Skip to content

A
AI Project
Commit 746088ed authored by Jonathan Poalses's avatar Jonathan Poalses
Browse files
Options

Added which test split is best

parent 9e255863
Show whitespace changes
Inline Side-by-side
Showing with 85 additions and 50 deletions
numbers_ml.ipynb
View file @ 746088ed
...@@ -22,7 +22,7 @@ ...@@ -22,7 +22,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 309, "execution_count": 525,
"outputs": [], "outputs": [],
"source": [ "source": [
"# Importing pyplot so we can visualize things\n", "# Importing pyplot so we can visualize things\n",
...@@ -61,8 +61,8 @@ ...@@ -61,8 +61,8 @@
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"ExecuteTime": { "ExecuteTime": {
"end_time": "2023-05-25T14:28:54.613430Z", "end_time": "2023-05-25T23:09:23.077805Z",
"start_time": "2023-05-25T14:28:54.538858Z" "start_time": "2023-05-25T23:09:22.993181Z"
} }
} }
}, },
...@@ -78,13 +78,13 @@ ...@@ -78,13 +78,13 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 310, "execution_count": 526,
"outputs": [ "outputs": [
{ {
"data": { "data": {
"text/plain": "array([0, 1, 2, ..., 8, 9, 8])" "text/plain": "array([0, 1, 2, ..., 8, 9, 8])"
}, },
"execution_count": 310, "execution_count": 526,
"metadata": {}, "metadata": {},
"output_type": "execute_result" "output_type": "execute_result"
} }
...@@ -99,8 +99,8 @@ ...@@ -99,8 +99,8 @@
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"ExecuteTime": { "ExecuteTime": {
"end_time": "2023-05-25T14:28:54.638244Z", "end_time": "2023-05-25T23:09:23.098977Z",
"start_time": "2023-05-25T14:28:54.545300Z" "start_time": "2023-05-25T23:09:22.999704Z"
} }
} }
}, },
...@@ -115,13 +115,13 @@ ...@@ -115,13 +115,13 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 311, "execution_count": 527,
"outputs": [ "outputs": [
{ {
"data": { "data": {
"text/plain": "array([[[ 0., 0., 5., ..., 1., 0., 0.],\n [ 0., 0., 13., ..., 15., 5., 0.],\n [ 0., 3., 15., ..., 11., 8., 0.],\n ...,\n [ 0., 4., 11., ..., 12., 7., 0.],\n [ 0., 2., 14., ..., 12., 0., 0.],\n [ 0., 0., 6., ..., 0., 0., 0.]],\n\n [[ 0., 0., 0., ..., 5., 0., 0.],\n [ 0., 0., 0., ..., 9., 0., 0.],\n [ 0., 0., 3., ..., 6., 0., 0.],\n ...,\n [ 0., 0., 1., ..., 6., 0., 0.],\n [ 0., 0., 1., ..., 6., 0., 0.],\n [ 0., 0., 0., ..., 10., 0., 0.]],\n\n [[ 0., 0., 0., ..., 12., 0., 0.],\n [ 0., 0., 3., ..., 14., 0., 0.],\n [ 0., 0., 8., ..., 16., 0., 0.],\n ...,\n [ 0., 9., 16., ..., 0., 0., 0.],\n [ 0., 3., 13., ..., 11., 5., 0.],\n [ 0., 0., 0., ..., 16., 9., 0.]],\n\n ...,\n\n [[ 0., 0., 1., ..., 1., 0., 0.],\n [ 0., 0., 13., ..., 2., 1., 0.],\n [ 0., 0., 16., ..., 16., 5., 0.],\n ...,\n [ 0., 0., 16., ..., 15., 0., 0.],\n [ 0., 0., 15., ..., 16., 0., 0.],\n [ 0., 0., 2., ..., 6., 0., 0.]],\n\n [[ 0., 0., 2., ..., 0., 0., 0.],\n [ 0., 0., 14., ..., 15., 1., 0.],\n [ 0., 4., 16., ..., 16., 7., 0.],\n ...,\n [ 0., 0., 0., ..., 16., 2., 0.],\n [ 0., 0., 4., ..., 16., 2., 0.],\n [ 0., 0., 5., ..., 12., 0., 0.]],\n\n [[ 0., 0., 10., ..., 1., 0., 0.],\n [ 0., 2., 16., ..., 1., 0., 0.],\n [ 0., 0., 15., ..., 15., 0., 0.],\n ...,\n [ 0., 4., 16., ..., 16., 6., 0.],\n [ 0., 8., 16., ..., 16., 8., 0.],\n [ 0., 1., 8., ..., 12., 1., 0.]]])" "text/plain": "array([[[ 0., 0., 5., ..., 1., 0., 0.],\n [ 0., 0., 13., ..., 15., 5., 0.],\n [ 0., 3., 15., ..., 11., 8., 0.],\n ...,\n [ 0., 4., 11., ..., 12., 7., 0.],\n [ 0., 2., 14., ..., 12., 0., 0.],\n [ 0., 0., 6., ..., 0., 0., 0.]],\n\n [[ 0., 0., 0., ..., 5., 0., 0.],\n [ 0., 0., 0., ..., 9., 0., 0.],\n [ 0., 0., 3., ..., 6., 0., 0.],\n ...,\n [ 0., 0., 1., ..., 6., 0., 0.],\n [ 0., 0., 1., ..., 6., 0., 0.],\n [ 0., 0., 0., ..., 10., 0., 0.]],\n\n [[ 0., 0., 0., ..., 12., 0., 0.],\n [ 0., 0., 3., ..., 14., 0., 0.],\n [ 0., 0., 8., ..., 16., 0., 0.],\n ...,\n [ 0., 9., 16., ..., 0., 0., 0.],\n [ 0., 3., 13., ..., 11., 5., 0.],\n [ 0., 0., 0., ..., 16., 9., 0.]],\n\n ...,\n\n [[ 0., 0., 1., ..., 1., 0., 0.],\n [ 0., 0., 13., ..., 2., 1., 0.],\n [ 0., 0., 16., ..., 16., 5., 0.],\n ...,\n [ 0., 0., 16., ..., 15., 0., 0.],\n [ 0., 0., 15., ..., 16., 0., 0.],\n [ 0., 0., 2., ..., 6., 0., 0.]],\n\n [[ 0., 0., 2., ..., 0., 0., 0.],\n [ 0., 0., 14., ..., 15., 1., 0.],\n [ 0., 4., 16., ..., 16., 7., 0.],\n ...,\n [ 0., 0., 0., ..., 16., 2., 0.],\n [ 0., 0., 4., ..., 16., 2., 0.],\n [ 0., 0., 5., ..., 12., 0., 0.]],\n\n [[ 0., 0., 10., ..., 1., 0., 0.],\n [ 0., 2., 16., ..., 1., 0., 0.],\n [ 0., 0., 15., ..., 15., 0., 0.],\n ...,\n [ 0., 4., 16., ..., 16., 6., 0.],\n [ 0., 8., 16., ..., 16., 8., 0.],\n [ 0., 1., 8., ..., 12., 1., 0.]]])"
}, },
"execution_count": 311, "execution_count": 527,
"metadata": {}, "metadata": {},
"output_type": "execute_result" "output_type": "execute_result"
} }
...@@ -133,8 +133,8 @@ ...@@ -133,8 +133,8 @@
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"ExecuteTime": { "ExecuteTime": {
"end_time": "2023-05-25T14:28:54.638690Z", "end_time": "2023-05-25T23:09:23.099274Z",
"start_time": "2023-05-25T14:28:54.563741Z" "start_time": "2023-05-25T23:09:23.029118Z"
} }
} }
}, },
...@@ -159,12 +159,12 @@ ...@@ -159,12 +159,12 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 312, "execution_count": 528,
"outputs": [ "outputs": [
{ {
"data": { "data": {
"text/plain": "<Figure size 640x480 with 6 Axes>", "text/plain": "<Figure size 1600x200 with 6 Axes>",
"image/png": "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\n" "image/png": "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\n"
}, },
"metadata": {}, "metadata": {},
"output_type": "display_data" "output_type": "display_data"
...@@ -173,7 +173,7 @@ ...@@ -173,7 +173,7 @@
"source": [ "source": [
"# Create a pyplot figure and set of subplots to show the images easily\n", "# Create a pyplot figure and set of subplots to show the images easily\n",
"# The figure is useless to me, but you need to have it because of how the subplots function works, sadly\n", "# The figure is useless to me, but you need to have it because of how the subplots function works, sadly\n",
"figure, axes = pyplot.subplots(1, 6)\n", "figure, axes = pyplot.subplots(1, 6, figsize = (16, 2))\n",
"# For every ax in the axes array, of which there would be 6, set the ax, image, and target as appropriate for where you are in the foreach.\n", "# For every ax in the axes array, of which there would be 6, set the ax, image, and target as appropriate for where you are in the foreach.\n",
"for ax, image, target in zip(axes, data.images, data.target):\n", "for ax, image, target in zip(axes, data.images, data.target):\n",
" # Disable the axis so it looks better\n", " # Disable the axis so it looks better\n",
...@@ -186,8 +186,8 @@ ...@@ -186,8 +186,8 @@
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"ExecuteTime": { "ExecuteTime": {
"end_time": "2023-05-25T14:28:54.815532Z", "end_time": "2023-05-25T23:09:23.268477Z",
"start_time": "2023-05-25T14:28:54.702672Z" "start_time": "2023-05-25T23:09:23.041900Z"
} }
} }
}, },
...@@ -202,13 +202,13 @@ ...@@ -202,13 +202,13 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 313, "execution_count": 529,
"outputs": [ "outputs": [
{ {
"data": { "data": {
"text/plain": "array([[ 0., 0., 5., ..., 0., 0., 0.],\n [ 0., 0., 0., ..., 10., 0., 0.],\n [ 0., 0., 0., ..., 16., 9., 0.],\n ...,\n [ 0., 0., 1., ..., 6., 0., 0.],\n [ 0., 0., 2., ..., 12., 0., 0.],\n [ 0., 0., 10., ..., 12., 1., 0.]])" "text/plain": "array([[ 0., 0., 5., ..., 0., 0., 0.],\n [ 0., 0., 0., ..., 10., 0., 0.],\n [ 0., 0., 0., ..., 16., 9., 0.],\n ...,\n [ 0., 0., 1., ..., 6., 0., 0.],\n [ 0., 0., 2., ..., 12., 0., 0.],\n [ 0., 0., 10., ..., 12., 1., 0.]])"
