keras_example.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n",
    "\n",
    "from sklearn.datasets import load_breast_cancer\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams[\"figure.figsize\"] = (8,10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "class CancelOut(keras.layers.Layer):\n",
    "    '''\n",
    "    CancelOut layer, keras implementation. \n",
    "    '''\n",
    "    def __init__(self, activation='sigmoid', cancelout_loss=True, lambda_1=0.002, lambda_2=0.001):\n",
    "        super(CancelOut, self).__init__()\n",
    "        self.lambda_1 = lambda_1\n",
    "        self.lambda_2 = lambda_2\n",
    "        self.cancelout_loss = cancelout_loss\n",
    "        \n",
    "        if activation == 'sigmoid': self.activation = tf.sigmoid\n",
    "        if activation == 'softmax': self.activation = tf.nn.softmax\n",
    "\n",
    "    def build(self, input_shape):\n",
    "        self.w = self.add_weight(\n",
    "            shape=(input_shape[-1],),\n",
    "            initializer=tf.keras.initializers.Constant(1),\n",
    "            trainable=True,\n",
    "        )\n",
    "    def call(self, inputs):\n",
    "        if self.cancelout_loss:\n",
    "            self.add_loss( self.lambda_1 * tf.norm(self.w, ord=1) + self.lambda_2 * tf.norm(self.w, ord=2))\n",
    "        return tf.math.multiply(inputs, self.activation(self.w))\n",
    "    \n",
    "    \n",
    "    def get_config(self):\n",
    "        return {\"activation\": self.activation}    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_importance(importances):\n",
    "    indices = np.argsort(importances)\n",
    "    plt.title('Feature Importances')\n",
    "    plt.barh(range(len(indices)), importances[indices], color='b', align='center')\n",
    "    plt.yticks(range(len(indices)), [features[i] for i in indices])\n",
    "    plt.xlabel('Relative Importance')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = StandardScaler()\n",
    "X = scaler.fit_transform(load_breast_cancer()['data'])\n",
    "y = load_breast_cancer()['target']\n",
    "features = load_breast_cancer()['feature_names']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Sigmoid + Loss "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_5\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_6 (InputLayer)         [(None, 30)]              0         \n",
      "_________________________________________________________________\n",
      "cancel_out_5 (CancelOut)     (None, 30)                30        \n",
      "_________________________________________________________________\n",
      "dense_15 (Dense)             (None, 32)                992       \n",
      "_________________________________________________________________\n",
      "dense_16 (Dense)             (None, 32)                1056      \n",
      "_________________________________________________________________\n",
      "dense_17 (Dense)             (None, 1)                 33        \n",
      "=================================================================\n",
      "Total params: 2,111\n",
      "Trainable params: 2,111\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "inputs = keras.Input(shape=(X.shape[1],))\n",
    "x = CancelOut(activation='sigmoid')(inputs)\n",
    "x = layers.Dense(32, activation=\"relu\")(x)\n",
    "#x = CancelOut()(x)\n",
    "x = layers.Dense(32, activation=\"relu\")(x)\n",
    "#x = CancelOut()(x)\n",
    "outputs = layers.Dense(1, activation='sigmoid')(x)\n",
    "model = keras.Model(inputs=inputs, outputs=outputs)\n",
    "model.compile(loss='binary_crossentropy', \n",
    "              optimizer='adam')\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "72/72 [==============================] - 0s 799us/step - loss: 0.4768\n",
      "Epoch 2/20\n",
      "72/72 [==============================] - 0s 813us/step - loss: 0.2149\n",
      "Epoch 3/20\n",
      "72/72 [==============================] - 0s 730us/step - loss: 0.1641\n",
      "Epoch 4/20\n",
      "72/72 [==============================] - 0s 742us/step - loss: 0.1429\n",
      "Epoch 5/20\n",
      "72/72 [==============================] - 0s 740us/step - loss: 0.1288\n",
      "Epoch 6/20\n",
      "72/72 [==============================] - 0s 779us/step - loss: 0.1174\n",
      "Epoch 7/20\n",
      "72/72 [==============================] - 0s 895us/step - loss: 0.1093\n",
      "Epoch 8/20\n",
      "72/72 [==============================] - 0s 747us/step - loss: 0.1023\n",
      "Epoch 9/20\n",
      "72/72 [==============================] - 0s 733us/step - loss: 0.0955\n",
      "Epoch 10/20\n",
      "72/72 [==============================] - 0s 840us/step - loss: 0.0902\n",
      "Epoch 11/20\n",
      "72/72 [==============================] - 0s 798us/step - loss: 0.0849\n",
      "Epoch 12/20\n",
      "72/72 [==============================] - 0s 740us/step - loss: 0.0788\n",
      "Epoch 13/20\n",
      "72/72 [==============================] - 0s 756us/step - loss: 0.0739\n",
      "Epoch 14/20\n",
      "72/72 [==============================] - 0s 785us/step - loss: 0.0696\n",
      "Epoch 15/20\n",
      "72/72 [==============================] - 0s 765us/step - loss: 0.0651\n",
      "Epoch 16/20\n",
      "72/72 [==============================] - 0s 716us/step - loss: 0.0615\n",
      "Epoch 17/20\n",
      "72/72 [==============================] - 0s 761us/step - loss: 0.0583\n",
      "Epoch 18/20\n",
      "72/72 [==============================] - 0s 821us/step - loss: 0.0541\n",
      "Epoch 19/20\n",
      "72/72 [==============================] - 0s 730us/step - loss: 0.0506\n",
      "Epoch 20/20\n",
      "72/72 [==============================] - 0s 743us/step - loss: 0.0483\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x1a3d6e5f50>"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X, y, epochs=20, batch_size=8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.23360944, 0.7166199 , 0.18183169, 0.0796449 , 0.36645526,\n",
       "       0.2475115 , 0.5372474 , 0.23054692, 0.2752572 , 0.3597356 ,\n",
       "       0.5012457 , 0.39871517, 0.2695111 , 0.21219146, 0.331213  ,\n",
       "       0.35613608, 0.24519855, 0.14574134, 0.3178066 , 0.57525223,\n",
       "       0.40728638, 0.5422141 , 0.42606953, 0.59417886, 0.65645874,\n",
       "       0.41864341, 0.500722  , 0.30290195, 0.4685865 , 0.34947008],\n",
       "      dtype=float32)"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cancelout_feature_importance_sigmoid = model.get_weights()[0]\n",
    "cancelout_feature_importance_sigmoid"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Softmax + No Loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_6\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_7 (InputLayer)         [(None, 30)]              0         \n",
