shallow-training/nncw.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"nncw.ipynb","provenance":[],"collapsed_sections":[],"authorship_tag":"ABX9TyPZ+y9ClSPYLOPAXqmN3d6g"},"kernelspec":{"name":"python3","display_name":"Python 3"}},"cells":[{"cell_type":"code","metadata":{"id":"TGIxH9Tmt5zp"},"source":["import numpy as np\r\n","import pandas as pd\r\n","import tensorflow as tf\r\n","import matplotlib.pyplot as plt\r\n","import matplotlib as mpl\r\n","from sklearn.model_selection import train_test_split\r\n","\r\n","%load_ext tensorboard"],"execution_count":null,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"fksHv5rXACEX"},"source":["# Neural Network Training\r\n"]},{"cell_type":"markdown","metadata":{"id":"l4zqVWyRAM0Z"},"source":["## Load Dataset"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":314},"id":"Hj5l_tdZuYh7","executionInfo":{"status":"ok","timestamp":1614723046534,"user_tz":0,"elapsed":681,"user":{"displayName":"Andy Pack","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GjA4K4ZhdArHXAFbAGr4n0aCv2HmyUpx4cy6zcUq34=s64","userId":"16615063155528027547"}},"outputId":"c7ffb838-3582-4e41-9075-5126b2dc323c"},"source":["data = pd.read_csv('features.csv', header=None).T\r\n","data.columns = ['Clump thickness', 'Uniformity of cell size', 'Uniformity of cell shape', 'Marginal adhesion', 'Single epithelial cell size', 'Bare nuclei', 'Bland chomatin', 'Normal nucleoli', 'Mitoses']\r\n","labels = pd.read_csv('targets.csv', header=None).T\r\n","labels.columns = ['Benign', 'Malignant']\r\n","data.describe()"],"execution_count":44,"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>Clump thickness</th>\n","      <th>Uniformity of cell size</th>\n","      <th>Uniformity of cell shape</th>\n","      <th>Marginal adhesion</th>\n","      <th>Single epithelial cell size</th>\n","      <th>Bare nuclei</th>\n","      <th>Bland chomatin</th>\n","      <th>Normal nucleoli</th>\n","      <th>Mitoses</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>count</th>\n","      <td>699.000000</td>\n","      <td>699.000000</td>\n","      <td>699.000000</td>\n","      <td>699.000000</td>\n","      <td>699.000000</td>\n","      <td>699.000000</td>\n","      <td>699.000000</td>\n","      <td>699.000000</td>\n","      <td>699.000000</td>\n","    </tr>\n","    <tr>\n","      <th>mean</th>\n","      <td>0.441774</td>\n","      <td>0.313448</td>\n","      <td>0.320744</td>\n","      <td>0.280687</td>\n","      <td>0.321602</td>\n","      <td>0.354363</td>\n","      <td>0.343777</td>\n","      <td>0.286695</td>\n","      <td>0.158941</td>\n","    </tr>\n","    <tr>\n","      <th>std</th>\n","      <td>0.281574</td>\n","      <td>0.305146</td>\n","      <td>0.297191</td>\n","      <td>0.285538</td>\n","      <td>0.221430</td>\n","      <td>0.360186</td>\n","      <td>0.243836</td>\n","      <td>0.305363</td>\n","      <td>0.171508</td>\n","    </tr>\n","    <tr>\n","      <th>min</th>\n","      <td>0.100000</td>\n","      <td>0.100000</td>\n","      <td>0.100000</td>\n","      <td>0.100000</td>\n","      <td>0.100000</td>\n","      <td>0.100000</td>\n","      <td>0.100000</td>\n","      <td>0.100000</td>\n","      <td>0.100000</td>\n","    </tr>\n","    <tr>\n","      <th>25%</th>\n","      <td>0.200000</td>\n","      <td>0.100000</td>\n","      <td>0.100000</td>\n","      <td>0.100000</td>\n","      <td>0.200000</td>\n","      <td>0.100000</td>\n","      <td>0.200000</td>\n","      <td>0.100000</td>\n","      <td>0.100000</td>\n","    </tr>\n","    <tr>\n","      <th>50%</th>\n","      <td>0.400000</td>\n","      <td>0.100000</td>\n","      <td>0.100000</td>\n","      <td>0.100000</td>\n","      <td>0.200000</td>\n","      <td>0.100000</td>\n","      <td>0.300000</td>\n","      <td>0.100000</td>\n","      <td>0.100000</td>\n","    </tr>\n","    <tr>\n","      <th>75%</th>\n","      <td>0.600000</td>\n","      <td>0.500000</td>\n","      <td>0.500000</td>\n","      <td>0.400000</td>\n","      <td>0.400000</td>\n","      <td>0.500000</td>\n","      <td>0.500000</td>\n","      <td>0.400000</td>\n","      <td>0.100000</td>\n","    </tr>\n","    <tr>\n","      <th>max</th>\n","      <td>1.000000</td>\n","      <td>1.000000</td>\n","      <td>1.000000</td>\n","      <td>1.000000</td>\n","      <td>1.000000</td>\n","      <td>1.000000</td>\n","      <td>1.000000</td>\n","      <td>1.000000</td>\n","      <td>1.000000</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["       Clump thickness  Uniformity of cell size  ...  Normal nucleoli     Mitoses\n","count       699.000000               699.000000  ...       699.000000  699.000000\n","mean          0.441774                 0.313448  ...         0.286695    0.158941\n","std           0.281574                 0.305146  ...         0.305363    0.171508\n","min           0.100000                 0.100000  ...         0.100000    0.100000\n","25%           0.200000                 0.100000  ...         0.100000    0.100000\n","50%           0.400000                 0.100000  ...         0.100000    0.100000\n","75%           0.600000                 0.500000  ...         0.400000    0.100000\n","max           1.000000                 1.000000  ...         1.000000    1.000000\n","\n","[8 rows x 9 columns]"]},"metadata":{"tags":[]},"execution_count":44}]},{"cell_type":"code","metadata":{"id":"w6BHSlP-EBe4","executionInfo":{"status":"ok","timestamp":1614724279812,"user_tz":0,"elapsed":687,"user":{"displayName":"Andy Pack","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GjA4K4ZhdArHXAFbAGr4n0aCv2HmyUpx4cy6zcUq34=s64","userId":"16615063155528027547"}}},"source":["def get_model(hidden_layers=2, activation='sigmoid'):\r\n","    layers = [tf.keras.layers.InputLayer(input_shape=(9,))] + [\r\n","        tf.keras.layers.Dense(9, activation=activation)\r\n","        for _ in range(hidden_layers - 1)\r\n","    ] + [tf.keras.layers.Dense(2, activation=activation)]\r\n","\r\n","    model = tf.keras.models.Sequential(layers)\r\n","    return model"],"execution_count":45,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"E9lVYI14AUMf"},"source":["## Split Dataset\r\n","\r\n","Using a 50/50 split, maybe use stratification?"]},{"cell_type":"code","metadata":{"id":"L83Ae5l9wM35","executionInfo":{"status":"ok","timestamp":1614720729256,"user_tz":0,"elapsed":725,"user":{"displayName":"Andy Pack","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GjA4K4ZhdArHXAFbAGr4n0aCv2HmyUpx4cy6zcUq34=s64","userId":"16615063155528027547"}}},"source":["data_train, data_test, labels_train, labels_test = train_test_split(data, labels, test_size=0.5, random_state=70)"],"execution_count":21,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"mQGAUtIPAd6e"},"source":["## Define Model\r\n","\r\n","Variable number of hidden layers. All using 9D outputs except the last layer which is 2D for binary classification"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"fYA34P0Vu_pX","executionInfo":{"status":"ok","timestamp":1614724295556,"user_tz":0,"elapsed":812,"user":{"displayName":"Andy Pack","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GjA4K4ZhdArHXAFbAGr4n0aCv2HmyUpx4cy6zcUq34=s64","userId":"16615063155528027547"}},"outputId":"92c91f75-3b29-4d04-a2d4-50333f4e552b"},"source":["hidden_layers = 9\r\n","\r\n","model = get_model(hidden_layers)\r\n","model.compile(tf.optimizers.SGD(), loss='BinaryCrossentropy')\r\n","model.summary()"],"execution_count":46,"outputs":[{"output_type":"stream","text":["Model: \"sequential_13\"\n","_________________________________________________________________\n","Layer (type)                 Output Shape              Param #   \n","=================================================================\n","dense_117 (Dense)            (None, 9)                 90        \n","_________________________________________________________________\n","dense_118 (Dense)            (None, 9)                 90        \n","_________________________________________________________________\n","dense_119 (Dense)            (None, 9)                 90        \n","_________________________________________________________________\n","dense_120 (Dense)            (None, 9)                 90        \n","_________________________________________________________________\n","dense_121 (Dense)            (None, 9)                 90        \n","_________________________________________________________________\n","dense_122 (Dense)            (None, 9)                 90        \n","_________________________________________________________________\n","dense_123 (Dense)            (None, 9)                 90        \n","_________________________________________________________________\n","dense_124 (Dense)            (None, 9)                 90        \n","_________________________________________________________________\n","dense_125 (Dense)            (None, 2)                 20        \n","=================================================================\n","Total params: 740\n","Trainable params: 740\n","Non-trainable params: 0\n","_________________________________________________________________\n"],"name":"stdout"}]},{"cell_type":"markdown","metadata":{"id":"KZSwFe-AAs1y"},"source":["# Train Model\r\n","\r\n","Example 10 epochs"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"s8U9Atu3zelS","executionInfo":{"status":"ok","timestamp":1614724302047,"user_tz":0,"elapsed":1283,"user":{"displayName":"Andy Pack","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GjA4K4ZhdArHXAFbAGr4n0aCv2HmyUpx4cy6zcUq34=s64","userId":"16615063155528027547"}},"outputId":"d00b31e1-f60e-4b48-f1b2-91f1873f51e7"},"source":["model.fit(data_train, labels_train, epochs=10)"],"execution_count":47,"outputs":[{"output_type":"stream","text":["Epoch 1/10\n","11/11 [==============================] - 0s 2ms/step - loss: 0.6393\n","Epoch 2/10\n","11/11 [==============================] - 0s 2ms/step - loss: 0.6428\n","Epoch 3/10\n","11/11 [==============================] - 0s 1ms/step - loss: 0.6519\n","Epoch 4/10\n","11/11 [==============================] - 0s 2ms/step - loss: 0.6451\n","Epoch 5/10\n","11/11 [==============================] - 0s 1ms/step - loss: 0.6403\n","Epoch 6/10\n","11/11 [==============================] - 0s 1ms/step - loss: 0.6371\n","Epoch 7/10\n","11/11 [==============================] - 0s 1ms/step - loss: 0.6461\n","Epoch 8/10\n","11/11 [==============================] - 0s 1ms/step - loss: 0.6620\n","Epoch 9/10\n","11/11 [==============================] - 0s 2ms/step - loss: 0.6497\n","Epoch 10/10\n","11/11 [==============================] - 0s 1ms/step - loss: 0.6439\n"],"name":"stdout"},{"output_type":"execute_result","data":{"text/plain":["<tensorflow.python.keras.callbacks.History at 0x7f59dff98a10>"]},"metadata":{"tags":[]},"execution_count":47}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"r0vxP3Ah42ib","executionInfo":{"status":"ok","timestamp":1614724304951,"user_tz":0,"elapsed":676,"user":{"displayName":"Andy Pack","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GjA4K4ZhdArHXAFbAGr4n0aCv2HmyUpx4cy6zcUq34=s64","userId":"16615063155528027547"}},"outputId":"6a8a63ad-9906-4128-ae62-d8eaf480b2e1"},"source":["model.metrics[0].result()"],"execution_count":48,"outputs":[{"output_type":"execute_result","data":{"text/plain":["<tf.Tensor: shape=(), dtype=float32, numpy=0.6451795>"]},"metadata":{"tags":[]},"execution_count":48}]},{"cell_type":"markdown","metadata":{"id":"z7bn8pKTAynt"},"source":["# Experiment 1"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"mYWhCSW4A57V","executionInfo":{"status":"ok","timestamp":1614726156371,"user_tz":0,"elapsed":33858,"user":{"displayName":"Andy Pack","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GjA4K4ZhdArHXAFbAGr4n0aCv2HmyUpx4cy6zcUq34=s64","userId":"16615063155528027547"}},"outputId":"3c318562-96ee-4465-8f04-e5d5213ee424"},"source":["hidden_layers = [2, 8, 32]\r\n","epochs = [1, 2, 4, 8, 16, 32, 64]\r\n","\r\n","results = list()\r\n","for hl in hidden_layers:\r\n","    for e in epochs:\r\n","        model = get_model(hl)\r\n","        model.compile(\r\n","            optimizer = tf.optimizers.SGD(),\r\n","            loss='BinaryCrossentropy',\r\n","            metrics=['Precision']\r\n","            )\r\n","        model.fit(data_train, labels_train, epochs=e, verbose=0)\r\n","        results.append(model.evaluate(data_test, labels_test, batch_size=128))"],"execution_count":83,"outputs":[{"output_type":"stream","text":["WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59dfeffc20> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 4ms/step - loss: 0.7152 - precision: 0.0000e+00\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f5a00355a70> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 4ms/step - loss: 0.7079 - precision: 0.5000\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59dc0ff9e0> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 4ms/step - loss: 0.6489 - precision: 0.5007\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59f0037dd0> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 4ms/step - loss: 0.6502 - precision: 0.6543\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59dfed69e0> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 4ms/step - loss: 0.6977 - precision: 0.5000\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59dd579c20> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 4ms/step - loss: 0.6402 - precision: 0.6543\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59dff24170> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 3ms/step - loss: 0.6121 - precision: 0.6543\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59dff0d9e0> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 6ms/step - loss: 0.6613 - precision: 0.6543\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59d98ab710> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 5ms/step - loss: 0.6614 - precision: 0.6543\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59e00d0560> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 4ms/step - loss: 0.8006 - precision: 0.3457\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59d95e5ef0> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 5ms/step - loss: 0.6792 - precision: 0.5000\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59dc260b00> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 5ms/step - loss: 0.6679 - precision: 0.6543\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59d6fb8680> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 4ms/step - loss: 0.6520 - precision: 0.6543\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59dd5a7d40> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 5ms/step - loss: 0.6455 - precision: 0.6543\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59dc2d8e60> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 5ms/step - loss: 0.6767 - precision: 0.0000e+00\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59f08c5440> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 4ms/step - loss: 0.7682 - precision: 0.3457\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59d9cd2290> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 5ms/step - loss: 0.8093 - precision: 0.3457\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59d6ff6560> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 1s 3ms/step - loss: 0.6968 - precision: 0.0000e+00\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59d9cd28c0> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 4ms/step - loss: 0.6512 - precision: 0.6543\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59d3e17680> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 4ms/step - loss: 0.6462 - precision: 0.6543\n","WARNING:tensorflow:5 out of the last 13 calls to <function Model.make_test_function.<locals>.test_function at 0x7f59d99d4290> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n","3/3 [==============================] - 0s 6ms/step - loss: 0.6449 - precision: 0.6543\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":518},"id":"X3MWHLxJElbc","executionInfo":{"status":"ok","timestamp":1614726224612,"user_tz":0,"elapsed":1797,"user":{"displayName":"Andy Pack","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GjA4K4ZhdArHXAFbAGr4n0aCv2HmyUpx4cy6zcUq34=s64","userId":"16615063155528027547"}},"outputId":"a76811a8-faa7-4ec5-f714-0b93cd277c32"},"source":["X, Y = np.meshgrid(epochs, hidden_layers)\r\n","\r\n","fig = plt.figure(figsize=(20,10))\r\n","\r\n","for i in [0, 1]:\r\n","\r\n","    result = [r[i] for r in results]\r\n","    shaped_result = np.reshape(result, (len(hidden_layers), len(epochs)))\r\n","\r\n","    ax = fig.add_subplot(1, 2, i+1, projection='3d')\r\n","    surf = ax.plot_surface(X, Y, shaped_result, cmap=mpl.cm.coolwarm)\r\n","    ax.set_title(labels.columns[i])\r\n","    ax.set_xlabel('Epochs')\r\n","    ax.set_ylabel('Hidden Layers')\r\n","    ax.set_zlabel('Accuracy')\r\n","    ax.view_init(30, -120)\r\n","\r\n","# fig.colorbar(surf, shrink=0.3, aspect=6)\r\n","plt.show()"],"execution_count":89,"outputs":[{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAABGoAAAIuCAYAAADnru5GAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeXijdb03/neWLkmatE23mS4zbafbdJ2ZTgcEBfTHg4I6P+UcEUE44JEjHERRUBEUETgCKnJ8BI5H4GE5CiOe5yjniKK4jP5kmW7TdjqdmXaYvdOZ0jRbl+ROct+/P4Y7pJ2mzXInuZO+X9fldck0ufNtmybffJbvRyNJEoiIiIiIiIiIKPW0qV4AERERERERERGdwUANEREREREREZFKMFBDRERERERERKQSDNQQEREREREREakEAzVERERERERERCrBQA0RERERERERkUroV/g6Z3cTERFlNk2qF0BL4h6MiIgos4Xdg7GihoiIiIiIiIhIJRioISIiIiIiIiJSCQZqiIiIiIiIiIhUgoEaIiIiIiIiIiKVYKCGiIiIiIiIiEglGKghIiIiIiIiIlIJBmqIiIiIiIiIiFSCgRoiIiIiIiIiIpVgoIaIiIiIiIiISCUYqCEiIiIiIiIiUgkGaoiIiIiIiIiIVIKBGiIiIiIiIiIilWCghoiIiIiIiIhIJRioISIiIiIiIiJSCQZqiIiIiIiIiIhUgoEaIiIiIiIiIiKVYKCGiIiIiIiIiEglGKghIiIiIiIiIlIJBmqIiIiIiIiIiFSCgRoiIiIiIiIiIpVgoIaIiIiIiIiISCUYqCEiIiIiIiIiUgkGaoiIiIiIiIiIVIKBGiIiIiIiIiIilWCghoiIiIiIiIhIJRioISIiIiIiIiJSCQZqiIiIiIiIiIhUgoEaIlrSjTfeiPvuuy/VyyAiIiJKO0eOHIFGo4Hf7wcAXHrppXj22WdTvCoiShcaSZKW+/qyXySi1Kuursbp06eh0+mQlZWF8847Dz/+8Y9RVVWV6qURUXrQpHoBtCTuwYhSqLq6GidPnsTJkydRXFwc/PfNmzdjYGAAhw8fRnV1ddj7HzlyBDU1NfD5fNDr9UlYcWQ0Gg3GxsZQV1eX6qUQ0TJ7MFbUEGWA//mf/8HMzAwmJiZQVlaGW265JdVLIiIiIkprNTU1eOGFF4L/vWfPHszNzaVwRUS0WjBQQ5RBcnNz8fd///cYGRkBAHi9Xtx+++1Yt24dysrKcOONN2J+fh4AsHPnTlRWVuLhhx9GaWkp1q5di6effjp4reuuuw7f+MY3gv/93e9+F2vXrkV5eTmefPJJaDQaHDx4MHjbm2++GR/+8IdhNptxzjnn4K233krid05ERESkrGuuuQbPPfdc8L+fffZZXHvttcH/fvnll7F582ZYLBZUVVXhnnvuCXutiy66CE8++SQAIBAI4LbbbkNxcTFqamrw6KOPLmiTuuiii/DNb34T559/PsxmMy655BJMTU0Fr/WJT3wCa9asQX5+Pi644ALs3bs3+LXl9mQXXHABAKCjowN5eXn4+c9/Hv8PiYgSgoEaogwyNzeHn//85zj33HMBAHfccQdGR0cxMDCAgwcPYnx8HPfee2/w9qdOnYLT6cT4+Dieeuop3HzzzbDb7Wdd95VXXsEPfvAD/OEPf8DBgwexc+fOs26zY8cOfOtb34LdbkddXR3uuuuuhH2fRERERIl27rnnwuVyYd++fQgEAtixYwc+/elPB79uMpnw3HPPweFw4OWXX8a//du/4Ve/+tWK133iiSfw29/+FgMDA+jv71/yPs8//zyefvppTE5OQhAEfP/73w9+7dJLL8XY2BgmJyexZcsWXH311QvuG25P9te//hUAMDg4iJmZGXzyk5+M6edCRInHQA1RBvjYxz6GgoIC5Ofn49VXX8VXvvIVSJKEn/zkJ3jkkUdgtVphNptx5513YseOHcH7ZWVl4e6770ZWVhYuu+wy5OXl4cCBA2dd/8UXX8T111+PlpYWGI3GJTNGH//4x7Ft2zbo9XpcffXVGBgYSOS3TERERJRwclXNq6++io0bN6KioiL4tYsuughtbW3QarVob2/Hpz71KfzlL39Z8ZovvvgivvjFL6KyshKFhYW44447zrrN9ddfj4aGBhgMBlxxxRUL9lWf+cxnYDabkZOTg3vuuQeDg4NwOp3Br3NPRpT+1HOyFRHF7Fe/+hUuvvhiBAIBvPTSS7jwwgsxMDCAubk5dHZ2Bm8nSRICgUDwv4uKihYccGc0GjEzM3PW9U+ePImtW7cG/3upg4rXrFmz4nWIiIiI0sk111yDCy64AIcPH17Q9gQAu3btwh133IHh4WEIggCv14tPfOITK17z5MmTC/ZS0eyrAoEA7rrrLvziF7/A22+/Da32TN59amoK+fn5y96XiNIHK2qIMohOp8Pll18OnU6HN998EwaDAXv37oXD4YDD4YDT6YzpzXrt2rU4ceJE8L+PHz+u5LKTQpIk+Hw+zM3NYW5uDitMvCMiIiLC+vXrUVNTg9/85je4/PLLF3ztqquuwvbt23H8+HE4nU7ceOONEe0v4tlXPf/883jppZfwhz/8AU6nE0eOHAEA1e5rJEmCKIoQBAEzMzMQBEG1ayVSEwZqiDKIJEl46aWXYLfb0dLSghtuuAFf+tKXMDk5CQAYHx/H7373u6ive8UVV+Dpp5/Gvn37MDc3h/vuu0/ppStOkiT4/X54PB643W44nU7Mzs5ifn4egiBwo0BEREQReeqpp/CnP/0JJpNpwb+73W5YrVbk5uaiu7sbzz//fETXu+KKK/DDH/4Q4+PjcDgceOihhyJei9vtRk5ODoqKijA3N4c777wzqu+lrKwMhw4diuo+0ZKTY/Pz83C73XC5XJidnYXX64XX64Xf7+cejGgFDNQQZYCPfvSjyMvLg8ViwV133YVnn30WLS0teOihh1BXV4dzzz0XFosFF1988ZJn0Kzk0ksvxRe+8AW8//3vD14PAHJycpT+VmImt3V5vV7Mzs7C5XJhZmYG8/PzEEURWq0WOp0uWCIsiiK8Xu+CVjAiIiKixTZs2LCgBVz2+OOP4+6774bZbMa9996LK664IqLr3XDDDbjkkkvQ3t6OzZs347LLLoNer4dOp1vxvtdeey3Wr1+PiooKNDc3B/dkkbrnnnvwD//wDygoKMCLL74Y1X3DCZcc83q9kCQJWq0Wer0eGo0GAOD3+5kwI1qBZoU/EP71ENFZ9u3bh9bWVni93gVn3CSbKIoIBALw+Xzw+XzBN3yNRhP832J+vx8ajQYGgwGSJEGSJOj1+gUbCKJVhk98deIejGiV+O1vf4sbb7wRR48eTfVSIiK3M/n9/uD/5D2VVqsNuwfz+XzIzs6GXq+HJEnQaDTIysqKKEBFlKHC7sFYUUNEEfnlL38Jr9cLu92Or33ta/joRz+a9CDN4oyNXEorCAI0Gg10Ol2waiaSoIu8kWBmh4iIiJJlfn4ev/nNb+D3+zE+Po5vf/vb+PjHP57qZS1LFMXgWX8ulwtutxvz8/PBBJhOp4Ner49oD6bRaIIVzoIgLEi2EdEZrKghooh86EMfwhtvvAGdTocLL7wQjz/+ONauXZvQxwzN2Ph8Pvj9/uDXlquaWU5oRU3o48iZnezs7ODmgWiVYEWNOnEPRpSh5ubmcOGFF2L//v0wGAz48Ic/jB/+8IewWCypXlqQ3FIu78FCW8UjTYgtJggCcnJykJWVteBx5Eqc7OxsVjfTahP2Cc9ADRGpSiztTNHw+/3QarXIzc0962vyZiErKyulLV1EScZdsTpxD0ZESZOI5NhicutTaKBGfmwmzGiVCvtHxU8iRJRScjuTvDEQRTH4Zq3VapP6Zi1vQOQAEc+tISIioky1XHIs1qqZWMhBIEmS4PV6g+fWcA9GqxkDNUSUVHLGRt4UTExMIC8vD0ajMSkbg5V6oEMnEoiiyDJcIiIiygih7UyCIODQoUOorq4GgKQlx5bbUzFhRvQuBmqIKKHkctbQUlpRFAGc2RS43W7k5uaq6sR/eVMgj/DmRAIiIiJKN4uTY6HnzGg0GthsNmzYsCGFKzwbE2ZEZzBQQ0SKCy2l9fv9wY2BXDGz+PwXNb4Bh5bhCoLAMlwiIiJStaWSY8u1M6l1T8OEGREDNUSkgJUmA6RzgCO0DFcURWRlZaXt90JERESZZamq5VSd9ackJsxotWOghoiiFm4ygDxeMZkH0CWD/L0EAoHgVKh03fgQERFR+lopOZapezAmzGi1YaCGiCKSyMkA6fCGK2d2QstwOcKbiIiIEik0OSb/T25xSrfAzEoDHcJhwoxWI37KIKIlLZ4MIB8ADCg7GSDWN+1U0Wq1kCQJIyMj2LhxIycSEBERkaKWS47JCbLVRv7evV4v3nrrLTQ0NDBhRhmNz24iAhC+nQlQpmpGTeL9PuRJCZxIQERERPGKpJ2JzpAkCQ6HgyO8KeMxUEO0ioWW0irdzhSNdH2DDW2Fys7O5kaKiIiIVrSakmOJIFfXMGFGmYyBGqJVRJ4MIG8MDh06hPXr16d0MkCyW58EQQAA5OTkxH2t0IkEoeMjuVkgIiKiUKHtTDMzM5iamkJ5eXlGBWaEt6fh+Es3nK/3o+Jzn4Kxseas28T7fcoTrRaP8GbCjDINAzVEGWypUtrQNzibzYYNGzakepkJFQgE4HA4MD09DYfDAa1Wi0AggPb2dlgsFkUegxMJiIiIKNTi5Fjo2GxRFOFyuVBVVZXqZcZF9ApwvjkA+5/fhP3Pb2J2eAx4JwE3+Z+/Q+Nj30LxRz+g6GPKP0OACTPKbAzUEGWQaCcDZOIbmSRJmJ2dxfT0dPAcmYKCAlitVtTW1kIURQiCgOHhYVRXV6O8vFyRx+VEAiIiotVL3oPJBwCHnjOTSVUzsyMHzwRmdnbD+UY/xDnPkrcLzMxi5LqvoerW61B9103QJHAIBRNmlIkYqCFKc6GltH6/H4FAIPjmlA4H0IVmRmLl9Xpht9sxPT0Nt9uNvLw8WK1WNDc3n9XiJIoiTCYTtm3bhj179sDpdKKxsVGRnxNHeBMREa0OciIs9JwZeULmSmOz0ymIIEzZ4di5653gzC4IE29HfmdJwvFHnsbM4H40PXE/YDLEvZ5w+0YmzCjT8BMEUZpZaTLAaij5DAQCcDqdmJ6eht1uh16vR2FhIaqqqpCXlxfR96/X67Fp0yYcPnwYvb296OjoUOTcGuDdEd6cSEBERJQ5lkqOAe9WzESTnEn2GX2RCm1ncuzchZk9o8F2pljZ//QGdn/gWtT/n+/AsLU9rmstl+BjwowyCZ+5RCoXbjJAuHamTBTazjQ9PQ2fz4f8/HxYrVbU1NRAp9PFdF2NRoPa2lpYLBb09vaipaUFBQUFiqxZ/p1wIgEREVF6Wi3Jsdl9b71zzsyuZduZ4uE5Oo69H/kcmn50N8qv/EjM14mkEpsJM8oEDNQQqVBoxiaVY7NTSRCEYGDG7XbDZDLBarWiqakJubm5ij5WcXExjEYjhoaGUFFRodjhfpxIQERElD5WS3JMbmcSfvFrvDl8d3TtTHEQ5z0Y+eydcPfvRf13boM2hmqXSFvmmTCjdMdADZEKhGZsBEEI9jgD6XHOTDzkN9zQdiZ5OpPVakVlZSXMZnPC31yNRiO6urqwd+9eOJ1ObNy4MeZKnVCcSEBERKReqyE5JnoFOHcNwvHOdKbQdqbACvdNhOOP/wzuoQNof+57yC4tiuq+0ZxtyIQZpTMGaohSIDRjIwhCxk4GWI7czjQ7O4sDBw5AFMVgO1N1dXVKeop1Oh3a2tpw/Phx9PT0oKOjAwZD/AffAQsnErAMl4iIKDVWS3IsGe1M8XD8rRe73ncl2n/2A+RvbYv4ftEOoWDCjNIVAzVESbB4MsDExAQsFkuwDHM1BGaApduZNBoNqqurUVQUXUYlUTQaDdatWwez2Yz+/n40NTUptjaW4RIRESXX4namY8eOoby8HEBmJccWTmfqhjAxmeolrcg7fhp9H7wejQ9/HRXX/V1E94l1WigTZpRuGKghShA5MCNvDERRDL65TE9Pw2g0KlatoVaiKC6YzrRUO9PIyIhi05aUVFhYiM7OTgwODsLlcqG6ulqRN/TFZbhyZoeIiIiUIQdm5D1YaDvTxMSEYmfRxUqJ/cRy7UzpRPQK2Pf5b8PVtxeND38d2uysZW8fa6AGYMKM0gsDNUQKWW4yQCZlbJYjSRLm5uaCVTNerxcFBQWwWq1Yv379ku1M8bzhxrrGSOXm5qKrqwv79+/H4OAgWltbFWnJCi3DFQQBer2emR0iIqIYLZccW9zOpJb32ljGc6u9nSke40//J2b2jqLtpw8jt7xs2dvG8ztkwozSBQM1RDEKLaWV/ye3OK00GUD+kJ4JBEGA3W4PtjMZDAZYrVY0NDSosmIo2jd3rVaL5uZmjI+Po6enB+3t7TAajYquhZkdIiKiyC2VHJMDM5mUHBOm7HD8ZRfsf9oF+85dadHOFA9n9xC633sl2v7j+yg8v3PJ24iiGPfvlgkzSgcM1BBFYfFkAPnNInRjEIl0fiNYqp2psLAQ5eXlsFgsaf29LaeiogJ5eXkYGBhAXV2dYt9naGZndnYWubm5KTlImYiISK3iSY6lE1HwwfnmQNq3M8VDmLSh/yM3oP47t2PdTVcteRul92B+vx9erzd4diKRGvDTANEylmtn0mq1cZ0any4VNUu1M8nTmcK1M0V7/XR5U8zPz0dXVxcGBwfh9XoVW7sc6Nu3bx/Wrl2LkpISBmuIiGhVC02O+f1+BAKB4HtuRk5n2rkLztczq50pVpLPj9GvPAhX/zA2/u+7oTPkvvs1hffP8nOqu7sb5513Hkd4k2rwkwBRiMWTAfx+f/BrSpbSqr31yefzLZjOpPZ2pmjF8zvMzs7Gpk2b8Le//Q27d+9GW1sbsrKWP/guUnJmkBMJiIhotUlkckxNfDYHbL///zD7P3/Em4OjGd/OFI9TL/wasyMH0f78IzCsrwDw7l5JSXLCDABHeJNqMFBDq97idqbQyQCZUkq7ksXtTBqNBlarFeXl5TCbzcwsLMFkMqG8vBw9PT1oa2uD2WyO+5ry5kOj0fDcGiIiymjhkmMZ3c60cxdmhg6sunameLgH96P7fZ9C6zMPoegD70loopMjvElNGKihVSc0YyMIAkRRDH4tWaW0qa6okSQJ8/PzEAQBQ0ND8Hg8irYzqZkSP3e55WnNmjXIy8vD0NAQamtrsWbNmrivG7ox5UQCIiLKJKslOTa7/9A705neZDuTAnzTDgx8/J+x4Vu3IOdTlyb0ecJBD6QWmftpjOgdcsbG5/NhampqwUFhmbYxWI7P5wtOZ3K5XMEWprq6OsWmGMUi2WfUKP1YeXl56Orqwp49e+B0OlFfXx9zsC90ksHiiQQswyUionQjJ8cEQYDNZltQfZpJ58z4bA7Yd765aqYzpYIUCODg3f8K8+u9KP725xP6WEyYkRowUEMZR54CEFpKK2dsxsbGsGnTppS/2CajokYURbhcrmA7EwAUFhZi7dq1aGhogFarRU9PT0qDNKkSb7BjcX90VlYWNm/ejEOHDqGvrw8dHR3Izs6O6bqL1xZahiuKIrKyshisISIiVQptZxIEIXjOTCAQwFtvvYUtW7ZkxHtYsJ1p564z05nYzpQ07lf+Bu/BY1j7n4/BWLc+YY/DhBmlGgM1lBEWB2ZEUQx+6A3N2GRK5mYpcjuTfAiwx+OBxWKB1WpFVVWVYgfe0tLtUxqNBhs2bIDFYkFPTw9aW1uRn58f9XWX2gDI/xYIBCBJErKysjL6uUxEROlhueRYaNWy/IFXDR9y5TVHu5YF7Uxv7IY4O5+gFdJKhIPH0H3hVWh54l9QctlFCX0sJswoVRioobS00mSAcO1MqT4bRul1LG5nys3NhdVqRV1dHQwGA99IEmS5DV5JSQlMJhMGBwdRVVWFyspKRa4rb3RDy3Az+SwhIiJSJzkwI+/BwiXHQqll/wVEXlXLdiZ18zvdGPzkF1Fzx+dQe+dNSTm3hgkzSibu8ikthJbSyv+TMyLRTAZQy0Yh1nWEa2das2ZNsJ0p3aglwxaNldZsNBrR1dWFvXv3wul0YuPGjRH9bkRRXPF2Wq0WkiRxIgERESXFcsmxSM/6U9P7VLjqHlHwwbVrMFg1w3amNCBJOPzAj+EeGEHrkw9Anx/bBM5I9uRMmFGy8dlFqrXcZAB5YxAttQRqojE3N8d2JpWJJLik1+vR3t6Oo0ePoqenBx0dHcjNzY37ugAnEhARUeIolRxLB2xnygxTv/0rui+8Cu3PP4K85rqYrhHpc5oJM0oWBmpINSJpZ4qXWgI1y63D7/cH25mcTifbmRSm5HjulWg0GlRXV8NisaCvrw8bN26E1WqN+7rytYF3JxJkZ2enZUUVERGl3uLkmDyFMJ7kmBr5bA4EdvZi9Olfw/GXHggnT6d6SaSQuYNH0fOBT6P58XtRdvklUd032upuJswoGRiooZQJzdjIB9DJEjU2Wy2BmlCiKMLtdgerZoAz7UxlZWVxjXpOJ+n25hbtG7rVakVnZycGBwexZs0arFu3bsn7x7JRkJ/ToeMj0+3nSUREybVScixT3kvCtTPxtJnMFJiZw55rb4dr9/Wou+cL0EQ45VUOTEaDCTNKNAZqKKmWa2dKRimtWgI1fr8fNpsNp06dwvz8PMxmM4qKilBZWbnq2pnU8PuIVizn6uTm5qKrqwsjIyPYs2cPWlpazhoTv3jsd6Q4kYCIiJaTiuRYqsjtTI6du+B4vZ/tTKvQ0UeehntgH1qf+S6yiwpWvH2s5yUyYUaJxEANJZScsQktpZUp1c4UjVQFaha3M4miiPz8fNTW1sJoNPIFPc3E+oau1WrR2tqK48ePo7u7Gx0dHTAajQuuGytOJCAiolCpTo4ly5npTLvOVM3s7GY7EwEApv/8JrovuBLtzz8CS8fGZW8b72ALJswoERioIUXJGRt5UxAIBPD222/D6/Wiqqoq5RuDZAVqJEkKTmeanp6GJEkoLCxEaWkp6uvrcfz4ceTm5sJkMiV8Lekg3d7M4n1Dr6qqgtlsxu7du9HY2Iji4uLg1+LdKHAiARHR6hTaziQIAkRRxNjYGNauXQuLxZIxwXtOZ6JIeY6eRO/F/4CN//ubWPupj4a9XSRTN1fChBkpjTt4ios8BSC0lFaumpErZuTT0dXwYpXIQM38/HwwMCO3M1mt1iXbmdTSgqUGqfg5xBsYUmKkeEFBAbZu3YqhoSE4nU7U1tbGdb1QnEhARJT5lkqOyUIPAJaD+OlsbvQI7H/phv0Pr7GdiaIiznuw94a74OwdRsODt0O7xBEDSuzrACbMSFl85lDUlgrMyC9wWq32rBckNQUllFzL4namnJwcWK1WtjOtEkr8fnNyctDZ2YnR0VEMDAwo+ncSOpFADhzyOUlElL6WSo6t1M6kpj1YNHzTDtj/vLCdydhcB2FikkEaismJf38BM3sOoO0/HkZOWdGCrykVqJHJCTOv1wuv18