}, },
"execution_count": 313, "execution_count": 529,
"metadata": {}, "metadata": {},
"output_type": "execute_result" "output_type": "execute_result"
} }
...@@ -221,8 +221,33 @@ ...@@ -221,8 +221,33 @@
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"ExecuteTime": { "ExecuteTime": {
"end_time": "2023-05-25T14:28:54.821812Z", "end_time": "2023-05-25T23:09:23.274159Z",
"start_time": "2023-05-25T14:28:54.816152Z" "start_time": "2023-05-25T23:09:23.271666Z"
}
}
},
{
"cell_type": "code",
"execution_count": 530,
"outputs": [
{
"data": {
"text/plain": "True"
},
"execution_count": 530,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Check to see if the first dimension of the data tuples match\n",
"(flat_images.shape[0] == data.target.shape[0])"
],
"metadata": {
"collapsed": false,
"ExecuteTime": {
"end_time": "2023-05-25T23:09:23.278716Z",
"start_time": "2023-05-25T23:09:23.275847Z"
} }
} }
}, },
...@@ -257,19 +282,19 @@ ...@@ -257,19 +282,19 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 314, "execution_count": 531,
"outputs": [], "outputs": [],
"source": [ "source": [
"# We'll start by splitting the data into training and testing, going with a 75% train, 25% test split, a 50/50 split, and a 25% train 75% test split.\n", "# We'll start by splitting the data into training and testing, going with a 75% train, 25% test split, a 50/50 split, and a 25% train 75% test split.\n",
"X_train, X_test, y_train, y_test = train_test_split(flat_images, data.target, test_size=0.25, random_state=42)\n", "X_train, X_test, y_train, y_test = train_test_split(flat_images, data.target, test_size=0.25, random_state=2023)\n",
"X_train2, X_test2, y_train2, y_test2 = train_test_split(flat_images, data.target, test_size=0.50, random_state=42)\n", "X_train2, X_test2, y_train2, y_test2 = train_test_split(flat_images, data.target, test_size=0.50, random_state=2023)\n",
"X_train3, X_test3, y_train3, y_test3 = train_test_split(flat_images, data.target, test_size=0.75, random_state=42)" "X_train3, X_test3, y_train3, y_test3 = train_test_split(flat_images, data.target, test_size=0.75, random_state=2023)"
], ],
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"ExecuteTime": { "ExecuteTime": {
"end_time": "2023-05-25T14:28:54.827791Z", "end_time": "2023-05-25T23:09:23.324546Z",
"start_time": "2023-05-25T14:28:54.824178Z" "start_time": "2023-05-25T23:09:23.282727Z"
} }
} }
}, },
...@@ -284,7 +309,7 @@ ...@@ -284,7 +309,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 315, "execution_count": 532,
"outputs": [], "outputs": [],
"source": [ "source": [
"# First the Gaussian Bayes\n", "# First the Gaussian Bayes\n",
...@@ -305,8 +330,8 @@ ...@@ -305,8 +330,8 @@
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"ExecuteTime": { "ExecuteTime": {
"end_time": "2023-05-25T14:28:54.895348Z", "end_time": "2023-05-25T23:09:23.364390Z",
"start_time": "2023-05-25T14:28:54.831677Z" "start_time": "2023-05-25T23:09:23.290016Z"
} }
} }
}, },
...@@ -321,15 +346,15 @@ ...@@ -321,15 +346,15 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 316, "execution_count": 533,
"outputs": [ "outputs": [
{ {
"name": "stdout", "name": "stdout",
"output_type": "stream", "output_type": "stream",
"text": [ "text": [
"0.8555555555555555\n", "0.8511111111111112\n",
"0.7686318131256952\n", "0.8209121245828699\n",
"0.8086053412462908\n" "0.827893175074184\n"
] ]
} }
], ],
...@@ -350,21 +375,21 @@ ...@@ -350,21 +375,21 @@
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"ExecuteTime": { "ExecuteTime": {
"end_time": "2023-05-25T14:28:54.906409Z", "end_time": "2023-05-25T23:09:23.378547Z",
"start_time": "2023-05-25T14:28:54.897176Z" "start_time": "2023-05-25T23:09:23.364622Z"
} }
} }
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 317, "execution_count": 534,
"outputs": [ "outputs": [
{ {
"name": "stdout", "name": "stdout",
"output_type": "stream", "output_type": "stream",
"text": [ "text": [
"0.9933333333333333\n", "0.9933333333333333\n",
"0.9755283648498332\n", "0.9810901001112347\n",
"0.9599406528189911\n" "0.9599406528189911\n"
] ]
} }
...@@ -386,22 +411,22 @@ ...@@ -386,22 +411,22 @@
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"ExecuteTime": { "ExecuteTime": {
"end_time": "2023-05-25T14:28:54.991884Z", "end_time": "2023-05-25T23:09:23.482178Z",
"start_time": "2023-05-25T14:28:54.906637Z" "start_time": "2023-05-25T23:09:23.378011Z"
} }
} }
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 318, "execution_count": 535,
"outputs": [ "outputs": [
{ {
"name": "stdout", "name": "stdout",
"output_type": "stream", "output_type": "stream",
"text": [ "text": [
"0.9866666666666667\n", "0.9866666666666667\n",
"0.9721913236929922\n", "0.9799777530589544\n",
"0.9629080118694362\n" "0.9688427299703264\n"
] ]
} }
], ],
...@@ -422,8 +447,8 @@ ...@@ -422,8 +447,8 @@
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"ExecuteTime": { "ExecuteTime": {
"end_time": "2023-05-25T14:28:55.205099Z", "end_time": "2023-05-25T23:09:23.724591Z",
"start_time": "2023-05-25T14:28:54.990989Z" "start_time": "2023-05-25T23:09:23.486829Z"
} }
} }
}, },
...@@ -439,20 +464,21 @@ ...@@ -439,20 +464,21 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 319, "execution_count": 536,
"outputs": [ "outputs": [
{ {
"name": "stdout", "name": "stdout",
"output_type": "stream", "output_type": "stream",
"text": [ "text": [
"Best single model : K Nearest Neighbour 1st Model\n", "Best single model : K Nearest Neighbour 1st Model\n",
"Best overall algorithm : K Nearest Neighbour Algorithm\n" "Best overall algorithm : Support Vector Classification Algorithm\n",
"Best split ratio : 25% Test Split\n"
] ]
}, },
{ {
"data": { "data": {
"text/plain": "<Figure size 640x480 with 2 Axes>", "text/plain": "<Figure size 640x480 with 2 Axes>",
