Stock_Price_Prediction_and_Forecast.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Stock Price Prediction and Forecast.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "gRe4ePVB1eYS"
      },
      "source": [
        "#Import required libraries\n",
        "\n",
        "import pandas_datareader as pdr\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Dense, LSTM\n",
        "\n",
        "from tensorflow.keras.callbacks import  EarlyStopping\n",
        "from tensorflow.keras import optimizers\n",
        "\n",
        "import tensorflow as tf\n",
        "\n",
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Q60JqsLU2XRP"
      },
      "source": [
        "#Get Apple stock ptrice data from Tiingo\n",
        "#Tiingo - https://api.tiingo.com/\n",
        "\n",
        "#key = \"Generate Key from Tiingo\"\n",
        "df = pdr.get_data_tiingo('GOOGL',api_key=key)"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OcTYtAK12_2z"
      },
      "source": [
        "#Save dataset as csv\n",
        "df.to_csv('GOOGL.csv')"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 200
        },
        "id": "hExXWERU3wxw",
        "outputId": "533d3a60-57cb-4c67-fcf9-d404e99ddfbf"
      },
      "source": [
        "#read the saved csv file\n",
        "df = pd.read_csv('GOOGL.csv')\n",
        "df.head()"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>symbol</th>\n",
              "      <th>date</th>\n",
              "      <th>close</th>\n",
              "      <th>high</th>\n",
              "      <th>low</th>\n",
              "      <th>open</th>\n",
              "      <th>volume</th>\n",
              "      <th>adjClose</th>\n",
              "      <th>adjHigh</th>\n",
              "      <th>adjLow</th>\n",
              "      <th>adjOpen</th>\n",
              "      <th>adjVolume</th>\n",
              "      <th>divCash</th>\n",
              "      <th>splitFactor</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>GOOGL</td>\n",
              "      <td>2016-10-19 00:00:00+00:00</td>\n",
              "      <td>827.09</td>\n",
              "      <td>829.810</td>\n",
              "      <td>823.2050</td>\n",
              "      <td>824.52</td>\n",
              "      <td>1500142</td>\n",
              "      <td>827.09</td>\n",
              "      <td>829.810</td>\n",
              "      <td>823.2050</td>\n",
              "      <td>824.52</td>\n",
              "      <td>1500142</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>GOOGL</td>\n",
              "      <td>2016-10-20 00:00:00+00:00</td>\n",
              "      <td>821.63</td>\n",
              "      <td>828.465</td>\n",
              "      <td>820.5500</td>\n",
              "      <td>827.56</td>\n",
              "      <td>1393870</td>\n",
              "      <td>821.63</td>\n",
              "      <td>828.465</td>\n",
              "      <td>820.5500</td>\n",
              "      <td>827.56</td>\n",
              "      <td>1393870</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>GOOGL</td>\n",
              "      <td>2016-10-21 00:00:00+00:00</td>\n",
              "      <td>824.06</td>\n",
              "      <td>824.290</td>\n",
              "      <td>818.3100</td>\n",
              "      <td>820.00</td>\n",
              "      <td>1615814</td>\n",
              "      <td>824.06</td>\n",
              "      <td>824.290</td>\n",
              "      <td>818.3100</td>\n",
              "      <td>820.00</td>\n",
              "      <td>1615814</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>GOOGL</td>\n",
              "      <td>2016-10-24 00:00:00+00:00</td>\n",
              "      <td>835.74</td>\n",
              "      <td>837.940</td>\n",
              "      <td>829.0400</td>\n",
              "      <td>830.09</td>\n",
              "      <td>1447616</td>\n",
              "      <td>835.74</td>\n",
              "      <td>837.940</td>\n",
              "      <td>829.0400</td>\n",
              "      <td>830.09</td>\n",
              "      <td>1447616</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>GOOGL</td>\n",
              "      <td>2016-10-25 00:00:00+00:00</td>\n",
              "      <td>828.55</td>\n",
              "      <td>838.500</td>\n",
              "      <td>825.3005</td>\n",
              "      <td>838.50</td>\n",
              "      <td>1890712</td>\n",
              "      <td>828.55</td>\n",
              "      <td>838.500</td>\n",
              "      <td>825.3005</td>\n",
              "      <td>838.50</td>\n",
              "      <td>1890712</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  symbol                       date   close  ...  adjVolume  divCash  splitFactor\n",
              "0  GOOGL  2016-10-19 00:00:00+00:00  827.09  ...    1500142      0.0          1.0\n",
              "1  GOOGL  2016-10-20 00:00:00+00:00  821.63  ...    1393870      0.0          1.0\n",
              "2  GOOGL  2016-10-21 00:00:00+00:00  824.06  ...    1615814      0.0          1.0\n",
              "3  GOOGL  2016-10-24 00:00:00+00:00  835.74  ...    1447616      0.0          1.0\n",
              "4  GOOGL  2016-10-25 00:00:00+00:00  828.55  ...    1890712      0.0          1.0\n",
              "\n",
              "[5 rows x 14 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "id": "KvP6S801368G",
        "outputId": "9cbda3d3-215b-4644-acf6-b258702d7f8f"
      },
      "source": [
        "df = df.reset_index()['close']\n",
        "plt.plot(df)"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7fae0b001f10>]"
            ]
          },
          "metadata": {},
          "execution_count": 5
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MAM2B22N4ttu",
        "outputId": "19908ccb-1da0-474e-f54c-30b694745c4f"
      },
      "source": [
        "### LSTM are sensitive to the scale of the data. So we apply MinMax Scaler\n",
        "scaler = MinMaxScaler()\n",
        "df = scaler.fit_transform(np.array(df).reshape(-1,1))\n",
        "df"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.03434073],\n",
              "       [0.03180248],\n",
              "       [0.03293214],\n",
              "       ...,\n",
              "       [0.92902668],\n",
              "       [0.96220986],\n",
              "       [0.96422744]])"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5ad6K0EI7x-K",
        "outputId": "7e564f45-3448-453c-c0eb-36fcea90298e"
      },
      "source": [
        "### Splitting data into Train and Test set\n",
        "training_size = int(len(df)*0.7)\n",
        "test_size =  len(df) - training_size\n",
        "\n",
        "train_data = df[0:training_size,:]\n",
        "test_data = df[training_size:len(df),:]\n",
        "\n",
        "print(\"Train Data Size\",training_size)\n",
        "print(\"Test Data size\",test_size)"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Data Size 879\n",
            "Test Data size 378\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HSPCyZjZ9vN8"
      },
      "source": [
        "#Create Train and test data set\n",
        "\n",
        "def create_dataset(dataset,time_step):\n",
        "  dataX,dataY = [], []\n",
        "  for i in range(len(dataset) - time_step - 1):\n",
        "    a = dataset[i:(i + time_step),0]\n",
        "    dataX.append(a)\n",
        "    dataY.append(dataset[(i+time_step),0])\n",
        "  return np.array(dataX),np.array(dataY)\n",
        "\n",
        "\n",
        "#reshape into X = t,t+1,t+2,....,t+149 and y = t+150\n",
        "time_step = 150\n",
        "X_train, y_train = create_dataset(train_data,time_step)\n",
        "X_test,  y_test  = create_dataset(test_data,time_step)\n"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9_7tqTvUCOJB"
      },
      "source": [
        "#reshape input to be [samples, time steps, features] which is required for LSTM\n",
        "X_train = X_train.reshape(X_train.shape[0],X_train.shape[1],1)\n",
        "X_test  = X_test.reshape(X_test.shape[0],X_test.shape[1],1)"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YXbcJxDduUlm"
      },
      "source": [
        "#Create Stacked LSTM Model\n",
        "\n",
        "#return_sequence=True : If We want to have a sequence for the output, not just a single vector as we did with normal Neural Networks,\n",
        "#so it’s necessary that we set the return_sequences to True. Concretely, let’s say we have an input with shape (num_seq, seq_len, num_feature).\n",
        "#If we don’t set return_sequences=True, our output will have the shape (num_seq, num_feature), but if we do, we will obtain the output with shape (num_seq, seq_len, num_feature).\n",
        "\n",
        "model = Sequential()\n",
        "\n",
        "model.add(LSTM(50,return_sequences=True,input_shape=(150,1)))\n",
        "model.add(LSTM(50,return_sequences=True))\n",
        "model.add(LSTM(50))\n",
        "model.add(Dense(1))"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AczzYpyRxrcT",
        "outputId": "f141eefa-202c-45aa-e8cf-f68f447aff49"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "lstm (LSTM)                  (None, 150, 50)           10400     \n",
            "_________________________________________________________________\n",
            "lstm_1 (LSTM)                (None, 150, 50)           20200     \n",
            "_________________________________________________________________\n",
            "lstm_2 (LSTM)                (None, 50)                20200     \n",
            "_________________________________________________________________\n",
            "dense (Dense)                (None, 1)                 51        \n",
            "=================================================================\n",
            "Total params: 50,851\n",
            "Trainable params: 50,851\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8g34ssd0xxkM"
      },
      "source": [
        "optimizer = optimizers.Adam(learning_rate=0.01)\n",
        "\n",
        "#Compile the Model\n",
        "model.compile(loss='mean_squared_error',optimizer=optimizer)\n"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ox4oDhivzM_u",
        "outputId": "4237a9f8-208f-40d6-d945-ed35fd50072b"
      },
      "source": [
        "#Stop training when a monitored metric has stopped improving.\n",
        "earlystop = EarlyStopping(monitor=\"val_loss\",patience=10,mode=\"auto\",verbose=1)\n",