vPBRQTBmpoRZGMzV7uxUer1S44myaV4tmwSJIEt9sNm80Gu90OURQXtDNFUzGUrhsnOkPJ57NWq0VTUxMmJiawZ88ezMzMIC8vT5Fry3+Xvb29OO+88ziRgIgozayUHFvpNV1Ne7DlBNuZ3jlrZmboALBo3XMjB5FTWQadJQ/eoydTtFJKZ47X+9H93k+i/WcPI39bR/DflQ7UAGf2YB6PBwcPHsSmTZs4wpuixkANnSXcZAC5fSnac2bUFJSI9gXS4/EE25lmZ2dhsVhgtVpRUVGB7OzsBK1ydUnEm+NKj6cEpde8du1aHDx4EENDQ9iwYQPKysoUuW5oyTsnEhARqVu8ybHF1LQHW2zuwOHgOTORtjN5T5yGLt8MU1sDZveMJmGVlGm8E5Po/dBn0Pi9O1D5j58AkLi9qPzZiSO8KRYM1BCAxE4GkHs11WCltfj9fjgcDkxPT8PhcATbmWpqahQtW1TzxinTSJKEmZkZ2Gw2TE9Pw+PxoKysDE1NTTH/PhP1hq7T6dDV1RU8t6a+vl7R5xwAnltDRKQiockx+X9yi1MsgZnF1LQHk1wzmPyv3wdHZ3vHY5vOFHC6Mbd/HpZzNsG1a0DhVdJqIAk+7P/ifXD1D6PpB3cl7CzN0L9jjvCmaDFQs0pFkrFRivzipAaLAyRyO5NcNSOKIgoKClBcXIy6ujpGvdOUIAiYnp6GzWYLthIVFRWhtbUVkiRhYmICfX196OjoOOug50gksgooKysLW7ZswcGDB9HX14f29nbFqrdCz60RRZFluEREKbBcciz0EGAlpHIPJvr8cO0aCJ414xnaj/2iMmuR3rl2/vlb4HytX5Fr0upz8tlfYmbvGGr+7dsJ2Q+Johi8LhNmFC0GalaJcO1MgDJVM8tRUzZHo9HA6/Xi5MmTwXYmeTpTa2tr0tqZWFGjLFEU4XQ6MT09DbvdDq1WC6vVinXr1iEvL2/Bc1sQBGzYsAFTU1Po6elBW1sbzGZzVI+X6HYtjUaD+vp6TE5OBtdosVgUuzaABRMJdDqdItcmIqKzJTM5tliy9xvBdqadu+B8vR+BmbmEPp7ztX6Yu9rhHtwHCL6EPhZlJlfvMPZe+llY7vs80Nqq6LUXV+owYUbRYKAmgyWynSkaqa6oCW1nmpycRE5ODsrKylBdXQ2TycQXyBSLNeghj0O32WzweDzIz8+H1WrF+vXrIxqFWFZWBpPJFNOZMMk6V6e0tBQmkwmDg4Oorq5GeXm5IteVs7aSJEEQBOj1emZ2iIgUksrk2GKJPkzYN+2AfWd33O1M8XD3DAUnQvntrqQ/PqU//5Qd05//Do46PVj/+WsUu25oRY2MCTOKFAM1GUSeDCCPjfb7/cGRvInO2Cwn2dmc5dqZtFot8vPzUVJSkrT1LEVNVUbpIhAIwOFwwGazLTg/qK6uDgaDIaZNb15eHrq6ujA4OAiXy4W6urqIrpPMA5BNJhO2bduG4eFhOJ1ONDY2RvW3vNzfHjM7RETKCE2OTUxMwGw2B19TkxmYWUzpPdiCdqaduzAzuP+s6ebOkB4AACAASURBVEypcGYi1BpOhKLYBQIYu+N7cPfvxcZHvwWd0RD3JcOdfcOEGUWCgZo0tlQprfwBcmZmBoIgoLCwMNXLTMpoyMXTmcK1MzmdzoSuIxpsfVqeJEmYnZ0NVs34/X4UFhaiuLgYGzZsUCz7kJWVhc7OToyOjmL37t1ob29fsSIn2ZOq9Ho9Ojo6cOTIEfT29qKjowM5OTkR3XeltS7O7HAiARHRyuQ9mFy1HLrPmZqagsFggMEQ/we9eCmxB5sbPRKczpSMdqZYeU+cgsZshKa2AtKh8VQvh9LUqRd/g5l9b6H9+UdgrKmM61pLVdSEYsKMlsNATRpZajKA/Oa7eDKAXq+Hx+NJ5XKDElFRI1dXyNOZsrKyYLVaV2xnUsvZMHwRXprP54PdbofNZoPb7YbRaITVasXGjRuRm5ubsMfVaDRobGzExMQEuru70dHRAZPJFPb2iQjUrPS81Gg0qKmpgcViQW9vL5qbmyMKxIqiuGLgJTSzw4kERERnk/dgcmAm9JyZxVUzOp1ONVWzsVTwqqGdKVaSew4arw/mczrg3jWY6uVQmprZcwDdF1yJtqcfQtHF58d8nUimSTFhRuEwUKNyoaW0fr8fgUAg+Aet1WrDZv6TUcUSKSXOqJFHLMtVM3J1RVFREWprayOurlDLz0UtAaNUkyQJPp8PR48ehcPhAAAUFhaivLwcFosl6YGCtWvXwmQyYWBgAA0NDWFb5BIVqInkmkVFRdiyZQsGBwdRXl6OqqqqZe8XSaBGxokERERnyCOyQ8+ZCZccW0wtew0gsv2G6PPD1T34TtWMetqZYiUJPrh3DSL/vC1wvs6JUBQbv92F3ZffjA3fvBnVt382pv3QShU1MibMaCkM1KjMSpMBIv2jVVMgINbzWLxebzAwI49YtlqtaG5ujrjtY6m1qOXnohbJbuPxer2w2WzBNjVBEGA0GrFu3bqYRmXHYrnv12KxBM+tmZmZQXV19Vm3T8TPLJqAisFgQFdXF0ZGRuByubBx48awwcpo18oyXCJarZYKzMivocslxxZT0zl04YJG6dLOFA/n6/2wbGuHa4AToShGooi3vv0juPpH0PKT+6E3h6+2XkokFTWhmDCjUAzUpFi4yQDyH3asB9ClWzYHOLudSa/Xhx2xnOi1JJpa1pEMoigGf692ux16vR5FRUXBNrX+/n6Ulpaq6sT77OxsdHZ2Yv/+/RgaGkJra+uC9aWyokam0+nQ2tqK48ePo6enBx0dHUuehxDtJgHgRAIiWh0iGZsd6x5MLe/x8n7DZ3fC8U47k33nLnhPnEr10pLC1c2JUBS/t//nj+i56DDaX3gEpoaaiO8XaUVNKCbMSMZATQokY2y2mgI14dayVDtTQUEBrFZrVO1M0VLL5ilTSZKE+fn5YNWM1+sN/l5ramrO+r0mu6InUlqtFs3NzThx4gS6u7uxadOmYCAkEc+hWH4OGo0G69atg9lsRn9/P5qamlBUVLTgNrFsEuRrh04kYBkuEaW7RCXHFlPDHkxuZ5p6+c9w/rUHx/YfTut2pngEJ0KZTfAem0j1cihNzR44hO6LrkbLT/4FpR95f0T3iaZaOhQTZgQwUJMUoRkbQRAWvHknamy2GjYJstDqkUS0M0W7FjXItIoaeST89PQ0nE4ncnNzUVRUhIaGBlVMvQgV7c+9srISeXl56O/vx8aNG2G1WgFA8b/beAJWhYWF6OzsDI4ZD23XijVQIwstwxVFEVlZWar5OyIiWkkykmOLpWoP5jkxAdtv/gL7n3fB+VpfRrYzxcp74hT0BWaYWusxOzyW6uVQmgq4ZjD0qVtRfftnseGbN0Ozwl4wlqpmGRNmxEBNAoTL2ACJ3RiEUkugRm5n8ng86OnpSUg7UzTU0jee7oEauRpKrpoRRRGFhYUoLS1FfX29qk+rj+U5V1BQEAyErFmzJuYMyXLieTMHgNzcXHR1dWH//v0YHBxEa2sr9Hp93NcF3v2ZBQIBSJKErKwsVf+OiWj1iqSdKdFStQezebMw8cLLmBvcl/THTgd+hxuBOQ+y2hrg2zOa6uVQupIkHPneE3AP7kPrUw8iq9CyzE2V24MxYbb6MFCjkMWBmWRkbJaTqk2CJEmYnZ3F9PQ0bDZbsJ1Jr9djy5YtKS/bS/cASSJEWskhCEKwGsrtdgeroVpbW5GdnZ2ElaaWHAjZu3cvZmZmUFlZqej14618Ad5t1xofH0dPTw/a29sVCyrJmZ3QMtxID9YkIkoUNSTHFkvVHkwsKETek/8G/Ve/Dtef30j646cDSfDBt2cUpnM6MMvx3RQH2+//hu4LP4X25x+BubVhydsosbcDmDBbrbjLjpE8GUDeGIiiiMHBQbS3t0On06X8jyeZB9l5vd5g20voB/jQdia73Z7yIA2gnkCNmtYRjiiKcLlcsNlssNvt0Gq1sFqtqKyshNlsVmzjq9Yzapai1WrR2tqK3bt348iRIygpKUFubq4i11by51BRUQGz2YyBgQGUl5cr+vOVX1s4kYCIUmVxO5MgCBgbG0NLS0tKAjOLpap61y8CMBiQ88j3Ufgv34H9Fy8nfQ3pYnbXIMzb2uHmRCiKw/yh4+j5wKfR/Ni3seYTl571dSUqamRMmK0+/O1GaLlSWjljI2euU71BkNeUqE1CIBCA0+lcMMWnsLAQVVVVKWlnioZaWsLUan5+Plg1Mz8/D4vFgqKiIqxfv55vBu/QaDQoKCiA2WxGX18fWlpaUFBQEPd1lQ5YyWPG+/r6FL++fJ39+/dj3bp1sFgsqv67J6L0tlRybPHYbJ/Pl/IkmSxVe43AOw+p0euR9a27UVxagqnHnkn6OtKFu3sIppZ6eMZPIeBwp3o5lKbEOQ+Gr/8aXP17UX//l6AJSUwrVVETSk6YDQwMoKOjgwmzDMZPXmGEltLK/5MkadnJAFqtFoFAQBUbBSU3CaHtTNPT0/D5fMjPzw87xUft1FLJooZ1SJIEm80Gh8MBh8OB7Ozs4NQto9GYkS/8SvzcJUmC2WxGZWUlBgYGUFVVFXcrVCIqi7Kzs1FfX4+xsTH09/ejvb0dWVlZilxbo9FgZmYGgUAAXq8X2dnZqnjtI6L0t1RyTH6NXKqdSd5/qUXKAjWLfgTam25CUZEVtvsfAcTU7znUaHbvGHKq1kJvzoP3OCdCUeyO/eg5uAf3oe3Z7yG75MzgiUScaQic2YO53W6O8M5w3FWH8YUvfAEjIyOYn5+Hz+eDRqOBTqeDXq8PWzWj0+lUU60R7yZBEAScOnUKIyMj6O7uxpEjR6DX69HU1ISuri40NDSguLg47YI0q/1FTA66HTt2DAMDA5idnYXdbkdRURG2bNmCjo4OVFVVwWQyJfVnlW6/F/kDg8FgwLZt22Cz2TAyMhLX35yS5bGLr1tUVISKigr09PTA7VYuayi3PgFnWiBDz+ciIorV5ZdfDrfbjfn5+eAHHb1eH2wtXypRpibJbD+XiaKEpR5Rd8UnYX7om9DmJn6yZrryHp9AwD0DU0t9qpdCac7+1x7set+VcPXvBZD49v7QVii1fAYl5bCiJgybzYb5+fmoAhFqaquJ9kUhtJ3J4XAk7DySVFNLJUsy1+Hz+WC322Gz2eB2u2EwGFBUVISmpiYMDw+jrq5OdZtctQt949XpdGhvb8eRI0fQ19eHjo6OmA5XTkR5LPDuWtesWYO8vDwMDQ2hpqYGa9eujfvai9s9OZGAiJQgJ4fSVSr2g/5lHs5w6YdhLC/H1D9/BQEnW3yWIk+EspzTDteuoVQvh9KY98Qp9F5yHZoeuQtiZ0PCknDAwhHe8rk1HOGdOdL3XTDBTCYTPB5PVPdRW+ntcpZrZ6qurk7rDdJy1BKoSSRJkuB2u4OjswGgsLAQ5eXlMJvNDMooYHGGRKPRoKamBnl5eejp6UFbWxsslvDjGiO5plJCy27z8vKwbds2DA0Nwel0oqEhvg1E6LU5kYCIlJLuHzJSEagJrPBwUsdmFP/0J7B/7osQTk4mZ1FpRhJ8cO0aQv55W+B8vT/Vy6E0Jnq8GLnpbhgvvxhrvv915a+/qKWKCbPMlJmfxhVgMBgwPz8f1X3U1Pq0lMXjlU0mE6xWK5qamhSbXBOOHCBJ9YuGWgI1Sq/D6/UGR6LPzs7CbDajqKgIlZWVip1JQu8K91wuKSmB0WjE4OBg1FUrifr7WHxdvV6PzZs349ChQ3FVAAFnVwFxIgERkToDNQAg1dSi4D+ehP1zt8B38GjiF5WmnK/3n5kItXsE8PlXvgNRGHP/9QeMHj2FzT//IXLWlCh23aXOvmHCLPNwBx2G0WiMOlCjptYn4MwHNHlsdiLHK0ezHgZqlCGKIhwOR7BVTafTBauhojlfJlN+HpGS/z7jfeNa7rlsMpmCVSsulwsNDQ0R/T6SUVEj02g02LBhAywWC3p6etDa2or8/Pyorx3uXB2O8CaieMgB33T9kJGaQE2E7+VlZch/9gnMfekrmOseTOyi0hgnQpFSZvqGsev8T6L9pw+j4D2bFblmuP0XE2aZhb+5MGKpqEl165MkSZibmwtWzczNzeHtt9+G1WpN+XjlVBystxS1BCZiWYf8u7XZbPB6vSgoKMj4VrVEUCJgsFJQRa5aeeutt4JVKytVNiXqQ8lyay0pKYHJZMLg4KAik6tCyY/JiQREFK3c3Fx4PB4Yjcao7qeGhBDwbqApmZY7o2YxbX4+8n78KLK+eQ+cL/8xcYtKc7N7x5CzjhOhKH7C6Sn0XfaPaHjwq6j63JVxX2+lPSMTZpmBn+7CMBqNmJubi+o+iWx9cr3eDct52876d0EQglUz8kGxVqsVDQ0NGB4eRkNDQ0LWE610DpCkah1+vx8OhwM2mw1OpxO5ubmwWq2or6+PevNKyorkw4BGo0FdXR1Onz6Nnp4etLe3Iy8vL65rxmKlN3Oj0Yht27ZheHgYTqcTGzduVCxgJH8/cmaHI7yJKBJyVXM073VyFYsaplEmOznl8XjgdAkATJHfKTsb2Q/cj6ISK2zP/CJha0t33mMT0BdaYGypx9zesVQvh9KY5PPjwG3fgat/GE0//CZ0cUxiiyS5x4RZ+mOgJgyj0RjTYcKJCtSM3vJtrP3UpVj75X8OTmeS25kKCwtRUVGxZDuTWrJLamkLU0ugZimSJGFmZiZYERUIBFBQUIDS0lLU19fzA26aKisrg9FoxNDQEDZs2ICysrIlb5fIM2pWeu7Ik6uOHj2Knp4edHR0KHZuFScSEFG04mk/V0ugJpF7nkAgsKD9OSsrC6biWiDKbYJGq4Xu9ttRUlqKt7/7WGIWmwH8dhcCs/OwbGuHq5sToSg+Ez/7b8yMHETH848gtyq2CZyRTgplwiy9MVAThtFoxNTUVFT3ScQbs9zO5LXP4q0Hn8EppwN5n/5ERO1MajnAN3Qtqaa2dSw+4DkvLw9WqxXNzc3IyYk90k6JFUnwI5TZbEZXVxcGBwfhdruxYcOGpAVV5dP/V6LRaFBdXQ2LxYK+vj5s3LgRVqtVsXWETiRgGS4RLSfW9nM1JIQA5dciT+qUpzn6/f5g+3NtbS10Oh2O2wKY9cZ2fc2116K0pBhvf/1+SP70mF6abJLgg6t7CJbztsDFiVAUJ/fuEex675Voe/a7sF50TtT3j6Zdngmz9MVATRixVNTodDpFzqjx+XwLPrzn5uYiMDMPiMDcj19CVX0dSq5dub9R3iioIXKqlg1UqgM1oijC5XJhYmICU1NTmJubQ2FhYcoOeKbYxPIcysrKQmdnJ0ZHRzEwMIC2trYFgdZogz+RijYAZLVa0dnZicHBQZSVlWH9+vWKPS9ZhktEkYilokapPZgSlNjzRJvIiWTq07IuvQylRVZM3fJ1BGaja/1fTVyv98NyTgdc/Xs5EYri4rPZsfv/vRF1934R6794XVT3jeXzHRNm6YeBmjBizeb4/dG/aIuiuKCdSaPRwGq1ory8HGazGQG7C2+8k+GQAhIO3P4wdHlGFF2+fcX1qCE4AqQ+QJLKdXg8nuAhwPPz87BYLDAajbBarWhpaUnqWkgZsVa/aDQaNDY24uTJk8Fza0ymM2cKRFrGGq1Y3sxzc3PR1dWFffv2Yc+ePWhpaVGsnWBxGa6c2SEikqV7RU0se43F0xyjndQZd6AGgLTtXJQ89zhsN34Zvren479ghnLtGoSptR6e46cQcHIiFMVOCgQwdtcP4OofQfPj90BniuxcrliTe0yYpRcGasJI5HhuSZIwPz8fzJR4PB7k5+eHbWeaP72wBUvyS9h30/1ozjXAetn/WnY9agiOAKsrULNU73hRURFqa2thNBqh0WjgdrujPqw6UdTwe0k38bYplZeXIy8vDwMDA2hsbERxcXFCz6iJ5bparRYtLS04ceIEuru70dHRodgh1qFluIIgQK/XM7NDREGxDHRQW6BmJYsndcY7zVGJQA0AiI0bYf3pE3DeeCs8h48rc9EMNDs8hpx15dBZTBCOn0r1cijNnf6/r2B2/1tof+FfYaytWvH28XRMMGGWPhioCUPpslufzxeczuRyuYLTmerq6lb88COcPvusHNEnYuQfv4G2HQbkX/jeJe+X6nHhodSygUpEoCZ0s2Wz2eDz+c7qHU/GOmLBD8axUSKoYrFYFpxbk6h+4XjbH+Vs7u7du9HQ0ICSkhIAyvwMmNkhoqWksv08keS9oM1mO2tSp8FgiPm6kiRBVHBLIVVUwvIfT0D3+S9jdmBEuQtnGO+xk2cmQjXXYW7kYKqXQ2luZu8Yui+4Eq1PPYjiD75v2dvGu7djwiw9MFATRrxlt/JZJHI7EwAUFhZi7dq1aGhoiOqPS5i0LfnvojeA4atuR/svH4N5W+dZX1dLMABQz1qUWoff7w9mwUIDb01NTRFNy1HLz4Nio1T1S3Z2Njo7O7F//344HA5UVFQosLqFlFhrfn4+tm7diqGhIbhcLtTW1ip2/hUnEhDRYgaDIa0ramTh9oJya7tSr3X+RHzbBYUwPfE4dF/9Olx/fiMBD5AZ/HYXAnMemLe1w82JUBQnv8ONgU/cgto7b0LN1/4p7P5NqXZ5JszUjYGaMKKtqJEjki6XC0NDQ/B4PLBYLLBaraiqqopo6ko4wumlAzUAEJj3Y8/f34KOl5+AqW3heSdq2rSoZS2xBkgkSYLb7YbNZoPdbocoirBarVizZk3UgTdKHUEQMDc3h7y8vLiuo2SbklarRXNzMwYGBnD06FGUlpbGlVldTKmASk5OzoLDkJuamhR73nMiARGFSmT7eaLJre3z8/Po7e1VbC+4HKXans5iMCDnke+j8DsPwP7irxP0IOlP8gpwdw8h//xOOF/rS/VyKN2JIg7d/xjcAyNo+cm/QG85e8+q5LAYJszUi4GaMCKpqFnczpSVlQVJklBXVweDwaDYh4ylWp9C+WcEDG3/HDb9/lkY6jcE/10tmxZAPRUk0azD6/UGq2ZmZmZgNpthtVpRUVGB7OzspK2DYidJElwuV3Ckqfw36XQ60dDQEPPfaCLOk8nPz4fJZEJ/f7+io7GVDio1NTVhYmIC/f39ivc0h04kkMeKM1hDtPoYjUa4XK6o7pOq1ie/3x/cCzqdTuTk5KCoqAg5OTno6upKymtYIr9tjV6PrLu/ieLSEkw9+nTiHigDOF/r40QoUszbv/4zui+8Ch0v/CtMTbULvqb0VN/QhJnH40FWVhZboVQgrcJlr7zyChobG1FXV4cHH3zwrK8fO3YM73//+7F582a0t7fjN7/5TfBrDzzwAOrq6tDY2Ijf/e53K15zqWyOfCL/oUOH0NfXh6GhIczMzGDNmjXYunUrGhsbkZubGzwwVinhWp9C+ZweDF16PbzH3j34TW2BGjWsZbkAiSiKsNvtOHjwIHp7ezEyMgJBELBu3Tps27YNzc3NWLNmTdxBGkosQRAwMTGB4eFh9PT04OTJk8jLy0NHRwc2b96M1tZWaDQa7N69O6YpbUBiAjWSJCEvLw+dnZ0YGxvD0aNHFQnmKf1mDiDYwjkzM4PTp08rem15s/C3v/0NgiAwoEm0Cqm5okZOABw5cgT9/f0YGBiAy+VCaWkptm7dio6ODlRWVkKr1SbtQ45fyQNqwtDeeCNKv/1VgJn2Zbl2DcLUWAtdvjnVS6EMMDd2BN0XXYXTv3p1wb/HOvVpJRqNJtgd4vf7uQdLsbSpqAkEArj55pvx6quvorKyEl1dXdi+fTuam5uDt7n//vtxxRVX4KabbsLIyAguu+wyHDlyBCMjI9ixYwf27t2LkydP4uKLL8bo6CgAhL2mvEnYt28furu7sXnz5hXbmRK1SYgkUAMA3uk5DF5yLTbt3IHsNWWqCtSoZQLV4kDN/Px8sNpCnr5VVFQU08SFeNZBsVtcNaPVaoO/Q5PJtGCjLPf0NjQ0YGJiAj09PTFNM0pUoEaj0SA3Nxdbt27FyMgI9u7di+bm5rjejBM1TUoeMX/ixAk4nU7U19cr9jhysIYTCYhWp1gOE07knsfj8QQrbGdnZ4MVtq2trcsmbxL1+rtYwlqfFvu7v0NZsRVv3/YtiB5vkh40/cwOjyJnXTm0JgN8JydTvRxKc4GZOez59G1w3/aP2PCtW6B557UuUe1JkiRBp9Px3BoVSJtATXd3N+rq6lBbe6b068orr8RLL720IFAjRwGBM60N5eXlAICXXnoJV155JXJyclBTU4O6ujp0d3cDwFnX/PnPf45Nmzbhd7/7HcbGxnDbbbfh4osvjqidKVFlt75lzqhZzDPpxuD/uhqb/vKiqgI1aglMiKIIQRAwOjoKh8MRLFGOZPoWqYcgCLDZbLDZbJibm4PZbEZRURHWrVu3YoBN/hteu3YtjEYjdu/eHXWrUaICNfKbrk6nQ2trK44dOxYMJkVySPVSEvVmLkkS9Ho92tracPDgQfT19aG9vV2RijP5tUIO8AqCwHNriFaRWAc6xFoluVggEIDD4cD09DQcDgeysrKCY7MXJwCWW48oikkJMictUANAuvD9KPk/hbD981fgd0TXnraaeI+dhMaSh+yGagijR1K9HMoARx5+Cq6BfWh7+qGEvrYEAoFgRSATZqmVNoGa8fFxVFW9O1e+srISu3btWnCbe+65B5dccgl+9KMfYXZ2Fn/4wx+C9z333HMX3Hd8fBwAgtecnJzEM888A5fLhc9+9rO4+uqrsWvXrgXtUytJWEXNCmfULDY/7sDQxZ+C9bkfQDKZFF9PLFIVNJIkCbOzs8FqC7/fD7/fj+LiYtTV1aXssCy1BK7Uso6VyJMz5MOcdTpd1JvmpeTn56OzsxODg4MoLy9f8BqzkkRV1IRef/369cjLy0NfXx9aWlpQUFAQ93WVIgeANBoN6uvrMTk5iZ6eHrS1tcFisShybYDn1hCtRkajMeqpT/EkyyRJwszMTLBqxu/3o6CgAFarFbW1tTF9QJE/5GRaoAYApPZNKPrpv8P+uVshjCvb/ppJJNcMfF6BE6FIMdN/fB3dF1wJ63dvQ87mlpXvEAO5ooYjvFMvbQI1kXjhhRdw3XXX4bbbbsMbb7yBa665BsPDwxHdt7i4GLfeeiv27t2L++67L6bHT0QwQvQK8DvdUd9v9ogN0vVfQc6Ljyu6nlglMyDg8/kwPT0Nm82GmZkZGI1GFBUVobm5GdnZ2ejt7VXsoNZY8UVuZfJhznLVjMViQVFREdavX69oW5rcarR371643e6IphklIvgRbtRiUVERtmzZgoGBAVRVVaGysjLq6yYiILl4vaWlpTCZTBgaGsL69euDFY2xXjt0zfLjBAIBSJKErKwsTiQgymCxVtREswcTBCEYmHG73cjLy4PVakVzczNycnKiXfKS60nWvicZZ9QsJlXXovCnT8J1062Y3/9W0h8/XUheAe6ePZwIRYqZPzKOk9feAc13voy1/3SV4teXK2pkoSO85T0YP8ckR9oEaioqKnD8+LsH5Z44cQIVFRULbvPUU0/hlVdeAQC85z3vgcfjwdTU1LL3lf9dq9XC4XCcdc1oJOJJG+n5NEuZe2sSp675Itb86RfQpvgA3ERW1IiiuGB0NgBYrVZUVlbCbDar9sUkHSpZkilRVTOR0Ol0aGtrw+HDh9HX14eOjo4Vzx1Q2nLBH4PBgG3btmHPnj1wu91obGyMOFCRqAPnlgoAmUwmdHV1YXh4GA6HI+YR3ktde6lzaxJ5jhQRpU4iDhOWB0LI7UxarTahewWtVpu0KVTJrqiRSSWlMD/zE+i+9BXMvNGfmkWkA0niRChSlOQRcOLLD0Jz8Djqv3MbtAruh8LtwQAmzJItbX7CXV1dGBsbw+HDhyEIAnbs2IHt27cvuM26devwxz/+EQCwb98+eDwelJSUYPv27dixYwe8Xi8OHz6MsbExbNu2bcVrquEDvnAquranxWZGxrH3o9dCTMHIylBKV9R4PB6cPHkSe/bsQW9vb3CyT3t7Ozo7O1FTUwOLxaKK3+FS1LquZPN6vfD5fBgeHkZvby9OnToFi8WCTZs2YfPmzcHWn2T8vDQaDWpra7F+/Xr09vZiZmYm7G0TeZhwODqdDh0dHcjJyUFfXx8EQYjouuEqdeIVrlJHr9ejo6MDBoMBvb298HqjP3ByuXYBud3K5/PB5/Mx4EmUgWI5THhx65Pc+nz8+HEMDg6it7cXU1NTKCgowKZNm7BlyxZUV1cnbK+QzIqaVAVqAAB5eTA+9kPkf/j/SeEi0oNr1yBMTbXQWTgRipRx/PGfof8j/xRXYn+xcPs7jUYTDIh7vV7FzgSj8NImHanX6/Hoo4/igx/8IAKBAD7zmc+gpaUFd999N7Zu3Yrt27fj4Ycfxg033IBHHnkEGo0GzzzzDDQaDVpaWnDFFVegubkZer0ejz32WPBDwFLXDJWsE/vDUeIPz95zEPv+7jNo/uUzKfte4h3PHQgE4HQ6YbPZFhzss+6rxwAAIABJREFUV1tbq/g49GRZjR8wRVGE0+nE9PR0sGpGkqSkVM0Akf3MS0tLYTAYMDQ0hPr6epSUlCx5nWQHaoB3g0lmszni82ASfUZNuHXKwdLe3l40NzejsLAw4msHAoFl1xxahsuJBESZx2AwRH1GjXyY8OTkJGw2G9xuNwwGA6xWKxoaGmAwGBK02vDrSdbZfCkN1ABAdjayH7gf1pIiTD/zYooXo26ze0aRu74cOosJwolTqV4OZQDH33qx672fRPvPfoD8rva4r7fS2VpyEFpOlvHcmsRJm0ANAFx22WW47LLLFvzbvffeG/z/zc3NeO2115a871133YW77roromvKsrOz4fP5FJliEithMr6KGpntr3tx4Kob0fj8j1PyxxTtNAZJkjA3NxfsHxcEITg6O9aD/dQkXQ7xVYLX6w1OaJqfn0d+fj6sVmvwrJmenh7k5eWlepkLmM1mbN26FYODg5iZmUF1dfWCv5tEnVETaRlpSUkJjEYjBgcHUVNTg7Vr1ypy3WhE0lIln68TelhzJD+3SA7glK8jZ3ays7NZhkuUISJtfZJbZqenpzE1NQWv1wuTyYTy8nKYzeaUviYkK1AjSVLqAzUANFot9LffhpI1pXj7oceAVbLHiYXn6EnoC/OhqS6HdORkqpdDGcB7chJ9H/oMGr9/Byqu//u4riUPblgOE2bJkVaBmmSTMzopDdREMZp7JZO/64P2hi+h4cl/VeyakYokMOH3+2G322Gz2eByuVKaCaPYyVUz8lkzer0+GGBLp+qn7OxsdHZ2Yt++fdizZw9aW1uXPFxNKdEGf0wmE7Zt24ahoSG4XC40NDQsef9UVNSEMhgM6OrqwsjICJxOJ5qbm1cMwkR67dCJBKHjI9PlOUZES1vuMOH5+fngQfMejwcWiwVWqxVlZWU4ePAgampqkrzapSUrUKOGIE0ozaevQVlJCSbvuA8Sz2IJy293AjOzMHe1wd2zJ9XLoQwgegXsu+VeuPr2ovEHd0KbvXywJex1IpxWx4RZ4vGnuYxYDrMDlG1piXY090pO/dff8NYX71T0mpFYasMiSRJcLheOHDmC/v5+DAwMwOVyYc2aNdi6dSva2tpQUVGRkUEaNVXUKLEOr9e74MygyclJFBQUYPPmzdi8eTPWrVuXlNYmpWm12uBY7J6enpjOW4lULAEVvV6PzZs3Q6vVor+/Hz6fb8nbJSpQE+l1dTodWltbkZ+fj56enhVfV6OtApIDNjy3hij5XnnlFTQ2NqKurg4PPvjgkrd58cUX0dzcjJaWFlx11cpTSvR6ffC8mdnZWbz99ts4cOAAuru7MTY2BkmSUFdXh66uLjQ1NaG0tBTZ2dlJazWKxGoN1ACA9MEPofTfH4bOZEz1UtTN54e7dxiW87ekeiWUQcaf+b/o/eB18Jw8HdP9F099Wo58bg2A4Lk13IMphxU1y4hn6oBSrTk+BQ+Hko3/x++hMxlQ/Z1vKn7tcOTAhCAIsNlsmJ6exszMDMxmM6xWK1pbW1NaubRaxfoBfnHVTCacGbQcOdDU29uLtra2hDxGrJUvGo0G9fX1OH36NHp6etDe3p6UVrJYginr1q2D2WxGf38/mpqaUFRUpMi15esDnEhAlEyBQAA333wzXn31VVRWVqKrqwvbt29Hc3Nz8DZjY2N44IEH8Nprr6GwsBCTk5MrXrO/vx+BQADve9/7oNVq8dRTT6G0tBT19fVh/66TeSZMJOI9my9SagzUAIC07VyUPPc4bDd+Gb63p1O9HPWSJLhe64flnE1w9e0B/Kkd/kGZwdWzB93vvRJtz30Phe/dGtV949mD+Xy+YOtUpn0WSAUGapaxXOltOPLUAaUCNUq2PoU69uOXoDOZUHXXlxNyfZn8gf7UqVNwOBxwOBywWq1Yt25d0qb5qJGaKmoi5fF4gkE2+ayZoqIi1NTUpNWZQbE+54qKirB582YMDAyErVyJR7xjtMvKymA0GjE0NIS6ujqUlpYquLqzxRqQLiwsDJ7/43K5zjr/R752LD8LjvAmSq7u7m7U1dWhtrYWAHDllVfipZdeWhCoeeKJJ3DzzTcHDxRf6rXJ6/XiZz/7GX7/+99j37592LJlC/x+P375y1+ivLw8orWoLVCTrKlPfvV8y2cRGzfC+rMn4bzxVngOHUv1clTNtWsAprZGeI6eRMDlTvVyKAMIkzb0f+SfUP+d27Dun6+O+H7x7MEAJsyUxJ/eMuKpqFGK0q1PoQ7/4HmM/+uPFb/u/Pw8xsfHMTQ0FGyDMZlMKCkpCY7DNJvNqzZIA6RHoEYURdjtdhw8eBA9PT3Yv38/RFFEbW0turq60NjYiOLi4rQK0sTLaDRi27Zt8Pl8wfJ7pSgxRttsNqOrqwvHjh3DwYMHE/ociyewlJOTg61bt2J+fh6Dg4NnHTQe7wHIHOFNlBzj4+OoqqoK/ndlZSXGx8cX3GZ0dBSjo6M4//zzce655+KVV1456zp6vR4ulwvf+MY3MDAwgKeffho5OTkRB2mAxLR4xiN5rU/qfn2Tyitgee4nMG1uWfnGq9zsngPIKjQju6Is1UuhDCH5/Rj96kMY/uydCMx7IrpPPHswjvBWFlONy4ilokbxQE0iy0Ul4ND9T0JnMmDNDf8Q82UCgQDsdjump6fhcDiQk5MDq9WKuro6GAwGaDQaOByOFcudKfXkqhn5kMZ0rZpJJL1eD4PBAFEUMTAwgLa2NkWqNpQ69DcrKwudnZ0YHR3FwMBAwoIUSgRTmpubMT4+HmzZMplMwWvH+3zjRAIidfD7/RgbG8POnTtx4sQJXHDBBdizZw8KCgqCt9HpdLj11ltTuErlreYzas5SUAjTTx6D5qtfx8yf30j1alRNnghl3LgBc/veSvVyKEOc2vFrzO47iPbnH4FhfcWyt1ViWihHeCuDgZplGI1GzM3NRXUfrVYbPAAvXj67E5KgfItFKEkExu58FDqTESVXfSKy+0gSZmdng1MX/H4/CgsLUVxcjLq6uiX/uNVWkpxqaqmokSQJDocDTqcTDocDWVlZKCoqWhBko7NpNBo0NjYGgwybNm2K+9BrJaczyes7efIkhoeHMTc3B6NR2UMdlRr7XVFRAbPZjIGBAdTX16O0tDSqg+yWw4kERIlVUVGB48ePB//7xIkTqKhY+CGgsrIS55xzDrKyslBTU4OGhgaMjY2hq6tr2Wvr9Xr4/f60bV9M1r5Hza1PCxgMMDzyPQh33Q3h5T+lejWq5rc7Ic7PcyIUKco9uB/d77sSrc98F0UfeE/Y2ym1v2PCLH7csS7DaDTC44msTEym0+kUe2NO1Pk0i0kBCftv/R5sv/p12Nv4fD6cPn0a+/btQ3d3N44cOQK9Xo/m5mZ0dXWhrq4OVqs17B+2WgITtLA1zeVywWazobCwEFu2bMGmTZtQVVWVkQcCJ0JFRQU2btyI/v5+2O32uK6ViDHa5eXlMBgM2L17N6amlG2jVKJVS2axWIItW2NjY4oFagBOJCBKpK6uLoyNjeHw4cMQBAE7duzA9u3bF9zmYx/7GHbu3AkAmJqawujoaPBMm+XEOnlTLZJWUZNGZ89q9FkoeOgBFN/ymVQvRfVEjwB37zDyORGKFOSbdmL3x27CkYefCnsbJYfiLD47kEn76DBQswyDwRBTRY1ygZrEnU+zmOQXse/Ge+H4/ZkshyRJcDqdOHToEPr6+jA0NIS5uTmUl5dj27ZtaG1tRXl5OXJyciK6PitqFkpmEEQURUxPT2NsbAw9PT0YHR2FKIqoq6tDQUEBNmzYsOrOmlFSQUEBtmzZggMHDuDEiRMxX0epDMZiOp0OW7duxaFDh3D48GHFghRKrzc7OxudnZ2QJAnHjx9XPJgSOsKbwRoiZej1ejz66KP44Ac/iI0bN+KKK65AS0sL7r77bvz3f/838P+z9+bxjdT3/f9rdFm3ZI3kU7bl+z7WtpZCyC4QlmU3CRCSkiUkhSaUkJCWQJqw3y8tSWgScgLfhKSEHCUhwDaFRwL9NaXdNhxtDixfko/1rr3re32sJNuSbcm65vfHZgbf1jEjjeR5/rUrjT7zto6Zz+f1eb3fbwBHjx4FSZKoq6vDtddei29961s7dnxbTzzp53wieV2f0u9aJvrkJ2H88ucBweG4OxSFpd91Q/tnLYBEmKMJsAMhEWPu16cx9ODXEFreus5lc7MMEDbMEiE9/aRJIh5HDZupTwEOWnPvRiQQwcBd/wfabz+I1XILtFotSJJEUVERpFJpQmPzzVHDhXuBT/h8PqZDk9/vh16vB0mSKCsr2yLI8OlzSQZcfO4KhQJWqxV9fX3wer2orq6O+SbHxXeS/mzp4r1DQ0NwOBxoaGhIWJjjQlgiCAJVVVXw+Xy4cOEC9Ho9NBoNq+MDQH9/P8rKyqDVajP6OiAgkAyOHz+O48ePb3js0UcfZf5NEAQef/xxPP744zGNG0/6OcCd6B0rbDqsdyMtatRsg+iDH0K2TovFh/4B1Fog1eHwGs8f6Y5Q0wh7llMdjkAaQhj00NaUgloLYvnMCLw9g/D2DGLhLRuaXnwCqkoLcyxX11B6vtXZ2Ym2tjahbk0UpP5OxmPimSSweWMOJin1aT1hfxieLzyJOokMtbW1yMnJSVikAZK3sxQNfBON2GA71wwAVFRU4ODBg6iqqgJJkoJrhkPEYjGam5shk8nQ3d0dcwtvroQaeky6eK/BYIDNZkt4pzrRduK7oVQqYbFY0NfXh5mZGVbHJggCq6uroCgKa2trrAnrAgIC7CKXy1Oafp4oyXPUcH4KzpBefwNynn0KEr021aHwnssdoXSQFQgdoQSigCCgqquA/up2KCtLQLkXsfT7Hni6+hFZfee6ujJ0HrZr7sCl37zBPMa12O3z+RAKhRAIBDJuPcY2glCzC6luz53M1Kf1hFeD6L/1PqwOnGFtTLr6Nx/IFKHG5/NhamoKdrsdXV1dcDqdMBgMaG1tRXNzM8xmM+sFZNMdrj93giBQXl4Os9kMm82GlZWVqF/LtVBDU1RUhOrqanR3d8Ptjr+rHJc38kgkApVKhYMHD2J2dpZpDc/m+LQAHQgEhBbeAgI8RKVSpTT9PFGSNe9Jm2LCO0A1NoN8/hnIzHmpDoX3+MenEfGvQVlbnupQBHiIWKeB7opm6P6sBdJsLVYGR7D4v51YHR7f9XWhJS/sH74f57/yfVAUxWnWAT32+ro1wobZzghCzS7st9Sn9QS9ATjefw98F0ZZGU9w1CROOByGy+Xa4pqpqqqC1WoVXDM8Ii8vDw0NDejt7Y2piC8XQs12Ykp2djba2towPDyMiYmJuH4PbBYT3kw4HIZYLIZEIkFLSwukUim6uroQCLBjj6fzr+nJgrCzIyDAP1I9B0uUZIhG6VifZjuoklJkP/cjKGorUh0K7wm5F+EfnYTG2pjqUAR4gKKsCPqr26FuqEJkeRVLb9ux9MdeBN1LsQ1EURj9+g9h//O/BrW8ytn8ji5UvL5ujbBhtjOCULML8RSyY7XrUwqFGgAILPrgOHoX1qamEx5LcNTEx3rXDO2A2OyaSbQttAA3aLVapojv2NhYSr5zu4kpcrkc7e3tWFpawsDAQMzXLa4dNfTYtEvJYrHAZrNhaSnGycc2rBewhI4EAgL8JJ6GDnxKfUqOUMPp8EmFMuVA808/hPqqtlSHwnuEjlD7FyJLBs2BOogaK5BVmAvfhUks/m8nlvvPgWJBpHa+9hb8n3kMywPDLES7lc2FioUNs90RhJpdSHnq02xqUp/Ws+Zagf2GjyFwKbFY+CSO8CmWzdCumXPnzqGjo2OLa6ayspJV14xQxItb6CK+Xq83LjEkUfYSU8RiMRoaGqDRaGCz2WLavU6WUENjMplw4MABDA4OJtRdazuEjgQCAvwj1XOwRElGLOme9rQFtRrKp56E/v3XpzoS/iN0hNo3yPJM0F/VCm1bAwixCN6eQUT6RrA2PcfJ+aiLl2C77qOYe/k/WB+bdkyvR9gw2xmh69MuxOOoyZTUp/X4Zz1wvOd2NL/5L5Bm6+Mag0+TJ74JNaurq0yHpkAgAL1eD6PRiPLyciGNiYfE+t0RiURoaGjA+Pg4Ojs70dLSAplMxlF0G4kmz5ggCJSUlECtVqOrqwv19fXQ6/f+nSdbqAEuL9wOHjyI/v5+LC0toba2lvUWkgAYC67QkUBAIHXEOwfjy1wjKY4afmR5sYtMBunXvgLSZITrp6dSHQ3vETpCZSAiEVS15ZBm6xCYvYTVkXEEZi8lNYTwig99d34enu5+VDz6WRAsrUd2mt/RYg3d6EEqlTIpUvsZwVGzC/Hs5rBlu40EgggtehIehy1WpxbgOHI7wiuxt8oE+OXcSLVQQ7tm/H4/Ojo6MDIyApFItME1YzAYkibS8Em0SgfiKbJGEAQsFgtKS0ths9ng9Xo5im4jsYgpJEniwIEDOHPmDKan90535LLr025xi8ViNDU1Qa1Wx+wCigbBhisgwA/inYPxpUZNMmrzZUqNms0QBAHxgw/AdPKvAR7NH/nKSt9ZSA1CR6h0RqzTQHtFM3RXNEOiVWNlYPhyIeCR3QsBc834//sZem6+FwHnAivj0TVqdoKegwWDQcHdDEGo2ZVU2m79s5cAnn05Vy5cguOG2xFheWGUbFIh1KyurmJychK9vb1MrRmJRIL29nY0NTWhsLBQqDWTRsQrPJpMJjQ3N6Ovrw9zc9xYVtcTq6hEO1YuXbqEM2fO7HotS4WjhoZ2AVVWVqKrqyuh7lU7jS90JBAQSC0qlSqtHTVisZjzuUYm1ajZDuKjH0Xut78EQiokAOyFf2wa1NoalDVCR6h0YXMhYM/bdiy9befVRj0AuN94Gx2HboendzDhsTbXqNkOYcPsHQShZheSabulKArLy8uYmJhAb28v+t78XcxjJAPvmWn0Hf8oIqFgqkOJm2QINeFwGE6nk6k1Q7tmqqurGdeMRCLhbKErwB2JdjtSq9WwWq2YnJzE+fPnOf0uxiOmiMViNDc3Iysra9dOS1x2fdprx4XGYDAw3auiLdgc7fu9uSOBsLMjIJBc0j31KRmOmoyrUbMN1NEbkfPM4xCrVakOhfcEXYvwj01C0y50hOIjhDwL2tZ66N/Viqz8HNYLAXOJf+IiOq+/ExeffyWhcaKd3wkbZpcRVom7wLXtNhgMYn5+HmfOnIHNZsPY2BgkEglqampQYcyJJ+SksNQ7hoH334lImv5ouBBqKIra4ppZXFyE0WgUXDM8gw1xIdExpFIpWltbEQgEYLfbObsBxZOmBVz++8rKyphOSx7P9rs7XLbnjlZgksvlsFqtWFlZgcPh2PO9jPU9WW/DFdpHCggkj3RPfRK6PrEHZb0Cxuf+EVKTIdWh8J6IPwBvVz90VwkdofiAJIe8XAi4tR4EAXi6B7D4u26szcynOrSYifjXMPjJv8fQg19DJBjfhn0s8zuhhbdQTHhXZDLZjrvJO7HbjZmiKHg8Hrjdbsaqn52djYKCAmg0mg1f3IU5fhQS3omFt89h6La/Qu1LP+FV/ZloYEuoCYfDWFhYgMvlwtLSEuRyOUiSRE1NDeRyOQuRCvCReMWPzYhEItTW1mJqago2m42TCX2i6UkmkwlKpRJ2ux2lpaXIz89nMbqdifU9FolEqK+vx9TUFDo6OtDc3AylUrntsdt1HNgLOpZwOAyKoiCVSgU3nIAAxyiVyphrUPHJUZMcoWb/LFqoymoYXvgJlj55P/wXJlIdDr+hKCz9/nJHKE9nHxDih3i5LxARIIrzoc7PgW96DqGJGSzypDkMW0w9cwrLfWfR+IvvICvXGNNr45mX0nMw2tkslUrTbu0ZL4JQswvxfAk235jX1tbgdrvhcrmwsrICrVYLg8EAs9kMqVS64ziBudS35t4L5+sOnPuL+1D93A9SHUpMxCvU0K4Z+vMMBoPIzs5GTk4OKisrhYUbT6EoCl6vFy6XCy6XCzKZDC0tLZBI4rv8sa3mm81mqFQq2Gw2LC4uRtVxKVrYEJVUKhUOHjwIh8MBr9eLysrKpNwg4zmH2WyGRqNBT08PqqqqYDKZthwTr3i12YYrlUrj/g4JCAjsjUKhwOpqbA0MRCIRQqEQRxHFRjKuk/sh9Wk9VH4BdM/9COK//hxWuvtTHQ7v8fyxF1m15VibuAisxOZOE4geiV4LVXUZAAorZ0cRGrsI79jFVIfFKYt/6EHH1SfQ+IvvQH9Fc9Svi2ezDNi/G2bCLJNlCIKA3+/HyMgIFhYWIJFIYDAYYLFYoFKpor5x86U1917M/VsHxPf+LSqe/naqQ4maWISaza4ZhUIBg8GQMa6ZVHfA4opQKAS32w2n0wmv1wuNRgOSJJGbmwuPxwObzbar62I32HLUrCc7OxtKpRJnzpxBSUkJCgoKWBmXrYK/EokEBw4cwMjICLq7u9HU1MRCdNyg0+lgtVpht9vh8XhQVla24fOKd5JAIxKJQFGU0MJbQIBj4nHUiMVirK2tcRQRPwiFQlhcXITL5UJYUQSRJCvVISUVSqeH6pnvQ3zyYXj+639THQ7vWTtzHrKSAkCnQeBi+qXb8BVFRQmy8k0IuZewMnQBS2/3pjqkpLM2M4+uYx9H9Tcfgvnu26J6TSLz0v24YZbZf12SWO+yWFtbQyAQgF6vR2lpadwLgnQRagDg4i/fgFj9CEq//WiqQ4mK3cQJ2jVDuy9CoZDgmkkDKIrCysoK87lFIhEYDAYUFhZCq9UyC+lgMIicnBzodDr09PSgrq4O2dnZMZ+L7YU53eraarWir68PXq8XVVVVCZ+HzVgJgkBlZSVmZ2dhs9l4UwdiO2QyGdra2nDu3Dn09PSgqamJuZmzIV6tt+FGIhHIZDJBrBEQYJl4HTV8SX1iC/r+Rs8z189L5n1ZyLytliiQy5H1nW8i+7GvY+HUq6mOhvcExi9CSuqhrCnD6tCFVIeTlhCKLGjqKiDKyoJvdBK+kXH4Utw6mw9QgSCGPvsVeLoHUPPEwxBlyXY9PhwO75pREg37acNMEGriIBQKYWFhAW63m6lNYjAYUFVVBblcjs7OThiNseXsbSbA8xo1m5n86W8gVilR/OWTqQ5lTzYLNZs/T4VCAZIkUVtbmxGuGT6TiJtnO7cTSZKor69HVtbuO4w6nQ6tra3o7e2N2cHClVBDEAQkEglaWlowMjKyRWCIBy5aaOfl5UGlUuEPf/gD5ufnkZPDz8LnIpEINTU1mJmZQUdHB5qamqBWqxN21NDQ3wF6Z0cmkwlCroAAi8RTTDhThJpgMMjc37xeL5RKJQwGA+rq6pj7WyRCYc6X/n9rvBBiMaR/9zCMOSY4v/uTVIfDe4KuRYRXfNC0N8DbKaSNRYOsMAdKSxHCPh9WBs/D0zWQ6pB4y8Wf/wrLA8Noev5xyM15Ox4XbdenvdgvG2ZpJ9S89tpruP/++xEOh3H33Xfj5MmNwsADDzyA119/HcBlp8v8/DwWFxfx+uuv44EHHmCOGxoawqlTp3DLLbfgrrvuwptvvgmdTgcAePbZZ9HS0gIATB93m82GvLw8LCwsIBKJcO6ySIcaNZsZ+95LEKmUMH/hb3Y8hotFbjysrq7C4/Fctg6Hw4JrJk3w+XyMayYQCCT0uSkUClitVjgcDiwvL0dde4VLoQZ4x7kyMzOTUIrW5nHZRKPRQKVSYWJiAl6vd0t6EZ/Iz8+HWq2Gw+FAWVkZsrKyWPuN0zZciqIYG65YLObteyEgkE7Em/qUjkINRVFYXl6Gy+WC2+1mXKHbNZug2S8dn/ZCdM89yMkxYf6L3wJ47PTkAxH/GrxdA9C9qxVLv+tOdTj8QyyGurYcEr0Gaxfn4bswicC0kC4WLZ6ufrx99YfR+PNvw3DIuu0xsXR92ov9sGGWVkJNOBzGfffdh9OnT8NsNsNqteKmm25CXV0dc8wTTzzB/Pt73/seenp6AADXXnstensv5w+63W5UVFTghhtuYI791re+hQ996EPM/y9duoTTp0/D7Xajra0NlZWV+OpXvxrVbj0bBC+5OT8HF1z4xs8hVimRf9/dW56jFzSpWMTQrhmXy4VLly7B6/UiPz9/w+6UQPLZy1ETiUSYXPzFxUXIZDKQJImqqipWWp3TtVfOnTuH3t5eNDU17an0cy3U0OTn50OpVKKnpwc1NTUgSTLmcblw1ADvxNva2sq8d42NjbzNFdZoNIwoJxaLWXfKrU+ti0Qi+6ojgYAAV8Sb+sTntMz1BAIBpguo1+uFWq0GSZJoaGiATLZ7+gCw/woJ78otH0AuacClB/8eEV9m1yhKGIrC0u/+1BHK5tj3ip9Er4W0zIxIKIzw+DSW+8+lOqS0JuhcQM9Nn0TFVx5AyWc+tuV5tuelmb5hxs9Z9Q50dHSgoqICZWVlAIATJ07glVde2SDUrOfFF1/El7/85S2Pv/TSSzh27Ni2u9SPPfYYXn31VchkMhw5cgS5ubl4/vnnk2rvDy15EfGn6Y2GAka++EOIVUrk3PWRDU/RluRkqJ2bc7pp10xeXh4kEgmys7PjWvgKsM/mi+na2hrjmvH5fNDpdDAajSgrK2PFLrnd+aurq5k22S0tLbsu5LkQanb6Xeh0OrS1tcFut2NlZQXFxcUxjcuVMErX1KHTiy5evJiw+4cmEolwErNUKkVrayscDgfm5uZQVlYW1WIoWvZrRwIBAa6Qy+UxFwbmY+oTfR2mKIpx8rrdbohEIqYLqEajifm6t8/X11ug3n0Ypp8+Bden/xahhaVUh8N7PH/shbyuAv7x6X3XEUpRUYKsPBOC7kWsDF1AqHsw1SFlFFQohOGT34Knqx913/8SxMp3NlbZSj/fTKZumKWVUDM9PY2ioiLm/2azGW+//fa2x46Pj2N0dBTXXXfdludOnTqFBx98cMNjDz/8MB599FHU1tbi17/+NXJzcwFcFocCgQAkCCz3AAAgAElEQVSLf8XeBGbTL+1pPVSEwtnPPwGRSgnjn9/CPM51h6H1rhmPx7NtTjdw2VGViZ2O0hWKorC0tASn04mFhQWIxWKQJImysjIolUrWL7Q7jWc2m6FUKtHV1YXGxkZotdod402Go4ZGLpejvb0dAwMDWF5eRk1NTdQCAFfC6GYxpaCgACqVCj09Paiurk6oRheXYi5BEMjNzQVBELDZbLt+zvGOT3ckCIVCrApBAgL7jXhEF74JNRRF4eLFi1hYWMDKygq0Wi1IkkRRUVHCBTXDEWEesxmqsQnk889g8ZOfxdrkTKrD4T3+wRGIzbkQZ3hHKEKRBXV9JcQyGVYvTAiFgJPE3L/8O1bOnEfTi09CWWoGwP0cD3hHDOJCEEo2aSXUxMKpU6fwoQ99aMuHNDMzg76+Phw9epR57LHHHkNeXh4CgQDuuece/PCHP8QjjzwCIL5idkBii7l06vi0E1SYwtBnvoY6lQKG45ffa7pKN2vnWNfpx+12b3DNVFVV7XghyNSW1PGQqvciGAzC5XLB6XRiZWUFFy9eBEmSKCkpSWn6jMFgwIEDB9Db24vy8nJGsF1PsoUa4HLdhcbGRoyOjqKrqwvNzc1RiQBcOWq2u9HqdDq0t7fDbrfD6/XCYrHEdW6uXXfhcBharRbl5eVwOBwoLi5GYWEhq+cQiUQZMUEQEEg3xGJxSlOfIpEIlpaW4HK5sLCwAL/fj0AgAIvFApVKxer1WEh92h6q2AL9cz+C59MPwDc4nOpweE94ag4iYzaU1aVYPTua6nBYI6swFwqLGeFVH1YGR4QCyiliuf8cOg6dQMNPvw7jkas5c9TQ0BtmmeJoTiuhprCwEJOTk8z/p6amdpxgnzp1Ct///ve3PP7LX/4SH/jABzbsZOTn5wMAsrKy8Jd/+Zf49re/zTwXj1CTaC2WwHx6O2poIsEIBj/+CBpeUEB/3SFmpzkRtnPNkCQZU60ZQahJPusLJbpcl4VI2vLt9/tRW1ub4gjfQalUwmq1MulGpaWlG37LqRBqgMvf27KyMqjVanR2djJdjHaDtn+yzU5iSlZWFtrb2zE0NIS+vj7U19fHfEPmWqihx1epVLBarejv78fS0lJMTiUBAYHkEOu1NhWOGp/Px6RZ+/1+6HQ6kCQJi8WC/v5+FBYWcuKuE1KfdoYymqD56dOQfO4heH/XmepweE/QuYDw8io0bQ3wdqWpoCESQVVXDqlei7XpOfhGp7A2PZfqqAQAhBY86PuLz+PAy99HRJE5IkoySCuhxmq1Ynh4GKOjoygsLMSpU6fwwgsvbDluaGgICwsLuPLKK7c89+KLL+Kxxx7b8NjMzAzy8/NBURR+/etfo6GhgXlOoVDE3R4y3i9iurXm3o1IIIyBj30BTS9/F4RKEbNAEq9rZjf4JNTwpQsWF9CimtPphMfjYQolNjY2MpPWcDic1L8/2s+drmdy5swZ9Pf3o76+fsN3jQuhJtrvck5ODhQKBRwOByorK2EymXYdN1mOGhqRSIS6ujpMTk4ydWtiKfycDEcNLV5JJBI0NzdjbGyMiZXtQsMCAgLJIxlCTTgcxuLiItxuN1Pk3mAwoKKiAgqFYsM1l20n8YY4BKFmd9RqKJ56EuIvfhmLr55OdTS8J+Jfg7d7ALqrWrH0+/ToCCUx6KCqKgUiFLxnRrDSLzio+ESWOQ+mGw/BeOwwsg8fhFiehUhXl+A4joG0EmokEgmeeuopHD16FOFwGB//+MdRX1+PRx55BO3t7bjpppsAXHbTnDhxYssCZWxsDJOTkzh8+PCGx++44w5cunQJFEWhpaUFTz/9NPNcPI4a2nobbwpHOrbm3o2wL4S+P78f2u+eRORPhaB3IxQKMbtTHo8HKpUqZtfMbvBJqMkkKIrC6uoq45qhRbX8/HxUV1fzQkGnLZHRQAsOExMT6OzsREtLC2QyGWfFhGMZU6PRMGlGy8vLO6YZcVmjZq9xi4qKoFar0d3djdraWhgMBtbGToRIJLJhkkAQBEpLS6HVatHV1YW6ujpkZ2cnfJ5MFWAFBJJNLNdcLoQRiqLg8/mYDaNAIAC9Xg+DwbBnkXuCIDhLxRJq1ESBVArpV/8BZI4Rrh+/mOpo+A9FYen3/O4IJbEUQJZrBOFdxcrQBSz9sTfVIQnQEAS0bfUwHH033GV50LXUoaqqasP1m8323PuBtBJqAOD48eM4fvz4hsceffTRDf//0pe+tO1rLRYLpqentzz+29/+dsfzJeKoiZdMqFGzmdBKEJ7PfhP+X1mgam3e8Nz6tBi3241IJAKDwYCCggJOFvh8EWpS2a6cLeidRbp9tlwuh9FozJi25wRBoKSkBEqlkkk34uK7E8/3QCaToa2tbdc0o1icOrEQ7bjZ2dlM16r8/PyoulZtFlLYZqdJAkmSaG1thd1uR0FBAYqKitL6tykgkAnIZDIEAoGk309CoRBzb1taWoJcLgdJkqiqqorJIcipoyY9upCnHIIgIP7sZ2HKycGlx74L8GD+x3c8f+yFurkGq+cnEFleTWksIqUc6roKiKRSrF6YQGDsIkJjF1Mak8A7iFUKGK69Esbjh2E8eghZuZc76lIUhfPnz6OrqwtNTU0bnPTJcNRkyvwt7YSaZBOPoyZhoSaDUp/WE/KuYfjPPwPVb5+HpLCAcc14vV6oVCoYDAY0NDRw3imFL0JNuuL3+5lCwGtra8jOzobRaERFRUXMokC6CFUmkwlyuZxZxKeiRs12bOf6Wb+g4arVdSzj0l2rBgcH0d/fj7q6ul2/J1zvtuzm2FEoFLBarRgcHMTS0hLq6uoEi66AQApRKpVYXV3lXKih06zpeUkoFEJ2djZycnJQWVkZ9zWJy1QsoZhwbBAf+QhyTSTmv/AoqGAo1eHwnmX7EOSlZkQ0KgRmLiX13FnmPChKChFe8WHlzAg8QiFgXiEvyofxxkMwHjsEw+ErIMraum4jCAIVFRWYn59HZ2cnGhoaoNVq02bezxcEoWYP4hVqErG6Zlrq03oCCz70HvkopE/9PbIrylFQUACtVpvUHy1fhBq+xLEXm7tYSKVSkCSJyspKKJXKVIeXNDQaDaxWK2w2G+sFehO9cRUXF0OlUjE3Q51Ox4ybqtSn9YjFYjQ0NGB8fBw2m22LoJTI2LGy124OHWu8NXYEBATYQ6lUwu/3czJ2MBhkmhN4vV4olUoYDAZWHaFcCjU8zErhPdSRo8h5JhvOv34I4RQ7RdIB/+gUpMnoCCURQ11XAYlWjbWpOfjGprA2Ncvd+QRiQySCtq0epmOHYTx+DTQNVVG/NCcnByqVCna7HRaLBQD3bpdMEoIEoWYPlEolPB5PTK8Ri8UJ3ZiDGZj6tJ6AcxXiB7+Oojd/CemfFpTJhI3uU5nEdmJRIBBgas2srq4yXSxKS0v3tcNAJpOhsrIS586dw+DgIGudgtgQJ0iSZFqLl5aWIj8/n1NHTazxEgQBi8UCjUazRVBKdOxYiGZ8giBQXFwMrVaL7u5uVFdXw2g0xnSeTJooCAikCoVCgdVVdhbUu6VZazQaTq47XAk1FEVBKFETH5T1IIzP/RDzn/gMKPdSqsPhPVx1hJKSeiirLKBCEawMncey4yxrYwskjlithOHaK2E6fhjGo++GLIeMeyyVSoWDBw+ir68Pfr+f83leJiEINXugVCoxPz8f02sSuTFToRCC++DG4bu4BMf1H0HLm/8CsVab1HPzxcnChzjoxSRFUfB6vXA6nXC73RCJREx7UZVKJSw610EQBEwmEyQSCbq7u9Hc3Jyww4YtK6hSqcTBgweZIsN8EmpoaEHJbrejuLgYhYWFrI0dDbHkR+v1eqZos9fr3bFos4CAADfE66ihr6mBQAButxtutxter5fpPpiMNGuAO6FGSHtKDKqyCvrnnoHvbz4P//mJVIfDe9jqCCWvKIE8z4iga/FyIeA/CIWA+YS8KB+iKxpReMsNKDp2zbYpTfEikUjQ0tKCN954A11dXWhubk7KNTjdEYSaPYi3mHC8qU9hQoSs4gKsjW8tepxprIy7YL/+djS//i8Qq5KXQsMHgYQPBINB+P1+nD9/Hj6fDxqNBiRJoqioiPXUnp1Ix1xVOp2ovLwcs7OzTCpPImlgbL4PEokEra2tOHfuHJxOJ/Lz81kZdz2JplQplUpYrVb09/fD4/FsKBrOB0fNerKystDe3o6zZ8/CbrejoaEh7o5+AgICsRGro4aiKITDYVy4cAGLi4sgCAIGgwFmsxkajSbp9xuuHLxC2lPiSEssIH76NIKfvh/hAaGt857E0RFKpFJAWVMOkVQC3/kJ+EfG4R8ZT0KwAlEhEkHX3gDjscMwHTsMdUMV+vr6kF1SwqpIQ0MQBLKysmCxWGCz2XZ0Vgu8gzDb3IN4dnMSSX0iCAK6O27F/Ne+F9fr043lkTn03fgRNP33LyFKkrLKF6Em2XHQxRLplCaKohAKhWA2m5GTk5N2gkmqWC+q5OXlQaFQoKenJ6G2zmwLVgRBoLq6GgsLCxgcHERbWxurdVbYEFMkEgmam5sxOjq6YXcl2e25o0EkEqG2thYXL15ER0cHmpuboVKpOIpQQECAJpo6gWtra0wR4NXVVYRCISgUChQXFydt02EnuOr6JAg17CAhSZA/+zH8Jx+G57/+N9XhpAV7dYSSFeVBUVSA8MoqVs6cxzKL6VICiSNWK2G47kqYjl8D4w1Xb0lpSkZakslkYurWFBUVwWw2szp+Jq1nBKFmD+iOA7GQqNVV/L73QvTEM4j41uIeI53wDEyh//hH0fAfL0KUpJZt+0WoCYfDWFhYgNPphMfjgVKp3GD7HhgYgFqtzqiLGtdsFlV0Oh3a2trQ09OzbSpPtGNycWNUKBTIyclBd3d3QkLSZthKqSIIAmVlZUzdmsbGRs67PiUyfkFBAdRqNXp7e1FZWYmcnJwdjxV+UwICibOdULO5wL1EItmQqutwOECSZMpFGoC71KewUKCGPeRyZH3nmzB8/Rtwv/hKqqNJC5btQ5CXFSGiViHgdENVWwGJRgn/1BzWJi4iMCkUAuYT8pICGG88DNPxw8h+txUi2c7XRi6FmvVzx83OarZqPmYaglCzB/EUsku06xN0WuhuOoKFf/7/4h8jzVjsvoAzt9yFuld/npRq4HwQarhidXWVcc0Eg0FkZ2cjNzcXVVVVvLsIpuPnsJ37RS6Xw2q1oq+vDysrK6isrIzpe8xVChhFUYyQ1NvbC7PZzMrOBds3cpPJBKVSCbvdztSQ4IpYatRsh1arhdVqhcPhwNLSEioqKgRRRkAAwGuvvYb7778f4XAYd999N06ePLnh+WeffRaf//znGTH7M5/5DO6+++5dx6TTz8+dO4fl5WVIJBL4/X6mwL3FYtmSipjwHIxFOKtRw48/L2MgxGJIHv6/MOaa4Hzyx6kOh/dkFRpBthdDmauA2zEP15uOVIcksB6RCLqDTTDeeOhySlN9ZdQv5VqoWT827aweGxtDZ2cnmpubWeu4lykIQs0exJv6FAgE4jofPd2Xf/iDwD4SagDA9bszGDpxD2pOPcPpwocvQg1bcUQiESwuLjK7i3K5HCRJoqamBnK5nIVIuSWZi1w23u+dRBW6UNrw8DB6e3vR2NgYdS0TLov+EgSxQUjyer0basLEOy7bN3K6K8Af/vAHhEIhFBQUcCZeJTquTCZDW1sbhoeH0d3djaampg2794JwI7DfCIfDuO+++3D69GmYzWZYrVbcdNNNqKur23Dchz/8YTz11FN7juf3+/HWW2/h9OnTmJiYQG5uLu677z4cOXIECoVi199Yop032YQ7Rw3rQwoAEN39V8gxmTD/yDcBnoh9vIAgoD9YBX1jIWSSNUQuXQRC06Cmgez8LEjffxCz/9qR6ij3NWKNCuR7roLx2CEYb3g3ZCZDXONwKdRst1FGEARKS0uh1WrR2dmJ+vp66PV6Ts6fjghCzR7EW0w44RtzdTVUrQ1Y6d5fuZ2XTvdA/PH7UfVP3+XsHHwRahJhbW0NTqcTLpcLfr8fer0eJEmirKxsX7fP3gs2FtC7LfQJgkBVVRWmp6fR2dmJlpaWqMQyLh019A1XLBajubkZFy5cSLhbVSQS4SStQCKRIC8vDysrK9sKIGzBVtpWVVUV5ubmYLPZ0NjYCI1Gw0J0AgLpR0dHByoqKlBWVgYAOHHiBF555ZUtQs1enD59Gk888QTm5uZw6NAhNDU14b3vfS/uvffeqMfgShyJB5FIhFAoxPq4glDDITffglzSgEsP/N2+KUGwHRKdGqZDdVAVaUAsXwK17AHcF7DlqxdYg1p2CSV3XoPxn72Rgkj3LwpLIXTXvwsL5Xkov/koCoqLEh4zUdfxbuwmApEkidbWVtjtdhQWFqKoKP6/JZM2ywShZg+iKWS3GbZst5o7PrjvhBoAmH3l9xDf9xDKv/8NTsbni1ATSxwURW3IyReLxSBJEuXl5VAqlXFflPjyXqQT0YgqhYWFUCgU6OrqiqqqPVdCzWanDkEQKC8vZ8SFeIviclVThx67sLAQ4XAYNpsNTU1NUKvVnJyLDXJzc5m6GKWlpZx02RIQ4DvT09MbJtZmsxlvv/32luNefvllvPXWW6iqqsITTzyxZTJusVjwgx/8ABaLBQDw3HPPYWpqKqZY9kPqk1Cjhluoqw8h55++D+enP4eQeynV4SQNTW0xsq1lkKspRJzTQGgemJ3Hnt82ioLUO4aKT1+P8z96HVSQH7+/jEMkgu6KZvgbytBw123Ibq4FcLmLq8PhwMqaP+F07GQ7atajUChgtVoxMDCApaUl1NXV8a5kQ7LZ3399FMQj1LBluyWuuxYSIzvFP9ON6Rf/G2MnH+VkbL6IE3vFEQgEMDs7i4GBAXR0dGBmZgZarRYtLS04cOAAiouLoVKp0lo5TkV7brbcFHthMBhw4MABDAwMYHZ298J6XKY+bXeTy83NRWNjI3p7e+F0Olkblw3osfPy8tDY2AiHw4H5+XlOzsUWarUaBw8exOzsLIaGhnizmy8gwCfe//73Y2xsDA6HA0eOHMGdd9655ZjKykpGpAGS33mTbTirUcOPPy+jiTQ0gvzFM8gqylzxnZBJoDtUj/J7r0fVJ6zIrRVBtjyGyOw4EI8TbHYElX/1bkj0/N1cSTfEWjVyPnAD6n/0VRy68Dqsp38G2YkboVlXd0YqlaK1tRWRSAQ9PT0Ju/i4mpdHM3cUi8WMQ9lms8V8/c80BEfNHsRbTDjuG/P634ZMBv1t74fzBz+Pb6w0Z+KZVyFSKVH893/L+th8EGo2Q1EUlpeX4XQ64Xa7AVy2AhYXFwudmXhELN8duqq93W7HysoKysrKtv0cuXKo7CaEaTQatLe3w263Y3l5GSUlJVF/x7gSluix6feCjtHhcMDr9e74/vEBukbRhQsXYLfbccUVV6Q6JAGBpFFYWIjJyUnm/1NTU1s64K0vEn733XfjC1/4wp7jpiz9nCUIghBq1KQxVLEF+l/8GN5PfxarA8OpDocV5EUmGN9VDQUpQ8Q1DSKwAsyPbE1pipPI7BjK/rweU/89jtULQveneFCUmmE8dhimY4ehv7oNok0p4NsJHgRBoLq6GhcvXmQc00qlMplh70m0aVUEQaCkpAQajQZdXV2sdi1NNwShZg8kEknMN1k2bbfSD94MPP0LgCeTjmQz9uQpiFVKFD74adbG5IuNjiAIBINBeL1euFwueDwepuMNV7U59jtcFhPeCXqnY2hoCH19fWhoaNjyHeQy9Wm373tWVhba29sxODiIgYGBqG2myewKIJPJ0NrainPnzsVcpDnZ0KllfEm7EBBIFlarFcPDwxgdHUVhYSFOnTqFF154YcMxMzMzTGrgq6++itra2j3HjcdRw7fUJy42hgShJnlQpBHqnz4N8ecegvd/O1MdTuwQBLKvqIa+oRBSsQ8R5wwQmAI1s3FvmE0irjkUXp2DSyYdFt8+y9FZMgdCLL7cpen4YRhvPAx1bfmux+82ZywoKIBKpUJPTw9qamo47aIZK7HOHQ0GA9ra2mC325Gfn4+ioqKo5sp83dCLB37OdtMcVm23efnQXX81lv7zLXbGSzco4PzXfgqxWom8e+5iZchUpj5RFMW0z15YWMDq6iqMRiMKCgpQU1OTUReXaEm3vzke94tIJEJdXR3Gx8eZIsMymSyhMaNlr/dXJBKhvr6eiS2a9ojJzmEWiUSoqalJeKeISyfQegSRVWC/IZFI8NRTT+Ho0aMIh8P4+Mc/jvr6ejzyyCNob2/HTTfdhO9+97t49dVXIZFIYDAY8Oyzz+45bjyu5kxPfaIoShBqko1KDcX3noT4S49i8ZX/THU0eyLRq2E6XA9VoXpdIeDzrLlmooFa9sBYmgV5jhWz/2pL4pnTA7FWDeP174Lx+GGQR66GjIyt09FucxmdTof29nb09vZieXkZxcXFvJhrRyKRmAsV011LBwcH0d/fj7q6un3VNEUQajiA7Ruz4sSt+1eoAYAIMPzw9yFSKZFzx20JD8eVFXknwuEwFhcX4XQ6sbS0BIVCAZIkodPpUF5eHlcx10wh2YIZ112f9qKkpARKpRKdnZ0biuSmolbPegiCgMVigVqtZmLbrYMRl8LSbiLQ+p2i6upqGI1G1sYWEBBIjOPHj+P48eMbHnv00XdqzT322GN47LHHYhoz3oYOmSzUCCJNipBKIf3KoyBzjHD96IW9j08ymoYSGNpKkaWKIHJpGgjPAbNzexcC5pLAGtQyJ3I+/GeY/+c/pjISXqAoK3onpeldrVtSmtgkKyuLKcwbi2OaS8LhcFwxiEQiNDQ0YHJyktmsUygUHETIPwShhgPYtt1S1naIi/IQnty/uZ5UmMLZB74NsVIJ8gPvS2isZDhqfD4fXC4XXC4XAoEA9Ho9TCYTKisrmYvUwsICpzEIbE+igkiioorJZIJCoYDD4UBlZSVMJhOnwkcsGI1GKBQK2O12lJeXIzc3d9vjkpn6tBl6pyie2jpctp1cDx92rgQEMoF4U5+4aIkdD4JQk1kQBAHx/ffDlGPCpa99F0hhvUNCIYPp6npoykmIg4uILDgB7xgi3pSFtD0UBW1wFup7r8OFH70J8CQtMSmIRdBf0QLj8cMwHbsGqurSpJ6eFjjGx8dhs9nQ0tKyp2OaS+Jx1KynqKgIarUa3d3dvEvr4gpBqOGAeGy3FEXB4/HA6XIBCvOG50QiEYx3ncDcPzzJZphpBxWK4My9j6JBqYD+6HviHocLoSYSiWxony2TyUCSJKqqqnZUffnQfYoPMaQbbLhf1Go1Y0tdXV1NWkpONKhUKqYA8vLy8rYFfJNVTHgn6No6Z86cQV9fH+rr66O6+Sc6SRAQEEgu8aY+8alGjSDUZB7E7R9BrtGI+YceBRUIJu28iuIckFdVQ0FKANdFUIEFYH4hqSlN8SKav4CKe96FsRe7EVpcTnU4nCHRa0C+513QXPdncJfmoPXQ1SmNh3ZMazQadHZ2oqGhATqdbttjuV4PxOuoWU92djZTt8br9W67WceX+TQbCEJNFIjFYoRCoagLWEZ7Yw4EAozrYnV1FRqNBgYDCc82LxW99xhE33kakdX93aYsEoxg4K7/i4ZfPgHdu6+Kexw2LkZra2vM5+fz+aDT6UCSJEpLS4XFYJQkI+UnEokwqWdutxsFBQUoLy+P+7xsxSyTydDe3o7+/n6srKygtDS5Oy27QRdAPnv2LOx2OxobGzd8p1PpqKGha+vEYoVlY5IgICCQPOJ11GRy6pPQmpsfUEduQI7BAOdnvoCwd4Wbk4hEMFxRDV1DAaSiVUQuzQCBSVAz3JyOc+YmUH5bAyb/ayyjOkIpyothOnYYRjqlSSKB1+uFZ3Q01aExkCSJAwcOoLe3FxaLBQUFBVuO4To9PBwOs1LDj65bMzQ0BIfDgYaGhoxddwlCTRTQ7SF3q9mwnp0WcZFIBB6Ph3FdiMViGAwGWCwWqFQqxt3gmd1GnddooL/lKNwvvJLIn5IRhNfC6D/xIJpf+QHU7a0xvz6RBTr9+bndbohEIpAkibKyMiiVypjHFdws3LFZRNPr9TAajSguLsb4+HhC+bpsiksikQiNjY2w2Wy4cOECSJLkTSFakUiE2tpaTE1NMZZZuVwOgB9CDc16K+xeLRyFGjUCAulFPI6aTBdqwhFh3sAXqLZ2GJ97Gu5PPoDgnJOVMSXZGuQcroeqUAV45kGteAFXcgsBc0nYOQvzu3NxyaTHwttDqQ4nPkQiKNvqUHjLDTAeOwxV1daNtlTXHtwOpVKJgwcPwuFwwOv1oqqqakOMXM+R2ByfbtIxNTWFjo4OXrYjZwNBqIkCekcnWqFmPWtra3C73YxrRqvVgiRJlJSUxNxiNuvDHwQEoQYAEPaF4Lj1M2j5zY+hbKiL6bWxCCTBYJD5/LxeL/P5FRcX87ZF8H6EoigsLy/D6XTC5XJdThc0GreIaMFgEDU1NZiamkJXVxdaWlpiFkbYvvkSBAG9/nK1f1oQ4dPNxmw2Q6VSoaurC/X19dDr9ZzezOOp10NbYXt7e1FQUIDi4uJtjxNq1AgIpBcymQyBQCCm1/Cp6xMXzQuE1Cd+QVVUwfD8j7B072fhHxmPawxtowXZbaXIUoYRmZ8GIrPp65qJgoh36U8doQ5i5l87Uh1OVEj0GpDXvwum44ehu+5KDE2Ow69UQllp2fZ4vm4MSSQSHDhwACMjI+ju7kZTUxMzD06GUMP2HMxsNkOj0cTdZILvCCvNKFAqlVHv6NCumbW1NXR2dm7rmombigqoDzZjucMe/xgZRGg5APv770HLf/4MisryqF8nEol2FGooisLKygqTJkNRFAwGA3MhYHuBvt8dNYn8/eFwmBHRlpaWoFarYTQaYTabdxRfaJGltLQUSqUyLmGEi12SSCQCk8mE3Nxc9PT0oLa2FuDQvCIAACAASURBVAaDgdVzJEJ2djZaW1vR29uLkpIS3hQ/Xg9thR0YGIDH49nWMcXXiZOAgMD2EAQR8/WW7YYOibDbfCNeBKGGf1B5BdD9/EcQ/83nsNLZt+fxhEIG06EGwCSFXhZCZNEFeEYR8SQhWJ5ABdagkjlhuetajD37eqrD2RZlZQmMN/4ppemqAxCt26BtIbMxPDyMnp4eNDU1bdm85cpRw8b1hCAIVFZWYnZ2lkkdV6lUSUl94mL89U0mPB4PqqurWT9HqhCEmiiQy+W7toek0yzcbjfjmhGJRGhpaWHddaG+/YOCULOO4JIfjuN/iZb/fh5ZxUVRvWazQBIKhbCwsMAs+FUqFUiSRENDA2QyGVehC7vufyKW94HupuV0OhEMBmEwGJCXl4fq6uqY38/c3FzI5XL09PQwTpFo4OLmS4+p0+kYZ4jZbIbZbN77xUlCoVDAarWir68Py8vLvPz+isViNDY27tjhIFmOGgEBAfaIdWHCp9QnLjZkhBo1/ITS6qD64VMQn/w7eE7/z5bnFSW5IK+qgtIgAeW6CCrgBlaBSGyZfZkFFYHEM4rK+45g5OnfgkqxwEpIJNBfeQDGY4cupzTt4JYBLv+2q6qqcPHiRWa+sb5OHleiB5vj5uXlQalUore3F9XV1ZDL5WnnqKGhm0wMDw/D7/fzypmeCGkp1Lz22mu4//77EQ6Hcffdd+PkyZMbnn/ggQfw+uuX1dnV1VXMz89jcXERwDsTeQAoLi7Gq6++CgAYHR3FiRMn4HK50NbWhueee45ZpKtUqg1CDd3hx+12M7Vm6CKydJqFzWbjJDWGuOYwpCYDgpfcrI+drqy5VmE/+hdoeeOfIcvN2fN4giAQDAYxOTkJp9OJUCjELPirqqqSuuO+3x01e0FRFNNNy+12QyqVwmg0orq6es/CsdGg0+kYp0hpaSny8vKieh1XQg3wjjPE4XBgZWVlSw5xLGOyjUQiQUtLC9544w3Y7XY0NzfzLgVwtw4HgqNGQCDz4VvqE9uEw8K8gbdkyZH17W/A8I1vwn3qVRiurIG+Lh8SYhUR5wywNolIBqc0xQs1M4yqTx3C+V/Ykt4RSpKtBdVSjdqPfRDkkashzdbG9PqCggIoFAp0d3dv2PTjylHD9jxGq9UybhTaaMAVXDd0EIlEqK6u5nSTPdnwa4YdBeFwGPfddx9Onz4Ns9kMq9WKm266CXV179QpeeKJJ5h/f+9730NPTw/zf4VCgd7e3i3jPvTQQ3jggQdw4sQJ3HvvvfjJT36CT33qU8xrJiYm8D//8z8oKytDTk4OdDodDAbDrrVmOPmRyqTQn7gZl773T+yOm+b457ywX/8RtLz1EqTZW50RkUiEcc0sLCxgbW2NKZZKF0hNNkLq0/bQdYGcTieWl5eZbloWi4UTJV6hUDA3KZ/PB4vFsuvvlosFwOZrhVgsRktLC4aHh9Hb24vGxsaYBRGuJgkEQSArKwsFBQWMZZaPOxd0hwO73Y7i4mIUFhYmrT03H91GAgL7BT6lPnGBkPrEbwixGLKHPoeG8hD8A/0ZVQiYS8IXR1H+4UZMnh7lvCOUstIC0/HDMN54GLorW/DHjg7kXRV/J9nN6eEFBQVp4aihod0ovb29WF5e5sx9LGyWxU7aCTUdHR2oqKhAWVkZAODEiRN45ZVXNgg163nxxRfx5S9/edcxKYrCb3/7W7zwwgsAgDvvvBNf/OIXUVdXh3//93/Hyy+/jLfeegtHjhxBXV0dSktL95yI09ZbLr7okltvBn7wcyCDJyLx4JtehOP6E2h58yWI1Wr4/X6m84/f72c6/1gsFvT396OwsDDVIaecVItFFEUxdYHolKZIJMIUbFar1awvercbj25HPTg4GFVHKLZvNNvVfKFttdPT0+js7NzQdSkauL4h5ufnQ6lUoqenBzU1NSBJkrNzxYtSqYTVakV/fz88Hg9UKpUwSRAQSDPogrzR/nb5lPrEBYJQw39MlwagyDXCP5DqSNKL8KWZyx2hjDosdJxlbVxCIoH+qgMwHjsM07HDUFaUsDY2DZ0eTruhNRoNJ/MNrmoEikQiWCwWjIyMbOn0yRZCQ4fYSTuhZnp6GkVF79QiMZvNePvtt7c9dnx8HKOjo7juuuuYx/x+P9rb2yGRSHDy5EnccsstcLlc0Ov1zI71b37zG7z11lsoLy/HsWPHIBaL0dDQgPe9731RxxmvUBPVlysnB/qjh7D4G34W30olK6NO9Fx3GyJPPgyZ8nKtmYqKig07/uFwmBdOllSLJKkkEolgcXERs7OzWFhYYFpT19fXb6gpkkxEIhHq6+sxNjaG7u5uNDc3b1uUmIvPLBKJ7PjbLywshFKpRFdX14Y0nr1IRmtIuoBbb28vVlZWduy2FC1cpWs1NzfjwoULGB8fTzhGAQGB5EJ33ozWuScINQKpRjPWBYLypzqMtCTiXYKxPAtZuQcxm0BHKKlBB/LI1TDeeAjkkXdBqo8tpSke6I5K586dw+joKEwmE+vn4HITLhKJQKfTwWQybej0yeb4wmZZbKSdUBMLp06dwoc+9KENYsn4+DgKCwtx4cIFXHfddWhsbNyy8Lnjjjvwq1/9Cv/4j/8IADh79mzUXZ9oxGIxwuFwzK1/o0Vx4lZBqNmB1ZF56P7Pd9D43/8MkWTr+7+fBZJUEggEmPbZPp8Per0eOp0OkUgEDQ0NqQ4PAJiOUAqFgnGxbK6Fw2Ux4Z3Izs7GgQMH0Nvbi7Kysqhq6STrhkhbZgcGBrC8vIyampq4z8tlulZ5eTlWVlYwNjYGg8EAjUbD+nkEBATYR6FQwOfzRS3UZNJu6mbCEQrC7IXfqFfnILo0CQCQ5ucjOCMUpYkVam0NapkTljuvxdjPol/rKKtKL6c0HTsM/Z+1gEhB8wCCIFBdXY3+/n5MT0/DbDaz6kzZbWOPjbFFIhEMBsOW1HE2SIajJtOu/2knaxUWFmJycpL5/9TU1I5foFOnTuH222/f8noAKCsrwzXXXIOenh6QJInFxUWEQiEAwMLCwoYx6d2cWOB8R6etDYqqUu7GT3OW+iYw8L6/QGSb9DC+CDV8iYMrKIqC1+vF6Ogourq60N/fj2AwiLKyMlitVlRVVUGv1/PyopqXl4fa2lp0d3djaWlpw3OpEGqAd9J4pqamcP78+T2/O8lw1NDQRdoVCgW6uroQCATiGofrGjJKpRKlpaXo6+vD7Cx3OfB8/E4LCKQrSqUy5s2yTCMcDsPpdGJ4ZDTVoQjsATnVzfxbVVGWwkjSHCoCifdyR6idBBdCKkH24YOo+sbncZXj33BV9yuo/MqDyH5XW0pEmvXo9Xrk5OSgq6tryzwyEbh21NBj03PO+fl5DA0NsbKmFRw1sZN275bVasXw8DBGR0cRCARw6tQp3HTTTVuOGxoawsLCAq688krmMbqILAA4nU787ne/Q11dHQiCwLXXXouXXnoJAPCzn/0MN998M/O6eCYJybDeaj9yK6fjpzsLHcM488FPbFnQ8mURlYlCDT2ZHBoaQkdHByYmJqBQKNDU1ITW1laUlJRApVIxn0EyxYRY0ev1OHDgAAYHBzE3N8c8niqhBninls7a2hr6+vp2vcYk+4ZIu5EsFgs6OzuxvBx75wauOwKEw2Fm8nHx4kWcO3cu436DAgKZhkKhiHmzjG/Ec51ZXV3F5OQkent70d3djcXFRWQbjBxEJ8AWonAQsjE783+ZPLViQSZAd4SS6NWXH9AokXvbcTT+7Fs4PPYm2v7txyi+72NQlhXtPlCSoSgKKpWKmUfOsOSsSpZQA7zT6VMikaC7uxvBYJDV8QX2Ju3eLYlEgqeeegpHjx5FbW0tbrvtNtTX1+ORRx5hWm0Dl900J06c2LD4OXPmDNrb29Hc3Ixrr70WJ0+eZIoQf+Mb38Djjz+OiooKuFwufOITn2BeF4+jhk594hLR8RshVvGv2wqfcL3Vj7N3fEpYjO0AG4KDz+fD1NQU7HY7M5nMy8uD1WpFfX098vLyOEsB5BqlUon29nZMTExgdHSUKX7MtlATa6HMuro66PV62Gw2RnzeTKpEMJPJhKamJjgcDszPz8f0Wq5v4rRjRyqV4sCBAyAIgpXJh4CAAHeku6Mm2o27cDgMl8uFc+fOoaOjAyMjIxCJRKipqYHVar1cb08tpGzyGaPrDIg1H/N/kccJkUqVwogyg/DFUVT/zTG0/+ezkDz/GJb/6mbo3nsYEh1/fw/0fGb95tDIyEjC6xEu50nbpSYRBIGKigoUFRXBZrPFtQm3eTwu4evmb7ykZY2a48eP4/jx4xsee/TRRzf8/0tf+tKW11111VXo6+vbdsyysjJ0dGxftEqpVMLn82373E4kpZidSgX9rcfgeu5lbs+T5sy/1gnxPZ9D5Y8eT3UoG0hXRw1FUfB4PHA6nXC73ZBKpTAajaiqqtpSzyUTkEqlaGtrw+DgIAYHBznJD45HVCkuLoZSqURnZyeampq21FxJ5c6FWq1mWp4vLy9H1SkP4D7m9Y4dgiBQWVmJ2dlZ2Gw2NDU1Qa1Wc3ZuAQGB+Ihns4xP7NZcwufzMd0p19bWkJ2dDaPRiIqKim2vheFI+s0Z9hPasa6ND0QiUNXWwNvZtf0LBKKCkCtgvPNuSHPzIfv971FTU4Ouri40NjZCq+W+SHA8rO/OJJFI0NraiqGhIdjtdjQ2Nsad5s1V1ydg9/Tz3NxcKJVK2O12VFZWIicnh5MYBDaSlkJNsqEL2cVCsroOyD58KyAINXsy8/JbEKseRtmTX011KLwjGrEoGAzC7XbD5XLB6/VCq9XCaDSipKSE6ZaWydAdoUZHRzE3N8e6Wy5e94vRaIRcLofD4UBlZeWGDgNcOWqiFapkMhna2towNDSEvr4+1NfX7zkxSYajZvP4eXl5UKlUcDgcqKioYGXykWk7OgICqUShUMTlqOFLau36+SDd8dDtdmNhYQEymQwkSaKysjKqYskcG7UFEkDpc0I0N7blcXm2Ft7kh5NRGD52D6S5+cz/s7Oz0dLSArvdnvB9m6sN081zJYIgUFtbi8nJyYTaX3Od+rSbA16j0cBqtca8CScQP5m/wmKBeB01XKc+AQBKS6G+shXLf+je+9h9ztTP/wNitRIlX3k41aEA4LejhqIorK6uMjt94XAYJEnCbDZDo9GwdmHmy0Q6GgiCQFlZGebm5tDf34/W1lbWHESJvA+0e6W3txerq6soLi4GQRCc3cxj2c2h07QmJibQ2dmJ5ubmXScmyXDUbCcWaTQaxgHk9XpRVlaWNt9LAYFMJ5E5GB82EiiKwszMDLxeL3w+H3Q6HUiSRGlpacy76kJrbv5inN5+Hi72LQIiEZDBLeO5RHnwamje/Z4tj6tUqg1zH4vFEtf4XM1Dd5orFRUVQalUxu0ISkbXp91YvwnncDjQ0NDAeSen/Uza1ahJBfE4asRicVIcNQCg/sgHk3KeTGD8B7/C1NefTHUYvCQSicDtdmN4eBg2mw3nz5+HRCJBXV0d2tvbUVpaCq1Wu+8XsHK5HBUVFdt2hIqXRCcKMpkM7e3t8Hg8THoWl46aWMWU4uJiVFRU7Nn9IBWOGhp68hEMBmG325kugLGy338fAgJsw/c52GZo18z58+fR2dkJr9eLUCjEdDysrq6G0WiMa3ETEtb6vISIhJA12rv9k2s+iEtKkhtQhiA2GEH+5ad3fH7z3Ceezc9UuI9JkkRLSwv6+/s3NKuIdtxkFRPeCXoTjiRJ2Gy2mK/PXJJpczBBqIkC3taooTn0bsjyTHsfJwAAuPCtX0D0q9+kOgxeOGro4oV9fX3o7OyE0+mEwWBAW1sbmpqaUFBQgKysrJTGyDcikQh0Oh0OHDiAgYGBmG+yO42Z6I1XJBKhoaFhQ4tsLm7m8e7mkCS5Z/eDnRwvbLHX+HThTpPJBJvNltYFTAUEMgXez8EABAIBzMzMoL+/HzabDbOzs9BqtWhpaQFJkkyKZaKLCKFGDT8xus6C8K/s+LyqqCCJ0WQIhAimex+EWLV77TiRSITGxkZkZWWhu7s75k2WVG1qqVQqWK1WTE5O4vz581GvB/gg1NCYzWZUV1eju7sbCwsLux6bTg56PiEINVGQ7NSnWL/GhEQC3e03732gwGUoYOWZf8XsT59LdSRJh6IoeL1ejI6OoqurC263G8FgEKWlpbBaraiqqgJJkkmzMaZaqIoXgiCYSv4TExMYGxtL6G9h6wZGp2cVFxfj7NmzcbtCdiORSQL9ns3MzGB4eHjLe5asrk97UVhYiLq6OvT09MDlcnEWj4CAwN6oVCrepZ/TRfUvXLiAzs5O9Pf3Y21tDRaLBQcPHmQEX4lEwqpoJKQ+8RPV6PbNSGgUYuGDixXdez8AeU1DVMcSBIHy8nLk5+fDZrPFVHycq+K80czrpFIpWltbsba2hr6+vqiuWVwLNbHO/7Ozs9Ha2oqzZ89icnJyx+PWN3MQiB7hHYuCdLDdSm65CYQ09bnYaUMEGH7ou7h0KnWFmJPlqAmHw3A6nRgaGkJHRwcmJiagUCjQ1NQEk8mEvLw8qNXqlCnd6aawr7/50h2hvF4vzpw5k9Bvns33ITc3F8XFxZifn4fb7WZtXCDxSYJEIsGBAwcQiUTQ29u7QUxKhlAT7fus0+nQ3t6O8+fPxyTEpdv3WUCA7/BlDhYMBjE3N4fBwUF0dHRgamoKKpUKzc3NaG1thcVi2fZeyqZQI6Q+8Q+RZwbSbYoIr4dYXoA0P3/XYwTeQVZaAf2td8T8uoKCAqYjlMfjieo1XNV8iVYAotOI9Ho9Ojs7sba2xsq48RCvmKJQKGC1WuF2u5n0+82kshNpOiO8Y1GQDrZbGI3Q3XhN8s6XAVBhCmf/5htwv/rvKTk/l0KN3+/H1NQU7HY7uru7sbi4iNzcXFitVtTX1yMvLw9SqXRfLioT/Zs3f2Z0ylFWVhZ6eno4cbHEg0KhQGFhIc6dO4epqSnWxmXjZksQBKqrq5Gbm7shvzkZN/JYPv+srCy0t7djZWUl6t0uAQEBdknVHIx2oI6NjaGrqwsOhwOrq6swm804ePAg6urqkJubu2uXFLZioREcNfyj0DkYlRNeVVHGeSyZAJElh+lTfwtim0Lg0cyZ6Y5Q/f39mJ+f3/P4VHfIpCkuLkZ5eTlT12q3cfmS+rQesViMpqYmyOVyJv1+PVynttNk2rpGsGBEgVwuj8lGBySx69M6FCduxeK//ldSz5nuREIRDN7zJTSoFNC/55pUhxM3tA3b6XTC7XZDKpWCJElUVVWx1plI4DLb7WbQttuZmRmm7WKq33e6zaLVaoXD4cDKygqqqqpYEarYmiQUFBRAqVSiu7sbdXV1vNxxoVuzJ9pSU0BAID4UCkXS5mChUAhutxsulwsejwdqtRokSaKxsREymSzm8ehY2BBqIhSFNM0W3pVwBAhGCIipCKTSNFtkRcKQj/ZEdahMLnTGiQbDR++GNG/7mj7Riip0Ryi73b5nR6hkd33aDaPRCLlcDofDsWPb8VR3fdoNOv1erVbDZrOhqakJGo2GlbH3K4JQEwUikShm50NKOg60tEBRUw7f0PnknjfNiQTCGPjYSTS99P+gueqKpJ03UUcNPaF0Op3wer3QarUwGo0oKSmJqSVpKuvEpGNxsd3er/z8fMjlcnR3d6OhoQE6nS6JkW2Efm/FYjFaWlowMjKCnp4eNDU1JdSylu2brV6vR1tbG3p7e6FQKGAwGFgbm02KioqgVqvR1dWFuro6ZGdnpzqk/5+9N4+O5Czvf79vVe+b1FJr3/d9G6llE7i2MbGN7csQ+DmO7WAHHPhdwDhhOYFcEhzMJSfOMYSQYzi5Jzlgs82EhIDtgCdAHPAlTiy11K1dGmlGy0ijGam19qpuddX9Q1M1Wnrvqu5qTX3+mlFXv/X2Vu9Tz/t9vo+MzC2BTqdL2Ng73hiMZVl4vV5sbGxgY2MDoVAIZrMZpaWlaGpqEuRaJ1Si5jQJ+hgG8AZpqBUMFBQLjYKFOuRDgFWDJdmT0LBszYJ4I6sfDkPtOkHp9WA8kU2Hb3V01t+C8c57Iz6eSMzIdXIcHx/H5OQkmpubw/6epaKo4TAYDHySyePxoLq6+sg4UlXUHKawsBA6nQ6jo6Ooq6tDUVFR2hQ1pw05tSUSKS3MKVwvTB+QW3UnQ8i/j7GHPwn38EimpxIVr9eLpaUl2O12OBwOuN1uXobd0tLCmxfKiEesRd1sNvMdoeKR3YrF4QWXEIKGhgYUFxcnbLQXblyhgxqNRgOr1QqPx4Pr169nrK1uLMxmM3p7e6Oa5mVb4lFGRuoIXfrE+bbNzMxgcHAQly9fhkKhQGtrK/r6+lBXV4ecnBzBboYES9RI87IYNwwD7PqVWNnRY37LhE2fDjTFgrtksoSCjnVndpIJkrM4FP/BDAN9S7N4k8lyDlpxPxX1mESTKoc7QkUqTRcr8ZHKuFySyePxYHx8/Mj1IxsSNcBBwonrajU3NycnapJETtSIRCZKnwCA3HcvaKM+7ec9Dex7gxh738fhnZpJy/niUdQwDIPNzU3Mzs5iYGAAc3NzRwLK2tpamEwm+eYwjcQTKHDdjRYXF7GwsJCeiR0jXELlsNHezs5O0uOKESTQNI3CwkIolUoMDw8jGAwKfg4h4JJKW1tbmJiYkGxSSUbmtJCMmfDxGMzr9eLKlStwOBy8b1thYSH6+vrQ0dGB0tJSqNVqoacO4GCtF+I6kY1GwiwLuPcUuLqrx8KWCeseHfz7ChACUAQIMTfXkhChoWPiU6dIAfXeDuirswk9R2M2iTSbLIdQsPzvT4I2GGMfmmC8G6sjlJilT6mMy5VdG41G2Gw23vNFTDNhoeM7rqvV/v4+Ll68KNi40Tht90NyokYkMlL6BIDodMh96MG0n/e0EHTtYfTBD8N3eT5jcwgEAlhdXcX4+DhsNhvW19eRl5eH3t5edHZ2ihpQZoJsu6jGu/ge7ggVyQVfTCIt5pziZ3JyEqurqwmPK+ZuDsuyKCkpQUVFBQYHB+F2C7e7KmQwRtM0Ojo6oNfr4+rSICMjkzzJKGoIIXC5XLh48SK/yUFRFJqammC1WlFfXw+z2ZwWz4RkyufDEWKyx6DGG6CxuqvD/JYJ1916+IIKsIfk4oobSpoQe6ikAzcSNVlixFNw1Z7wXGnfNkDkW6/jmO5/L7StnTGPS2Udj7RRJWbpkxCNF6qrq1FTU8PHRGLGYGKoXiiKQnNzM3JycnD9+vWEy1hvdeQaiQRI5Mec9q5Ph1D97u8A3/5hRs59Gghs+zD67g+i5z/PQVUW3tBMCDhFDcuycLvdfI08IQR5eXmorq6GXq/PukRGImTSHycVErkOtLe34/Lly7Db7ejq6kpbaVq0XRdO8cMZ7dXW1sb9mtKxm2OxWKDT6TAyMoKmpiZYLJaUx0627WQkuACK2+3KtCeRjMxpJV5Fjc/n442AXS4XtFotqqqqUFdXl1HJPUVRgnQDlHrpkz9IYduvgS+oAMNGXk8owoK+cSlmmJvHsYQCBQZq1oc9ohN7uqnBMtDODyf+vD0fdI2N8M5MCz+nLEVVXQfzQ4/HdWyqMSO3UeVwOHizXqkqag5TUFAArVaL0dFRqFQq0e4LxPSNzM3NRSgUgt1uR0tLi2T9CKWGnNaNE5VKlZAUP5XSp5R/IlXVUPbIdbCpsOf0YOTexxFYd4oyfigUgsfjwbVr1zAwMIDFxUVoNBp0dnbizJkzqK6uhsFgED1JI2aLcCmSidcaS3YrFrG8ZBQKBXp6ehAIBDA6Ohr39Spd9dFGoxFWqxWXL1/GwsJCyp8dwzCi3Kzl5+fzCqWrV6+e6sSqjEwmiKSoYRgGW1tbmJubw+DgIGZmZsAwDOrr61FfX4+8vDzk5+dn3BdBqI07KZY+7e1TWHNpMb9pxMquAZ6AMmqSBmChoG9ey0PHjmVAQ58F5U95W5dAPMmVD2vKiwSeTfZCVGoUfOwzYVtxh0OIjaLjpelixTRCb2pxJsO7u7u4du2aaPGsmB2ldDodent7MTs7i6WlJVHOc9qQEzVxkmjXgUyVPnHk/+ETGTv3acG3uoPRex7D/vauIOP5/X6srKxgZGQEw8PD8Hq9/M1oe3s7iouLoVQqBTmXjPQ4LLvd3RXmOxWNeIIEiqLQ0tICs9kcdwlPOltDqlQq9PX1we12nzDUSxShFTWH4QK/tbW1iCbDMjK3AhcuXEBTUxPq6+vx3HPPRTzuRz/6EQghsNlsMcc8rKjZ2NjA1atXMTY2BpvNhrW1NeTm5uLMmTPo7u5GRUUFdDpdxmOww5w2M+FgCFh3a7CwZcTyjgGugAoMSyGebUbqhjcNR4g9ek0OEQV0rPQTNblXEjARPoaSCQg4k+wm7/f/EMqS8riPFyo5cbg0fX5eHKsDMWIllUqFvLw8eL3erPPI48qqNBoN+vr6sL29LcprOG2bZXKiJk4SNbPLuFLhjv8DCoMqc+c/JXiXNjF67yMIeRKvqWRZFjs7O7h06RIGBwcxNTUFhmHQ0NAAq9WKoqIiaLXatNTIS5VsbM+dCmazGd3d3RgfHxe9I1QiQUJlZSXq6upgs9ngckUPktPdcYAz1DOZTCn5wYg5b+BAodTV1YXSUvHKJWVkpEwoFMJTTz2F1157DZOTkzh37hwmJydPHOdyufD1r38dt912W1xjDgwMYG9vD7fddhuefPJJ7O/vo7a2FlarlS+NPK6ayWT5+XGES9RkLqbcZ4ANjxqLW0YsbZuwu6e+YQScyPrNQkkffQ2HS5+AA0WNmvVDwUo3maEKuKC4knzTCeLegrKkRMAZZSe63tthfOe7E3qOkDEjV5quVCqxvr4uSHniYcRU6jQ1NUGv12NoaIg3GZY6h98P2ecvfiR9hxgKhSSz0CZqZpfpm09C08g5U5PROZwW3HNrGLvvUTBxlKzswDSnbgAAIABJREFU7+9jbW0Nk5OTGBgYwNWrV2E0GtHT04Oenh5+t08m/ezv78PlcmW81Euv16Ovrw8LCwtYXFwUbT6JBgkWiwVdXV0YGxvD+vq6YOMmQiTVCyEEVVVVqK2thc1mS0qRlI7WkISQjJdZyMgIQTLx18DAAOrr61FbWwuVSoVHHnkEL7/88onjvvCFL+Bzn/scNBpN2HGCwSC+973v4bHHHkNPTw++/e1vg2VZvPbaa3j55ZdRWVkZ079NSokaobo+pbuRaIgBNr1qLG0bsLhlwrZfg/2EkzM3oSng+Ed2vPQphIPrp5S7PxWsOgA2tc9TX1cr0GyyE9qch/wnP5Hw84Te3COEoKSkBAaDQfDSdDG9b2iaRk1NDaqqqmCz2QRtvCAWx8vPDxslJxvX3QpI2kz43nvvxTvf+U7cddddyMvLQ0lJCcxmc0bmkkx7yGRgGObGjVvqP279HWew+ZuZVNcTGQC7kyuY/7//H9R97S9PPOb1erGxsQGn04lQKIS8vDyUl5fDaDRGvUhnXHV1C+D3++F0OrGxsYFAIACaprG9vY2WlpaMJlO5kp7x8XFMTU2Jco5kggSuBtrhcMDj8aCqqurEGAzDiFaixwUgkbBYLNBqtRgZGUFdXR2KiuKv9RfLo+Y4mU7Sy8gIQTLx18rKCioqKvj/l5eX46233jpyzPDwMK5cuYIHH3wQzz//fNhxaJrG9evX8bnPfQ6dnZ0ghKCnpych80mappP2CRQamqYFWevT4VHDMMDungquPRUCoeSTMidhoaBOvgcnS59ogAX0jAu7dL5A5xYQloVuPvmyJw6VVtK3X+JCCCwf+SRoY+KtysVIfrAsC6PRiIKCAgwNDQnaGEDsblKFhYV8TCRU4wWxiLQRV1BQwDePqK2tRXFxcQZmJ10kfaX46le/ildffRV33HEH7r33Xtx2222ora1FU1MT31s+XSTTHjJe/H4/3/HH7/cjt7wDhAhwQ1FWCUOZHq4rntTHkuE7KjIMg52dHTidTmxtbUGj0SA/Px+tra2nqm12OhA6UcWyLFwuF5xOJzY3N0HTNCwWCxobG6FWqxEIBHD9+nU4HA50dnZmvBNIR0cH5ubm4PV6sb+/L2hHqGSVL1wSaWJiApOTk2hpaTkyjphdn+LxkdHr9XzHKpfLhbq6uriCITE9amRkThtixF8Mw+DTn/40XnzxxajHURSFz3zmM0nO/OYYp05RI9LLYRjAHVBid0+FvX0awiVnbsK14z557pOlTwCgZT0gbAisELGwgJh35kFcWymPQ+2ug9LrwXhuvfjcdN9ZaNu7k3quWIkaQkjYjlBS5HhZu9Fo5DfYvF4vKisrMzi7yERTNev1evT39/NxXX19fdKf82nbLJN0oqa7uxvd3d344Q9/iOeeew5jY2P48Y9/jLW1NZhMJvzzP/8z9Hp9WuYiZKKGYRjs7u5iY2MDm5ubUCgUyM/PR11dHXQ6HS5fD0KIUmTWXICcOpOcqBEIKleH8fFxeL1e5OTkwGKxoLa2NumbfVlRc0CqF9VQKIStrS04nU7s7OzAYDCgoKAAlZWVRxIfXOKioaEBKysrsNls6O7uzmhyjRCCmpoarK6uYnBwED09PRHLARIllYCGq91eWFjA0NAQuru7eRVNuj1qwsEZAc7MzGBkZAQdHR0xf4die9TIyJwmkom/ysrKjphpLy8vo6ysjP+/y+XC+Pg47rrrLgDAtWvXcPbsWbzyyivo6+uLOp9Er2VSStQIMReWZQWJC2+OB7j3FNjdU2NvnwYrQnKGg+BmO+4T8wABw940GA4RBf8cLeuBlySuuhCTVEyEj8Aw0Lc0w2UTaLwsgS6vgvnh5JudiJGoORwbcI0BuKRHOFVxpgkXy6jVan6Dze12o7m5OeF4R8xGEdz40eakUChw5swZzM7Owm63o7OzU9DNy2wlK94BTrJ+5513orCwEC+++CJmZmbg9/vTlqjRarUJdX3i4C4qgUAAm5ub2NjYgNvthslkgsViQVVVlWhfRCa3ADqLBupcFfa2s8NsSsoo83NRXV0dszY+XqSQqJHaAhQve3t7fLnZ3t4ezGYziouL0djYGNfiVFZWBrVajaGhIXR1daXtOhIOlmWh0WhQX1+PoaEhdHR0wGRKPThNNTHBJZH0ej0GBwf59ymdXZ9iza+5uRnLy8sYHBxEd3d31CRXOjxqZGROG4nEX1arFbOzs5ifn0dZWRnOnz+PH/zgB/zjOTk5cDqd/P/vuusufOUrX4mZpAEOyoeCwWDcZZdSKn0SIlEjhJqGZQFvUIEdvwr+oELU5MxhFHR4NQ1HiCGgbpgMc4oa4KD8yUtJJ1GjDHqgXBKuVFljNkK6TjzCw9AKrN7+Lpg83qRLi8RU1HBwG0ETExOYmppKKukhJpHiJM6gd35+HsPDw+jq6kqoTF3szax4ys8JIWhsbMTq6ioGBgbQ3d2dkK9ntt7TREPSiRqv18ubM7700ku4evUqNjY28Oijj+K73/1uWjNtOp0uIZMplmXBMAzm5+extXUgk8zPz0dlZSUMBkNavkys+UC2l1NnwtqQM8bRMrFQmk3Q6XSn7kKQ6WRRPLAsC4/Hg/X1dWxuboIQAovFgvr6+qTNmS0WC9RqNRwOB1pbWzPmf8UFCXl5eeju7sbIyAgaGhpQUFAgyLipUlhYCI1GA4fDgebmZkkoag5TXl7Odz9oa2tDbm6uYGMnw2m7PsjcmiQTfykUCrzwwgu47777EAqF8OSTT6KtrQ3PPPMM+vr6cPbs2aTnw6ma473xOG2KmlT8aXxBCtt+DfxBBRg2vdcnAvZIO+5wHLT2PkiqhQ6VOukY10F2SSLX1IJrIwAjXPKP8u2AJQQkC2IwITA9/ARyb7sDExMTSZcWpSNRA9xUFV++fBl2ux1dXV2SUndEeg8IIaitrT2xwRYPYsdIiZSfl5SUQKfTwW63S957R2yk8607BPej+dGPfoTvf//78Hg8WFhYwNNPP40777wzI3OKR1Gzv7/Pq2Z2d3cRCASg0WjQ2dkpmvlmNFi9CaxSDWMFA+foJpigNIKWbIUQFiGvF5TBINB4mVfUZJpor59hmCMlTTqdDhaLRdDfk9FoRG9vL+x2O2pqajJiYnZ4ceQ6QjkcDvh8vpRqjYVcdE0mE3p7e+FwOEBRVEImvomSTBBmNptx5swZOBwOVFZWHim34JAVNTIysUk1/nrggQfwwAMPHPnbl770pbDH/upXv4p7XlqtFn6/P2614WlL1CSqqPHvU9jxqeENKm4kQjJDLDUNcLTz02FFjQL7ULM+7BFpdMrUC2AifBiy5wOpqACWlgQdV4qou/qgu/NeEEL4GMfv9ycV46QjUcOdp66uji9N7+7uhlarFfTcYlFUVAStVstvsOXnxzbmloKi5jA5OTn8d8XtdkuyDC0dSEfLFYbV1VV0dnbiPe95D8xmM86fP4//+I//yEjP+HCKGm6Xf2lpCXa7nf8ylZaWor+/ny9vSvSmUrDvISFgci2gFBRM1cIkF25liEKBkGsn09M4dRy+8AYCAayurmJsbAw2mw0bGxsoKChAX18f2tvbUVxcLHjSU6PRwGq1Ynl5GfPz82lPnh0PEjgz362tLUxNTSU9H6F3nrj3yefzYWlpSXJJRq1WC6vVirW1NczMzJyYn+xRIyMTP1KKv4CDGCyR8nOKoiRzjRImURP7tQT2Kay5tVjYNGJlxwB3QJXRJA1FInvTHCZ0yFA4dMw8WCptunN2F0F2hFemK4tOv1KAyslF7gc/xv9fpVKht7cXm5ubYdfqaIhReh2rQUJJSQmam5sxPDyMnZ3suQfgNthmZ2ePeIdFQkqKGg61Wg2r1Qq3243x8XHJJN/TiaSjVqPRiPn5edx33334i7/4C3R1deH555/H9PQ0Lly4gKamJtTX1+O555478dxPfepTvBleY2MjL4d3OBx429vehra2NnR2duKf/umf+Od88IMfRE1NDf88h8PBP6bX6+H1euF2uzE9PY2ZmRkMDg7i8uXLUCgUaG1tRV9fH2pra5GTkwNCCCiKyniNNGs+KJ/IqZNOnW+2QpQKMF63cOPJihqwLItgMIjFxUUMDw9jbGwMe3t7qKmpgdVqRWNjI/Ly8kS/weZMzNxud0rJkWSIJLvljNTsdjv29/cTHleMRZemaZjN5pTmJSYKhQLd3d2gKArDw8NH5peu9twyMqeBaPEXkP6SWU5Rk40I0fUpkqJmPwSsezRY3DLiyo4Brj3VjZbXmd55ZqGg4/uOHE4mMceE/npWGomavHhMhAkF0IkVKuiVpzwGJAS5H/oEaONRTxqaptHV1QUAGBkZSej3IYaZcKwxuY5QExMTuH79uqDnFxNug21zczNmbJsORU0y41MUhba2NphMJgwODmJvby/isadRcSPJRA33g71w4QI+/OEP4/7770dRURE++tGPwmQyYWZmBk899RR++tOfYnJyEufOncPk5OSRMb72ta/B4XDA4XDg6aefxvvf/34AB7sy3/nOdzAxMYELFy7gk5/8JLa3t/nnPf/88/zzursP2sctLCzgjTfewLlz53DHHXfg9ddf53f5Ozo6UFpaGrZzjBSkt0zuQaJGZVBCV5Qdkj2pQhQKMMHsDBSlBFfSNDs7i4sXL2J7exsqlQrt7e3o7e1FdXV12nycDsPVJKtUKjgcjrQlWaPJbhsaGlBUVASbzZbwTYoYtdzcuFVVVSguLsbg4KBg3fCEgnvfysrKMDAwAM+N9qfpas99GgMFmVuHWPHX3NwcgPQnahJV1EgJIdQ9+4eWo30G2PCqsbRlwOK2Cbt+NfYZKSRnbkIRxPSm4TiiqDl2W6Jm/aDZoJBTSxjFvg/KxcnYBzb1AG23JTS2JuiDMgMl1+lC/64HoInQipsQgqamJpjNZthsNgSDsT9nMa478cZKXEeopaUlLCwsZM1GK03T6OzshEqlOrGBdRixN7NSKT8nhKCqqgr19fWw2WxZpWxKFUkmajhKSkqwu7sL4GbwW1RUhKWlJdTX16Ourg4qlQqPPPIIXn755YjjnDt3Do8++igAoLGxEQ0NDQCA0tJSFBYWYn19/cRz5ubm8Cd/8ifo6+vDxz/+cSgUCrzjHe/A8PAwPv7xj8e1yy+FRA2be9OQVFbVpAZRKgBGOAXBraSoCQaDuHbtGiYmJjA4OIi1tTXk5eXxhnIlJSVQqVSZniYIIfycbDZb1My9UMSS3ZaVlaGxsRFDQ0NwueLfXRRrd4Qbt7S0lJcDH052S4Xi4mK0tbXB4XBgY2ND9qiRkUmASPEX11kt3QlJzkw4GxHivdrbB7a8KixtG7C4ZcK2T4MgQ0NKyZmbsFDGqaYBjnrUgFBgjt2a6DNc/lRwfQwIxUgiGHLBNHaC9NwJqCJ3HwyHvr4uhdlJF0V5FUz/6wMxj6uqqkJ1dTUGBwdjJmPTZSYcCa4jlMvlwtTUVMR7PKnF9pzfTllZWcT3WezNLCFi0vz8fPT09GBychJXr14VaGbSRpKJGu6DfPTRR9Hc3AwAvNv2Rz7yERgMBlRUVPA/rPLycqysrIQda3FxEfPz87j77rtPPDYwMIBAIIC6upsXyT/7sz9DZ2cnvvzlL+Ptb3873njjDfzsZz/D7/3e7yUc5EuhPSRjvumqri/RQqGTpH90VkAU8k1eIni9Xt6/aWRkhDfH7e/vR1NTE/Lz8yXrGVJWVoa6ujoMDQ3xigxAnMU3Htkt1xFqbGwsbGI5HGIpag4vtpyJ79TUFFZXVwU/V6pwZnRzc3PY2dkR/fsmq2lksp1Y8RdXriAnasQnxACX1xX49UUtRlaM2PRpEQxJNTlzE5pKzGvxeCeqEKTlU6Oft8U8hunoB9EZQTQ6kK63JzS+SnsK43KlCuaPfBIkTk/BwsJCtLa2wm63R1VLZDpRA9xUX6vV6ogl4FL1xCsuLubf583NzSOPpWPOQnx2nLLp+vXrCXscZSPS+xbh5gd55513oqqqCj//+c/x3e9+Fz/84Q+Rn58fl3s1x/nz5/HQQw+dSLKsrq7i8ccfx7e//W3+i/lXf/VXmJ6exuDgIFiWxejoKN/6V6vVJhwkSEJRY76pqCGEIKfWmMHZZDdEqRA0WSMFRY2Qc2BZFtvb25ibm8Pg4CBmZ2dB0zTv31RTUwOj0Zg1N7MWiwUdHR1wOBzY2tri/56pIIHrCHX58mUsxdElQmxFDQdn4nv16lXMzc2lZH4sBpwZnd/vx+LiYsavyTIyUiZW/BWuo1o6SCYGA6S3sx0LhgGWNhT4zawW/z6hx9Q1Ddx70k/O3ISFgkrsPQ8xR9cp5pihsJZ1g7CZuW6bXMugtmJ4klS3gDXlgKhudKfqeDugjb+BB7W7Dkonjc5WQpHzu49DWVaR0HNyc3N5H5hIG1JiJGqSiZU4hUppaWnYEnAxS89TJScnB729vbh48SKWl5f5v0s1uRSO436EXNlcttxfJIKkP5H19XV84hOfwOc//3n8+te/xgsvvIDPf/7zCAaDuHLlCv+FXV5ejhg8nD9/ni974tjd3cWDDz6Iv/zLv8Ttt9/O/72kpASEEKjVanzoQx/CwMAA/1gyuzlSSNQwh0qfACCnxggSb+GwzBEIrQBRqTL+mUqJ/f19rK2tYXJyEgMDA1hdXUVubi7OnDmDrq4ulJWVhfVvyha49t3T09O4du1axndzDneEmp6ejrpop0NRw8GZMQeDQYyOjialJBQzSKAoCjk5OdBqtRgaGspY5xoZmWwhUvyVKTPhZGIwKWyGxAPLAle3abx5SYMLk3qMXdVgx0+DzZrkzE0UVOx23MdhQXC4qdVxRQ0FFlpWuEYOiRDTRFijQ6juQHlG1Ac+kESpAjlzV/wnYRjoW5qTnKH0UHeegf7u+5N6rk6n4zekwnUqynQMdpxIHaHEimeEev2cybDT6eRVKdmUqAGO+hEODg4eUb+fJiT5iXA3wi+99BJycnJgs9nwj//4j3jjjTdQXl6OyclJzM7O4tKlSwgEAjh//jzOnj17Ypzp6WlsbW3hbW97G/+3QCCA973vfXjiiSfw0EMPHTmek+6zLIuf/OQnaG9v5x9LVlGT6dInaPVgNTcz9bSahqFCn8EJZS9EqQChaTACdbrJliDyOD6fD8vLy7zpttvtRnl5Ofr7+9HS0gKLxRJ3maBYyQQhOdy+e2FhQfDxE30POGM4mqbhcDiidl4Sa0cn3GJOCEFLSwvy8vKS8vcR28iOZVlUVlaiuroaNpstIb8fGZlbhVjx17lz544cly6yNgaLwpqLwluXNbgwoYf9ihZbXgVYVtrrYXTia8cdjsPlTyFyshQoE+VPdGgPqsXxqMcw7f2AQgEo1SDUofWrxQoYzXGfS5OXE/ugLIAy5SL3g09FfDyemITbkHI6nbh48eKROFlqiRogfEeoWN6DySJkMoXrvEVRFOx2O4LBYFYlajiKi4vR3t4Oh8MRtzVANiHpT0ShUPDGdW73QTY9NzcXLpcLL7zwAh544AG0tLTg4YcfRltbG5555hm88sor/PPPnz+PRx555MgP8Ic//CHeeOMNvPjiiyfacP/+7/8+Ojo60NHRAafTiT//8z/nn5fMbg5N05JQXxxX1eTKpsLJceMmMhgUZjde6gkKDpZlsbu7i0uXLmFwcBAzMzMAgKamJr4lvclkyprXkwycYsTlcsHn8wmaYEtWdtvQ0JBW02OOWJ46FRUVvDN/IskQsY3suPELCgrQ2dmJsbExrK2tCXqO0/wbkLm1iBR/Zarbhl6vT7jrk1RisMNseigMLqhxYUKPwQUdnB7FCY+WbEVJJ66m4Thc/sTgZMJez7gOpEdppGBtHNiPEu+V1oDNP4iv+bKnGxCaBul7V9znonzbB+29s5zcD30ctCn1pBNN0+ju7gbDMBgdHeV/x1JM1AAnO0KFQqG0KZpTgYsli4uLcenSJUkntqNhMplgtVp5P7XThKRfUWlpKe8NYTAc1Hu2t7ejuLgYDzzwAO6///4jP4QvfelLR57/xS9+8cSYH/jAB/CBD4R3IX/99dcjziVbS5+AGz411xb5/2vy1FCbVdjbkuX/iUBuXABYvw/Qx19/HA2pKmpCoRA2NzfhdDqxu7sLo9EIi8WCqqqqrL4QprJwUhSFlpYW/Pd//zfsdju6uroEUYCkEiSUlZVBo9HAZrOhs7MTRmN6PKhizTc/Px9dXV0YHR3lu2jFQmzZ7WHFjsFgQF9fH0ZGRuB2u1FTUyMnWWRkDhEt/gKQ9p1XnU4Hv9+f0HOkEoPt+gl2VW349wndjTbapw8CNu523OE43PnpeOkTACiwDxXrR4Bokz9JghgWopQ9KVQINXXy/yXqMB4zDZ3AyG+AzWsxz0X2fNA1NsI7M53MVCWB/u77oek4E/HxRONdQgiam5uxsLCAoaEhdHd3i5aoEeJ6xnWEmpiYwM7OTlYkajhKS0vhcrmwvLwMi8UCszl+NVg8pONeR6VS8b6ypwlJ3nFxX8KHH34YAOD3+xEMBhEMBtHf38/fjKQzsE5WdssZHCWC0K/quKIGOFDVXLc5BT7T6YZL1DA+D4CT72nC40ms9Glvbw9OpxNOpxOBQABms5mvvxXrtyal1x8vOp0ORUVFsNls6O7uTtmDJ9XA43BSpLGxERaLJaX5CMXhZIjH40F1dXXU1yl2oua4YkelUvH+Q2NjY2hra5Pbd8vc8kgx/gKyr/TJGwDm1lS4vqtAIEQASg9kPmckGooU1DQAwDA3n3zcTJhDz7oQQHoSNQb3Kihn+G6yAMC29QEqFf//cIkaQiig/7fBXvheXOfUlBVlbaJGUVYJ0+8+LsrY1dXV0Gq1sNlsvJeokMTTeTNeuI5QnK/h/v6+oJubYsZJarUaVVVVmJ6eRlVVFUpLSwUbW+zS9tOMJBM1h5mcnMR//ud/wu12Y39/H9vb2+js7MTjjz+eVn8LlUqVsAElTdNpLUmIBGs+uZttqNBjfXQTTOAURw5Cw11kgpn/TIWAZVkEAgFcu3YN8/PzoGkaFosFDQ0Nac1KZ5OSgbvmcCbJQ0ND6Orqgl6fvO+TENcxLilit9vh8/lQUZFYtwWx4JIhk5OTmJiYQGtra8QgIx2KmuPjUxSF1tZWLC0twWazoauriy/3kJG51ZFK/AUkp6hJd+mTPwhcWldhdUeBvX2C7OnUlBoUSd6bhiOWogY48KnZomOrM4Ugf2U48oOWUjBFh29iCYgq/LpBqprBFlcdUbVHQoksVbkrlDB/5I9BlKrYxyZJUVER1Go1HA4Hr+oTCqGvZYQQlJeXY3d3F4ODg+ju7oZWK0yCUczrLsMwfBfPsbExuN1uNDQ0CHI+sUvbTzOSfNe4XfbR0VE8/vjjGBo6kB9qNBoUFBQgNzcXQHpv8JI5l1Rkt4z5pPqDoinkVMutuhOBU9SAEWaHLhOKmlAohI2NDczMzGBgYAC7u7tQq9Xo6upCT08PKioqTqV0UCgOL5KR2ncnM6YQCxhnwMd9vlJRK1EUhba2Nuj1+qgdl9LRcSDSdbyyshINDQ0YGhpKyYMjm5KOMjLhkGL8BRwoahL1qElHDBbYB6avKfH6tA7/Ma3HwoYKe/sUbpUkDcBCQae+1oTYQx41ERQ1GtYHmk1cpZ4oVCgA1cJo+AcJhVDrGRyWDxGV5kA9EwHSf09c5yWuLSgFTkKkA9NDH4CyvEr08+Tm5qKiogKrq6uCmsaK1fJbr9ejpaXlREeoVMcVK07ixuZaXwOI2bAikbFlRU1ySFpRs7Ozg4qKCnzrW9868Vg2tBGTSseBcKVPAJBTZ8TWxcwYA2YlNy4yBMLcAKcr0A0EAnA6ndjY2IDP54PZbEZhYSEaGhqwsLAAk8kEpVKZlrlkO8cXdK59t91uR01NTVI7PULKbjkX/9nZWfh8PoRCIUksjoQQ1NTUQK/X88qV4yqkTF/T8/Ly0NPTg5GREVRXV6OkpCRjc5GRyTRSi7+S9QkUIwbbZ4AFpwLL20p49m6lpMxJKIKUvGk4Dpc+haLcmugYF1x0XuonjELB+iRIILx6i23uAY6pI8L60xx+vKQabGUTsDQT89yqqkoEr8X2tJEK6vZu6N/1QPrOd6M85/Lly9jb20N5eXnKY4rpe5Obm4uenh44HA7U1dWhqKgopXHTkagBDmK2xsZGrKysYHBwED09PSmpjdOlqDmNm2WSTNRwb3RRUREUCgUGBgZgMBjg9/vh8XigVqvR39+f4VnGJmnZrcDfMzY3vGeFUq+ErlgL77XEgp9bEkJAuEQNLdwHJIbqgWVZeDwePjlDCEF+fj5qa2uh0+kkdSFLp+pDiHOFW9C59t0OhwM+ny+mF0s8Y6YCt8BevXqVX2BT9dERisLCQmg0GjgcDjQ3NyM/P59/TArSWK5zw+joKFwul2CyXxmZbEGq8VemO28yDHBlS4E1F40Nt+JIqc6tizBqGuBo6VMkRQ1w0P1J7ESNcTGCibApD0x59Yk/x0rUAAeqGvbKxZidqxTq7PleUUYTcj/0VFrXSIZhoFAo0NfXh9HRUfh8PtTX16c0BzGSH4c34Li4wuFwwOv1JhwjHh83HYkajrKyMuh0OgwNDaGtrY1XVAoxtkx8SDJRw2EymTA/P4/3ve99OHPmDJRKJYLBIOrr69Hf35/2GulEkUrpE1QaMHoTKM/uiYdy60xyoiYOKPVNxQmhaYT2Q6AVqSkVhCx9YhgG29vbcDqd2N7ehlar5UtzVCrx6oaFQMq/4eNEuuZw7bsnJiYwNTWFlpaWuF+XWNcxlUqFhoaGtHeEioXJZEJfXx8ftHB+OizLSkL9o1Ao0NPTg9nZWdjtdnR2dmZ1pzMZmWSQWvyVrJlwKjEYywJON4WrO0qsuWgE9uUbjcPQlDBqGuBoe+5IHjUAoGXdICwDVqRW1nrvGqi1pTCPEDAdViAAn6v1AAAgAElEQVTMzebx1tzhIPnFYOs7gdmRqMdp99xw6XRgEizzywS5H/w46BxhuwPFAyGEb9/NNQNob29POhGQjk5SXEeoyclJTE5OoqWlJan5pjtRAwBmszlltbFU1N3ZiKSjT4vFgnPnzvEZPLVaDYVCwZdppPsGjxCS0I9EKqVPwI0W3WESNbpiLZR6BYKe1GsQTzOU5uhPZT+wB1qRWS+XYDCIjY0NOJ1OeL1e5ObmwmKxoL6+Xs5ci0S0BZ1z+7906VJC7bvFChIA4TtCCZVYVKvV6Ovrw9jYGDweD5qamiShqOE4rkrq6uqKy7spm5KOMjLRkFr8lWx77mRjMO8ewf/Ma+ELSuOaJD1YKCjhFLEsCFj2wPqFiZKoocBCy3rgJeJsPEQ0Ea5rA2swnfw7oQBldMVqKBTC1tYWts01qCZjoNgoyUMmBH1LM1xDUcyMJYDurvug6epL+HmpXjcOJ0AIIWhpaTnSvjuZMn6xPGqOj8n59c3Pz2N4eBjd3d0JbwJlIlED3FQFjYyMwO12J6xikhU1ySPpRA1N06AoCmtra/B4PHC5XPD7/aBpGvfff3/a56PRaOD3++M2W013x4FoMLmFoJcvnfg7IQQ5dSY4RzczMKvsgVYfvfgz+6kb2iWjqPF6vXwLbZZlkZ+fj+rqauj1+qy8SZS6Ku44seZLCEF9fT1WVlbibt8tpEfN4TG5RdFgMKC3t5cvzUqlI5RQxsfATT+dubk5DA8Po6CgQLSFPNkEU2lpKXQ6Hex2O1paWpCXJ67kXkZGKkgt/krGTDiVGCzEEjlJEwUFlVo77nCEWAIFOcjWhECBjtDPXMe44KWET9RQoX2o58MoXrQGhGoawz6HqMOXk3PegE6nE36/H2azGZbqOlBuKzD5VtR5aPJy4ErqFaQHRWk5ch5+IiPnDheDVVdXQ6PR8DFXoh2W0qGo4SCE8DYEyXSEErvrU7TNRU45PjMzg5GREXR0dMStkkmXoiab7ifiRdKJGo/Hg97eXuTl5UGpVIJlWayurqKsrCwjgQJXIx1vokYypU+4oaiJgKnagI2JLbAhaXSJkSLUsURNKJieNooMw2BnZwdOpxNbW1vQaDSwWCxoa2uTjO/IrUY8C0Ei7bvFKPk5vpgfVrB4vV40NjYmtaAJnVQihKChoQFXr17F7OwsysrKBBv7MKkECbm5uXyiq6ysTDKtz2VkxERq8Re3UZYIFEWl0LFEjocik3o77nCEGMKrdBjQURM1vPxGQCybUyB7J5OBTMdtfDOJ4xDVwU12JG/Aurq6I+s/2/dOsLN2IEoMSfm2D5Q60ZQ3mUKhgPkjnwRRZSb+jJSoKC4uhlqtxvDwMDo6OmAyhVE/RRlTTI+acBQXF0Oj0WB4eBjt7e3IycmJe1yxNrTiUTUTQtDc3Izl5WU+0RSPybDc9Sl5JJ2oMRgMJ1qajY+P4+tf/3pG5pOomZ2USp8idX4CAFpFw1ihx+6CO40zyi6OJ2rYwF7KY0ZS1Ozv7/MlTW63Gzk5ObBYLKitrZUvdBkmkd0Mi8UCtVoNh8OB1tZWmM3ha7nFkt0eX3APd4RKdDck2rhCUFpait3dXSwvL8NisSRtWBeJVOfNGUaPj4/D7XajqalJlvHKnGqkFn8l83tLJQY7hRuzgqGkhVfTAADDUsCN5EyIKKCM0IpbiSBU7B4CJPkuNOEwLoQxEa6oB2uOrKT0BBlcu3gxbm9AojWA7fgtYPhXEcckez5oGxvgm4ndJSrdmN7/+1BWVCf1XLEaOnCYzWZ0d3djZGQkoVLvdCpqDpNMR6hMlT4dp7y8nDcZjifRJKXS9mwj69619vZ2eDyejJw7UTO7ZGW3YsQH0RQ1AJBTF3/2+VaEHEvUkH1hFTU+nw9XrlyB3W6Hw+GAx+NBRUUF+vv70dzcDIvFIkqSRkhD42xAiProRMbg2ndPT0/jWoSWm+mW3XIBjM1mw95eYglHMYMEtVqN6upqTE1NYXV1VdCxhdjNoWkanZ2dUKlUGB4eRjB48ibiNMpuZWQ4Mhl/JYOUys9PCwSsYAbCxznS+SmKTw0A6NiTnoupoPVtgL42f/SPSjVCDW1Rn7e+40Z+fj76+vrQ0dGBkpKSmA0cSNc7AE10Zb62LHHDVrFRt3ZCf8//mfTzCSGix2B6vR69vb2Ym5vD8vJyXGOKVX4ez5ic98vS0hLm5+djxuNSSdQAQF5eHnp6ejAxMRExvk12bJmbSP5d83q9ePPNN/HGG2/A5XIhGAzi6aefzohSJVFFjZSC9miKGgDQmNXQ5MmlNJGgVMcMypjUzJdZloXb7YbL5cLg4CAuXrwIiqLQ0tKCvr4+1NbWwmQySeo7JAanzaMmHJwaY3l5OexCnAnZbXl5Oerr62Gz2eB2x6+kEztI4N4rrhRKqCSiULs5hBDU1dWhsrISg4ODCb13MjLZhpTir2RIpfw8e1al9KIQSU0DHJQ+8f+O0qIbOGjTLSQFYUyE2fZ+QBkl6UIrUNfYjPz8/ITWF6LSgHTfEfUYJVJXbQsJZTAh98mnMx6vxRODqdVqWK1WrK2tYW5uLmYckSlFDQfXEcrj8WBycjLqNUtKiRrgZqJpeXkZly5divheyx41ySP5RM0zzzyDj3zkI/jDP/xD/M3f/A38fj++9a1vJbwTLATJtIeUCmyOBWyML7CsqokMpT5aJUgi1E5HY39/H+vr65iamsLAwACuX78OpVKJnp4edHV1oaysLK5aT5nMkWzigDNhc7vdmJqaOjJOpoIEriPUyMgINjY24hpX7CCBpmn+vQqFQhgZGRHkplDo+ujCwkJ0dHRgZGQE6+vrgo0rIyMlpBR/AYlff6XkE3gaoIg43jQcB6VPN/4dQ1GjZn2gWYG6lTIhqOftR/4Uyi9ByBK9FCWettwRabsNMEQuFyGuLSiLi5MfX2By/uCjoHPT34r7OPHGSzRNo6enB4FAAOPj41GvA+nq+hQNriOUVqvF8PBwRG8tsc2Ek4nvlEolzpw5g0AggNHR0bAxm6yoSR7Jvmvcgvzqq69idHQUs7Oz+PGPfwyj0YixsTG4XOn3RE9UUSMpFAqwpugdSwzletBqyX4lMspxRQ0hiCsA9Pv9WF5exsjICOx2O3Z3d1FaWor+/n40NDRApVIl3J5PJnlSVWikskhy7btVKhXsdju/mGVSdmswGNDX1xe3TFgM9Q/HYdULZ1iXn5+fVIlWtLGFwmg0wmq1Yn5+HgsLC7dUCaHM6UaK8dfxucUDTdMJJXo5pevCwgLGx8eTmd4phoWCFvcad0RREyNRQ3DDVFgACjZnQPw3S/oYQiHY1AUSo8aLqJNP1BCFEqT3nVGP0dfXJj2+kOjuuAfanv5MTwNAYjEY177bYDBETX6IkURIZkyuI1R5eTkGBwfD3m+KmfBIJb7jKgLMZnPYmC0diprTqKYBJJyo4WhpacGFCxcQCATAsix+85vfQK1Wh/UHEJusTtQAYGOUP1E0galG+JaHp4HjZsKEEOwHTvrUsCyL3d1dXL58GTabjVdPNDY2wmq1oq6uDjk5OYLU6p4GxL655T6PS5cuYXh4GHa7PeEWr8fHS+Vz49p3FxUV8YuZWIqaeMfkOkKtr6/j4sWLUT8TMZJKh8c+HiRUVFTwJVq7u8l7EogV3KhUKvT19cHtdsfctZORyTakFH8BB2WkgTDrbiTiUdSEQiE4nU7MzMxgYGAACwsL0Gg0aG5uSnW6pwqKQDRvGo4jHjUk9gaWUIkaxfT/d/QPrX1QGmPHwqkkagAAjWfg1xgiPqzSKiM+li4UxWUw/d4HMz0NnkTjJUIIampqUFZWhsHBwbCd4zIdgx2nuLgYLS0tGB4ePmHoLnVlSmVlZdiYTerzljKS38rv7e3FU089hXe/+91Qq9X4yle+grNnz0bsoCIm2Vz6BACMuQD04nTUY3JqTdia2ZE7Ux6DUp38qYT2AwA0CIVC2NragtPpxM7ODgwGAywWCyorK2OqZeRdeOGz4AzDYGtrC+vr69jZ2YHRaITFYkFFRQU8Hg/sdnvC7Rs5hFrQuTK3oaEhGI3GtHR9igZN0+ju7sbFixejdoTKRH10fn4+38mhvr4ehYWFCY8t5m4OJ1leXl5GMBiUFXIypwYpxV/AQQzm9XqhVsfnpxcpUeP3+/nOioFAAGazGYWFhWhoaOCvQb6AvJFyE/HVNADAsPEragBAx7oPWliT5Nck2rUGw/Yh8/rcArCllXE9l2vNnSyEonC1rBW1lwbCPk7tOkHpdGBS2FxKCYUC5v/9x6Di/L2lg2RjsJKSEj7m6uzshPFYIi7TMdhxuI5QIyMjqK2t5TtCZUPC43DMxnWzSpdHzWlE8hFlXV0dnn32WWg0GrznPe9BQ0MDGhsbMzKXZBU1iV5YDm7ehQ8SYilqAECpU0BfooPnaoYWBolyXFEDAO6dXcxeXoDf70deXh6Ki4vR2NgY90X0Vuu4JCbBYBBOpxNOpxM+nw9msxlFRUVHPo9QKASTycS3Q0ykfSOHkDsv+fn56OjowMDAAJ9MEopkZbdNTU24cuUKbDYbenp6TnSvyJSRnV6vR19fH0ZGRuDxeFBdXZ3Q5yC0R81xCCGoqKiI2e1DRiabkFL8BRzEYOF2xCPBlT5xykqn04nNzU0oFApYLBY0NDRAp0tRFXELQFPiq2mAg0QNy94oLY9hJgwAFBhoWQ98JHkleN7yEAgXhxECpt0a34tVqEDo1G+hdsxlYC2lIM6rJx9kQtC3NMM1dNLoOB2YfudRKCulUX7FkUoMZjabeV8+rrRaLISIFXU6Hfr6+uBwOOD1elFdXZ0ViRrgIGazWq18F9t0tOc+rVUKkk3UcG/4Y489BpfLhdnZWWxvb2N1dRVFRUUxe7aLgU6nS1h+z+3oRLtJYBiGDyK2trZgKm4GpRA+g83EaNHNkVtnkhM1xyDKkz8VwjKor6/P6kAvm5NFPp8P6+vrcDqdYFkW+fn5qKmpgV6vD3vB5hZOnU6H3t5e2O12BAIBlJaWxn1OoSWyRqMRBQUFuHLlCjQaDYoFMg9MZZ4VFRXQarWw2Wzo7OyEwXBTmp3JjgMqlQq9vb2YnJzExMQEWltb455LOoIEGZnTghTjL+CmoiYe9vf34XQ64Xa7MTAwAJPJhPz8fFRVVcWlejulMX8SsFBQ6YoRCEIsgYKwcSlqgIPyJx+VZKKGDcF4Zezmf+s7wer1cT015bInbhxCwFp/G+S174R9XJOXg0w4QqlaOqC/72wGzhybVGIwzpfPbrfD7/ejrKxMwJndRKhYiesINTk5icnJSRBCsiaW4eY+NTWFjY0NVFRUZHpKWYlkEzUcKysr+MxnPoOJiQnk5eXB4/Hg/e9/Pz72sY+lXX6r0+mwtraW0HMiJWqCwSA2NjawsbEBt9uNnJwc/kbzykYIQRG6X8Zq0c2hLdRAaVQi6MpMHboUCaeoUdNUSkmabE6SCEUir59lWbhcLqyvr2NzcxMqlQoWiwVtbW1xS+E5OF+WkZER+P1+1NTUxLX4i1HLTNM0WlpasLCwAJ/Pl7BaJBypmv5aLBao1eoTO09imgnHo3rhyowWFxcxNDSErq6uuFQs6dqFOq07OjK3JlKKv4DoqmaWZeH1evm4KhQKIS8vD0qlEv39/fJvM0kUlHjtuMPBMASg2JjtuTn0jAsbnAwnQUzXp6DYu2EirDeBqaqL+7kpdXw6TmktUFYHrFw68RDl2zko7WLT539G9AaYn/yEJH8zQsTMx+M/MeJwIWNFLu6Zn5/H0tISCgriu5eTAtzc/+u//guTk5Po7e2VlccJItm0HOfU/+UvfxmdnZ0YGxvDr3/9a9hsNrz55pt48803AcTXeUcokil9Oiy99Xg8WFxcxPDwMEZGRuDz+VBZWYn+/n40NzejoKBAVHk+a47P24EQgtxa2VT4MFQ4RU1ITmQJQbTFLBQKYWNjA9PT0xgYGMDy8jKMRuORluaJJmk4FAoFenp64PF4TrTMjoRYpnNce8Nw7buTQQjTX6PRiL6+PszOzvIdocRMeMSreiGEoLq6GlVVVbDZbPB4PDGfI9dHy8jEjxDx14ULF9DU1IT6+no899xzJx7/+7//e3R0dKC7uxvveMc7MDk5GXNex2MwhmGwubmJ2dlZDA4O4tKlS1AoFGhtbUVfXx9qa2tB07QkbzizA3HbcYcjdKNFNxPnPrISQSjZ5LoC5lwZ4v/NdNwGJLBGCKWo4ce77V6Eszwge15oGxsEPVcscp/4KGiz8GVBQiREhIrBuPjP7/fD7/cLfi8pdKzEdYTS6XSYnp4W3C9V7E1jrVaLqqoqDA4OZrRrYDYieUWN3+9Hbe1BjaTL5YLRaEReXh7W19fTPhetVptQfTTDMNjf38fly5fhdruh1WqRn58fUwEgVkzBGs1gaRokjnaVxmojnONbYEO3tuKDg6hOKmoIG77VX9xjyoqasMTjNyMUXMvsubk5OBwOdHZ2xryhF6s7ADeXS5cuwW63o6urK+nkglDKF7VaDavVitHRUfh8Puh0urR2fYpGYWEhtFotHA5HzHpzhmGgVIrbQUO+GZQ5bSQbf4VCITz11FP4xS9+gfLyclitVpw9exatra38MY899hg++tGPAgBeeeUVfPrTn8aFCxeijqvVanH16lV84xvfQE1NDSwWC3Jzc5Gfn88nZYRC/jUDCjq9ahrgZucnBhTidWvUsy5sQ5PQedR7u9Bcnz/4T1UT2JzchJ6fqpHwifEKysDWtgKXJ048pi0rgW9mRtDzRUL3jndB23t7Ws6VDEIrVVpbW7G2tobh4WF0d3cL1gxAjE094KDzXXl5OYaHh9He3i5YGapY8+UIhUKwWCwwmUwYHR1NujFENE5rDCbZRA33hldUVGB//+CGWKM5uBC//e1vR01NzZHj0oFOp4tZH723t8dLb30+HxiGQVFREZqbmxO4CSEQpe0SRYHNsYBsXo95KK2kYKwyYPeynPkEIihqWCalrPlpvagkQ6J+M0JCCEFDQwOuXLmCoaEh9PT0RLypF6M99eGkCte+e2VlBYODg+jp6UlKMSTkPLmOUDMzM1haWhLMR+c4yfyWONUPZ7YXqQZaVtTIyMRPqvHXwMAA6uvr+STPI488gpdffvlIouZw1z2PxxNxLJZl4XA48G//9m/4l3/5F/ziF7/APffcg3e/+92ora0Vb30gt/YmCgELOgMhCsPcOCkhYECDRuyNRR3jwjadWDlIwaodhGURUqqButbYTziMSgMigrKUWO8BuzAFHFN3KJGcYihR6KJSmB79kGjjC/FbFTqhQAiBWq1GaWkpH3Nx17pUEEt9zDAMcnJy+I5QNTU1gsRkYpeHc6XtnEcQZzIsRKn/aUeypU/cF+ZP//RPcfbsgaEVd/PU39+PqqoqAMC///u/R5XXfupTn0J3dze6u7vR2NiI3NybWfOXXnoJDQ0NaGhowEsvvcT/fWhoCB0dHaivr8cf/dEf8aqHcEZ2nG/G/Pw8bDYbJiYmEAwGUVtbC6vVCrPZDIPBIBnzp3h9aoADU2GZA0iY9tyEAKH91FQ1t6qihuvCsb6+jsXFRVy8eBE0TaOtrQ29vb2orq6GwWBI6wW8oqIC1dXVsNlsUWWlYrRxPD5mWVkZGhoaMDQ0BLfbndSYQstum5ubodPpcOXKFQQCAcHG5kg2AOPqzTc3NyOWjWVLpwQZGSkQb/wV6fe6srJyJGlaXl6OlZWVE8d94xvfQF1dHT772c/i7/7u7048/p3vfAfd3d346le/ivr6ejz++OP47Gc/i2effRZ1dXWirg+3+q1DJtQ0wE1FDRBfi24A0LBeUIkonFkG2vmDTkqBxh4gzEZcNCgh/WkOQXItQOOZk393bUEp0gYJD62A+SN/BEqdepJCTMRSfpSWlqKpqQlDQ0OClOaINU9uY4/rCLW8vIz5+fmU7yXE3sw6XNquUqnQ19cHj8eDiYmJtFqYZCOSjVwdDgdGR0eh0+lgMpkQCAQwMTGBf/3Xf8U3v/lNjI6O8vLa1157DZOTkzh37tyJOuevfe1rcDgccDgcePrpp/H+978fALC5uYlnn30Wb731FgYGBvDss89ia2sLAPCxj30M//AP/4DZ2VnMzs7yclyuNeTu7i7eeust3jdjaWkJOp0OXV1dOHPmDKqqqnglAGcmLBXYODs/AYA6RwWNRfjuU9kIFSZRAwChQPI+Nbda6VM4vxm1Wo3KysqU/WaEorCwEC0tLRgeHg67WKfTdI5r3z0yMsJfm1IdM1VycnJQWFgIm82WVAIpFsnOmaZpdHZ2QqlUYnh4mFcBcKRDUSPvCsmcFuKJv4DUr4dPPfUULl26hL/+67/Gl7/85ROPv+c978HQ0BC+973v4dFHH4XFYom765NM8lAk/d40HJxHDRBfi27gIKmmY+K/uc7bvgzi3sa+pQyq8sQ70QjtT3P4d0T67gYUJxW9+npx22Qb3/t7UFXXizJ2MBjEtWvXMD09nbJthZglOnl5eejq6sLo6Cg2NjZSGktMRQ03LudtKETCIx2bWYc/N85k2GAwwGazibL5d1qQbKLm1Vdfxd/+7d8iEAhgdnYWL7/8Mr761a/i+9//Pu68807cddddR+S1KpWKl9dG4ty5c3j00UcBHChx7rnnHuTl5cFsNuOee+7BhQsXsLq6it3dXdx+++0ghOCJJ57AT37yEywuLuJHP/oRhoeHcffdd+OXv/wliouLYbVa0dbWhqKiorDlElJL1CSiqAFkVQ1HuPbcABDaz+6Li9g3l8FgEKurqxgbG8PQ0BA2NzdRVFQEq9WK1tZWGI1GySkdcnNz0d3djbGxsROLtVhmwpHGNBqN6O3txfT0NK5duxb3mGItuizLIicnh08gbW5uCn6OZOHKxjgJ8+EbOllRIyMTP/HEX0Dk9aOsrAxXrlzh/7+8vBy1De4jjzyCn/zkJyf+bjabj3hGcJtliZJMQunWzbumsx33SfjSJ8SvqAEOfGriJXdpCKCVYFp7Epobh5CJmuO/IaI3AW23nThOpRXPY03R0ALDu98r6Jh+vx/Ly8uw2+1885TS0lKsrKxgcXEx6XHF9lIxGAzo7e3F7Owsrl69mvQ4Ys3zuAKbS3jo9XoMDw8jGExu8zgTMRLXGIJTsqe6+XdaN8sk61HzB3/wB/jiF7+I22+/HV6vFxRF4fnnn8eDDz7IH7O8vHxCXvvWW2+FHW9xcRHz8/O4++67AUSW5q6srKC8vBwAsLW1hZ/97Gd49dVXsbS0hDvuuAOlpaV45ZVX4n4dFEXxHRSkwH5rP1S/eRVkLz7HcEOZHrRmEyG/dF5DJoiUqGH3s19RI/QcMuk3E4t4z6/X69Hb2wu73Y5AIICSkhIA6U/UAAfeEFarFQ6HI+723WIGCRRF8Qkkh8OBioqKqDdh6aakpARarRZ2ux2tra0wm82yR42MTALEE39Fu8ZYrVbMzs5ifn4eZWVlOH/+PH7wgx8cOWZ2dhYNDQfdbH7605/y/46GTqdLODnMxWBCmYSedigCZDKnfbj0iSGKuO0atYz7oIU1iT55VdANxfIM2PZ+0NokDIEJBSjFLQ8iPXeCnbIBgZtJSWrXCUqnAyOwooxRa2B44qMpe+6wLAu32w2n04mNjQ3QNI2CggK0tLTwni+BQABdXV2Ynp7G3t4eGhoaEo5TxE7UAAcxF9e+2+fzJeWFlQ5FDQchBDU1NdBqtbDZbOju7oY2we92JjezuMYQIyMjaGxszKr24+lAsitXZWUlvvWtb2FtbQ0vvvgi3nzzTZw/fx4XL15EZWUlenp6EvrhnD9/Hg899FBCwbpGo0FLSwvW19fx2muvweVy4bXXXkvoddA0LSlFDWspgf9/fRyac38DEscNOqEIcmqM2JzaTsPspAuJYC7Lyi26eZ+m9fV1bG5uQqVSwWKxxOxuxj0308mbSHDeJw6HA3t7e6iurhZlvvEskAqFAmfOnMHExASmpqbQ0tISdR5idTk6vJvDBTOjo6Pwer2or6+XzGeZm5uLM2fOwOFwoKqqSlbUyMgkQDzxF2cUHA6FQoEXXngB9913H0KhEJ588km0tbXhmWeeQV9fH86ePYsXXngBv/zlL6FUKmE2m4/4BEYiGUVNsqpmaVzJ0g0LBZ3ZzSOGJWDZGx6ACShqaDDQsl74iCHqcQWrDiDXAqY4uc0FotKK3+BArQW63gF28Jc3/8iEoG9phmtoWNBzue5+EAV5lqSeyzAMtre34XQ6sb29DZ1OB4vFgq6urrDxB2cH0dnZiZmZGYyNjaG9vT2htVnoGCzSRiXXvntychKTk5NoaWnJ6Dw5osUyxcXF0Gg0GB4eRltb2xFP1lTGTQdGoxFWqxV2ux1erxeVlZWSiSczjWQTNRyFhYX47Gc/CwCYmJjAiy++iFdffRWPPvoo2tvbsbS0xB8bTV57/vx5fOMb3+D/X1ZWhl/96ldHnnvXXXehrKwMy8vLAA7Mg2tqaniZXjgz4VhIrfQJAEJNZxB450NQv/7PcR2fU2vE5vS2KI2osoVIihqEkjcTzuaLUCgUwtbWFpxOJ3Z2dmA0GmGxWFBVVSXZnctklENcgmR8fBzT09PQaDSCv754F/RE2nenK0jggpmZmRmMjo6io6NDMgkRrVbLtxZ3u91prb+WkTkNRIu/amtrowb3DzzwAB544IEjf/vSl77E//vrX/96wvPRarVRjd7DIbXNMilDUweKmsxCwLAENGHj9qjh0DEu+KjoiRrdogOhzv6ka9uE9qeJSMdvARP/A3hvloNo8nIgZB9W7W+9E+v1iXW82t/fx+bmJtbX1+F2u5GbmwuLxYL6+vq41lhCCN+cYH5+no9l4o2rxIhtIo3HlRVdvnw54XmKWX4e7fUf3qRKpCNUphM1wIHJsNVqxfj4eFLJsdNKVrwDLMuCYRi0tR5kmjcAACAASURBVLXh+eefx+uvv47f/u3fhtVqxdzcHObn5xEIBHD+/Hm+Q8FhpqensbW1hbe97W383+677z78/Oc/x9bWFra2tvDzn/8c9913H0pKSmAymfA///M/YFkW3/nOd/De9x7UbioUioQXfCkmagAgeMd7sd/aH9exCq0ChtI0LU4SJVx7bgAgTGpdn7KJ434zW1tbR/xmCgsLJZukSQWKovgExMrKiuClYokEHpwPS1FREQYHB7G3F75tZ7plt83NzTCbzUmbwolVAsglkhiGwdTUlKTKUGVksoFI8ReAtAfROp0u4URN0oqajCcs0k1mvWkOw5U/JaKoAQA94wKirCXm7XmguALQJh/PEnUS5VLJnEepAjnzziN/o3w7MUu74oUuLEbOY0/Gdeze3h5WVlYwMjICu90Ol8uFiooK9Pf3o6mpCfn5+UldC2pqalBSUgKbzRYxlgmH0IqaaOMRQlBXV5fwPLnuTGIQ6/VrtdqEO0Jx7bPFIJH4jou3tVothoaGEvLcOa2bZVmRqOHkcgD4QLumpuaIvLalpQUPP/wwL6897CNz/vx5PPLII0c+xLy8PHzhC1+A1WqF1WrFM888g7y8PADAN7/5TXz4wx9GfX096urqcP/99yc992Q8atLyVSME/t/5vxAqis/xPucWNxWOpKih2JAkfGbEwufzYWlpCcPDwxgdHcXe3h5qampgtVrR0NAAs9mc0mKULe8dIQSNjY3QarVYXFxM2rAtHMnsEMVq350J2W1lZSVqampgs9ng8XgSGlfMEjhCCLRaLSwWCwYHB5MyI5WRuVWJFH9lgmQTNXKCNjYKKjPtuMMRumEozCSYqFEiACUibxTkbM6CqahOZWogIrXmDktzH2DKu3nuPS+0jbG9nGJC0zB/+I9BabRh116WZeHxeLCwsIChoSFMTEwgFAqhoaEBVqsVdXV1MJlMgqzZpaWlfCwTT8XCcTPdVIl3PG6e8ZreCj3PREm0I9Th9tlCk+jGISEEtbW1qKysxODgYMLx5Gkj67a/j2f8YslrAeCLX/xi2LGefPJJPPnkyYxyX18fxsfHU5voDSQtu1Vr4H/009D9v18A8UW/8OgKtVCZlAjs3pqeLJESNQQsmFAIdJYqSY4bGqfiN5PKHLKF3Nxc6HQ62Gw29PT08CZ5qZDszsvh9t2caS6HmLLbaOMWFBRArVbD4XCgpaWFT37HQmzZbSgUQkVFBfR6PYaGhtDR0QGT6dZOPsvIJEqmDbnl0iexyFw77nAwLAUghFCCpU/AQfnTDn0yVlHue0H//+y9eXgrd33v/54ZabRZkmV53/d9k2Un+REIpNwSkl5CgDQQ6K+0TaFLWmj7tKRPey8X6KUhTW8eShPCJaQt0BLgUhpCe0lZQgIEwvEi+3i3z7F9vBz7WPJua9fM/cPMHPlY24xmpJHOvJ4nz5Oj5auvZGm+n+/7+/l83jYrQKbxHSYpEFp54qBYEBQFDL4ZbFSbAkNVOXzz82mNa37bA6Abzwo+LMvi4OAAHo8Hu7u7/OFGT08PaJpO6/WSYbfb0d3dDZfLhe7ublitVllfLxohh0R2ux29vb2YmJhIGt8oof8iV7q1srKCsbGxuL2DAHljMLFmDmVlZTAYDBgfH0d7ezvsdrsMs1M+Cro05ydKLX3iYG2l8D/wh2BT+IHezFk1cXvUAAiLKPVQEgzDwOPxYG5uDhcuXMD6+jrMZjMcDgf6+vpQVVUlm0iTa7Asi8LCQrS3t8fNZhEzplji2XdnI6OGw2KxwOl0YmFhIWV7y0zURxMEAbvdjv7+fkxNTeHatWuSj6+ioiIfGS19EvyM3EVDKSebBrieURMRcZZsYg5j3l5EHgImU1rzkqM/TVL3z+ZewH69z0iijKFUoFs6UHDPOwCcbqCDwSAWFhZw4cIFbG5uorCwEE6nEz09PaioqJBdpOGwWCxwOByYnp6Gx+OJ+zipS4qExkpczDU/P4/NzU3JxpULzhGquroaIyMjcbOW5IzB0hnbYrFgcHAQly5dOtOT9mZCFWpkJhfSbiON3Qi+5b1JH2epLQCpyf6FJxsQCTJmIhKWwWQKrt/M5uYmVlZW8qrfDMuyCATlKaniFl+bzYa+vj5MTEwItouNRToLOmffHV2PLGePmlTmyjlCbW1t4dKlS0nFqEw2sjOZTBgaGsLq6mrK9dsqKirZR0xGjegY7CYJdQiwoBT2XrkeNUKbCQOAnvWCZM/2DoyEIoj40re1zmjZE/eaBAHill++/u+jPWhTbBB7biyjCab3/x42t7Zw8eJFvgdIeXk5brnlFnR0dKC4uDhrmXNGoxGDg4O4fPly3EMeqddrMbEHF3NtbGwkjCGUINRwlJeXo7OzEy6XC/v75118lZhRw8E5sO7t7WF2djau8K6kz1tKVKFGIBRFIRxOvYGsqLTbLHzXQre9FaG+NyR8DKklYa5L3FE/X0kk1DDh3MioidVvpqioCA0NDZL0mxGDVIuuP8BibCaAf/i3I/zJ3+ziv31mF1se6Rs9R5+SFBQU8Ccr0dks2YBzpzo+PuYXsmxl1ETPyeFwIBQKYXJyMuF1UM5GdrHQarVwOp3wer2YmppSdNajiorKKWIyasSWPuVnyH8epWXTAFzpk/BmwsDp383IRGW6Mgz+6tE5mOn047SMOT7d+Lq1bUBFPf9vU3OjqHH2bv9lzGxsIhQKoampCbfccguMRiOsVmtGNrmpxHs0TcPpdGJzczOmCCKHPbeY8biY6+TkBLOzszlx4GO1WjEwMIDZ2dlzMatSM2o4KIpCb28vaJrG2NiYpH0ilY4q1AhEaKCg9NInHoJA4G2/hUhl4gXgZi1/IjTxAwY2DYtuOWFZFoeHh7h8+TKGh4exsLAAiqLQ1dUFp9OJ+vp66HS6rKvQYl9/ezeC7/3Mi7/9p3384aMe/P1XDvHjUT/2jxh49hl88vP7uLx29mIuxXuNHoPLHFlbW8OVK1fSHjsdOPtumqbhdrtlCRzENIXr6OiA1WpN6AglZyO7eJAkic7OTpjNZtFuVSoqKpkjk6VPNwMEoazeNBxc6RMLEmJWMSN7amJNsCw+9fg8igtxLstGDJlyfIr52re8hf9/2iAi27lvCE33/SqcTifq6upgSrMMTE64Q57j42PMzc2d66OoBKEGuN4DRqfTweVyCTrEzxbRjlBLS0v8Zyu3UCPFQRznwFVdXY3h4eGUmk/nAwq8RCsbg8EgyDUkF0qfeLQ0/O/5IzAF8Rt56Sw0DCXpN1DNNRL1qFGSUBOJRPK230wkwmJuKYivvXiMv/i7XTzyxC6+8h8nmL4UQqz18djL4m/+YR8T86nbPiYj1qLOZWfs7+9jfn4+qycrnH23wWDA7OysIMvLVBC7mNfV1aG+vj6uI1QmS5+iIQgC9fX1vFtVOj2Hsi14qqjkO1qtVrCgKjYGuxl+zlqF2HHfCFf6BIIQlVVjZI5AsAw+/feLuOYO4vY+CWIeSguCit2INRMQ5bVAXTsAgDzcAWlMPbuHKilD+Qc+lFOxH3fwpNFoMDExwYutShJqgOviQVlZmWCbcaFI9d45Ryifz8c7QqVbnpQIqQ/iysvL0dXVBZfLJUnrAaWjCjUCMRqNglS8XHMcYK12+N/9YbAJfrA3ZVYNFV+oIZjsCjVcv5nJyUmMjo7mVL+ZZKLG0QmDn7r8+OxXD/GhR3fw2D8c4MWf+LDpTi3wDoaAz/zLIX40IuwUNh7xFkqSJNHb2wuWZZOW+WQCvV6Puro6yRoec6TTyK+0tBTd3d0YHx/H3t7emfvkFGpSEc5KSkp4B61EjQxVVFSyh5hNSnoxmDKFDCkgCRZZ0MZTgmEJcJdtRkRDYQoMxn66ipXV03W/qyH9N5qtsqczc7jll08VRCYCU0d7ak+KsuKOhZJLdgiCQEtLC4qKivh+OtztUiGV+FFVVSXIvlsMUpa0cxnFJpOJLyWSs0eN1GNbrVbetGJ9fR1A/h6WKXP3pmD0en1+lj5FwdS2IXDPb0D/7Wdj3l9QaYTGQCHsy5FMIQlIlFFDZkGo8Xq98Hg88Hg8YFkWdrsdDQ0NMJlMgi5WSriw3TiH1c0wJuYDmJgPYmk9jHTjCIYB/vH5Y+zs63Hfm9PrsZQoqCEIAu3t7bwVYn9/f9YEMpZlYbPZYLVaY9p3iyXdQIFzhHK5XKirq0NlZSU/brbro81mMwYHBzE+Pg6v14va2lpZ5qOiopI50onBCOSrVMNCQyn5nRFgWAIUwZ5adIuYakvl9b95dXH671UuoUZQvFZUhnBDN6ilSYR1qWU/mP/r/aCbWiWbQ7qIea3a2lrodDqMjIxIvp+S0kUq2r5bjkoKqeMkzhHKaDRienoaZrNZsrGjkasHIdfUeXJyEsfHx+jt7VXEnkZqVKFGICaTSbBQkzOlT1GEB38Joc0VaEd+cO4+giRgaTBjd+Z85/B8hUhwkSHAIMJEQJHyNUNlWRZHR0dwu93Y3d0FTdMoLi7m62PTHTubBEPA+NypMDMxH8TeoTzC5gsv+3F4Ajz0LhokKe5insrpC9f7Z3h4GAMDA1lJN+YWdJPJxAsjDQ0NKBfpFsEhRVDDLa4TExPw+XxobGxUjOMA5y4wNTWFk5MTtLW1ZaUkS0VFRRpyNQaTE4oERC6BGSPCEqDAghFR+gQALeXX+9MVaINpK25yZtQkisFYlsXx8TEf+xks1Wgnp2Ehw/ARJMDGj5fo5nYU/Mo75ZhyxikrKwNN0xgeHsbx8TEKCqQxNpHaeIGz7/7Rj36Era2ttGOuaOSKk8rKyrC5uYnLly/DZDKhsLBQ0vHlNIugKAp9fX24evWqYizRpUYVagQippmw0I2wUr5mgbt/HeT2OqjV+XP3WRvN2J3dz9fjpvMkyIwgAESCIVB6aS9EkUgEe3t78Hg8ODg4gNlsRnFxMerq6hRbypQqnr0ILi4E8TOXDiubQDhymJHXfXnYjyPvLv7gvUWgtcJ/aakuBBUVFfwJUF9fn2RBRapEz5MTRsbHx+Hz+VBfX5/1xYxzTJibm8Pk5CRKSkqynlHDwbkLLC0tweVyoa+vL6XfW7Y/UxWVmwGhv7N0Sp8IAmlndCoPFhqF9qaJhmEIgILojBojFUJnkxYUCRBs+kIdQWeukTDDMNjf34fH48He3h5MJhNKSkpQW1sLjUYDJrwPTP0MhtZm+OYXYs/XYEThb38IhIwHiJnGZrPBYDBImiUsx+Zer9fDZDJhfX1d0phLzgMtrVaLtrY2zM7OSnKoF43cZhEEQaC6ujpvD9Xy813JiMFgECTU5HTwrtHA/+4Pg7EUnb9Lr0FBtXK7xktNoowaAIiEpHGMyeV+M4lgGBYLKyH8n/88xn/7+1382f/axZe/fYxL6xqEI5n9jYxO+/HXn/fg2Cs8eBeyqBcVFfFpsDf2ZJGbGxd0JVpJco5QFosFly9flu11xDTJ4xoEVlVV3VTuAioqSocgCEHCSy6Wn8uJhlSeHXcsIr+w6BabUQMAd71Oh9t6JMho1eplFzzC4TCuXbuG6elpDA8Pw+12o7i4GENDQ+ju7kZZWRkf+xEDbwK0NAxVFXHHs77vt6EpLk36urmWhUBRFJxOJ+bm5rC9vZ32eHK8f27MgYEBHB8fSxZzSVmmdSMMw8R1hJJibLkyam4Gcm/Hl2XE2EPmMmyBFf4H/wSGZz8OInzW6riwyYLjtfMOLvkIkUQcYUKhhPcnQqp+M0rjxMfg4kIQF+eDmFwM4sSXfXGAY/FKEB//rBuPPGRHsS31y6DQRd1sNmNgYADj4+NoampCaWnywEkKYi3onJUklynS29urCNGvvr4eXq8X6+vrqKyshFGAm0UqpHMKVV5eDoPBAJfLJdkJnoqKini4w7JU7YXV0qdolGnHHQvO+SlCiN/g9dSz2D+RopGwPNk0gUAAfr8f09PTiEQisNvtqKmpgdlsThhnEAYT2J7boV0cj3m/4dbXw3jbHbLMWQno9XoMDg7C5XIhEAigpqZG9FhSlz4B1+NEzrnq0qVLfHZuOoKFHHONHpskSd4RanZ2FtPT0+js7ExbHIpEIoqINXMV9ZMTiMFguOlOV5nKBgTe/gHo//WzZ243FOtBW2kED6TJJlE0SS6uTCR1oYZlWRweHsLj8eDk5ASLi4uS9ZvJNuvXwr/oNRPA5bUwlHyQeXU7jI895cZHHipGbUVqtptiTl+4UwqXywW/35+RJrXxFnQuU2RjYwMjIyNwOByK+M4VFBRAo9HIIoike5pjtVp5sS26AbKKikrmMRgM8Pv9KQs1aZU+iXqWctFQuZFNA/yi9AniXJ847AY/jHT66xtBS3N4wLIsTk5O+IM5rjVCU1MTrFarsDn1vR6YuQBteTlCW1v87VRxKay/9kFJ5qtktFotnE4nLl68CL/fj+bmZlEihhxZKtFjcs5V6+vrfMxF07SocTNlusA5Qq2srGB0dBT9/f3QasVb08s5b45cPtROhirUCMRoNMLv92d7Ghkn3Hs7gpsroH/6f8/cXthkxvbYTnYmlUGSZdSwkcTOT/H6zRiNRvT19Uk51YwSCrGYXQ7yjYB39hWszMRg75DBXz3txh+/347OpuQBndg0WS6omJycRCAQ4IMKuUqQks2zqqoKer0+az10boRhGJhMJtTW1sLlcqG+vh4VFfHTuoUgRX20wWDA0NAQLl68iOPjY7S0tJz7fPM5UFBRUQpGoxFerxd2uz2lx6dV+pRHtk8EWFA5dImSIqOGAGCg0j9IJNNoJMyy7Jl+MwaDAcXFxejr64NWq8XExAT0er3gcQlaBzjugIn9OfY5oYYkUfjQh0AaUp+vEsqgxUJRFPr7+9PK/pCj9CnWQVl1dfWZmCtVofnGcTNluhDtCDU8PIz+/n7R2c5iys9VrqMKNQIxmUxwu93ZnkZWCP6X94C8tgrN5Sn+NnNtATyTe2BCubVBFwJJJ7/AEDGEmlAoxJ+c+Hw+2Gw2lJWVobW1lb/YXrlyRfL5ys3eYQTjc0FcXAhi5nIQQfFVX4rA62fxN8968LvvtuG2vsQLUTqLOtedfm5uDlNTU+jq6gIgzwY/lXlGW0lmu6yHC0CiHaG8Xi8aGxvT/nykChI0Gg0cDgcWFhYwPj6O3t5eNfhQUckwQg/L0rXnzhdyKZsGACJM+j1qgFOBKj0IQGAj4Ugkgt3dXbjdbhwdHcFqtaK4uBiNjY3n1oy01reuW6FdmuP/WfAr74KupV3wMLl8yMD1ultaWsL4+Ljg8iK5etTEElSKi4tB0zTGx8dFxVyZyqiJpqysDHq9Hi6XC11dXaIcoTKRUZPPqEKNQLi0WyFwze9y/otKUfDf/4cwfv6/g9w7beJFakhY6gqwfykzrj3ZgNAnT1Mk2FOhJla/mcbGRhiNRkUuhqlkdTAMi6X16yVNa1v5V+8fCgNPfmUP+0cM3vr6+Nkl6S7qBEGgvb0dKysr/MIn13UhlXlyVpJSZ7EIJfr6yAki0YJWOp+RlNdegiDQ1taG9fV1/pRJr9cr8retopKPCC0/pyjqpu9RQxC505uGg+EyatIUatKFoA0pXd+DwSAf+wUCARQVFaGqqgoWiyXp88VmtRCUBlSHA+RLr4CqqIb5bfeLGifXiVXSnWp5UaYyajgsFgtfSi3UXUnuZsLxxo4u/xbjCKVm1KSHKtQIREwzYe5EJ+eFGgAwFpw2F/7C/wARDAAArE2WvBZqSF3y2kySjWB4eBg0TedFvxmvn8HU4mk508WFII69uZsemyosC3z5hQPsHkTw4D3xg6t0F3UupVSn08HlcmV9kx9t3+33++NaSWaikR0HSZLo6OjAlStX0q6RliNIqK6uhtFoxOjoKHp6egT3F1BRURGH0BgsrYyaPNFftTlgx30jEZYAywJMGqVPUkAkKHvyer1wu93Y2Tkt/7fb7Whubpa8IX4iiJZ+GPoHYHrbg3llxS2Gqqoq0DTNizUGQ/JMKDnEj2T7Pa5vYXTMJcW46ZDKnLk40ev1oqGhIeV4UO1Rkx6qUCMQofbcgIhAQeHfN6asBv53/B4MX/s0AIA2a2Es1cO7nZ+9e6gUhBoCQE9XN/RGedwBMsGmO4zx+VOXpsUrIUTyt5otIf/xyjH2jyL44K/aoLmhqF/K05fKykqQJInJyUmcnJyIqlmWCs6+e2ZmBrOzs+jo6Dj3PuU8zYnVR4YgCNTX18NgMKRVIy1XkFBUVASHw4Hx8XE0Nzen5TyhoqKSGkJjMDl7geUCRm0IGgoIRnJtE0+AYQlEiOxuU6KFmmgjiN3dXeh0OhQXF6O7u1t0g9i050eSMP/a74MU+fr59tsoKSkBTdMYGxtDT08PLBZLwsfL6fqUCK5v4dTUFGZnZ9He3p70OXIelgHJxQ4uThTaE0jNqEmPPEjxyCxiMmqEpN4Gg0EEg8p3UYp0DiH4xnfw/7Y2J74Y5jIknaIjEJNb6dXhMItLawT+/cckHnliB3/xd3v4+osnmFu+eUUajlfHfPjbf9yBP3D2g5A6TdZms8FsNmN8fBz7+/uSjSsGzr6by/QJh8/2XZL7NCfeQl5WVoauri64XC5Rn1G6rk+JMBqNGBoawsHBQd4FvCoqSkRoDJZWqaroZyoFFnaTH3ajFzSV2PBAiTAsARan/2VtDhodPB4P5ubmcOHCBWxsbPDlK729vaisrMyaSMMhVqThyLdsBKvViv7+fkxNTfHZTvHIZI+aGyFJEj09PdBoNBgfH0+6T1RCZQbnCGUymTA6OopQKHmTSiXMO5dRPzmBpFP6FAvOru/KlSsYGxvD5OQkWCV7GkcRfNM7EW4bAACYKozQGPNTMSX1qQk1TAoXrGyzfxTBj0Z8+Pt/OcAf/vUOvvgfNH42SWF7Nze+c5lkciGA//k5Dw6Ori+eUm/GWZYFTdNwOByYmZnJeqNyrta7rKwMIyMjCAQC/H2ZLH26Ea5GenZ2FltRVqSpIIXrUyK0Wi3a2tryLthVUVEiYmIw8eS2+GrRBUFTDEgCsBt9OSfWRFgCIIis9amJsMDYxSns7e2hrKwMQ0ND6OzsRElJiZohoHBMJhOcTicWFxexubkZ93GZ7lFzI5x9d3FxMUZGRhIe1CtF8ODK92trazE8PJy0Z5iaUZMeaumTQKQofWIYBgcHB+fs+rj0yfWdEHzBHAgQSBL+d/4ejF/4HyDdV2FttGBnai/bs5IcUpfaz4SJiBNq5Fgoosde3jhtBHxxPogrm2GoB/+ps7wRwseecuOR3y5GebFG8vIfbkE3Go0YHByEy+VCIBBAdXW1ZK8hBs6+e3R0FL29vSgoKMhqfTRwvUaac4RKtUZaKcGNiopK+oiJwcSSy9orQbCwGa8L7ZxYs+M1IBjJjdA/wpz+ARiCAtjMi0wEbcDQUK8qwktIJj9LnU7H94IJBAIxe8FkM6MmmpqaGt6+O16Zt9JimVQdodQeNemhnL94jiC29CkQCGBrawvT09MYHh7G9vY2ioqK4HQ60dPTg4qKiqynT4pCb4TvPX8CVm+EpcEMgsy/H0sqzYQBACKFGqnxBRiMTAfw7DeP8EeP7eCvPrePF37oxcpVVaQRw/ZuBB9/yo3La0FZMmq4BYamaQwODsLtduPSpUtZL6Wx2+3o6enBxMQE9vb2si7UANcdoXw+H6anp1Pq/aWe5qio5A8mk0mQ65NQWJbF0dERlpeXU0rrVyqF+gA0NzQRzrXMGoY9XROylVGjMZozsgHM9lqfK5tcMZ8T11fl8PAQ8/Pz58aQI64Rm31cUlKC7u7uuGXecvYJFEsq2c5ylp/fDCjrL54DCLHn9nq9WF1d5TdePp8PtbW1uOWWW9DW1ga73Z4XX162uAL++/8AlF6DgurMdbvPFCSd4ulTRHjwI1Wjw2s7EXz3p148/g/7+NBf7+Cp5w7xkzE/Do9VZUYKDk8YfPJ/e7B8lZY0qLnxNIeiKPT39yMYDKYsRMgJZ989NzeH7e3trAs1gPAaaaWdQqmoqIhHSAwWTaJ1NhKJYGdnB/Pz87hw4QJWV1dhNBpFO81lG4pgUGgIxLwvl8QaLqMmW0INQcsfz+aKSCIF6ca6YrNfuF4wADA5OXkmrpKr9ElszMH1P5qZmcG1a9ckG1dOuGzn9fV1LC0tnfs7y1k1cDOQG/mPCsJoNMY9zWFZ9kxJE2fVXFJSgqKiIhQXF2d4tpkj0tKH4JvfDavnn3C0epLt6UhKqs2ECSZzgU84wmJhJcTbZ295cquRcS4SCLL49k/tKCn3485bpQngYy1gBEGgo6MDS0tLGB8fR19fnyBBV+rTOc6+e2RkBCzLKiKw4WqkjUZjUkeoTGTUqEGIikpmEJPVzB2IRP9Og8EgPB4PdnZ24PP5YLPZUFpaipaWFv5aNLUr6dQzhs0YQKLk5lwpg4qwXOmTJivtghJZc6sIR4p1UuwYBEGgra0NKysrGBsbQ39/PzQajWylT+mMyQkfLpcLfr8fdXV1AOTrEyhFzCjWEUolOXnxKb744otoa2tDc3MzPvWpT8V8zNe//nV0dnaiq6sL733vewEAP/zhD9Hf38//p9fr8fzzzwMAfuM3fgMNDQ38fePj4wBONy3RpznhcBjb29uYmZnBhQsXsLm5yaeC9fX1oaqqCjRNp+z6lMuEXv9foXnjndAV5mAJVwKIFIUako0IvuAJyag5PGbwkzE/nnruAB/66x08/o8H+O5PfapIk0EYlsAX/vUQz//gSJrx4ggUNzb1FeIEJ0fgodFo0NrainA4jNnZWcnFILGpsak4Qqlptyoq+YPRaBScUUOSJMLhMI6Pj7GysoLR0VFMTU0hFAqhsbERQ0NDaGlpgc1mO3s9zkH9VUtFYNElXy9yIbMmq6VPlAaEJjczqvIVKYSK+vp6VFVV8WYJSjh4ioVWq8Xg4CD29/cxNzcHlmVly6iRalwu27mgoCBl3P83rwAAIABJREFURyipyOfDMuVK6SkSiUTw8MMP43vf+x6qq6sxNDSEe++9F52dnfxjFhcX8eijj+LVV1+FzWbD9vY2AODOO+/kBZjd3V00NzfjLW95C/+8xx9/HPfff/+Z1yNJElqtFo899hgmJyfx4Q9/GHa7HdXV1TCbY9ezUhSV9RKGjEAQCLz9A7D+eAzbP5jL9mwkI9UeNQRYRMIRaLSp/6ySXVxWroZwcT6IifkgljfUHjNK4f/85yH2DiN4/9utINPoy5QsSOCE3kQN5mKNKddJht1u5+27e3t7odFIs4Sk48zECePj4+NoaGhAeXm5ZGOrqKgoC4PBkHKPGoZhsL+/D7/fD5fLBaPRiOLiYvT09KTUEzAXQ3+70Z9yE2SlZ9acaSacYdRsmvyF6wk6OjoKm82WUlwlBKnEH5Ik0dvbi8XFRUxMTECv1ytaqAFO9zT19fUwGAx8trNKeijvyiyQCxcuoLm5GY2NjQCA97znPfjWt751Rqh55pln8PDDD8NmswEASktLz43zjW98A3fffXfcTtsjIyP49re/jf/8z//E+vo6CILARz/6UbS2tiadYyJ77ryD1oH4i0+A/MmDYAL5kelBalMPEiKhgCChBjibdhgIspi+HMTE3GlJ0/7RTfK9yUG+/7MTHBxF8PsPFoHWiluUU1nQS0pKQNM0XC4Xuru7YbVaEz5ervRYLjOlqakJGxsbGBkZgcPhgE6nk2T8dFOFBwcHMTExAZ/Ph/r6en48pdZ1q6ioCCdZ6VMoFMLOzg48Hg9OTk5QWFgInU6H9vZ2mEymDM408+g1YZhoYRkynFiz6zUgoDCxhit9ykZGTSb606gIQ8rsF7vdju7uboyMjECv10syJofUwkdraytWV1dx+fJlfh8rJXKUh0c7Qt0MFSVykvPR68bGBmpqavh/V1dXY2Nj48xjFhYWsLCwgNtvvx233XYbXnzxxXPjfPWrX8WDDz545ra//Mu/RG9vL2pra/GZz3wG/f39+MEPfoDy8nJ85CMfSUmkAU6FGiFf1Fw3TiLLq6G96y3JH5gjpOz6BCAiMNWPIAi4d8P4/s98+F9f3Mcf/LUHf/8vh/jRqF8VaXKA4Sk/PvWMBydecX+rVAMPq9UKh8OB6elpeDyepGPKfepSVVWFlpYWjI6O4vj4WPLXEoNWq8XAwAC8Xu+ZRsxqjxoVleyQrCz9iSeeQGdnJ3p7e/HmN78ZV65cSTpmLHtuzrhhbGyMF2vr6up44wa9Xp91Z51MYDcKb7IMnMacRUYfdIorgyLAMGpGTb6QrWbC8bBYLCgrK8Pa2hrcbrdk48pRTlVbWwuLxYKFhQXJXe/kOszi4la/34/NzU3Jx79ZUJZ8LhPhcBiLi4t4+eWXsb6+jjvuuAOTk5O85/vm5iYmJydx11138c959NFHUV5ejmAwiA9+8INoamrCu971LlGvz9lzp0qxRQOfJwQmh+MK/QcfQuDbLyIfanVS7VEDAEw4uVATibBYXD0taXptogx7R9L0O1HJDvMrQXz8aTceecgOe6HwbKpUF3Sj0YjBwUG