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAfIAAAHgCAYAAABej+9AAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAA9hAAAPYQGoP6dpAABwPElEQVR4nO3deVhUZf8G8PsAwwy7gOwii7kvuWCKey70umVZpqmlYYuplVlqZiqWiloZmm+aZkrm1qKmLS6VYr6pAWlu5PLTlFQEFBgEWWbm+f1BTI6gzjAwZ4a5P9d1rpozZ7k9A3znec5zzpGEEAJERERkkxzkDkBERERVx0JORERkw1jIiYiIbBgLORERkQ1jISciIrJhLOREREQ2jIWciIjIhrGQExER2TAWciIiIhtmV4V8zZo1kCQJKSkpBvOzs7MRFRUFd3d37N69+47r7927F5IkQZIkHDhwoML7o0ePhru7e7XnlsNHH32ENWvWGL18eHg4JEnC2LFjK7xXfty++uork3P89ddfkCTJpCy3kiQJEyZMuOdycXFxkCQJ2dnZVdqPpX322WcYNmwYGjduDAcHB4SHh1d5W+vXr0dCQoLRy/fo0QOSJCEyMhKV3Rhy3759+t+Tqn5ulSn//f3rr79MXrf88zXWL7/8gieeeAIhISFwdnaGl5cXOnXqhGXLlqGgoMDk/Zvip59+QlRUFNzc3CBJErZu3Vqt2zf3d8oc5Z+Dg4MDzp07V+H9goICeHp6QpIkjB49ukr7mDdvnsnHzJyfLWtgV4W8Mn///Te6du2Kc+fO4ccff0SfPn2MWm/KlCk1nExephbycqtWrcKpU6eqLUdQUBAOHDiA/v37V9s2a4O1a9fixIkTeOCBB9CgQQOztmVqIQcADw8PnD9/Hj///HOF9z799FN4enqalUlOs2bNQrdu3XDp0iW888472L17NzZu3IhevXohLi4Ob731Vo3tWwiBJ554AgqFAtu2bcOBAwfQvXv3at2HNfxOubu7Y/Xq1RXmf/nllygtLYVCoajytqtSyPv3748DBw4gKCioyvuVk10X8jNnzqBz587Iy8tDUlISOnbsaNR6//nPf7B//35s3769hhMa5+bNm5W2jCwtOjoabm5uePPNN6ttm0qlEh07doSfn1+1bdMWlJaWQqPR3PH9nTt34tixY1i7di0aN25swWRl6tevj44dO+LTTz81mJ+fn48vv/wSQ4cOtXim6vDll1/i7bffxpgxY7B//37Exsaie/fu6Nu3L9555x2cPXsWffv2rbH9X758GdevX8ejjz6KXr16oWPHjvD29q7WfVjD79TQoUORmJgInU5nMH/VqlV49NFH4ezsbJEc5X87/fz80LFjRyiVSovst7rZbSE/cuQIunTpAicnJ+zfvx8tW7Y0et3Ro0ejWbNmmDZtGrRa7T2X37Rpk77Iubu746GHHsLhw4cNlklJScGwYcMQHh4OFxcXhIeH48knn8SFCxcMlivvAtq1axdiY2Ph5+cHV1dXFBcXG72vc+fOYdiwYQgODoZSqURAQAB69eqFI0eOACjrJj9x4gSSkpL0XaTGdN36+PjgjTfewObNm3Hw4MF7Ln/mzBkMHz4c/v7+UCqVaNq0Kf773/8aLHOnbsBvvvkGrVq1glKpRGRkJBYvXnzX7tO1a9eiadOmcHV1xf33349vv/220uXS09MxePBgeHp6wsvLCyNHjkRWVpbBMjqdDgsXLkSTJk2gVCrh7++Pp59+Gn///bfBcuHh4ZV2D/bo0QM9evTQvy4/9bB27Vq89tprCAkJgVKpxNmzZ+9w5AAHB+N+dbOysvD8888jNDQUSqUSfn5+6Ny5M3788Ud9lu+++w4XLlzQf9bGdkHHxsZi8+bNyM3N1c/buHEjAGDYsGGVrrN//3706tULHh4ecHV1RadOnfDdd99VWO7gwYPo3LkzVCoVgoODMW3aNJSWlla6TWN+5o319ttvw9vbG0uWLKn0OHh4eCAmJkb/uqioCNOmTUNERAScnZ0REhKC8ePHGxwToOxnYcCAAdixYwfatm0LFxcXNGnSxOCLUFxcHOrVqwcAmDp1qsHv3ejRoyv9HazsZ/7LL79Ehw4d4OXlBVdXV0RGRiI2Nlb//p1+p4z5bMr//uzZswcvvvgi6tatC19fXwwePBiXL1++43G9XWxsLNLT0w1OZZ4+fVr/5el2RUVFeO2119C6dWt4eXnBx8cH0dHR+OabbwyWkyQJBQUFSExM1P8sl/+u3e1v5+1d62fOnIGnpyeGDBlisP2ff/4Zjo6OmDFjhtH/Vkuwy0K+f/9+9OjRA/7+/ti/fz8iIyNNWt/R0RHx8fE4ceIEEhMT77rsvHnz8OSTT6JZs2b44osvsHbtWuTn56Nr1644efKkfrm//voLjRs3RkJCAnbu3IkFCxbgypUraN++faXnbWNjY6FQKLB27Vp89dVXUCgURu+rX79+SE1NxcKFC7F7924sW7YMbdq00f/x2bJlCyIjI9GmTRscOHAABw4cwJYtW4w6Nq+88gpCQkLueerh5MmTaN++PY4fP473338f3377Lfr374+XX34Zs2fPvuu6O3bswODBg+Hr64tNmzZh4cKF2LBhwx0/i++++w5Lly7F22+/ja+//ho+Pj549NFHKz1H9+ijj+K+++7DV199hbi4OGzduhUPPfSQQRF58cUXMXXqVPTp0wfbtm3DO++8gx07dqBTp05mnWOfNm0aLl68iOXLl2P79u3w9/ev8rbKPfXUU9i6dStmzpyJXbt24ZNPPkHv3r1x7do1AGWnUDp37ozAwED9Z13Z+I/KDBs2DI6OjtiwYYN+3qpVq/D4449X2rWelJSEnj17Ii8vD6tWrcKGDRvg4eGBgQMHYtOmTfrlTp48iV69eiE3Nxdr1qzB8uXLcfjwYcyZM6fCNo39mTfGlStXcPz4ccTExMDV1fWeywsh8Mgjj+C9997DU089he+++w6TJk1CYmIievbsqf9yXe6PP/7Aa6+9hldffVX/RXTMmDHYt28fAODZZ5/F5s2bAQAvvfSSSb935Q4cOIChQ4ciMjISGzduxHfffYeZM2fetXcHMP6zKffss89CoVBg/fr1WLhwIfbu3YuRI0canbNhw4bo2rWrwReZTz/9FOHh4ejVq1eF5YuLi3H9+nW8/vrr2Lp1KzZs2IAuXbpg8ODB+Oyzzwz+/S4uLujXr5/+Z/mjjz4y2FZlfzsry7dy5Up89dVXWLJkCQAgIyMDw4cPR9euXREXF2f0v9UihB1ZvXq1ACAACC8vL5GZmWnS+nv27BEAxJdffimEEKJLly6iXr164ubNm0IIIUaNGiXc3Nz0y1+8eFE4OTmJl156yWA7+fn5IjAwUDzxxBN33JdGoxE3btwQbm5uYvHixRX+DU8//bTB8sbuKzs7WwAQCQkJd/23Nm/eXHTv3v2uy9wqLCxM9O/fXwghxMqVKwUAsX37diFExeMmhBAPPfSQqFevnsjLyzPYzoQJE4RKpRLXr18XQghx/vx5AUCsXr1av0z79u1FaGioKC4uNvh3+vr6itt/pAGIgIAAoVar9fMyMjKEg4ODiI+P18+bNWuWACBeffVVg/XXrVsnAIjPP/9cCCFEWlqaACDGjRtnsNyhQ4cEAPHmm28aHJNRo0ZVOFbdu3c3OLblx6dbt24VljVG//79RVhYWKXvubu7i4kTJ1Z5/cp0795dNG/eXAhR9jMfFRUlhBDixIkTAoDYu3evSE5OrvC5dezYUfj7+4v8/Hz9PI1GI1q0aCHq1asndDqdEEKIoUOHChcXF5GRkWGwXJMmTQQAcf78eSGEab9f5Z/v3Rw8eFAAEG+88YZRx2HHjh0CgFi4cKHB/E2bNgkAYsWKFfp5YWFhQqVSiQsXLujn3bx5U/j4+IgXXnhBP6/85/3dd9812OaoUaMq/Yxu/3e99957AoDIzc29Y+7KfqeM/WzK//7c/vO/cOFCAUBcuXLljvu9NW9WVpZYvXq1UCqV4tq1a0Kj0YigoCARFxcnhBDCzc2t0t+dW7OVlpaKMWPGiDZt2hi8d6d17/S389b3yn+2yr344ovC2dlZHDhwQPTs2VP4+/uLy5cv3/XfKAe7bJE//PDDyMvLw8SJE43qGr+TBQsW4O+//8bixYsrfX/nzp3QaDR4+umnodFo9JNKpUL37t2xd+9e/bI3btzA1KlTcd9998HJyQlOTk5wd3dHQUEB0tLSKmz7scceq9K+fHx80KBBA7z77rtYtGgRDh8+XOE8lbmeeeYZNGvWDG+88Ual2y4qKsJPP/2ERx99FK6urgZ5+/Xrh6Kiojt2zRcUFCAlJQWPPPKIwXk0d3d3DBw4sNJ1HnzwQXh4eOhfBwQEwN/fv8JpCwAYMWKEwesnnngCTk5O2LNnDwDo/3t7l/kDDzyApk2b4qeffqo0gzFu/0yrwwMPPIA1a9Zgzpw5OHjw4B27p6sqNjYWKSkpOHbsGFatWoUGDRqgW7duFZYrKCjAoUOH8Pjjjxtc2eHo6IinnnoKf//9t36Q5J49e9CrVy8EBAQYLHf7eXdTfr9qQvlAv9t/FoYMGQI3N7cKPwutW7dG/fr19a9VKhUaNWpU6c9hVbVv3x5A2c/tF198gUuXLt1zHVM+m3IPP/ywwetWrVoBgEn/liFDhsDZ2Rnr1q3D999/j4yMjLuOVP/yyy/RuXNnuLu7w8nJCQqFAqtWrar07+PdmPJ79sEHH6B58+Z48MEHsXfvXnz++edWOSDOLgv5jBkzMHPmTKxfvx4jR46scjHv1KkTHnnkEcyfPx85OTkV3r969SqAsl8uhUJhMG3atMmgG3b48OFYunQpnn32WezcuRO//fYbkpOT4efnh5s3b1bY9u0/TMbuS5Ik/PTTT3jooYewcOFCtG3bFn5+fnj55ZeRn59fpeNwO0dHR8ybN++Opx6uXbsGjUaDDz/8sELWfv36AcAdu6hzcnIghDD4I1+usnkA4OvrW2GeUqms9LgGBgYavHZycoKvr6++K7r8v5X9MgcHB+vfr4qa+AOxadMmjBo1Cp988gmio6Ph4+ODp59+GhkZGdWy/W7duqFhw4b4+OOPsXbtWsTGxlZ6brn8c7vTcQNgcIxv/xyAip+NKb9fxigvsufPnzdq+WvXrsHJyanCoDFJkhAYGFjhZ8GUn8Oq6tatG7Zu3ar/glOvXj20aNHC4PTH7Uz5bMrd/m8pHyRmyr/Fzc0NQ4cOxaeffopVq1ahd+/eCAsLq3TZzZs36y8H/Pzzz3HgwAEkJycjNjYWRUVFRu8TMO33TKlUYvjw4SgqKkLr1q2NvqrJ0pzkDiCX2bNnQ5IkzJ49GzqdDuvWrYOTk+mHIz4+Hi1atMC8efMqvFe3bl0AwFdffXXHH1AAyMvLw7fffotZs2bhjTfe0M8vPy9Umdv/WBq7LwAICwvDqlWrAJQNMPniiy8QFxeHkpISLF++/K7rGmvQoEHo3LkzZs2ahRUrVhi85+3trf+2P378+ErXj4iIqHS+t7c3JEnS/xG/VXUUp4yMDISEhOhfazQaXLt2Tf+Hq/y/V65c0Q9MKnf58mX95wCUtbhuP08KlH1JuXW5cqZc52ysunXrIiEhAQkJCbh48SK2bduGN954A5mZmdixY0e17OOZZ57BW2+9BUmSMGrUqEqX8fb2hoODA65cuVLhvfJBUuXHxNfXt9LP8vZ5pvzMGyMoKAgtW7bErl27UFhYeM/z5L6+vtBoNMjKyjIo5kIIZGRk6FvH1eFuP0u3GzRoEAYNGoTi4mIcPHgQ8fHxGD58OMLDwxEdHV1heVM+m+oWGxuLTz75BEePHsW6devuuNznn3+OiIgIbNq0yeD3pLJjci+m/J4dP34cM2fORPv27ZGcnIxFixZh0qRJJu+zptlli7xcXFwcZs+ejS+++ALDhw+/54CQyjRp0gSxsbH48MMPcfHiRYP3HnroITg5OeH//u//EBUVVekElP1gCSEqXPrwySefGN1bYOy+bteoUSO89dZbaNmyJX7//Xf9/OpoKSxYsADp6en6wSLlXF1d8eCDD+Lw4cNo1apVpVkra70AZd/io6KisHXrVpSUlOjn37hx444j0U1x+x+TL774AhqNRj/ytWfPngDK/rDcKjk5GWlpaQYDdcLDw3H06FGD5U6fPl2t19mbon79+pgwYQL69OlTrZ/1qFGjMHDgQEyePNngS9Ct3Nzc0KFDB2zevNlgXzqdDp9//jnq1auHRo0aASg7FfLTTz8ZfFnTarUVBl1V9Wf+bmbMmIGcnBy8/PLLlV7SeePGDezatQsA9J/17T8LX3/9NQoKCiodtFVV4eHhyMzMNDgmJSUl2Llz5x3XUSqV6N69OxYsWAAAdxzJb8pnU92io6MRGxuLRx99FI8++ugdl5MkCc7OzgZFOCMjo8KodaD6ejkKCgowZMgQhIeHY8+ePZgwYQLeeOMNHDp0yOxtVze7bZGXmzlzJhwcHDBjxgwIIbBhwwaTW+ZxcXFYt24d9uzZAzc3N/388PBwvP3225g+fTrOnTuH//znP/D29sbVq1fx22+/wc3NDbNnz4anpye6deuGd999F3Xr1kV4eDiSkpKwatUq1KlTx6gMxu7r6NGjmDBhAoYMGYKGDRvC2dkZP//8M44ePWrQG9CyZUts3LgRmzZtQmRkJFQqlUmX6AFA586dMWjQoEp/2RYvXowuXbqga9euePHFFxEeHo78/HycPXsW27dvr/RGI+Xefvtt9O/fHw899BBeeeUVaLVavPvuu3B3d79jD4axNm/eDCcnJ/Tp0wcnTpzAjBkzcP/99+OJJ54AADRu3BjPP/88PvzwQzg4OKBv377466+/MGPGDISGhuLVV1/Vb+upp57CyJEjMW7cODz22GO4cOECFi5cWC3X7548eVI/KjsjIwOFhYX6O+c1a9YMzZo1Q15eHh588EEMHz4cTZo0gYeHB5KTk/Wj/su1bNkSmzdvxrJly9CuXTs4ODiYVASDg4ONugFHfHw8+vTpgwcffBCvv/46nJ2d8dFHH+H48ePYsGGD/o/0W2+9hW3btqFnz56YOXMmXF1d8d///rfCHdWM/Zk3xZAhQzBjxgy88847+PPPPzFmzBg0aNAAhYWFOHToED7++GMMHToUMTEx6NOnDx566CFMnToVarUanTt3xtGjRzFr1iy0adMGTz31lEn7vpuhQ4di5syZGDZsGCZPnoyioiIsWbKkwhf9mTNn4u+//0avXr1Qr1495ObmYvHixVAoFHe9sYyxn01NKO8dvJsBAwZg8+bNGDduHB5//HGkp6fjnXfeQVBQEM6cOWOwbMuWLbF3715s374dQUFB8PDwqNK9FsaOHYuLFy/qf5bef/99HDhwAMOGDcPhw4eN/ttsETIOtLO48pGJycnJFd6bO3euACAGDx4sSkpKKl2/stHX5d58800BwGDUermtW7eKBx98UHh6egqlUinCwsLE448/Ln788Uf9Mn///bd47LHHhLe3t/Dw8BD/+c9/xPHjxyuMfL7bv8GYfV29elWMHj1aNGnSRLi5uQl3d3fRqlUr8cEHHwiNRqPfzl9//SViYmKEh4eHAHDPUc23jlq/1cmTJ4Wjo2Olx+38+fMiNjZWhISECIVCIfz8/ESnTp3EnDlzDJbBbSNshRBiy5YtomXLlsLZ2VnUr