        "\n",
        "history = model.fit(X_train,y_train,validation_data=(X_test,y_test),epochs=100,batch_size=64,verbose=1,callbacks=[earlystop])"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "12/12 [==============================] - 9s 389ms/step - loss: 0.0096 - val_loss: 0.1810\n",
            "Epoch 2/100\n",
            "12/12 [==============================] - 3s 275ms/step - loss: 0.0021 - val_loss: 0.0644\n",
            "Epoch 3/100\n",
            "12/12 [==============================] - 3s 272ms/step - loss: 6.3818e-04 - val_loss: 0.0519\n",
            "Epoch 4/100\n",
            "12/12 [==============================] - 3s 274ms/step - loss: 4.4688e-04 - val_loss: 0.0100\n",
            "Epoch 5/100\n",
            "12/12 [==============================] - 3s 272ms/step - loss: 3.8179e-04 - val_loss: 0.0193\n",
            "Epoch 6/100\n",
            "12/12 [==============================] - 3s 270ms/step - loss: 3.6201e-04 - val_loss: 0.0121\n",
            "Epoch 7/100\n",
            "12/12 [==============================] - 3s 272ms/step - loss: 3.3480e-04 - val_loss: 0.0117\n",
            "Epoch 8/100\n",
            "12/12 [==============================] - 3s 273ms/step - loss: 2.9703e-04 - val_loss: 0.0087\n",
            "Epoch 9/100\n",
            "12/12 [==============================] - 3s 272ms/step - loss: 2.6431e-04 - val_loss: 0.0130\n",
            "Epoch 10/100\n",
            "12/12 [==============================] - 3s 275ms/step - loss: 2.8521e-04 - val_loss: 0.0124\n",
            "Epoch 11/100\n",
            "12/12 [==============================] - 3s 271ms/step - loss: 2.5541e-04 - val_loss: 0.0051\n",
            "Epoch 12/100\n",
            "12/12 [==============================] - 3s 271ms/step - loss: 3.2134e-04 - val_loss: 0.0063\n",
            "Epoch 13/100\n",
            "12/12 [==============================] - 3s 275ms/step - loss: 2.5390e-04 - val_loss: 0.0032\n",
            "Epoch 14/100\n",
            "12/12 [==============================] - 3s 275ms/step - loss: 2.4431e-04 - val_loss: 0.0078\n",
            "Epoch 15/100\n",
            "12/12 [==============================] - 3s 273ms/step - loss: 2.6585e-04 - val_loss: 0.0120\n",
            "Epoch 16/100\n",
            "12/12 [==============================] - 3s 274ms/step - loss: 3.7103e-04 - val_loss: 0.0013\n",
            "Epoch 17/100\n",
            "12/12 [==============================] - 3s 278ms/step - loss: 2.5272e-04 - val_loss: 9.3273e-04\n",
            "Epoch 18/100\n",
            "12/12 [==============================] - 3s 276ms/step - loss: 2.1224e-04 - val_loss: 0.0053\n",
            "Epoch 19/100\n",
            "12/12 [==============================] - 3s 271ms/step - loss: 1.8935e-04 - val_loss: 0.0043\n",
            "Epoch 20/100\n",
            "12/12 [==============================] - 3s 271ms/step - loss: 1.9000e-04 - val_loss: 0.0052\n",
            "Epoch 21/100\n",
            "12/12 [==============================] - 3s 273ms/step - loss: 1.8821e-04 - val_loss: 6.3400e-04\n",
            "Epoch 22/100\n",
            "12/12 [==============================] - 3s 273ms/step - loss: 1.9246e-04 - val_loss: 5.4395e-04\n",
            "Epoch 23/100\n",
            "12/12 [==============================] - 3s 277ms/step - loss: 2.0120e-04 - val_loss: 0.0075\n",
            "Epoch 24/100\n",
            "12/12 [==============================] - 3s 276ms/step - loss: 2.5019e-04 - val_loss: 0.0054\n",
            "Epoch 25/100\n",
            "12/12 [==============================] - 3s 275ms/step - loss: 2.2412e-04 - val_loss: 0.0030\n",
            "Epoch 26/100\n",
            "12/12 [==============================] - 3s 274ms/step - loss: 1.5870e-04 - val_loss: 0.0034\n",
            "Epoch 27/100\n",
            "12/12 [==============================] - 3s 272ms/step - loss: 1.6178e-04 - val_loss: 0.0015\n",
            "Epoch 28/100\n",
            "12/12 [==============================] - 3s 275ms/step - loss: 1.9278e-04 - val_loss: 0.0025\n",
            "Epoch 29/100\n",
            "12/12 [==============================] - 3s 272ms/step - loss: 2.1601e-04 - val_loss: 0.0042\n",
            "Epoch 30/100\n",
            "12/12 [==============================] - 3s 271ms/step - loss: 2.0310e-04 - val_loss: 0.0011\n",
            "Epoch 31/100\n",
            "12/12 [==============================] - 3s 273ms/step - loss: 1.8602e-04 - val_loss: 0.0018\n",
            "Epoch 32/100\n",
            "12/12 [==============================] - 3s 274ms/step - loss: 1.5229e-04 - val_loss: 7.4360e-04\n",
            "Epoch 00032: early stopping\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 621
        },
        "id": "J4yh_zcF0Uwa",
        "outputId": "3322b05d-8dba-4760-84c8-7910385fb42e"
      },
      "source": [
        "#Plot Model Loss\n",
        "plt.figure(figsize=(10, 10))\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.title('model loss')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.legend(['train', 'val'], loc='upper left')\n",
        "plt.show()"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RlScWGwy13kr"
      },
      "source": [
        "#Prediction\n",
        "train_predict = model.predict(X_train)\n",
        "test_predict =  model.predict(X_test)"
      ],
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rW21I9Vs2wP_"
      },
      "source": [
        "#Transform to original form\n",
        "train_predict = scaler.inverse_transform(train_predict)\n",
        "test_predict  = scaler.inverse_transform(test_predict)"
      ],
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 592
        },
        "id": "6covPAen3ms4",
        "outputId": "5d36ad85-a540-4390-edd2-f6429ee51652"
      },
      "source": [
        "#Plotting\n",
        "\n",
        "#Shift train prediction for plotting\n",
        "look_back = 150\n",
        "trainPredictPlot = np.empty_like(df)\n",
        "trainPredictPlot[:,:] = np.nan\n",
        "trainPredictPlot[look_back:len(train_predict)+look_back, :] = train_predict\n",
        "\n",
        "#Shift test prediction for plotting\n",
        "testPredictPlot = np.empty_like(df)\n",
        "testPredictPlot[:,:] = np.nan\n",
        "testPredictPlot[len(train_predict)+(look_back*2)+1:len(df)-1, :] = test_predict\n",
        "\n",
        "plt.figure(figsize=(10, 10))\n",
        "plt.plot(scaler.inverse_transform(df))\n",
        "plt.plot(trainPredictPlot)\n",
        "plt.plot(testPredictPlot)\n",
        "plt.show()"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlwAAAI/CAYAAACifAdEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeZwcV33v/e+p6m2mZx+NdlmS5UWWjDFgjMGEODiYNRgID5jVXLjxJQFCQsh9IA9cEm4IhATIzZOAMQFiiIkxqw2Y2GY1YDAWljdZxpZlWbs00uzTa1Wd+0f19DJrjzQ93T3zeb9efrnqVHXPGaNX9M3v/OqUsdYKAAAAtePUewIAAABLHYELAACgxghcAAAANUbgAgAAqDECFwAAQI0RuAAAAGosMtcNxpiEpDslxQv3f91a+yFjzGZJN0rqlfQbSW+y1uaMMXFJX5L0DEknJb3WWruv8F3vl/Q2Sb6kP7XW3jbbz16xYoXdtGnTKf5qAAAAi+c3v/nNCWtt33TX5gxckrKSnm+tHTPGRCX93BjzfUnvkfQpa+2NxphrFQapzxT+PWitPcsYc5Wkv5f0WmPMNklXSdouaa2kHxhjzrHW+jP94E2bNmnHjh3z+FUBAADqwxjz5EzX5lxStKGxwmm08I+V9HxJXy+MXy/pFYXjKwvnKly/3BhjCuM3Wmuz1tonJO2RdPE8fxcAAICmU1UPlzHGNcbcJ+m4pDskPS5pyFrrFW45KGld4XidpAOSVLg+rHDZsTg+zWcAAACWrKoCl7XWt9ZeKGm9wqrU1lpNyBhzjTFmhzFmR39/f61+DAAAwKKZ11OK1tohST+W9GxJXcaYiR6w9ZIOFY4PSdogSYXrnQqb54vj03ym/GdcZ629yFp7UV/ftH1nAAAATWXOwGWM6TPGdBWOWyS9QNJuhcHr1YXbrpZ0c+H4lsK5Ctd/ZMM3ZN8i6SpjTLzwhOPZkn69UL8IAABAo6rmKcU1kq43xrgKA9pN1trvGmMelnSjMeZvJe2U9PnC/Z+X9GVjzB5JAwqfTJS1dpcx5iZJD0vyJL1jticUAQAAlgoTFp8a00UXXWTZFgIAADQDY8xvrLUXTXeNneYBAABqjMAFAABQYwQuAACAGiNwAQAA1BiBCwAAoMYIXAAAADVG4AIAAKgxAhcAAECNEbgAAABqjMAFAABQYwQuAACAGiNwAQAA1BiBCwAAoMYIXAAAADVG4AIAAKgxAhcAAGgq195/rW7YfUO9pzEvkXpPAAAAoFp539e/3vevkqQ3nPeGOs+melS4AABA0/jcr35V7ymcEgIXAABoGg8de7x47AVeHWcyPwQuAADQNHYeLQWuk+mTdZzJ/BC4AABAUxgcz2kge7R4fmDkiL74iyf0jM++Qe/5yXvqOLO50TQPAACawhMnx2UiY8Xz/337nbp/9xa1n/eA7niyjhOrAhUuAADQFPadGJdxxxXkeiVJe/U5RXvurPOsqkPgAgAATWEknQ8DV3ZlcSzRfU8dZ1Q9AhcAAGgKGS+QcVMK/KTG9rxX3tg5stHmaJwncAEAgKaQznky7risl5TNr5D12mRMULye9/N1nN3sCFwAAKApjOSHZBxfLz7vHL378rNlg0TF9bH82AyfrD8CFwAAaAqD2X5J0su3b9fVz9k0NXDlCFwAAACn5WRuvyRpdXK14hFH1q8MXGk/XY9pVYXABQAAmsKTuZ9I1mhDxwbFI44UxCuu5/ycrLX1mdwcCFwAAKApjHsnFc9doI5YhyKuI+snK65/58F9uuBLT9W1919bpxnOjMAFAAAa3v0HhpQJUhpNlZYRg9yKinu+eO/tkqy+tOtLizy7uRG4AABAw3v4yIiMk61YRpzYcX6CEwv35Dqz68xFnVs1CFwAAKDhpfM5GScvW963ZaPKHH2ZOsfeKEkybviUYsyJVXw2k/fr3ttF4AIAAA3v4PCgJKklUtm3lR98rjp0tiTJuOOSJMdEKu758Hcf1qs+c5eCoH6hi8AFAAAa3tGREUnS+664YMq1eCTs6zKRsMLl+aVrWc/XzTsP6cwVbXIcU/uJzoDABQAAGt5gJgxcva2dU64louEyoxMJK1wZL1u8NprxNJ7zdeGGqZ9bTJG5bwEAAKiv4eyYlJBaIi3Fsf/5onNlZLTjycMV9+aD0jsV8374rsWoW98aExUuAADQ8Eay4S7yiUhpW4g/uews/fFlW9Qaq9xxvjxw5TwCFwAAQFXGCoEr7sanXGuJxGRtqT8rH+RKx4UKVyxC4AIAAJiRtVapfEbSDIEr5kq21CU1UeF6cuRJXfPjV8lJHKDCBQAAMJusFyhQGKKmC1yJqCsbRIvnExWumx64S/2Zo0qsvkWxSP2eUJQIXAAAoMFl8r6MEwau8h6uCYmoW7EDvVeocH32Z48VRgLFXLfm85wNgQsAADS0rBdIZq4K19TAZZyJ7SGMoi4VLgAAgBll8r5kPEkzBS5HCkqv8/FsXsOpfGXgqnPTPPtwAQCAhpbJB8UlxekCV8R1JIUVLBtE5CmvoXROKgauQDGa5gEAAGaW9cIKl2sicp2pvVjHhjOSwvckWj+pQL6G07lShcvJsy0EAADAbDL5QMbkFXVi015/07M3qi0ZvkcxyK6SJA2l08XAZZwc20IAAABMx/MDvfj//Ezff+iI5HiKOVOXEyVpVUdCna1hGPPTGyRJw+l0aUnR5KhwAQAATOfJgZR2HxnRF3+xT8bkFZumf2vCtb9/rf7swvfL+m2SpKFsWsYJ9+MKK1w8pQgAADDF3v5xmehJuS37wgqXO/2SoiRt6dqiV571asmGPV4jFUuKnlzHLsaUZ8RTigAAoCE93j+m5JZPyJhA3uh5SrhTNz0t5zpGNgijzUg2U/aUouTZjKSWWk53VlS4AABAQ9pzfEzGhC+fNrGTSkRmXlKUpIhjiu9UHM1mZNxs8aXWaX+0tpOdA4ELAAA0pMf7x4rHbvz4nIHLdUxxSXEsm5FxsrL5bknSaG64dhOtAoELAAA0nG/ee1A7D/RXjCWjsy8JRl1HtlDhGsuFgSvId0mShrJDtZlolQhcAACgoQyn8nrPTffLiZ2sGJ/uxdXlHKNihWs0Ny6ZoFjhInABAACUOTmelYmeVPLMf6oYn+0pRUkyxsg1UUnSzkOHJElBIXCN5EZqMNPqEbgAAEBDuePhY4q075oyPtdTipLkmnBJ0bjjkqQg36Uz/T/V5WdcvrCTnCcCFwAAqLuTY1n9+Vfv03Aqr49+/xE58eOSpBd2fkLWD3u35qpwSZKr8J6JwPXp1z1H33rLf9fq5Ooazbw67MMFAADq7nsPHtG3dh5SayzswXJj/XpKz9N18Zrz9P3+hIybrqrC5ZiIrCQTSUmS2mJJOU59d5mXqHABAIAGsLI9DFM33L1fkmRiJ7Spa6M6ElHZIKxwxefYFkKSou5EhSsMXMloshbTnTcCFwAAqDsvCGTcUUXadsm4Y3Ii4zq7e3P4DsTCk4cdsY45v6ctFoayiSXF1khr7SY9DywpAgCAusv7geJrvqVo+8PK9v++JOni1RdrU3uvena4Gg5UVR9WR7xFAyoFLipcAAAABXnfypi8JCnW+1NFTEzn9Z6nZDyilZ3hVg99LX1zfk97IlyaZEkRAABgEs+3xV3ijZPXxo4NckwYU7Z2b5UkrWhZMef3dMYLgStSWFKMsqQIAAAgKVxSdCKldyduaN9QPP7AJR/QlWddqfXt6+f8ns6WuOyokTG+4k5SUSdak/nOFxUuAABQd3k/kImMFs+v2npV8bg12qpnrXlWVd/TnogWm+yTbtfCTvI0UOECAAB199ChYRl3VK8869V659P+WCtbV57S93S2RCUbkeSpNdI4gYsKFwAAqKsDAyl9+4E9Mo6vMzs3n3LYkqSVHYliL1gy0r1