"
+ ]
+ },
+ "metadata": {
+ "needs_background": "light"
+ },
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "#Compare Machine Learning Algorithms Consistently\n",
+ "# Compare Algorithms\n",
+ "\n",
+ "from sklearn.model_selection import KFold\n",
+ "from sklearn.model_selection import cross_val_score\n",
+ "from sklearn.linear_model import LogisticRegression\n",
+ "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
+ "from sklearn.neighbors import KNeighborsClassifier\n",
+ "from sklearn.tree import DecisionTreeClassifier\n",
+ "from sklearn.naive_bayes import GaussianNB\n",
+ "from sklearn.svm import SVC\n",
+ "from sklearn.ensemble import RandomForestClassifier\n",
+ "from sklearn.linear_model import Perceptron\n",
+ "from sklearn.linear_model import SGDClassifier\n",
+ "\n",
+ "array = df.values\n",
+ "\n",
+ "#split the dataset \n",
+ "X = array[:,0:11]\n",
+ "Y = array[:,11]\n",
+ "\n",
+ "# prepare models and add them to a list\n",
+ "models = []\n",
+ "models.append(('LR', LogisticRegression()))\n",
+ "models.append(('LDA', LinearDiscriminantAnalysis()))\n",
+ "models.append(('KNN', KNeighborsClassifier()))\n",
+ "models.append(('TREE', DecisionTreeClassifier()))\n",
+ "models.append(('NB', GaussianNB()))\n",
+ "models.append(('SVM', SVC()))\n",
+ "models.append(('FOREST', RandomForestClassifier()))\n",
+ "models.append(('PERC', Perceptron()))\n",
+ "models.append(('SGD', SGDClassifier()))\n",
+ "models.append(('RVM', SVC()))\n",
+ "\n",
+ "\n",
+ "\n",
+ "# evaluate each model in turn\n",
+ "results = []\n",
+ "names = []\n",
+ "scoring = 'accuracy'\n",
+ "\n",
+ "for name, model in models:\n",
+ " kfold = KFold(n_splits=10, random_state=7)\n",
+ " cv_results = cross_val_score(model, X, Y, cv=kfold, scoring=scoring)\n",
+ " results.append(cv_results)\n",
+ " names.append(name)\n",
+ " msg = (name, cv_results.mean(), cv_results.std())\n",
+ " print(msg)\n",
+ "\n",
+ "# boxplot algorithm comparison\n",
+ "fig = plt.figure()\n",
+ "fig.suptitle('Algorithm Comparison')\n",
+ "ax = fig.add_subplot(111)\n",
+ "plt.figure(figsize=(20,20))\n",
+ "plt.boxplot(results)\n",
+ "ax.set_xticklabels(names)\n",
+ "plt.show()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "> **##Exporting BEST Results to CSV**"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 21,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "MCC: 0.8402064856722555\n",
+ "Accuracy: 90.0\n",
+ "370\n",
+ "388\n"
+ ]
+ }
+ ],
+ "source": [
+ "from sklearn.metrics import matthews_corrcoef\n",
+ "\n",
+ "selected_features_train, selected_features_test, Y_train, Y_test = train_test_split(X, Y, test_size = 10, random_state = 10)\n",
+ "modelG = GaussianNB()\n",
+ "modelG.fit(selected_features_train, Y_train)\n",
+ "result = modelG.score(selected_features_test, Y_test)\n",
+ "\n",
+ "Prediction = modelG.predict(selected_features_train)\n",
+ "print(\"MCC:\", (matthews_corrcoef(Y_train, Prediction)))\n",
+ "print(\"Accuracy:\", (result*100.0))\n",
+ "\n",
+ "outputG = modelG.predict(Test.values) #Assigning the preditions of the model to variable AssignmentOutPut\n",
+ "\n",
+ "outputG = pd.DataFrame(outputG)\n",
+ "# Converting AssignmentOutPut to a dataframe since its output is an array\n",
+ "outputG.columns = ['Class'] # Naming the out\n",
+ "outputG.index.name = \"Index\" #Creating a column called index\n",
+ "outputG['Class'] = outputG['Class'].map({0.0:False, 1.0:True}) # Converting 0.0 to false and 1.0 to true\n",
+ "\n",
+ "outputG.to_csv('outputG.csv') # Writing AssignmentOutPut1 to a csv file\n",
+ "#Printing the number of False and Trues\n",
+ "print(outputG.groupby('Class').size()[0].sum())\n",
+ "print(outputG.groupby('Class').size()[1].sum())"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.7.4"
+ },
+ "varInspector": {
+ "cols": {
+ "lenName": 16,
+ "lenType": 16,
+ "lenVar": 40
+ },
+ "kernels_config": {
+ "python": {
+ "delete_cmd_postfix": "",
+ "delete_cmd_prefix": "del ",
+ "library": "var_list.py",
+ "varRefreshCmd": "print(var_dic_list())"
+ },
+ "r": {
+ "delete_cmd_postfix": ") ",
+ "delete_cmd_prefix": "rm(",
+ "library": "var_list.r",
+ "varRefreshCmd": "cat(var_dic_list()) "
+ }
+ },
+ "types_to_exclude": [
+ "module",
+ "function",
+ "builtin_function_or_method",
+ "instance",
+ "_Feature"
+ ],
+ "window_display": false
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/Assignment Colab/lwasampijja-baker.ipynb b/Assignment Colab/lwasampijja-baker.ipynb
index df98d6b..64c04b0 100644
--- a/Assignment Colab/lwasampijja-baker.ipynb
+++ b/Assignment Colab/lwasampijja-baker.ipynb