4XC4EAgFUVVXFfJycGTXRi3m0fXdnZ6csJz1C4WqkuYaBfX19akaNikoWSKUs3eFwYGRkBEajEU8//TQ+8pGP4Gtf+1rCcbmMmu9973tYWlrCwMAAb9zQ1dUVM8NPbFZzLumvJjoEvVb8yTUn1igtsybCkojkuVBzM4iIUiBn/72lpSUEAgFUV1enPaZcMYder0d5eXnK2dWpImeMpNPpYDabcfXqVXi9XjQ2NsoSn+bzYVnOR69VVVVYW1vj/72+vn5uA1NdXY17770XWq0WDQ0NaG1txeLiIn//17/+dbzjHe84Y8VYUVEBgiCg0+nwm7/5m7hw4cKZMYVcWIUGCbSGQLnADZ/SMDZUInLL7dmehiSkbM8NgI1j0X3sZfDTcT+e/tohPvToDh579gDf+YkPe0dqs7p8YONaGB97yoP1LWEZVUIXSJqm4XQ6sbW1FdMGUcyYqRIrUyfavlspJyacI1R1dTVGRkYQCoXUZsIqKhkmuiydpmm+LD2aO++8ky83v+2227C+vh53vKOjI/zrv/4rHnnkEbz22mv40pe+hLKysjPGDfHKMMX2Ccyd0J9Fkchsmmi4MiglZdZEWAJMps+UNToQZGbWjHzeYAKnccP+/j4WFxcxOTmJozQOJuVyaOIa925vb+Py5cuKy/zhYBgGZrOZz67m+q1KMa5cQg1XMu9wOBAIBDA1NXXztAKRiJwXaoaGhrC4uIjl5WUEg0F89atfxb333nvmMffddx9efvllAIDH48HCwgLf0wYAnnvuuXNlT9ymg2VZPP/88+ju7ubvo2laUDdroaVPAGDSk7AX5PbmwviguAwkpZGqPTcAIHL9e7G2Fca/v3yCT35+Dx96dAfPfOMIFyYD8PrV05N8ZPcggo8/7cbsUurZc2IWdI1GA4fDAa/XG9OBSc4gIda4nH33xsYGlpeXFXM6WF5ejo6ODni9XhweHmZ7OioqNxWplKVH8+yzz+Luu+8+d/vBwQHe+ta34s1vfjNcLhd+93d/Fy0tLfjyl7+M++67LyURVkwMlktYdEHQlDSbH0JhYg3DEJnPqKGl7Vlys8EwDHZ2djA3N8e74dpsNjQ2NmJqagp7e3uixpXToYmiKPT398Pn86XtbCm3O5PRaMTQ0BBWVlawuroq2bhywJWekySJjo4OmM3mjDtC5Tq5nbaB003Lk08+ibvuuguRSAS/9Vu/ha6uLnz0ox/F4OAg7r33Xtx111347ne/i87OTlAUhccffxx2ux0AsLKygrW1NbzxjW88M+773vc+uN1usCyL/v5+fO5zn+PvMxqN8Hq9KbkGAOJPc4rMFAJhBsc5urEv/KUhXC2vBrUV/5QsFyAENAdmwmF86YUjTMwHsXugqsY3G14fi8e+4MHvP1iEW3oMSR8vNvAgSRJdXV24fPkyxsfH0dvby29YsmHhqNFoMDAwgJmZGczOzqKjo0MRJ4WFhYUwGo2YmZlBY2MjysrKZHkdJbxXFZVc5Z//+Z8xMjKCV1555dx9FosFn//851FbWwvgtJQ9UTPhWIg2dMiBnzUBFjZj6ocDKY0Z5QaV7TKoCEuCyXAzYcp63nBEJTHhcBg7Oztwu93wer2w2WwoLy9HW1sbvz4GAgE4HA5MTEygubkZJSUlgl5DrowabsxEcVW25wmcjcG4LKCLFy/C5/OhtbVV9GvKnVHDjc05QhmNRt4RSmrHrXwk54UaALjnnntwzz33nLntE5/4BP//BEHgiSeewBNPPHHuufX19TFPeV566aW4r2cwGOD3p55mmo7rU5lVg2A4jGA498QakiKhuf9dYJ/8u2xPJS0IAaVPFBvBDy+kn4KskruEwsBn/nkXv/52K97yuoKEj01nQScIAs3NzVhbW+MdmGialjVISBS0cEHO0tKSIPtuuTNwSJLkHaG8Xu8ZRygVFRV5SKUsHQC+//3v45Of/CReeeWVmKVLBEHwIg1wKgoLjafyufSp0BCAhpT+GqoUsSbCEmAJEgwIkJA/DiZMNpBqI+GUCAaD8Hg8cLvdCAaDsNvtqKurQ0FBwbk1lvv9GY1GOJ1OuFwuhEIhVFZWpvx6cgs1wNm4anR0FA6H40xbjFTI1GEZSZLo6+vDwsICJiYm0NvbK+p1M5FRE01paSl0Oh1cLpdk/Q3zOabL+dKnbMBl1KRKOkINSRKotGly1gmq6IG7wdLS2PdmCyKFzSYHRTAw6nJPVFORFpYFvvj8Ab72nYOEj5NigaypqUFDQwNGRkbg8/ky4voUD4Ig0NTUhLKyMoyMjKTURD0TzX45R6iTkxPMzMyoNdIqKjKTSlm6y+XC7/zO7+CFF15Aaal8WQziYzBlr+UUwaDQIG02TTRKKIOKMKfBb0ayakgKVHFN8sfdxPh8Pt5hbWpqCuFwGC0tLRgaGkJjYyPMZnPcTTN3O9drb3NzEysrKym/diaEGo6amhrU19djZGRE0MF8ojHTJVZsRxAE2traUFRUxPfkE0qmMmqisVqtGBgYUFR/Q6WiCjUi4FwHUiXd+mithkC5LTeTn2i7Beyb70r+QAUjqEcNgNoyZQd3KpnjhR8e43Nf20MkEvs7IdWCXlpais7OToyNjeHo6Cgjrk+J4Oy7R0ZGktp3Z8qVicv4MRgMGBsbQzisjB4MKir5SHRZekdHBx544AG+LP2FF14AAPzZn/0Zjo+P8au/+qvo7+8/J+TEQ+j1TWwMpvTzMZsxIPshHi/WaLJzvWTY0zcYIeSPgalSNdvyRliWxfHxMZaXlzEyMoL5+XlQFIWuri4MDAygtrY2bvkKwzCIRCIIBoP8essJCVyvvYODAywsLKSUWZtJoQY4java29sxOjoqqAlyNsrPa2trUV9fj+HhYcH23cmypdMhVkYNh8FgwNDQEK5evZpWE+d8/83m5u4/yxgMBkFCjdi022hMOhLFZgqeo9xriGd+3ztx8p0Xsj0N0QjpUQMAVcUM5lZVDVTllB+PenFwHMGH//8i6Omz3wspA4/CwkL09/djdHQUZrNZkjGjEZqpY7fb0dvbi4mJCXR0dKCoqCjm4+QUam5c+AmCQGNj45kaaYMheS+hZOR7oKCiIoZkZenf//73RY0rNKCnKArBYFD4Cyn4Z60lI7DoRLwnERAEYDf4sOMzIBDO7LYhwpyuDXJn1BBGK0h94lLlmwWWZXFwcAC32429vT0YjUYUFxejv78/aTkzwzBgWfbMnoeiKH6zzgk2Wq0WJEmit7cXc3NzmJ6eRmdnZ8JYQM5mwvGw2Wzo6+vDxMREymU6mTZ04IguKRJi3x2JRDKeUcPBCXZzc3OYmppCV1dXRg7ucgn10xCBmIwaKVLtbQUUzPrc+5NZHK2ItPdkexqiESrUlBeqZRUqZ7k4H8AnP+fBwfFZoVXqBd1kMqGlpQX7+/u4evWqZOMCyYOEWHD23fPz83HTW7ORdss5Qo2NjeHgIHF5moqKSm4jNgZTsE6DIqMfmdSHObEm05k1fEaNnEINQYIsrk3+OJlQglNiPKcmp9OJ7u5ulJeXxxVpuMyZUCiESCTCr7tarRY0TYOmaV6s0Wg0fJYNF1O0t7fDYDBgYmIiYeZbpjNqOAoKCuB0OjE3N4dr164lHTMbGTUcVquVt+92u92SjSuWVLJ1VEeoxOTerl8BZLr0KZrSQgo6jZLDh9jo3n1/tqcgGqFCjd2iCjUq51laD+HjT7lxbed6oCtH4EFRFCorK3H16lVJ7bLFLubJ7LuzFSQUFhbyAU0qwZeKiooy0Gq1goL5fLPn1mvCKNBlvhSJE2v0GRRrWBBgWMhm0c2yLKiS+qyd4mczGzMcDiMUCmFqagojIyPY3d1FWVkZbrnlFnR0dKC4uDju+hld1sSJMxRFgaZp6PV6Xpy58XPlrLAZhkE4HObFmqamJhQXF2e8LDnVGEyv12NwcBCrq6tJLbHl6hMIpPZ9MRqNGBwcxNLS0pmG7vGQu5lwKmNzjlB1dXWCy7fyPaNZFWpEILT0ScofAEkQqCjSgMqxv1zR298Ixpp+Z+9sQAoUaqxGVahRic21nQg+9pQby+unKetyLJAsy4KiKAwMDOD4+Bhzc3OSiDXpzJWz7z45OcHs7OyZ0+1sBglGoxFDQ0NYXV3FysqKIk42VVRUEpON8nMlUWTMnrMkQQBFGRZrIox8Ft2HQQakUfpSYaUSDAZx9epVTExMwOVygWEY1NXVYWhoCC0tLbDZbHE3vpy4wmXOsCwLjUbDizNcOVMyOLEmEonwYg1w2sC3pqYGw8PDMY0I5BBAhByWabVaOJ1O7O7uYnFxMW68ICb7WGpomsbg4CB2dnaS9gDKtOtTIkpLS9HV1QWXy4W9vT1Z5pRr5Nh2XxkIdX2SGi1FoLwwt9oLUToaxNvuy/Y0RCHE9QlQhRqVxBweM/ifn/Pg4rxf1pprkiTR3d0NiqKSphQLGVcsXDNfnU6H8fFx/tRM6EIuhFTmzAVfR0dH50SkVMl2UKaicjNhNBoFObGILn1S4M/apA3BoM1udtB1sSYzJQoMS8jTTJggsXyYP5lW8Yh2apqcnDzj1KTT6RI6NTEMg1AoxIszBEHwmTM6nS5lceZGSJKERqPhx+d+n+Xl5WhtbcXo6Oi5fZYcBylCxR/OEjscDmNqairmdUWuHjVCoSgKfX19YBgGFy9ejHsNzEb5eSJUR6izqEKNCAwGg2C7Nqkx6kiUWDJgVyghhe99G9gcbBIltPRJS0bg6LUgB9+qSobwB1n87T/tYGyOkLXmmiAItLa2oqioCGNjY2nV/kqxmHMpzuXl5bx9txJOczhRS6/Xw+VyqY5QKioKxmAwCDosy5/SJxZFpuzGnhynYo0/I2JNhCEkz6hhAVDFtXmZRZnIqcnpdCZ0agJiizNc5gxN06LFmRvhMmtYluV71gCnRgRcVkW025JcB1tCx+T66hQUFMSMFzLlZJkK3FwLCwvj9n+R0/VJ7NhSOULlA8r4JuUYJpMpqxk1HIUmChZD7vwJjXXlYG67I9vTEIzQjBoNEcHVHaCnywqBzt4qNxGRCPDNH2rx0rC0C1CsIKG2tha1tbUYGRkRLTJLGSRVVlby9t0nJyeKOM3hHKEqKysxPDyccmmFEk7OVFRuJjJV+qS0X7ZFFwRNKSdjN1NiTYQlJO9Rw+pMIE2pueLkApxT06VLlzA8PIyVlRUYDAb09/ejv78fVVVV0Ol0MZ8b3W+GE2eimwHrdDpoNBpZ1mlOrCEI4oxYY7Va0d/fj4sXL2J3d5d/j9loJhwLgiDQ0NCAiooKjI6OnnGVU0pGTTR1dXWora2NGdvI6fqUzticI1QgEIibvQTkfwyWO7t8BSE07VZOSqwU9Nrc+ZIaH3xntqcgGKEZNQSAssIw1j0EmlsKYTKqPzOV+Pzfn7D44rf2wTDSCDbxgoSysjK0t7djdHT0zClVqkh9SsTZdy8tLcWsR5cCMWVVFRUVaG9vVx2hVFQUislkyozzpoJCKwIsbEZ5rpPpkAmxhmGl7VHDgABhb5BsvGxxo1PT1atXUVhYKMipKbqpL0VR55yaMpEZwpVB3SjWmEwm3jVye3tbUUINR2VlJZqamjAyMsIf4CspoyaasrIydHZ2YmxsDIeHh/zt2XZ9SgRJkujs7LypHaGU903KAYSe5nDIkbpFEgQqbLnTXLjwTU5EquqyPQ1BEBrhF5mKwtNUyGt7QGW1BfbC3CpTU8ks3331BE9+ZQ+hsLxNf202G/r6+s6cUkkxrljMZjMaGxuxu7srSy2y2DnbbDbeEWp7e1vyeamoqIhHjKGDmNInBek0KDQEoCGVmf7PiTUmWp5NVISRLqOGBbBNVYGrTVfCabyQvUE4HMb29jamp6cxMjKCnZ0dlJWVYWhoSJRTU2lpKRYWFjIuzsQiWqzhfq+c29Ly8jLcbrfihBoAKC4u5ku1Dg8PZXV9ShfO7XJqaoq371aC61MyxDpC5QPK/CYpHKH23EAaJzopoKFOxZrsLzfJIUkCmvvfle1pCENg6RMAlJivByy7R4DBYkCRNR9q5FXk4ucXfXjsCx54feldJ5IFHgUFBfwp1dbWVsrjyrWYUxSFmpqauPbd6ZDOaQ7nCHXlyhWsrKzEfZwSAn0VlZsJoTFYrrs+UQSDQoPysmmi2dryob86gJ0t6bMQI6x0PWr8MOCEKpRkLClIZf240anp+PgYtbW1GBoaQmtrK2w2W9y1OZlTU3NzM2w2mySGA1IQS6zRarUYHBzE3t6e5FmuUmXpWK1WXgDx+/2yCEpSEW3fvb6+rrhmwvEoLS1Fd3f3TecIpQo1IhCTUSOnUAMANMXCQAnP8skGRQ+8FazOkO1ppIzQHjUAYDOdbS7mC2lgKLSjsTZ2jbCKCgDMLgXxiafd2D0QHzClsjDq9XoMDQ1hbW0toQghdFwxMAwDrVYb1747HdI9zYl2hJqZmcnpzZ6KSr4gJqMml3vU2IwBkEqZTAzcbj+cbXroaRLveL1GcrGGYUlEJBBqWBDY1JxmdHP7aKU2KfX5fFhbW+OdmkKhEO/U1NjYmNSpKRwOnxE7Ejk11dbWoqysDC6XSxGlJRqNBhqNhu+ZA5zOv7GxEcFgEPPz85L+3aQSVYxGI5xOJ7xer+SZuFKXfXH23W63GwcHB7IdOEnt6mmxWOB0Os84QuX7YZkq1IhAKRk1wWAQm5ubmJycPLWyO9iGnsr+RTYZdGEB2F++O9vTSBkxpU9W/XnXGH8QOGEM6GjJHZFKJfOsbYXxsafcWL8m7rec6oKu0WjgdDpxcHCQUuAjl1DDiSnR9t1SOS9JESRwjlBSzktFRUU8RqNRUPq76EBeAfG/lozAogsmf2CW2N0NorOOhqXg9EBLqyEkF2siDAEQJJg0tiwsADdVAZaUweZbAmI5NXFrotPpRF1dnWCnpuhmwMmcmiorK1FbWwuXy3WmMW624PrWcNlAHGVlZWBZNmFz2Wyi0+lQUFCAra0tLC8vSzauHPEXRVHo7+8HwzBYXFyU5fOUY97cQePN4gilCjUiENNMmKIoSdIKvV4vVldXeZU9EAigsbERQ0NDaGlpQXWJMSeaC5vfd1+2p5AyYjJqTLrYm7lwBHCf6NDbaUp3Wip5zM5+BJ/4rBvzK8LT3YUsjCRJore3FwBw8eLFhAu1XE4G0fPl7LsrKip4+26pxk4Hbl6cI5RSmsmrqNyMZM7QIfsbgCKjH0o9MD48DKG+lIS98Ky9pVZD4B1vkE6sYUGAYZFWVk0AehxRRfy/lfCRcrbUq6urGB4exvLyMgwGA/r6+tJ2aqJpWrBTU2lpKZqamjA2NqaINS5arOGaDJMkifb2dphMJoyPjyuiXOtGWJaFw+HA0dER5ubmJBES5DooIwgCJpMJFotFlma9UmfUcEQ7Qq2vr0s+vpJQhRoRGAwGwc2MxGbUcJZ7ly9fxvDwMBYXF6HRaHiVvb6+HiaTid9AEb9oLqxR+F/W0tuMcGdftqeRGiKEGgMVf/FgWWBjTwtHjzmdWankOSc+Fo9+3oPhKWHZe0IFFYIg0NbWBqvVitHR0YQZI3ILNRzR9t3Hx8eix5Y6SOAcoUZHR3nXhHxPu1VRURpCM2rEku1ftk4TRkGcQ59sc+INo6SARWVpbCFBS0kr1jAMAUZkQ2EWwKamPuZ9mb5+3+jU5PP5YDab4XQ60dPTg/Lycmi12rjPjUQiCIVCsjk12e12tLe3w+Vy4eTkRPQ4UsGJNVw5Fyd6NDY2orS0VJFOQCzLgqIo9PT0gCTJpIdgqSB3w9+6ujrU1NRgZGRElFlOPOScN+cIVV1dLcv4SkHh23llInfpUyQSgcfj4S/kGxsbsFgsGBgYQF9fHyorK+Oq7EDuNBcu+423ZXsKKUGI2OhpyQiSncateii0t2gh0zVMJQ8IhYG/+/Iuvv+z1MUKsQtjfX09ampqMp4xEm++nH33xMSEYIeqZGOnA+cINTU1pTpCqahkgcxl1GQXu1GZ7zEQiMBERtBQnbiM+7pYs5/2a0bS6FPjIcvAZLHkKZFTk9Vqhc1mS+rUxGXNcGsaJ8zI4dRUWFiI7u5uXLx4EUdHR5KNKxaWZbGzs4PNzc0zjcGrq6tRV1eHkZERxV0PCIIAQRBobW1FYWFh0kOwZGSi4W95eTna29vP2Xeng1wZNdEo1WFLKpRZrKlwxDQTTlb6FAwGsbOzA4/HA5/PB5vNhvLycrS1tYlS/PU0iVIrhWtpNCWVG+0v/xdoSz6LkFvcJixjiMioIcDCbo5g5yjxc93HJrS3hXD58gkC2S8LVlEgLAv8478dYPeAwQNvtaTwePHWkOXl5aBpGqOjo+jr60NBQYGocYSQKADhThpdLhfq6+tRUVEheGw5ggTONWF8fByhUAiNjY2Sv4aKikpsxMRgYshmspxRG4JBq7z4LRRigEAQ7e2prQ2nYo0W//bjfdjLxbstRdhfZNQIrCIJgMahpuTc7XL/bYPBIDweD9xuN4LBIOx2O2pra1FQUHAupr+xNIZhGDAMc+Z2iqIyap1tNpvR19eHixcvor29HYWFmXXKCoVC8Hg82N7ehs/nQ3FxMZqbm2E0GhEKhfieO2VlZdBqtRgbG0N/f3/CPj7Zoq6uDjqdDsPDw3A4HNDr9YLHkKv0/MaxbTYb+vv7MTExgdbWVhQXF6c1tpwC082CKtSIgKZpwc22YmXUeL1eeDweeDwesCwLu92OxsZGGI1GSX6QFiOFQIjFvld5DbcAgKBpFL777XA/+Y/ZnkpCxGTUAEBFYSipUAMA1/a1aGwqxNrqIY5PlPm3Usk+33rpCPtHETz0zkJQVPzrA8MwaV0/ioqK+EyWzs5O2Gw20WOlQrKFnGscNz4+Dp/Ph4aGhpTfX7quT4ngXBPS7aOjoqIiDDFZzYC8mx1pYRWZTROJsPAf+DHUK6xsWwqxhmEIRChhWxYWwNU4JU/8YyRsROrz+XhxhmVZFBcXo6WlJSXx4EZxhiCIjIszN2I0GvlNe3NzM+x2u6yvFwwGsb29je3tbYTDYRQXF6OpqenMgVF0hhEn1hQVFfG2zT09PbBYkh9oZZp0D8HkFjyir4smkwlOpxPj4+Pw+/1plRalc3Cocooq1IiAS2kTAkmSiEQiODg4gMfjwe7uLmiaRklJCe90IgfFFgqBMAtfMPtN8WKheee9wNNfAhTYEIxHpFBTagkBSM3haXsfKK+0YH/nCJ5dBX8WKlnllWEvDo4i+NCvFUFHx178pNiMRGeyNDY2oqysLK3xEpFKaqxGo8HAwABmZmYwMzODjo6OlBZ/uYMbkiRhNqu9plRUMkk65edCMuyyJemYdSHQGmUd2jAMiz23F68fELcJ1lIE3vUGLb75430UiRBrIiwJRmDp0x5ZCoakEz6fqFICAAAgAElEQVSGIAjRaybLsjg5OYHb7cbOzg40Go2gmJ4TZqKbAZMkmVVx5kb0ej0cDgfGx8cRDocljwV8Ph/cbjdfRlxSUoKOjg4YDLFjZ+5zYRgGwWAQNE2DJElYLBZ+nu3t7SgqKor5/GwSfQjW1dUlKEsp05kpOp0Og4ODmJiYgN/vR1NTU46I3PmHMq4EOUiqKjzXb2ZnZweLi4uC+82kC99cWN4SQfGUlcH6ljdkexbx0ZAgRF4c7QXC6lH3jgFToRlVFYkDC5Wbm/G5AD75vz04Ookt6Em1oOv1egwODmJ1dRWrq6tpjxePVOfLWZXq9fqUbbIzUR+toqKSWcSUPnGHZYLIwr6EAIsiBWbTbF89ES3ScFAUgXe+QYtdET1rIgyBiIBmwiFosacpjXu/2D8tZ/Bx6dIl0U5NoVAIwWAQkUgEJSUlWFpaAnCapZnMRjsb0DQNh8OBtbU1bGxspD3eyckJlpeXceHCBczOzoIkSfT09GBwcBB1dXVxRRoOTswiCIJ3gwJOBVyn04n5+Xlcu3Yt7XnKgdlsxsDAAGZnZwX1uMtGCRFFUXA4HAgGg4q1Qwfy39BBWVeDPCEYDGJzcxOTk5MYHR3F/v4+LBYL6uvr0dnZiZKSkoxuHiiSQKVNo1iLR+N73pntKcSF0okXTQoNwhuHHfsAVmtEU7184p1K7nN5LYSPPeXG9u7575iU6f1arRZOpxO7u7vw+/2SpolzCAlAhNp3q/XRKir5h5iMmugmpKmSjZCp0BCAhlRWBvTG6jHedItVkrHEijURlkg5o+a05Kkh8YN+8cflMmoSwTk1zc/P48KFC7h69SqsVqtgp6ZYzYArKirQ3NzMZy4oFa1WC4fDge3tbVy5ckXQc1mWxeHhIS5duoTXXnsNi4uL0Ol0cDgcGBgYQHV1teADa4IgoNFoeLGGE2G5TJArV64o1rbZYDDA6XRiZWUl5TlmK5YhCAIdHR0oKCjA2NhYWg2RVcSRNxHsiy++iLa2NjQ3N+NTn/pUzMd8/etfR2dnJ7q6uvDe976Xv52iKPT396O/vx/33nsvf/vy8jJuvfVWNDc3493vfnfCvjRerxerq6sYGxvD5OQkAoEAGhsbMTQ0hObmZhQUFMiyyUkVnZZEmVWhJ8uDg9C31Gd7FjEhdeKrAwt04kqYAiHgKGRAZ6vymqKpKIctTwQff8qNlY2z1yWpF3SSJNHb2wsAmJyclPxURcx8U7XvVjNqVFTyD7EZNUo9EeYgCQaFBmX1vFpbPcYv3Sptzw9OrLm25kn5OQybekbNPlmMcJKSp2TEcmoqLS3F0NAQOjo6Eh64xhNn4tlo2+12tLa28n3YlApFUejr68PBwQEuX76ccE/Dsiz29/cxPz+P1157DVeuXIHZbMbg4CD6+/tRWVkZV9wSQiyxhjtg2t7extLSUlb3XvGgaRpOpxNutzvpZwlk99CJIAg0NDSguro6466gKnki1EQiETz88MP4zne+g5mZGTz33HOYmZk585jFxUU8+uijePXVVzE9PY1Pf/rT/H0GgwHj4+MYHx/HCy+8wN/+yCOP4I//+I9x6dIl2Gw2PPvss2fGfOmll/Cnf/qnvEKs0WjQ1dUFp9OJ+vp6mEwm/mRbVNqtxJgNFCyptUzJOJb3KjOrhtSLX+wNGvHKc4QBto9o9HXJ77qjkrvsHzH4q895MLV4feGUq3mb0WiE2WyW/FRFbACSin23XK5P0eR72q2KitIQ26NGaAyW6V92kSEAUkGXk7XVE9w5ZJblGkdRBF7X4sH2empiTYQhEUmhrWYYGuxqypM+jntH0Rk1wWAQV69excWLF+FyuXB0dISamhoMDQ2htbUVNpst7lrFiTOcYMCtPTRNQ6/XJ7XRttls6OjowPj4OLxeb9L5ZwuuTCkQCGB+fv6MwMBlHs3OzuK1117DxsYGioqKcOutt6KnpwdlZWXQiHBRTYZGowFFUWfEGu4A/uTk5Nw8ObIt4HBz9Pv9mJmZSSp8yfE7FPIZcPbdo6OjKdm2Z/vzzRfyQqi5cOECmpub0djYCJqm8Z73vAff+ta3zjzmmWeewcMPP8w7mJSWxq9dBU6/YC+99BLuv/9+AMD73/9+fPOb38S///u/44Mf/CBWVlbwzDPPwOFwwOFwJO03IybtVg6KTATC/uQ/sExD/crdoEzKyyBJJ6OGptLbzLIA1nc1cPSaFVu2ppJ9/AEWf/MPO3h17DS4S9f1KRacmNLQ0ICqqqqUyo6Eji0Grunx/Pw8Njc3z90vp+uTiopKdshU6VMmlZqTIz8semFuonKyseHFG50maCj5rp8aDYk3tO+lVAbFgkCYTSy6p+LydCMMw2B9fR0ulwuTk5MIhUJobm7G0NAQmpqaYLFY4q6nDMMgHA7zmTMsy0Kj0fDijNB+M1arFd3d3ZiYmEiYKZptuHIYiqIwNTWFa9euYWpqCj//+c/hdrtRVlaGW2+9FV1dXSgpKcnIGkxRFDQaDYLBIEKhEIBTUam7uxsEQcTMBlaCCxxBEOjs7IRer8f4+HhcMVmujBqhn4HNZkNfXx8mJyexs7Mj6dhiyfbfUG7yIoLd2NhATU0N/+/q6upzDa8WFhawsLCA22+/HbfddhtefPFF/j6/34/BwUHcdttteP755wEAOzs7KCws5NXfL33pS/jJT36CV155Bb/+67+OgYEBPPnkk3jf+96XUm2lUtJuKYrCye4VaJVWDWAywfqOt2Z7FucgdeJTMymwMIosf4pm1U2hr8ci1nxK5SYgEgGe/toe/uOVI1kyaqLFn4qKCrS0tGB0dBQnJyeSjJ1O1gtn33316tVzac5qjxoVlfxDr9cLTr8XE4OxGYzZVhe3FHMgs7Xlw//XbQQtc6B46qAKvOuO1HrWBJnEB2eHRBFCpD6l115ZWcHIyAh8Ph+/WXY6nairq0top32jOMPZaNM0DZ1Ol3YzYLPZjN7eXkxOTuLw8FD0OHISDodx7do1+Hw+7O7u4tKlS6iqqsJtt93GOy5lY90lSRIajQbhcJhvMkwQBNra2mA2m+Fyuc4IIXLESmKySLjeeyUlJRgdHY3ZZkOuWEbMuAUFBXA6nbxBTjzU0nNpuGki2HA4jMXFRbz88st47rnn8IEPfAD7+6cLw5UrVzAyMoKvfOUr+KM/+iNcvnz53PMffvhhNDU14fHHH8frX/96mEwmQSc6Sih94ubBMhFU2LSKCQo4dO9WXvkTSaeXplllC0kyjzU3iY52C9KoxFLJc1gW+Mp/HOJll/R20Tcu5na7HT09PRgfH+evo+mMne6JiEajgcPhgNfrxczMDL8hk6sMTEVFJXuIuV6kGoN5vV5cuXIFY2NjODw8EDM9wWxdPYT/KH3RWwrcbj8cLXoYDfJvsLiyI5IkTsWazb2Ejw8nKH0Kg4JHU5HS67Isyzs1FRQUoKKiIuGBK8MwCIVCZ8QZLnNGDqcmk8mEvr4+zMzMpL2+SkUwGMTGxgZcLhd/SNPQ0IA77rgDtbW1WF5eVsweR6PR8IIaFws0NDSgvLwcIyMjfMaNHNnH6cQc1dXVqK+v5wXEaJQk1ADXmzZfu3Ytbo8dtfRcGvIigq2qqsLa2hr/7/X1dVRVVZ15THV1Ne69915otVo0NDSgtbUVi4uL/PMBoLGxEW9605vgcrlgt9uxv7/P92I4Ojo6M6bQZnZKKX3i0GkJlBdKXyuaFo2NMN/myPYszpCO6xMAlFml6+WxuUuivrEQVrOqUKvEZ/ySGU9/7RDhsHT1wbEWc85mcmZmRpDNZCykWGzF2HdLwc0QKKioKA2hJ9fxMmputFu+dOkStFoturu7UVhok2q6CfnpS8soMGU/HtvbC6K9hobVnJm5RPeHIUkC73ojnTCzJgIKkRjbFhbApqYOqZ4+EgTBOzXFun5H95vhxJkbmwFrNBpZDwGMRiP6+/sxNzcXtweb3Pj9fqytrWF0dBQTExMIhUJoa2vDrbfeiqamJpjNp/2LampqUFFRAZfLxYsg2YSz775RrKmqqkJDQwNGRkZ4F0s5ysTTGbO0tBSdnZ0YGxs70wdGaUINcP2ALBAIYHp6+tz1VS09l4a8+ASHhoawuLiI5eVlBINBfPWrXz3j3gQA9913H15++WUAgMfjwcLCAhobG7G3t8f3WvB4PHj11VfR2dkJgiBw55134hvf+AYA4Itf/CLe/va38+MJrZFWSulTNAV6EkUFCvsKvP3ubM/gDOmUPgFAsVnazaL7ACguN6PUnv2gTkW5/PyiH48964HXL801J95ibjAYMDg4iOXl5TNieba40b5baddcFRWV7BB9WBaJRODxeDA3N4cLFy5gY2MDVqsVAwMD6O3tRWVlJWiaBkHI3wxzeXEHV9cOYCrI7pp+eBhCbTGJkqL0nXhS5UZrbC6zZm8rdmYNw8S26D4irAiSqfc4jN5Gc3OIdmriNvcURcV1asoEer0eDocDi4uL8HhSd8hKB6/Xi5WVFQwPD2N6ehoA0NXVhaGhIdTX18ctC6uoqEB9fT3GxsYk61+XDtFiTSgU4n/7paWlfEPck5MT2fr5pUNhYSH6+/sxOTnJi3RKFGqA6/2KTCbTOaOJTGTU3AwobJcuDo1GgyeffBJ33XUXOjo68MADD6Crqwsf/ehHeRenu+66C3a7HZ2dnbjzzjvx+OOPw263Y3Z2FoODg+jr68Odd96JP//zP0dnZycA4LHHHsMTTzyB5uZm7Ozs4KGHHuJfU4xQo4S0wBspKqBg0innRNh0913QlhVnexo8RJqlT4VG6U/1948BvaUAtVXJeyOp3LzMXA7ir552Y+8g/etOonRemqYxODgIj8eDxcVFRXT65+y7vV6vopsyqqioiEPoBotlWezu7mJychKjo6PY29tDeXk5brnlFnR2dia0W5YLhmHxs5eXASCrGTVebxjFJqCqLLMxxY1CDXAq1rzzDhq7McSaCEsgQpz9nCIg4dZUi3p9hmHAsmxMp6ZsiTM3otPp4HA4sLS0lHbmaixYlsXR0REuX76Mn//855ifn4dWq0VfXx+cTidqamqg16fW96ekpAQtLS1wuVyKsBnnxBqWZc+INTabDb29vZiampI881aqLB2TyYSBgQHMz89ja2tLtsa8UvSR4ey7q6qqzth3qxk10pA3x/L33HMP7rnnnjO3feITn+D/nyAIPPHEE3jiiSfOPOZ1r3sdJicnY47Z2NiICxcuxLwv10ufOAiCQFmhBmueEEIK0JEIrRaFD94H96e/kO2pAEg/o8ask6f84sQP0BoDWhsJLCwJa6qocvOwuhnGxz7rxiMP2VFZKv67nCydl7OZnJ2dxfT0NDo7O7O+QNvtdhgMBkxMTKCjowNFRUVZnY+Kioq0JNu8+Hw+eDweuN1uBAIBGI1GNDU1wWQypbTpkfsIa/biFnY9p259BQXZOXkOBCIwEBE01mTedTOWUANwmTU0vvmjPdjKr5efRVjyTEYNC2BLQMnT9Rdm+RIdq9WKtbU13sVIidA0DYfDwbsCVVSk1osnHizL4vDwENvb29jZ2YHRaERpaSnq6urSts8uKiribcZ7enpQUFCQ1njpwsUhDMMgGAyCpmmQJAmz2YzOzk64XC7s7OzAbrdL8npSZr5wRgkulwskSZ5r6SEFUs6X6/U0OjqK3t7ejJg53Ayl56rUJZJ8KH3ioEgClUVakAr5vmve8TYQGmUsmOk2EzbR8vXJCIaBvYAe3e3KszVXUQ6evQg+/lk3Fq+IT0dOZcHlUmCNRmNGe8TEg2VZUBSV0L5bCm6GQEFFRWnodLpz7ijcBnRpaQnDw8OYn5/ne1c1NjaisLAQBQUFivjNhoIRXPjxFf7fJmPmz03DYQZsIIiOpuzEEPGEGoDrWaM7UwbFMAQixPXY8JiwwE+aRL02lznT3NwMvV6Pubk5RWSDxkOr1cLhcODq1asJnXbiwTAMdnd3MTc3h9deew1ra2uwWq0YGhpCb28vysvL0xZpOKxWK3p6ejA5OYmDg8w05E4ESZIgSRIEQfBuUMCpEGKz2bC4uIitrS1JXkvqzBeNRgPn/2PvTYMjyctz3yezsrL2fdMulXapJJWklmaa7WLAZmDAY4bhHA8wxiz2jXtizLUdvnFN4DAGn2vDAR8uYTCHMz4EeIEZCI7NhJeZIGyuD3iIXrRUSSp1t/a9VYvWUlWpqnK5H0TmSGottWRWVavz96m7VZWZXar85/t/3vd93mvXkEqlcP/+fcm/o1KLKXa7HX6/HxMTE9jZ2alY8fNhQhFqCiTfippKbX0SoKkKMhd2umB51y+U+yoAAESRFTUaUt7NKscB9/doDPRKP+lH4epwkOTxpy9sYTRUWDlyrg9zgiDQ3NyMqqqqc8dMlgqhXeui8d0KCgry8+qrr6KjowOtra34whe+8MDPf/KTn2BwcBAURYm+gJeh0+mQTCaRTqexvb2Ne/fu4datW1hdXYXRaMTAwAD6+/tRW1sLjUZTUAwmp54TvL2OxMHr66OxxB41LMsjuraF/s7yVTxcJNQciW578NeuYmNxA8BR65NQUcOCRERVWMsTAUKc1CT4mmm1WoRCoYp+PlAUhf7+fkQiEaysrFz6eo7jEI1GEQqFcPPmTYTDYbhcLjz++OPo6emB2+2WbSNtNBrFKttymSEfR5gGJYg1LMuC53lQFIWhoSGsrq7m9JlehhxVJCRJwuk8soSYnJyUNOkvx/UajUYMDQ1hfX0diURlTLN7mFGEmgJ52CtqznoYGbQkHGUqvz2N7tnKGNVNqosLnlQEB4qU//e+ElNh0G+uuJHrCpVDJsvjK3+zjR/fyP/Bme/DvLa2Fi0tLRgZGUEymbzwuHJxvD/6vPHdxVIJmXkFhUqGZVk8//zzeOWVVzA9PY0XX3wR09PTJ17T0NCAb3/72/jQhz6U0zH39vaQTqfx0Y9+FI899hjm5ubgcrkwPDwMn88Ht9v9QHVAJcVgqWQGozdOmq8bSuxRs7YUQ3tjeZOHp4UaoepDEN3W19dhtVrwwV8yYWdzByxHgoUKPIAwVQ8UuME8vWwLYo1er5d8Iyw1KpUKfr8fOzs7WFxcfODnDMMgHA5jcnISN2/exM7ODmpra3H9+nV0dXXB4XCUrC1Zp9OJZsjRaLQk57wMYVpXJpMBwzDiqPXBwUFsbW2dO2o6V+TykuF5Ho2NjTCZTJJWLMvVnqTRaOD1epFIJIr+TB91FKGmQAwGQ15CTSUF9BcFLHaTCkZtBVzrwAB0nS3lvgqQdHEVNQSAGltpWkCWIyT8vVYUqS0pXGE4Dvjm3+3if/5oP8/35f8wdzqd6Onpwfj4+Lnlz3JOBThtkleu8d0KCo8yt27dQmtrK5qbm0HTNJ599lm8/PLLJ17T1NSEvr6+C9eY9fV1fP3rX8cTTzyBd77znYjH4/j4xz+O8fFxPPbYY7Db7Re+vxCfQLkioduvrSCbeV0k0elUUKlKF3fdX0ugt4Ur++aJIAiwLItIJIJQKITbt28jFouJoptg8kxRqp+3Qe2C4wkkCSNSpPRVxM3NzTCZTBUv1pAkid7eXhwcHGB+fh6ZTAYbGxsIBAIYGRlBPB5HY2Mjrl+/jvb2dlit1rLtQTQaDQYHB7G0tCRb+3G+qFQqqFSqE3s4wWcvlUrhzp07Bd8bcgkfggDk9XrFqZZSTNeS00eGIAjU1dUhlUpJmiA7fY6rjiLUFEi+rU9A5XyhLssseSwUaKr812r+cPmraoqd+gQAVdasBFeSG6tRoKPTCn0liG0KFcvf/Usc/+MHO+C43IKRi6Y+XYTZbMbAwABCodCZGTU5g4Szjn16fHcljBFVULjKrK+vo76+Xvx7XV1dQR4bP/zhD8GyLF544QXcvHkTAwMDaGlpyXn9KKiiRobH6O5OClNjJzesRkPpKpnXVg7wtsfMF7YdyU02m8X9+/exuLiIcDiMeDyO+vp6PPbYY2hvbz9TdCMI4P1v1SDFa7GpapDt2rxeL6xWa8WLNdlsFlarFevr67hx4wYODw/R1taG69evo7W1FWazuWL2HMf9ddbW1sp6LYeHh1hZWcH4+DiWl5fhdDpFY2mCIODz+UBRFCYmJgr6/ctVUXM8nhGmWo6Ojl5YsZzrceVMllEUpSTIikTJvRdIvq1PlcRlAQtJEqixUViJZZHjPk4WiHc9AdUXvw42Xr4ex2IragDAaSrtwnR/C6hvsiByPy7JaGaFq8n/dyuJvTiHTz5nB62+OLC4bOrTRej1egwNDWF8fBzpdBp1da/7Csgt1JwXgNTU1ECj0WBkZAR+v7/skykUFBQu5vnnnz/x90LazyvBJ/DG/1p6QCAvlT/N6vIBfvG6GUDpW8EODw8RjUYRi8XAcRwcDgdqa2uxv7+PlpaLq6eFMdosy4LQmwCmuI3wZe9ubGwEQRCYmJi4tNKrlKRSKUQiEUQiERAEAbfbLfqrCFPNKhXBX2dychLZbBZer7dk506lUgiHw4hGoyBJEm63G729vdBoNOA4DgzDgOd50DQNgiDQ3t6OpaUljI2Nob+/Py+TZblimtPHdTgcYsVyT08PLBZLQcdlWVY2QY/jOKjVajFBtrGxgZGREQwMDECj0chyzqtIZaw+DyE6nU6cFf+wcdkDmmVZ7O7EkNlfL2tpLKHXw/rMk5e/UM5rUBevNNsMpVeQY3uA1WVClat4oUnh6jJ25xB/+kIM8cTFG5higw+apjE0NIRIJHKiX1lOoea4R81ZOBwO9PX1IRgMFmx2WCkZSwWFSqW2tharq6/7saytrUkyZjbfquZKaH3aiR1g7s6DlYWl8KdZW03gbY+ZxDWrFBU1iUQCS0tLGBkZwZ07d0AQBLq7u3Ht2jU0NTVBp9Odew3CBjqbzZ4wfqUpCZ4XOfxiGxoa4HA4EAwGyyrwHRwcYGFhAbdu3cKdO3egUqnQ19eHoaEhNDQ0QK/Xo729HWq1GqFQqKKrgIRrTyQSmJ2dlfX7l0gkTnxuFEXB7/fj2rVrqK+vF4UCwWRYGN8tfH5NTU2oqanJeyhCKSpqBI5XLMdisYKPW6r2c6ESaGRkBAcHB7Kc8yqiCDUFotfriy45KxdnCTVCKerExARGR0exv7+PGo8dTnN5zYU1//Hpsp5fiooai7Y8pX77CYAyGNBUpyjXCuczu5zB574eQ3Tn/O+pFILK8R7w6elp8Dxf8tan05hMJtnHdysoPMoMDw9jdnYWi4uLyGQyeOmll/DUU08Vfdx8Y7BKMBOOLJ49Aljuipr19STeMmAApXp9PZTj8+B5Hnt7e5ibm8OtW7ewsLAAjUYDv9+PgYEB1NXVncikn2UmnM1mRXGGIAhxjLZGo4FarYZagrb8XI9QX18Pl8tVUrFGGDE/NzeHGzduYH5+XjTlHRwcfOAzBI4+x9bWVuj1ekxNTZX9e34Rglccy7KSjkTneR7xeBzz8/O4efMmZmdnodVqT3xuNE2fe02CkMswjPj51dTUoLm5GSMjIzmLwqWqqBEQKpbn5+exsbEh2XGl4KxjS5EgO86jkCxTWp8KpNDWJzlvilwRHo6pVAqxWAzRaBQ8z8PhcKClpQUGg+HE6zNZBvHD8iz8fGMTTG8eQvzfR8pyfkICZ14DXb6ezFQaUFM6dLSSuDf3cLbqKcjP/SiDz34tiv/7E0401jwoTgolrMUiBGnz8/MYHx+H1+uVtaIml0yRML47GAwilUrB6/U+Eg9/BYVSQFEUvva1r+GJJ54Ay7L4+Mc/Dp/Ph8985jMYGhrCU089hdu3b+Ppp5/Gzs4O/uEf/gF/9Ed/hFAodOFx9Xp9XlXN5R7P7TIx2DgnFpCzomZzM4U39OihoU+uhVJV1HAch93dXUSjUezu7sJkMsHlcsHr9V66/hIEIYozx/9NpVKBJMkznw1UCU2XgSNPJZIkEQgE0N/fL0v1Ac/z2N3dRSQSwfb2NoxGI9xud06f4XGam5uxvLyMiYkJ9Pb2ylYpUSwEQaCjowPz8/OYmpqCz+crKA4QRK1IJIKtrS3o9Xq43W40Njbm1a4EQDw/y7LIZrPi+PYjM2sKY2NjObVJyzn16bzPiKZpXLt2DcFgEIeHh3nFMOVoPxcSZOPj42hqakJ1dbUs578qKEJNgRRiJiwotuUSanieRyKRQCKRQCgUgkajgcvlgs/nu7Bf0G1VIRPjkWbK0wZl+NAz5RNqJKio0VLlNc/KMsA2q0FvN4nJ6fL5/ShUNrtxDv/5G1H87kcc8LWeXA+kXLeE7N/a2hqmp6dhNpslOe5p8rlmYXz39PQ0pqen0dXVVXZBXUHhqvDkk0/iySdPtjH/8R//sfjn4eHhvE1GdTpdXhU1hbQ+SQePrqoMIp6z4yyjTEJNLJbGQJsWet2Dm6VihBqWZbG9vY1oNIqDgwNYrVa4XC60tbVdum4e95sBjlp7OI6DRqMBQRCXvl+Kzqd8qampAUEQGB8fz9uz5Dw4jsPOzg7C4TD29vZgsVjgdrtz+gwvorGxESqVCsFgEH6/v6LFmtbWViwvLyMYDKKvry+naxWqtsLhsChqeTweNDc3F/1/PU+ssdlsYhWIz+eD1Wo99xhyVtRcJL4IMUwoFMLdu3fR2dmZk1hTrvZzJUGWO4pQUyCFVNSUo/RWUOpjsRh2dnag1+uhVqtRV1cHp9OZ0zFIgkC1ncJqLAu2HHHOm98MusaNzEak5KcmJXggqwkWAIdydhpyPLCxo0Zvlw6Td5TKGoWzSR3y+OI3Y/g/ftWGN/S/bkxY6NSni6irq0M6ncby8jKSyaTkRoi5VtQICNU+CwsLGB8fh9/vvzQgVwILBYXykG8MVogwQUCa5FSdjYFJy6H6PKHGKP1memcng446NSyms9ewfOPRbDYrVmCn02nY7XbU1dXBZDJdug4K4szx86lUKphMJni9XoRCIQwMDOS0XktRUVPIsl1dXUhO7nAAACAASURBVA2CIMTKmkLEGpZlsbW1hUgkgng8DrvdjurqanR1dUn6LKmrq4NKpZJUWJILofolEAic+8zleR47OzuIRCLY2dmB2WyWRNQ6i+NiTSaTAU3TIEkSJpMJg4ODGB8fR3t7+7n7JzliJSA3QYUkSfT09GBubk4Uv3IRTssx0AFQEmS5Url3b4VTSEVNqaYOsCyLnZ0dRKNR7O/vw2KxwOVyiaMs5+bm8r4Z1CoCVVYK69ulrw4hVCpYP/g+RP7rC6U/twTjuQkALjOL6H75F6CNXQ16u4HQ3RQquI1ZoYwwLPAXL+5gZ5/Fk/+bCUBxU58uwmw2w+l0Ynx8HL29vZJW1xQSgCjTCRQUHg70en1eHgflElVJgke758iMtOYcoUbq1qd4PIt6BwmX/fyK4FyEq8PDQ1GcESY1ndUefxYnJjX9/LMXWpqOb96qqqrEapWBgYFLRQUJ5jsUzPFr7e/vz6kdWBC4IpEIUqkUHA4H6uvrZR+fXV1dDZIk87rWclFbWyu2F/X394Om6TMrjjweD9rb22XfzAvHFwyGBbFGp9NhaGgIY2NjyGazZ7bsyBUr5RrPEASBtrY2rKysYHR09NLf/WVDF4ohl2MfT5AFAgH09fXlJSw+CskyRagpkEIqauQsvRUeBrFYDKlUCna7HTU1NWeWvxVa2aPXkHCZVYjul94BX/W+p0B89VvgM9nLXywhBCVNVFBjZRDdr4wH5caOBn09GoSmd5Etb1eWQoXC88B3/nEf23scPvxes6zlvHq9Hi0tLQgGgxdmqvIl34qa4yjjuxUUKpt8PWoKQYo9gNeZhU59JIi4nTRUKuB0vk5KM+FkkoFdB9RVXSwwnxcHJhIJRKNRbG1tiR4dXV1d0Gq1l56b4zhRoAGONlEURYkCzXl4PB4AyKlapdQeNafxeDwnhKWzNsGZTEYco53NZsVEqcFgKOnG0uPxnBBrzjPTrQSEz/XWrVswmUxIJpOw2WyyVBzlguCRxDDMCbFGmGA5Pj6ObDaLhoaGE++Tq6ImX++bhoYGMYYZGBg49/7leV629rh8xKWWlhasr68rCbIzUISaAim0okZKoeZ4toNlWTidTjQ3N0Ov1194QxdzHVaDCuksj/1Uicsx7HZYn3wbdn74o5KelpAoC+EyZwHoJDmWFKxGgfYOK5YW95BIlm8Eu0Jl88pPD7AXZ/HWPnknGRgMBtFcLpPJoKamRpJjF5NFPD6doKurC3a7vehrUlBQkIZCBzqUErWKR4vr9dG+KhUBj1ODjXD6xOuk8qhJp1nowKCl4fKKF6GiRpiWI4gzWq0WLpcLfX19Oa2fgjgjVBIIZsCXiTOn8Xg84Hn+UrFGRR5VKRcTtRBFDl53u90nKmtomsbh4aEozvA8D5fLhc7OTslbevPF5XKBJEmxWiUXwa2UsCwrVhwdHBzAbDYjHo+jv78/p8otuaEoCgzDIJ1Og6ZpqFQqUBSFa9euYWJiAul0Gq2treKeS66KGiD/6hGPxwOapjE6Onpuwknuipp8RKDa2lpotVolQXYKRagpEIqi8m5jKlaoEcyAhQeqSqXKK9sh1XW4LCpkGB6H2dJu8LW/+n6g5EKNNLeIw1h5pSub20BtvQWxcBzbu6WvklJ4OPhZIIWNsA6f/JC85bwajQZDQ0MFTS44C5Zliw5KhekEgUAADQ0NDwhIj0LZrYJCJVJIsixfir2729yZB1p1qj0PCjUGCSpqGIYDn8qiqzu3tqR4PI69vT3cvn0bRqMRLpcLTU1NOW2sOI47Ef+SJCkarxaz6auqqgJwcWXNkRB01KJbMBIs2y6XC+l0Gjdu3IBGowFFUXC73ejp6ak4McThcEClUok+MDpdeZOGDMMgGo2K7WBOpxONjY2i31E8HsfExAR6enpgMpnKeq3A6/s9obJGECH9fj+mp6dx584dsepHroqaQrHZbPD7/QgGg+ju7obNZjvx83J61JzF8QTZWdf7KKIINSVEpVLlLe4IDuexWAzb29vQ6XRwuVzw+/0FZ4uLHctIEgSqbRRWSm0u3NcHXXcbUtOzJTslKZFQY9FVnlADAFv7gMVhgkaTxP1w5vI3KDySLN1X4yvfOcSnfpOF1SRdmezpIOH45ILjwY8Uxy4UrVYrqYCkoKBQPHq9Pq+pTwJ5tRAUcZvr1Bwa7Q+2atd4NBg99neaJkGri1unOI5HYucQj/vP39Qen9QUj8dhNBpB0zQGBgbymtQkfHaC10wuk5ryIRexRq0iwLCFx7CF/lp5nsfBwQEikQhisRg0Gg2qqqqwtbWFnp6eim7XsFqt6OrqEn1ASl2tks1mEY1GEQ6Hkclk4HQ6z20HM5lM6Ovrw8TEBLq6ui6cslQqhO96JpMBRVFQq9UgCALd3d0nzHuLreSVA6PRKFYst7S0wO12iz8r19Snizg+vtvr9Yprwlk8CrGYItSUkFwrWU6bAZvNZrhcLni9Xkl6CaVowaJUR2LN+hYj0VyE3DA/9wxSn/5Cyc4nlUeNUVOZQg0A7CcBrVaP5kYCC8vpy9+g8Eiyusnhs38Rxe9/woFqlzSBCMuyDwTixycXCEFlIeteMR41p1GmEygoVBY6nS5vjxohSVWK4L6jKoOzlojqU94xBkPxa9RWOIm3XHvQiP2iSU2ZTAb37t07dx07PqlJ+MyEliapxZnTXCbWSBSW5YSQLI1EItja2oLBYIDb7RanFQHA9va2eK2VLNZYLBb09PRgYmICvb29sreWHPfqYRgGLpcL7e3tOYlEBoMB/f39CAaDaGtrg8PhkPVac4EkSbEVCoAo1rS1tWF5eRljY2OwWq0VGRsICafx8XGk02nU19cDkFeoKWatFcZ3BwIBpFIpNDU1PRKizFkoQk0JuUggyWaz2NraQiwWO2GidZYZsBTXISw0xaCjSbgsKkT2Stc2QzzxTlD/5S/A7MVLc0KJlHEdVdmtRYcZgFHp0NVG4s5sZff9K5SP6DaLz309hv/rYw60NhRvTHiekZ0Q/KyurmJ0dPRc08aLkDoAKWR8t4KCgjwU4lEjxGC5rguFRl4WHYsay9kxVrX75LpZrJHw/bUDvO0xi/j3dDqNaDR66aQmgiAeiEfPGqMtV+XMZVRVVYHneQSDQfT39594TlBksS41FyNMHIpEItjd3RXHQbe0tJz5vLLb7Whvbxc9ayqt9ek4x6tVfD6fpJMWAYhePcL3z+12o6urq6B2K51Oh8HBQQQCAbAse6ISpFwIYg3LsuB5HhRFgSRJNDY2gqZp3Lt3D21tbeW+zDNRq9Wit87h4SFaW1tlFWqA4ipeKIrC4OCgJBXWDzNKlFkE+X5hTrc+HX+gsiwLh8OBpqYm2Z3hpTQ1tuiPzIX3kqXpgSK0Wlj/w3sR+x8vluZ8EqVuaLJyK2oEGBaIJjTw+1QIhg7KfTkKFUo8weFP/3sMn3zOjoGu4gLSy0pj6+vrodFocPv2bQwMDOQV7BXSH30Zp8d3Dw8Pl90sUkHhUaTQyZtnVfFJTWdV5tyJUTWnKmqKMRJeWznAOx43i96FsVgsZ+9CobrorElNhZgBy4Ew/lioVhHW82InP531uxFawyKRCPb392Gz2eDxeNDR0ZHT52Cz2dDZ2VkxPjAXcbxapbOzs+jWolQqJVbOEAQBt9sNn88niWAltOcFg0EwDCPJoIFiEb4PLMuCYRhRrKmurkYkEsH8/DzsdntFfgdUKhX6+/tx584dhEIhWc2EpUCosJ6fnxcTZHJNqapUFKGmCITKlFwf+gRBIJVKYWlpSRx96HQ60dnZWdIbWurpUy7zkVhTKnNh9X94GvjmS0czhGWGkCigI8HDpGMRT1X2AsPzwNo2hcE+M8Ym9st9OQoVSjrL48t/tYVPvN+KX3is8F73XLI5brcbarUaY2Nj6Ovry9lcUM4ARBjfnc0+6EGhoKAgP6WYvFlIvsxlZOA0nl9BW+XSnJhaVGhFzfJSHC3uGEZGZqDRaPLyLhTEmUwmA4ZhxKqZShBnTnOWWHPaoLlQGIYRJw4lk0nY7XbU1taiu7u7oGSp4AMTDAYrXqzR6XTo7+9HIBBAe3t73lMNE4mEWDkjGCn7/X5ZRoCr1eoTYs3pkdjlQLhPGIYBz/OimbZWq4XFYsHY2FjFTi4iCAJdXV1YXFzE5uZmUZ6lpYAgCLS2tmJtbU0c3y18zx6FChtFqCkCIaNz0caB53ns7++LJloURaGhoSHn0YdyILVQQ/zcXHg1lgVTisKaujqY3/o49v/thuynkqqiBgBqbVncrXChRmA5SmLAb8HE5F5pDaMVHho4DvjLH+xiZ5/F079YWPl0rmW3xycXdHZ25tSvLkdFzXEcDkfFmQYqKDwqFGImLHXs8yA8OqsvNuWnaRJWC4GdvaPNkaGAipqF+R002MKwWNw5exeentREURSsVitWVlbQ3t5e0Rue02JNsRU1mUwG4+NToqmt1+uF0WiU5DOwWCzo7u4WzWUrueJSq9ViYGAAgUAAzc3NcLlc575WmDobDocRi8VA0zQ8Hk9BbcmFIFSCTE5OgmGYijD1P94Glc1moVarwXEczGYznE4ngsEgfD5f0RVLcvhqEQSB5uZmrKysYGxs7IT4UanU1dWdGN9dCePbS0FlSecPGeeZ2XEch62tLdy9exe3bt3C+vo6LBYLWlpa4PF4UF1dXdYAX45gRTAXLtWyafjg+0tyHqkqagDAbX64su8rEQI9PitodeUGcArl5wc/iuOb/3MHHJd/Viaf/mij0YihoSHMzMzg/v37l76+0kt6FRQUCketVudd0UaSZF6TN/N98tVZGZi1l8dWLvvr61K+FTXrawm8fdiKnh4fXC7XuSKNIMxks1lkMhlxYhNFUaBpGjRNo6urCyzLYn5+Pq9rKAfV1dWorq5GIBAAieLiV57n0N7ejscffxwtLS3iWGipMJvN8Pl8CAaDSCQSkh1XDjQaDQYGBrC4uIhwOHziZ0KieW5uDjdv3sT8/Dz0ej2uXbuGgYEB1NTUlHQvQ5Ik+vr6kEqlMDMzUxGVIEI1Gs/zyGQy4DgOBEHAaDRicHAQ09PTiEajRZ1DOKYc0DSN5uZmjIyM5F2heBFy/W6cTid6enoQCASwu7sryzkqjSsXxb766qvo6OhAa2srvvCFs6cDff/730d3dzd8Ph8+9KEPAThS6t/whjfA5/Ohr68P3/ve98TXf/SjH4XX60V/f79YKgiczOgwDINwOIypqSmMjIxge3sbHo8Hw8PD6O7uhsvlAkVRMmdzcoMkSVluIi1NwmUuzVeKf+Mboamvlv9EEgo1DlPl+9ScZi0GtLZZYDJeuaVCQUJ+fDOJr/zNNjJ5tj/ma2Sn0WgwPDyM9fV1LC0tXXpsuXuZy53RU1B4VCnk3lOpVLLFYCTBo91zcTWNgNP++rqUj0fN5mYK13sN0GrPfs9xcYZlWXF9FYQZjUYj+mkAr7dApNNpLCws5Hwd5aKmpgbV1dW4ezdU1HG0Wq3s2XiTySROWKp0sUbwgVldXcXGxgZ2d3cxMzODGzduYHl5GSaTCcPDw/D7/aiuri6rib4wEhsApqenK0qs4TjuhNih0+kwNDSEhYUFbGxsFHx8uQ1/XS4XfD4fxsbGsL8vjeWBnNP1zGYzrl27hqWlpYr4/cvNldp9sSyL559/Hq+88gqmp6fx4osvYnp6+sRrZmdn8fnPfx6vvfYaQqEQvvKVrwA4El3++q//GqFQCK+++ip+53d+54Ra96UvfQmBQEAsvQSOMjrf+ta38Eu/9Et49dVXkUgk0NjYiOHhYbS1tcFms524ueQvu82Ns9z+C0UIDDKZDLLZLPQ0D7NO/s0LQZKwfEj+qhopK2qsuodPqAGA8A7gqTHDYVc6JRXOZzR0iD99IYaDPIzFCwlAhEkA+/v7uHv37rkPaqWiRkHhapPvRkBOj5omRxY6OrdNg9t5TKgx5iYmx2Jp9LdqYdCdfP154oxarRYFmou8Z4SNbyKRuFT8rgRqampQ5XYVtUErlbxuMpnQ29uLiYkJHBxU7oAGnudxcHAAg8GAu3fvYmZmBjabDY899hh6e3vh8XgqysCVIAi0t7dDp9NhYmKirPsqlmURiUQQCoUwNjYGnU4HmqbFa6JpGteuXcPGxkbB95fcQg1w1LLX39+PqakpbG1tFX08uRNlQtveo5Asu1JR7K1bt9Da2orm5mbQNI1nn30WL7/88onX/OVf/iWef/552Gw2ABDHvbW3t4sj1