01-DNN-Regression.ipynb

      "Epoch 80/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.7503 - mae: 1.6151 - mse: 4.7503 - val_loss: 9.2102 - val_mae: 2.0407 - val_mse: 9.2102\n",
      "Epoch 81/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.9437 - mae: 1.6686 - mse: 4.9437 - val_loss: 9.1291 - val_mae: 1.9641 - val_mse: 9.1291\n",
      "Epoch 82/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.7192 - mae: 1.6199 - mse: 4.7192 - val_loss: 9.6963 - val_mae: 2.0608 - val_mse: 9.6963\n",
      "Epoch 83/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.6980 - mae: 1.6101 - mse: 4.6980 - val_loss: 9.2010 - val_mae: 2.0322 - val_mse: 9.2010\n",
      "Epoch 84/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.7077 - mae: 1.5825 - mse: 4.7077 - val_loss: 9.1720 - val_mae: 1.9802 - val_mse: 9.1720\n",
      "Epoch 85/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.5014 - mae: 1.5738 - mse: 4.5014 - val_loss: 10.2397 - val_mae: 2.1321 - val_mse: 10.2397\n",
      "Epoch 86/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.6795 - mae: 1.5714 - mse: 4.6795 - val_loss: 9.5360 - val_mae: 2.0136 - val_mse: 9.5360\n",
      "Epoch 87/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.3851 - mae: 1.5927 - mse: 4.3851 - val_loss: 11.9117 - val_mae: 2.4633 - val_mse: 11.9117\n",
      "Epoch 88/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.6168 - mae: 1.5938 - mse: 4.6168 - val_loss: 9.2050 - val_mae: 2.0299 - val_mse: 9.2050\n",
      "Epoch 89/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.4263 - mae: 1.5820 - mse: 4.4263 - val_loss: 8.7735 - val_mae: 1.9424 - val_mse: 8.7735\n",
      "Epoch 90/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.3609 - mae: 1.5683 - mse: 4.3609 - val_loss: 9.5918 - val_mae: 2.0796 - val_mse: 9.5918\n",
      "Epoch 91/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.4027 - mae: 1.5585 - mse: 4.4027 - val_loss: 9.0182 - val_mae: 1.9578 - val_mse: 9.0182\n",
      "Epoch 92/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.1592 - mae: 1.5116 - mse: 4.1592 - val_loss: 8.9067 - val_mae: 2.0204 - val_mse: 8.9067\n",
      "Epoch 93/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.1746 - mae: 1.5005 - mse: 4.1746 - val_loss: 9.7606 - val_mae: 2.1310 - val_mse: 9.7606\n",
      "Epoch 94/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.2409 - mae: 1.5045 - mse: 4.2409 - val_loss: 10.2962 - val_mae: 2.2091 - val_mse: 10.2962\n",
      "Epoch 95/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.2110 - mae: 1.5291 - mse: 4.2110 - val_loss: 10.3401 - val_mae: 2.2458 - val_mse: 10.3401\n",
      "Epoch 96/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.0498 - mae: 1.5090 - mse: 4.0498 - val_loss: 9.4759 - val_mae: 2.0933 - val_mse: 9.4759\n",
      "Epoch 97/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 4.0777 - mae: 1.5070 - mse: 4.0777 - val_loss: 9.0798 - val_mae: 1.9662 - val_mse: 9.0798\n",
      "Epoch 98/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 3.8911 - mae: 1.4699 - mse: 3.8911 - val_loss: 9.3206 - val_mae: 1.9965 - val_mse: 9.3206\n",
      "Epoch 99/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 3.8871 - mae: 1.4670 - mse: 3.8871 - val_loss: 9.4710 - val_mae: 2.0691 - val_mse: 9.4710\n",
      "Epoch 100/100\n",
      "36/36 [==============================] - 0s 3ms/step - loss: 3.7977 - mae: 1.4708 - mse: 3.7977 - val_loss: 9.7039 - val_mae: 2.0753 - val_mse: 9.7039\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(x_train,\n",
    "                    y_train,\n",
    "                    epochs          = 100,\n",
    "                    batch_size      = 10,\n",
    "                    verbose         = 1,\n",
    "                    validation_data = (x_test, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 6 - Evaluate\n",
    "### 6.1 - Model evaluation\n",
    "MAE =  Mean Absolute Error (between the labels and predictions)  \n",
    "A mae equal to 3 represents an average error in prediction of $3k."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_test / loss      : 9.7039\n",
      "x_test / mae       : 2.0753\n",
      "x_test / mse       : 9.7039\n"
     ]
    }
   ],
   "source": [
    "score = model.evaluate(x_test, y_test, verbose=0)\n",
    "\n",
    "print('x_test / loss      : {:5.4f}'.format(score[0]))\n",
    "print('x_test / mae       : {:5.4f}'.format(score[1]))\n",
    "print('x_test / mse       : {:5.4f}'.format(score[2]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6.2 - Training history\n",
    "What was the best result during our training ?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>loss</th>\n",
       "      <th>mae</th>\n",
       "      <th>mse</th>\n",
       "      <th>val_loss</th>\n",
       "      <th>val_mae</th>\n",
       "      <th>val_mse</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>16.860632</td>\n",
       "      <td>2.365933</td>\n",
       "      <td>16.860632</td>\n",
       "      <td>16.798294</td>\n",
       "      <td>2.488858</td>\n",
       "      <td>16.798294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>56.393275</td>\n",
       "      <td>2.309748</td>\n",
       "      <td>56.393275</td>\n",
       "      <td>41.353652</td>\n",
       "      <td>1.842335</td>\n",
       "      <td>41.353652</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>3.797741</td>\n",
       "      <td>1.467033</td>\n",
       "      <td>3.797741</td>\n",
       "      <td>8.773514</td>\n",
       "      <td>1.933441</td>\n",
       "      <td>8.773514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.222468</td>\n",
       "      <td>1.684371</td>\n",
       "      <td>5.222468</td>\n",
       "      <td>9.394964</td>\n",
       "      <td>2.038982</td>\n",
       "      <td>9.394964</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>7.079284</td>\n",
       "      <td>1.910565</td>\n",
       "      <td>7.079284</td>\n",
       "      <td>9.958539</td>\n",
       "      <td>2.131567</td>\n",
       "      <td>9.958539</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>10.276749</td>\n",
       "      <td>2.210366</td>\n",
       "      <td>10.276749</td>\n",
       "      <td>11.525946</td>\n",
       "      <td>2.266691</td>\n",
       "      <td>11.525946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>491.377869</td>\n",
       "      <td>20.047050</td>\n",
       "      <td>491.377869</td>\n",
       "      <td>394.252594</td>\n",
       "      <td>18.079798</td>\n",
       "      <td>394.252594</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             loss         mae         mse    val_loss     val_mae     val_mse\n",
       "count  100.000000  100.000000  100.000000  100.000000  100.000000  100.000000\n",
       "mean    16.860632    2.365933   16.860632   16.798294    2.488858   16.798294\n",
       "std     56.393275    2.309748   56.393275   41.353652    1.842335   41.353652\n",
       "min      3.797741    1.467033    3.797741    8.773514    1.933441    8.773514\n",
