{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# CauseClusterer and EffectClusterer Hyperparameter Tuning\n", "\n", "This notebook gives a quick demo of how to tune a CauseClusterer and/or\n", "EffectClusterer directly through CFL parameter specification.\n" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "collapsed": true }, "outputs": [], "source": [ "import numpy as np\n", "from sklearn.model_selection import train_test_split\n", "\n", "import visual_bars.generate_visual_bars_data as vbd\n", "from cfl.experiment import Experiment" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Load Data\n", "We will be using the visual bars example data for this tutorial. To learn \n", "more about this included dataset, please refer to the visual bars background\n", "[page](https://cfl.readthedocs.io/en/latest/more_info/Visual_Bars_data.html).\n", "If you would like to see a standard example of running CFL on this dataset\n", "without hyperparameter tuning, please refer to the main CFL code \n", "[tutorial](https://cfl.readthedocs.io/en/latest/examples/cfl_code_intro.html).\n", "\n", "Here, we create a dataset of 10000 samples, where the cause is a 10x10 image\n", "and the effect is a binary 1D variable. For sake of simplicity, we will be\n", "tuning a standard feed-forward network (as opposed to a convolutional \n", "neural network), so flatten the images down to vectors of shape (1,100). \n", "\n", "Since we want to evaluate our models on the same subset of the data across\n", "all hyperparameter combinations, we generate in-sample and out-of-sample\n", "indices to pass to CFL. If these are not provided, CFL generates this\n", "split randomly, and it would look different across every trial. " ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(10000, 100)\n", "(10000, 1)\n" ] } ], "source": [ "# create visual bars data \n", "n_samples = 10000 \n", "im_shape = (10, 10) \n", "noise_lvl= 0.03\n", "random_state = 180\n", "\n", "vb_data = vbd.VisualBarsData(n_samples=n_samples, im_shape=im_shape, \n", " noise_lvl=noise_lvl, set_random_seed=random_state)\n", "\n", "# retrieve the images and the target \n", "X = vb_data.getImages()\n", "X = np.reshape(X, (X.shape[0], np.product(X.shape[1:]))) # flatten images\n", "Y = vb_data.getTarget()\n", "Y = np.expand_dims(Y, -1)\n", "print(X.shape)\n", "print(Y.shape)\n", "\n", "# define train and validation sets to remain constant across tuning\n", "in_sample_idx, out_sample_idx = train_test_split(np.arange(X.shape[0]),\n", " train_size=0.75,\n", " random_state=42)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Set up CFL pipeline\n", "Here, we assume we have already tuned the CDE and will focus on tuning\n", "the CauseClusterer (as can be seen from the loss plot below, this assumption\n", "is very wrong). \n", "The CauseClusterer expects the user to specify the `'model`' they would\n", "like to use (usually an sklearn model), along with any associated \n", "hyperparameters they would like to specify through a `model_params` dict:\n", "\n", "```\n", "CauseClusterer_params = {'model' : 'DBSCAN', # sklearn model\n", " 'model_params' : { \n", " 'eps' : 0.1, # dbscan hyperparam\n", " 'min_samples : 10}, # dbscan hyperparam\n", " 'verbose' : 0, # CFL verbosity setting\n", " 'tune' : False # default is False\n", " }\n", "```\n", "\n", "This syntax can be extended to tune the clusterer over a range of hyperparameters\n", "by making all model parameters iterable:\n", "\n", "```\n", "CauseClusterer_params = {'model' : ['DBSCAN'], # sklearn model (formatted \n", " # as list for compatibility)\n", " 'model_params' : {\n", " 'eps' : np.logspace(-5,2,10), # try 10 values of `eps` \n", " 'min_samples : np.linspace(10,60,5)}, # try 5 values of `min_samples`\n", " 'verbose' : 0 # CFL verbosity setting (does \n", " # not need to be iterable because \n", " # filtered out as cfl param)\n", " 'tune' : True # default is False\n", " }\n", "```\n" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "All results from this run will be saved to visual_bars_cfl/experiment0017\n", "Block: verbose not specified in input, defaulting to 1\n", "CondExpBase: kernel_initializers not specified in input, defaulting to None\n", "CondExpBase: bias_initializers not specified in input, defaulting to None\n", "CondExpBase: weights_path not specified in input, defaulting to None\n", "CondExpBase: show_plot not specified in input, defaulting to True\n", "CondExpBase: tb_path not specified in input, defaulting to None\n", "CondExpBase: optuna_callback not specified in input, defaulting to None\n", "CondExpBase: optuna_trial not specified in input, defaulting to None\n", "CondExpBase: early_stopping not specified in input, defaulting to False\n", "CondExpBase: checkpoint_name not specified in input, defaulting to tmp_checkpoints\n", "Block: user_input not specified in input, defaulting to True\n" ] } ], "source": [ "# the parameters should be passed in dictionary form \n", "data_info = {'X_dims' : X.shape, \n", " 'Y_dims' : Y.shape, \n", " 'Y_type' : 'categorical' #options: 'categorical' or 'continuous'\n", " }\n", "\n", "# the optimal hyperparameters for the CDE can be found by following the\n", "# Optuna tuning tutorial\n", "CDE_params = { 'model' : 'CondExpMod',\n", " 'model_params' : {\n", " 'dense_units' : [160, data_info['Y_dims'][1]],\n", " 'activations' : ['relu', 'linear'],\n", " 'dropouts' : [0.1, 0],\n", " 'kernel_regularizers' : [None] * 2,\n", " 'bias_regularizers' : [None] * 2,\n", " 'activity_regularizers' : [None] * 2,\n", " 'batch_size' : 72, \n", " 'n_epochs' : 100,\n", " 'optimizer' : 'adam',\n", " 'opt_config' : {'lr' : 4e-05},\n", " 'loss' : 'mean_squared_error',\n", " 'best' : True, \n", " 'verbose' : 1}\n", " }\n", "\n", "# CauseClusterer_params = {'model' : 'DBSCAN', \n", "# 'model_params' : {\n", "# 'eps' : np.logspace(-3.5,-1,15),\n", "# 'min_samples' : np.arange(30,300,30).astype(int)}, \n", "# 'verbose' : 1,\n", "# 'tune' : True\n", "# }\n", "\n", "CauseClusterer_params = {'model' : 'KMeans', \n", " 'model_params' : {\n", " 'n_clusters' : np.arange(2,11).astype(int)},\n", " 'verbose' : 1,\n", " 'tune' : True\n", " }\n", "# steps of this CFL pipeline\n", "block_names = ['CondDensityEstimator', 'CauseClusterer']\n", "block_params = [CDE_params, CauseClusterer_params]\n", "\n", "# folder to save results to \n", "save_path = 'visual_bars_cfl' \n", "\n", "# create the experiment!