@@ -7,6 +7,21 @@
"**#STEP: 1 - IMPORTING LIBRARIES AND DATA**"
]
},
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "\n",
+ " \n",
+ " For your final report please just imagine this:\n",
+ " Different people are going to have a look at your report to try and see if you have some sort of proof of understanding the concepts you applied.\n",
+ " Your explanation and elaboration in the notebook will tell them, so you can't assume anything.\n",
+ " \n",
+ " Give the explanations more meat.\n",
+ " \n",
+ " "
+ ]
+ },
{
"cell_type": "code",
"execution_count": 1,
@@ -19,9 +34,9 @@
"name": "stdout",
"output_type": "stream",
"text": [
+ "/kaggle/input/merged-data/merged_data.csv\n",
"/kaggle/input/ace-class-assignment/Test.csv\n",
- "/kaggle/input/ace-class-assignment/AMP_TrainSet.csv\n",
- "/kaggle/input/merged-data/merged_data.csv\n"
+ "/kaggle/input/ace-class-assignment/AMP_TrainSet.csv\n"
]
}
],
@@ -490,7 +505,7 @@
{
"data": {
"text/plain": [
- ""
+ ""
]
},
"execution_count": 7,
@@ -554,6 +569,102 @@
"cell_type": "code",
"execution_count": 9,
"metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {
+ "needs_background": "light"
+ },
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "# Visualize the count of class for columns 'FULL_Charge', 'FULL_AcidicMolPerc', 'FULL_AURR980107', 'FULL_DAYM780201', 'NT_EFC195 ', 'CT_RACS820104'\n",
+ "cols = ['FULL_Charge', 'FULL_AcidicMolPerc', 'FULL_AURR980107', 'FULL_DAYM780201', 'NT_EFC195', 'CT_RACS820104']\n",
+ "\n",
+ "n_rows = 2\n",
+ "n_cols = 3\n",
+ "\n",
+ "#Number of rows/columns of the subplot grid and the figure size of each graph\n",
+ "#NOTE: This returns a Figure (fig) and an Axes Object (axs)\n",
+ "fig, axs = plt.subplots(n_rows, n_cols, figsize=(n_cols*3.2,n_rows*3.2))\n",
+ "\n",
+ "for r in range(0,n_rows):\n",
+ " for c in range(0,n_cols): \n",
+ " \n",
+ " i = r*n_cols+ c #index to go through the number of columns \n",
+ " ax = axs[r][c] #Show where to position each subplot\n",
+ " plt.figure(figsize=(40,20)) #This is used to change the size of the figure\n",
+ " sns.countplot(df[cols[i]], hue=df[\"CLASS\"], ax=ax)\n",
+ " ax.set_title(cols[i])\n",
+ " ax.legend(title=\"CLASS\", loc='upper right') \n",
+ " \n",
+ "plt.tight_layout() #tight_layout automatically adjusts subplot params so that the subplot(s) fits in to the figure area\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {},
"outputs": [
{
"data": {
@@ -833,7 +944,7 @@
"CLASS 0.033432 0.267652 1.000000 "
]
},
- "execution_count": 9,
+ "execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
@@ -845,16 +956,16 @@
},
{
"cell_type": "code",
- "execution_count": 10,
+ "execution_count": 11,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- ""
+ ""
]
},
- "execution_count": 10,
+ "execution_count": 11,
"metadata": {},
"output_type": "execute_result"
},
@@ -880,16 +991,16 @@
},
{
"cell_type": "code",
- "execution_count": 11,
+ "execution_count": 12,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- ""
+ ""
]
},
- "execution_count": 11,
+ "execution_count": 12,
"metadata": {},
"output_type": "execute_result"
},
@@ -914,16 +1025,16 @@
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": 13,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- ""
+ ""
]
},
- "execution_count": 12,
+ "execution_count": 13,
"metadata": {},
"output_type": "execute_result"
},
@@ -967,7 +1078,7 @@
},
{
"cell_type": "code",
- "execution_count": 13,
+ "execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
@@ -978,7 +1089,7 @@
},
{
"cell_type": "code",
- "execution_count": 14,
+ "execution_count": 15,
"metadata": {},
"outputs": [],
"source": [
@@ -989,7 +1100,7 @@
},
{
"cell_type": "code",
- "execution_count": 15,
+ "execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
@@ -1005,7 +1116,7 @@
},
{
"cell_type": "code",
- "execution_count": 16,
+ "execution_count": 17,
"metadata": {},
"outputs": [],
"source": [
@@ -1087,7 +1198,7 @@
},
{
"cell_type": "code",
- "execution_count": 17,
+ "execution_count": 18,
"metadata": {},
"outputs": [
{
@@ -1102,7 +1213,7 @@
"[6]Decision Tree Classifier Training Accuracy: 1.0\n",
"[7]Random Forest Classifier Training Accuracy: 0.9947322212467077\n",
"[8]Perceptron Training Accuracy: 0.8762071992976295\n",
- "[9]Stochastic Gradient Descent Training Accuracy: 0.9100087796312555\n",
+ "[9]Stochastic Gradient Descent Training Accuracy: 0.8955223880597015\n",
"[10]Linear Discriminant Analysis Training Accuracy: 0.9135206321334504\n"
]
}
@@ -1114,7 +1225,7 @@
},
{
"cell_type": "code",
- "execution_count": 18,
+ "execution_count": 19,