      "_________________________________________________________________\n",
      "cancel_out_6 (CancelOut)     (None, 30)                30        \n",
      "_________________________________________________________________\n",
      "dense_18 (Dense)             (None, 32)                992       \n",
      "_________________________________________________________________\n",
      "dense_19 (Dense)             (None, 32)                1056      \n",
      "_________________________________________________________________\n",
      "dense_20 (Dense)             (None, 1)                 33        \n",
      "=================================================================\n",
      "Total params: 2,111\n",
      "Trainable params: 2,111\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "inputs = keras.Input(shape=(X.shape[1],))\n",
    "x = CancelOut(activation='softmax', cancelout_loss=False)(inputs)\n",
    "x = layers.Dense(32, activation=\"relu\")(x)\n",
    "#x = CancelOut()(x)\n",
    "x = layers.Dense(32, activation=\"relu\")(x)\n",
    "#x = CancelOut()(x)\n",
    "outputs = layers.Dense(1, activation='sigmoid')(x)\n",
    "model = keras.Model(inputs=inputs, outputs=outputs)\n",
    "model.compile(loss='binary_crossentropy', \n",
    "              optimizer='adam')\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "72/72 [==============================] - 0s 916us/step - loss: 0.6312\n",
      "Epoch 2/20\n",
      "72/72 [==============================] - 0s 740us/step - loss: 0.3737\n",
      "Epoch 3/20\n",
      "72/72 [==============================] - 0s 819us/step - loss: 0.1710\n",
      "Epoch 4/20\n",
      "72/72 [==============================] - 0s 724us/step - loss: 0.1125\n",
      "Epoch 5/20\n",
      "72/72 [==============================] - 0s 785us/step - loss: 0.0935\n",
      "Epoch 6/20\n",
      "72/72 [==============================] - 0s 723us/step - loss: 0.0888\n",
      "Epoch 7/20\n",
      "72/72 [==============================] - 0s 711us/step - loss: 0.0783\n",
      "Epoch 8/20\n",
      "72/72 [==============================] - 0s 806us/step - loss: 0.0734\n",
      "Epoch 9/20\n",
      "72/72 [==============================] - 0s 752us/step - loss: 0.0695\n",
      "Epoch 10/20\n",
      "72/72 [==============================] - 0s 745us/step - loss: 0.0670\n",
      "Epoch 11/20\n",
      "72/72 [==============================] - 0s 723us/step - loss: 0.0655\n",
      "Epoch 12/20\n",
      "72/72 [==============================] - 0s 737us/step - loss: 0.0635\n",
      "Epoch 13/20\n",
      "72/72 [==============================] - 0s 732us/step - loss: 0.0650\n",
      "Epoch 14/20\n",
      "72/72 [==============================] - 0s 727us/step - loss: 0.0604\n",
      "Epoch 15/20\n",
      "72/72 [==============================] - 0s 725us/step - loss: 0.0616\n",
      "Epoch 16/20\n",
      "72/72 [==============================] - 0s 734us/step - loss: 0.0602\n",
      "Epoch 17/20\n",
      "72/72 [==============================] - 0s 742us/step - loss: 0.0579\n",
      "Epoch 18/20\n",
      "72/72 [==============================] - 0s 751us/step - loss: 0.0579\n",
      "Epoch 19/20\n",
      "72/72 [==============================] - 0s 722us/step - loss: 0.0571\n",
      "Epoch 20/20\n",
      "72/72 [==============================] - 0s 790us/step - loss: 0.0560\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x1a3e05f450>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X, y, epochs=20, batch_size=8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.8646685 , 0.9297946 , 1.041775  , 1.0581877 , 0.87913954,\n",
       "       0.7880292 , 1.070631  , 0.9703354 , 0.71175075, 0.79631186,\n",
       "       0.8345016 , 0.67362726, 1.0588152 , 1.2686845 , 0.5851457 ,\n",
       "       0.8804763 , 0.66059697, 0.5979381 , 0.7027532 , 0.890007  ,\n",
       "       1.3074998 , 1.2685775 , 1.2207824 , 1.1551926 , 1.0438159 ,\n",
       "       0.5558124 , 1.0898336 , 1.0736428 , 1.0018368 , 0.7529111 ],\n",
       "      dtype=float32)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cancelout_feature_importance_softmax = model.get_weights()[0]\n",
    "cancelout_feature_importance_softmax"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### RandomForest Feature Importance "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "rnd_clf = RandomForestClassifier(n_estimators=500, n_jobs=-1, random_state=42).fit(X,y)\n",
    "rf_importances = rnd_clf.feature_importances_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sigmoid + Loss\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkoAAAJcCAYAAAAGrIsfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOzdeZifVX3//+crISRCQlIJUhYxCsjiQmQGlJRAQH60bhAUpCmLwVZEW1CWWFr52ahViUCxWC0Ni1FMhQKyGJVFIAskJswQsrDL4ooRFUICJJLk9f3jPiMfZj6fWULCZDKvx3XdV8597rPdn7muzPs658znyDYRERER0dGA3h5ARERExKYqgVJEREREAwmUIiIiIhpIoBQRERHRQAKliIiIiAYSKEVEREQ0kEApIiIiooEEShGxSZL0hKQXJK2suXbcAG0etqHG2I3+Jkv67qvVX2ckTZR0Z2+PI6KvSaAUEZuyD9geWnP9pjcHI2mL3ux/ffXVcUdsChIoRUSfImm4pMskPSnp15L+XdLA8mxXSbdL+oOk30uaLmlEeXYFsAvwgzI79RlJ4yT9ql37f551KjNC10j6rqRngYmd9d+NsVvSJyU9ImmFpC+WMc+V9Kyk/5O0ZSk7TtKvJP1reZcnJB3X7nP4jqSnJP1c0jmSBpRnEyXdJelCSX8ArgIuBg4o7/5MKfc+SQtL37+UNLmm/VFlvB+R9Isyhs/WPB9YxvZoeZdWSa8vz/aUdKukP0p6SNKHa+q9V9L9pc6vJZ3V7R9+RC9IoBQRfc00YA2wG/AO4HDgH8ozAV8BdgT2Al4PTAawfQLwC16apfpqN/s7ErgGGAFM76L/7vhroAl4F/AZYCpwfBnrW4EJNWX/EhgJ7AR8BJgqaY/y7OvAcOBNwMHAicBJNXXfCTwGbF/aPwWYV959RCnzXKk3Angf8AlJ49uN90BgD+DdwOck7VXyzyhjfS+wDfBR4HlJWwO3Av8LvA74W+CbkvYu9S4DPm57WHnf27v1qUX0kgRKEbEpu17SM+W6XtL2VL+YP237Odu/Ay6k+mWM7Z/ZvtX2attPAf9BFUS8EvNsX297HVVA0LD/bvqq7Wdt3wcsBW6x/Zjt5cCPqYKvWv9/eZ9ZwA+BD5cZrL8F/sX2CttPABcAJ9TU+43tr9teY/uFegOxPdP2EtvrbC8GvkfHz+vztl+wvQhYBOxT8v8BOMf2Q64ssv0H4P3AE7a/VfpeCFwLHFPqvQjsLWkb20/bvqcHn13Eqy7r1hGxKRtv+ydtN5L2BwYBT0pqyx4A/LI83x74T2AsMKw8e/oVjuGXNek3dNZ/Ny2rSb9Q5/4va+6ftv1czf3PqWbLRpZx/Lzds50ajLsuSe8EzqWa2dkSGAxc3a7Yb2vSzwNDS/r1wKN1mn0D8M625b1iC+CKkv4QcA5wrqTFwNm253U11ojekhmliOhLfgmsBkbaHlGubWy/pTz/MmDgbba3oVpyUk19t2vvOWCrtpsyU7NduzK1dbrqf0P7i7KU1WYX4DfA76lmZt7Q7tmvG4y73j1Uy2M3Aq+3PZxqH5PqlKvnl8CuDfJn1Xw+I8py3ycAbN9t+0iqZbnrgf/rZn8RvSKBUkT0GbafBG4BLpC0jaQBZTN023LRMGAlsFzSTsCkdk0so9rT0+ZhYEjZ1DyIaqZj8Cvof2P4vKQtJY2lWta62vZaqgDjS5KGSXoD1Z6hzr6KYBmwc9tm8WIY8Efbq8ps3d/1YFyXAl+UtLsqb5e0LTADeLOkEyQNKtd+kvYq73GcpOG2XwSeBdb1oM+IV10CpYjoa06kWia6n2pZ7Rpgh/Ls88C+wHKq/Tzfb1f3K8A5Zc/TWWVf0Cepfun/mmqG6Vd0rrP+N7Tflj5+Q7WR/BTbD5Znp1KN9zHgTqrZocs7aet24D7gt5J+X/I+CXxB0grgc/Rsduc/SvlbqAKey4DX2F5BtcH9b8u4fwtM4aUA9ATgifJXhKcAxxGxCZNdbzY2IiJ6k6RxwHdt79zbY4nozzKjFBEREdFAAqWIiIiIBrL0FhEREdFAZpQiIiIiGsgXTkYHI0eO9KhRo3p7GBEREa+K1tbW39tu/x1qQAKlqGPUqFG0tLT09jAiIiJeFZJ+3uhZlt4iIiIiGkigFBEREdFAAqWIiIiIBhIoRURERDSQQCkiIiKigQRKEREREQ0kUIqIiIhoIIFSRERERAMJlCIiIiIaSKAUERER0UACpYiIiIgGEihFRERENJBAKSIiIqKBBEoRERERDSRQioiIiGgggVJEREREAwmUIiIiIhpIoBQRERHRQAKliIiIiAYSKEVEREQ0kEApIiIiooEEShERERENJFCKiIiIaGCL3h5AbHpaW0Hq7VFERER0ZL+6/WVGKSIiIqKBBEoRERERDSRQioiIiGgggdKrRNJESTu+gvr/uiHHExEREV1LoPTqmQisd6AE9DhQkpTN+hEREa9AvwmUJI2S9KCkaZIeljRd0mGS7pL0iKT9S7mtJV0uaYGkhZKOrKk/R9I95RpT8sdJminpmtL+dOnlfzMm6WigGZgu6V5Jr5HUJGmWpFZJN0vaQdJwSQ9J2qPU+56kj0k6F3hNqTu9jGVpTftnSZpc0jMlfU1SC/Cpev28Ch93RETE5sF2v7iAUcAa4G1UAWIrcDkg4Ejg+lLuy8DxJT0CeBjYGtgKGFLydwdaSnocsBzYubQ7DziwTv8zgeaSHgTMBbYr98cCl5f0/1fa+Fvgppr6K9u9y9Ka+7OAyTX9fLOrfuqM72Sgpbp2cfUHmLly5cqVK9emdW0Mbb/T6139bWnmcdtLACTdB9xm25KWUAUfAIcDR0g6q9wPAXYBfgP8l6TRwFrgzTXtLrD9q9LuvaWtOzsZxx7AW4Fby+TTQOBJANu3SjoG+Aawz3q+51Vd9dOe7anA1Oodmr2e/UZERGxW+lugtLomva7mfh0vfRYCPmT7odqKZWlrGVXwMgBY1aDdtXT9uQq4z/YBHR5IA4C9gOeBvwB+Vaf+Gl6+bDqk3fPnuuonIiIiutZv9ij1wM3AqW37jCS9o+QPB560vQ44gWp2pidWAMNK+iFgO0kHlD4GSXpLeXY68ADwd8C3JA0q+S/WpJcBr5O0raTBwPsb9NlZPxEREdGFBEodfZFqb8/isjz3xZL/TeAjkhYBe/LSrE13TQMuLktzA4GjgSmlvXuBMWUT9z8AZ9qeA8wGzin1p5YxTbf9IvAFYAFwK/BgvQ5t/6lePz0cd0RERL+lspE34s+qPUotvT2MiIiIDjZG2CKp1XZzvWf9bY9SdENTE7QkToqIiMjSW0REREQjCZQiIiIiGsjSW3TQ2gov/27xiIjor/r7VubMKEVEREQ0kEApIiIiooEEShEREREN9KtASdJ4SXtvAuOYKGnHmvsnJI3szTFFRERER5tloCSp0fEi44FeD5SAicCOXRWKiIiI3rVJBUqSJkk6raQvlHR7SR8qaXpJT5C0RNJSSVNq6q6UdEE5quMASedKul/SYknnSxoDHAGcJ+leSbu26/uY0uYiSbNL3kRJ10u6tcz6/JOkMyQtlPRTSa8t5UaX+8WSrpP0F43yJR0NNAPTyzheU4ZwqqR7yrvtWepPlnS5pJmSHmv7bMqz4yUtKG38j6SB5ZpW3mOJpNNL2dNqPosrN/xPLiIiYjNle5O5gHcBV5f0HKqzzAYB/wZ8nGoW5hfAdlRfbXA7ML6UN/Dhkt6W6kDYtiNaRpR/pwFHN+h7CbBTu/ITgZ9RHWa7HbAcOKU8uxD4dEkvBg4u6S8AX+sifybQXNP3E8CpJf1J4NKSngzMBQYDI4E/lM9jL+AHwKBS7pvAiUATcGtNu23v8RtgcG1enfc/merckhbYxdUfhObKlStXrv5+9QdAi10/NtmkZpSAVqBJ0jbAamAe1ezLWKrAaT9gpu2nbK8BpgMHlbprgWtLejmwCrhM0geB57vR913ANEkfozq0ts0dtlfYfqq0+4OSvwQYJWk4VfAxq+R/GzioUX4n/X+/5jMYVZP/Q9urbf8e+B2wPfBuqqDo7nLI7ruBNwGPAW+S9HVJfwM8W9pYTDWDdTywpl7ntqfabrbdXMWEERERsUkFSrZfBB6nmsmZSxUcHQLsBjzQRfVVtteWdtYA+wPXAO8HbupG36cA5wCvB1olbVsera4ptq7mfh0b9gs729pd267d2v7bngn4tu3R5drD9mTbTwP7UM1YnQJcWuq9D/gGsC9VcJUvGo2IiOiGTSpQKuYAZwGzS/oUYGGZGlsAHCxpZNmwPQGY1b4BSUOB4bZ/BJxOFTwArKBaRutA0q6259v+HPAUVcDUJdvLgacljS1ZJwCzGuV3NY5uug04WtLrythfK+kN5S/nBti+liro21fSAOD1tu8A/hkYDgx9BX1HRET0G5vizMIc4LPAPNvPSVpV8rD9pKSzgTuoZlV+aPuGOm0MA26QNKSUO6PkXwlcUjZFH2370Zo650navZS/DVgEjO7mmD8CXCxpK6rlr5O6yJ9W8l8ADuhmH39m+35J5wC3lEDoReAfgReAb5U8gH+hWkb8blkKFHCR7Wd62mdERER/1LbZOeLPpGZX+7ojIqK/6w9hgqTWao9uR5vi0ltERETEJmFTXHqLXtbUBC2ZUIqIiMiMUkREREQjCZQiIiIiGsjSW3TQ2gpSb48iIqL/6Q8bp/uazChFRERENJBAKSIiIqKBBEoRERERDSRQ2gRIGi9p794eR0RERLxcAqVXUTmfrp7xwHoHSjnkNiIiYuNIoNQNkiaV8+GQdKGk20v6UEnTS3qCpCWSlkqaUlN3paQLJC0CDpB0rqT7JS2WdL6kMcARVGfN3Stp13Z9f0DSfEkLJf1E0vYlf7KkKyTdBVwhaTtJ10q6u1x/VcrtL2leqT9X0h6vxmcWERGxOchMRPfMAc4ELgKagcGSBgFjgdmSdgSmAE3A01SH1Y63fT2wNTDf9pmStgUuA/a0bUkjbD8j6UZghu1r6vR9J/CuUv4fgM+UsUA1C3Wg7Rck/S9woe07Je0C3AzsBTwIjLW9RtJhwJeBD7XvRNLJwMnV3S6v8OOKiIjYPCRQ6p5WoEnSNsBq4B6qgGkscBqwHzDT9lMAZZbpIOB6YC1wbWlnObAKuEzSDGBGN/reGbhK0g7AlsDjNc9utP1CSR8G7K2XvgBpG0lDgeHAtyXtDhgYVK8T21OBqdX4m/NNHhEREWTprVtsv0gVoEwE5lLNMB0C7AY80EX1VbbXlnbWAPsD1wDvB27qRvdfB/7L9tuAjwNDap49V5MeQDXzNLpcO9leCXwRuMP2W4EPtKsfERERnUig1H1zgLOA2SV9CrDQtoEFwMGSRpYN2xOAWe0baJvhsf0j4HRgn/JoBTCsQb/DgV+X9Ec6Gd8twKk1fY2uU39iJ/UjIiKinQRK3TcH2AGYZ3sZ1RLaHADbTwJnA3cAi4BW2zfUaWMYMEPSYqq9R2eU/CuBSWXD9a7t6kwGrpbUCvy+k/GdBjSXTeL3UwVyAF8FviJpIVlqjYiI6BE5B8tEO9UepZbeHkZERL+TX8m9Q1Kr7eZ6zzLDEB00NUFL4qSIiIgsvUVEREQ0kkApIiIiooEEShERERENZI9SdNDaCi99b2VERLwS2aDdt2VGKSIiIqKBBEoRERERDfQ4UJJ0mqQHynlmr4ikieVA2a7KTZN0dBdlRklaWtLNki56peNbH5Lm9ka/ERERseGtzx6lTwKH2f5VbaakLcpZZj0xEVgK/GY9xtGQ7RZ66RsTbY95tfpq/5l392ewnj+riIiIfqdHgZKki4E3AT+WdDnVOWK7lrxfSPoX4Apg61Lln2zPLXX/GTgeWAf8mCqQaQamS3oBOACYRHVw62uoDp/9uDv56nBJTcDl5faWmvxxwFm23y9pMvDGMsZdqM5YexfwHqoz0D5g+8XS1n8AQ6mOCplo+0lJM4H5VIfgjgD+3vYcSW8BvgVsSTUz9yHbj0haaXuoJFEdH/IewMC/276qjG1y6eOtQCtwfPv3LEeZfAPYDnge+JjtByVNozo+5R3AXZJe2+7+O8DFwFbAo8BHbT9d3uNe4EDge8AFjT7XiIiIKGz36AKeAEaW9GSqX/SvKfdbAUNKenegpaTfQxX4bFXuX1v+nQk017T92pr0FVRBDMA04Og6Y1kMHFTS5wFLS3ocMKNmjHcCg6gOoX0eeE95dh0wvjybC2xX8o8FLq8Z4wUl/V7gJyX9deC4kt6y5jNYWf79EHArMBDYHvgF1Vlx44DlwM5UAdY84MA673YbsHtJvxO4veazmAEMbHC/GDi4pL8AfK3mPb7Zyc/1ZKrgtQV2cfV3Grly5cqV65Veselri1fqXRvi6wFutP1CSQ8C/qucXL8WeHPJPwz4lu3nAWz/sUFbh0j6DFXA9VrgPuAH9QpKGgGMsD27ZF1BFZDV82NXs0ZLqAKXm0r+EmAUsAfV7M6t1UQQA4Ena+p/v/zbWspDFeB8VtLOwPdtP9KuzwOB79leCyyTNAvYD3gWWOCydCnp3tLmnTXvNhQYQ3UYblv24Jq2ry7tvuxe0vDymcwq+d8Grq4pd1XdTwewPRWYWvXf7EblIiIi+pMNESg9V5M+HVhGNXMzgGpJqFskDQG+STXD9MuyZDZkA4wPYDWA7XWSXizRI1TLgFsAAu6zfUBn9amCvy1KW/8raT7wPuBHkj5u+/aejKd9mzUGAM/YHt2g/nNd3DfS3XIRERHBhv96gOHAk7bXASdQzcxAtQR1kqStAMq+GoAVwLCSbguKfl9mVDr9KzfbzwDPSDqwZB33Csb9ELCdpAPK+AaVPUgNSXoT8Jjti4AbgLe3KzIHOFbSQEnbAQcBC7ozGNvPAo9LOqb0JUn7dKPecuBpSWNL1gnArE6qRERERCc2dKD0TeAjkhYBe1JmMGzfBNwItJSlprNK+WnAxSVvNXAJ1V/B3Qzc3Y3+TgK+Ueqv93dJ2/4TVWA2pYz9Xqqlr858GFha+n4r8J12z6+j2i+0CLgd+Izt3/ZgWMcBf1/Gcx9wZDfrfQQ4T9JiYDTVPqWIiIhYD3ppFSqiUu1R6pVvV4iI2Ozk1+ymT1Kr7eZ6z/LN3BEREREN5FDc6KCpCVoyoRQREZEZpYiIiIhGEihFRERENJClt+igtRW03n9DGBHRt2XzddTKjFJEREREAwmUIiIiIhpIoBQRERHRQAKlDUjSeEl79/RZN9v+dNsRMBEREfHqSKC0HiQNbPBoPNAoGOrsWXd8GuhRoCQpm/UjIiJegX4VKEmaJOm0kr5Q0u0lfaik6SU9QdISSUslTampu1LSBeXstQMknSvpfkmLJZ0vaQxwBNU5a/dK2rWmbodn5bpJUqukOZL2lLSFpLsljSv1viLpS2XMOwJ3SLqjbTw17R8taVpJT5N0saT5wFfr9bPxPuGIiIjNS3+bcZgDnAlcBDQDgyUNAsYCsyXtCEwBmoCngVskjbd9PbA1MN/2mZK2BS4D9rRtSSNsPyPpRmCG7WtqO7U9t/0zSbcBp9h+RNI7gW/aPlTSROAaSacCfwO80/afJJ0BHGL79914z52BMbbX1usHOLR9BUknAydXd7t08+OMiIjYvPW3QKkVaJK0DbAauIcqYBoLnAbsB8y0/RRAmWU6CLgeWAtcW9pZDqwCLpM0A5jRk0FIGgqMAa7WS19YNBjA9n2SrihtHmD7T+vxnleXIKlhP+3ZngpMrcbXnG8RiYiIoJ8FSrZflPQ4MBGYCywGDgF2Ax4Adu+k+irba0s7ayTtD7wbOBr4J+rM0nRiAPCM7dENnr8NeAZ4XSdt1AYzQ9o9e66b/UREREQn+tUepWIOcBYwu6RPARbaNrAAOFjSyLJhewIwq30DZaZmuO0fAacD+5RHK4BhDfr98zPbzwKPSzqmtCdJ+5T0B4HXUs1kfV3SiAZtL5O0l6QBwFH1Ouysn4iIiOhafw2UdgDm2V5GtYQ2B8D2k8DZwB3AIqDV9g112hgGzJC0GLgTOKPkXwlMkrSwdjN3g2fHAX9fNoffBxwpaSRwLvAPth8G/gv4z1J/KnBT22buMs4ZVDNjT3byvh366fzjiYiIiDZyDrWJdqo9Si29PYyIiF6RX4v9j6RW2831nvXHGaWIiIiIbulXm7mje5qaoCUTShEREZlRioiIiGgkgVJEREREA1l6iw5aW+Gl76eMiOgfsok76smMUkREREQDCZQiIiIiGkigtBmQ1CzpopIeJ2lMb48pIiJic5A9SpsB2y289A2R44CVVN/YHREREa9An51RkjRK0oOSpkl6WNJ0SYdJukvSI+XQWiRtLelySQvK8SFH1tSfI+meco0p+eMkzZR0TWl/utRxa7Ok3ST9RNKiUn/XcpbaeZKWSloi6diu2pS0n6S5pZ0FkoZ1MrYrJb2vZgzTJB1d2p8haRTV2XWnS7pX0lhJj0saVMpvU3sfERERXbDdJy9gFLAGeBtVwNcKXA6I6jyz60u5LwPHl/QI4GFga2ArYEjJ3x1oKelxwHJg59LuPODAOv3PB44q6SGlvQ8BtwIDge2BX1CdK1e3TWBL4DFgv9LONlSzfI3GdhTw7ZLeEvgl8JrS/oySPxk4q2ac3wLGl/TJwAUNPs+TqWalWmAXV3//kStXrlz954r+q+33bL2rz84oFY/bXmJ7HdWBr7eVF15CFUgBHA6cLeleYCZVULMLMAi4RNIS4Gpg75p2F9j+VWn33pq2AJA0DNjJ9nUAtlfZfp4q+Pme7bWuDtydBezXSZt7AE/avru086ztNZ2M7cfAIZIGA+8BZtt+oYvP6FLgpJI+iSpw6sD2VNvNtpthuy6ajIiI6B/6+h6l1TXpdTX363jp3QR8yPZDtRUlTQaWAftQzfKsatDuWjbM59STNk+vNzbbqyTNBP4aOBa4sqtObd9VlvLGAQNtL12v0UdERPRDfX1GqTtuBk6t2RP0jpI/nGo2Zx1wAtVyWbfYXgH8StL40uZgSVsBc4BjJQ2UtB1wELCgk6YeAnaQtF9pZ5ikLboY21VUM0NjgZvqtLkCGNYu7zvA/9JgNikiIiLq6w+B0heplrIWS7qv3AN8E/iIpEXAnsBzPWz3BOA0SYup/sLsL4HrgMXAIuB24DO2f9uoAdt/opoZ+noZx61US4Odje0W4GDgJ6V+ez8AjmrbzF3ypgN/AXyvh+8YERHRr6na0hObM0lHA0faPqF75Zv90rcNRET0D/l12H9Jaq326HbU1/coRRckfZ1q4/d7e3ssERERfU0Cpc2c7VN7WqepCVoyoRQREdEv9ihFRERErJcEShERERENZOktOmhthY6HtkREbBqy6TpeTZlRioiIiGgggVJEREREAwmUIiIiIhpIoLQBSBonaUZJHyHp7N4eU0RERLxy2czdQDkbTuW8tW6zfSNw48YZ1ctJ2sL2mkb3ndQbaHvtxh1dRERE35cZpRqSRkl6SNJ3gKXA6yX9t6QWSfdJ+nxN2b+R9KCke4AP1uRPlPRfJT2tHB/S9mxl+XcHSbPLeWxLa85kqx1Lk6RZklol3Sxph5I/U9LXJLUAn6pz/25JCyUtkXS5pMGl3hOSppTxHrMxPr+IiIjNTWaUOtod+IjtnwJI+qztP0oaCNwm6e3Aw8AlwKHAz4CretjH3wE32/5SaXer2oeSBgFfpzqf7SlJxwJfAj5aimzZdiaNpA+03UsaAjwCvNv2wyXg+wTwtVLvD7b3rTcgSScDJ1d3u/TwdSIiIjZPmVHq6OdtQVLx4TILsxB4C7A3sCfwuO1HXJ0q/N0e9nE3cJKkycDbbK9o93wP4K3ArZLuBc4Bdq553j4wu6qm3uO2Hy733wYO6qTen9mearu5CsC269HLREREbK4yo9TRc20JSW8EzgL2s/20pGnAkB60tYYSjEoaAGwJYHu2pIOA9wHTJP2H7e/U1BNwn+0Duhpjg/tGulsuIiIiyIxSV7ahCi6WS9oeeE/JfxAYJWnXcj+hQf0ngKaSPgIYBCDpDcAy25cAlwLtl8MeAraTdEApP0jSW7ox3ofKuHYr9ycAs7pRLyIiIurIjFInbC+StJAqMPolcFfJX1X29PxQ0vPAHGBYnSYuAW6QtAi4iZdmdMYBkyS9CKwETmzX75/KJvCLJA2n+jl9Dbivi/GuknQScLWkLaiW+C7u+ZtHREQEVH/+3ttjiE2M1Gxo6e1hRETUlV9bsaFJam37I6n2svQWERER0UCW3qKDpiZoyYRSREREZpQiIiIiGkmgFBEREdFAlt6ig9ZWkHp7FBER9WUzd7yaMqMUERER0UACpYiIiIgGEihFRERENJBAqROSxkvau7fH0RVJO0q6pqRHS3pvb48pIiJic5BACZA0sMGj8cAmHyjZ/o3to8vtaCCBUkRExAbQpwMlSZMknVbSF0q6vaQPlTS9pCdIWiJpqaQpNXVXSrqgnMN2gKRzJd0vabGk8yWNoTrI9jxJ99YcgNtWf3tJ10laVK4xJf+M0tdSSZ8ueaMkPSDpEkn3SbpF0mvKs90k/aS0cY+kXSUNlXRbuV8i6chS9lxJ/1gzhsmSzirtL5W0JfAF4Ngy5mMlPSJpu1J+gKSftd1HRERE5/p0oER1GO3Ykm4GhkoaVPJmS9oRmAIcSjXTsp+k8aX81sB82/sADwBHAW+x/Xbg323PBW4EJtkebfvRdn1fBMwq9fcF7pPUBJwEvBN4F/AxSe8o5XcHvmH7LcAzwIdK/vSSvw8wBngSWAUcZXtf4BDgAkkCrgI+XDOGD5c8oDpMF/gccFUZ81XAd4HjSpHDgEW2n2r/QUo6WVKLpBbo8DgiIqJf6uuBUivQJGkbYDUwjypgGksVRO0HzLT9lO01VEHJQaXuWuDakl5OFZxcJumDwPPd6PtQ4L8BbK+1vRw4ELjO9nO2VwLf56VA7nHb99aMe5SkYcBOtq8r7ayy/Twg4MuSFgM/AXYCtre9EHhd2ZO0D/C07V92Mc7LgRNL+qPAt+oVsj3VdnN1KGAmnCIiIqCPB0q2XwQeByYCc6mCo0OA3ahmiTqzyvba0s4aYH/gGuD9wE0bYbira9Jr6fzLPsYvxNAAACAASURBVI+jilaabI8GlgFDyrOrgaOBY6mZTWqkBFLLJB1K9Y4/7vnQIyIi+qc+HSgVc4CzgNklfQqw0LaBBcDBkkaWDdsTgFntG5A0FBhu+0fA6cA+5dEKYFiDfm8DPlHqD5Q0vPQ/XtJWkramWs6b02jgtlcAv2pbDpQ0WNJWwHDgd7ZflHQI8IaaalcBf0sVLF1dp9l6Y76Uagnu6rbgMCIiIrq2uQRKOwDzbC+jWkKbA2D7SeBs4A5gEdBq+4Y6bQwDZpSlrjuBM0r+lcAkSQvbb+YGPgUcImkJ1VLa3rbvAaZRBWjzgUvLcllnTgBOK33PBf6SaomwubR9IvBgW2Hb95Xx/rq8X3t3AHu3beYueTcCQ2mw7BYRERH1yTk0Z7MnqRm40PbYLgsDUrOhZSOPKiJi/eTXVmxoklqrPbod5VDczZyks6mWCI/rqmybpiZoSZwUERGxWSy9RSdsn2v7Dbbv7O2xRERE9DUJlCIiIiIaSKAUERER0UD2KEUHra0g9fYoImJzkg3Y0VdlRikiIiKigQRKEREREQ0kUKpD0nhJe/f2OLpD0ghJn+ztcURERGyO+nWgVI41qWc80CcCJWAEUDdQkpQ9aBEREa9AnwyUJE2SdFpJXyjp9pI+VNL0kp4gaYmkpZKm1NRdKekCSYuAAySdK+l+SYslnS9pDHAEcF45BmTXdn0fU9pcJGl2yZstaXRNmTsl7SNpsqRvS5oj6eeSPijpq2VcN0kaVMo/Iekrpb8WSftKulnSo5JOaffed5exfr5knwvsWuqeJ2lc6e9G4H5JX5D06Zo2viTpUxvy5xEREbG56pOBEtVZbm3HcTQDQ0vQMRaYLWlHYApwKDAa2K/t4Flga2C+7X2AB6gOrn2L7bcD/257LtXZaJNsj7b9aLu+Pwf8dal/RMm7DJgIIOnNwBDbi8qzXcs4jqA6mPYO228DXgDeV9PuL2yPLu82jerQ23cBny/tHg7sDuxf3qlJ0kFUZ9k9WsY6qbS1L/Ap228GLqc6Lw5JA6gO1P1u+w9U0sklSGuBp+p95hEREf1OXw2UWqkChW2A1cA8qoBpLFWgsR8w0/ZTttdQHTJ7UKm7Fri2pJdTHaJ7maQPAs93o++7gGmSPga0Ld1dDby/BGsfpQp02vzY9ovAklL+ppK/BBhVU+7Gmvz5tlfYfgpYLWkEcHi5FgL3AHtSBU71LLD9OIDtJ4A/SHpHW33bf2hfwfZU283VWTfbdeNjiIiI2Pz1yT0stl+U9DjVLM5cYDFwCLAb1SxRowACYJXttaWdNZL2B95NNYPzT1SzP531fYqkd1LNBrVKarL9B0m3AkcCHwaaaqqsLvXWSXrRL51CvI6Xf/6ra/JX1+S3lRPwFdv/UzseSaPqDPO5dveXUn1Wf0k1wxQRERHd0FdnlKCaOToLmF3Sp1DNlhhYABwsaWTZsD0BmNW+AUlDgeG2fwScDuxTHq0AhtXrVNKutufb/hzVGtXry6NLgYuAu20/vYHesdbNwEfLmJG0k6TXdTbWGtcBf0M103bzRhhbRETEZqlPzigVc4DPAvNsPydpVcnD9pOSzgbuoJqJ+aHtG+q0MQy4QdKQUu6Mkn8lcEnZMH50u31K50navZS/DVhU+myV9CzwrQ39oqX9WyTtBcxT9bXZK4HjbT8q6S5JS4EfAz+sU/dPku4AnmmbTYuIiIiuyfle+Q2ibCCfCexpe10vD+dlyibue4BjbD/SdflmQ8vGH1hE9Bv5VRObMkmt1R7djvry0tsmQ9KJwHzgs5tgkLQ38DPgtu4ESREREfGSzChFB83NzW5pyYxSRET0D5lRioiIiFgPCZQiIiIiGujLf/UWG0lrK1R/WBcR/UF2YEQ0lhmliIiIiAYSKEVEREQ0kEApIiIiooEESg1IGl++g2hjtT93A7UzTtKYDdFWREREvFy/D5TKWXD1jAc2eKAkaQsA2xsquBkH9KittjFERERE5/psoCRpUjmLDUkXSrq9pA+VNL2kJ0haImmppCk1dVdKukDSIuAASedKul/SYknnlxmaI6jOdbtX0q7t+p4m6WJJLZIelvT+kj9Q0nmS7i5tfbzkj5M0R9KNwP1tY6h5NkvSDZIeK2M5TtKCMvZdS7ntJF1b2r5b0l9JGkV1GPDpZZxj65Ur9SdLukLSXcAVG+nHEhERsVnpyzMLc4AzgYuAZmCwpEHAWGB2OXttCtAEPA3cImm87euBrYH5ts+UtC1wGdUZbZY0wvYzJaiZYfuaBv2PAvYHdgXukLQbcCKw3PZ+kgYDd0m6pZTfF3ir7cfrtLUPsBfwR+Ax4FLb+0v6FHAq8GngP4ELbd8paRfgZtt7SboYWGn7fABJ/9u+XGkbqhmyA22/0H4Akk4GTq7udmn0mUdERPQrfTlQagWaJG0DrKY69LWZKlA6DdgPmGn7KYAyy3QQcD2wFri2tLMcWAVcJmkGMKOb/f9fOdftEUmPAXsChwNvl3R0KTMc2B34E7CgQZAEcLftJ8s4HwXagqslwCElfRiwt176gqNtJA2t01Zn5W6sFyQB2J4KTK3G0JxvVYmIiKAPB0q2X5T0ODARmAsspgoqdgMeoApQGllle21pZ42k/YF3A0cD/wQc2p0h1LkXcKrtm2sfSBoHPNdJW6tr0utq7tfx0s9oAPAu26vatd2+rc7KdTaGiIiIaKfP7lEq5gBnAbNL+hRgoauTfhcAB0saWTZsTwBmtW+gzLYMt/0j4HSqZTCAFcCwTvo+RtKAsofoTcBDVMtcnyhLgEh6s6StN8B7QjXLdGrNuEc3GGejchEREdFDm0OgtAMwz/YyqiW0OQBlKets4A5gEdBq+4Y6bQwDZkhaDNwJnFHyrwQmSVrYfjN38QuqYOzHwCllBudSqs3a90haCvwPG27W7jSguWwSv58qKAT4AXBU22buTspFRERED8k55KfHJE2j843efVq1R6mlt4cREa+S/BqI/k5Sq+3mes/67B6l2HiamqAlcVJEREQCpfVhe2JvjyEiIiI2vr6+RykiIiJio0mgFBEREdFAlt6ig9ZW6Pj1TBGxOcjG7YieyYxSRERERAMJlCIiIiIa2CwDJUnjJe3d2+Poqb467oiIiM1Vnw6UytEk9YwH+mLA0VfHHRERsVnqlUBJ0iRJp5X0hZJuL+lDJU0v6QmSlkhaKmlKTd2Vki6QtAg4QNK5ku4vR3acL2kMcARwXjnWY9d2fW8v6TpJi8o1puSfUfpaKunTJW+UpAclTZP0sKTpkg6TdJekR8phukiaLOkKSfNK/sdK/lBJt0m6p7zLkTXjOLGMeVGp22HckmZKmiJpQel/bKk7UNJ5ku4ubXy85O8gaXapv1TS2FJ2WrlfIun0jfJDjYiI2Az11l+9zQHOBC4CmoHB5SDZscBsSTsCU4Am4GngFknjbV8PbA3Mt32mpG2By4A9bVvSCNvPSLqRxkeMXATMsn1UmZEaKqkJOAl4JyBgvqRZpe/dgGOAjwJ3A38HHEgV1Pwr1SwQwNuBd5XxLZT0Q+B3wFG2n5U0EvhpGdvewDnAGNu/l/Ra239sP25Vf3q2he39Jb0X+DfgMODvgeW295M0GLhL0i3AB4GbbX+pvNtWwGhgJ9tvLW2OqPcDkXQycHJ1t0tnP7uIiIh+o7eW3lqBJknbAKuBeVQB01iqIGo/YKbtp2yvAaYDB5W6a4FrS3o51UG4l0n6IPB8N/o+FPhvANtrbS+nCnyus/2c7ZXA98tYAB63vcT2OuA+4DZXB+QtAUbVtHuD7Rds/57qIN79qYKuL5cDd38C7ARsX8ZwdSmL7T92Mt7v13xmbf0dDpwo6V5gPrAtsDtVIHeSpMnA22yvAB4D3iTp65L+Bni2Xie2p9purs662a6T4URERPQfvRIo2X4ReByYCMylCo4OoZq9eaCL6qtsry3trKEKSK4B3g/ctBGGu7omva7mfh0vn5Fr/+0kBo6jijqabI8GlgFD1rP/tTX9CTjV9uhyvdH2LbZnUwWUvwamSTrR9tPAPsBM4BTg0h72HxER0W/15mbuOcBZwOySPgVYWGZrFgAHSxpZlpAmALPaNyBpKDDc9o+A06kCAoAVwLAG/d4GfKLUHyhpeOl/vKStJG0NHFXyeuJISUPKcuA4qtmd4cDvbL8o6RDgDaXs7cAxpSySXtuNcde6GfhEWa5E0pslbS3pDcAy25dQBUT7liW/AbavpVru27eH7xUREdFv9XagtAMwz/YyqiW0OQC2nwTOplrCWgS02r6hThvDgBllaetO4IySfyUwSdLC9pu5gU8Bh0haQrWctbfte4BpVAHafOBS2wt7+D6Ly3h/CnzR9m+olgybS18nAg+W97sP+BIwq2xK/49ujLvWpcD