19fzJ8/X7z88svC29vbYDkAYvz48ZXmvfW4lo+mTU1NFQMHDhTu7u7Cw8NDPPnkk+Lq1asG62q1WrFgwQLRqFEjoVAoRN26dcXIkSNFenq6wXI6nU4sXLhQREZGCpVKJaKiosTPP/98x1Hrlf1c3Ul53sqmWbNmCSGEKCoqEmPHjhWtWrUSnp6ewsXFRTRu3FjMmjVLFBQU6Ld1/fp18fjjj4s6deoISZLuObr71lHrd1LZqHUhhPjll19Ez549hZubm3BxcREdO3bUX91wq//973+iY8eOQqlUisDAQDF58mSxYsWKSkcWG/P7Zcyo9VslJSWJxx9/XAQFBQmFQiE8PT1FdHS0ePfddw2ugLh586aYOnWqCAsLEwqFQgQFBYkXX3xR5OTkGGzvTr8ft/8s3GnUuhBCfP/996J169bCxcVFREZGiqVLl1b4d3377beib9++IiQkRDg7Owt/f3/Rr18/8csvv1TYR1U+mzv9/Sn/Gd6zZ8+dDqkQwnDU+t1UNvJ8/vz5Ijw8XCiVStG0aVOxcuXKSj/XI0eOiM6dOwtXV1cBQH987/a38/ZR6+VX3tx+jM6ePSs8PT3FI488ctf8liYJYQW3BCMyU2lpKVq3bo2QkBB91ycRkT2w+651sk1jxoxBnz59EBQUhIyMDCxfvhxpaWl3vBSQiKi2YiEnm5Sfn4/XX38dWVlZUCgUaNu2Lb7//nv07t1b7mhERBbFrnUiIiIbZteXnxEREdk6FnIiIiIbxkJORERkw1jIiYiIbBgLORERkQ1jISciIrJhLOREREQ2jIWciIjIhrGQExER2TAWciIiIhvGQk5ERGTDWMiJiIhsGAs5ERGRDWMhJyIismEs5ERERDaMhZyIiMiGsZATERHZMBZyIiIiG8ZCTkREZMNYyImIiGwYCzkREZENYyEnIiKyYSzkRERENoyFnIiIyIaxkBMREdkwFnIiIiIbxkJORERkw1jIiYiIbBgLORERkQ1jISciIrJhTnIHMIdOp8Ply5fh4eEBSZLkjkNERCYSQiA/Px/BwcFwcKi5tmVRURFKSkrM3o6zszNUKlU1JKo+Nl3IL1++jNDQULljEBGRmdLT01GvXr0a2XZRUREiwtyRkak1e1uBgYE4f/68VRVzmy7kHh4eAICBW5+Ews1Z5jT/ut4nV+4IREQ2QYNS7Mf3+r/nNaGkpAQZmVpcSA2Hp0fVW/3qfB3C2v2FkpISFvLqUt6drnBztqpC7iQp5I5ARGQbRNl/LHF61N1DgrtH1fejg3WewrXpQk5ERGQsrdBBK8xb3xqxkBMRkV3QQUCHqldyc9atSbz8jIiIyIaxRU5ERHZBBx3M6Rw3b+2aw0JORER2QSsEtKLq3ePmrFuT2LVORERkw9giJyIiu1BbB7uxkBMRkV3QQUBbCws5u9aJiIhsGFvkRERkF2pr17rdtshvfnYT1zrnoCChUD+vcNVN5DyZh2u9cnD9P7lQv5KP0hMaWfINGJWNxINp2H7uKJbuOI0WD9yQJQczMRMzMZMtZDJG+ah1cyZrZJeFXJOmQdG2Ejje52gw3zHUEW6TXFHnM094fuQBh0AH5L+aD12OZa8d7P5wDsbOvowNS/wxLqYRjh9yw5x15+EXYv4j+JiJmZiJmWpbJnsneyH/6KOPEBERAZVKhXbt2uGXX36p0f2JQoH82QVwm+oK6bab5ytjnOHcXgHHEEc4RTrC9WVXiAJA+3/mP/rOFIOfz8bODT7Ysd4X6WdVWD4rBFmXFRjw9DWL5mAmZmImZrKFTMbSVcNkjWQt5Js2bcLEiRMxffp0HD58GF27dkXfvn1x8eLFGttnwfuFcI5WwLn93Z9QJkoFir8phuQuVWi51yQnhQ4NWxUiNcnwkX6pSR5oFlVgsRzMxEzMxEy2kMkU2n9GrZszWSNZC/miRYswZswYPPvss2jatCkSEhIQGhqKZcuW1cj+in8sgea0Bq5jXe64TMn/SnCtdw6uP5iLm5uK4JngDoc6ljtMnj5aODoBudmG4xBzs5zg7S/P+XpmYiZmYiZrzWQKrTB/skayFfKSkhKkpqYiJibGYH5MTAx+/fXXStcpLi6GWq02mIylvapDQUIh3Ge6QVLe+ZmyirYK1FnjCc/lHnDuqED+jAKLnyMHgNvHVEgSIPeXQWYyDjMZh5mMw0x0L7IV8uzsbGi1WgQEBBjMDwgIQEZGRqXrxMfHw8vLSz+FhoYavT/tKQ1EjkDemHxc65aDa91yoDmsQdFXxbjWLQfin69akosEx3qOULRwgvs0N8ARKN5eXPV/qInU1x2h1QDefobfbr3qapCTJc/VgszETMzETNaayRQ8R15DJMmwdSyEqDCv3LRp05CXl6ef0tPTjd6Pop0CXms94bXm38mxiSOcY5zhtcYTkuMdWukCEKVG78ZsmlIHnDnqirbd8g3mt+2Wj5MpbpYLwkzMxEzMZAOZTKGDBK0Zkw537s2Vk2xfoerWrQtHR8cKre/MzMwKrfRySqUSSqWySvuT3CQ4RRoOWpNcJDh4ls0XNwVuJhZB0UUBh7oSRJ5A0eZi6LJ0cH7QuUr7rKrNK+pi8pJ0nD7qgrQUN/QbeQ3+IaX47jNfi+ZgJmZiJmayhUz2TrZC7uzsjHbt2mH37t149NFH9fN3796NQYMGWT6QA6C9oEXRD8UQeQKSpwSnpk7w+sijwheAmpa0zRse3lqMePUqfPw1uHBKhbdGRiDzkmW/UDATMzETM9lCJmPpRNlkzvrWSBJCvlvVbNq0CU899RSWL1+O6OhorFixAitXrsSJEycQFhZ2z/XVajW8vLwwePcoKNys54foWuccuSMQEdkEjSjFXnyDvLw8eHp61sg+ymvFoROBcPeo+hnlG/k6dGieUaNZq0LW0QlDhw7FtWvX8Pbbb+PKlSto0aIFvv/+e6OKOBEREVnBQ1PGjRuHcePGyR2DiIhqufJBa+asb41kL+RERESWoBMSdKLqxdicdWuS7JefERERUdWxRU5ERHaBXetEREQ2TAsHaM3oiLbsczCNx0JORER2QZh5jlzwHDkRERFVN7bIiYjILvAcORERkQ3TCgdohRnnyK30Fq3sWiciIrJhbJETEZFd0EGCzoz2qw7W2SRnISciIrvAc+RW7HqfXDhJCrlj6F3Z2lTuCBUEPZImdwQiIqoBtaKQExER3Yv5g93YtU5ERCSbsnPkZjw0xUq71jlqnYiIyIaxRU5ERHZBZ+a91jlqnYiISEY8R05ERGTDdHColdeR8xw5ERGRDWOLnIiI7IJWSNCa8ShSc9atSSzkRERkF7RmDnbTsmudiIjIfsTFxUGSJIMpMDBQ/74QAnFxcQgODoaLiwt69OiBEydOmLwfFnIiIrILOuFg9mSq5s2b48qVK/rp2LFj+vcWLlyIRYsWYenSpUhOTkZgYCD69OmD/Px8k/bBrnUiIrILcnStOzk5GbTCywkhkJCQgOnTp2Pw4MEAgMTERAQEBGD9+vV44YUXjN+HyalqoQGjsjHkxSz4+JfiwmkVls8MxvHf3C2yb9cfcuC6IweOmaUAAE19JW48URfF7cr2f6eHnahH+aPgUV+LZCwn53FiJmZiJmayFmq12uC1UqmEUqmsdNkzZ84gODgYSqUSHTp0wLx58xAZGYnz588jIyMDMTExBtvp3r07fv31V5MKud13rXd/OAdjZ1/GhiX+GBfTCMcPuWHOuvPwCymxyP61vk7If8of2e+FI/u9cJS0dIV3fDqcLhYDAK6ubmgw5b4UBCEBRdEeFslXTu7jxEzMxEzMZC4d/h25XpVJ9892QkND4eXlpZ/i4+Mr3V+HDh3w2WefYefOnVi5ciUyMjLQqVMnXLt2DRkZGQCAgIAAg3UCAgL07xlL1kK+b98+DBw4EMHBwZAkCVu3brV4hsHPZ2PnBh/sWO+L9LMqLJ8VgqzLCgx4+ppF9l/8gAeKo9yhDVFCG6JE/kh/CJUDFKduAgB03k4Gk+pQPkpauEIb6GyRfOXkPk7MxEzMxEzmKr8hjDkTAKSnpyMvL08/TZs2rdL99e3bF4899hhatmyJ3r1747vvvgNQ1oVeTpIML2kTQlSYdy+yFvKCggLcf//9WLp0qSz7d1Lo0LBVIVKTDFu3qUkeaBZVYPlAWgHVL3mQigRKm7hUeNshVwNl6g0U9q5j0VhWd5yYiZmYiZlk5OnpaTDdqVv9dm5ubmjZsiXOnDmjP29+e+s7MzOzQiv9XmQ9R963b1/07dtXtv17+mjh6ATkZhsehtwsJ3j7ayyWw+mvIvi+8RekEgGhckDOG/WgCa34g+Hycx6Ei4PFu9Wt5TgxEzMxEzOZw/x7rZvX9i0uLkZaWhq6du2KiIgIBAYGYvfu3WjTpg0AoKSkBElJSViwYIFJ27WpwW7FxcUoLi7Wv759wEFV3X4ffEkCLHndvyZEiewPIuFQoIXqQD68llzG9blhFYq560+5uNnNC3CWpyNF7uNUGWYyDjMZh5mMY42ZjGHp55G//vrrGDhwIOrXr4/MzEzMmTMHarUao0aNgiRJmDhxIubNm4eGDRuiYcOGmDdvHlxdXTF8+HCT9mNThTw+Ph6zZ8+utu2prztCqwG8/Qy/SXrV1SAny4KHRiFBG+QMLYDS+1ygOHMTrtuvQz0u6N9FThTC6VIJcl4PsVyuf1jNcWImZmImZjKDpVvkf//9N5588klkZ2fDz88PHTt2xMGDBxEWFgYAmDJlCm7evIlx48YhJycHHTp0wK5du+DhYVqvq02NWp82bZrBAIP09HSztqcpdcCZo65o283w4vu23fJxMsXNrG2bRQBSqeHXW9cfc1HSQAVNhMricazxODETMzETM1m7jRs34vLlyygpKcGlS5fw9ddfo1mzZvr3JUlCXFwcrly5gqKiIiQlJaFFixYm78f6v0Ld4m7X6lXV5hV1MXlJOk4fdUFaihv6jbwG/5BSfPeZZa7R9libiaK27tDVdYJ0UweX/Wo4nyjE9Zmh+mWkQi1Uv6qR/4xpAyCqk9zHiZmYiZmYyVzm3xDGOtu+NlXIa0LSNm94eGsx4tWr8PHX4MIpFd4aGYHMS5a5vMshV4M6CZfhmKOBzs0BmjAlrs8MRUnrf2+uoPpFDUkAN7t6WiRTZeQ+TszETMzETObSCQk6M55gZs66NUkS4vZhC5Zz48YNnD17FgDQpk0bLFq0CA8++CB8fHxQv379e66vVqvh5eWFHhgEJ0lR03GNdmVrU7kjVHCnO8QREclJI0qxF98gLy8Pnp4101gprxULk7vCxb3q7debNzSY0v6XGs1aFbK2yFNSUvDggw/qX0+aNAkAMGrUKKxZs0amVEREVBvpzOxa17FrvaIePXpAxg4BIiKyI1V9gtmt61sj60xFRERERrH7wW5ERGQftJCgNeOGMOasW5NYyImIyC6wa52IiIisDlvkRERkF7Qwr3tcW31RqhULORER2YXa2rXOQk5ERHZB7seY1hTrTEVERERGYYuciIjsgjDzeeSCl58RERHJh13rREREZHXYIq8B1vikseztjeSOUEHdgafljkBEdqS2PsaUhZyIiOyC1synn5mzbk2yzlRERERkFLbIiYjILrBrnYiIyIbp4ACdGR3R5qxbk6wzFRERERmFLXIiIrILWiFBa0b3uDnr1iQWciIisgs8R05ERGTDhJlPPxO8sxsRERFVN7bIiYjILmghQWvGg0/MWbcmsZATEZFd0AnzznPrRDWGqUbsWiciIrJhbJEDGDAqG0NezIKPfykunFZh+cxgHP/N3W4zqb7PheqHXDhc1QAAtPWdUTjMF6VRbmUL3NTBLTELzgcL4JCvhdZfgaKBdVDUr45F8t2Knx0zMRMzGUtn5mA3c9atSdaZyoK6P5yDsbMvY8MSf4yLaYTjh9wwZ915+IWU2G0mXV0nFIyqi9wP6iP3g/oobeUKz7mX4HihGADg/kkmnH8vxI3XApHzUThuDqoDt48z4XzwhkXylZP7ODETMzGTdWQylg6S2ZM1krWQx8fHo3379vDw8IC/vz8eeeQRnDp1yqIZBj+fjZ0bfLBjvS/Sz6qwfFYIsi4rMODpaxbNYU2ZSh5wR2mUO3QhztCFOKPw6boQKgc4nSoCADj9WYSinp4obekKXYACxf+pA22EEk5niyySr5zcx4mZmImZrCOTvZO1kCclJWH8+PE4ePAgdu/eDY1Gg5iYGBQUFFhk/04KHRq2KkRqkofB/NQkDzSLskwGq8+kFXDep4ZUJKBpogIAlDZzgfOhG3C4VgoIAcXRQjhcLkFJGzeLxbK648RMzMRMsmQyRfmd3cyZrJGs58h37Nhh8Hr16tXw9/dHamoqunXrVuP79/TRwtEJyM02PAy5WU7w9tfU+P6tOZPjX8WoM/kiUCIgXBygnh4EbX0lAKDgeX+4L70Kn9HnIRwBSBJuvBQATXMXi+WzluPETMzETPJmMkVtPUduVYPd8vLyAAA+Pj6Vvl9cXIzi4mL9a7VaXS37FbddUiBJAGS+zEDuTNoQZ+QsDoNDgQ7Ov+bD44OryItXQFtfCZftOXA6dRPqGcHQ+imgOFEIt+VXofNxRGlry7XKAfmPU2WYyTjMZBxmonuxmq8XQghMmjQJXbp0QYsWLSpdJj4+Hl5eXvopNDTUrH2qrztCqwG8/Qy/SXrV1SAnS57vOFaTSSFBF+wMTUMVCkf5QROhhGpbLlCsg+vabBSM8UPJA+7QRihRNMAbJV084LIlx2LxrOY4MRMzMZOsmUyhg6S/33qVJg52u7sJEybg6NGj2LBhwx2XmTZtGvLy8vRTenq6WfvUlDrgzFFXtO2WbzC/bbd8nEyxbMvSmjMBAISAVCogaQUkDf75Cn7L2w4SoLNcHGs8TszETMxk3YSZI9aFlRZyq/gK9dJLL2Hbtm3Yt28f6tWrd8fllEollEplte5784q6mLwkHaePuiAtxQ39Rl6Df0gpvvvMt1r3Y0uZXD/LRkk7V+jqKiDd1EG5Lx+K4zehjvOBcHVEaQsXuK3OQoFSKutaP14I1R41Csb4WSRfObmPEzMxEzNZRyZj8elnNUAIgZdeeglbtmzB3r17ERERYfEMSdu84eGtxYhXr8LHX4MLp1R4a2QEMi85WzyLtWRyyNXAY1EGHK5rIdwcoAlXQh0XgtJ/RqWrpwTBLTEb7u9dgcMNHbR+Tih4qi6K+npZJF85uY8TMzETM1lHJnsnCXH7sAXLGTduHNavX49vvvkGjRs31s/38vKCi8u9R0Cr1Wp4eXmhBwbBSVLUZFSbl729kdwRKqg78LTcEYhIZhpRir34Bnl5efD09KyRfZTXikd3PwOFW9W/cJQWlGBLn9U1mrUqZG2RL1u2DADQo0cPg/mrV6/G6NGjLR+IiIhqLXat1wAZOwOIiIhqBasY7EZERFTTzL1furVefsZCTkREdqG2dq1bzXXkREREZDq2yImIyC7U1hY5CzkREdmF2lrI2bVORERkw9giJyIiu1BbW+Qs5EREZBcEzLuEzFrvfMJCTkREdqG2tsh5jpyIiMiGsUVORER2oba2yFnI7YQ1Pmms3WGd3BEqSG3DTiqi2krOQh4fH48333wTr7zyChISEgCUPW9k9uzZWLFiBXJyctChQwf897//RfPmzU3aNv9qERER1aDk5GSsWLECrVq1Mpi/cOFCLFq0CEuXLkVycjICAwPRp08f5Ofnm7R9FnIiIrIL5S1ycyZT3bhxAyNGjMDKlSvh7e2tny+EQEJCAqZPn47BgwejRYsWSExMRGFhIdavX2/SPljIiYjILgghmT0BgFqtNpiKi4vvuM/x48ejf//+6N27t8H88+fPIyMjAzExMfp5SqUS3bt3x6+//mrSv4uFnIiIyAShoaHw8vLST/Hx8ZUut3HjRvz++++Vvp+RkQEACAgIMJgfEBCgf89YHOxGRER2obqeR56eng5PT0/9fKVSWWHZ9PR0vPLKK9i1axdUKtUdtylJhnmEEBXm3QsLORER2YXqGrXu6elpUMgrk5qaiszMTLRr104/T6vVYt++fVi6dClOnToFoKxlHhQUpF8mMzOzQiv9Xti1TkREVM169eqFY8eO4ciRI/opKioKI0aMwJEjRxAZGYnAwEDs3r1bv05JSQmSkpLQqVMnk/bFFjkREdmFWwesVXV9Y3l4eKBFixYG89zc3ODr66ufP3HiRMybNw8NGzZEw4YNMW/ePLi6umL48OEm5WIhJyIiu2Btd3abMmUKbt68iXHjxulvCLNr1y54eHiYtB0WciIisguWbJFXZu/evQavJUlCXFwc4uLizNouz5ETERHZMLbIiYjILggzu9bNbZHXFLbIAQwYlY3Eg2nYfu4olu44jRYP3JA7EjPdxZVVZQ83SX/X8Jfq5jng7CsSDneVcLizhD+fllByxfL5rOU4MRMz2VMmYwgAQpgxyf0PuAO7L+TdH87B2NmXsWGJP8bFNMLxQ26Ys+48/EJKmMkKMxWcALI3S3BpaPgrVZwOnIqVoIoAGq8UaLZJIOg5AanifRpqlLUcJ2ZiJnvKZO9kLeTLli1Dq1at9BfXR0dH44cffrBohsHPZ2PnBh/sWO+L9LMqLJ8VgqzLCgx4+ppFczDTvWkLgfNvSgibIeB4270YLi2V4NUFqDdRwLUJoKwHeHUFFD4WiwfAOo4TMzGTvWUyVvmd3cyZrJGshbxevXqYP38+UlJSkJKSgp49e2LQoEE4ceKERfbvpNChYatCpCYZDvVPTfJAs6gCi2RgJuNdjJfg1RXw7Gg4X+iAvP2Aqr7AmXES/ugpIe0pCbl7LBYNgPUcJ2ZiJnvKZIrqemiKtZG1kA8cOBD9+vVDo0aN0KhRI8ydOxfu7u44ePBgpcsXFxdXeOqMOTx9tHB0AnKzDcf85WY5wdtfY9a2mal6Xd8BFP4JhLxU8SyV5jqgK5SQsVqCZyeBhssEvB8U+L/XJOSnWCQeAOs4TszETPaWiazoHLlWq8XGjRtRUFCA6OjoSpeJj483eOJMaGhotexb3FYbJAmyj2pgpn+VZADp70qImCPgUMk5b6Er+69XDyBgJODaGAiMLetaz/rK8t+g+dkZh5mMw0zVR47nkVuC7JefHTt2DNHR0SgqKoK7uzu2bNmCZs2aVbrstGnTMGnSJP1rtVptVjFXX3eEVgN4+xl+k/Sqq0FOljyHhpkqKkwDNNclpI24ZaZWwo3fBTI3SWjzqwCcBFwiDf+SqCKBG4drPJ6e3MeJmZjJHjOZonz0uTnrWyPZW+SNGzfGkSNHcPDgQbz44osYNWoUTp48WemySqVSPzDOmKfP3Ium1AFnjrqibbd8g/ltu+XjZIqbWdtmpurj8QDQ7Esdmm0U+sm1mYBPP6DZRgEHZ8CtGVB0wfDbcvEFwDnoDhutAXIfJ2ZiJnvMRFbQInd2dsZ9990HAIiKikJycjIWL16Mjz/+2CL737yiLiYvScfpoy5IS3FDv5HX4B9Siu8+87XI/pnp3hzdAJf7DOc5uABOXv/ODxglcH6qBPe2gEcUoP4VyN1XdimaJfGzYyZmsl5y36K1psheyG8nhEBxcbHF9pe0zRse3lqMePUqfPw1uHBKhbdGRiDzkrPFMjCT+bx7AtrpAhmfSkhfCKjCgAbvCri3sWwOazxOzMRMtT2TsWprIZeEkK/X/80330Tfvn0RGhqK/Px8bNy4EfPnz8eOHTvQp0+fe66vVqvh5eWFHhgEJ0lhgcRUndod1skdoYLUNrKfbSKyKxpRir34Bnl5eWafLr2T8lrReP0bcHSt+l2itIXFODV8fo1mrQpZW+RXr17FU089hStXrsDLywutWrUyuogTERGRzIV81apVcu6eiIjsSG0dtW5158iJiIhqQlkhN+cceTWGqUY8IUhERGTD2CInIiK7UFtHrbOQExGRXRAw706yVtqzzq51IiIiW8YWORER2QV2rRMREdmyWtq3zkJORET2wcwWOay0Rc5z5ERERDaMLXIiIrILvLMbERGRDeNgN6JqZo1PGlt+Yb/cESoYG9ZF7ghEZMVYyImIyD4IybwBa2yRExERyae2niO3vr5NIiIiMhpb5EREZB94QxgiIiLbZdej1pcsWWL0Bl9++eUqhyEiIiLTGFXIP/jgA6M2JkkSCzkREVkvK+0eN4dRhfz8+fM1nYOIiKhG1dau9SqPWi8pKcGpU6eg0WiqMw8REVHNENUwWSGTC3lhYSHGjBkDV1dXNG/eHBcvXgRQdm58/vz51R6QiIiI7szkQj5t2jT88ccf2Lt3L1QqlX5+7969sWnTpmoNR0REVH2kapisj8mXn23duhWbNm1Cx44dIUn//qOaNWuG//u//6vWcERERNWG15GXycrKgr+/f4X5BQUFBoXdlgwYlY0hL2bBx78UF06rsHxmMI7/5s5MzHRX2z+oj+8S6hvM8/QrwcKU3wAAa15riINfBRi8H9FGjalbj1ok36342TFTbc9kz0zuWm/fvj2+++47/evy4r1y5UpER0dXXzIL6f5wDsbOvowNS/wxLqYRjh9yw5x15+EXUsJMzHRPwY0KsCD5kH6asfN3g/ebd79u8P6ENSctlq2cNRwnZmImq8DBbmXi4+Mxffp0vPjii9BoNFi8eDH69OmDNWvWYO7cuVUOEh8fD0mSMHHixCpvoyoGP5+NnRt8sGO9L9LPqrB8VgiyLisw4OlrFs3BTLaZycFJwMu/VD95+BpexeGkNHzfrY7lr/KwhuPETMxkFcqffmbOZIVMLuSdOnXC//73PxQWFqJBgwbYtWsXAgICcODAAbRr165KIZKTk7FixQq0atWqSutXlZNCh4atCpGa5GEwPzXJA82iCiyahZlsM1PmeRdMbd8e0ztH4ZMJjZF1UWnw/umDXpjc9gHM7NEOa6feB3W2wmLZAOs5TszETFRzqnSv9ZYtWyIxMbFaAty4cQMjRozAypUrMWfOnLsuW1xcjOLiYv1rtVpt1r49fbRwdAJysw0PQ26WE7z95bk+nplsJ1NE63yMXnQaAZE3oc5W4PsP6+Pdwfdj5u7f4e6tQYseOWjXLxs+9YpxLV2Fbe/XR8KTLTDt2yNQKC3TR2cNx4mZmMla1NbHmFapkGu1WmzZsgVpaWmQJAlNmzbFoEGD4ORk+ubGjx+P/v37o3fv3vcs5PHx8Zg9e3ZVIt/V7R+OJEH2cyHMZBw5M7V4MEf//yEAItuewIxuUTj4lT96P3cZUQOz/32/cSHCWubjzc7tcfxnH7Tpa9luSH52xmEm41hjJqNw1HqZ48ePY9CgQcjIyEDjxo0BAKdPn4afnx+2bduGli1bGr2tjRs34vfff0dycrJRy0+bNg2TJk3Sv1ar1QgNDTXtH3AL9XVHaDWAt5/hN0mvuhrkZMnzYDhmst1MSlcdghsXIPMvl0rf9woohU9IMTL/UlX6fk2wxuPETMxE1cvkc+TPPvssmjdvjr///hu///47fv/9d6Snp6NVq1Z4/vnnjd5Oeno6XnnlFXz++ecGN5a5G6VSCU9PT4PJHJpSB5w56oq23fIN5rftlo+TKW5mbZuZ7C9TabGEjLOu8PKvfPTujRwn5FxRwsu/1GKZrPE4MRMzyaaWDnYz+SvUH3/8gZSUFHh7e+vneXt7Y+7cuWjfvr3R20lNTUVmZqbBADmtVot9+/Zh6dKlKC4uhqOjo6nxTLZ5RV1MXpKO00ddkJbihn4jr8E/pBTffeZb4/tmJtvO9NWccLTqfR0+wcXIv1Z2jrzohiM6PpaJogIHfPtBfbTtew2e/iW49rcK3ywMg7t3KVo/ZNludbmPEzMxk7WQRNlkzvrWyORC3rhxY1y9ehXNmzc3mJ+ZmYn77rvP6O306tULx44dM5j3zDPPoEmTJpg6dapFijgAJG3zhoe3FiNevQoffw0unFLhrZERyLzkbJH9M5PtZsrNUGLVS41xI0cBd59SRLbJx5Qtf8C3XjFKihxw+ZQbDm32R6HaCV7+JWgUnYdn//snVO5ai+QrJ/dxYiZmshq19By5JMS9x+HdOjp8//79mDJlCuLi4tCxY0cAwMGDB/H2229j/vz56NevX5XD9OjRA61bt0ZCQoJRy6vVanh5eaEHBsFJsuxlPVQ7Lb+wX+4IFYwN6yJ3BKIaoxGl2ItvkJeXZ/bp0jsprxWhCW/DwaXqY1R0N4uQPnFmjWatCqNa5HXq1DG4/aoQAk888YR+Xvl3gYEDB0KrtWxrg4iIyCjmnue25XPke/bsqekcAIC9e/daZD9ERGSHamnXulGFvHv37jWdg4iIiKrA5MvPyhUWFuLPP//E0aNHDSYiIiKrZOGHpixbtgytWrXSXy4dHR2NH3744d84QiAuLg7BwcFwcXFBjx49cOLECZP/WVV6jOkzzzxjEOZWPEdORERWycJd6/Xq1cP8+fP1V3QlJiZi0KBBOHz4MJo3b46FCxdi0aJFWLNmDRo1aoQ5c+agT58+OHXqFDw8PO6x9X+Z3CKfOHEicnJycPDgQbi4uGDHjh1ITExEw4YNsW3bNlM3R0REVCsNHDgQ/fr1Q6NGjdCoUSPMnTsX7u7uOHjwIIQQSEhIwPTp0zF48GC0aNECiYmJKCwsxPr1603aj8kt8p9//hnffPMN2rdvDwcHB4SFhaFPnz7w9PREfHw8+vfvb+omiYiIal41jVq//YFdSqUSSqWysjX0tFotvvzySxQUFCA6Ohrnz59HRkYGYmJiDLbTvXt3/Prrr3jhhReMjmVyi7ygoAD+/v4AAB8fH2RlZQEoeyLa77//burmiIiILKL8zm7mTAAQGhoKLy8v/RQfH3/HfR47dgzu7u5QKpUYO3YstmzZgmbNmiEjIwMAEBAQYLB8QECA/j1jVenObqdOnUJ4eDhat26Njz/+GOHh4Vi+fDmCgoJM3RwREZFNSU9PN7ghzN1a440bN8aRI0eQm5uLr7/+GqNGjUJSUpL+/Vvv0QKUDYC7fd69mFzIJ06ciCtXrgAAZs2ahYceegjr1q2Ds7Mz1qxZY+rmiIiILKOaBruZ8tAuZ2dn/WC3qKgoJCcnY/HixZg6dSoAICMjw6ARnJmZWaGVfi8mF/IRI0bo/79Nmzb466+/8Oeff6J+/fqoW7euqZsjIiKyG0IIFBcXIyIiAoGBgdi9ezfatGkDACgpKUFSUhIWLFhg0jbNfoCsq6sr2rZta+5miIiIapQEM59+ZuLyb775Jvr27YvQ0FDk5+dj48aN2Lt3L3bs2AFJkjBx4kTMmzcPDRs2RMOGDTFv3jy4urpi+PDhJu3HqEI+adIkoze4aNEikwIQERHVRlevXsVTTz2FK1euwMvLC61atcKOHTvQp08fAMCUKVNw8+ZNjBs3Djk5OejQoQN27dpl0jXkgJGF/PDhw0ZtzNQT9ETWxhqfNLbz8hG5I1TwUHBruSNQFUkK63rcqCQkoNRCO7PwQ1NWrVp11/clSUJcXBzi4uKqnglW9tAUIiKiGlNLH5pS5XutExERkfzMHuxGRERkE2ppi5yFnIiI7MKtd2er6vrWiF3rRERENowtciIisg+1tGu9Si3ytWvXonPnzggODsaFCxcAAAkJCfjmm2+qNRwREVG1EdUwWSGTC/myZcswadIk9OvXD7m5udBqtQCAOnXqICEhobrzERER0V2YXMg//PBDrFy5EtOnT4ejo6N+flRUFI4dO1at4YiIiKpLdT3G1NqYfI78/Pnz+hu830qpVKKgoKBaQhEREVU7C9/ZzVJMbpFHRETgyJEjFeb/8MMPaNasWXVkIiIiqn619By5yS3yyZMnY/z48SgqKoIQAr/99hs2bNiA+Ph4fPLJJzWRkYiIiO7A5EL+zDPPQKPRYMqUKSgsLMTw4cMREhKCxYsXY9iwYTWRscYNGJWNIS9mwce/FBdOq7B8ZjCO/+bOTMxkc5nWvheIzxcFGszz9ivFxj9OAABuFjhg1dwgHNjpBXWOEwLqlWDQmCwMHHXNIvluxc/O9jK1eCAfj79wBQ1bFsI3oBSzn7sPB3Z5y5KlKnhDmFs899xzuHDhAjIzM5GRkYH09HSMGTOmurNZRPeHczB29mVsWOKPcTGNcPyQG+asOw+/kBJmYiabzBTW+CY2HDmun5b//Kf+veWzQpCy1xNTPryIlUl/YvDzWfjorXr4dYenxfIB1nGcmMl0Klctzqe54qOZ9WXZv9lqade6WXd2q1u3Lvz9/au8flxcHCRJMpgCAwPvvWI1Gvx8NnZu8MGO9b5IP6vC8lkhyLqswICnLd9CYSZmqg6OjoCPv0Y/1fHV6t9LS3VFnyHXcX+nGwgMLUG/kdcQ2ewmzhx1tVg+wDqOEzOZLmVvHSS+Vw//2+Ejy/6pclUa7BYZGXnHyVTNmzfHlStX9JMlL2FzUujQsFUhUpMMH+KemuSBZlHyjMBnJmYy16XzzniyTXM83aEp5o0Nw5UL/z5/uvkDBTi4ywvZVxQQAjjyP3dcOqdEu+75FstnLceJmeyQuZeeWWmL3ORz5BMnTjR4XVpaisOHD2PHjh2YPHmy6QGcnIxuhRcXF6O4uFj/Wq1Wm7y/W3n6aOHoBORmGx6G3CwnePtrzNo2MzGTHJmatC3A5CU3US+yGDlZTtiwOBCvPtwQK/b8CU8fLca9cwkJk0Mxol1zODoJODgITHwvHS06WK4wWMNxYiY7VUtv0WpyIX/llVcqnf/f//4XKSkpJgc4c+YMgoODoVQq0aFDB8ybN++OLfv4+HjMnj3b5H3ci7jtw5EkyP6BMZNxmMlQ+57/tqwjmgLNos5hdHRT7P7SB4+9kIWtq+riz1RXzF5zDv71SnDsoDuWTqsHH/9StO12wzIh/8HPzjjWmImsS7U9/axv3774+uuvTVqnQ4cO+Oyzz7Bz506sXLkSGRkZ6NSpE65dq/z8z7Rp05CXl6ef0tPTzcqsvu4IrQbw9jP8dutVV4OcLHmeJ8NMzFSdVK46hDcpwqXzShTflLBmfhCej7uMjjFqRDYrwqDYbHR/OBdfLa/6WBdTWeNxYiY7wcFud/fVV1/Bx8e0ARB9+/bFY489hpYtW6J379747rvvAACJiYmVLq9UKuHp6WkwmUNT6oAzR13Rtpvh+cG23fJxMsXNrG0zEzNZQ6aSYgnpZ5Xw8S+FRiNBU+oABwfDv0YOjgJCZ7lM1nicmMk+8Bat/2jTpg0k6d/b1AkhkJGRgaysLHz00UdmhXFzc0PLli1x5swZs7Zjis0r6mLyknScPuqCtBQ39Bt5Df4hpfjuM1+LZWAmZqouK2YHo2NMHvxDSpGb7YT1CQEozHdEnyeuw81Dh1bRN7DynWA4qy4hoF4Jjh5wx49f+eD5WZcskq+c3MeJmapG5apFcPi/45QCQ4sR2awQ+bmOyLqslCUTVaGQP/LIIwavHRwc4Ofnhx49eqBJkyZmhSkuLkZaWhq6du1q1nZMkbTNGx7eWox49Sp8/DW4cEqFt0ZGIPOS871XZiZmsrJM2VcUiB8XDvV1R3j5atCkbSESvj2NgHqlAIBpy/7Cp/OCsGBCfeTnOsE/pASjp16x+OVMch8nZqqaRq0KsHDTKf3rF2aWnd7c/aUv3n/d9KuWqHpIQtw+lOLONBoN1q1bh4ceeqharvd+/fXXMXDgQNSvXx+ZmZmYM2cOkpKScOzYMYSFhd1zfbVaDS8vL/TAIDhJCrPzEFmjnZePyB2hgoeCW8sdgapIUsj3xaQyGlGKPaVfIi8vz+zTpXdSXisaTJsHR5WqytvRFhXh/+LfrNGsVWFSi9zJyQkvvvgi0tLSqmXnf//9N5588klkZ2fDz88PHTt2xMGDB40q4kRERKaorbdoNblrvUOHDjh8+HC1FNuNGzeavQ0iIiJ7ZnIhHzduHF577TX8/fffaNeuHdzcDEdPtmrVqtrCERERVSsrbVWbw+hCHhsbi4SEBAwdOhQA8PLLL+vfkyQJQghIkgStVnunTRAREcnH3u/slpiYiPnz5+P8+fM1mYeIiIhMYHQhLx/czoFoRERkizjYDTC4EQwREZFNsfeudQBo1KjRPYv59evXzQpERERExjOpkM+ePRteXl41lYWIiKjGsGsdwLBhw+Dvb7mnJBEREVWbWtq1bvTTz3h+nIiIyPqYPGqdiIjIJtXSFrnRhVyns+ADi4mIiKoZz5ETkSys8UljjVOs72mDp6JK5Y5gE0RpidwRDAhhwc+tlrbIjT5HTkRERNaHLXIiIrIPtbRFzkJORER2obaeI2fXOhERkQ1ji5yIiOwDu9aJiIhsF7vWiYiIyOqwRU5ERPaBXetEREQ2rJYWcnatExER2TC2yImIyC5I/0zmrG+NWMiJiMg+sGu99howKhuJB9Ow/dxRLN1xGi0euCF3JGZiplqZ6dpqLU5FlSLzfa1+3pU4DU5FlRpMF0ZrZMlnLceJmWpG+eVn5kzWyO4LefeHczB29mVsWOKPcTGNcPyQG+asOw+/EPmeEMRMzFQbM908oUPeFh2UDSu+59ZJQoMdTvqp3mJHi2YDrOc4MROZSvZCfunSJYwcORK+vr5wdXVF69atkZqaarH9D34+Gzs3+GDHel+kn1Vh+awQZF1WYMDT1yyWgZmYqbZn0hUKXJmhRcB0Rzh4VDzTKCkAp7qSfnL0svzZSGs4TsxUw0Q1TFZI1kKek5ODzp07Q6FQ4IcffsDJkyfx/vvvo06dOhbZv5NCh4atCpGa5GEwPzXJA82iCiySgZmYyR4yXV2ghXtnB7h1qPxPTmGqwNk+pTg3uBQZczTQXLfsX0xrOU7MZAG1rIgDMhfyBQsWIDQ0FKtXr8YDDzyA8PBw9OrVCw0aNKh0+eLiYqjVaoPJHJ4+Wjg6AbnZhmP+crOc4O0vzzk6ZmKm2pZJvVOHoj8F6k6o/M+NWycHBM1xROgyJ/hPdETRSYH0sRroSiz3l9MajhMz1T7x8fFo3749PDw84O/vj0ceeQSnTp0yWEYIgbi4OAQHB8PFxQU9evTAiRMnTNqPrIV827ZtiIqKwpAhQ+Dv7482bdpg5cqVd1w+Pj4eXl5e+ik0NLRacojb/l5IEmT/9sVMxmEm48iVqTRDIPN9LYLecYKDsvLucs8YB7h3cYDyPgnu3RxQb4kTSi4CBfstf9D42RnHGjMZw9KD3ZKSkjB+/HgcPHgQu3fvhkajQUxMDAoK/u29WLhwIRYtWoSlS5ciOTkZgYGB6NOnD/Lz843ej6yF/Ny5c1i2bBkaNmyInTt3YuzYsXj55Zfx2WefVbr8tGnTkJeXp5/S09PN2r/6uiO0GsDbz/CbpFddDXKy5Lkyj5mYqTZlKvpTQHsduPCUBqc6lOJUh1Lc/F0gZ6MOpzqUQmgr/mV0qitBEQSUXLRcZZD7ODGThVTTOfLbe4aLi4sr3d2OHTswevRoNG/eHPfffz9Wr16Nixcv6seBCSGQkJCA6dOnY/DgwWjRogUSExNRWFiI9evXG/3PkrWQ63Q6tG3bFvPmzUObNm3wwgsv4LnnnsOyZcsqXV6pVMLT09NgMoem1AFnjrqibTfDbz5tu+XjZIqbWdtmJmZiJsCtvYTwjU4IX/fvpGomwfM/EsLXOUFyrNhK1+YKaK6WFXRLkfs4MZNtCQ0NNegdjo+PN2q9vLw8AICPjw8A4Pz588jIyEBMTIx+GaVSie7du+PXX381Oo+sX6GCgoLQrFkzg3lNmzbF119/bbEMm1fUxeQl6Th91AVpKW7oN/Ia/ENK8d1nvhbLwEzMVFszObhJUN5nOE9SAY51JCjvk6ArFMheoYNHz7LR6qWXBbI+0sGxDuDxoGVHrvOzs91Mxqqux5imp6cbNCSVSuU91xVCYNKkSejSpQtatGgBAMjIyAAABAQEGCwbEBCACxcuGJ1L1kLeuXPnCif+T58+jbCwMItlSNrmDQ9vLUa8ehU+/hpcOKXCWyMjkHnJ2WIZmImZ7C2TngNQfFZA/Z0O2nzAqS7gGiUheJ4THNwsW8it8TgxUzWrpju7VaVHeMKECTh69Cj2799f4T1JMvxZF0JUmHc3khC3D1uwnOTkZHTq1AmzZ8/GE088gd9++w3PPfccVqxYgREjRtxzfbVaDS8vL/TAIDhJCgskJiIAaJxifb9vp6JK5Y5AVaARpdiLb5CXl2f26dI7Ka8VLcfMg6Ozqsrb0ZYU4diqN03O+tJLL2Hr1q3Yt28fIiIi9PPPnTuHBg0a4Pfff0ebNm308wcNGoQ6deogMTHRqO3Leo68ffv22LJlCzZs2IAWLVrgnXfeQUJCglFFnIiIyBSWHrUuhMCECROwefNm/PzzzwZFHAAiIiIQGBiI3bt36+eVlJQgKSkJnTp1Mno/sg8zHDBgAAYMGCB3DCIiqu2qqWvdWOPHj8f69evxzTffwMPDQ39O3MvLCy4uLpAkCRMnTsS8efPQsGFDNGzYEPPmzYOrqyuGDx9u9H5kL+REREQWYeFCXn4FVo8ePQzmr169GqNHjwYATJkyBTdv3sS4ceOQk5ODDh06YNeuXfDw8ICxWMiJiIhqgDFD0CRJQlxcHOLi4qq8HxZyIiKyC9V1+Zm1YSEnIiL7YOGudUuR/TGmREREVHVskRMRkV2QhIBkxq1TzFm3JrGQExGRfWDXOhEREVkbtsiJiMgucNQ6ERGRLWPXOhEREVkbtsiJyGTW+KQx6ecQuSNUIHpekjsC3YJd60RERLaslnats5ATEZFdqK0tcp4jJyIismFskRMRkX1g1zoREZFts9bucXOwa52IiMiGsUVORET2QYiyyZz1rRALORER2QWOWiciIiKrwxY5ERHZB45aJyIisl2SrmwyZ31rxK51IiIiG8ZCDmDAqGwkHkzD9nNHsXTHabR44IbckZiJmZjJQsT6fIielyCW5v47b99NiCnZEI9cKXvvbIks2QDrOU7WnskoohomK2T3hbz7wzkYO/syNizxx7iYRjh+yA1z1p2HX4h8v7jMxEzMZBnizxLg2wIg8razjEUCaOEMPOdp0Ty3s5bjZO2ZjFU+at2cyRrJWsjDw8MhSVKFafz48RbLMPj5bOzc4IMd632RflaF5bNCkHVZgQFPX7NYBmZiJmayfCZxUwfMuw68VgfwMPxTKMW4QnraE2intFieyljDcbKFTEYrv47cnMkKyVrIk5OTceXKFf20e/duAMCQIUMssn8nhQ4NWxUiNcnDYH5qkgeaRRVYJAMzMRMzyZRpcS7QQQWpncpy+zSB1RwnK89EMo9a9/PzM3g9f/58NGjQAN27d690+eLiYhQXF+tfq9Vqs/bv6aOFoxOQm214GHKznODtrzFr28zETMxkvZnEz4XAmVJgmb9F9lcV1nCcbCGTKXhDmBpWUlKCzz//HLGxsZAkqdJl4uPj4eXlpZ9CQ0OrZd+395ZIEmQf1MBMxmEm4zDTLfvN1AD/zQPe9IHkXPnfGmvCz64acbBbzdq6dStyc3MxevToOy4zbdo05OXl6af09HSz9qm+7gitBvD2M/wm6VVXg5wseTormImZmKmGnS4FcnTAC5kQvS9B9L4E/FECbCkoe621jr/Wsh8nG8lEVlTIV61ahb59+yI4OPiOyyiVSnh6ehpM5tCUOuDMUVe07ZZvML9tt3ycTHEza9vMxEzMZKWZ2iqBVf7Aylumxgqglwuw0h+So3W00mU/TjaSyRS1ddS6VXyFunDhAn788Uds3rzZ4vvevKIuJi9Jx+mjLkhLcUO/kdfgH1KK7z7ztXgWZmImZqr5TJKrAxBh2IYRKgnwdIAUoSh7rdYBmRog+59beaVrynpVfRwh+TjWeMZy/OyqGZ9+VnNWr14Nf39/9O/f3+L7TtrmDQ9vLUa8ehU+/hpcOKXCWyMjkHnJ2eJZmImZmEm+TAZ+vQkszP339Ts5Zf992gMYbblry63xOFljJnsnCSHvVwydToeIiAg8+eSTmD9/vknrqtVqeHl5oQcGwUlS1FBCIrIF0s8hckeoQPS8JHcEq6cRpdiLb5CXl2f26dI7Ka8V0X3fhpOi6pcbakqLcOCHmTWatSpkb5H/+OOPuHjxImJjY+WOQkREtRmfflYzYmJiIHOnABERkc2SvZATERFZQm29IQwLORER2QedKJvMWd8KsZATEZF9qKXnyK3mhjBERERkOrbIiYjILkgw8xx5tSWpXizkRERkH2rpnd3YtU5ERGTD2CInIiK7wMvPiIiIbBlHrRMREZG1YYuciIjsgiQEJDMGrJmzbk1iISeiWsEanzSWvb2R3BEqqDvwtNwR5KP7ZzJnfSvErnUiIiIbxhY5ERHZBXatExER2bJaOmqdhZyIiOwD7+xGRERE1oaFnIiI7EL5nd3MmUyxb98+DBw4EMHBwZAkCVu3bjV4XwiBuLg4BAcHw8XFBT169MCJEydM/nexkBMRkX0o71o3ZzJBQUEB7r//fixdurTS9xcuXIhFixZh6dKlSE5ORmBgIPr06YP8/HyT9sNz5ERERDWgb9++6Nu3b6XvCSGQkJCA6dOnY/DgwQCAxMREBAQEYP369XjhhReM3g9b5EREZBcknfkTAKjVaoOpuLjY5Cznz59HRkYGYmJi9POUSiW6d++OX3/91aRtsZATEZF9qKau9dDQUHh5eemn+Ph4k6NkZGQAAAICAgzmBwQE6N8zFrvWiYiITJCeng5PT0/9a6VSWeVtSZJk8FoIUWHevbCQExGRfaimG8J4enoaFPKqCAwMBFDWMg8KCtLPz8zMrNBKvxcWcgADRmVjyItZ8PEvxYXTKiyfGYzjv7kzEzMxEzNZLJPq+1yofsiFw1UNAEBb3xmFw3xRGuVWtsBNHdwSs+B8sAAO+Vpo/RUoGlgHRf3qWCTfrazxszOGNd2iNSIiAoGBgdi9ezfatGkDACgpKUFSUhIWLFhg0rbs/hx594dzMHb2ZWxY4o9xMY1w/JAb5qw7D7+QEmZiJmZiJotl0tV1QsGousj9oD5yP6iP0lau8Jx7CY4XygZSuX+SCeffC3HjtUDkfBSOm4PqwO3jTDgfvGGRfOXkPk625MaNGzhy5AiOHDkCoGyA25EjR3Dx4kVIkoSJEydi3rx52LJlC44fP47Ro0fD1dUVw4cPN2k/shZyjUaDt956CxEREXBxcUFkZCTefvtt6HSWe1bc4OezsXODD3as90X6WRWWzwpB1mUFBjx9zWIZmImZmImZSh5wR2mUO3QhztCFOKPw6boQKgc4nSoCADj9WYSinp4obekKXYACxf+pA22EEk5niyySr5zcx8ksFr6OPCUlBW3atNG3uCdNmoQ2bdpg5syZAIApU6Zg4sSJGDduHKKionDp0iXs2rULHh4eJu1H1kK+YMECLF++HEuXLkVaWhoWLlyId999Fx9++KFF9u+k0KFhq0KkJhketNQkDzSLKrBIBmZiJmZipgq0As771JCKBDRNVACA0mYucD50Aw7XSgEhoDhaCIfLJShp42axWFZ3nEwl8O8zyasymdiz3qNHDwghKkxr1qwBUDbQLS4uDleuXEFRURGSkpLQokULk/9Zsp4jP3DgAAYNGoT+/fsDAMLDw7FhwwakpKRUunxxcbHB9Xpqtdqs/Xv6aOHoBORmGx6G3CwnePtrzNo2MzETMzGTqRz/KkadyReBEgHh4gD19CBo65eNiC543h/uS6/CZ/R5CEcAkoQbLwVA09zFYvms5ThVlTWdI69OsrbIu3Tpgp9++gmnT58GAPzxxx/Yv38/+vXrV+ny8fHxBtfuhYaGVkuO2z8bSYLsj6tjJuMwk3GYyThyZ9KGOCNncRjy3quPor5e8PjgKhwvljVeXLbnwOnUTahnBCP3gzAUjKkLt+VXoThi+Zaw3MeJDMnaIp86dSry8vLQpEkTODo6QqvVYu7cuXjyyScrXX7atGmYNGmS/rVarTarmKuvO0KrAbz9DL9JetXVICdLnkPDTMzETHacSSFBF+wMHQBNQxWczhRDtS0XBc/5wXVtNtRvBqO0fdnocG2EEk7niuGyJQelrS3TvW41x6mqBMx8jGm1JalWsrbIN23ahM8//xzr16/H77//jsTERLz33ntITEysdHmlUqm/fq86ruPTlDrgzFFXtO1meIP6tt3ycTLFcuedmImZmImZKiUEpFIBSSsgafBP0/eWtx2ksnO3FmK1x8lYFh7sZimyfoWaPHky3njjDQwbNgwA0LJlS1y4cAHx8fEYNWqURTJsXlEXk5ek4/RRF6SluKHfyGvwDynFd5/5WmT/zMRMzMRMAOD6WTZK2rlCV1cB6aYOyn35UBy/CXWcD4SrI0pbuMBtdRYKlBK0fgoojhdCtUeNgjF+FslXTu7jRBXJWsgLCwvh4GDYKeDo6GjRy8+StnnDw1uLEa9ehY+/BhdOqfDWyAhkXnK2WAZmYiZmYiaHXA08FmXA4boWws0BmnAl1HEhKP1nVLp6ShDcErPh/t4VONzQQevnhIKn6qKor5dF8pWT+ziZRQfAtLufVlzfCklCyNdXMHr0aPz444/4+OOP0bx5cxw+fBjPP/88YmNjjbqzjVqthpeXF3pgEJwkhQUSExEZL3t7I7kjVFB34Gm5IxjQiFLsxTfIy8sz+3TpnZTXil4tpsDJser3Rddoi/HT8YU1mrUqZG2Rf/jhh5gxYwbGjRuHzMxMBAcH44UXXtBfLE9ERER3J2sh9/DwQEJCAhISEuSMQURE9sDcAWsc7EZERCSjWlrI7f6hKURERLaMLXIiIrIPtbRFzkJORET2oZZefsZCTkREdoEPTSEiIiKrwxY5ERHZB54jJyIismE6AUhmFGOddRZydq0TERHZMLbIiYjIPrBrnYiIyJaZ+0xxFnIiqycpbOBRjFZAlJbIHcEmWNuTxgCg3WHruhi6+IYOe7vIncK2sZATEZF9YNc6ERGRDdMJmNU9zlHrREREVN3YIiciIvsgdGWTOetbIRZyIiKyDzxHTkREZMN4jpyIiIisDVvkRERkH9i1TkREZMMEzCzk1ZakWrFrnYiIyIaxRU5ERPaBXetEREQ2TKcDYMa14DrrvI6cXesABozKRuLBNGw/dxRLd5xGiwduyB2JmWw0U4sH8hG36jTW/XYEOy4kIzomR9Y81poJsL7Pjpnu7soqILWNA9LflQzm3zwHnH1FwuGuEg53lvDn0xJKrsgS0W7ZfSHv/nAOxs6+jA1L/DEuphGOH3LDnHXn4Rci39OdmMl2M6lctTif5oqPZtaXLcPtrDGTNX52zHRnBSeA7M0SXBoadi0XpwOnYiWoIoDGKwWabRIIek5AUlo0nvHKu9bNmayQrIU8Pz8fEydORFhYGFxcXNCpUyckJydbNMPg57Oxc4MPdqz3RfpZFZbPCkHWZQUGPH3NojmYqXZkStlbB4nv1cP/dvjIluF21pjJGj87ZqqcthA4/6aEsBkCjp6G711aKsGrC1BvooBrE0BZD/DqCiis50fNEAt59Xv22Wexe/durF27FseOHUNMTAx69+6NS5cuWWT/TgodGrYqRGqSh8H81CQPNIsqsEgGZqo9mcg41vjZMdOdXYyX4NUV8OxoOF/ogLz9gKq+wJlxEv7oKSHtKQm5eywWjf4hWyG/efMmvv76ayxcuBDdunXDfffdh7i4OERERGDZsmWVrlNcXAy1Wm0wmcPTRwtHJyA323DMX26WE7z9NWZtm5nsLxMZxxo/O2aq3PUdQOGfQMhLFVuimuuArlBCxmoJnp0EGi4T8H5Q4P9ek5CfYpF4ptMJ8ycrJFsh12g00Gq1UKlUBvNdXFywf//+SteJj4+Hl5eXfgoNDa2WLLf3lkgSZL/wn5mMY42ZyDjW+Nkx079KMoD0dyVEzBFwqOScd/mDwLx6AAEjAdfGQGBsWdd61ldSxRWsgBA6sydrJFsh9/DwQHR0NN555x1cvnwZWq0Wn3/+OQ4dOoQrVyof8jht2jTk5eXpp/T0dLMyqK87QqsBvP0Mv9161dUgJ0ueK/OYyXYzkXGs8bNjpooK0wDNdQlpIySkRpVNN1IlZG4AUqMkONUB4CTgEmn4rUIVWfYlwCoJM1vjPEde0dq1ayGEQEhICJRKJZYsWYLhw4fD0dGx0uWVSiU8PT0NJnNoSh1w5qgr2nbLN5jftls+Tqa4mbVtZrK/TGQca/zsmKkijweAZl/q0Gyj0E+uzQR8+gHNNgo4OANuzYCiC4at7+ILgHNQjcejW8jadGnQoAGSkpJQUFAAtVqNoKAgDB06FBERERbLsHlFXUxeko7TR12QluKGfiOvwT+kFN995muxDMxUezKpXLUIDi/Wvw4MLUZks0Lk5zoi67I81+RYYyZr/OyYyZCjG+Byn+E8BxfAyevf+QGjBM5PleDeFvCIAtS/Arn7yi5Fs0rCzMeYWmmL3Cr6IN3c3ODm5oacnBzs3LkTCxcutNi+k7Z5w8NbixGvXoWPvwYXTqnw1sgIZF5ytlgGZqo9mRq1KsDCTaf0r1+YWXb6Z/eXvnj/9Uhm+oc1fnbMZDrvnoB2ukDGpxLSFwKqMKDBuwLubeROdgc6HSCZcZ7bSs+RS0LI9xVj586dEEKgcePGOHv2LCZPngylUon9+/dDoVDcc321Wg0vLy/0wCA4SfdenuheJIV1/IG0dqJUvpukkHnaHbauYlR8oxQJXbYjLy/P7NOld1JeK3p5jICTVPXfcY0owU/562o0a1XI2iLPy8vDtGnT8Pfff8PHxwePPfYY5s6da1QRJyIiMgm71qvfE088gSeeeELOCEREZCeETgdhRtc6Lz8jIiKiamcVg92IiIhqHLvWiYiIbJhOAFLtK+TsWiciIrJhbJETEZF9EAKAOdeRW2eLnIWciIjsgtAJCDO61mW87cpdsZATEZF9EDqY1yLn5WdERER256OPPkJERARUKhXatWuHX375pVq3z0JORER2QeiE2ZOpNm3ahIkTJ2L69Ok4fPgwunbtir59++LixYvV9u9iISciIvsgdOZPJlq0aBHGjBmDZ599Fk2bNkVCQgJCQ0OxbNmyavtn2fQ58vKBBxqUmnWNP1E5SUj3XoggRKncEaiKim9Y13ne4oKynyVLDCQzt1ZoUJZVrVYbzFcqlVAqKz4SuKSkBKmpqXjjjTcM5sfExODXX3+tepDb2HQhz8/PBwDsx/cyJ6Fag/WJarm9XeROULn8/Hx4eXnVyLadnZ0RGBiI/Rnm1wp3d3eEhoYazJs1axbi4uIqLJudnQ2tVouAgACD+QEBAcjIyDA7SzmbLuTBwcFIT0+Hh4cHJMm8lpRarUZoaCjS09Ot5vF0zGQca8tkbXkAZjIWMxmnOjMJIZCfn4/g4OBqSleRSqXC+fPnUVJi/uN3hRAV6k1lrfFb3b58Zdswh00XcgcHB9SrV69at+np6Wk1vyzlmMk41pbJ2vIAzGQsZjJOdWWqqZb4rVQqFVQqVY3v51Z169aFo6NjhdZ3ZmZmhVa6OTjYjYiIqAY4OzujXbt22L17t8H83bt3o1OnTtW2H5tukRMREVmzSZMm4amnnkJUVBSio6OxYsUKXLx4EWPHjq22fbCQ/0OpVGLWrFn3PNdhScxkHGvLZG15AGYyFjMZxxozWauhQ4fi2rVrePvtt3HlyhW0aNEC33//PcLCwqptH5Kw1pvHEhER0T3xHDkREZENYyEnIiKyYSzkRERENoyFnIiIyIaxkKPmHzFnqn379mHgwIEIDg6GJEnYunWrrHni4+PRvn17eHh4wN/fH4888ghOnTola6Zly5ahVatW+htSREdH44cffpA10+3i4+MhSRImTpwoW4a4uDhIkmQwBQYGypan3KVLlzBy5Ej4+vrC1dUVrVu3Rmpqqmx5wsPDKxwnSZIwfvx42TJpNBq89dZbiIiIgIuLCyIjI/H2229Dp5P3Xun5+fmYOHEiwsLC4OLigk6dOiE5OVnWTPbO7gu5JR4xZ6qCggLcf//9WLp0qWwZbpWUlITx48fj4MGD2L17NzQaDWJiYlBQUCBbpnr16mH+/PlISUlBSkoKevbsiUGDBuHEiROyZbpVcnIyVqxYgVatWskdBc2bN8eVK1f007Fjx2TNk5OTg86dO0OhUOCHH37AyZMn8f7776NOnTqyZUpOTjY4RuU38BgyZIhsmRYsWIDly5dj6dKlSEtLw8KFC/Huu+/iww8/lC0TADz77LPYvXs31q5di2PHjiEmJga9e/fGpUuXZM1l14Sde+CBB8TYsWMN5jVp0kS88cYbMiUyBEBs2bJF7hgGMjMzBQCRlJQkdxQD3t7e4pNPPpE7hsjPzxcNGzYUu3fvFt27dxevvPKKbFlmzZol7r//ftn2X5mpU6eKLl26yB3jrl555RXRoEEDodPpZMvQv39/ERsbazBv8ODBYuTIkTIlEqKwsFA4OjqKb7/91mD+/fffL6ZPny5TKrLrFnn5I+ZiYmIM5lf3I+Zqm7y8PACAj4+PzEnKaLVabNy4EQUFBYiOjpY7DsaPH4/+/fujd+/eckcBAJw5cwbBwcGIiIjAsGHDcO7cOVnzbNu2DVFRURgyZAj8/f3Rpk0brFy5UtZMtyopKcHnn3+O2NjYan2wham6dOmCn376CadPnwYA/PHHH9i/fz/69esnWyaNRgOtVlvhnuUuLi7Yv3+/TKnIru/sZqlHzNUmQghMmjQJXbp0QYsWLWTNcuzYMURHR6OoqAju7u7YsmULmjVrJmumjRs34vfff7eac4YdOnTAZ599hkaNGuHq1auYM2cOOnXqhBMnTsDX11eWTOfOncOyZcswadIkvPnmm/jtt9/w8ssvQ6lU4umnn5Yl0622bt2K3NxcjB49WtYcU6dORV5eHpo0aQJHR0dotVrMnTsXTz75pGyZPDw8EB0djXfeeQdNmzZFQEAANmzYgEOHDqFhw4ay5bJ3dl3Iy9X0I+ZqkwkTJuDo0aNW8e27