\n", "my_exp = Experiment(X_train=X, \n", " Y_train=Y, \n", " data_info=data_info, \n", " block_names=block_names, \n", " block_params=block_params, \n", " results_path=save_path)" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "#################### Beginning CFL Experiment training. ####################\n", "Beginning CondDensityEstimator training...\n", "No GPU device detected.\n", "Epoch 1/100\n", "WARNING:tensorflow:AutoGraph could not transform .train_function at 0xf1fc38ef0> and will run it as-is.\n", "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n", "Cause: 'arguments' object has no attribute 'posonlyargs'\n", "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n", "WARNING: AutoGraph could not transform .train_function at 0xf1fc38ef0> and will run it as-is.\n", "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n", "Cause: 'arguments' object has no attribute 'posonlyargs'\n", "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n", "104/105 [============================>.] - ETA: 0s - loss: 1.3172WARNING:tensorflow:AutoGraph could not transform .test_function at 0x1a53234320> and will run it as-is.\n", "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n", "Cause: 'arguments' object has no attribute 'posonlyargs'\n", "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n", "WARNING: AutoGraph could not transform .test_function at 0x1a53234320> and will run it as-is.\n", "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n", "Cause: 'arguments' object has no attribute 'posonlyargs'\n", "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n", "105/105 [==============================] - 3s 11ms/step - loss: 1.3166 - val_loss: 1.0101\n", "Epoch 2/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.9030 - val_loss: 0.8055\n", "Epoch 3/100\n", "105/105 [==============================] - 0s 5ms/step - loss: 0.7817 - val_loss: 0.7412\n", "Epoch 4/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.7290 - val_loss: 0.6998\n", "Epoch 5/100\n", "105/105 [==============================] - 1s 5ms/step - loss: 0.6916 - val_loss: 0.6633\n", "Epoch 6/100\n", "105/105 [==============================] - 1s 7ms/step - loss: 0.6529 - val_loss: 0.6295\n", "Epoch 7/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.6201 - val_loss: 0.5971\n", "Epoch 8/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.5878 - val_loss: 0.5661\n", "Epoch 9/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.5597 - val_loss: 0.5354\n", "Epoch 10/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.5273 - val_loss: 0.5058\n", "Epoch 11/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.4965 - val_loss: 0.4775\n", "Epoch 12/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.4689 - val_loss: 0.4501\n", "Epoch 13/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.4414 - val_loss: 0.4234\n", "Epoch 14/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.4170 - val_loss: 0.3978\n", "Epoch 15/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.3897 - val_loss: 0.3736\n", "Epoch 16/100\n", "105/105 [==============================] - 0s 5ms/step - loss: 0.3658 - val_loss: 0.3501\n", "Epoch 17/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.3458 - val_loss: 0.3283\n", "Epoch 18/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.3241 - val_loss: 0.3075\n", "Epoch 19/100\n", "105/105 [==============================] - 0s 5ms/step - loss: 0.3038 - val_loss: 0.2886\n", "Epoch 20/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.2865 - val_loss: 0.2714\n", "Epoch 21/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.2692 - val_loss: 0.2551\n", "Epoch 22/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.2552 - val_loss: 0.2406\n", "Epoch 23/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.2398 - val_loss: 0.2276\n", "Epoch 24/100\n", "105/105 [==============================] - 1s 8ms/step - loss: 0.2288 - val_loss: 0.2160\n", "Epoch 25/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.2171 - val_loss: 0.2057\n", "Epoch 26/100\n", "105/105 [==============================] - 1s 7ms/step - loss: 0.2093 - val_loss: 0.1971\n", "Epoch 27/100\n", "105/105 [==============================] - 1s 8ms/step - loss: 0.1993 - val_loss: 0.1895\n", "Epoch 28/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.1919 - val_loss: 0.1826\n", "Epoch 29/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.1857 - val_loss: 0.1771\n", "Epoch 30/100\n", "105/105 [==============================] - 1s 7ms/step - loss: 0.1809 - val_loss: 0.1723\n", "Epoch 31/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1774 - val_loss: 0.1682\n", "Epoch 32/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1748 - val_loss: 0.1648\n", "Epoch 33/100\n", "105/105 [==============================] - 0s 5ms/step - loss: 0.1719 - val_loss: 0.1622\n", "Epoch 34/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1675 - val_loss: 0.1598\n", "Epoch 35/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1654 - val_loss: 0.1577\n", "Epoch 36/100\n", "105/105 [==============================] - 1s 5ms/step - loss: 0.1638 - val_loss: 0.1561\n", "Epoch 37/100\n", "105/105 [==============================] - 0s 2ms/step - loss: 0.1623 - val_loss: 0.1550\n", "Epoch 38/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1602 - val_loss: 0.1536\n", "Epoch 39/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1599 - val_loss: 0.1525\n", "Epoch 40/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1578 - val_loss: 0.1518\n", "Epoch 41/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1556 - val_loss: 0.1510\n", "Epoch 42/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1551 - val_loss: 0.1500\n", "Epoch 43/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1544 - val_loss: 0.1496\n", "Epoch 44/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1527 - val_loss: 0.1492\n", "Epoch 45/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1525 - val_loss: 0.1486\n", "Epoch 46/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1518 - val_loss: 0.1483\n", "Epoch 47/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1513 - val_loss: 0.1478\n", "Epoch 48/100\n", "105/105 [==============================] - 1s 5ms/step - loss: 0.1514 - val_loss: 0.1476\n", "Epoch 49/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.1490 - val_loss: 0.1472\n", "Epoch 50/100\n", "105/105 [==============================] - 0s 2ms/step - loss: 0.1492 - val_loss: 0.1467\n", "Epoch 51/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1487 - val_loss: 0.1466\n", "Epoch 52/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1475 - val_loss: 0.1465\n", "Epoch 53/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1470 - val_loss: 0.1461\n", "Epoch 54/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1460 - val_loss: 0.1457\n", "Epoch 55/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1461 - val_loss: 0.1455\n", "Epoch 56/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1461 - val_loss: 0.1454\n", "Epoch 57/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1445 - val_loss: 0.1452\n", "Epoch 58/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1452 - val_loss: 0.1450\n", "Epoch 59/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1436 - val_loss: 0.1451\n", "Epoch 60/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1429 - val_loss: 0.1449\n", "Epoch 61/100\n", "105/105 [==============================] - 1s 7ms/step - loss: 0.1436 - val_loss: 0.1447\n", "Epoch 62/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1419 - val_loss: 0.1446\n", "Epoch 63/100\n", "105/105 [==============================] - 1s 5ms/step - loss: 0.1432 - val_loss: 0.1445\n", "Epoch 64/100\n", "105/105 [==============================] - 0s 5ms/step - loss: 0.1398 - val_loss: 0.1443\n", "Epoch 65/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1414 - val_loss: 0.1442\n", "Epoch 66/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1411 - val_loss: 0.1441\n", "Epoch 67/100\n", "105/105 [==============================] - 0s 2ms/step - loss: 0.1397 - val_loss: 0.1441\n", "Epoch 68/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1410 - val_loss: 0.1440\n", "Epoch 69/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1390 - val_loss: 0.1439\n", "Epoch 70/100\n", "105/105 [==============================] - 0s 5ms/step - loss: 0.1387 - val_loss: 0.1438\n", "Epoch 71/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1395 - val_loss: 0.1437\n", "Epoch 72/100\n", "105/105 [==============================] - 0s 2ms/step - loss: 0.1397 - val_loss: 0.1438\n", "Epoch 73/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1384 - val_loss: 0.1436\n", "Epoch 74/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1379 - val_loss: 0.1437\n", "Epoch 75/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1382 - val_loss: 0.1437\n", "Epoch 76/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1370 - val_loss: 0.1436\n", "Epoch 77/100\n", "105/105 [==============================] - 1s 7ms/step - loss: 0.1369 - val_loss: 0.1436\n", "Epoch 78/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1359 - val_loss: 0.1435\n", "Epoch 79/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1346 - val_loss: 0.1435\n", "Epoch 80/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1367 - val_loss: 0.1434\n", "Epoch 81/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1357 - val_loss: 0.1432\n", "Epoch 82/100\n", "105/105 [==============================] - 0s 5ms/step - loss: 0.1347 - val_loss: 0.1433\n", "Epoch 83/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.1357 - val_loss: 0.1432\n", "Epoch 84/100\n", "105/105 [==============================] - 1s 7ms/step - loss: 0.1338 - val_loss: 0.1431\n", "Epoch 85/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1346 - val_loss: 0.1432\n", "Epoch 86/100\n", "105/105 [==============================] - 1s 7ms/step - loss: 0.1334 - val_loss: 0.1432\n", "Epoch 87/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.1330 - val_loss: 0.1431\n", "Epoch 88/100\n", "105/105 [==============================] - 1s 5ms/step - loss: 0.1348 - val_loss: 0.1430\n", "Epoch 89/100\n", "105/105 [==============================] - 0s 4ms/step - loss: 0.1348 - val_loss: 0.1430\n", "Epoch 90/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.1312 - val_loss: 0.1431\n", "Epoch 91/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.1324 - val_loss: 0.1432\n", "Epoch 92/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.1315 - val_loss: 0.1429\n", "Epoch 93/100\n", "105/105 [==============================] - 0s 5ms/step - loss: 0.1327 - val_loss: 0.1430\n", "Epoch 94/100\n", "105/105 [==============================] - 1s 6ms/step - loss: 0.1318 - val_loss: 0.1430\n", "Epoch 95/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1332 - val_loss: 0.1430\n", "Epoch 96/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1323 - val_loss: 0.1428\n", "Epoch 97/100\n", "105/105 [==============================] - 0s 2ms/step - loss: 0.1327 - val_loss: 0.1429\n", "Epoch 98/100\n", "105/105 [==============================] - 0s 2ms/step - loss: 0.1307 - val_loss: 0.1428\n", "Epoch 99/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1312 - val_loss: 0.1429\n", "Epoch 100/100\n", "105/105 [==============================] - 0s 3ms/step - loss: 0.1309 - val_loss: 