9wD2SlgL/QzXbNA5YJGkhcCzwn1TLfTPLMt13gX/p4XtFRET0W3K+z/4VK3uCVto+v7fHsiFIzYaW3h5GRGwE+S8/oiNJrdUe3Y769PcoRURERGxMORR3A7A9ubfHsCE1NUFLJpQiIiIyoxQRERHRSAKliIiIiAay9BYdtLZC9aXgEdEXZcN2xIaTGaWIiIiIBhIoRURERDSQQCkiIiKigQRKG5mk8ZL23sBtzpTUXNI/kjRiQ7YfERERlQRKG0g5k66e8UCXgZKk9dpYb/u9tp9Zn7oRERHRuX4fKEmaJOm0kr5Q0u0lfaik6SU9QdISSUslTampu1LSBeW8tgMknSvpfkmLJZ0vaQxwBHCepHvbn98maZqkiyXNB74qaX9J88pZb3Ml7VHKvUbSlZIekHQd8JqaNp4ohwePKue+teWfVY5WQdJpNeO6cuN8khEREZuffD1AdRDvmcBFQDMwWNIgYCwwW9KOwBSgCXgauEXSeNvXA1sD822fKWlb4DJgT9uWNML2M5JuBGbYvqZB/zsDY2yvlbQNMNb2GkmHAV8GPgR8Anje9l6S3g7c08N3PBt4o+3VjZbpJJ0MnFzd7dLD5iMiIjZP/X5GCWgFmkqQshqYRxUwjaUKovYDZtp+yvYaYDpwUKm7Fri2pJcDq4DLJH0QeL6b/V9te21JDweuLjNDFwJvKfkHAd8FsL0YWNzDd1wMTJd0PLCmXgHbU203V4cCbtfD5iMiIjZP/T5Qsv0i8DgwEZhLFRwdAuwGPNBF9VVtQU4JovYHrgHeD9zUzSE8V5P+InCH7bcCHwCGdLMNqAKg2p9nbd33Ad8A9gXuXt/9UBEREf1Nvw+UijnAWcDskj4FWGjbwALg4LIPaCAwAZjVvgFJQ4Hhtn8EnA7sUx6tAIZ1cxzDgV+X9MSa/NnA35V+3gq8vU7dZcDrJG0raTBVsIakAcDrbd8B/HPpY2g3xxMREdGvJVCqzAF2AObZXka1hDYHwPaTVHt87gAWAa22b6jTxjBghqTFwJ3AGSX/SmBS2aC9a516tb4KfEXSQl6+f+y/gaGSHgC+QLVc+DJlZuwLVIHdrcCD5dFA4LuSlgALgYvyV3IRERHdI+dQoGhHaja09PYwImI95b/1iJ6R1Frt0e0oM0oRERERDWRTb3TQ1AQtmVCKiIjIjFJEREREIwmUIiIiIhrI0lt00NoKUm+PIiK6Ixu3IzauzChFRERENJBAKSIiIqKBBEoRERERDfTrQEnSCEmffAX1R0t674YcU0RERGw6+nWgBIwA1jtQAkYDPQqUVNlgn3s5f67hfXfrRUREREf9PVA6F9hV0r2SzgOQNEnS3ZIWS/p8yTtK0m0lyNlB0sOSdqE6W+3YUv9YSZMlndXWuKSlkkaV6yFJ3wGWAq+v1097kg6XNE/SPZKuLgfvIukJSVMk3QMcU+d+gqQlpf8pNe2tlHSBpEXAARvnI42IiNh89PdA6WzgUdujbU+SdDiwO7A/1WxRk6SDbF8HPAn8I3AJ8G+2fwF8Driq1L+qi752B75p+y3AHvX6qS0saSRwDnCY7X2pDl87o6bIH2zva/vK2ntgNjAFOLS0vZ+k8aXM1sB82/vYvrNdfydLapHUAk9146OLiIjY/OV7lF7u8HItLPdDqQKa2cCpVLNBP7X9vfVo++e2f9qNftq8C9gbuEvVlxptCcyred4+MGu73w+YafspAEnTgYOA64G1wLX1Bmd7KjC1qtOcb2aJiIgggVJ7Ar5i+3/qPNsZWAdsL2mA7XV1yqzh5bN0Q2rSz3Wzn9oyt9qe0OD5c13c17PK9tpulIuIiAiy9LYCGFZzfzPw0Zq9QDtJep2kLYDLgQnAA7y0BNa+/hPAvqXuvsAbG/Rbt592ZX4K/JWk3UqZrSW9uRvvtAA4WNLIsmF7AjCrG/UiIiKinX49o2T7D5LukrQU+HHZp7QXMK8sd60EjgdOAebYvrNshL5b0g+BO4CzJd0LfIVqWetESfcB84GHG/R7S4N+fldT5ilJE4HvSRpcss9p1GZNvSclnV3GJuCHtm/o8YcTERERyDkoKNqp9ii19PYwIqIb8l94xCsnqdV2c71n/XpGKepraoKWxEkRERH9fo9SREREREMJlCIiIiIayNJbdNDaCtUe84jYGLKvKKLvyIxSRERERAMJlCIiIiIaSKAUERER0UACpV4g6dOStqq5X9mb44mIiIj6Eij1jk8DW3VZKiIiInrVJhMoSRol6UFJ0yQ9LGm6pMPKESOPSNq/lNta0uWSFkhaKOnImvpzJN1TrjElf5ykmZKuKe1Plzr+TZek0yTdL2mxpCtL3mRJ3y7t/lzSByV9VdISSTdJGlTKvbuMZUkZ2+BG+ZJOA3YE7pB0R03/X5K0SNJPJW1f8qZJukjSXEmPSTq6pvwkSXeX8X6+5rP5YWlnqaRjS/65Ne92/sb4+UVERGyWbG8SFzAKWAO8jSqAa6U6iFbAkcD1pdyXgeNLegTV2WdbU83QDCn5uwMtJT0OWA7sXNqdBxxYp//fAIPb2i3/TgbuBAYB+wDPA+8pz64DxgNDgF8Cby7536GaMaqbX9JPACNr+jbwgZL+KnBOSU8Dri7j3hv4Wck/HJhaPpsBwAzgIOBDwCU17Q4HtgUe4qXjakY0+PxPpjq3pAV2cfUHzLly5doYV0RsWtpihnrXJjOjVDxue4ntdcB9wG3lBZZQBVJQBQltB9HOpApIdqEKZi6RtIQquNi7pt0Ftn9V2r23pq1ai4Hpko7/f+zde5xXVb3/8ddbRFFETCOPlkSpRd5AZ7TES3jJc8pSLMpS8daRrA6mHktP9isz63g5Hk07ZmiEF4568JZZeUNBRAVm5DKgoie1ND1mpQgaqDOf3x/7M7Gd+X5nBhic2/v5eHwfs/faa6/12XsKP4+11nwXRcLW7LcR8WbG0A+4I8ubY/pwxt28We1VFElLtfJK3qBIdqBIEMvx3RoRTRHxKLBl6R0cBMwDHgGGUySHDcAnJJ0naZ+IWEqRJK4Afi7psxTJXisRMTEiaiOiFoZUCdPMzKxv6W5fOLmydNxUOm9iVawCPhcRS8o3SjoLeJFi5Gc9iuSgUruNVH7ugykSmc8AZ0rauXxvRDRJejMTt5Yxra1yuy3jK8eu0s9/j4iftWxI0m7Ap4BzJE2LiLNz2vIAYCzwL8D+nRS3mZlZr9bdRpQ64k5gQvM6I0m7Zvlg4IUcNRpHMfrTIZLWA7aJiPuA07OtTTp4+xJgmKTt8nwcMKONcoBlwKCOxlfBncDxkjbJ+N8r6T2StgZej4hrgQuA3bLO4Ij4DXAKRSJpZmZmHdDdRpQ64gfAxcDCTHCeBj4NXAbcJOloiumx11ajzX7AtZIGU4zWXBIRr1RY891KRKyQdBwwVdL6wFzg8ohYWak8b5sI3CHp+YjYbzXibO7zLkkfAR7KGJcDRwHbARdIagLeBL5KkZD9UtKAfLZTV7c/MzOzvkqrZnzMClJtFOu6zWxd8D+7Zt2LpPpijW5rPXHqzczMzOwd0ROn3mwdq6mBOg8omZmZeUTJzMzMrBonSmZmZmZVeOrNWqmvhw78wZ+ZdZAXb5v1XB5RMjMzM6vCiZKZmZlZFU6UzMzMzKro1omSpOGS5kuaJ2nbtWxrpKRPdaDeaEm3d6DedEm1efwbSZutTXxrQtLZkg58p/s1MzPrK7p1ogSMAW6MiF0j4nfNhSqsbuwjKTaL7XQR8amIeGVdtN1Ov9+NiHve6X7NzMz6ijaTDUnDJD0uabKkJyRNkXSgpFmSnsxd6ZE0UNIkSXNy9OfQ0v0zJT2Sn1FZPjpHZG7M9qc0b3Jb6vtTwMnAVyXdl20tkXQ1sAjYRtJPJdVJWizp+6V7d5f0oKQFGdNg4Gzg8ByhOlzSHpIeyngflPThdt7FRpKul/SYpFuAjUrXnpH07k54X8dKulnSHVn//Czvl20uktQg6ZQsnyxpbB4fkG01ZNsblmL7fr7/BknD2/nfhJmZmTWLiKofYBjwFrAzRVJVD0yi2Fz1UODWrPcj4Kg83gx4AhgIbAwMyPLtgbo8Hg0sBd6X7T4E7F2h/7OA00qxNAEfK13fPH/2A6YDuwAbAE8Bu+e1TSm+BuFY4CelezcF1s/jA4GbSrHdXiGWU4FJebxLvpfaPH8GeHcnvK9jM/bBwADg98A2QA1wdymWzfLnZGBs1n0W+FCWXw2cXIptQh5/Dbiyyu96PMUGb3UwNIo/aPbHH38642Nm3VtzflLp05Hpq6cjoiEimoDFwLRstCETA4CDgDMkzc+EZQAwFOgPXCGpAZgK7FBqd05EPJftzi+11ZbfR8TDpfMvSHoEmAfsmO1/GHghIuYCRMSrEfFWhbYGA1MlLQIuyvvbsi9wbba5EFhYpd7avC+y/tKIWAE8CryfInn6oKRLJf0T8GqLPj+c/T6R51dlvM1uzp/1VHnPETExImojohaGVH8LZmZmfUhHvnByZem4qXTeVLpfwOciYkn5RklnAS8CIyhGWFZUabexg7G8Vmr7A8BpFCNHL0uaTJFwdNQPgPsi4jBJwygSls6wNu/ro1R4L/l8I4B/BE4EvgAcvwYxdfQ9m5mZGZ23mPtOYELzOiNJu2b5YIrRnSZgHMUUWWfZlCJxWippS+CTWb4E2ErS7hnLIEnrA8uAQaX7BwN/zONjO9Df/cAR2eZOFNNva6ra+6pI0ruB9SLiJuA7wG4tqiwBhknaLs/HATPWIj4zMzOj8xKlH1BMsy2UtDjPAS4DjpG0ABhOaURobUXEAoopt8eB/wZmZfkbwOHApdnv3RQjTfcBOzQv5gbOB/5d0jw6NsryU2ATSY9RLAyvX4vwq72vat4LTM+pumuBfytfzGm64yimEhsoRq8uX4v4zMzMDFCxfMZsFak2inXdZtYZ/M+sWfcmqb5Yo9ua16tYKzU1UOc8yczMrNt/4aSZmZlZl3GiZGZmZlaFEyUzMzOzKrxGyVqpr4e3byhjZmvCi7jNej6PKJmZmZlV4UTJzMzMrAonSu2QtJmkr3V1HGZmZvbOc6LUvs2AbpsoSerX1nmVeyTJv3szM7N2dJv/WEo6WtJCSQskXZNlwyTdm+XTJA3N8smSfirpYUlPSRotaZKkx3Jz3OY2l0u6SNLivH9Ilp8gaW72dZOkjbN8S0m3ZPkCSaOAc4Ftc+uTC7Kv6ZJulPS4pCmlPdtqJM2QVC/pTklbZflJkh7N57g+yz6ebc6XNE9SeR+65viPkjQn6/ysOQnK57owt2jZs8L5qZIW5efk0rtcIulqYBGwzTr5RZqZmfUmEdHlH2BH4Ang3Xm+ef78FXBMHh8P3JrHk4HrAQGHAq8CO1MkfvXAyKwXwJF5/F3gJ3m8Ranvc4AJeXwDcHIe96PYOHcYsKhUfzSwFHhf9vcQsDfF3m0PAkOy3uHApDx+HtgwjzcrPdteebwJsH6Ld/KRrNM/zy8Dji491xdKdf9+DtQADcDAbHcxsGs+RxPwsSq/g/EU+5bUwdAo/l7HH3/8WZuPmfUMQF1E5Rylu4wo7Q9MjYg/A0TEX7N8T4oNbwGuoUhImv0qH64BeDEiGiKiiSIxGJZ1miiSHyg2k22+fydJM3MD2SMpErXmOH6aMTRGxNIq8c6JiOeyv/nZ34eBnYC7c/Pa71AkUwALgSmSjgLeyrJZwH9KOokieXqLtzuAIumZm+0dAHwwrzUCN5Xqls/3Bm6JiNciYjlwM7BPXvt9RDxc6YEiYmJE1EZELQyp8thmZmZ9S0/+HqWV+bOpdNx8Xu25In9OBsZExAJJx1KMEq1J31AkKetTjG4tjog9K9Q/GNgX+AxwpqSdI+JcSb8GPgXMkvSPEfF46R4BV0XEv1Vob0VENLZxXs1rHahjZmZmqbuMKN0LfF7SFgCSNs/yB4Ev5vGRwMzVbHc9YGweHwE8kMeDgBck9c92m00Dvpox9JM0GFiW9duzBBgiac+8v7+kHXPR9DYRcR9wOsV03iaSts1RsPOAucDwFu1NA8ZKek+2t7mk93cgjpnAGEkbSxoIHMbqvzczMzOjmyRKEbEY+CEwIxck/2demgAcJ2khMA74xmo2/Rqwh6RFFNNqZ2f5/wNmU0x/lUdxvgHsl1Ny9cAOEfEXihGfRZIuaOMZ3qBIys7LZ5gPjKJY63RttjkPuCQiXgFOzjYXAm8Cv23R3qMU03d3ZZ27ga3ae+CIeIRixGxOPuOVETGvvfvMzMysNRXLfHonScsjYpOujqOnkWqjWNdtZmujF//zatarSKov1ui21i1GlMzMzMy6o568mLtdHk1aMzU1UOcBJTMzM48omZmZmVXjRMnMzMysil499WZrpr4eik1ZzGx1ePG2We/jESUzMzOzKpwomZmZmVXhRMnMzMysii5PlCSNkbRDlWtDJM2WNE/SPpXqrEY/wyQd0cF6izpQb7KksXl8ZbVnWJcknSjp6He6XzMzs77iHUuUJPWrcmkMUC3JOABoiIhdI+Jt+5W10V41wyj2e+t0EfHPueXIOyoiLo+Iq9/pfs3Mw0ZNzwAAIABJREFUzPqKdhMlSd+UdFIeXyTp3jzeX9KUPP6SpIbcu+y80r3LJV2Ye5/tKelcSY9KWijpPySNAg4BLpA0X9K2pXtHAucDh+a1jSq0911Jc7PfiVLxt1qStpN0j6QFkh7Jds8F9sm2TsmRo5l5/ZGMpa33IEk/kbRE0j3Ae0rXpkuqLT3zBZIWZwx75PWnJB2Sdfplnbn5Lr6S5aOz7o2SHpc0pfRMb3t3WXaWpNOa35ekh/P6LZLeVYrtPElzJD2xtiNzZmZmfUpEtPkBPgZMzeOZFJut9ge+B3wF2Br4AzCE4usG7gXGZP0AvpDHWwBLWLW/3Gb5czIwtkrfxwI/KZ3/vb0837x0fA3wmTyeDRyWxwOAjYHRwO2l+hsDA/J4e6Auj4cBiyrE8lmKjWn75TO/0hw3MB2oLcX4yTy+Bbgr39cIYH6Wjwe+k8cbUmys9oGMcSnwPook9iFg7zbe3VnAaXm8EPh4Hp8NXFyK7cI8/hRwT5V3PT7jqIOhUfyhsz/++LM6HzPrmZpzgEqfjky91QM1kjYFVuZ/vGuBfSgSp92B6RHxUkS8BUwB9s17G4Gb8ngpsAL4uaTPAq93oO+Wyu0B7JdrmBqA/YEdJQ0C3hsRtwBExIqIqNRXf+CKvHcq1af/mu0LXBcRjRHxPEVCWMkbwB153ADMiIg383hYlh8EHC1pPkVStwVFsgYwJyKei4gmYH7e0+a7kzSYInmakUVXsep3AHBz/qwvxfA2ETExImojorbIec3MzKzdRCn/I/80xejOgxTJ0X7AdsBj7dy+IiIas523gD2AG4FPsyqZWB1/b0/SAOAyilGdnYErKEaPOuoU4EWKkZ5aYIM1iKeSNzM7BWiiSC7JxKf5Cz4FTIiIkfn5QETclddWltpqBNbvhHfX3GYj/pJRMzOzDuvoYu6ZwGnA/Xl8IjAvE4I5wMclvTsXWH8JmNGyAUmbAIMj4jcUScqIvLQMGLQGsTcnRX/OtscCRMQy4DlJY7LfDSVtXKGfwcALmcCMo5hSa8v9wOG5vmgrimRxTd0JfFVS/4zxQ5IGVqvcxrsDICKWAi+X1h+No8LvwMzMzFZPR0cXZgJnAg9FxGuSVmQZEfGCpDOA+yhGSn4dEb+s0MYg4Jc5EiTg1Cy/nmIK7CSK0aHfdSSgiHhF0hXAIuD/gLmly+OAn0k6G3gT+DzFGp7GXAg+mWI06qb88/o7gNfa6fIWium9RynWZD3UkTiruJJiCuyRXKz9EsVf/1VT7d2VHQNcnknhU8BxaxGfmZmZsWpxsNnfSbVRrOs2s9Xhf07NeiZJ9cUa3da6/AsnzczMzLorL+y1VmpqoM4DSmZmZh5RMjMzM6vGiZKZmZlZFZ56s1bq66HYOMXMvEDbrG/ziJKZmZlZFU6UzMzMzKpwogRIGibpiNL5sZJ+0pUxmZmZWddzolQYBhzRXqXuRtL6bZ23cV9727WYmZkZ3ThRkjRQ0q8lLZC0SNLhWf6MpH+XNF9SnaTdJN0p6XeSTsw6knRB3tdQurdiOXAusE+2eUqWbS3pDklPSjq/FNdyST/MuB6WtGWWD5F0k6S5+dkryz+e7c6XNE/SIElbSbo/yxaV9mgrP3+NpBmS6vP5tsry6ZIullQHfKPC+QHZT4OkSZI2LL238yQ9QrGli5mZmbWjO//V2z8Bz0fEwQCSBpeu/SEiRkq6iGLftr0oNsldBFwOfBYYSbF57LuBuZLuB0ZVKT8DOC0iPp19HZv1dgVWAkskXRoRzwIDgYcj4sxMoE4AzgF+DFwUEQ9IGkqx8e1HKDYT/npEzMrNbVcA44E7I+KHObqzcfnBc7PcS4FDI+KlTOh+CByfVTZo/qp1SZ9pPs+94J4EDoiIJyRdDXwVuDjv+0tE7FbpZUsan3EBQ6v+UszMzPqS7pwoNQAXSjoPuD0iZpau3Vaqs0lELAOWSVopaTNgb+C6iGgEXpQ0A9i9jfJXK/Q/LSKWAkh6FHg/8CzwBnB71qkHPpHHBwI7aNXf1W+aidEs4D8lTQFujojnJM0FJmVCdGtEzG/R94eBnYC7s71+wAul6ze0qH9D6b6nI+KJPL8K+DqrEqWW9/1dREwEJhbPW+s/iDYzM6MbT73lf+x3o0iGzpH03dLllfmzqXTcfN5ZyV+53cZSu2/Gqp2Ey+XrAR+LiJH5eW9ELI+Ic4F/BjYCZkkaHhH3A/sCfwQmSzq6Rd8CFpfa2jkiDipdf61F/Zbn1XS0npmZmdGNEyVJWwOvR8S1wAUUSVNHzQQOl9RP0hCKpGROG+XLgEFrGfJdwIRS/CPz57YR0RAR5wFzgeGS3g+8GBFXAFdWeLYlwBBJe2Yb/SXt2IEYlgDDJG2X5+OAGWvzUGZmZn1Zd5562xm4QFIT8CbFWpuOugXYE1gABPCtiPg/SdXK/wI0SlpAsebp5TWI9yTgvyQtpHiv9wMnAidL2o9itGsx8Fvgi8A3Jb0JLAfeNqIUEW9IGgtckmuz1qeYPlvcVgARsULSccBUFX8BN5dizZaZmZmtAYW/n99aKNYo1XV1GGbdgv+JNOv9JNU3/5FUS9126s3MzMysq3XnqTfrIjU1UOcBJTMzM48omZmZmVXjRMnMzMysCk+9WSv19bDqezPNejcv1jaztnhEyczMzKwKJ0pmZmZmVThRMjMzM6vCidI6IGmMpB26Og4zMzNbO06U1o0xQMVEKbcW6RQt2+po250Zg5mZWW/W4xMlSQMl/VrSAkmLJB0uaX9Jt5bqfCL3eUPSckkXSFos6R5Je0iaLukpSYdknWMl3SrpbknPSPoXSadKmifpYUmbZ71tJd0hqV7STEnDJY0CDqHYp25+1pku6WJJdcCZkp6W1D/b2LR8Xop5iKSbJM3Nz15ZfpakayTNAq6pcD5M0r2SFkqaJmlo3jdZ0uWSZgPnr+vfi5mZWW/QG0YW/gl4PiIOBshNZF8FLpM0JCJeAo4DJmX9gcC9EfHNTJ7OAT5BMQJ0FXBb1tsJ2BUYAPwvcHpE7CrpIopNbC8GJgInRsSTkj4KXBYR+0u6Dbg9Im7MmAA2aN5HRtIw4GDgVooNcm+OiDdbPNePgYsi4oFMdu4EPpLXdgD2joi/STqrxfmvgKsi4ipJxwOXUIxwAbwPGBURjS1foqTxwPjibGh779zMzKxP6A2JUgNwoaTzKJKTmQCSrgGOkvQLYE+K5AbgDeCO0r0rI+JNSQ3AsFK790XEMmCZpKXAr0r37CJpE2AUMFWrvnRowzbivKF0fCXwLYpE6TjghAr1DwR2KLW9afYJcFtE/K1Ut3y+J/DZPL6Gt48eTa2UJAFExESKxC83xTUzM7MenyhFxBOSdgM+BZwjaVpEnA38giK5WUGRILyVt7wZ8fevmGsCVmY7TS3W7qwsHTeVzpso3tt6wCsRMbKDob5WinlWTpGNBvpFxKIK9dcDPhYRK8qFmTi91qJuy/N2YzAzM7P29YY1SlsDr0fEtcAFwG4AEfE88DzwHYqkqVNFxKvA05I+n3FI0oi8vAwY1E4TVwP/3UZsdwETmk8kdTQhe5BiOg/gSGBmB+8zMzOzFnp8ogTsDMyRNB/4HsWao2ZTgGcj4rF11PeRwJclLQAWA4dm+fXAN3Px97ZV7p0CvAu4rsr1k4DaXJT9KHBiB2OaABwnaSEwDvhGB+8zMzOzFhS9eKMjST8B5kXEz7s6lpYkjQUOjYhxXR1LS8UapbquDsPsHdGL/wk0sw6SVN/8B1ct9fg1StVIqqdYk/OvXR1LS5IuBT5Jsa7KzMzMuqlemyhFRE1Xx1BNRExov1bXqamBOg8omZmZ9Yo1SmZmZmbrhBMlMzMzsyp67dSbrbn6elj1PZdmPZsXa5vZ2vCIkpmZmVkVTpTMzMzMqnCiZGZmZlZFj0uUJI2RtENXx9FVJJ0t6cB26oyWNOqdisnMzKy36raJkqR+VS6NAfpsohQR342Ie9qpNhpwomRmZraWOj1RkvRNSSfl8UWS7s3j/SVNyeMvSWqQtEjSeaV7l0u6MPdO21PSuZIezf3O/iNHSQ4BLpA0v+U+apK2lHSLpAX5GZXlp2ZfiySdnGXDJD0m6QpJiyXdJWmjvLadpHuyjUckbStpE0nT8rxB0qFZ91xJXy/FcJak00rvYm7G//0q72t5vqfF2f6QLB8p6eG89xZJ78ryybn9CZKekfT9UkzDJQ2j2BfulHxH+0j6fD77Akn3r9Uv2MzMrA9ZFyNKM4F98rgW2ERS/yy7X9LWwHnA/sBIYHdJY7L+QGB2RIwAHgMOA3aMiF2AcyLiQeA24JsRMTIiftei70uAGXn/bsBiSTXAccBHgY8BJ0jaNetvD/xXROwIvAJ8LsunZPkIipGZF4AVwGERsRuwH3ChJAE3AF8oxfAF4AZJB2X7e+Rz1kjat8L7GgjUZQwzKDb2BbgaOD2fvaFU3tKfM6afAqdFxDPA5cBF+Y5mAt8F/jGf55BKjUgaL6lOUh28VKUrMzOzvmVdJEr1FEnBpsBK4CGKhGkfiiRqd2B6RLwUEW9RJCXNCUQjcFMeL6VITn4u6bPA6x3oe3+KhIGIaIyIpcDewC0R8VpELAduZlUi93REzC/FPUzSIOC9EXFLtrMiIl4HBPxI0kLgHuC9wJYRMQ94j6StJY0AXo6IZ4GD8jMPeAQYTpE4tdREkWwBXAvsLWkwsFlEzMjyq0rvqKWby/FXqTMLmCzpBKDilGZETIyI2mJTwCFVmjEzM+tbOv0LJyPiTUlPA8cCDwILKUZgtqMYJaqULDRbERGN2c5bkvYADgDGAv9CkQh1ppWl40ZgozbqHkmRQdTkMz4DDMhrUzPGf2BV0iPg3yPiZ6sZ0+p+PV7zMzRS5fcZESdK+ihwMFAvqSYi/rKa/ZiZmfU562ox90zgNOD+PD4RmBcRAcwBPi7p3blg+0sUU05vI2kTYHBE/AY4BRiRl5YBg6r0Ow34at7fL0dmZgJjJG0saSDFdN7MaoFHxDLguebpQEkbStoYGAz8KZOk/YD3l267AfgiRbI0NcvuBI7P50DSeyW9p0KX6+V9AEcAD+RI2MuSmke+xlHhHbXhbe9I0rYRMTsivksxr7bNarRlZmbWZ63LRGkr4KGIeJFiCm0mQES8AJwB3AcsAOoj4pcV2hgE3J5TXQ8Ap2b59cA3Jc1ruZgb+Aawn6QGiqmoHSLiEWAyRYI2G7gyp8vaMg44Kft+kGKkaApQm20fDTzeXDkiFme8f8znIyLuAv4beCjvuZHKCd5rwB6SFlGMmJ2d5cdQLFpfSLHG6ewK91bzK+Cw5sXc2U5D9vEgxXs3MzOzdii8EVKXkrQ8Ijbp6jjKpNqAuq4Ow6xT+J84M2uPpPpijW5r3hTXWqmpgTrnSWZmZt33Cyf7iu42mmRmZmarOFEyMzMzq8KJkpmZmVkVXqNkrdTXg9TVUZh5IbaZdT2PKJmZmZlV4UTJzMzMrAonSj2cpGGSjujqOMzMzHojJ0o93zCKrU9akeQ1aGZmZmuhxyVKOYLyuKTJkp6QNEXSgZJmSXoyN9JF0kBJkyTNye1ODi3dP1PSI/kZleWjJU2XdGO2P0VqvaRZ0kmSHpW0UNL1ktbLfofk9fUk/a+kIRnjTyU9LOmp7GOSpMckTS61uVzSBZIWS7pH0h4Zy1OSDsk6/bLO3Oz7K3n7ucA+uV3JKZKOlXSbpHuBaZKubt63LtuZ0vwuzMzMrG09LlFK2wEXAsPzcwSwN8VGvN/OOmcC90bEHsB+FPudDQT+BHwiInYDDgcuKbW7K3AysAPwQWCvCn2fAewaEbsAJ0ZEE3AtcGRePxBYEBEv5fm7gD0pNva9DbgI2BHYWdLIrDMwY92RYkPbc4BPUGzg27zH25eBpRGxO7A7cIKkD2Q8MyNiZERclHV3A8ZGxMeBnwPHAuQmwaOAX7d8KEnjJdVJqiv2zTUzM7Oemig9HRENmaQsBqZFsWldA8VUFMBBwBmS5gPTgQHAUKA/cEVuVDuVIilqNicinst255faKlsITJF0FPBWlk2i2CgX4HjgF6X6vyrF9mKLuJvbfwO4I48bgBkR8WaF5zk6n2c2sAWwfZX3c3dE/BUgImYA2+eI15eAmyLirZY3RMTEiKgt9roZUqVZMzOzvqWnrmFZWTpuKp03seqZBHwuIpaUb5R0FvAiMIIiUVxRpd1GKr+fg4F9gc8AZ0raOSKelfSipP2BPVg1ulRusxxny1jfjFW7E/+9XkQ0ldYZCZgQEXe2eJ7RFWJ8rcX51cBRwBeB4yrUNzMzswp66ohSR9wJTGheZyRp1ywfDLyQozrjgH4dbVDSesA2EXEfcHq21bxX25UUU3BTI6Kxcx7hbe4Eviqpf8byoZxKXAYMaufeyRRTikTEo+sgNjMzs16pNydKP6CYZlsoaXGeA1wGHCNpAcX6ppajL23pB1yb03bzgEsi4pW8dhtF0vSLajevpSuBR4FHJC0CfkYxIrUQaJS0QNIplW6MiBeBx9ZhbGZmZr2SwnsEdApJtcBFEbFPV8fSkqSNKdY77RYRS9uvXxtQt+4DM2uH/3kys3eCpPpijW5rvXlE6R0j6QzgJuDfujqWliQdSDGadGlHkiQzMzNbxSNK1kptbW3U1XlEyczM+gaPKJmZmZmtASdKZmZmZlX01O9RsnWovh5ab95i9s7wagAz6048omRmZmZWhRMlMzMzsyqcKJmZmZlV4UQpSTpR0tHt1+xQW9/ujHbMzMysa/l7lABJ60fEW53Y3vKI2KT9mm+7p9/q7hHXMu6OPkd79fzN3NaV/E+Smb3Tev33KEkaJulxSVMkPSbpxty2A0k1kmZIqpd0p6Stsny6pIsl1QHfkHSWpNNK1y6SVJft7S7pZklPSjqn1O9RkuZImi/pZ5L6SToX2CjLplSrl+XLJV2Y+87t2eKZtpV0R8Y9U9LwLJ8s6XJJs4HzK5yPlPSwpIWSbpH0rkrPu25/I2ZmZr1Dr0iU0oeByyLiI8CrwNck9QcuBcZGRA0wCfhh6Z4NIqI2Ii6s0N4bmV1eDvwS+DqwE3CspC0kfQQ4HNgrIkYCjcCREXEG8LeIGBkRR1arl30MBGZHxIiIeKBF/xOBCRn3aRSb+TZ7HzAqIk6tcH41cHpE7EKxv9v3OvK8ksZnYlgHL1V4HWZmZn1Pb/oepWcjYlYeXwucBNxBkdzcreKLgfoBL5TuuaGN9m7Lnw3A4oh4AUDSU8A2wN5ADTA3294I+FOFdg5oo14jxR5xbyNpE2AUMFWrvtBow1KVqS2m6aZGRKOkwcBmETEjy68CpnbkeSNiIkVyllNvZmZm1psSpZb/cQ9AFEnOnhXqA7zWRnsr82dT6bj5fP1s+6qIaG8j3LbqraiyLmk94JUcgaqkZdxtPcea1DMzMzN619TbUEnNCdERwAPAEmBIc7mk/pJ27KT+pgFjJb0n295c0vvz2ps57ddevYoi4lXgaUmfz3skaUR7AUXEUuBlSftk0ThgRhu3mJmZWRt6U6K0BPi6pMeAdwE/jYg3gLHAeblgej7FlNZai4hHge8Ad0laCNwNbJWXJwILJU1pp15bjgS+nHEvBg7tYGjHABdkXyOBszv6TGZmZvZ2veLrASQNA26PiJ26OJRewV8PYF2pF/yTZGY9TFtfD9Cb1ihZJ6mpgTrnSWZmZr0jUYqIZyj+us3MzMys0/SmNUpmZmZmncqJkpmZmVkVvWLqzTpXfT2s+p5LszXnhdlm1tN5RMnMzMysCidKZmZmZlU4UepmJB0raeuujsPMzMycKHVHxwJOlMzMzLqBLkmUJA2T9LikyZKekDRF0oGSZkl6UtIeWW+gpEmS5kiaJ+nQ0v0zJT2Sn1FZPlrSdEk3ZvtTpNbLkiVtJ+keSQvy/m1zP7ULJC2S1CDp8FKbMyT9UtJTks6VdGTG1CBp26w3WdLlkurymT7dVqx57fRsY0G2OxaoBaZImi9pI0nPSPp+3tsgaXg772bHLJsvaaGk7bPur7OfRc3PZmZmZu2IiHf8AwwD3gJ2pkjW6oFJgCj2NLs16/0IOCqPNwOeAAYCGwMDsnx7oC6PRwNLgfdluw8Be1fofzZwWB4PyPY+R7EPWz9gS+APFHuyjQZeyeMNgT8C3897vwFcnMeTgTuy3+2B50ptV4r1k8CDwMZ5vnn+nA7UlmJ9BpiQx18Drmzn3VwKHJnlGwAb5bNdUWpzcIV3Mp5i35I6GBrF3yv548/afczMeoLm/zZX+nTl1NvTEdEQEU0Um75Oy2AbKBIpgIOAMyTNp0ggBgBDgf7AFZIagKnADqV250TEc9nu/FJbAEgaBLw3Im4BiIgVEfE6sDdwXUQ0RsSLwAxg97xtbkS8EBErgd8Bd2V5Q4v2/ycimiLiSeApYHgbsR4I/CL7JiL+2sa7ujl/1nfg3TwEfFvS6cD7I+JvGecnJJ0naZ+IWNqyg4iYGBG1EVELQ9oIxczMrO/oyu9RWlk6biqdN7EqLgGfi4gl5RslnQW8CIygGMFZUaXdRjrnGTsSK0C0uC+AU9qIdXX7Lz9PxXcDPCZpNnAw8BtJX4mIeyXtBnwKOEfStIg4ew3iMDMz61O6+2LuO4EJzeuMJO2a5YOBF3LUaBzFdFmHRMQy4DlJY7LNDSVtDMwEDpfUT9IQYF9gzmrG+3lJ6+W6pQ8CS9qI9W7guOwbSZtn+TJgUAf6qvhuJH0QeCoiLgF+CeySf0X3ekRcC1wA7Laaz2VmZtYndfdE6QcUU1cLJS3Oc4DLgGMkLaCY3nptNdsdB5wkaSHFOqF/AG4BFgILgHuBb0XE/61mu3+gSK5+C5wYESuqxRoRdwC3AXU5fXZatjEZuLx5MXcbfVV7N18AFmWbOwFXU6wFm5Nl3wPOWc3nMjMz65NULAuytSVpMnB7RNzY1bGsLak2inXdZmvH/7yYWU8gqb5Yo9tadx9RMjMzM+sy3hS3k0TEsV0dQ2epqYE6DyiZmZl5RMnMzMysGidKZmZmZlV46s1aqa+H1hu/mHWcF3GbWW/hESUzMzOzKpwomZmZmVXhRMnMzMysCidK65CksyUdmMcnN29XYmZmZj2DE6V1KCK+GxH35OnJQKcnSpL6tXXe0fvMzMystV6XKEk6WtJCSQskXZNlwyTdm+XTJA3N8smSLpH0oKSnJI0ttXO6pIZs59wsO0HS3Cy7SdLGkgZL+r2k9bLOQEnPSuqf7Y+VdBKwNXCfpPskHS/p4lJfJ0i6qMKzHCTpIUmPSJoqaZMsf0bSeZIeodiIt+X5lzL2RZLOK7W3XNKFue/cnuvg9ZuZmfUqvSpRkrQj8B1g/4gYAXwjL10KXBURuwBTgEtKt20F7A18GmhOiD4JHAp8NNs5P+veHBG7Z9ljwJcjYikwH/h41vk0cGdEvNncQURcAjwP7BcR+wH/A3xGUv+schwwqcWzvDuf5cCI2I1i87VTS1X+EhG7RcT15XPgfuA8YH9gJLC7pDFZZyAwOyJGRMQDLfobL6lOUh28VPkFm5mZ9TG9KlGiSA6mRsSfASLir1m+J/DfeXwNRWLU7NaIaIqIR4Ets+xA4BcR8XqLdnaSNFNSA3AksGOW3wAcnsdfzPOqImI5cC/waUnDgf4R0dCi2seAHYBZkuYDxwDvL11v2Ufz+e7A9Ih4KSLeokgM981rjcBNVWKaGBG1xaaAQ9oK38zMrM/wF07CytJxe1+zOBkYExELJB0LjM7y24AfSdocqKFIgtpzJfBt4HHgFxWuC7g7Ir5U5f7X2jmvZEVENHagnpmZmdH7RpTupVijswVAJi4AD1KM9EAxEjSznXbuBo5r/iu1UjuDgBdyyuzI5so5QjQX+DFwe5VkZFne33zPbGAb4Ajgugr1Hwb2krRdxjBQ0ofaiRtgDvBxSe/OBdtfAmZ04D4zMzNroVeNKEXEYkk/BGZIagTmAccCE4BfSPomxQKc49pp5w5JI4E6SW8Av6EY/fl/wOxsYzalxIdi6msqq0aZWpoI3CHp+VynBMVapZER8XKFGF7KUavrJG2Yxd8Bnmgn9hcknQHcRzEq9euI+GVb95iZmVllCm/K1GUk3Q5cFBHTujqWMqk2irXjZmvG/6yYWU8iqb5Yo9tab5t66xEkbSbpCeBv3S1JMjMzs1V61dRbTxERrwAdWW/UJWpqoM4DSmZmZh5RMjMzM6vGiZKZmZlZFZ56s1bq60HtfaOU9XlesG1mfYFHlMzMzMyqcKJkZmZmVoUTJTMzM7MqnCj1QJKekfTuPH6wq+MxMzPrrZwodROS1mhhfUSM6uxYzMzMrNCnEyVJwyQ9LmmypCckTZF0oKRZkp6UtEfWGyhpkqQ5kuZJOrR0/0xJj+RnVJaPljRd0o3Z/hSp9d+RZZ2LJdUB35D0GUmzs497JG2Z9baQdJekxZKupNjDrbmN5aU+by+V/yT3ikPSuZIelbRQ0n+ssxdqZmbWy/jrAWA74PPA8cBc4Ahgb+AQio1wxwBnAvdGxPGSNgPmSLoH+BPwiYhYIWl74Dqgea+YXYEdgeeBWcBewAMV+t+geX8ZSe8CPhYRIemfgW8B/wp8D3ggIs6WdDDw5Y4+nKQtgMOA4dnuZlXqjQfGF2dDO9q8mZlZr+ZECZ6OiAYASYuBaZlQNADDss5BwCGSTsvzARTZxPPATySNBBp5+7YkcyLiuWx3frZVKVG6oXT8PuAGSVsBGwBPZ/m+wGcBIuLXkl5ejedbCqwAfp4jTrdXqhQRE4GJRby1/oYcMzMz+vjUW1pZOm4qnTexKpEU8LmIGJmfoRHxGHAK8CIwgmIkaYMq7TZSPSl9rXR8KfCTiNgZ+ApFQtZRb/H23+cAgIh4C9gDuBH4NHDHarRpZmbWpzlR6pg7gQnN64wk7Zrlg4EXIqIJGAf0W8t+BgN/zONjSuX3U0wJIumTwLsq3Pt7YAdJG+ZSibMWAAAaQElEQVT02gFZfxNgcET8hiKxG7GWMZqZmfUZTpQ65gdAf2BhTs/9IMsvA46RtAAYzttHh9bEWcBUSfXAn0vl3wf2zb4/C/yh5Y0R8SzwP8Ci/DkvLw0Cbpe0kGLq79S1jNHMzKzPUHjDJmuhWKNU19VhWDfnfzrMrLeQVN/8h1UteTG3tVJTA3XOk8zMzDz1ZmZmZlaNEyUzMzOzKjz1Zq3U10Pr7xE3W8Xrk8ysr/CIkpmZmVkVTpTMzMzMqnCiZGZmZlaFE6UeRtKVknZop86Y9uqYmZlZ+5wo9TAR8c8R8Wg71cYATpTMzMzWUqcmSpKGSXpc0mRJT0iaIulASbMkPSlpj6w3UNIkSXMkzZN0aOn+mZIeyc+oLB8tabqkG7P9Kc37rrXofztJ90hakPdvq8IFkhZJapB0eHttStpd0oPZzhxJg9qI7XpJB5dimCxprKR+2e9cSQslfaWN9zVF0mMZy8Z57YB8Nw35rjbM8umSavN4uaQfZpwPS9oy4zoEuEDS/HwHJ0l6NOO4vjN/52ZmZr1aRHTaBxhGsYv9zhRJWD0wCRBwKHBr1vsRcFQebwY8AQwENgYGZPn2QF0ejwaWAu/Ldh8C9q7Q/2zgsDwekO19DribYsPaLSn2SduqWpvABsBTwO7ZzqYUX6NQLbbDgKvyeAPgWWAjYDzwnSzfkGJPkA9UeF8B7JXnk4DTMvZngQ9l+dXAyXk8HajN4wA+k8fnl/qbDIwt9fM8sGHz+67yuxufMdbB0Cj+ANwffyp/zMx6k+b/plf6rIupt6cjoiEimoDFwLQMoiETA4CDgDMkzc//8A8AhlJsPHuFpAZgKm+fPpoTEc9lu/NLbQEgaRDw3oi4BSAiVkTE6xTJz3UR0RgRLwIzgN3baPPDwAsRMTfbeTUi3mojtt8C++WIzyeB+yPib/mMR+czzga2oEiwWno2Imbl8bUZ74fzPT6R5VcB+1a49w3g9jyub/lOShYCUyQdRZHIthIREyOiNiJqYUiVZszMzPqWdfGFkytLx02l86ZSfwI+FxFLyjdKOgt4ERhBMcqzokq7jXRO7KvT5imVYouIFZKmA/8IHA40T20JmBARd7YTQ7Rz3pY3MwmFtuM/mCLR+gxwpqSdM/kzMzOzNnTVYu47gQmlNUG7ZvlgitGcJmAcxXRZh0TEMuA5SWOyzQ1zvc9M4PBcMzSEImGY00ZTS4CtJO2e7QyStH47sd0AHAfsA9xResavSuqf7XxI0sAK/Q2VtGceHwE8kDEMk7Rdlo+jGAnrqGXAoOx3PWCbiLgPOD2fY5PVaMvMzKzP6qpE6QcUU1kLJS3Oc4DLgGMkLQCGA6+tZrvjgJMkLQQeBP4BuIVi6mkBcC/wrYj4v2oNRMQbFCNDl2Ycd1NMDbYV213Ax4F78n6AK4FHgUckLQJ+RuURnyXA1yU9BrwL+GlErKBIvKbmVF8TcPlqvIfrgW9Kmkcx3XdttjMPuCQiXlmNtszMzPosrZq5sXeapGHA7RGxUxeH8jZSbRTrus0q8z8bZtabSKov1ui25u9RMjMzM6tiXSzmtg6KiGeAbjWaBFBTA3UeUDIzM/OIkpmZmVk1TpTMzMzMqvDUm7VSXw+tN4ix3s4LtM3MWvOIkpmZmVkVTpTMzMzMqnCiZGZmZlaFEyUzMzOzKpworSOSOrxP3Vr0sX5b5x29z8zMzCrzfzDXgKRbgW0o9oD7cURMzPLlFHu6HUixf9sw4CRgA2A28LWIaJT0U2B3YCPgxoj4XoU+tgX+CxgCvA6cEBGPS5oMrAB2BWZJ2rzF+dUU+8JtDPwOOD4iXpY0HZgP7A1cB1zYya/FzMys1/GI0po5PiJqgFqKTXi3yPKBwOyIGAH8hWJz3b0iYiTQCByZ9c7MPWV2AT4uaZcKfUwEJmQ/p1FsytvsfcCoiDi1wvnVwOkRsQvQAJSTsA0iojYiWiVJksZLqpNUBy+t5uswMzPrnTyitGZOknRYHm8DbE+RGDUCN2X5AUANMFfFlxJtBPwpr31B0niK978VsAOwsLlxSZsAo4CpWvWFRhuW+p8aEY0tzyUNBjaLiBlZfhUwtVTvhmoPlKNiOTJW62/UMTMzw4nSapM0mmJqbc+IeD2ntAbk5RWlBEbAVRHxby3u/wDFCNHuOSU2uXR/s/WAV3IkqpLX2jmvpqP1zMzMDE+9rYnBwMuZJA0HPlal3jRgrKT3AEjaXNL7gU0pEpalkrYEPtnyxoh4FXha0ufzXkka0V5gEbEUeFnSPlk0DpjRxi1mZmbWBo8orb47gBMlPQYsAR6uVCkiHpX0HeAuSesBbwJfj4iHJc0DHgeeBWZV6edI4KfZRn/gemBBB+I7Brhc0sbAU8BxHX80MzMzK1N4gydroVijVNfVYdg7zP8UmFlfJak+/8iqFY8oWSs1NVDnPMnMzMxrlMzMzMyqcaJkZmZmVoUTJTMzM7MqvEbJWqmvh1Xfc2ndnRdhm5mtOx5RMjMzM6vCiZKZmZlZFU6UuoCkrSXd2EltjZG0Q2e0ZWZmZm/nROkdJmn9iHg+IsZ2UpNjKDbVXa0YOqlvMzOzXq1HJkqShkl6XNJkSU9ImiLpQEmzJD0paY+sN1DSJElzJM2TdGjp/pmSHsnPqCwfLWm6pBuz/SlS62XNWefHkuZLWtSB/o6VdJuke4Fp