"metadata": {},
"outputs": [
{
@@ -1220,14 +1331,14 @@
"Model 8\n",
" precision recall f1-score support\n",
"\n",
- " 0 0.88 0.96 0.92 353\n",
- " 1 0.96 0.89 0.92 407\n",
+ " 0 0.88 0.93 0.90 353\n",
+ " 1 0.94 0.89 0.91 407\n",
"\n",
- " accuracy 0.92 760\n",
- " macro avg 0.92 0.92 0.92 760\n",
- "weighted avg 0.92 0.92 0.92 760\n",
+ " accuracy 0.91 760\n",
+ " macro avg 0.91 0.91 0.91 760\n",
+ "weighted avg 0.91 0.91 0.91 760\n",
"\n",
- "0.9210526315789473\n",
+ "0.906578947368421\n",
"\n",
"Model 9\n",
" precision recall f1-score support\n",
@@ -1262,7 +1373,7 @@
},
{
"cell_type": "code",
- "execution_count": 19,
+ "execution_count": 20,
"metadata": {},
"outputs": [
{
@@ -1272,12 +1383,12 @@
"('LR', 0.835359128018065, 0.2708521320979506)\n",
"('LDA', 0.8535044293903076, 0.2571395669719574)\n",
"('KNN', 0.8027933385443807, 0.2521136100771112)\n",
- "('TREE', 0.7368681605002605, 0.2833452298920867)\n",
+ "('TREE', 0.7282949018586069, 0.2897498144639564)\n",
"('NB', 0.880815746048289, 0.11642272449162755)\n",
"('SVM', 0.8350280093798853, 0.25836507020625044)\n",
- "('FOREST', 0.8087067917318048, 0.2909686308113511)\n",
+ "('FOREST', 0.8096968907417057, 0.2888139504088565)\n",
"('PERC', 0.8350964043772798, 0.2645271720286034)\n",
- "('SGD', 0.7037432690637486, 0.3108465910214562)\n",
+ "('SGD', 0.8736299287823519, 0.27869983960880124)\n",
"('RVM', 0.8350280093798853, 0.25836507020625044)\n"
]
},
@@ -1295,7 +1406,7 @@
},
{
"data": {
- "image/png": "iVBORw0KGgoAAAANSUhEUgAABIQAAARiCAYAAADVzyypAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3U+IpPd95/Hv19NjBEtGqUGzsFgS8kGBNnNJaJyAdfCQ7CL7YF/C4rmEhCa6xMrFBBw6ZEZe+rKXhQXvLoIOYRe2jU+7YhE4lwnZXtbBLUJCbGEQzh8PDkS704kOQXgsfjnoj0fjlrvVXVVPV31eLxCoqp/u5/t0qaqat57fUz3GKAAAAAByfGTqAQAAAABYLkEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACDMxlQ7fuyxx8ZTTz011e4BAAAA1s4rr7zy/8YY107abrIg9NRTT9Xh4eFUuwcAAABYO939N6fZzpIxAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACDMiUGou/+gu/++u//yA77e3f0fu/u17v6L7v6F+Y8JAAAAwLyc5gyhP6yqZ3/K1z9TVU+/889zVfWfzz8WAAAAAItyYhAaY/xJVd37KZt8vqr+63jbN6vqZ7v7X81rQAAAAADmax7XEPpYVX3/gdt337kPAAAAgAtoYw4/o4+5bxy7Yfdz9faysnryySfnsOvF6z7u8BZrjGN/fZzBsh8/j938eO4BfHheO1ebv1tgGl47V5fH7nzmEYTuVtUTD9x+vKp+cNyGY4wXq+rFqqqtra2V+C2e9cHu7rX6D2VVneUx8NhdDJ57AB+e187V5u8WmIbXztXlsTufeSwZe6mqfu2dTxv7par6xzHG383h5wIAAACwACeeIdTd+1X16ap6rLvvVtWtqrpcVTXG+C9V9XJVfbaqXquqf6qq31jUsAAAAACc34lBaIxx84Svj6r6rblNBAAAAMBCzWPJGAAAAAArRBACAAAACDOPTxkDgPfxEaAAAHCxCUIAzJ2PAAUAgIvNkjEAAACAMM4QAgAAOAdLpYFVJAixEq5evVpHR0dL298y39Rns1ndu3dvafsDAGC+LJUGVpEgxEo4Ojpa2zfLKf6PEgAAANligtCyzzCpcpYJAAAAcDHFBKF1PsOkylkmAAAAwOnFBCEAAFgG1z4EYBUIQgAAF4yl7qttnc9Md1Y6sAje96YhCAEAXDDrHBSqRAUA3s/73jQ+MvUAAAAAACyXIAQAAAAQxpIxAOA9U5zSvM6niAMAXFSCEADwnrPGme4WdgAAVoglYwAAAABhBCEAAACAMJaMAQAAsPKuXr1aR0dHS93nMq+9N5vN6t69e0vbH+tPEAIAAGDlHR0drfX17Kb44AfWmyVjAAAAAGGcIQQAAFCWHAFZBCEAAICy5AjIYskYAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACOOi0sBC+bQOAACAi0cQAhbKp3UAAABcPJaMAQAAAIQRhAAAAADCxCwZG7euVN1+dOoxFmbcujL1CAAAAMCKiAlC/cIba38dk3F76ikAAACAVWDJGAAAAECYmDOEAABWhaXuACTxvjeNnmoZ1dbW1jg8PFza/rp7/ZeMOb6VtM7HVuX4+HD8PleXx26+1v336fhW1zofW5XjW3WOb3Wt87FVLf/4uvuVMcbWSds5Q4iVsM7F+KLWYgAAANaXIMRKWOeLgrsgOAAAAMvmotIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwG1MPAMDFdfXq1To6OlrqPrt7afuazWZ17969pe0PAAAuCkEIgA90dHRUY4ypx1iYZcanZRPzYDrj1pWq249OPcZCjFtXph4BgDkRhABgDYl5MJ1+4Y21ff51d43bU08BwDy4hhAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAML42HkAAICqGreuVN1+dOoxFmbcujL1CMAFIggBAABUVb/wRo0xph5jYbq7xu2ppwAuCkvGAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACLMx9QAAwPyNW1eqbj869RgLM25dmXoEAICVJggBwBrqF96oMcbUYyxMd9e4PfUUAACry5IxAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwG1MPAKy3cetK1e1Hpx5jYcatK1OPAAAA8KEJQsBC9Qtv1Bhj6jEWprtr3J56CgAAgA/HkjEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEGZj6gEAAAAuiu6eeoSFmc1mU48AXCCCEAAAQFWNMZa6v+5e+j4B3mXJGAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwG1MPAKfV3VOPsBCz2WzqEYA1ta6vm1VeOwEAzksQYiWMMZa2r+5e6v4AFmHZr2NeOwEAVoslYwAAAABhnCEEAABwDudZonvW73VWJnBeghAAAMA5iDPAKrJkDAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgzMbUAwBwcY1bV6puPzr1GAszbl2ZegQAAJiEIATAB+oX3qgxxtRjLEx317g99RQAALB8lowBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQZmPqAQAAAOC8xq0rVbcfnXqMhRm3rkw9AmtGEAIAAGDl9Qtv1Bhj6jEWprtr3J56CtZJVBDq7qlHWJjZbDb1CAAAAMCKiAlCyy7F3b3WdRoAAABYXS4qDQAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQZmPqAQAA+EndPfUICzObzaYeAYALxvve8glCAAAXzBhjqfvr7qXvEwDe5X1vGpaMAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgzKmCUHc/293f7e7XuvvLx3z9ye6+091/1t1/0