       "25%      5.222468    1.684371    5.222468    9.394964    2.038982    9.394964\n",
       "50%      7.079284    1.910565    7.079284    9.958539    2.131567    9.958539\n",
       "75%     10.276749    2.210366   10.276749   11.525946    2.266691   11.525946\n",
       "max    491.377869   20.047050  491.377869  394.252594   18.079798  394.252594"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "df=pd.DataFrame(data=history.history)\n",
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "min( val_mae ) : 1.9334\n"
     ]
    }
   ],
   "source": [
    "print(\"min( val_mae ) : {:.4f}\".format( min(history.history[\"val_mae\"]) ) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAg4AAAGdCAYAAAB6oftJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdeZxkdX3v/9fnVFVXd88KDAPDsIqyCDOAMgrhRmW5osFIgl6DAaNGxFiiQvxpJEZA9KoxQVBCaYDcRBQUUeRCXBi5MoBBWcRBgizDMggzLDPDLL1V13K+vz++p7pP95zuqd6qTvW8n49HPU71Od86/a0zPV3v/m7HnHOIiIiINCJodQVERESkfSg4iIiISMMUHERERKRhCg4iIiLSMAUHERERaZiCg4iIiDRMwUFEREQapuAgIpNmZv9hZi56VMxs8Q7K/1msvDOz9yWU2dXMPmNmd5vZy9F5XzSzB83su2b2ITN7RcLrLhp17vEel03jZRDZqWRbXQERmTWywF8C430o/9V4JzCz1wM3A/EAsg2YAyyPHqcD/xf4szFOEwIbdlDXbTs4LiJjUHAQkenwB2BffDBIDA5mtitwCtALlIFdRx1fyHBoWANcBNzsnOuNju8JvBF4Fz4cjOVZ59z+k38rIjIeBQcRmQ6/AirAUWZ2mHPu4YQypwMdwHeBk8Y4vhgYBE5wzj0XP+icewG4HrjezLqms/Ii0jiNcRCR6fLtaDtWd0R9/zVjHF8WbVePDg2jOecGJlg3EZkmCg4iMl3qweEMMxvxu8XMDgJeDzwLrNrBeZaYmU1/9URkOig4iMi0cM49BfwXsBQ4YdThemvDtc65scYn3B9t9wX+t5nlp7+WIjJVCg4iMp3q3RBD3RVR68GZo44n+S7wWPT8fOAFM/uhmX3KzI43s+4G67CPmb2wg8dhE3lTIjJMwUFEptP3gRJwmpnNifa9EdgPuN8598hYL3TOlfAtFT+Odi0ETgP+EfgFsMXMbjazY3dQhwDYYweP3MTfmoiAgoOITCPn3BbgFvy6C++Idu9oUGT89eudc28DDsW3OvwYeD46nAP+FPgvM/v4OKd5xjlnO3isnsTbExEUHERk+tUDwnuiaZPvwE/V/G6jJ3DOPeqc+7Jz7m3Oub3wQeJzQD9gwFfN7DXTXG8RaYCCg4hMt58BL+G7Hc4B5gM/dc5tnOwJoyBxEfBWwOF/d7136lUVkYlScBCRaeWcqwLfw/9++d/R7m+P/YoJnftO/KqSAAdNxzlFZGIUHERkJtS7K3LAZvy4h+nSF23L03hOEWmQlpwWkWnnnPuNmV0EzAN+55wb3NFrzGwF8Lhzbus4ZQ4Djoi+1ABHkRZQcBCRGeGc+9wEX/IXwFlm9l3gh8A9zrkeADPbDX9zqwvwLaV9wNXTWF0RaZCCg4ikRQVYAPxN9MDMtuF/T8UXf9oCnO6ce3aM8+xjZi/s4Hvd7Zw7bYr1FdkpKTiISFr8PfCfwFuAY4FDgN3x0y83AI8AtwJXOec2jHOe+gJQ49l1B8dFZAzmnGt1HURERKRNaFaFiIiINEzBQURERBqm4CAiIiINU3AQERGRhik4jKNQKLhCoaDRoyIiIhFNx2yMwoOIiOwsbLyDanEQERGRhik4iIiISMMUHERERKRhqQgOZubGePQmlD3YzG4ys81m1mdmd5nZCWOcNzCz88zsUTMrmdmzZnaJmc2Z+XclIiIy+6RpcORdwJWj9lXiX5jZgcDdQBX4CrAV+CBwq5m91Tl326jXXwp8DPgRcAlwaPT1UWZ2knMunExFK5UKzz33HKVSaTIv3+l0dnay9957k8vlWl0VERGZojQFh6ecc9/ZQZkvAQuB1zrnVgOY2TXAw8AVZnaIi26+YWaHAR8FbnTOvaN+AjN7Gvg6cDpw3WQq+txzzzFv3jz2339/zMYdfLrTc86xadMmnnvuOQ444IBWV0dERKYoFV0VdWbWYWZzxzg2B3g7sKoeGgCcc73A1cBBwIrYS96Nn1Jy2ahTXQX0A2dOtp6lUonddttNoaEBZsZuu+2m1hkRkVkiTcHhnfgP9B4ze8nMLjezBbHjy4E88KuE1/462saDwwogBO6NF3TOlYDVo8pOmEJD43StRERmj7QEh3uBi/Dh4b3AL4BzgLtiLRB7Rdt1Ca+v71sa27cXsNE5NzhG+UVm1pFUGTM728zun9A7EBER2QmkIjg4517vnPtn59xNzrlrnHOnA58BlgEfj4p1R9ukIFAaVab+PKnsWOXj9bnSOXd0w29ARERkJ5GK4DCGfwLKwCnR1/3RNp9QtnNUmfrzpLJjlW8ra9eu5ZBDDuGss87i8MMP54wzzuC2227juOOO41WvehX33nsvd9xxB0ceeSRHHnkkRx11FD09PQD80z/9EytWrGD58uVceOGFLX4nIiLSTtI0q2IE51zFzNYDi6Jd66Pt0oTi9X3xboz1wKvNLJ/QXbEU341Rnmo9T/78j6d6ijHd+tlTxj3+xBNPcMMNN3DllVeyYsUKrrvuOn75y19y880388UvfpFarcYVV1zBcccdR29vL52dnaxcuZI1a9Zw77334pzj7W9/O3feeSdveMMbZux9iIjI7JHaFgcz6wT2Bl6Mdj2E73o4NqH4MdE2Pi7hPvz7e13CeY8cVbYtHXDAASxbtowgCDjssMM48cQTMTOWLVvG2rVrOe644/jbv/1bvv71r7Nlyxay2SwrV65k5cqVHHXUUbzmNa/h0UcfZc2aNa1+KyIi0iZaHhzMbLcxDn0e3yJyCwxNu7wFeJOZHRF7/VzgLGANI2dQXI+/q+W5o877QfzYhmuno/6tlM8P98QEQTD0dRAEVKtVPv3pT3P11VczMDDAMcccw6OPPopzjvPPP5/Vq1ezevVqnnjiCT7wgQ+06i2IiEibSUNXxT+Y2THA7cAfgLnAnwDHA/cAl8fKng+cCKw0s0uBbfggsBQ4pb74E4Bz7iEzuwI4x8xuBH7C8MqRdzDJxZ9G21F3Qt2mnhKbevyYzF3n5lk0v2s6vv24nnzySZYtW8ayZcv41a9+xaOPPsrJJ5/MZz/7Wc444wzmzp3LunXryOVyLF68eMbrIyIi7S8NwWEV8Gr8NMzdgBq+9eAzwFejdRcAcM49YWbHAV8GPg10AA8Ab0lYbhp8a8Na4Gz8IMuN+CBywWSXm56sILaWQejGKTiNLrvsMm6//XYymQyvfvWreetb30o+n+eRRx7h2GN9j8/cuXP5zne+o+AgIiINsdgf6TJKoVBwAMViccT+Rx55hEMPPXRC59rSN8hLWwcAWNDdwR4LE2eCzlqTuWYiItIS467a1/IxDjuL+OqJimoiItKuFByaJL7qslp5RESkXSk4NElA88c4iIiITDcFhyZRi4OIiMwGCg5NMmKMg3KDiIi0KQWHJlGLg4iIzAYKDk0SqMVBRERmAQWHJom3OIRNnpA5d+7cpn4/ERGZvRQcmsRQi4OIiLS/NCw53d7OektDxXLAQRM999U/S9z9d3/3d+y3334UCgUALrroIsyMO++8k82bN1OpVPjCF77AqaeeusNvsWrVKi688EL22GMPVq9ezWmnncayZcv42te+xsDAADfddBMHHnggN9xwA5/73OfIZDIsWLCAO++8k1qtxqc//WlWrVrF4OAgH/nIR/jQhz400XcpIiJtRC0Obej000/n+uuvH/r6+9//Pu9///v50Y9+xAMPPMDtt9/OJz7xiYYHYT744IN87Wtf46GHHuLb3/42jz/+OPfeey9nnXUWl1/u7zF28cUXc+utt/Lggw9y8803A/Bv//ZvLFiwgPvuu4/77ruPq666iqeffnr637CIiKSGWhza0FFHHcVLL73E+vXr2bBhA7vssgtLlizhvPPO48477yQIAtatW8eLL77InnvuucPzrVixgiVLlgBw4IEH8uY3vxmAZcuWcfvttwNw3HHH8b73vY93vetdnHbaaQCsXLmS3/3ud/zgBz8AYOvWraxZs4YDDjhgJt62iIikgILDVI3RnZBkzfNbh1oBXrlkwYiZFhP1zne+kx/84Ae88MILnH766Vx77bVs2LCB3/zmN+RyOfbff39KpdKOTwTk8/mh50EQDH0dBAHVahWAb37zm9xzzz38+Mc/5sgjj2T16tU457j88ss5+eSTJ/0+RESkvairooniMWGqazmcfvrpfO973+MHP/gB73znO9m6dSuLFy8ml8tx++2388wzz0ytsqM8+eSTvP71r+fiiy9m0aJFPPvss5x88sl84xvfoFKpAPD444/T19c3rd9XRETSRS0OTWRmQ1Mqpjqz4rDDDqOnp4elS5eyZMkSzjjjDP70T/+Uo48+miOPPJJDDjlkGmo87JOf/CRr1qzBOceJJ57IEUccwfLly1m7di2vec1rcM6x++67c9NNN03r9xURkXQxrWI4tkKh4ACKxeKI/Y888giHHnrohM/31IvbqNZCAA5YPI9cNjMNtWwPk71mIiLSdOP2o6urooniYxp0h0wREWlH6qpoolber+Khhx7iPe95z4h9+Xyee+65p6n1EBGR9qbg0EQj7pDZ5O+9bNkyVq9e3eTvKiIis426KiZpMi0G0zmrop3sTO9VRGS2U3CYhM7OTjZt2jThD8SdcYyDc45NmzbR2dnZ6qqIiMg0UFfFJOy9994899xzbNiwYUKv29I3yGDFz6ro29hBZ27nmFXR2dnJ3nvv3epqiIjINFBwmIRcLjepZZW/dONvWfXw8wB86tQjOPFQfZiKiEh7UVdFE+Vzw5e7HK3nICIi0k4UHJqoI7bgU7lSa2FNREREJkfBoYk6srEWh6paHEREpP0oODRRvMVhUMFBRETakIJDE41scVBXhYiItB8FhyYaMcZBLQ4iItKGFByaaMSsCrU4iIhIG1JwaKKRsyrU4iAiIu1HwaGJNMZBRETanYJDE2lWhYiItDsFhyaKtzhU1OIgIiJtSMGhifI5tTiIiEh7U3BoIo1xEBGRdqfg0ESaVSEiIu1OwaGJRrQ41NTiICIi7UfBoYnU4iAiIu1OwaGJNMZBRETanYJDE2lWhYiItDsFhyYa0eJQqeGca2FtREREJk7BoYkyQUAmMAAcUA0VHEREpL0oODTZ6FYHERGRdqLg0GQjZlZonIOIiLQZBYcmi7c4DGpmhYiItBkFhybLq8VBRETaWOqCg5l1m9nTZubM7F8Sjh9sZjeZ2WYz6zOzu8zshDHOFZjZeWb2qJmVzOxZM7vEzObM/DtJltMYBxERaWOpCw7AxcCipANmdiBwN3As8BXgk8Bc4FYzOynhJZcCXwV+D3wUuAH4GHCLmbXkvcfXcijX1OIgIiLtJdvqCsSZ2WuAc4FPAZckFPkSsBB4rXNudfSaa4CHgSvM7BAXLY5gZofhw8KNzrl3xL7H08DXgdOB62bw7STSrAoREWlnqWlxMLMMcBXwM+DGhONzgLcDq+qhAcA51wtcDRwErIi95N2AAZeNOtVVQD9w5nTWv1GaVSEiIu0sNcEBOA84BDhnjOPLgTzwq4Rjv4628eCwAgiBe+MFnXMlYPWosk2jWRUiItLOUhEczOwA4HPAxc65tWMU2yvarks4Vt+3dFT5jc65wTHKLzKzjjHqc7aZ3b/Dik9CvMWhohYHERFpM6kIDsA3gKfxAxnH0h1tk4JAaVSZ+vOksmOVH+Kcu9I5d/Q4dZm0joxh0T0q1OIgIiLtpuWDI83sTODNwBucc5VxivZH23zCsc5RZerPF49xrqTyM+uma2DlDzmvXGbBwmO5YcFrNcZBRETaTkuDg5nl8a0MPwFeMLNXRofqXQ4Lon0bgfWjjsXV98W7MdYDrzazfEJ3xVJ8N0Z5qu+hcQ7KgwRAzvmWBs2qEBGRdtPqroouYHfgFGBN7LEqOn5m9PVZwEP4rodjE85zTLSNj0u4D//+XhcvaGadwJGjys68bG7oaUc9OKjFQURE2kyruyr6gP+VsH93oIifmvlvwO+cc71mdgtwmpkd4Zx7EMDM5uKDxRpGzqC4Hvh7/LoQd8X2fxA/tuHaaX4v48sNj8PM4YODxjiIiEi7aWlwiMY0/GD0fjPbP3r6pHMufvx84ERgpZldCmzDB4GlwCn1xZ+icz9kZlcA55jZjfjukEPxK0feQbMXfxrR4lAFNKtCRETaT6tbHCbEOfeEmR0HfBn4NNABPAC8xTl3W8JLzgXWAmfju0M2ApcDFzjnmvupHW9xcGpxEBGR9pTK4BCt5WBjHHsEOLXB89TwS1cnLV/dXLEWh5zGOIiISJtq9eDInUesxaFDsypERKRNKTg0S04tDiIi0v4UHJolsatCLQ4iItJeFByaZURXhZ9VoRYHERFpNwoOzZLQ4qBZFSIi0m4UHJolYXCk1nEQEZF2o+DQLGpxEBGRWUDBoVmSpmOqxUFERNqMgkOzxKdjonUcRESkPSk4NEt2+1kVg2pxEBGRNqPg0CwJC0BVqjVi9+USERFJPQWHZkkYHBk6qIYKDiIi0j4UHJolk4HAX+4MjiC6OadWjxQRkXai4NBMSatHVjTOQURE2oeCQzPpfhUiItLmFByaKWEtB82sEBGRdqLg0EyxFoeO2MwKERGRdqHg0EyxFoecWhxERKQNKTg004gWh/qttdXiICIi7UPBoZkSl51Wi4OIiLQPBYdmShjjoBYHERFpJwoOzZQwxkF3yBQRkXai4NBMI+5XoTEOIiLSfhQcmimrdRxERKS9KTg0U8IdMtXiICIi7UTBoZmSxjhoVoWIiLQRBYdm0qwKERFpcwoOzZTYVaEWBxERaR8KDs2klSNFRKTNKTg0k+5VISIibU7BoZniLQ7U746p4CAiIu1DwaGZklocKuqqEBGR9qHg0ExJYxxqanEQEZH2oeDQTInrOKjFQURE2oeCQzMlruOgFgcREWkfCg7NlHh3TLU4iIhI+1BwaKaEBaAGFRxERKSNKDg0U1YrR4qISHtTcGimXMKsCg2OFBGRNqLg0EyJYxzU4iAiIu1DwaGZxrg7pnOuVTUSERGZEAWHZoq3OERLTocOaqGCg4iItAcFh2aKtTjk3fDYBs2sEBGRdqHg0EwJLQ6gG12JiEj7UHBopvh0zDDW4qCZFSIi0iYUHJppxKyK6tBzzawQEZF20fLgYGYHm9m1ZvaImW01s34ze9TMvmpmS8Yof5OZbTazPjO7y8xOGOPcgZmdF52vZGbPmtklZjZn5t9ZgkwGzF/yDI7A+cCgZadFRKRdZFtdAWBvYAnwI+A5oAosA84GTjezI51zLwGY2YHA3VGZrwBbgQ8Ct5rZW51zt40696XAx6JzXwIcGn19lJmd5Jxr/p/6uRyUBwG/CFTJOtTiICIibaPlwcE59/+A/zd6v5ndCXwfeB8+JAB8CVgIvNY5tzoqdw3wMHCFmR3iokURzOww4KPAjc65d8TO+zTwdeB04LoZeltjyw4Hh5yrUUKzKkREpH20vKtiHM9E210Aou6FtwOr6qEBwDnXC1wNHASsiL3+3YABl40671VAP3DmzFR7B2LjHOqLQGlWhYiItIvUBAcz6zSzRWa2t5m9GfjX6NBPou1yIA/8KuHlv4628eCwAgiBe+MFnXMlYPWoss2TdIdMzaoQEZE2kZrgAJwFbACeBW7Fd0mc6Zy7Kzq+V7Rdl/Da+r6lsX17ARudc4NjlF9kZh0Jx2ZWdvsWB41xEBGRdpGm4HAT8D+BPwcuBrYAu8eOd0fbpCBQGlWm/jyp7Fjlh5jZ2WZ2fwN1nrh4iwPD96sQERFpB6kJDs6555xztznnbnLOXQi8F/hHMzs/KtIfbfMJL+8cVab+PKnsWOXjdbnSOXd047WfgBFjHKJba6vFQURE2kRqgsNozrnfAb8FCtGu9dF2aULx+r54N8Z6fHdEUnhYiu/GKE9HXSckmzDGQS0OIiLSJlIbHCJdwK7R84fwXQ/HJpQ7JtrGuxfuw7+/18ULmlkncOSoss2T2/7W2ppVISIi7aLlwcHM9hxj//HA4UQzJqJpl7cAbzKzI2Ll5uIHVq5h5AyK6wEHnDvq1B/Ej224dprewsQktThoVoWIiLSJli8ABXwjWlr6F/i1GzqB1+IXaOoBPhErez5wIrDSzC4FtuGDwFLglPriTwDOuYfM7ArgHDO7ET+ts75y5B20YvEnSB7jUFOLg4iItIc0BIfv4gdCvgc/i8LhA8S/Av/knPtDvaBz7gkzOw74MvBpoAN4AHhLwnLT4Fsb1uKXrz4F2AhcDlzQkuWmIbHFoawWBxERaRMtDw7Oue/jl5ZutPwjwKkNlq3h71FxyeRqNwNG3CGzPjhSLQ4iItIeWj7GYaejFgcREWljCg7NlnCvCrU4iIhIu1BwaLakFget4yAiIm1CwaHZRqzj4GdVaDqmiIi0CwWHZovd5Gr4XhXqqhARkfag4NBsCStHqsVBRETahYJDsyVMx9QYBxERaRcNBYdCofBXhUJh+ah9HYVCYf4Y5d9YKBQumI4KzjrZ7cc4qKtCRETaRaMtDv8B/NmofecDm8co/ybgwslVaZZLWgBKXRUiItIm1FXRbAljHNTiICIi7ULBodkSWhwqtZBw+P5cIiIiqaXg0GyxMQ55hrso1OogIiLtQMGh2eJdFQyHBd2vQkRE2oGCQ7ON0eIwqCmZIiLSBiZyW+2FhUJh3/jXAIVCYR/ARpedasVmrYQxDgDliroqREQk/SYSHD4ePUZbOz1V2Ulkt59VAWpxEBGR9tBocPgDoGH/02HEbbWrQ8+1eqSIiLSDhoJDsVjcf4brsfOItThk4y0O6qoQEZE2oMGRzRYf4xCqxUFERNqLgkOzjdnioOAgIiLp11BXRaFQ6AKWABuLxeK2Ucf2Ay4FTsDPrrgD+P+KxeLj01zX2SGTAQvAhQTRI7RAC0CJiEhbaLTF4RxgDfDq+M5CoTAPHxROBeYD84C3AasKhcJu01jP2cNsxCJQQze6UleFiIi0gUaDwx8DzxaLxV+P2v9hYF/gV8ArgT2Ay4E9SZ66KZB8a211VYiISBtodDrmq4H7E/afhp+m+dfFYvGpaN/HC4XCKcBbgQumXsVZKLHFQV0VIiKSfo22OOwOPB3fUSgUcsBRwGMJ4xl+gW+BkCQj1nKIbq2tFgcREWkDjQaHPJAZte8wIAfcm1D+JaB7CvWa3WJdFTnU4iAiIu2j0eDwAnD4qH1/hO+mSOrCmAe8PIV6zW5JLQ4aHCkiIm2g0eDwX8AJhULhTTA0PfOD0bGfJ5Q/HFg35drNVtmEMQ7qqhARkTbQaHC4NNquLBQKD+DHOywHVhWLxcfiBQuFwnzgOGD0DAypS7hfhdZxEBGRdtBQcCgWi/cD7wMGgCOBxfguivcmFH8v0AGsnJ4qzkIJLQ7qqhARkXbQ8G21i8XidwqFwg/x3RCbYtMvR7sFuBN4ZBrqNzvltr+1tgZHiohIO2g4OAAUi8UB4L4dlFk7lQrtFJJaHDTGQURE2oBuctUKCWMctOS0iIi0g0ZvcvVXkzl5sVi8ZjKvm/USWxzUVSEiIunXaFfFf+DXbGiUReUVHJLEWhx0kysREWknExnjUAX+E/j9DNVl55HdfnCkpmOKiEg7aDQ43AG8Afgz/FTMq4DvF4vF0kxVbFZLanHQ4EgREWkDja7jcDxwMPDP+JtX/TvwfKFQuLxQKCyfwfrNTrnt71WhdRxERKQdTGQdhyeAvysUCp8BTsUvOf1hoFAoFH4D/CvwvWKx2DcjNZ1NsgmzKjQ4UkRE2sCEp2MWi8VqsVj8YbFYfAtwIPBFYAlwJbC+UCgcO811nH0SFoAKnaNaU3gQEZF0m9I6DsVi8ZlisfhZ4Gz8Ta3mArtPR8VmtVhw6LThsKCZFSIiknYTWjkyrlAo7AX8dfTYDygB3wEemJ6qzWLZ5OBQroTMybeiQiIiIo2ZUHAoFAoB8DbgLOAt0esfAj4OfLtYLG6d9hrORrFZFXmGWxnU4iAiImnX6MqRBwAfAN6PH8/QB3wLuKpYLN47c9WbpWItDnniLQ4KDiIikm6Ntjg8EW3vBy4EvqvZE1MQv1fFiBYHDY4UEZF0azQ4GFDBtzZcAFxQKBR29BpXLBb3m0LdZq+ElSNBazmIiEj6TWSMQw7Ye6YqslMZcXfMWIuD1nIQEZGUayg4FIvFGbv9tpkdBJwJvBm/LkQn8CRwA3CZc65vVPmDgX8E3gh04GdxXOic+0XCuQP8wM0PAfsDG4DvAxeMPm9TjWhxqA49V4uDiIik3YwFggn4a+A8fFi4GPgk8BjwBeBuM+uqFzSzA4G7gWOBr0Rl5wK3mtlJCee+FPgq/sZcH8WHkY8Bt0ShojVi6zhkR3RVqMVBRETSbdLrOEyjHwBfcs7Fp3J+08zWAJ/Bz+b4l2j/l4CFwGudc6sBzOwa4GHgCjM7xDnnov2H4cPCjc65d9RPbGZPA18HTgeum9F3NpbYktO5cLjFQTe6EhGRtGt5i4Nz7v5RoaHu+mh7OICZzQHeDqyqh4bo9b3A1cBBwIrY69+NH9R52ajzXgX047tHWmPMFgcFBxERSbeWB4dx1AdivhhtlwN54FcJZX8dbePBYQUQAiPWmXDOlYDVo8o2Vzw4xFsc1FUhIiIpl8rgYGYZ/LTPKsPdCXtF23UJL6nvWxrbtxew0Tk3OEb5RWbWkXBs5sVmVWRiwUELQImISNqlMjjguxeOwc9+eCza1x1tk4JAaVSZ+vOksmOVH2JmZ5vZ/Y1Xd4JisyoytXiLg4KDiIikW+qCg5l9HjgHuNI596XYof5om3QbqM5RZerPx7plVFL5Ic65K51zRzdW40nIZMEMgMCFBM53UWhWhYiIpF2qgoOZXQT8A/DvwN+MOrw+2i5le/V98W6M9fjuiKTwsBTfjVGefG2nwGxEq0MuGiCpWRUiIpJ2qQkOZnYh/j4Y1wBn1adVxjyE73o4NuHlx0TbePfCffj397pR36cTOHJU2eZLWD1SsypERCTtUhEczOwC4CLg28D7nXPbtdlH0y5vAd5kZkfEXjsXf5vvNYycQXE94IBzR53qg/ixDddO41uYuBEtDn6cg5acFhGRtGv5AlBm9hHgc2Sf9jEAACAASURBVMAfgNuAv7So/z/yonPu59Hz84ETgZVmdimwDR8ElgKnxFspnHMPmdkVwDlmdiPwE+BQ/MqRd9CqxZ/qctt3VajFQURE0q7lwYHh9RT2Bb6VcPwO4OcAzrknzOw44MvApxm+V8VbnHO3Jbz2XGAtcDZwCrARuBw/W6O1f97H71cR3Vpb6ziIiEjatTw4OOfeB7xvAuUfAU5tsGwNuCR6pEtsjMNQi4MGR4qISMqlYozDTmnEHTLrLQ4KDiIikm4KDq2S2OKgrgoREUk3BYdWGdHiEM2qUIuDiIiknIJDqyS1OCg4iIhIyik4tEouYYyDuipERCTlFBxaJWHJabU4iIhI2ik4tEpu+zEO5WrI9itti4iIpIeCQ6vExjjkbbiLolJTd4WIiKSXgkOrxLoqumLBQeMcREQkzRQcWiXW4tBpw90TGucgIiJppuDQKiNaHIbDwqCWnRYRkRRTcGiVMcY4lHWjKxERSTEFh1aJd1UQa3FQV4WIiKSYgkOrxIJDF9Wh57pDpoiIpJmCQ6t0dQ8/DctDzwfVVSEiIimm4NAqXXOGnnbXhoODWhxERCTNFBxapTPW4lAbHHquMQ4iIpJmCg6t0j3c4tAZCw6aVSEiImmm4NAqsRaHfLU09FzrOIiISJopOLRKbIxDR1VdFSIi0h4UHFqls3PoaUd1kMD5Loqy7lUhIiIppuDQKkFm5ABJVwHU4iAiIumm4NBKsbUc5oS+u0KDI0VEJM0UHFopFhy6o0WgNDhSRETSTMGhlTrjLQ4+OOi22iIikmYKDq0UXz3S1YODuipERCS9FBxaqSuhxUFdFSIikmIKDq0Ub3Goj3FQi4OIiKSYgkMrdSbNqlCLg4iIpJeCQyt1J7Q4qKtCRERSTMGhleItDhocKSIibUDBoZUSxzioxUFERNJLwaGVklaO1L0qREQkxRQcWmnEypG6V4WIiKSfgkMrdQ53Vcxx9RYHBQcREUkvBYdW6t7+XhXV0FELXatqJCIiMi4Fh1ZKuFcFaC0HERFJLwWHVkq4VwVoLQcREUkvBYdW6shD4P8JOlyNnPOBQWs5iIhIWik4tJJZ8loOanEQEZGUUnBotaS1HDTGQUREUkrBodV0h0wREWkjCg6tlnS/CnVViIhISik4tJruVyEiIm1EwaHVEsc4qKtCRETSScGh1RJaHNRVISIiaaXg0Gpd249x0OBIERFJq5YHBzM738xuMLOnzMyZ2dodlD/YzG4ys81m1mdmd5nZCWOUDczsPDN71MxKZvasmV1iZnOSyrdE1/b3q9B0TBERSauWBwfgi8AJwJPA5vEKmtmBwN3AscBXgE8Cc4FbzeykhJdcCnwV+D3wUeAG4GPALWaWhvc+oqtiztACUGpxEBGRdMq2ugLAgc65pwDM7L/xQWAsXwIWAq91zq2OXnMN8DBwhZkd4pxz0f7D8GHhRufcO+onMLOnga8DpwPXzcD7mZhOtTiIiEj7aPlf3fXQsCNR98LbgVX10BC9vhe4GjgIWBF7ybsBAy4bdaqrgH7gzClUe/qMuNGVn1WhJadFRCStWh4cJmA5kAd+lXDs19E2HhxWACFwb7ygc64ErB5VtnW6tr+1tqZjiohIWrVTcNgr2q5LOFbft3RU+Y3ORX/Gb19+kZl1JH0jMzvbzO6fdE0nQgtAiYhIG2mn4FD/0zwpCJRGlak/Tyo7VvkhzrkrnXNHT7iGk5E0q0JdFSIiklLtFBz6o20+4VjnqDL150llxyrfGvFZFVrHQUREUq6dgsP6aLs04Vh9X7wbYz2+OyIpPCzFd2OUp7F+kzN6VoVzbOkbq6FERESktdopODyE73o4NuHYMdE2Pi7hPvz7e128oJl1AkeOKts62Sx0+GyTwdHpKmzcVtrBi0RERFqjbYJDNO3yFuBNZnZEfb+ZzQXOAtYwcgbF9YADzh11qg/ixzZcO6MVnohRrQ4be0rUQtfCComIiCRr+QJQZvYeYL/oy92BDjP7h+jrZ5xz344VPx84EVhpZpcC2/BBYClwSn3xJwDn3ENmdgVwjpndCPwEOBS/cuQdpGHxp7qubtjmF82c48q8HDo29w6yaH7nDl4oIiLSXC0PDsAHgDeO2vf5aHsHMBQcnHNPmNlxwJeBTwMdwAPAW5xztyWc+1xgLXA2cAqwEbgcuMA5l54RiAkzKzZsG1BwEBGR1Gl5cHDOvWmC5R8BTm2wbA24JHqkV8L9KjZsK3Foq+ojIiIyhrYZ4zCrJdyvYsO2gVbVRkREZEwKDmkwosXBT8XcoJkVIiKSQgoOaRAf4xAtLbFhq1ocREQkfRQc0iBhjMNL6qoQEZEUUnBIg4RZFVoESkRE0kjBIQ1iLQ5zo+CwuXeQSi09M0ZFRERAwSEdYrMqFmaqgF/ycpNaHUREJGUUHNKge7jFYYFVh55rSqaIiKSNgkMaxFoc5lIZeq4pmSIikjYKDmkQGxxZX8cB4CVNyRQRkZRRcEiD2ODIztpwcFBXhYiIpI2CQxrEWhw6KsPdE+qqEBGRtFFwSIN8F5gBkKkMEkQ37lRwEBGRtFFwSIMggM6uoS91oysREUkrBYe0iM2smBfNrOgZqFAqV8d6hYiISNMpOKRFbIDk0uEMwUvqrhARkRRRcEiL2ADJJZ029FzdFSIikiYKDmkRa3HYo2P4HhW62ZWIiKSJgkNaxMY47J4bDg4btAiUiIikiIJDWsTuV7Frtjb0XFMyRUQkTRQc0iLW4rAg0I2uREQknRQc0iI2OHK+bnQlIiIppeCQFrHBkXNceej5S1sHcM61okYiIiLbUXBIi1iLQ65cIp/LAFCq1OgtaREoERFJBwWHtOieN/TU1j/D7vM7h77WOAcREUkLBYe0eNVhkM355394giN5eeiQgoOIiKSFgkNazJ0PK94w9OWbNj4w9FwDJEVEJC0UHNLkTacMPT103YPMrfnAoEWgREQkLRQc0uQVh8I+rwAgW6twUt+jgFocREQkPRQc0sQM3vS2oS/f1vMQOMea57dSCzUlU0REWk/BIW1ef/zQKpL7VLdwROk5/rCxl5vufbrFFRMREVFwSJ/OLjj2xKEv39b73wB86/bHeH5zf6tqJSIiAig4pFNskORxA0+xa7WPwWrI1378kFaRFBGRllJwSKOl+8OrDgcg40Le0fNbAH779EZ+/rvnWlgxERHZ2Sk4pFWs1eGd237Lh16+i8CF/OvKR9jcO9jCiomIyM5MwSGtXvvH8MpXD315Ws9qLtzwY2r9fXz2e/fx9IvbWlg5ERHZWSk4pFU2C+d9EV5z3NCuYwbWcskLP2Twmaf5yNW/5P/84lEGK7UWVlJERHY2Cg5plu+Ev/kMvOVdQ7sOrGzkquev5Uvrf8gLK39K4Ru3s+rh9VRqYQsrKiIiO4tsqysgOxAE8M6/hj2Xwre/DjXfwnDE4DqOGFzHlpfv5LdP7MN35+3F7suX89rjj2Xx4l1aXGkREZmtFBzaxf84GfbaH356Pe7BX2Ohb2FYGA5wfP/j0P84/HwVtZ9fzjNz9uTFJQdTPWgZ85cfxYH7LqarQ//UIiIydaZ1AcZWKBQcQLFYbHVVRtq8EX55K+EdPyHYsmncohUCSkGODkKyhAQupDZ3Iey5D5ml+2B77gP7HggHHAy5jia9ARERSTEb76D+DG1HuyyCPz2D4E9Oh2fWUHvqMTb+7kFYu4bd+zeMGLiSIyQXjpy+me3ZDD2bYc3vhvZVgwybdtufbfseQnbxHsynwpxwkHx5AMt3wasO82tLdM9p0psUEZE0UnBoZ5kMvOIQMq84hD1OOhWAnk2b2Xj/vdR+/1sW/uERFvW82NCpsmGNPTY8yR4bnkwu8DMIMV6cvxcvL9qXbFcX+a5OOrs66ezKY7kOPxMkmyMIjK5yH9nerdCzBUol2Ht/OGiZn2Ia3YtjTGHob/hl44be1nAOHvkt3HUrLF4CJ/05zFvQ6lqJiDSNuirGkdquiokY6KO3f5DHX+zl9+u38tjz26hueIk5m59nj4FN7Fd5mUMHn2ef6pamVKdmAc/PW0Jf1wJqHV24zi6sI8+cwR7m9m+mu2cTnb2bCXN5qrssJtx1MbZoD7KdnWRqFaxagfIgVMpQrfhHpeJDRkfez0Tp6IR582HvA2DfV8Ke+/iQFReGUKv6R7Xqv+7s8t01YwWWJ38PP/oWPPrg8L7Objj5HfA//3xkIKpW/XlGf1+RqXIOtmyCXB7mzE1nwE4b5+APT8C6Z2CfV/iHjGfcHyoFh3HMiuAwjr7BCpu2ldiwrcSW518g8+TDzHvucdxAP5vDHBtrGbaGORbVelleWseB5Q1kaL+fl0qQZducXcnWquRqZTqqJbK1amLZMMhQy3cTdnUTds7xj65uMqV+up98aMzv4ebOh4OWYVs2waaXYNtmyGRhyb6wzwE+xCzaEywYbk2pVWGgDwb6ob/Xh6B8lw8wnV3+eSbjX5PJQJAZtY06pZwDF0Lo/HOirXNQGoi+Rx/09/nvGdb8sTD0QWneApi30G+DDPT3QF+vr1NYgznzYe58H8ZyHbBtC2zd7N/j4ADstgfsuTfssbc/R60GvVuhZ6s/Rzbrw1y+04e7Ws2Hv3IJymXo6oaFixr7EAxD/156t/nz92715woy/npnAv88zgJ/Pbvm+K62XAeU+qGvxz8G+v1rO/LQ0eHrOmeefy8zOe4nrMEL6+CZNT4I7LYYFi+FPfbydY2rlOGx38GD98Dv7vE/YwDZHCzcDRbu6j8MDzsaDjnCv98xv2/oz+ecf38TDbfO+eu/bXMUvGv+vdR/ruLmzIMFu/ifoWACs//DMPr5C/05xqtjpQwvroPnn/UtnPMX+p+nXXbzf1Tcdwfcczu8EFuu/4CD4Y1/Aive6H8ux1Kt+HqP/plK4lz0f2MT9GzzP5u92/zPlwuH/09mMrDrYv/7YNEe/t+ukfPH1f8N639EVcr+52C89zIxCg6TNduDQyMGylW29pXZOlCm7+XNBE/+HrfheUr9g5QGSpRLg9QGy2RclUxYI+tquDBkY5hnc6aTLUE3NQs4dPB5lpfW8YrK+IM520EN4xdzDuag8ovsV9nc6uqkS0fe/zKbjFyH/+VXDxf1D6NKZfiXZLUyvfXdka5uH6o6uyGf96GiI+8/CAYHoTIY/fKujGwBqwe0+vsIMj581QNJtQLPPgWDpeTvO3e+f039w2ZwwF+DRmSyfkzS/F38B1f9A6w0kHwNMxnIdkShKT8c8HL56EMz+uCsVf3A7M0bJ/5vHAT+Onbkh89XP7cFw88HS/7Dt2+b/3Ac+neY469dVzdg0cea+RC58UX/7zEZXd3+3kDd83yo7Or2gXTTS8N/AFgAc+f5+s9f6ENZEAvw/b2+DptenNzPfibj31/90dkV/RzFgkG16vfVW0iT/vA59wtw+NGTuw7b0+BImbyujixdHVn23KUb9loIhx/Q0OtC5+gZqLClb5Bt/WVqoWNz6Higv4fO59dS6+2hNtBP2N9PWBqgN9vFpvwCNnYsYEMwl2qpn66tG5nT9zIL+jfjqlX6XYZBy1K2DGXLUrGMfxBgQN5V6XQVOsMKi2s9HFjeyCvLG9i91ptYxzIBVctQs4AQoyss08HYv4BC4PY5B/OdBa9jfW4hgQs5oe8x3rPlHvas9WxXdqdcXW2yoQH8L8gNz09fXabDQL9/TFUY+laFHcyCGtK7gyXl812+daaUULdadWR32o7UalAb8OFkpoQhbH158q+vt5pNRb7TtzSseXj4g3egH574/fivc6EPEz1bYf0zU6tDklotCnhTvI1Ao8FyGszq4GBmAfBx4EPA/sAG4PvABc65Kf4UyngCMxZ0d7Cge3RT7+6wbHL9i845ytWQwUqNSi0kdC5qcXdUauHQo1wNKVdrlCshD1druJ4tBFtfppLJU850UM7lKVuWav11Vf+6/sEq5YEBXF8vNtBPvloiXx2gq1IiW6vweNcSnsnuwmClRr5cJXQBq+a/mrvmHcRr+5+hq1bmxew8XsrOY1NmDl1hhf0rm3hFeSOvqGxkQW0AAwyH4Qeb9gUd9AV5+qyDsmXpcmW6wgpdrkKnq5BxIRkcgatPp3VkCIf2+44JI4z+CgujPxTC6LuUgiz9lqc3yNMXdFC2jC9vFoWlCgvCARbUBlgYDhDg6Any9ASd9AZ5QgLmhwPMr5VYEA7Q4WpsyXTxcmYOmzPdlMmwpLqVvatbWFrZQrerUMPoCTrZmumiJ8iTcaF/P2GVvKtQtQwly0YhMMu8sMRutT66XWOtCf2WY2umi61BF1sznZQsRwZHxoUEhGRGtaJmCOkOy/7hynSGFfqD/ND77A86CAjJhzU6ovC5ICyxsNY/411zW3NzeHbOXmzu2oVdBrey+8AmdittJuu2X0p+U9euPL7oYJ7c81CeX/QKyGbpqJWZV9rGwoHN7LvxSfZ/6TEWbd1x+KpmcgBkalVsEu+x2tFJee5CXLYDy0TdREGAM4saSRwuDMkO9JLr20pucOLhK+yaAxZgA73YOC3jzgx22wNbso+fcbZtC27zRh/SBku4g5cTvP54OOL1Pjz0bIG7b8Pd8VPspXXjV8KCibVmdM2BXXf3rUpz5/tt15yomybqoqyUfQvFxhf8tndr4+ePy3X4R0c+6nJq3sf5rO6qMLOvAR8DfgT8FDgU+ChwF3CSc+P/RKirQiaiFjoq1ZoPN9UaYTjy/1Y1dJQrNQarNUqVGoMVX7YSBZ1KLaQWuhGPMHQ45wgd1MJwxLYahaeh8tHrq7VwKBT1D1bpHajQU6rQW/IfzB3ZgFwmIJcNMIxqGPrX1KLv6Ya/54Q4R7crU7IcoU28vaU7LLNbrZeMC6lZQI2AGkYlamEqR61NzRoMaM4xNxxkYdhPZ1ih0/ng0xlWqZkxaDkGYwGoakFU1ww1AqoWEBJQMyPrQuaHJeaHJRbUfEB7KrcbL2fnbvd9AxeyIBzAHITmP9ZDAnqCfEPvfddqL8sH15NxIVsznWwLutgaBaRBy1GxjP+wBYiCaIerkXdV8tH77HQVOlyNoB5ccYQYmzJz2JCdS3+Qn9C1zLkqC2sD5FyNTBSAg+i8gQsJcGScYzDIsjnoYlumi6plhq7HnLDMvLBEl6sMhQjDUbYsz2cXUA6yZAIjE1j0/2Lk988GRmdHlnwuoFpz/v9epcrSymYW1gZYQJkFlJlHhf5sJy93LGBjx3y2dszHcMyp9DOv2s/8Sh+d1MgHjnzGyJujluvg5c6FbMrvQinbSejciD9qHMM9K4YRBJDLBGSjR95VyVcG/R8qlQE6amU/dqUeCDo6cJmcH38VZAiDDC6TI8gEZAIjMCMwOP7wpey/eN6E/l3GsXN2VZjZYfiQcKNz7h2x/U8DXwdOB65rUfVkFsoERqYjS+csWUfLRb8AqzUfRuqtOQPlKv2DVfrLVQbLNf9LEsBBzfmw48OIoxYOh6H61wAW/bIDv80ERhD9EqyG/hd7/QFEv2SNXMb/4u8brNAz4MNQuRqSq4ehTEBgxFqeotanqKWq/nz79xpt8b/wK1H467FOejLDA84sqks9YDWqZhk2BDk2sONf7KEFbM5Mfr2Ul7NzWZU9qLHCZtTIMGAZBuiAGZoEVLEsG7KT+1ALLaAnM/LfIUn95yxJNXT0lir0jhhSYjyb25Vnc0nfFCgBpXrXWxaYHz2i4/UfozLQB9AfPaaiI3rUlaPHjh2818LpDA7jmrXBAXg3/v/5ZaP2XwV8GTgTBQeRMZkZGTMyAeRzO+e00loYMlgJCQIfWjI+7eCcb9EpVWqUyjWcc2SCwI/xi5rr691g5WqNWugIor+Ig+iv/VrUKlRvISJq5amHl/iffA6Gwlu9a805RrQOhVFoq3+A1oNfGP3lO+J49Lz+17F/vv37t2gcopnhohBZbx2rVEMqUStbJWqxygZGLjv813Q2es+ZIPBDMso1+gYr9Jaq9EUtYPXWr1w2wDlGBL16EKi3jA/99R4JnYuucTiiwyUwf94gMAYr4ZghLzCbUABMsyBo3rTc2RwcVuAz4b3xnc65kpmtjo6LiIwpEwR057fvdjEzOrIZOrIZ5o8z81Gaw0VddtXQkcv4oBI/VqmFlMq+mzCXCcjnMnRkAzJBELWqDQeyWjgyTMXDk5kPfPVuxlJlZJdkvWwQWNSq5l/joia5etiLt+JVa6E/NwylomptOJDVQ+LQ+2H7IBg6xz67bd/tNVNmc3DYC9jonEsa5r0O+CMz63DObdcOZGZnA2d/+MMfnuk6iojIFJkZ2YyRTWgYi4e8JMEOjsv2ZvOMsW5grLlhpViZ7TjnrnTOTduEWBERkdliNgeHfmCsob+dsTIiIiLSoNkcHNYDi8wsKTwsxXdjNG/FDBERkVlgNgeH+/Dv73XxnWbWCRwJ3N+KSomIiLSz2RwcrscPQD131P4P4sc2XNv0GomIiLS5WTurwjn3kJldAZxjZjcCP8GvHPkx4A60hoOIiMiEzdrgEDkXWAucDZwCbAQux9+rYpK3UxMREdl5zerg4JyrAZdEDxEREZmi2TzGQURERKaZgoOIiIg0TMFBREREGjarxzhMl0Kh0OoqiIiINIsrFotj3m5TLQ4iIiLSMHOz5F7k7cLM7tcNtKZO13F66DpOD13H6aHrOD1m+jqqxUFEREQapuAgIiIiDVNwaL4rW12BWULXcXroOk4PXcfpoes4PWb0OmqMg4iIiDRMLQ4iIiLSMAUHERERaZiCwwwzs8DMzjOzR82sZGbPmtklZjan1XVLIzM7yMwuNrNfm9kGM+sxs9Vm9pmka2ZmB5vZTWa22cz6zOwuMzuhFXVPMzPrNrOnzcyZ2b8kHNd1HIeZ7Wpm/2xmT0T/jzeY2e1m9sejyuk6jsHM5prZ35vZQ9H/641mdreZvc/MbFTZnf46mtn5ZnaDmT0V/b9du4PyDV+zqX4uaeXImXcp8DHgR/i7dB4afX2UmZ2k23tv56+BjwA3A9cCFeB44AvAu8zsGOfcAICZHQjcDVSBrwBbgQ8Ct5rZW51zt7Wg/ml1MbAo6YCu4/jMbD9gFTAX+DfgcWABsBxYGiun6zgGMwuAnwJ/BHwLuBzoBt4N/Dv+9+LfRWV1Hb0vAi8DDwALxys4iWs2tc8l55weM/QADgNC4Iej9n8UcMBftrqOaXsARwMLEvZ/Ibpm58T2fR+oAUfG9s0FngEeIxr8u7M/gNdEv1D+NrqG/zLquK7j+NfvLuBZYMkOyuk6jn1tjo1+9i4dtb8DeArYouu43TV7Rez5fwNrxynb8DWbjs8ldVXMrHcDBlw2av9VQD9wZtNrlHLOufudc1sTDl0fbQ8HiJrU3g6scs6tjr2+F7gaOAhYMcPVTT0zy+B/3n4G3JhwXNdxHGb2BuB/AF9xzj1vZjkz604op+s4vvnRdn18p3OuDGwE+kDXMc4591Qj5SZxzab8uaTgMLNW4JPdvfGdzrkSsJqd5D/ANNk72r4YbZcDeeBXCWV/HW11feE84BDgnDGO6zqO70+i7R/M7BZgAOgzs8fNLP4LVtdxfPcCW4BPmdn/MrN9oz75LwGvBS6Kyuk6TtxEr9mUP5cUHGbWXsBG59xgwrF1wCIz62hyndpO9FfzBfjm9uui3XtF23UJL6nvW5pwbKdhZgcAnwMuds6tHaOYruP4Do62VwG7Au8FPgCUgW+b2fuj47qO43DObcb/Vfwyvln9GeBR/HimdzjnroqK6jpO3ESv2ZQ/lzQ4cmZ1A0n/OAClWJlyc6rTti4DjgH+3jn3WLSv3lycdH1Lo8rsrL4BPA18dZwyuo7jmxdte4Djo6Z1zOxH+L75L5rZt9B1bEQvvq/+ZvxAvl3xweE6MzvVOfdzdB0nY6LXbMqfSwoOM6sfWDzGsc5YGRmDmX0e38x+pXPuS7FD9euWT3jZTn9to2b0NwNvcM5Vximq6zi+gWj73XpoAP8XtJndDPwVvlVC13EcZrYMHxbOc859M7b/u/gwcVU0M0DXceImes2m/LmkroqZtR7f7JP0D7oU31yk1oYxmNlFwD/gp2v9zajD9UFWSc2W9X1JTXezXvTz9lXgJ8ALZvZKM3slsF9UZEG0byG6jjvyXLR9IeHY89F2F3Qdd+Q8/IfSDfGdzrl+4Mf4n8390XWcjIlesyl/Lik4zKz78Nf4dfGdZtYJHAnc34pKtQMzuxC4ELgGOMtF84ViHsI3tx2b8PJjou3Oen27gN2BU4A1sceq6PiZ0ddnoeu4I/UBZHsnHKvvewldxx2pf4BlEo5lY1tdx4mb6DWb+udSq+eqzuYHsIzx58ue2eo6pvGBHwjp8KEhGKfcDfi5y0fE9tXnLj/OTjLfO+G65IB3Jjw+HF3Xn0ZfH6TruMNruQuwDd/yMDe2fwm+z/7x2D5dx7Gv46XRz96nRu2vt3q9DGR1Hce8fjtax6HhazYdn0u6O+YMM7PL8X30P8I3HddX6Pov4ASnlSNHMLOPAP8C/AH4LP4HPO5F5wdRETW/34tfXfJS/C/4D+L/Y5zinLu1WfVuB2a2P36w5BXOuXNi+3Udx2FmZwP/CjwM/B/8okUfxoeHtznnVkbldB3HEK2++QA+iF2L//23K/767A98xDlXjMrqOgJm9h6Guxc/iv+5uyT6+hnn3LdjZSd0zab8udTqJDXbH/imuU/gV+8axPc1fZXYXy96jLhe/4FPvWM9Vo0qfyjwf/FzxPuBXwIntfp9pPGB/wW93cqRuo4NXbvT8HPi+/AzLFYCx+k6TugaHohfbvq56ANuG3AncJquY+L1WtXo78GJXrOpfi6pxUFEREQapsGRIiIi0jAFBxEREWmYgoOIiIg0TMFBREREGqbgICIiIg1TcBAREZGGKTiIiIhIw3R3TBGZ9QqFwkX4e58cXywWV7W2NiLtTcFBRHaoUCg0slKcPpRFdgIKz5SXCAAAA5ZJREFUDiIyEZ8b59jaZlVCRFpHwUFEGlYsFi9qdR1EpLUUHERk2sXHFODv8HcucAj+BlH/Cfx9sVh8IeF1r8LfFfVEYHdgI3Ab8PlisbgmoXwGfxfA9wCH4+8guA5/g6B/HOM17wQ+FZUv4W9Y9YlisbhuKu9ZZGehWRUiMpPOA74JPAhchr8b3/uBuwuFwu7xgoVCYQVwP3AmcB/wz/g7Up4B3F8oFI4eVb4D+BnwDWAf4Drg68BvgD8HjkuoTwH4Dr5b5Qrgv4G/AG4rFAr5Kb9bkZ2AWhxEpGFRS0KSUrFY/HLC/rcCry8Wi7+NneNSfAvEl4EPRPsMuAaYD5xZLBavjZX/C+B7wHcK/3979/OiUxTHcfw9KbIaVn5lLRZCoTRhIdmIjURhq+9eScr8B8rim7JRitggykIpIilpysKPlVKEjSklwlic8+h2ey73mcyG96ueTvO95565dzWfufec80Ssycwf9dAksAO4AezLzC+NcxbUsdp2ARsz82mj70XgALAHuNJ585IAnzhIGs2pjs/xjv4XmqGhmgSmgYON//K3UF5lPGyGBoDMvAzcB1YBE/DrFUUAn4GjzdBQz/mSmR+GXM+ZZmioztV2U8c9SGrwiYOk3jJzbMRT7g4ZYzoipoBtwGpgCthQD9/pGOcOJTSsB+5RQsY48Cgz34xwPY+H1F7XdvEI40j/LZ84SJpL7zrqg4mR4632bUf/QX1Rqx11QuPHIbVvtZ034ljSf8ngIGkuLemoL63tdKtdOqQvwLJWv0EAWDH7S5M0GwYHSXNpW7sQEePAOspSyGe1PJgHsb1jnEH9SW2fU8LD2ohY/jcuVFI/BgdJc+lQRKxv1SYpryYuNSY1PqAs1Zyo+yz8Un/eCrykTJIkM78DCSwEzraXUkbE/PZyT0l/h5MjJfX2m+WYANcyc6pVuwU8iIgrlHkKE/XzisZKjMyciYgjwG3gckRcpzxVWAXspWwcdbixFBPK9tebgd3Ay4i4WfutBHYCx4Dzs7pRSZ0MDpJGceo3x15RVkg0nQauUvZt2A98ovwxP5GZ75sdM/NR3QTqJGV/ht2UnSMvUXaOfNHq/zUidgFHgcPAEWAMeFN/5/3Rb0/Sn4zNzPT50jtJ6s+vsZb+Xc5xkCRJvRkcJElSbwYHSZLUm3McJElSbz5xkCRJvRkcJElSbwYHSZLUm8FBkiT1ZnCQJEm9GRwkSVJvPwEwmbUD7EBc7QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAg4AAAGdCAYAAAB6oftJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdeZxkZX3v8c+v9u5ZYYZtBmQzMMhqYBRigoBEjUS5BmNA8aoRiFYAIbkuuCAuV4wGMBrKBXJVEjCgIpELgnKVxaAC4uCIbAozMAswPWtv1bWc5/7xnOo+3XO6p7qnu+pUz/f9etXrVJ/znNPPOdPT9evfs5lzDhEREZFmpNpdAREREekcChxERESkaQocREREpGkKHERERKRpChxERESkaQocREREpGkKHERERKRpChxEZIfM7Jtm5szs7imca2Z2hpn9p5k9Y2YDZrbVzB4zs6+Y2fFNXuc4M/s3M3vCzPrNbNDMVpnZ/WZ2lZmdbmZzxjl3dzP7aFh2k5lVzewFM3vEzL5tZn9nZgdN9t5EdkWmCaBEZEfM7JvAO4F7nHMnTeK8/YHvAMsju3uBLFCI7Pt34DznXHmc63wS+Dhg4a4A2ALMC6/V8Gbn3C1jzn0l8ANgz8jubUAaiAYa/+Wc+x/N3ZnIrksZBxGZEWZ2APBzfNCwDfhfwN7OufnOuS5gGfBlfBDwDuAOM8vGXOcs4FJ80HATcByQd84tArqAI4EPAr+NOXchI0HDU8DbgXnOuQXOubnAPsCZwM1AdbruXWQ2U8ZBRHZoshkHM0sDPwOOBzYAJznnfjdO2bcC38b/IfMF59wHxxz/BfBK4Dbn3F/u4PsWolkLM3sv8BVgCHipc27NBOd2OecGd3RvIrs6ZRxEZCb8FT5oACiOFzQAOOduAr4afvl+M1sypsiR4fb/7uibxjR1NM5dMVHQEJ6roEGkCQocRGQmnBdun3DOfbeJ8p/DN1nkgHeNU2bpTtRnHzOzHRcTkR1R4CAi0yrsp/An4Zf/1cw5zrnngF+FX5405vBD4fYCM/uzSVance5LgP9tZvlJni8iYyhwEJHpdgDQHb5/ZBLn/SbcHjZm/6fx2YgFwL3hMM6vmtl7zOyIHWQSvg08Eb6/BHjezL5nZh80s5PNrHuCc0UkhgIHEZluu0feb5zEeT3hdlF0p3PuLuB/AKvDXcuAvwOuBVYC683sC2Y26rzw3DJwCnBbuGshvv/FPwE/AbaY2Q/M7IRJ1FNkl6bAQUSm21T7Eox7nnPuVuClwGnAvwC/ABqdGffCD/V8xMyWxZy7LhyNcRg+63AbsD48nAXeCPy3mb1/ivUW2aUocBCR6RbNMmyXBZhAo+ymuIPOuZpz7nbn3EXOuRPw2YM/Z2S0xVLghvGaLpxzjzvnPuec+0vn3BJ8IPFJYAAftFxpZn88ifqK7JIUOIjIdFuF/zAGOHoS5x0VbscduhnlnKs45+5yzr0R+Ldw98uBY5o8/3Hn3GXAXwAO//vwnZOor8guSYGDiEwr51wVuD/88vRmzjGz/YBjwy/vmcK3/bfI+0Mmc6Jz7l78rJKTPldkV6TAQURmwtfD7aFm9pYmyn8Y//uoCnxjCt+vP/K+shPnT+VckV2KAgcRmQk3Aw+G70tm9rLxCoZTTr83/PJLzrl1Y46f2sTkTW+LvF8ROXe5mS2Y6EQzO5yRJpUVE5UVEQUOIjI5WTNbvINX1jlXxy8e9QKwB3C/mf2DmQ2vUGlmh5jZvwA34H8X/Qz4SMz3/E9gpZl9wMyONLNUeH7azF5mZl/FL3IFcKtz7pnIuX8DrA6X7z7VzOZFvv8iM3sfcFf4/fvxQzxFZAJa5EpEdiiyyFUzTnbO3R2edyB+We1jI8e34YdBdkX23QCcE7dehJmtB/aO7KoDW/ETQqUj+/8beJNzblPk3MvxzSBR24AMI5NUgV+i+0zn3J1N3J/ILi3T7gqIyOzlnHvGzJYDbwH+Gr/K5Z5ADXgSuBv4pnPu5xNc5lDgDcDJ+CW1D8QHDUPA88DD+OW2v+u2/0voI/jhmq8HTsBPHrUHfvjlBuAx4E7gGufchp28XZFdgjIOIiIi0jT1cRAREZGmKXAQERGRpilwEBERkaYpcBAREZGmKXCYQLFYdMViUb1HRUREQhqO2RwFDyIisquYcKZWZRxERESkaQocREREpGkKHERERKRpiQgczMyN8+qLKXuomd1iZpvNrN/M7jOzU8a5bsrMLjazx82sbGbPmdkVZjZn5u9KRERk9klS58j7gK+P2VeNfmFmBwP34+e5/zx+oZtzgTvN7C+cc3eNOf8q4ELg+8AVwGHh1y83s1Odc8FUKlqtVlmzZg3lcnkqp+8yCoUC++67L9lstt1VERGRaZKkwOFp59x/7KDM5cBC4Fjn3AoAM7sOeBS42syWNRa5MbPDgQuAm51zZzQuYGbPAF/CL/l7w1QqumbNGubNm8cBBxyA2YSdT3dZzjk2btzImjVrOPDAA9tdHRERmSaJaKpoMLOcmc0d59gc4E3A3Y2gAcA51wdcCxwCLI+cchZ+SMkXx1zqGmAAOHuq9SyXyyxatEhBwwTMjEWLFikrIyIyyyQpcHgL/gO918xeNLMvm9mCyPGjgDwQt/zuL8JtNHBYDgTAA9GCzrkysGJM2UlT0LBjekYiIrNPUgKHB4DL8MHDO4GfAOcD90UyEEvC7dqY8xv7lkb2LQF6nHND45RfbGa5uMqY2Xlm9tCk7qDF5s6NTcyIiIjMqEQEDs65Vzrn/tk5d4tz7jrn3JnAR4EjgfeHxbrDbVwgUB5TpvE+rux45aP1+bpz7rimb0BERGQXkYjAYRxfACrAaeHXA+E2H1O2MKZM431c2fHKdyTnHB/4wAc44ogjOPLII7nxxhsBWL9+PSeeeCLHHHMMRxxxBPfddx/1ep13vetdw2WvuuqqNtdeREQ6TZJGVYzinKua2TpgcbhrXbhdGlO8sS/ajLEOeJmZ5WOaK5bimzEqO1vP1336tp29xLju/PhpOyxz8803s2LFCh555BF6enpYvnw5J554IjfccAOve93r+OhHP0q9XmdgYIAVK1awdu1afvvb3wKwZcuWGau7iIjMTonNOJhZAdgXeCHctRLf9HBCTPHjw220X8KD+Pt7Rcx1jxlTtmP97Gc/46yzziKdTrPXXnvx6le/mgcffJDly5fzjW98g8suu4yVK1cyb948DjroIJ5++mkuuOAC7rjjDubPn9/u6ouISIdpe+BgZovGOfRpfEbkVhgednkrcJKZHR05fy5wDvAUo0dQ3Ihf1fKiMdc9F9+34frpqH+7hdNWbOfEE0/k3nvvZenSpbzjHe/guuuuY7fdduORRx7hpJNO4uqrr+acc85pcW1FRKTTJaGp4mNmdjzwU+BZYC7wBuBk4JfAlyNlLwFeA/zIzK4CtuEDgaXAaS7yKeqcW2lmVwPnm9nNwO2MzBx5D1Oc/GmsZpoTADb2ltnY6/tk7j43z+L5XdPx7TnxxBP52te+xjvf+U42bdrEvffeyxe+8AVWr17N0qVLOffcc+nv7+fhhx/mDW94A7lcjjPOOIODDz6Yd73rXdNSBxER2XUkIXC4G3gZfhjmIqCOzx58FLgynHcBAOfc783sVcDngA8DOeBh4PUx002DzzasAs7Dd7LswQcil051uumpSkXmNAjikwRT8uY3v5mf//znHH300ZgZn//859l777351re+xRe+8AWy2Sxz587luuuuY+3atbz73e8mCPytX3755dNXERER2SXYeKlugWKx6ABKpdKo/Y899hiHHXbYpK61pX+IF7cOArCgO8deC2NHgs46U3lWIiLSVhPO3tf2Pg67iugsigrVRESkUylwaJHo7MvK8oiISKdS4NAiKWamj4OIiEgrKXBoEWUcRERkNlDg0CKj+jgobhARkQ6lwKFFlHEQEZHZQIFDi6SUcRARkVlAgUOLRDMOQYsHZM6dO3fcY6tWreKII45oYW1ERKSTKXBoEUMZBxER6XxJmHK6s53z+qaKZYFDJnvta++I3f2hD32I/fffn2KxCMBll12GmXHvvfeyefNmqtUqn/nMZzj99NMn9e3K5TLve9/7eOihh8hkMlx55ZWcfPLJPProo7z73e+mUqkQBAHf+973WLJkCW9961tZs2YN9Xqdj3/84/zN3/zNZO9QREQ6jAKHDnTmmWdy0UUXDQcON910E3fccQcXX3wx8+fPp6enh+OPP543velNo0Zz7MjVV18NwMqVK3n88cd57Wtfy5NPPslXv/pV3v/+9/P2t7+dSqVCvV7n9ttvZ8mSJdx2220AbN26dfpvVEREEkdNFR3o5S9/OS+++CLr1q3jkUceYbfddmOfffbhIx/5CEcddRSnnnoqa9eu5YUXXpjUdX/2s5/xjne8A4Bly5ax//778+STT3LCCSfw2c9+ln/6p39i9erVdHV1ceSRR3LXXXfxoQ99iPvuu48FCxbMxK2KiEjCKOOws8ZpTojz1Pqtw0MxX7rPglEjLSbrLW95C9/97nd5/vnnOfPMM7n++uvZsGEDv/rVr8hmsxxwwAGUy+UdXyhivGGib3vb23jlK1/Jbbfdxute9zquvfZaTjnlFH71q19x++23c8kll/Da176WSy+9dMr3IyIinUGBQwsZIwtcOedGD7WYpDPPPJNzzz2Xnp4e7rnnHm666Sb23HNPstksP/3pT1m9evWkr3niiSdy/fXXc8opp/Dkk0/y7LPPcuihh/L0009z0EEHceGFF/L000/zm9/8hmXLlrH77rtz9tlnM3fuXL75zW9O+V5ERKRzKHBoITMbHlKxsyMrDj/8cHp7e1m6dCn77LMPb3/723njG9/IcccdxzHHHMOyZcsmfc1isch73/tejjzySDKZDN/85jfJ5/PceOON/Md//AfZbJa9996bSy+9lAcffJAPfOADpFIpstksX/nKV3buhkREpCOYZjEcX7FYdAClUmnU/scee4zDDjts0td7+oVt1OoBAAfuOY9sJj0NtUy2qT4rERFpmwnT4eoc2ULRPg1aIVNERDqRmipaqJ3rVaxcuXJ4xERDPp/nl7/8ZUvrISIinU2BQwuNWiGzxd/7yCOPZMWKFS3+riIiMtuoqWKKppIxiDYa7Qp9S3aFexQR2dUocJiCQqHAxo0bJ/3BuCv1cXDOsXHjRgqFQrurIiIi00hNFVOw7777smbNGjZs2DCp87b0DzFU9aMq+ntyFLKze1RFoVBg3333bXc1RERkGilwmIJsNsuBBx446fMuv/nX3P3oegA+ePrRvOYwfaiKiEhnUVNFC+WzI4+7Es7nICIi0kkUOLRQLjLhU6Vab2NNREREpkaBQwvlMpGMQ00ZBxER6TwKHFoomnEYUuAgIiIdSIFDC43OOKipQkREOo8ChxYa1cdBGQcREelAChxaaNSoCmUcRESkAylwaKHRoyqUcRARkc6jwKGF1MdBREQ6nQKHFtKoChER6XQKHFoomnGoKuMgIiIdSIFDC+WzyjiIiEhnU+DQQurjICIinU6BQwtpVIWIiHQ6BQ4tNCrjUFfGQUREOo8ChxZSxkFERDqdAocWUh8HERHpdAocWkijKkREpNMpcGihURmHah3nXBtrIyIiMnkKHFoonUqRThkADqgFChxERKSzKHBosbFZBxERkU6iwKHFRo2sUD8HERHpMAocWiyacRjSyAoREekwChxaLK+Mg4iIdLDEBQ5m1m1mz5iZM7N/jTl+qJndYmabzazfzO4zs1PGuVbKzC42s8fNrGxmz5nZFWY2Z+bvJF5WfRxERKSDJS5wAD4FLI47YGYHA/cDJwCfBz4AzAXuNLNTY065CrgS+B1wAfAd4ELgVjNry71H53Ko1JVxEBGRzpJpdwWizOyPgYuADwJXxBS5HFgIHOucWxGecx3wKHC1mS1z4eQIZnY4Pli42Tl3RuR7PAN8CTgTuGEGbyeWRlWIiEgnS0zGwczSwDXAHcDNMcfnAG8C7m4EDQDOuT7gWuAQYHnklLMAA7445lLXAAPA2dNZ/2ZpVIWIiHSyxAQOwMXAMuD8cY4fBeSBn8cc+0W4jQYOy4EAeCBa0DlXBlaMKdsyGlUhIiKdLBGBg5kdCHwS+JRzbtU4xZaE27Uxxxr7lo4p3+OcGxqn/GIzy41Tn/PM7KEdVnwKohmHqjIOIiLSYRIROABfAZ7Bd2QcT3e4jQsEymPKNN7HlR2v/DDn3Nedc8dNUJcpy6UNC9eoUMZBREQ6Tds7R5rZ2cBrgROdc9UJig6E23zMscKYMo33e45zrbjyM+uW6+BH3+PiSoUFC0/gOwuOVR8HERHpOG0NHMwsj88y3A48b2YvDQ81mhwWhPt6gHVjjkU19kWbMdYBLzOzfExzxVJ8M0ZlZ++heQ4qQ6SArPOZBo2qEBGRTtPupoouYA/gNOCpyOvu8PjZ4dfnACvxTQ8nxFzn+HAb7ZfwIP7+XhEtaGYF4JgxZWdeJjv8NtcIHJRxEBGRDtPupop+4K9j9u8BlPBDM/8N+I1zrs/MbgX+ysyOds49AmBmc/GBxVOMHkFxI/AR/LwQ90X2n4vv23D9NN/LxLIj/TCz+MBBfRxERKTTtDVwCPs0fHfsfjM7IHz7B+dc9PglwGuAH5nZVcA2fCCwFDitMflTeO2VZnY1cL6Z3YxvDjkMP3PkPbR68qdRGYcaoFEVIiLSedqdcZgU59zvzexVwOeADwM54GHg9c65u2JOuQhYBZyHbw7pAb4MXOqca+2ndjTj4JRxEBGRzpTIwCGcy8HGOfYYcHqT16njp66Om766tSIZh6z6OIiISIdqd+fIXUck45DTqAoREelQChxaJauMg4iIdD4FDq0S21ShjIOIiHQWBQ6tMqqpwo+qUMZBREQ6jQKHVonJOGhUhYiIdBoFDq0S0zlS8ziIiEinUeDQKso4iIjILKDAoVXihmMq4yAiIh1GgUOrRIdjonkcRESkMylwaJXM9qMqhpRxEBGRDqPAoVViJoCq1upE1uUSERFJPAUOrRLTOTJwUAsUOIiISOdQ4NAq6TSk/ONO40iFi3Nq9kgREekkChxaKW72yKr6OYiISOdQ4NBKWq9CREQ6nAKHVoqZy0EjK0REpJMocGilSMYhFxlZISIi0ikUOLRSJOOQVcZBREQ6kAKHVhqVcWgsra2Mg4iIdA4FDq0UO+20Mg4iItI5FDi0UkwfB2UcRESkkyhwaKWYPg5aIVNERDqJAodWGrVehfo4iIhI51Hg0EoZzeMgIiKdTYFDK8WskKmMg4iIdBIFDq0U18dBoypERKSDKHBoJY2qEBGRDqfAoZVimyqUcRARkc6hwKGVNHOkiIh0OAUOraS1KkREpMMpcGilaMaBxuqYChxERKRzKHBopbiMQ1VNFSIi0jkUOLRSXB+HujIOIiLSORQ4tFLsPA7KOIiISOdQ4NBKsfM4KOMgIiKdQ4FDK8WujqmMg4iIdA4FDq0UMwHUkAIHERHpIAocWimjmSNFRKSzKXBopWzMqAp1jhQRkQ6iwKGVYvs4KOMgIiKdQ4FDK42zOqZzrl01EhERmRQFDq0UzTiEU04HDuqBAgcREekMChxaKZJxyLuRvg0aWSEiIp1CgUMrxWQcQAtdiYhI51Dg0ErR4ZhBJOOgkRUiItIhFDi00qhRFbXh9xpZISIinaLtgYOZHWpm15vZY2a21cwGzOxxM7vSzPYZp/wtZrbZzPrN7D4zO2Wca6fM7OLwemUze87MrjCzOTN/ZzHSaTD/yNM4Us4HDJp2WkREOkWm3RUA9gX2Ab4PrAFqwJHAecCZZnaMc+5FADM7GLg/LPN5YCtwLnCnmf2Fc+6uMde+CrgwvPYVwGHh1y83s1Odc63/Uz+bhcoQ4CeBKltOGQcREekYbQ8cnHP/D/h/Y/eb2b3ATcC78EECwOXAQuBY59yKsNx1wKPA1Wa2zIWTIpjZ4cAFwM3OuTMi130G+BJwJnDDDN3W+DIjgUPW1SmjURUiItI52t5UMYHV4XY3gLB54U3A3Y2gAcA51wdcCxwCLI+cfxZgwBfHXPcaYAA4e2aqvQORfg6NSaA0qkJERDpFYgIHMyuY2WIz29fMXgt8LTx0e7g9CsgDP485/RfhNho4LAcC4IFoQedcGVgxpmzrxK2QqVEVIiLSIRITOADnABuA54A78U0SZzvn7guPLwm3a2PObexbGtm3BOhxzg2NU36xmeVijs2szPYZB/VxEBGRTpGkwOEW4M+BNwOfArYAe0SOd4fbuECgPKZM431c2fHKDzOz88zsoSbqPHnRjAMj61WIiIh0gsQEDs65Nc65u5xztzjnPgG8E/gnM7skLDIQbvMxpxfGlGm8jys7XvloXb7unDuu+dpPwqg+DuHS2so4iIhIh0hM4DCWc+43wK+BYrhrXbhdGlO8sS/ajLEO3xwRFzwsxTdjVKajrpOSienjoIyDiIh0iMQGDqEuYPfw/Up808MJMeWOD7fR5oUH8ff3imhBMysAx4wp2zrZ7ZfW1qgKERHpFG0PHMxs73H2nwwcQThiIhx2eStwkpkdHSk3F9+x8ilGj6C4EXDARWMufS6+b8P103QLkxOXcdCoChER6RBtnwAK+Eo4tfRP8HM3FIBj8RM09QL/GCl7CfAa4EdmdhWwDR8ILAVOa0z+BOCcW2lmVwPnm9nN+GGdjZkj76Edkz9BfB+HujIOIiLSGZIQOHwb3xHyHfhRFA4fQHwN+IJz7tlGQefc783sVcDngA8DOeBh4PUx002Dzzaswk9ffRrQA3wZuLQt001DbMahooyDiIh0iLYHDs65m/BTSzdb/jHg9CbL1vFrVFwxtdrNgFErZDY6RyrjICIinaHtfRx2Oco4iIhIB5uWjEOxWMzgOzIa8NtSqVSdjuvOSjFrVSjjICIinaKpjEOxWDywWCz+bbFYPCTm2Bvwy2H/Cj/EcV2xWDxjbDkJxWUcNI+DiIh0iGabKt6DX1Vy1IRJxWLxQOA7wJ74NSYex8+7cEOxWDx8Gus5e4yax8GPqtBwTBER6RTNBg5/CqwslUqrxuy/ED9J0zXAgaVS6XDgb4AscMF0VXJWiSxyNbJWhZoqRESkMzQbOBwI/DZm/+uBGvChUqnkAEql0nfxS1+/elpqONvEzBypjIOIiHSKZgOHPYBnozuKxeI84BDgoVKptGVM+YeBfXe+erNQzHBM9XEQEZFO0Wzg4IAFY/Ydgx9F8XBM+a0kYI6IRMps38dBTRUiItIpmg0cVgGvGrPvZHxA8cuY8nsAL0y9WrNY3ARQaqoQEZEO0WxW4EfAhcVi8WPAl/BNFO/D92+4I6b8cYxp2pBQTB8HZRxERKRTNJtx+DywBfgksBmfZdgL+EapVNoQLVgsFvcHXo5fSErGisk4VOsBwcj6XCIiIonVVOBQKpXWAycCPwXKwFr8+g8XxhR/J76Pww+nqY6zS6SPQ56RJgplHUREpBM03YGxVCo9CpzaRLlPAZ/amUrNatGmCkaChUq1TiGbbkeNREREmqZFrlptnIzDkIZkiohIB5iuRa6Oxo+yMOC+Uqn00HRcd1aK6eMAUKmqqUJERJKv2UWuTiwWi9cVi8XjY45dhp/L4Qrgn4FfFovFK6e1lrNJZvtRFaCMg4iIdIZmmyr+Gngr8Fh0Z7FY/DPgUiAArge+CmwE3h+umiljjVpWuzb8XrNHiohIJ2g2cDgB+GWpVNo6Zv/f4SeBurBUKv3PUqn09/jRFzXg3dNXzVkkknHIRDMOaqoQEZEO0GzgsAR4Kmb/KUA/fnVMAEql0uPAncDyna7dbBTt4xAo4yAiIp2l2cBhEWOmkC4Wi3sDewP3l0ql2pjyT+EniJKxxs04KHAQEZHkazZwGGT7QOCPw+2vY8oP4ZsrZKx0Gsw/9pQLSDnfRKEJoEREpBM0Gzg8DvxFsViMDt88Dd+/4f6Y8vsB63eybrOT2ahJoIYXulJThYiIdIBm53H4Ln69ih8Ui8Wv4he5eg9+aukfx5R/FWNGYEhEJguVIcCPrBgiS0VNFSIi0gGaDRy+DJwFvB54XbjPgP9VKpXK0YLFYvGVwAHhORInNuOgpgoREUm+Zhe5GsIPs7wUv4z29cAbS6XSv8YUPwb4L+DW6arkrDNqLodwaW1lHEREpANMZpGrfuAzTZT7GvC1nanUrBcZWZFFGQcREekcWuSqHeIyDuocKSIiHWDSi1wVi8U5wFvwHSCX4EdWrAf+G/humJmQiWRi+jioqUJERDrApDIOxWLxbGAV8H+Ac4A34IdlnhPuW1UsFt8+zXWcfWLWq9A8DiIi0gmaDhyKxeIFwLfws0g+BHwWeB9QDN8/GB67rlgsnj/9VZ1FYjIOaqoQEZFO0FRTRbFYPAS4EtgMvL1UKt0ZU+xjxWLxz4EbgCuLxeKPSqXSk9NX1Vkku/3S2uocKSIinaDZjMP78fM2nD5O0ABAqVT6MXA6kAYu3PnqzVJxGQf1cRARkQ7QbODwGuDuUqn03zsqWCqV7gd+Cpy6MxWb1WL6OGjKaRER6QTNBg77Ar+axHV/FZ4jcWIzDmqqEBGR5Gs2cHD4popmTabsrieScdAiVyIi0kmaDRzWAMdN4rrHAs9Nvjq7iMz2nSM1HFNERDpBs4HDT4ATi8Xiq3ZUsFgs/glwUniOxInLOKhzpIiIdIBmA4cv4psrvl8sFsft9FgsFl8D3ALUgX/Z+erNUtnt16rQPA4iItIJmprHoVQqPVUsFv8RH0DcWSwWHwDuwjdHOOAl+FEUr8D3b/gHzeEwgUzMqAp1jhQRkQ4wmdUxv1QsFrfgg4dX4oOEKAO2AP9YKpW+MX1VnIViJoAKnKNWD8ikte6YiIgk16Q+pUql0nX47MLfAt8A7gDuBL4JvAfYX0FDEyKBQ8FGMg0aWSEiIkk36dUxS6VSHz5Q+OZ4ZYrFYgHIlUqlbVOu2WyWiQ8cKtWAOfl2VEhERKQ5M5UX/wqwaYau3fkioyryjGQZlHEQEZGkm8kGdU0CNZ5IxiFPNOOgwEFERJJNPfHaIbpWxaiMg0ZWiIhIsilwaIeYmSNBczmIiEjyKXBoh1GrY0YyDjsldb8AACAASURBVJrLQUREEq7tgYOZHWJmnzKzX5jZBjPrNbMVZvZRM5sTU/5QM7vFzDabWb+Z3Wdmp4xz7ZSZXWxmj5tZ2cyeM7Mr4q7bUqMyDrXh98o4iIhI0rU9cMDPCXEx8AfgU8AHgCeAzwD3m1lXo6CZHQzcD5wAfD4sOxe408zipsK+CrgS+B1wAfAd4ELgVjNr371H5nHIjGqqUMZBRESSral5HIrF4kz+Kfxd4HLn3NbIvq+a2VPAR/ETS/1ruP9yYCFwrHNuBYCZXQc8ClxtZsuccy7cfzg+WLjZOXdG48Jm9gzwJeBM4IYZvK/xRaaczgYjGQctdCUiIknX7F/dNoVXU5xzD40JGhpuDLdHAITNC28C7m4EDeH5fcC1wCHA8sj5Z4X1+OKY614DDABnN1vHaTduxkGBg4iIJFuzi1y1I62/b7h9IdweBeSBn8eU/UW4XQ48EHkfRL4GwDlXNrMVjA4yWisaOEQzDmqqEBGRhEtCH4ftmFkauBSoMdKcsCTcro05pbFvaWTfEqDHOTc0TvnFZpaLOTbzIqMq0pHAQRNAiYhI0iUycMA3LxwPXOqceyLc1x1u4wKB8pgyjfdxZccrP8zMzjOzh5qv7iRFRlWk69GMgwIHERFJtsQFDmb2aeB84OvOucsjhwbCbdwyUIUxZRrvx1syKq78MOfc151zxzVX4ylIZ8B8N5CUC0g530ShURUiIpJ0iQoczOwy4GP4JbvfO+bwunC7lO019kWbMdbhmyPigoel+GaMytRruxPMRmUdsmEHSY2qEBGRpEtM4GBmnwA+AVwHnNMYVhmxEt/0cELM6ceH22jzwoP4+3vFmO9TAI4ZU7b1YmaP1KgKERFJukQEDmZ2KXAZ8O/Au51z2+Xsw2GXtwInmdnRkXPnAucATzF6BMWNgAMuGnOpc/F9G66fxluYvFEZB9/PQVNOi4hI0jU1HHMmmdnfA58EngXuAt5mNmoaiBeccz8O318CvAb4kZldBWzDBwJLgdOiWQrn3Eozuxo438xuBm4HDsPPHHkP7Zr8qSG7fVOFMg4iIpJ0bQ8cGJlP4SXAt2KO3wP8GMA593szexXwOeDDQA54GHi9c+6umHMvAlYB5wGnAT3Al/GjNdr75310vYpwaW3N4yAiIknX9sDBOfcu4F2TKP8YcHqTZevAFeErWSJ9HIYzDuocKSIiCZeIPg67pFErZDYyDgocREQk2RQ4tEtsxkFNFSIikmwKHNplVMYhHFWhjIOIiCScAod2ics4KHAQEZGEU+DQLtmYPg5qqhARkYRT4NAuMVNOK+MgIiJJp8ChXbLb93Go1AK2n2lbREQkORQ4tEukj0PeRpooqnU1V4iISHIpcGiXSFNFVyRwUD8HERFJMgUO7RLJOBRspHlC/RxERCTJFDi0y6iMw0iwMKRpp0VEJMEUOLTLOH0cKlroSkREEkyBQ7tEmyqIZBzUVCEiIgmmwKFdIoFDF7Xh91ohU0REkkyBQ7t0dY+8DSrD74fUVCEiIgmmwKFduuYMv+2ujwQOyjiIiEiSKXBol0Ik41AfGn6vPg4iIpJkChzapXsk41CIBA4aVSEiIkmmwKFdIhmHfK08/F7zOIiISJIpcGiXSB+HXE1NFSIi0hkUOLRLoTD8NlcbIuV8E0VFa1WIiEiCKXBol1R6dAdJVwWUcRARkWRT4NBOkbkc5gS+uUKdI0VEJMkUOLRTJHDoDieBUudIERFJMgUO7VSIZhx84KBltUVEJMkUOLRTdPZI1wgc1FQhIiLJpcChnbpiMg5qqhARkQRT4NBO0YxDo4+DMg4iIpJgChzaqRA3qkIZBxERSS4FDu3UHZNxUFOFiIgkmAKHdopmHNQ5UkREOoACh3aK7eOgjIOIiCSXAod2ips5UmtViIhIgilwaKdRM0dqrQoREUk+BQ7tVBhpqpjjGhkHBQ4iIpJcChzaqXv7tSpqgaMeuHbVSEREZEIKHNopZq0K0FwOIiKSXAoc2ilmrQrQXA4iIpJcChzaKZeHlP8nyLk6WecDBs3lICIiSaXAoZ3M4udyUMZBREQSSoFDu8XN5aA+DiIiklAKHNpNK2SKiEgHUeDQbnHrVaipQkREEkqBQ7tpvQoREekgChzaLbaPg5oqREQkmRQ4tFtMxkFNFSIiklQKHNqta/s+DuocKSIiSdX2wMHMLjGz75jZ02bmzGzVDsofama3mNlmM+s3s/vM7JRxyqbM7GIze9zMymb2nJldYWZz4sq3Rdf261VoOKaIiCRV2wMH4LPAKcAfgM0TFTSzg4H7gROAzwMfAOYCd5rZqTGnXAVcCfwOuAD4DnAhcKuZJeHeRzVVzBmeAEoZBxERSaZMuysAHOycexrAzH6LDwTGczmwEDjWObciPOc64FHgajNb5pxz4f7D8cHCzc65MxoXMLNngC8BZwI3zMD9TE5BGQcREekcbf+ruxE07EjYvPAm4O5G0BCe3wdcCxwCLI+cchZgwBfHXOoaYAA4eyeqPX1GLXTlR1VoymkREUmqtgcOk3AUkAd+HnPsF+E2GjgsBwLggWhB51wZWDGmbPt0bb+0toZjiohIUnVS4LAk3K6NOdbYt3RM+R7nwj/jty+/2Mxycd/IzM4zs4emXNPJ0ARQIiLSQTopcGj8aR4XCJTHlGm8jys7XvlhzrmvO+eOm3QNpyJuVIWaKkREJKE6KXAYCLf5mGOFMWUa7+PKjle+PaKjKjSPg4iIJFwnBQ7rwu3SmGONfdFmjHX45oi44GEpvhmjMo31m5qxoyqcY0v/eIkSERGR9uqkwGElvunhhJhjx4fbaL+EB/H394poQTMrAMeMKds+mQzkfGyTxlFwVXq2lXdwkoiISHt0TOAQDru8FTjJzI5u7DezucA5wFOMHkFxI+CAi8Zc6lx834brZ7TCkzEm69DTW6YeuDZWSEREJF7bJ4Ays3cA+4df7gHkzOxj4dernXP/Hil+CfAa4EdmdhWwDR8ILAVOa0z+BOCcW2lmVwPnm9nNwO3AYfiZI+8hCZM/NXR1wzY/aeYcV2FT4NjcN8Ti+YUdnCgiItJabQ8cgPcArx6z79Ph9h5gOHBwzv3ezF4FfA74MJADHgZe75y7K+baFwGrgPOA04Ae4MvApc655PRAjBlZsWHboAIHERFJnLYHDs65kyZZ/jHg9CbL1oErwldyxaxXsWFbmcPaVR8REZFxdEwfh1ktZr2KDdsG21UbERGRcSlwSIJRGQc/FHODRlaIiEgCKXBIgmgfh3BqiQ1blXEQEZHkUeCQBDF9HF5UU4WIiCSQAockiBlVoUmgREQkiRQ4JEEk4zA3DBw29w1RrSdnxKiIiAgocEiGyKiKheka4Ke83Kisg4iIJIwChyToHsk4LLDa8HsNyRQRkaRR4JAEkYzDXKrD7zUkU0REkkaBQxJEOkc25nEAeFFDMkVEJGEUOCRBpHNkoT4SOKipQkREkkaBQxJEMg656kjzhJoqREQkaRQ4JEG+C8wASFeHSIULdypwEBGRpFHgkASpFBS6hr/UQlciIpJUChySIjKyYl44sqJ3sEq5UhvvDBERkZZT4JAUkQ6SS0diCF5Uc4WIiCSIAoekiHSQ3Kdgw+/VXCEiIkmiwCEpIhmHvXIja1RosSsREUkSBQ5JEenjsEd2JHDYoEmgREQkQRQ4JEVkvYrdM/Xh9xqSKSIiSaLAISkiGYcFKS10JSIiyaTAISkinSPna6ErERFJKAUOSRHpHDnHVYbfv7h1EOdcO2okIiKyHQUOSRHJOGQrZfLZNADlap2+siaBEhGRZFDgkBTd84bf2rrV7DG/MPy1+jmIiEhSKHBIij86HDJZ//7Z33MMm4YPKXAQEZGkUOCQFHPnw/ITh788qefh4ffqICkiIkmhwCFJTjpt+O1hax9hbt0HDJoESkREkkKBQ5IcdBjsdxAAmXqVU/sfB5RxEBGR5FDgkCRmcNJfDn/5l70rwTmeWr+VeqAhmSIi0n4KHJLmlScPzyK5X20LR5fX8GxPH7c88EybKyYiIqLAIXkKXXDCa4a//Mu+3wLwrZ8+wfrNA+2qlYiICKDAIZkinSRfNfg0u9f6GaoF/MttKzWLpIiItJUChyRaegD80REApF3AGb2/BuDXz/Tw49+saWPFRERkV6fAIakiWYe3bPs1f7fpPlIu4Gs/eozNfUNtrJiIiOzKFDgk1bF/Bi992fCXf9W7gk9suI36QD8f/88HeeaFbW2snIiI7KoUOCRVJgMXfxb++FXDu44fXMUVz3+PodXP8PfX/oz/85PHGarW21hJERHZ1ShwSLJ8Ad77UXj9W4d3HVzt4Zr113P5uu/x/I9+SPErP+XuR9dRrQdtrKiIiOwqMu2ugOxAKgVv+VvYeyn8+5eg7jMMRw+t5eihtWzZdC+//v1+fHveEvY46iiOPfkE9txztzZXWkREZisFDp3iT18HSw6AH96Ie+QXWOAzDAuDQU4eeBIGnoQf3039x19m9Zy9eWGfQ6kdciTzj3o5B79kT7py+qcWEZGdZ5oXYHzFYtEBlEqldldltM098LM7Ce65ndSWjRMWrZKinMqSIyBDQMoF1OcuhL33I710P2zv/eAlB8OBh0I216IbEBGRBLOJDurP0E6022J449tJveFMWP0U9aefoOc3j8Cqp9hjYMOojitZArLB6OGbmd7N0LsZnvrN8L5aKs3GRQew7SXLyOy5F/OpMicYIl8ZxPJd8EeH+7kluue06CZFRCSJFDh0snQaDlpG+qBl7HXq6QD0btxMz0MPUP/dr1n47GMs7n2hqUtlgjp7bfgDe234Q3yBOyDAeGH+EjYtfgmZri7yXQUKXQUKXXksm/MjQTJZUimjq9JPpm8r9G6Bchn2PQAOOdIPMQ3X4hhXEPgFv2zCoLc9nIPHfg333Ql77gOnvhnmLWh3rUREWkZNFRNIbFPFZAz20zcwxJMv9PG7dVt5Yv02ahteZM7m9ew1uJH9q5s4bGg9+9W2tKQ6dUuxft4+9HctoJ7rwhW6sFyeOUO9zB3YTHfvRgp9mwmyeWq77Umw+57Y4r3IFAqk61WsVoXKEFQrUKv6V7Xqg4xc3o9EyRVg3nzY90B4yUth7/18kBUVBFCv+Vet5r8udPnmmvEClj/8Dr7/LXj8kZF9hW543Rnw528eHRDVav46Y7+vyM5yDrZshGwe5sxNZoCdNM7Bs7+Htathv4P8SyYy4Q+VAocJzIrAYQL9Q1U2biuzYVuZLeufJ/2HR5m35knc4ACbgyw99TRbgyyL630cVV7LwZUNpOm8n5dqKsO2ObuTqdfI1ivkamUy9Vps2SCVpp7vJujqJijM8a+ubtLlAbr/sHLc7+HmzodDjsS2bISNL8K2zZDOwD4vgf0O9EHM4r3BUiPZlHoNBvthcAAG+nwQlO/yAUyhy79Pp/056TSk0mO2YaOUc+ACCJx/T7h1DsqD4ffoh4F+/z2Duj8WBD5QmrcA5i3021QaBnqhv8/XKajDnPkwd74PxrI52LYFtm729zg0CIv2gr33hb329deo16FvK/Ru9dfIZHwwly/44K5e98FfpQyVCnR1w8LFzX0IBoG/l75t/vp9W/21Umn/vNMp/z7KUv55ds3xTW3ZHJQHoL/XvwYH/Lm5PORyvq5z5vl7mcl+P0Ednl8Lq5/ygcCiPWHPpbDXEl/XqGoFnvgNPPJL+M0v/c8YQCYLCxfBwt39h+Hhx8Gyo/39jvt9A3895/z9TTa4dc4//22bw8C77u+l8XMVNWceLNjN/wylJjH6PwjCn7/AX2OiOlYr8MJaWP+cz3DOX+h/nnZb5P+oePAe+OVP4fnIdP0HHgqvfgMsf7X/uRxPrerrPfZnKo5z4f+NjdC7zf9s9m3zP18uGPk/mU7D7nv63weL9/L/ds1cP6rxb9j4I6pa8T8HE93L5ChwmKrZHjg0Y7BSY2t/ha2DFfo3bSb1h9/hNqynPDBEebBMpTxEfahC2tVIB3Uyro4LAnqCPJvTBbakuqlbisOG1nNUeS0HVSfuzNkJ6hg/mXMoh1ReYP/q5nZXJ1lyef/LbCqyOf/LrxFcND6MqtWRX5K16vTWd0e6un1QVeiGfN4HFbm8/yAYGoLqUPjLuzo6A9YI0Br3kUr74KsRkNSq8NzTMFSO/75z5/tzGh82Q4P+GTQjnfF9kubv5j+4Gh9g5cH4Z5hOQyYXBk35kQAvmw8/NMMPznrNd8ze3DP5f+NUyj/HXH7keo1rW2rk/VDZf/j2b/MfjsP/DnP8s+vqBiz8WDMfRPa84P89pqKr268N1D3PB5Vd3T4g3fjiyB8AloK583z95y/0QVkqEsAP9Pk6bHxhaj/76bS/v8ar0BX+HEUCg1rN72tkSOP+8LnoM3DEcVN7DttT50iZuq5chq5chr1364YlC+GIA5s6L3CO3sEqW/qH2DZQoR44NgeOhwd6KaxfRb2vl/rgAMHAAEF5kL5MFxvzC+jJLWBDai618gBdW3uY07+JBQObcbUaAy7NkGWoWJqKZaha2r9IYUDe1Si4KoWgyp71Xg6u9PDSygb2qPfF1rFCipqlqVuKAKMrqJBj/F9AAfDTOYfyHwtewbrsQlIu4JT+J3jHll+yd713u7K75OxqUw0awP+C3LB++uoyHQYH/GtnBYHPKuxgFNSwvh1MKZ/v8tmZckzd6rXRzWk7Uq9DfdAHJzMlCGDrpqmf38ia7Yx8wWcannp05IN3cAB+/7uJz3OBDyZ6t8K61TtXhzj1ehjg7eQyAs0GltNgVgcOZpYC3g/8HXAAsAG4CbjUObeTP4UykZQZC7pzLOgem+rdA46cWvuic45KLWCoWqdaDwicCzPujmo9GH5VagGVWp1KNeDRWh3Xu4XU1k1U03kq6RyVbJ6KZag1zqv58waGalQGB3H9fdjgAPlamXxtkK5qmUy9ypNd+7A6sxtD1Tr5So3Apbh7/su4b94hHDuwmq56hRcy83gxM4+N6Tl0BVUOqG7koEoPB1V7WFAfxADDYfjOpv2pHP2pPP2Wo2IZulyFrqBKl6tScFXSLiCNI+Uaw2kdaYLh/b5hwgjCv8KC8A+FIPwu5VSGAcvTl8rTn8pRsbQvbxYGS1UWBIMsqA+yMBgkhaM3lac3VaAvlScgxfxgkPn1MguCQXKuzpZ0F5vSc9ic7qZCmn1qW9m3toWl1S10uyp1jN5Uga3pLnpTedIu8PcT1Mi7KjVLU7ZMGARmmBeUWVTvp9s1l00YsCxb011sTXWxNV2gbFnSONIuIEVAekwWNU1Ad1DxL1ehEFQZSOWH73MglSNFQD6okwuDzwVBmYX1gRlvmtuancNzc5awuWs3dhvayh6DG1lU3kzGbT+V/Mau3Xly8aH