0.1430\n" ] }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAakAAAE0CAYAAACW1MzGAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAABSFElEQVR4nO3dd3wUdf748dd7d9NDL9KLojQ7iIIgWLFg+6Li6aHYFVTOduLpeXjqgb33hlh+nqKnYkelqCgaEFQElBJEBKSXkL7v3x8zCZuw2WQnm2yyeT8fj3kMu/OZ2fdMhn3vZ+Yzn4+oKsYYY0xd5It3AMYYY0xFLEkZY4ypsyxJGWOMqbMsSRljjKmzLEkZY4ypsyxJGWOMqbMsSdUBItJFRLSaU5dajHFIDX1GyfZH1cT2zS4iMiTac6fcOp6mGt6tGlPu/B8f73gaEktSxpiEZT986r9AvAMwAKwEGlWwbBDwgfvvy4FXKiiXE+ugjCnnCyo+T88FnnT/faJb1phqsyRVB6jT7ceOcMtEJDfkZb6qhi1X01Q1G5Aa/owa3b6pHlUtpuLzND/kZW68zlOTeOxynzHGmDrLklQ9JiKjQm9Ii0hHEXlIRH4RkZ3usqYh5fcVkX+IyAwRWScihSKyTUR+EJH7RKRjhM+K2HBCRLJDbyqLyCki8omIrBeRPBFZIiJ3ikjjCJ9R4f0DN2YVkUnu60Ei8raIrBWRfBFZISKPiMgelRwzEZELReRrd9+3iUiWiFwlIv7yxzRaIpIqIieIyBPucd3mHuc/ReRTEblERJIjrD/J/fwZ7uv9ReRlEfnd3c/f3TJ7VSGWk0VkmohsEpEcEflJRG4VkXQv+xYLIrKP+3f6WUS2u+fpEhF5WEQ6RVhPROQvIvKBiKwRkQIR2SoiS919/Hvo+iXnS8gmXgjTkGNIDe5quH04WUT+JyJ/uH/LDW6co0UkKcJ6Ue17yHoZ7rLZ7jlQ6H7mIhF5wz0XM2p2r2NAVW2qwxMwBFB3GlVu2aiQZYcBm0Jel0xN3bIHhFlWftoGHFtBHF1Cyg0JszzbXTYeeCDCZ8wHMiv4jLD76S6b4S6bBFwDFFew/ZVA+wq2nwS8HSG2acAlJa89/r0i7XvJ9HXJ3yXM+pPcMjOAEUBeBdvYBOwXIY77Inz+D8DpIa+7xOA8DT0Xdzs/3DLXAYUR4soBhoVZz1/J361kuj7M+RJpChtnFc7/8VEemxTg9UpiWQC0i8W+u+u1BX6pwnp9a+J7K5aT1aQSxxSc+wXnAx2BNsDJOF9y4JyQc4AbgSOAfYAWQA+cm95ZODfFXxeRNtWIYyTwN+BZ4BD3M3oCT7nLDwD+UY3tH4HzBfwuMBBoCewJ3I6zj52AeypY9z/Aqe6/3wMGuOv3drd5NDCuGrGB8zd4A+eYHoLzt9gD6AP8G9iM84PiiUq20w14EedvdhzQ2t3W34ACoBm7jmkZInIxcK378jvgWKAVsDfOj4geOPtba0RkDHAvzn3w94ChOOdoK5z9+wJIB94Qkf3KrT6KXX+3J3COXzucL+J+OOf8BzgJsMQJlG3kcbn7OnSqrcYdjwBnuv9+Bzgc57zrhXOuFgP7A++HqWWPIvp9B7gb5++9E/g7sK/7mV1wGmNdC8zF+T9Tt8U7S9oUeaLqNakNQIdqfE4A+NLd1m1hlncJ+awhYZZnhyy/uYLPeNddvqaC5VWpSSnwdAXrP+wuzwcal1vWEShyl78DSJj1bwz5DK2hv+d+bhxBYK8wyyeFxPAhEAhT5tqQMj3KLUt1zwXFqbWmh1n/PMr+mu4Sg/0KPReHlFvWll01wqcinH/T3TLvlVv2pvv+/zzEVeE5FeV2Qs//8VGsd1DIeq9UcN6NDilzdSz2Hdjorje2Js7j2pysJpU47lHV372urKpFwKvuy2OrEccqYGIFy15w520kwv2vSpT8Moy0/WScGluoc3AunQBcp+7/5HLuxYm/xqjqj8A8nJaSR1dSfKz7dylvUsi/Dym3bBhO7RVgnKruDBPDZJyac225HOeS10bg6nAF3P281X15ooTcS2VXK+TVNRVgDbrInRcAfwt33qnq4ziX+wAuLrfY677X52NWhiWpxPF+VQqJyOki8rqILHNvpof2BvCYW6x7NeKYpk5T5XCWhPzb6yXFb1R1i4ftD3DnP6vq0nAru3FX6ThGIiLNReRG2dVApaDccS5JLJGO83JV/aWCODcB692X5fdzoDvPwbnHVpG3KtmNWDrGnX8BJIlIZrgJWOyWE5zLoyW+d+cXishIEUmppbhjoeTvMUNV10coN8Wd71suQXvd95L1JojIYBGpt4932HNSiWN5pIVui663cO4FVKZJNeL4I8Ky0F/1XluYVbh9Vd0Z8n+x/Pa7uPMlRLa4kuURichhOJc1W1WheKTjHOk4wq5jWdF+/hrhxwLAokq2H0slyfg0YHsV1wk9fg/gXE7sDEwGnhCRr3AuT88AZleyr/HU2Z3/XEm5he5ccO6rbnFfe933v+P8KOjmllsnIrPc9z5X1YVh1qmTrCaVIMJd1innPnYlqBdxLgt1w7mZWnIjebS73L/b2lVX1S8Lr7/svG6/pKltZT1zeH4IVZzm9W/jfMGuB24C+gPtgabsOs5fuatE+pHodT8z3Xll+1GbD9t6+dGTWvIPVd2K00jgEZyGJxk4jS3+DcwCVonIWBGpi99nJY03Kjveocm7tMGH131X1W/d9f6Hc6lxD5zGGw8DP4nIfBE53utO1SarSTUAbi1qlPvyLlUN24JNRFLDvZ8gSpJTZc+FZFayPJIzcL4MgsCRFf1aFZGKuhaKhZIvw8r2ozr7Ga0dOEn6XlW9wcsGVPVP4GoRuQY4ECf5Hwkcj9Mw40GcWuQ11Y42trbj7Hs0f48ytU2v+66qC4D/c///H4pzyftYnNZ9BwAfiMhwVf1f9LtVe+riLw8Tez3Y9cv0/0UoV77pbyJZ6c73qaRcde7HHejOf4iQoJKrEEN1ZLvzvUUkUo24Zw3GUF7JpeiDqrshVS1W1bmq+qiqDsdptfmlu/gqEWkRYfV4yHbnvSop19udK/BbuAJe911Vd6rqdFW9U1WH4DR3X49TC/9XlfckTixJNQyhN1vDfnG5N65Pq5Vo4qPkEltvqaC3BvdL/aRqfEbJcY6UHM4k5FJWDSj50sogcivN/6vBGMr7xJ0PFpGusdyw24jkfveln91/ABSFLIuHkr/HYBFpGaHcGe78pwgNg8qowr5XtN5Cdv1Yrc0fK55YkmoYVoT8+9TyC92WPw/jPCCaqF5l132eeyto7XQtzk1rr0pqDD1FZLcvDBFpD9xVje1XxXs4Tb0BJkqYLpBE5Dygbw3HEeoxnGfXAsBL7g+iColI93Kve1Sy/dAfHRvLLSt53a4KcdaE59x5CrsSShkichm7auHPlFsW9b673SFV9ohHyXrlj1edY0mqAVDVtcBM9+VNIvIvEekuIi1E5AhgKnABlbdAqrdUdRXwkPvyNOBtETnMbS7eQ0TuwXm+K2IryUq8iZMIAzi9B5wuIm1FpIM4/RF+g/NDYGWEbVSLquaxq9eMA4DpInKM+7feS0RuxekNZEWFG4l9TL/j9JQBTm8L80XkMnH68WvqHqMBInKdiMxhV3PsEh+KyDfu8v4i0sb9u/UWkZuBO9xy88I025/rzs93180UkYA7eW2808E9dyJNB7n7Pp9diWekiLxZ7rybyK5HP+azey8iXva9FbBcRN4RkQtEZD8RaSkie7jbeIFdVwwiXf6vG+L9NLFNkSeq2ONEFbbTg109EYSbXgMurGh7RNF3X4QYKttGVXqcmFTJfkbaRjJOQq7oGHwKXOr+u8jj3+uGCNvPBYZH2hdC+u6r5HMiHm8i9933I/Hpu+9SKu6LMHSaW8G+RppWAPuE+cyhEdYJG2cVzt2qTNkh61a1777d+pz0su9RxDoNyKju376mJ6tJNRCquhjnAclncZ5CL8S5efo5MFJVz8ZplZawVLUA53LnJTh94u3AaUn1Pc6lvuPZdV+pqs/zlP+Me3D6TPwcp8PefJwvmueBQ1T1Te97EFUc1wGnAJ/hPHOzE+fZqNtx+n/bXBtxlIvpaZx+Fu8EvsXpJLcY51j/jJOgT8WpbYUaitNTxds4+7AZ517TBpxm2NcB+2qYh59V9WOcv+uHwJ/sukdVa1Q1X1XPwtm3d4C1OP//NuHEfyXOuRGudwgv+/4bzjH8F87ffxlO69YC4Hec5/hGAMepap0fLFXczGuMAUTkIZwvhZ9UNZFbOxpTL1hNyhiXe4+i5Fr93EhljTG1w5KUaTBEJLmSlmV/Z1erp9drISRjTCWsxwnTkDQHFojIUzhj8CzFeaCxO05v1aPccl8BH8UjQGNMWVHfkxKRt4C1qjq60sLG1CHiDOa4ppJiC4GhFdzENsbUMi9JKh94W1VH1ExIdVPLli21S5cu8Q7DVIOqsmnTJrZu3Upubi6FhYUUFxcTCARIS0ujadOmtGzZEp/ProIbEytz587doKpVGRUgLC+X+1YDSV4/sL7q0qULWVm1OU6cMcbUfyJSrYfXvfxkfA8YFK67FWOMMSaWvCSp8TgPKU6pQv9QxhhjjGdeLvfdi3NzeRjwq4jMw+mLLDdMWVXVi6oRnzHGmAbMS5IahdPvEzh9oR3mTuEoTtNeY4wxJmpektQFMY/CGGOMCSPqJKWqL9ZEIMYYY0x59kCIMcaYOqta3SKJSDLO8A/t3bdW44wFU1DdwBLFkrXbmbrgDy4c2JXmGcnxDscYY+oVTzUpEUkSkTtxxmf5EvivO30JrBeRO0SkwT3wG87y9Tt4dPpS1m3Li3coxhhT70RdkxIRP84DvcfgdM65BmfIbQG6Am2Bm4BDROREVS2OXbj1T3qKc4h3FjTow9AgFRUVlXbDVFRU62PtGRNTgUCAJk2a0Lx5cwKB2uub3MsnXQocC/wC/E1Vy/QWLSJDgQdxktglwJPVjLFeS0/2A5BrSapBCQaDrFq1ipSUFDp16kRycjLOcFXG1D+qSkFBARs3bmTVqlV07ty51vq49PIp5+EMRXx0+QQFpcM1H4MzXPX51Quv/itJUjkF9ku6Idm8eTOBQIC2bduSkpJiCcrUayJCSkoKbdu2JRAIsHnz5lr7bC9JqhcwPdJQBu6y6W7ZBi092amsWk2qYdmxYwdNmza15GQSiojQtGlTcnJyau0zvSSpJJxaUmV20gB7Sy/PalINU15eHunp1gezSTzp6enk5obrBa9meElSK3F6Qa+wPbW7bKBbtkGze1INUzAYtHGpTELy+XwEg8Ha+zwP67yL04LvRRFpWn6hiDQBnnfLvFOt6BJAyeW+nHxLUg2NXeoziai2z2svrfvuBv4CnAWcICJTgRU4ncnuCZwMNAJ+d8s2aH6fkBLwsbPQLvcZY0y0vPTdt1FEjgJeBfoC57KrV/SSFPsdcI6qbopJlPVcerLfLvcZY4wHnp7IUtWlQD8RGQgMxukWSXBqTzNV9cvYhVj/pScH7HKfMcZ4EPU9KRG5X0RuBVDVL1X1TlUdrapXuP+2BFVOerKfXLvcZ0ydMGrUKESESZMm1fpnjx8/HhFh/Pjxtf7Z9ZWXhhNXAQfEOpBElp7st5qUMeV06dIFESE7OzveoZg6zEuSWgtYtSAK6ckBuydlTB0xYcIEFi1axOmnnx7vUEwVeLkn9SlwrIgEVNWSVRWkJ/tZt916QTemLmjbti1t27aNdximirzUpP4FpAFPikhGjONJSGnJfnba5T5jAJg0aRIiwsqVzrP+Xbt2RURKp+zs7NIyo0aNYuPGjVx99dV07dqV5ORkTjvttNJtvfnmm1x44YX07t2bpk2bkpqaSrdu3RgzZgyrVq0K+/kV3ZMKvV+0bt06LrvsMjp06EBKSgpdu3Zl3Lhx5OXV7I/N999/nxNOOIGWLVuSnJxMx44dOf/881m0aFHY8n/88QdXXnkl3bp1IzU1lfT0dDp16sTxxx/P008/vVv51157jaOOOormzZuTlJREy5Yt2W+//RgzZgzLli2r0X3zyktNahTwIXABcKqITMPpWSJcPxmqqrd7Dy8xZCQHbKgOY1zdunXj/PPPZ8qUKeTk5DB8+HAyMzNLl4f+e8OGDRxyyCFs3bqVQYMG0bdvX1q0aFG6fMSIEaSmptKrVy+OOeYY8vPzmT9/Po8//jivv/46X331Ffvss09U8a1atYo+ffqgqgwYMIBt27bx5Zdfctddd/Hzzz/z7rvvVv8ghHHTTTcxceJEfD4fAwcOpH379vzwww9MnjyZ119/nSlTpnDSSSeVll+zZg19+vRh7dq1dO7cmeOPP56UlBRWr17NN998Q3Z2Npdeemlp+fHjx3PbbbeRlJTEgAEDaNeuHVu2bCE7O5vHH3+cQYMGsddee9XIvlWHlyQ1Hue5KAFaAGeHKVOyXIEGn6TSkv3Wd58pddvUhfz8x7Z4hxGVXu0a86+Te8dkWwMHDmTgwIHMmDGDnJwc7r33Xrp06RK27Pvvv89xxx3HlClTaNSo0W7LX331VYYNG1amn8SioiJuu+027rjjDsaOHcuHH34YVXzPP/88F198MY899hjJyU7vb4sWLaJfv35MnTqVr776isMPPzyqbVbmgw8+YOLEiWRkZPDBBx9wxBFHlC675557+Pvf/865557LL7/8QuvWrQF45plnWLt2LZdddhlPPPFEmZ4g8vPzmTNnTpnXd999N5mZmcydO3e3xP3rr7/W6hhR0fAS1b/Z9fCuqYKMFHuY1xgvkpKSeOqpp8ImKICzzjprt/cCgQC33347zz//PJ988gnbt2+vcP1wOnbsyMMPP1yaoAB69uzJyJEjeeKJJ/jss89inqTuu+8+AMaOHVsmQQHccMMNvPnmm8yZM4dnnnmGm2++GYB169YBcPzxx+/WVVFKSkqZ7Wzbto3c3FwOOOCAsDXLvffeO6b7E0teepwYXwNxJLT05ABFQaWgKEhywDodbehiVSNpCA4++OAKa1klfvnlFz766COWLl3Kjh07Sjs/LSoqIhgMsnTpUg466KAqf+ZRRx1FWlrabu/36NEDcO4DxVJRURFfffUV4NwvC+eCCy5gzpw5zJgxozRJ9evXj8cff5wbb7wRgGOPPZaMjPDNBFq1akWXLl1YsGAB1113HZdccknp/tR1XoaPnwcsV9UzaiCehJSW5PSEvrOgiORAhZ3HG2PK6dy5c4XLioqKGD16NM8++yyqFV/c2bYtukurnTp1Cvt+48aNAWLeeGLjxo3k5+fj8/kq3N+Se0WrV+8axm/kyJF88sknvPrqq5x++un4/X723XdfjjjiCM4++2wGDBhQZhuTJ09m+PDh3H///dx///20atWKww47jKFDh/LXv/6VJk2axHS/YsXLz/oeQGGsA0lkGSklScou+RkTjXA1mhIPPfQQzzzzDG3btuW1117jt99+Iy8vD1VFVenfvz9AxAQWTm0PsRIaX0U9jIfbB5/PxyuvvMKPP/7IhAkTOOGEE/jtt9945JFHOPzww7nooovKlB80aBDZ2dn897//5YorrqBdu3a89957pa0Dv//++9juWIx4HU8qs9JSplSaO1zHTms8YUzMvPHGGwA89dRTjBgxgo4dO5KSklK6fOnSpfEKLSotW7YkJSWFYDBYYe8bK1asAKB9+/a7Ldt3330ZN24cU6dOZcOGDUydOpVGjRqV3pMLlZ6ezllnncXjjz/O/PnzWb16NSNGjGDDhg2MGTMm5vsWC16S1JvAESLSMtbBJKqMZKtJGVNeScOEoiJvP942bXIGWejYseNuy6ZNm8b69eu9B1eLAoFAaUOMyZMnhy1T8kzXkCFDIm7L5/MxbNgwTj31VAAWLFgQsXzbtm258847q1Q2XrwkqTuBxcDHInJojONJSGklQ8jbA73GlCqpFVT0oGplSm78P/HEE2VGil22bBmXX3559QOsRddeey0ADz74YGkjihL3338/X3/9NU2aNOHiiy8ufX/y5MnMmzdvt21t3LiRr7/+Gth1T2/lypU8++yzYe/PTZ06tUzZusZLE/T3gWLgEGC2iKwj8sO8R1cjvoRQMjqv9YRuzC6nn346M2bM4Nxzz+W4446jadOmANx1111VWv+mm27io48+4qmnnmL69OkcdNBBbNq0iZkzZ9K/f3/atGnD7Nmza3APYuekk07ixhtv5K677uKII45g0KBBtGvXjh9//JGffvqJ1NRUXn75ZfbYY4/Sdd566y3OP/982rdvz4EHHkjTpk3ZuHEjX3zxBTk5OQwaNKi0f8LNmzdzySWXMGbMGA488EC6du1KMBjk559/ZuHChSQlJXH33XVzjFovSWpIyL8FaONO4djzVNjlPmPCufLKK9m2bRuvvPIK7733Hvn5+QDccsstVVq/f//+fPvtt9xyyy1kZWXxzjvv0LVrV26++WZuvPFGhg4dWpPhx9zEiRMZOHAgjz76KN999x2zZ8+mdevWjBw5knHjxtGrV68y5a+77jq6dOnC7NmzycrKYvPmzbRs2ZKDDz6YUaNGce6555KUlAQ4rQMfeOABZsyYwcKFC1m4cCE+n4/27dtz6aWXMnbs2N22X1dItC1fRGRwNOVVdWZUH1BH9e3bV7Oysjyt+/vmnQy8azp3D9+fsw7Z/fq5STyLFi2iZ8+e8Q7DmBoRzfktInNVta/Xz/LyMG9CJJ3alGGt+4wxxhPr/qAWlDacsMt9xhgTFc89CoqIDzgB6A+0Auao6vPuslZAM2CZqjb4b+aUgA+fYP33GZNAJk6cyOLFi6tUduDAgWVa5pmq85SkRORg4DVgL3b1dp4EPO8WOQV4GjgNmFrtKOs5EbHhOoxJMB999BEzZ1b97oclKW+89N3XGZiGU1N6H5gJlG+7+CbwGJakSqUl++2elDEJZMaMGfEOoUHwck/qZpwEdaWqnqyq95YvoKpbgEU4z1IZICPFalLGGBMtL0lqKLBIVR+vpNwqoK2H7SektCSrSRljTLS8JKk9gJ+qUC4PqPpIYwkuI8VvNSljjImSlyS1HSdRVaYrsMHD9hNSWnLAmqAbY0yUvCSp74G+IlLhpTwR6Q4cCHzrMa6Ek57kJ9cu9xljTFS8JKnngXTgFRFpUX6hiDTGaX7uY1eT9AYv3S73GWNM1KJOUqr6GvA/nI5ml4vIu+6iw0Tkv8AKYBDwuqq+F6tA67v0ZEtSxhgTLa/dIo3AeTYqAAxz3+sBnIkzau+DwMjqBpdInId57XKfMcZEw1OPE6paBIwTkbuAI4E9AT9Os/NPVfXP2IWYGNKS/eQVBikOKn6fxDscY4ypFzz33QegqpuBt6paXkROBQ5Q1X9X53Pro3S3k9ncwmIyU6p12I0xpsGo7V7QTwP+VcufWSek23AdxtS6SZMmISKMGjXK8zZmzJiBiDBkyJCYxWWqzobqqCUlNamd+dZ4whhjqsqSVC3ZVZOyJGWMMVVVp5KUiHQXkbEi8rKILBaRoIioiJxRze2eIyJfiMhWEdkhIlkiMsYdE6tWlNak7HKfMcZUWZ1KUsAVOM3XzwW644xVVS0i8hjwCtAX+AJnmJF9gEeBKSLir+5nVMWuJGU1KdOwLV68GBGhdevWFBYWhi1TXFxMmzZtEBEWLlwIwJw5c7jhhhvo27cve+yxB8nJybRr144zzjiDb775pjZ3oYyFCxdy3nnn0bFjR1JSUmjZsiUnnngiH374YdjyeXl5TJw4kYMPPpjMzExSUlJo27Yt/fv355ZbbiEvL69M+W+//ZYzzzyT9u3bk5SURJMmTejWrRvnnHMOn3/+eW3sYlzVtST1E3APznNY3XDGqvJMRIYDo4G1wP6qOkxVTwf2xhlK5HTgympFXEV2uc8YR48ePTj00ENZv349H3zwQdgyH330EevWraNv37707t0bgJtvvpkHHniAwsJC+vXrxymnnEKLFi148803GThwIG+88UZt7gYA7777Ln369OGll16iSZMmDB8+nF69evHxxx9z4okn8s9//rNM+WAwyEknncRNN93E8uXLGTx4cOk6q1at4s4772TLli2l5adNm8bAgQOZMmUKrVu35vTTT+eoo46iWbNmTJkyhddff72W97j21am20Kr6bOhrkWpXpG5y5zeq6q8hn7NORK4AZuA87/WIqgar+2GR2OU+U+rDcbD2x3hHEZ02+8EJE2O2uVGjRjFnzhxefPFFTj311N2Wv/jii6XlSlx//fW88sor7LFH2f6tp06dyvDhw7n88ss56aSTSE9Pj1mckaxdu5aRI0eSn5/Pfffdx7XXXlu6bMaMGZx00knccccdDBw4kKFDhwLw5Zdf8vnnn3PwwQcza9YsMjIyStdRVWbPnk3jxo1L35swYQKFhYW8+uqr/OUvfynz+Rs3biQ7O7tmd7IOqGs1qZgRkQ5AH6AA2O0nlqrOBFYDbYDDaiyQjcvgmyfIIAewmpQxAGeffTapqam89957bNhQdrCEzZs38+6775KcnFzmi/n444/fLUEBnHzyyZx55pls2rSJ6dOn13jsJZ555hm2bdvGgAEDyiQogCFDhnDllc5Fmnvv3TUu7Lp16wAYNGhQmQQFzo/yww8/vEySLSl/wgkn7Pb5LVq0oE+fPrHZmTqsTtWkYuwgd75QVXMrKPMd0N4tO7tGoli3ED4aR8aFTh60mpSJZY2kvmratCmnnXYar732Gq+++ipXX3116bLXXnuN/Px8zjjjDJo3b15mvQ0bNvDee+/x008/sWXLFoqKnP9PP/3kDHH3yy+/cNJJJ9XKPsyc6dyNqOgZrAsvvJC7776bL7/8kuLiYvx+PwcffDB+v5/nnnuOffbZh+HDh4dNvCX69evHzz//zDnnnMPNN9/MYYcdht9fK7fR64yErUnhjGcFsDJCmd/KlY29zNYApORtBKwmZUyJki/3kkt7JcJd6gN46qmn6Ny5MxdccAH33Xcfzz33HC+++CIvvvgiP/zwAwDbtm2r8bhLrF69GoCuXcN/fXTt2hWfz0deXh4bNzr///faay8eeOABCgoKGDNmDG3atGGvvfZi5MiRTJkyheList8PEyZM4MADD+TDDz9k4MCBNGnShMGDB3PbbbexfPnymt3BOqK2k9RGdiWGmpbpznMilNnhzsOOICwil7rN1bPWr1/vLYqMVgD4d64nNclnScoY17HHHkuHDh2YN28eP/7o3KNbsmQJc+bMoU2bNhx//PGlZbOysrjiiisoLCzknnvuYfHixezYsYNgMIiqctNNzu1nVa21+Es+K9p751dddRUrV67kiSee4Nxzz6W4uJiXX36ZM888k759+5ZJtG3atGHu3Ll89tlnjBs3joMPPpg5c+Ywfvx4unfvzvPPJ/5oSLWapFT1elWtuVpLWSVnjuezVlWfVtW+qtq3VatW3jbi1qTI+ZN06wndmFI+n4+RI53BEiZNmlRm/te//rXMZa0pU6agqlx99dVcf/31dO/enYyMjNIEsXTp0lqNHaBDhw4AFdZosrOzCQaDpKam7nbZsk2bNlx++eW8/PLLZGdnM3/+fPbbbz/mz5/PxIllLwf7fD6OOuooJkyYwKxZs9i4cSMTJ06kqKiIMWPG1GrtMR4qTVIicl51ptrYiQpsd+eZEcqULNseoUz1JGdCIA12/GljShlTTsklvVdeeYWCggJefvnlMu+X2LRpEwAdO3bcbRvr169n2rRpNRpnOIMHDwZg8uTJYZe/8MILAAwcOJBAIPLt/wMOOICxY8cCsGDBgohlMzIyuPHGG+nQoQN5eXksWbIk2tDrlao0nJiEt9qIuOuF/wvWvGx33jlCmZIzPjtCmeoRgcxWkLPeSVLWd58xpfbZZx8GDBjA7NmzueGGG/j999/LPBtVokePHoCTEC666CIyM53fl9u3b+fCCy8s82xRbbnkkku45557+PLLL3n44YfLNP6YNWsWjzzyCADXXXdd6fuff/45eXl5HHfccWUSV3FxcekzY5077/rKuvfeexkxYsRuyTkrK4s1a9bg8/lKa3SJqipJajK7J6lmwCnu+z+w60u+C7C/++93gc3VjtC77915bxFJq6CF3yHlytaMjNaw40/SkgPsLLQkZUyoUaNGMXv2bB5++OHS1+VdcMEFPPjgg8ybN48999yTgQMHoqrMmjWL5ORkLrzwwlq/P9OmTRteeuklRowYwdixY3n22WfZd999+eOPP/jiiy8IBoPccsstZe6t/fDDD1xzzTU0adKEgw8+mLZt27Jz507mzJnDmjVraNOmDTfeeGNp+TvuuIMbbriBnj170rNnT1JSUli1ahWzZ88mGAwybtw42rZtW6v7XdsqTVKqOir0tYg0B+bgNNm+QlV/LLd8X+BxoDdwaMwijZKqrhKRecDBOCMGl6nRichgoANObxRf12gwma1hy29kJPvZmW/3pIwJVfIln5ubu9uzUSWaNWtGVlYW