgzcO7eKthN54qqTwTCAEClk08EObyiXNYWDO1z1VYe6qTX777Mg+v/hmHWptIZFJ5V1nx3Oz53otu5/5L7mJ6dV3e9jz1eGPPBh79069Zl9MgSuVRiw3nJjRXzLqA5RfcwaK6vivQt9Z/xL8+/1OeXP0slgjtyZODlWPbBOwAZ9TN4KHb7+PWi68hHMz/WWluP8ZP317J91b/nLZkB+19aHzFQ1HCgTDfuHwZX198O7Xl1Xlfl8lmeP6dV3n4+cf4YN82sq5HJs8zx0JRPPW48+Lr+eYNX2XB9Pwn01WVd3e8z/dXPs7Kt1/BEl91uDdBJ4glwqIzzuGh25Zz7aLLcez8K8X9TYf40e+e4j9efJKsm80brESEaChCdfkYHrxtOXdfeSsVsf7lYEZFAOhJU2szT7z6DD944Qnak520Jzv86GsHeOCGv2b555cxsaq24N+dcbO8Xr+Jp7a8wp7Wg2Q9l4Dt4Kly2ZRz+dKZV3Fm9bSC21NVtjTv5emtq1l74P2ckJqLZVnMrqxj2bwlLJ509qCW/64qyaxL0v1kSLGAsG0Rdqw+HVY+ktk0bx3+iBf3baQl2U7ayxKyAyjK5yacyeennM2kQVT8qvqrgax3csgT/Nm+xeDy8/uON/Dcrjd5Zd8GRCCZzRCyHWoildw15zKumLyAyCCW/yk3y+62o+xobSLjeWTVwxELEZgdr2JmRRWxQbTneh4fHj3ImoPbOdBxzNfOsSxUlfljJ3LZhDlMLStcH0pVebd+C99/7UlWvr8GSywy2QyObfOZCbN46Mp7uPnsKwk6hafzDh1v4qlNL/Kzd1eS8bJ0pBPEghFigQh/e8HtfPGc6xgbrSy4vc5kgmfeeImHVzzK3ob9JNMpoiFfpvzLl9/CN27+Gz4z9YyC21NVXtu+jkf++2es3r7uE4qai2eex98tuYer512EPYjPyo6Du/n33z7JT9f8BhA6Up1Eg2FqK2t46OZ7WXbpTcQjhUktQy4L8d/P8shzj9HS5mchysJ+MF5+7Ze5f+ndTKudPHBDObJulpffWcPDKx5lw0cf+OJ9to3neVy/6Aq+fevXuWjewoLHzagLAF14nsfq997kN2++zNLzl3DtX13WZ/QtlL2th1i5449MLKviupkXEQ8OLx94PNXBS7vfpjlxnFtmL6YuPm5Y7akqaU9Jux4h2yIwROGvnuxqbeC1g39mdsV4LqydTXCYNuxaFUD/s/1CSbkZXtv/Hltb6rl22iLmjin8w9ZX/xqTHexrP8bEaJyJ0fJBBc98NCXaeKthN+XBCBeMm0Z0mHnpY53HefKdF9jbcoh7L7qNueNnDKu9rJdlzc51vPrR2yyddxmLp5037H2rTTs/5Cev/JrzZs3nzs8tHXbu/HBrEz9+cwWpbIZ7F3+ByWMnDKu9ZDrFs2//jrVb13PPFbdz/uwFwxqLqsraLe/w1JqVLDl3MTd99mqCgeG9z7sP7eNHv/0F1eVj+Zur76C6YvAyK6M2ABgMBsNoZ0QEABFpAvad7n6cIqqBI6e7E58CjB1OYGzhY+zgMxw7TFXVAQupPtUBYCQjIu8WEqFHOsYOJzC28DF28CmFHT7dFUYGg8FgOGWYAGAwGAyjFBMATh+Pne4OfEowdjiBsYWPsYPPKbeD2QMwGAyGUYpZARgMBsMoxQQAg8FgGKWYAFACRGSyiLwmIltFZIuIfDP3/bEi8qqIfJT7e8zp7uuppB87/C8ROSAim3N/rjvdfT2ViEhYRN4Rkfdydvjfue9PF5F1ufHw65xi7oilHzv8TET29BgPC053X0uBiNgisklEXsp9fcrHg9kDKAEiMgGYoKobRSQObABuBu4BWlT1X0TkO8AYVf2H09jVU0o/drgDaFfV757WDpYI8bUHYqraLiIBYC3wTeDbwHOq+rSIPAq8p6r/dTr7eirpxw73AS+p6orT2sESIyLfBhYC5ap6g4g8wykeD2YFUAJU9ZCqbsz9uw3Yin8V5k3Az3Mv+zm+Mxyx9GOHUYX6tOe+DOT+KHAF0OX0RsN46MsOow4RqQOuB36c+1oowXgwAaDEiMg04FxgHVCrqofAd47A8FTh/oLoZQeAB0TkfRF5YqSnwqB7ub8ZaAReBXYBx1S1S0f6pPuyRyK97aCqXePhn3Pj4fsiEjqNXSwVPwD+Hrqv06iiBOPBBIASIiJlwLPAt1T1+Onuz+kijx3+C5gJLAAOAY+cxu6VBFV1VXUB/rWo5wNn5ntZaXtVenrbQUTmA/8IzAUWAWOBEZsWBRCRG4BGVd3Q89t5Xlr08WACQInI5TifBX6hqs/lvt2Qy4t35ccbT1f/SkU+O6hqQ84ReMDj+A5xVKCqx4DXgc8ClSLSpbs96Puy/5LpYYdrcqlCVdUU8FNG/ni4GLhRRPYCT+Onfn5ACcaDCQAlIJfP+wmwVVW/1+NHLwBfyf37K8Dzpe5bKenLDl1BMMctwIel7lspEZEaEanM/TsCLMHfD3kNuD33stEwHvLZYVuPSZHg571H9HhQ1X9U1TpVnYZ/T/oaVV1GCcaDOQVUAkRkMfAG8AEncnz/hJ//fgaYAtQDX1DVltPSyRLQjx2+iJ/+UWAvsLxrb2QkIiJn42/q2fiTsGdU9f+IyAz8GeBYYBPw5dwseETSjx3WADX4aZDNwH09NotHNCJyGfBg7hTQKR8PJgAYDAbDKMWkgAwGg2GUYgKAwWAwjFJMADAYDIZRigkABoPBMEoxAcBgMBhGKSYAGAwGwyjFBACDwWAYpfx/G7IEy2Y3tR8AAAAASUVORK5CYII=\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f537c151940>"
+       "<matplotlib.figure.Figure at 0x1112d1a58>"
       ]
      },
      "metadata": {},
@@ -336,7 +336,7 @@
      "data": {
       "image/png": "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\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f533f4f14e0>"
+       "<matplotlib.figure.Figure at 0x1a19493320>"
       ]
      },
      "metadata": {},
@@ -387,7 +387,7 @@
      "data": {
       "image/png": "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\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f533f05edd8>"
+       "<matplotlib.figure.Figure at 0x1a2236cd30>"
       ]
      },
      "metadata": {},
@@ -422,7 +422,7 @@
      "data": {
       "image/png": "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\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f533efeeef0>"
+       "<matplotlib.figure.Figure at 0x1a2255a6d8>"
       ]
      },
      "metadata": {},
@@ -449,7 +449,7 @@
     {
      "data": {
       "text/plain": [
-       "Text(0.5,1,'Scatter plots of Ranking vs Height')"
+       "Text(0.5,1,'Scatter plots of Height vs Ranking')"
       ]
      },
      "execution_count": 9,
@@ -458,9 +458,9 @@
     },
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzsvXt4XFd97/35zUVzkUYjjSxfZUvO1XbiJA65QIASEkiahAO0XEsPDbemfQ/08JZLCzzlLact57R9ymlpOAeaQk+gp4VSaAI0LYRCXC4FQm4EEicksWVblmXLGmkkzWju6/1jj2RZ3nu01+zZoxlrfZ7Hz2hm9p69Zkte37V+V1FKYTAYDAbDSgJrPQCDwWAwtCdGIAwGg8FgixEIg8FgMNhiBMJgMBgMthiBMBgMBoMtRiAMBoPBYIsRCENbIyJKRC5Y4zH8kogcFZF5EdnX4mvfLSJ/5PDep0Tkw60cTysRkV8VkftdHvsWEfme32NabxiBOAcQkReJyH+ISEZE0iLyfRG52uNnnvUfrt5ktdb4PEH8GfAupVSPUupRm2srEcnWBOSYiPxPEQn6NJYllFK/qZT6Q7+v0ygisl9E3rHitetFZMzN+Uqpv1NK3eTXWAyrYwSiwxGRXuCfgTuBFLAN+G9AYS3HZYeIhNZ6DA0yDDyxyjGXK6V6gJcAbwDe5vuoDAafMQLR+VwEoJT6vFKqopRaUErdr5R6fPEAEfl1ETkgInMi8qSIXFl7/QMi8tyy13+p9vpu4FPAC2qr4hkRuQP4VeB3aq99rXbsVhH5sohMisghEfmvy677ERH5koj8XxGZBd6ycvC1XcmnROSbtXH8u4gM231REUmKyOdq1zosIr8nIgG78daOv7X2veZqK/v3OXxuoPZZh0XkZO0aSRGJiMg8EAR+IiLPrfbLUEo9C3wfuGLZ57912f0/KCK/sey960VkTETeW7v2cRF5q8M4EyLygIj8pVgs7ehW+xwRGRCRr4nIrIj8WET+yGnHJSJfF5F3rXjtJyLyy7Xr/nntGhkReVxELl3tvjhRu8+fqY33WG1cwdp7Z+wKReQmEXm6dt3/XftbWblD+TMRma79Ld5Se+2jwIuBT9T+Pj7R6HjXHUop86+D/wG9wBTwWeAWoH/F+68DjgFXAwJcAAwve28r1kLhDUAW2FJ77y3A91Z81t3AHy17HgAeBv4/oAs4DzgI3Fx7/yNACXh17diYzfjvBuaAXwAiwMeXXxdQwAW1nz8HfAVIACPAz4G31xnvceDFtZ/7gSsd7uHbgGdr4+8B/gn4W7sxOJy/fIy7atf97WXv3wacX7v/LwFyi2MBrgfKwB8AYeDW2vv9y+85MAA8uOL+L/0+XHzOF2r/4sAe4OjK+7Xsc38N+P6y53uAmdrv5+ba77yv9n12L/7N2HzOfuAdK167Hhhb9vxe4K+AbmBj7Tv+xsrfKbABmAV+GQgB78b623rHsmNLwK9jCfr/A4wD4jQW88/F/LLWAzD/mvBLtP6T3g2M1SaJrwKbau99A3i3y895DHhV7eel/5zL3l+akGrPrwWOrDjmg8D/qf38EeA7q1zzbuALy573ABVge+25whK1IJbZbM+yY38D2F9nvEdqx/SuMoZvAf9l2fOLa5NNaPkY6pyvapNXtvbz54FInePvXfyd1CbMhcVr1V47CTx/2f35G+BnwPudfh/1Pqd270rAxcve+6OV92vZe4nadxmuPf8o8De1n2/AEubnA4FV7ut+LJGaWfZvnppAAJtqv9PYsnN+BXhg5e8US7R+sOw4wRK55QLx7LL347XfxeZlYzECofnPmJjOAZRSB5RSb1FKDQGXYu0K/qL29nbA1jQiIr8mIo/VTEgztXM3aFx6GNi6eH7tMz6E9R9/kaMuPmfpGKXUPJCufYflbMDapRxe9tphLJ+LE6/BWkkfrpkjXuBw3Fabzw1x5vdYjSuxxO0NWMLZvfiGiNwiIj8UK4Bgpjam5fd5SilVXvY8V/usRW4DYlhmtHo4fc5g7fss/104/l6UUnPAfcAbay+9Efi72nvfBj4B/C/ghIjcJZYfzIn/qpTqW/wHvGLZe8NYu53jy/5+/gprJ7GSrZz5d6KwFkTLmVj2fq72Yw+GhjECcY6hlHoKa2W5aBc+imXeOIOanf+vgXcBA7X/vD/DWpmBtfo66+NXPD8KHFo+ASilEkqpW+ucY8f2ZePqwXK2j6845hTWKni5f2IHlvnM9jpKqR8rpV6FNeHcC3zR4frjNp9bBk64GPvy6yml1BeBH2CZ3RCRCPBlrEioTbX7/C+cvs9u+Gvg68C/iEj3agfbMIn1fYaWvbbd4dhFPg/8Sk1UY8ADi28opf5SKfU84BIsH9j7GxgTWH8/BWDDsr+fXqXUJTbHHl8+fhERzvw+q2HKVjeAEYgOR0R21RyTQ7Xn27G26T+sHfJp4H0i8ryag/GCmjh0Y/2nmayd91ZOiwpYk+OQiHSteO28Zc8fBGZF5HdFJCYiQRG5VPRDbG8VK1S3C/hD4EdKqTNWuEqpCtYE/9Gas3YYeA/wf+3GKyJdYsXRJ5VSJSwTUMXh+p8HfltEdtYE6r8D/7BiNa7DHwN3iMhmrF1PhNokXXOcNhK6+S7gaeCfRSSmc2Lt3v0T8BERiYvILiyTTT3+BUs0/wDrXlQBRORqEblWRMJYZqg8zvd1tXEdB+4HPiYivWIFC5wvIi+xOfw+YK+IvFqsaLh3Aps1Lrfyb9fgAiMQnc8clknjRyKSxRKGnwHvBVBK/SOWDfnva8feC6SUUk8CH8Na7Z4A9mJF3yzybazQzgkROVV77TPAnpo54N7axPOfsCJ2DmGt8j8NJDW/w98Dv49lWnoeVrSUHb+FNSkdBL5XO+9v6oz3zcCoWBFUvwn8Z4fP/Rvgb4Hv1L5HvnathlBK/RT4dyyfwRzwX7HEbRp4E5aPSPczFXAH1qr7KyIS1fyId2H9XiawvuvnqRMKrZQqYInKy7Du8yK9WDuaaSxT3BTW7qhRfg1LRJ+sfeaXgC024zmFFVTxp7Vr7gEeqvcdVvBx4LW1CKe/9DDedcWih99gWBNE5G4sp+XvrfVY1hMi8idYDtzb13osjSAiASwfxK8qpR5Y7XhDY5gdhMGwDqiZIi+rmRmvAd4O3LPW49JBRG4Wkb6aX+dDWH6cH65ymsEDnZrZajAY9EhgmZW2YoW/fgwrp6STeAGWuWvRJPVqpdTC2g7p3MaYmAwGg8FgizExGQwGg8GWjjYxbdiwQY2MjKz1MAwGg6GjePjhh08ppQZXO66jBWJkZISHHnporYdhMBgMHYWIHF79KGNiMhgMBoMDRiAMBoPBYIsRCIPBYDDY0tE+CIPBsL4plUqMjY2Rz+fXeihtSTQaZWhoiHA43ND5RiAMBkPHMjY2RiKRYGRkBKvAq2ERpRRTU1OMjY2xc+fOhj7DVxOTiIyKyE9rPQceqr2WEqu95DO1x/7a6yJWK8VnxWpjeKWfYzMYDJ1PPp9nYGDAiIMNIsLAwICn3VUrfBAvVUpdoZS6qvb8A8C3lFIXYnXy+kDt9VuAC2v/7gA+2YKxGQxrSiZX4tCpLJlcaa2H0rEYcXDG671ZCyf1q7D6J1N7fPWy1z9Xa7ryQ6BPRM4q+2toIgvTMPWc9djudNBY3U76mVyJ+346zveeneS+n46vmUjoiJQRtPWF3wKhgPtF5GERuaP22qZao5DFhiGL7QW3cWYbxDFs2kmKyB0i8pCIPDQ5Oenj0M9xFqbhiXvh4H7rsZ0n3g4aayZX4h8fPsq9jx7jHx8+WnciTeeKVJRiS2+MilKkc8WGr9nopK0jUu0iaO2GiPDmN7956Xm5XGZwcJBXvOIVdc6C/fv3r3rMWuO3QLxQKXUllvnonSLyC3WOtdsL2bWRvEspdZVS6qrBwVUzxQ1O5NKgqtC71XrMpYE2XSE6jLUdOTyV5fGxGU7NF3h8bIbDU1nHY1PxLgqlKj89lqFQqpKKdzke64TXSTudKzKfLxMQYT5fritSzRK0c43u7m5+9rOfsbBgFZb95je/ybZt9Vqldw6+CoRSarz2eBKr9vw1WI3OtwDUHk/WDh/jzD65Q5zdl9jQLOIpkADMjluP8VT7rBBXmpPiKXIlxeT4IXIlZY3dRzyJpJxe1aja81UOR0RpNahejtdJOyTCk8cz/OC5KZ48niFUx2adincRFOH47AJBkYYErR3wYxF0yy23cN999wHw+c9/nl/5lV9Zeu/BBx/kuuuuY9++fVx33XU8/fTTZ52fzWZ529vextVXX82+ffv4ylfaoxK7bwIhIt0iklj8GasP78+w2i0udrG6ndM16b8K/Fotmun5QGbRFGXwgVg/XPJqOO966zHW3x4rRBtzUkb1cF/1Wr5fvYT7qteSUT2+Xd6rSA6nurliqI/BRBdXDPUxnOp2PDadK9IVDnDp1j66woGG7rfXSbusFFuTMTb0dLE1GaNcp/x/Mh7mtr1bedEFg9y2dyvJeGOx9WuJX4ugN77xjXzhC18gn8/z+OOPc+211y69t2vXLr7zne/w6KOP8gd/8Ad86EMfOuv8j370o9xwww38+Mc/5oEHHuD9738/2azz7rNV+JkHsQm4p+ZFDwF/r5T6uoj8GPiiiLwdOILVZxasJum3As8COeCtPo7NAJZIxPqXnrbFCnG5OWl2HHJp0qqLfChBf2ojx2cXSOeKvk1Oy0WykWsl42Fe+7ztpHNFUvGuuucuNzEloqGG7ncyHubFFwwyms4ykurWvi8LhTLff+4UlSoEA/CKy7auer1OFIZFvP5+nbjssssYHR3l85//PLfeeusZ72UyGW6//XaeeeYZRIRS6WxRuv/++/nqV7/Kn/2Z1d47n89z5MgRdu/e7XlsXvBNIJRSB4HLbV6fAm60eV0B7/RrPIbVWVwhupncfMPG9JVSrROuVLyLTK7Ecyfm2ZCINDxpu713Xk1MmVyJbzw5wVy+zNMTc7zuedu1fm9TuSKbemNsTEQ4OVdg6hz3K/i5CHrlK1/J+973Pvbv38/U1NTS6x/+8Id56Utfyj333MPo6CjXX3/9Wecqpfjyl7/MxRdf3LTxNAOTSW04gzVfIS6avnJpSyxi/SShZcI1u1DiyfEMuVKVk3N5ZhdKvu5WusIBLh3oa3g1e3gqy0OjacLBAKVKlWtGUlwW73N9/kiqm3hXgJmFEvGuACN1TGKdRrlSpVJVBANCKGhZ0/1cBL3tbW8jmUyyd+9e9u/fv/R6JpNZclrffffdtufefPPN3Hnnndx5552ICI8++ij79u1r2tgaxRTrM7QfsX4YOP8M81cyHmbnBn0Tii6j6SyBQIBLtyYJBAKMpv2zAzdjNTufL3MknWP0VJYj6Rzz+bL1hsu8ke0Dce548fm85KJB7njx+WwfiNc9vi2j3GwoV6pkFkrMF8pkFkqUK9Wl9/z6WxoaGuLd7373Wa//zu/8Dh/84Ad54QtfSKVSsT33wx/+MKVSicsuu4xLL72UD3/4w00dW6OYHYTBsIyRVDehAIxOZQkF8HVF3bTVrKx4XHT0q6plpqsFIdiRyZV4bGyGilI8NjbDtv644zgWHbwVpQiKtLWjulK1nO2LO6tKVREK+nOt+fn5s167/vrrl0xJL3jBC/j5z3++9N4f/uEfnnVMLBbjr/7qr/wZoAeMQBjWBZlcydVEvH0gzntfvmvJ6bvaitrLtZpBTyzE9v44XaEAxXKVnlgIclNnOfqdBELHaeuXg9cPggFLLUu1ncPic4MeRiAM5zyL2c1z+TKJaGhVR+72gXhDwrB4LberbN1x2TGc6uaakRRzhRKJSNgKqxV1lqPfCR0zV1tEubkkFAyQjIXP8kEY9FifArEwfYYT1HBu49WRq8NiZnJ3NLSUmew06S9mXSeiYQ6dmm9oXMl4mJv2bF4R5nq2o7/e+W7NXG0R5aZBKBjwzay0Xlh/AqFhn20bjKB5YtGRGxShotRpR64PLGYml6sQCsAtl9apNylQLFeZzZcolaurZl3bkcmV+O6zk1SU4mg6x22x2o5lRY5LPXQi19Y8ys3QUtafQNgkYrX1pNuJgtaOrHTk+kRZKXZu6LGuo6ibmdwf66JSVUzP5umNhumP6ZtsOskvYOg81p9A2CRitTWdJmhtiK0jtw5enMwhEQ6dml/aQdSrbTSdKxIMCBt7o5TKVaZzRW3fR6v9Akencp4c+IbOYv0JhE0iVlvTaYLmE14mbVtHbp3reAnlLCvFnq1JurtCZIvlujsIBMKhAL3RMLP5UkO7m1b6BY5O5fjYN59aEr/3vnyXEQkgGAyyd+/epef33nsvIyMjvlzr7rvv5qGHHuITn/iEL5+/kvUnEKBln11zOk3QfMBrtI9OfSSvJptUvIueSIiKUvRE6tdXWizsN1cocd6G7rrC1Q6MprOUqzAy0M3oVJbRdNYIBFYOw2OPPbbWw/AFE/vVCdhkFq8nFqOQDk7O89Boum6PBSfcZs96NdnoVDxdFK5XXTHEaxsIcQW9BkVeaWUSoa+0oDthpVLh/e9/P1dffTWXXXbZUhLc/v37eclLXsLrX/96LrroIj7wgQ/wd3/3d1xzzTXs3buX5557DoCvfe1rXHvttezbt4+XvexlnDhx4qxrTE5O8prXvIarr76aq6++mu9///tN/x7rcwdh6Cjm82Uy05NskHkyqof5/Ihv12qGyaaVUUHNCJVtZRLhmuND0MfCwgJXXHEFADt37uSee+7hM5/5DMlkkh//+McUCgVe+MIXctNNNwHwk5/8hAMHDpBKpTjvvPN4xzvewYMPPsjHP/5x7rzzTv7iL/6CF73oRfzwhz9ERPj0pz/Nn/7pn/Kxj33sjOu++93v5rd/+7d50YtexJEjR7j55ps5cOCAp++yEiMQhrYnGchya+BHdAWqFKsBkoE9wAb/rtdJoZyaDYpWoutz6Y2FGeqP0xvrkPuzEh+CPuxMTPfffz+PP/44X/rSlwCrYN8zzzxDV1cXV199NVu2WOHP559//pJw7N27lwceeACAsbEx3vCGN3D8+HGKxSI7d+4867r/9m//xpNPPrn0fHZ2lrm5ORKJhKfvsxwjEIa2Z0dkgZmBKKcCGxiqnmJHZEH7M1pZ/kIHr+Py6sfQ8bk0I/N7zWlR0IdSijvvvJObb775jNf3799PJBJZeh4IBJaeBwIBymUrR+e3fuu3eM973sMrX/lK9u/fz0c+8pGzrlGtVvnBD35ALBbz5TuA8UEYOoDEwCb2DW/guo1F9g1vIDGwSet83S5irapY2gz/gVc/ho7PRaffdtti00nRD26++WY++clPLjUH+vnPf67VIW55ifDPfvaztsfcdNNNZ0Qz+eEoNzsIw5rhevUc6ye+7zXEG4zk0l0le61Y6vZ7NcN/4BUtn4tHc1bb0IIoxne84x2Mjo5y5ZVXopRicHCQe++91/X5H/nIR3jd617Htm3beP7zn8+hQ4fOOuYv//Iveec738lll11GuVzmF37hF/jUpz7VzK+BqHpx2m3OVVddpR566KG1Hsb6xUMJkFaWjtYxjRw6leV7z04uicmLLhhk5wb3Zhudaz0+NsNff/fgUh7Er7/4PC4b0ncwt/I+funho0v5JI1GXjWTAwcOrHlbznbH7h6JyMNKqatWO9fsIAyN4TEapNUlIty29vQa5qqzKxhOdTOcijM+vcBwKt5QHkQr76NOPonh3MAIhKExPEaDtLJEhE5rT89hrhpmmNmFEgcn58mVquTLlYbam7a61EZHRXgZPGMEwtAYHqNBWlkiQncS9TIJ6kQVjaazVKow1B9jcrZwZmayS/NdM+5ju0Z4uUUphdSpebWe8epCMAJhaIxYP5nzXsFseoLe1GaSDTj9WrUabaUY6ZhhBuJdTMwucGxmgWDAeg5om++83MdWtxFtthhFo1GmpqYYGBgwIrECpRRTU1NEo9GGP8MIhKEhMrkS9z2To6ISBKdy3LZX3zzSSlppGnF7rVgkxA27N1m2KLGeAy2t4KvT4MgrumLkRkyGhoYYGxtjcnLSlzF3OtFolKGhoYbPNwJxrtGi5kKd1oegHc0oqXgXgz2RpQlzyfTVwgq+Wg2OPKIjRm7FJBwO22YZG5qDEYhziRY2F+qk/sSZXIn7H3mKcCFNKZLipit3eV65rnY9Ty08NSv4ehmvVnlyj+iIUactQM5VjECcS7TQNNFJ/Yln0icZGv86fdEgM1MVZkZSJOPbbI/1apPXPd/RHOUymcvreFPxLgIinJwrkIjWL0/uFR0x6qQFyLmMEYhziRY3F0rKPElJg6SA9i1FnmKOo5UKP1/oZ6B6ihRzjsd6Xbmmc0VOzRVQAqLwfeXbjJW22xwRr+j0yuikBci5jBGIc4lWNhfqoF7Z1fgAxQoEsscpRsNU4wOOx3pduS4UynzrqRNUqhAMwMt269WN0sXreHVyRMCbOSsZD/PiCwaXyoWvdr7JuVh7jECca7SqW14H9co+nA3zDV7App4sJ8rdbMiGucxhc6U7ia1kKldkU2+MjYkIJ+cKTOWKTfgGznhdaesITDPMb999dpKKUhxN57gt5m9IrcE7ppqroTGaYc5qQWcvAARywQRTke3kgom62c2ZXIlvPDnBQ6PTfOPJCe0Kq1bXNeHYzAKhgLSk65rbbnlO5774gkFGBrp58QWDrtuxVpQirSl+Xs83tB6zgzA0hldzVgtNVDrZzV4rrPbGwuwc7ObY9ALb+mMtaayjZfZZEQats6r3as4yjufOw3eBEJEg8BBwTCn1ChHZCXwBSAGPAG9WShVFJAJ8DngeMAW8QSk16vf4DGtEiyOuXBeZ81jS+vBUlqPpHIlomKPpHIensr6W8NZq4mMjyulcl2snt1dzlnE8dx6tMDG9G1jeKPVPgD9XSl0ITANvr73+dmBaKXUB8Oe14wztyuJkc3C/9ahrJmp1xJVLM8zibmMw0cUVQ336FVYFopU5BgpHiVbmfO+ZcHgqy0OjaQ5OzvPQaLp+E5/loqyqkEs3VKeqUXNWM843tBZfdxAiMgTcBnwUeI9YxVJuAN5UO+SzwEeATwKvqv0M8CXgEyIiqpMbVpzLeN0BtDLiCr3kNS8lrYfjJX6RH5CdL3FFNMxwfI/XoddlPl9mPLNAJBikUKkwny87H2wjysmY3qq+HTPSDf7ht4npL4DfARa7aA8AM0qpxb/iMWAxY2kbcBRAKVUWkUzt+FPLP1BE7gDuANixY4evgzfUoRk7gBZFXOn2UvYSXhnITREJwlxiC5HqKQK5KUgN1h2blwm3JxZiSzJGVyhAsVylJ1bnv7SDKLv9vq0u7GdYe3wTCBF5BXBSKfWwiFy/+LLNocrFe6dfUOou4C6wOso1YaiGRmjxDsALrWztmSZBIBjkougsM/kgaRJLq6OVNGPCHU51c+m2JJNzBQYTkdVNYjai7FakTPmL9YefO4gXAq8UkVuBKNCLtaPoE5FQbRcxBIzXjh8DtgNjIhICkkDax/EZvNKqnAuvaDqevazq+1IbeXDrL3KikKaUSrEntdHx2GZVUo2Hg6S6w8TDQe1zdUTKRCGtP3wTCKXUB4EPAtR2EO9TSv2qiPwj8FqsSKbbga/UTvlq7fkPau9/2/gfDM1AJ8zV66o+GQ9z05W7XAlMMyqppnNFKlXFYCJKtgGR0dkVmOZE64+1yIP4XeALIvJHwKPAZ2qvfwb4WxF5Fmvn8MY1GJvhHETH8dxKM0pZKXZu6LF2NIqGKql6FZlWdtszPozOoyUCoZTaD+yv/XwQuMbmmDzwulaMx9AkWtR7AryvPN1ObF7NKDoO8ZAIT0/MsVCqEAsHCTXQEc1rue5W5iYYH0bnYTKpDY3RjExolwLTypWn11pMOg7x6VyRqlL0REOUylWmc8XTPaldolMh1YlWFcUzPozOwwiEoTG85kFoCEwzVp5udyCLtZjm8mWenphbNST2LHQc4gLhUIDeaJjZfKmhpLpOyk7upLEaLIxAGBojnoJSHsZ/ApGEfh6EhsA0w+zjdgfiNSR2ONXNRZsSTM4VuGhToq5DXMd5Xo9OKovdSWM1GIEweKG0YE3sgQb+jDQS7byuPLXCST3WYgL3YafJeJib9mxu2JxlMPiNEQhDY6QPwfGfQLAL5o5bz7c1qdSGjW/Cy8pTJ9LH66pepwGP6Y9gaHeMQBgaozBHafoopUAX4WqRcMG5jacjdol2PpQBrxvps0KMkvEwr70kwWx6gt5UqqEGPIVSlZ8ey6za47lZUT0mt8DgF0YgDA0xp2Kk1QAVFSKoyqRUzLGkhBY+lAF3jPSxEyMg/NRXiBaKhE92wb7XaF/fbY9nHTFxwuQWGPzECIShIdKRbZzov5xUsMCpSgQi25ojED6UAXf0YdiI0dxCiYeem+S46meLTHLVjhMkhtwLhG6PZ7diUu96zSjXYTDYYQTC0BB9qY08uP2XGC+kKUVSXFyn5pAWPhUBtPVhxFOwkIbJp6FnEOIpnpxI89PjswSZ5RQQz4S5dsj9dXRNTDpiYkdIhMeOziwl2zVSrsNgcMIIhKEhdGoOadOqIoD5DEz8FIoLMD8B+QxjC13cV76WzeEcE6U42xa6uFbzY3VMTF4Tx5qRbKeD8XesL4xAGBqm42Pa04dAQrD1cpg6COlDbOnbC7F+jtOHhIQtfbHTx7vI/NbZFXjN2gaakmznFuPvWH8YgTCsGWu+Gk3thEDAEodAAFI72V6NsyMVI5Mvk4yG2N5fW427jK7S2RU0I8y1Wcl2bjC1lNYfRiDWMy0streSTK7E/Y88Rbjmw7jpyl2tn2z6R+CG37N2Eqmd0D9C+VSWq3cOnB0S6zK6SieprxkOZq8tUnVoRtRVu7Lmi5U2xQjEesWHfAMdZtInGRr/On3RIDNTFWZGUiTj21Y/sdn0j1j/ajiGxPoQXdWMfhDQWlOf16irdsSYzpwxArFeaUa+gYcdSIo5TqCYIEU3J0nRQKKdDzjuAFxGV+mU+/ZaqrvVNCPqqh0xpjNnjECsV7yuiD3uQBIDm7h0KEW2kKN7Q4r4wCbNL+AfjityF9FVOsX+mlGquxm4Na+cqyYmU4bcGSOno3pcAAAgAElEQVQQ7UQrfQJe8w287kBi/cT3vYZ4q75vq+6tRrG/dih/rWtemckVGZ9eYGt/zPGYTqMdfg/tihGIdmEtfAJe8g2aYZNvVb5DC++tTrlvaI7/wIuDVce88sR4hu/8fJJIOMizk/Ncd/4Grrtgg6extwsdH7LtE0Yg2gUfahD5Sqwfzn/p6Qigdh5rLk2uUGS2ayO9hZPWrsXH8bot9w3eo2e8Olh1zEaz+RIK6/vlSxUr78JwTmMEol3wIUrGVxam4cB9UJiDE0/Cvjf5Oul6mUgz0stT43Oo6gwSCLJrdy9Jn8apW+7brUO73vW8OljdRiZdsiXJ9v4YuVKV7f0xLtni1100tAtGINoFn2oQ+Ub6EIw/AtEkTD0Dwy/Q6weBXhtQL6vkdDXOgf4bSQXmSFcTbKzGfRMIHYfn4aksD42mCQcDlCpV7e51utezQ0fQtg/E+eAte5Yyv/0s6WFoD4xAtBOtssk3BbH+KXX6Zw10Jv10rkiwmGG4a4GxYox0boOWQIREeHCiykIpSixc5UbxL4pfx+E5ny8znlkgEgxSqFSYz5d9vZ4dupFJ2wfiRhjWEUYgDI2RGoGt+6A4BwMXWM810DGNpAI5RibuR1UrjASCpC65Hajj/F0RsdTqgnZu6YmF2JKM0RUKUCxX6Ykt+++oEXXl1cG6UKownSsSCrZX+pvJbl57jEAYGiPWb/kdGk2U0zCNJNUse7cmLCdz8SRxNQsM2h9sF7Ek0pYF7YZT3VwzkmKuUCIRCZ+OeGph1NXhqSxPHMsQDgY4OZvn8FS2rpmrVZO2yW5uD4xAGBrHg0lMyzQSTxGPdBFXaYh01Xfg20SDDad2OBe0s1mpH53KNWxn19kZOdZRamFEm46Zq5WTtslubg+MQBjWDNemER0Hvk00WDLmMBHbrNSP5iL8j389sNSA54O37NYSCV2nsWMjoxZFtPXEQgz2RKgCvYTONHOtoJWTtslubg+MQBg6A7vdip2d3kFMbCdim5X6E+NxDk1l6YmEmJjN88R4pr5ArBiDrtPY1mSjGdHmxezTH+siEgqQK1WJhAP0x5wn4lZO2ia7uT0wAmHoTOrZ6W3ExHYStVmpB6TA7EKRbL5CRVUI1It4chiD251RXZONS/OdV7NPWSmu2NHvqmBgUxocaWCym9ceIxDnGi2s57SmUSYadnrHSdRmpb65b4aLN/eyGLy7uS/alDHY0Yx+EF7NPql4FwERTs4VVg1zbUaDI0NnYQTiXKKF0S9NcVhqiNlZYqRhp687ia5YqQ+nunnh+RvOjiyyw6OvoBn9IJph9nGbSd0MQTN0Fr4JhIhEge8Akdp1vqSU+n0R2Ql8AUgBjwBvVkoVRSQCfA54HjAFvEEpNerX+M5JWhj94tlhqSFm9mLk3k6vFVKr06HNwVfgdmfVjH4QXm316VyRSlUxmIiSXWXSb1aDI0Pn4OcOogDcoJSaF5Ew8D0R+VfgPcCfK6W+ICKfAt4OfLL2OK2UukBE3gj8CfAGH8d37uFX9IvNSt9x0nW7K9AQM0cxcmmn99XhuWIMOjsrHfOOX+hM+p3W4MjgHd8EQimlgPna03DtnwJuAN5Ue/2zwEewBOJVtZ8BvgR8QkSk9jkGN/hRz6mOI/asSVfHxKUhZs0wo+g4jr30ytY1w3ht4em14J/OpN8uDY4MrcNXH4SIBIGHgQuA/wU8B8wopRazccaAxUbE24CjAEqpsohkgAHg1IrPvAO4A2DHjh1+Dr8zaXY9pzor/bMmXR0Tl4aYtTLk0WuvbJ0VeTNaeOp0sLNDZxdjQk/XH74KhFKqAlwhIn3APcBuu8Nqj3aLqLOWM0qpu4C7AK666iqzu/AbHbOVrolLQ8xaFfLotVe27orccwtPgWK5ymy+RKlcbaiMiM4uxoSeri9aEsWklJoRkf3A84E+EQnVdhFDwHjtsDFgOzAmIiEgCaRbMb51ixt/gY7ZqtNKltuQGNjEpZui5OePEN3Up90rW9cM49XE1B/rolJVTM/m6Y2G6ya62dGMXYxfNKOZktnteMPPKKZBoFQThxjwMizH8wPAa7EimW4HvlI75au15z+ovf9t43/wER1/gY7Zqk1LlutMFvFwkHh3F7joCLcSHTNMMybn6VyRYEDY2BttqFJtq0tatKoHiCn21xxcCYSI2NkK5pRS9XoObgE+W/NDBIAvKqX+WUSeBL4gIn8EPAp8pnb8Z4C/FZFnsXYOb3T7JQwN0M4tTpuc7Kc1WeTSEI7CwHm+35emTM6Cp0q1un4FL4UMdXuAeAmjNsX+moPbHcQjWOafaaw/wT7guIicBH5dKfXwyhOUUo8D+2xePwhcY/N6Hnid+6EbPNGuLU59SPbTmiw83hedSbAZpSuGU93OlWpd4tavcHQq56mQoVYPEI/iaYr9NQe3AvF14B6l1DcAROQm4BeBLwL/G7jWn+EZAH/KZ7Srv8CHnY2WMzjWD+e/1GqpmtqpfW2dMNdmlK7QSuzzyBPjGb1ChivQTVj0EjFlIq6ag1uBuEop9ZuLT5RS94vIf1dKvaeWAW3wCz/LZ7Sjv8CnnY1rZ/DCNDz3gHW/p0e177dumGsnmUECIu4LGdqgO2l7jZgyEVfecSsQaRH5XSzHMlgZztM1/0LVl5EZLNrYV+A5yiQ9yWx6gt7UZpKpWoc4H3Y2Ws5gj/dbN8zVqxmklc7YzX1R94UMHTCTdmfhViDeBPw+cC/W38b3aq8Fgdf7MzQD4K+vwIPpynOUSXqSp771WVS1wnggyK4bbz9TJJooglomJo/3WyfMtRlmkFYW0NMqZGg4J3AlEEqpU8BvObz9bPOGYzgLv3wFTqYrl6Lh1Twym55AVStEB3aQnzrCbHritED4gGsTk8f7nYyH+YWhMMcnjrFl8zbfV8shER47OrPkOPazgF4r/R2G9sBtmOtFwPuAkeXnKKVu8GdYhjPww1dgZ0oB1/4Or+aR3tRmxgNB8lNHkECQ3tRmr9/IkXSuSA/z7EosMFaMrS5mHu53Jj3Jsf/4e1S1wrGDQRLLd0Yrj/VYRwmsPIiqUvREQw3lQSyOo1V+AUNn4dbE9I/Ap4BPAxX/hmOwxY8oJjtTiob93XOUSWqQXTfefrYPwgdSgRznH/sKodI854d7SF3ym4A/5pHZ9ATFYplKYgvBueN1d0Ze6ygBZ+c9aOZBmIQyQz3cCkRZKfVJX0disMevKCYnU4qG/T0p8yQlDZIC9MeUTA36KgxL11k4yl4Oko/2EK2cJLJwFPDnuhIb4Mh0nvKpZwkFg+yMDdQ5+HSxMVV7rkt/rIuAwFy+THyVntJ2dFoklaG1uBWIr4nIf8EquFdYfFEpZWol+Y2fUUwrTSk69vcWdq/zjtQmX1V7bLTy0eqkq3Ee7nkJfTLHjEqwtxpnyOHY4VQ3F21KMDlX4KJNiYacvjo9pe1oSsFAwzmLW4G4vfb4/mWvKeC85g7HcBatznh2a39v4/DblSa5TGyIZ9R5hPLzlMObuDA2RNKvawvkgglKoaSr6qrxcJBUd5h4A3WfoDk9GhZKFaZzRUJB/4TT0Jm4jWLa6fdADA60a8Zzm4bf2u1s0tU4RzbewDAnOMImBqpx3wRCx+STzhXJFsqICNlCYyGqXst1HJ7K8syJORLRMM+cmOPwVFbfD2I4Z6krECJyg1Lq2yLyy3bvK6X+yZ9hGc5grTOe7SbsVoffusVmZ5MK9NI//l3G8yX6oz8ndcVF+OWk1jH5LBTK3P/EBIVKlUgwwMt265UWhyaU62iCH8Rw7rLaDuIlwLeB/2TzngKMQJzr1JuwWxV+q3MNm51NbvwYEzM5TsgAm/JT5KZPNOYcd7Gz0enQdiSdo1hVRMNBCuUqR9I59mzT29t4dTI3o9if4dylrkAopX6/9uNvKqUKy99zKAFuWGuaHRLro6/BNv7ewXTlOlbfZmczujCFCgS5pHueyWyQ0YUo2ulkGjsbt0l5ARGyhRL5YrCh2kbg3clskt8M9XDrpP4nEXnVYi9pEdkM3Ac8z7eRGfTxI7LIJ1+DY/y9zQSfyZW4/5GnCBfSlCIpbrpyl1ai29CWrdzfcxVThWOke7bx4i1b9QfsUih16j41o7YReHcym+Q3033OCbcCcS/wJRF5DVZfiK9iZVYb2gk/Vvs++RrqmkZWTPAz6ZOMHL2X/mCB6UqEmZG3kYxvc32t7fEC79oxRjq7QKp7jIF4AdDLNnYrlDor+uFUN/t29DM5V2AwEWnIvNNqJ/O5OJGaZEFn3EYx/bWIdGEJxQjwG0qp//BzYIYG8CuyyAdfg06pjlRxnOrsExSCPWyozJMqjgPuBYJcmoHuEANbLm5cODWEUmdF7zXMtZVO5mZMpO0oMCZZ0JnVopjes/wp1u7hMeD5IvJ8pdT/9HNwBk3aNSTWBp1SHYlIiK6+GPlAlGi1QiSi2Uq9WcLpQih1VvTpXJFKVTGYiJJtsBLrcKqb4VSc8ekFhlNxX53MXifSdl2pm+5zzqz2Py2x4vk9Dq8b2oV2DIl1wLXtOzVCZPtVRIpz0JWA1IjemDSF09MqV6BYrjKbL62aKKfTXMiJ2YUSByfnyZWq5MsVZhdKvk26XifSdl2pm+5zzqwWxfTfWjUQwzmAn3Wj9r3J287IpXB6XeX2x7qoVBXTs3l6o+G6iXI6zYWcGE1nCQQCXLo1wehUltF01rdqrl4n0nYu62Ec9faYct+G1XG7K2hl3SidceF+EvTagGc6VyQYEDb2Rlctv92MMhkjqW5CARidyhIKWM910BVErxOp674chrbAlPs21EdnV9DKulEa49IJk/Vs9hEoVaoUKlUCtedONMO0sX0gzntfvmup1Ibu7qGVHem0Wr8a2gJT7ttQn3q7gpUr+FY6yTV2KzPpkwyNf52+aJCZqQozIynHMFmvZp+QCOPTC+RKFeLhIKEGkt902T4Q1xaGRZrhB3GLcQZ3Hqbct6E+8RSU8jD+E4gkTu8KnFbwOk5yL1nfGruVFHOcQDFBim5OkmLO+ViNUhl2HEnnqAC9sfCq5TOaFdXjxaneDD+IW4wzuPMw5b4N7li5EPbqb1iYJvfol8kWinRHuojve43e+Rq7lcTAJi4dSpEt5OjekCI+UL8oXqQ0Syg/STCoX68pIMLsQgkRUIq65TOaEdXjVWSa4QfRwTiDOwtT7rsT8KPlqFtyaQhHYeC8M4XAo79hbuoEB8bSZKMb6Z48ye4dJ0gMOZiunHC7W4n1E9/3GuIuPnMmfZLhifuJhKAwBzPpQa2s7Z5IiJ5IiHKlSigYoKdOzkYzTC5efQhmVW+oh+uMIxG5FNgDLBWMUUp9zo9BGZax1p3bnITAo78hTYIKwmbSzCCkSVjJNT5934zqIa26SKmuur0gugrTTGTmSQc3kqqc5MLCNFpZ20BXOEAsHKRCfXON114O0BwfglnVG5xwG+b6+8D1WALxL8AtwPcAIxB+s9ad23xyPPelNvLg1l/kRCFNKZViT2qj9YbD9/ViZ8/kSvzjw0eZy5dJREO87nnbHT+jGOlnc7KH4WCOQqWHYkTv+/bEQmzvj9MVClAsV+mJOf8Xy+RKfOPJCebyZZ6emKs7Lida6UMwrD/c7iBeC1wOPKqUequIbMIKeTX4TatbjtrhlIOgs9JfYTZKxsPcdOUuV+W+vdrZD09leXxshkQ0zKFT81wzknIsf9GX2sijW15Id/YI2e4dXL4oXC4ZTnVzzUiKuUKJRCRct/TF4aksD42mCQcDlCrVuuNyotU+BEN70KqaVm4FIq+UqopIWUR6gZMYB3VraNf6Sjo7GwcxsTVt2Hzf9KmsN2euRkG7pMxzbeEHTGWmGAgdIyl7APf3XKe/wny+zHhmgUgwSKFSYT5fdn2d5dfzaqYydBatrGm1qkCIiACPi0gf8NfAw8A88KAvIzKczVrXV7JDZ2ejayZb8X29OnN1uqZNHDrAkZ9+j3m6yY4fQHZcx+Y912ldz61NvycWYksy5soc5YTnlqOGjqOVNa1W/YtUSikRuUIpNQN8SkS+DvQqpR6vd56IbMfyUWwGqsBdSqmP1zrR/QNW2Y5R4PVKqemaEH0cuBXIAW9RSj3S+Fc7x1nLyCbQ29l4NJN5jbTRWdUfnJwnX6xQClXJliscnJxns9bV4OhUzlVms445yol2LYBn8I9WJhy6XbL8UESuVkr9WCk16vKcMvBepdQjIpIAHhaRbwJvAb6llPpjEfkA8AHgd7Ec3xfW/l0LfLL2aFiJn5FNfghPE8xkXiNt3J6fTwzzk/J59FRyzKuNXJ4Y1rrO0akc/+NfD7BQqhALB/ngLbsdRaIZ7T5bnZ3cjv0c1hutDE12KxAvBX5DRA4DWSwrrlJKXeZ0glLqOHC89vOciBzAihd8FVZEFMBngf1YAvEq4HNKKYUlSH0isqX2OYbl+BXZpCM8uiLVjmYyG6KJAb6XuIlEJcNcMMm1iQGt858Yz3BoKktPJMTEbJ4nxjN1dxHNEL5W+SDatZ/DeqRVocluBeIWLxcRkRFgH/AjYNPipK+UOi4ii2Ei24Cjy04bq712hkCIyB3AHQA7duzwMqzOxa/IJh3hWevwW5/oiYXYMLiJrtAWIg34BQIiTGeLzOSKq2ZSg3tzlBOt9EEYc9b6w20m9eFGLyAiPcCXgf9XKTUrzv9h7N44K6hbKXUXcBfAVVddtT6Dvv2KbNIRnnYIv/UBr36BnkiIZDyMqiokIHUzqXXMUU60ctJu534OBn/QD5vQQETCWOLwd0qpf6q9fGLRdCQiW7BCZsHaMWxfdvoQMO7n+DoaJ5ONFx+CjvD4GX67hg74ZDzMC84b4Injs1yypVd7su2JhRjqi1EFArXnTjwxnuHodI5UdxdHp3OrmqPsaLUPwvRzWF/4JhC1qKTPAAdW9K7+Klbxvz+uPX5l2evvEpEvYDmnM8b/oEkznNc6vgI//AprXFrk6FSOu777HOUqfP/ZSd778l1ak3Z/rItIKECuVCUSDtTtKNcbC1OpKtJZqzd1b0x/5d9Kh2Uz+jkYJ3dn4ecO4oXAm4Gfishjtdc+hCUMXxSRtwNHgNfV3vsXrBDXZ7HCXN/q49jOTc4Fv0AuDcU56OqxHlv8HUbTWRaKVQZ7I0zOFrRbeJaV4uLNvbUwDuqWvtjeH2dbX5RMvkwyapXoaIRWOSy97lZ0Sp4Y2gPfBEIp9T2cc1ZvtDleAe/0azzrgk7zC9iZkgJBOP44VKsQCMDuV7Z0SAPxLiZmFzg2s0AwYD3XISTCoVPzS8Xz6jUMKivF1TsHOqaOkteIqWaUFjG0Fl99EIYW0wy/gF/2/5Wf62RKqlZgy+XQ1Q3FrPW8hcQiIa47f8OS4zhWx8lsh07xvGbVUfJqtnF7vteIqWaUFjG0FiMQ5xpe/AJ+2f/tPtfJHBZPWeYlVbUeW7wLColwPLPgagdgh05Humb4D7zmJuic7zViqhmlRfzC+EbsaZ/fkGHt8cuHYfe5PvWZcMLtBFBWiisGhRRzpEk0ZPbRifTx6j/wOmnrNBxqRk0sr6VF/MD4RpwxAmE4jV8+DLu+1vWEoMnRUTqr5FQgx+70t1DVCpsCQVKBIcD9RJbOWRFJg4ko2QY6vOniddIOifDY0Zklk1q9hkOtrInVSnTKwa83jEAYTqO5es+kJ5lNT9Cb2kwy5aJ/88oldYvKb+isspNqlr1bE8x2baS3eJK4mgXc96bWmXDBu2nD66Q9nStSVYqeaIhSucp0ruh7aZB2EYYlNMrBrzeMQHQCrUwcczlpZ9KTPPWtz6KqFcYDQXbdeLuzSDj1tW4RWqvseArKBbpmfgo9fdq7KJ0Jt1m1jTxNugLhUIDeaJjZfGnVyfFctNXrlINfbxiBaCZ+TOQL0/Do30NhzjLP7HtTW+Q2zKYnCJazxLuT5LIZZtMTzgKxxuG3OqvszEKZZ45nCJWylOeECxfKJGMaF1s5wdaZcHXs/36hMzmeq7b6djV9tQNGIJqFXxFA6UMw/ghEkzD1DAy/ALb5KBAuRa43HqM0+xTMVIgGgvTG68yisX44/6XWd0ntXBOBc7vKnk1PUJYuQluuoDx1pL7w2dAf6yIgMJcvE18lk1rXHOUHOpNjM2z17boDaUvTVxtgBKJZ+JbFLNY/pU7/7BcaIpeMCOGLryFLlG7yxCN1xrUwDc89YH3u9GjLy2csjcGN8KU2Mx4Ikp86ggSC9Kb02gXpZFLr2v+daIYfw9V5Hm31plx452EEoln4ZUZJjcDWfVbZiYELrOe6uDV96YhcPEW8p4+4qoJEm9tytNnUE74V9yaZGmTXjbfrOd+XERLh6Ym5pV1B3TwKTfu/Ha2cdL3a6nVDctt1t7GeMALRLPyqbhrrt/wOjX6ujulLR+Ra2HJ06Xs0eg+cBMrh3syqbkbVZkZUN0nNYU7nihTKVvZ3oVypuysYTnVzzaYAlewEwU2Dbd9y1KutXidYwOw22gMjEM3Er7BNL5+rs3rXFTm34/Iqnl4d9U4CZXNvjuYizj0aXIjUfL7M8UyeoAgVpeqWk0jKPLcGfkQ6sEAqECMpw4DevWl1uW8vtnqdYAHTnKg9MAJxrtMuBfy8iJxXR72TQNncmyeedejRoLMTkxWPDsxNneDA8WmmAhsYmD3F5VMnSAzp3aNWlvtuJa0WPoM9RiDOdeqt3t0W0FtzhEKxQLEwRZdUiDRirLcTKJt70xs7Zd+jweVOrCdmle12U2/oSD7G0ZkC8dA4R8uK/nyMS+p9B4cdTCsjcLz4BXTMRueq8HUaRiDWA3aTo04BvWbgwYeQkQTTmTyB8gLVUIx+SWj7BhxZcW+298fZmIgwnSsykIic7tHgcic2nOrm0m1JJucKDCYidf0KlWiSRxPXszmUZaLczSXROt+qDcTbq19A12xkQk/XHiMQ6xWdAnpOuJ30PU5us7kFJpOX0p3oIzs3QzC3oC0QY8fGOT5xjC2btzG0bavjcdO5ItFwkB0D3WeGnmr4UeLhIKnuMPFwsO6YhlPdXDy8nblCiYtXK1631pFgePcLmJ7WnYcRiPWKnRjoOJN1HMceJ7fe1GbGwz3M5gpIuEc7N2Hs2Dj/fs9dlCsVfh4M8pJfusNZJOqFnrrwo+i05dSKCvLRl+TWbNQMv4Dpad1ZGIFYTitrHq01TmLg1pms4zj2OLl5zU04ePgw6VyebNdGunMnOXj4sKNAeI3198256mMZdLflM7z6BZrR09rQWoxALNIGNt6W4yksVyPDuwmTWzI1qC0Mi5SjKXLFKqHSBDllPXe8jsdYf626T7q1jXR+Xy4XO7rlM7z4BYyJqfMwArFIG9h42wK3uyjdDG8/ckRcjnXDxk0c3XIzPdUM84Ekt2zcVPdjvTpHZxdKjE3nCImsWtvIlx7NmiG5rSx1bUxMnYURiEXaJV9gLdGZWLxmeC9erwUZ4sOpbvZdtLMlncyOTuX42DefWmpZ+t6X73LMpG5Wj+azfAgai53hVDcXbUowOVfgok0JX++NMTF1HkYgFnEyg6wnv4TuLmot+19rjDUZD/PaSxI1H0bK10lpNJ1loVhlsDfC5GyB0XTWUSCa0aPZ3kzVnosdk/zWeRiBWM7KCa9d/BKtEimnicWP63s16elMggvThJ/6CtFCkfDJLtj3Gt/u40C8i2PTOQ6emicSDDBQZxJsRo9mWx/Cdvc+n8NTWZ45MUciGuaZE3Mcnsr61m7TJL91HkYg6tEOfolWipTdLsqvhkWaq9yz2ptqOL7npk5wYCxNNrqR7smT7N6xSkkLD4JYrio2JaNUgUDtuRNNaVTj5ENwu7sTKJarzOZLlMpV3zvKmeS3zsIIRD3aYau+1iLlV8MijQnesb2py0kwTYJyYYFU4SmyxEiTIOF0sFdBFohHQq5LeHudML2G5eo0ODpXO8oZnDECUQ+/Snjr0EqRstst1Atn9Wp6cjnBz6YnUNUK0YEd5Bvo8hYUYWpmhkJ5hvlQHzvq9WjIpckVisx2baS3cJK4piC3ur9xMh7mpj2bGU1nGUl1a0/YOg2OmtFRztBZGIFYDb9KeOtcv1UiZbdbSO2EwV0wf9J6XAxnbaHpy2uXN5k5xK7AMaqJXgKFY8jMIXBIlMtIL0+Nz6GqM0ggyK7dvVplPbxO2LpkciW+8eQEc/kyT0/Maa/qQyIcOjW/FHW1WoOjVobEGtYeIxCdQMtEymG3EI5Bd8p6XKQZpi+XOxCvmdS90TDFYICiQCholdFwGkO6GudA/42kAnOkqwk2VuNaApHJlfjus5NUlOJoOsdtsWUF7Xxw9ntd1ZeVYmsyRq5UIR4O1t1BtHp3ZFh7jEAYTpMagf4dkDlmPaZGrAktHIWB884UAq+mL80diJdM6sTWCwnuvo78/AzRnj7iWy90HENIIjx2SlGu9hAKKG6st6K2IZ0rMp8v0x0NMZ8vn471d/i+nttqelzVLxTKfP+5U1SqEAzAKy5zLmTYFKe6oaMwAmE4TT4DU89CcQHKC9ZzJyGI9cP5L7XMUqmd+iviVjrfY/3Er3mz5U9Yvnq3GUNZbeaKQSHFnOXcrrOiBs7aFYRE+PFomtl8id5omFsu3eJ4rYzq8ez09bqqn8oV2dQbY2Miwsm5AlO5Yt3jTRTS+sIIhOE06UMgIdh6OUwdtJ6f/1LnBMLnHrAmvelRfR+EX6XFnbAz09mMIbWQY3f6W6hqhU2BIKnAEOAw6drsCo5OVzg2s0AwIMzlyxydzlmJcjbXOnyqjnnIrfnNo89jJNVNvCvAzEKJeFeAkVUExvOOx9BR+CYQIvI3wCuAk0qpS2uvpYB/AEaAUeD1SqlpERHg48CtQA54i1LqEb/GZnAgtRMCAUscAgHrOdhPrl53ALF+Mue94rRfwcd+EvXGsFL8krnn2NtfIl8sE+0KEVezgINpy+YezObj9FTm2RTMcqLSbYW6OuTbmJwAABi5SURBVFwLmbHPQdD4vnV9Hi7YPhDnvS/ftSQwTlnfi9fy0jDI0Hn4uYO4G/gE8Lllr30A+JZS6o9F5AO1578L3AJcWPt3LfDJ2qM/rKfyGTr0j8ANv3fabNQ/4nxsPAWlPIz/xAqJ1fRBZHIl7nsmR0UlCE7luG1vyXmy8dMctVL8ilnCz34DKZcJhUJwyW3O59rsCkbis7yw+F3KC2UuCIYYiS9rIrriWv2xLipVxfRsnt5o+HQOgsb39drEByyRqCcMzbyWobMI+PXBSqnvAOkVL78K+Gzt588Cr172+ueUxQ+BPhHZ4svAFldnB/dbjwvTvlymY4kmoW+H9bgapQVr8iotaF9m+WRTUYp0Pdt3PEWupJgcP0SupHzNBZlPTzBe7edwbDfj1X7m0xPOBy/uCs67/vQqP5emNxqg3LOF3mjAuj8OTOeKBAPCxt4owYAwvXgPNMxvraxvZGoprT9a7YPYpJQ6DqCUOi4iG2uvbwOOLjturPba8ZUfICJ3AHcA7NixQ38Ea52Z3M7omHLSh2DygCUkkwes5xoZ1jqTTUb18K9zFxPOHKKU3Mktqqd5PalXMB0doipBUqUTZKWL6egQPfVOWLEryNBLYWGBzdWnWQh0k6HX+Vyn7nUauS+trG9kaimtP9rFSW0XnGcbPqKUugu4C+Cqq65aJcTEhnYon9EM1ryAnkbDIBt0Jpux8XHUwW8TCgnFqUOMnb+D5AXDWtdzS2//IAd7L2Yhf4pCdAPn9euF1mZLZXLFCqgKOamQLTmX8K4bgbTWCZoOmCim9UWrBeKEiGyp7R62ACdrr48B25cdNwSM+zKCdiif4ZVmOW1XioyOeKZG7DOsNXCcbFaMK1hIU61UmAhtpKdykmAhDegJhNvom6Sa5eKRHcx2XUVv8WR9J7UNpcwkpUCY0eCF9FcmKWUmgYvtr9WEvAJTH8ngJ60WiK8CtwN/XHv8yrLX3yUiX8ByTmcWTVG+0KarM0dWTuTNymK2E5m1Fk+bcfX0bQICRHMnIBisPXePVvRNPEU80kVcpSHSpb3D7BvcQiwSpocZKqEwfYP+uNIWaUZ9JBO6anDCzzDXzwPXAxtEZAz4fSxh+KKIvB04Aryudvi/YIW4PosV5vpWv8bVcdgV0GuGmcxJZNyKp0cfhM64SpHNlHa/ail5rRSpeSBcmtnSuSLBYobhrgXGijHSuQ3OE6FHkdy1cwc/uuSXmJ46Qe/AJnbtdPaTNWX1r1mu224MJnTV4IRvAqGU+hWHt260OVYB7/RrLB2NbbntK72v9D2LjDcfRN1xrQifTakuunpSZFQ/XYsObQ0zWyqQY+vYN8iXSmwNh0ld8hYck9/A8w6zu3cD1UiS7kj9ifbwVJZnDh9lUyjLM+VuDjew+ncMlXWJCV011KNdnNQGRxwmYp1JzG6l7bXFamoEtu6D4hwMXNCQD6Iuy/QmGQ9z24Xx00l18TBMuTezBXJTBKgwH9nIhuopArkpaLCu02ro9F0OFjPsnXmAcKDKxmqAYHEE0BOI6VyRqlLEu0JUlWI6V3SV07BIKt5FoVTlp8cyJKIhE7pqOAMjEO2OV2dwvZW2lxarsX7YfVvjtZicsCsOCCQP/jNJVYWZAMRerbUDSpMgEAxyUXSWmXywfsMgj+iE76aYI0iV42qATUyRYk77evP5MpPzBSLBIIVKhfm8c9SUEwKIKFd7QOOvaA9a9XswAtEJ2JXbdouOQ1vnWN1aTG53JnYTv924Bs53bWbrS23kwa2/yIlCmlIqxZ7URsdjwaa9qQY64bvFSD/DfUH2qONkJUYxoi+yPbEQW5IxukIBiuUqPTG9/9LpXJFsoYyIkC2U6+54jL+iPWjl78EIRLvjVG7bLTq+Bp1jdcVEZ2diN/HblfVwaWZLxsPcdOUuV5O2Y3tTDdyG7/bFu5gMBVAlRVcoQF8D5p3hVDfXjKSYK5RIRMLa1VwXCmXuf2KCQqVKJBjgZbudI8Ra7a8wuxV7Wvl7MALR7niteqoTlaNzrNO47HYKumG5dhP/YlmPQGN/sm4TvLy2NwWHic1GJJNqlvM2beA4F7KDaZKaOReL38tLLsWRdI4q0BfvIlsocySdY882+zz1VvorzG7FmVaWPDEC0e7oTNpOK3Udh7bbY+36QThd32NhP99Cam3w2t7UcWKz6wchvTxzKouqzjIfCLJL9NqbNoPeWJhQ0Oo6FAoKvbH6k/BCqcJ0rmid4yMmusqZVpY8MQLRCbidtFvYBtTWB7Ha9RueU3wKqbXBa3tTx45yNjuudLaLI7E9bCs8w7HIhdrtTcH7SvuSrUlu2rOZyfk8gz1RLtnqPILDU1meOTFHIhrmmRNzHJ7KaoflusVEV9WnVSVPjECcS7SyDaidGDhdP5cGKtCzEYrz+sLld0jtChzbm7oQz5AITx7PUK5CKMDpjnI2O8Gu4wcYefrTUK0yEvh3uvbshA27tcbqdaWdjId51eXb3DUc8tjeFPT8CjrRVQZ/MAJxLuG1VEYubU3CXT3W4+JEbjcx2omB0/UDQTj+OFSrViOi3a/U/15+hNTq4FI8y0qxc0OPNbspzmxZumInGJg5RCISJBMbJrkwRmDmEKAnEF5X2joNh7y2N9XZ7ejkkxj8wwjEuYaXLGC7ibyeX8NODOyuX63AlsuhqxuKWet5PVYKklNIbSsbP7k034VEOHRqfmkHERLn9W+1byeZhQrV3EEyEiTRt7OhoXnxC+jsQLw6xHWuZXpPtAdGIDqBVk2EdhN5vYnRrRjFU0AA5k6u7qS2EyS7MYA/FW1xMIO4NN+VlWLP1iTdXSGyxfKZO4gVTIW38OW+2xksHWMyvI3/HN7CNs3he/UL6O5AvNi+dSZ903uiPTAC0e741Y/ZjnjKMi+pqvW4OAk2q3+GmwWunZnLbfKcbrkQm3ubUT32ZhCX5rtUvIueSIiKUvRE6k+48/kyTxVSHAoOUig0lgWNQLZQZq5QttpDNmCwb5WtX3fSN70n1h4jEO1OKzvgOU2CXgsD5tJQmgfEeqzn27AzczmNyykPw4OjPa26nM0gLnZMOpOg1yxosExY49ML5EoV4uFgXZOWHe1s6zeJcmuPEYh2p9Ud8OwmQaeJ0e1KvZiFp79+etK+6BbnidzJX7FyDE6i4SSoLh3tKaVp+/Zg/vOaBQ1WolsFK5+hUK7WTXSzo12T30yiXHtgBKLdaXUTHw/mGeeV+hT0boPEJpg7YT3v6rafyB3MXK5Xk3aCquFoT4J7M4iOicqGZnSUC4iQLZTIF4NUVIWA5g4C2jP5zSTKtQdGIDoBj/0JXOM1D8Lp2NROq55Ubtp6TO20sqLtdkY2GdqZXIn7H3mKcCFNKZLipit3kZR599FVU8+RKxSZ7dpIb+Ek8UYc7S7vQV0TlQ1e7eyb+6LsGxR6qtPMB5Js7otqna/r5PZi9tGqdGuimNoCIxCG0+hM+jqmr/4RuOH3Tk/6/SPW6079KFaEtM6kcwyNf52+aJCZqQozIymSsbzr6KqM9PLU+ByqOoMEguzavaykxYodU13Those3romKq929uF4iTfGH6FQLhIJdTEcv0zvAzSS37yafXT8M8l4mBdfMOgugc/gG0YgDKfRmfR1TV/9I6eFoR42IpWixAkUE6To5qTVNyG+yXmsKybydDXOgf4bSQXmSFcTp0ta2OyY0jmHHYBd61ePJqpmtBxNqln2bO7hOP1saaDgn07yWzPMPm53TDoJfAb/MAKxVrQyycstupO+F/NMvcJ+Kyb+RBwuHUqRLeTo3pAiPrDJvligw+eGJMJjpxTlag+hgOLGRTu9nRjFd9jvAGxbv9oXQnQ7CR6eyvLQaJpwMECpUuWaBlqOZqSXJybmKZamSYfDXHa5XsE/HT9IK80+xgfRHhiBWAtamdugS6v8HU7mLIeJPz18K8cnjrFl8zbi9bKrbfIoymqzffKajRglYzbtTQE/CgbO58uMZxY8dYM7nA1zT+Eq+phjppAgmQ1zmWagm1tBa2XymvFBtAdGINaCVuY2tCtOJcAXpuHAfZYp58STsO9NHM1F+Nh3JyhXg4SemeC9L+9jOw730CaPIhVxSF6z2zEtTJ/d3jTW70vBwGbkQcznyzw7HyIS3NCwyLRjvoHJpG4PjECsBTrNdpxoRxNVI6xciNuYckZz2ylXYWSgm9GpLKPpLNuHHO5htQIbLrQ8rmI9rzvZrNwx5dL2EU+xfsvv0MR7Ppzq5tJtSSbnCgwmIg3lQXgVGd3cBK8+Ex1MJvXaYwRiLXBYubo2O+maqNpRTBxbqZ5tyvn/2zv/GLmqKo5/zuzM/pjd7rKzbaHQpW2wgFRDgYLYiMQqKmCsP4KgJEAgIRAlQMSkVWJM9I8KaqIRQ6qgEFGQSLQmKCCKmAgFSqBQCrTQli5baLs/uz+6M7t7/OPd7Q7T96bzOjsz723PJ3mZ++67b+Z+92bfeffce89dnGkmmYCdPcMkE7A40wxN6eDIsfu3TfcgEnVAiMFRaWXb7r0kc2/TlWphaf6Mpwq439KpOjLNKdKpuqO6f1GmmaXzW+juG2XR/JbQRiaMr39XzzCbu/qZ05hix/6hoxozMeKFGYha4fPmWrLbqVL7QVeToF6UjyunsynNty86/dCUx86OtFc2ROTYUt0o/SNZEuMHaWeYnvEk/SNZ2or59EsN9udTdibCXAyO5nh73xAjuUkOjk8wOJoL9R2hfP1V3g/CqD1mIKJCmCmmYcoG7fFQa4qFC/dx5XR2pKcNQzF8VmKHcaNkst1wYCsTkqJDu73zoBirYVZS+5TNpFvKHojd2TvMxCQsbG9i3+CY534r5e/kCOPrn4n9IKrpojLKxwxEVAgzxTRM2XI364FAF1XZb4NBLptyXDk+f5ve/cPBbpQCbXMYoZFeclJPSrOkGAn+rYCV1L5bjvqUbetoD5gxVTod6XreGxzl3f5R6hLeeaUoNzSIuajihxmIKBHmwVhq2bCb9RQS4KLyDX8RlbfBgr9NJl1P4/gBhvbsorEhQybtegR+2hpaSWVOJpVIwuQ4NLQG/45PTy45ErDlaECMKN8ZU2GkNiRZecpcRnMTNKXqaGqo3CA1lDlwHNJFZe6o2mMGYrYTtMdDqQSMd/T37j08/EU65HY3VRo8b5MhLk1sZDiRpTlRT1oWAu3+2jKLGelYRnbgPerbTyBdbDqrz5qN8eFh/zUXYWNElUhShF09I4cMxNGE+67WgrQwLiqL5hoNzEDMdsqNBhsw3pHhAPsnRxk42ET95KgX/iIM1Rw8H+klnRLSHUsOjxxboG1gdJxt+7Mkc/WM57IsHR2nramIhoLFepl0S/CGQT4xogJnTJVI30iWSVVaGpPkxifpG8mGGoOo5oK0MC6qaq+krmZvJU49IzMQxwIz7NMHmNPcyEdkF9lsjvpUiobmcFFEK7pYsG/nBwMDBg3q+2gbfPcVxqWe5ILljPe8w2Dve7RlAmIbBY0rlDjo2z+SJZubJImQzU1+cMZUqb0rgVQyQWtjisGDudAzi6odFK9UF1VU96mI02/NBJEyECLyeeDnQB3wG1VdV+MqGRA4nbShczkNRzu2UWwmVjmup76d8K8fTQ/Kr7rdMxJBvagCba2ZE+hO1HGw5x0kUUdr5oTiGnxWg5f6EKwf66NraILeuoVkJvayZKwPOClU72omZhZFNShe/0iW7r5RTmwP6sLNDNXsrcQtxlRkDISI1AF3ARcBXcDzIrJBVV+rbc2OMUp9OJc7thHk+irX9dS7A3Kj05sT9e7wDESJvai2zDxO//TV0zOLgnoP+RxlWKZsQzutTY00TfSRqm8k2+DqF6J3Ve7Moqg+sLZ0D/D0m/toSNWxfd8QK0+Zy8oPza3Ib1XTzRa3GFORMRDAecB2VX0bQEQeBFYDZiCqRZiHc7ljG1PfUXhfua6ndId330CX65l0hK5WW2ZeaYYhcDV4aQxLC/cPnkXLxABDdW3cLi1OQ7htZsuZWRTVB9bgwRyKt9L8YG7Cc59ViGrGfYpbjKkoGYiTgN15513AxwoLicj1wPUAJ598cnVqdqwQ9uFciciv5e7BXd8Mp13sheoQ8c4rRZl17RnJ0nLcPObPWQgHxugZyXoXZsL4lkhUH1jLFrTR2d7ESG6SzvYmli0IO3wfjmrGfYpTjKkoGQi/jroelqG6HlgPsGLFisOuG2VQ7sN5JpiJWVfN86Z7QZXUUGZdF2eaSdcn6B/Nka5PeDGm8r+7Siveo/jA6uxIs/biMw4Pr2JUlSgZiC6gM+98IdBdo7ocm1TxzfWI9ZjhWVcVo4y6dnYExJgygBDhVYyKESUD8TywVESWAO8CVwDfqG2VjkGq+OZaMWKkwR6CRpSJjIFQ1XER+RbwGN4013tVdUuNq2UYhnHMEhkDAaCqjwKP1roehmEYBiRqXQHDMAwjmpiBMAzDMHwxA2EYhmH4YgbCMAzD8EVU47vWTET2AbtqXY8SmAvsr3UlKsRs1TZbdcHs1Wa6SmeRqh4xnkysDURcEJEXVHVFretRCWarttmqC2avNtM185iLyTAMw/DFDIRhGIbhixmI6rC+1hWoILNV22zVBbNXm+maYWwMwjAMw/DFehCGYRiGL2YgDMMwDF/MQMwAInKviOwVkVcL8m8SkTdEZIuI3JGXv1ZEtrtrn6t+jUvDT5eIPCQiL7ljp4i8lHctFrogUNtyEXnWaXtBRM5z+SIiv3DaNovI2bWreXECdJ0pIs+IyCsi8jcRac27Fos2E5FOEfm3iGx1/083u/yMiDwhItvcZ7vLj0WbFdF1mTufFJEVBfdUr81U1Y4yD+CTwNnAq3l5nwL+CTS48/nu8wzgZaABWAK8BdTVWkOpugqu/xT4ftx0FWmzx4GLXfoS4Km89N/xdj08H9hY6/qH1PU8cKFLXwv8MG5tBiwAznbpOcCbrv53AGtc/hrgx3FqsyK6PgycBjwFrMgrX9U2sx7EDKCqTwO9Bdk3AutUdcyV2evyVwMPquqYqu4AtgPnVa2yIQjQBXhvaMDXgD+6rNjogkBtCky9XbcxvaPhauB+9XgWOE5EFlSnpuEI0HUa8LRLPwF81aVj02aqukdVX3TpA8BWvH3sVwP3uWL3AV9y6Vi0WZAuVd2qqm/43FLVNjMDUTlOBS4QkY0i8h8ROdflnwTszivX5fLixgXA+6q6zZ3PBl23AHeKyG7gJ8Balx93ba8CX3Tpy5je2jeWukRkMXAWsBE4XlX3gPewBea7YrHTVqAriKrqMgNROZJAO1739jvAn9xbt/iUjeNc468z3XuA2aHrRuBWVe0EbgXucflx13Yt8E0R2YTnxsi6/NjpEpEW4M/ALao6WKyoT15ktUVVlxmIytEFPOK6uM8Bk3hBt7qYfoMDWMi0KyMWiEgS+ArwUF527HUBVwOPuPTDTHfdY61NVV9X1c+q6jl4Rv0tdylWukQkhfcQfUBVp9rp/SnXkfuccuXGRluAriCqqssMROX4C7AKQEROBerxIjJuAK4QkQYRWQIsBZ6rWS2Pjs8Ar6tqV17ebNDVDVzo0quAKffZBuAqNzPmfGBgyq0RB0RkvvtMALcDd7tLsWkz1/u+B9iqqj/Lu7QBz7DjPv+alx/5NiuiK4jqtlmtR/Fnw4H3VrYHyOFZ+OvwDMLv8fy/LwKr8sp/D+8t7g3crJkoHn66XP7vgBt8ysdCV5E2+wSwCW+WyEbgHFdWgLuctlfIm1UStSNA1814s2PeBNbhIijEqc1c2yiwGXjJHZcAHcCTeMb8SSATpzYrouvLrv3GgPeBx2rRZhZqwzAMw/DFXEyGYRiGL2YgDMMwDF/MQBiGYRi+mIEwDMMwfDEDYRiGYfhiBsIwAhCRxYUReo9Q/gYRueoIZa4RkV8GXPtu2DoaRiUxA2EYM4Sq3q2q95fxFWYgjEhhBsIwilMnIr92sfkfF5EmETlFRP4hIptE5L8icjqAiPxARG5z6XPdPgTPiMidBT2RE93928TtEyIi64AmtxfFA9WXaRiHYwbCMIqzFLhLVZcB/XihstcDN6kX2+g24Fc+9/0Wb7X5x4GJgmvLgcuBjwKXi0inqq4BRlV1uapeWSEthhGKZK0rYBgRZ4eqTu2atwlYDKwEHvbC6ADe5i2HEJHjgDmq+j+X9QfgC3lFnlTVAVf2NWARHwzhbBiRwAyEYRRnLC89ARwP9Kvq8iL3+IVkLvad9n9oRBJzMRlGOAaBHSJyGRza+/jM/AKq2gcccFFEAa4o8btzLvSzYUQCMxCGEZ4rgetE5GVgC942kIVcB6wXkWfwehQDJXzvemCzDVIbUcGiuRpGBRCRFlUdcuk1wAJVvbnG1TKMUJjv0zAqw6Uishbvf2wXcE1tq2MY4bEehGEYhuGLjUEYhmEYvpiBMAzDMHwxA2EYhmH4YgbCMAzD8MUMhGEYhuHL/wG0YMlVfhTBDAAAAABJRU5ErkJggg==\n",
+      "image/png": "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\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f533f0afc18>"
+       "<matplotlib.figure.Figure at 0x11135ef28>"
       ]
      },
      "metadata": {},
@@ -468,12 +468,12 @@
     }
    ],
    "source": [
-    "plot(\"height\", \"ranking\", data=data_male, linestyle='', marker='o', markersize=3, alpha=0.3, label=\"Male\")\n",
-    "plot(\"height\", \"ranking\", data=data_female_clean, linestyle='', marker='o', markersize=3, alpha=0.3, label=\"Female\")\n",
-    "xlabel(\"height\")\n",
-    "ylabel(\"ranking\")\n",
+    "plot(\"ranking\", \"height\", data=data_male, linestyle='', marker='o', markersize=3, alpha=0.3, label=\"Male\")\n",
+    "plot(\"ranking\", \"height\", data=data_female_clean, linestyle='', marker='o', markersize=3, alpha=0.3, label=\"Female\")\n",
+    "xlabel(\"ranking\")\n",
+    "ylabel(\"height\")\n",
     "legend()\n",
-    "plt.title(\"Scatter plots of Ranking vs Height\")"
+    "plt.title(\"Scatter plots of Height vs Ranking\")"
    ]
   },
   {
@@ -555,7 +555,7 @@
      "data": {
       "image/png": "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\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f5347ac1cf8>"
+       "<matplotlib.figure.Figure at 0x1a22467cc0>"
       ]
      },
      "metadata": {},
@@ -599,7 +599,7 @@
      "data": {
       "image/png": "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\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f533cb5cac8>"
+       "<matplotlib.figure.Figure at 0x1a228f30f0>"
       ]
      },
      "metadata": {},
@@ -743,30 +743,289 @@