8Ye/DWL/4IMhkyNUrzCtvY+HgZl7S8wcOePEJFm/dcfBVS2cBSNdr2BTusZYrUJm7EJfJYemwmSiVwpmFSRKHCwIyg31k+7eSHZp88BV0zQFLYYN92ASZcWcGi/bC9tnPjzjbtgW3uccHaUNl3KFHkXrlyXD0K33w0LsF7r8Ld88PsRfXTlwJS00um9E1B3bfw2eV5s732645YTNN2ERZrfgMRc/zftu3tfnrR2Vz/pXLh01Orfs4n9VNFWb2L8CFwPeBHwKHARcA9wGnOjfxT4SaKmQy6oGjWqv74KZWJwhG/9+qBY5Ktc5QrU65Wmeo6stWw0CnWg+oB27UKwgczjkCB/UgGLWthcHTcPnw/Fo9GA6KBoZq9A1W6S1X6Sv7D+ZcJkU2nSKbSWEYtSDw59TD7+lGvuekOEe3q1C2LIFNPt/SHVRYVO8j7QLqlqJOijpGNcwwVcJsU6s6A5pzzA2GWBgMUAiqFJwPfApBjboZQ5ZlKBIA1SwV1jVNnRQ1SxGQom5GxgXMD8rMD8osqPsA7ensIjZl5m73fVMuYEEwiDkIzH+sB6ToTeWbuvfda30cNbSOtAvYmi6wLdXF1jBAGrIsVUv7D1uAMBDNuTp5VyMf3mfBVcm5OqlG4IojwNiYnsOGzFwGUvlJPcusq7GwPkjW1UmHAXAqvG7KBaRwpJ1jKJVhc6qLbekuapYefh5zggrzgjJdrjocRBiOimVYn1lAJZUhnTLSKQv/X4z+/pmUUchlyGdT1OrO/9+r1lha3czC+iALqLCACvOoMpApsCm3gJ7cfLbm5mM45lQHmFcbYH61nwJ18ilHPm3kzVHP5thUWMjG/G6UMwUC50b9UeMYaVkxjFQKsukUmfCVdzXy1SH/h0p1kFy94vuuNAKCXA6Xzvr+V6k0QSqNS2dJpVOkU0bKjJTByUcs5YA9503q32UCu2ZThZkdjg8SbnbOnRHZ/wzwJeBM4IY2VU9moXTKSOcyFGbJPFou/AVYq/tgpJHNGazUGBiqMVCpMVSp+1+SAA7qzgc7Phhx1IORYKjxNYCFv+zAb9MpIxX+EqwF/hd74wWEv2SNbNr/4u8fqtI76IOhSi0g2wiG0ilSRiTzFGafwkxV4/329xpu8b/wq2Hw12sFetMjHc4srEsjwGpW3dJsSGXZwI5/sQeWYnN66vOlbMrM5e7MIc0VNqNOmkFLM0gOZmgQUNUybMhM7UMtsBS96dH/DnEaP2dxaoGjr1ylb1SXEuO57O48l437pkAZKDea3jLA/PAVHm/8GFWAfoCB8LUzcuGroRK+duzQJQunM3CY0KwNHICz8P/Pvzhm/zXA54CzUeAgMi4zI21GOgX57K45rLQeBAxVA1IpH7SkfbSDcz6jU67WKVfqOOdIp1K+j1+Yrm80g1VqdeqBIxX+RZwK/9qvh1mhRoaIMMvTCF6if/I5GA7eGk1rzjEqOxSEQVvjA7QR+AXhX76jjofvG38d+/fb37+F/RDNDBcGkY3sWLUWUA2zbNUwY5VJGdnMyF/TmfCe06mU75JRqdM/VKWvXKM/zIA1sl/ZTArnGBXoNQKBRmZ8+K/3UOBc+IyDUQ0uKfPXTaWMoWowbpCXMptUAJhkqVTrhuXO5sBhOT4mfCC60zlXNrMV4XERkXGlUym689s3u5gZuUyaXCbN/AlGPkpruLDJrhY4smkfqESPVesB5YpvJsymU+SzaXKZFOlUKsyqjQRk9WB0MBUNnsx8wNdoZixXRzdJNsqmUhZm1fw5LkzJNYK9aBavVg/8tWE4KqrVRwKyRpA4fD9sHwgGzrHfou2bvWbKbA4clgA9zrm4bt5rgT8xs5xzbrs8kJmdB5z3vve9b6brKCIiO8nMyKSNTExiLBrkxUnt4LhsbzaPGOsGxhsbVo6U2Y5z7uvOuWkbECsiIjJbzObAYQAYr+tvIVJGREREmjSbA4d1wGIziwseluKbMVo3Y4aIiMgsMJsDhwfx9/eK6E4zKwDHAA+1o1IiIiKdbDYHDjfiO6BeNGb/ufi+Dde3vEYiIiIdbtaOqnDOrTSzq4Hzzexm4Hb8zJEXAvegORxEREQmbdYGDqGLgFXAecBpQA/wZfxaFVNcTk1ERGTXNasDB+dcHbgifImIiMhOms19HERERGSaKXAQERGRpilwEBERkabN6j4O06VYLLa7CiIiIq3iSqXSuMttKuMgIiIiTTM3S9Yi7xRm9pAW0Np5eo7TQ89xeug5Tg89x+kx089RGQcRERFpmgIHERERaZoCh9b7ersrMEvoOU4PPcfpoec4PfQcp8eMPkf1cRAREZGmKeMgIiIiTVPgICIiIk1T4DDDzCxlZheb2eNmVjaz58zsCjOb0+66JZGZHWJmnzKzX5jZBjPrNbMVZvbRuGdmZoea2S1mttnM+s3sPjM7pR11TzIz6zazZ8zMmdm/xhzXc5yAme1uZv9sZr8P/x9vMLOfmtmfjSmn5zgOM5trZh8xs5Xh/+seM7vfzN5lZjam7C7/HM3sEjP7jpk9Hf6/XbWD8k0/s539XNLMkTPvKuBC4Pv4VToPC79+uZmdquW9t/O3wN8DPwCuB6rAycBngLea2fHOuUEAMzsYuB+oAZ8HtgLnAnea2V845+5qQ/2T6lPA4rgDeo4TM7P9gbuBucC/AU8CC4CjgKWRcnqO4zCzFPBD4E+AbwFfBrqBs4Bv4H8vfigsq+fofRbYBDwMLJyo4BSe2c59Ljnn9JqhF3A4EADfG7P/AsABb2t3HZP2Ao4DFsTs/0z4zM6P7LsJqAPHRPbNBVYDTxB2/t3VX8Afh79Q/iF8hv865rie48TP7z7gOWCfHZTTcxz/2ZwQ/uxdNWZ/Dnga2KLnuN0zOyjy/rfAqgnKNv3MpuNzSU0VM+sswIAvjtl/DTAAnN3yGiWcc+4h59zWmEM3htsjAMKU2puAu51zKyLn9wHXAocAy2e4uolnZmn8z9sdwM0xx/UcJ2BmJwJ/CnzeObfezLJm1h1TTs9xYvPD7broTudcBegB+kHPMco593Qz5abwzHb6c0mBw8xajo/sHojudM6VgRXsIv8Bpsm+4faFcHsUkAd+HlP2F+FWzxcuBpYB549zXM9xYm8It8+a2a3AINBvZk+aWfQXrJ7jxB4AtgAfNLO/NrOXhG3ylwPHApeF5fQcJ2+yz2ynP5cUOMysJUCPc24o5thaYLGZ5Vpcp44T/tV8KT7dfkO4e0m4XRtzSmPf0phjuwwzOxD4JPAp59yqcYrpOU7s0HB7DbA78E7gPUAF+Hcze3d4XM9xAs65zfi/ijfh0+qrgcfx/ZnOcM5dExbVc5y8yT6znf5cUufImdUNxP3jAJQjZSqtqU7H+iJwPPAR59wT4b5Gujju+ZbHlNlVfQV4BrhygjJ6jhObF257gZPD1Dpm9n182/xnzexb6Dk2ow/fVv8DfEe+3fGBww1mdrpz7sfoOU7FZJ/ZTn8uKXCYWQPAnuMcK0TKyDjM7NP4NPvXnXOXRw41nls+5rRd/tmGafTXAic656oTFNVznNhguP12I2gA/xe0mf0AZ+higQAABm9JREFU+J/4rISe4wTM7Eh8sHCxc+6rkf3fxgcT14QjA/QcJ2+yz2ynP5fUVDGz1uHTPnH/oEvx6SJlG8ZhZpcBH8MP13rvmMONTlZxacvGvrjU3awX/rxdCdwOPG9mLzWzlwL7h0UWhPsWoue4I2vC7fMxx9aH293Qc9yRi/EfSt+J7nTODQC34X82D0DPcSom+8x2+nNJgcPMehD/jF8R3WlmBeAY4KF2VKoTmNkngE8A1wHnuHC8UMRKfLrthJjTjw+3u+rz7QL2AE4Dnoq87g6Pnx1+fQ56jjvy/9u791ApqjiA499JM4LIDCStBCF6GL2MTEjJHhJGSSWGVFpEFHosKCQJKbzRPwaVYXkygpDykfaHSYJlIiZaVGZGRWX/2ENLi0jpoVFNf5xZG7fde2e9Xq96vx8Yhj17dpz5se787pnzqHUgO73Be7WynRjHjtRuYL0avNe7tDeOrWs1Zp2/L3X3WNWjeQPOp/3xshO7+xwPx43UETInJQ3HtFPvVdLY5QtLZbWxy1voIeO9G8TlWGB8g21KEdeVxeuzjGOHsewH7Ca1PJxQKh9Iema/pVRmHJvHcXbx3ZteV15r9foZ6G0cm8avo3kcKsfsYNyXXB2zi2VZ9gzpGf0yUtNxbYauDcBVuTNH7ifLsqnAs8A3wCOkL3jZjjx1oqJofn+fNLvkbNIP/N2k/xjX5Xn+5qE67yNBlmWDSZ0l5+Z5fm+p3Di2I8uye4Dngc+AF0mTFk0hJQ/X53m+qqhnHJsoZt/cRErEFpJ+/04mxWcwMDXP81jUNY5AlmWT+O/x4n2k792Txeuv8zx/uVS3pZh1+r7U3ZnU0b6RmuamkWbv2kt61vQUpb9e3PaL13xS1ttsW1tXfwiwnDRG/HdgPTC6u6/jcNxIP9D/mznSOFaK3TjSmPjfSCMsVgEjjGNLMTyDNN30d8UNbjewDhhnHBvGa23V38FWY9bZ+5ItDpIkqTI7R0qSpMpMHCRJUmUmDpIkqTITB0mSVJmJgyRJqszEQZIkVWbiIEmSKnN1TElHvRBCG2ntkytjjGu792ykI5uJg6QOhRCqzBTnTVnqAUwcJLXi0Xbe23qoTkJS9zFxkFRZjLGtu89BUvcycZB00JX7FJBW+LsfOIe0QNQKYEaM8YcGnzuTtCrq1UB/4CdgNfBYjPGrBvV7kVYBnAScR1pBcBtpgaDHm3xmPDC9qL+HtGDVtBjjts5cs9RTOKpCUld6AJgHfAw8TVqN707gnRBC/3LFEMIwYCMwEfgAeIK0IuVtwMYQwiV19fsAbwDPAYOARcAc4EPgJmBEg/MJwALSY5W5wKfABGB1COG4Tl+t1APY4iCpsqIloZE9McZZDcqvBYbHGD8qHWM2qQViFnBXUZYBLwEnAhNjjAtL9ScArwALQgjnxhj/Kd5qA0YDrwM3xxj3lj5zXHGsemOAYTHGT0p1FwG3ADcAS5tevCTAFgdJrZnZZHuoSf2Xy0lDoQ3YBdxa+iv/MtKjjHfLSQNAjHEJsB44GxgJ+x5RBOAPYHI5aSg+szfG+GOD85lTThoKLxT7S5tcg6QSWxwkVRZjzFr8yNsNjrErhLAZGAUMATYDFxdvr2lynDWkpGEosI6UZPQF3osxbm/hfDY2KPu22Pdr4ThSj2WLg6SutKNJea1jZN+6/fdN6tfKT6rbt9qh8ZcGZX8V+14tHkvqkUwcJHWlU5qUDyj2u+r2AxrUBRhYV6+WAJx24Kcm6UCYOEjqSqPqC0IIfYGLSEMhPy+Ka/0grmhynFr5pmL/BSl5uCCEcOrBOFFJ1Zg4SOpKk0IIQ+vK2kiPJhaXOjVuIA3VHFnMs7BP8fpyYAupkyQxxr+BCBwPzKsfShlC6FM/3FPSwWHnSEmVtTMcE+C1GOPmurKVwIYQwlJSP4WRxbaV0kiMGGMeQrgDeAtYEkJYTmpVOBu4kTRx1O2loZiQpr8eDowFtoQQVhT1BgHXAA8C8w/oQiU1ZeIgqRUz23lvK2mERNlsYBlp3oYJwK+km/mMGOPOcsUY43vFJFAPk+ZnGEuaOXIxaebIL+vq/xlCGANMBm4H7gAyYHvxb65v/fIkdSTL8yqL3klSdS5jLR297OMgSZIqM3GQJEmVmThIkqTK7OMgSZIqs8VBkiRVZuIgSZIqM3GQJEmVmThIkqTKTBwkSVJlJg6SJKmyfwHJz2rUhv6dcwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ooo.plot_history(history, plot={'MSE' :['mse', 'val_mse'],\n",
    "                                'MAE' :['mae', 'val_mae'],\n",
    "                                'LOSS':['loss','val_loss']})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 7 - Make a prediction\n",
    "The data must be normalized with the parameters (mean, std) previously used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "my_data = [ 1.26425925, -0.48522739,  1.0436489 , -0.23112788,  1.37120745,\n",
    "       -2.14308942,  1.13489104, -1.06802005,  1.71189006,  1.57042287,\n",
    "        0.77859951,  0.14769795,  2.7585581 ]\n",
    "real_price = 10.4\n",
    "\n",
    "my_data=np.array(my_data).reshape(1,13)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction : 9.31 K$\n",
      "Reality    : 10.40 K$\n"
     ]
    }
   ],
   "source": [
    "\n",
    "predictions = model.predict( my_data )\n",
    "print(\"Prediction : {:.2f} K$\".format(predictions[0][0]))\n",
    "print(\"Reality    : {:.2f} K$\".format(real_price))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "<img width=\"80px\" src=\"../fidle/img/00-Fidle-logo-01.svg\"></img>"
   ]
  }
 ],
 "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}