QUzxtBC4AAFBXR4YzcgrLiStb526Mn83K9tLmqF3RUw9uC43ABQAA6mo858lpCXeY72ude+uH2VxyZm+x+f6Ksy467bktFAIXAACoq6FUTpH23Qq8pJ7a99TT+q5E1C0eX7j63NOd2oIhcAEAgLo6OZaTMXnZfI9ihXchLoSeRM+CfdfpInABAIC6Gsl4kvFlrTv3zfPQGe9c0O87HQQuAABQV54fyBhPChZ2t6pG2fRUInABAIA68wIrGb+wf9bStHR/MwAA0BTyfiAZb8GWFK+54BodGz+2IN+1UAhcAACgrjzfyixghetdT3vXgnzPQmJJEQAA1JUXBJLjqa+ttd5TqRkCFwAAqCvPt3KdQJdvXVvvqdQMgQsAANRV2DTvLegeXI2GwAUAAOoq7weyxlPMIXABAADUhOdbSZ6ibuPsm7XQCFwAAKCu8oEnmYAKFwAAQK3kfE+SqHABAADUSj7ISmqsV/EsNAIXAACoq5yflySeUgQAAKiVfBAGLipcAAAANeIHVLgAAABqaqLCxVOKAAAANeJZlhQBAABqypvo4WJbCAAAgNrwbE4SPVwAAAA14wWFjU9ZUgQAAFg4x0cyeu/X7lcm7yvjhRufLuWm+Ui9JwAAAJafv7t1t75932E9a3OP0l5OLWJJEQAAYEGN53xJ0snxnGTCY5YUAQAAFlDWCyRJT55MSYaXVwMAACy4nBdWtR7vH5MpBK6l3MNF4AIAAIvu6HBGknRgIFVaUqTCBQAAsDAGx3PadzIlSToynCkuKS7lChdPKQIAgEV1bDSsbrWuu0l+vlXWT0pa2k8pErgAAMCiOT6S0fu+8aAkK7fjXrmScgOXSpIiztKNJSwpAgCARfP2//iN7jswJDmZ4lgkcUStkVY5ZunGkqX7mwEAgIZz7/4hSZKJjBXHTOKwktFkvaa0KAhcAABg0TnuePHYOBkClzFmgzHmx8aYh40xu4wx7y6M/7Ux5pAx5r7CPy8p+8z7jTF7jDG/Nca8sGz8RYWxPcaY99XmVwIAAI0sedbfKb7mGxVjrdHWOs1mcVTTneZJ+gtr7b3GmHZJvzHG3FG49ilr7T+W32yM2SbpKknbJa2V9ANjzDmFy/8q6QWSDkq6xxhzi7X24YX4RQAAQGMbz3qSrJzoyJRrS73CNWfgstYekXSkcDxqjNktad0sH7lS0o3W2qykJ4wxeyRdXLi2x1q7V5KMMTcW7iVwAQCwDJwcy0lOtmLMBlEZJ6/WyNKucM2rh8sYs0nS0yTdXRh6pzHmAWPMF4wx3YWxdZIOlH3sYGFspnEAALAMpPKejJMunifcFlk/DFpLfUmx6sBljGmT9A1Jf2atHZH0GUlbJF2osAL2iYWYkDHmGmPMDmPMjv7+/oX4SgAA0ACy+UDGLQUua61sEG522hZtq9e0FkVVgcsYE1UYtm6w1n5Tkqy1x6y1vrU2kPQ5lZYND0naUPbx9YWxmcYrWGuvs9ZeZK29qK+vb76/DwAAaFBZrzJwGWOkIC5J6op31Wtai6KapxSNpM9L2m2t/WTZ+Jqy214p6aHC8S2SrjLGxI0xmyWdLenXku6RdLYxZrMxJqawsf6Whfk1AABAo8t6fkXgWt++Tuu7E5IIXJJ0qaQ3SXr+pC0gPm6MedAY84Ck35P055Jkrd0l6SaFzfD/JekdhUqYJ+mdkm6TtFvSTYV7AQDAMpDNB8Wm+a09W/WJ3/2EupPhkmJXYmkHrmqeUvy5JDPNpVtn+cxHJH1kmvFbZ/scAABYurJeIGPykqRPX/5p9bX2qTsRPnPXEmmp59Rqjp3mAQDAosh6vuSEgSsRCZcSL1hxgaSlH7iW7mu5AQBAQwkrXDlJpcB1zQXX6Nzuc3Xp2kvrObWaI3ABAIBFkc2HFa6IiSjqRCVJESeiyzdeXueZ1R5LigAAYFFkvUDGyRWrW8sJgQsAACyKrBdIJq/EEu/Xmg5LigAAYMGNZvJyHaPWWET7T6Z0w6+f1I8fOS4nklfLMqxwEbgAAMCCe8pf365VHXF944+fo+f9w4+L421neCwpAgAALJRjI1ld+9PHK8ascmpxl9+SIoELAAAsKD+wxePxrC9J+vLbLtZzz1qhjX2RJb/n1nQIXAAAYEH91TcfLB5/a+chXXpWr37n7D79x39/lhLxvJLRZB1nVx8ELgAAsKC+uuNAxXlXS6x4nMqnCFwAAACn6+JNPRXnLTG3eDyeH1drtHWxp1R3PKUIAAAWVMbz1RaPaCzrSZKMpG/v+bZaIi0az48vywoXgQsAACyo4XRe56/r0K/2DkiSBv1H9cFffLh4fTkGLpYUAQDAghpO57Wyd0iJNV+VTF79/n0V15dj4KLCBQAAFkwQWI2k83rCu0XRrp3yUls0GhzSWV1nyVqrx4cfV2tk+fVwUeECAAAL5sZ7DiiwUiJqJElu/IhS9qjOaD9Df/nMv9RlGy7TxWsurvMsFx8VLgAAsGD+6lvhHlyjwWFJkomMKmf6taH993Xpukt16bpL6zm9uqHCBQAAZhUEVjv3D85537GRTOHI17F0uBfX9k0Z5YOc+lr7ajjDxkfgAgAAs/rsnXv1yk/fpV8/MTDrffsHUpKkv3/dOnlBuCXEo4OPSpJ6W3prO8kGR+ACAACz2nV4WJJ0ZDg9633pXPjexJR/TJJ02YbLitd6Ej3TfWTZIHABAIBZ2cK7qB1jpr2+5/iYbrrngD783YclSSPeCUnS285/W/Ge3sTyrnDRNA8AAGYVFBLXTIHr9z/504rzkXy/HONo+4rtxbF1betqN8EmQIULAADMaiJwZT2/qvuPpPZrTXKNok60ONYWa6vJ3JoFFS4AADCroLCkmMpVE7gC3de/Q5dvfL4k6ZZX3KKIQ9zgvwAAAJiVLVS4UjlvyjV/Io0VmMiYRvMj2t4bLidu7txc+wk2AZYUAQDArCZC1XQVrm/vPFRxbqLhfl3LvWdrMgIXAACYVdYLJJW2fSh3eKhyqwiHwDUtAhcAAJhVzgvkJA7q10Nf1XP//kf6xO2/LV7L+2EYu/3PnydJMpERSVr2O8tPRuACAACzyvmBkpv/RXvy39DBoVH9/z/aU7yW9QPFYhld/9jHZKIDMm5KERNRW3R5P5U4GU3zAABgVrnCkqIkGTcl63VUXIu3HtAtj9+i8y58SJuT2/XQcIfMDHt2LVdUuAAAwKxyfnngGldL1C2e5/1AbjTs4xrLDyqRyKoz3rnoc2x0BC4AADCr/KTA1ZOMFc9zXiA3EgaujJ/RYHZQnTEC12QELgAAMKusV9p/y7jjSsZLFa6cF8iJpCRJaS+te47eQ4VrGgQuAAAwq5yfKx4bN62WWKTsWiBTCFwTVrauXLS5NQsCFwAAmJUXlAKXnLyCst3lc56VnLTOaD9D77jwHZKkmBub/BXLHoELAADMKhdki8fG5CpeYp3zAxknq2Q0qUvWXCJJeu665y76HBsd20IAAIAZDafyytus4hMDTr5im4ic50uRjJLRTl248kLd9bq71B5rr8tcGxkVLgAAMKNXfeYXMiZfPDdOrviqn7Gsp1/tHZCvjJLRpCQRtmZA4AIAADN6vH9cMqWnFHvaTLHCdcOvnpQkpf2UWqOtdZlfsyBwAQCAWRmn1DTf01Z6mXVrrLA9RKGHCzMjcAEAgNk5pQpXTsPKuwckSVE3jBHGyao1QoVrNjTNAwCAKe7ac0I/33NCUvhkoiS1Rlp1NPeQYhsf0p7+V+mGu/dLCmScPBWuORC4AABAhYODKb3+3+4unm9YYXRCUl9rn54cCfu2rvzszRofWykVlhsJXLNjSREAAFT4wcPHKs7PX5+QVLmDfNaE9xgn3KOLpvnZEbgAAEAFY0zFua/w1T2rW1eX7omMhP8uBK5khArXbAhcAACgaNfhYR0byUiSWqLhU4ie0kpGkxUvpS4+uTgRuFhSnBU9XAAAQJIUBFYv/eefF8839rbqkaOj8mxabdE2dcQ7SjcXAhdLitWhwgUAACRJQ+nSjvItUVfxQoUrb1Nh4IqVAtdEhSsaDT9D4JodgQsAAEiSTo6VXlLdEnMVj4QxYTQ/qM54p8ZyY6WbC5WtVz+zTxI9XHMhcAEAAElSf1ngyuR9ZaMPS05WB8b3aGvPVj115VOL143J6wMvPU9PKTzBSA/X7AhcAABAknRyrPQKn4yOaK/7T0pu+bhyQUZbe7bqkjWXaMcbdsjPrJVxsopHXQ1lhyQRuOZC4AIAAJKkg4Pp4rGJDUiSnMi4JKm3pVeSFI/EZYOY5OSUiDjadXKXNnZspIdrDgQuAAAgSXq8v9Sj5UQHK66VN8wriMk4OSWirnad3KXzV5y/WFNsWgQuAAAgSdp3YlwXb+7Rx//wAr3gKfGKa+V7cNkgKpm8Hhi6XcdTx3VW11mLPdWmQ+ACAACSpMFUTqMtN+tzT75VPR3ZimvlgUs2ImN83bjvE5Kk1cnVwuzY+BQAAEiSRjOe0vZ70rh0YPSA2qPtGs2PSpq0pGhdyfGUjHRo3BvRZesvq8+EmwgVLgAAIEkazWSKx/cev1ebuzYXzyNOqUZjbUQynjyb1xvPe6PaYm2LOs9mROACAAB67Niosu6BirH1beunv9m6Mu64sn6a5cQqsaQIAAD0gk/dqWjX0YqxnkSPPvCsD2j7iu2VN9uIjLGS2H+rWgQuAAAQcsIlxbZom8byY+pr7dNrt752ym3WluJD3I1PuY6pWFIEAADa0NMi42ZkZNTXGr4fcWXryulvtm7xMBFJLMb0mh6BCwAA6IL1XYpHc2qLtqktGjbB9yZ6p7+ZCte8EbgAAIA8P1BLIq+2WJuu2HiFJGlt29pp74060eJxwqXCVQ0CFwAAkOdbycmoLdamq7dfrTtefYc2dmyc9t73XlFqomdJsToELgAAoHxgZU1G7dF2GWNm3e6hI14KWSwpVofABQAA5PmBZDJqjbbOeW/MjRWPqXBVh8AFAADk+VbWeFVVrKJuqYeLCld1CFwAAED5IJCMp5gTm/Pe8ntomq8OgQsAAIQVLuUrlgtnUl7VikeocFWDwAUAAJT3A1njVRW4yl/nQ4WrOgQuAAAgL7AKqqxwlQcuY0wtp7VkELgAAEBY4VJ1Fa6JnehRPQIXAABQ3g/CClcVTfPJWHLOe1CJwAUAAJT385JU3ZJihMA1XwQuAAAgz4aBq5p9tVzHrfV0lhwCFwAAkBfkJFW+mBoLJ1LvCQAAgPrzgrwiqm5JUZI++/ufnfV9i6hE4AIAAPIVBq5qX9XznHXPqe2ElhiWFAEAWOastfKCsIer/D2JWDgELgAAlrm8byXHk0QPV60QuAAAWGIODqZ064NHqr4/7weSfEkErlohcAEAsMS88tN36U9uuFeZvF/V/TkvkDGBJCni0N5dCwQuAACWmP7RrCTpwECqqvtzfiAZKly1ROACAGAJCQJbPN53ssrA5RG4ao3ABQDAEnFk7Ii+s+cHxfODg1MD1737B5XKeRVj5RUulhRrg/+qAAAsEVd844rC0UclGR0byRav7dw/qL394/qLr92vK7at0nVvvqh4LezhInDVEv9VAQBYapyMFLTo+EimOPTKT99VPL7/4FDF7eFTioWmeUM0qAX+qwIAsAQcHU4Xj9tastrSvVpHywJXubI2L0mTerjY+LQm6OECAGAJ+OqOvcXjS89NqLc9ooeOnNTN9x2acm//aFZ37TlRPC8PXCwp1gaBCwCAJSARD4rHF2yy2qG3K9/7Bb37xvvkTy5pSXr9v92tdC4MWVm/rIeLJcWaIHABALAEDGfGi8fHcrtl5SvS9qikQgVrGmPZ8GnFvBdIbHxaUwQuAACWgNFsqYfr8eFHyq742nlgcNrP5Pyg7N8sKdYSgQsAgCWgPHDtOrmreGwiY3r95+6e9jPZwqt/yreFYOPT2iBwAQCwBIzlSoHLt6V3KJrI6Iyfmahw5Xm1T80RuAAAaHLpnK8fP3pw2mvGyU47LpV6u4bTeZ5SrDECFwAATW7fyXHJCRvg22PtkqSk2xNenCVwZQuB68hwRrGIlWtcGWNqO9llisAFAECTc4yRMTlJ0rnd50qSzuk5U5JknNyMn5uocB0dzigZN1S3aojABQBAk/MDW1wSfN3W1+ns7rP1noveLWn6JcVnbQ6rX1nP14mxrL7/0FG1Jgz9WzVE4AIAoMmVB65tvdv0zZd/U1u6toQXp6lwvev5Z0uS/uG2R/Xh7zwsSVrXHaPCVUP8lwUAoMn51spM2ri0NdIqafoKV1sivGf3kRHtPjIiSeprd3U4R4WrVqhwAQDQ5MJX91QGLtdxFXPi0/Zw9bTGpoxZk1NLpKWm81zOCFwAADS5wJaWFMvfhdgSaZWcTMW9V164Vono1L/+A0vgqiUCFwAATc7zbfFdiK7jFsfbou0ybmXgSkRcxVxH0e5fykQHiuNpP61EJLE4E16G6OECAKDJBdYWX81T3vjeHuuQcdIV9yaijobyx5RYfbPa7BZ95WU36MhQWp959EYCVw3NWeEyxmwwxvzYGPOwMWaXMebdhfEeY8wdxpjHCv/uLowbY8w/G2P2GGMeMMY8vey7ri7c/5gx5ura/VoAACwf0/VwSVJHrEPGTVXcmzXH9Ac3v0SStL43ps0rknrOWSuU8TNqcVlSrJVqlhQ9SX9hrd0m6RJJ7zDGbJP0Pkk/tNaeLemHhXNJerGkswv/XCPpM1IY0CR9SNKzJF0s6UMTIQ0AAJw6f4Yerr5ktxKtJ7SuO14cO+7tLB6val1ZPM54GXq4amjOwGWtPWKtvbdwPCppt6R1kq6UdH3htuslvaJwfKWkL9nQryR1GWPWSHqhpDustQPW2kFJd0h60YL+NgAALEN+oYfLmfRqnv5Uv/I2qzddcVgXFzY7bUnki9fzQek47dHDVUvzapo3xmyS9DRJd0taZa09Urh0VNKqwvE6SQfKPnawMDbTOAAAOA2+DZcUXeNWjK9tWytJum3/t7Rx6zf1T6/bqo72Ea1sXalL116qoexQ8V4CV21V3TRvjGmT9A1Jf2atHSlP0NZaa4yxCzEhY8w1CpcidcYZZyzEVwIAsKQFQdg0Pzlwvf/i9+vbe76tPUN7tGdoj/5L/6WEm9Dmzs3qSfRo38i+4r0ZL0PgqqGqKlzGmKjCsHWDtfabheFjhaVCFf59vDB+SNKGso+vL4zNNMcce7EAACAASURBVF7BWnudtfYia+1FfX198/ldAABYlsIerkCuqayjtEZbtbFjY8VYxs8oGU2qM96pwcygJCmVTykX5NQR61i0OS831TylaCR9XtJua+0nyy7dImniScOrJd1cNv7mwtOKl0gaLiw93ibpCmNMd6FZ/orCGAAAOA3huxSDad+F2B2f+nxaW7RNXfEupbyUcn5Oh8cOS5LWJtfWfK7LVTVLipdKepOkB40x9xXG/krSxyTdZIx5m6QnJb2mcO1WSS+RtEdSStJ/kyRr7YAx5n9Luqdw34ettaUd1wAAwCkJt4Xwp1S4JKk7MTVwJWPJ4vhQdkiHxsIFp4meLyy8OQOXtfbnkswMly+f5n4r6R0zfNcXJH1hPhMEAACz8yd6uBx3yrWeRM+UsWQkXFKUwsC1f3S/JGlD+4Yp92Jh8GofAACaXGBnXlKcLnAZY9QV75IkDWeH9cjAI1rRskK9Lb01n+tyReACAKDJeRM9XGZqhSvuxqeMPTr4aDFwffiXH9Ytj9+i83rOq/k8lzMCFwAATS4o9HBNV+Eaz49Lkl5zzmv09T/4up6+8ul619PeVQxcE1tDXLz64sWa7rJE4AIAoMmFPVyB3GkC1ws3vVCS9IZtb9C5Pefq+hdfr2eufqa6El3Fe15x1iv0hm1vWLT5LkdVb3wKAAAak28lmUDRaZ5S3L5iux68+sEp4+VLje+96L2KOtFaTnHZo8IFAECT84NA0vRPKVZj4olF1A4VLgAAmpwfKKxwzbNK9ZpzXlOxtIjaIXABANDkwm0hfEXd+f21/sFnf7BGM8JkLCkCANDkJprmp9sWAo2BwAUAQJPzAispUGSeFS4sHgIXAABNLgjCJcXINE8pojEQuAAAaHK+LSwpTrMPFxoDgQsAgCYXzPLyajQGAhcAAE1u4l2KbF7auAhcAAA0uVTOZ0mxwfG/DAAATSqd8zWayWsknZfjBHLZFqJhEbgAAGhSr772Lu06PKJLz+qVIlS4GhlLigAANKldh0ckScPpvIyocDUyAhcAAE3KbdmraPfPNZzOyxqfpvkGRu0RAIAm9NixUbVuuk6SNLTvUhmxLUQjo8IFAEATuuX+g8XjjA7LyqeHq4ERuAAAaEL3HjxWPLbRfkni1T4NjMAFAECT6R/N6u59R4rnxslIEkuKDYzABQBAkzkynJavbPHcuGHgomm+cRG4AABoMnnfyphcacAJwxfbQjQuAhcAAE3GD6yMky+eTywp0jTfuAhcAAA0Gc8PilUtiR6uZkDgAgCgyeQDK+OULSkWerh4SrFxEbgAAGgyYYWrFLhModrFkmLjInABANBkvMDKGE+SZINo8SlFAlfjInABANBkPN9KxpckWT8h46QkSXE3Xs9pYRYELgAAmowXBKXAFbTIuGHgSriJek4LsyBwAQDQZMJ9uMLAJT8h44b9XDE3VsdZYTYELgAAmoxfUeEqVbUSESpcjYrABQBAnR0YSMlaW/X9+UIPl5GRDUp9W1S4GheBCwCAOjoxltXvfPzH2vz+W3VkOF3VZzw/kOTLMREpKIUsergaF4ELAIA66h8t7Rh/20NHq/pMuC2Er4iJyAalF1ZT4WpcBC4AAOpoNOMVj3/6aH9Vn/GCcEkxHonpaRtWFsepcDUuAhcAAHU0mglfQv3UDV365d6TCoK5e7k8P2yajzpRPXfLmuI4Fa7GReACAKCOJipcua7/lNf6a6Xz/pyfmdgWIuZGK55MZOPTxkXgAgCgjkYzeclJ6aD3U7Ws/brGs96cn/GCQMYJK1wtkZbiuOu4tZwqTgOBCwCAOhrJeHJbDhTPT6TGpr3P8wNtet/39PmfP1Fsmo86Ufq2mgSBCwCAOhrLeorGRornTw4fnPa+4XTY6/VPdzwqz7dyHF8RJ6JkNLko88TpIXABAFAnqZyn6+7cq2istP/W/hkC12AqDFzxqCPPD2RMoKgTVVeia1HmitND4AIAoE4+9v1H5AdWOVuqcB0ZPzLtvUOp8H2J8YirfGDDHi43qu5496LMFaeHwAUAQJ0cGc5IkkxkXN3xXknSUHZk2nvLK1x5L5BT6OHqThC4mgGBCwCARfTAwSF9bUfYJJ/3A0mScce1unW1bBDRrbue0FjhScW8H+iLv3hCOS/QYFmFK+sFMk6giBNRV5wlxWYQqfcEAABYTl7+L7+QJL3iaetKgSsypr7kWbL9CRk3o5/+tl8vvWCNbtpxQH/znYeVzvuKR8ItHyKOUdbzi08pxtyYuuPdunr71XX7nTA3AhcAAItkImBJ0uGhtJ7oH5cUVri6492yfouMU2qg9/xw1/mDg2mt7w7327KyynqlneYl6c6r7lysXwGniCVFAABq7DdPDmrT+76n99x0f3Hs7r0DOjyckWRl3HH1tPTo7BV9Mm5Gn7j9t7LWKhkP6yIj6bwy+TCseb5VNh8UK1xoDgQuAABq7Ie7j0mSvnP/4eLY//zGA+GBk5VxPPXEe7SqrUvGTWvviXHd/cSAcl4YsobTeWULr/zJ5H1lPV8y4T5caA78LwUAQI2lctO9HzFcLox1/0qStK59nTrjHZITPrl4zxMDaomFfVs/e+xE8VMjGU/JeERW4bYQaA5UuAAAqLFjI5mK80j7A2o793/JRIbVsfpnkqTnrX+e2mPtSsSzkqTDw5mKoDYRugbGc9p1eEQyHkuKTYTABQBAjR0fzVacW79VxsnLiR9XNhjXnzz1TxR342qPtcs4GZ2/rkNHhtNTK2PGU2zF7ZKTDStcBK6mwZIiAAA1lsn7ikWcYk9WkO+RJLmJcD+u3pZw09P2WLtyQU6rOiI6OJDRxp7Wiu+Jdt2jeN+PJDmyooermVDhAgCgxvzAam1nQpL0Vy/Zqi+9+QpZ68htCQPXxOalHbEOSdKKDl+/PTaqnQeGKr7HRErngVhSbCZEYwAAaizvB9q+rlPXv/VindHTKmOM7E+65bSEL6qeeD1Pe6xdktTZFi4lPnBwWL3JmE6Oh7vMG3dijy7LkmKTocIFAECN+YFVxDHa2JuUMUaSFOR65ETGJKn4AurOWKckaTRXqmS97II1xWPjhL1gJjosSTyl2EQIXAAA1Fjet4o4lX/lvmzbU4rHq5KrJElr29ZKkrZuyBevPXtLrz7w0vMklQLX1nVhL1jEsFDVLAhcAADM4LFjo7r5vkOn/T1eECjimIqxrSvOKB63RdskhYHLyGg8OK7/dukmSdJ45F79n71X6u//n7O1bV3YB+ZERyRR4WomBC4AAGbwgk/dqXffeN9pf