    "metadata": {},
    "outputs": [
     {
-     "ename": "NameError",
-     "evalue": "name 'dataM' is not defined",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m--------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mNameError\u001b[0m                                Traceback (most recent call last)",
-      "\u001b[0;32m<ipython-input-21-284410c792e4>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mgroups\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m4\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mincr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mmaxR\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mminR\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m/\u001b[0m\u001b[0mgroups\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 6\u001b[0;31m \u001b[0mobservations\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m \u001b[0mdataM\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"hand\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"L\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m>=\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m<\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgroups\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m]\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m[\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m \u001b[0mdataM\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"hand\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"R\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m>=\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m<\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgroups\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      7\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[0mlh_prod\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfloat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"hand\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"L\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m/\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m<ipython-input-21-284410c792e4>\u001b[0m in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mgroups\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m4\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mincr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mmaxR\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mminR\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m/\u001b[0m\u001b[0mgroups\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 6\u001b[0;31m \u001b[0mobservations\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m \u001b[0mdataM\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"hand\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"L\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m>=\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m<\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgroups\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m]\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m[\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m \u001b[0mdataM\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"hand\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"R\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m>=\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m<\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgroups\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      7\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[0mlh_prod\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfloat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"hand\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"L\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m/\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;31mNameError\u001b[0m: name 'dataM' is not defined"
-     ]
+     "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>M: 168 - 188</th>\n",
+       "      <th>M: 189 - 210</th>\n",
+       "      <th>F: 155 - 171</th>\n",
+       "      <th>F: 172 - 189</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>1 - 99</th>\n",
+       "      <td>67</td>\n",
+       "      <td>32</td>\n",
+       "      <td>38</td>\n",
+       "      <td>60</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>100 - 199</th>\n",
+       "      <td>69</td>\n",
+       "      <td>30</td>\n",
+       "      <td>31</td>\n",
+       "      <td>32</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>200 - 299</th>\n",
+       "      <td>75</td>\n",
+       "      <td>17</td>\n",
+       "      <td>18</td>\n",
+       "      <td>22</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>300 - 399</th>\n",
+       "      <td>61</td>\n",
+       "      <td>21</td>\n",
+       "      <td>11</td>\n",
+       "      <td>17</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>400 - 499</th>\n",
+       "      <td>59</td>\n",
+       "      <td>22</td>\n",
+       "      <td>7</td>\n",
+       "      <td>7</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "           M: 168 - 188  M: 189 - 210  F: 155 - 171  F: 172 - 189\n",
+       "1 - 99               67            32            38            60\n",
+       "100 - 199            69            30            31            32\n",
+       "200 - 299            75            17            18            22\n",
+       "300 - 399            61            21            11            17\n",
+       "400 - 499            59            22             7             7"
+      ]
+     },
+     "execution_count": 21,
+     "metadata": {},
+     "output_type": "execute_result"
     }
    ],
    "source": [
     "# Calculate observations\n",
     "minR = data[\"ranking\"].min()\n",
     "maxR = data[\"ranking\"].max()\n",
-    "groups = 4\n",
-    "incr = (maxR - minR)/groups\n",
-    "observations = [ len( dataM.loc[np.logical_and(np.logical_and(data[\"hand\"] == \"L\", data[\"ranking\"] >= i*incr), data[\"ranking\"] < (i+1)*incr)] ) for i in range(groups) ] + [ len( dataM.loc[np.logical_and(np.logical_and(data[\"hand\"] == \"R\", data[\"ranking\"] >= i*incr), data[\"ranking\"] < (i+1)*incr)] ) for i in range(groups) ] \n",
+    "minH = {}\n",
+    "maxH = {}\n",
+    "minH['M'] = data_male[\"height\"].min()\n",
+    "maxH['M'] = data_male[\"height\"].max()\n",
+    "minH['F'] = data_female_clean[\"height\"].min()\n",
+    "maxH['F'] = data_female_clean[\"height\"].max()\n",
+    "rank_groups = 5\n",
+    "height_groups = 2\n",
+    "genders = [\"M\", \"F\"]\n",
+    "incr_rank = (maxR - minR)/rank_groups\n",
+    "incr_height = {'M': (maxH['M'] - minH['M'])/height_groups, 'F': (maxH['F'] - minH['F'])/height_groups }\n",
     "\n",
-    "lh_prod = float(len(data.loc[data[\"hand\"] == \"L\"])) / len(data)\n",
-    "group_nrs = [ len(dataM.loc[np.logical_and(data[\"ranking\"] >= i*incr, data[\"ranking\"] < (i+1)*incr)]) for i in range(groups)]\n",
-    "expected = [ lh_prod*group_nrs[i] for i in range(groups) ] + [ (1-lh_prod)*group_nrs[i] for i in range(groups) ]\n",
-    "stats.chisquare(observations, expected)\n"
+    "obs_data = []\n",
+    "for i in range(rank_groups):\n",
+    "    row = []\n",
+    "    ranklb = minR+i*incr_rank\n",