WpqauB2u88sV5uZmcEXv/hFvOUtb8Frr72GRCKBr371q3jve9+L5ubmC0so8y27lYtiBaOzAgOVSiUGBtV2DfQa+b9aql95DwiNzD2VlHRlnWbtwynUAMBOHDBajaitruweVoXyMrucwee+HkVsJ7fvOsuyBT3MSZJEb28vSJJEMBg8cz3jeV4RahQUrjD5btblisHUKh6t7tyqaQDA43w96ZGLR83ubgZttRSs5qPXHk+QHY/BaJqGVqu9VJw5DUEQ8Pl82Nvbw8rKSs7/j3JRW1sDkih/0jMXjEYj+vr6MDk5iXg8Xu7LEREsGu7cuYMbN27g/v37cLlceMtb3gK1Wo1UKlVR4sxpBI8Vm80mju8uFSzLIhwOY2JiArdu3cLe3h4aGxtx/fp1dHZ2gqIopNNp8ZqE5NLe3h5mZ2fzXrdKIdQAR9PArl27htnZ2Zzayy+iFImyR0GkAa6YULO+vo76+nrx73V1dVhfXz/xmpmZGczMzOBNb3oTrl+/jldfffWB49y6dQuZTAYtLS3iv/3BH/wBqqur8e53vxtWqxV/93d/h2eeeQZPP/00Ojs7c7o+Octu86EQoSafrA1BEKix01AXafh2KRYrbL/8i7KegtRLVx6rV5dfpCuGgxTAq/VobSp+5KLC1WUjwuCzfxHFyv3L/SOKCUCErJrNZsPo6KgygUlBQeFC8q6oQW4bqlZX5tJJRILfx/z8PNjs60nAy4SaeDyLOhsBl40CwzBiDMbzvOg3o9VqxakzhSKI39vb21hbWyv4OHLD8zy2t7fBs0Ukvkq8vzMYDOjr68PU1JRkrSWFwHEcotEoQqEQbt68iWg0Co/Hg8cffxzd3d1wOp1Qq9Xw+/3Y2dnB4uJi2a41VxoaGlBVVYWxsTFZYwCWZbG5uYmJiQncvn0b8XgcTU1NuH79Otra2mA2m0XhQGgvFIRU4HV/nWw2i1AolJdYU8qkk0ajwdDQUE7t5RdRaBJO4UGulFCTCwzDYHZ2Fv/2b/+GF198Eb/5m795osXp/v37+LVf+zV861vfEm+Mz3/+87h79y6Wlpbw5je/GZubm3C5XHmPh6yU1qdcr4PjuBOBQT5ZG0pFoNZBFzTeMh80vypv+xNBSvcfoMmHt6JGIJ0F9rNadLdX7iQDhfKzs8/hP/+3KKbn0xe+TopMUWNjI+rr6zEyMnKmubucPCoZHQWFSiTfmEqOZJlOzaHRcfYGkeM4bG9vY2ZmBrdv38bq6ipMJhP8PXXiay7yqEmlWFhoBh6nSjQDFmIwjUZTtDhzGmEzGYlEivLUkBqO4xCLxTA9PY0bN25gc3MTmiJmdJdj1TYYDPD7/Ziensbe3l7Jziu05kxOTuLmzZvY3t5GbW2tWP1ht9sf+A4Jol0ikcDc3FzF+4DU1NSgsbER4+PjSKcvjjnygWEYbG5uIhgM4vbt2zg4OIDX68Xjjz+O1tbWE+LMaSiKAkVRoi0E8LonlE6nQyAQyHktkqui5rzf6/H28nv37hX0+y9VFdCjwJUynaitrcXq6qr497W1NdTW1p54TV1dHR5//HGo1Wp4vV60t7djdnYWw8PD2N/fx3ve8x78yZ/8Ca5fvy6+RxgNqNFo8LGPfQx/9md/BuD18dy58jC0PnEcB47jxBtTCAzyKaMV0NIkqm1qbGzLmOnu7obB34VE8I4shz8yxJLmWCqCA01xyDAP9+LFckAkTqO/h0RgqnL7rhXKS/KQxxe/GcN/etaOx/t0Z75Gqod5VVUVNBoNRkdH4ff7YTQaiz6mgoJCZSMky3I1hs07Bsvh2d/hyUB1bAljWRbb29uIRqOIx+OwWq1wuVxobW0V1zqO24TZSCCRAnS6swWHTIaDKnuI5la9GH+VYuNDkiT8fj8CgQAIghDj31LDsixisRgikQgODg5gt9tRXV2Nrq4uEASBjW0G6UT5E5/5oNfr4ff7EQwG0dXVBYvFIst5GIYRP7tkMgmHw4GGhoYLhYXTkCQJn8+HO3fuYGZmpuJHuLvdbqhUKoyPj8Pv90OnOzvmuAyGYRCNRhEOh3F4eAiXy4WWlhYYDIaCPLEo6qgSDjjyNSUIAi0tLVhZWcHo6CgGBgYuNWeWa+rTRfGXINbNzMxgYmJCbDXP59hKRY00XCmhZnh4GLOzs1hcXERtbS1eeuklfPe73z3xmve973148cUX8bGPfQyxWAwzMzNobm5GJpPB008/jY985CP4wAc+cOI99+/fR3V1NXiexw9/+EP09PQAyF+oqdTWJ6Gt6fjPBRfzYgMDs57CYYbH9oF81SSmDz+DRPD/kf7AFClpRRABoNqWxXJUI91BywQPYHWLwmCfGeOT+6jwhItCmcgywFe/s42DhAnveIP5gZ9LmXWx2WxiECxMhZCTSg5aFRQeBfR6PQ4PD/MSavJpj7jsDjdrWdRYj6qOt7a2EI1GkUqlYLfbUVtbe+7GmCRJOO0qEOdMl2UYDmwyjUGfpSxZaZVKhf7+foyPj4MkSXg8npKcN5vNigJDKpWC0+lEY2Pjmd6PVBF7wHKu3DqdTpwe29nZCavVKslxs9ksotEoIpGIKDB4vV4YjcaCn1VCBcjMzAzu3LkjimSVisPhAEVRCAQC6Ovry3ldED67cDiMTCYjCqtSJHwEseZ4uyJJkmhoaABN0xgZGcHAwAA0mvP3BXJVp1x2XIIg0NHRgaWlJYyNjaG/vz/niV+KR410XCmhhqIofO1rX8MTTzwBlmXx8Y9/HD6fD5/5zGcwNDSEp556Ck888QR+9KMfobu7GyqVCl/60pfgcDjwt3/7t/jJT36Cra0tfPvb3wYAfPvb30Z/fz8+/OEPIxqNgud59Pf34xvf+AYAPLStT8DRDXo8YBEqZ+TI2rgsFNJZDom0PP934hffAcr2VTA70paTqozSj2+ssjBXQqgRWI6SaG9RY3Y+A45/NBZNhcsx6jjY9Cyyh2msbyTwZ19dw+N91x4o85d6hKPRaMS1a9cwPj6ueNYoKFxxdDqdWDGQCyqVStLWCCe1jkBgGSzLwul0orm5GXq9Pqc1zWkncMg8qDZwHI/EbhpvHJCn2iJXjos1BEGIgzekJp1OiwIDwzBwOp1oaWm5dJNMFeOBWOZQRavVimJNR0eHONwkXzKZjCgwCJ9dW1ubpKPHBS+4+fl5hEIhdHd3V3RLi8ViEcei+3w+mM0PJoiAs8WZ9vZ2Wca2C58Xy7JgGEYUa6qqqqBWqzE6Oor+/n7o9WdbCpRLqBFoamqCRqPJSVQ6fmylokYarpRQAwBPPvkknnzyyRP/9sd//MfinwmCwJe//GV8+ctfPvGa5557Ds8999yZx/zxj3985r8bDIYzJ0OdRzlbn4R2JuH8DMPg8PBQzFTIufASBIEaB42lSBpZRobSC40G1v/4y4j997+V9LCUQXrTXJfp4fepOU0sYYDPZ8DszB4OM0ppzaOIXsPDpDnEwd4uEgkScwsPTkD515/G8Cvvqnrg36XOimi1WgwODuKnP/0plpeX0djYKOnxFRQUKgO5288vWpo03A6MVALeri5otZfHCsdjMLVaDaM+DaPhwU3kdiSFtw5LU2VRLBRFnaiscTqdkhw3lUohEomICVC32y16d+R8bXIPq5AZYbxwIBBAW1sb7HZ7Tu9Lp9OIRCKIRCLgOA4ulwudnZ3nbvKlgCAItLa2YnFxEZOTk3m3wZQak8kEv9+PiYmJE0LYaWFLTnHmNMLnJbRBCWKNw+GAz+fD+Pg4+vr6YDKZHnhvuYUa4MgChKZpsb38ss+sFBU1jwpXTqgpJTqdLi/zSpIkS2rKJQQGx6t4BL+Znp4e3Llzp6heznxQkQTqHDSWI2lwMnwE9AfeB/zldwEJK5ZIvfSfi81wNbP8G1uAt8WC9bU49uPl92FSkBeNmofDxIJgMghHklhcTF3a/vbKj6NnCjVyoFKpYDKZsLu7i8PDQ1n66x+VslsFhUpF9qrmcxc1Ho+1a2DW1Z/z8yPO8vyjKAp2ux2t3h3snrr0+2sJ/OIbCquukAu1Wn2isibX6qXTJBIJUZyhKAputxu9vb05ZefPvK4i9oCVsnJrNBqxsqa1tfXcz/a4sAUc+bH4fL6cBEIp8Xq9WFlZQTAYRF9fX0VXTOj1evF7a7PZkEwmwTAM3G637MLWeRxvg8pms6IZuMViEb8H3d3dD1RYyTX1KV8ByOFwoLe3F4FAAD09PRd6LClmwtKhCDVFkG82pxQcz9oIG4mzzIAtFgu6uroQDAZzLmUrFo2aRLWdxvrWg9n2YuFramB52xuw96+vSXZMlQxCjdtydUWMyC7gqjJBSx8gsnX1KoceZUiCgcdKgOQz2IomsbKUxHKemujsQgJziwm0euXPXgmjIfv6+nDv3j1MTk6ip6dHCRwUFK4QcvgEchyHvb09RKNRbByYAHXbA6+ptTIw6y4eyCAYgJ43kKGnuxp3l1+vyF5bSeCdb6yMSprT0DSNgYEBsbIml1YdnucRj8cRiUQQi8Wg1WrhdrsxMDAAtVpd9DUVV1FTKVLNkVgjVNbwPC9WLSWTSbFyRqVSFS1sSUVDQwNIkkQgEIDf78/Zs6SUHK86IkkS0WgUTU1NqK+/WFgtBcfboDKZDGiaBkmSMBgMYtt2S0vLiVbDcpgJn4fJZMLg4CDGx8fR1tYGl8t15utKMZ77UUmWVd4d4LCMPQAAIABJREFU9hAhGNmVm/OyNpeZAVssFrS3tyMQCGBwcFCSh+dlmHQqOEwUtuLSb+T1H3xGYqFG+myFhmTxjjcZ8dObB8hcQS1j9wAwmI1o0CSxsiG9IKdQGlQkD5eZA4UMdraTWFpNYkMCjfHVH0fxW58onVBDEAQ6OzsLMsNTUFCobKRqfTpvUpPK6sbd8KljEDw6PCefbWcNZLgsBqvxaGH4uWfX/NwO3vtWV0VvPGiaFqdBnWeCy/M89vb2EA6Hsb29DaPRCLfbDa/XK/mmrRgz4QrSaQAcfbb9/f0YHR3F/fv3kUwmQdM03G43+vv7QdN0uS/xBHV1dVCpVKJYU4q9w2UI4kw4HBbb6bq7u6HT6cAwDILBIEiSfGAScDl4ffobd0Ks0Wq1Jzz2hGuVs/WpkPtSp9NhaGhIHIdeV1d35rGVxJg0KBFrEQhGduVAqjHadrsdzc3NCAQCOY2JkwKn+chc+OBQWnNh7rFhkDVucBsRSY5H6qTPXKiJLA6yNN583YKpOweIbF29CpvEIUCr9WjzkphdLL+QqXA5BHEkzGhUWeztprC8ksR6Rnrz73/5aQz/+0caQBdTt54Dp4MEwQzv9u3bGBwcLHtWUkFBoXiKaX3KZVLTXuzBHX2TIwsNxYJlT1Yu5zuQwWRUwWpVY20tidaqXczO7lb8COTjJrjd3d0wm83gOA47OzuIRCLY3d2F2WyG2+1GW1ub7N6HKhJgC3hMVconzPM8Dg4OEA6Hxaqj/f19NDc3l20seq5UV1eL47DLJSYdHh6KlTPAUUtYT0/PAy1hgtfSxMQEstksmpqaSn6tpxHWCYZhTog1NE1jaGgIgUAAmUwGXq9XNtGjGB8Z4TqDwSDS6TSam5tPrF2CF5dC8ShCTRGUuvXprKyNFJOaXC4XWJZFMBhEf39/ScrVqu1HfjUZCc2FSZUKro8+i/Cf/rk0x9NJX1GjAguAQyKjQnu7Ga5oAqG5q1d5kskCe6wWvg4SoXvlETMVLoKHw8RBR2VxsH+I5bUE7s/LP5EufsDgtVvbeNubnLL6dZ1VdiuY4Y2MjMDv9xc9erOSN1QKCo8Cer0+r2QZy7JIJBIIBAJgGAYOhwNerxcGgyGn+1lNcmi0JcGyR2uXkBwrZCADQRBw2Si8ddAEDW3BnTt3sLCwgJaWlryOU2q0Wi16enowPj4Oo9GIw8ND2Gw2eDwedHR0lDSLrlYRYOUwPZQRnuexv7+PSCSCra0t6PV6eDweNDU1gaIoZLNZBAKBko5FLxS32w2SJEWxphQJkMPDQ4TDYUQiERAEAY/Hc6Y4cxqVSgW/349QKIS5uTm0tLRUxDOcoihRrFGr1eKaMjAwgKmpKdy7dw86na4iPGpOI0yGm56efmB8u1JRIx2KUFME+WZzBHIdSXuW34wgzkg9qamqqgosy2JiYgJ+v1/2G0wwF16S2FyY/OUnQf6/L4BLFV/JQcrQ+kQAMKnTiGd1yLIETHYD3jJE4d9Hkni4wo3LYTkgvE+jv0eFwFS83JfzyGMzsjDQDFIHh1hZS2BzoTzVXK/8a1QUauQKlM4LEhwOB/r6+hAMBs807VNQUHh4yGWgQzKZRDQaRSwWA3C0NnR2duZkxMrzJ8Vrr+MQGjUBlYqSJAZ7zy84xDWwq6sLU1NTWFpaqoiM/2kYhkEsFkM4HEYqlYLD4cD29jb8fv+Zk2pKAaUCUMB8hlJvz3mex+7uLiKRCLa3t2EymeB2u9Hc3PxAQkGtVp/wrKmqKo0Bf6E4nc4TYo0cBseCmbLgOeN2u9HX15e3MESSJHp6enD37l3cu3cPHR0dFSfW0DQtJuB7e3tx9+5drK+vn9leVCxSiCkkScLn82F+fh6BQEA0mVY8aqRDEWqKoJCKGqH09rwv8PFJTcJGppisTT7U1taCZVlMTU2ht7dX9puAVpOosdNYk9Jc2GyG9alfwvb3/qHoQ6l08pRymtSHiGePjIp5EMiQWrzjzSr8bCSO5BXrFOIBrG6pMOg3Y3xi/9LJQArSYdFzMGkZHCYPsb6exJ2Fypg4Nj65h3A0DYctvzbNfLhojT1uhtfa2nrCtE9BQeHhwWAwYHd398S/CSa20WgU29vb0Gg0cDqd4kjhYDB44WbyZAz2+vqkU3NoqyagIqWLC47HWARBwOfzYWJiAmtra7JszPJFGGcciUSQyWTgdDrR0tIiViAdHBxgcnISfX19JRlxfJojQ+H8gwq+BGkxjuOwu7uLcDgstoR5PJ6cWsKEVp1gMAie5yu+Dcput6Ozs1PcqEsxUSmVSomVM8KUML/fX3SLleBbNzc3h1AohO7u7oqo/KAoSvSsoSgKarVavNaxsTGsrq6ipqZGUvFDqqoXYXz76uoqRkdHMTAwoFTUSIgi1BRBIWbCwtSB4zfbeWO0SZKUXZw5TUNDAxiGwfT0NLq7u2UXa4w6FZxmCrF96Zx1tb/6fkACoYaUqYzTSD34nYln1Hj8mhVz83Gsbl4935rlCAl/rwVToT0wV++/VxEYtRysegbZwzQ27idxb6EyW+o4HnjlX8P40PurZMu4XNZ7rdPpMDw8jPHxcRweHqKhoUGW61BQUDji1VdfxW//9m+DZVn8xm/8Bj71qU+d+Hk6ncZHPvIRjI6OwuFw4Hvf+96llSV6vR6bm5vIZDLY3NzE4eEhdnd3YTAY4HK50NjYeMJ373ScJXCe5x91bH3qrGagkjkUE7LowWAQKpWqLBt0wfcjGo2C4zi4XC50dHScufk2Go3o6ekRK7FLPfK4UEPhne0d1NicksfWHMdhe3sb4XAY+/v7sFqtcLvdBbWECWJNIBAAx3EVYYJ7EVarFd3d3eLo7kKEu2QyiXA4fGKEuxz+NwRBoK2tDYuLi5iYmEBvb29FjBoXTMgZ5mg/JIg1wnhxQQSRyvtFajGlvr4eGo0GIyMjMBgMFfGZXgUUuasICjETFipqOI4DwzDIZrNgWRY8z0OlUoGmaWi1WrFXsRyKpNfrhVqtxr1792T1kRCw6AAK0m0q+Y4OGK/1FH0cOcyEAcBEn/1/TWZJ1DeZMdh9NY1OV6MEOrus0GkejXJFudFpeNTZs6gyHgAHEcxNr2NkJIzg1C6iUlapycA//GgDyWRS1oqay46tVqtx7do1bG1tYXZ2Nu+17lEpu1VQKBaWZfH888/jlVdewfT0NF588UVMT0+feM03v/lN2Gw2zM3N4Xd/93fx+7//+xceM5VKYXJyEv/8z/+M4eFh/OM//iMcDgeGhobg8/ngdrsfGI5AEIR4n3Mch2w2K8ZggjhD0zQ0Gs3RJok8usfNOg51ttJkGAQvjfX1ddEkVW6SySSWlpZw+/ZthEIhsbpneHgYTU1NFwowJpNJrAQqpWcjAKjJwtZgWqPG5OTkpaPac4FlWUQiEUxNTeHmzZvY2tpCTU0Nrl+/jq6uLjgcjoKfc4IHSDgcxtraWtHXKjdmsxm9vb2YmJhAPJ5bu3sikcDi4iJu3ryJu3fviqPgBwcHUVdXJ6tJsdfrhdPpFD2rKoHjYk0mkxH3i4LwPDIygnQ6Lcm55Kh6cbvd6OzsRDQarYipyFcBpaKmCLRabV5fRI7jQBAE0um0WC1TyKQmuRHU5rt372J+fh6tra2Sn0MYpReJRMCyLJwuN9RaN7KsNJsf44eewcHoVFHHILXyPCA0ZBZH5boP/l9ZjoDaaMDbrlP4X7cSkCCOkB2KIsDkaAp9fwto9FqwuRHH7r5SWpMPWjUPu4kFn00jEkliafEQiw9pK9nuPvDyP4fga5dHlMy1P1oIhO/cuYOpqSn4fL6KWosVFK4Ct27dQmtrK5qbmwEAzz77LF5++WV0d3eLr3n55Zfx2c9+FgDwgQ98AL/1W7/1gI/VwcEBfvjDH+Lv//7vsbCwgLa2NrS1teGf/umfcrrfhaqZbPaoDfSySU3CmbursyilLiuINePj41CpVHA4HJIeX5g2FIlEEIvFxFHQhbaWmM1mdHV1iQMp5PApOYuj1qf8MRlN0HAWTE5Oii1x+cCyrOjXk0gk4HQ6UV9ff2JimFQI34WJiQnwPI/6+npJjy81RqMRfX19mJiYQHd3NywWywOvSSQSYuUMTdPweDwYHBwsy5Sguro6UBQleuxUwqQiQaxhWRYMw4gVwi6XCxRFYXR0FP39/UVXsMnlI2Oz2WC1WjE3NwetViubF+CjkixThJoiyGVxPz2pyel0YnFxEX19fSUZhV0oQm9kKBSSzNwumUyKJbUEQcDtdqO7uxs63ZFfS4bhjsyFJRAniLe/DWrnnyMb2yn4GKRGHqGGJhho1RwOs+cvkElOg7e9kcJIYB97B5W5G9dqSTQ2GsGDwMb9FA72c6viiO4BdrcJGvoA4VhlZDEqEUrFw2VmQXIZxKJJrCylsPQQCHe5MrNkQq07gng8LrkZZT6ZIoIg0NXVhaWlJYyPj8Pv91f02qyg8LCxvr5+YoNZV1eHmzdvnvsaiqJgsViwtbUFp9MpviYajWJxcRGf+9zn4PP58Nprr+G73/3upZ5/xwcyGI1GbG5uoqGh4dI1giAAl4mF21z6hVetVqO/v18Ua6xWa1HHO2vakNvtfqA9rFAsFgs6OztFsaYUE4AKbX0CATQ1NWFpaQlTU1Po6em59LuQzWYRi8UQiUSQSqXgdDrh9XphNBpl3zCeFmsqvVXXYDCIHjsdHR2w2WziGPJoNAqtVltWceY0VVVHbdhjY2Ml++5ehvB9ZFn2xB7S4XCIU9d6e3thNpsLPgfHcbJ9/seNm5ubmyt+glklo0SjEnOWGfDxrE19fT1YlsXdu3fh8/kqWhEkCALd3d2YmJjA6upq3ko+z/NIJBKiOKNWqy/M2tAUiVo7jdWYBG0bNA3rrz6F6F/8VcGHkEuooZCFRc/gcO/iKOMgo0JvjwlLCztYi1ROr6daTaKxUQ+7TQ0ewGaUhdV2lEHLVazZSwB6oxGN2hSW16Qp43zYURE89FQcOjWPgwMWK8tJrLGVKdJJwehECk/+gh2Tk5Po7OyE3W6X7Ngsy+YVgBAEAa/XK/ZXDwwMVESwpqCg8Dperxd/+Id/KP79rMmb53n+CQMZenp6EAwGQdM0ampqLjwfgaNqmnJB0zT8fj8CgQC6u7vz3pQJhraRSAQ7OzsXThuSAqvVira2NgQCAQwMDMjatgIUXlEjvOsysea0mbLL5UJLSwuMRmORV54/JEmir69PbNmqxMlgx9FqtWhra8PExARUKpX43RsaGqrIRIhQrSIka4QEcrngOA5bW1vY3NxEPB5HTU2NmIAym82iEFZM7CSn4S/HcdDr9RgaGkIgEEA6na54gbFSqby75SHkrMDgojHaTU1NmJ2dxczMDNrb2ytarBEeDoFAABRFXWpudzpro9Pp4Ha7c1bODVoVXBYK0b3iKy1U73sK+G9/g0JLdAiNPEqzGlkYNAwoUgWGu/gWzHIU6rxO1NWkcCNQ3n5PlQpoaDDA6aBB/rw3PJnixElO+Yo1yTSgpnRobyYxs1Da3vZKgCR4OEwcNKoM9nYPsbKaRDpzhUpmLiHL8Ji4R+JjHzyawCRl1uWiqU8XUVNTA5qmMTo6Cr/ff6EhYiWv2woKlURtbS1WV1fFv6+trT1gjiq8pq6uDgzDYG9v79KWH2Hy5lmVM0Jy7Kx1oK+vD2NjY1Cr1XC5XOce32NhoSlzlKzVatHX14dgMIje3t5LRQJhgxeJRE4Y2ra3t5ekrdNut6O1tVUUa+SsmCi4ouYYp8WabDaLaDSKcDgMlmUvNFMuNYLZ9NTUFBYXF+H1est9SScQWurC4TBisRh0Oh1aWlqwvr6OmpqaC++1SsBms6G7uxuBQCCne01qhHs3HA4jHo/D4XCgoaEBRqNR9NRSq9UgSRIGg0GcXllo7CSnUCO0awlegBMTE0in02htbVVipzxRhJoiiUQiODw8FG+eXMdot7a2Ynp6GktLSxW32J6GJMkT/dKnx9nyPC9mbba3t4vO2jhMaqQzPPZTxXmYEDXVML39jYj/y78X9H5SpmyQiuCgycRhoVhsZS7PkPE8AdB6vONNFH568wCZEncLEQRQV6eHx62B6lQGK5k6KSxYbVqABw7iuYk1WQbY4TTo7SIxeSch2TVXJjycZg5aVRYH+4dYXk1g4/DREWbO4sZYFv/pY1oMDQ1hfHwcmUxGkh78y6Y+XYTT6QRN0wgEAvD5fEW3HCgoPOoMDw9jdnYWi4uLqK2txUsvvYTvfve7J17z1FNP4a/+6q/whje8AT/4wQ/w9re//dKAPpvN4t69e6JAQ1FUTp5/gjfV2NgYKIo610Oh3CKNgF6vR29vLyYnJ8+crsQwjLjBSyQScDgcqK2tLcnkzrNwOBzgOA6BQEBW3w+SIKAiATbPx+jpT6Sqqgo7Ozv4yU9+AoPB8EBbfiUhtJSEQiEsLCyIvk/lgud5xONxhMNhsaXO4/HA6/WK8b/H4xGnV1V6C4xgiDw5OXmux46UnCXOnOV3RBCEOL6bpmmQJAmt9vXYKZvNoq6uLu9zyynUCL9/YQ959+5dSb0AHxXBRyWYt53DhT9UAGZmZvDyyy/j6aefBkVRoknwZRAEAafTidXVVTAMU1SfYSkQjKzu3r0LrVYLrVaLra0tLC0tYX5+HtlsFi6XC21tbfB4PDAYDEXdiAYdiYNDNu8H8GmyWgrpf/5xQe91ve8XoKmXPgOQBg2aTyPO6BBncs/SZDgVWhtpJBIMEqnStMTU1OjQ1maC1aIWq2gEsgyP/TP8c7Q6ChzHI5PJTWjjeSB+qEKnV43NSGVPK8oXm5GFy5AFxSYQvb+NtdV9bNxPIradydmA+Sqzf8DjjUM2uJxaVFVVYXFxEclkEjabraiHcCwWg8FgKDgLqtFo4HQ6MTk5Ca1We2ZljSDKXxE+V+4LUDiTz5b7AqSAJEm0tbXhueeew1e/+lU899xzeOaZZ/CZz3wG8XgcHR0d6Ovrw3e+8x18+tOfRiAQwDe+8Y1LTSh1Oh2+//3vQ61WY3BwMK97UqVSwel0IhQKwWKxVHyrI03TsFgsmJqagtPpBM/zCIfDmJ+fx8rKCiiKQl1dHZqbm+F0OqHVasu6PhkMBlAUhXv37sHj8ci2IdxNcnnHiXoNARWfxvr6OmZnZxGLxeByucTnRWtrq+xtW8UgeDxubGwgHo8X/bzMF6FyfmVlBbOzs0ilUrDb7Whra0N1dTWMRuOJ37dKpYLH48HMzAwASO5JJzU0TcPpdGJqagp6vV5ywU4QZxYXF7GwsADgqJq3tbX13HtXKAAQKgePD6SpqqrCwsICUqlUXsa90WgUJpNJFkFSsMsQ/h/CnjeZTGJxcRFut7voNaESW+iK4NwYjLhkJKmyk8iBT33qU0gkEvjCF76Q92LJsiwCgQDq6uoqXmlmWRabm5uYnZ0FRVFwuVxwu92wWq2yPCSyPzcXLkas4XkO4Xc9A34tnPd7O7/5WZiG2go/+ZnXA8QoD8jDJCKHZswnLm4lOwu1isdONIHQnHyihsutRW2NFhr6/IV0d5+9UDDa3T7MubJGoNHNY3xiDyWYCi8LFj0HkzaLw0QaaxsJ7ErQwnfV+ZV3efB//sZRVSHP8wiFQiBJEl1dXQWvK9PT06iuri562kAmk8H4+Dhqa2sfyFYJVZRXhCujOF0xHtKVsHQkk0m8+93vxic/+Uk8+eSTBb0/GAyir6/vwlbHSiCdTmN5eRlra2ti5Yfb7a7o697c3MT6+jr6+/tl8cVZimaROMzvNskchMGnouLnd1yUWVxcxMHBAXp6eipeiOd5HtPT06BpWvaWEkGcESpnhMp5h8OR8++VZVkEg0G43e68qz/KQTqdRiAQQHNzc9FtWxzHYXt7G+FwGPv7+7Db7fB4PLBYLHn/3hiGAc/zUKvV4mfPcRxCoRDUajU6OjpyOmYoFEJdXZ0sVUM/+9nP8MY3vvHMn21sbGB1dbUoHyuCICpaTC2Ac39hilAjATzP4xOf+AS8Xi9+7/d+L+/3MwyDsbExtLS0SD6GsVjOcrq3WCyYm5uDz+eTvRIomWaxEi1OkMh+9yXE/uTP835f99/8CQx9TUWd+zQ7hA2MWg8ukUQyo0YoXpi5FgEeNJ/GT28nJb1JbXYa9XV66LQXb0B5nsf9KHupoFKIWFPvAkLTu8g+BBqHUcvBqmeQPUxjfeOoUkYhP0xGFb7/wjXQPxcFeZ7H3NwcEokE+vr6ChJDpqamUF9fL0kAIgSXZrMZLS0tYgBE03TFB/J5cGX+I1cMJQbLgZ2dHbzzne/E5z//+XM3BxcRj8cRCoVKOlo6V1KpFCKRCCKRiFhJoVarxY1OJUzNuYyNjQ1sbm7C7/dLLtasbzHYTeaXzbMZgBr7+Zu8hYUFJBKJh0asuXPnDiiKQltbm6TXy/M89vb2EA6HRVsDj8cDh8NRcJKCZVlMTk7CZrOhsbFRsmuVi2w2KybTL/PoPI2U4sxpGIYBx3GgaVq8p3iex8zMDNLpdE6TzCYmJuD1emWpcLpIqAGOqp5nZmYKHjOuCDWvowQJOcIwDJ555hm8613vwq//+q/n/f5MJoOxsTF0dXXJ3hOZy7UIgQHDMHA6nXC73SeMtVKpVMkMt7YPGER2i5i8cHCA8Nt+BVwyP8Na30v/Bfou6VT/FKdBnHaAIAlkE2kQLIvR3Zaijmmis/jZyAGSeWaUTmM2UWhoNMCgzy2ISqY47OznFhwVItbYjRmsLB8gnamsigWdhofdwIDNpHH/fhLhqDKxSgr+4Hda8fY3O0/828rKCsLhMAYGBvIucQ0Gg5JO5xAylzzPo7u7GyRJKkKNQilQYrAcWV9fx3ve8x688MIL6Onpyfv9Ozs7mJmZqYiRwQcHB4hEIojFYqAoCm63Gy6X60R7ViQSwfLystjyVemsra0hGo3C7/dLWokY3mUQi+cn1DhNJDzWi58pCwsLSCaTFT+dFTh6Pt27dw8EQRQ9oETwnAyHw9jZ2YHZbBYrZ6T6vXEch6mpKRiNRni93or/fBmGwcTEBFwu16UeehzHYWdnB5ubm9jf34fNZkNVVZUk4sxphNHdarX6RIy0uLiInZ2dS4XRQCCAtrY2WSryLhNqAGBvbw9TU1MFjRl/lIQaxaNGIkiSxPve9z58+tOfhtlsRkdHR17vV6lUcDgcotJc6i9gKpXC/fv3MTs7i83NTej1ejQ1NaGpqenM61Gr1bDZbJicnITD4ZA1sNHRJLIMj3S2wJiVpkHHIkhN3s3rbZ4PvhtquzQbPQ4kaFcj9g85kARAgoeBSGGbc/y8cqSwBTzDqtBUpwGbzZ7pF3MZOr0Kba0m1NZoxYqGXNiN5+4flK9nDQCkMv8/e28e5chZ3vt/S7vUau0qqfd91l7HCxgcm802ZvA2xrnG+RHisAZuAuGSjSRcO+CTEw44JtxgwPlhwAZiG3uyEBzAjm9Ogjc80+p9enpVqxetrda+1Hb/aFdNq1eV9p6pzzkcjkdV0qvq0lvP+32f5/vI0eDUgGNopIoUoYpBreTgNNHQK9JIRSLwLK5jdTUOry+NRLI4s2uJSyQSNG5+R25qsdFohFwux8TEBEiSFCXWrK2twW63l2xeIggCdrsd8Xgcbrdb2NWu9QBTBJJHTW3yQLUHcFgwGAx417vehQ996EO46aabRJuAa7VaqNVqXLhwAU6ns6Jljbwhq8fjwcWLFxGLxWAymdDV1YXm5mYYDIYd8x/vAzg7O1sSv4dyYzAYQFGUMH+Wau7MUBziImMEnYaA/oCsYbPZjFgshtXVVdjt9pqe6wmCgNVqRSgUQjAYhNVqFTVejuMQDofhdrsxOzuLbDYLkiRzPCdL+f35zLC1tTVEIhFYLJaavr4ymQwOhwNut3tXHxg+c4b37GRZFg0NDeju7obdbi+bXxTvU0PTm+nnvChjNpvBMAwuXrwIkiT3FGtWV1eFWKbU8B41+6HRaGCz2TA6OgqtVisqs4b357mMkDxqKsXGxgZuueUWfOlLX8L1118v+vx4PI6xsTEMDg6W3XE+kUjA7/cjEAgI3ZxIkhRlqheNRjE5OVn2lGGW47AUyCCdLeyWJObmsHr7h0Sd0/8vj0DdVhoz4YyuAVbSjClPBhp5GtkMCwsbwouBfsgIDnVqDmolBwJAlgYSGRloNv+JXS7jkIkmcW4yvwwPlVqG9rY6mIwK0Q8QiubgD4kXKQrJrDHogHQsgTV/ERlVIlAqONj0DGRsBoFgCssrKbDSLFh2CAJ48u8H4SR3ziGhUAjT09OiUmTPnTuH3t7eshiELi8vY2VlBW9961svpx2d2o2Sr2yk2Uckb7zxBj7+8Y/jueee29GhMh/4Mp3BwcGyih/bu2Xq9XqQJAmbzSZqAbK8vIxgMFhwmWilWVxcRDQazas0Yz94ccEXSoDViPN3tBtkII35Cf9zc3NIpVKHJrNmZmYGNE0f6PHGsqyQObOxsQGj0QiHwwGz2Vyx+4jjOFy4cAEEQeTtq1JNWJbF5OQk1Go1Ojs7hesXiURgNpvhcDjK5tl50LhomoZcLhea2gCbWXdzc3MYGhradX32xhtvoK+vryxxUj4ZNTyZTAbDw8NobW1FY2NjXufwrb8vI6TSp0qytraG973vfXj00UfR398v+vxIJIKpqSmcOnWqpAsBfteGT6nVaDRCSm0xN/zGxgYuXLhQ8vFuh2I4LPrSBZsLx3/3U4i95sr7+P6ffQPqJkthH7aFtEwPa0srCAJYj7OIRGNgaA4ks4af+U5h998nB52Kg1bFgaUpJNJZEAo9MvT+D1CdPIP/fC0Bdo9rpFAQaGurg8W8s4tTvkRiDOLJwqaG8HoKiZg40UWrBpRsCgtLpS8zUsg4GHUZUMkIslk5llfSoBlp2qsGH7qnCb/zP3bfgYlGoxgbG8s7Rfb1118vq39DJBKp+V1WkVw2X+Qvm6SVAAAgAElEQVQyQ5qMCuCFF17AX/7lX+Ls2bMF+egtLS1hY2MDfX19Jf2N8zvvfr8fkUgERqMRJEnCYrEUtTheXFxELBY7FJ4qwGZpRiKREC1+8GUl/OLYaDTCbGvAekZc6YbdIAdpzF8Mm5ubQzqdrlq7czFwHIe5uTlkMpkd491+/UwmkyAuVEvkEyMuVRv++k1PTyOTycDpdAriVrXHzbKs0A1qq1izvr6OCxcuYGBgYEeJU7niJJZl8dprr+G6667L+xyapjEyMgKLxYKOjo4Dj5eEmktIQUKBzM7O4u6778aPfvSjvG667YRCIczNzeHUqVNFtSDbbgbG79pYrdaStjbjx1tuc7tkhoEnkC3oxuReeBHez/xl3scP/uJRKB3FmSXTkEPX0AWV+tK1nnQnoVVkYcmu4cVAP7Js/tdLpWBRp+agkAEsC6QpAoksga2/cb2KwbmRKDZil66STAa0tNbBblNBXqBAA2zeT94AU1SWSSFijUIOmLUZXJgR5zO0HRnBwWZgoSKyiGyk4fYkkC20pE6ipJA2FX74zaE9BcRkMgmXy4WjR48eaLr+yiuv4C1veUtZg89ab+crktqNjq9spMmpQJ5++ml85zvfwTPPPFPQb3V2dhY0TRe9088wDEKhEPx+P2KxGCwWS1m6ZfIlK7W+2OXJd7x8K2Ofzye0oiZJUlgcZ2kOM2vi4gnSIIddhFCzn/hRi3AcJ7RrPn78OMLhsCAObr9+tQA/Xt4TqJYyw4TMrTczj0wmE0iSRDgcRjKZLDozrJTsJdbwG129vb05HqivvvoqrrnmmpKXENE0jXPnzuEtb3mL6PGPj49DpVIdOO9KQs0lpCChCM6fP4+PfvSjePbZZwtqve3z+bC8vCy6rSGv+vr9fmxsbJTFDGw3eHO7Qsw/xbARp+EtwFyYo2iEbjkDyhfM6/ijP30EhpbiSp8oQzPMllyxZ8FHA0wCumQQ5zY6EaWLM/JSyDZLp1QKDhwHZGgCFA2sLMcxu0SjuVkHh0MNhbz4h3IyzSIcKaJf+puEQykk4uL+hgQBNJoojE4mRJy1Kcxo5BTi0U1hJpUufvwS5eFv/uIYrh7c21uCT5Ftb2+H0+nc87iXX34Z1113XdkC0SvJyE6iqkgxWBF885vfxIsvvojvfe97ohcjfFmGSqVCV5c40//t3TKtVitIkoTBYCjbnMR3fCEIouTdf8oB392PYZgdizKGYYTrF4/Hheu3myEry3GYWhYp1BjlsBvE3w+HRazhxS2+bMvpdMLpdFalLEcMi4uLiEQi6Ovrq6r4sZs4s1vmjNvtxvr6Ovr7+2vGL4UXawiCgFKpFK7jbhtd5YqTMpkMxsbGcPXVV4s+l8+wSqVS+94HV5JQI5kJl5GGhgb09PTgU5/6FO68807RHi56vR4sy2JxcfFA8zX+wbawsID5+XmwLAuHw4Hu7u6ymIHtxlZzO4fDUbaJVqOSgWY4pEVmQhByGdSZJBKvDed1fONH7oBCV/iueUphhNWxU+ip1xJYj2VBUBSitA4JpjhvH5YjkKEJJLMyJLMyZGkCLLeZodDg1EBfX3iZ03Y2omzBpWdb0eqUYBgWVFbcm8XSchzrUMHr37sMyqJnYdVnoaATCKytY9kTxeraZutsmpbWPbUMTbO48bq9s2UUCgWcTiemp6fBMMyeHfI8Hg9aW1vLNcwryshOoqo8UO0BHGauueYaTE1N4Sc/+Qne+973ioqBCIKAzWbD8vIyMpnMgd04s9ks1tbWMDs7i9XVVahUKrS0tKCjowNWq7VshqJbx2u1WuHz+RCPx3cYntYaBEHAYrFgfX0doVAIRqMRfr8f8/PzcLvdkMvlaGpqQldXF2w2257XjyAIrMcZ7L/nnEudRoY6tbj4lCAImM1mRKNReL3emit9ZVk2Zw0AAO3t7VCr1Uin02htba2ZzI+9MJlMoCgKCwsLFTfI5sWZxcVFIZvO6XSiu7sbJElCq9Xu+HubTCawLIuZmRnY7faaiAkIggBBEGBZFizLCobDSqUSJElifHwcSqUSer2+bHESRVEIhUKi25nz47darchkMvuuJWUyWc3fzyKRzISrybPPPou///u/x09+8pOCDHcXFhaQTCZ3qPg0TQu7DslksiK7NvlQCXM7juOwFMgiJXKhj2AA3nffDe5Nl/T9OPXf34W8vjChhoIS+uZOKBW7T9xTngzUqRC8KQOWUuIND6tBnZrF9d0ppCggnQHSWQ7hSArrGwlEYmnIFVoo1XWQKbSgWQI0A1AMkMkySKayUKq04ECAYTeFJI4DuGwGFA3QDAeK4pClOGRp7Omxw9NGcnCNRcCygFHHol5DIZPMYGk5gUj04L+tRG2iVBB46rFTMNbvv1PCsixGR0dRV1eH7u7uHfOdGCO7QpAyaiQqhBSDFQnHcfj0pz8Ns9mMP//zPxd9PsuycLlcaGho2LHwSKfT8Pv98Pv94DgOdrsdJEmK6l5SajiOw9jYGIxGI9ra2qo2jnygKEoQZ1iWRXNzc0Ebi7NrFDIiNmEcRjlsIjNqePjMmlooM9teVrdX5pHb7RY8lw7D4nZlZQVerxcDAwNlzc7f3oq8UENln88Ht9uNwcHBmokLeKGG4zioVCrh+1AUheHhYTQ0NMDj8ZQlTorH45ibm8PAwEBR7+P1erGwsLCrGTJvnHwZIZU+VZtvf/vbeP755/HEE08UlII7MzMDYFMhDwQC8Pv9yGazsNlsIEkSer2+ptT9Spjb0QyHRX9GtPlr7HN/gvjP/+vA46565XuQ6cSn1nEcwFnaYDDuXdIUjrMIrgURyyhxMd4k+jOqwUBLFscaqRzPo/r6ejgcjgPL6niD7IGBgZxuZhwHvDAqx+hi7m9CRnBQyAG5bLOjlVyGzbbmBEAQHGQAQGfw/NkJBNfFdZKSqG0+fX8bzpw+eCeG4zhMTU2BZdkdppSSUCOa2nl4SGxFisFKAMMw+OAHP4jrrrsOn/jEJwo6//z582hvb4dOp9vRLZNvwVsrsCyLkZERkCSJpqbaii+y2awgbtE0DbvdDrvdDrfbDY1GI7rMDAAW/RQSmcoINcClsi2Koiou1uxWFuZwOA7coF1aWhLKdA6DWOP1euHxeDA4OFjSEpdSiTPbCYVCmJ2dxcDAQM3MBXuJNQzDYHR0FBsbG3jHO95R8vs3Go3C7Xajr6+v6PdaX18X1g56vV74d0mouYQUJJSQL33pS1hYWMA3vvENUT8MftdmYWEBMpkMLS0tVd+1yQe+nrecD7JUlsWSPyPqRs288grWP/q/Djzuqtd+AJlG/MM8pbTA3rS3fwbP9FwYTIbCaLRd9GdUGgIcThjGEI8ECvY84sWa/v7+nHuX44AXRuQYdYu/1smVJfzXf3pEnydRu3S26fDY1/LrlsebEEaj0Zw68XILNVdSfbREVZFisBKRyWRw22234UMf+hDuvvvuvM/jOA7xeBxra2vweDwwGAxoamoqultmuWEYBi6XC01NTfv6eVUCPoYNBAJgWRZ2ux0Oh2Pbpg2H8fFx6PV60Q04lkM0Isn8s6udJjms9cWVqVRSrOHFGZ/Ph0QiAZvNBofDgfr6elGf6/F4hGz3WijTOQi/34/FxcWiM1X4piper7ek4sx2+A64fX19OzosVROapsFxHJRKpfB3Z1kWL730EhobG3Hs2LGS3r/hcBhra2s4ceJESd4vFothdHQUJ0+ehMm06WF4JQk1kkdNBbnhhhvw4osv4pVXXsGNN96477HJZBIrKyuYmZlBMBiEXq9Hd3c34vE49Hr9gV1PagGz2YxwOIxQKASr1VqWB5lSTkApJxAXYRCraGkB8+JLoEPhfY9r+uQZECINeLNQw9zUnJcnTDrLgcjGsZKu/b+lURHG0SYZenp6CvY80mg0MBgMGBsbg8ViER68BAF0OjjE04A/Iu6habTVY2XOj2yWEXWeRO0SjlC47ioTrJaDAzPe54CmaVy8eBEkSUIul8Pj8aClZfdW36VA8qiRqBAPVHsAlwsKhQJ33nknPvOZz6C5uXlfMYBf2C0tLWFmZgaJRAJWqxUdHR3w+Xxobm7OERlqEZlMBpIkMT09DZVKVfGFYyqVwurqKmZmZhAIBFBXV4eOjg60tbXBZDLtELkIgoDdbofH40EmkxEWZPmQzLJIZfPXNPUaGXQiPWq2wz97+Axjm81W0hiXpmn4/X7Mzc3B7XZDoVCgubkZnZ2dsFqtUKvVoj/PaDSCYRjMz89X3AOmEOrq6qBWqzE5OQm73S5qYc7/ht1uN2ZmZpDNZkGSJHp6eoSN7lKvSTQaDYxGI8bGxmAymWqmMyT/d6ZpWvB2IQgCa2tr0Ov1WFlZOdAHVQzJZBKpVAo2m60k76dWq2G32zE+Pg61Wo26ujrI5fKav39FsmcMJgk1FYQgCLz3ve/Fd7/7XayuruKaa64RXuN3bZaXlzEzMyN0a+rs7ERLSwuMRiOUSiXsdjvm5uYgl8tz0sBqEd4UamtbynKgUcnAsBzSIh7UKgWQeOnlfY9p/MTdIESY8LIcAYW9FRpNfrtsajkDJr6BxWTte9S89QgBh2WnmZpY1Go1TCZTycQaiiHQe8KIiVFfUeOSqC3kcgJvvSp/M0yj0QiFQoGJiQnBUFMSakQhCTW1yQPVHsDlhFqtxu23346PfOQjGBgYQGNjo/Aay7JYX1+H2+0WWkfbbDZ0d3ejoaEBer0eKpUKZrMZY2NjsFqtNZ1RA1wSayYnJ6HT6couLiUSCWGDkS+N7urqEmLYgxbaBEGAJEm43W5QFHWggTNPhuIQT1dWqAEuiTUbGxsIBAJFizU0TcPn82F+fh5LS0tQKBRoaWkpSpzZjsFgELKBytn0o1TodDrodDrh2b7fb243ccZutwsbjOUQZ7ajVqthsVgwNjaG+vr6mimD2k2s4bsKp9Ppkop3iUQC2Wy2pAkFSqUSDocDFy5cAMdxMJlMNX/vikQyE64lstksbr/9dtxzzz1wOp147rnncM8998BkMoEkSdhstn0faLwZVHd3d9nEj1JSTnO7TCaDQCAAn88HtakVCnWe4lUyAf+77gQT27vVc/+rj0OtzT/dMq0hYXPmryCzLIvY0gye9w7lfU41qFOzeP9gGqV8vsXjcYyNjaGvry9HcOQ44JcjcoyJKoPiEJiew/lzklhzuVCnk+OZx66CWmQwvb6+jsnJSchkMqn0SRxS6VNtIsVgZWBxcRF33HEHvvWtb2FiYgKTk5M4ffo0zGYzSJI8cBEQjUYxOTmJoaGhmtk1349MJoPh4WEcO3ZMVKbKQXAch0QiAZ/Ph2AwCJVKBYfDUXRZmFiPnUiSwXIo/6zaBpMcliJLn7bC+0gyDCO6jIRv5e7z+ZDJZISypnJvxK6trWF1dbXshr2lgi8r2lk6z+X4JhoMBjgcDlgslqou5NPpNFwuF3p6emqqAoJlWdA0DblcjjfeeEOIk1ZWVrC8vIyhoaGi/fe8Xi8SiURBflMHwfvrNDU1lXUzrgpIpU+1Ak3T+O///m+srKzg4YcfxuLiIm699VbcdNNNaGpqgl6vP3BykcvlQhpYLSm2e8GntC4uLoJhGBgMhqLeL51OCym1Pp8POp0O7e3tcFgNiKUYsPmEtkoVVOtBJEcm9zyk4RN3Qy7Pb6LPEFpYGhtFt//MRkKYi9vxpj1uTXKskQJpKEFP7i2oVCph12FriiifWRNLicmsIeBsNmJ2chWMSGNpidqEoji0NmvR2SbOh0ur1aKurg5LS0tCS9dycCW1hpSoKg9UewCXG9FoFC+99BJmZ2fxta99DTKZDHfccQeuv/562O32XdvwbketVkOn02Fqakoot6xlFAoFrFYrJiYmYDQaixKXOI5DLBaDx+PBzMwMYrEYTCYTuru70dTUhPr6+qKvB59ZMzs7C4IgUF9fv+/xDAtsJPKPUeq1MmhVpZu/t7YazyezhqIo+Hw+zM7OYnl5WWjl3tnZmZNpXE7q6+shk8mEkuFaf55tLytKp9NCaWImk6l45sxBKBQKkCSJqakpKJXKmvGs4dt3J5NJBINBQewwGAxQqVSYmJgQXWa2nWg0KmS9lBqZTAan01mSeabGkDJqaoGvfOUrePLJJ3H99dfjzJkzOHnyJN7//vfj4YcfzimDypdUKgWXy4X+/v6amQT2oxhzu1QqBZ/Ph0AgIKTz7tZpIZ1l4Q5ksP9t/SaLi1g7fd+eL18z8iPk8xNgIYPS0QmtiOwbno2lOcylWzGznl+Kb6UhCA63D6VQwFfLi0QigdHRUfT29uYEYxwH/MIlx/hS/hNxHRPGc0/tLbxJHC4Gew342gPizehSqRTGx8dBURSOHDlSsjrprVxJRnYSVUWKwUoEx3G4++674fF4cPr0aZw5cwaRSASf+9zncPbs2YKyk/1+P5aWljA0NHQoFg3JZBIjIyOiY0aO4xCNRuHz+RAKhVBXVweHwwGbzVbW780wDIaHh9HS0gKHw7HncRmKw6yXyvt9G8xyWPSlHzfHcbh48SI4jsPRo0dzxAK+FTnfsZUkSZAkWfXY3efzYWlpqeTdlUoNfw96PB74fD5YLBY0NzeLbmpRaSiKwsjICBobG3NKLas1Fr/fD5/PB5qmhbWYQqEQrmE4HMbk5OSOLkti8Hg84DgOra2tpRx+Dkqlsqb/7gUgdX2qBXw+H+x2e87NtbS0hNtuuw2PP/44jh07Jvo94/E4xsfHMTg4WPOZNcBmRtHw8DDa29tht9v3PTaRSAidAnh1miTJA3cbokkaq+v5PbTjH/t9xF4+t+tr14z8GMDBuzQZXQOsZP5+GlvZWHEjTtTjv5eaCzq/3DSZafzG0fK2v04mkxgdHcWJEydysq04Dvi5S46JPMUaGcFh7o0pXJze3yRa4nBAEMAP/s8gGh3i5rV4PI65uTkcP34c58+fR2tra8kDJEmokagQUgxWQlZXV3fMBf/6r/+Kr3zlKzh79mxBnTRXV1fh8/kwMDBwKBYOfNnx4ODgvp4129sY19fXC90eKylK8TFjW1sbSHJ3Pz+W5TC1kr9Q02iWw1wGoQbIFWs6OjqEsia+FTmf8VFL+P1+uN3umhNrtguE9fX1QrewsbExnDhxIm8fo2rCl+tYrdayihe7QdM0AoEAvF4vKIrKuQdZlgXDMJDJZDliDd9lqbe3t6Dru7i4KJhflwtJqLmEFCRUgMnJSdx33314+umnC7qx+drNU6dOVSRlslh4j52urq6c2k3eUJkXZzQajZA5I/bh4d+gsB6nDz7wpf+Ltf/5hV1fumb0RzgoNSct08Pa0lqwf0vEu4oEReBcsA2JTO0t/G44mkajubRlT7uRSqUwMjKC48eP5zwYOA74+bAcE578giqTlsYz330tv/I3iZrn/7u7Cfd/UFwdcjQahdvtRl9fH2iahsvlgs1mQ3t7e8nGJQk1EhVCmskqwPe//308/fTT+NGPflTQQtXtdiMWi+HkyZNVL7nIh0gkgqmpqR0eOyzLIhwOw+/3Y2NjA0ajESRJVt3vg48ZOzo69tzgm1rO5v3cL6dQk81m4fP5sLi4CJZl0draCqfTWfNdwgKBABYWFopuhV0sfGkd73uk1+vhcDh2CIR8zHj06FGYzYVtlFYSlmUxPj6Ouro6dHZ2lr2dO+/dmUqlQJKk0K11t3ExDAOCIHLEj2QyCZfLVVBW8vz8PLRaLRoaGkryfXZDEmouIQUJFeLll1/GH/zBH+Ds2bMFGU8Fg0EsLCxgaGjoUCwgstkszp8/j6NHj0IulwuKuVarFVJqi/keHMdhOZhFIrO/yMDRNNbf+wFk1/w7Xjuo9ImGHLqGLqjURdRyBgNIpjIIw4GRldraGdCpWNw2VFoT4f3gH7zbDQ/FijWKhB//9s8z5RqmRAWxW1X40aNDebW75wmHw1hbW8OJE5tlUyzLYmxsDBqNBkeOHClJgKRQKA5FqYMIan91eWUixWAV4qtf/SqGh4fx7W9/u6AFAG8mu73kpVYJh8O4ePEiBgYGEIvF4Pf7EY1G8zZUrjR7bfDxzKxlkc1jbw4AmixymOpKN39nMhmhrIllWdjtdpAkiaWlJRAEUbLnTrkJBoOYn5+vuFizVZzZWlp3UPZWJpOBy+VCd3d3TRn27gXHcZiamoJcLi/5PcEwjJC9lUwmYbPZ4HQ68ypf2kus4U3I29vbRdlVzM7OwmAw7JkBVwquJKFGMhOuEVpaWtDc3IzPfvazOHPmjOhJUqfTQS6XCy33avmhwGfOZLNZzMzMIJVKCUZgfBvMYn+ABEFAr5Vvmgvvo9UQMhnUVBqJV3aWPzX93pl9P4MxNEFfX9wuCU1RoNMpEColljd0qKX10tEGCg5j+bNpeJRKJWw2m7DrwO9AEQTQ5eQQSQKB6MH3hVyrQ2I9jFi0vCVbEuUnmWJw4kg9mhryL39KJpNIpVLCLhBBEHA4HEL6r91uL3p+lMyEJSrEA9UewJXC2972Nrz22mv4xS9+gXe/+92i5wiLxYJAIIB4PF7zO/wMwyAejyMWi2FmZgZyuRyNjY3o7u7O21C50vBNNCYnJ3PiA55oigWVZ+Mng1YGTZFmwplMBmtrazlNLTo6OtDW1gaTyQSlUgmr1YpgMIj19XVYrdaau6bb0el00Gg0mJycLNpQ9iB4cYY3BE4mk7Bareju7obT6URdXd2Bz9haNezdC4IgYLPZsL6+Dp/PV3Q7d5ZlhU36hYUFyOVyNDc3C+3c811H8gbDW0uhCIKAQqGA0+nE9PQ0GIbJuwwqEAigrq6urCV+CoWi5n9PIpHMhA8LTz75JJ588kk89dRTBaXgejwerK+vo7+/v6ZuYo7jhJTacDgsqK18rWm5DJEzFAu3P7N/SmwoBO+77wJHbdmOIQhc43pyz1NiXB0aOopvNZ5JJhEL+hBTWOBNmOAO1UZ6LAEOtw2loVNXfgrgd0l6enpyDB45Dvj3YTkm88isMWsy+MfvvlHOYUpUiBveasH//vyRvI/3+/2IRCLo6enJ+XeO47CwsIBwOIzBwcGiMmKkjBqJCiHFYBWE4zjcf//96OnpwR/+4R8WdP7o6CgsFkvNtY6laRrBYBB+v19YFJMkiVQqJbTlPQxz2l6txpdDNCLJ/DaWCs2oSafTQuYMAME3cT9/SI7jMD09fagya9bX1zEzM4PBwcGStp/nN2n5siadTlcSU2qKouByudDS0iK6UUk14DgOi4uLiEaj6OvrE7Xpw7KsIPREo1FYLBY4nU4YDIai7y2WZcGyLDiOg0qlEsbFsixGRkZQX1+Prq6uAz9nampKaJFeLlQq1aH4LYlAKn06TPzt3/4tXn/9dTz22GMF7drOz88jnU7j+PHjVb2R+Xpnn8+HSCQCo9EIh8MBs9mc873yNbcrlFiKwUpo/+yK9Bf+N8L//Evhv+V6HU796rFdj6WgxFIkDYvFVHQwxtAMoisL2FDaQciUeHWx9B1qCqHRxOCGY5mqff5eKa0cB/z7eTkmlw9+qFOBFfzHLxfLOEqJSqBUEHjqO6dgNOQnXHu9XiQSCXR1de36+vLyMlZWVjA0NFRwerck1EhUCCkGqzA0TeOuu+7C6dOn8du//duiz2dZFi6XC42NjVVfNFIUhUAgAL/fj3Q6DZvNBofDAb1enxMbrq6uwuv1YnBw8FBkCqbTabhcrhxPO+8GjVAsP6Gm2SqHUZff/J1Op+Hz+eD3+4XsTJIkRQkYh1GsCYfDmJ6eLrpRSbnEme3QNI2RkRE0NDRUvbtSvng8HgQCAQwMDOx7LfiNbq/Xi0gkArPZDKfTCaPRWJZ7iabpHWINx3GYnJwEQRAHri3Hx8fR0tJSVqPnK0mokUqfapDrrrsO586dw89+9jO85z3vEX0zmkwmhMNhhMPhitdtbk3Fm5+fB8uycDgc6OnpAUmS0Ol0O76PSqWCwWDA+Ph40d40u6FWysBxQCq790NcYbMi/uxPL/232QDnh27dcRzHAYSlBS0tjVhYWADDMDmdisQik8mQjoSQltcBHItIWossU/1AabCVgkFbvTUCn9I6OTkJjUYjpFASBNDdwGEjAQQPKIPSGnTwun3IHOBTJFHbsCxgNatw4kj9wQdj00yY47ic3datGAwGqFQqjI+PF5zeLZfLL7cgQSp9qk0eqPYArjRkMhnuvPNO/Nmf/RlMJhOOHMk/mw/YLCOw2+2Ynp6GWq2ueIefbDYLr9eL2dlZrKysQK1Wo7W1FZ2dnbBYLFCr1Tvmrvr6elAUBbfbDZIka35uUygUsFqtGB8fh9FohFqtRjrLIZHJL2Yx6OTQKPf+jqlUCisrK5iZmREMbTs7O9Ha2gqj0Sj6mUEQBKxWKwKBAMLhMCwWS81fY61WC71ej/HxcVitVlEZ/rw44/F4MDMzg0QiAYvFUlJ7g+3IZDI4HA7Mzc2BpulD0Q2KH+P09DRIkswRa/iua263G7Ozs2AYBk6nEz09PbDb7dBoNGW7h2QyGTiOA03TQpk3P69FIhF4PJ595wmv1yvMNeXiSorBpIyaGoXjOHz0ox9FW1sbPv/5zxd0/sTEBPR6fUm7newGb2Ll9/sRj8eFlFqxai9vblfMTvdecByHlVAW8fTei/bIPb+N5OQsAEDT0YS+f/rKjmNSSgvsTZu7ZAzDYGRkBE6nsygFf2NpDmG5FSAIBCMM5iLVbdWtfdNEWIR/a9nIZrNwuVw7uj1wHPD8eTmmDsisMSjieOYHI+UepkSZaW/R4v//24G8jvV4POA47sA2mOFwGJOTkxgYGMjLcG8rV5KRnURVkWKwKhEOh3HLLbfgoYcewtvf/nbR5/MNE7aX6JSDrWa2DMMIJTliRaKFhQUkEolD070qmUxiZGRks8MfocXKen4mNS1WBQy63Pk7mUwK11AulwuZM6WMRTmOw4ULFyCXy9HT03MorjHfIWxgYODAdgfO8LEAACAASURBVO6JRAJerxfBYBBarRZOp7PkmTMHwbIsRkdHYTKZyr72KRV+vx+Li4vo7+8XMrjW19dhMBjgdDp3VCFUCoZhwDAMVCpVzt/Q7XYjEAhgcHBwV9FyeHgYR48eLatIXU4RqEpIGTWHDYIgcPr0afzd3/0dYrEYBgcHRZ9vs9mEFoHFZH3sBk3T8Pl8mJ+fh9vthlwuR1NTE7q6umCz2QpSe7VaLTQaDaampnaoy8VCEATqNHLEUwyYPbQatUqG+H/8CgCganaAvPudOa9noYa5qVnoQCOTyUCSJGZnZyGTyUQv9iiKgtfrBUFnQMm1ACFDnVaGlUgdqrluOuqk4aygifB+yOXyXc3iCALoauCwHmURiu/TFYBVwaZnsLIcq9SQJcrARpTGtYMG2K0HP5zD4TDkcvmBc55Wq4XFYsHIyAgMBoOo9O4raTdHoqo8UO0BXKlotVqcPn0av/M7v4Nrr70WDodD1PlyuRw2mw1jY2Mwm80l33xKp9NYXV3FzMwM/H4/6urqdpjZisVsNiMSiSAQCBwK81ulUgmLxbJ5jS02JLL5LWaNOhnUSgLJZBLLy8uYmZkRFsZdXV1oaWmBwWAoucDAx+W8j9phyKzRaDSor6/H+Pg4LBbLjvuKz5y5ePEiYrEYLBYLuru70djYWJbMmYMgCAIkScLj8QjG3rV8jTmOA8uyiMfjuHDhAiiKAkmS6OnpgcPh2LUKoVLwmTQURQGA8HvghecLFy7sulZbXV2Fw+Eoqxn1YehuLBIpo+awkk6nceutt+ITn/gEbr/9dtHnMwyD4eFhtLa2Ft0qja939vl8yGazQr1zXV1dSScSn88Hj8dTFnO7LMVicQ9zYS6VQuBdd4CJxlF/bS+OPfZnwmssR0Bm74Bev3MxR9M0hoeH0dbWduA1zmazwq4NTdOw2+0wqeSIcTowss0HoC9hxGKosunSPAQ4vH8ojboqmAjvB28W19raCrPZLNSMMwyLxXQ/POG9y2Lq1Cyef+rXSCXz7N0pUZPc+i4rPvfJrgMDv/n5eWi1WjQ0NOT1vqlUCsPDw0JKcT5IGTUSFaK2JuIrkJmZGXzgAx/Aj3/844J26BOJBEZHR0viwbc964PPnCm14ev09LSQ9XEYiMfjmJi6CL2zN6/jVXQQ634PVCoVHA4H7HZ7QcJWofBtmpVKJbq7u2taSOCJRqOYmJhAf38/OI6Dz+dDIBCAVqsVPGdqafHMe6oolcqay17a7tvDtyNXKpW4cOFC2ZqrFArLsqBpGgqFIud3EggEMDMzg6GhoZy57fXXX8fQ0FDZflMEQVS0fXyFkMyEDzORSAQ333wz/uqv/gq/8Ru/Ifp8iqIwPDyM7u5u0S7cuwkLJEmWfRIpp7ldPMVgeQ9zYeprX0fwu0/B9JEPoucP3i/8e1pDwubc2+iXFxLa29t3LPb4a+jz+YS0ZIfDIUxs0YAPsQyQlW2KQNU0FW4wMrjxePVMhPcim81ibW0N8/PzUKlUaG5uhsPhgEajAcsBz5+T48LK3qJePSL4yY/GKzhiiVKj08rww2/2QV+n3ndOmJ2dFbrK5Us2m8Xw8DCam5vR1NR04PFXkpGdRFWRYrAa4Ny5c/jYxz6G5557rqANL758ZGhoSLSoEo/H4ff7EQgEoFQqBWGhnAsVvnSez9I5DGxEoliJ5pcVWSeLoIksLOuoVBw2sSaRSGBpaQmrq6swmUxobGwsewvvYuFFRwA4evRo1a8xL87wAtdupWF8c5WTJ0+WvBKiGLa27lYoFEIMtrGxgYmJiZwS8ldffRXXXHNN2UreJKEmFylIqBG8Xi9uvfVWfPOb38TAQH5eDVvhWxqeOHHiwB9/Op0WOgWwLAu73Z4jLFQKvtW42PZ1+RCMUghGd2ZYEEtLWH3fB2H55x+jq22z3jlDaGFuaRdKnvaCF8S6urqg1+sFgYtlWWHna7drGA+HEYunNg2F32TCZ0M8U/kg4u1HMmix5FfnXW62Clz8NbRarZienkZjY2NOxgTLAT87J8f0HmINQXDwjE5jYixUqeFLlIH/9clWvOt6S043gu1MT0/DarXCZhMndjIMA5fLBYvFcuDiRBJqJCqEFIPVCC+++CL+4i/+AmfPni1oAbW+vo7Z2dkDd5q377ZrNBohY6GSwgLLskLZ1kF+X7XCpCcDLo+prNWmQL22+hmRvFijUqnyantcaRKJhCAsqNVqYR0wNTWFvr4+0eX+1YDjOMzOziKbzeLEiRMVv8bJZBJerzfnGh4kcCWTSYyOjuLo0aMwm80VHO3+7CXWxONxjIyM4OTJkzCZTHj55Zdx3XXXle1aS0JNLlKQcADpdBo33HADMpkMaJrGBz7wATz44INYWFjAvffei1AohKuuugpPPPFE0TfW3Nwczpw5gx/+8Ifo7OwUfT5vvLZbWl0qlRKEBb7GkyTJotrylYJymtuthDKIpXZ6scQf/T7UH/0wOogFsJBB6eiEVnvw3y6TyWB1dRULCwvQarVobGwUsj72I51IILK+jqT8UvCXYXVwLVfWtV6jZHH7qeqaCO8mzmwXuPgF9fY2jAeJNUYNjed+8DpoWprWDit9x/X4m7/o2dE6citTU1OCAZ9YWJbF+Pg4VCrVvjtwklAjUSGkyeoAKhmDPfPMM/jWt76FZ555pqDYaK+ybo7jEI1G4ff7hVIIkiSrXk7CtxovtmFCJeA4DhdXM6DZgwWYNpsC+hoQaoBLJTpqtbomxJrdxJntwgJfztfb24v6+vy6MVYTjuOwsLCAeDyO3t7espctp1IpoTxfoVDA6XSKLq9Lp9MYGRkRfD9rBb4MSiaT5ZSAby0hn5mZwdve9rayjYH/7MsMSagpF7zTuV6vB0VRuP766/H1r38dDz/8MM6cOYN7770Xn/zkJzEwMIDf+73fK/rzhoeH8bu/+7t49tln4XQ6RZ8fi8UwMTGBwcFBMAwjpNTy7ZDtdnvNuWnPzs6CoigcO3aspA8xluXgDmSQoS7d5hwHJGMs1FoC7cQiMroGWMm9F3zpdFoQuACAJEmYTCZMTk7m3e2BpiiE11YQV1z6HJmMwKuLJDhULpgwy5bwnqvMFXXoBzbFGd77iKbpHaVhu8F33CJJEs3Nl7pksRzwb2/IcXF19++gyYbwzz+5UPLvIFEZCAJ4/JE+kDbFnmLN+Pg4WlpaCm7PyXEcLl68iEwms2dQJwk1EhVCisEOoNIx2KOPPopf/OIX+MEPflDQs3JlZQWBQAB9fX2IxWJCh5f6+noha7TSz+D94H0OW1paRBsqlxuO4xCLxQSBS08eA6E82N+vza6AXlMbQg1QfbEmkUgIcaxKpRJKcvZbDPNiTa2V6OyH2+1GOBxGf39/ycUavluT3+8XWoUX2zUsm81iZGQELS0tBa33ygWfWUMQRI5Yw3e6y2QyuPHGG8v2+ZJQk4sUJIggmUzi+uuvx6OPPorTp0/D6/VCoVDglVdewQMPPICf//znJfmcl156CX/6p3+Kf/qnfxK1GOEDGrfbDa/XC5PJhIaGhoobqYmlnOZ2WfpNc+E3E2uyrBwZRgUdEnDIQ7C2tGL7M3PrhEwQhDAhbxW40uk0XC4Xjh8/fuDfiOM4rHsWEVXk+gf5k0YsBCtlKszhmkY3NgIeDA4Olj1QpChKyJzJV5zZDsMwGB0dhc1mQ0tLi/DvLAv827ndxRq5jMPkr8axuBAtyfeQqDwfvKsB99/bDJqmwbLsjtaRo6Oj6OjoKHqnb3FxEaFQCAMDAzt2tWtNzC4BklBTm0gxmAgqFYM9+OCD8Hg8+PrXvy5qUc2yLDY2NjA7O4tEIiFkjFqt1po2J6dpGufPn0dnZ2fVd/d5cWa7EavNZsNqmEN0lyzp7dSaUANcEms0Gg26urrK/nnJZFKIYws1VeZLdPKJc2sFj8eDQCCAgYGBouPcTCYjXEMAcDgccDgcJS3LoWkaIyMjcDgcOZuS1YYXawDkbJjRNI2XXnoJR44cQVtbW1k++0oTamprpjqkMAyDwcFBkCSJm266CV1dXTCZTEJw39zcjJWVlZJ93jvf+U788R//Me677z6k0+l9j+VTamdnZ/Haa69hbm4OFosFvb29QmlJrd/wBEHg6NGjyGazWFhYKOl7qxQyNFk2J1WOA7LM5rVgGQ4GR6Mg0qRSKSwuLuLXv/41JiYmIJPJ0N/fj6uvvhotLS07Fm4ajQYDAwOYmppCNLq/KEAQxK7BnkOfLME3zA+HkUVXK4mGhga4XC5hAi4lFEVhZWUF58+fx/DwMCiKwvHjx3Httdeivb1dtAeSXC7HwMAAQqEQlpaWhH+XyYDTVzE40rgzYGNYAm9959Giv4tE9fjlfwbBsBwUCgXkcjmy2WzO/cowTEmExvb2djQ0NODcuXPIZnc3H5eQkKg+lY7BvvjFL0KtVuOhhx468FiWZREMBjE5OYnXXnsNPp8PXV1daGxshEKhgM1mq2mRBthshTs4OIi5uTmEw+GKfz7HcYhEIrh48SJeffVVuN1uGI1GXHvttejv74fD4YBcLociz2m/FlVpgiBw4sQJpFIpzM3NleUzkskkFhYW8Prrr2N6ehpqtRqnTp3C0NAQGhsbRa8FdDqdEOdGIpGyjLnU8Nkpw8PDoGnxnUCz2SyWl5dx7tw5jI6OAgB6e3uFtUCpvVP4314wGMTi4mJJ37sYZDIZ5HI5CIJANpsF++Zut0KhgE6nQzgcxszMDA5IBpHIg9q16z5EyOVyuFwubGxs4K677sKFC+Uvrbj77rsRCoVw//3344knnsjZ8eXFGZ/Ph1AoJOw4dHR05CxgWJYV2kbWeqBAEASOHz+OsbExLC0tldTcrk4jh92owPI6oJBzMOtYNJjqQGdTWF1cFupMSZLEwMBA3hOxVqtFf38/RkZGDqzl3U2o4VgKBg2FaLr8QloXufnA4uvQXS5XSTJrtrZ0pygKdrsdx44dg05XmkwhXjAbGxsDx3GCgr8p1tAAFLi4mntvh9Mq3PK+Lvz8Z+UJhiTKS3CdwrmRCK4dMuUECnzrSJZlSzaf8cHrG2+8IbSgvMxKniQkDj2VjsEIgsA3vvENfPCDH8S3vvUtfPKTn8x5nWEYhEIh+P1+xGIxWCwWNDQ04Pjx48L8YbFYMDk5icXFxUPRWUmlUmFwcBDDw8MVyaDgxZntpWFdXV17xiVK+eGemwmCwMmTJzExMYG5ubmSZNbwmTO8xYHT6Sxp62StVovBwUG4XK68y/2rTWNjI2QyGYaHhzE4OHjgtdgtA/zkyZMV8/CUy+Xo7+/HxMQEZmdna8LLCNiMv2UyGWiaRjabFTJrCIIQBLzJycmSmzjXwnevJJJQU0JMJhPe+c534pVXXsHGxobQd355eTmvlq9i+fjHP45AIIDPfOYzeOSRR/DCCy8gmUyisbExr4daQ0MDKIrC+Pg4+vr6av7ml8lk6Ovrw8jICBQKRUnN7az1StRpOLDUpgmYa37zoeZwODA4OFiwSq7T6dDf34/R0dFdTZx5CJlsM6Vn29+g05Ysu6mwWsmh2XwpI6GxsREcx2FkZKSg9NDdxJmjR4+WTJzZDn9fjI+PY2FhQQh6ebGG4xSYWctduKttDpDkGvz+ymUtSZSOf/+PIK4d2gwI+Q4E/O5YqTJqePh08PPnz6O/v//Q1ONLSFxpVDIGk8lkeOKJJ3DbbbfBarXilltuwbPPPovOzk6oVCpYrVa0tLTAYDDsGltt3XxaXl6uqbKGvVCr1RgYGIDL5SpL15/dxBmHw4Genp68xPd8M2pqMqXmTUoh1mw3s3U4HCUVZ7aj0WgEsabWOhXthdPphFwuF8Sa7TE+TdMIBALwer3IZrMgSbKkm4xikclk6O3txfT0NC5cuFByz85i4OOvTCYj3GP8/DY3NweXy4WBgYGaTwioVeQPPPDAfq/v+6IEEAgEQFEUNBoNUqkUHnzwQdx5551CGmBvby8eeugh3HjjjbjmmmtK+tkURYGiKDz++ON4+OGHEQwG8fa3vx1vf/vb4XA4UFdXd+APw2g0Ih6Pw+v1wmaz1cwPfy/4jlSzs7OQy+UlCRQSiQSWl5cxPzeDjY0NGI1GdHV1obm5GQaDoehFn0qlgtlsxtjYGKxW664Py0wqiQwrA0fk/r1UchorGzqUM7LocdBoMOeWCRkMBrAsi7m5OZAkeeB9RFEUfD4fZmdnsby8DI1Gg9bWVnR0dMBsNpe9vI4gCNjtdqysrCCRSAiBAkEAPQ0sQjEC6/FL15BhCZw4bsa4a62s45IoD6veDE7fZIdGvfnb5MsHGYbBysoKmpqaSirWaDQaWCwWjI6OQq/XH4q2pCJ5sNoDkNiVB6o9gFqnmjFYPB4HAHzhC1/AE088AZ1Oh5tuugl9fX2w2WzQaDT7xlT8c2tubq5k8Uy5USqVsFgsGBsbg8ViKbrUg+M4bGxswO12C22U7XY7enp6hDg237iUZoFI8mCPGnOdDEpF7ca6/H2xurqKeDyel/CRSqWwsrKCixcvYn19HQaDAd3d3SWLYw9CoVDAbrdjYmICOp1OdCl7Nairq4NGo8Hk5KSw/vH7/Zibm4PH44FKpUJbW1vF4tiDIAgCVqsVkUgEq6ursNvtNbNmk8lk4DhOaA7S3NwMgiBgsVhAURRmZmbgcDhKItbwmTyXGXvGYJKZcJGMjo7iwx/+MBiGAcuy+M3f/E188YtfxPz8PO69916sr69jaGgITz75ZMkMKOfm5vDQQw/hjTfewA033IC77roLjz32GK6++mp86lOfEv1+vFmvQqFAd3d3ScZYbmiaxvDwMNrb22G320WfH4/HhY5XhRqpiYXvuDUwMLDjIRZfDyGcZEDLdt4jgaQR82U0FT49kEK9dvef+srKCnw+366ZNdt3HOx2uxBYVQuO4zAxMQGtVovOzk7hIcawwE/fUGB2W2ZNcmUJ//WfnmoMVaJIPv6hFnzg/ZudEPjua16vF0qlEn19fTndCEpFOp3G/Pw8BgcHayZAKhGX1Ze5jJBisAOoRgz2y1/+Eo888ggCgQBuu+02vOMd7xAym6+++mrR78eb9XZ1dcFqtZZkjOVmawdRsSUgHMchHA7D5/MJm2MkScJisRQ1Z6cpDnNe6sDj2u1y1Glqp7PWXnAch/HxcdTV1aGzs3PH67tlztjt9pL7pIghk8nA5XKhu7v7UNzLDMPA7XbD7XZDrVaDJEk4nc6aF00XFxexsbGBvr6+qneJ4zhO2PAPBoMgSRIdHR0566m1tTW43W6cOnWq6PtTLpfvaPBwGSB1fbqc4DMXrrvuupy2aHfccQfuuece3HvvvaLfk38gGAyGsjl1l5psNovh4WH09PTAYrHseyzf8Yqv1dVoNEKngEqq5JFIBFNTUzuCm1QshvVIAlnZzl0IQqbEq4vl6bRAGhi860Rm32O2ijUcxwllTZlMpibEme3s1eqSYYGf/lqBWe+lQFCnYvHCs+cQi0lmsYeN1iYNHvicGT6fDxzHCR0X+DRc/mFearHmSus4IFFVpBisBhkZGUFdXV3Oxpbb7cbtt9+O733vezh6VLxhPd/a9jB10NnY2MCFCxcwNDR0oAjGsizC4TD8fr8gzjgcDpjN5pLN0QzL4cLKwUINvTGPvpNHDsWuPB+b6/V6dHR0IJVKCa205XK50DmsmuLMdrLZLFwuV010CdsNlmURCoXg9XoRj8dhs9lQV1eHxcVFDAwM1FQ8ux/Ly8tCbF4N4SKRSMDr9SIQCECn08HpdMJqtYIgCKHsdGusFAwGcfHiRcHvr1D4RhKXGZJQcyWQSCRwyy234POf/zxuvvlm0eezLCu0gSul/0s5yWQye5rb8Sov38ZRq9UK4kw11djdghsqk0HIH0RKvruKf8FvQ6QMpsLXdWfQZtu/wxNN05ienhYELr6Vdi0/zDiOw9TUFBQKBXp6evYVa4zyOJ5+YqRaQ5Uogj/7n2Zcd03zjh1dvnUkQRAlz6yRhBqJCiLFYIeIiYkJ/NZv/RaeeeaZgjxxUqlU2fxfysX6+jpmZ2d39UDhxRmfz4dIJAKTyQSHwwGTyVQ2kWRyOYuDGs1o2QDikRB6e3sPhVjDt8GmKAo6nQ4Oh6PmxJnt8GJNR0dHQVnvpYZlWayvr8Pn8yEajcJqtcLhcOR4SMViMcGz87D8/rxeLzwej6hGJ8WwNYtLqVTC6XTCbrfvWFPxMRjvIcj/ziKRCMbHx9Hf379vg5X9kISaXKQg4ZARCoVwyy234Ktf/SquvfZa0eczDIPz588XXFJUDVKpFEZGRnDy5Eno9XrEYjH4/X4Eg0HodDqQJFl1cWY74XBYUJZVKhVYlkVgeRlJxe47aVlOi2FPad30VQoOd5xKQb5LnELTNILBIHw+H1KplHAvRKNR9Pf3H4pJkuM4XLhwATKZDEeOHMkRa/711wrMCWINh8D0HM6f81VvsBIFceu77fjDj7fv+hofKAAQuhGUAkmokaggUgx2yPjVr36Fz372szh79uyBmb67EY/HMTY2hsHBwUPh8wFs+gQtLi5iaGgIMpksZ0FsNptBkiTMZnNFykUvrmZB7b/3hE6HAt4VN+LxOHp7e2uyjDWdTgsLYplMBrvdjvX1dRiNxkPRJQzY9C50uVxoa2sDSZIV/3yO44R7MRKJwGw2w+l0wmg07vk3TyQSGB0dxcmTJw9N44BgMIi5uTkMDAyUpRNVJpOBz+eDz+eDTCaD0+kESZIHxkF7iTXxeBwjIyM4ceJEQcbTklCTixQkHEI8Hg/e//734/HHH8exY8dEn09RFM6fP48jR44cCvd2juPg9/sxNTUFpVIpdAqw2Ww1/WMOhULCTpRKpUJg2YO4bPcHg0xG4NVFEhxKt/tz1ElhqP1SmvBu4ozD4cjZWVheXkYgEDg0Du4cx+HixYvgOA5Hjx7dU6yp1zD4lx++jmzmYCNCidpBp5XjH789IJgKb4dlWbAsC47jSibWXGn10RJVRYrBDiE//elP8Td/8zc4e/ZsQV1i+BLpUvg5VAKWZTE7O4u1tTXBbJjPnKm0CDLvo5DK7v+z6XIooFHJMDc3h3Q6XfL2wYXCe63xC2I+e5m/B1iWFSwK2tvbqzvYPOHFmtbWVjgcjrJ/Hm9O7fV6sbGxIWRxiREK+Qymw9JuHNjc/J2enkZ/f39JOlPxLcm9Xi9YlhVKy8X6fLEsu2spejqdxvnz59HT0yM6KUASanKRgoRDyuTkJO677z48/fTTBbV9TKfTcLlcOHnyZMHpaeWE4zhEo1H4fD6EQiHo9XoYDAasrq4eqp2oYDCI+fl5DA0NIer3IUrsfa2DKQPmAqUrN3rfQAp1KlrwnNlLnNmOx+NBMBg8VGLNzMwMaJrG8ePHc8Saf/m1AvNvijV1TBjPPTVZzaFKFMDnP9WBm2/cvw6epumSiTWSUCNRQaQY7JDy/e9/H0899RR+/OMfF5SBFwqFMDc3h1OnTtXkfMP7fPh8PsRiMcEIOBaLYXBwsGqxgSdIIZrKT6jhOA6zs7Ogabpq7Y63ijMEQQhlTXstiA+jWEPTNFwuF5qbm+F0Okv+/tvbupfC/4hfAx05cqSgzLhqEI1GMTExUXDp1tYGIRRFCUJhseupvUrReZ/RlpYWUXYb5WgUUQNIQs2VyCuvvILf//3fx9mzZwtyX08mkxgZGcHAwEBJFNpi2T4Z19fXgyRJWK1WQV3ld6LyMberFfx+P9xuN7qaGxHh9p5cZTIFXlksTTmaUZ1Go2wYqVQKNpsNTqdTVBvMwyjW8K0/t+6eMSzwL68rMO+TQUZwmHtjChenw1UerYQYeo/p8fCDxw88jqZpsCwLlUpV1G6MJNRIVBApBjvEfO1rX8O5c+fwne98p6DnpM/nw/LyMgYHB2tiB5lhGEGcicfju/p8uN1uRCIR9PX1VUX4WAvTWI/vnxnb5VRCo9wcG591CyCnRLqc8KUkfr8fAA4UZ7ZzWMWakZERNDY2oqGhoej3275ZW19fD6fTWXTnsK3wHay6urpq0hR5N/jSrRMnTuRlSs7/pr1eL5LJJOx2u7AeKCV7laLz94XVas37XpaEmlykIOGQ8/zzz+PLX/4yzp49W5DCWkwLxlKwmzjjcDhgtVr3/KGur69jZmYGp06dOjReEj6fD9lYBLSWBPYJFEplKtyhX8KJNjX0en3BgclhFGvm5+eRSqVw8uTJXcUak5bGM999Daw08x0qHn+kD00NB89PpRBrJKFGooJIM9Eh50/+5E+QSqXw13/91wU9a5eXlxEKhdDX11eV5yzDMEJJdCKRgM1mA0mSOeLMdqpZUhSIMvBH9jep6XYqoVZeGhfvZ6dQKNDd3V2WMWcyGSFzBoCQrVDohuJhFGsYhoHL5YLT6SzIbHt7G2i9Xi+sB8olZPKmyO3t7VXx2SkE3rdzr2wg3ljZ6/UiFosJm7XFrAfyYa9SdJZlMTY2Bq1Wm9P8Yy8koSYXKUi4DPjhD3+IJ554Av/4j/9YUL0zb3xbKeGDrzH1+XwIh8MwGAxwOByilPKt5naHZVHl8ywhJasHR+z9HSlOi/NFmgqr5BzuuGp3E2GxeDwehEIh9Pf3H5qJc35+HolEAidPnhTGvFWsUST8+Ld/nqnyKCXE8D/ucOIj97XkdSxfM729dWS+XGn10RJVRYrBDjkcx+EjH/kIOjs78bnPfa6g91hYWEAymayY8LFVnEkmk0LmTH19fV6fz5cbcxxXsSwVno0Eg5V1cUINsDnmyclJaDQadHV1lWQsvDjj9/vBcZzQSrtUm578AtdkMqGtra0k71luGIbByMgISJLM25aB794aCASg1WrhdDor6kHJ++w0NzeXJBuoEmQyGYyMjAgCE8dxQge2jY0NwUdqP2PlcrFbKTovljIMgxMnTuy7npCEmlykIOEy4ZFHHsGrr76Kf/iHfyjoBueFj1OnTpVlBivEAAAAIABJREFUctw+iZSixtTr9WJlZaVm0oYPIhmNIBilwMj2XjzKZARedZPguMInqR4nhau2mAgXy9LSEtbX1w+VWLOwsIBYLJbTnpNmNj1rloMEzv/HCFZXElUepUS+WM1KPPnNAchl+QUcxYg1klAjUUGkGOwygKZpnDlzBrfeeis+/OEPiz6fL8+RyWTo6ekpwwh3NhOw2WyCX10hCzmO4zA1NQW1Wl0y4SMf4mkW7gC97zE9DUqoFDu/E8dxmJiYQF1dXcGdlbLZrFDWxJuwllKc2c5hFWtGR0dhs9nQ0rL7BksikRDEGbVaDYfDsWsb6ErBl+gUmg1UDbLZLM6dOwe1Wo1MJgOj0Qin01mxDmz7wYs1SqVSiKf4rPeDusuqVKqqj78MSEKNBPCFL3wBkUgEX/nKVwq6yVdXV+Hz+UpW6sKyLMLhMPx+f8nEme2srKzA7/cfivKcTCqFQCgKSrb/Az2UMmC2CFPh9/anYNKV9qd9GMWaxcVFoZZeJpOBoih4fQG8MGFEPMHhX5+aqvYQJUTwV3/cg7delX+22V7dCA5CEmokKogUg10mpFIp3Hrrrfj0pz+N06dPiz6fFxH0en3JSl1481C/3y80EyBJsmQlEBzHYXx8HPX19RUrz0lnWcz5ChNqgEslRUajMW/hI5vNCmVNlRBntnMYxRqWZTE6OgqLxYLW1lYAm78Rr9eLQCAApVIpiDO1YmHAC0xWq1UYcy0Si8Xg8/kQDAah0+mQTCbR2NhYc2PeqxTd4/HA6/XuWREhCTW5SEHCZQTHcfjYxz6GlpYW/NEf/VFB77G0tIRIJILe3t6Cfii8OOPz+RCJRITWeSaTqWwL/KWlJWxsbFTN3C5fGIaBb9WHjHx/EaYYU2GrnsFNvZmCzj2IwyrWeL1eqNVqZLNZkCQJm92BX44ZcHFiDf/xy8VqD1EiT952jQkPfF7cbjNvcCeTyfIWayShRqKCSDHYZcTGxgZuvvlmfPnLX8b1118v+nx+cWu32wve1acoSsicyWQyOZkz5YAfs81mK6gDqVhohsP06v4Zw/sJNcClMVut1j0zPraLM7znTDW8HIFLYo3ZbK65BflesCwLl8sFgiBAURQUCoUgctWKOLMd/jobDIaCs67KQTKZFEQujUaTUx62dczt7e01tQ7aK7vZ6/ViYWEBp06d2uHjJAk1uUhBwmUGTdO455578J73vAf3339/Qe/BtzM8evRoXj8W3rjK5/MhGo3CbDaDJMmKpt/xJrLVMLcTg295DUnZ/h22OI7D+KoBSUZ8YHVtZwad5P7128XgdrsFUaxWxRq+/t7r9SKVSkGtVoOm6ZyyPpoB/u0NAj/+3ijW19NVHvGVhYwA5HICMjkBhZyAXE5ALgPkCgJyGQHFm/8vl2/5n4yAWi3Dl/64ByqVuPtOrFhzpdVHS1QVKQa7zFhbW8P7/h97Zx4fRX3//9fmTsh97YYrJwRCIJtARa0cWiHcwQjk8isVLV9trYpaS39Whfoo3ha1onxFtPWo7gbCoRU8EcVitWQTEkhIQkKunclmN5tsNnvP/P5IZ0xCjt3NHjPJPB+PPvowx+yHye7Me16f9/v1WrMGr732GhYsWODw7zOGrNOnT4dYLLbrdywWC1QqVX9ogdmMuLg4iMVilye7jATjS5KQkOARj48LLeZRPzizE/zhP4pQA/TfFyoqKhAXF8cKTMOJM/Hx8eOOL3YVw3WpcJGhqVdWqxWxsbFIS0vz8srsg6IoVFdXIzg4GKmpqV57pjAajSAIAh0dHfD394dEIhlxPIyiKFy8eBH+/v52GfZ6kpHEGrVajdraWkil0kHJw4JQMxihSJiAGI1GrF27FnfddRfy8vIc/n3G9CkgIGDE2WOKoqBWq9HR0cGKM0znjDc+YIy5HUVRdgtM3kDVTqAXY9/0nTEV9velkZdjgJ+bmwG4KNYMjSAcOn/f2trKjsgNFGsOf6nHgQMXvLz6fhjBghEqRCIaAIXAQH/4+/nAx3egkNH/cz7/FTT8fK8WNn4SPDDoe34Dfmbw1/oDyQwGPfS9PbBazYiKDEdUdCSmTAlif575WV8f0VXiilargVLZhvnzMxAUFMj+rM+An/Gx02fGlTBijUgkGlOIEYQaAQ8i1GATkLq6OmzatAnvv/++U7vyFosF5eXlSEtLGzbVhfkZRlSwWq3sWJOnxJmh2Gw2nDt3DomJiW5Pz7nUboZllP2o2VP94e879iXPZrOhvLwcwcHBMJlMsNlsrMjFFXFmKFwVa4YbD2NSrxjhY8qUKUhJSfH2Uu2C8WDy9fX1qGE244FEkiREIhHbgWRPUAzjdWWz2TB37lxOPQeNNIre3d2NqqoqLFiwAGFhYQAEoWYoQpHgIVpaWnD77bezH77t27fj/vvvx65du/DGG28gLq5/1GXPnj1Ys2bNuF+vu7sbK1euxO7du7F06VKHf5+maXYmlrkZ2Gw2tnNGp9N51VV8OBiByd/fn7PKvZrsQI9t7LhGH5EI/26Oh80BU+E0sQWLkl1nIjwaXBBrGLGQeT/GxMRAIpGMmFzR1tYGgiAGmU9bbcAHR9rRZ7D9V/zw6e/uGEn0GCB+jC6ODPyffT87HCRJ4sqVK5BKpU4lutl7Hl0Z5ahWq1FXVwepVOq1NvHhsFesEYQaAQ8i1GAewtM12Llz53DnnXfi8OHDdnfGDMRkMqG8vBxz585FREQEgJ8ehjs6OlhxRiwWD9qN9iaMwJSamoqYmBi3vc5l0gKDeeSPzlhCjdlsZjuQrFYrzGYzZs6cySnhYzTsGd3yBAPFQpvNNup4GOPBFBwcjJSUFE48M4wFI3xQFIU5c+a4bc1Dz+NAkctRGMNevV4/KEyDC4zU3azX66FQKDB37lxER0c7HWnPcQShhusolUoolUrk5ORAp9Nh4cKFOHLkCGQyGUJDQ/Hwww+7/DUJgsDq1avx6quvQiqVOvz7FEWhvLwcoaGhMJvN6O3tZWMcw8PDOXmhdYchnytgRAWDTgdbkH0FjMYQjjoHTIVz5xsQNcVzH+mhZr2egPFAIggCPT09DouF7e3tUCqVyMrK4k2se2dnJxoaGiCVSl12A2NS2AiCQHd3N6KjoyGRSFz2udZqtbh48SIWLFjgtR3e4aAoChRFXRUdORBBqBHwIEIN5iG8UYN98cUX+OMf/4iysjKEh4c7/PsGgwHnzp2DRCKBVqsd9DDM1Y4Ps9mM8vJypKenIzLSsa5ge2nutEBnGPmjkz7VH35DhJqhHUgDz6PVakV5eblHuoFcBTO6NVqykjuwWq3sebRYLIiLi4NEIrHr/chEpAcEBCAtLY2TzxBDoWkaDQ0NMBqNY8ZKO4LNZoNKpQJBEDCZTIiPj7f7PNpDc3Mz1Gr1qOlK3mAkscZoNLIirzfFRzciCDV8Iy8vD/feey/OnDnjtiIBABoaGpCfn493333X7ghFxuODJEno9XpYLBYkJiZi5syZvLiwcmW3Yah3T3R0NKLDw9FL27f75YipcPQUG1bOd4+J8Gg0NTWhp6fHrcr9UFFhvGN2SqWSjXXni1ij0Whw6dKlcXWp0DSNnp4eEAQBjUbj9ihHnU6HqqoqZGZmsi2tXIGJjhxOrJlsbbcCXkWowbyEp2qw0tJS7Nu3D6WlpXZfu00mE9s5Y7FYYDKZkJWV5Tbhw9UYjUYoFApkZGQ4JVCNRXuXFV291IjfZ4QaxruHIIgxO5CYbqDk5GS2u4rreEqsGSoqjMcDiRkp8vPz45yXymhcvnwZvb2946p1KYpi/RP7+vrY8+guo+/29na0t7cjKyuLU+bNI3U3WywW1NTUICcnZ1JtlglCDQdpamrC0qVLUVVVhRdffBFvv/02wsPDsWjRIrzwwguIiopy6espFArccccdKC0tHdHobaA409fXx3bOhIWFsTcwd+6QuBrGbd5T5nYDX3eoODOw46NXq4VKj34zEDu41BmDrr6xR15+lmxCqth9JsKj4Q6xhqZpdHd3gyAIdHV1uVxUIEkSzc3NkEqlnLqBjYZWq0VNTQ0WLFhgd6s7TdPo7e0FQRDo7OxEWFgYJBIJoqOjPXIj7OvrQ2VlJebMmcO5awcj1vj7+w/acRKEGgEPItRgXsDTNdjrr7+OkydP4u9///uIu9tGo5EVZ2iaHhQBzVz7c3Jy3DYC62r6+vpQUVGB+fPnu/xBVNVtQ0fPyPVOhI8aqg6C7fiwdzzMU6NbrsRdYs3QZwJXigo0TaO2thYikcij/i/j5cqVK2zyqb1dKkO7wMca0Xc1HR0daGpqcuv4vDOMJNaIRCJOrdOFCEINX+jt7cWyZcvw6KOPIj8/HyRJIjY2FiKRCI899hiUSiUOHjzo8tc9deoUfv/736OsrIx9YLJareyF2GAwXGXAOhBmh2TevHmc2x0fCcYobubMmW5tZx1LnBmIUa8HqTWBEtl3kbfSQfhPy+hFo99/TYT9vdjd2NjYCJ1ONy6xZmjHR3h4OMRisdtEBeYGlp2dzRuxpqenB9XV1WOOFOn1ejZ1ITg4eFCUo6cxGo2oqKhAWloa54pfq9UKiqIQEBDAnhtBqBHwIEIN5mG8VYM9+eSTaGpqwssvv8xeXxhxhvHNYVKGhuu86ezsRGNjI7Kzs3nTCdrb24vz588jKyvLpT46Xb02tHeNLNQEW9shEcc79ZrM6NasWbNGNHLmGjabjY11H09EOtPxQZIkent7x+1XNxqM/wtN05wOABnKcMEUQ6FpGlqtFgRBQKvVIioqChKJxGt+nlz2DRw6ii4INVcjFAkexGKxYN26dcjNzcWDDz541febmpqwbt06VFVVueX1y8rKsHfvXqxfvx7Hjx/H2rVrsXr1asTHx9t1Idbr9aisrHT5TdeduGuHhFHJSZJ0aBzHarVCSXTB6mPfhcgeU+HUeAt+luIZE+HRcEasoWkaOp0OJEmyHR9isRgxMTEe6fhQqVRobGzk3G7DaDDF79CRIkeiHD2N2WyGQqFAYmKiU8aa7sRms8Fms8Hf3x9+fn6CUCPgSYQazIN4swajaRq//e1v4ePjgylTpuDEiRN46qmnkJycjPj4eLv8xwiCQFtbG7Kzs3kzGtDT04MLFy649CFRZ6DQ3Gkd8ftzpvnDdxzpgoyRMxc7QUfCWbFmaJiAJ30ombRWq9XKuZSi0RhufJ7ZaCRJEmq1GuHh4WwXOBc+q1z1DQQGj6IPjfCeQAhCDdehaRpbt25FdHQ09u7dy35dqVSyozl/+ctf8P333+ODDz5w6Wt3d3fj+PHjkMvlqKioQFJSEnbv3o2f/exnDh+LuelmZ2fzxpmb2SGZPXv2uFqanRVnBkLTNNrbO2AW2V+waIxhqOsYud10ZaYR0aEjz2t7ksbGRvT29mLevHmj3pwGjuOEhIRAIpEgJibGKx0fjFlvdnY2b8QaRjSdNWsWDAYDCIKAj48PmxbAxRudxWJBRUUFEhISMG3aNG8vZxBMdKSfn5/b5sW9DD8q4MmHUIN5CG/WYE1NTTh06BDKysrQ1NSEG2+8EX/84x+dGlVpaWlhxy/48mCr1WpRW1s77nus1WrtT2vq1MI/cmTPxbnT/OEzDqEG+KmLfGDqFtex2WyoqKhAfHz8qGLN0FrWmwmujFmvyWRCRkYGb97TTCLn7Nmz0dnZic7OTkyZMoWtZbkgzgyFD76BgYGBnOr6cSGCUMN1vv32WyxZsmRQQs6ePXvwj3/8AwqFAiKRCElJSdi/f79LPVXefvttvPbaa9iwYQNuvfVWzJkzB3v27MGlS5fw6quvOnVR1Gg0qK+v59XIiLPmdq4QZ4aiVKpgpO0vVmw2Cj+2Df9gGxVCIXeB0al1uIuB0YADz5FerwdBEFCpVAgODoZYLEZcXBwnHOnVajXq6+tdmqzkLpj0ivb2duh0OkydOhXJycmcXzfw065fdHQ0EhMTvb0cFpvNhjNnzuDDDz9EcXExVqxY4e0luRp+VL+TD6EG8xDeqsF+9atfobm5GZs2bcLGjRsRERGBDRs2oLCwEFu2bHHqmJcvX4bRaORVF4JarWY3RBypG1lxhiRZI9vYODFatCMfY+50f/i44LwYDAZUVFS4zRTZHTBijVgsHrQhwoQykCQJrVaLyMhIiMVit4UJOAITKW0wGDBv3jyvr2cs+vr6QJIk2tvbYTabMWvWLCQkJHCilh0LrvoG6vV6fPzxxygrK8O7777LOSHJBQhCjcDwMLvEA6FpGjt27IC/vz92797t1HE7OjrQ3NyM7OxsXlycgP6brkKhGNPcjqIodr60u7vb5Te0jg419Fb7x1FomkZtRxS6TVfH9i1MMmOWZOQWYG/BiDUpKSmsQWJAQACnxnGGotFo2DlerokeNpuN9TIwm81sJKZIJIJCocCsWbM45/8yEhRFoaqqClOmTEFKSorXijLGhFEul+Pzzz/HokWLUFxcjJtuuomT789xwu3Kd/Ii1GATnOFqsN7eXuTm5uKRRx5xShRm/D18fHwwa9YsVy3V7XR0dODKlSvIyckZtW4cKs4M9E8E/hvz3DryuHfGdH+X3VfcaYrsLgZ21oSGhg4KZWBqWS52fDB141gd2d7AaDSCJEmQJAk/Pz+2lu3p6eGk/8tocMU30Gw248svv4RMJsOFCxewdu1aFBUVYcGCBV5bkxsRhBoBx6AoCv/zP/+DrKws3HvvvU4do729HSRJIisri3MX1ZEYydzOk7sN6s4u9JgdO19anQW1XTMHfc3P578mwhx7pjQYDGyqEgCkpqYiPj6eF91XXV1dqK2t5cRNd7goR4lEctV8MeP/wqdYUSai08fHx6NGgjRNo76+HqWlpfjoo4+QmpqKwsJCrFu3zut/bzcjCDXcRKjBJimdnZ3Izc3Fiy++6NQYOk3TqK6uRmhoKJKSkly/QDehVCqhVCqvMmMdGG5hNBqvEmeGUttuhnUEP2FXCjXAT6PGXPT3GArjlaJUKtHe3o7Q0FCkpKR4LOlxvLgimMJVmM1mVpwBwI6WDx3fY1LZsrKyEBx89YYqF/GWb6DNZsO//vUvyOVyfPfdd1i+fDmKi4tx3XXXef3v7WYEoUbAcSwWC/Ly8nDrrbeiqKjIqWNcuXIFOp2OF+2KDIzPTlZWFisqDBRnIiMj3XrB0Hb1oMvg2EdPJAL+fSUeFH4qbJLjrFicanb18pxiYHoF45USHx+PlpYW3rSzMnjzpjs0QYwpVseKcrRYLFAoFJg5cybnzHpHghFNmNl0d33maJoGQRCQy+U4cuQIIiIiUFhYiPz8fN54D7gAfnz4Jh9CDTaJaW5uxvr16/H2228jPT3d4d9nugLFYjGmTp3qhhW6h9bWVnR2diIjI4O93xkMBocioBsIC4yW4T8+rhZqAPclWLkCmqYH+f6FhoayG43nz5+HRCLh1fvjypUr0Gq1g8YUPYXFYoFKpQJBELBaraw4M9ZGTnd3Ny5cuMALMY/BU76BTBe1TCbDZ599huzsbJSUlOCmm27ixQauixCEGgHn0Ov1WLVqFR588EHk5uY6dYy6ujrYbDZeROwxsXnNzc3o7OyEWCxGQkKCR1tBe3t0UPU6bv7bZQzDpQGmwjfPMyI2zHsmwsPtNgyXXtHQ0CCINaPgqihHq9UKhUKBqVOn8qooa2xsRE9PDzIzM106RtnV1YWjR4+itLQUZrMZW7ZsQUFBAW+ELBfDjw/e5EOowSY51dXVKCkpgVwud+phyWazoby8HDNnzkR8fLwbVuhabDYbm7ZoNBoxY8YMtlPUkftds8oCnXH4j8+8Ge4JBdDpdKiuruZM50Rvby9IkoRKpUJISAjEYjFiY2MH3UdtNhsUCgUSEhJ4VRc0Nzezptnurs2Z9yTTzRUfHw+xWOywIMeY9fJtTM4dvoGM71BpaSmOHz+OpKQkFBYWYv369Zz47HgBQagRcB61Wo3c3Fw899xzWLx4scO/z4wxBAUFISUlxQ0rHB/MgzBJkoPmdJkLiadNkQ16PYhux31lfHz88K+m/tGW0AAz1uV43pvGbDaznTMURbHizFi7DXwUa7q7u9k4Q1fvoDHtyQRBQKPRsO/J6OjocZ0fe1MfuEZLSws6OjqQlZU1Ln+Yvr4+fPLJJ5DJZFAqldi4cSOKioq86oXDESb1P57DCDWYAL777jvcd999KCsrc8ozwmKxoLy8HGlpaYiOjnbDCseHzWZjx5r6+vrYTlGCIGCxWJwyRW7XWNGlH36jyl1CDfBTXeCt8Wi9Xs+KM4GBgZBIJIiNjR31vslXsaalpQWdnZ1YsGCBy70wKYqCWq0GQRDQ6/WIjY2FRCIZt8DCjMnNmzePNwbUrvINpGkaJEni0KFDOHz4MKZMmYKioiLk5+ePK3F3giAINQLjo6WlBevXr8eBAweQkZHh8O/TNM2qss5ETrqakcSZoZ0zjLlddna2xwxELWYT2lR9oEWO7xLUq2Og1gcgBpewcHaIR4qyoa2gzG6DI6o4XyMYe3p6UF1d7bJ21oHtye6KcrTZbDh//jyioqI4law0FkqlEq2trZBKpQ4JpxaLhTWkq6qqwpo1a1BcXMyr+FoPIJwIbiLUYAIAgH/+85/Ys2cPysrKnLrXmEwmlJeXcyahaCRxJjQ0lL0uM6bIIpEIs2bNcuh63dFtg6rnapMaEYAMNwo1wE8dt9nZ2R4JHjAYDGxipr+/P7tB5kjNylexprW1ld3EGa9YQ9M0NBoNCIJAT08PYmJiIBaLER4e7tJagUkL41qy0miMxzdQq9Xi2LFjkMvlMBgMbPcyE3ghAEAQagRcQU1NDQoKCiCTyZwSWyiKYkcvJBKJG1Y4OjRNo7u7m3W4Dw8Ph0QiGXOsiTFdk0qlHkmwoigKrcou2ESOC0NWBEHRGolV87pRff4c0tPT3aJUM6kLBEHAbDY73Qo6EL6KNUw7q7NiTV9fH1tkBQUFsTtg7nyvMTskoaGhSE5O5s25VqlUuHz58pjJWxRF4ezZs5DJZDhz5gyWLVuG4uJiXH/99RPdkM5Z+PEGmHwINZgAyzvvvIP3338fH3zwgVNdvkxCkbd8MuwRZ4ZC0zQuXLiA4OBghzqyNb02KLu8I9QA/aO1ly5dQnZ29lXmsq5gaMoQI86Mp/ubEWumTp3q0gh6d9Pe3g6lUulUjT70uSAyMhISiQSRkZFurYuMRiMUCgVmz57NyS634XDEN9BgMODEiROQy+VoaWlBXl4eioqKkJaWxpt608MIQo2Aazh79ix+85vfoKysDLGxsQ7/vtVqRXl5OZKTk536fUdhLsIkSUKj0SA8PJwdIXHkgY0xt/PEPCwAtCq7YKEdfx0fkQgdhmj8LMXK3ghcpdoPnNNljP2GSxkaD3wVaxgjQXtnj0eKcvTkiB1TAAcEBPDq5qnRaHDp0iXMmzcPYWFh7NdpmmYN6T799FNIpVIUFxfj5ptvnkyGdM7Cjz/+5EOowQQG8eKLL+KHH37AG2+84VQtwvioeGo0x2azQa1WgyRJ6PV6xMTEsCMk9t5zaJrG+fPnERERYXcXqM5Aobnz6vFvkQjImO5+oQbotw2or693mVhjMpnYumFgKIMrhSC+ijXMhqo949E0TUOn04EgCKjVars3bV2NyWSCQqFASkoKbxI5gZ98A+fMmTNow8xiseDrr7+GTCZDRUUFVq1ahZKSEmRlZfGmvvQiglAj4DpOnDiBJ598EmVlZU7Na5rNZpSXlyM9Pd0tbX+uEmeG0tTUxMYCuvuiQ3RoYbA6/hrBgX6IDAtGUED/v5MRa5xtd2aKLCYC2p4dsPHCqPbOzqZ7C0asyczMHCQgMAw0VxaJRG4pshyFpmnU1NRAJBLxwuyboaOjA2vXrsX+/fsRGRnJGtLNmDEDRUVFWL9+PeeSNzgOP/7wkw+hBhO4ip07d0Kv1+Ppp5926pqt1WpRW1uLnJwct4jYA8WZ3t5eu9MJR4NJsIqPj7fLVNlgpnCZvFqo8REBcz0k1AD9Mevj8Tpk6oaOjg7QNM12L7tzpIqvYg1BEOx49HBijV6vZ7uXQ0JC2O5lb3bZ8jGREwBOnjzJjmJeunQJMpkM33zzDZYsWYKioiLccMMNHplAmEAIQo2Aa3n//ffxt7/9DR9++KFTD5qMgJCZmekS9/Oh5qvh4eGIj493ub8H0+3hbgFBpdah12RfYpMIwJQQf0RMCUSA/9UXRoPBwJ7r4QSEoTAmaiRJQqfTuaTIchS+ijWMURxzrgf699hsNrbI8obJ4EjQNI26ujpYrVbenGuSJPHGG2/gzTffRGJiIrZv345bb71VMKRzHu7/0ScnQg0mcBU0TePOO+9EcnIyHnroIaeOoVKp0NTU5DL/PXeIM8O9hkKhwLRp08Ycn7fYaFxqt1z1dU8LNYDjXocWi4UNZfBW3cCkRE6bNo1XYg1Jkmhubma97Bj/no6ODtZcOS4ujlMiAhODzZdzzXRjv/jii/jyyy+xZMkS3HHHHVixYoVXNx55jiDUCLiel156CWfOnMHBgwedEkOYh1qpVOpUHBsjzpAkCbVajbCwMIjFYpeLM0Nf89KlSwCA2bNnu+2htqtbD61+9NQmH5EIYVMCEDElAL6+o/97mXM90mgORVHo6upiTdSio6MhkUhcbqLmCHwVa3Q6HcrLyxESEjKoyOJyhweTcNbX14d58+Zx0selu7sbR48exaFDh6DX67Fp0yZcf/31+PWvf429e/di6dKl3l4in+HHh2vyIdRgAsNitVqxadMmrFy5Er/85S+dOoZSqWS9PZy55g/d1HGX+epAmPH5pKSkUcdFaJrGhVZuCDVAv4DQ0tKC7OzsYUUCZlOHJElYLBanQhlcDSPWTJ8+3Su+ks7S1taGxsZGBAQEuMy/x90wIqRYLOZkIidN02huboZcLsexY8cwdepUFBUVwd/fH8899xyOHj3Kq/cIBxGEGgH38Oijj0Kj0eAqIr5VAAAgAElEQVT555936sbMRBna647vDXFmuDVcvHgRgYGBSE1Ndctr6HqN6OwxDfs9P18fRIQGIDQ4AD4+9p9zZjSHMRKkaZoVZ7q7uxEVFQWxWOx2EzVH4ItYMzTKMTIyEmq1GpmZmYiIiPD28uymqakJ3d3dmD9/PifEGoPBgJMnT0Iul6OpqYmN0x6YAKJUKpGXl4fdu3dj9erVXl4xb+HmB0tAqMEERsRgMGDNmjW4++67sX79eqeO0dzcDK1Wi/nz59t1f/WGODMUi8WCc+fOYdasWaMasda2mWEd0pjs6wPMmeadXf+hwRRMKANJkjCZTIiLi4NYLPaK0fNI8EWsMZvNbBcSTdOYMmUKuru7kZOTw5suD5vNhsrKSsTExGDmzJneXg6A/m6wsrIyHDp0CAEBASgoKMCmTZsQExPD/szXX3+NHTt2oKysjFdJohxDEGomEy0tLbj99ttZL4zt27fj/vvvh0ajQUFBAZqampCUlASZTDbuUQGaprF9+3ZMnToVv//97506hkajYQ3XhlO8uSDODLcmR83tHKHPaAapMQz6WqC/LyJCAxES5Od0QdTT04PKykpERkZCp9MhIiKCNVHjqgjCjObYbDbMmTOHM+sc2oXEmCQyrd58jGAE+q8fjHG2N9qDrVYra0hXXl6O3NxclJSUjLrrq9Fo0NraigULFnh4tRMGbnyoBIYi1GA8xJM1mFarRW5uLv70pz9hyZIlTh2joaEBZrN5xPsrF8SZoTBx46PdXxsIC4yWwR8hbwo1QP97o7W1FSEhIWwoA+P7x1W4KtYMTB9lupAkEgk7Isb4A0mlUt6INUwiZ1hYGJKTk72yhp6eHhw/fhylpaXo7u7Gpk2bUFhYiGnTpo34eVcoFEhKSuJVrcsxBKGGb7zyyiu47bbbnLqJM+2sOTk50Ol0WLhwIY4cOYK3334b0dHR2LlzJ55++ml0dXXhmWeeGfdabTYbNm/ejJtuugnbtm1z6hgkSbImYL6+vqwrO0mS6OzsRGhoKCvOcGW2lDG3i4uLc3mrotliRZtKDwAICfJDxJRABAU6N0c+1L8nODgYvb29yMnJ4fQ4zkC4ItYMjXIcqwuJ8WJyV0y6u2hrawNBEHYlKLgCiqLwww8/QC6X4+uvv8YNN9yAoqIiLFmyhDOf9wmOINRwE6EG8xJ8qsGUSiXWrFmD1157zSmxmqZp1NbWws/PD2lpaQB+Emc6OjrYcWixWIyIiAjObJYwmyFDEwAZrqgs6DUO/gj5+QDpHhZqKIpiY8l7e3sRGBgIiqJGHIPiIszImbdNb5mId4IgYDAY2BGxkbqQmOQtqVTqVgNmV0JRFC5cuICgoCCkpqZ65PNmNBrx6aefQi6X4/Lly1i/fj2Ki4t5FTLBcwShhk/s2LEDr776Ki5fvuwSASAvLw/33nsv7r33Xpw6dQoJCQlQKpVYvnw5amtrXbDi/g/52rVrceedd2Ljxo1OHaO1tRVKpRIRERHQaDSYMmUK58SZoThibucIFEVB021ERFgg/P0c/7cPFbqGdiExI2eeiuh0BYw/EEVRHhVrxhvlyIg1s2fPHrVNm2solUpWPHXHbDczQiiTyXDixAlkZmaiqKgIubm5vNn9mkAIlRg3EWowL8DHGqy+vh633nor3n//fad24pkuYT8/P3Zzh4vizFD6+vpQUVHBjnQPpE1jhVY/ePbJU0LNWF1IV65cYUfOuDBmbA/eEmuGnsuBXUj2vC81Gg3q6up4JdYwiZw+Pj5u88O0Wq349ttvIZPJ8OOPP2LFihUoKSlBTk4Ob96TEwhBqOEL99xzD06fPo3//Oc/LnmAbmpqwtKlS1FVVYWZM2dCq9UC6L8IREVFsf/tCnp6erBy5Uo88cQTWLZsmV2/Q9M0ent72Ydgmqbh7+8/YrweF7HX3M7dMOeSEWeY+MGRhK6uri7U1tba7Q/EBRixhqZptyv9zPty4Ll0NsrRZDJBoVAgLS1t0Gwv1+no6EBTU5PLWodpmkZLSwtrSCcWi1FUVIS8vDxOt35PArj5FCYg1GAehs81WHl5ObZt24ZDhw7ZvXFEURQ0Gg1IkkRPTw+sVisSEhI8tpPvChj/vaHBFB3dVqh6hgg1vkD6VPcINcOFMowmdDU2NqK3txeZmZm8OdeeEmuGeiiON+CCqXf5uDlps9lc5tFIURTOnTsHmUyGU6dO4brrrkNxcTGWLl3K2Q3xSYIg1PCBVatW4YsvvsC2bduwf/9+AP0fqvr6esyePdvh4/X29mLZsmV49NFHkZ+fj8jIyEFFQVRUFLq6uly2fqB/hGnVqlX461//iuzs7GF/ZjhBQSwWIzY2Fr6+vuyDuDtTlVyNxWJBeXk5UlNTPf4grtfrQRAEVCoVgoODWUHBnosus9OQnZ3Nmy4Gd4o17oxyNJvN7HskNjbWBav1DJ2dnWhoaBjXbpRKpWIN6Xx9fVFYWIhNmzbx6jxMcPhxoZ18CDWYB5kINdiXX36J//f//h/KyspGNLIfKihERUVBIpEgIiKCfZDz9saTowwXTKHptUHZZRv0c/6+wGwXCjWMoECSJLRaLSIjIyGRSOwOZWhoaIDRaERGRgZv6l13iTVDR8sjIyMhFotd5qGo1WpRU1ODrKwsr6ZpOQJN04PeI85sFDKjjUz3cnp6OoqLi7Fq1SrebNJOAgShhstQFIVly5YhMjISx48fR0FBAdLS0vDnP/8Zp0+fxjvvvIPVq1cjPz/f7mNaLBasW7cOubm5ePDBBwEA6enpbmu7Hcjly5dxyy234N1332VTkcYSZwZC0zQuXLiAkJAQr5lpOYM95nauoq+vDyRJsoKCWCxGXFycU11IjOHaSGbOXMSVYo3RaGTTAnx9fVlxxh3nwmw2Q6FQIDk5mVdFsEajwaVLlxwqcHQ6HY4fPw65XI7u7m7k5+ejqKgI06dP501BOokQ/iDcRKjBPMBEq8EOHTqEV199FaWlpWz3ACPOkCQ5Zsojk6o0e/ZsXnmrDd146jFQaOm0DvoZVwg1QwWFiIgIVlBw9EGaL8mWQ7FYLFAoFEhMTER8fLzTxxnYVc+M6UskEkRHR7tl/IYR9Pgk1gD93Vc9PT12j8rRNI22tjaUlpairKwMcXFxKCoqwsaNG4f1cxLwOoJQw1UsFgvS0tJwzTXXQC6XA+hXq1esWIGkpCSIxWL4+/vjoYcegtFotKudlaZpbN26FdHR0di7dy/79d/97neIiYlhjew0Gg2effZZt/y7KioqsHXrVjzxxBM4ceIEKIrCnXfeOaI4MxSKonD+/HnExMS43KjXnTB+JCOZ240Hg8HAijN+fn4uFRRUKhWampqQnZ3Nm5EzZpdAJBI53H01NMpRLBZDLBZ7pKuI6b5KSkoaV4HjaZjdqAULFoxoQm0ymfDpp59CJpOhoaGBNaTjUlqXwLAIfxxuItRgbmai1mD79+/HP//5T2zZsgWlpaW47rrrcPPNN49qfj8Qd9Yy7oTZeMrJyYHZ5oPGDtcINUPTR8PDwyEWi10iKHhypNuVjEesGdgJPtaYvqthxJrRahku0tzcDLVaPWoip1qtRllZGUpLSyESiVBQUIDNmzfzamNwkiIINVzmu+++w/XXXz/oa3//+9/x2muvobCwEHfffTcCAwOxf/9+ZGVl4dprrx31eN9++y2WLFkySHnds2cPFi9ejC1btqC5uRmJiYmQyWQuNzelaRrV1dWQyWT48MMP2fju2267DeHh4Q4dizHqnT59uldd5h1lNHM7RzEajaw44+PjA7FYjPj4eLcICiRJoqWlhVf+QI6INVarFR0dHSAIAlar9aooR0/CFDjeTlBwFJ1Oh7NnzyI+Ph5ZWVkA+s/rmTNnIJPJ8MMPP+Dmm29GSUkJFi5cKBjS8Qd+PBlMPoQazANMpBrMarXi66+/hkwmw9GjR5GWloYHHngAK1eudPi+ztQyWVlZvHqgZWqZzAVSNJCDPWoC/IBZCfbVT0O7PUJDQ1lBwdX3NqaW8fHxwaxZsyakWDNws9Hf35/dbPRGvdnT04Pq6mqX1OmepK2tDWfPnsUvfvELtnO/t7cXH330EUpLS6FWq3HLLbegsLAQiYmJvHkfCQhCDSehaRoikYj9f6Bf6f3kk0/wzTffYP78+bj77rtx9OhR2Gw25OXlQaVSuTRdyJVUV1fj9ttvR1JSErZs2YK1a9fi888/x969e3Ho0CGn2gytVivOnTvnFe+X8TCSuZ09mEwmttsDACvOeGKWVKlUor29nY1J5wOjiTXDRTlKJBJOFJ3MnPeMGTM4+5kejoqKCtx22234wx/+gIqKCnz11Ve45pprUFJSgmXLlvFG5BMYhFDNcROhBnMjE60Ge+edd/Dss89iyZIl2LJlC2644QY8+OCDCA4OxhNPPOHUMXU6Haqrq3llwgoA7e3tIAgCgXGZg75uj1AzNEjA3k7w8cIkIQYEBLAx6XxgNLHGZDKBJEmQJAkfHx9IJBLEx8dzYsxep9OhqqqKd2LNgQMHcPDgQezYsQMff/wxamtrsW7dOhQXF/PK60hgEIJQwwf0ej3eeecd/Oc//8GyZctw22234ejRozhx4gSmT5+O++67j21BpSiKc7vVRqMRFovlqjbZAwcO4OjRo3jvvfeceogzm804d+4c5s6dO6I5HhdxJAJ74CgORVGsOOONwogpcLKysngl1jBRhmlpadBoNCAIAnq9HrGxsZBIJJxMFbJarWy8e0JCgreXMypMe7ZMJsNXX32F+vp6PPLII7j//vsFQzr+I1R23ESowTwI32uwrq4uhIeHD7pvUxSFkpISLFq0CPfcc4/Tx7106RJycnI48YBtL83NzeihYyHy+anuDPQD0oYRavR6PUiShEqlQlBQEOv75+kaiOlKDwkJQUpKikdfezwwI93JycmIjIxku5eZelYsFnOyTmA2VefPn8/JGnEgNpsNZ86cgVwuR2VlJdra2vDWW29hxYoVnLsWCTiMINTwhYMHDyIoKAjFxcV4//338e2332LRokXYtm0bent7YTKZeNVZwvDUU0+hpqYG+/btc0rtNRgMUCgUvLiYDoQpcIZLVbJYLKw4w4ziiMViThictba2QqVSISsrixc3AJqmodFoUFtbC7PZjISEhHFFOXoSZsQvISEBU6dO9fZyBkHTNNrb21lDupiYGNaQrqurC/n5+di7dy+WLFni7aUKjA9uf0gmL0IN5mEmYg1mNpuxYcMGFBQUoKCgwKljMD52OTk5vNnAAYALzb2gRT/VXoH+IqRJ+sUmJuVRpVJ5fRRnIBRFoaqqCuHh4UhKSvLqWuzFarWCIAjU1dUhICAA06ZN40w9OxZ6vR6VlZXIzMzknB8TRVGoqKiAXC7H559/jkWLFqGkpAQ33ngjPvvsM+zatQvHjh3j1Qi9wLAIQg3XGdh6S9M0XnrpJVy6dAlLlixBXl4e/vSnP+HUqVPw9/dHfn4+duzYge7ubuj1es493A0HTdN48MEH4efnh927dzt1jPGME3mTgalKIpEIKpUKBEHAbDaz4gwXRnGG0tzcjK6uLrtd5j3NSFGOBEHAz8+PV3PeNpsNFRUViI+P54R5tlqtxpEjR3Do0CHYbDYUFBRgy5YtV7U1t7e3Iy8vDx9++CGvdv8EroIfH5TJh1CDeYiJXoP19vZi1apVePjhh7Fy5UqnjtHe3g6SJHmzgQMAV1QW9Bp/+hgF+NLwNyvZUAame5lrnUJMoEZUVBRmzpzp7eUMi81mg1qtBkEQ6OvrQ1xcHGJjY1FbW8u7ZEsuiTVMElhpaSk++ugjpKamoqioCGvXrr2qy/6rr77C448/jlOnTvFKQBW4CkGo4Rt/+ctfIBaLUVxcjF/96leoq6vDP/7xDxgMBtx///3YuXMnnnzyScyYMQNvvPGGt5drFxRF4fbbb8f8+fPx29/+1qljMONEOTk5HknocQU2mw319fVQKpUICgpCXFwcJBIJL2Zim5qa2EhALoge9kQ5MnPe/v7+SEtL48S67cFms6GyshKxsbGYMWOGx19fr9fj448/hlwuR0dHB/Lz81FYWIikpKRRz6Fer0dISAhvzrPAsAh/PG4i1GBeYiLWYJ2dnVi1ahVeeOEF/OxnP3PqGM3Nzeju7kZmZiYvrvltGiu0+p8MhSmLAVEBOreFMrgSppsiLi6OExs4QP+amNFynU43aLSceT8MHIPik1jT19eHyspKZGRkOBx+Ml5omgZBEJDL5Thy5AgiIiJQWFiI/Pz8MS0f9Ho9L54nBEZFEGr4ArOrw8w/l5WV4dlnn8W//vUv9me2b9+O77//Htdddx1ef/11L67WcSwWCzZu3IhbbrkFxcXFTh1DrVajoaEBOTk5Xm9RHYmhOw2xsbHw8fGBVquFVCrlzW4UAFy+fBl9fX2YN2+e1wozR6Mc+SrWMIVZTEyMR3bRzGYzvvjiC8hkMtTU1GDt2rUoLi726t9awCsIf2xuItRgHmai12DNzc1Yv3493nrrLcyZM8epY9TX18NqtXI6StpsNvdHafcC/qE/jYWIaDMyZvJnfJ7ptpVIJF7r3KJpGl1dXSBJElqtFlFRUZBIJIiIiBjx789XscZgMKCiosJjnphdXV04evQoSktLYTabsWXLFhQUFAijTJMPQajhK2+99Rbq6uqwZ88edse9qKgIW7duxR/+8Af252w2G2/a3vR6PVatWoUdO3Zg1apVTh2DJEm0trZyKp2Ioiio1WqQJMnuNIjFYoSFhbE3s5aWFmg0Gs6OE41EfX09zGYz5s6d67HCjIlyJEkSAQEBDs+P81msOX/+PCIjI5GYmOjy49tsNnz33XeQy+X417/+hRtvvBHFxcW49tprefWeHIhWq8Vdd92FqqoqiEQiHDx4EOnp6SgoKEBTUxOSkpIgk8kQFRXl7aVyFX58OCYfQg3mZSZiDXbhwgUUFRVBLpc71anBmPcHBAQgNTXVDSt0DovFwtYMFEX1BzKEi9HZ+9N9jbYaEO7bzdlxouFgfOymTp3qsdABmqbR09MDgiCg0WgQEREBiUSCqKgou2sps9kMhUKBlJQUxMbGunnFroMRa+bMmcNGYLuSvr4+fPLJJ5DJZFAqlbjllltQVFSE5ORk3tSpQxFqsHEjCDV8paamBhs2bMDOnTvR0tKCo0eP4u6778b27dtRWVmJy5cv45prruHFjPRANBoNVq5cieeeew6LFy926hitra1Qq9VeFT2YNlCSJNHT04Po6OgxTWwbGxuh1+t51bVA0zTq6upAUZRbd9FcHeU4MO4yNTWVN+ebEWvCw8ORnJzssuPJZDJ89tlnWLhwIUpKSnDTTTdxtivNEbZu3YolS5bgrrvugtlsRl9fH/bs2YPo6Gjs3LkTTz/9NLq6uvDMM894e6lchR8fjMmHUIN5mYlag3333Xe47777WJN4R6Fpmt1Q8KboYbFYWN+/4UIZevootKit7M8H+QPd7VVe7VBxBiYhcsaMGW7tttDpdGz6VVhYGMRiMWJiYpyusfkq1hiNRigUCqSnp7tEXLBYLPjyyy8hk8lQXV2N1atXo7i4GAsWLOBNXToaQg02bgShho8wOzR1dXV44YUXUFZWhv3792Pp0qWsx4tOp4NSqcQPP/yA8vJynDx5Ejt37vTyyu2jtbUV69atw4EDB5CRkeHUMRobG9HX14eMjAyPXeyYNlCCINDd3W1XG+hQ6uvrYbFYMGfOHN5cpGmaRm1tLUQiEWbPnu2ydQ/dBXN1lCNN07hw4QICAwN5J9ZUVVUhNDTUKaNemqbR0NDAGtIlJyejsLAQ69at45UZ91h0d3dDKpXi8uXLg/626enpOHXqFBISEqBUKrF8+XLU1tZ6caWchh8fismHUIN5kYleg/3zn//Enj17UFZW5pTHBUVRbGKhpzo9gH7RQqVSgSRJmEwm1vdvuFCGPhOFxo6fhJrgABESY31QXl7udtHD1VgsFigUCiQlJbl0nKivrw8EQaCjowPBwcGQSCSIjY11WYcYX8Uak8kEhUKBWbNmITo62uHfpygKZ8+ehUwmw5kzZ7Bs2TIUFxfj+uuv52338nAINZhLEIQavsLMS3/zzTdQKpXYsmULfv3rXyM1NRX5+flITk7Gq6++im+//RYkSSI3NxcPPPCAyx5y3U1NTQ0KCgrw4YcfOrUrQ9M0Ll26BB8fH8yaNcsNK/zpdbRaLUiSHJQw5Egb6NDj1dbWwtfX163rdjWu6lBhCi2CIGCxWNweTc5nsebChQsIDg5GSkrKmOumaRokSbJx2mFhYawhnTtaeLmAQqHA9u3bkZGRgYqKCixcuBAvvfQSpk2bBq1WC6D/vERFRbH/LXAV/PhATD6EGszLTPQa7N1338W7776LDz74wClzXavVynqRuPMh3GazobOzEwRBwGAwIC4uDmKxGKGho/vNmK006pQW9r9DAkRIFvvDarXi3LlzvBMPGO+X8a7baDSy4oy/vz+bfuWuDluz2Yzy8nKkpaXxKt6eEWvsXTdN06iqqoJMJsOnn34KqVSKkpIS/OIXv+BcspirEGowlyAINXxmYGykSqXCPffcgyeeeAIZGRnw9fXFzp07IZPJ8NRTT2H9+vWcjHoejX//+9+45557cPjwYad2CWiaRnV1NUJDQ5GUlOSydTHxzyRJQqPRIDw8nJ3RdYUazqx7ypQpLhlv8RSM6MGIB/bCFFokSbJRjp5Mv2LWHRQUxKm5+rGwR2TSarU4evQo5HI5TCYTa0gnkUi8sGLP8uOPP+Laa6/FmTNnsHjxYtx///0IDw/HK6+8MqgoiIqKQldXlxdXymkEoYabCDUYB5joNdjevXtx9uxZHDhwwKnahnkIT09Pd+mGwMBQBr1eP0icsXezhaJpXGwdINQEipAc7z9o3bNnz+aVdwazbkc7PRiDZZIkIRKJWHHGU+lXfBVrmHWnpqYOK47RNI3GxkaUlpbi+PHjmDFjBoqKinh5LXAGoQZzCYJQM1H44YcfcNddd0GhUEAkEuH//u//sG/fPjz88MO47bbbBv0sk1rAB06ePIndu3fjyJEjY+6QDAdFUaisrERcXBymTZvm9DpomoZOpwNBEFCr1S6Z0R0NxjskKiqKV+Z2zK5BWFjYqOKYPVGOnoQRx4KDg3kn1lRWVuLkyZN4+OGH4ePjg76+Ppw4cQJyuRxtbW3Iy8tDUVERrzqGXAFBELj22mvR1NQEAPjmm2/w9NNPo76+Xmi7tZ/J84bhF0INxjEmag32hz/8ATqdDs8884xT9w/G02PevHkICwtzeh1DQxliYmIgkUgGhTI4Sk2bGbb/JnQPFGqA/o6J8vJyj6X8uApm3WN5qFgsFnR0dIAkSdhsNlacCQoK8uBqf4LPYs0nn3wCs9mMzZs3A+gPNTl06BAOHz6MkJAQFBYW4tZbb+WV6OcKhBrMJQhCzURi165daG5uhlKpBEEQeP7557FgwQKcP38ejY2NEIlE2LZtGwB+FQrvv/8+3n77bchkMqcUfsYZf8aMGYiPj7f792iaRm9vL0iSRGdnp13xz66EiWQWi8W8MrcbSWQa6uETHR0NsVjskIePO+GrWGOz2bBt2zaYTCaEhobi/PnzrCFdVlYWJ86tt1iyZAkOHDiA9PR07Nq1C3q9HgAQExPDGtlpNBo8++yzXl4pZ5m8bx5uI9RgHGQi1mA0TeOuu+5CYmIiHn74YaeOodfrUVlZiaysLIc6CSiKYmsGJpTBlTVDPWGBydL/UZoSKEJS/OARFIPBAIVCgfnz5zu1UegtGHFsqMhks9nY0XKTyYT4+HhIJBLOeNPxVaxRqVRYu3YtbrzxRtTU1ECv12Pz5s0oKChAQkKCUIMJNdh4EISaicDAG/7zzz+P9957Dy+//DKSk5Px1ltvYe/evdi3bx+OHTuGlJQUPPnkk15eseO8/PLL+Pbbb3Hw4EGnihtmfjctLW3MllBGnFGpVG4xUHMERmSaPn06r8ztmE6m6OhoREREjCvK0ZPwSayhKArff/89ZDIZvvnmGwQHB7NRh3yJg3U3CoWCTRtISUnBW2+9BYqisGXLFjQ3NyMxMREymcwpQ8BJAvc+pAKAUINxioleg1mtVmzevBkrVqzAL3/5S6eO0dPTgwsXLiA7O3tUnx5mQ4ckSWi1WkRFRUEsFiMyMtLlNUNThwV608hCDfCTyLRgwQKPjWO7goFR0mazGQRBsKPl9nj4eAtnx7e8gcFgwMmTJyGXy9Ha2gqVSoWHHnoI9957LyfrW28g1GDjRhBqJgoDCwUmm/7ll1/GmTNnAABpaWl4/PHHsWHDBjz//PPIzMwc94Vk27Zt+OijjxAfH4+qqioA/TtKb7zxBusps2fPHqxZs2Zcr8Pw2GOPobOzE88//7xTax+tlZVxt1epVAgMDIRYLEZcXBwnIooZUz5XO/q7C6YTSalUorW1FaGhoUhOTnbbmJirYcSakJAQp1KV3AmzNsaQbsGCBSguLsaKFSvg5+eHRx55BDqdDvv27ePFuRbgPEK1yU2EGoxjeLoG83T9ZTQasWbNGmzfvh0bNmxw6hgajQb19fXIzs4eZKDK+P4RBIGuri6Pbei0qq3o7uuffQoNEiExbnhTV51Oh+rqakilUq+NBjkC04nU2tqKzs5OiMVizJw5c1xjYp6Ey2KN1WrF119/DZlMBoVCgdzcXJSUlCArKwu9vb3YsGED7rnnHhQUFHh7qQITA0GomUgMNLYDgLvuugt33HEHfv7zn2PLli3Q6/WYO3cunn/++RF/xxFOnz6N0NBQ3H777YMKhdDQUKdbZEeDpmncfffdEIvFTsdcMq2sCxYsgI+PD0iSREdHB/z8/CCRSBAXF8dJB3Yu37gY9Ho9ez6ZTqSoqCicP38eU6dO9WhM53hhvHamTJnidbGGpmk0NTWhtLQUx44dw/Tp01FUVIQNG/dujpkAACAASURBVDZc1UZO0zQeffRRZGRkXOWLICDgBNyv6icnQg3GQTxZg3m6/gL643Zzc3Oxa9cuLF261KljdHR0oLm5GVKpFHq9nu22DQ8Ph1gsRnR0tMc2GUitFZ26sYUaoN+Yv6amZsyOIG/BJJASBMF2IkkkEvj5+eH8+fO8G98abwS2K6EoCj/88APkcjlOnz6NG264AYWFhViyZMlV3cu9vb3Iz8/Hm2++iRkzZnhpxQITiBFvDt5vIxBwmKE3e5vNhs8++ww///nPIZPJcOedd7KdJDt27EBeXh6WL1/udKGwdOlS1iTKE4hEIuzbtw9btmzBgQMHcNdddzl1jNjYWHz//fcICwtDQkICpFKpx9ztnSUgIABSqZRz5nZDoxwlEgkWLVo0qBOJWbePjw9vxrdEIhEyMzNRVVWFxsZGr6RvkSSJw4cP4/DhwwgKCkJhYSE+++yzUYsWkUiEP//5zx5cpYCAgIAA4NkazNP1FwBERETgyJEjWL16Nfbt24esrCyHfp+maQQFBcHHxwenT59mEx5nzZrllQ5QP1/7z3lkZCRmz56NioqKqzqCvAVN0+jp6QFJklCr1WwCaXp6+qDzOX/+fJw/f55X41uBgYGQSqVeE2tomsbFixchk8lw8uRJzJs3D8XFxdi7d++ozwuhoaE4efIkLzqXBPiNINTwGOam/+KLL2L16tUICQnBI488gjfffBMA8PDDD+Nvf/sbwsPDERsbi8zMTJca2/31r3/F3//+dyxatAgvvPCCS53OfX198d5772HdunWIiYnBLbfcMubvmEwmttMDAMRiMebPn4/Lly97NIJwvAQGBiIrKwsKhQKZmZnjSlAYDyaTCR0dHSAIghVfRhO7fH19WbFGJBI5ZOjsTUQiEebNm4fq6mqPiTXd3d04duwYDh06BJ1Oh02bNkEmk2Hq1Kl23/iFAkFAQEDAe3izBnNn/QUAEokEpaWlyM/Px3vvvWdXx2lvby8IgmBDGaZPn47IyEj09fUhNjbWa/esgUKNPSuIjo5GcnIyFAoFsrOzvTYaP/B8TpkyBRKJBGlpaSO+f0JDQ5GZmYnKykpIpVLOmAePBSPWMFHp7hZraJpGS0sL5HI5jh07BolEgqKiIjz++OMOdSMJNZiAJxBGn3iOzWaDr68vrly5gnfeeQf33HMPYmJisH79ethsNjzyyCMwm8248847cerUKaeNU5uamrBu3Tq29ZYkSfbG+9hjj0GpVOLgwYOu/KcB6Demy83NxWOPPYbly5df9X2z2cxGD1IUNWz0YGdnJxobG716w3UGb5jbDRflKBaLHWoBZgydU1JSEBsb68bVuhaKotjIcXeINUajkTWka2xsZOO0Z8+ezesbflJSEsLCwuDr6ws/Pz/8+OOP0Gg0KCgoYD0cZDLZpIus5Bn8fQNObIQajON4ogbzVv0FAOXl5di2bRsOHToEiURy1feZsSYmlIHx/Rs4KlJXVweKorx2r9ObKDR1WAEAYcEizIy1r0uGIAi0tbVBKpV6zLi/r6+P3XAMCgoa9nyOBWPonJWVxRuxBrA/ctxZVCoVysrKcOjQIfj5+aGgoACbN2/mVfLUcAg12IRA8KiZyDA7NDRNo6+vD8uXL0dGRgbeeustVnkvKCjAAw88gOuuu86p1xhaKNj7PVdAkiRWr16NV155BdnZ2VCpVOjs7ITBYIDVakV8fDzEYvGoNySCINDe3g6pVMor81XG3M6dN1ybzcaKM66KcuRr/KKrxRqr1YrTp09DJpOhvLwcK1euRHFxMbKzs3n1PhyNpKQk/Pjjj4NEuUceeQTR0dFsLGNXVxeeeeYZL65SYAwEoYabCDUYD3B3DebN+gsAvvrqK+zcuRNHjhxBREQEampqEBAQAJVKhYCAANb3b6SNMGa8JCgoyCtecGYrjTqlBYBjQg0AtLW1oaOjA1lZWW67ZxuNRpAkCZIkXeajyHjt8MUYmcHVYo1Op8Px48chl8vR3d2NW2+9FUVFRZg2bRqvN8gGItRgEwLBo2Yiw9w8RCIRamtrsXDhQrz++usA+tX5jz76CAaDwaXthEqlkjWNLSsrQ2ZmpsuOPRSxWIy3334b+fn5iImJQU9PDx5//HGsWLHiKpPVkZBIJLBYLOz8Ll8u0GFhYZg7dy47L+0qczuKotDZ2TkoyjEtLc1lJnQBAQHIzs5mPWv4ouT7+PiwnjXMToSjUBSFH3/8EXK5HKdOncLPf/5zbNu2bVhDuonK0aNHcerUKQDA1q1bsXz5cqFIEBAQmJB4ugbzZP0FADfeeCO2bduGtWvXwmQyITw8HPv27UNOTo5dYoJIJMLcuXNRWVmJlpYWj5uv+g3QV0QOatLTpk2D1WpFVVUV5s+f77La0Ww2s+IMgDFHyx0lMjIS6enpqKiogFQq5aQx8nAM9KxxVqwxmUz49NNPIZPJ0NDQgA0bNuCll15Ceno6b2r/8SLUYBMHoaNmAjHUqE6n0+Hw4cN4++23cf/992Pjxo1X/Y5OpxvTA6WoqAinTp1i4/92796NU6dOQaFQQCQSISkpCfv373d52o9er8dHH32EDz/8EC0tLbjmmmtw7tw5vPvuu06/1uXLl2E0GjF37lxeXbA1Gg3q6ursLoyGg6IoaDQakCSJnp4exMbGQiwWuzXKkdkdmTNnDiIjI93yGu6A6awJDw+3S6yhaRo1NTWQyWQ4ceIE5s6di5KSEuTm5vLGG8lZkpOT2XjV//3f/8X27dsRGRkJrVYLoP/cREVFsf8twEn4czGcXAg1GI9wRw3mrfoLANrb2yGXy3Ho0CEEBgYiISEB3d3deOedd5waI6coCuXl5Zg2bdqwY1Tu5GKrGRQNhAf7YEas42tvaGiA0WhERkaG0/WSxWKBSqUCQRCwWq3saLk7O16Y2jE7O5tXtYjRaIRCobC7drRarThz5gxkMhl++OEH3HzzzSgpKcHChQsnTPfySAg12IRAGH2abFgsFuzduxfHjh3D7t27cdNNNwHAICO7K1euYN++fbj++uuRl5fnzeUOy8svv4zOzk4UFBRg3rx5APqjKh9++GEcOXLEqQd/mqZRW1sLX19fzJo1y9VLdisqlQpNTU0Oee2MFOUYERHhMaGKueFyKcXKHiiKwvnz5xERETGsWEPTNFpbW1FaWoojR45ALBajsLAQeXl5XjOA9gZtbW2YNm0aOjo6sGLFCrzyyivYsGHDoKIgKioKXV1dXlylwBgIQg03EWownjIRarDt27dDKpXi1ltvZZMc//znP6O+vh5//etfnaohrFYrzp0753EPu3qlGSYrEBHig+kxjgs1NE2jrq4ONE075LVjs9mgUqlAkiSMRiM7qm9vN7gr6OzsxOXLlzmTYmUvY4k1jPAnk8nw1VdfYfHixSgpKcGyZcsmTfcyINRgEwRBqJmMfPDBB0hMTGRnogcWCPX19Th48CAuXbqEZ555xmmTYW9w9OhRvPjiizh8+LBTPio0TbM+JM6MtngTe8ztmChHgiCg0WjYKMeoqCiv7SwYDAavp1g5A0VR+Oabb3Du3Dns2LEDQH/RwxjSiUQiFBYWYvPmzbwyTnYXu3btQmhoKN544w2cOnUKCQkJUCqVWL58OWpra729PIGREYQabiLUYDxmItZgNE3jgQceQGBgIHbt2uXUMcxmM86dO+fRTtumDgv0JtppoQb4yWsnMDBw1L8XRVFQq9UgCAJ6vR6xsbGQSCQuGy13ho6ODly5coV3oRpGoxHvvfceZs6ciRUrVoCmaVy6dAkymQyffPIJZs+ejZKSEqxatYo3413uRKjBeIsg1ExmKIqCSCSCSCQCRVG4ePEi9u/fD51Oh2effRZxcXEuje32BAcPHkRZWRnee+89p1twKyoqIBaLMXXqVDes0H2MZG43XJRjTEwMZ/6ufX19qKiowPz5871asDiKyWTCxo0bMWPGDKjVamg0GtaQbvr06bwaoXM1er0eFEUhLCwMer0eK1aswOOPP44vvvgCMTExrJGdRqPBs88+6+3lCozM5H0TcxuhBpsATLQajKIo3HbbbcjOzsZvfvMbp45hNBpRXl7usXqgVW1Fdx81LqEGGHmjj6ZpaDQaEASBnp4exMTEQCwWIzw8nDM1AkmSaGlpgVQq5ZVY09DQgPz8fKxevRpnz55FTEwMioqKsHHjRoSHh3t7eV5FqMEmDIJQI9DfclpZWYl9+/YhMDAQzz33HEJCQth4Sb7xzDPPoKqqCq+//rpTN0KbzYby8nLMnDkT8fHxblih+2huboZWq0VqaipIkoRKpUJQUBAkEgliY2M5+/f0RuS4s5hMJnz++eeQyWSor6+HyWTChg0b8NRTT3Gm8PI2ly9fxi233AKg//pSXFyMRx99FGq1Glu2bEFzczMSExMhk8lcamYu4HKENzQ3EWqwCcREqsHMZjPy8vKwefNmFBYWOnUMph6QSqVuj5EmtFaodRQiQ3wwbRxCDdAvVFVWViImJgZhYWEgCAJdXV2IjIyERCJBZGQkZ2sEpVLJJqBy/T2nVqtx5MgRtnu5oaEB+/fvx+rVq729NM4g1GATBkGomexQFIXTp0/jb3/7GxISEvCnP/0Jfn5+vCwQGGiaxkMPPQSRSIQnn3zSqWNYLBY2RpovFzEmyvHKlSsAgNTUVMTHx/Nm9lin06GqqgpZWVkendO2B5vNxhrS/fvf/8YvfvELlJSUYNGiRbBarSgsLMTy5ctx3333eXupAgKuhJtPFQJCDTZBmIg1mF6vR25uLh566CHk5uY6dYzu7m5cvHjRpamWw9Gps4HU2hA5xQfTosfXUaPT6aBUKtHW1oawsDCkpKR4dbTcUdra2kCSJLKysjj33uvt7cXHH3+M0tJSqFQq5Ofno7CwEImJiWhtbUVeXh5ee+01LF682NtLFRBwJYJQIwDccccdOH36NBoaGgD0P/AzbvNVVVXQ6/VYvHjxVckFXIaiKGzduhXz5s1z+uGZSSbKyMjgbBvlcFGO8fHxuHLlCiiK4l3sYE9PDy5cuICsrCy376SNBUVRUCgUkMvl+OKLL3DNNdewhnRD24PNZjMKCwvx+OOPQyqVemnFAgIuhz8Xj8mFUINNICZiDaZWq5Gbm4vnn38e11xzjVPH0Gg0qK+vd6vZbXefDa1q54UavV4PgiCgUqkQEhLCds5UVFQgMTGRd13Zra2tUKlUV43QewOz2YwvvvgCMpkMNTU1WLt2LYqLizFv3ryrPgfNzc3YunUrTpw4IXjSCEwkBKFGoP9Gs3TpUhQWFuJ3v/sd+3WbzYba2lr8+te/xiOPPII1a9bwqlCwWCzYuHEjNm7ciJKSEqeOwfincGkkZ2CUo81mY9MCBkY5MrHQ/v7+SEtL8+JqHYfZSZNKpW6NpxwOJsFBLpfj448/xqxZs1BUVIQ1a9aMuRY+eQkICNgJPy72kw+hBptATNQarKWlBevXr8ebb76JuXPnOnUMkiTR2trqtpEcvZFCk8qKqCk+mGqnUGMwGEAQBDo6OhAYGMiOlg/cwGG6slNTUxETE+PydbuT5uZmdHV1Yf78+R6vaWw2G7777jvI5XKcPXsWN954I4qLi7F48eIx1yLUYAITEEGomexYrVb4+flBqVRi/vz5eO2117B582bcd999CA4OxvLlyyGVSrFu3TocOnSId2lIfX19WLVqFR544AGsWrXKqWPodDpUV1d7RThgYKIcCYKAyWSyK8qRpmlUV1cjNDSUd383rVaLmpoat7c9A/3nSalUorS0FGVlZYiKimIN6fgUGy4g4Ab48UQ4+RBqsAnCRK/BLl68iKKiInz44YeYMWOGU8doa2uDSvX/27vzsCbPtG3gZwAZRJBNNhFEBBQVAi6j07pgreKCiqiQgBWtVm2njq9aaHU6VTvtvC4dtYdoO+NS16IJLiiHVasjr627I0FxRZayCJFVQEEgeb4//MjouBKBhHD+/lJJ7lxBSM5cz70Uws/Pr9E/iD+qFXCnoPaVjZpHjx5pZi8bGxtrZi+/bKZPTU0NkpOT0a1bt2Y7xaqxZGVloby8HL6+vk3eGKzf20cul+P48ePo3bs3IiMj8c4777SozY2JmgAbNQTNWuiMjAzcvXsXu3btwvnz57FmzRrMnz8fM2fOxPnz5/GXv/xFMzujJXWuS0pKEBQUhOXLl2uOw2yosrIy3Lp1C7179262PV/UajWKioqgVCq1Psrxyc3ttA1JulJSUoK0tDQEBATA1NS0Scav35Curq4OYWFhCAsLg6OjY6M/VnNTqVTo27cvXFxckJiYiMzMTEgkEhQXF6NPnz7YsWNHk3xPyeCwUaOfmMEMiKFnsHPnzuGPf/wj9u/fjw4dOmg1RlZWFiorK5+77OVNqNQCbubVwtbCCM42zy5pvnfvHpRKJdRqNZycnODg4NCgi0fV1dVQKBR6vYT+RTIyMvDw4cNG/54Djy+QpaenIz4+HocOHYKHhwekUimCg4N1dkG0MTGDUSN54S9ey3j1p0ZhbGwMtVoNDw8P+Pv7o6ioCCtWrMCQIUOwZMkSfPPNN3BwcICnpyeuXr2K8vLyFhMQAMDW1hb79+/HwoULce3aNa3GsLa2hoeHBxQKBerq6hq5wv9Qq9UoLi7GtWvXcP78eZSVlaFz587o378/PD09G3xcpZGREfz8/FBYWIj8/Pwmqrpp2NrawtPTEwqFArW1tY0y5oMHDyCXyxEWFoaQkBCUl5djy5YtOHXqFObOnWsQTRoA+Pbbb5+aav7pp59i/vz5uHPnDmxsbLB582YdVkdERPUMPYMNGDAAX331FSQSCSorK7Uaw93dHaamprh9+zZecSG5QYyNRDB64qNQXV0d7t69i+TkZE3e69mzJ/r16wdXV9cGz/A1MzODn58frl27pvVz1xUPDw+YmZnhxo0bjfI9FwQBBQUFiI2NxbvvvotPPvkE7u7uOHnyJPbu3YtJkyYZRJMGYAajpscZNa1YQkICvvzySwQFBSEjIwOOjo744IMP4OHhgf79+2PhwoWYNm2arstssFu3biEsLAx79uyBm5ubVmPk5+cjPz8f/v7+jRaUBEHA/fv3m/Qox5Z85HhhYSEyMzO13lCwpqYG//rXvyCTyXD9+nXNhnS9evVqMWv9GyI3NxdRUVH485//jNWrV+PQoUOwt7dHQUEBTExMcPbsWSxduhRHjx7Vdamk/wzvF8QwMIMZMEPNYLt27cKOHTuwe/durWYTCIKA69evw9zcHF26dGm0utLu1sBIqEZlUSaqqqo0S8sbc1/CyspKXL16VS9PtXyZ+n373uRwirKyMiQkJEAul+PRo0cICwtDeHg4nJycmqBi3WMGo0b0wl84LgpspQRBwPjx42FlZYXY2FjY2tpCKpWiV69e2LdvH0aOHPlMQGgpx0h269YNmzZtQmRkJPbt2wd7e/sGj+Hs7Iza2lqkpqa+0drd+qMcCwoKUFxcjPbt28PJyQne3t5NcqXM2NgYYrEYycnJMDY2blGb29nb20MQBCgUCgQEBLzWmmWVSoWzZ89CLpfjzJkzCAwMxNy5c/GHP/yhRV2J1Mb//M//YOXKlaioqADw+PQNa2trzfetU6dOyMvL02WJRET0HIacwSIjI1FUVITZs2dj8+bNDX4vFolE8PHxwdWrV5Gbm4tOnTppXUv97GWlUgmVmTNMjQV4eHjAwsKiSS7gWFhYoGfPnrhy5YpO9ztsKJFIBC8vL9y6dQtpaWnw8vJ6re/Pw4cPceTIEcjlcuTl5SEkJAQbN26Eh4eHQV4gexIzGDUHNmpaqfoX0MDAQJiZmcHU1BS9e/cGAMTHxyMsLAzA493809PTERgY2CICQr1+/frhf//3fyGRSHDgwAFYWlo2eAw3NzfU1tbi5s2b6N69e4PedCorK1FQUICioiLNUY6enp7N0jxo06YN/P39kZycDCMjI9jY2DT5YzYWBwcHqNVqpKSkvPD0B7VajdTUVMhkMvz8888ICAhAZGQkNmzY0Gz7CulaYmIiHBwc0KdPHyQlJem6HCIiagBDz2Dz5s3DvXv3EBMTg1WrVjX4Q7uRkRF69eoFhUKBNm3aNGi5siAIKC0tRUFBAe7fvw9bW1u4urqivNYcbUxEsLRs2o8+7du3R/fu3ZGSktJke+81BZFIhG7duuHGjRtIT09/4UmitbW1SEpKgkwmw9WrVzFy5EgsW7YMYrHY4Jsz9ZjBqLmwUdOK1R//OGDAAACPPwDHxcUhMzMTffr0QWRkJCwtLZGSkoJt27ZpOuxvcmzk+++/r3mBS01NBfB4s9fw8HBkZWXB3d0dMpmsUZoLI0aMQElJCaZMmQKZTKbVqUIeHh64devWS9+06j3vKMcuXbroJFyZmppCLBa3yM3tnJycoFar8d1332H69Olo164dBEFARkaGZkM6d3d3SCQSfP3112jbtq2uS252p0+fxsGDB3H48GFUV1ejvLwc8+bNQ1lZmeZ0kdzcXLi4uOi6VCIieg5Dz2BfffUVZs2ahVWrViEmJqbB96+fIXz58mWYmJi8dIbwfy8tt7KygpOTE3x8fDTfq6rSumZb42ltba3Ze0/b5dy6UD+bKSUlBZs2bcLMmTMBPP7ZPH/+PGQyGX799VcMHjwYc+bMwdtvv23ws5efhxmMmgv3qCGcPXsWDg4OMDc3x4IFC3D58mUMHz4cpqamWLp0Kdq3b4+amhoUFxfD2dkZALQOCqdOnYKFhQWmTp2qCQkxMTGwtbXFZ599huXLl6O0tBQrVqxotOe3bt06nDp1Clu2bNGqaSIIAlJTU2FlZfXMnjfV1dWa0wJe9yjH5lRVVQWFQgFfX98Gb1Csa6tXr0ZiYiLGjx+PgwcPol27dpBKpQgNDW1Rs4SaWlJSEr755hskJiZi8uTJmDhxIiQSCebMmQM/Pz989NFHui6R9F/ruAza8jCDtQKGnMHq6uowefJkvPvuu5g+fbpWY9TU1ODy5cvw8fGBlZWV5t8FQXhq9rKlpSWcnJxga2v73OZBUbkKKrUAR+vmu0Z97949/Pbbb+jdu3eLmhFVV1eHSZMmwd3dHe3bt8exY8cgFosRERGBd999V28yrj5gBqNGwD1q6MXS0tIwefJkhIaG4v79+9i6dSs6deoEV1dX5Ofn45NPPsHdu3dhZWUFsViMmJgYra/mDB48GFlZWU/9W0JCgmbqYFRUFAIDAxu1UTN37lwUFhYiOjoaf//73xtcu0gkQs+ePZGSkoI2bdrAzs5O05wRBAGOjo4Qi8V6Ob21bdu28PX1bVGb25WVleHgwYM4efIkampqEBcXh8TERLi6uraaabXaWrFiBSQSCT7//HMEBARgxowZui6JiIhewpAzmImJCeLi4jBmzBjY2tpi/PjxDR7jyRnC9XsGFhQUoLCwEObm5nB0dISHh8crGyEmxoCqmVufDg4OUKlUSElJgVgs1vtmjSAIyMrKQnx8PEpKSnDnzh0MGjQI586daxH5UdeYwaixcUYNAQCuXLmCLl26oE2bNk9tfhYREYHTp09jzJgx2LBhAwYPHowFCxYgJCRE68fKyspCcHCw5mqOtbU1ysrKADx+k7CxsdH8vbEIgoAPP/wQ9vb2WLRoUYPvX1dXh4KCAqSlpcHU1BQuLi5wcnJqMRvFlZeX4/r163q7uV1VVZVmQ7qcnByMHz8eUqkUnp6eWLNmDc6fP49du3a91gbDRNRg7IDqJ2awVsLQM9j9+/cRFBSEJUuWYMiQIQ2+f1VVFbKzs5GbmwsrKyu4uLjA3t6+QZngQbUaDx6p4WDV/DkiJycHxcXF8PPz08ulQkqlEvv27cO+fftgZmYGiUSCiRMnwsLCApMnT8aoUaMwZ84cXZdJZKhemMH079WCdMLX1xeWlpYwMzNDffNu+/btmmmbBQUF+PrrrzFt2rSnrgi8otHXYCKRqElmTYhEIqxfvx43btzAxo0bX+s+KpUKSqUSKSkpuHTpEmpqatC7d2+YmJjAyspKLxseL1K/uZ1CoUBNTY2uywHweEO648ePY9asWQgMDIRCocDSpUtx6dIlfPHFF5r1+AsWLIC/vz8WL16s65KJiIganaFnMCsrKyQkJGDx4sVQKBSvdZ9Hjx4hOzsbFy9exPXr19GuXTv4+/ujrq4OdnZ2Db5wY2Isgq560q6urrC2tsa1a9ca/f9MW/fv38eOHTsQEhKCyMhIiEQiyOVyHD9+HB988AFsbW1hamoKmUyGgwcP4tixY7oumajV4eVpAoCn3pjr/2xubq45hWD79u0IDQ3Fb7/9hri4OJSWlqKoqAheXl5vtLEdADg6OiI/Px/Ozs7Iz8+Hg4PDmz2ZFzA2NsauXbsQHBwMOzs7hIaGPnOb+qMcCwoKUFlZCXt7e3h4eDx1alT9FNyePXtqdZqUrlhbW8PLy0unm9up1WpcuHABcrkcv/zyCwYOHIgPPvgAAwcOfOmU4EWLFqG8vLwZKyUiImoerSGDOTo6Yu/evZgwYQJ27tyJrl27PnOb2tpaKJVKKJVKqNVqODo6ws/P75nDIBQKhebC2esy0fGqI3d3d9y5cwc3btx4apPj5lRdXY2jR49CLpcjMzMT48ePR2xs7EuP4/7d736HvXv36v2yLSJDxKVP9EK5ubmYMGEC5s2bhylTpqCsrAxpaWno0aMHJBIJLCwssGbNGjg5OTUoKPz3tNvo6GjY2dlpNrIrKSnBypUrm+x5lZeXIygoCJ9//jmGDh2Kuro65OTkoKqqSnOUo5OTE9q3b//C5/Tw4UPNmuOWtm63/gpdQEBAsywlEgQBN27cwJ49e3D06FH4+voiIiJCs1kiEekclz7pJ2awVsxQM5hCocD06dMRHx8PZ2dnlJWVoaysDKWlpaitrYWDgwMcHR1feqKjUqlEbm4u/P39G9RAKK5Qwc5Sdw0HQRBw+/ZtiESilzZHmIjFFwAAGrNJREFUGlNdXR1OnToFmUyG5ORkjBgxApGRkfD399fLZVhErdALXwjYqKHnUqlUMDY2RnJyMubOnYvo6GiMHz8e165dQ3R0NIyMjNC1a1dcvnwZcrkcTk5OrzWuVCpFUlISioqK4OjoiGXLliEkJARhYWHIzs5G586dIZPJYGtr26TPT6lUYvDgwejevTuuXLmC6dOn4/3334eNjc1rv3FWVFTg2rVrCAgI0Orob13Kz8/H3bt3GxxyXpcgCMjOzoZcLsfBgwfRsWNHSKVSjBs3Du3atWv0x2su1dXVGDx4MB49eqQ5FWHZsmXIzMyERCJBcXEx+vTpgx07drAJRS0JGzX6iRmslTL0DHbs2DHExMTAxsYGubm5WLlyJQYOHNigfJCbm4vi4mL4+vq+dsPh/kM1rMx125wQBAHXr19H27Zt4eHh0SSPoVarcenSJcjlcvzf//0f3n77bUilUgwaNKhFz4xhBiMDxUYNNZxarYaRkRFycnLQqVMnnDt3DkuWLMGoUaMQHh6Ojh074uzZs6ioqMDw4cMhEok04UIfCYKAlJQU7N69Gz/99BO8vLyQk5OD2NhY9OzZU6sxS0tLcfv2bfTu3bvFHVeYm5uLoqKiRt3c7t69e9i/fz/27t0LU1NThIeHY9KkSbCzs2uU8XVNEAQ8ePAAFhYWqK2txcCBA/Htt99i9erVCA0N1RzJKBaL8eGHH+q6XKLXxUaNfmIGa8UMLYPV1tbixIkTiIuLg0KhQNeuXVFRUYEff/xR6ws4mZmZePjwIXr06PFaF9kePlLD/He6n0UiCAKuXr0KKysrdO7cudHGvHHjBuRyOY4cOQIfHx9ERkYiKCjIYJoWzGBkoNiooTdTWFiI4OBgREVFYerUqbCwsAAA/PDDD/jll18gEomwefNmANDboLBz504cOnQIUqkUI0eOhJmZGa5evYr33nsPe/fuhbOzs1bjFhYWapYS6ePzfpmsrCxUVFSgV69eWk/BLS8vx6FDhxAfH4/y8nJMnDgREokELi4uBn2c9sOHDzFw4EB89913GDNmDAoKCmBiYoKzZ89i6dKlOHr0qK5LJHpdhvuL2rIxgxEAw8hgc+fOhbGxMSQSCfr37w+RSIRNmzbh4MGD2Llzp1ZLseuXEhkZGcHLy+uVt6+pE2Bqoh8vd2q1GikpKXBwcICLi4tWYwiCgNzcXMjlciQkJMDR0VEze7kl7aGoDWYwMiBs1NCbS09Ph4ODg+bFv7y8HFKpFPPnz8fJkydx69YtxMfH67jKhvvll1+wcOFC7N+/HzY2NlqNcffuXSiVSojF4ha35jc9PR2PHj1q0OZ21dXVOHbsGORyOTIyMjBu3DhERETA29vboJszwOMQ3KdPH9y5cwd//OMfER0djQEDBuDOnTsAHh/DOWrUKM36f6IWwLB/aVsuZjDSMNQM9re//Q23bt3Chg0btMoPgiDg2rVrsLCwgLu7+ytvq08ZRaVSQaFQwMXF5bWXrwFAUVGRZvaykZERwsPDMXnyZHTo0KEJq9UPzGBkgHg8N2mvvpnXtWtXTUDIy8tD+/bt0bdvXyQnJ+Prr79G27ZtcfHiRV2WqpVBgwbh888/h1QqxcOHD7Uao2PHjrC1tdWroxdfl4eHB4yNjXH79u2X1l5XV4ekpCR89NFHGDx4MC5cuIBFixbh3//+N5YtW4Zu3brpVQBqKsbGxlAoFMjNzcWFCxdw8+ZNXZdEREQGytAz2KJFi2BjY4OlS5dqdX+RSIQePXqgrKwMeXl5r7ytPjE2NoZYLEZOTg4KCwtfetuKigrExcUhNDQUYWFhePToEXbt2oWTJ0/iww8/bBVNGoAZjFoXNmrolf77jS0pKQmzZ88GAEyaNAk5OTkAgK1bt6Jfv37NXl9jGDduHKZPn45p06ahtrZWqzE6d+6Mtm3b4tatWy2qWSMSieDt7Q2VSvVMyKvfkC4mJgZvvfUW9u/fj6ioKKSkpGD16tXo27dvi5tB1Fisra0xdOhQnD17FmVlZairqwPweO8fbacxExERPcnQM5hIJMLq1atx9+5dxMbGajWGkZERfH19kZ+fj3v37jVyhU3LxMQE/v7+yMjIwO3bt5/62qNHj5CYmIioqCgEBQUhKysLa9aswdmzZxETEwNXV1e9az41F2Ywag1a5ycseiOBgYFwcXHBnDlzkJaWht27dyMnJ6fFf2CfPn06Bg8ejI8//ljrRkvXrl2hVquRkZHRyNU1LZFIBB8fH6xduxZffvklbt68iS+//BIDBw7E+vXrMWLECPz73//Gxo0bMXToUL1c/94cCgsLUVZWBgCoqqrCzz//DB8fHwwdOlQz5Xzbtm0YP368LsskIiIDZYgZzMjICFu3bkVSUhLi4uK0GqN+dkpWVhZKSkoaucKm1aZNG3Tv3h0SiQQnTpzAqVOn8PHHH2PQoEE4ffo0oqOjcfnyZfz1r39t0DJ1Q8MMRq0N96ihBqk/hQAA/vSnP6GoqAjBwcGIiIjQcWWN55NPPoFarcZXX32l1f3rd/O3traGm5tbI1fXNARBQF5eHvbs2YNt27bB3Nwc0dHRCAkJMfgN6RriypUriIqKgkqlglqtRlhYGL744gtkZGRAIpGgpKQEAQEB2LlzZ5Me2a5v6+ypxeMPk35iBqOnGHoGe/DgAUaOHIn58+dj5MiRWo3x6NEjJCcnw8fHB1ZWVo1cYdNQq9VQKBTYsWMH9u7di4EDB2Lu3LkYMmSIVpssGypmMDJQ3EyYGs+TJwrU1dUZ3JuIWq3GtGnT4OPjg3nz5mk9hkKhgLOzs9anSTWH4uJizYZ0ABAeHo7g4GDMmDEDM2fORHh4uI4rpCc9GdKJGhETp35iBqNnGHoGKy4uRlBQEFatWoX+/ftrNUZVVRUUCgX8/Py0Pvq7qQmCgLS0NMhkMhw+fBheXl6aQxkiIyMhk8ng7e2t6zLpCcxg1ETYqKHG9WQ32RA7y7W1tZgwYQLGjRuHKVOmaDVGXV0dkpOT0aVLF73a5K2yshKJiYmIj49HcXExQkNDIZFI4Obmpvl/rKiowOjRoxETE4OxY8fquOLWSxAECILwTDBITU3F0aNHERkZ2aCTIohewLBewA0HMxg9l6FnsJycHIwdOxabNm1Cjx49tBqjsrISqamp8Pf3h5mZWSNXqB1BEJCfn4/4+HjNSaNSqRQTJkxA+/btNbe7evUq3nvvPSQkJKBz5846rLh1YwajZsJTn6hxPRkKDC0gAI/XC8tkMuzatQs//fSTVmOYmJhALBYjPT1ds6ZWV2pqanD48GFMmzYNw4cPR3p6OlatWoVz587h008/RefOnZ/6f7S0tMTBgwdhbW2tw6pbp9raWnz//feora2FSCR6KiAUFRUhMjISM2bMQGlpKVQqlQ4rJSIiXTD0DObq6ordu3djxowZms2SG8rCwgLdu3dHSkoKampqGrnChikpKcGWLVsQHByM6dOnw8LCAocOHcKRI0cQFRX1VJMGAHx9fbF58+ZWux+gLjGDkT4xrPmSRI3I3NwcBw4cwIgRI2BlZYW33nqrwWOYmppCLBZDoVCgZ8+ezbrfi0qlwunTpyGXy3H+/Hm88847WLhwIfr16/daUzdtbGwwaNCgZqiUgP9MZzcyMsI//vEPuLi4YOzYsVi/fj26dOmC0aNH4/bt21AqlTh//ryuyyUiImoy3bt3x8aNGxEREYH9+/drNTPZ2toaXbt2RUpKCgICApp1mdiDBw9w+PBhyOVyKJVKhIaGYsuWLXB3d3+t5lqfPn2aoUqqxwxG+ohLn4heIS8vD2PGjME///lP9OrVS6sxHjx4gCtXrkAsFsPc3LyRK/wPtVqNK1euQCaT4cSJE+jbty8iIiIwdOhQg1vHbiieN239H//4B7Zv347u3bujpKQElZWV+PjjjzF27Fj4+PjA29sbXl5eyMvLw8KFC/H73/9eR9WTgTC8S/KGgRmMWr2jR49i2bJlOHDgACwsLLQao6CgAHfv3oW/v3+T7jFSU1ODf/3rX5DJZLhx4wZGjx6NiIgI9OrVyyBnPhkCZjDSA9yjhuhN3L59G5MnT8bu3bu1Xi9cXl6O69evIyAgoFF3oxcEAXfu3EF8fDwSExPRtWtXSKVSjBkzRm/WZb+JnJwcTJ06FUqlEiKRCLNmzcK8efNQUlKC8PBwZGVlwd3dHTKZDDY2Nrou95XUajVEItEzwWDv3r34+eef8fHHH8PMzAx9+vTBgQMHMHToUKxfvx537tzB4sWLIRKJUFJSApFIhJUrV+L+/fuQyWQGuU8BNRv+4OgnZjAiAD/++CO2bt0KmUwGU1NTrcbIyclBSUkJ/Pz8GvW9UqVS4ezZs5DL5Th79iwCAwMRERGBAQMGGMTGs8xgzGDU5LhHDdGb8Pb2xubNmxEZGYl79+5pNUb79u3h7e2NlJQU1NbWvlE99RvSrVu3DsOGDUNMTAy6du2KpKQkxMfHY+LEiQbRpAEe7/Xz97//HdevX8e5c+ewfv16XL9+HcuXL8ewYcOQlpaGYcOGYfny5bou9bUYGRk982Y+bdo0bN++HQMGDICzszM8PT3h7++PmzdvAgD69u0LAPjll1/QoUMHdOjQAWq1Gm3bttV8jQGBiIgMUUREBMaMGYNZs2ZBrVZrNYarqyssLS1x48YNvOIi9SvVz17+/PPP8dZbb2HXrl2YNGkSFAoFYmNj8dZbbxlEkwZgBgOYwUh3DONVhKgZ9O3bFytWrIBUKkVFRYVWY9ja2sLd3R0pKSlabUJWWlqKrVu3YuzYsZg2bRratm2LhIQEHD16FNOmTYOVlZVWdekzZ2dn9O7dG8DjTY59fHyQl5eHhIQEREVFAQCioqJw4MABXZb5DEEQnhsoL168iCVLluDSpUtQqVS4ePEiVCoV4uLiMG3aNNjZ2QEA3nvvPSQmJgIAfHx8YG9vjytXrkClUiEqKgozZsyAvb095syZ06zPi4iIqLnNmzcP3t7eiI6O1rrR0qVLFxgZGSE9Pb3B9xUEAenp6VixYgWGDBmClStX4g9/+AMuXLiAbdu2ISgoyCCXmDODMYOR7nDpExkUd3d3WFpawtjYGCYmJrh06VKjP8aePXuwceNGyOVyrZcw5eXl4d69exCLxa+86vLw4UP89NNPkMlkKCgoQEhICKRSKbp06dLqOvhZWVkYPHgwUlNT4ebmpjlNSxAE2NjY6Px0LeDxNOjnXbEBgLVr12LXrl0YOHAgiouL0atXL4wcORJTp06FQqFAZWUlzM3NYWRkhNzcXIwdOxaHDx+Gs7MzfvjhB2RmZuKzzz5DVVWVJkwQNYLW9ULScjCDUYvS1BlMEATMnj0bzs7O+PTTT7UeIzU1Fe3bt3/lUnZBEKBUKrF3717s378f7dq1g1QqRWhoaKs8FZMZjBmMmgT3qKHWwd3dHZcuXdLqdICG2LBhA06cOIGtW7dqfXxiVlYWKisr0bNnz2feUGpra3Hy5EnIZDKkpqZqNqTz9fVtdc2ZepWVlRgyZAj+/Oc/a0LSk6HAxsYGpaWlzV5XdnY2duzYgYqKCixYsAAODg4AgMLCQnz//ffo0KEDpkyZAgCYNWsW/vKXv6BHjx64efMmhgwZAqVSCUdHRyQlJcHHxwfA440PO3TogAkTJmD06NH48MMPUVVVhbZt2z712C8LJEQNwB8g/cQMRi1Kc2QwlUqFsLAwBAYGYsaMGVqNoVarkZKSAkdHR3Ts2PGZr5eVleHgwYOQy+Worq7G5MmTER4eDicnp1b7fssMxgxGTYaNGmodmqtRAwBLlixBfn4+1qxZo/WL9MmTJ7F//36sXbsWAHDu3DnIZDKcOXMGQ4YMgVQqNai1ztqqra1FcHAwgoKCsGDBAgBAt27dkJSUBGdnZ+Tn5yMwMBC3bt1qtpouXbqEmTNnwsbGBt27d8ecOXMgFouRmZmJmJgYuLm5oby8HObm5khPT0diYiK6dOmC06dPa4Jh79698d133+HMmTO4ePEivLy8cO7cObi5uWHt2rV4+PAh7O3tn9qkjhvWURPgD5R+YgajFqW5Mlh1dTXGjBmDGTNmICQkRKsxVCoVvvnmG7i6uiIiIgJVVVU4cuQI5HI5cnJyNLOXu3bt2urfc5nBmMGoSbFRQ61Dly5dYGNjA5FIhNmzZ2PWrFlN9liCIOCjjz6CnZ0dFi9erNUY9VeGBEHA3bt3IRaLERkZiWHDhqFNmzaNXHHLJAgCoqKiYGtrq2loAUB0dDTs7Ozw2WefYfny5SgpKcHKlSubra6srCz07t0bJSUlz3ytY8eOiIiIwDfffAMA8Pf3R2xsLLZv3w5HR0f89a9/BQDMnj0b48aNw6hRo5CcnIwff/wRb7/9NkJDQ5vteRCBjRp9xQxGLUpzZrDy8nKMGDECS5YswZAhQ7QaIz8/H6NHj0a3bt2QlZWFkSNHIiIiAmKxmB/G/z9mMKImx0YNtQ55eXlwcXHBvXv3MHz4cKxbtw6DBw9ussdTqVSQSCR4++23XzuQCIKAzMxMxMfH49ChQ3Bzc0N+fj4iIiLwpz/9qclqbal+/fVXDBo0CL6+vpqZRX/729/Qv39/hIWFITs7G507d4ZMJoOtrW2z1ubm5gaFQoFLly7hn//8J/z8/PDFF19g9uzZsLKy0oSW1atX49q1a1i0aBHmzJmDoUOHIiMjA3l5eThw4MBzT+jiVRtqRvxB00/MYNSiNHcGUyqVGDVqFNatW4eAgIDXuo9arcaFCxcgk8nw66+/om/fvjhz5gzi4uLQp0+fJqu1pWIGI2pyL/xBM7ztyalVc3FxAQA4ODhgwoQJuHDhQpOGBGNjY+zcuRPBwcGws7PDxIkTX3jbJzekMzc3R3h4OI4fPw4bGxtUV1dj7NixcHJyQlhYWJPV2xINHDjwhSc8nDhxopmreVp0dDRcXFwQGhqKYcOGITIyEgAwadIkfPrppygoKICTkxOcnZ2RnZ0NT09P/PDDD4iNjUVAQABWrVr1TEBQq9Vc80xERC1Oc2cwR0dHxMfHY8KECdixYwc8PT2feztBEHD9+nXIZDIcPXoUfn5+iIiIwLp169CmTRukp6dj4sSJkMvl8PLyarJ6WyJmMCLd4YwaMhgPHjyAWq2GpaUlHjx4gOHDh+OLL77AyJEjm/yxKyoqEBQUhMWLF+Odd97R/Pv9+/eRkJCAvXv34sGDB5g8eTIkEslzN6SrqKjA5MmTsWfPHoM8ZtsQFRYWon///sjIyHjma76+vujbty88PT2xefNmbNq06amfjXq8akN6gD+A+okZjFoMXWawK1euICoqCvHx8XB2dgbw+L01OzsbcrkcBw8ehIuLC6RSKcaNGwdzc/NnxkhJScHKlSuxa9euJq+XGgczGBkIzqghw6dUKjFhwgQAQF1dHSIiIpolIACApaUlEhISEBQUBDMzM9y7dw9yuRy//fYbQkJCsGHDBnh6er70zcDS0hI//fQT3zBaEHt7exgbGyM7Oxtubm4A/jP1e+LEiThz5gxGjRqFffv2wd/f/6n7qlQqGBsb8/+biIhaPF1mMD8/P6xduxZSqRSbNm3C8ePHsW/fPpiamkIikeDYsWOvXJYjFouxc+fOZqmXGgczGBk6zqghakSZmZn4/e9/j6lTp2LKlCkQi8Wt/sQmQ7dlyxZcvHgRQUFBWL9+PWpqavDtt99CEARMmTIF165d03WJRK/CpKqfmMGIGuD777/HV199hfnz50MikaBjx478IG7gmMHIAHAzYaLmYijTKN9//30kJibCwcEBqampAICSkhKEh4cjKysL7u7ukMlksLGx0XGlulVeXg5ra2vMnDkTYWFhePfddzVfCwgIwI4dO9CrVy8dVkj0Si3/BcswMYMRNRAzWOvCDEYGgI0aImqYU6dOwcLCAlOnTtWEhJiYGNja2mqOYywtLcWKFSt0XKnu1W8+V6+mpgampqbP/DuRnmr5n2oMEzMYUSvFDPb6mMGohWOjhogaLisrC8HBwZqQ0K1bNyQlJcHZ2Rn5+fkIDAzErVu3dFylflCpVBCJRAwF1BKxUaOfmMGIWjFmsNfHDEYtGDcTJqI3p1QqNScqODk5QalU6rgi/WFsbKzrEoiIiMhAMYO9GDMYGSK2HYlIKyKRyCDWgRMRERG1JMxgRIaPjRoiem2Ojo7Iz88HAOTn58PBwUHHFREREREZPmYwotaFjRoiem3jxo3Dtm3bAADbtm3D+PHjdVwRERERkeFjBiNqXbiZMBE9l1QqRVJSEoqKiuDo6Ihly5YhJCQEYWFhyM7ORufOnSGTyWBra6vrUonozXD+vH5iBiNqpZjBiFoNnvpEREREz8VGjX5iBiMiIjJsL8xgXPpERERERERERKQn2KghIiIiIiIiItITbNQQEREREREREekJNmqIiIiIiIiIiPQEGzVEeurIkSPo1q0bPD09sXz5cl2XQ0RERNQqMIMRka7x1CciPaRSqeDt7Y2ff/4ZnTp1Qr9+/RAXF4cePXroujQiMjw89Uk/MYMR6QAzGBE1I576RNSSXLhwAZ6envDw8ICpqSkkEgkSEhJ0XRYRERGRQWMGIyJ9wEYNkR7Ky8uDq6ur5u+dOnVCXl6eDisiIiIiMnzMYESkD9ioISIiIiIiIiLSE2zUEOkhFxcX5OTkaP6em5sLFxcXHVZEREREZPiYwYhIH7BRQ6SH+vXrh7S0NGRmZqKmpga7d+/GuHHjdF0WERERkUFjBiMifWCi6wKI6FkmJiaIjY1FUFAQVCoV3n//ffTs2VPXZREREREZNGYwItIHPJ6biIiodePx3PqJGYyIiMiw8XhuIiIiIiIiIiJ9x0YNEREREREREZGeYKOGiIiIiIiIiEhPsFFDRERERERERKQn2KghIiIiIiIiItITbNQQEREREREREekJNmqIiIiIiIiIiPQEGzVERERERERERHqCjRoiIiIiIiIiIj3BRg0RERERERERkZ5go4aIiIiIiIiISE+YvOLromapgoiIiIiexAxGRETUSnFGDRERERERERGRnmCjhoiIiIiIiIhIT7BRQ0RERERERESkJ9ioISIiIiIiIiLSE2zUEBERERERERHpCTZqiIiIiIiIiIj0xP8DGTMlwIg9nnQAAAAASUVORK5CYII=\n","text/plain":["<Figure size 1440x720 with 2 Axes>"]},"metadata":{"tags":[],"needs_background":"light"}}]},{"cell_type":"code","metadata":{"id":"sdG7b3tjJBnA"},"source":[""],"execution_count":null,"outputs":[]}]}