d2fnf+oAAAAAMzDiUGouy9V1Ver6jNV9Ymqutndn3hos9+rqq+PMX6+qr5QVf9p3oMCAAAAMB+nOUPok1X12hjje2OMH1bV16rq8w9tM6rqyjv//mhV/WB+IwIAAAAwTxun2OZjVfX9B27frapffGib21X1R939fFX9i6r6lblMBwAAAMDcneYMoT7mvvHQ7ZtV9YdjjMer6rNV9d+6+yd+dnc/192H3X34+uuvf/hpAQAAADi30wShu1X1xAO3H6+fXBK2XVVfr6oaY/zfqnqkqh57+AeNMV4cY2yNMbauXbt2tokBAAAAOJfTBKFvVdXT3f3x7v5ovX3R6Jce2uZvq+qXq6q6e7PeDkJOAQIAAAC4gE4MQmOMH1XVF6vqG1X1ar39aWLf7u6vdPfn3tnsS1X1m93951W1X1W/PsZ4eFkZAAAAABfAaS4qXWOMl6vq5Yfu+/0H/v07VfWp+Y4GAAAAwCKcZskYAAAAAGtEEAIAAAAIIwgBAAAAhDnVNYQAgAzdvfTv9TkUAADLJwgBAO8RZwAAMlgyBgAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhfMoYa+2sH4Hso5MBAABYZ4IQa02gAQAAgJ8kCAEAwJyd9Wzji242m009AgBzIggBAMAcLfMM5e52RjQAZ+Ki0gAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhNqYeAFh/3T31CAszm82mHgEAAOBDE4SAhRpjLHV/3b30fQIAAKwaS8YAAAAAwghCAAAAAGEEIQAAgCXa39+v69ev16VLl+r69eu1v78/9UhAINcQAgAAWJL9/f3a2dmpvb29euaZZ+rg4KC2t7erqurmzZsTTwckcYYQAADAkuzu7tbe3l7duHGjLl++XDdu3Ki9vb3a3d2dejQgTE/1aTxbW1vj8PBwkn0vg086gml47s3Xuv8+1/344LQ8F1aXx271XLp0qd588826fPnye/fdv3+/HnnkkXrrrbcmnGz1rfvzYd2Pb5nW/XfZ3a+MMbZO2s4ZQgAAAEuyublZBwcH77vv4OCgNjc3J5oISCUIAQAALMnOzk5tb2/XnTt36v79+3Xnzp3a3t6unZ2dqUcDwrioNAAAwJK8e+Ho559/vl599dXa3Nys3d1dF5QGls41hBZk3dckwkXluTdf6/77XPfjg9PyXFhdHjv4sXV/Pqz78S3Tuv8uXUMIAAAAgGMJQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEGZj6gEAAJiP7l76944xzrxPADgP73vnIwgBAKyJdfojFQBO4n3vfCwZAwAAAAjjDKETOAUNAAAAWDeC0AnEGQAAAGDdWDIGAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAmI2pBwA4Tncv/XvHGGfeJwAAwCoRhIALSZwBAABYHEvGAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwpwpC3f1sd3+3u1/r7i9/wDb/tru/093f7u7/Pt8xAQAAAJiXjZM26O5LVfXVqvrXVXW3qr7V3S+NMb7zwDZPV9XvVtWnxhhH3f0vFzUwAAAAAOdzmjOEPllVr40xvjfG+GFVfa2qPv/QNr9ZVV8dYxxVVY0x/n6+YwIAAAAwL6cJQh+rqu8/cPvuO/c96Oeq6ue6+/909ze7+9l5DQgAAADAfJ24ZKyq+pj7xjE/5+mq+nRVPV5V/7u7r48x/uF9P6j7uap6rqrqySef/NDDAgAAAHB+pzlD6G5VPfHA7cer6gfHbPM/xxj3xxh/VVXfrbcD0fuMMV4cY2yNMbauXbt21pkBAAAAOIfTBKFvVdXT3f3x7v5oVX2hql56aJv/UVU3qqq6+7F6ewnZ9+Y5KAAAAADzceKSsTHGj7r7i1X1jaq6VFV/MMb4dnd/paoOxxgvvfO1f9Pd36mqt6rqd8YY/3+RgwOwHN3HrRxeD7PZbOoRAABgEj3Gw5cDWo6tra1xeHg4yb4BuJi6u6Z6XwJYRV434cfW/fmw7sfH/HT3K2OMrZO2O82SMQAAAADWiCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCbEw9AAAAUNXdS/2+McaZvg+A9SAIAQDABSDQALBMlowBAGe2v79f169fr0uXLtX169drf39/6pEAADgFZwgBAGeyv79fOzs7tbe3V88880wdHBzU9vZ2VVXdvHlz4ukAAPhpnCEEAJzJ7u5u7e3t1Y0bN+ry5ct148aN2tvbq93d3alHAwDgBD3VWuWtra1xeHg4yb4BuJi62zU0VsilS5fqzTffrMuXL7933/379+uRRx6pt956a8LJAEi07n9HrPvxMT/d/coYY+uk7ZwhBACcyebmZh0cHLzvvoODg9rc3JxoIgAATksQAgDOZGdnp7a3t+vOnTt1//79unPnTm1vb9fOzs7UowEAcAIXlQYAzuTdC0c///zz9eqrr9bm5mbt7u66oDQAwApwDSEALgxr4wGAs1r3vyPW/fiYH9cQAgAAAOBYghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAmI2pBwBg/XT30r93jHHmfQIAQBpBCIC5E2cAAOBis2QMAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAECYjakHAAAAgHno7qlHWJjZbDb1CKwZQQgAAICVN8ZY6v66e+n7hHmyZAwAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCnCoIdfez3f3d7n6tu7/8U7b71e4e3b01vxEBAAAAmKcTg1B3X6qqr1bVZ6rqE1V1s7s/ccx2P1NVv11VfzrvIQEAAACYn9OcIfTJqnptjPG9McYPq+prVfX5Y7b7d1X176vqzTnOBwAAAMCcnSYIfayqvv/A7bvv3Pee7v75qnpijPG/5jgbAAAAAAtwmiDUx9w33vti90eq6j9U1ZdO/EHdz3X3YXcfvv7666efEgAAAIC5OU0QultVTzxw+/Gq+sEDt3+mqq5X1R93919X1S9V1UvHXVh6jPHiGGNrjLF17dq1s08NAAAAwJmdJgh9q6qe7u6Pd/dHq+oLVfXSu18cY/zjGOOxMcZTY4ynquqbVfW5McbhQiYGAAAA4FxODEJjjB9V1Rer6htV9WpVfX2M8e3u/kp3f27RAwIAAAAwXxun2WiM8XJVvfzQfb//Adt++vxjAQAAALAop1kyBgAAAMAaEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACtVr0qAAAWOElEQVSAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAPDP7d1/bOT5fdfx11uzDm5SoJBeq5JLmiCiMqcRTYJ1CsQKuAkogSoHiEpnqShCI8IfrUkBCQVGKqKSJQqIH1pViKgujaBMSK+tOKEjbZQOoPmjob4kpZu6UUNomyOhd6ilBSI3vuXDH/Yte3t72fGe976e/TwekuXx19/beUsfOeN55vv9GAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ1ZKQhV1buq6rNV9bmq+sBtvv/XquoXquo/V9XHq+qbL35UAAAAAC7CHYNQVY2S/ECSdyd5KMluVT10y2mfSrLVWvtDSR5L8vcuelAAAAAALsYqVwg9nORzrbXPt9a+kuTDSR65+YTW2qK19uWzL38myYMXOyYAAAAAF2WVIPSaJF+46eunzo69mGmSf/dShgIAAADg3rmywjl1m2PttidWfWeSrSR/7EW+/74k70uS173udSuOCAAAAMBFWuUKoaeSvPamrx9M8sVbT6qqdyaZJXlPa+23b/cPtdY+2Frbaq1tPfDAA3czLwAAAAAv0SpB6GeTvLGq3lBVr0jyaJLHbz6hqt6c5J/lNAY9ffFjAgAAAHBR7hiEWmvPJvnuJD+Z5CjJR1prn6mq76uq95yd9veTfG2SH62qT1fV4y/yzwEAAAAwsFX2EEpr7YkkT9xy7HtvevzOC54LAAAAgHtklVvGAAAAALiPCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQcF+Yz+eZTCYZjUaZTCaZz+dDjwQAAHBpXRl6AICXaj6fZzab5eDgINvb21kul5lOp0mS3d3dgacDAAC4fFwhBKy9/f39HBwcZGdnJxsbG9nZ2cnBwUH29/eHHg0AAOBSqtbaIE+8tbXVDg8PB3lu4P4yGo1yfHycjY2NG8dOTk6yubmZ69evDzgZAAD3q6rKUO+n4aupqidba1t3Os8VQsDaG4/HWS6Xzzu2XC4zHo8HmggAAOByE4SAtTebzTKdTrNYLHJycpLFYpHpdJrZbDb0aAAAAJeSTaWBtffcxtF7e3s5OjrKeDzO/v6+DaUBAABehD2EAAAA4JzsIcRlZQ8hAAAAAG5LEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIA6NR8Ps9kMsloNMpkMsl8Ph96JAC4Z7zuPd+VoQcAAODlN5/PM5vNcnBwkO3t7SyXy0yn0yTJ7u7uwNMBwMXyuvdC1Vob5Im3trba4eHhIM8NANC7yWSSq1evZmdn58axxWKRvb29XLt2bcDJANZDVWWo99OcX0+ve1X1ZGtt647nCUIAAP0ZjUY5Pj7OxsbGjWMnJyfZ3NzM9evXB5wMYD0IQuulp9e9VYOQPYQAADo0Ho+zXC6fd2y5XGY8Hg80EQDcO173XkgQAgDo0Gw2y3Q6zWKxyMnJSRaLRabTaWaz2dCjAcCF87r3QjaVBgDo0HMbaO7t7eXo6Cjj8Tj7+/vdbqwJwP3N694L2UMIAAAAzskeQlxW9hACAAAA4LYEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhCCM/P5PJPJJKPRKJPJJPP5fOiRAAAA4J64MvQAcBnM5/PMZrMcHBxke3s7y+Uy0+k0SbK7uzvwdAAAAHCxXCEESfb393NwcJCdnZ1sbGxkZ2cnBwcH2d/fH3o0AAAAuHDVWhvkibe2ttrh4eEgzw23Go1GOT4+zsbGxo1jJycn2dzczPXr1wecDAAAuIyqKkO9n4avpqqebK1t3ek8VwhBkvF4nOVy+bxjy+Uy4/F4oIkAAADg3hGEIMlsNst0Os1iscjJyUkWi0Wm02lms9nQowEAAMCFs6k05P9vHL23t5ejo6OMx+Ps7+/bUBoAAID7kj2EAAAA4JzsIcRlZQ8hAAAAAG5LEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlB6ILN5/NMJpOMRqNMJpPM5/OhRwIAAAB4nitDD3A/mc/nmc1mOTg4yPb2dpbLZabTaZJkd3d34OkAAAAATrlC6ALt7+/n4OAgOzs72djYyM7OTg4ODrK/vz/0aAAAAAA3CEIX6OjoKNvb2887tr29naOjo4EmAgAALhvbTFwuVXVXHy/1v4WhCUIXaDweZ7lcPu/YcrnMeDweaCIAAOAyeW6biatXr+b4+DhXr17NbDYThQbUWnvZP+AyEIQu0Gw2y3Q6zWKxyMnJSRaLRabTaWaz2dCjAQAAl4BtJoDLooaqk1tbW+3w8HCQ576X5vN59vf3c3R0lPF4nNlsZkNpAAAgSTIajXJ8fJyNjY0bx05OTrK5uZnr168POBlwv6iqJ1trW3c6z18Zu2C7u7sCEAAAcFvPbTOxs7Nz45htJoAhuGUMAADgZWKbCeCycIUQAADAy+S5uwn29vZubDOxv7/vLgPgZWcPIQAAAID7xKp7CLllDAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCIDBzefzTCaTjEajTCaTzOfzoUcCAID72pWhBwCgb/P5PLPZLAcHB9ne3s5yucx0Ok2S7O7uDjwdAADcn6q1NsgTb21ttcPDw0GeG4DLYzKZ5OrVq9nZ2blxbLFYZG9vL9euXRtwMgAAWD9V9WRrbeuO5wlCAAxpNBrl+Pg4GxsbN46dnJxkc3Mz169fH3AyAABYP6sGIXsIATCo8Xic5XL5vGPL5TLj8XigiQAA4P4nCAEwqNlslul0msVikZOTkywWi0yn08xms6FHAwCA+5ZNpQEY1HMbR+/t7eXo6Cjj8Tj7+/s2lAYAgHvIHkIAAAAA9wl7CAEAAABwW4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAwJqZz+eZTCYZjUaZTCaZz+dDjwTAmrky9AAAAMDq5vN5ZrNZDg4Osr29neVymel0miTZ3d0deDoA1kW11gZ54q2trXZ4eDjIcwMAwLqaTCa5evVqdnZ2bhxbLBbZ29vLtWvXBpwMgMugqp5srW3d8TxBCAAA1sdoNMrx8XE2NjZuHDs5Ocnm5mauX78+4GQAXAarBiF7CAEAwBoZj8dZLpfPO7ZcLjMejweaCIB1JAgBAMAamc1mmU6nWSwWOTk5yWKxyHQ6zWw2G3o0ANaITaUBAGCNPLdx9N7eXo6OjjIej7O/v29DaQDOxR5CAAAAAPcJewgBAAAAcFuCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnVgpCVfWuqvpsVX2uqj5wm+//jqr612ff/0RVvf6iBwUAAADgYtwxCFXVKMkPJHl3koeS7FbVQ7ecNk3yG621P5DkHyX5/oseFAAAAICLscoVQg8n+Vxr7fOtta8k+XCSR24555EkHzp7/FiSd1RVXdyYAAAAAFyUVYLQa5J84aavnzo7dttzWmvPJvnNJK++iAEBAAAAuFirBKHbXenT7uKcVNX7quqwqg6feeaZVeYDAAAA4IKtEoSeSvLam75+MMkXX+ycqrqS5Hcn+fVb/6HW2gdba1utta0HHnjg7iYGAAAA4CVZJQj9bJI3VtUbquoVSR5N8vgt5zye5L1nj/98kp9urb3gCiEAAAAAhnflTie01p6tqu9O8pNJRkl+qLX2mar6viSHrbXHkxwk+RdV9bmcXhn06L0cGgAAAIC7d8cglCSttSeSPHHLse+96fFxku+42NEAAAAAuBdWuWUMAAAAgPuIIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQmWqtDfPEVc8k+ZVBnvzl8fVJ/sfQQ3BXrN16s37rzfqtL2u33qzf+rJ26836rTfrt77u97X75tbaA3c6abAgdL+rqsPW2tbQc3B+1m69Wb/1Zv3Wl7Vbb9ZvfVm79Wb91pv1W1/W7pRbxgAAAAA6IwgBAAAAdEYQunc+OPQA3DVrt96s33qzfuvL2q0367e+rN16s37rzfqtL2sXewgBAAAAdMcVQgAAAACdEYQuWFX9UFU9XVXXhp6F86mq11bVoqqOquozVfX+oWdidVW1WVX/qap+7mz9/s7QM3E+VTWqqk9V1b8dehbOp6p+uap+vqo+XVWHQ8/D6qrq66rqsar6xbPXvz8y9Eyspqq+5exn7rmP36qq7xl6LlZXVX/17HeWa1U1r6rNoWdiNVX1/rN1+4yfu8vvdu/Rq+r3VtXHquqXzj7/niFnHIogdPF+OMm7hh6Cu/Jskr/eWhsneWuS76qqhwaeidX9dpJva619a5I3JXlXVb114Jk4n/cnORp6CO7aTmvtTf6E69r5J0k+2lr7g0m+NX4G10Zr7bNnP3NvSvKHk3w5yU8MPBYrqqrXJPkrSbZaa5MkoySPDjsVq6iqSZK/lOThnP7v5rdX1RuHnYo7+OG88D36B5J8vLX2xiQfP/u6O4LQBWut/cckvz70HJxfa+1LrbVPnj3+Xzn9pfg1w07Fqtqp/3325cbZh03S1kRVPZjkTyf5waFngV5U1e9K8vYkB0nSWvtKa+1/DjsVd+kdSf5La+1Xhh6Ec7mS5Guq6kqSVyb54sDzsJpxkp9prX25tfZskv+Q5M8OPBNfxYu8R38kyYfOHn8oyZ95WYe6JAQhuI2qen2SNyf5xLCTcB5ntxx9OsnTST7WWrN+6+MfJ/kbSf7v0INwV1qSn6qqJ6vqfUMPw8p+f5Jnkvzzs9s1f7CqXjX0UNyVR5PMhx6C1bXW/luSf5DkV5N8KclvttZ+atipWNG1JG+vqldX1SuT/Kkkrx14Js7vG1trX0pOLwxI8g0DzzMIQQhuUVVfm+THknxPa+23hp6H1bXWrp9dOv9gkofPLunlkquqb0/ydGvtyaFn4a69rbX2liTvzunttm8feiBWciXJW5L809bam5P8n3R6yfw6q6pXJHlPkh8dehZWd7ZfySNJ3pDk9yV5VVV957BTsYrW2lGS70/ysSQfTfJzOd16AtaOIAQ3qaqNnMagH2mt/fjQ83B3zm55+Pexn9e6eFuS91TVLyf5cJJvq6p/OexInEdr7Ytnn5/O6R4mDw87ESt6KslTN11N+VhOAxHr5d1JPtla+7WhB+Fc3pnkv7bWnmmtnST58SR/dOCZWFFr7aC19pbW2ttzeivSLw09E+f2a1X1TUly9vnpgecZhCAEZ6qqcrqPwlFr7R8OPQ/nU1UPVNXXnT3+mpz+ovWLw07FKlprf7O19mBr7fU5ve3hp1tr/l/SNVFVr6qq3/nc4yR/MqeX03PJtdb+e5IvVNW3nB16R5JfGHAk7s5u3C62jn41yVur6pVnv4O+IzZ1XxtV9Q1nn1+X5M/Fz+A6ejzJe88evzfJvxlwlsFcGXqA+01VzZP88SRfX1VPJfnbrbWDYadiRW9L8heS/PzZPjRJ8rdaa08MOBOr+6YkH6qqUU5j90daa/58Odx735jkJ07fz+RKkn/VWvvosCNxDntJfuTstqPPJ/mLA8/DOZztX/InkvzloWfhfFprn6iqx5J8Mqe3G30qyQeHnYpz+LGqenWSkyTf1Vr7jaEH4sXd7j16kr+b5CNVNc1poP2O4SYcTrXmj/AAAAAA9MQtYwAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM78P+plfcpViurVAAAAAElFTkSuQmCC\n",