+    "    rankub = minR+(i+1)*incr_rank\n",
+    "    for g in genders:\n",
+    "        if g == 'M':\n",
+    "            cur = data_male\n",
+    "        else:\n",
+    "            cur = data_female_clean\n",
+    "        for j in range(height_groups):\n",
+    "            heightlb = minH[g]+j*incr_height[g]\n",
+    "            heightub = minH[g]+(j+1)*incr_height[g]\n",
+    "            row.append(\n",
+    "                len( cur.loc[np.logical_and(\n",
+    "                    np.logical_and(cur[\"ranking\"] >= ranklb, cur[\"ranking\"] < rankub),\n",
+    "                    np.logical_and(cur[\"height\"] >= heightlb, cur[\"height\"] < heightub)\n",
+    "                )] )\n",
+    "            )\n",
+    "            \n",
+    "    obs_data.append(row)\n",
+    "    \n",
+    "    \n",
+    "observations = pd.DataFrame(\n",
+    "    data = obs_data,\n",
+    "    columns=[ \"M: %d - %d\" % (minH['M']+j*incr_height['M'], minH['M']+(j+1)*incr_height['M']-1) for j in range(height_groups)]\n",
+    "    + [ \"F: %d - %d\" % (minH['F']+j*incr_height['F'], minH['F']+(j+1)*incr_height['F']-1) for j in range(height_groups)],\n",
+    "    index=[ \"%d - %d\" % (minR + i*incr_rank, minR + (i+1)*incr_rank -1) for i in range(rank_groups)]\n",
+    ")\n",
+    "observations"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "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>M: 168 - 188</th>\n",
+       "      <th>M: 189 - 210</th>\n",
+       "      <th>F: 155 - 171</th>\n",
+       "      <th>F: 172 - 189</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>1 - 99</th>\n",
+       "      <td>72.807018</td>\n",
+       "      <td>26.754386</td>\n",
+       "      <td>42.172131</td>\n",
+       "      <td>55.426230</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>100 - 199</th>\n",
+       "      <td>72.078947</td>\n",
+       "      <td>26.486842</td>\n",
+       "      <td>27.110656</td>\n",
+       "      <td>35.631148</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>200 - 299</th>\n",
+       "      <td>67.710526</td>\n",
+       "      <td>24.881579</td>\n",
+       "      <td>17.213115</td>\n",
+       "      <td>22.622951</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>300 - 399</th>\n",
+       "      <td>59.701754</td>\n",
+       "      <td>21.938596</td>\n",
+       "      <td>12.049180</td>\n",
+       "      <td>15.836066</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>400 - 499</th>\n",
+       "      <td>58.973684</td>\n",
+       "      <td>21.671053</td>\n",
+       "      <td>6.454918</td>\n",
+       "      <td>8.483607</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "           M: 168 - 188  M: 189 - 210  F: 155 - 171  F: 172 - 189\n",
+       "1 - 99        72.807018     26.754386     42.172131     55.426230\n",
+       "100 - 199     72.078947     26.486842     27.110656     35.631148\n",
+       "200 - 299     67.710526     24.881579     17.213115     22.622951\n",
+       "300 - 399     59.701754     21.938596     12.049180     15.836066\n",
+       "400 - 499     58.973684     21.671053      6.454918      8.483607"
+      ]
+     },
+     "execution_count": 22,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "exp_data = []\n",
+    "for i in range(rank_groups):\n",
+    "    row = []\n",
+    "    ranklb = minR+i*incr_rank\n",
+    "    rankub = minR+(i+1)*incr_rank\n",
+    "    for g in genders:\n",
+    "        if g == 'M':\n",
+    "            cur = data_male\n",
+    "        else:\n",
+    "            cur = data_female_clean\n",
+    "        gnr = len ( cur )\n",
+    "        for j in range(height_groups):\n",
+    "            heightlb = minH[g]+j*incr_height[g]\n",
+    "            heightub = minH[g]+(j+1)*incr_height[g]\n",
+    "            rank_group = len( cur.loc[np.logical_and(cur[\"ranking\"] >= ranklb, cur[\"ranking\"] < rankub)] )\n",
+    "            height_group = len( cur.loc[np.logical_and(cur[\"height\"] >= heightlb, cur[\"height\"] < heightub)] )\n",
+    "            row.append( gnr * (rank_group/gnr) * (height_group/gnr) )\n",
+    "            \n",
+    "    exp_data.append(row)\n",
+    "\n",
+    "expected = pd.DataFrame(\n",
+    "    data = exp_data,\n",
+    "    columns=[ \"M: %d - %d\" % (minH['M']+j*incr_height['M'], minH['M']+(j+1)*incr_height['M']-1) for j in range(height_groups)]\n",
+    "    + [ \"F: %d - %d\" % (minH['F']+j*incr_height['F'], minH['F']+(j+1)*incr_height['F']-1) for j in range(height_groups)],\n",
+    "    index=[ \"%d - %d\" % (minR + i*incr_rank, minR + (i+1)*incr_rank -1) for i in range(rank_groups)]\n",
+    ")\n",
+    "expected"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "7.697606186049128"
+      ]
+     },
+     "execution_count": 23,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "stat_height = stats.chisquare([item for sublist in observations.values for item in sublist], [item for sublist in expected.values for item in sublist]).statistic\n",
+    "stat_height"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.8082925814979871"
+      ]
+     },
+     "execution_count": 24,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "stats.distributions.chi2.sf(stat_height, 12)"
    ]
   },
   {
@@ -780,7 +1039,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 25,
    "metadata": {},
    "outputs": [
     {
@@ -811,7 +1070,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 26,
    "metadata": {},
    "outputs": [
     {
@@ -832,34 +1091,203 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [
     {
-     "ename": "NameError",
-     "evalue": "name 'dataM' is not defined",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m--------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mNameError\u001b[0m                                Traceback (most recent call last)",
-      "\u001b[0;32m<ipython-input-24-284410c792e4>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mgroups\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m4\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mincr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mmaxR\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mminR\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m/\u001b[0m\u001b[0mgroups\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 6\u001b[0;31m \u001b[0mobservations\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m \u001b[0mdataM\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"hand\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"L\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m>=\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m<\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgroups\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m]\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m[\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m \u001b[0mdataM\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"hand\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"R\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m>=\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m<\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgroups\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      