//znP5k2bRrvv/8+rVu35v/+7//497//zVNPPRWHyOHUU08lKyuLu+66i88//5wpU6bQqFEjjjvuOK666ipOPPHEMuVPPvlktmzZwqxZs1i6dCmzZ88mMzOTTp06cfnll3PFFVcQev/7scceY9q0aWRlZTF9+nRyc3Np27YtJ598MqNHj+a4446r7V2udRJtaxi3L7wRwJ6qGvaOnYg0AZYBr6vqaM/BicwABgNnquqUCspMwOne6H+qelO5ZWfgPMi7Fhikqkvd91sD04FewN9U9aHKYunbt69mZWV525F3r4YlH3Jx61f5Y0seH4wd5G07pt5YtGgRPXv2jHcYxtSIaM5vEZmrqn29fpaX1n0nA9MrSlAAqroVJwkMi2bDInKwiHxTMuHUggD+U+79UG1xOqPdrc7rJrYncHqV+FFEporIW8CvOAnqbZyOZmtWZmvYuYGMJCHXLvcZY0yVeelxojVQlUee/UC07bYbE/4S4d5RbqeUqo4WkS+BMTi1Mj+wGHgeeKKm++wDnHtSGqSVbwc5drnPGGOqzEuS+h04UkRaqOrGcAVEpCVwFPBHNBtW1RlEOTyHe89sVCVlXgVejWa7MZXp5OoWspXcgtS4hWFMQ/f222/z9ttvV6lsy5Yty/S7Z+LDS5L6L07v4p+KyNWq+kXoQhEZCDyE04vDY9UPMQFkOA/0tmQrOQUBVDUWPbwbY6I0f/783bphqkjnzp0tSdUBXu5J3QlkAQcAM0TkNxGZ6U4rccaAOgiY55Y1bq8TTXULQYX8opq/wmiM2d348eNR1SpNDWEYjPog6iSlqjuBIcADOP3idQAGuVNHYCdOTWqIW9aUJKli57GxXOt1whhjqsTTUB1u8rlORG7GGbOp5JHn1cDcCENjNEwpjcGfQiM3SeUUFNEsIznOQRljTN1XrfGkVDUP+CpGsSQuEchsTUaR0/+Y1aSMMaZqqj3ooYh0w2lqvlFVf6l+SAkqoxXpBU6SyrEk1SBYAxmTiGpzOBTwOFSHiARE5FYRWQcsAb4ExoUsHyUis90ukgxAZmvSCpwW+xt35Mc5GFPT/H4/hYWF8Q7DmJgrLCys1dGBo05SIhIAPgD+BTQFFrH7s01ZwGHA8GrGlzgyWpGa7ySp5esjjcNoEkGjRo0Sfpwf0zBt27aNRo3CjhNbI7zUpK4EjgE+A7qo6m61JVX9Cadn9MTv/bCqMlvj27mBFukBlm/YUXl5U681b96czZs3s2HDBgoKCmr9EokxsaSqFBQUsGHDBjZv3rzbII41ycs9qZE4w8CfpapbIpRbQTW6M0o4Ga1Bi9mvZZBlVpNKeCkpKXTq1IlNmzaRnZ1NcbHdhzT1m9/vp1GjRnTq1ImUlJRa+1wvSao7MKOSBAWwDhjgYfuJye0aad/Geby20pJUQ5CSkkLbtm0TfrwfY2qSl8t9ClSly4Q2QJ6H7Scmt2ukfTJz2bAjn215dlPdGGMq4yVJrQAOEJEK1xWRNJwRehd5DSzhuL1OdEl1alHWeMIYYyrnJUm9i9PDxPURytyIM8T8O16CSkgZzuW+toHtACxfb40njDGmMl7uSd0PXABMEJEDgZIRc1uKyAk4Q7WfD/yGM4y8AUhrBr4kmusW/D5hxQarSRljTGWiTlKquklEjsepJZ2NM5S8Aie5kwCrgJNVdXsMY63fRCCjFf6dG+jYLM0u9xljTBV47WD2RxHphVOjOgHYE2fE21XAh8DTqmrfwuVltoYdf7Jnq0yW2eU+Y4ypVNRJSkROAQpV9UPgCXcyVZHZGnasY8/2GcxetoFgUPH5rG83Y4ypiJeGE/8D/hbjOBqGjNawYz17tsokrzDIH1ttRBNjjInES5LaBGyIdSANQmYryFnPni3TAWuGbowxlfGSpL4FrHdzLzJaQ7CQvRo5D/JaM3RjjInMS5K6C+gtIhfFOpiE5z7Q25KtNEoJsNyaoRtjTEReBz18EnhaRM7AuUe1Egh7g0VVZ3n8jMTTuB0AsnkFe7ZqZJf7jDGmEl6S1Ayc56IEGErk4TjU42ckpvZ9ICkdln7Gnq3+ypzlG+MdkTHG1GleEsgsnORjohVIga6DYek09ux9Kf/7fjU7C4pIT7Y8bowx4XjpcWJIDcTRcOx9DPzyIfulrQdgxYYcerdrEuegjDGmbvLScMJUR7djAei981sA5q/aEsdgjDGmbrMkVduadYaW3Wm5dhZ7tspg6oI/4h2RMcbUWdW6GSIiGUA3oDFOQ4rdWOu+MPY+Fvn2aYb3u417Z6xi7dY82jRJjXdUxhhT53iqSYlINxF5H9gCzMNp8Tc9zPR5TKJMNHsfC8UFDG++HFV47werTRljTDhRJykR6QDMxun9fB2wHqcW9Q1Od0klNaqvgS9iE2aC6dQfkjJos24W+3dowjvzLUkZY0w4XmpS44CWwO2q2gFnaA5V1cNVdQ+cZ6dWAAVEfoaq4QqkwJ5DYOk0Ttm/LT+u3mpdJBljTBhektRQnHGjbgu3UFWnuWUGAH/3HlqC2/sY2PIbp3bIQQTetQYUxhizGy9JqgMwX1WD7usggIgklRRQ1WXATOAv1Y4wUe1zPIifVkvf4NCuzXl3/h+o2jPSxhgTykuSygPyQ16XXKdqXa7cJqCrl6AahMbtoNepMPdFhu/blOUbcvhx9dZ4R2WMMXWKlyS1GugU8nqpO+9f8oaICHAQYN+6kfS/EvK3MqzoM9KS/Ez6KjveERljTJ3idTypXiJS8mDPR+78ARE5QUT2Ax4D9gayYhBj4urQBzoeRtq8pxh5aHvenr+abBu+wxhjSnlJUu8DacAwAFX9FXgOaA+8B8wHLgcKgVtiEmUi6z8GtvzGmLZLSPL7eGz60srXMcaYBiLqJKWqb6pqkqpOCXn7CuAGnFrWUmAqMERVf4hNmAmsx0nQrAtN5j/NOYd24q3vV/Pbxp3xjsoYY+qEmPTdp6rFqnqfqvZX1e6qeqqqfhOLbSc8nx8OGw2r5nDV3pvx+4THZ1htyhhjwDqYrRsOPBfSmtF87kOcfUhHpsz9nd83W23KGGMsSdUFKZkw4Gr49ROu2mcLIvDkzGXxjsoYY+Iu6l7QRSSaTmNVVY+O9jMapH6XwteP0irrfoYfPJ7Xs37n6qP3pnUj6x3dGNNweRmqY0gVyihOR7PWhUJVldSmPv0XVw+/mtezgjz35QpuOqFnvCMzxpi48XK578gKpqOBC4E3cRLUXcBRsQmzgeh3CaS3pN33D3DS/u14+euVbN1ZGO+ojDEmbrw0QZ9ZwTRdVSep6pnAlcA17OoyyVRFcgYM/Bssn8F13TeQU1DMi19nxzsqY4yJmxppOKGqjwPZwPia2H5C63sRZO5Bl+/v5ejurXj+qxXk5BfFOypjjImLmmzd9yPOcB0mGsnpMOQmWPUNN+21nC07C3llzsp4R2WMMXFRk0mqDU73SSZaB42ElvvQbcG9DNm7OY9NX8aWnQXxjsoYY2pdjSQpETkbpxa1uCa2n/D8ATjmNtj4KxO7zmd7XiEPffZrvKMyxpha5+U5qecjLM4EegC93dcPewnKAN1PgE79aTP3fs7r8zIvfb2SkYd1Zs9WmfGOzBhjao2X56RGVaHMduDfqjrJw/YNgAgcezs8dww3NPqEKUn9mPDhYp45r2+8IzPGmFrjJUldEGFZAc6giN+paq63kEypjodAr9PIyHqcG/ofxb9mrGP2sg0M2KtlvCMzxphaIarWKURV9O3bV7Oy4jCG4+ZseLQfxb1O44hfz6ZJWhJTrxqI3ye1H4sxxkRJROaqqudLQNbBbF3XrAv0H43/x/8y8dACfl6zjSlzV8U7KmOMqRWWpOqDgddCRmsGLr+fvp2acs/Hv7A9z7pLMsYkvpruBb086xXdi9TGcNQtyNSrue/IZQz+sAWPz1jGjcf3iHdkxhhTo6rTC3rJzazyN0cqej90mYnWQX+F756h89yJ/OXA53nuixWc068THZunxzsyY4ypMV57QX8IJwnNw+lI9nR3usZ9T9wy5XtKt17RvfL54cR7Ydtqbmn8Pn6fcNvUn7GGL8aYROalJqU4vZxfr6r3h1n+kIhcA9wNvK2qM6sToAnR6TA44C9kZD3B+MOP5MYZ65j6wxpOOaBdvCMzxpga4aUmdQvwcwUJCgBVfQD4GbjZa2CmAsfcBklpnLX+MQ7o0ITx7y5k4478eEdljDE1wkuS6ovTw3llfgQO8bB9E0mjPWDITciyT3m87xp25BVx67sL4x2VMcbUCC9JKhnoXIVynfF2OdFUpt+l0LoX7b/+N9cO6cD7P6zho5/WxDsqY4yJOS9JagEwQEROrKiAiJyA0wv6Aq+BmQj8AacRxdbfuMT3Dr3bNeZf7y5kZ4ENjmiMSSxektQ9OK33/iciz4nIUSLS1Z2OFJFngbfdsvfGKlBTTpfDYf8R+Gc/xMQhGazbls9TM5fHOypjjImpqJOUqr4NjHPXHQVMA5a606fAhYAfuNkta2rKsbdDIJX9fvgPw/Zrw1OzlrFmq/Xra4xJHJ66RVLVu3EaRUwCluP0fl4ArABeAPqp6kSvQYnIOSLyhYhsFZEdIpIlImNEJKp4RWSSiGiEqX4PythoDzjyH7B0GuO7ZxNUuPujJfGOyhhjYsZzwwZVnQ9cFLtQHCLyGDAayAM+AwqBo4FHgaNF5ExVLY5ys1/h1PTKq/+tDQ65BOa9RMtZtzJ6wEs8OGs15w/owoEdm8Y7MmOMqbY61fpORIbjJKi1wBGq+qv7/h7AdJxeLa7E6c0iGs8m7ACM/gAMewCeP47RTOGVRkdx29SFvHn5AHw2nIcxpp6LWS/oIuITkYtF5BERuV5EGnnYzE3u/MaSBAWgquuAK9yX46K97JfwOh0KfUaR/N2T3Nlf+f63Lbw57/d4R2WMMdUW9Ze9iIwTkZ0iMqTcoveBp4AxwF3A1yKSEcV2OwB9cO5tvVF+udu90mqgDXBYtHEnvGPGQ3pzjl02kb4dGzHxw8VszbXhPIwx9ZuXGslQYBtQ2iefiBznvr8auAP4FuiJ09Kvqg5y5wsjDD3/XbmyVXWkiNwvIk+LyO0iMjThamNpzWDof5DVWTy8zw9s3lnA/Z9YIwpjTP3m5Yu6G07ffaHdbw/H6Xj2bFW9Fae3883AOVFst6s7XxmhzG/lylbVeTg9tF+C0/fgR8CPIrJflNup2/Y7E/YcQrusuxh9cBovfbOShX9sjXdUxhjjmZck1YLdW8UNBNaq6mwAtyY0G+gSxXYz3XlOhDI73HlV73fNB64GervbbwcMw+kJoxfwqYi0r2hlEbnUbf6etX79+ip+ZByJwLAHIVjE2NzHaZaWxD/f/onioA3nYYypn7wkKQVK7zWJSBOgB04z71BbgaZRbLekKVrMvlFV9UFVfURVf1bVHFVdo6rvA/2Ab4DW7GqsEW79p1W1r6r2bdWqVazCqlnNu8JR/yRp+TSeOGA5837bwqTZ2fGOyhhjPPGSpFYAh4bc0xmGk2C+LFeuFbAhiu1ud+eZEcqULNseoUylVLUAmOC+rLAPwnrr0MugQz8OWTyR0/dO4p6PF7NiQ6QKqjHG1E1ektS7wB44ffddjdOXXzHwTkkBERGcxg0rothutjuP1MN6x3Jlq6Okt4kKL/fVWz4/nPooUpDDhNTJJPt93PDGArvsZ4ypd7wkqbuARcDJwIM4TcLvVdXQBg8DcWpS5WtXkXzvznuLSFoFZQ4pV7Y6WrjzHRFL1VetusPgG0n9dSrP9FlF1srNvPBVNL8ZjDEm/rx0MLsVZ+DD84G/A0NUtfx9nRY4vUK8FsV2VwHzcMarOrP8chEZDHTA6Y3i62jjDuMsd/5dxFL12eF/g3YH0e/nO/m/fZK45+MlLFufmDnZGJOYvHYwm6uqL6nqvao6K8zyt1X1GlX9IfR9ETlVRG6NsOmS+0R3iUi3kPVaA4+7LyeqajBk2QQRWSwiE0K2g4gcKCLDRMRf7v2AiFyL0+oP4IHK9rfe8gfgtCeRgh38J/kFUgN22c8YU7/U9gOtpwH/qmihqk4BnsC5hPijiEwVkbeAX3GajL+N09FsqLZAd3ceqgswFfhTRL4WkTdE5COc57Duc8vcqKofV2eH6rzWPeDIm0ld+gHP9lnJvN+28PyXdtnPGFM/1LleF1R1NHAuzqW/wTg9WSzF6Vh2eBQ9oC/AueS4BOiEcw9tMLCTXcOJ3B3b6OuoAVdBh0Pou/BOztjHzz2fLGHpn3bZzxhT90nZjiNq+MNEXgDOU1V/pYXrmL59+2pWVla8w/Buw1J4ciD5HQdwaPZldG6ZyZTL+5Pkr3O/U4wxCURE5qpqX6/r2zdUQ9GyGxx3OykrPmPyAQtZsGoL933yS7yjMsaYiCxJNSSHXAx7HcX+C+/hqgOFJ2cu47NF6+IdlTHGVMiSVEMiAqc+Bv5krtl+H/u2yeDa1xfw++ad8Y7MGGPCsiTV0DRuB8Pux7c6i5f3nklxULny1e8pLA5Wvq4xxtQyS1IN0b7D4YBzaJr1IE8PLmD+qi08Nn1pvKMyxpjdWJJqqE68G5p2ZsD8cfxlv0Y8+vlSflptY08ZY+oWS1INVUojOOM52LGW23zP0jw9ieteX0B+UVUfQzPGmJpX20lqI7tG1zXx1r4PHHULyUve4aUDF7Jk3XYe+vTXeEdljDGlajVJqer1qhrt0O+mJg0YC92Opfv3d3JN7xyenLmMb1dsindUxhgDVKPHCXfY9SNxhmRPraCYqurtHmOrU+p9jxOR7NwETw4iKH5OKfoPfxak8t7VA2ndqKI/qzHGVE11e5yIOkm5Axo+CIxmV01MyhVT9z2tj10ghZPQSQpg1bfwwgls73QUhyy7gAM6NOOViw8lYN0mGWOqobpJKuBhnRuAq4Ag8BHOCLfbvAZg6oiO/eDY22n08U28se++nPx9X+75ZAk3ndAz3pEZYxowL0nqAqAQOFpVoxl519R1h10Bv3/Hfj8/yK297uPfM6Fnm8acdlD7eEdmjGmgvFzL6Qp8aQkqAYnAKY9Ay+5csOZ2hnUq5Po3FjB98Z/xjswY00B5SVJbAOuVNFGlZMLZryDBIh6S+9i/TTJXvDKXrGxr8WeMqX1ektTnwCGxDsTUIS32gtOfwr/uB15t/QptG6dy4aTvWLJ2e7wjM8Y0MF6S1D+BViLyz1gHY+qQHifC0beSuvgt3t7vK1KT/Fw46TvWb8+Pd2TGmAbES8OJw3GGXx8vIicCH+L0IhG2G21Vnew9PBNXA6+FDUtp8s09TDnyUYZ+2opLJmfx2qWHkZqUEE8WGGPqOC/PSQXZ9RwU7r8rZM9J1XNF+TD5NFg9l2+OeJG/fBTkxH3b8shfDsLnK/94nDHGlBWP56QmU0liMgkkkAIjXobnjuGwb0Yz4YjnGDdzDR2ap9kzVMaYGhd1klLVUTUQh6nLMlrAX9+EZ49lxJK/kd3nYZ6cuZxWmSlcPGjPeEdnjElg1ueNqZrme8K5ryM567lx062c1qsxd7y/iLe/Xx3vyIwxCcySlKm69n3gzEnI2p+4r3giR3RJ5/o3FvDZIntszhhTM7zckyolIj2BfYDG7N7JLGCt+xLOPkOdZ6jeuoTnuwQ4q81YLntpLvePOJBTDmgX7+iMMQnGU5ISkQHA00CkO+eC08DCklSi2f9MCBYSeHs0/90rwPlJVzP2te/ZkVfEOYd2ind0xpgEEnWSEpEewCdAOjAbaIPTn99rQDfgIMAPvA1sjVWgpo458BwoLiBp6lhe2gtGdxvLP/73I9vyCrl88F7xjs4YkyC83JMah5OgLlPVgcAXAKp6rqoeChwAzMW5DHh1rAI1dVCfUXDKI/iXf84T3MlZ+zZi4oeLuefjxXgdTNMYY0J5SVJDgF9V9ZlwC1V1ETAM6ITThZJJZAefB2e8gG/1XO7a8Q8uOSiDx6Yv41/vLiQYtERljKkeL0mqDfBTyOtiABFJKXlDVf8EZgKnVys6Uz/0Pg3OeQ3ZuIx/rB3LuEMCTP56JVf+v3lszyuMd3TGmHrMS5LaQdmWfCWj8rYtVy4XsNHyGopux8D5U5H8HVy29HIeOjyfjxeu45RHv2LRGhu42RjjjZck9TvQMeT1Ynd+ZMkbIpIEHAqs9x6aqXc69IWLpyFpTTl1wRV8dOxGcvKLOO2xr3gja1W8ozPG1ENektRXQG8Raey+fh/nkt8DInKFiJwMvAl0AGz03oam+Z5w0TRosz97z7ySGQfNpG+nxtww5Qf+88Eiiu0+lTEmCl6S1FvAapwGFKjqamACzgO9j+I0PR+G0/z85lgEaeqZjJYw6j3ocwHp3z7MSyl3c8UhTXh61nIunZxl96mMMVUW9VAdFW5IZDhwBtAc5xLgg6q6IiYbrwMa7FAd1TXvJXj/OkhvwbTu47l8diM6Nktj/Cm9GdK9dbyjM8bUsOoO1RGzJJXoLElVwx/z4c2LYeOv/NHrYi5YeTxLNhZwXK89+OewXnRsnh7vCI0xNaS6Sco6mDU1r92BcNks6HsR7X5+lg/Tb+H+/nl88esGjn1gJk/PWkZRcdiBnY0xDZznJCUiTURkjIi8LCIfi8jfQ5Z1F5HjRCQtNmGaei85HYbdD+e8gS9/B//3/YV8d+D7HNs1lf98sJhTH/uKBau2xDtKY0wd4ylJicjxwHLgYeAc4BigR0iRg4APgVOqG6BJMPscB2O+gcNGk/nTyzy88RLe6/8LG7ft5NTHvmLEU1/z/g9rKLSalTEGD0lKRPbFaeHXCHgcGMHuw3S8i/Mw76nVDdAkoJRGcPwEuORzpEU39v1+PF81/SdP9lvH6s07GfPqPAbfPZ3Xs1ZZk3VjGjgvNal/ACnAGap6laq+Ub6Aqu7EaeF3QDXjM4ms3UFwwYcw4mX8wUKO/+EavmjyT6YOWcsejZL4+5QfOOGhWUz7eZ31A2hMA+W1g9nvVfXdSsqtYveukowpSwR6ngyj58CpjyPFhez3zbW8VTSGj/rOI61oK5dMzmLwvdN5bPpS/tyeF++IjTG1yMughy2AWVUoFwSs4YSpmkAyHHQuHPAXWPIB8s0T9PjpXt72p/D73scxaefh3PtxDvdP+4XDu7Vk2H5tGdq7DU3Sk+IduTGmBkX9nJSIrAWWumNJlbwXBCap6oUh780Hmqpql9iEGl/2nFQc/LkIvnsOfvgv5G+jMLM93zU+lhc29uLTre3w+/wc0LEph+/VgsP2asF+7ZvQKNWSljF1Sa0/zCsi7wJDgX1V9Vf3vTJJSkQOAeYA/09Vz/UaXF1iSSqOCnNh8fuw4P/Bss9BgxSmtWJRZn8+zuvJ6xu6sF6bANC5RTq92jZmr1aZdG6RTucWGezVKoMWmSmVfIgxpibEI0kNxWle/iNwlqouCU1SIrIn8A7QCxisqgnRyawlqToiZyMsnQa/fARLP4N8ZxiQHU32ITt9X+YG9+GTbZ2Ys7UJRcFdjU6bpSfRrXUm+7VvSr+uzejbpTktLXEZU+Pi0i2SiDwEXAUosBDojdPp7BqcZ6QCwP2qer3XwOoaS1J1UHERrFkAK2ZC9hfwe1Zp0tLkRuS36M7GjH34zdeBxQWtmLu9GdPXpZJT5LQXapmZQrumqbRrksYejVNokZlCi8xkWmQk0zwjheYZSbTKTKVxWgCR8k9ZGGOqIm5994nI5cCtOCP1htoI3K6qD3sNqi6yJFUPBIth/RL4/VtY+xOsW+hM+VtLi6j4KMhox4ZAW9ZIK34vbsaKgiYsz2vMyvwMNmpjNtOInaRQ8vhferKfNk1SaZmRQsAv+H1CWpKfdk3TnCTXNI12TdPo0DSNlpkp+HyW0IwpEdcOZkXEBxwI7An4cZqdf6uqRZ43WkdZkqqnVCFnPWxaDhuXweZsd1oBW1fDjrWgu/duoeKnMJBBXqAx26Uxm2jEZs1gp2SwnXS2FqeyLs/PlqIAeZrMTlLYSSp5pFDgTyffn06xPx1fUiqBpBRSk5NolplCS7e2lhLwE/A5CS8l4CMt2U9qwE9Kko8kv49kvw8R51IFQJO0JDo2S6dlZrLV6ky9Ut0k5aUJeilVDQLz3MmYukcEMls7U6fDdl9eXATb18COPyHnT2eeuxnJ30Zy3jaS87bQeOcm2u/cCLlrIG8b5G8HLXYqWpEaExa7Ux4EEYo3+ikkQKH6ySeJfE0inyTnPdxlBChQZ16En2J8FONjC36WqR/1BSCQTIEmka8BisWPz+fD5/Pj9/tJDvhJSfKTFEhC/AHwJ+MLJJGUlEJScjJJScmo+AiKD8WdRBDxkZGaTOP0ZDJTUyjGR14xFASF9GQ/TVKTaJwaICngB/GDz+fMxRcyiTNHQl5LuTIhU2i50B/LImG2J2XL77auu35oLCYhVCtJGVPv+QPQtKMzVZWq0+KwMBcKdzpTQc6uef52KNjh/Lu4AIoL8RUX4CsuJClYBMWFaFE+WpiLFuYSLC4kWFRIsCgfLS6E4kIoyodgLqLFSLDYKVNcCEUF+LQIf0la06JdydDGkiyl4vZToIqw+9UixUnMuymTBEMTXfmE65ZBd1V3hXLJNySJatAtWy4Zh0u8IVGW7EPpaw06r0XcHwsBZ53S7Ze/KlA+wZcEGrL90HIlCV+17OcechEMum7341ULPCcpEekIDAbaAakVFFNVvd3rZxhTJ4k4vbonp+M82+5hE+z6qvBXN56SLxQt3vUlEyyGYKFTUywuoKgwn9z8fPLyC5BgMUIQKfliI0iwOMj23AK27sxj+848knxKqg+SfMrOgmK25RexLbeI/MIiCgsLKSwqwi9Kih+SfUphUTHb8wrZkVuAqpKR4icz2Ud+YSFrtuSyOScPH07CSAsIaQEnblWluLiY/GJFS4+LU8cL+JRgUPERLH3fmQfLlHP22FnqI0hAgvhwvqxDl4Uee59Aml/x+/27/g4+SAn4SPVDSkBIDvhICfjxoeTkF5GTX0hBUTHJPiHFDyl+CPj9JAV8+H0+ioqDFBUVUxwsJi3JR3qSj7QABPFRFFSKgri1XR/JAR8BAZ84+1BQVExeYTH5BYUUheSZlCQ/zdOTaZaZQpJPKEYoLFaCqgQkSJI4f+vCoFAQhGJVJyafICIUFBVTUFhEYXExAZ8QECXgc9J2UBVFSA34SPKDKOFrrwDN96ruWepZ1ElKRAI4w8RfzK7/Z+Xr1iXnmwKWpIypSSWX1SL0chbA6RG6UYTNtIxxWKF25BexLbeQ5hnJpCbtnpbzi4rZmltIbkExjdxLi36fsDW3kDVb8/hzez6FRUGK1UlcqUl+0pP9pCX7Cfh8+Nxdz8kvZltuIVtyC8grDFJQFCS/qBhBCPiFgE/ILwqyZWch63ILyC8MltYn8ouC7MgrZHteEdvyCtm6rZCtuYUEg9CmSSptm6TSLD2ZnIIituUVsSOvkJzcYnLyi8grKiYtyU+j1CRSAj42bitga663qm2y39kZRSksdqITgbQkPzsLincr7xOobteWKQEfezR26hoFRUEKip0fAn6fc8z+2rQzo6v3EZ55qUmNBy4FioAPgF+BHTGMyRiTYDJTAmSmVPx1kxLw07rR7smraXoyTdOT6RnHXkBV1VNjldyCYjbsyCfJ7zaMSfKRVxBkS66TwAqKghQWK0XBII1Sk2jdyGlYkxzY9WPjz+15LFy9jR9Xb2VbbiFN0pJonJZEwC/k5BexI78YVS19PzXJT26B835hcZCmaUk0y0imUUqA3MJicgqKyS0owidCkts4Z8OOAtZuzeXP7fkIlNb0VJ3aVlGx0rl5RgyPaHS8JKmRQA5wuKr+EON4jDGmTvHamjIt2U/H5ull3ksJ+KPqb7J1o1Ra90jlyB6tPcWQCLz0gt4amGkJyhhjTE3zkqR+A/JjHYgxxhhTnpck9RowWEQyYx2MMcYYE8pLkvoPsAR4X0T2iXE8xhhjTKmoG06oar6IHAd8DSwUkZXA78Dufcs4z0kdXc0YjTHGNFBenpNqCUzD6flccPrt27OC4tVsvW+MMaYh89IEfSJwAM4lvyeBpdhzUsYYY2qAlyR1Es64UYep6tbKChtjjDFeeRmZdwfwoaqeWTMh1U0ish5Y6XH1lsCGGIbTENgxi54