+ "image/png": "iVBORw0KGgoAAAANSUhEUgAABIQAAARiCAYAAADVzyypAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3U+IpPl93/HvV9MjFoJn3cNOIGh3WR3W0GIuDo1i8B40OIGVDtLFGM3FJDTei7W5mIBCG8+sQl9yCQSUhIU2JgG30MlezIJ8GWO3sYx6MTGRBsEi29GggDaejvdgFo2WXw77h9GoV13TXVVPV31eL1iYqn66n291bXXNvPv5PU+PMQoAAACAHB+begAAAAAAlksQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIszHVjp966qnx3HPPTbV7AAAAgLXzxhtv/N8xxrXTtpssCD333HN1dHQ01e4BAAAA1k53/90s21kyBgAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEOTUIdffvdvcPu/t/fcTHu7v/c3e/2d1/3d3/fP5jAgAAADAvsxwh9HtV9eLP+Phnq+r59/97qar+6/nHAgAAAGBRTg1CY4w/rar7P2OTL1TVfx/v+WZV/Xx3/7N5DQgAAADAfM3jHEKfqKrvP3T73vv3AQAAAHABbczha/QJ940TN+x+qd5bVlbPPvvsHHa9eN0nPbzFGuPEbx9nsOznz3M3P157q83zB9Pw2ltt/t4C0/Czc3V57s5nHkHoXlU989Dtp6vqBydtOMZ4taperara3t5eie/iWZ/s7l6r/1FW1VmeA8/dxeC1t9o8fzANr73V5u8tMA0/O1eX5+585rFk7LWq+vX3rzb2S1X1D2OM/zOHrwsAAADAApx6hFB3H1TVZ6rqqe6+V1W3qupyVdUY479V1etV9bmqerOq/rGq/s2ihgUAAADg/E4NQmOMm6d8fFTVb85tIgAAAAAWah5LxgAAAABYIYIQAAAAQJh5XGUMAAAAJnX16tU6Pj5e6j6Xednzzc3Nun///tL2x/oThAAAAFh5x8fHa30p8WXGJzJYMgYAAAAQxhFCAAAA5zDFkRvrfCQMsByCECth2euBrQUGAGBWZ40z3S3sAJMRhFgJ67we2FpgAAAAli0mCDnjPAAAAMB7YoLQOh9hUuUoEwAAAGB2rjIGAAAAECbmCCEAgFVhqTsASbzvTUMQAgC4YCx1ByCJ971pCEIAADBHy/5Nt99yA3AWghAAAMzROv+m+6L+lhuAxycIAQAfmuIfe+v6D2cAgItMEAIAPnTWONPdwg4AwApx2XkAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACOMqYwAAAFV19erVOj4+Xuo+u3tp+9rc3Kz79+8vbX/AxSYIAQAAVNXx8XGNMaYeY2GWGZ+Ai8+SMQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABDGSaWBhXK1DgAAgItHEAIWytU6AAAALh5LxgAAAADCCEIAAAAAYWKWjI1bV6puPzn1GAszbl2ZegQAAABgRcQEoX7l7bU/j8m4PfUUAAAA03AQADyemCAEAADA+nIQADweQQgA4ILxW24Aknjfm0ZPVVC3t7fH0dHR0vbX3etfiz2+lbTOj63K4+Px+H6uLs/dfK3799PjW13r/NiqPL5V5/GtrnV+bFXLf3zd/cYYY/u07RwhxEpY52J8UWsxAAAA60sQYiWs83pga4EBAABYto9NPQAAAAAAyyUIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAizMfUAAFxcV69erePj46Xus7uXtq/Nzc26f//+0vYHAAAXhSAEwEc6Pj6uMcbUYyzMMuMTALB46/zevrm5OfUIrBlBCADWkKO7YDrj1pWq209OPcZCjFtXph4BPtKyf4nV3Wv9izPWnyAEAGvI0V0wnX7l7bV9/XV3jdtTTwHAPDipNAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwrjsPAAAQFWNW1eqbj859RgLM25dmXoE4AIRhAAAAKqqX3m7xhhTj7Ew3V3j9tRTABeFJWMAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACE2Zh6AABg/satK1W3n5x6jIUZt65MPQIAwEoThABgDfUrb9cYY+oxFqa7a9yeegoAgNVlyRgAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQJiNqQcA1tu4daXq9pNTj7Ew49aVqUcAAAB4bIIQsFD9yts1xph6jIXp7hq3p54CAADg8VgyBgAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIszH1AAAAADCV7l76544xzrxPmBdBCAAAgFjiDKksGQMAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQZmPqAQAAAC6K7p56hIXZ3NycegTgAhGEAAAAqmqMsdT9dffS9wnwAUvGAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAITZmHoAmFV3Tz3CQmxubk49ArCm1vXnZpWfnQAA5yUIsRLGGEvbV3cvdX9wkY1bV6puPzn1GAszbl2ZeoSFWfbPMT87AQBWiyAEwEfqV95e63/kd3eN21NPAQAAyycIAQAAnMN5luie9XPX+Rc2wHIIQgAAAOcgzgCryFXGAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAmI2pB1im7p56hIXZ3NycegQAAABgRcQEoTHGUvfX3UvfJwAAAMAsLBkDAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAizMfUAAAD8tO6eeoSF2dzcnHoEAC4Y73vLJwgBAFwwY4yl7q+7l75PAPiA971pWDIGAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwMwWh7n6xu7/b3W9295dP+Piz3X2nu/+qu/+6uz83/1EBAAAAmIdTg1B3X6qqr1bVZ6vqU1V1s7s/9chmv11VXx9j/GJVfbGq/su8BwUAAABgPmY5QujTVfXmGON7Y4wfVdXXquoLj2wzqurK+39+sqp+ML8RAQAAAJinjRm2+URVff+h2/eq6l88ss3tqvrj7n65qv5JVf3LuUwHAAAAwNzNcoRQn3DfeOT2zar6vTHG01X1uar6H939U1+7u1/q7qPuPnrrrbcef1oAAAAAzm2WIHSvqp556PbT9dNLwnaq6utVVWOMv6iqJ6rqqUe/0Bjj1THG9hhj+9q1a2ebGAAAAIBzmSUIfauqnu/uT3b3x+u9k0a/9sg2/7uqfqWqqru36r0g5BAgAAAAgAvo1CA0xvhxVX2pqr5RVXfrvauJfbu7v9Ldn39/s9+qqt/o7v9ZVQdV9a/HGI8uKwMAAADgApjlpNI1xni9ql5/5L7feejP36mqX57vaAAAAAAswixLxgAAAABYI4IQAAAAQBhBCAAAACDMTOcQAgAydPfSP9d1KAAAlk8QAgA+JM4AAGSwZAwAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwrjKGGvtrJdAdulkAAAA1pkgxFoTaAAAAOCnCUIAADBnZz3a+KLb3NycegQA5kQQAgCAOVrmEcrd7YhoAM7ESaUBAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwmxMPQCw/rp76hEWZnNzc+oRAAAAHpsgBCzUGGOp++vupe8TAABg1VgyBgAAABBGEAIAAAAIIwgBAAAs0cHBQV2/fr0uXbpU169fr4ODg6lHAgI5hxAAAMCSHBwc1O7ubu3v79cLL7xQh4eHtbOzU1VVN2/enHg6IIkjhAAAAJZkb2+v9vf368aNG3X58uW6ceNG7e/v197e3tSjAWF6qqvxbG9vj6Ojo0n2vQyudATT8Nqbr3X/fq7744NZeS2sLs/d6rl06VK98847dfny5Q/ve/DgQT3xxBP17rvvTjgZ5Fj3n53d/cYYY/u07RwhBAAAsCRbW1t1eHj4E/cdHh7W1tbWRBMBqQQhAACAJdnd3a2dnZ26c+dOPXjwoO7cuVM7Ozu1u7s79WhAGCeVBgAAWJIPThz98ssv1927d2tra6v29vacUBpYOucQWpB1X5MIF5XX3nyt+/dz3R8fzMprYXV57gAe37r/7HQOIQAAAABOJAgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAECYjakHAABgPrp76Z87xjjzPgHgPLzvnY8gBACwJtbpL6kAcBrve+djyRgAAABAGEcIncIhaAAAAMC6EYROIc4AAAAA68aSMQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMJsTD0AwEm6e+mfO8Y48z4BAABWiSAEXEjiDAAAwOJYMgYAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhJkpCHX3i9393e5+s7u//BHb/Fp3f6e7v93dvz/fMQEAAACYl43TNujuS1X11ar6V1V1r6q+1d2vjTG+89A2z1fVv6+qXx5jHHf3P13UwAAAAACczyxHCH26qt4cY3xvjPGjqvpaVX3hkW1+o6q+OsY4rqoaY/xwvmMCAAAAMC+zBKFPVNX3H7p97/37HvYLVfUL3f3n3f3N7n5xXgMCAAAAMF+nLhmrqj7hvnHC13m+qj5TVU9X1Z919/Uxxv/7iS/U/VJVvVRV9eyzzz72sAAAAACc3yxHCN2rqmceuv10Vf3ghG3+cIzxYIzxN1X13XovEP2EMcarY4ztMcb2tWvXzjozAAAAAOcwSxD6VlU9392f7O6PV9UXq+q1R7b5g6q6UVXV3U/Ve0vIvjfPQQEAAACYj1OD0Bjjx1X1par6RlXdraqvjzG+3d1f6e7Pv7/ZN6rq77v7O1V1p6r+3Rjj7xc1NAAAAABn12M8ejqg5dje3h5HR0eT7BuA2XR3TfU+sQzr/viA9efnGACP6u43xhjbp203y5IxAAAAANbILFcZAwAAFqz7pIv7Lu7zHFkEkE0QAgCAC0CgAWCZLBkDAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEA4MwODg7q+vXrdenSpbp+/XodHBxMPRIAADPYmHoAAGA1HRwc1O7ubu3v79cLL7xQh4eHtbOzU1VVN2/enHg6AAB+FkcIAQBnsre3V/v7+3Xjxo26fPly3bhxo/b392tvb2/q0QAAOEWPMSbZ8fb29jg6Oppk3wDMprtrqveJZVj3x7doly5dqnfeeacuX7784X0PHjyoJ554ot59990JJwMAyNXdb4wxtk/bzhFCAMCZbG1t1eHh4U/cd3h4WFtbWxNNBADArAQhAOBMdnd3a2dnp+7cuVMPHjyoO3fu1M7OTu3u7k49GgAAp3BSaQDgTD44cfTLL79cd+/era2trdrb23NCaQCAFeAcQgB8pHU/x866Pz4AAPI4hxAAAAAAJxKEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgzMbUAwBwsXX31CMszObm5tQjAADAJAQhAD7SGGOp++vupe8TAAASWTIGAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEmSkIdfeL3f3d7n6zu7/8M7b71e4e3b09vxEBAAAAmKdTg1B3X6qqr1bVZ6vqU1V1s7s/dcJ2P1dV/7aq/nLeQwIAAAAwP7McIfTpqnpzjPG9McaPquprVfWFE7b7D1X1H6vqnTnOBwAAAMCczRKEPlFV33/o9r337/tQd/9iVT0zxvijOc4GAAAAwALMEoT6hPvGhx/s/lhV/aeq+q1Tv1D3S9191N1Hb7311uxTAgAAADA3swShe1X1zEO3n66qHzx0++eq6npV/Ul3/21V/VJVvXbSiaXHGK+OMbbHGNvXrl07+9QAAAAAnNksQehbVfV8d3+yuz9eVV+sqtc++OAY4x/GGE+NMZ4bYzxXVd+sqs+PMY4WMjEAAAAA53JqEBpj/LiqvlRV36iqu1X19THGt7v7K939+UUPCAAAAMB8bcyy0Rjj9ap6/ZH7fucjtv3M+ccCAAAAYFFmWTIGAAAAwBoRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwgBAAAAhBGEAAAAAMIIQgAAAABhBCEAAACAMIIQAAAAQBhBCAAAACCMIAQAAAAQRhACAAAACCMIAQAAAIQRhAAAAADCCEIAAAAAYQQhAAAAgDCCEAAAAEAYQQgAAAAgjCAEAAAAEEYQAgAAAAgjCAEAAACEEYQAAAAAwghCAAAAAGEEIQAAAIAwghAAAABAGEEIAAAAIIwgBAAAABBGEAIAAAAIIwj+qGOhAAAWJElEQVQBAAAAhBGEAAAAAMIIQgAAAABhBCEAgP/f3v2H2J7fdx1/vTl34m1StZpuS80mTaShnnCwSRxCNEN0miiJlqyKhQxUghyMf7RjqoJED1QsHLAq/uBSxNBTG7SemKYtLpKmDelROX80djZJ7abT0BjbZk3srrS2aphmcv34x8ze7o+73jN35+53zv08HnCZc77zvfe84cPsmXnu9/sZAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAANAZQQgAAACgMxsFoap6W1V9pqo+W1Xvvc3n/3pV/UJV/eeq+lhVfePljwoAAADAZbhjEKqqUZLvT/L2JK9JclBVr3nGaZ9Mstta+8NJPpTk71/2oAAAAABcjk2uEHpDks+21j7XWvtykg8keeipJ7TWVq21L50//ZkkD17umAAAAABclk2C0MuSfP4pzx87P/Zcpkl+4vkMBQAAAMC9c22Dc+o2x9ptT6z6jiS7Sf74c3z+3UnenSSveMUrNhwRAAAAgMu0yRVCjyV5+VOeP5jkC888qaremmSW5B2ttd++3T/UWntfa223tbb7wAMP3M28AAAAADxPmwShn03y6qp6VVW9KMk7kzz81BOq6nVJ/nnOYtDjlz8mAAAAAJfljkGotfaVJN+V5CeTHCf5YGvt01X1vVX1jvPT/kGSr07yI1X1qap6+Dn+OQAAAAAGtskeQmmtfTjJh59x7Hue8vitlzwXAAAAAPfIJreMAQAAAHAfEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQh4L6wXC4zmUwyGo0ymUyyXC6HHgkAAODKujb0AADP13K5zGw2y2KxyN7eXtbrdabTaZLk4OBg4OkAAACuHlcIAVtvPp9nsVhkf38/Ozs72d/fz2KxyHw+H3o0AACAK6laa4O88O7ubjs6OhrktYH7y2g0ysnJSXZ2dm4dOz09zfXr13Pz5s0BJ+OiqipDvS8BAMD9oKoeaa3t3uk8VwgBW288Hme9Xj/t2Hq9zng8HmgiAACAq00QArbebDbLdDrNarXK6elpVqtVptNpZrPZ0KMBAABcSTaVBrbekxtHHx4e5vj4OOPxOPP53IbSAAAAz8EeQgBcGfYQAgCA58ceQgAAAADcliAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEANCp5XKZyWSS0WiUyWSS5XI59EgAcM9433u6a0MPAADAC2+5XGY2m2WxWGRvby/r9TrT6TRJcnBwMPB0AHC5vO89W7XWBnnh3d3ddnR0NMhrA3A1VVWGel+C3kwmk9y4cSP7+/u3jq1WqxweHubRRx8dcDIAuHw9ve9V1SOttd07nicIAXBVCELwwhmNRjk5OcnOzs6tY6enp7l+/Xpu3rw54GQAcPl6et/bNAjZQwgAoEPj8Tjr9fppx9brdcbj8UATAcC9433v2QQhAIAOzWazTKfTrFarnJ6eZrVaZTqdZjabDT0aAFw673vPZlNpAIAOPbmB5uHhYY6PjzMejzOfz7vdWBOA+5v3vWezhxAAV4Y9hAAA4PmxhxAAAAAAtyUIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghCcWy6XmUwmGY1GmUwmWS6XQ48EAAAA98S1oQeAq2C5XGY2m2WxWGRvby/r9TrT6TRJcnBwMPB0AAAAcLlcIQRJ5vN5FotF9vf3s7Ozk/39/SwWi8zn86FHAwAAgEtXrbVBXnh3d7cdHR0N8trwTKPRKCcnJ9nZ2bl17PT0NNevX8/NmzcHnAz6UlUZ6n0JAADuB1X1SGtt907nuUIIkozH46zX66cdW6/XGY/HA00EAAAA944gBElms1mm02lWq1VOT0+zWq0ynU4zm82GHg0AAAAunU2lIb+zcfTh4WGOj48zHo8zn89tKA0AAMB9yR5CAFwZ9hACAIDnxx5CAAAAANyWIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOC0CVbLpeZTCYZjUaZTCZZLpdDjwQAAADwNNeGHuB+slwuM5vNslgssre3l/V6nel0miQ5ODgYeDoAAACAM64QukTz+TyLxSL7+/vZ2dnJ/v5+FotF5vP50KMBAAAA3CIIXaLj4+Ps7e097dje3l6Oj48HmggAALhqbDMBXAWC0CUaj8dZr9dPO7ZerzMejweaCAAAuEqe3Gbixo0bOTk5yY0bNzKbzUQh4AUnCF2i2WyW6XSa1WqV09PTrFarTKfTzGazoUcDAACuANtMAFdFtdYGeeHd3d12dHQ0yGvfS8vlMvP5PMfHxxmPx5nNZjaUBthQVWWo9yUAeCGMRqOcnJxkZ2fn1rHT09Ncv349N2/eHHAy4H5RVY+01nbvdJ7fMnbJDg4OBCAAAOC2ntxmYn9//9Yx20wAQ3DLGAAAwAvENhPAVeEKIQAAgBfIk3cTHB4e3tpmYj6fu8sAeMHZQwiAK8MeQgAA8PxsuoeQW8YAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABsaLlcZjKZZDQaZTKZZLlcDj0S3JVrQw8AAAAA22C5XGY2m2WxWGRvby/r9TrT6TRJcnBwMPB0cDGuEAIAAIANzOfzLBaL7O/vZ2dnJ/v7+1ksFpnP50OPBhdWrbVBXnh3d7cdHR0N8toAXE1VlaHelwAA7mQ0GuXk5CQ7Ozu3jp2enub69eu5efPmgJPB76iqR1pru3c6zxVCAAAAsIHxeJz1ev20Y+v1OuPxeKCJ4O4JQgAAALCB2WyW6XSa1WqV09PTrFarTKfTzGazoUeDC7OpNACXrqpe8L/rVjMA4F57cuPow8PDHB8fZzweZz6f21CarWQPIQAAAID7hD2EAAAAALgtQQgAAACgM4IQAAAAQGcEIQAAAIDOCEIAAAAAnRGEAAAAADojCAEAAAB0RhACAAAA6IwgBAAAW2a5XGYymWQ0GmUymWS5XA49EgBb5trQAwAAAJtbLpeZzWZZLBbZ29vLer3OdDpNkhwcHAw8HQDbolprg7zw7u5uOzo6GuS1AQBgW00mk9y4cSP7+/u3jq1WqxweHubRRx8dcDIAroKqeqS1tnvH8wQhAADYHqPRKCcnJ9nZ2bl17PT0NNevX8/NmzcHnAyAq2DTIGQPIQAA2CLj8Tjr9fppx9brdcbj8UATAbCNBCEAANgis9ks0+k0q9Uqp6enWa1WmU6nmc1mQ48GwBaxqTQAAGyRJzeOPjw8zPHxccbjcebzuQ2lAbgQewgBAAAA3CfsIQQAAADAbQlCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ0RhAAAAAA6IwgBAAAAdEYQAgAAAOiMIAQAAADQGUEIAAAAoDOCEAAAAEBnBCEAAACAzghCAAAAAJ3ZKAhV1duq6jNV9dmqeu9tPv+7qurfnH/+41X1ysseFAAAAIDLcccgVFWjJN+f5O1JXpPkoKpe84zTpkl+o7X2TUn+cZLvu+xBAQAAALgcm1wh9IYkn22tfa619uUkH0jy0DPOeSjJ+88ffyjJW6qqLm9MAAAAAC7LJkHoZUk+/5Tnj50fu+05rbWvJPnNJC+9jAEBAAAAuFybBKHbXenT7uKcVNW7q+qoqo6eeOKJTeYDAAAA4JJtEoQeS/Lypzx/MMkXnuucqrqW5Pcm+fVn/kOttfe11nZba7sPPPDA3U0MAAAAwPOySRD62SSvrqpXVdWLkrwzycPPOOfhJO86f/wXkvx0a+1ZVwgBAAAAMLxrdzqhtfaVqvquJD+ZZJTkB1trn66q701y1Fp7OMkiyb+sqs/m7Mqgd97LoQEAAAC4e3cMQknSWvtwkg8/49j3POXxSZJvv9zRAAAAALgXNrllDAAAAID7iCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAzghAAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0JlqrQ3zwlVPJPmVQV78hfG1Sf7H0ENwV6zddrN+2836bS9rt92s3/aydtvN+m0367e97ve1+8bW2gN3OmmwIHS/q6qj1tru0HNwcdZuu1m/7Wb9tpe1227Wb3tZu+1m/bab9dte1u6MW8YAAAAAOiMIAQAAAHRGELp33jf0ANw1a7fdrN92s37by9ptN+u3vazddrN+2836bS9rF3sIAQAAAHTHFUIAAAAAnRGELllV/WBVPV5Vjw49CxdTVS+vqlVVHVfVp6vqPUPPxOaq6npV/aeq+rnz9fu7Q8/ExVTVqKo+WVX/buhZuJiq+uWq+vmq+lRVHQ09D5urqq+pqg9V1S+ev//90aFnYjNV9c3nX3NP/vmtqvruoedic1X1186/Z3m0qpZVdX3omdhMVb3nfN0+7evu6rvdz+hV9fur6qNV9UvnH3/fkDMORRC6fD+U5G1DD8Fd+UqSv9FaGyd5Y5LvrKrXDDwTm/vtJN/aWvuWJK9N8raqeuPAM3Ex70lyPPQQ3LX91tpr/QrXrfNPk3yktfaHknxLfA1ujdbaZ86/5l6b5I8k+VKSHx94LDZUVS9L8leT7LbWJklGSd457FRsoqomSf5ykjfk7L+b31ZVrx52Ku7gh/Lsn9Hfm+RjrbVXJ/nY+fPuCEKXrLX2H5P8+tBzcHGttS+21j5x/vh/5eyb4pcNOxWbamf+9/nTnfM/NknbElX1YJI/k+QHhp4FelFVvyfJm5MskqS19uXW2v8cdiru0luS/JfW2q8MPQgXci3JV1XVtSQvTvKFgedhM+MkP9Na+1Jr7StJ/kOSPzfwTPx/PMfP6A8lef/54/cn+bMv6FBXhCAEt1FVr0zyuiQfH3YSLuL8lqNPJXk8yUdba9Zve/yTJH8zyf8dehDuSkvyU1X1SFW9e+hh2NgfTPJEkn9xfrvmD1TVS4YeirvyziTLoYdgc621/5bkHyb51SRfTPKbrbWfGnYqNvRokjdX1Uur6sVJ/nSSlw88Exf39a21LyZnFwYk+bqB5xmEIATPUFVfneRHk3x3a+23hp6HzbXWbp5fOv9gkjecX9LLFVdV35bk8dbaI0PPwl17U2vt9UnenrPbbd889EBs5FqS1yf5Z6211yX5P+n0kvltVlUvSvKOJD8y9Cxs7ny/koeSvCrJH0jykqr6jmGnYhOtteMk35fko0k+kuTncrb1BGwdQQieoqp2chaDfri19mNDz8PdOb/l4d/Hfl7b4k1J3lFVv5zkA0m+tar+1bAjcRGttS+cf3w8Z3uYvGHYidjQY0kee8rVlB/KWSBiu7w9ySdaa7829CBcyFuT/NfW2hOttdMkP5bkjw08ExtqrS1aa69vrb05Z7ci/dLQM3Fhv1ZV35Ak5x8fH3ieQQhCcK6qKmf7KBy31v7R0PNwMVX1QFV9zfnjr8rZN1q/OOxUbKK19rdaaw+21l6Zs9sefrq15v+SbomqeklV/e4nHyf5Uzm7nJ4rrrX235N8vqq++fzQW5L8woAjcXcO4naxbfSrSd5YVS8+/x70LbGp+9aoqq87//iKJH8+vga30cNJ3nX++F1J/u2Aswzm2tAD3G+qapnkTyT52qp6LMnfaa0thp2KDb0pyV9M8vPn+9Akyd9urX14wJnY3DckeX9VjXIWuz/YWvPry+He+/okP37280yuJfnXrbWPDDsSF3CY5IfPbzv6XJK/NPA8XMD5/iV/MslfGXoWLqa19vGq+lCST+TsdqNPJnnfsFNxAT9aVS9NcprkO1trvzH0QDy32/2MnuTvJflgVU1zFmi/fbgJh1Ot+SU8AAAAAD1xyxgAAABAZwQhAAAAgM4IQgAAAACdEYQAAAAAOiMIAQAAAHRGEAIAAADojCAEAAAA0BlBCAAAAKAz/w8BCRVIbRrZzgAAAABJRU5ErkJggg==\n",
"text/plain": [
""
]
@@ -1375,7 +1486,7 @@
},
{
"cell_type": "code",
- "execution_count": 20,
+ "execution_count": 21,
"metadata": {},
"outputs": [
{
@@ -1432,7 +1543,36 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.6.6"
+ "version": "3.7.4"
+ },
+ "varInspector": {
+ "cols": {
+ "lenName": 16,
+ "lenType": 16,
+ "lenVar": 40
+ },
+ "kernels_config": {
+ "python": {
+ "delete_cmd_postfix": "",
+ "delete_cmd_prefix": "del ",
+ "library": "var_list.py",
+ "varRefreshCmd": "print(var_dic_list())"
+ },
+ "r": {
+ "delete_cmd_postfix": ") ",
+ "delete_cmd_prefix": "rm(",
+ "library": "var_list.r",
+ "varRefreshCmd": "cat(var_dic_list()) "
+ }
+ },
+ "types_to_exclude": [
+ "module",
+ "function",
+ "builtin_function_or_method",
+ "instance",
+ "_Feature"
+ ],
+ "window_display": false
}
},
"nbformat": 4,