7\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[0mlh_prod\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfloat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"hand\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"L\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m/\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m<ipython-input-24-284410c792e4>\u001b[0m in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mgroups\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m4\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mincr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mmaxR\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mminR\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m/\u001b[0m\u001b[0mgroups\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 6\u001b[0;31m \u001b[0mobservations\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m \u001b[0mdataM\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"hand\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"L\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m>=\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m<\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgroups\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m]\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m[\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m \u001b[0mdataM\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_and\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"hand\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"R\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m>=\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"ranking\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m<\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mincr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgroups\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      7\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[0mlh_prod\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfloat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"hand\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"L\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m/\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;31mNameError\u001b[0m: name 'dataM' is not defined"
-     ]
+     "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>1 - 99</th>\n",
+       "      <th>100 - 199</th>\n",
+       "      <th>200 - 299</th>\n",
+       "      <th>300 - 399</th>\n",
+       "      <th>400 - 499</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>L</th>\n",
+       "      <td>22</td>\n",
+       "      <td>23</td>\n",
+       "      <td>17</td>\n",
+       "      <td>6</td>\n",
+       "      <td>8</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>R</th>\n",
+       "      <td>174</td>\n",
+       "      <td>139</td>\n",
+       "      <td>117</td>\n",
+       "      <td>105</td>\n",
+       "      <td>88</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   1 - 99  100 - 199  200 - 299  300 - 399  400 - 499\n",
+       "L      22         23         17          6          8\n",
+       "R     174        139        117        105         88"
+      ]
+     },
+     "execution_count": 27,
+     "metadata": {},
+     "output_type": "execute_result"
     }
    ],
    "source": [
     "# Calculate observations\n",
     "minR = data[\"ranking\"].min()\n",
     "maxR = data[\"ranking\"].max()\n",
-    "groups = 4\n",
+    "groups = 5\n",
     "incr = (maxR - minR)/groups\n",
-    "observations = [ len( dataM.loc[np.logical_and(np.logical_and(data[\"hand\"] == \"L\", data[\"ranking\"] >= i*incr), data[\"ranking\"] < (i+1)*incr)] ) for i in range(groups) ] + [ len( dataM.loc[np.logical_and(np.logical_and(data[\"hand\"] == \"R\", data[\"ranking\"] >= i*incr), data[\"ranking\"] < (i+1)*incr)] ) for i in range(groups) ] \n",
-    "\n",
+    "observations = pd.DataFrame(\n",
+    "    data = [\n",
+    "        [ len( data.loc[np.logical_and(np.logical_and(data[\"hand\"] == \"L\", data[\"ranking\"] >= minR+i*incr), data[\"ranking\"] < minR+(i+1)*incr)] ) for i in range(groups) ],\n",
+    "        [ len( data.loc[np.logical_and(np.logical_and(data[\"hand\"] == \"R\", data[\"ranking\"] >= i*incr), data[\"ranking\"] < (i+1)*incr)] ) for i in range(groups)]\n",
+    "    ],\n",
+    "    columns=[ \"%d - %d\" % (minR + i*incr, minR + (i+1)*incr -1) for i in range(groups)],\n",
+    "    index=[\"L\", \"R\"]\n",
+    ")\n",
+    "observations"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "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>1 - 99</th>\n",
+       "      <th>100 - 199</th>\n",
+       "      <th>200 - 299</th>\n",
+       "      <th>300 - 399</th>\n",
+       "      <th>400 - 499</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>L</th>\n",
+       "      <td>21.466476</td>\n",
+       "      <td>17.563481</td>\n",
+       "      <td>14.527817</td>\n",
+       "      <td>11.92582</td>\n",
+       "      <td>10.407989</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>R</th>\n",
+       "      <td>176.533524</td>\n",
+       "      <td>144.436519</td>\n",
+       "      <td>119.472183</td>\n",
+       "      <td>98.07418</td>\n",
+       "      <td>85.592011</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       1 - 99   100 - 199   200 - 299  300 - 399  400 - 499\n",
+       "L   21.466476   17.563481   14.527817   11.92582  10.407989\n",
+       "R  176.533524  144.436519  119.472183   98.07418  85.592011"
+      ]
+     },
+     "execution_count": 28,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Calculate expected values\n",
     "lh_prod = float(len(data.loc[data[\"hand\"] == \"L\"])) / len(data)\n",
-    "group_nrs = [ len(dataM.loc[np.logical_and(data[\"ranking\"] >= i*incr, data[\"ranking\"] < (i+1)*incr)]) for i in range(groups)]\n",
-    "expected = [ lh_prod*group_nrs[i] for i in range(groups) ] + [ (1-lh_prod)*group_nrs[i] for i in range(groups) ]\n",
-    "stats.chisquare(observations, expected)\n"
+    "group_nrs = [ len(data.loc[np.logical_and(data[\"ranking\"] >= minR + i*incr, data[\"ranking\"] < minR + (i+1)*incr)]) for i in range(groups)]\n",
+    "expected = pd.DataFrame(\n",
+    "    data = [[ lh_prod*group_nrs[i] for i in range(groups) ], [ (1-lh_prod)*group_nrs[i] for i in range(groups) ]],\n",
+    "    columns=[ \"%d - %d\" % (minR + i*incr, minR + (i+1)*incr -1) for i in range(groups)],\n",
+    "    index=[\"L\", \"R\"]\n",
+    ")\n",
+    "expected"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "6.467312944404331"
+      ]
+     },
+     "execution_count": 29,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Chi Square test\n",
+    "stat_hand = stats.chisquare([item for sublist in observations.values for item in sublist], [item for sublist in expected.values for item in sublist]).statistic\n",
+    "stat_hand"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.1668616190847413"
+      ]
+     },
+     "execution_count": 30,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "stats.distributions.chi2.sf(stat_hand, 4)"
    ]
   },
   {