48fWEXcysA1UdWSVFxmjLkxrU6u1r6RfXr/i8/Tz/7n7+m7+74uSers3atY1JMkHRsPd66nh6t5ELgAAKixcElxUuBqDQNXMpqsGD+n+xw9NviYjPG1oadVGzs2SpLuP36/Ul5KkjSaH5VE4GomBC4AAGosrHBV/pW7oX2DJOntF7y9Ynxb7zbtGdqjN3//zZKk0VwYrh4ZeKQYuCawD1fzIHABADCHILCn9fm8P7WHa3VytX722p/pLee/pWL89VtfL0l66ORDOjh6UEPZ8InFvcN7lfbSFSGLClfzIHABADCHfBCc1uen6+GSpK5E17RjX3nJVyRJe4b2aDA7KEkayAxoODus9W3ri/dS4WoeBC4AAOaQ90+9wmWtlRdYuU71f+WuaQv33jo8dlhDmSE5pvTZde3riseT+7/QuAhcAADMwfNPvcLlF5Yjo87UCtdMehI9ijkxHR47rMHsoM7tPrd47eyus4vHE68EQuMjcAEAMIfTqXB5hcDlTrOkOBPHOFrbtla7Tu6SF3i6ZM0lxWtndZ1VPO6Md57yvLC4CFwAAMzBO40eLq9Y4ZrfX7nbV2zXjmM7JElndZdC1sQ2ERIVrmZC4AIAYA7e6VS4CsuR7jyWFCXpKStKr/7pincVX2zdk+gpjrdEWk55XlhcBC4AAOaQP40ermKFax5LilJl4NrStUVffOEX9fItL9fatrV6xqpnSCrtUI/Gx/OkAADM4bR6uAqfnc9TipK0tWerzmg/Qx2xDq1rC59M/MhzPyJJuu4F1ynrZ095Tlh8BC4AAOZwehWu8LPT7cM1m5gb0y2vuEWBnfqzY25MMTd2ynPC4iNwAQAwB+80dpqfqHBN3mm+Gq7jypV7yj8bjYMeLgAApmFtKWSdzj5cE2Ft8rsUsbzwvz4AANPI5Esh6/T24SosKZ5ChQtLB0uKAABM8qNHjulbOw8Xz09nH668d+pLilg6CFwAAEzy1n/fUXF+Ok3zg6mcJKmrlSb35YwlRQAA5nAqS4pHhzPKeYEGxsPA1dtG4FrOqHABADCH+e40n/V8XfLRH+oPn75e29Z2SJJ6kwSu5YwKFwAAk7TFK+sR811SHE7lJUk/fOSYBsazch2jjkR0weaH5kPgAgBgkjP7klrf3aJfvO/5ksKK1XwMFgJXS9TVkeGMVrTF5NA0v6wRuAAAmCSV83XB+k4lIuFfk+VbRFRjom+rJerqwYPD2r62c8HniOZC4AIAYJJU1lNrLKJ4NNzlfb4VrqHCk4mxiKM9/WM6b037gs8RzYXABQDAJOM5X60x95QrXBNLiq5jZK2UjPOM2nJH4AIAoIy1VqlcWOGKuI5cx5xCD1dY4TKFtq14hPchLncELgAAyoxlPeV9q+7W8KnCRMSZd4VrYknxoUMjksKlRSxv/AkAAKBMaaPSuCQpHnXnXeEaGM9XnMcJXMsefwIAAChzYqxyZ/jTqXBNIHCBPwEAAJQ5OZaVJK1IhhWuRNRV1ptv0zyBC5X4EwAAQJljIxlJUl97GLhiEUeZ/PyWFIfTk5cUaZpf7ghcAACU+a9dR9Uac7Wq49QrXONZX2euSBbPqXCBPwEAABQ8cWJcv9hzUqs6EjKFPR3ip1DhGs962lq22SlPKYKd2AAAKDg6HC4nXvO8M4tjiairoUlLhDNJ53wdHcloNOupsyVWHGdJEQQuAMCyMJzK6/pf7tOZfUm97IK1094zmgmD1VPWld59GI84ylZR4fruA4f1zq/sLJ53FfbxkqhwgcAFAFgm/t9vPKD/2nVUkmYMXGNZT5LUVvYqnmp7uG6573DFeVdLKXDRwwX+BAAAloUjhacPZ/OlXz4pSWpPlAJXtRUuP7AV51S4UI4/AQCAZWFbWRP7dPJ+oPsODEmS2hKVFa5MFRWuyVWwzrIKV0uUHq7ljsAFAFhWkrHpw08qV6pilTe5xyOOBsZzevTY6Kzfm/V8PWdLb/G8ozxwzfAzsXwQuAAAy0K6EKgyXiBr7ZTrqZw37efi0fCvyis+dees35/JBxW9Wr2FneolerhA4AIALHH/9rO9uuvxE0oX+rD8wCqV8/VvP9urXNky4ESF69XPWF/xeb/KPU+znq9E2dLhxLsYJRX39MLyReACACxpf/u93Xr95+6ueAH1v/x4j/72e7v11R0HimMTFbArtq2q+Hy6rPJ19Rd+XXyScbKsV1nh6m6NTXsflicCFwBgyUqX9WWly540/MxPHg8PypYWxwtBKhmv3DGp/HM/fbRfP9x9bNqflcn7Fb1frkNVCyXswwUAOC13PX5CaztbtKns3YGN4sRYtng83et52hOlxvZU4frkBvfyZnpJCqbp/5LCClci6uhvX3G+Hjw4fMpzxtJE4AIAnLL0/d/W7Tf9UN+Kvkz3//UL6z2dKU6O54rHBwZSU66X752VyobBKhmr/KvxvDUd+u4DR8o+M/3PyuYDxaOu3njJxtOZMpYoAhcA4JR8+c6H9aYfXa2/jkpneYdk/efLuNG5P7iIToyWKlyDqbw6EhGNZEo9WBkvDFk5L9Anbv+tJKl1UoXr7b+7RdZa/ePtj0qShqd5r6K1VhnPV4KnETGDOf9kGGO+YIw5box5qGzsr40xh4wx9xX+eUnZtfcbY/YYY35rjHlh2fiLCmN7jDHvW/hfBQCwmL5220+Kx2+M/FD5nV+t32RmcGQ4XXF++XmVDfETjfRf+80B7T0xLqly/ywp7MV61dNLTy4OpXKaLO9bWSvF2eAUM6gmiv+7pBdNM/4pa+2FhX9ulSRjzDZJV0naXvjMp40xrjHGlfSvkl4saZuk1xXuBQA0ocNDaa1VvyTpi9s+L8868o4/WudZTXVwMAxc56xq0y3vvFQffdVTKq5P9HUdHS699qezZWqVbm1XS/F4cJrANVEpm26/rUSUqheqWFK01t5pjNlU5fddKelGa21W0hPGmD2SLi5c22Ot3StJxpgbC/c+PO8ZAwDq7p59A1qn45KkvjO26uiuHnUOH6rzrKY6OJTW5hVJ3f7nv1sce+H2VbptV/ik4cHBlF7wyZ9OWUaczj+/7mn60//cqaHU1CXF4cLY5OrYzg++QBGXpxVxettCvNMY80BhybG7MLZO0oGyew4WxmYaBwA0odGMp1VmUDbSolhbr47YHpmRw/WeVoUHDw7rew8c0bqy6pQk/cvrn66nbuiSJP1g93E9dnxM9xeeKvziW5454/e9/Klr9bQzuqYPXIW+rsnVse5krOJJSCxfpxq4PiNpi6QLJR2R9ImFmpAx5hpjzA5jzI7+/v6F+loAwAIayeS1wgzLJvvUloiq33bJSTXW/81+yxd/LUlKxiurV1HX0c3vuFSdLVF5ZY8cXrihS7+3deWs39ndGpt2SXEicHVNsxwJSKcYuKy1x6y1vrU2kPQ5lZYND0naUHbr+sLYTOPTffd11tqLrLUX9fX1ncr0AAAL7b6vSA/cVDwdzXhaaYZl2vqUjEeUVlzGS8/yBYtvYsOHmXaGT0QdDZZVq0YzUytXk3W1RmetcHWxuzxmcEqByxizpuz0lZImnmC8RdJVxpi4MWazpLMl/VrSPZLONsZsNsbEFDbW33Lq0wYALBYvMy59+4+lb/5RcWw0k1efMyqTXKm2RERpG2u4wLWlL9yI9c3P3jTt9ZZJTxSmc1M3Rp1scoXr+rv2adP7vlfc72u6hntAqqJp3hjzn5Iuk7TCGHNQ0ockXWaMuVDh/wOxT9L/kCRr7S5jzE0Km+E9Se+w1vqF73mnpNskuZK+YK3dteC/DQBgQVlr9eEvfEMfLpwfPnpYH/nhUeX9QCvMsJRcobZChcttsMAVizi6aGO3Xrh99bTXn35Gt/adLG2GmppmJ/rJOhJRpXK+/MDKdYz+5jvhX2Uf/HZYd+hqJXBhetU8pfi6aYY/P8v9H5H0kWnGb5V067xmBwCoqx89clzHDj4uFVbKrv/ez/S9x9pkFOgziWGpbaUSEVdpxRTx0+G7CU1jPJU3kva0om3mJb7nndOnb+4sdbe84sK5n+Wa6Acbz3nqSERVtlG9YhFHCfbhwgzYaR4AMKMDAymtMyeL5+12RFKbujQmV4GU7FMs4ihj4zKykpeVoon6TbjMSCavM/tmfr/jtrUdxeP7P3SF2uJz/5U48WLrVNZXx6SnD2mYx2wIXACAGY1kPK0pC1wdwbCkteo1I+FAIXClFA/P86mGCVyjGW9KKCp35oqkXrh9ld7ynM1V915N7Nc1XSM+/VuYDdvfAgBmNJrJa70zoBHTLklqC8L9qvpM+G8l++Q6RjkzEbgWt4/r0FBaH711t4LC2t69+we1c/+grLUaSefVnpi5rhBxHX32TRfp2Vt6q/55E1WwVM7Tw4dHKq5Vs3kqli8CFwBgRqMZT+vdkzoYO1OBjNr9MGSsUaHq1bFWkpR3ClWtfGq6r6mZt3/5N/rsnXv1eP+YJOlVn75Lr/z0XUrnfXmBnbLz++lqjYWB659/uEe37Tpacc11GqN3DY2JJUUAwIxGM57W6KQejW/RSGaPDh8Jm8zPcQ7JOlGZ7s2SJM9NhM+tL3Lg2lsIWs6ksDOaCZf8ZltSPBUTFa4f7D6mH+w+Jtcx8gvVNRrmMRsqXACAGY2nU+qxgxqPr9aAbVePGdVb3e/r7ZHvKOjcILlhAPGKFa7FXVIcL+yd9cvHT+r5//iT4vhIYSPS2ZYUT0XrpF3rn7mpu3i8ecXMDfoAgQsAMCM7clSOrFItazSodnVrVP8r+uXw4pmXFe8L3EIPl5dZtLkdGyn9rA98+yHtPTFePB/JTP8y6dM1OcCt7WrRT957mTpbonrrczcv6M/C0sKSIgBgWnk/UH5gvxSRBiN9arXtOtuES4p3B1v1rBf9XfFeG4lLOYXbQiyCHz1yTD/YfXzG60eGwzDWscAVrp5Jr+7pa4tr04qk7v/QFQv6c7D0ELgAAFMEgdXZ/9/39WZ3nyTpmNOnpG3TJveYJOn8V/6lFG0p3m8XscJlrdVb/33HrPe88ys7JS18hSviVi4MrWiLL+j3Y+kicAEApjgwmNJWs18fiPyHMj3n6oDWqlU9xevJtdsqPxAp9HB5OdXaXY+fnPumgskVqYW2qrMx9hxD46OHCwAwxSNHR/VG9w55ish53Y06c1WHvuH/jiTpP8wfSKtmCly1r3D956/3V3WfYxa+wjXZuq6WuW8CRIULADCNR4+O6lnOQY10b9PqvjP1rucH+p2zV8hb/xZdZab+/+omsnhLioeH0opHHGW9YNrrnS1RDafz6mqN1WRvrI+96il63zcflETgQvWocAEApnjk2KjOcQ5r9ZkXSJKirqNnbupRJBKZ0scU3jBR4apt0/x1dz6ue/cPaeuajinXNvS0aGNvq7pbw6rWxL8X2h8+Y33xeGU7PVyoDoELADDFiaOH1KVRqe/cqu53orVfUsx5gf7u1kckSZt7W6dc//tXXaCf/uXvFQNhT7I2/VvRssA5ecNVYCYELgDAFC0je8KDFedUdX8inlAgU9MK10OHw/c3XnnhWr3vxedJkraVVbou2NAlqRSIahW4gFNB4AIAVBjJ5LUmfyA8qTJwtSWiyipaswqX5wd61afvkiR94KXbtLozoe++67m6/q0Xl+ZQeO1OzA2rTgQuNBKa5gEAFQ4OpHWWOSzPbVGkc0NVn2mLR5S1USW8jBZyke0rd+/XphWtOndVe3Gsr9A3df66zuJ7DMtNLCl213BLiK/80bO0vmvqsiYwEwIXAKDC135zQJeZQ/K6tyjiVLcQ0paIKKuo/FxmQf9i+atvhU8D3vmXvydJ+vgfXlBxfeIpxBdtX10cy+TD9yvWssL1nC0ravbdWJoIXACAIs8P9MVf7NObYscV7buk6s8lCxUuL5tasL9YPL+07cNY1pMkdbRM/fadH3yB2spe4TNeuLerxpueAvNBDxcAoGg850uyWm0G5XaurfpzbXFXKSV0fGBwweYyMF7atX48F4aoZHxq4OpOxiqeHJzYn6ttmnuBeiFwAQCKxrOeOpRSq8lK7Wuq/lwQSCnFte/wcX35l/sWZC79Y6UnHn/2aL8kqTU2d4jKEbjQgAhcAICiVM7TKlOoUnVUX+H63XP7lLJxJU1GH7x5l379xMBpz+X4aClw/fOPwm0qqglRE4ErGXdPew7AQiFwAQCKxrK+VptCWJpHhWtFW1xOvE2tCkPSaz77y9Oey5GhqVtMVBOiWFJEIyJwAQCKUlmvFLg6qg9ckpRzW9SqhduH6/BQespYVRUuf6LCReBC4yBwAQCKrv/lPq1SYUlxHhUuSco5LWHv1wKZHLje8Kwz5vXkIYELjYTABQBQ3g90YCCl23Yd02ozID/eJUVb5vUdOWdhK1zlTfOS9D+et2Ven0/G6OFC4yBwAcAydWwko6uu+6WODKf1qTse1e98/MeSpFVmUHae1S1JyrstSpqsjIK5b67CUCqvSNnLoVd3Jqr63OVbV0oq7TgPNALqrQCwTH3/wSP61d4BPfujP6oY3xgdVqRr87y/z3PDV90klFNa1YWj2Qymcjqjt1V7+8clSbFIdQHq0298usaz/mn/fGAhEf8BYJkKNzmVJFv4J7TOHZp3/5YkeW64BJnU6fdx+YHVwcG0NvUm5/3ZeMTlxdVoOFS4AGAZOjmW1T/c9ls937lXn4xeq37bqev8l2qFRtSaH5A618/7O71IWOE6xzmgPzR3SvYlkpn/q6x/uPuYPvb9RyRJLdGwD+vpZ3TN+3uARkLgAoBlZv/JlK75/E/0j9Fr9Wr3TtmWbnWlD+kfnOtKN6192ry/N++Ey4gfi3xOZzj90oG3SGdU/z5GKaxsve36HcXzy87t0x88dY2efSYvi0ZzY0kRAJaZa+98XJcPf0uvdu+UnIjMa29Q/6V/U3nT+mfO+3vTJlxSjJrCUuWTd837O/rLdpfvSET06mes14vOX6PO1ui8vwtoJAQuAFiihlI5ffKOR+X5lU8NPnRoWC9q2yOtOl/6wHFp06Xqe8Gfae8fPaone58r/ymvlVp75v3zRvywb6pPQ+FAev4vsvaC0lzXdrXInMKSJNCIWFIEgCXqH2//rf7jV/t17qp2vfSCsAneWqsnToxrc2SvtPYPJKe0V9WZ61ZJ7/reKf+8IS+sQkVMITRlhub9HX5Qat7n1TxYSqhwAcASNRFeTo6XlulOjueUy6TU5g1J3RsX9OcNepOeDEzPP3Dl/VLgch2qW1g6CFwAsERNVIiGUvni2P0HhrTWnAxPOjcs6M8bzFVWpOwpLCmWV7giLoELSweBCwCWqEw+kFGgzj3fkjLDkqSHD49onTkR3rDAgas/X1nhsqdQ4Srv4bp4U+9pzwloFCyQA8ASZK3VYCqnN7o/0NVH/l2P3jz8f9m77/A4qnPx498zW9V7tWTJlnvBBRvjS7ExvYTeIYHQSSAkkEtICEmAm9BSuCQ3hPALgRBCCL2aZsCm2Bgbd1sustV7X+1q68zvj1mtJCQ3WbLa+3keHu/OnJk9u8jeV+855z38sPo00uIcjLeHM099qLW1L9+aNxFWmo/LjVSyva0HfY+ODNf8/CRuWjy+P7snxKCSDJcQQoww/15dylG/WUZRnZt52g4A2nZ9wdaqVpbvqGOctRGUBvHZ/fq6t58yOfJ4j56J7mW5fgAAIABJREFUHjj4jayD4YDr+ydMwGGVzafFyCEZLiGEGGHuemUTYNa0muysACDPvxMnPk7TviLPUg/RWWDp39pWXUs4lBoZLAyWHvQ9OjJcVk3yAWJkkYBLCCFGGKum+B/tCSZr5UykHL9hIUW5eMr2CP9l2Qo+IOO/BrQPLqJRwYPfUzEYXqUoKxTFSCO/QgghRqXPd9WTf9fbbK92DXZX+pU/qBOvt3Cp9RPmaLuwEOJz5/EAZrDVIXXCwHTggr/hOe5n+LBh0X1gGPu/pouOSfM2WaEoRhgJuIQQo9LLa8sB2Fh+8CvphqKvihtpdPu5+9VNHKEVdTuXdczlPS9InTQwHZl5IdrxP8ZrhFcsHmSWq2MOl2S4xEgjAZcQI8TXpU288nX5YHdj2Gjy+AFIjrHvp+XQF9INLvrLSi776ype/rqcxTGlGEqDcx+HY37IlGPOi7T9e/BU88GMCwesPzaLho/w/LBg+0FdGwrJHC4xMskcLiFGiPP/bG4UfP7c/l3qP1I1houB2q3D/4u9td18L9trzOHRUxIrUGoKzO6S2frBOqhcz5VTzgHDD7aoAeuPRVP41cFluLyBEE6bRTJcYsQa/v/SCCFEHzSHM1zB0MHNMRqKOrJ1HZLbdkLmEd0bJY+HGedjs1oGNNjqEOgIuAL7z3C9uaGSKfe8y9qSJh5bthOQKvNi5JGAS4gRwOMPAuDAT8B3cEM4o1GLJ0BJg4dTtdVENWwZ7O4csub2zq174mnD2V4DGdMGsUfgDpkDKCu27X+Y+90t1QBc/ffVbK0yi6VKhkuMNBJwCTEC1LSawzabHdeinjplkHsz9M26730UOk/YH+XoD84d7O4cspbw8KhCZ7IKBzjpgxtweTEzXLur6vfb1hYOrlzeYOSYVQIuMcJIwCXECFDb6sWBH5sKYa3ZCMAbGyr55eube7S9+9VNvLmh8nB3ccgIhsyyAzmqbpB70n8a3X7Gq0q2Oq7hVvsb5sH0qYPap45J8zbDv892z64s5rX1PX8eJcMlRhoJuIQYAWpdPiaoishzXyDID55fxzMrS/iiqDPDoOsGz31Zyq3PrxuMbg4JdW1mNvB4bdMg96T/fLKjjosty4lSfo5X68GRAPFjBrVPHWUhrKF9b+9z/1vbej1us8jXkxhZ5CdaiBGgoc1Hsuos4Llm2y4ALrF8zDN/+xO7as1z//3SxkHp31BiDr8aXGN9DwCfLX5wO3QI/EGdrZWtvLmhknnR1Z0n0iaBGtwMkRsnAA5933MKZ+Yk9HpcMlxipJGyEEKMAJ5AiATckedVFWVMUmU8ZHsSgHVtN0J6HC9/XU4Gjbzs+BWhXTFYJiwepB4PnqrmdiapcgpUBZVGMtmBRggF+n1fwcPhqN98SHN4/taMqEZImgHtTTD7ikHuGbgwV0I6Qu59tstLjmZtSVOP4zKHS4w0kuESYgTwBnQSVOcXm7elhtmWPZHnUdtfizy+3PoROaqe4Kq/HNY+DhV7GtzM1czSA8tCc82D/rYBeS1dN/h4e21kQ+a+aPeHeHLFbryBULfjgZAeCbZsBHG0lUPBErh9K8z77iH1uz+0GWbAZd9PwOUL6kxIj+1xXDJcYqSRgEuIEcAbCBGPJ/J8887dzHVUELI4qDUSia34NDJZfIYyAzHdVTMofR1sWypa+S/7bvSoZLYaeeZB/76Dgr76qriR7/79K371xsGVngiEdOrDc80+2FbDr9/Zxn1vbe3WpmNrovO0T7na8i4q5IOxR/dPx/uBO5zhsuuefbbzBkJMZzdTVUm341JpXow0MqQoxAjgDYTIt3V+sVm9jUyJKkVPnca6CgsLGzdR32auFpvrKIMgaK6qwerugKpu8WLRFGlxjh7nqlraWbq5intid6Fnz8PdYs4zwr/voKCvOjJbL60t5/5zZxzwdT9+cQOvr6/k1OkZNIT/v339jWG3T3fVM1WV8Af7450H84879E73Ez82fIZ1v0OKvqDOD9ufwmfzcqH/V5HjkuESI438CiHECNDuD5GstdOMOTSTqlqYokrRMmdQaIwlzlNGXXMLsXhICpqrFm3t9WAM/yrr33T0A8s4+oFlvZ5bU9xEnNFGZqAUI2c+7YSDssDAZLgC4YCrPRAEXT/g614Pl0l4b0sNa8KBlsfffUix2ePniuTCzgPH3AbOobUAoI0o7MH9BVwhko0mao3EbsdlDpcYaSTgEmIEaA+ESNQ8NBjx1BnxXJBZhzPQhCVrJjXWMSgMXFW7GKfMlWxf6lPQjIA5wXoECYSHTfc2Z2rV7gYW2HYDoOUeFVlJN3AZLrM/v7c9jvFw/gFd4w+a12joxHYZJo711XQL2po9AQqoMMs//LwOTr6v/zreT9qMKKzBfc+P8wV1kkKN1BnmasXLjsoFQJOAS4wwEnAJMQJ4AzrxeLDFJFFlpJBb/6l5ImM6trQJAPhrd1KgzMzJKj1cFLNtZM3jKmnYezbFGwjx3pZqzk2rAKWh5R5JuxHOcA3QHK6OfRrPt3yG8rZAW+1+r6lqMcso/DP1GTY5r+fn1me5yfIG74RuxL/yiUi7Zk+AnFAppE0Gq31A+n+o3ERh3U+GS/d7iDbc1BpJANx24iSKHzzzcHRPiMNKAi4hRgBvIEQcbeRmZTFhwmTzoMUOmTNJyzO3ePnsy9WM1yoxlIVCa3jblxEWcNW6fD2ObSxvZn1ZM8t31FHf5mehfTekT0c54vBr4QzXAA0phnQDB10qrVdt2O81FU3tgMH8wFoUBtdZl3KX7d8AuNf8K9KuxeMl01cCaYNbUX5fXAcwpBgTaASgFnNI0WmTryUxMslPthDDnK4bZsBltKGiEokeE56cnX8sOBMYn5dLkxHLOFXNsYlNqKQ8PNHZZpsDyLgMJx1lEqBzSPHsP33Ouf/3OXvq3WjoJDZugNyjAPBpAzukGNQN4rvUR6N1/1sqlTe3k6Pqsfka4Yzfcq9+HQAew0Fsy3bQQ2yvdpEUqMZm+MwM1xDVZkRh+Fw9Slp0FR9oAKDO6Ai4LIelb0IcbhJwCTGMLd1UxfifvcOakiZiDDdEJcLRN8PkM+H0RwA4bmIqdfYcjktqZI5WBJlHYMRkmDfYS4arpT3AvW9uodUb6PX8UNXsCWAjyE7Hdwh8/DC1LnNbmTSa2FZWx7zoGpS/LRJwBbRo88LAwK1SjFdd7r2flaG/eH0zT322h9mauVMAOfP54U8fpPHG9fzBdh023cfrn6zk+dWlTLOGg7dB3jNxb/7n3Bm0EYXyt3Hbv/e+lVR8yMxwdQRcDqt8LYmRScpCCDGMrS9vDj8yiNbd4EyAmFS4rHPoKc5pI+7IE2HV/5kHxv2IuEAyvjo7jl4CridX7ObX75j7280Zm8TZs7IH+m30myaPn2O0zdhUCJb/mpVJV+DEx3LH7TTujOOt+EvBBeTMBzCHFA0Gbg6XbnTbAaCtvoztJY0cmZfca/t/rDRrUV0VvQe0KMiYToLFBlHj8MTlQxO88uEKluuzeDCjEVqA1EkD0vdDdeXRebz2fgxxoXY+39XQa5sX15QR7a8HG/z+2lNYXqmhBnlLIiEGivwqIcQw9fmuep5Ybq64S8KFhRBEp/beeNo55p/2ODjiEgrS46g1Egi19gy4HvvIrMKuoQ+7shHNHj+TVFnk+apddZzs3Ea08pGj6rlUvQ/RKZA8HoBgZEhxoCbN690yXCs3bOOCx1cC5orKjlWVYJZH6DDPWgTZs7ttN6Ql5ACQocyM0Hx7iblCMap7OYWhxKOiiaWdvYVQz6wsJl01YyiNyePHc8PxBYeze0IcVhJwCTFM/d/H5rDT9y2vscLxI/NgwpjeG49dAFe9Bd9fBY44CtJiqDMS8DZ1n1PU5gvi8ga5b3o12xzfZerX9w7kW+g3u2pd3PCPNWwoa2GM1plN2bFrJ4uTGyPPE1t3QM5RkY2drVYLPi1qwIYUzTlc5r2rjGQSwxuMb6lsYd7/fMhJv18eaevyBgG4dbaFCYHt5hy8ruIy0Q1FFo1E4SW/6XOYfPqA9Lu/uFU0UcqPRfU+h8vlDTI3JYSKSgZN5m6JkU2GFIUYpiyawomP/7b9p/NgfM7eLxjXWYW8IC2WciOJiV2299F1g1e+LgcMzqn5PxwqQHrdFwPQ8/73u/d38P5W873cmdgK5tQtrK0lTIqvg5g0c7i1YRdMODFyXZTdgi/goLi0mnFBHXs/zx8y53CZ2bNSI50UWgE487HPAHOunDcQYlNFC09/UQzAJa6nUbZomHdtt3vFREdTTwKZqpHTtK+wBNthxgX92t/+5lHm9j4xtPd6vtkTIDHBBdG9D7EKMZJIwCVGvUa3n8Qo27ArtOjyBrkopZiui+BIHHtA145Pi2GdkYDuKuTZlcV8e2E+43/2DgD5qpqEtiLqjXhS2svNFXz26H7vf3/qyA4BZIcqKbOOJTdYSg51pAYqzCHEbz0Ga/8Osy6LtE2MstHcYmNLaTXF22s5dXpmv/YrqBvEhoONciON8VrPVYqnPboCp81CYbULK0Eya1bArIshPqtbu/goK1VGMkdou1liWYeRPg2VO3T2TuyNJ7yfYrzy9jgX0g1avQHi41wQJQGXGPlkSFGMahXN7Rz9m2V877mvB7srvWr3h1hZ1EBRXc9q3S5vgBOt68EWDXfsgOs+gti0A7pvtN2Kz5lKguHivtc3YHSZq3XNmAoAXjROQmFAS9nebjNk1Lq8zE0O8JH9drIDpexOXkTIUIzVaknwVkDSOEifAqc/BI7YyHUJUTY8OInBG9mz8P0t1TS09azn1RchXSdK+TBQVBtJJOKma8kKgOIGD9F2czhtiio1C4WOO77HveKdNmqMJKZrJaRpbajz/wpDfINntzID9d4yXA1uH4YBMXqrZLjEqDC0/7YKMcDe21yNP6Tz7pZqvvPU6m6Bx1Bw75tbuOzJVZz4u+U9zrm8QaZ515tzfeIyIOfIg7q3Fm9mc1Jo4cF3O/fkm6VvhehU1ltnmQdaK/r+Bg6TWpeP6xLXMF4zty4K5h1LFSlMUOVEtVdHJsl/U2K0HQ8OovBR0eyh2ePnhmfXctM/1/ZLv4K6QQxeDFs0TUYcNhUiHjcPW5/gGG0TBaqC9+x3kuvbAcAMrdi8MHt2j3vphkGVYQYm2vhFkDmzX/o4kNyYAVes6hlwPf5JEQBRwRbJcIlRQQIuMWq1+YLc99ZWLIRYpG3gix1V+IIHvsHw4bC5smWv59q9XlJ95X3+4nUmmgFXmmqJrHa0EGJi2xrIP4ZWR7hW1wEU6xxMgZBOsydArl4BUUnwo60kzjiVMj2dY7QtZpYueVyv1yZE2fAYDqKVj5IGD2WNZmCws3bf+/8dqGDIIBov2GMiG4tfZFnBxdbl/Mn2Rx6I+ieTtXKmt35OLB5+YH0F3ZFgZuS+QVOKciOcwRw7tIcSO3jCe1XG9pLhMveMNHAGWiTDJUYFCbjEqPVsuObR1Zb3eMb+EI/b/hdfILifqw6v3KRoHPi5wvIhbU3VkeOBkE5asAqNEKRM7NO9Z02dAkCaamau2oGNIMdqm826SDMuxJ4UXvE4xAOu1nazOGuqr9T8LBLGMGNMPKVGOonhCet7y3AlRNloDw8pbixvoazJXFFo7af5fEHdIFr5UI5Ymg0z4PqB9RUAklQb8/WNAGQEK/id7S9kq0a0rCMiqyi7umR+LgWn30roomdh4ff7pX8DzaPtfUgxpBuMjQUV8knAJUYFCbjEqOUKV1G/PvErAE62rEVteXW/19W6vN1qJg2kZk+Asy1f8GvbU9Q9/e3I8b99tofxKly1PLVvAdf0SWbBzBusb/OK41f8yPoS98a8iBGVDJNOZdKYVBqMePTm8kN+HwPlieVFkdIKiZ6SyGfhsFrImzCts2EvGSOACemxNBmxjFF1/I/rHja//mj4TP8EXCFdJ1b5ULYYmsIBV4Ly4DKiwq9i0GpEM1WVcLS2Fb9hgW/9b6/3slk0Lj12GpbpZ4M9pl/6N9A8+xhSDIQMUi3hgFiGFMUoIAGXGLVKGjzMSNbJdO9g64Qb0Q0FdYX7vMblDXDUr5fx67e3HZY+Nnn8nJ1gDvfltnwNITMD95flRYxX4cxTyoS+3Tw+G6KSOVoz38v3rG+QH9iNOvN3YHUwOTOeSiOZ9oahO2n+gaWFNHkCxOIhylfX7bM4ev5RnQ33kkFZMD6ZapKIV+0cb9nEncG/kEkDvn3s/XcwOuZwdR1SBPhPaHHk8b9CJzJJqyBBeWhY9BtIGTnFP93hshCxhhlwtbQH+M9XZWypbCEQ0klS4aFbyXCJUUDKQohRq7rVy2LHTvAYNGYeQ8POf/e61U1Xy3fUAfBVcdOA929XbRuF1S7GJxQDYCUIrkraorJp9gQ4L98DrrS+VxpXCsYdT7B0NY82HcOPbS8CCqafB5jZn2ojhfyWoZvhsls1/EGdfBUebu0afI5bZP6ZNK7XITqA9DgnN515LLz3WuTYfG07wWARevMMtMS9FJI9QKGQOaSIPSMypAiQNedUGowsPCqW9Ws7f+/NmjI85mYdqHaiCRgWksL1x+56eSNLN3cOjV+YZBaClQyXGA0k4BKjVkObj1mOHaDZcKfOpsZIIm8/AVdNq1kuIC3OMSB9cvuCNLT5GZsSzardDWjoZATKKLHmkxcshpZySj1mgJXr3gRZPVezHZRz/4wl5Mf6cTGeulaip5wUCU7Gp8Ww0UjG7t5xiO9q4CigQFVwtfV980DX4dXoZLjyFUibvM972BPMxQHFCQsY27yaiy2fcJxlM3see5Wsn67DaTu4Cugub4Dt1S7m5SebGS7lA1s0LXQOA56x6DhI/S4bC2v59KsVnRenT+vljsOXoTSqjGSyMH9RqWntXo8ruaOInGS4xCggQ4pi1Gpw+xkbKoOUCdicTmqNRCzu6n1e0xYusLl5RxEvr+n/obZH3tvO8Y98zOaKFgqrWxmr1WPVfRTGmJst01JOaaObRFzEunZD/jGH9oL2GFRUEj88Yw7RV/0HFtwQORXvtNFqT8cRdIGvf1btHawvdzeQf9fbVDT3nANU1ujBF9R50X4vF1rCQcs3J8dPOBES9lF9v6PNCXfz0REPU0kKx1k2AzBOL2Fn6cEvGLj4iVVc+JeV1Lq8tPmCxOIBZyLfP3FqZ6OkPPOl02NxE8Vni1+A7y4Fq/2gX2+oqzDSyDJqAXpsTN2x1ZFkuMRoIAGXGJV8wRAub5AMXwmkTcJusdBCDJrP1aOtNxBC1836XC5vgImqnK+dN+F67Y5+79eOGvP1731zC/9cVco4zBpYexIWmA1ayihp8DBBhWtjpU/v9z50E59t/umqGtjX2YsnVpjz1zaWNfc491FhLbF4SA7PAzKiU8Hah8yjPQYW3YkjNolKI6XbqbYt7x3UrcoaPWyrMofPjvr1Ml5aW0680QbOBH508iQ4/RE44pLIptS5ydHs/s0ZHLv4NMj7r4Pv+zBQSQrphrm/5TcHdhPoCLiSDm+nhBgEEnCJUammxYcDP/HeCkidjMOm4TacaIHumRx/UGfKPe/y2/e3A2btrhO0dQBcbX3P3PamH/mDOjmqjgvLHyKN5khg1ZI4nSbiqK/YzQNLC5npNDMGpE3q19f/JktiODs0SPO4qlrMISirpec/VbUuL9Ms4X4t/inqlq8O6bViHVZqDPOLv0jPosmIRdv57gFfbxgGP3xhPTaLoqOavJWgOYerY57dghvg/L92u264bSl1sJqNWOLCQ4ffnEqXgAvscSMysyfEN0nAJUYdwzA4/pGPyVM1aOiQNhmHVaONKCxBd7e2tS7zC/+5L0sBcPmCTFCdw0yhskP7kv+mWpePU7XVXGL9hLtt/+TH09sgPgdbXCoVegobt5jDXZMtlWB1QkJuv77+NwWjUs0HnoYBfZ296aix1fFnV41uP1Md9eaTmRcd8jygeKeN6nAl97zJs1nvnM/E1pWs3VN/QNdXtnhZW9LEXSeNZW3yPXzp+B6TVTggdCYcUt+GK6Wg1Yg263CFgqhwjutiy8ecq31GguGCaMluidFBAi4x6hQ3mFmpySo8ByttMnarhsdwYgn5IqUXoHOSfFR44rTLGyTP6abFaa5ec+9e1a99q2/zMdVqDt/NsZXiqFgJ+ccwKyeBKiOFbGUGPnlGhVnkUzu4Cd0HS3eaX4Z7Sst4YnnRgL5Wb0K6gULH39pzMUOj288kWx0oS78EnkkxdkqNdACssSmkHXkOyaqN5h2f7/faNl+Qu1/dBMBC3xekeHaToZq52hLOkI3SgAugNVyLC19reEzR4GHbkzxq/zPxhmxcLUYPCbjEqPPTV8zq3j+b3W5midKm4LBacIe3IcHfOY+rY1WV02b+VWnzBkhVrfjix1FjJKI37O63fgVDOh5/iPlxjQDk6eXgroO8YzgyL4kqI5lMZZ7L10v7XPD0YOhOcyjs7S838cDSQmq/scpsoAV1nSdsf+Cy5Uv4/dKN3PzPtbSEs11N7oBZ/DUxt1+GpMalxPBS6HieC54I865lwsJzCRgWEsuX7ffa3763nU+2myvx8kpfwRuXx9f6BC6yhifzO/tYumMEaDXM1ZmV1dWs3tNIFo2Rc5nBClmhKEYNCbjEqNPkDpAV7yCzYZVZVsFiwx4eUgTA3zmsWN3SEXCZmaSaVh9JRgt6dCrVRjKqHyeTu/1msc0EvQXiu9R/Gr+YhCgb1UYSicpNNvVkGXWQPaffXntv7HYHLUY0qRYzK/h16cDXH+vwxPIi6tv8nGIxN5J+f8XnLN1czax738flDVBU18Y4vbjfFg4kRJvb/NwdvBbGzMUZl8RappJbt6LX9lf+vy95dmUxAKWN5uczVtUQU7kS51FXMe2M75kNUydD1hH90sfhqKMcxrY9ZkZ5rrYzci4jWCkZLjFqSB0uMerUtfm4aoIHtX0LnPFbABxWc9I80K0EQk14DpdSimBIp7rVS5yzmYYYM+AqaOvHgMtnDmVGBZpg6nnmhsuJeZCUhwJqwvOLIiUQcub122vvjcOq0WjEER00Vwm2th++vSYf/XCnWVIhbLIqpdAYC8C3/7aaNncbGVHlkHlZv73mQxfMJCshKvL8g+Bs7vH+E3dtMTHp+ZHjbl+Qz3bV89muer69MD+SdbvM8hGG0lCzLsMZnw1zLx022/AMlFYjvL2P0QY4ma3t6t5AMlxilJAMlxg1AiGdT7bX0uj2M9dnZk2YfDpAeNJ8x5BiZ8D12c56pqs9/Ln5Jur3bMCut2PTfWix6VQZyTg8+67bdTDcviAaOvZAC8SkwTG3wfRzI+fLjDQAbre9RJklF3IHviq502ahmTgSMT+TdWVNLN10YEFmcb2b8//8OfVtvj699qzcBLJU5/BTnqrlVO0rVjpuIaViGbfODKIMHTL6rzTGJfPHcvyktMjzuhQzqG3Y1j3Ltae+++KK0kYPp2mrudn6Jmr6eZ3lNEZ5sAXQSBwAW4v2ADBRVVCmd37GkuESo4UEXGLUeGJ5EVf/3VxVOMG1CtKmRopidkyaByBci2tlUQNbKlu5y/o844xy/BteJkWZNZYscenUGMnYAq3dhiAPhdsfIpE2FAbEpPY4f/9NnZmcqNPvBW3g//p2ZLiSlYt7rX/n9g1ncc9zHx/Qtfe9tZWvS5v5oqhvKxxjHVbmJXV+trlaLX8cu5ws1cit1te4ZnyLeSJjRp/ufyBuuOhs3IYDo7T74ojdXQKuOpePOpePm+M+JehMhrMe/eZtRi2Lpqg3zAUDpaXFAIxXlawzumzBJBkuMUpIwCVGjQ+2mbWrJsWHyGz6GiaeFDlnt2i4I3O4zGzOpgpzGG2cZmaxguVfkxreE86eYGa4AGg9+GHFj7fXcs9rm2n1dpY7cPuCJHVU3o5O6XHN5LwxZnBhiyZ15ikH/Zp9YWa4YklUbVxl/YA01cJpltWR801uP/l3vc37W3pm+rZUmgFRnOPAZy78Y2Ux/1hZDJjDl+Nt5pyxciOVqZYKbLXmgofZWhHRK39vrtRMGte3N3cAxiTHsU6fQGzNmm7Hn/+yFIXOJZaP2V5iln6Ypu/EOv0ccMYPWH+Gmye/M48WYggYFlJVCxo6OaqeYiOjs1EvP+tCjEQScIlRo6LJwy+nVPC+/9soPQDTzoucU0rhs3QsXzcDrq2VrSyIqSZHmXWYNFcVKcoMIpwJmVTTEXBVHFQ/2v0hvvv3r3h2VQnLtnWWO2jzBUkJB3R7/RK6/iO4bQM44g7qNfuqI8PV8RkAnG/5lB+/YBZ/LW4wMz3/u2xnj2vr2/yAWdX/QP3i9S384vUtLN1URY3LS7ZqQFcW1uqTmEERSg/yYOBSs3FrBSz8/oBm+hKjbewklzhP5zZOe+rdrNzdwCJtIw/ZniT18/uIw2NmO5PyB6wvw1FBWiwGGg3Ek0orCbShKYMGo0uZjP3sdSnESCEBlxgVvIEQ9W1+lrS8ah44+vuQc2S3NkFrOOAKZ7i2VLbyvehl6FYnb4eOIi5Qx4RocxK3LT6NJks4KDrIlYpVLZ37Anr8ncFItwxXL0OKgLl1TWz6Qb3eoXDaLDQZsZHnq/XJzNV20bjhrW7tGsLBVVch3WCKKj3glZzNns573Pzc15Q0eMgw6iEui1BifuTc2df9nLYjroIpZ8HsKw7yHR0cpRQuFY9d90DQ7F9heOue6zPMTb2jmgrJ7ghIEwe2EO1wVWskkqGaSA7/fOfl5FAdrupPqgRcYnSQgEuMCturXYDBmNZ1MP96OO03PdoELOEJzv42mtx+iupczG//nPYJZ7JdH0uKcrEgoQVQEJfF1MlTAGiqLj6ovlSHa1kt0jbgrOicG9TkCZCyjyHFweCwatTRWUPqUct3AXPiM5jZOoAGd/eJ8Q3hifLvOu7i1PdOOKDX2lzR2uNYWqgGLTGX808I7zPoTGDauFxiz38MLn3usGyI3ZP+AAAgAElEQVQJ49LC2cR2c3izqM4MyGfp2wCI91Uz3hae3J8wdsD7MxyVG2nkqtrI4osrl8zhQv8vucr/E9nWR4waEnCJUeGdzVXkWFqwhtr3PoRhjUJHA18bb2+qYizVRAebUXnHUI352/g0YyfEZYLFxtWLptFsxNBYtafX2+m6wW/e2dZjRVtHMdVn7A9xwcYboN2cK9bk9g+5gCsj3slavXO/xjNOOpkmI5ZcVcvCB5bxdnjFYiBkdLvu1XUVWOgylBhoZ382lPfcoDrFX2HO0cpdAPE5sOgnfXwnfdemhYe/2htZsaOO376/gzHUEduynVYjmiSjmblR4aFhyXD1qsxII0fVRRad2ONSKTfSmXb8+YPcMyEOH6nDJUakz3fVs7mihRsXFQBQ2exlTlwTeIHk8b1fpBRthoNYn4syv4epFjOYsI85ghrD/EJNa9kYKWKZneCkykgm1tNz2xmA3fVt/HXFbj7bWc87tx0XOV7v8pPRpdr2c6+8yt+qxlHe1M699jawxZlDh0NAfmoMP//Ot2hc9xXJ2eOx2x2UGunkqRqqWryRPSa/qdHtJ7PLe6SpGNKn9tr2F69v5sSpGfzry1Ic+Pkf61N8oB/Jcn0Wsf46sx5Z2iS4fcsAvMP981jiQYdgWz0vfGUGjjfGrMAIKf4veA4/tT3PUWorWOwQc/iGe4eT7XouDmuQG6xvmweiUyh+cOAL9woxlEiGS4xIV/y/L3lgaWFkaKvO5WWKPTzPJrn3VW0Vze24iaK4soY6l4+pTrOcgTW1IDLfxBLyRSZGR9ktVBvJONt7r8X1YXhVZNe5SQAub4BJWudE+8qtX7C73o0/pDPG0txZw2mIWDI1k+TL/wqL78Ju1dioj2eOtqtbBmuB2kZ7fUnkuccfYry9S1X6xt63QKpt9fKPlSVc9dRqKprbOd2ymousK/ir/Q/kq/DnOoCrEA+E22KuOnxr1Rbe3lTFUWobV/I2TD+XLc7ZAEzzbTB3BzgMpTqGo9f0Y9mlZzNP24FhsUPc0PoZF+JwkH8dxIhk0RQAHxWaQU+dy8d4S+1+NzpuNaKx+FrM9tZ6cMRDVBLVRpdaQeFJvtF2K9VGMtHe3jNcDy4tBMzVh125fEHGhcsd+AwrszRzU2iFTrpRP+QCrq7sVo3P9BnEKi+LtfUAHKGKeMFxP8Zbd0TaefydJR0AfHVFeAM9Vysu/u0n3Z7Pj6mNPH7PcZf5YC8B8uHSbjOHFMsrK3Dg52nn79Bi01FL7sGWamZQbYZ/0Ps5lOlofKybwamRlA8WGVwRo48EXGJY+6iwJrKfXVdxTvMf9BfXlKPrBvVtfnKMakgcCxbbXu/XTCzOoBlw5apac2sdpSL7wQGQbk6Wt2iKOpVCdKApsoLtm6wEWeL/hJ17OjM8Lm+QfGsDKI0P9HnM1PZQoCpY5biFKaEd3fdRHGI0pVimz6XViGaxtgGAY7XNAIRqtvLlbjMr6PaHyLF0Dim++slq5t7/QY+gq2OV5lhVQ46q47jERgKx2WyxdakeP8gZLm844IoOtXJ91h6iDQ+c/RikFHDfJcdEMmCy2m7fdhtZAKhRvJG3GN0k4BLD2jVPr+Ge17eg652Tttt8QZo9AeYk+ZhR9hxPLy+kpT1ARrBiv1mIJiMOu98MuDL1GkjKA+C8OTmsTznTbDTh5M721lSzMnxb92HFjv78MHMTj9r/TNJHd0XOubwBcrUGiMtmt3MaWaqRF/LeIEOFJ41nzuzz5zHwDIJYKTHSyVF1HKW2RQqhWjz1XPLXlTS0+Wj3hxij6mk0YtmtZxLrr8PjD/XY5md8agznz0xhheNHLLf/kCTPHmw5c/lq0bMU6rm4xxw76JXIdUsUfmyo9kYW2Haac7XyzFWTucnRxCSES3hkTBvEXg5dS287jr9cOZea8LC8ypk/yD0SYnBIXleMCHsa3BSkmfWiyhrNWlmPJ/6TzPZl/HVZI3A5Se2lkLJon/dpMmKx+XfT5PGSElUFSWcB8IdLZkPoGQj6wOaMtG+xpkEAaK00s2dhbr85jDg/ZA67Wbyd29u4vEGyqIeEHK4+60L411OkVn/K7ti5ZE6YTfScKw/58xgoHXFtuZHGTG0P/3HcHzkXrXwk46K4wYPHHySTeqqMFFqNGDLCeyI2ewLkJHXezxsIMcm/FQCLMohr2wOp53LVMeMpn7aamOTow/be9sZutdCmxRHtb2Wcr8IMiLsualhyD6x+EmZcMHidHMKmZsUzNSuem/XZ/CRwPQ+d+IvB7pIQg0IyXGLYKGlw8581Zb2e6wiyAMqb2tHQSas3a1xdavmYfFWNLeSB1Il7vf8Fc3PMIcVAC2lGM1bd171yuMUGjthu17Taw6vSWiu7He+Yt5UbMvtr93TOJfuiqIE0vRYSc4mfsBAmnwlRSYy//lmiz/1Dj9cYSnTDjLjKjPRu1ec/DZn7GWaretr9ITz+EKl6A5VGCtUkkYk5n6vpGwsIfEGdsYGi7i+SOgmlFLlDINgCc95aoxFLknKR7CuH1EndG8w4H65ZKhtV74eBxguhE7r9wiLEaCIBlxg2zv7T59z50sbI/oPVLWY9q3GqCm+TGfAYhsGmihYmqnIsATf6vGuJVx5+bH3RvEnKhF7vDfC7i2cRdCRiJcgULRzY7WerFrfDDLhqK7vX4nL7goBBis/cZ8/ZXg2GwfIddWjopOr15uR9TYPL/gV37olspD2UdWQRy43OSvjnBe7n4aC53c4Y1cCG8mY2lreQEKjlyJnTqTGSSVdNXGz5mIw1jwBmgPzy2nIa3H5yvLtoUl22esnvLKExVFQH4xij6ony1g76nDIhxPAkAZcYFt7bUk1Luxlo7axpo7C6lYffK0RD52PHHSz50Bz6u/X5dTy2bCenJpgBk7bw+9TFTeUsS7ii+z4CLoB2qzmhd6YKT3LfT8BlOOLxGA7eWPFVt+Mub5AkXDhCbZQbqVh0P7Q3sbG8mQJ7ExYj2P3eSh3ApzD4pmbF89atx1JmdNabevT2aykNP5+r7WD19nKc+Mx98xLMLVwcKsjDtieZtP0JjICX4x7+mDteNCfdZ7XvpDJmGtf676B63p1DrnjopzvrqSGJ6VqJOV9P9kvsk1+cNY0zj8ga7G4IMWhkDpcYFm58dm3k8RvrK3hmpVnz6YZxjVAF9qCLnXt289ZGs1jppSlF0JQKyeNJO/l2eOV68+L9rAD0WMxtXL6VVg3Nap8lJAAcNitVRjKZqqnbcbcvRL4yy0V8rU8kx1KP4a6nusXLvLgmcLP3AqxD3JjEKIqMcOmK/OPIS4nBGZ9Coy+WG61vs7B2N7ep6wCwJuVQbXSvMl+ybXXkcQotpLQXk3jMuZyddi0Zs4ZmSYyarmVBpPxDn1xz7DiuQT47MXpJhksMec+v7qxmrqHz3MrOOT+LozofF24wv8g3LVpDdsW7MPtyM3M0/TxzmOq0B/dbmLI+ZM7DyW0vNOthHcB8k2ojmUzV2O1YWZOHvC4BF8Avnv+ElUUNTLCEVzSmFOz33kNRlN1CqZHByb6H4cpXAFh2x2KcaWYAeYS+jQJlDvHGpI3lxGOP6Xb99rXLAZiqSljrvBmNELa8BZwzewxqiGb6Ihstg2S4hBB9IgGXGPJW7KgD4KHj7RSm380H9v8mGXNPtgLXGtyYQZFRu4286ABxa/4EU78FJ9xt3sBig6vfgqNv3u9rNWNmuKJ9dWYNrv1QCnNSuGrkla/N+VpvbKjkp69sIl+rxkCxQTcDq9rqSly+IJONYohKhrjhObzisJr/bOw0ciIbD8c6rFgvfpq3Q0cBsET7GgAVP4aLTj8ZbNGs1wuoM+LxFpvDr7ekb+i8ae6Cw/gODk5GvMN8rx1i0gavM0KIYUsCLjGk/f797SzdXM0d40q55KuLsbeWME6r4afWf5FOE+n1K1mVdhEtRjSqrpBT4vZAyA8LburTaqiOgAs44CG/SiOVTBq58z9rue/Nrfzg+XUATLTWoRJyqQoPR3Vs3Ds+sAOyZw+beVvf1JGFmpIZ1+24Pb2AXwWuBuBy68egNDNLqBTcUcgl/nvYqBcw1dgFQEFgJwB/mfI0RA3dYphv3HIsP7s+XKojfdqw/f8mhBhcModLDGmf7TJLD3wn+JKZcbriJVj3LBd9/igXWVeAAXELvs2ON74k3VfMrI5aWBkz+vR6DXqXIOIACpAqoMTIwKp0clQdT33e+VfqhPQ2iBlHfU0CAcPCGFWHAz8Zvj2QfU6f+jdUfHj78WTE9wxo60hktT6Zo7TtMGYe2KLME84EfNjZoBdwgnU9cXjI8RXxQnAxpY59L2QYbBnxTjLis+GGT2Q4UQjRZ5LhEkOaN6Bz3iQbCXVrYNZlkDoBFt1pzssauxAW/4zMcTPZpY+hQFUygTKIz+lzxiREl+zFmLn7bb94cjp79EyAzs2WgTg8RDUWQvrUcGX2DApUFTPVbixGCLJm96l/Q8WE9DjinD23SFowLpmtSSeaT6Z1DyqtmmK1MQVNGZykrSU21EyRkUV6nKPHfYak7DkQlbT/dkII0QvJcIkhrT0QYmLQHIIi/1jzT3sMXPR0pM0Y3aDYyCRVtRLdtgFy+pbdArOS+kOBS7m5oJ74A9iC5DsL81i+bgrUwThVTbrlY7JVAzPGJKJq2mHWpcwqclNUk81sbRcPxb2AEbKjxu+74v1w9cKNCyE0D3YcA5NP73ZuxZ0n8PiyDAIbH+AMy5cAnHbsUcw8YWhnuIQQoj9IwCWGHF038AV1ouwW2v0hxgXDNbEyew+kLJqi2DCzTNHtlZB+Ud9f2zB4PHQ2l5y9mPgDmKujlCIhJYvW2ihmaMVcYPnUPFEDTD8fsufw8s06ZcsvJ3PF7eBvgtMeAmfCPu87rFlsMPWsHoezE6OYV5BJ5YYUFmiFAMydMQMskmgXQox8+/2XTin1lFKqVim1ucuxZKXUB0qpneE/k8LHlVLqMaXULqXURqXU3C7XXBVuv1MpddXAvB0x3LX7Q5z1x8848XefYBiGuSdfsAJiM/cZpBQaXeplZc/p8+t3bF0Tbbcc8DVKmQFfR7DVYMRhTDkLTn8YAKtFY9yiq2Dc8XDUjXD0TX3u33AX57RSbqQRr8JbMQ2D6vpCCNEfDuRXy6eB075x7C5gmWEYE4Fl4ecApwMTw//dADwOZoAG/BJYABwF/LIjSBODJ6SbAY07vO/fYCuqa+OSv65ka1UrlS1evi5txhvQSQlU73ey8h+/d37nk0MoMZCTZE7ydh5EwBXUDf4cPIedjuk8m3kX12e8gLr0OYjtUj7AYoWr3oQzHu5z30aCWIeNMsP8XAzNCrEZg9wjIYQ4PPY7pGgYxgqlVP43Dp8DLA4/fgb4BPhJ+Pg/DMMwgFVKqUSlVFa47QeGYTQCKKU+wAzinj/kdyD6xDAMCn72DgBH5CTwxi3HDnKP4MkVu9lY3sI5s7N5fX0lFzz+BQBJ/gpIWrzPa4/ITYLvvgt6wCxF0EdPXT2fL3c3Et/LhPC9OXdONtdsOIrLLriFb09K49t9fvWRL9Zh7dwWyJkI2oEHtkIIMZz1dfJEhmEYVeHH1UDHr6ljgLIu7crDx/Z2XAyCVm+A97Z0rqjbWN5ywNfWuXxc8/RXVDS377/xQapu9TJzTAK/PjrIh/Yfc7llGTaCxPpqD6gIKXkLzWG7Q5Ae5+RbB7m9zJIpGRTefxqLJklBzP2Jc1ojGS5l6IPcGyGEOHwOedK8YRiGUsroj84AKKVuwByOZOzYsf11WxHmC4Y45sGPcHk7hxFTYw9sWX6ty8sJj3yC2x9i0dYarvqv/H7r17Mri/lkex2nTkok9qUrmaBV8xvtb1QYqcNiw2CnTTI1ByLWYY0scOhtYr0QQoxUfc1w1YSHCgn/WRs+XgF03e03J3xsb8d7MAzjr4ZhzDMMY15ammQM+tu7m6u7BVs/nhXkN8FHoL15v9f+7dM9uP0hAGz9uLLMMAzueX0LAPm170NbNR9N/DkAV1neMxslHUCGSwx5sU4rG43xXOD7JZz16GB3RwghDpu+fmu+AXSsNLwKeL3L8e+EVyseDbSEhx7fA05RSiWFJ8ufEj4mDrM99W4AovBy7njFzTuu4RRWYWw1/xeuK20i/663Kapr63Htip31HJGTgEKnrd3bb30qb+ocnrzGvgxSJrI793x26GNYYllvnkgenhs9i+7MQF2x1pgs87eEEKPKgZSFeB5YCUxWSpUrpa4FHgROVkrtBE4KPwd4B9gN7AKeBL4HEJ4sfz/wVfi/+zom0IvDq7ypnYK4EJvS7uHRystQ4TII+u4VANz9qln944vwljodgiGdoto2jsmP4w37zzlm8y/6rU9ljWaJgFfPiyGjdRPMv5Zz5+bQOvakzkZxmf32emJwPXbZHL63WAJoIcTociCrFC/by6kTe2lrAN/fy32eAp46qN6Jflfa4OHM6C1YWyrAmciX43+Ad9PrHFe7jX+vLmVrlbnBcnsg1O26R97fjj+ks8T/ETO1Yqgvhra67qUP+qDNF6Q0HHAVFD0D1iiYdRmpUQ5SL74LHvsPFCyRDYNHkLNnZXP2QS5MEEKI4U5KPI8i3kCI9eXNLLIXgiMe7txNZcElbDdyUfU7uPsVc/juSLWdpF2vdbt2xY56FDpzyp/rPFix5pD7dOT9H3DXK5tIopW4na/BvGs690GMz4a7SuHiZw75dYQQQojBJAHXKBEM6awtacIf1JkY2gVZs0CzkBRjY4eeg8UIkK+qSaKVlx33clHpfdBirmvYUNbMtqpWHplVg7VxJ3+MuRUdDSrWHlKfvIEQvqBZGuAEbb1ZJmDmhd0bWe3mVjFCCCHEMCYB1yiwqbyFCXcv5fcf7MCphYhr2QHZswGYnp3A9vC2OFNUGQ/ldWatgiUrAVi2rQaAs90vQ3wO2uwr2K7n4Cs5tAzXhrKOlZEG1yWuhbgsyJp9SPcUQgghhiIJuEawD7fWcP9bW3liRREAa0uaOCmlERXyR/YbzIh30hA9nlYjmtMtX7K45XXW22bjMRys/XQpYE60Pz6+Bnv5F3D0Tcwel8Z6vYBA2Vow+l6C7b0tNThtGjuO+5Rpnq9gzrdBkx9JIYQQI498u41Q68uaue4fa/jbZ3t4a2NV5PjxceHyZ10ySavuOQP/tAs4y/Ildm8dz9ouZINeQFKjOaertNHD6fZweYYjLmFscjQbjfHE6q2EGvbssx83/GMNf/hgR6/nyps8TE1W2L9+CiadBot+cgjvWAghhBi6DrnSvBia7ntzS+TxFQvG8tyXpQA4510OmWdC0rhu7VPP+hXUfAG5C/hs21QmG+O4OvQ+u2uaWVvaxMNp2yBjBsSmk60bbNDNZf1lmz8lf/H4XvvgD+q8v7WG97fWUNLgZnxaLC5vgDtOmUxlcztNHj/HWrZCyA8LbzE3eBZCCCFGIPmGG6GaPAEA3rr1WGaMSYgEXKfMHAu9bUMTkwq3fAVK41u2bWz/Ihc7AYq2bwRDJ8+7DSZdBIBFU/zg0m/hfeWXLPtwKU+vzuV3F81mVm4CHxfWcer0DAwDShvdkdu/tr4y8rjZE+DFteUAXJ+1HTQb5B41UB+FEEIIMegk4BqB/EGd0kYPty6ZwIwxCd3O7XPPv3Dl75+eMZVbtk0GN3jLNzLZYsPid0HO/EjT46Zks8M6mSXG1zzc2MKP/r2OyVnxfFRYy1UL83hpbTm3njgRgHSaiFFe9hhZACzd3LlxdkFgF2RMB+uB7ecohBBCDEcScI1AxQ1uQrpBQVps5Njz1x+N3XpgxUMtmoK0KYTcGtb6QpbExIOfbgFXjMPKjAvvRnvhMtZH3UTIC7fvuBmYzzMrSwB4cGkhCRYfy6J+QXSwidXpl7C+so0P/XPZzDgWalvJc2+EaVf347sXQgghhh4JuEagtzdWoRQcmZcUObawIOWg7pGbnsSe3ZlY6rZydHwqWJIhZUK3NtrUM+Dbr+Je+S+sO9/hZuubvO+fFz5rBndXp+4krqUBolNZWPs8C61wpeVDVlrmcYr+KRjAhJMQQgghRjIJuEagL4rqmZObSG5ydJ/vMT07nkIjl5lqD3ZvFUxY2Pv2OgVL8CYfzXO/tfAT27/Z6LieGiOJXwW/w7WWpczzlEN0Kvx4B5Sv4c9vfsb36u4zgy2AedfChJP73E8hhBBiOJCyECNMSDfYVNHCrNzEQ7rP9Ox4CvWx5Gm1ZIUqIW/hXtumxzl5I7QQ3VDEKw8TtQqesz/AEst64gP1MO0cc37Y2AVcdd1t1Ey4BFInwx074KzfS+0tIYQQI55kuEaYNm8Qb0AnJ6nv2S2AcamxLNPn8mNeNA+MX7zXtnarxkkL57E29UXsug/vB/ezQCuEM35rNph9RaRtjMNKzBVPmE9kQ2ohhBCjhARcI4zLZ5aDiHMc2v9ai6Y497RTKaxvZEqyBTJn7rP9vefMAGYAsCJhKt7oCpwFx/UeVEmgJYQQYpSRgGuEcftCgJlJOlQ3LioAfn7Q1x1/RAFQcMivL4QQQowUMnlmhGkLZ7hinRJLCyGEEEOFBFwjjMsbBCC2HzJcQgghhOgfEnCNMB1DihJwCSGEEEOHBFwjjAwpCiGEEEOPBFwjSKPbz+OfFGGzKJKibYPdHSGEEEKEScA1gny+q57iBg+3nzyZaLtkuIQQQoihQgKuEWR7tQuLprjm2PzB7ooQQgghupCAawTZU+8mLzkah9Uy2F0RQgghRBcScA1xId3gjv9s4J7XNu+3bV2bj9Q4x2HolRBCCCEOhgRcQ5hhGPzi9c28/HU5z64qoc7l22f7+jYfabEScAkhhBBDjQRcQ9izq0p47stSJmfEAVDc4N5rW8MwKGnwkBprP1zdE0IIIcQBkoDrMKlz+fjB8+vYUePab7stlS34giH++NEuFoxL5tFLZwNQ22pmuEK6wYoddRiGEbnuz58UEdINkmMkwyWEEEIMNRJwHSbvb63mjQ2Ve52LFQjpNLr9XPbkKs587DP+s6acOpePW5ZMICPeCUBNqxeAl9eW852nVvOv1aUA/P797Tzy3nYALpyXcxjejRBCCCEOhhRrOkw2V7QAnXsddqXrBjc+u5aPCmsjx+55bTNzxiZy7IRUAGwWxZOf7mZiRix3vrwRgH+uKuXyo8by/tYaAL64awnZiVED/VaEEEIIcZAkw3UY6LrBx4V1gDkPq+tQIMCr6yq6BVsd7jlrGkoplFIcPT6FqhYvT322J3J+W1UrK3c34PIGOX/OGAm2hBBCiCFKAq7DoKK5nepWLwVpMXj8IVrbO7Ncn+6s444XNzAhPZbN957KmUdk8fer5/PRHYuYOzYp0u6648YD8PH2Oo6flMb6X5xMZryTy5/8kormdjISnIf9fQkhhBDiwEjABXy5u4EVO+oG7P7V4blXs3ISAWhwm5Pfy5s8fPtvqwE4b84YYh1W/u/yuZwwJZ3xabHd7jE9Oz7y+IwZmSRG2/nb1fOYkmmuYDxhcvqA9V8IIYQQh0bmcAGX/HUVAHseOAOlVL/dt7K5nc921hPQdQCmZcfzyroKGtx+8lMMHv+kCIDLF4zlOwvz9nmv1FgHZ87Motbl5ZL5uQBMz07g3R8e32/9FUIIIcTAGNUBlzcQ4n+X7Yw8L2nwkJ8a0y/3fnltOXe8uKHbsenZCQA0tPm5980tPPdlKSdNzeA35808oHv+6fI5GAb9GhQKIYQQYuCN6iFFh1WLZJkAvihq6Jf71rZ6I8FWnLMzps1PjTbPu7x8sqOOiemx/G+4xtaBUEqhaRJsCSGEEMPNqA64lFJ8/4SCyPOfvbqJ9WXNB3x9IKTz+voK2nzdSz1sDxc3/cuVR7LpV6fy+BVzeecHx5ER5yQvJZpfvL6FkgYPVx6dR4xjVCcZhRBCiFFhVAdcAD86aRIbfnkKt504EYA3N1QC5nDj/jz12R5u+/d6ZvzyPW59fh2BkDlXq7je3IJndq45Sf70mVlMy45H01Rk/tXxk9K4YsHYfn8/QgghhBh6Rn3AZbVoJETZ+OFJE8lNjqK43s0XRfVMuedd1pY07fW6OpePB5YWRp6/uaGSufd/wOaKFpZuribOYSUjvuc2O1csyOOmRQX88bI5WC2j/uMXQgghRgUZzwpTSnHshDReW1cRKSD61sZKjsxL6rX9q+vKAZg7NhGrRcNh1fh0Zz1n/fEzAE6bntnr5PaEKBt3nT5lgN6FEEIIIYYiCbi6OHV6Bs+vLuXj7WbV9zqXb69tK5vN2lr/uv5onDYLvmCI655Zw6c76wH42RlTB77DQgghhBgWZEyrizHhzFZ5UzsALe2BvbataG5nckYcTpsFAIfVwuNXHhk5PzYlegB7KoQQQojhRAKuLtLju2+P07qXgCsQ0tlU3kLeN4KqWFlxKIQQQoheSITQRbzTSpTNQnt4hWJze4CQbtDg9pEe5+ShdwspaXBz7bHjqG71cs/saT3u8ZPTpmBg9DguhBBCiNFLAq4ulFIUpMewuaIVgGZPgF+9sYVnV5WgFBjhOOqUaZkATMqI7XGPmxcX9DgmhBBCiNFNhhS/YdGkNACSY+y0egM8u6oE6Ay2AP6y3KxOnxWe8yWEEEIIsS+S4fqG206cxLSsBBo9fu55bXO3c5nxTqpbvRRWm5XkZc6WEEIIIQ6ERAzfYLdqnHlEFrtqXZFjp8/I5KenT8ViURzz4EcALJmSPlhdFEIIIcQwIwHXXuSlxEQe333mVHKSotH1znHFW5ZMGIxuCSGEEGIYkjlce2Hrsu1OcowdAE1TZIZLR+R3CciEEEIIIfZFMlwHINre+TG9dPNCWtuDkSBMCCGEEGJ/JOA6SDlJ0fD/27vfUD3rOo7j7w9bzjRwW4XUJjppFGtUiuSiiNDQaeIKfLAYtGrQkyCLIBx7ID3oQRRZQRmhpYVotKyG0J+1hB650gpbTpiXF48AAAYMSURBVNsxyz/MtnKuKEilbw+u35m38zh37+Y6933O/X7Bxbmv33Udzu/+8j3n+pzrzzlz/3tFSZKkORm4TuBzH1h/7N/8SJIknSoD1wlsufjccU9BkiQtAt40L0mS1DMDlyRJUs8MXJIkST0zcEmSJPXMwCVJktQzA5ckSVLPDFySJEk9M3BJkiT1zMAlSZLUMwOXJElSzwxckiRJPTNwSZIk9czAJUmS1DMDlyRJUs8MXJIkST0zcEmSJPXMwCVJktQzA5ckSVLPDFySJEk9M3BJkiT1zMAlSZLUMwOXJElSzwxckiRJPUtVjXsOLynJYeCv8/ClXgP8fR6+zmJmDUdj/UZnDUdnDUdnDUez0Ot3blW9dq4NEx245kuSe6vqonHPYyGzhqOxfqOzhqOzhqOzhqNZzPXzkqIkSVLPDFySJEk9M3B1vjnuCSwC1nA01m901nB01nB01nA0i7Z+3sMlSZLUM89wSZIk9WyqA1eSjUkeSjKT5Lpxz2dSJTknyd1JHkjyxyTXtvGVSXYnOdA+rmjjSfLVVtf7k1w43ncwGZIsSfK7JHe19TVJ9rY6fS/JaW18WVufadvPG+e8J0WS5Ul2Jnkwyf4k77AHh5PkU+17eF+S25Ocbh+eWJJvJTmUZN/A2NB9l2Rr2/9Akq3jeC/j8hI1/EL7Xr4/yQ+TLB/Ytr3V8KEklw+ML+hj9tQGriRLgK8BVwDrgA8mWTfeWU2s54BPV9U6YAPw8Var64A9VbUW2NPWoavp2rZ8DLhx/qc8ka4F9g+sfx64oareABwBtrXxbcCRNn5D20/wFeCnVfUm4K10tbQHT1KSVcAngIuqaj2wBNiMffhybgE2Hjc2VN8lWQlcD1wMvB24fjakTYlbeHENdwPrq+otwJ+A7QDt2LIZeHP7nK+3X1YX/DF7agMXXdPPVNWfq+oZ4A5g05jnNJGq6mBV/ba9/hfdgW4VXb1ubbvdCry/vd4EfKc69wDLk7xunqc9UZKsBt4H3NTWA1wC7Gy7HF+/2bruBC5t+0+tJGcB7wZuBqiqZ6rqaezBYS0FXplkKXAGcBD78ISq6lfAU8cND9t3lwO7q+qpqjpCFzaODyCL1lw1rKqfV9VzbfUeYHV7vQm4o6r+W1WPADN0x+sFf8ye5sC1CnhsYP3xNqYTaJcVLgD2AmdX1cG26Ung7Pba2r7Yl4HPAP9r668Gnh74gTNYo2P1a9uPtv2n2RrgMPDtdln2piRnYg+etKp6Avgi8Chd0DoK3Id9eCqG7Tv78cQ+CvykvV60NZzmwKUhJXkV8APgk1X1z8Ft1T3u6iOvc0hyFXCoqu4b91wWsKXAhcCNVXUB8G+ev4wD2IMvp13C2kQXXl8PnMkUnWXpi303miQ76G5buW3cc+nbNAeuJ4BzBtZXtzHNIckr6MLWbVV1Zxv+2+xlmvbxUBu3ti/0TuDqJH+hOw1+Cd39SMvbpR14YY2O1a9tPwv4x3xOeAI9DjxeVXvb+k66AGYPnrz3Ao9U1eGqeha4k6437cPhDdt39uMcknwYuArYUs//japFW8NpDly/Ada2J3ROo7tJb9eY5zSR2n0bNwP7q+pLA5t2AbNP22wFfjww/qH2xM4G4OjA6fepU1Xbq2p1VZ1H12e/rKotwN3ANW234+s3W9dr2v5T/Rt0VT0JPJbkjW3oUuAB7MFhPApsSHJG+56eraF9OLxh++5nwGVJVrQzjZe1samVZCPdbRZXV9V/BjbtAja3p2TX0D2A8GsWwzG7qqZ2Aa6kezriYWDHuOczqQvwLrpT5vcDv2/LlXT3c+wBDgC/AFa2/UP3NMnDwB/onooa+/uYhAV4D3BXe30+3Q+SGeD7wLI2fnpbn2nbzx/3vCdhAd4G3Nv68EfACntw6Bp+FngQ2Ad8F1hmH75szW6nu+ftWbozrdtOpe/o7lOaactHxv2+JqCGM3T3ZM0eU74xsP+OVsOHgCsGxhf0Mdu/NC9JktSzab6kKEmSNC8MXJIkST0zcEmSJPXMwCVJktQzA5ckSVLPDFySJEk9M3BJkiT1zMAlSZLUs/8DRMD0Q/yYG6gAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 720x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CddNEPoaIYsi",