ds+jZMYtetMess6q28vphXmpSi4jcBVhCqs5BFpGs6oyn0hDZMYueHbPo2TGLXm0fMy9N0B8Dhljzc2OMMTUt6iSlqpOAB4EZInKRiHSIdVDGGGMMeGuCHtpX/NPuexUVV1W1gRXd42SiYscsenbMomfHLHq1esy8NJwI99BuhVTVyyVFY4wxJvokZYwxxtQWq+UYY4ypsyxJ1SAROUdEvhCRrSKyQ0SyRGSMiDS44y4iSSJytIjcJyLfiMgaESkQkdUiMkVEhlSw3iQR0QjT4trdk9pVnf1viOefiAyp5HiFTp1C1kvo80xEuovIWBF5WUQWi0jQ3a8zqrCup/MoVuefNWqoISLyGDAayAM+AwqBo4FHgaNF5ExVLY6wiUQzGKfPR4C1wFycEZ57AcOB4SJyu6reWsH6X+F0wVXemlgHWkdFtf8N+PxbC7wYYXk/oCewDFgVZnminmdXAGOjXcnreRTT809VbYrxhPOlqzgn9t4h7+8B/OwuGxvvOGv5mBwFTAEGhVk2Aihyj8uR5ZZNct8fFe99iNNxi3r/7fyLeGwWuvv/j+oe5/o0ARcDdwNnAXsBM9z9PSPW51Gsz7+4H7xEnIAs9w9xXphlg0P+gL54x1pXJuBZ97g8V+79hP7yqMJx8ZKk7PwLf1z6u/teBLSv7nGuz1MVk5Sn8yjW51/CXpuOF/fh5j5AAfBG+eWqOhNYDbQBDqvd6Oq07925PRxeDXb+RXShO/9IVVfHNZI6zut5VBPnn92Tir2D3PlCVc2toMx3QHu37Oxaiaru29udV3Tt/0gR2R/IBNYBXwLTVDWq5/bqsaruv51/YYhIOs5lZYDnIhRt6OdZCa/nUczPP0tSsdfVnUfqMf23cmUbNBFpA4xyX75ZQbHzwrz3s4icrao/1khgdUtV99/Ov/DOxOkY+0/gvQjlGvp5VsLreRTz888u98VepjvPiVCmZJDIBtebfHkiEgBeBpoAn6nq1HJF5gNX44wEnQm0A4YBC3BaBn4qIu1rLeDaN5/o9t/Ov/BKLvVNVtXCMMvn07DPs/K8nkcxP/+sJhV7JR0ZWlceVfMkTtPUVcBfyy9U1QfLvZUDvC8i04CZONe1bwKurNkw48PD/tv5V46IdAOOcF8+H65MQz/PwvB6HsX8/LOaVOxtd+eZEcqULNseoUzCE5GHgItwnm05WlXXVnVdVS0AJrgvT6yB8Oq0CPtv59/uSmpRX6vqomhWbMDnmdfzKObnnyWp2Mt2550jlOlYrmyDIyL34VxeWY+ToH71sJmSXgAa0mWYUOH2P9ud2/kHiIifXfeZIjWYiKQhnmfZ7jza88jrehWyJBV7JU2pe4tIWgVlDilXtkERkbuBa4GNwLGq+rPHTbVw5zsilkpc4fbfzr+yhuIklxzgvx630RDPM6/nUczPP0tSMaaqq4B5QDJOi6IyRGQwzrNAa4Gvaze6+BORicANwGacBLWgGps7y51/V+3A6qfd9t/Ov91c5M7/q6pek0yDO8+8nkc1cv7F+8nnRJyAM9j1VHW3kPdbs6tblrHxjjMOx+V2d983A32qUP5AnBZW/nLvB3BqYsXu9obGe99q6Hh52n87/0r3tyWQ7+7vADvPyuzbDCrvccLTeRTr88/Gk6ohIvI4TqeOecCn7OpgsTHwNs7JkYgdfIYlIqcA77gvs3BO1nAWq+pEd53TgP8Bm4BfgN9xmq3uh9NEOAjcpKp311zk8VOd/bfzD0TkGuB+nHOqZ4Ryp5Hg55mIHAw8HvJWL5x9/BVnvwFQ1cPKrefpPIrp+RfvjJ7IE3AOTq/K23Cuic8FxtDA+kxzj8UonF9QlU0zQtbpCjyI81T6aveEz3X/Yz1PFWpj9Xmq7v439PMP+ME9p26oyeNcHyZgSFX+/8XyPIrV+Wc1KWOMMXWWNZwwxhhTZ1mSMsYYU2dZkjLGGFNnWZIyxhhTZ1mSMsYYU2dZkjLGGFNnWZIyxhhTZ1mSMiYKIpItIlqFaUi8Y60KERnvxjs+3rEYE44NemiMNx/jdJJZkSqPjWWMqZglKWO8maiqM+IdhDGJzi73GWOMqbMsSRlTg0Ski3vPJ1tEAiIyTkQWiUieiKwTkRdFpFOE9XuLyGQRWSUi+SKyQUQ+EJETKvncoSLyloj8ISIFIrJWRL4SkRsrGoxORPYQkadE5Hf3s1aIyEQRSQ1T1i8il4vIbBHZ6n7GOhGZJyL3iUir6I+WMbuzJGVM7fkvcBvwG85wBQU4Q5t/JyLdyxd2hzeZC4wEtgJvAj/jjDb7gYjcHmYdEZEngI+A03F69X4TWIAzbPdEYI8wsXV0P2sYzmB0M3DG/7kReD1M+eeAJ3DGYpoDTHE/ownOGEx7RTwSxlRVvLuQt8mm+jQB2TjDGgypYvku7BoKYR3QK2RZMvCSu+zbcuu1wUlMClxbbtkQnKEPwg14eI37/lrgsHLLBDgSaBLy3viQ+J4BkkOW9QS2u8sOD3m/s/veb8AeYfb5QKB1vP9WNiXGZDUpY7yZHqH5+ZYK1rldVX8ueaGqBcCVOMnoEBE5PKTsJTgDxM1W1ftDN6JOg41H3ZfXl7wvIgHgH+7LUar6Tbn1VFWnq+rWMLGtAq52YyopvwgniYIzYF2J1u58nqquK78hVZ2vqn+G+Qxjomat+4zxJlIT9J0VvP9y+TdUdauIvAeci1ND+spdNNidT6pgW88DfwcGiohfnVFO++IMmf67qn5U2Q6U87mq5oZ5f7E7b1fuve3ASSLyD+AVVV0Z5ecZUyWWpIzxJtom6FtUdUsFy7LdeYeQ99q78xUVrLMCZ1jzVKAF8CfOZTiAJVHEVeK3Ct7f5s5LG0+o6nYRuRAnUd4J3Ckiq3HuZb0PvKaqeR5iMGY3drnPmLojdJhsCfNeTQpGU1hVpwCdgFE4yWoHcAbwArBYRDrGOkDTMFmSMqZ2NBWRJhUs6+LO/wh573d3vmeEdXxAHrDJfa/kkttuLQVrgqpuUdUXVfUiVe0BdAOm49To7qqNGEzisyRlTO05t/wbbuIa5r6cEbJopjs/r4JtXeDOv1TVIvffc4ENQAcRGVq9UKOnqstwLv8BHFDbn28SkyUpY2rPrSLSs+SFiCQBD+E8WzRXVb8MKfsMTuOEgSJydehGROQI4Cr35X0l76tqITDBffmCiPQrt56IyJAINboqEZGDRGREBQ8Fn+zOrSGFiQlrOGGMN+NEZFSE5a+q6ichr3/DqenMF5HPcZqd98e5r7OBcjUmVV0rIiNxHgB+SEQuBn7CaWU3COcH5h1hWvE9gPN808XANyKSBSwFmgO9cB7a7ep+vledgdeAnSIyD6f5ejJwEM7lye3ArdXYvjGlLEkZ401ll9PmA6FJSoGzgHE4PUh0xmk59zLwT1XNLr8BVX1HRPri9PpwFE7DhO3udh9R1Q/CrKPAJSLyDnA50A/n4dpNwK/AI1S/h/ZvgJtwmsn3APrg9J6xCqdm94g1STexIs45bYypCSLSBae5+EpV7RLfaIypf+yelDHGmDrLkpQxxpg6y5KUMcaYOsvuSRljjKmzrCZljDGmzrIkZYwxps6yJGWMMabOsiRljDGmzrIkZYwxps76/ylcHdLzDeVYAAAAAElFTkSuQmCC", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "WARNING:tensorflow:AutoGraph could not transform .predict_function at 0x1a53234a70> and will run it as-is.\n", "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n", "Cause: 'arguments' object has no attribute 'posonlyargs'\n", "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n", "WARNING: AutoGraph could not transform .predict_function at 0x1a53234a70> and will run it as-is.\n", "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n", "Cause: 'arguments' object has no attribute 'posonlyargs'\n", "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "0it [00:00, ?it/s]" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Loading parameters from tmp_checkpoints01042022185134/best_weights\n", "Saving parameters to visual_bars_cfl/experiment0017/trained_blocks/CondDensityEstimator\n", "CondDensityEstimator training complete.\n", "Beginning CauseClusterer training...\n", "Beginning clusterer tuning\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "9it [00:06, 1.46it/s]\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Please choose your final clustering parameters.\n", "Final parameters: {'n_clusters': 4}\n", "CauseClusterer training complete.\n", "Experiment training complete.\n" ] } ], "source": [ "train_results = my_exp.train()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Visualize the conditional probability learned by CFL with the optimized hyperparameters" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(
,\n", " )" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "from cfl.visualization.cde_diagnostic import pyx_scatter\n", "from importlib import reload\n", "import cfl.visualization.cde_diagnostic as cvc\n", "reload(cvc) \n", "# first by ground truth macrostates\n", "cvc.pyx_scatter(my_exp, vb_data.getGroundTruth(), colored_by='ground truth')" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(
,\n", " )" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# now color by learned macrostates\n", "cvc.pyx_scatter(my_exp, train_results['CauseClusterer']['x_lbls'], \n", " colored_by='cause macrostate')" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "interpreter": { "hash": "a3e2eae926291f18d1afe7b431ab5303b6e9ec8df44d60cf0ff75f0c781aceac" }, "kernelspec": { "display_name": "cfl_env", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.7.9" } }, "nbformat": 4, "nbformat_minor": 2 }