2f+i0rVbJd0t6RlJ/yLp1Lz/YUmbZ71tJd0hqT5jH55xHwJckLFsW6le3j9Z0uWSZgPnr+vfkZmZWW/Qk0cWtgM+DxwPzAWOAPamSBy+TTHSciZwb0QcL2kzYI6ke4A/AZ+IiBWStgeuA5q/unxXYEfgeYp92PYCHqjQ/8YRMVLSvsAkYKc2+gPYDdglIv4qaViLtnbKfgcA/wucHhG7SroIOBq4GJgInBgRT0r6KHBZROwv6Tbg9oi4EUDStJb1gP2zn/cBoyKiseXDSBoPjC/OhlZ+42ZmZn1MT06Uno6IBgBJi4FpERGSGoBhWecg4BBJp+X5AIos4HngJ5JGAo3Ah0rtzomI57Ld+dlWpUTpOoCIuF/SppkYVesP4O6I+GuVZ7kvIpYByyQtBX6V5Q3ALpI2AUYBU0sDXBu2bKQD9aZWSpLyOSZSJGO515uZmZn15ERpZem4qXTexKrnEvC5iFhSvlHSWcCLwAiK6ccVVdptpPo7aplMRBv9fRR4bS2eZT3glYgY2UYbdKBeWzGYmZlZCz1yjdJquBOY0LzOSNKuWT4YeCEimoBxQL81aPvwbHNvYGlELG2jv7USEa8CT0v6fLYrSSPy8jJgUAfqmZmZ2Wrq7YnSD4D+wMKcnvtBll8GHCNpATCcNRtpWSFpHnA58OV2+usMRwJfzpgXA4dm+fXAN3Px97Zt1DMzM7PVpPD+B6tN0nTgtIio6+pY1oVijVKvfLReyf8XNjNbO5LqI6K20rXePqJkZmZmtsZ68mLuLhMRo7s6hnWppgbqPKBkZmbmESUzMzOzapwomZmZmVXhqTdrpb4eWm/cYmvLi67NzHoejyiZmZmZVeFEyczMzKwKJ0pmZmZmVThRWkuStpZ0YwfqffudiMfMzMw6jxOltRQRz0fE2A5U7fRESdL6bZ139D4zMzOrbJ0nSpKOlrRQ0gJJ12TZMEn3Zvk0SUOzfLKkSyQ9KOkpSWNL7ZwuqSHbOTfLTpA0N8tukrSxpMGSfi9pvawzUNKzkvpL2lbSHZLqJc2UNLxCvGdJukbSQ5KelHRClkvSBZIWZRyHl55lUR4fK+nm7ONJSedn+bnARpLmS5qSMf06417U3FaLOCrGmu/ockmzgfMrnI+U9HC+21skvSvvmy7pYkl1wDc66/drZmbWm63TkQVJOwLfAUZFxJ8lbZ6XLgWuioirJB0PXAKMyWtbAXtTbFZ7G3CjpE9SbO760Yh4vdTOzRFxRfZ1DvDliLhU0nz+f3v3H2RVed9x/P1RQSsQcSR1NCZiSBzHsYAgxJBU0dCMbQwwIlFqM5ChOqaRjGPQOqO2iLYdtSbTxqihbSRpjVDQJERbGBVEJAos4i7FiL9wGiNN/IkoSnD59o/zbDyu9+w9C3fv3b37ec3s7D3nPOec7/eeu7vffc5z7wOnA6uAs4EVEbFH0gLg4oh4RtJnyCbHPbNC6COBU4FBwCZJ9wGfBUYDo4BhwAZJD1fYdzRwMrAb2CrpuxFxpaRLImJ0inUa8FJEfCktH1bhOF3Fekx6TtslLey03AbMiYjVkuYDfwtcmvYbWDSXjaSLgIuypU9UamJmZtbv9PQtmDOBJRHxCkBEvJbWfxY4Jz3+d+DG3D4/jYi9wJOSjkzrJgF3RMSuTsc5KRVIQ4HBwIq0fjFwHlmhdD5wq6TBwARgid7/kKCDC+L+WUS8A7wjaRUwnqx4uysi2oHfSFoNjAPaOu37YETsAJD0JHAs8KtObTYDN0u6Abg3ItbkN5aIdUmK4wPLqeAaGhGr0/ofAkty7RYX5EtELCArztKkuGZmZtYbx6rszj2u9rGHC4GpEdEqaRYwMa1fBvx96nkaC6wk6x16o6NXp4rOhUJ3Cod8/O1UeI4j4mlJY4A/A66X9GBEzM81OaBKrG9XWS5Stp2ZmZnR82OUVgLTJR0BkLtl9guynh6AC4A1FfbNux/4mqRDOx1nCLBd0oB0HAAi4i1gA/BPZD027RHxJrBN0vR0DEkaVXC+KZIOSXFPTMdaA5wn6UBJHwVOA9aXeRKSPSlOJB0N7IqI/wBuAsbkG3Yz1vx+O4DXJf1xWvVVYHUXu5iZmVkXerRHKSK2SPo7YLWkdmATMAuYA9wh6XLgZeBrVY6zXNJooEXS74D/InsX2TXAunSMdWSFU4fFZLedJubWXQDcJulqYACwCGitcMo2stt2w4DrIuIlST8hu2XYStbDdEVE/J+k4aWejOy2Vpukx4EfATdJ2gvsAb5eoX3ZWDubCdyeisrnqfLcmpmZWTGFJ6D6AEnzgLci4h8bHUujZGOUWhodRtPxj5qZWe8kaWPRm538OUpmZmZmBXrjYO6Gioh5jY6h0caOhRZ3KJmZmblHyczMzKyICyUzMzOzAr71Zh+ycSOo2idY9VIeMG1mZrXkHiUzMzOzAi6UzMzMzAq4UDIzMzMr4EKpCUjyWDMzM7Me4EKpCknDJT0laaGkpyXdKWmSpLWSnpE0PrUbJOkHktZL2iRpSm7/NZIeT18T0vqJkh6StDQd/07pw0OoJV0oaYOkVkl35+a7WyjpdknrgBsljZC0XNLGdL4TUrsvS1qXYnpA0pF1e/LMzMz6OBdK5XwKuBk4IX39OfB5YC7ZnHMAVwErI2I8cAbZXG6DgN8CfxIRY4DzgH/OHfdk4FLgROCTwOcqnPueiBgXEaOAXwKzc9uOASZExGVkc8nNiYixKa5bU5tHgFMj4mSy+eKuqJSgpIsktUhqyabOMzMzM9+yKWdbRGwGkLQFeDAiQtJmYHhq80VgsqS5afkQ4BPAS8AtaVLfduD43HHXR8SL6bhPpGM90uncJ0m6HhgKDAZW5LYtiYh2SYOBCcCSXKfUwen7McBiSUcBA4FtlRKMiAVkxVaa683MzMxcKJWzO/d4b255L+8/hwKmRcTW/I5pkt3fAKPIevDeLThuO5Wvx0JgakS0SpoFTMxtezt9PwB4IyJGV9j/u8C3I2KZpInAvAptzMzMrALfequdFcCcjnFGkk5O6w8DtkfEXuCrwIHdPO4QYLukAcAFlRpExJvANknT07klaVTu/L9Oj2d289xmZmb9mgul2rkOGAC0pdtz16X1twIzJbWSjW96u2D/ItcA64C1wFNdtLsAmJ3OswWYktbPI7sltxF4pZvnNjMz69cUnvPBOsnGKLU0Oox94pezmZl1l6SNEXFKpW3uUTIzMzMr4ELJPmTs2Kxnpi9+mZmZ1ZILJTMzM7MCLpTMzMzMCrhQMjMzMyvgQsnMzMysgAslMzMzswIulMzMzMwKuFAyMzMzK+BCyczMzKyACyUzMzOzAi6UzMzMzAq4UDIzMzMr4ELJzMzMrIALJTMzM7MCLpTMzMzMCrhQMjMzMyvgQsnMzMysgAslMzMzswIulMzMzMwKuFAyMzMzK+BCyczMzKyACyUzMzOzAi6UzMzMzAq4UDIzMzMr4ELJzMzMrIAiotExWC8jaSewtdFx1Mkw4JVGB1En/SlX6F/5Otfm1Z/ybWSux0bERyttOKjekVifsDUiTml0EPUgqcW5Nqf+lK9zbV79Kd/emqtvvZmZmZkVcKFkZmZmVsCFklWyoNEB1JFzbV79KV/n2rz6U769MlcP5jYzMzMr4B4lMzMzswIulMzMzMwKuFDqxySdJWmrpGclXVlh+8GSFqft6yQNr3+UtVEi19MkPS7pPUnnNiLGWimR62WSnpTUJulBScc2Is5aKZHvxZI2S3pC0iOSTmxEnLVQLddcu2mSQlKve6t1WSWu6yxJL6fr+oSkv2xEnLVQ5rpK+kr6ud0i6cf1jrGWSlzb7+Su69OS3mhEnL8XEf7qh1/AgcBzwCeBgUArcGKnNn8F3J4enw8sbnTcPZjrcGAk8CPg3EbH3MO5ngEcmh5/va9e127k+5Hc48nA8kbH3VO5pnZDgIeBx4BTGh13D17XWcAtjY61Trl+GtgEHJ6W/7DRcfdkvp3azwF+0MiY3aPUf40Hno2I5yPid8AiYEqnNlOAH6bHS4EvSFIdY6yVqrlGxAsR0QbsbUSANVQm11URsSstPgYcU+cYa6lMvm/mFgcBffUdLGV+ZgGuA24A3q1ncDVWNtdmUCbXC4HvRcTrABHx2zrHWEvdvbYzgLvqElkBF0r918eAX+WWX0zrKraJiPeAHcARdYmutsrk2iy6m+ts4L97NKKeVSpfSd+Q9BxwI/DNOsVWa1VzlTQG+HhE3FfPwHpA2dfxtHQLeamkj9cntJork+vxwPGS1kp6TNJZdYuu9kr/jkrDAo4DVtYhrkIulMz6KUl/AZwC3NToWHpaRHwvIkYAfw1c3eh4eoKkA4BvA99qdCx18nNgeESMBO7n/d7vZnQQ2e23iWQ9LP8iaWhDI6qP84GlEdHeyCBcKPVfvwby/4Edk9ZVbCPpIOAw4NW6RFdbZXJtFqVylTQJuAqYHBG76xRbT+jutV0ETO3RiHpOtVyHACcBD0l6ATgVWNZHB3RXva4R8WrutfuvwNg6xVZrZV7DLwLLImJPRGwDniYrnPqi7vzMnk+Db7uBC6X+bAPwaUnHSRpI9oJc1qnNMmBmenwusDLS6Lo+pkyuzaJqrpJOBr5PViT15bEOUC7f/B+ULwHP1DG+Wuoy14jYERHDImJ4RAwnG382OSJaGhPufilzXY/KLU4GflnH+GqpzO+nn5L1JiFpGNmtuOfrGWQNlfp9LOkE4HDg0TrH9yEulPqpNOboEmAF2S+Y/4yILZLmS5qcmv0bcISkZ4HLgMK3I/dmZXKVNE7Si8B04PuStjQu4n1X8rreBAwGlqS33/bZorFkvpekt1Q/QfY6nllwuF6tZK5NoWSu30zXtZVs3NmsxkS7f0rmugJ4VdKTwCrg8ojoi7373Xkdnw8s6g3/nHsKEzMzM7MC7lEyMzMzK+BCyczMzKyACyUzMzOzAi6UzMzMzAq4UDIzMzMr4ELJzJqCpPb0cQf/I+nn1T65WNI8SXOrtJkq6cTc8vz0YZ37G+tCSefu73G6ec5LJR1az3OaNQMXSmbWLN6JiNERcRLwGvCNGhxzKvD7Qiki/iYiHqjBcetK0oHApYALJbNucqFkZs3oUdJEm5JGSFouaaOkNekTfz9A0oWSNkhqlXS3pEMlTSD7xOebUk/ViI6eIElnSVqS23+ipHvT4y9KelTS45KWSBrcVaCSXpD0D+kcLZLGSFoh6TlJF+eO/7Ck+yRtlXR7mtsNSTMkbU49aTfkjvuWpJvTBzJeBRwNrJK0Km2/LZ1vi6RrO8VzbYp/c8fzJWmwpDvSujZJ0/YlX7O+xoWSmTWV1HvyBd6fFmEBMCcixgJzgVsr7HZPRIyLiFFknxY8OyJ+kY5xeeqpei7X/gHgM5IGpeXzgEVpeomrgUkRMQZoIfs08Gr+NyJGA2uAhWRTBp0KXJtrMx6YQ9bDNQI4R9LRwA3AmcBoYJykjrnsBgHrImJURMwHXgLOiIgz0varIuIUYCRwuqSRuXO9kuK/LT1nANcAOyLij9JEtCv3I1+zPuOgRgdgZlYjf5CmKfkYWbFzf+rdmEA2XUtHu4Mr7HuSpOuBoWTTu6zo6kQR8Z6k5cCXJS0lm0PuCuB0skJmbTrfQMrNVdVR1G0GBkfETmCnpN25sVbrI+J5AEl3AZ8H9gAPRcTLaf2dwGlkc4O1A3d3cc6vSLqI7O/AUSnutrTtnvR9I3BOejyJbFqJjufgdUln72O+Zn2GCyUzaxbvRMToNGB5BdkYpYXAG6m3pisLgakR0SppFmkC0ioWkc1Z9RrQEhE7lVUL90fEjG7Gvjt935t73LHc8Xu683xT1eafejci2ittkHQcWU/RuFTwLAQOqRBPO13/ndjXfM36DN96M7OmEhG7yCZJ/RawC9gmaTqAMqMq7DYE2C5pAHBBbv3OtK2S1cAY4EKyogngMeBzkj6VzjdI0vH7mVKH8cpmXD+A7FbfI8B6sttmw9ItxxkprkryuXwEeBvYIelI4E9LnP9+cgPkJR1Oz+Zr1iu4UDKzphMRm8huI80gK3xmp0HNW4ApFXa5BlgHrAWeyq1fBFwuaZOkEZ3O0Q7cS1Zk3JvWvUw2i/1dktrIbkN9aPD4PtoA3EJ2W3Eb8JOI2A5cSTajfCuwMSJ+VrD/AmC5pFUR0QpsIsv1x2R5V3M9cHgaNN5KNt6pJ/M16xUUUa331szMGknSRGBuRJzd6FjM+hv3KJmZmZkVcI+SmZmZWQH3KJmZmZkVcKFkZmZmVsCFkpmZmVkBF0pmZmZmBVwomZmZmRX4f6bwIRz2oKd5AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Softmax + No Loss\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Sigmoid + Loss')\n",
    "plot_importance(cancelout_feature_importance_sigmoid)\n",
    "print('Softmax + No Loss')\n",
    "plot_importance(cancelout_feature_importance_softmax)\n",
    "print(\"Random Forest\")\n",
    "plot_importance(rf_importances)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}