        "outputId": "911a314b-a098-4218-8df9-f9149073de7b"
      },
      "source": [
        "#Predict Stock price for the future 30 days\n",
        "X_input = test_data[228:,:].reshape(1,-1)\n",
        "X_input.shape"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1, 150)"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pU5SuszvKaE8"
      },
      "source": [
        "temp_input = list(X_input)\n",
        "temp_input = temp_input[0].tolist()"
      ],
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7DyO3MiHLoIg"
      },
      "source": [
        "lst_output = []\n",
        "n_steps = 150\n",
        "i = 0\n",
        "\n",
        "while(i <30):\n",
        "  if(len(temp_input)>150):\n",
        "    x_input = np.array(temp_input[i:])\n",
        "    x_input = x_input.reshape(1,-1)\n",
        "    x_input = x_input.reshape((1, n_steps, 1))\n",
        "    yhat = model.predict(x_input)\n",
        "    temp_input.extend(yhat[0].tolist())\n",
        "    lst_output.extend(yhat[0].tolist())\n",
        "    i = i + 1\n",
        "  else:\n",
        "    x_input = np.array(temp_input[i:])\n",
        "    x_input =  x_input.reshape((1, n_steps, 1))\n",
        "    yhat = model.predict(x_input)\n",
        "    temp_input.extend(yhat[0].tolist())\n",
        "    lst_output.extend(yhat[0].tolist())\n",
        "    i = i + 1"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "41DVrCE3N9Eo"
      },
      "source": [
        "predict_df =df.flatten().tolist()\n",
        "predict_df.extend(lst_output)"
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        },
        "id": "yC3nURdsYGAP",
        "outputId": "1f465c6c-2209-4b40-b24b-4b9671573a1a"
      },
      "source": [
        "plt.plot(scaler.inverse_transform(np.array(temp_input).reshape(-1,1)))"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7fae06969510>]"
            ]
          },
          "metadata": {},
          "execution_count": 22
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD4CAYAAAAAczaOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3dd3hU17Xw4d8a9d4LCCEJkADRQTT3ju3YwXZix05c45g4sW/iFOfGaU7PdeLYN86Xa8e995Lg2ImNG26A6VUIhISQhHrvZWZ/f8yRLIEEkhhpitb7PHoY7XPOzDoDrNmzzzp7izEGpZRS44PN3QEopZQaO5r0lVJqHNGkr5RS44gmfaWUGkc06Sul1Dji7+4AjiU+Pt6kp6e7OwyllPIqmzdvrjbGJAy0zaOTfnp6Ops2bXJ3GEop5VVEpGiwbTq8o5RS44gmfaWUGkc06Sul1DiiSV8ppcaR4yZ9EQkWkc9EZLuI7BaRX1ntGSKyQUTyReQFEQm02oOs3/Ot7el9nusOqz1PRFaM1kkppZQa2FB6+h3AWcaYecB84HwRWQbcBdxrjJkG1AE3WvvfCNRZ7fda+yEi2cCVwCzgfOD/RMTPlSejlFLq2I6b9I1Ts/VrgPVjgLOAl632J4BLrMcrrd+xtp8tImK1P2+M6TDGFAL5wBKXnIVSSqkhGdKYvoj4icg2oBJYAxwA6o0x3dYuJUCK9TgFKAawtjcAcX3bBzhGKaWU5V87DrN6++FRee4hJX1jjN0YMx+YhLN3PmNUogFEZJWIbBKRTVVVVaP1Mkop5ZHyK5v40cs7eGrdQRwO1693MqzqHWNMPfA+sByIFpGeO3onAaXW41IgFcDaHgXU9G0f4Ji+r/GgMSbHGJOTkDDgXcRKKeWTWjq6ufnpLYQE+PHXqxZis4nLX2Mo1TsJIhJtPQ4BzgVycSb/L1u7XQf803q82voda/t7xrk812rgSqu6JwPIBD5z1YkopZQ3M8bwo5d3cKCqmb9cuYDkqOBReZ2hzL0zAXjCqrSxAS8aY/4lInuA50Xkt8BW4BFr/0eAp0QkH6jFWbGDMWa3iLwI7AG6gVuMMXbXno5SSnmn+97N542dZfzkwhmckhk/aq8jnrxGbk5OjtEJ15RSvu4/u8q4+ektXLYghT9fMQ9nwePIichmY0zOQNv0jlyllHKjPYcb+d4L25mfGs3vL5tzwgn/eDTpK6WUmxyub+MbT2wkKiSAB69ZRHDA6N+vqklfKaXcoK6lk2sf/Yym9m4euT6HxMjRuXB7JI9eREUppXxRa2c3Nzy+kUO1rTxxwxJmTYwas9fWnr5SSo2hzm4H33p6CztK6rnvygUsnxo3pq+vPX2llBojHd12vv30Ftbuq+IPl83h/NnJYx6DJn2llBoD7V12Vj21mQ/3VfGbS2Zz1ZLJbolDk75SSo2y1s5uvvHEJtYV1HDXl+bwlcXuSfigSV8ppUZVTXMHNz25iW3F9fz58nlctnCSW+PRpK+UUqPkQFUzNzy2kYrGdv721YVcMGeCu0PSpK+UUqNhQ0ENq57ajL9NeG7VMhZOjnF3SIAmfaWUciljDI9/epDfv5nL5NhQHr9hCamxoe4Oq5cmfaWUcpHG9i7+++Ud/HtXOefMTOLPl88jKjTA3WH1ozdnKTUObS6q5aev7cQ+CiszjVfbi+u5+K8f8/aeCn564UweunaRxyV80J6+UuNOcW0r33hiE3WtXVy9LI2ZEyLdHZJXa++y85d39/P3tQdIigzmhVXLyEmPdXdYg9Kkr9Q4UdvSyXt7K3nwwwO0dTnXL9p6qF6T/gnYeqiOH728g/2VzVyRM4mfXZRNZLDn9e770uEdpcaB2pZOLvjLh/zwpe3UNHfy4DU5xIYFsq24btjP5XAYnvj0IHUtnaMQqXeoae7gv1/ewWX3f0pzRzeP37CYP355nscnfNCevlI+r2ft1bqWLp79xlKWTYnDZhPmp0az9VB9737v763kT2/l8devLmBqQvigz7e1uJ47V+9mb3kjf7hs7licgsfosjt4Zn0R96zZR2unnW+cksF3zs4kwguSfQ/t6Svl417bWso7uRX86PzpnDQtHpvNuTLTgtRo8quaaWzv4mB1C995fit7yhq57fltdHY7Bn2+rYec3w5e2lTCoZrWMTkHd7M7DP/YWsrZf17LL1/fw7zUaP5z26n89AvZXpXwQZO+Uj5v7b4qJkQF8/WTM/q1z58cjTHwaX41Nz+9GZsIv7gom52lDfzsHzspa2gb8Pm2HKojPjwQfz/hL+/uH4tTcBu7w/DmzjIu/MtH3PbCNsKC/Hn0+hye/PoSpiVGuDu8EdHhHaV83P6KZqYnR/T28HvMS41GBL73wnY67Q4evX4xp2clUFrfxiMfF/LS5hLuuWIely7oP1fMlqJ6TpoaT3x4EI9/WsgvLs4mKsS7ervH095l5+XNJTz8UQEHa1rJiA/jr1ct4AtzJhz1Pnob7ekr5cPsDsOBqmYyE48eo48MDmBaQjhtXXb+fPk8Ts9KAODnF2Xz/g/PYGJUCP/ZVd7vmMP1bZQ3trNwcjTLp8bhMFBQ1Twm5zIWcssa+fXre1j+h3f52T92ERUSwP99bSHvfP90Lp430esTPmhPXymfVlLXSke3g2kDJH2An3xhJm2ddi48YiKwjPgwZk2M5EBVS7/2LdZ4/sK0GMKCnOmjsLqFBR4yr8xIFFQ1s2ZPBf/aUcbO0gYC/ITzspO5ZnkaSzNiEfH+RN+XJn2lfNj+CmcvfLDx5zOnJw567NTEcN7Pq6TL7iDAzzkosKWonuAAGzMnRGIM+NmEwuqWQZ/DEzkchm0l9by9u4I1e8p7P9hmp0Ry58XZrJyfQmxYoJujHD2a9JXyYfsre5L+4CWYg5maEE6X3XCotpUp8WGs3VfFv3eVMTcluvdDIDUmhIIq70j6O0rqeXFTMW/vrqCyqQN/m7B0SizXLEvjnOwkJsV4zqRoo0mTvlI+yBiDiLC/somkyKARXWjt+aA4UNnM+3sr+e0buaREh3DbuZm9+2TEh1FwnJ5+W6edxz89yLXL03qHhMaKw2F4e085939wgO0lDYQE+HHG9ATOm5XEWdOTPHJunNGmSV8pH9PeZefyB9Zxzswk8iubyRxhaeGUhDAADlS18ObOMuZNiuKlm08i0P/z+o+M+HDWF9TicJhBL3I+s6GIu/6zl+AAGzccUTY6mkrqWrnt+W1sKqojPS6UX6+cxSULUrzirtnRpElfKR/z3GeH2FnawO7DDfjZhK8tTRvR80QGB5AYEcT6ghp2ljZw+4rp/RI+OD8Y2rrslDe2MzE65KjnsDsMT6w7CMCrW0rHLOnvLW/k8gfWYQz88ctzuWxBCv5+WqwIWrKplE9p67Tzt/cPsCgthuTIYLrshsyk4Y/n95iaEM7afVUAnDE94ajtU+Kd3wYGu5j7Tm4FxbVtLJ8Sx87SBvZVNI04luPp6LbTbk0kd++afQjw5ndO5YqcVE34feg7oZQPeXp9EdXNHfz4ghnc9eW5BAfYTmiZvp5x/cSIILIHmI0zwxoCGmxc/7FPCkmJDuF/r5yPn014dUvpiGMZTLfdwcMfFbD8D+/xhfs+YkNBDW/truD6kzOYHDc+Ls4OhyZ9pXyEMYbnNh5iSXosi9NjOTUzgd2/Ov+Epk6eaiX1M6YnDFivnhQRTEiAH4UDVPDkljWyvqCWa5enkRQZzBlZCaze5vqk//LmEn77Ri7TEsMpqWvjaw9vICTAjxtOSnf5a/kCTfpK+Yg9ZY0UVLWwcsHE3ja/E7yDdHqy8wPjrBlJA2632cSq4Dn6rtzHPikkJMCPKxdPBmB+ajSHG9rp6LafUExH2l/ZTEiAHy+sWsZ9Vy3AYQxXL5tMjA/X2p+I4yZ9EUkVkfdFZI+I7BaR71rt80VkvYhsE5FNIrLEahcRuU9E8kVkh4gs7PNc14nIfuvnutE7LaXGn9e3l+FvEy6YPeH4Ow/RsimxPL9qGStmDZz0AaYnR5Bb1tivraa5g39sO8xlC1N6yyJ7knBdS5fL4gPn1BATo4MREVbMSmbt7Wfy4wtmuvQ1fMlQevrdwA+MMdnAMuAWEckG/gj8yhgzH/iF9TvABUCm9bMKuB9ARGKBO4GlwBLgThHx3nu3lfIgxhhe336YUzLjXXo3qYiwbErcMacimJMSRUVjBxWN7b1tz312iM5uBzecnN7bFmfFVevixVecSf/zyqHU2NAT/objy46b9I0xZcaYLdbjJiAXSAEM0DNYGAUcth6vBJ40TuuBaBGZAKwA1hhjao0xdcAa4HyXno1S49S24npK69u4eO7E4+/sYnMmRQGws6QBcH4Avbq1lKUZsf2mf4gZpaRfWt9GygDlompgwxrTF5F0YAGwAbgN+JOIFAN3A3dYu6UAxX0OK7HaBms/8jVWWcNFm6qqqoYTnlLj1hZrBazTso4uqxxt2RMisQnsLHUm/b3lTRRUtXDxvP4fQL09/dbhJ/3yhnZ+9o+dVDa192tv77JT3dw54D0CamBDTvoiEg68AtxmjGkEvgV8zxiTCnwPeMQVARljHjTG5BhjchISxv4fsFLeKL+ymZjQAOLDx/7iZViQP9MSw3uT/hs7yrAJnD87ud9+vT395o5hPX+33cF/PbeFp9cf4v+9l99vW1mD80NAk/7QDSnpi0gAzoT/jDHmVav5OqDn8Us4x+kBSoHUPodPstoGa1dKnaADlc1MSwx32zTAs1Oi2FnagDGGN3aWsXxqHPHhQf32iQ4JQARqW4d3Iffed/ax8WAdWUnhPL+xuF9v/3C9c3UvHd4ZuqFU7wjOXnyuMeaePpsOA6dbj88CetZNWw1ca1XxLAMajDFlwFvAeSISY13APc9qU0qdoPyq5hHNpOkqc1OiqGrq4LWtpRRWt/CFOUdfW/D3sxEVEkBty9B7+u1ddh76sJBL5k/kwWty6LY7eOSjwt7tpZr0h20oc++cDFwD7BSRbVbbT4CbgL+IiD/QjrNSB+BN4EIgH2gFbgAwxtSKyG+AjdZ+vzbG1LrkLJQax2pbOqlt6WRqgvuS/pxJ0QB8/8XtpESHcOGc5AH3iw0LHFbJ5paiOjrtDlYuSCE9PoyL5k7kmQ2H+NH5M/CzCYfr2xCBpKig4z+ZAoaQ9I0xHwODfWdcNMD+BrhlkOd6FHh0OAEqpY4t/wTmzHeVWRMjSYgIYt6kaO6+fC7RoQNfW4gNDaRmGD39dQU1+NmExemxAJwyLZ7V2w9TXNtKenwYh+vbSAgPIsjfzyXnMR7oLJtKeThjDPe+s5+spHAuGqAk0xOSfnCAH+vvOPu49fGxYYEU1bQO+XnXHahhTkoU4dY8/FnJzhLQvIomK+kPPLunGpxOw6CUh3t5cwn3vbuf+z84MOD2/ZVNhAT4MTHKvclvKDdExYYFUjPEOv3Wzm62l9SzfGpcb1vPAu/7rdk6D2uN/rBp0lfKgxVWt3Dn6t0E+tnYU9ZI/QA17vlW5c5gi5h4ktiwQOpaO3GOAh/b5qI6uuyGZVM+T/phQf5Migkhr6IZYwyl1hQMaug06SvloVo7u/nW05sJ9Ldx9xXzMAY2FB5d+9BTrukNYsMCsTsMjW3dx9133YEa/G1CTlr/2VqykiLYV95EdXMnHd0OHd4ZJk36SnkgYwx3vLqTvIom/nLlAlbMSiI4wMa6AzX99mts7+JwQ7tXJX0Y2l25n+RXMz81+qh1dbOSIiiobubtPeUAzEuNdn2gPkyTvlIeoLPb0e/3d3Ir+ee2w/zg3CxOz0ogyN+PnLRY1hf0T/qbD9YBsGCydyS+z+ffOXYFT11LJztKGzg18+i78rOSwumyG/72Xj5T4sNYoEl/WDTpK+Vm7+2tIOtn/+bce9by0IcFADyw9gCTYkK4+fSpvfstnxrH3vImavpMY/DZwVoC/IQFqd4xYe3nM20eu1b/0wM1GAOnZMYftS0ryVnBc7ihncsWprjtLmRvpUlfKTd7dkMxsWGBRIcG8Ls3c/npazvZXFTHN07J6Le2a08Vyyd9hng2FtYyJyWKkEDvqFOPCR24p//pgWo+2v/5BIsf7a8iItifedYMnn1NSwzHJiACly6cNLoB+yBN+kq5wcHqFgqrW6hr6WTtvkq+tDCF525axinT4nlmwyGiQwO4YnFqv2PmTYomMSKIN3Y4ZzFv77KzvaSexRmx7jiFEYkLH7inf9e/9/Kbf+0BnNczPtpfzUlT4wZc0Dw4wI/MxAhOmRav5ZojoDdnKeVCd7+VR1JUMFcvnTzosIMxhhse30h1UweX56TSZTdcsiAFfz8bf71qAdc99hmXLUghNLD/f08/m/CFuRN4ZsMhGtu72HO4kS67YakXJf2QAD+C/G39evrGGAqqW+jsduBwGA7WtFBa38a3zpg66PM8dsNiggO849uNp9Gkr5SLdNsd/O2DfIyBdQequeeK+QMmps1FdRRWt+BnEx79pJDMxHCyrcXLY8ICWX3rKYO+xsXzJvLYJwd5e3dF77wzi9K8J+mLCHFH3KBV29JJU7uzhLOssZ2NB51lqX1vyjqSlmmOnA7vKOUi1c2dGAM5aTG8ubOcV7aU9G6rbGznqgfX8+G+Kl7ZUkJooB+PXJdDoL+Nq5YM/q3gSAtSo0mNDeGBtQd4cl0RM5MjiQoJGK1TGhUZCWHsLv18Td3C6pbexwerW9hb7rzDOCMuzB3h+TxN+kq5SM8asd88fSozkiN47rNDvdv+9939rCuo4ZZntvD69jLOn53MGdMT2fyzc/qtI3s8IsLKeSnkVzYzMTqYP355rqtPY9SdlplAXkUTZQ3OaZEL+iT9wuoW9pY1kZUc4RV3GHsjTfpKuUhP0k+KDOKrSyezq7SRnSUNHKhq5oWNxVw0dwJBAX40d3Tz5UXOqpOI4IBhlxzeetY0Xli1jH98+2Rmpxxd3eLpzpieCMDaPGe1zsHqFvxtQpC/jcLqFvIqmpiZHHGsp1AnQMf0lXKRyibnxcmkyGBWzk/h92/mctd/9tLU3kWwv41ffnEWFY3tvL+3kmUZg49XH09wgB9Lp4z8eHfLSgpnQlQwH+RVceWSyRRWtzA5NpRAfxsbD9ZS29LJdE36o0aTvlIuUtnYjk2cNyD5+9lYOS+FFzYVEx8eyK9WziY+PIj48CBmTfS+3rkriQinZyXwxo4yuuwOCqtbSI8PI9DPxn92O6dW0KQ/ejTpK+UilU0dxIUH9daW/+LibK49KY2ZyZE6Pn2EM6Yn8PzGYjYW1nKwpoWTp8UT0Kcmf0ZypBuj8206pq/UCWjvsrNmTwXGGCoa20mK/HzZvrAgf2ZNjNKEP4BTMxOICwvkZ//cRXuXg/T4MDLiQwFIjAjqnZhNuZ4mfaVOwEMfFnDTk5vYfbiRisYOEiN0bvehCAvy56dfmElBlbNyZ0p8GBnxzplCZ0zQXv5o0qSv1AjZHaa3LHNnaQOVTR39evrq2C5dkMJy64J0RnwY6VZPf4aO548qHdNXaoTe21vJ4QZnmeb24npqWrSnPxwiwr1fmc/7eZW9d9j+6ouzOGtGopsj822a9JUaoafWF5EcGUxqbAgf5FVhDCRqT39YkqOCuWrJ5N7frzsp3X3BjBM6vKPUCDR3dPPhviq+vGgS81OjKe+5MUt7+srDaU9fqRHouft2amIYtj531GpPX3k6TfpKjUC1dfdtQngwyVGf9+6TIrWnrzybDu8oNQLVzc6pgeMjApkSH0ZYoF/v3bhKeTJN+kqNQFWTc3gnPjwIm03InhjZ725cpTyVDu8oNQLVzZ342aR3zdebTp1CcV2bm6NS6vg06Ss1AlVNHcSGBeJnTbFw3qxkN0ek1NDod1GlRqC6uYOEcK3UUd5Hk75SI1Dd3EF8hCZ95X006Ss1AlVNHcSHa6WO8j7HTfoikioi74vIHhHZLSLf7bPtv0Rkr9X+xz7td4hIvojkiciKPu3nW235IvJj15+OUqPPGEN1cycJ2tNXXmgoF3K7gR8YY7aISASwWUTWAEnASmCeMaZDRBIBRCQbuBKYBUwE3hGRLOu5/gacC5QAG0VktTFmj2tPSanR1djeTafdoWP6yisdN+kbY8qAMutxk4jkAinATcD/GGM6rG2V1iErgeet9kIRyQeWWNvyjTEFACLyvLWvJn3lVap67sbVnr7yQsMa0xeRdGABsAHIAk4VkQ0islZEFlu7pQDFfQ4rsdoGaz/yNVaJyCYR2VRVVTWc8JQaE9XNzqQfrz195YWGnPRFJBx4BbjNGNOI81tCLLAMuB14UUROeF04Y8yDxpgcY0xOQkLCiT6dUi7X09PXpK+80ZBuzhKRAJwJ/xljzKtWcwnwqjHGAJ+JiAOIB0qB1D6HT7LaOEa7Ul6jp6evwzvKGw2lekeAR4BcY8w9fTb9AzjT2icLCASqgdXAlSISJCIZQCbwGbARyBSRDBEJxHmxd7UrT0apsVDd3IGfTYgOCXB3KEoN21B6+icD1wA7RWSb1fYT4FHgURHZBXQC11m9/t0i8iLOC7TdwC3GGDuAiNwKvAX4AY8aY3a79GyUGgM9Nfo22wmPZio15oZSvfMxMNi/7qsHOeZ3wO8GaH8TeHM4ASrlaZxJX4d2lHfSO3KVArrtDr760Hpe3Fh8zP1qmjtYV1DDrImRYxSZUq6lSV8p4I2dZXx6oIZHPyk85n6PflJIR7eDVadNHaPIlHItTfpq3DPGcP8HB7AJ7C1vIr+yacD9Gtq6ePLTIi6Yncy0xPAxjlIp19Ckr8a9D/Kq2FvexO0rZiACr28vG3C/lzYV09TRzbfPmDbGESrlOpr01bi2o6Se21/eQUp0CDeeksHSjFhe33EYZyFaf//eVc6siZHMTolyQ6RKuYYmfTVu5ZY18pW/ryfI38bjNywm0N/GRXMnUlDVwhl3f8B/v7yjd9/Kpna2HKrjvGxdIUt5N036atx6ZXMJdofhtW+fRGZSBACXLkjhm6dNITo0kBc2FdPY3gXAu7mVGAPnzUpyZ8hKnTBN+srnGWN4dsMhbn9pO912R2/7B/uqWDollsTI4N62sCB/7rhwJt85yzluv6/ceVH37d3lpMaGMCM5YmyDV8rFNOkrn9beZef6xzbyk9d28tLmErYW1wNQXNtKfmUzZ0xPHPC46VZy31veRHNHN5/k17AiOxkXzCmolFtp0lc+7eXNJazdV8XtK6bjbxPe2+tc9uGDPOefZ04feCbXlOgQIoL8yStvYkNBDZ12B2fNGPgDQilvoklf+SyHw/Dox4XMSYni22dMJSc9hvd7k34Vk2NDyYgPG/BYESErOYK88ibWHagh0N/GwrSYsQxfqVGhSV/5rHf3VlJQ3cJNp01BRDhrRiJ7y5vYdLCWTw5Uc+b0hGMO10xPjmBveSOfHKghJy2G4AC/MYxeqdGhSV/5rMc/LSQlOoQLZzvLLHuGZ7728AYCbDauWZ5+zONnJEfQ2N5NblkjJ02NG+1wlRoTmvSVTzLGsKWonnOzk/D3c/4zn5oQzuTYUBzG8PdrFx13KoXpSZ9X6iyfGj+q8So1Voa0cpZS3qaisYO2LjtTEz4fsxcR7rtqAXaHg0Vpscd9jhnJzpk0QwP9mDtJ78JVvkGTvvJJBdXNAExJ6N+bn58aPeTniAoNICU6hKykcAL89Eux8g2a9JVPKqxuARi0OmeoHro2h6hQXRZR+Q5N+sonFVa1EBxgI7nP3bYjka2LpSgfo99ZlU8qrG4hPS5M17FV6gia9JVPKqhuYUrCiQ3tKOWLNOkrn9Nld3CotvWEx/OV8kWa9JXPKa5txe4wTInXJQ2VOpImfeVzeit3dHhHqaNo9Y5ym/0VTdS3dbE4/fg3SvVobO/ij//Zy6HaNto6u2nttJMeF8aPL5hBamwoAAVVzqQ/RYd3lDqKJn3lFsYYbn12K6X1bWz4ydmEBQ3+T/Hhjwr4+4cFXLVkMm/vLie/splZKVGEBviRFBnA+3mVvLe3khtOTmfGhEjue3c/U+LDiA4NHMMzUso7aNJXbvHh/mryKpyrUq3efpirlkwecL/391byuzdzSYkO4b539xMe5M9jNyzm1MzP58E/XN/G797I5f61BzDGOVHaI9cvHpPzUMrbaNJXY8bhMNz7zj7mpETx1PoiEiOCiA4N4On1RVy5OBURYUdJPdtLGrhmWRq1LZ1857mtzEyO5JVvnURJXSuB/jbS4voP20yMDuFvX1vIoZpW1u6r5NKFkwg/xjcHpcYz/Z+hxsyWQ3X89b383t9vXzGdyJAAfv6PXWwvaWB+ajQ//+dudpTUc/HcCXycX01TRze/vXQ2IYF+vYuXD2ZyXOhxp0tWarzT6h01ZtbsqSDAT7jjghmcMT2Bq5emcemCFMKD/LlnzT62HKpje3E9xsD6glo2FNQQHuTP3BSd4VIpV9Gevhoza/ZUsGxKHN88fSrfPH1qb/vtK6Zz5+rdHKhsJiLIn26HYX1BDRsKa1mUFtM7H75S6sTp/yY1JvIrmymobuHc7KSjtl2zLI2ctBhK69u4YnEqizNi+c8uZ5XOsim6YpVSrnTcpC8iqSLyvojsEZHdIvLdI7b/QESMiMRbv4uI3Cci+SKyQ0QW9tn3OhHZb/1c5/rTUZ5qzZ4KAM6ZeXTSt9mEP10+jwtmJ/ONUzNYPiWO8sZ2AJZOGXoNv1Lq+IYyvNMN/MAYs0VEIoDNIrLGGLNHRFKB84BDffa/AMi0fpYC9wNLRSQWuBPIAYz1PKuNMXUuPB/lQYwxiAg1zR08+1kRs1MimRgdMuC+GfFh3H/1IoDe9WhDA/2Yo+P5SrnUcZO+MaYMKLMeN4lILpAC7AHuBX4E/LPPISuBJ40xBlgvItEiMgE4A1hjjKkFEJE1wPnAc647HeUpPt5fzS3PbmH5lDjKGtupbOzgf7+yYEjHzpoYSUSwP/NTo3XFKqVcbFgXckUkHVgAbBCRlUCpMWa7SL85y1OA4j6/l1htg7Uf+RqrgFUAkycPfMPOeOVwGK+YH37jwVpuenITCRFBfHrAWXb5wNWLWJQWM6Tj/f1sPHD1IhIigkY5UqXGnyEnfREJB14BbsM55PMTnEM7LmWMeRB4EB1tNzkAABXqSURBVCAnJ8e4+vm91ft7K7n12S18cPuZHp8Mb39pO8lRwbx083JCAvwoa2hnWuLwZrw8eVr8KEWn1Pg2pO/OIhKAM+E/Y4x5FZgKZADbReQgMAnYIiLJQCmQ2ufwSVbbYO1qCN7eU05Lp51NB2vdHcoxNbR2cbCmlctzJhEfHkRYkP+wE75SavQMpXpHgEeAXGPMPQDGmJ3GmERjTLoxJh3nUM1CY0w5sBq41qriWQY0WNcF3gLOE5EYEYnB+S3hrdE5Ld+zvsCZ7LcV17vl9Y0xPPZJIaX1bf3a7n4rjw/yKnvb9pY3AjBzgq4tq5QnGkpP/2TgGuAsEdlm/Vx4jP3fBAqAfOAh4NsA1gXc3wAbrZ9f91zUVcdW3tDeO0f8Vjcl/ff2VvKr1/fw+zdye9vyKpr4f+/n862nt/Qm+9wy55/ZmvSV8khDqd75GDjm1UOrt9/z2AC3DLLfo8CjwwtRbSisASAnLYadJQ102x1jepeqMaZ3zpw3d5VRUNXMlIRwVm87jJ9NCA/255tPbeZf/3UKuWVNxIQGkOjh1x2UGq+0Hs4LrC+oISLIn6uWTKaty87+yuYxff1PD9Swrbie75+bRZC/jQfWHsAYw+s7DnPS1Dj+etUCimpa+ce2w+wtb2TmhEiOqOhSSnkInXvHC2woqGVJRiwLrZLHbcX1YzJmnlfexG/f2MPWQ/UkRQax6rQp1LZ08vT6ImJCAymubeO/zspkaUYs05MieHlzCXkVTXxtadqox6aUGhnt6Xu4upZOCqpbWJwRS3pcKNGhAWw7NDbj+v/cVsqnB2r44vyJPHztYoID/PjO2ZnkpMfw9w8LCPSzsWJWMiLCpQtT2F5cT3uXgxnJx54CWSnlPpr0PVxxXSvgXO9VRFiQGs37eZXUtnQOuH9VUwc3Pr6Rg9aF3xNRUtdGSnQIv790DnMmOadDiA0L5NlvLOPXK2fx84uziQoJAOCS+Sn0jOho5Y5SnkuTvocrqXOWSKbEOOesue2cLOrburj12S102x1H7X/3W3m8u7eSj/ZXueC1W5kUc/RcOTabcO3ydK5Z9vkwTnJUMCdPjcffJmQmaV2+Up5Kk76HK7WS/qSYUADmpUbz+0vn8OmBGh7/9GC/fXeVNvDiZudMFwdrWk/4tUvq2gZM+oP5+UXZ3H35PIL8/U74tZVSo0OTvocrqWslIsi/dxgF4MuLJjFzQiRr9/Xvzf/ujVxiQwOZHBtKUc2JDe+0d9mpbOro/bAZiunJEVyy4KjplJRSHkSTvocrrW/rHdrpKycthi1Fdb1DPAVVzawrqOGm06Ywc0IERSfY0++583Y4PX2llOfTpO/hBhtiyUmPoaXTzt7yJgBe2VKCTeCyBSmkxYVRVNuKwzHy+ep6riWkxg69p6+U8nya9D2YMYbSurYBh1h6pineXFSHw2F4bUspp2UlkBgZTFpcKJ3djt7Vp0aixKoa0p6+Ur5Fk74Ha2zrpqmje8DEmxIdwoSoYDYV1bGuoIbDDe18aeEkANLjwgBOaIinpK6NAD8hMSJ4xM+hlPI8ekeuB2rvslPW0E5LRzfgTPBHEhEWpcWwvqCGPYcbiA4N6F10fLI1JFNU08LyqSNbWLykro2J0SH4ecGiLUqpodOk72G67A5ufGIjGwpq+fEFMwAGraDJSYvhXzvKaOu08+j1zjtmASZGhxDgJyMq23z4owImxYQOWqOvlPJumvQ9zK9e380n+TXYhN6ZLQeq3gG4YM4EPs6v5jtnZzJ3UnRvu59NSB1B2WZHt50/vpWHv03wE+HCORNGfiJKKY+kSd+D5JU38fT6Q9x0agbVzZ28trWU0EA/YkIDBtw/KTKYh69bPOC2tNjQYY/p7yxpoLPbQc8ED9rTV8r36IVcN6hsbB9wCoVdpQ0AfGXxZG48JQNwJt6RTFOcFhdGUU0LzuUNBnbkts+spRi/f26W8zniw4b9ukopz6ZJfwx02x1sLqqlrdPOixuLOfmu97j56S1H1dHnVTQR6G8jPS6U2SlRnJudRE567Ihec05KFC2ddrYcqhtwe1lDG2fe/QFPrTvY27axsJZpieHceuY0HrthMStmJY3otZVSnkuT/hh4bWspX7p/HfN+/TY/emUHqTGhvJNbwf99kN9vv9yyRjITw3tXxXrwmkX8/tI5I3rN82cnExboxwsbi4/a1t5l5+ant3CwppWn1x8CwOEwbCqqY3F6DDabcOb0RJ1DRykfpEl/DGwtrici2J+rl6bxw/OyeOt7p3HJ/Inc/fY+Zvz831z36GcYY8grb2JG8ufTEp/I6lNhQf5cNHci/9pRRrNV+tnjL+/uZ3txPWdMTyCvoomD1S3kVTTR1N7N4hF+s1BKeQdN+i5ijOGzwlrsA0x9sLu0gTkpUfzi4mxuPSuTAD8b//Olufzy4mxOz0pg7b4q1hXUUNnU4dIFSK5YnEprp503dhzu1/5JfjUnTY3jNytnA/DW7nI+K3SO52vSV8q3adJ3kQ/3V3PF39fx8ub+wylddge55U3Mmth/YZHgAD+uPzmD314yBxH433f2A86ZKl1l4eRopiSE8cbO8t42h8Owv6KZGcmRpMaGMjslkhc2FXPvO/vISgrXih2lfJwmfRfpuSD6Zp8EC3CgqpnObgezU6IGPC4hIojF6bG9PW1X9vRFhOwJkRzqU69fUtdGW5edLGuhkxXZyRRUtRDoZ+PhaxfrguZK+ThN+i5QXNvKu3sriQz259MD1TS0dWGMwRjD7tJGgKN6+n1dMDsZcC5FmBAR5NLYUqJDONzQ3lsptK/COStnlvXhcnlOKitmJfHUjUuZHKczairl6zTpu8Bznx1CgD9cNpcuu+H17Yf50v2fcuMTm9hZ2kBIgB8Z8YMvIXi+lfSnJ0W4vKedEhNCZ7eD6pYOwFkWCpCZ6IwnOSqYv1+T49JhJaWU59I7ck9QcW0rT60v4pyZSVwwO5mkyCDuXL2794JukL+N2SlRx5y4bEJUCF8/OYPZKa5fULxnsrbSujYSI4LZX9FESnQIEcED3+WrlPJt2tM/AV12B995fisY5/qwNptwwewJ2B2G36ycxelZCXR0O445tNPjFxdnc5k1NbIrTexJ+tZKWHkVzb3j+Uqp8Ud7+ifg4Y8K2Xqonr99dWHvClM/XDGds2cmcmpmAudkJ/G1hzZw5vREt8XYM1nb4fo2uu0ODlQ1c1pmvNviUUq5lyb9E/DGzsPkpMXwhbmfz0YZHuTPqZkJgHPY5r0fnuGm6JwigwOICPantK6NotpWOrsdZCXp+L1S45UO74xQVVMHu0obOXOG+3rxQ5USHUJpfRt51nq6mvSVGr806Y/Qh/uqADg9K8HNkRxfSnQIJXVtfJxfTXiQv1bqKDWO6fDOMK16chOnT09gQ0Et8eGBZE9wfcWNq6XEhPDZwVoa2ro4ZVo8gf76Wa/UeKVJfxhaO7t5e08Fa3IrCPK3ceGcCdi8YA3ZlOgQmtq7aWrv5rZzPP+biVJq9By3yyciqSLyvojsEZHdIvJdq/1PIrJXRHaIyGsiEt3nmDtEJF9E8kRkRZ/28622fBH58eic0uipbnKuKRXs70d7l8Mrhnag/3KLZ7ixkkgp5X5D6el3Az8wxmwRkQhgs4isAdYAdxhjukXkLuAO4L9FJBu4EpgFTATeEZEs67n+BpwLlAAbRWS1MWaPi89p1FQ1twPw+8tmU9PcyYpZyW6OaGh6avVnTYwkKTLYzdEopdzpuEnfGFMGlFmPm0QkF0gxxrzdZ7f1wJetxyuB540xHUChiOQDS6xt+caYAgARed7a13uSfpNzKoPpSZFkL/D8sfwek617CM72gkojpdToGtaYvoikAwuADUds+jrwgvU4BeeHQI8Sqw2g+Ij2pQO8xipgFcDkyZOHE96oq7SSvqsnRRtt8eFBPPn1JSxKi3F3KEopNxtyGYeIhAOvALcZYxr7tP8U5xDQM64IyBjzoDEmxxiTk5DgWWPmVU0d2MQ5G6a3OS0rgbAgvW6v1Hg3pCwgIgE4E/4zxphX+7RfD1wEnG2M6VkyqhRI7XP4JKuNY7R7haqmDuLCg445eZpSSnmyoVTvCPAIkGuMuadP+/nAj4AvGmNa+xyyGrhSRIJEJAPIBD4DNgKZIpIhIoE4L/audt2pjL6qpg4Swr1raEcppfoaSk//ZOAaYKeIbLPafgLcBwQBa6w54NcbY242xuwWkRdxXqDtBm4xxtgBRORW4C3AD3jUGLPbpWczyqqaO0iM1KSvlPJeQ6ne+RgYaDzjzWMc8zvgdwO0v3ms4zxdVVMH03XeGqWUF9P78YfI4TBUN3d4XeWOUkr1pUl/iBrauuiyG036Simvpkl/iKqavbNGXyml+tKkP0Q9d+Nq9Y5Syptp0h+iyibnvDva01dKeTNN+kNU5aVTMCilVF+a9IeoqqmDkAA/wnUqA6WUF9OkP0RVTc5yTetGNKWU8kqa9IfocEM7yToXvVLKy2nSH6LSujYm9VmBSimlvJEm/SHosjsoa9Ckr5Tyfpr0h6C8oR2HgUkxoe4ORSmlTogm/SEornPOHK09faWUt9OkPwQldW2A9vSVUt5Pk/4QlNS1YRNIjtLqHaWUd9OkPwQlda0kRwYT6K9vl1LKu/lkFnM4DH94M5fXtpZQXNt6/AOOo6SuTYd2lFI+wSfnFChrbOeJdQdp73LgZxPW3XEWiREjH5oprWtjaUas6wJUSik38cmefkp0CLt+uYJ7rpiH3WHIr2we8XN12x2UN7aTopU7Sikf4JNJH8Dfz8bidGfv/FDNyId4yhrasTuMlmsqpXyCzyZ9gAlRwQT4CUUnMK6v5ZpKKV/i00nf38/GpJhQimpaRvwc20vqAZgcq0lfKeX9fDrpgzNZF41weKe5o5uHPizglGnxpGrSV0r5AJ9P+mlxoRyqacUYM+xjH/u4kJqWTn64YvooRKaUUmPPJ0s2+0qLC6Opo5u61i5iwwKHdMx97+5n9fbDFNW0cG52EvNTo0c5SqWUGhu+39O3hmUOHjGu/8jHhXz98Y00tnf1a69p7uCv7+0nyN/G15am8ZuVs8csVqWUGm3joKfvTPqHalpZODkGgHdzK/jNv/YAcMNjG7nvqgWEB/kTFRLAK1tK6LIb7v3KfLKSItwWt1JKjQafT/o9F2CLalrp6Lbz+vYyfvX6bmZNjOSmU6fw/Re3cfL/vIcIfOesTF7ffpictBhN+Eopn+TzST84wI8JUcG8k1vBCxsPcbihnRnJETxw9SJSY0NJjQ1lb3kj6wtq+cu7+wG45cxpbo5aKaVGh88nfXCWbW4orCUzMZwnvr6E0zLjEREAFqXFsCgthq8umUxqTAgf7q/iwjkT3ByxUkqNDhlJKeNYycnJMZs2bTrh51m7r4rcskauPymd4AA/F0SmlFKeS0Q2G2NyBtp23OodEUkVkfdFZI+I7BaR71rtsSKyRkT2W3/GWO0iIveJSL6I7BCRhX2e6zpr//0icp2rTvB4Ts9K4ObTp2rCV0qNe0Mp2ewGfmCMyQaWAbeISDbwY+BdY0wm8K71O8AFQKb1swq4H5wfEsCdwFJgCXBnzweFUkqpsXHcpG+MKTPGbLEeNwG5QAqwEnjC2u0J4BLr8UrgSeO0HogWkQnACmCNMabWGFMHrAHOd+nZKKWUOqZh3ZwlIunAAmADkGSMKbM2lQNJ1uMUoLjPYSVW22DtR77GKhHZJCKbqqqqhhOeUkqp4xhy0heRcOAV4DZjTGPfbcZ5NdglV4SNMQ8aY3KMMTkJCQmueEqllFKWISV9EQnAmfCfMca8ajVXWMM2WH9WWu2lQGqfwydZbYO1K6WUGiNDqd4R4BEg1xhzT59Nq4GeCpzrgH/2ab/WquJZBjRYw0BvAeeJSIx1Afc8q00ppdQYGcrNWScD1wA7RWSb1fYT4H+AF0XkRqAIuMLa9iZwIZAPtAI3ABhjakXkN8BGa79fG2NqXXIWSimlhmRc3JyllFLjybFuzvLopC8iVTi/RYxUPFDtonBGk7fECd4Tq7fECd4Tq7fECd4T62jFmWaMGbASxqOT/okSkU2Dfdp5Em+JE7wnVm+JE7wnVm+JE7wnVnfE6fOLqCillPqcJn2llBpHfD3pP+juAIbIW+IE74nVW+IE74nVW+IE74l1zOP06TF9pZRS/fl6T18ppVQfmvSVUmoc8cmkLyLni0ietZDLj49/xNg5xqI0vxSRUhHZZv1c6AGxHhSRnVY8m6y2ARfPcXOc0/u8b9tEpFFEbvOE91REHhWRShHZ1adt2AsQuTHWP4nIXiue10Qk2mpPF5G2Pu/tA26Oc9C/axG5w3pP80RkxVjFeYxYX+gT58GemQ7G7D01xvjUD+AHHACmAIHAdiDb3XH1iW8CsNB6HAHsA7KBXwI/dHd8R8R6EIg/ou2PwI+txz8G7nJ3nAP8/ZcDaZ7wngKnAQuBXcd7D3FOX/JvQHAuWLTBA2I9D/C3Ht/VJ9b0vvt5QJwD/l1b/7e2A0FAhpUb/NwZ6xHb/wz8YizfU1/s6S8B8o0xBcaYTuB5nAu7eAQz+KI03mKwxXM8xdnAAWPMidzJ7TLGmA+BI+eYGu4CRGNioFiNMW8bY7qtX9fjnB3XrQZ5TwezEnjeGNNhjCnEOSfYklEL7gjHitWazPIK4Lmxigd8c3hnSIu1eIIjFqUBuNX6Gv2oJwyb4Fwj4W0R2Swiq6y2wRbP8RRX0v8/kae9pzD8BYg8xddxfhPpkSEiW0VkrYic6q6g+hjo79qT39NTgQpjzP4+baP+nvpi0vcKcvSiNPcDU4H5QBnOr33udooxZiHOdY9vEZHT+m40zu+kHlPzKyKBwBeBl6wmT3xP+/G093AwIvJTnOtlP2M1lQGTjTELgO8Dz4pIpLviwwv+rgdwFf07KGPynvpi0vf4xVpkgEVpjDEVxhi7McYBPMQYfgUdjDGm1PqzEngNZ0yDLZ7jCS4AthhjKsAz31PLcBcgcisRuR64CPia9SGFNVxSYz3ejHOsPMtdMR7j79pT31N/4DLghZ62sXpPfTHpbwQyRSTD6vldiXNhF49gjeMdtSjNEWO3lwK7jjx2LIlImIhE9DzGeUFvF4MvnuMJ+vWcPO097WO4CxC5jYicD/wI+KIxprVPe4KI+FmPpwCZQIF7ojzm3/Vq4EoRCRKRDJxxfjbW8Q3gHGCvMaakp2HM3tOxuoo9lj84qyD24fyk/Km74zkitlNwfp3fAWyzfi4EngJ2Wu2rgQlujnMKzqqH7cDunvcRiAPeBfYD7wCx7n5PrbjCgBogqk+b299TnB9CZUAXzvHkGwd7D3FW7fzN+ne7E8jxgFjzcY6J9/xbfcDa90vWv4ttwBbgYjfHOejfNfBT6z3NAy5w93tqtT8O3HzEvmPynuo0DEopNY744vCOUkqpQWjSV0qpcUSTvlJKjSOa9JVSahzRpK+UUuOIJn2llBpHNOkrpdQ48v8BSmnVWMaydZEAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}