cePGOHLkCHJzc/H1119j1KhRSEpKkq2Yp6en45VXXsGuXbss/pSlO+nbt6/+/1u2bIno6Gg0aNAAiYmJmDRpkiyZdDodoqKiMG/ePABAmzZtcOLECSxbtswqCvmqVavQt2/fGn2spjE2bdqEzz//HOvXr0fz5s1x5MgRTJw4EcHBwRg1apRsudauXYvY2FiEhITA0dERbdu2xfDhw/H777/Llsne2XUht9Qj5mqLl156Cdu2bcO+ffuq/fGxVeHs7Iz77rsPABAVFYXk5GQsXrwYH3/8sSx5UlNTkZmZiXbt2unnabVa7Nu3D0uXLkVxcTEcHR1lyVbOzc0NLVu2xJkzZ2TLEBQUVOHLVtOmTfH111/LlOhfFy5cwI8//ojNmzfLHQWTJ0/GG2+8gWHDhgEo+yJ24cIFxMfHy1rIGzRogKSkJBQUFECtViMoKAhDhw5FRESEbJnsnV2fI7fUI+ZsnRACEyZMwObNm/Hzzz9b7S+sEALFxcWy7b9Xr144duwYjhw5op+ioqIwYsQIHDlyRPYiDgDFxcVIS0tDUFCQbBk6d+5c4fLF06dPV+tDJKpq9erV8Pf3R//+/eWOgsLCQjg4GP6JdnR0lP3ys3Jubm4ICgpCTk4Odu7ciUGDBskdyW7ZdYscsMwj5kx148YNnD17Vv/6/PnzOHLkCHx8fFC/fn2L5xk/fjzWr1+Pb775Bh4eHvoeDC8vL7i4uFg8DwC8+eab6Nu3L0JDQ5Gfn4+NGzdi79692LFjhyx5gLLzh7ePG3Bzc4Ovr69s4wlef/11DBw4EPXr10dmZibmzJkDtVota4vu1VdfRadOnTBv3jw88cQT+O2337BixQqsWLFCtkxAWZf/6tWrMWrUKDg5yf+nceDAgZg7dy7q16+P5s2b4/Dhw1i0aBFiY2NlzbVz504IIdC4cWOcPXsWkydPRuPGjfHMM8/ImsuuyTpm3kr897//FWFhYcLZ2Vm0bdtW9suq9uzZIwBUmEaNGiVLnsqyABCrV6+WJY8QQsTGxuo/Mz8/P9GrVy+xa9cu2fLcidyXnw0dOlQEBQUJhUIhgoODxeDBg8WJEydky1Nu+/btokWLFkKpVIomTZqIFStWyB1J7Ny5UwAQp06dkjuKEEIItVotXnnlFVG/fn2hUqlEZGSkmD59uiguLpY116ZNm0RkZKRwdnYWgYGBYvz48SI3N1fWTPaOjzElIiKyYXZ9jpyIiMjWsZATERHZMBZyIiIiG8ZCTkREZMNYyImIiGwYCzkREZENYyEnIiKyYSzkRERENoyFnMhMcXFxaN26tf716NGj8cgjj1g8x19//QVJknDkyJE7LhMeHo6EhASjt7lmzRrUqVPH7GySJGHr1q1mb4eIKmIhp1pp9OjRkCQJkiRBoVAgMjISr7/+OgoKCmp834sXL8aaNWuMWtaY4ktEdDfyPxmAqIb85z//werVq1FaWopffvkFzz77LAoKCrBs2bIKy5aWlkKhUFTLfr28vKplO0RExmCLnGotpVKJwMBAhIaGYvjw4RgxYoS+e7e8O/zTTz9FZGQklEolhBDIy8vD888/D39/f3h6eqJnz574448/DLY7f/58BAQEwMPDA2PGjEFRUZHB+7d3ret0OixYsAD33XcflEol6tevj7lz5wKA/pGwbdq0gSRJ6NGjh3691atXo2nTplCpVGjSpAk++ugjg/389ttvaNOmDVQqFaKionD48GGTj9GiRYvQsmVLuLm5ITQ0FOPGjcONGzcqLLd161Y0atQIKpUKffr0QXp6usH727dvR7t27aBSqRAZGYnZs2dDo9GYnIeITMdCTnbDxcUFpaWl+tdnz57FF198ga+//lrftd2/f39kZGTg+++/R2pqKtq2bYtevXrh+vXrAIAvvvgCs2bNwty5c5GSkoKgoKAKBfZ206ZNw4IFCzBjxgycPHkS69evR0BAAICyYgwAP/74I65cuYLNmzcDAFauXInp06dj7ty5SEtLw7x58zBjxgwkJiYCAAoKCjBgwAA0btwYqampiIuLw+uvv27yMXFwcMCSJUtw/PhxJCYm4ueff8aUKVMMliksLMTcuXORmJiI//3vf1Cr1Rg2bJj+/Z07d2LkyJF4+eWXcfLkSXz88cdYs2aN/ssKEdUwmZ++RlQjRo0aJQYNGqR/fejQIeHr6yueeOIJIYQQs2bNEgqFQmRmZuqX+emnn4Snp6coKioy2FaDBg3Exx9/LIQQIjo6WowdO9bg/Q4dOoj777+/0n2r1WqhVCrFypUrK815/vx5AUAcPnzYYH5oaKhYv369wbx33nlHREdHCyGE+Pjjj4WPj48oKCjQv79s2bJKt3WrsLAw8cEHH9zx/S+++EL4+vrqX69evVoAEAcPHtTPS0tLEwDEoUOHhBBCdO3aVcybN89gO2vXrhVBQUH61wDEli1b7rhfIqo6niOnWuvbb7+Fu7s7NBoNSktLMWjQIHz44Yf698PCwuDn56d/nZqaihs3bsDX19dgOzdv3sT//d//AQDS0tIwduxYg/ejo6OxZ8+eSjOkpaWhuLgYvXr1Mjp3VlYW0tPTMWbMGDz33HP6+RqNRn/+PS0tDffffz9cXV0Ncphqz549mDdvHk6ePAm1Wg2NRoOioiIUFBTAzc0NAODk5ISoqCj9Ok2aNEGdOnWQlpaGBx54AKmpqUhOTjZogWu1WhQVFaGwsNAgIxFVPxZyqrUefPBBLFu2DAqFAsHBwRUGs5UXqnI6nQ5BQUHYu3dvhW1V9RIsFxcXk9fR6XQAyrrXO3ToYPCeo6MjAEAIUaU8t7pw4QL69euHsWPH4p133oGPjw/279+PMWPGGJyCAMouH7td+TydTofZs2dj8ODBFZZRqVRm5ySiu2Mhp1rLzc0N9913n9HLt23bFhkZGXByckJ4eHilyzRt2hQHDx7E008/rZ938ODBO26zYcOGcHFxwU8//YRnn322wvvOzs4Aylqw5QICAhASEoJz585hxIgRlW63WbNmWLt2LW7evKn/snC3HJVJSUmBRqPB+++/DweHsuEyX3zxRYXlNBoNUlJS8MADDwAATp06hdzcXDRp0gRA2XE7deqUSceaiKoPCznRP3r37o3o6Gg88sgjWLBgARo3bozLly/j+++/xyOPPIKoqCi88sorGDVqFKKiotClSxesW7cOJ06cQGRkZKXbVKlUmDp1KqZMmQJnZ2d07twZWVlZOHHiBMaMGQN/f3+4uLhgx44dqFevHlQqFby8vBAXF4eXX34Znp6e6Nu3L4qLi5GSkoKcnBxMmjQJw4cPx/Tp0zFmzBi89dZb+Ouvv/Dee++Z9O9t0KABNBoNPvzwQwwcOBD/+9//sHz58grLKRQKvPTSS1iyZAkUCgUmTJiAjh076gv7zJkzMWDAAISGhmLIkCFwcHDA0aNHcezYMcyZM8f0D4KITMJR60T/kCQJ33//Pbp164bY2Fg0atQIw4YNw19//aUfZT506FDMnDkTU6dORbt27XDhwgW8+OKLd93ujBkz8Nprr2HmzJlo2rQphg4diszMTABl55+XLFmCjz/+GMHBwRg0aBAA4Nlnn8Unn3yCNWvWoGXLlujevTvWrFmjv1zN3d0d27dvx8mTJ9GmTRtMnz4dCxYsMOnf27p1ayxatAgLFixAixYtsG7dOsTHx1dYztXVFVOnTsXw4cMRHR0NFxcXbNy4Uf/+Qw89hG+//Ra7d+9G+/bt0bFjRyxatAhhYWEm5SGiqpFEdZxsIyIiIlmwRU5ERGTDWMiJiIhsGAs5ERGRDWMhJyIismEs5ERERDaMhZyIiMiGsZATERHZMBZyIiIiG8ZCTkREZMNYyImIiGwYCzkREZEN+39tPbmkEZU0zwAAAABJRU5ErkJggg==\n" "image/png": "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\n"
}, },
"metadata": {}, "metadata": {},
"output_type": "display_data" "output_type": "display_data"
...@@ -499,14 +525,23 @@ ...@@ -499,14 +525,23 @@
"average_dictionary = {gnb_average : \"Gaussian Naive Bayes Algorithm\",\n", "average_dictionary = {gnb_average : \"Gaussian Naive Bayes Algorithm\",\n",
" knc_average : \"K Nearest Neighbour Algorithm\",\n", " knc_average : \"K Nearest Neighbour Algorithm\",\n",
" svc_average : \"Support Vector Classification Algorithm\"}\n", " svc_average : \"Support Vector Classification Algorithm\"}\n",
"# And get the average scores for each of the different train_test_split settings used\n",
"first_settings_average = statistics.fmean((gnb_score, knc_score, svc_score))\n",
"second_settings_average = statistics.fmean((gnb2_score, knc2_score, svc2_score))\n",
"third_settings_average = statistics.fmean((gnb3_score, knc3_score, svc3_score))\n",
"average_settings_dictionary = {first_settings_average : \"25% Test Split\",\n",
" second_settings_average : \"50% Test Split\",\n",
" third_settings_average : \"75% Test Split\"}\n",
"# Get the highest of those values\n", "# Get the highest of those values\n",
"highest_score = max(dictionary.keys())\n", "highest_score = max(dictionary.keys())\n",
"highest = dictionary.get(highest_score)\n", "highest = dictionary.get(highest_score)\n",
"highest_average = average_dictionary.get(max(average_dictionary.keys()))\n", "highest_average = average_dictionary.get(max(average_dictionary.keys()))\n",
"highest_settings_average = average_settings_dictionary.get(max(average_settings_dictionary.keys()))\n",
"\n", "\n",
"# Print the best\n", "# Print the best\n",
"print(\"Best single model :\", highest)\n", "print(\"Best single model :\", highest)\n",
"print(\"Best overall algorithm :\", highest_average)\n", "print(\"Best overall algorithm :\", highest_average)\n",
"print(\"Best split ratio :\", highest_settings_average)\n",
"\n", "\n",
"# Now let's generate and look at the confusion matrix from the best model\n", "# Now let's generate and look at the confusion matrix from the best model\n",
"display = metrics.ConfusionMatrixDisplay.from_predictions(test_dictionary.get(highest_score), prediction_dictionary.get(highest_score))\n", "display = metrics.ConfusionMatrixDisplay.from_predictions(test_dictionary.get(highest_score), prediction_dictionary.get(highest_score))\n",
...@@ -518,8 +553,8 @@ ...@@ -518,8 +553,8 @@
"metadata": { "metadata": {
"collapsed": false, "collapsed": false,
"ExecuteTime": { "ExecuteTime": {
"end_time": "2023-05-25T14:28:55.437282Z", "end_time": "2023-05-25T23:09:23.963528Z",
"start_time": "2023-05-25T14:28:55.208964Z" "start_time": "2023-05-25T23:09:23.724208Z"
} }
} }
}, },
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment