From e5af800dc35a776e605eb58e22d0b99cce415c4f Mon Sep 17 00:00:00 2001 From: MariaLukina Date: Mon, 12 Feb 2024 16:45:50 +0700 Subject: [PATCH 1/6] Add requirments.txt for HW1 --- requirements.txt | 5 +++++ 1 file changed, 5 insertions(+) create mode 100644 requirements.txt diff --git a/requirements.txt b/requirements.txt new file mode 100644 index 0000000..1fc5ddd --- /dev/null +++ b/requirements.txt @@ -0,0 +1,5 @@ +numpy +matplotlib +pandas +scikit-learn +seaborn \ No newline at end of file From f1150eb63d09c3bc235cc7aafdba00f7babe92e9 Mon Sep 17 00:00:00 2001 From: MariaLukina Date: Mon, 12 Feb 2024 16:48:45 +0700 Subject: [PATCH 2/6] Add knn.py and metrics.py with part of the functions --- knn.py | 143 +++++++++++++++++++++++++++++++++++++++++++++++++++++ metrics.py | 79 +++++++++++++++++++++++++++++ 2 files changed, 222 insertions(+) create mode 100644 knn.py create mode 100644 metrics.py diff --git a/knn.py b/knn.py new file mode 100644 index 0000000..c73ba6e --- /dev/null +++ b/knn.py @@ -0,0 +1,143 @@ +import numpy as np + + +class KNNClassifier: + """ + K-neariest-neighbor classifier using L1 loss + """ + + def __init__(self, k=1): + self.k = k + + + def fit(self, X, y): + self.train_X = X + self.train_y = y + + + def predict(self, X, n_loops=0): + """ + Uses the KNN model to predict clases for the data samples provided + + Arguments: + X, np array (num_samples, num_features) - samples to run + through the model + num_loops, int - which implementation to use + + Returns: + predictions, np array of ints (num_samples) - predicted class + for each sample + """ + + if n_loops == 0: + distances = self.compute_distances_no_loops(X) + elif n_loops == 1: + distances = self.compute_distances_one_loops(X) + else: + distances = self.compute_distances_two_loops(X) + + if len(np.unique(self.train_y)) == 2: + return self.predict_labels_binary(distances) + else: + return self.predict_labels_multiclass(distances) + + + def compute_distances_two_loops(self, X): + """ + Computes L1 distance from every sample of X to every training sample + Uses simplest implementation with 2 Python loops + + Arguments: + X, np array (num_test_samples, num_features) - samples to run + + Returns: + distances, np array (num_test_samples, num_train_samples) - array + with distances between each test and each train sample + """ + + """ + YOUR CODE IS HERE + """ + pass + + + def compute_distances_one_loop(self, X): + """ + Computes L1 distance from every sample of X to every training sample + Vectorizes some of the calculations, so only 1 loop is used + + Arguments: + X, np array (num_test_samples, num_features) - samples to run + + Returns: + distances, np array (num_test_samples, num_train_samples) - array + with distances between each test and each train sample + """ + + """ + YOUR CODE IS HERE + """ + pass + + + def compute_distances_no_loops(self, X): + """ + Computes L1 distance from every sample of X to every training sample + Fully vectorizes the calculations using numpy + + Arguments: + X, np array (num_test_samples, num_features) - samples to run + + Returns: + distances, np array (num_test_samples, num_train_samples) - array + with distances between each test and each train sample + """ + + """ + YOUR CODE IS HERE + """ + pass + + + def predict_labels_binary(self, distances): + """ + Returns model predictions for binary classification case + + Arguments: + distances, np array (num_test_samples, num_train_samples) - array + with distances between each test and each train sample + Returns: + pred, np array of bool (num_test_samples) - binary predictions + for every test sample + """ + + n_train = distances.shape[1] + n_test = distances.shape[0] + prediction = np.zeros(n_test) + + """ + YOUR CODE IS HERE + """ + pass + + + def predict_labels_multiclass(self, distances): + """ + Returns model predictions for multi-class classification case + + Arguments: + distances, np array (num_test_samples, num_train_samples) - array + with distances between each test and each train sample + Returns: + pred, np array of int (num_test_samples) - predicted class index + for every test sample + """ + + n_train = distances.shape[0] + n_test = distances.shape[0] + prediction = np.zeros(n_test, np.int) + + """ + YOUR CODE IS HERE + """ + pass diff --git a/metrics.py b/metrics.py new file mode 100644 index 0000000..2896fd2 --- /dev/null +++ b/metrics.py @@ -0,0 +1,79 @@ +import numpy as np + + +def binary_classification_metrics(y_pred, y_true): + """ + Computes metrics for binary classification + Arguments: + y_pred, np array (num_samples) - model predictions + y_true, np array (num_samples) - true labels + Returns: + precision, recall, f1, accuracy - classification metrics + """ + + # TODO: implement metrics! + # Some helpful links: + # https://en.wikipedia.org/wiki/Precision_and_recall + # https://en.wikipedia.org/wiki/F1_score + + """ + YOUR CODE IS HERE + """ + pass + + +def multiclass_accuracy(y_pred, y_true): + """ + Computes metrics for multiclass classification + Arguments: + y_pred, np array of int (num_samples) - model predictions + y_true, np array of int (num_samples) - true labels + Returns: + accuracy - ratio of accurate predictions to total samples + """ + + """ + YOUR CODE IS HERE + """ + pass + + +def r_squared(y_pred, y_true): + """ + Computes r-squared for regression + Arguments: + y_pred, np array of int (num_samples) - model predictions + y_true, np array of int (num_samples) - true values + Returns: + r2 - r-squared value + """ + y_mean = np.mean(y_true) + return 1 - ((np.square(np.subtract(y_pred, y_true))).sum())/np.square(y_true - y_mean).sum() + + + +def mse(y_pred, y_true): + """ + Computes mean squared error + Arguments: + y_pred, np array of int (num_samples) - model predictions + y_true, np array of int (num_samples) - true values + Returns: + mse - mean squared error + """ + #return np.mean((y_pred - y_true)**2) + return np.square(y_pred - y_true).mean() + + +def mae(y_pred, y_true): + """ + Computes mean absolut error + Arguments: + y_pred, np array of int (num_samples) - model predictions + y_true, np array of int (num_samples) - true values + Returns: + mae - mean absolut error + """ + return np.absolute(np.subtract(y_pred, y_true)).mean() + + \ No newline at end of file From d9103eacb3ed6b19b9ee3b0b37e3000c0c1b95f5 Mon Sep 17 00:00:00 2001 From: MariaLukina Date: Mon, 12 Feb 2024 16:49:37 +0700 Subject: [PATCH 3/6] Add KNN.ipynb file --- KNN.ipynb | 1622 +++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 1622 insertions(+) create mode 100644 KNN.ipynb diff --git a/KNN.ipynb b/KNN.ipynb new file mode 100644 index 0000000..a4f5dd8 --- /dev/null +++ b/KNN.ipynb @@ -0,0 +1,1622 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "39a37345-99a6-4b16-9be7-ebdca1414c7f", + "metadata": {}, + "source": [ + "## Домашнее задание №1 - Метод К-ближайших соседей (K-neariest neighbors)\n", + "\n", + "Сегодня мы с вами реализуем наш первый алгоритм машинного обучения, метод К-ближайших соседей. Мы попытаемся решить с помощью него задачи:\n", + "- бинарной классификации (то есть, только двум классам)\n", + "- многоклассовой классификации (то есть, нескольким классам)\n", + "- регрессии (когда зависимая переменная - натуральное число)\n", + "\n", + "Так как методу необходим гиперпараметр (hyperparameter) - количество соседей, то нам нужно научиться подбирать этот параметр. Мы постараемся научиться пользовать numpy для векторизованных вычислений, а также посмотрим на несколько метрик, которые используются в задачах классификации и регрессии.\n", + "\n", + "Перед выполнением задания:\n", + "- установите все необходимые библиотеки, запустив `pip install -r requirements.txt`\n", + "\n", + "Если вы раньше не работали с numpy или позабыли его, то можно вспомнить здесь: \n", + "http://cs231n.github.io/python-numpy-tutorial/" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "9638c464-806f-41b5-9dfe-1ea2048a1fa1", + "metadata": {}, + "outputs": [], + "source": [ + "import time\n", + "import matplotlib.pyplot as plt\n", + "import seaborn as sns\n", + "import numpy as np\n", + "import random\n", + "import pandas as pd\n", + "\n", + "\n", + "from sklearn.datasets import fetch_openml\n", + "from sklearn.model_selection import train_test_split\n", + "from knn import KNNClassifier\n", + "from metrics import binary_classification_metrics, multiclass_accuracy" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "43bd8dc9-c430-4313-a6a1-4d3e5a7e9c47", + "metadata": {}, + "outputs": [], + "source": [ + "plt.rcParams[\"figure.figsize\"] = 12, 9\n", + "sns.set_style(\"whitegrid\")\n", + "\n", + "SEED = 111\n", + "random.seed(SEED)\n", + "np.random.seed(SEED)" + ] + }, + { + "cell_type": "markdown", + "id": "2867b963-214c-49ea-9460-5b427b56544d", + "metadata": {}, + "source": [ + "## Задание 1. KNN на датасете Fashion-MNIST (10 баллов)" + ] + }, + { + "cell_type": "markdown", + "id": "60a90da7-87ac-42e6-b376-bb34dac2b10b", + "metadata": {}, + "source": [ + "В этом задании вам предстоит поработать с картинками одежды, среди которых можно выделить 10 классов. Данные уже загружены за вас: в переменной X лежат 70000 картинок размером 28 на 28 пикселей, вытянутые в вектор размерностью 784 (28 * 28). Так как данных довольно много, а наш KNN будет весьма медленный, то возьмем случайно 1000 наблюдений (в реальности в зависимости от вашей реализации можно будет взять больше, но если будет не зватать ОЗУ, то берите меньше)." + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "54fa4253-ea6a-4ec4-b914-f7cb2346b195", + "metadata": {}, + "outputs": [], + "source": [ + "X, y = fetch_openml(name=\"Fashion-MNIST\", return_X_y=True, as_frame=False)" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "3a188c83-6bf3-485d-9995-9f71d0868d30", + "metadata": {}, + "outputs": [], + "source": [ + "idx_to_stay = np.random.choice(np.arange(X.shape[0]), replace=False, size=1000)\n", + "X = X[idx_to_stay]\n", + "y = y[idx_to_stay]" + ] + }, + { + "cell_type": "markdown", + "id": "4a9e7f89-97f9-4257-94aa-989826258726", + "metadata": {}, + "source": [ + "Давайте посмотрим на какое-нибудь изображение из наших данных:" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "277e132c-b89f-4dbb-8efd-cbcea015876d", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# возьмем случайную картинку и сделаем reshape\n", + "# 28, 28, 1 = H, W, C (число каналов, в данном случае 1)\n", + "image = X[np.random.choice(np.arange(X.shape[0]))].reshape(28, 28, 1)\n", + "plt.imshow(image)\n", + "plt.axis(\"off\");" + ] + }, + { + "cell_type": "markdown", + "id": "236e593f-595e-45f1-a794-4069a16637d7", + "metadata": {}, + "source": [ + "### 1.1. Посмотрим на все классы (0.5 баллов)" + ] + }, + { + "cell_type": "markdown", + "id": "8cdf3ab2-47a4-492f-bf9a-25c4b00eb945", + "metadata": {}, + "source": [ + "Возьмите по одной картинке каждого класса и изобразите их (например, сделайте subplots 5 на 2)." + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "362137fb-4577-4a21-8088-98cba79206f2", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "labels = []\n", + "for i in range(0, 10):\n", + " labels.append(np.where(y == str(i))[0][0])\n", + "plt.figure(figsize = [12, 7])\n", + "for i in range(0, 10):\n", + " plt.subplot(2, 5, i+1)\n", + " t_image = X[labels[i]].reshape(28, 28, 1)\n", + " plt.imshow(t_image)\n", + " plt.axis(\"off\");" + ] + }, + { + "cell_type": "markdown", + "id": "866ea214-4de8-41b8-a2aa-c86ac0a04b74", + "metadata": {}, + "source": [ + "### 1.2. Сделайте небольшой EDA (1 балл)" + ] + }, + { + "cell_type": "markdown", + "id": "1fe3abdf-2c95-4ce8-8fd3-2445d815ea3c", + "metadata": {}, + "source": [ + "Посмотрите на баланс классов. В дальнейших домашках делайте EDA, когда считаете нужным, он нужен почти всегда, но оцениваться это уже не будет, если не будет указано иное. Делайте EDA, чтобы узнать что-то новое о данных!" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "74595ef7-06ab-4700-b9d9-42db3fdd36d5", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "уникальные значения и количество вхождений\n", + "обозначения типов одежды: ['0' '1' '2' '3' '4' '5' '6' '7' '8' '9']\n", + "количество картинок разного типа: [ 93 102 95 121 96 98 101 79 109 106]\n", + "диаграмма баланса классов\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "#посмотрим на размерность данных (хотя мы и так сами отобрали 1000)\n", + "np.shape (y)\n", + "# проверим вхождения уникальных значений, в данном случае, видим, что данные сбалансированы\n", + "print ('уникальные значения и количество вхождений')\n", + "labels, counts = np.unique (y, return_counts= True )\n", + "print('обозначения типов одежды: ', labels)\n", + "print('количество картинок разного типа: ', counts)\n", + "print ('диаграмма баланса классов')\n", + "labels = ['0', '1', '2', '3', '4', '5', '6', '7', '8', '9']\n", + "plt.pie([ 93, 102, 95, 121, 96, 98, 101, 79, 109, 106], labels=labels)\n", + "plt.show()\n" + ] + }, + { + "cell_type": "markdown", + "id": "68e8f61e-32d5-4ad7-9f2c-f7e0de050d32", + "metadata": {}, + "source": [ + "### 1.3. Разделите данные на train и test (0.5 баллов)" + ] + }, + { + "cell_type": "markdown", + "id": "25cf3d30-6bd6-4bbb-bba6-4e249da33475", + "metadata": {}, + "source": [ + "Разделите данные на тренировочную и тестовую выборки, размеры тестовой выборки выберите сами. Здесь вам может помочь функция `train_test_split`" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "1932bd43-16d6-4201-8950-7dbe720a9fa1", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "870\n", + "870\n" + ] + } + ], + "source": [ + "train_X, test_X, train_y, test_y = train_test_split(X,\n", + " y,\n", + " test_size = 0.13,\n", + " random_state=SEED)\n", + "train_y\n", + "test_y\n", + "print(len(train_X))\n", + "print(train_X.shape[0])" + ] + }, + { + "cell_type": "markdown", + "id": "7c8e4cd4-b3d7-49b7-9b10-be02991ecfa7", + "metadata": {}, + "source": [ + "KNN для бинарной классификации (6 баллов)" + ] + }, + { + "cell_type": "markdown", + "id": "aac2e121-639a-4b0c-8e9c-471ad5a8fac6", + "metadata": {}, + "source": [ + "Давайте возьмем для задачи бинарной классификации только объекты с метками классов 0 и 1." + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "f40beae7-54a4-4323-b467-3173737dfd84", + "metadata": {}, + "outputs": [], + "source": [ + "mask_train = np.logical_or(train_y == '0', train_y == '1')\n", + "binary_train_X = train_X[mask_train]\n", + "binary_train_y = train_y[mask_train]\n", + "mask_test = np.logical_or(test_y == '0', test_y == '1')\n", + "binary_test_X = test_X[mask_test]\n", + "binary_test_y = test_y[mask_test]\n", + "#print(binary_test_y)\n", + "#print(len(binary_test_X))\n", + "#print(binary_test_y.shape[0])\n", + "\n", + "\n", + "#for i in range(0, 25):\n", + "# plt.subplot(5, 5, i+1)\n", + "# t_image = binary_test_X[[i]].reshape(28, 28, 1)\n", + "# plt.imshow(t_image)\n", + "# plt.axis(\"off\");\n" + ] + }, + { + "cell_type": "markdown", + "id": "7df7db35-8832-47ec-9955-d0656695e7cd", + "metadata": {}, + "source": [ + "И вот мы подготовили данные, но модели у нас пока что нет. В нескольких занятиях нашего курса вам придется самостоятельно реализовывать какие-то алгоритмы машинного обучения, а потом сравнивать их с готовыми библиотечными решениями. В остальных заданиях реализовывать алгоритмы будет не обязательно, но может быть полезно, поэтому часто это будут задания на дополнительные баллы, но главное не это, а понимание работы алгоритма после его реализации с нуля на простом numpy. Также это все потом можно оформить в виде репозитория ml_from_scratch и хвастаться перед друзьями." + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "44d468e9-2a00-4268-bfdc-fbae3857ce90", + "metadata": {}, + "outputs": [], + "source": [ + "knn_classifier = KNNClassifier(k=5)\n", + "knn_classifier.fit(binary_train_X, binary_train_y)" + ] + }, + { + "cell_type": "markdown", + "id": "c5817a1d-161e-4242-bea5-821f416b3eec", + "metadata": {}, + "source": [ + "### Настало время писать код!" + ] + }, + { + "cell_type": "markdown", + "id": "61c760bb-63c9-426d-9f4e-8fab02536da5", + "metadata": {}, + "source": [ + "В KNN нам нужно для каждого тестового примера найти расстояния до всех точек обучающей выборки. Допустим у нас 1000 примеров в train'е и 100 в test'е, тогда в итоге мы бы хотели получить матрицу попарных расстояний (например, размерностью 100 на 1000). Это можно сделать несколькими способами, и кому-то наверняка, в голову приходит идея с двумя вложенными циклами (надеюсь, что не больше:). Так можно делать, то можно и эффективнее. Вообще, в реальном KNN используется структура данных [k-d-tree](https://ru.wikipedia.org/wiki/K-d-%D0%B4%D0%B5%D1%80%D0%B5%D0%B2%D0%BE), которая позволяет производить поиск за log(N), а не за N, как будем делать мы (по сути это такое расширение бинарного поиска на многомерное пространство).\n", + "\n", + "Вам нужно будет последовательно реализовать методы `compute_distances_two_loops`, `compute_distances_one_loop` и `compute_distances_no_loops` класса `KNN` в файле `knn.py`.\n", + "\n", + "Эти функции строят массив расстояний между всеми векторами в тестовом наборе и в тренировочном наборе. \n", + "В результате они должны построить массив размера `(num_test, num_train)`, где координата `[i][j]` соотвествует расстоянию между i-м вектором в test (`test[i]`) и j-м вектором в train (`train[j]`).\n", + "\n", + "**Обратите внимание** Для простоты реализации мы будем использовать в качестве расстояния меру L1 (ее еще называют [Manhattan distance](https://ru.wikipedia.org/wiki/%D0%A0%D0%B0%D1%81%D1%81%D1%82%D0%BE%D1%8F%D0%BD%D0%B8%D0%B5_%D0%B3%D0%BE%D1%80%D0%BE%D0%B4%D1%81%D0%BA%D0%B8%D1%85_%D0%BA%D0%B2%D0%B0%D1%80%D1%82%D0%B0%D0%BB%D0%BE%D0%B2)).\n", + "\n", + "$d_{1}(\\mathbf {p} ,\\mathbf {q} )=\\|\\mathbf {p} -\\mathbf {q} \\|_{1}=\\sum _{i=1}^{n}|p_{i}-q_{i}|$" + ] + }, + { + "cell_type": "markdown", + "id": "c32db2d0-355c-4d74-961e-22d6f42aa11b", + "metadata": {}, + "source": [ + "В начале я буду иногда писать разные assert'ы, чтобы можно было проверить правильность реализации, в дальнейшем вам нужно будет их писать самим, если нужно будет проверять корректность каких-то вычислений." + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "01b1ef27-4284-4d6c-978b-25f0fefd39be", + "metadata": {}, + "outputs": [ + { + "ename": "TypeError", + "evalue": "'NoneType' object is not subscriptable", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mTypeError\u001b[0m Traceback (most recent call last)", + "Cell \u001b[0;32mIn[13], line 3\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[38;5;66;03m# TODO: compute_distances_two_loops\u001b[39;00m\n\u001b[1;32m 2\u001b[0m dists \u001b[38;5;241m=\u001b[39m knn_classifier\u001b[38;5;241m.\u001b[39mcompute_distances_two_loops(binary_test_X)\n\u001b[0;32m----> 3\u001b[0m \u001b[38;5;28;01massert\u001b[39;00m np\u001b[38;5;241m.\u001b[39misclose(\u001b[43mdists\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;241;43m0\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m100\u001b[39;49m\u001b[43m]\u001b[49m, np\u001b[38;5;241m.\u001b[39msum(np\u001b[38;5;241m.\u001b[39mabs(binary_test_X[\u001b[38;5;241m0\u001b[39m] \u001b[38;5;241m-\u001b[39m binary_train_X[\u001b[38;5;241m100\u001b[39m])))\n", + "\u001b[0;31mTypeError\u001b[0m: 'NoneType' object is not subscriptable" + ] + } + ], + "source": [ + "# TODO: compute_distances_two_loops\n", + "dists = knn_classifier.compute_distances_two_loops(binary_test_X)\n", + "assert np.isclose(dists[0, 100], np.sum(np.abs(binary_test_X[0] - binary_train_X[100])))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "505e2c4b-1cfe-4e4a-8002-d9ce089d100e", + "metadata": {}, + "outputs": [], + "source": [ + "# TODO: compute_distances_one_loop\n", + "dists = knn_classifier.compute_distances_one_loop(binary_test_X)\n", + "assert np.isclose(dists[0, 100], np.sum(np.abs(binary_test_X[0] - binary_train_X[100])))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "dd81b766-5de2-4f62-82cf-8b6bd52002db", + "metadata": {}, + "outputs": [], + "source": [ + "# TODO: compute_distances_no_loops\n", + "dists = knn_classifier.compute_distances_no_loops(binary_test_X)\n", + "assert np.isclose(dists[0, 100], np.sum(np.abs(binary_test_X[0] - binary_train_X[100])))" + ] + }, + { + "cell_type": "markdown", + "id": "64d108b7-b132-42b5-bdca-bc91af8ed3d7", + "metadata": {}, + "source": [ + "Проверим скорость работы реализованных методов" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "8ed08354-d0ef-497a-9d9c-b17c92b7bef9", + "metadata": {}, + "outputs": [], + "source": [ + "%timeit knn_classifier.compute_distances_two_loops(binary_test_X)\n", + "%timeit knn_classifier.compute_distances_one_loop(binary_test_X)\n", + "%timeit knn_classifier.compute_distances_no_loops(binary_test_X)" + ] + }, + { + "cell_type": "markdown", + "id": "8180ecc3-8c24-4564-8963-1a0123b06043", + "metadata": {}, + "source": [ + "Реализуем метод для предсказания меток класса" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9ca2679e-c731-467b-94b5-91a3a0641d7e", + "metadata": {}, + "outputs": [], + "source": [ + "# TODO: predict_labels_binary in knn.py\n", + "prediction = knn_classifier.predict(binary_test_X)" + ] + }, + { + "cell_type": "markdown", + "id": "d746796d-6ca0-4828-be8a-8b9e22999f54", + "metadata": {}, + "source": [ + "### Метрика" + ] + }, + { + "cell_type": "markdown", + "id": "c29f2abf-be34-4273-a7cd-d2c62ba1eecc", + "metadata": {}, + "source": [ + "Теперь нужно реализовать несколько метрик для бинарной классификации. Не забудьте подумать о численной нестабильности (деление на 0)." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "94268969-5334-4672-a939-65733068a89a", + "metadata": {}, + "outputs": [], + "source": [ + "# TODO: binary_classification_metrics in metrics.py" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "7842978b-a328-4035-b9c4-eb3711cb58c4", + "metadata": {}, + "outputs": [], + "source": [ + "binary_classification_metrics(prediction, binary_test_y)" + ] + }, + { + "cell_type": "markdown", + "id": "71dbca1f-9ad4-4059-9b1c-0c0c85d7d816", + "metadata": {}, + "source": [ + "Все ли хорошо с моделью? Можно проверить свою реализацию с функциями из библиотеки `sklearn`:" + ] + }, + { + "cell_type": "markdown", + "id": "5982d081-ddf9-4522-99b0-459a5cee087c", + "metadata": {}, + "source": [ + "" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "5caeb94f-6464-4adf-b3b6-8e12bf4a50b4", + "metadata": {}, + "outputs": [], + "source": [ + "from sklearn.metrics import precision_score, recall_score, f1_score, accuracy_score" + ] + }, + { + "cell_type": "markdown", + "id": "6e9b0e1a-3c67-4cc3-bf75-6848f43e8256", + "metadata": {}, + "source": [ + "### Подбор оптимального k" + ] + }, + { + "cell_type": "markdown", + "id": "4069069e-f200-4673-a99c-745e7a5b6b36", + "metadata": {}, + "source": [ + "Чтобы подрбрать оптимальное значение параметра k можно сделать следующее: задать область допустимых значений k, например, `[1, 3, 5, 10]`. Дальше для каждого k обучить модель на тренировочных данных, сделать предсказания на тестовых и посчитать какую-нибудь метрику (метрику выберите сами исходя из задачи, но постарайтесь обосновать выбор). В конце нужно посмотреть на зависимость метрики на train'е и test'е от k и выбрать подходящее значение.\n", + "\n", + "Реализуйте функцию `choose_best_k` прямо в ноутбуке." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "374d6fcf-21f2-433c-b011-76019201ce52", + "metadata": {}, + "outputs": [], + "source": [ + "def find_best_k(X_train, y_train, X_test, y_test, params, metric):\n", + " \"\"\"\n", + " Choose the best k for KKNClassifier\n", + " Arguments:\n", + " X_train, np array (num_train_samples, num_features) - train data\n", + " y_train, np array (num_train_samples) - train labels\n", + " X_test, np array (num_test_samples, num_features) - test data\n", + " y_test, np array (num_test_samples) - test labels\n", + " params, list of hyperparameters for KNN, here it is list of k values\n", + " metric, function for metric calculation\n", + " Returns:\n", + " train_metrics the list of metric values on train data set for each k in params\n", + " test_metrics the list of metric values on test data set for each k in params\n", + " \"\"\"\n", + " \n", + " \"\"\"\n", + " YOUR CODE IS HERE\n", + " \"\"\"\n", + " pass" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "a2418dd8-93f1-4a11-8488-a8bce98e7d5f", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "params = [1, 2, 4, 5, 8, 10, 30]\n", + "train_metrics, test_metrics = find_best_k(binary_train_X, binary_train_y, binary_test_X, binary_test_y, params, accuracy_score)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "7053ce06-854c-412b-8559-833595b1d6c0", + "metadata": {}, + "outputs": [], + "source": [ + "plt.plot(params, train_metrics, label=\"train\")\n", + "plt.plot(params, test_metrics, label=\"test\")\n", + "plt.legend()\n", + "plt.xlabel(\"K in KNN\")\n", + "plt.ylabel(\"YOUR METRIC\");" + ] + }, + { + "cell_type": "markdown", + "id": "73fdecd7-a4b6-4ca9-8688-d1ae2195647f", + "metadata": {}, + "source": [ + "На самом деле, это не самый лучший способ подбирать гиперпараметры, но способы получше мы рассмотрим в следующий раз, а пока что выберите оптимальное значение k, сделайте предсказания и посмотрите, насколько хорошо ваша модель предсказывает каждый из классов." + ] + }, + { + "cell_type": "markdown", + "id": "0bc98c29-3217-407c-a466-58072bb7b8cc", + "metadata": {}, + "source": [ + "### 1.5. Многоклассоввая классификация (2 балла)" + ] + }, + { + "cell_type": "markdown", + "id": "aa0fa9bf-3002-4b0e-8e52-1d9e70795092", + "metadata": {}, + "source": [ + "Теперь нужно научиться предсказывать все 10 классов. Для этого в начале напишем соответствующий метод у нашего классификатора." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "88b37ef1-fa15-4cc5-8558-0254890c899d", + "metadata": {}, + "outputs": [], + "source": [ + "# TODO: predict_labels_multiclass in knn.py\n", + "knn_classifier = KNNClassifier(k=1)\n", + "knn_classifier.fit(X_train, y_train)\n", + "predictions = knn_classifier.predict(X_test)" + ] + }, + { + "cell_type": "markdown", + "id": "fa8ed7ad-e347-4f78-b34f-88febee9e94b", + "metadata": {}, + "source": [ + "Осталось реализовать метрику качества для многоклассовой классификации, для этого реализуйте функцию `multiclass_accuracy` в `metrics.py`." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "65887922-a799-4042-902e-f63f69508beb", + "metadata": {}, + "outputs": [], + "source": [ + "# TODO: multiclass_accuracy in metrics.py\n", + "multiclass_accuracy(predictions, y_test)" + ] + }, + { + "cell_type": "markdown", + "id": "a6fe44f6-e056-4f17-beb6-0ce65fab268f", + "metadata": {}, + "source": [ + "Снова выберите оптимальное значение K как мы делали для бинарной классификации." + ] + }, + { + "cell_type": "markdown", + "id": "daa8ee4a-88c1-4967-84f1-6f7ec223db7e", + "metadata": {}, + "source": [ + "## Задание 2. KNN на датасете diabetes (10 баллов)" + ] + }, + { + "cell_type": "markdown", + "id": "d3dac1f9-42ef-406f-988c-2d663b8b2806", + "metadata": {}, + "source": [ + "Теперь попробуем применить KNN к задаче регрессии. Будем работать с [данными](https://scikit-learn.org/stable/datasets/toy_dataset.html#diabetes-dataset) о диабете. В этом задании будем использовать класс `KNeighborsRegressor` из библиотеки `sklearn`. Загрузим необходимые библиотеки:" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "8ab4c84e-6036-4fe2-b81f-8115bf0a4774", + "metadata": {}, + "outputs": [], + "source": [ + "from sklearn.datasets import load_diabetes\n", + "from sklearn.metrics import r2_score, mean_absolute_error, mean_squared_error\n", + "from sklearn.preprocessing import StandardScaler\n", + "from sklearn.neighbors import KNeighborsRegressor" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "4136b3bb-e482-4102-ad5f-af5b3b1a030a", + "metadata": {}, + "outputs": [], + "source": [ + "X, y = load_diabetes(as_frame=True, return_X_y=True)" + ] + }, + { + "cell_type": "markdown", + "id": "9ce6cd71-dd17-40af-b7cc-c48e61d31719", + "metadata": {}, + "source": [ + "### 2.1. EDA (2 обязательных балла + 2 доп. балла за Pipeline)" + ] + }, + { + "cell_type": "markdown", + "id": "4397b3ad-63f9-4b25-ab7e-d572216e21c5", + "metadata": {}, + "source": [ + "Сделайте EDA, предобработайте данные так, как считаете нужным, нужна ли в данном случае стандартизация и почему? Не забудте, что если вы стандартизуете данные, то нужно считать среднее и сдандартное отклонение на тренировочной части и с помощью них трансформировать и train, и test (**если не поняли это предложение, то обязательно разберитесь**).\n", + "\n", + "**Дополнительно**:\n", + "Попробуйте разобраться с [`Pipeline`](https://scikit-learn.org/stable/modules/generated/sklearn.pipeline.Pipeline.html), чтобы можно было создать класс, который сразу проводит стандартизацию и обучает модель (или делает предсказание). Пайплайны очень удобны, когда нужно применять различные методы предобработки данных (в том числе и к разным столбцам), а также они позволяют правильно интегрировать предобработку данных в различные классы для поиска наилучших гиперпараметров модели (например, `GridSearchCV`)." + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "id": "a4aa413d-d830-4355-a063-9c5d6d5a6c9a", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + " In dataframe X (Number or rows, number of columns) = (442, 10) \n", + "\n", + "Number of dublicates = 0 \n", + "\n", + "In this table you can find data types for all columns: \n", + "\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
0
agefloat64
sexfloat64
bmifloat64
bpfloat64
s1float64
s2float64
s3float64
s4float64
s5float64
s6float64
\n", + "
" + ], + "text/plain": [ + " 0\n", + "age float64\n", + "sex float64\n", + "bmi float64\n", + "bp float64\n", + "s1 float64\n", + "s2 float64\n", + "s3 float64\n", + "s4 float64\n", + "s5 float64\n", + "s6 float64" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Summary for basic statistics for all numeric columns:\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
agesexbmibps1s2s3s4s5s6
count4.420000e+024.420000e+024.420000e+024.420000e+024.420000e+024.420000e+024.420000e+024.420000e+024.420000e+024.420000e+02
mean-2.511817e-191.230790e-17-2.245564e-16-4.797570e-17-1.381499e-173.918434e-17-5.777179e-18-9.042540e-189.293722e-171.130318e-17
std4.761905e-024.761905e-024.761905e-024.761905e-024.761905e-024.761905e-024.761905e-024.761905e-024.761905e-024.761905e-02
min-1.072256e-01-4.464164e-02-9.027530e-02-1.123988e-01-1.267807e-01-1.156131e-01-1.023071e-01-7.639450e-02-1.260971e-01-1.377672e-01
25%-3.729927e-02-4.464164e-02-3.422907e-02-3.665608e-02-3.424784e-02-3.035840e-02-3.511716e-02-3.949338e-02-3.324559e-02-3.317903e-02
50%5.383060e-03-4.464164e-02-7.283766e-03-5.670422e-03-4.320866e-03-3.819065e-03-6.584468e-03-2.592262e-03-1.947171e-03-1.077698e-03
75%3.807591e-025.068012e-023.124802e-023.564379e-022.835801e-022.984439e-022.931150e-023.430886e-023.243232e-022.791705e-02
max1.107267e-015.068012e-021.705552e-011.320436e-011.539137e-011.987880e-011.811791e-011.852344e-011.335973e-011.356118e-01
\n", + "
" + ], + "text/plain": [ + " age sex bmi bp s1 \\\n", + "count 4.420000e+02 4.420000e+02 4.420000e+02 4.420000e+02 4.420000e+02 \n", + "mean -2.511817e-19 1.230790e-17 -2.245564e-16 -4.797570e-17 -1.381499e-17 \n", + "std 4.761905e-02 4.761905e-02 4.761905e-02 4.761905e-02 4.761905e-02 \n", + "min -1.072256e-01 -4.464164e-02 -9.027530e-02 -1.123988e-01 -1.267807e-01 \n", + "25% -3.729927e-02 -4.464164e-02 -3.422907e-02 -3.665608e-02 -3.424784e-02 \n", + "50% 5.383060e-03 -4.464164e-02 -7.283766e-03 -5.670422e-03 -4.320866e-03 \n", + "75% 3.807591e-02 5.068012e-02 3.124802e-02 3.564379e-02 2.835801e-02 \n", + "max 1.107267e-01 5.068012e-02 1.705552e-01 1.320436e-01 1.539137e-01 \n", + "\n", + " s2 s3 s4 s5 s6 \n", + "count 4.420000e+02 4.420000e+02 4.420000e+02 4.420000e+02 4.420000e+02 \n", + "mean 3.918434e-17 -5.777179e-18 -9.042540e-18 9.293722e-17 1.130318e-17 \n", + "std 4.761905e-02 4.761905e-02 4.761905e-02 4.761905e-02 4.761905e-02 \n", + "min -1.156131e-01 -1.023071e-01 -7.639450e-02 -1.260971e-01 -1.377672e-01 \n", + "25% -3.035840e-02 -3.511716e-02 -3.949338e-02 -3.324559e-02 -3.317903e-02 \n", + "50% -3.819065e-03 -6.584468e-03 -2.592262e-03 -1.947171e-03 -1.077698e-03 \n", + "75% 2.984439e-02 2.931150e-02 3.430886e-02 3.243232e-02 2.791705e-02 \n", + "max 1.987880e-01 1.811791e-01 1.852344e-01 1.335973e-01 1.356118e-01 " + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Overall number of missing values: 0\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Summary for y data\n" + ] + }, + { + "data": { + "text/plain": [ + "count 442.000000\n", + "mean 152.133484\n", + "std 77.093005\n", + "min 25.000000\n", + "25% 87.000000\n", + "50% 140.500000\n", + "75% 211.500000\n", + "max 346.000000\n", + "Name: target, dtype: float64" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import seaborn as sns\n", + "import pylab\n", + "\n", + "\n", + "def run_eda(df): \n", + " print(\"\\n In dataframe X (Number or rows, number of columns) = \", df.shape, \"\\n\")\n", + " print('Number of dublicates =', len(df) - len(df.drop_duplicates()), \"\\n\") \n", + " print(\"In this table you can find data types for all columns:\", \"\\n\" )\n", + " display(df.dtypes.to_frame())\n", + " print('Summary for basic statistics for all numeric columns:')\n", + " display(df.describe())\n", + " print()\n", + " print('Overall number of missing values:', df.isna().sum().sum())\n", + "\n", + " # создание heatplot корреляции\n", + " tr = np.triu(df.corr())\n", + " pylab.subplot (1, 2, 2)\n", + " sns.heatmap(df.corr(), annot = True, cbar_kws= {'orientation': 'horizontal'}, vmin=-1, vmax=1, center= 0, square=True, cmap=\"vlag\", mask = tr)\n", + " pylab.title (\"Корреляция между численными переменными\")\n", + " pylab.show()\n", + " return\n", + "\n", + "\n", + "run_eda(X)\n", + "print('Summary for y data')\n", + "display(y.describe())\n", + "\n", + "#from sklearn.pipeline import Pipeline эта часть будет в конце ноутбука" + ] + }, + { + "cell_type": "raw", + "id": "35da5f1f-c18c-4164-ade1-5ee2793ce2bd", + "metadata": {}, + "source": [ + "Мы видим, что датасет хорошего качества, нет пропущенных значений, а данные уже стандартизованы ( все колонки, в том числе возраст имеют числовой тип,по всем колонкам одинаковое стандартное отклонение, а среднее значение равно нулю), некоторые переменные сильно коррелирую между собой.Наверное (хотя я не уверена вообще) в данном случае это может быть не очень хорошо, в частности коллерируют содержание холестерина и каких-то липопротеинов, вроде, это довльно ожидаемо, и как-то будто увеличивает размерность, не давая новой информации. Но я дилетант, и интересно было бы про такое поговорить.\n", + "\n", + "\n", + "Работа с исходным датасетом и составление pipeline будет в конце ноутбука" + ] + }, + { + "cell_type": "markdown", + "id": "8e1fa5ec-18db-4e7b-88f1-6d010f449488", + "metadata": {}, + "source": [ + "### 2.2. Регрессионная модель (1 балл)\n" + ] + }, + { + "cell_type": "markdown", + "id": "3d79f583-5feb-4eeb-b748-9b4cdea9150d", + "metadata": {}, + "source": [ + "Создайте модель `KNeighborsRegressor`, обучите ее на тренировочных данных и сделайте предсказания." + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "75690c55-b60b-435f-a6e0-a3a39975863e", + "metadata": {}, + "outputs": [], + "source": [ + "X_train, X_test,y_train, y_test = train_test_split(X,\n", + " y,\n", + " test_size = 0.15,\n", + " random_state=SEED)\n", + "knn = KNeighborsRegressor(n_neighbors=1)\n", + "knn.fit(X_train, y_train)\n", + "y_pred = knn.predict(X_test)\n", + "#type(y_train)\n", + "#y_test" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e0ea1d90-fd85-4304-8eed-fa5e9c68cb32", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "id": "5ed4a2ea-4f42-43cb-95d7-50cdeab5f284", + "metadata": {}, + "source": [ + "### 2.3. Метрики регресии (3 балла)" + ] + }, + { + "cell_type": "markdown", + "id": "cbf563ad-1b71-464c-8359-75fa4da268d3", + "metadata": {}, + "source": [ + "Реализуйте метрики $R^2$, MSE и MAE в `metrics.py`. Примените их для оценки качества полученной модели. Все ли хорошо?\n", + "\n", + "Напомню, что:\n", + "\n", + "$R^2 = 1 - \\frac{\\sum_i^n{(y_i - \\hat{y_i})^2}}{\\sum_i^n{(y_i - \\overline{y})^2}}$\n", + "\n", + "$MSE = \\frac{1}{n}\\sum_i^n{(y_i - \\hat{y_i})^2}$\n", + "\n", + "$MAE = \\frac{1}{n}\\sum_i^n{|y_i - \\hat{y_i}|}$" + ] + }, + { + "cell_type": "code", + "execution_count": 45, + "id": "580f604f-e9c5-4109-8b9f-235369836057", + "metadata": {}, + "outputs": [], + "source": [ + "# TODO: r_squared, mse, mae in metrics.py\n" + ] + }, + { + "cell_type": "markdown", + "id": "a6568d8d-a9ec-4639-aac1-0ea18a644f9e", + "metadata": {}, + "source": [ + "### 2.4. Подбор оптимального числа соседей (2 балла)" + ] + }, + { + "cell_type": "markdown", + "id": "d82f145d-41f5-4bbe-b553-1d05d90ec2a2", + "metadata": {}, + "source": [ + "Мы почти дошли до конца. Теперь осталось при помощи реализованных нами метрик выбрать лучшее количество соседей для нашей модели.\n", + "\n", + "!!! Обратите внимание на то, что значат наши метрики, для некоторых хорошо, когда они уменьшаются, для других наоборот." + ] + }, + { + "cell_type": "code", + "execution_count": 46, + "id": "138a7969-8079-4a38-b92e-8eae5d246325", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "MSE, MAE, R_squared for k =1 5078.2835820895525 54.223880597014926 0.014816337639925625\n" + ] + } + ], + "source": [ + "from metrics import r_squared, mse, mae\n", + "\n", + "#a = np.mean((y_test - y_pred)**2)\n", + "#mse(y_pred, y_test)\n", + "#mae = mae(y_pred, y_test)\n", + "#r2 = r_squared(y_pred, y_test)\n", + "#y_mean = np.mean(y_test)\n", + "#e = 1 - ((np.square(np.subtract(y_pred, y_test))).sum())/np.square(y_test - y_mean).sum()\n", + "print('MSE, MAE, R_squared for k =1 ', \n", + " mse(y_pred, y_test), mae(y_pred, y_test), r_squared(y_pred, y_test))\n", + "\n", + "#видим, что пока значения не очень, MAE составляет порядка 30 % от среднего значения, а R2, и вовсе отрицательный,\n", + "#что не похоже на прогностическую ценность. Надо подбирать k \n" + ] + }, + { + "cell_type": "code", + "execution_count": 53, + "id": "2d5d2005-c7fd-4562-8b7b-789565d3c6ca", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "k r2 r2 delta (test and train) mae mae delta(test and train) \n", + "\n", + "1 0.014816337639925625 0.9851836623600744 54.223880597014926 54.223880597014926\n", + "2 0.2617753256003845 0.48178636278171694 47.88805970149254 17.94939303482587\n", + "3 0.25599129801966247 0.3889634556994489 47.32835820895522 11.583469320066335\n", + "5 0.38465042477168876 0.18995196681989002 42.27164179104477 2.4801751243781\n", + "7 0.40045541093882087 0.15505910223614883 44.366737739872065 3.482166311300638\n", + "9 0.38919995987749045 0.15890463872661642 44.93366500829188 3.7964798231066936\n", + "10 0.41525631076956926 0.11179650737362745 44.076119402985064 1.8713194029850584\n", + "11 0.4287646963413788 0.09020873142717345 44.49253731343283 1.861749434644942\n", + "14 0.4179365360380808 0.10689235837425559 44.55863539445629 1.8997782515991517\n", + "17 0.40953642363404297 0.10346161199556103 45.34679543459174 1.983658179689776\n", + "20 0.4218166735193798 0.08652367631892344 44.31641791044776 0.3664179104477583\n", + "25 0.4164088787996362 0.08576495420845631 44.862686567164175 0.15409990049750633\n", + "30 0.4077641109586141 0.07991535134962957 45.08855721393035 0.3607761194029848\n", + "35 0.382184398292995 0.09964317441701842 45.70959488272921 0.3688813077469817\n" + ] + } + ], + "source": [ + "\n", + "knn = KNeighborsRegressor(n_neighbors=1)\n", + "knn.fit(X_train, y_train)\n", + "y_pred = knn.predict(X_test)\n", + "n_neighbors = [1, 2, 3, 5, 7, 9, 10, 11, 14, 17, 20, 25, 30, 35]\n", + "r2_scores = []\n", + "mse_scores = []\n", + "mae_scores = []\n", + "#functions = [mse, mae, r_squared]\n", + "print('k r2 r2 delta (test and train) mae mae delta(test and train) ')\n", + "print()\n", + "for i in n_neighbors:\n", + " knn = KNeighborsRegressor(n_neighbors=i)\n", + " knn.fit(X_train, y_train)\n", + " y_pred = knn.predict(X_test)\n", + " y_pred_train = knn.predict(X_train)\n", + " #mse = mse(y_pred, y_test) уберем, так ка дублирует по сути mae в данном случае\n", + " mae_v = mae(y_pred, y_test)\n", + " r2_v = r_squared(y_pred, y_test)\n", + " delta_mae = abs(mae(y_pred_train, y_train) - mae_v)\n", + " delta_r2 = abs(r_squared(y_pred_train, y_train) - r2_v)\n", + " #mse_scores += mse\n", + " mae_scores += [mae_v]\n", + " r2_scores += [r2_v]\n", + " print(i, ' ', r2_v, ' ', delta_r2, ' ', mae_v, ' ' , delta_mae)\n", + " " + ] + }, + { + "cell_type": "raw", + "id": "19648cbd-274a-403d-87a6-947a0ebd514c", + "metadata": {}, + "source": [ + "Выберем оптимальное k исходя из максимального значения r_squared и минимального mae, видим, что это происходит при k = 10 - 11, \n", + "Также можно еще посмотреть на разницу между метриками, посчитанными для train и выборками, она становится минимальной для k = 20, \n", + "Но после k = 10, скорость изменений очень снижается\n", + "Поэтому я выбрала бы 11\n" + ] + }, + { + "cell_type": "markdown", + "id": "1b4bbef7-35a4-4f05-abf8-5d2b450c86f2", + "metadata": {}, + "source": [ + "Для поиска лучшего k вы можете воспользоваться функцией `find_best_k`, которую вы реализовали выше." + ] + }, + { + "cell_type": "markdown", + "id": "2cb77960-fa30-4a29-9a1b-7c195cc867cb", + "metadata": {}, + "source": [ + "### 3. Социализация (0.5 доп. балла)\n", + "\n", + "Так как у нас теперь большая группа, то было бы здорово всем познакомиться получше (так как выпускной не за горами). Соберитесь с одногруппниками в зуме, познакомьтесь, а сюда прикрепите скриншот с камерами всех участников." + ] + }, + { + "cell_type": "markdown", + "id": "e116a42f-fae8-499c-a985-dc09e66a29b0", + "metadata": {}, + "source": [ + "## Therapy time" + ] + }, + { + "cell_type": "markdown", + "id": "031c493b-26f3-4622-a1f4-affee64f81db", + "metadata": {}, + "source": [ + "Напишите здесь ваши впечатления о задании: было ли интересно, было ли слишком легко или наоборот сложно и тд. Также сюда можно написать свои идеи по улучшению заданий, а также предложить данные, на основе которых вы бы хотели построить следующие дз. " + ] + }, + { + "cell_type": "markdown", + "id": "6d1c75b8", + "metadata": {}, + "source": [ + "**Ваши мысли:**" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9b873437-f42d-43fc-8222-327eef80fc8a", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "raw", + "id": "13ad8747-c1a2-468f-a451-c76a131ad18a", + "metadata": {}, + "source": [ + "РЕАЛИЗАЦИЯ Pipeline" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "f76ed0e4-9817-4ab0-89a0-016df7d47628", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
agesexbmibps1s2s3s4s5s6
count442.000000442.000000442.000000442.000000442.000000442.000000442.000000442.000000442.000000442.000000
mean48.5181001.46832626.37579294.647014189.140271115.43914049.7884624.0702494.64141191.260181
std13.1090280.4995614.41812213.83128334.60805230.41308112.9342021.2904500.52239111.496335
min19.0000001.00000018.00000062.00000097.00000041.60000022.0000002.0000003.25810058.000000
25%38.2500001.00000023.20000084.000000164.25000096.05000040.2500003.0000004.27670083.250000
50%50.0000001.00000025.70000093.000000186.000000113.00000048.0000004.0000004.62005091.000000
75%59.0000002.00000029.275000105.000000209.750000134.50000057.7500005.0000004.99720098.000000
max79.0000002.00000042.200000133.000000301.000000242.40000099.0000009.0900006.107000124.000000
\n", + "
" + ], + "text/plain": [ + " age sex bmi bp s1 s2 \\\n", + "count 442.000000 442.000000 442.000000 442.000000 442.000000 442.000000 \n", + "mean 48.518100 1.468326 26.375792 94.647014 189.140271 115.439140 \n", + "std 13.109028 0.499561 4.418122 13.831283 34.608052 30.413081 \n", + "min 19.000000 1.000000 18.000000 62.000000 97.000000 41.600000 \n", + "25% 38.250000 1.000000 23.200000 84.000000 164.250000 96.050000 \n", + "50% 50.000000 1.000000 25.700000 93.000000 186.000000 113.000000 \n", + "75% 59.000000 2.000000 29.275000 105.000000 209.750000 134.500000 \n", + "max 79.000000 2.000000 42.200000 133.000000 301.000000 242.400000 \n", + "\n", + " s3 s4 s5 s6 \n", + "count 442.000000 442.000000 442.000000 442.000000 \n", + "mean 49.788462 4.070249 4.641411 91.260181 \n", + "std 12.934202 1.290450 0.522391 11.496335 \n", + "min 22.000000 2.000000 3.258100 58.000000 \n", + "25% 40.250000 3.000000 4.276700 83.250000 \n", + "50% 48.000000 4.000000 4.620050 91.000000 \n", + "75% 57.750000 5.000000 4.997200 98.000000 \n", + "max 99.000000 9.090000 6.107000 124.000000 " + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 442 entries, 0 to 441\n", + "Data columns (total 10 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 age 442 non-null float64\n", + " 1 sex 442 non-null float64\n", + " 2 bmi 442 non-null float64\n", + " 3 bp 442 non-null float64\n", + " 4 s1 442 non-null float64\n", + " 5 s2 442 non-null float64\n", + " 6 s3 442 non-null float64\n", + " 7 s4 442 non-null float64\n", + " 8 s5 442 non-null float64\n", + " 9 s6 442 non-null float64\n", + "dtypes: float64(10)\n", + "memory usage: 34.7 KB\n", + "None\n", + "age 48.518100\n", + "sex 1.468326\n", + "bmi 26.375792\n", + "bp 94.647014\n", + "s1 189.140271\n", + "s2 115.439140\n", + "s3 49.788462\n", + "s4 4.070249\n", + "s5 4.641411\n", + "s6 91.260181\n", + "dtype: float64\n", + "age 13.109028\n", + "sex 0.499561\n", + "bmi 4.418122\n", + "bp 13.831283\n", + "s1 34.608052\n", + "s2 30.413081\n", + "s3 12.934202\n", + "s4 1.290450\n", + "s5 0.522391\n", + "s6 11.496335\n", + "dtype: float64\n" + ] + }, + { + "ename": "NameError", + "evalue": "name 'train_test_split' is not defined", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", + "Cell \u001b[0;32mIn[2], line 19\u001b[0m\n\u001b[1;32m 14\u001b[0m category \u001b[38;5;241m=\u001b[39m [\u001b[38;5;124m'\u001b[39m\u001b[38;5;124msex\u001b[39m\u001b[38;5;124m'\u001b[39m]\n\u001b[1;32m 15\u001b[0m preprocessor \u001b[38;5;241m=\u001b[39m ColumnTransformer(transformers\u001b[38;5;241m=\u001b[39m[\n\u001b[1;32m 16\u001b[0m (\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mscaler\u001b[39m\u001b[38;5;124m\"\u001b[39m, StandardScaler(), numeric),\n\u001b[1;32m 17\u001b[0m (\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mohe\u001b[39m\u001b[38;5;124m\"\u001b[39m, OneHotEncoder(drop\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mfirst\u001b[39m\u001b[38;5;124m\"\u001b[39m), category)\n\u001b[1;32m 18\u001b[0m ])\n\u001b[0;32m---> 19\u001b[0m X_train, X_test,y_train, y_test \u001b[38;5;241m=\u001b[39m \u001b[43mtrain_test_split\u001b[49m(X,\n\u001b[1;32m 20\u001b[0m y,\n\u001b[1;32m 21\u001b[0m test_size \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0.15\u001b[39m,\n\u001b[1;32m 22\u001b[0m random_state\u001b[38;5;241m=\u001b[39mSEED)\n\u001b[1;32m 23\u001b[0m knn \u001b[38;5;241m=\u001b[39m KNeighborsRegressor(n_neighbors\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m1\u001b[39m)\n\u001b[1;32m 24\u001b[0m knn_pipeline \u001b[38;5;241m=\u001b[39m Pipeline(steps\u001b[38;5;241m=\u001b[39m[\n\u001b[1;32m 25\u001b[0m (\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mpreprocessor\u001b[39m\u001b[38;5;124m\"\u001b[39m, preprocessor),\n\u001b[1;32m 26\u001b[0m (\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mknn\u001b[39m\u001b[38;5;124m\"\u001b[39m, KNeighborsRegressor(n_neighbors\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m8\u001b[39m, n_jobs\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m8\u001b[39m))\n\u001b[1;32m 27\u001b[0m ])\n", + "\u001b[0;31mNameError\u001b[0m: name 'train_test_split' is not defined" + ] + } + ], + "source": [ + "from sklearn.preprocessing import (StandardScaler,\n", + " OneHotEncoder,\n", + " LabelEncoder)\n", + "from sklearn.compose import ColumnTransformer\n", + "from sklearn.pipeline import Pipeline\n", + "\n", + "from sklearn.datasets import load_diabetes\n", + "X, y = load_diabetes(as_frame = True, return_X_y = True, scaled = False)\n", + "display(X.describe())\n", + "print(X.info())\n", + "print(X.mean())\n", + "print(X.std())\n", + "numeric = ['age', 'bmi', 'bp', 's1', 's2', 's3', 's4', 's5', 's6']\n", + "category = ['sex']\n", + "preprocessor = ColumnTransformer(transformers=[\n", + " (\"scaler\", StandardScaler(), numeric),\n", + " (\"ohe\", OneHotEncoder(drop=\"first\"), category)\n", + "])\n", + "X_train, X_test,y_train, y_test = train_test_split(X,\n", + " y,\n", + " test_size = 0.15,\n", + " random_state=SEED)\n", + "knn = KNeighborsRegressor(n_neighbors=1)\n", + "knn_pipeline = Pipeline(steps=[\n", + " (\"preprocessor\", preprocessor),\n", + " (\"knn\", KNeighborsRegressor(n_neighbors=8, n_jobs=8))\n", + "])\n", + "knn_pipeline.fit(X_train, y_train)\n", + "knn_pipeline.predict(X_test)\n", + "y_pred = knn_pipeline.predict(X_test)\n", + "#print(mse(y_pred, y_test))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "2351cb18-78a3-416c-aebc-ab269d835602", + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "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": "3.11.7" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} From a52ef2807ff815e033836a70c772c8ce15e738f6 Mon Sep 17 00:00:00 2001 From: MariaLukina Date: Mon, 12 Feb 2024 16:50:30 +0700 Subject: [PATCH 4/6] Add photo --- socialization.jpg | Bin 0 -> 172350 bytes 1 file changed, 0 insertions(+), 0 deletions(-) create mode 100644 socialization.jpg diff --git a/socialization.jpg b/socialization.jpg new file mode 100644 index 0000000000000000000000000000000000000000..935043d51b300b2b5a44ae1b38129dcf476820a3 GIT binary patch literal 172350 zcmeFZcU)6Xw=Wz7M4E`w5u$(;BVeHm#PXwy2+~_rR0IS>dW{GI(jp)rARVL=k={E2 z0qMPl&}%{sq`lkUdD?x>dGCAfeb2q0^T!i1u=nh}XV03oX3bjPwFb&KWgc`+Q&mG1 zL`6jfY65;hlqt}C5cR23zrVnT2KdsRp{1pvp=F?>JAHSDg_Dbejq5xcJKOI@sHlNwXlUtaY3bRR8JXGsmp_yy5bK#!t29g0 zR05z=tW?yjRFpOl7zCoC1$z6t!+-fuodWuJn(hoe10!%l`8m)jDr)LeG}OO)4czSu zTnEvx(w@J3DTx=-l@oY-#qN2Q+;yjRl9t~Y=cy7lbsdwK>APA+bqE5ah8VppZ4 zWn|^#74H9`qN=8@q4`+fz|hFp#MJ8fi0_`5YS; z|0N+YBQq=eTTX6X{*ThK@`}o;>YA3;w)PHGXV=fcq2ZCyvGIvX^a5saX?bOJZ5_9} zw|{VWgg-v{O&1l2`rpX<8)g4Q7b`&5DHwtG?OCC&7LxAd^=&)yC&a0p4Eui$=@_9tckIl|umZ&CI)!v0RzB#4Qc3Sb^J zD+mfA-O6aNRX&ePjb}Lwior0>F?Ai{ve#t1=HlTmcZ8tU1pQPSy>&K?%A4x!X1ovY zUYH_E;%2iwzfEt{q=sB+Ade3{YmCtu{T4d2X{7bhAwgG&+RofT9u)2T>z0M=t&)A= z?0l2eikI8uWr&u(i(%r+miImEs=iv+sW)_OYPNz(r1xdqoo2eNWG5WzDrS~`g6zZL zb<}?4i|%B4oM90`u?t<|69zi=hY>V-D(2Q(iQskAWhi5KwGJ@}a;yUN$PHxOL zKnG`OvX&FyWcPe>=nk3wZY0?M`L!=(zYcSIkg(2p$D#A0e^ z+ys*y40Hp{XaYOPPiQXmI@hQL6YVG<#wmUE!h=;s{frLph-VuHbPovh!^nv)#ILEP z@(Omr-CrGok!bU4zOE`hzE){OEo`Izi)x$_1r(CikU|07Ev+r{pXHXwQ3!AF%eX3NzPd zcSX-q-BZ81&7oY9QAbOJ;7p;cCO=QAs{3QwVzjU1Hx0~4S01`_Mh=DjNP(qR9(mv7 z3YyuS-+jC>cY)6(S!Sr!QY6eGtkp{|i?h<^qO5yG<>>Ox&1*&P20E*SeAeo8-;^9! zh*(iTqr^u3)CivhOydVQr>i8xMvZo4Y5tsTpG=-tB?aV|$mnh(HDz@!@a|1Hh}-i> zp=Va(SA?u>_(|TLMff2!8R>}bnz}ni-JR14&ucS>bdb${%E(X4ditBm|I)FQ;E=F| zvMXNBUaK_yFcz}EdTq<``Oxz!S;h}FceGmjaJiOhY09yKrnC3q_RPIyC4Kfc=>_6A zrX#aXaq-WugdQiMDWFz3kB{W2)rdr${8?m<%}Gbb_q7u-UBPyxq8jo3J@Zqtch<*` zOK!QHuIYV@-U;7_`oqr>mC(oEDsgAu$7xHsjS0wgRhGXMp9!MA_re%+>8sDM`8Drk z_FYo(2VAsSUOD57VR1)k1gAJgd<%QsM!{$DkAd%EwM=olb?{oS*Nek%6i^EtF0{2p zAvrR&$o!9})wt0n+X(UAk)D!Si_$OOI2iFPC!W`sUr)BUsJ9Z>oIzo@QhsetbF2j@nlos86oo=PaST!`7U zSu=W-#sIU@(Nyk8iFFGFWgklIXZ|0m{&sEbdp3cK-60fIT+!^M|FhO+SidAKBUCK? zI`HeAt_mAtuC5BZ5+guyBK!ML(QlVBLPbIUdqvC@c<|3gK|sTQFKT;RYs7-S14;h| zjuB+>j)O2xpwAidkZcrsU~DzrRxIBjUtaljD>W0)n(-g z_bbC+OQ2HGx8$LZO{JPdS1L3%shXh=nc1u_VgwtDp@NU-!in0XJRfoFj%mD>mS^34 zXUA$uXsP2hjF}$$y`5A1ttiiy(=ahMCH+a(S`vyzVaTZWIu;GEZYm zYOtGv(3tA_YN@$-bS6FIor!+C`PHm#{fZdNzq8^qg@ARHdGLbd$kwsNdW~S`1XK? zQx1;nRW9~i`~U9M{lE5#0*V)*fZpo<)49>|0h{ypA5iuG$so*TM7nXPt+5c5CCPc5 zWb;%~m)buZ!;*$|f3l3k?G(_600qR&nr}Z+ytBj#o`-JvS7z= z;YU}%|8VT*sH=9HQE)$)4CVxQ4h*J*4tX*2e>g+-u8Xb!WA+J9#|sqDdK7%AivpVd z4FAWo*x#eG`LEIW*XaC9cK)1Y|C(9VK|E+yAd>x03Pyq~b38*JAUp#pWLu z=6@~4|BVclm5bNQ8d|;)cuu|&{l3K_tsd>rrr9mZ-lSuZ|iQg@*ga^|G~OT zhFZZ4iMHHDJz8IJU;k|T;y<_jTND6GR@w0&;z5pPbJjDlaz4%f5|{G7!Y%!8-iCxJ z1(dgV>h`uA*AcSbL5hW1yuO*ov4zxallIAH`nvcVcMJ)+GuG_(`hhVWojf-;IMzEb zwVVQKs}}T_(kq)1GJJ#Cys=w>tdS_A%JW30GOy2-i+tQ)dreeuVf1ii3=ADuCU+YK zb!a>yFCN!YUlOram7JGL1Z)>a-p zD-U+^$rXR|4ql~(9s4}v%y`jCsU9-4Ov_i`tpAr0h4d>k5E*dsTeJ%W$g_AsZG`?v(d*Plaoi48 z@6H%t!YkIesP{Aj_&g7hHh|W!BKN26i<(eC^UHk{P~;=JpTkBYQ0yjr@&fq-Bb0b? zf&#+*grdhOpfV8>0|y*0UrGUCK2z>OlXHled=;|Zj;pW~{3g}`cnGZp@2k%d}rho#Hpk(U2z*(GSGLg#`dU`EsS20;*hvcBC|x z?6h;NrB4Bp=mRDia5C~ss(3#kMKnNUwHc-ngfnJ6g^?lO!MIBm&_e)*rhwW9 z-jtyCjcSmak8w7Q%diR*DbOB8?$d#d2QTa=D?;r7@q)m~5@8`&FUV~C;YED&RO=`p znKuC(dtqIL^nrXsR_E*#nDFuU8(hp;s4^P-!QT-bs`DOir*Oktzgq);hjqMMLwCc7 zJZA9UH>Ip1(}OozaM?8;Z-)WALl5Y9SRUsg6@GMXPka)fV(@oO21eO6qahrI7uo3; zo%gW)+pRh9)_SrmyuTEfgaKNPWQi|`udk3N_?bhQdBGXPVsBwpylgoXpi_YGU&KxD;kRn#t!_?e;FRkdcKOQDC{A7i7YnRDAn<8vurjXs~5sd;-fcD7yh zeA18muktXZiG~aX(Y`o~u>dkPz7O?Q;8l%_JJSq%u@wSQz`B0*tEu>49^b&qu+gv~ z6srwK)Fl7tb3|*&3pfN5p7n!A*5Ut78)-!SO>YJ4pm{GT9fSdZTsb5br?x4Ai)qHi zjVMCx7?653$)#lfJo7=eCEZD}(k9X3$qgXtP-uN!>&eGU>&pFK$gW=9^K=pAASQ+9 zIw?uTnc$wGsuj)bD?1&iuFETt=3ma4{!-f`{)XpmGBxqt~m!V6NoCbvJ%Z_0?-6UBKL2>bdAY-c)RP; zvRdGPq+j^p&-7h zaZhr}n%p9@0!U=$23unS6r(kgw)-5yXfZ!Nm~{EF`}sc+u;`xLg*&jCxz> zg!O_{G_3{yoIA;{1zO~iXtXJb3pTbGO#xTD0kp=a*zXk3M6SK~I6t+D_hAWviTJc1 zV_l0S71}CP9>$jXB^7)ehFUjXv?=*{#PeLeM9jKrQ;k@9Bg}to1I;9;stpN{34!E))X2Vzxjg#azp&+CQX9=dKSO=_Nz~In$nq9 zNNw!SCI*vw!Q=c?`yr2OVIq55FOAW8SlfpL=cw`|rTVL*F$78CAEQN!DMRY-)jvJ2 zoJvFO6?;AtZC!|l-EzH`6+3LEm=Stl2A>FCpj|@6@W;P?LEf24qkwqoT?vB3N2TGV ztB}fKv^PCCy|AG+hyqFt+kvbf)Toe9k-C<^LdjH#BUZC2gNoR^*DL0T z7Sv?=BV73~E4usE{)l}ez9!*WZo8fIQE*8vxm)>dM#KBz#;4HmM5O}2s&$yLdi>r< zuksmG7%7W8YKC(9+$yiE#+^=yi!z7CvzmQb>}pnGTPax%S$L&AdA>?PTA&z{-ZG%~ zn?nPKCh%GdVIh6}#@{UR1*-aK{^_J;6t+G5IvZ?rv7h=?UPHIA+(o_r@Wy}FMEL-o z6O9Z}7F-{)sbS8ru)I%gUZa@D#TsNFa?NRE4B-`jh24``cwc>K6=@Q4Xo5%Z+ABy` zK@SL2*AY&j211g*(PWa*fgEky#p@E8@#rw2^n(TnLwfZxV4W?p0i4l{Qj6rlkV2S9+`jS zo*cNtr}HK2@ku*p*(Xv2Y{^gnmpj40>?PvM<$8 zdH;B97HohV1#Yhi-M7Dg=!F$D_mR zEy$Mb7YfcfILNtw;FoDalUM?ATm zHH$pf`EAH|5EvhQ~5Yt#yd`D7>>a%H?mSd_I5!vr2OQh+^~l_Kc(0S zF*k$4^H%TD?IPP?Q{6f|?pv`1cgK}tB>NM`l8zpVNac|av^r0Iap7TUK6|^B^A3<)9+eS&uw{;8R5?c8Sc(p|gdBa!q_1O&Dnfvmh=-gpTPR zc#;M4D+6=i8+V=%n>xK5O%if6NzF9_d712_k z*#y6!cmpRel8rq?+U3%4zsh{A9`H@`J3b2j=6Pj#H7we|JeZGEkb!r{g_&Sy z8|`Fbb7$A~{+RIG`o&_I`d}esx9oAY)se6qJ}=M!2sZuLL3Jui;dNx0FPbhT=9{bX z3O^m{1%Ct)5|3VlUA^MIvX)9J$^|Fbp`j$emMgrAj( zu_4N5BqU0mIQAY)I!>}QTl;BXY z!GW29e<{vSrC%Mx>x@G|+AAJz<0qGjv@)4S$n%W_uISerrg6X$I{EA84IpM z?lK%3Z*mvwxjXDmZ9LwMzeqk+x3|EF0y7eCR_-a@FHob|ycyT&uGu=q*HDhScRo*9 zKQcjO{J_1=0wp4?XP1QxC*B1{_59$aS@z>3+U#7quqqi@X0M)j)V5h0 z24;I8!hMtnZEC_Z%g)P^73?Ohv4jE{Pj>$WUvWV!IVY;KA_mC+k&e#nRI6wRUM!D4 zS>|iO92Asmw%LF2A9BPSOa_$wRaJ?p!MrpL2;+R-x{L1!yoWA{qbvbl&ug;YYq8aQ z4KtFX*-TC0feAJf5)Nj)D?bX#OpHZ%xR1>}h%c&DuO-ImkMA>wkS}haC1b*0d z;H)z&t+Hk$Ct!cW_vTNki`Oqj+}j78h3xrQ#ro(Et@c&4wBHMaMTs}Q^SP0R;=S_K zUdrl}A-Bke(cqls7j_SU7i&jOmhQP8BQIlioY(Xl(Grj&O|S^BgQD_p%pA$ z@PQ~7qjAUfh9xu38LgD^p2yRk!~67WZo?bZxB5wtvJxDoRUzuKxsjc4whcy!!%X9| z)^afyYYGI_)_>^j;Hde~49u(M4Q}q<+1j{0!CVschuwEj(KOihq#NDW0cOGHG(!qC zq}+C!+j~4w${7v!aIrP3ma;Y2*@{LSBvF;I`VBjMGc-#2C!Wr<9^2T)?A=lm1$`O% zs4F1wMs?4FMg!^fCs*Zj=&IWzGUkU7C3Oab`(`>dd^`i+n66(DqUAp>u0-pE7L{*Z zl**EtswlOVv$CXqiuR6w9XQR>b6O#?qaK1j`rs%v6(HAhJ!Y?+v~Ky+K&S_E<-&w3 zv$DKEEY}P3Cx}OO^Dw+8_xizP;jp$R212dh*CMw2Fs>ag%qT z_r1XF9}kvha;=5*bthJ+xrDqf0l8PXkc?AAIZW^$(_ z?@lJUOtFo{=vfY=%qaMTcBr)Z-=uxfE5BT1Nm{Vuw~>i?DEhG!+%Xu3jjV#MbvTkZ z?9e+p*}erB4Ox%Jrr`Z0S#p>yKF@E<*FguXGXj=O0;qbb!e-RZf95YaK~$Cm(jyAU zE6-qR=iokRJ^K!je)XR!;J1Je1*U)&m+G`Apl1<%C5c{_U@N z8*P^y|30Y9=&O$&ekhZuFp%P!!aKD!WjHHOb!q-pxnhR=}(jOI;gbeLSdW)<}x6Zrcy?Q;FQhD^D`qB_}1$;b`d%OmO_e*dbSX9-%j3^^ z@EsSm*Vv5wXbHhfVj#?SS`Rgzei7(Q%h#xy&qN&77;VB@Qu!x*^suL+7VJccAv@1{ z;#SBe>N$3o4r%<#caa^E6RFxjhS=23d7DSqykZM29iuY^RH%jEshaBmvXg9i4Pl0V zTsRKdCUE4Aa}#Q#JOGr`ZX*5hl*+eq#ft{)%(@4#IMLlbLt;s#z-KiJwa-6a^5|SG zK6dFquo*|kw6+^W^Nj2s=`(|SL&K=Bb3}GqrN+{}U<659h5|a#aEHbAQ$PjMCx8o2 z`x4=}xq}{}fF2uElA|q+mKcCbMPd$K)o`4a8{E^cd>^%mk(hZlMM4wXeY&iVf=3&o zNy*PCppPS!gx2HG^*n4!kz*jqAAd^>&eQ^(U|)4zZlp1D4+4xh7o%*W~UbnrR)~Ap&*Chbo?1wLM7*vLIUR0*C@%$J{wB@-f+U%m-RC9m4#k z@xt@ngJT52`grUzCv|SHTfGISnUP~)Hzvvmp zS10tds{%<$*XpQ1&mP4d^RlO)-Ap6MGlyPyvnQLX*OsqTeLI`>rs*IvrJm& zSM$~)R$qS%?g*ax!|;hbmDd=+z*BIW2w0Ol$D~A+*`xkS+8J3%$LP~{_bDLOhj{0# zafe8mk0#E0)`L}&p=1^qi^q>$vs*PCy7fw+9ZMIdH*Gr6YN~>gE>C;>KCRUR@?_Ut zXd}PAb;$+($dkR=Dl;qc3^EaL0x6(SU6015{t3F??bt9IBT_0FnYnNA$|orjGM+Bl zcAWTe(+l=te zD?fO)FXMv!rTC~E`+&q_$e|S;>5T0@v>F)g?j{&4Ee$6)G$}({svWug24|&hu0Ye;^t#*OV_ZlTy7E!}T*Ub zvBe7?>>c(!7ph`x7T>EY`WhR>+nP#LF7yRCkB?80GJcI7GERNX5xB0!L02Isjf(rJ zTf~71w`u)M0Haf?bIkD~4~KUgPK993Y=tkyPnJo8BTX&BXCKBty|Rb-$Yl>B*S~F~ zD`Oub^G`O55mf3j`$Dw|BeCz~PnTUIgYWtl$cJx6MCx!OAG%(=+9i$h^k|h-{Ou9F%$R!p0*8jpagylhc9M4*) zSzvOR1pc-P{G6R`sn`vO(NF_;aBDwFI^u!@t>;lIk^z-DkyUh5Vo^BR(z<(CY&saL z%{p~g)YfIqg@roXpzhO?QNC|!&sP);MysdmzK)&O$(!SR`(e?);5^#>Mo}J4@Z8l8 zZM^2T#5oBtHKlQIXPDXN+7`du z;FA1`yW%Z|NSx5kt7koV8{>6 z+8Pf$S|1cPqsdWL6W@23 zMap|oGO~h220PSME!>uH5L7IMyRVVi*Wef6b_hDWcgtkb_>6S#Ty!_b2u6rCk+_f5 zNx(rnhI++b&RmWv-5&P2y40e$Te1V7DQ1Xs6Ww`sH@OENvZ*^;TsiV5D!CYF=fBsH z%RMhUu)kkXdHY$!8*S&Ug%EVqXYK07TJ!VyQ_4XpJU0xj-80C)yjVgd;$B zlO;q0?3e|f9ciXa%tr-~uaDz3cg1#Oe@H&3Nj%|I!N;P(2!7t0 zL8iy~+T4bOKmMSVBQ%}tS$tlcU5O3N?i^A@H9LN!Ln-ig$36BwvjXm@W1h=JYHY!h zQh0EOMOYXro}-3&176c5n?K=7jBUE8+cgH}NC z4`{&3R2w-P=d)6Gy%U2;Zvg*gU+)p!s4fzY+$;DbfqB$X`wqNPWD%sO_ag48McwFE zsEffFlx8E#-AC{08Ak9bdckSEmcbVvvxhm2bSh)>`$sll_4r3p7J)8Bysm4Bc?=_s z3cSXaCRxVM?#4x{Qt5le+;ywYWV%ny?#Bk{B!MM8Mqk}!gw|s8;)L!|B-!q}UAw3VJ^SL15pFKS@iy|l8 zW-2{iy}coy%r%GO&#WC@p}SWNna+O<2)mqgNd9yFhVu3)iy+=43fQV6cFMe z+(+$iyj)QsF1*gnBb)JW9XrNazs(`d_X`EtuXra97+N>3PG;8+>vmlz44vV2BA$$; zdz?R!bA3fG$Gg!Sz5C5zge@K;|6RA_JiT+;J&=At~U zf5%lpe(MYgJzcd8&iBu?5VZM?GCQoy`zUC(CnxbjV>5JInFTlcEiBAJr_G%i2oak1 zV~!K&DWI3&^Z4!#{&OQ`Q!M9`Um8?Ii`qM+jdu$QvzBhYvD-uhq@GQ~n_*Y}z)p)V zOMeDwU0U6h?b*VYIR32nx%Pr%V(-g(lvBz0Ra_`cBFCUtEV7=q1bvf=Pf)F0XK5j# ztY_Qc#;`rR*^6)XeQb%;Xe1uu;tjV3p9P;Ji8t;??luZ@ZX0bl64?jb$h~A<=z)wG ze2s6(Pi-itdSj{YDDZHr95vbyMoJN$Dll95ZNeURupD$BAbs$(u9Zd=;4Q>5BtNIg ztK9V?K@cm}>hba-+{Gx@D9W$T*)xSZ1pXbG?X~4uGL5|o*tNFUzR8<>lT0h5XuIjD zau3uNJxc=mxQ`%m%#&R`@y@Lf(d8I+{xyy!lT>O7JL^D~a0a@P_zT|-jzvCzPfD}A zb8NfSq0!W9*X_K1BW3@Bax301^l(+%>sKnnmbuY854!eG_ZjkH*f$bk5!VLvk29rh zqob9TwPvj|TCEAm6IxTz9e&p8f z1wps)h9fCO%!NM(Wrq8t=SRXTZs+|{xSdjo4nuwyNS@N6fGqQT^oE~R4fR*^R@01F zfZHau7tcgZ=YGag9Td)8H0|}wMhQ{(K6h+QM@ocS^K_N^m}e@meBR#~FKsNqGtXFIsl+Tg^vf!F!aoQE&b>=Z5QOq*Gp|hvvqU}Zbb{p zidi>~MGGys+5)dP3r3qo;f?0zN zb&IlkrV^-^ZXX^XgR-G7!t+GRCyw+NRssW!tmgYpqYAt7$Fn$v*M9Z8xjpkTT2pI* z2aDzTD1C5!Wwv$UvR==v@FU_zU&BQoMyy+K5tj=#W~PI>CV%B(q}>`x@}bP+^GC?e z&>P`ap$k1FN3nUCrS!oz-YbhcC>_T2241{>?wb24Gh=k}yBM2Ru~oYiEN7dzx8^Xq zEJ;0d`OMsswPC~wE{Km@e!`0zLO%>xh`7fw@-31+>sO!j8#5ddicQ^&NjIIl(pfc~ z>OnP>TO4Y9;(EO=QQ?agi`=m3*Ubj-sPs-ygJM0oyQ*}r%vteKrAUu<@f{`nW6KtSI)w=Q{oiyWJx!yX!;I#T${yrr_#hJOQ#RIeHIqsM#o=|sGQT)2IS`?6598W#oB zP|!HAe>)(prkDB-!MTwO8sS;l3^-|;l`j3#ZL)*}7|% zR0CjuXmfFXnSK^pwtA6Zvi;`7+F#s3B60g|K9#)h^jHrJs|(?i7;2h5FT*KdO&<7k zb489cu-vc~-O)0q68CdtK|rO&(^_7n-p2%R^o6lsuq&-_TIFe}n$P}-s*o|m(|S3k zQb)rETUQ@GAf9A|LN*XI|4A!3niO!2JcV;NxldGJ>ZO2UIDot{n9LZG++%qyrg`@E zFw)9Cd%s!$djFal;6xTHKC7pf6wF~JzB)-wGg7ZOqSfetgyUY8RTPNL9I3pu1mg=` zzdhNB)^>)m);+G68Q`VC!T5OGtcMEtT`_&{Y+6_W8ST@Jg5pY;`s7v_Dx* zG(R~kl$CfaFI~2PsWg?aq)xQ6kbJsZ|C67tpJ;@W z*gd$$!zg}whUwYLOQwh+1~~L*6_VSB1OIWJkRW-Dc^{Yai+>h&-D=us4uY2 zj(_=S!ekQI2Yi0~8K?uZ{m?E1=N=|i#L%G?)*W@xcs0RYd7RI#bmm);Ju_NuQqmJh zA7H_J9$de#HkLt9THgA2o%gQ87r`+iWW31LS(dias@@a6#W&9<+kOtrwtBA5F7MJF zzij2J4^4}U*7zE~vK@c2l(INB|0CkWHDUL~fnW*rfzerkK>NN2F6(kyQ7N^2zc4yc zf$tVddV0#G+?i+V2(R2M30x@(!OEPYIwW~OX=Y3|2=~qKj~woVPq_ykG@js60VJ^8 zM`djvncqdGKgnEnhKDML1DrphZu)DcBV-Pq58g4jTI);Bh1cun{%5yrQg6fXf)o(( zYvo7yrY?~GHQ0TXdIK!N{14t+Yg1l`w zlFSn2;ICEb}g{B622A3y40NkJ>Po2wo7q zR^FKwWK*oz!C0N7r0t#a(T{D7U*Oq5&19!nXB*FgJ&_mPcxuD2pKmgGLyjs8#MuDX z8LjgD76HEB1<4$(b4MouQC*gD-06=`PGSiUO3c~sbx2=I9zXFv9%_846J{wsu;r>G zgVBl59%KvBxO6$b!fvABl#kZHQl0WelcMDymn);HG!k?^W6}81h5ZyNd|vJvXHeUh zZ*xo)Rg(iTC`{-^j>ykQg|gvq!@qPe@$aQs&Q49jX0~ol->`PoQ}{Y2TWB`DuFMjZ zvAo;~NjoqHFKhBv4d}ZI4?zaITi!2H#jch(upnx=ovG*(Kyh`U}oCZ5{U zJj=2U$&aZmSvEpk^Q;9Ib3}Vq;Ht=%n6-^lm>S4~*p8&PXMH-^R1^BUaE45l8TL%a$#>Q1v?>c0 z=fSKV*7}1heX*YQLf6!EV+92YTaQ$3S=Nzzu-iHA6p*6$^O#*d8x2#5j;PqgSI0Sr zH}CRw61sgnEz7VZo#%2` zvm3{}yCS!LRNyY_P{n%~U}E&C=SPD0;uM6*p~J4TRykR8>hY@N_{mQxeAJ6;1D;WJ z4k@q_J!lbhA$YSCZLhT;{VO}Ymu*l@`KXe6I%56C9MLWsf5U0Pk!9A@u0o5a+jf52 zt?9LM-;(Je`TDTYf?K+uTJ%`F_#0f%&-cACp_4a1^M|+SW0d-UXxvNqX%oD4>$!!F z+uj{q(ar%v0nB&@smssi zo_)zG)L>1^LM`$7o<>DhMuN;`NKL}hLg5g%&2)mAHcJq&Ynd8c>Ifu&ZQ28Qn1i<` zi9xm`AHe^t2O@sucsp{(lp{&79@$A(5ZvZ9;TB6T=s3%8NCCaugSV|uk${c4ke1Bb zmi4Lg59V-+0l$%YR}JqRq%I+6b8HhMyu;iA9jAa84x=XPL=J6|;K4;Gf;`z^dt^z? z&4bJ`tuwiqtXgkaw}-jXn1HVCBhVf5l)k=JZTrnn(Z_@ItNM-3OXz4_ zeqRT*_Ue?AWg|hujH^5>16l_!@+m5|P$;HjAENP9y*!)T_w+j+eI@LP)==mwhLo9e z?um8)OjN^Dc)JXLKyOeprm1YQ?F-5G@u{oT%a~uNs(-vd+mY2{($y!uRB_ZEvXa=y zH*ipBDjlYvIudriy;^E+OVF8+aQ4}g`+f#}qyoOxy^o9hwgTk}0*6X=7=wda562b- z--MR-oDLNEm;%}u3z;^pT8yzkeUvsFxh6Gg6~kQ9eoSQ^(T>87x(Y4b8vSe~RBVi$ zNiw%(6?vp1oxb2JvoMP}=fti(k1fdk!z_ICC{)N8v&!|dEzDDPAaZLUBh}ZN62L z?!Nsi|4xx+n01baHD;;>+qz7vLUm^!ndx_tc#mhM?s->Lirt>|{W!x01kFJc5wbDo zUwLz?ne~|yz+OU<4Xy2NO#GBR)ZK|W3O%&!SI(1@9;{iri_74hZ zZnRJbOng%RBI=qj2@@__>^R^18Hg=W?dN`qN?2JV9~a+hL(siD(8w7}F)V+{RCJzm zTuGgMS1&8s5uGr(r zvnhL9cQw|$jae}~+7lPhBq?-D?aESQ8uH?R7@Y$bYwAN4S1=>_xuxy&Ws>jRjRw2#VbEAfu=@UlyHNd={oq$p8h>^opW4Ywnu8alA2k=nE1DPoJD=N zS;dRQ>-j?4a{Hi#n9pXf4p#@hW_(Tf<>B0P>WgyOUqlssBKl)>$Qb-R>mC7TF=OBS@vRs zxHos=`?SIfuFIj=>DnCqufM9W&zTH|tYW`9ceA1B_d}d5xG~84;FR z8W^KCtSpY(ukphN)|x>vmN)V|S}=w1tSlNSv4^}hALniHojg|Q zt?}HhgMd;Zkjqz0EfC=n?^~o=!%4m6@0s zGZHgt=Ux*Tvt%)SE92e0%5pEm)I7+0bK_sJ!YCp9k1!OEK(3q7-Re_`cui29;q&lEki-=MO&KePod!PUvO( z&oT>C3402I&m6b*)bkvegl5JWMO_`yiJbA*DqsFk@`pk5JXJn zkV(&+9q*FZDhX!9`;+W84hkV1Loe|fdwN>m*TrrRj>%77Mz-(Z#lXZAKn(s8A2HgZ z|Fcqrn6jMy7jNV<)d8L`w=z78ODq7x_ohd3YliMvgq2$c<*M3l3dc}u^1nykH|aMF zyf}Er)@&c}Hb#FG^$8$(b6;=G+AJEU-HneL7uDT}dDxc6`^E*SOv)_pa$yPJ7fj(0 z6B-czE=W#PNGH)3LGh+!=F)^qN$m=n>T@oib*YCr_;hry8G1UABdNg5mUC}f zwI^0ZDL( zD+31*>oSYIsnitc@oMteF}S6UdLbtyvK&B*W`7(zDrY)L3=LC5p7_O1_&{)Uhb*Du zmF>$@rRXv4`J-aH!_QcMEWFJw;ga4pSM_M}OiB1QSazM(K)!NYDJ=QQuY8nqVDyY+ zCPbts@U4;^!=A=)*|EOO7*)|nLL~jvmZ^erTJ51mbGwUSg^WA%UgsnJ4VAB!D05o~ ze;!tKs%^}aX@lCvxg#xAK!e*PgpZ1EBMq;GPHGG>MJs%WZg*vF4$1NfU!n3PH;l-( z_%h6CmtAQ0XuF*+WIr;#ue)xy&N89XS}_Q-OZ+@~`lp>qUC6W4rBBs;U)ueuR%UC~ z9>3<(I%kd7N`&WxZ`5JHH#W?=AEAys)`n7Tqx=rM2SPPF z@oO4g72S+l>Wyx3QJ2++%N|`W(0pa9l4hyanu;~rIhJW;hL2AHQJpOJa{RIE9C93n zhpZ5jU_p(W4uA1bJgO)lW7{>NT8=-yV|&r4$md+u6!LEB4iGy*H6L#bDW9|OG+pE( zke z+Qql;02%%AhwROp{$IylyD8`EBql74>3&(1&t-PC6zHEEE~N@xG7}W6ceJ)zwnfGL zfo&LMwrLh#I`yM)J{o-}z^U{GVmmu~o5QHq=UwZKA^oLG;l;&5gTBf4*{3I({vX=j zJF2O#YZpaPQE4I_gsAi`U1~(63rGze1q2}i(mPR+-a&d3X;LG-6FSn75~P>VYeEf> zSFdP)%Fg@v^Nh-|zY* z5hY+2vtSCThg^O93{5a>9ma2zMY{9}Mjusg7`M;1r5&1+gcFUoFx^*ZQcHt?&2qN& z;rxrJ9$i&sYbA4pdN*dH@occBHoBTdt`%df7n$`!>oL4Ke`3dm^Bb~WM}z4{M^6AP zx@fg9q0)5LVq;_rT_(RRetcb62Yaf4$K zTafN>@?!o!PXWv~F{b3WYf-b}JJV4+OYZ2R!tM#&bh7QSf_Rg7%c`o4`G8-n!+!f% z)Z5}DCAXg#P{h!*x5}uv)2Km$Bi(W4#-cW!XgXSDmCeN?;exTEbFN!Ktn>?sRCpf% z!G<$(yUsB9$$;cd+&*jqT(e*ZZ&ne`UlPDXNjF+b7y&rYkN*l|Uq;28=*Hy9*LW!W z&e0=`1U~&d*nVarShCH6$YC?l+|h6?Cro3RcH16%J53uQ4ANfw`el$(dBduOq@$kn zAEMjr_cGnsv$SL`r#;Yft2D_?!*wYY^BIwhL=lTSipMWhl$>>@a~v)P;*+F-j9d|= zj%$xd4B(B922ZmFM#{B#_TWaVExI`(pR%EM&6N!aPA^S?lZ|6rOrW4~zg@{v z;(gRm$jHNRY^ECkj9Xa>z4ZFBrs(wParl#`YuvriTr^jR;K`pVSpbf}FI3h;-p`)uBxhSk}8Eq(09*X8oxR?cx3_iFCj7 zj(?CITrX>HbYUqd=iUl7df6PTcT9U|a08+)qkfBW6IqVkM%2}mMvJg0Zz{u*=3a$A zFMEAY%LsL{(Z-5p!N82Lu2liCg9*r$_@WT))mh;_g!pLye4lZpRj;dR;kF=6nVr8W z$pg2^hzkdw*qX>9X_c0_?F$M-^y za93f4g|8Dv;m<;w&9q&s=HF*~^N){8;w5C>M7tMpT{6U~8PS-4oI>Z%cW^$)ZbtzK zFCc)E>b=>u$csJp~n+E9?kgMhsRIt+d5?!0)pYWA$JJ*}ThXL7v&DocQ0NkW!ie z2pZP8dNy*4O(qlzsZd*?nf;7xeQ*0kOm(S7U$rp zi5)|qd{3G=lH8{XY?6x}PEh+<_7qm?hUZ(A!8@M7sZAW{ev!`o=#gd)EfAU6k5cEgf`V^7coEj7 zD##XPJro#$m`1~c_dQC7P5bC)hv%r}=F(Cx>0Up>1^fO()H~->A$IYiiji3L8g6&k zNenyQaq#+q*CmQ}DL5in;woENIfcRwF0-uYMMB_(t$fmE^ zBu#U9m3z~Fv74|lI;eP>sT>NRF?S(vO1xK%9P3TxC$*BCveQ{%H5zWPRh8fxO+n&=?FF}f*L^k19njp@V{cxPrY787{}0%rC%JPQo`A$!|J{e zcECx2Em7mdz48>^jWvky`k0-CJ*~k{EEycV^mdQ_;YD2^B zzrq~M>u37?L!{o1uY=j&AuwVEDJlNEEHASlQ&s4Hv6-{3=a-?;cwCR85k(c$hreVc zpRiG`txa?vLM#$tE~hs`8kX9ucSjD@4{#o_`Q=>_-A z(zYz`%Cn1C!>o7iALtATVPp8_Py8YsmFmR0Ehac z#~Y8Z#FQBLSQH_pVWEzODF1{g;`q~wkkwu73HQkGsCHRvHn93d3V}`{zY;w}Qfsvv z_+(04!!v6BSC}tSe|chZKbup)_#PJCOJ0m^GVU;(#WGw@SDIkM9G1w}%$J@Jq2?ij z31$q3Ri=T=i53&fb8BsAx%E1i4^k$(;VAtS(Z($kib{dgvPWjc26v&eC#^SYW6ARO zh+FvL%74ce_#ZD{7UCXusJ>3_HkUiF1CyBXX;o_oXU^QOQwn9uXR6%K^>k1 zwgzvHR_d6H~{;=@o=mhch^lTMdjNDyEI+P~H0m`-i{Aqb!bf4KrA5;)-cdT$S zRxsb92talHY-{yc5eo0mTMGHKVHl0n>LbSSog_=lmniwu!%#UCX8teGg~@g3NW2IV ziS$cVW>w3)-qG5C=Q}7DT$K{&wtrmx;`&dLBHCo$9D#hTyG00YdV6~w#?RB)e)PAN zAppo6M-xlX29_|P!5((W!v}~YKXDsJ02%4!@6+)XgxSl)&WYgWjOM69qxcUA<nzQTcyPeCgKNm^Jb_ye_oy?zbJAme#m-Sn%54hjw0}Ek z5PGn`@dwNbEAX;#qif%`;IxYXG)VPF-S|5X4c}~dEQ}Y_+blHCX1-r*>TTDqm>p`H zd_2xM zfX`ndM^W+U{iE&lJyySPGN>EttpE|qTK6W(zStYP(-S6~^*77cxS$=bbGPNYF_!oJ z4Y5&|$(7P&sn6D2E+BW_xVvx78N@w_)-okwZY_2AVEpWl?t`jINrt#$~qT zOZpS9O6#%dEC>lVl*FxsA#85RP(Eafd*M}C5U4;XRR)q{?PO)blxr^y8Ca@6%Pp~A zqO9?%YHO&gO<2!silOJNBD>xkcKParSN^!N2N27R#aaJ7`u#gO_4nv!Y^gyQGpWe) zKwtJQqI8-Y>&(?9?+wdH6R)dZBf7krmO zfaE@ay{94Q1M{YC@FB$E&tWm289ogk%UW_{;bP z5^B=cG#4U`vIWgAX-BURW60lS+Vtvdznw^~z~d!61BM*(hG`d?SLP-M$O^w>g7~pe zh`+>T@C1)$mnU@Ba`yYAIh6X-=rBQiGv_XjiOX@Xk#@H;IDWrOeXVjeP!1dT+c=rZ zIs$E9qyPELvPMdE@`MFndcuWb{Zxdt>!@$a!WiX7$VL%VT$^ zk}=$1KVl+Y4-&sIeEhRz@_XHvKgSD^uCbkafciUxQvP>GXHLaxw7QKG|t)3PjVwgr~@tz@)X!SMD2y zk>(!;au%Of_(u0%Yqle_#{S>}N1K&MfOoZv0W>UctC@-n5XJrUF9(M8t1f(zar+rd zMX0}Hi;!c(vXg(HN4!Uwv}%tTs%>1MJeGWz@{FZLQX@6Bj^T2n8mF|3th1~!y@$?* z4mt?5S&v5*FcijWU@b96hg}~c<#fO`dwq8@!|(a|P8Vul>0vES^yL!`K`mWyB zE5z!G=)=HtLZxA!_E(D2pW!EN~X@MSm$? zv`Do>;pK~_YYGruRrHsQUa5~VVfdZ*=Ivc&%bVJZveS7gR4T-`e(KZ|OpfI5tiJVa zW(Cqk`-JOQkp9#mn}%UST?}%hTh(jc?B4TcKO!T7!2phB+*C8_&d0?e%b;wvezrOG zY0^~zl|D=Q)7|!-4PG9XxcO7?D%7tNd{>^$JN=C#EI=aGp}(fZrSSLm z^kNI#b}h4HU_Ya&&QN7MU80e;kxKPB>&MC-FKKYt8~(b=#e6qOzR(9(Oc%~d_^&OB zyo2~g6ja8WzNE*-j%%UnS-BkZ@=~b#;97mG9_ALJ1V47^?=st)dGB^7HP*Ie;E$?P znoB81m3#xR|J!QI&Byhn;d$K6rdpWLS1-NVJ`3NIiYQjrV8$Z);$Yt-4!TQLQows2XI1^T zbi4sG9D7kZ^_O=0?)o`aM+N&IEK%Q-y^YHHZj$@>r78*eHXD!i+AYL&{rc$C*p)h4 zz_5)sU?NU9Q>;tx?wbhY70=k7V1Gy1))D_> zUy!LDv$HG7KsMi$#gd-3%sT#n)R)2)#unR~m(z7*VcJr~2-0bgu~`)HCQxIM$l~R* zDOm;UqWLo2$TAH_7iNhWHpYR?N!o3TQHmgY_DT(wCi$t`2yTX_pjd~)ZrfLb37Y2F zR-J=1Cp;{ubka{5%0>f;o+v zF?s=V0H-$4o)&cmc$|{6E;8!hEBzdpTJfNWiho$E05ZATt+woAtzCJ92$v^baIkn( zUpKv%CUy?-JX*hzynjh;L5c!o*3VgQwzNw49oFy=Fis|vFRx(Hf#Lp!Mj1wxs^}ia zwfL{T34EJ}_e?kjhHRO?X43Y0%Q9UIdAeBMOjD1tY$$j4oX90pX3_V{;l;^dWmdJr zB&QA2nH4r?_iZEW_!=pd836Fibg5b30D|hI^}6xKm`$sJuKLU`4(k((!U{rvg859~ zc)jPDt_n**RD$V13=H`kC8N^ty7n@taHbpS@{4UQ>4BHi)TKFSNkOCWC^golCbny) zW-g23;9%rv)=wjtjEND&-M^=>29f7G6Fp1C%B)z10cr`_9;Mr_w13!}7z|Gp?39#M zin)k>G;)iP4&cL#!nqz^+Q}~8hnbwbVb%K-Orv9R|KF@)$p0bL7HW$$cHgYhO51XgjXucRm*NaG0wO!Vg)q?4CS=IS)#X$WpuY z$vOb%y58NRzBAiV9LMYd1$b2pAdBiaTk3khZ<-sSy=AnVxOJ_JPybA3L-o{k=)S|wrz6xX0$V_6)yIq-;Sm1>6GnrEjE%X7-1r^V`*Ujq`~T>02vs)YFGE?<3G>&xoIy`XhEsb)!E%lD8eh z1#KktnjS#DKy%2&u>O2rKe|@$0I{?aQ%7GY$;4;HHBEk<_xCT#db(Z~-6ZPuLET)O zjb>$8g%-xFqbcYC+2s3 zCgT{-N^4ktQ`b|k-Ma69M1hL8O@`+OH;W?lppo#R0-fuFeE=?fgWn)js$!{wo}8k< z;i_)`1N4Q8;+8>(CP-=W9;__5I}g8f1f6Q34`3}=d1|9eWHUN;(zk1&QVyl+@Pf8_ zDexO5V#+CR?oQal-c2NpCX}()kLI#~O{#gwT#-#DP3{Sf!39p*t?#{^yPBM0EBZWK zN;(9h0R~ipTm#!Xb~5}2Fw>g&ZUMjDSI4ZkGNVVOp4t>}r?cwe^*(orGD2ttSiPaY z@C~2QiE~h!^cxJR9;UaU=t@+xAyL=mQG3G6Rc>IB%;5NTa%T8Yz2zuN7HCspuJ4$U zkM0irjw$!Ido%%oZ!C$i6_%l_;&aYLX1HW*C%l&iE^r#1xWCYT#66pQnsnf`h$#q>n!ffbqXfV@dUI&3Y zw%dZadg@`Ujc9C(Up(^bwzA+htH4X;$D~PVuXZhI+-1I~)hvL!B`RRvn32aGtnJh5 z+WR9fz_sOj2%F?lK5I5p&4*SL-RySb2osM(T8_+@H&#p#YzQX}RB_-G{c*VL@Yo;R zVotNkN;yr^@9p)Zb;El8X4W%AjYo^OUar*B(LvW+e(oJ7fB<81{Wks+4JeJQV-2Ea=Q{QOn zcJAti*xK+@CY>pfHflXH^m=%_6yqx8XuV@tl&5l~X$l9SV+a2s`lJXHcPdUf%dXc0 zB=B$uEDabNal8Ot$qGJeEw+!Wco~`3y{VVJy2$WcwJP5zbSO)4JvG5NZ$*2oYx$7! zj}qqop=NV-pKf(^1UCGD8P6E?N_A(?T8f*2=R(`5Fx1s+@>CV(;OE^a&yvWndzRZr zkg)(^y3W2dVpD#o+%VLBTVxy^-~M(X`CtJ&0xfDlvsBa;I?N8GkHqG8PG~@F1=#g# zeb|DH!>ETJL=^d=NVyibd}AcFvD$@;#tT~*2Ul}!GU`f1^0@4-=`3>^%?6KGa^QnC z8nev`K9j(Md^Jn!k=Gcy!GP#HvYQnS-;iM$VFYE_SY&e+cqQgjiT$QgJw^_V#nZe*TT zq;y)6l*y_$ef5b^_^SjzwO**H&v-ktq0qF{e}x`FAp;=w?&nEFxjQ~ zFgTVh4qb!mmNU7U^lpmFn)`ygl448&CeUr1oQHq3%#t=8Y^tW68vDuOcfrgSfwmGF zK&lZlS$}gn8E13^rF@yNx={6__@Lybc`G8xntwl}mq$04m+#zVlFwRLe5z!44lR5t zplr~FU(2@M6hEQtfKi!ttqdkqUAQIjs;pQ(^#MCH%tN@OhTYphh#V~|t3;~^Wab6? zOc(c+>Ysi87qRjc<1(Kc`!5YyTUWb^MT|_d>b5%^dNK52c1K}?YQBBrw)jQ0wPPI~ zdpFe%vH0n}67{(JH#2dbtW&4;dngyj6$Itnf+iDUj`J0yjfOfMJD~~0ftyq@S!+TY zE|wUxU*Y@{ikM8sd!a6iPmR}bPyM;1)(0mUlE!usxWXjg<&DU(dBdRC9 zQ<90X1YwU+wP%s5ITWVv5~^QPaE6=8q{#7Y-}dNj8Q(v6!e*a$pN6GM3J>k;d4mty zFl9+~s3?O-Z0d3TC>`4bp)BrP*_n|0ttb|~D!p~XG^IbVXcpf#9UWIpCKs@~MM)2x%}znXm3ZuF6YPO>ZQq8q@cf057i*O( zinvg_9H1F32+nmc(Y6Q`?pTjjw@p0^*^+uFQu9$_&u!^wSPMigJ)pbimsUsjlu`Mv zm#f;o<7frKPH5H^6k0d-;*usM<2r88qZ*9eh8Il`?q0L5vFq@<0f>h_gJGJmDcb1h z?AUULbdIm@1l(JB`Myk`+^uixMP$i>yIR}}l-k;sX~TaDb#&+t*9wg&Y)m`$^pYPVx$ zpaZsf;hcH^BRzA|f49h_hbyYQJW67msEnXFw-=oH%!c*Nug+v**_c)bzZ~K}-sS11 z9-2ampFC#~GhE7bR}g;H^;y)NddUI%P_JxV3fanwf1&OOH>#UsrJ3hae@jd z$~c3rnkkQIPlVcM;N|X`7K@8>8;Cf{ntQi&Xbuhfo{5O`#VqXpD%YD)Nn@KBK=TQA z^WJyjSu(ZEUD;Gjf1037Rtdhnu1X%OZg-{!3#_kgwL%%S9ly0Ph&}(%Lm~0vBRVoh zxEIEBDnzp7%Ypc`|K3C37V=ns%H7f0m_E^`&HY}N)9R8F$oX;*sC0UAAQrt&%Qbmi z();Ac3WkvTE`X(zm#P@u^Vh^gbKBmN-2{_*7o; zLHH974r zA(#(pdv=kF#U`<~`x=5maKxOAWl8}RW^z*{nW;&hhllmsmd~(~XVw!fFQ!vSdPD?@ zCG4~ptjQcdq)6}yD$T-vGF5A}RiDRk zj%gV;t+yZZY}A~6D6DxQHDa3O6#B#7Q(7)HJORnZKuY0OLotQis;@iO?W;_RyM$zI zA;>+<6&0(qrKsP>ekKaKjuP;RmdkGjDQM>Q``qORyEIMf ztz=PjDb(B3(HiF*io1YF{q6AuOmicmoJ0$U=KI6VHK`pkG1p^GH>P6wre;`9)X4jbffR%%_)1!opl`@%MQ-E&viI@#w?Rnb}Ma&zv*O^ z{&|pHJG6s%U!3o!kEU7wm@MMzFkV}Z7@a(FDag6lVxA|{`MZYU=2*zv+`i>eI)~Y+ zcUC*ISK4)-13dW}A73{*H2zJuv_a8J)8U*Ll&xCOfMpY}NLevOPQFcRQU1-W;m4=% zT6ytYZLnIkB$!x`HcjouamT3kj(6+K8hAz%e+jP2xH{cW@9T)Acn5{o5CkR#A5GM# zddwN~*Q9dNzv_1*|7rNPsZ-)2<=}vUe8gK)TzdvNtuo1fauv&9QjuS+c5_ddqh*+w zqpbdBUW`mg{8t|iX5XHWc;Da1u1bLS__vuud9npcO7F%EZ>IgXb??;qZuXi}xW%5q zOt#AdZ!1;Ibo|&dS(}!YMcM}n0*SQu7==v@vQgeUHblw~ZCBb}OGC6*5TEL$L1E&3 zHAn&HviHv7(N71(;uZMtVR`(eREd7hH??ThES|2da?y&d4is^wC&*ZaFR=)SI}^{m zGn5vcF;VVv5e-?2+Klp@j__{oqbzw=N@MXtJ;T{RpcL1Z_7V&LZW8 zQeMq1{%PuFB1x7=5KUcoRk#~?S?V@keTd`8H&ZNitTuS(&qz@~wGgG{{7CWPW&%wE z4jMbtAy(#bWnXUkRGWBrxM)YybKrpXNjM_7f7jwu9`0Gl$V#@kY%@sh83}FmcYn&D z9;VaJM&>Ht?L(GcXcVtM?SMnp$hg16d}(k(EV;~dr`0W zhEs|ZP{aevcBNqaYh&PCJ9|VGMV|UQ|M+%$`&~-2Q&~uong5;S%i}$i*_RAQVR4z& zpLA58_$!C+6Se=cKB6GnarJ{}Ts`S6>P`>pVoy+xLk~Ssoxjaa@g*}sZnyf{r-DzB zU)Y@Tt@uW%Fz@}6%PTCgMY>JRjpK#m*1|XIV^%_rLa`Ynv4woSvOcShTwUMw8(WMw zNLcl`xY&?Qu$aZCZJ{O&_RPFLKLFIhhl+i?8(Sx;bxeKkMQ<-KFqYg{S1 zy8ZnAw(n{5FUP8aUskYXgDW94R}XRbB2{htk{d5K8=CgCX1qvO%(@oA7E<<>V`LZ{ zNd{gq#3^a*u3R1BIeq`C&)&s1VaMf%(Sv>3vzoWCu(jstRpCCc$)4%_>ir1r683*H z-r@h7>0GjZewj-+)%_20@_)!f{TrP1AAc#8upksUUq|5MDZx{p^xIq$>5CFZGe;Zm zjL5ESMOz>dH7dc1^5XQfkk`YdnUn)<2DQOX)SU{fL9AgL^R}>E{P&Cj;;+zKyVx8mIh? zWKA9#8@dwhuNw)vUh;`KXX&-V3koD*&W}S@veq>#h^sk!_ZNJxRx2}5XF23cvjQHa zKh};{n?!uHD6muGK#nwidaJkPnmcA61(zc{!*tA|_tE8w z6aAI?a{gJVwa`$PM^n?UN?6m~B4j1|c%EKYTflan#&eYs)V7KUlZFKkX;*gDp6@ce<*P7Th{^=dmbr~8#{c7@XE%~M! zTX5JxT+H+7drb~roatv&KV;Jj4El5pXnb1`Oo?VwEep>tHB8M1;!Apm9)5B8Nb-%~ zaq1elWI;9I=uS?mHOW!?+r891x{$7_T6JNRxX1)|e8oMlti`iyTaIjEi%QsXoov_5 zJ_o-%rJM^|;jd!j3P0eor{5UKaD8UoD{a*#guPnLaE@AfASA1{!e=^csXp_b$rQ>H z&J4+nZtSB)QJXrR_5>{b6_$HIAu+5HfD?uz{_SXnYqE@B zOSHw3!KB@FU}RY{o0oey0@eX{Da_x5<`!b03XV@BX}M z?O?;eCk-YU^{XFn@9L-oc(}LOu$t+(4}DC~-c&abRV{xfgd-&onTsCv9`{;sS5`Jm zoJ5)jB0E=LFs46RRQ=|S4`V1O=C9Jf`@cmMpUgQ*yJJL+O~byt3P@gaJQ}g?1GNWk zpNn6%)YCi6B&%ln^2Wikq&2qbZRVKv-(RvG?o!Qyyim-at}(Xd9RP2N4Hp@_cB^19IhsD*5J}%`k+u zbHrtJ*y`B6{3zXv*E@8m*sFuK0n?QC)P^gTqQ!mVb;ij74EedlpwbDagU!vJbD-i+ zh{^(Mux;O*9>7vq-3b7lfv{7BLuIlu?*_*<_3F0&Aqtq@(wE?2dz1F@tTad^yfT~} zhcUnx<+MsY!u$7~I%u9=jfUjmW8``E2e%}4{%jIB4gUbjKRcv{fBIHyyS+6j<^X1w zApl!54-mjuqU=KD8T0F0njM-Bz+Ow`mf%4*U9GWWzqaYG$-}9h?h`>-9JRf2RHfo>H3Bd_iD`Dx>$K9 z#bogGy7W!{!(>Z#mB}{qPqbKQf{;~Z74U~T&B{mw8UkeYx*?Nsgqb$M_%=tb7k8G< z-3U1?@E2(3{Wtc-{bar)>Bx=-AIFS!(5Fq>#cHxyq&piBT8FiBjOoZrjCmYZ?mSVG zPb}wm=<>p;J5dt)&25`8pRe>V>=QHFzEkDZkGm89A+ni&z#HQ|V_C>bP3XcDrq?%_ z?sAVX)GWJWL*dJ&cFH$_(Y5WXjG5P)7p3N7PBeMnq z_n)cPm?;OtY}QtJ!=&@J%@-UTd_6Ncj~rN~bM$}jpKQ<4)z{g+eOv!(J$8|!Df9fa z{iLJ`6dn^?q5xD_=!beh9}m-@iq$rSH&inxlW|L}@sR9(7elmtY&b-<*D`g!Gh zD_SP~w6FK1eZpc-uQ#0n1u7(lSZe)3;?wQ;1=ghO<5V7PPleuUZ962~>00In`_YHV zs_>IUT_7(2VL{p(FtpZTsOtmJJ#3On&A+-*yV>{)<>?)X^F|M9ZuXRnu)jU(!e_tS z=c@+A_vqAO@kQ@`&U2=fnVTxQ<_*U@&E(@Ic|bLJU1b%z6$V?*v?O19O&Nv*$j`8H zV8QelIb%7Oy3GH`vtd;5ggQX^h_=i@tSsp3#b*>ME3?VLez*wH|kHAv8i#N(3+{s0cNcQ2_vw7JF%|N}Xh*B;bwdG3FQQKn-1SJj_Vul|cT}e5~ zTncUb_rT~!fKz-M(7eDMfzHR^lWE}P3?QVy#wh>jzutzrGnNa|)o{W32qVmT{P&)g zMxg5Wi-;!j%k0O%&&s2vWp?S6No!N)nea3~AC8xA6Alooia=4AzpwaL33Xsc;Pp^~ zb=za`p{Da+{DPE$?)+k6vF4V)aEXJi0~7SQoUvBw4eZPDpmb45m1&QHrvwc$x`k+|oLo5KA>;4)?Su4*43OnNbl z{{*DSWJB>MJP?J}(MupN8gLiyItE@&LXe%B!bB|ZV=M14UU1cduAbLY;_ln6SRMaE zbVoaK6yIV3Fy7@rrmAw8vAa{-Bnzp8(84b@sM|+6y z+yY(V0aeVG&G7iMtJ$ zn>Am!O$lFZ?pK%b&icKz)zKlhauYke?KLU3H*+WLd%l64@qx+Zg@NKEgH^7vo%fZE( z_({a~oDOI%6<=Je{6jQLYB_#bvkez-ReTks9i2S8u*v%qF_XP0rIPpfSG(UsmYrxs zADDV>r)tLqn~Zl}k9f+e+nrUJ^(#p3dL1+22K2-hjAhsEELsLw0&Iju0oVm=Ss$>^ z!?0S)uoZqx3gN~q>`2ab!i0rpr?>D2uD1mn4J%d@la%gR=oabxDLt^)fr5wx%T>09 z|8Q*a@UY`d5ojy;kYMlJ)^{WJQN7bZ_KXwkE*>}$hk|ijKv_?CuO*h+{_iuwx^mj6 z0T*QkC4{?E^LHSXLpLmO0?|$&N|ot+{tgx@! zhso)FjCvX8o$*SI|FHyGqa~5)$!)!<6hK`$|8GM;tQ2TT9Q*HtiIoIl&if8sgSxl?}LCvU3FWeXrZsFR_AaPP}WH%N;HQ z1n+;AF_sirA?EkD#u}tj+KOLZ18fZp3v307c2Nj4=fpom@jzxH6zu95ykHMNW%zm;0BmIS53 z3&1O*z~TgodA$GkD`$gsfC$vrk}-BsA5X9T^%T?09TTmm5|`cO)+iYu2tzS;C>WDf ztp5ASEji>k<+jBI8h+M^Rm|V28(kHmdO4eL^gCK)rH=3jS_y2zpI;M3XU#VFM#Hlo z)qQVL}gqJbFUUD zL)7TYfW?=)MEvIMX!t=Q0};$-{_-E9MoY=SrJPzwS57P|DY`CedBMt2d8Ze@nZ;h2 zevhUXB9rw@HongSx&{9qQ$PhLb+rPtJPV zZkZIjGBveFWxR@KGV5osk~gqie`XIA4WGcdyAO)h)hBR-z2tz2JUC=yR=f-z-HeDd z!+ghvzDj)6>^8{)aU;3wp`6FTApgsq&Lj&l_?MLFd*Z&%^;{q{Fzuf_Pj(Byt-JQzc11D z2vN^%^``RAz2Bv`M?^mkuuK=Ju(kRYb&bwAD`J#dv|+H4XGS&`BFf+scJUn-LPy}O zctsYsZrPYfF$Tc%^2n1)Zpz_-f@s9eU7?M&EKWyfrp*RFGl#mv1&~T#U%+PWKfWhJ zOWXJj=ZAD}g{U_-BZWM;Pk!0N{I;+vqbiXDp{2^;)h#TYIKRi}J%T7; zvU+#pF`&Jm!yJXpPv5gfDK(W{>=P>7WrY}?`3#F^cn| zQJ1_`p~REO(2zFX8sDkmjpEV;x;Qh<#voy8Eq(RC^7@nVXpLp9BzsYINytG^vj_|$ zHY~ZU8GJIx(C}BR)d!4CH&=fV_y{`#4&%4_**YzF=)T2h!R%8{W2nR(%7D5%BzpL2 z3yAMUxU&c{GlOX?6rnVhw)ePizVxf7EQ9VY1b5Y5`*7+BerSf^f&3!PZv7 zQ^X>`I}-q*Zb-A&Rn6>5CRfd>v5#poH3-XmtmsnLb#K%91IlSJq(XULSY2q zS2=~lXL<4BuJ6}Dfo>y_+w87~&2GZ&2WsGGKZ#sme_BDyedXUY!_bkO7uNR? zBH>~f`n_26l|uaQ+F>ugzF{#C&Z#zrMakHMx|GZA&xXGrW~r0FhEXp5p%{2b*7E|^ z5O*Jd9n!7`KE%{z_=(xxF`pjNoiV)nwsv|6DeAwH$H_&-iMQJPL)7~BX2tzz`u2BA zg4Lyx_`1+b74xdh0!mxgA4=Le#j8Pv)uH+c`c?wC{Q*zFQ14AIAg+PmkA})$n+WVL zgGx$V={&phRxA^PAA@efv9H?hIyaZ@+a!54f^MqlA7lFvMb9 zT!xQjq$Sftz8hP55xw8o`KsO2|VZo&Z+Ia}&YZ z&f0PY4-7r?MrT$zGF6np_VZylC3h3;y|fiMbtUqWT_V;%<3}Wb5NG<9+H9YsM$VIf zD)6Ds*)QNZEnTtBXV=gYwU7B{&%;pUqX(Pypb@4sOLE|x&q1KurCT-j*IQlV%1KzA z{)Z^az;%XdC<812G=Bc$$_BOwM_OXt4m%xzT-%rGUce8B|Fx-kga`P)+u`WHx5LTg zK+XSXwG8n;>>{vP|KGOu87LtILdyS|r+>tuJYeF=Bd~*T8=ulfwavS4q|gS}R4k#c_0c!tZ?8rujHbFO znIvd+9V|Qlo4Ur;hX;(E&_Nxx6r5Cqv{@_n)wMQR3#e6N^Ae9dVTn2RHNNMw}YM~lry>)|CTl%CdBaOuVk#z(H!L;># z(W|{rZ1Ur=J)1b`bJKFD*n`UMl1+-#`*b{JLJXp!AE zC6?R-4O2;a|mEckwy4c~~GuzM{FslcW?o>kCSfWN^xF;tfP01ENo z$*i752xnjh*IE8Jb!YCfVM{!1JAtVOjToqAohjeuxPWGtojx!oL6!_YH1R(|dUVf2 zWF#0}p3J-GPt~jGjEbv}Er~xenu|UwfAJ%Mf2P(#$zJ(s*_FZU z_nJp#Wpcx$p0gwh7UF;&f(z`XsYGaA4mLxZ>evqCNh=4GO#!iZcTTqx=$ogCz+!CY z_W1LThktBkP*+^qbw)_mrC=I?2pDqP^;&GI%NhOLr+ zLZ_nnRR>s_{Wax`5nsM43-{Q+kjf5|F&WkSPWpwuyo2~DGMyHLkFZKxTZY!{kNJeT5qhruCE zN+r|CIc*n;D4c_S77N0kIzx|=vMFJg+znYSy>zG_lGO*!zB@Na2h+Iv32F+y9^rHA z8ODBotP*>L{ndbHI+99)`__M#^S#_ax%nf3Jkgv94aLdjVsW-nhK#zmXcF(ZOo8Hp z`s+4azIk7H;vba0Ue|MznyhA4L@c&rvVq^UXLqpd+df*~T%=JflS(~WBgN@IIro#uGv zf%A{Q=8DUxUq}hmmj&o4Borx6qwQNLvIc0KsXRy;)FtC;HuyCpt>X+%cl^cKxQNwL zC^saXfi+-#QMg^K%~2l3enw>)m56WZgpDjI>S023EaD+F;sbA2MuVrN>PExQlbg*H zp&;_ycKL3rPk-)IRcMqfgqlhaFV>0;)nk?9e99x3nwRnM?`q+%`0L zi@Y7jBi7+sJEJgGp=Sv}Pq0{-X2BJ`Rx==Nk7|-nUOZDT>n$IQs#YnGv6poHrT-$; zK*}(^EmZuTQvPdK;f{@}{R)}L$t3NTr2W5}JHj7q$M(;;-sGG%eQ;-R(%|cl^(oK% z5#s@32c#uVKD*`<2_$8}_#Pb>44Fo&t|_=}&a|?w9*n-cKvKEd(3=#h4SSOuJXYO( z@;lCtizc!knhR;07dW*qmIu=Y)IFCk19=bXu8$d1X<_0a@%Ro>gcJx&3hx7w!uxie zwFe+S@QAC(i{)!y?f>}H1!erDf0LOmc{~$vQB#v?^A9)s{HFQO|$IQL7Io1xNRfpfw~Q|W&W8c9JyScup~gQ zqwzloj*P_;O&_7Xz)LkF;wa#rbz@kn%L51IP)+zKSEc*YQz#C@ws4nj1WCi(LSzb4`aRpe~my3C)sZHdBx)g=nV3Q8I_5=Mg zB50ct!pcktX+6j!Q`5x(n`eq4YMa1!2giHlqncrs;asGfVGWWid)~}ZEv65bvkBls z@7Y8nC2P3%4b@-p>_C=-ZV<2=D7Xllwv zng(=Ditnk=?AuGlQ~&Eml6?0g%0D!&b)m-YdhtJedARN@rH3Vwx8<09_JJW>?)zxT zZYxZ?G|gOyHu6NTCda)a{incnp&dRShzs>8#K@c^TF&@NiT+u!!uDO2`FzkD+bmJ* zGK)ZZ=z+E)Jsm#&?-mmYWfn|$@y*g@UJ(?1?5`=VjiHRgn#0N48G$PWfJ1-dxc^On z;kH-A86ZiPOZ^0_sa6!uY=)Wsdzkl5r!<+uavw;sP@`F@j5BX_nG%uT8~K~stD;3C zzUS!i4Wk@K{Shs`hbDMVueyD2D|*`WKiUs{{>U1p>z~8;raV)p3O}+wK{DW}(A@JS z4CZn(c41dG^f|J1n(B+mCh%eg3-3`9Cv;XyKxy51L-cmo)96xN@EjsPkm=$&2>psV1%)f&8Gy+Kv`MeA#Z zskoryUGBIUq3=Z25Tr^<$}3szL-BK`OKqQwKvumb!}(9QqT_nol5v15dMXVM@)B)hT8;$Ko?)M&di z>#-c5n_giymVufgD9ze)m0_!y+5>izz5(MmuhC6Wv`oXXCdV-L$)dj|NKnzDAU)t+ zK(Tb3*;oLnu|s#&ePsUHhkU)z(5P%Z=Wc(~4!_56pe$}GrgC%Uj}G!x7<=CA3wzX1G1Fd~?`dbR66{C(~Eg^03 zO;vSzyXu~EVn~P-5wucV-qljd=uT7q4F0P0wVZ98C)UR@!p937`y0ePDYuNh!6Tlu zfQ_tvGrxm;*lC$iSLi+|ZfJba@a-ixw<92pVai_SMRVLptwvuSmup5tlfazQugt(EqAY)=~97EhqU3qPG&9f#|dxk%45bZ2E&5JvLP8$<#{ z;IeYQZTfP} zHoO$PCj)NE&j)rcAoZqY4|hO2EeRv+`i0e zQIS}W_Ai9LpUQU1%&AZ+>>3~Sr^om59&HaHz_f9_{@Rm! zXj&E$w`Abv7b4@6KM*-y8~oFNrZzL#0A6Orz#XCPJRffm9iS{5C^Mt^D~C5SuKVJU z;L?R;`J$Y$sv5gQe>pLPaCmZ&G+$?5@z*D9?*>#dvPH*bb<5;B9Pz;Hw$xxCI&}N1 znfYoQl@guesUX?UmmopC3PxZgqGJm+EdY*{FwG0ubj)*+F{rT zvh*gCy*)ziTVg}-F1rFw2G>-4O9hGE&al>t@OrXRd4e^Vj#G|NaMEDK$hp5eO$zgp zty_D1vyZ*EaN?w-;&m)qzFO#ZE9A!v#|*z+%7nYjBOYo63s4JNsp$@*J^fh1uk|wC zOg3Ch`n>Ab^WwTNIo%}_D#QK}m7C{CQO^o9(FF1(8QV&oW*K7{HCBg2ND}^)nxox9 zbEU|OaYDwvTvT(wW6wR4v+wX1@H?obVYxY*6+tQXO;oant!NC!YPf=X{xisHviYC3 zuIq0tmRP&;PlxK^A%^SG>;;ck;)W>k!PEq>Moo3>%(cM&TCz<#GN+>%YN)bBjo5go|_YG!N|5J znt`s*k7$Rw(ZyG92wzZPm}f&1_mtH}g5aBF#mt-|zTyJE9AvT_n=&=#gpzdU7aypI z4h0JAxV~P*^_2ERZTw>s20ghm+|i%+Aly{?f`0Afr|dP z^OfLZv5KNofqsG{sRwUy4f>ywIkjt)_(+(JxJ$j=e)>a*V_vRxoNh!o+Q4J{Pp=}| zZ#ND3hg;9;4o)iWE(X2bJTutSXu ztpl=1H2qu1$rd@o(9@pU$RTGC9#ZjrP4d=pnI1!I>=J7P)nzsCu|W?s`T!vEjMhD4 zs758gI%Bb_p8wU8`iZi1iWg#jz@NXxNH66ZWqf~7Z~8Y)R@fDA3{jXY^-I3a`jXqb z65fYz-*SygV@WWxYb$h6T~oaOU=Ey#>D%@8+67*pbLn1AfTNl!<7%yN#6fML#ZPPs#MesADtp6V@pi z3uP$HMwFI#wuW0$Cwn%)cZ@muKJ_XoN}Tqu(4M)PO9VW*E_ha{DwcC!=^H1%_eTsn z0@apL;~!`_EVyd}j`Eeh7xAoNeVDNXlH8rnRx@n9^kyLh0&S|MEP7UXFhk|`0P-`> z((REEnYg5ww0c>9wE^3jJfy0j-pK8J1bgXghS6o?6sv?tvh@m%8;#qhfG;P@JiP`P z5+cUrS(#e{!X%8Ufzp(Pq-fxHKTMXugW-S$O@P$eU-r<23m^08)2rJpUr7H;rV_Ws zXH{GgtoXD1#g>xGqdxNDA=qa>q25axEAnE4PcZ+L z`y5RaRp+GP+^qsqNYHn@kfV71e zw`VLpE^AV@v9tKnCWi;rI(nLTT`U0BZ5<2X5JG)6egS_nM|MVK)=2}gToxvT$@#Vk?bQ}86GS+jrC(w|3MYCS<3i-Zxjn`W1k9pXdfZLWsR z)L(O^K$@!ujE`mnPN?X(vaV|*F~KI6mj-6jCjt|uv`C10@md@o0+S7A*r;MC`*>u| zO$a_6n>;?Lt3w`_w2VG?ZmM_hXWC@&?>2`ohqjG!o`&xqer!|~u1(r*7~ApUnCrW! z9>EK7P#2biqKG8v{rEU8(qLj9!FcLiz;Ym=tIAdvD;Z=R#gzwf6ByO-kFt2uZ!hp_*I+gPS)TBtgIysx5TIrxew>=NR^!fDAMAHo{n*j} z;_h#W6Y5DM{pYg5PBE(4!Y&0lSKm@DRUvo~krKj_c4-R{nsmy}SWe~*F8Id|u2%Bp zoT3f*+W1i$XC%Y<&wG5E+5}?-Yp$z<&c-Y|j7F+;>F~LJVj!@LfWD*O?#DMM=Ykh_ zbeF+q&m6b|V98!zJ}VOM!7kIWW4#X~Z3|*I7141FYJtUxoQp|dB5R_(KnD$};j(>^ zWM)D!;tSIrgU^aJt7HeCe9m=0hzySfl}R7`$d+mxkD8rVuUkbd9W!7a^c_ii+uB+s zB1)h_bdX6bYI^;+NI*7GKHR4)nD$CWP8(guN3kId*u2npdRXzxOXW50V3oWEgXXRe zOB~mt9-iDbtIMe#f!?YtzaLn|8lvLjceZ?U>ontD&h4z0XHvQPI%cq3IbA0vK$ve* z_rL|-LS?j($I;Ms4Ds@^<~ zn=`H2e>IDCDg-t|q}_a}MP8`TiRSGjqr8DEaxZL32tXY^e;a|?Dz-w+d=TtYFkLyk z%xlu~``*6CsZ+IB_dS`9{z&ZJ+3wK_-~rgB9k-XCD78NQc>%cw+QVLo4dYbMRZ%J` zEcXg7=WRwPHwh}|duCAHSX2&B_r6ncwm80PRlYscbIW?7;NiC9`1zRA$AI=z$N*T& z2Yid=LbR*NbfyNkV7T)&V^cd>Ea>BT$j7)HzHi7D-h?6Tv|h*`-5och6Dd}v<*%k( z(T?;FjZx0wU*9D6IIHC5e|q`S{wUY2Dx10OQoh^z*GJ?O;O>cq>-v3nb3Qf4nY2$I zFd1iYA@vXByE&vcFCy(1G>rg$0sQ7KdzUxifhwfO7~v?|A2bBK^Do(S2R6Z6i{MX` zn~2Q0`%QtvGV-1An(P~rs6ZLBnv&v_JaIRJ$E!Q;LZSPeNUz(ld%5c2E+5Zx58_5 zOMjeGW__Z%gdiMOKbb2GNQB0f$lPQz1^aOJ6VJTDbB|vASU^UR;cE_YUKQQa-Sg&q zwf)ttBL;_4J0Q>-<@xoHM>SWZU<*US*YYLQcDr3Ho4nj*W6`hYhKDyeXYmGD z>F&r5OPhY^!510%(>38e3kaLAmNZ#B(sw3b%!7|ZDfnNA$1-ES8 z>$4itg2oMG3&8dF0Ka;=`GHJ#1aBbr*c?Og4LuR`*huC!OW5!C)d|0(DL9FwOrIQ5 z`!}68_B`%6FFV}D|9b)Aek_`SzlCoHO|;7v!gGBs(W8Bh_3^4b%H-3xlFWplF?u~qg6&gJBf+cu z`?)FZji3u^Pyw?$$2y)qFgoEFxjX7Azn=YvxjzQC;*SGD{X6Hf;D3n&M0$Vo@Rz3A zF6-Lvd~DtK-^HfvF3)9~-_k$)rJFD$kl+k=ke9=7WNbMd|6aW90N(PiN{9?74Cw2f zaC%pbH|>qj^qMm8yRX4A9X_QP=^FdTtE$#gUHwh3}lhmuDBd7qsg}j1ka48{jHpJV?cDwyWuJMt92r*gH$pa>@H* z`lTH(KniHQ3CsM%-Nd&a-dIK~%uk{ux+G6i8M%voxJYt?$c!j%0-%a7*z|wpJcmMz zBwv=xs@*k4R=F$*SkL878Yb0RW1^7)_;;FbMzx7EHM{_om;>}Z5q8JYVnnrU<(ZDD zX2Q8|g+IEWd&cb6G@bS(J*LzzBATc-?qC)VORW;w=3 z!Bm6_Qw7ZKVUElTR%Ur6zpxhkLT|pCOk__F^hcuRW7rQ;@bd-)CZ-xQPO6-q%j z!xLCoJs94oaxI&4O~Y4m9kc!~S=4Ib}CyHS4o$t=~1%V3@JD(?l_uS(W<#-jgD6%gV;dT>F_%)Tu@eg{LU+u9$d&G+3U4E%y-G*z7 z)UzfwO1e{qf64sPL}LrUdS)X_@_AGl^ELX>Lj9yo=G$2fD20@QL*p*a;m+RZP24S# zsNFoRPO2(MII`OlC^2!kn6OXX+^E-Ik&t?V{8As1e^euARMS5K8LQ1Q)@BKKkH1$Y zn;Ko?3)#KHP{SQx5Orf}qm@CDp1znjdL^v`e@=yvDv#0Y1)oliF%`>PndNUIE@$EHWRN{8aCDJAsvOmxJaS$Wr6K zI%{1o{H@<28X1FqB$SJN_(n^j{htca7|nZ>$!r4JQ;;a1je4Uz3r@$}@%A~nNW$4^ z-;ra0er~O96M4PMiQ|V8hJ)pr(HL%gL}wJ8^zd-3!2FgBC?sh9n_$W6CPv|HRuFt} z(Dp6cXtoWhWn9!$Yn-y6SaqZcz|~(FSv@&~&xgn}k&~(_Njn?gm6#WWwc*N#jH%iV z=m4G7EN;c8m((Q7fW>FXryfuGQmoa``t|P?v$4Tj^{zwF1zPK5hkW2u-KqQtj8~ed zw7_SuSD@;!yV*Qba`5GpE3o~%&gTDx zH+`A3KI|rQ^LdQ>NkyvU^)egZy>BrY8XJw;Yyd(5Jm=yt=f8|nkHf%8lbqQ0HT{`j z>_G7Kqp8#_(Ad=nf)F}JrHlQ^bf#95zg}Qrzn!S^XJd&}ZoczBMEwPL$4buM)PC{x zU|K`LMq%JlNBc2X#l?;->hotlWLr3W?(hLU+`eF1C~=tU@GSdJvy-H5Ao~(mLbmyH zD;}ut#jMGw>uK?gb8>f`;7%hsWakXtY}&us!^L65I1j?!xO(l!rectI)c^KluB4># zc7@?-huE?nvPH90 z?OyyDT{pK+8t22Wk3DQi2?ON z!Wboikw~c$Y=av_Ln|t0mfkp|g6ny%;GRzaSTl88@3H|Q9;J}^JGEuXla_E6X>{lA z!X~PACcDqAh-EPS(`dt${-tYgJ-d0tv4qDWu?c^=&)(T;w<@$plN+EFBS5NbY_U>Z z#zU^8XHqo;1A7!SpzsVA9KwyfQb#+B<(T6t9Hb&w7NpK3Sv5tUNhxR;>ua? z?HqrM$jA6rw4$Gj$)+_n$AS}Fj>N1IwJXa`V>teZC8rnUZjH|qg8R}zBQ}GN?S9!- zie<52x($+M2f`n+0y3HU#7WkW2epHNjF)fchu0Yvbl#`x7!%oNb}Lw3#507i-|U@z z0-LFc&PXUQ_Kq1J*#4J{Du3Ko=vF~tpR4H+f8JZ=6oro+oL6pqY83*;Ch!G;uQ5w@ zz;`H!ERfE}me||>J*VI9C@sb09=xcwxkj3c*5jgvco$nd=!rAfN4=3i?WEEDK3z0_ zkKCl_82y0++zg=lJ!-}GZeoaLi0ogo3f)=p3$8o|wQ_r}wDFP;jX!Mk<-d}ln_0Y? z-@=BLv8%Z~%%0XewlD#TPi<;w9=2by2n?C8qE+Em% zSM(AqCP9;^-3Zw19`N|xE&`(*zp3`_g6a9(&gYO;$M~@$SC?LId zzbD|*7b8S(??^4fDQR#mQvGFm+D8_BsOeU@gwiY;ULJZ9eG#1V4u39_qy1>^I1D~l zuYZ1oYE@jFbE)rn0`wt8>1;j^&^hg5?Vb-;=4+p9d(pKrYn8gem!1<^C$*CY$*kQC zJSnK#Q+SxO3M0XBR!ZV$5vwkgCJRa%CYcc=`LXu!nh!}r!_Q=xM~kyM(QjRR6RDcz zwmDbw^}b?oP!9yIAi_Q0+-P7K03({~w(?2cK18tbsr}sSIb~&vOiT?)YQMrqLX6?x z>fiCBJ^ZCH#C_q)oFLx)lGQg>PJMSxB9B;sWQ4(P8h1d1&PiEWU;0{I6c^I|a`2!J z5aVzD4WwIzQ%;AVWqtYK#g@XOw|4Y<750@p1_U@_26hkUv!wR~CBFSFJDr*6`{Fe0 zkpXY9U5S5P)+8tKH-^BFP&&K1FD7mGGlw5p9l$nW9fE$=UmJYVV|Sf8lp;-$aSVvS zI;T<9Ury(1Yy^7ZiiEpJN>;vXTZnr2=LZnU5Pr|)n`^fW zQ#pV7Qs^TVoRzdxXL!^OG~zl9sRj8DM^;eK25LMxwG>_GW+x~X^Us`wK;)g zK`k*U)ZsAvdL{*&Xp!x$&#{|%3Xe&&aLlw0z=@*TeqTo(2yEc$zi)gOG(f-fx~BPT z#o?fFRvo4NAW0z)&!6EOXL}}M1}){?=ltS&Tp_ZIPDz;HnPSv#Xq@~A6zF!{7hGxy z5=WZ@NN>d@pVGm1$;0F!B^x7S|ES-Xy|@P{!I?pU-L$klE~`;AZ_d=<>}r=X=MI5i zIWln1^2aYgwcpIX2lr^lEitv>Q1d^zqletyM!LzIR`_@LofJiROxM@6$NWQ6TFnii zz<@fzDR4FOS2Io(XlBt{g5UgCJYkIHk0e9aE+vj=_C*pLXS>+{)alq8uE7ih)uv|8 z0Bl?J#x^nAqECgW1hI$)_mOjl+RtHV9Vj2ZmK(DET{m*uOw9WZZm`X!&JPTZ17 zv>0Bk;hH{@aOg{KzCbr_>qsPZ)o%0f6$UuFKkpsk*`e(OJp8o8L-5aE{?xyYV!EoK zlNPQdJ~Y)RH^fQ7=&zG@mk=u(8H*J>iLVJ6r>DZbzw60cdl;-RHaxDW=LY@ILD?^n zk{%TX*zQMl>HQnkf^vbk_+L4&eKk3eg5Q}bM|@6X=j)kA^X{L6872rWy#QxV)-!j* zwvD>6lRYlQJ!lV{0_q;bmNP`D>4{%~F-%gpF{|x$mWZ!pGbTK}E^gxrcWQ4WB`Y1@ zE!&c&Uu}yx2Mv190pOC|fQ!vP!CQ@xbwWjsgr2VaLhy0E?@xq-2=Fye1FOoDOCm#p zr)z(SpkN)I=ttS#)F2SgxBX&UQ;5?~ef2Gf5g=S1)$b!NH^~lB*dgiLf)Z|pT<)Ds zmpJ3O_9WO$Nm2Mopyfk>tZ3x@eZGxi{r7<)stgj(G2%ikK`B{w#7d+d)JO6|rLJ>O z&)xlMb8uM@7%PbA@}A&N>=jsKgr)gX5xh_30vVh7e^{v|^q4#y3$<^flk%e&+{s&aDLQOsZH&z^psPLX*h zEkk59mD1Mgk_br=A_!3)bcoX2d`6@6?BfqRy1A+^Rb@G zXSL(9w`!viHL=&>PM`fR zky)SGxa#l}``JM5)!>MkrkEFC%_)Y6ftX(T?fyBVNV)21(Wz8(X+WIcx}2uX<+3Of z6EsHQ$Zu1eJMT^-JV9G}+vGt&#hMlidk~V$0#pEo|E!^&W*zjZYp}MM$gg)}vaH=` z+;E#carvhA$0uZ^*$H6t5qOXfdvzUq5YQgxt=2mUbBF+*9-m;AXxxzB0QEJu4|NNKljWB*G2Q=jse2<_Cb@ zOmVaY}DJShl;Xkwrngl6)l2kGT;74zXXTKmt~1F z+WI+Zy;lmR+yMnU>dWMNJ1odtP?ljx=b?wiqu_J(3yQG>Q{}i7A1I|rFGE_^Qp=Q? zyWkwUGH0BO|IxU zTm#HhdVDwXgxvW%6eHul*I;nASL8gtMP&10;x!6Z1RKQhHLg!C(g)D;Wa0iLQ+x93 z4vvqioA0pIK0r8ggy2#H(CjRRbh(NPOP2;{9GbIn?@726xB)2+6cG4_T?gAfIU~)% zEGo@f4Wnm}!7FAjfdHKX;F(|^kbr~A2qUacDPA`TI5+Gp5JQ@LY_4b7upv0;W|j?v z=M1BFL5U3-gNW%5lP5L?v#SDE8A;lB#O39#%Ki3+0A}sBM!jpXd=9wuWF>~eNr`0 zgtS8vsbw~xM;(#qPm)s2+7S|;eY|^Fl#5Cb>?L;gpR4-_7hu*F$0Tdqa{yLe+6#$3@M_oU6@Y4YLzn@I5zOgen>p$}_^waSRgQD8> z3twDtFNz|rMHEu(ocq{cS2 zIQl-RXC}j1cHczkGim&$&rjq~Ea{t2e0Hvm7+h+&hIAVXY$Wyv%ISc4!<`TmvD3Xi zjJgvoIv6A6sK;d%pBinglKEA)4`yl>mQ##3Cu9@~o%0?Xk3MS%er)e`+JyIIc&i7? z$iFcDUe+zk7jf1O<4>8C_{S&v*1L z=4^^pD8H>yz_kC`^5-Jns2Y-rPf8;vJ>4uI&P5ylRC@kFSOOjzW)?haXKY{jGgCNu z`WuPk@_!s7Xn_f0m#eCT_ zm)puve;w3#LWCdxOV;JOzKy>*$si&TodOrV(#y{@oWIGi5=^(3EsOP)|9%bGj$_ei zQ{0@4JYPmzCy3pTUcDNfB#-n@;$wH=u=yeH8hJ6e(i_1~vQzIWjJWdgy%ax;#%KM; zg!V$P%X2v&7=y^`^x_spRYs@u(LoOb0Qh@~rW045c7e}BjvT>l2XJuB_({~@&uld? zWqZr0>Gbj8wwI)iOX6*-$hCl1GZLbKsY}ZqBd%X({g9c%1#$UBnoFC_K|LgO=X7OZ z|8y!B5_-$AuU&(h2tOL#d+LeN6Kdo%Jf?Ua17CpHBtR{97; zsWKV4NSL!)Az6jzFBMKvfIYq-PRD?SnMk_#8^ZcFy!$C}|6Ip8os2TtOK6u6wA4)% zrPT-3n8cVkQb%?EC7Y=jBvQ5(47*;+`Dm$F9)huB=MhMW!)PQ)>XEnb6cOCjZ7O37 zL|R0UL%}%xgGb=GD}G#k#1B%Nx^-G!r*X8E^}1eC6;5Th9k1RE zxJLx~1lEd$*gMlfH^E^Gk3^JJ&-wsan)ohs{>51jjLu|GDxu|w?vU^vRNUhpd_(AI z6+Z6AmA=sZ?MIF!#PcU{Ymgo|ExlR^L=rjV+tlD2cXbC@(&8y+aE(De>?U9JcHvIv z`C%OzKU+bijSS_bUYiN)!F{|_p{BI3sK(~h9#z4Wu6(ErnQde7iBwJy@f{ABbnzu7 z<`|#4grChDiymm2mxy>PEZ}Irr+@sO)TySHoG6?@Sng!mOrs7!w>JM(;Oy=n zJx4bk48^K($5rim(r3_1Z!-8v-#nIv@s7)B*G;=YU}K&I4No3sY^FTJ=pd9m>1(Gy zE=-pF-2HFj9P&4~zm{b5?ZvuWbF{+xSgwQq4;RV)7w7hW`@{w@=O&r`x{-8hOIm$T zZ8f8kX5EL04+IS>u#uK>(hzBN4I{%cYQPtkfSI<^N#euRh<+nA2VJ>$NVP1|UqrUG zF*hWA$w!W%ygLnNX_GQ`_=)tdN`c-{n12mzvb1%?);$LjElDA)o8ln z>oKUmd4v~!<>%j0RoY~MAch6CxJ8zQT7rtG)tL!$YKBNv7eTlUj_XPtKh|dm_#H5A zhYM^NGK59}L$}JAdgAY?_PX(HY40}sAQp|mV8e=^`Oho5CbGU<;R%>aPQnZ+dHMnt z0R_TqqjEQH&%|=mA!-P1zGjTIWZ|Hu;09QCl)L;R2L6y2{4!VM<6G(wDk_ zp2zEK%_Jp$x*XZYvc=93GjZA}mmPP%i}IX=+92Hz7~~PJ99=QEqn2&V^z-mMb8$&n zj-r10cqt45kRrf9!rREQ!ui?~dxwu9Oz)dz>0(27Ae%titB3gIAA(DK=CtNYoY6Ps zyEVSv*l`PjpVsf5;nK`-6^#|z=&4xwH`8G)FV`6OrYf)6jnCkd)KRnEoyGq&RUR*L z&MWu>Of5MRXOBiB+Sr20efU(yW90^AJ=7%rCG&qg6I)gy=UTZ2GUfi>dxB05i+Yib2XS+-|8Y71+_R&_OtCZ@o9@Wie%#f7N?BoxCj}~~(K$>J$ z0;IoVG(frAE=^DP9Q?Cw90HnXC7M(IQ}Y9U@Uq~p4*^OX1k3T@V-xtvWZYyj|Vici3JR^D~K-Tat6A)x+aX8{z(+!iQ=nDRti4f9WxeIHT^wOlif5kr3R^>{LW8xb?H`%3)py-` zUrp`)EAXJV%>mDaKnX5}+W1i>N4;gteri9PL@89r_e^~~VbF;(P|2WiZQAKl$ux?` z>Gj@hW#I2ibY7CNC^fSp;51YG9eV4w^hT9XIyj#E(>hy>fAjZ+t0l1eq!g4!WJc(| zkNf+oS?lb@u!}u_;8-rF-hPn5#nTFI2(%>HvR;l6%{{R{R};;5n6@k?>bR~^< zNYgVH2HJas`rak>&t|d>!&j7k&Y>jw3p;HNfyUkkgFb~8`*Pjw)_0ua9VV*YwMkD z0!Uo_SpFBk#n9TvWqVhRBEC?p$s!_r)s8bd`p7h_fAQ!Ax)QdZTW8=4UT zlF6%&|16{;0xqUW6i#Tq!qZ9f_dnKajK)SvMFLu`y6+Jl6B!or2I4657ZyjIuA{Fg zY|sY)2S%2xCKJ>w3Qr&Ue5jk6ZaV~0Lzu=Dx^J~7a=a>Cx%^Ea~J0mqOB&H{7nV{^1_BoV&kALJq_mz%UGCS zQjHcK1-$z{n2RvWFwTSG-YRP7_du4h95eL}z0MI&bsdwfbUQ zysSF#_hv#y#^IYW(w7fW@Qp}r#76!dr&H%};=jl8+dfJu74Y)l{Q4kej2gez+oZ+U{uo{eh?$JosKlHHvb3SomwQ7U~)%+t?^ zOK~aP33Lbj6c->xG}TZ3<8wRBLi5#8D^u-Q(z2T?X%+tTAY3S@|FLwcKQrUH^SyV- zM4v>+Ae1J9Y(06USf1AJxJXx=ZRCnNF-lL+KIr)bzw7PwF4k1tAz6JYsNcZh;S#!| zzaa)sa=g|MT@4K+SVHRe9W4wn?!!~R^vklI?WNRM(Ys@=J)EKq++t4T4UddBOAL47 zO^te2Y%Y-$cvgehQQgUQAI2M*8EIW5-jIT-#*Zcvl6G)^yT8Fy_GX-QGtO`bRiN_A zfrRulse@}?`@vNEUsCmc8kjapf{+eA8){Qld01O!cf?}1)$K(Bsf1}?x+pbw$xw&> zZg5u7dW|@Q72NV_GL1Bzn=dG6qUL|zz;FcIE8#x4VXqa0YbG(9&FX$#=QXZOE?FIS=IydUb=l+cvK!ugZFA z`Ep1teo_`aZ#~uwfre%Dy6!@{_Llpn3@;RdjprC>;*dfS^+dY}k2K;*x!rNaKI@<1 z9$&_eOT6cb)7?>j(t%C!vnlWLUVRUiAdYzCz(@i!3iqHrk}yS28L-aM4)+j+s&qG+o$N)$Gg(Kv81x={A=@%N$eguYuRx58j^)cZHd8(e;v z-zH%sEJyqC5)n>8e=Y2+|IuGLc5cNo-fqO|bvGbVjq}!hNUM>l6J{?-g{BB*cm6wI zo2s3;;!^#(v4sx3Cy=PzH15I_T-^9RSjS8pUqrmSDH;B;VMdMudS3Zl;5x0?O-1ie zavjkckl=ltyV|O&E^j-$>x_}Iw&>uPsBTG^zq7hg5m!nSzfzu*3oIJ8a}`-KY{$hR zn2yCs48`bUuq@SfaAj^w?2-~0!eeJyLYkwE?+?rA)`z!O8G4eLq0dhC?I8k8Q&yT= zzBGtXmlGEP!F@e6u?@Zjq^k7^00xY61{o(zE;af48?Nn2mmFu_fo?RWr^&i1&r_bs zMp5Pec+EcD>R1LfT9|DZ7=8Hs;sh|z(99Z^D$aKfQpy{$@QiUMMMr%z_A%~Q3fIaYxEy@NfIiNIKx z{Q#Vo5d5xT8uq)UCMLfjz4KXdf4>xz1nz?LOkD9+jbG6eHnreTFWWriHF88>oH3Es z>Mqvfb}_47c(W1;REvsFx!9ocR?hL*C3O%Se-CIZlhtQ;0IHB=Uf^cYl z`$}ir#VRnNpJ#5JWI9iX+T!k{snQ##Pv1`rz}d8f0isELEn4ABAy!FFi|MO zTdEm-3E~?*5aiC3{Za0ls%+ISjtR8YjLOViGzrkiR^jmNu^UEXGo!N*x8DLW;A^@85izoL*p~5+Vlxh^sum2Fm67jehy8)y0yJIh zEN@uE%YcUBm*qsr9~&&Ni=;Ank$05pGzuEXha$4&V?x6ma0he_2%SdIBFXo~k_s2c zJ}|6v$f(%$Lw>L~?9>Ht{@`lQ9X*Eq{+G-q3(R6x1o>5~TOPWl6mYg4hh78y4xqv( zD?G$kBV;WHP6EP#i;mHN$&+Fzt;crf)Ov=(jHWfh_O4U~&=}mj0m(E9C>mX2PpR#( zotSoQ5W#GdHaksS8Cf*f!#30H-Cu;o4L=f5fUopjI_QoAtb+Q9+{mYuO_lX+V9htG zxWuM#GFMXO8{o0n=mUEqXFZ;4YU4cpk+dM)KcOTi|FoGRPU})1J|*zhxO1LdO4f?C zDxo)c^xcI^58&P!Ui;fVPvE;9HQi_9%0n-QV;iov=Vlvw^b#``j%nwT5Zh=?u}-i% z@6nEMB!PlR3B_A@0#Wz6x_Z8bcA>m^{cC4ku2b6(*EwsGL-;=J03p zabMy`(3)M(D7_|r9uwY-wmT|xg1on)NEGS3KdTYwbA_GmG*cPaxXT;+I(O+D_>X4_ zJU+dm=*YQ{D(rFm|3lb!M>Q3FYoe$qRa8J~R1~BtB{YGkND~18=}katsM2dx1f)hl ziuB$gQbLC)y@wtMHS`Xl1_*JkzxU>?nYCuE`2)x$;c(AA`|PvJx4(TG5(T5gH-t*p zIYXq5QBm=p9MrWHdy!d*hs9KGagF7lhJT#`bF319cQlIOPWsyD;dV!szII;5D#uBg z@XIw0NdMoLU5$K>ir>yK{LTp;I8jYhaZW^768xsbKt6jk*UnGE=-yze35)}meaj7a-C37hXltFLSnt3fFs4-byrtNae z_YO=&)|2pX5ZIH2A~fvjvvPy!)#VL;+pNt(Z^;o_yeMjgM<+z;KvH0L4lFH-McIr4O4gWm&L;RH8!On-niJ00DvSXMZlB z`*wA?OiWhp9k1=yfN(-g%m2ap{vVEj|Hn057GV6_w52m5om)B2X)S=NC&r@_(Wf25 zLB9bSk@N{e((7^HCf)-_VAgE&diB4SLzgp=AEU=@t3#I%3<#Q{$0ey^JEG4;40`?V zGUvJpo3TXwaG)BmK={cqG%u(#4A_C_1q6~4C~a!lt0EX)oD}Jv<*>NP+M}zi?V`43 z17~Z$1#sz4PhUA`iOM;4k8L6}=)*HM7$Ad?x+kPH(F-jSGqYyjJ3;WU&sF>+DS>dt!)G<@~606p(yG8=Y z{q^#1S!DMP3uzlC7ix`d=M;Pge-8(rrOqV4k;M`x870R6a;gves!NW|(h%$+<42(@ z$O=Bl;bYIH_&hC~+WMpe>e#v<-`~J%f@OH_vXbp`RJv&JM8SLPsaAdoo%OhADnw7W zmTze$?v!hb4uy2BciXgfITm7anV87Dq7A9Mba$%vWiF;t#8-H+eB8_{ z!#7(GO>PLpC|c-=rQdquSDwJn$FS4DY_jT#XMOW!weQ0p#A9SR)VG3c5sY20Z>uMM zHccJFtf$kdj$=AR`aKj3D6~@c`b+xOu1g?H_h!CEcjSDLt-2up^AuYYyunvn{;Q@Y zg07`9wr)qGP2p+fqoJR%Hi+j#AGp9}veS9fCJ_U|&Zu&<&#_XJmbufd| z2Kir1Fr;F;%f*!H4b=Xfqr;X@LlHi(*QW<9&J-7gBq5<~i%bwP(8WajR`GJINqUlW zQu#`_V)gVJq0ZBV0_o}VcbX=CQ@)giF5WfYMNGyZK|Y*?e0dFLC)5N`we(X4=$iD(kvEL)r^52OWK}-n0S0+TAB^(LjcP3 z19R2;zK7&dY24VA>W_L%ziBN)CVyQ5MVBZy#5`3ucyHaW?|CI+0hwch{OKlZ@glb` zx%-<5lyPGb8UQZ!pWD>)R2dI^$pa+Sa~6t3JD31OncnDazy9a-(}Qr0j}Z~e(RJi( zfg606H$)g(RUP=_D}|O9Ik`7z)azmqucNHHvK1Vd!9^cv5Kx!MtQ{t`pfCofy^@B8 z-nzsp3@dtFC(j}_Y@pV>u3zN~4<=nRFZb!WEt8&-eKCY_T_;Jg5WvV!{ zQ(awKd~%yktqOpMviNO0%$6pj>g=@NsiEA^+plta@X&A2Zm{JCcDsjy&xIY>6{k-4#9J5h`;PhmT*frddK(W=sCKvN7)JPmtT6$d z1upbkVLo)z&&}7e!(26gsOXShSx{w!$gM^5mSze6DyKtb2M96~5lj~&oVxM^QmjkXJ zzxt0t<#TUQe>==2*I+F4$=j;xMy=JUhizvV$sD5MAh_#9e`6d?>Xnc)pee=HFFR(K zk@~WlJB(u1)KRyPJAt{_^&0;8f|8qQvT+M1CzqD4+$@loGWmm;yEcgUX8GPi0_$qR z3T_b_yu_XI*+iC?$B#QZBsRvA%eDseV}vW(uPI}>VVnzDDu?Uku4}ZU_Eu65mKR1Z z)U@6u3S5*Tpb0GQ`;1!ozsbArvy%Y`t*_1pb}PfBSGH`For{sMaPXQesP>nkN_hBG zewdyIS%x(juNV%oDW1{X_jY^hc7D}f9(7Gq z@Do+96}uKB_kLq_*1zfC)TNt82tUqj8n4n9o!DypEF7PvMCgm`HrpnFh5sBhF+=p; z$}Uyb>1ub;R?sy>>|Sd&Cj4u^+R}jkI{1@Qj3tmclX+Qk8QQ)LvsGtWJ7Zaoua6F6 z4-zL6Cs~y@Lds0Nh6U98U+zPD`+Hu+N?JsKS}M+}3U>xJB)+ECBV`=o*tYWF5n^;62Khh3^TloJ!hW6+tmfnCs_ zs(ARRSwh29hX@)m8zfTYKZJ&I1zG(#NBZ5*156|DmFd$iw1vsVUiNX>;0K^+#j~yW zBHY{V6}@_A!h@)lat%)j7Y6Sve*s@N{m&E(K<3D5sCVxet4oP{!W6PTR6Hr*zR7Oa zXEb5*$DiCSW~Pbz2Y z)HxG#ZD@@zoV*Q@8dBKHfAu`@F|R zFI&!HyGcmj$-bn%DVBtBpXR#{<$oa~_B>tz9d0l-mgSKY$K+e&3d*on!OkjiP5Z!ysbhRex{y8#NZ~KIH~RcwUrGr-oVC z-3h<3Nt@<}yuBd^M$(wgH07FGq*-`IH|TZodLg>E&mV;QOGYoU-3kPprA8_c;DhF% zg>ZIZhWN8%R=p6-ZFoy{&PBE~dnOME8o%_JNDd#UD>CQNn$;1@4=#iRf!^qE@&YEM zE=DWd%-T}kL`CQ~1HJ=!?rdWqrFbkuYnVGA&ioufPe7gPc_l1ZcE;wdDvp>Nb=pS= zcaEm3LHkRS-sY91q7CU17Ob|$;3!!>s8Ta~T&8=BnAQPLmoy8phOjG9kIDBt01BB5 z!c_>%3r*k;hNwH!{E~l=ckhL@V~9{-9t@{38#LZD-K59A?~xdOxEg59u%jlC7Dd*{ zWw`?3!Rm>H>&CLl3oraWr+VC!ED7azV~9>Pp6yMqPj)fZ3CUdBTVF8zmSZ)bphhb- zoM!p@Q3<;TlCQPoYVyy^v>o-Tf*X=0g8J!#-bGAlcZb|ta;e8Hb^C7dbeDaU%o$lt z8gld-JI|8Ca2(Aj-lp7~c{CKqw9DQfG#_!aP@l?6kezik(a-7Fo96L_OTauD`gzq8 zfk38T#LCFTu0!d2jDTfjYzU|^;tU8TY~D6#romTq!>%PI9<|d@0ibn0^p#wf zkVRVSjg=B5-JuKIrwlhkoOH>-l~`QP5u2?uvlPQsPLyC!6W4gzWsKM>$= zF_Kl)T5R4O8^@Ik&?dW$XH`}|Oyg%W7A3)qGAC(~9Pq}EMzQFso*6lMQwG0B1TWUM1gV_; z^osS}o8celyS8p{@3s3U&o>;nL(@&My>7h zb435#Q~-p~RjTTakIDkmjk4;mg2!04q5}rRt$HRR+>QsnJh!EaIHw|m)>jnz(+9iy-+&Z zb;e_SiF20UK;O(AcTvS(o7yX(r*%My1uvnrkf5{QiW%pvf)4$H&|R}*E3&Mk1L;Dsne2RpLw?GSQexB$Z%K9`)Py7Dt-L=* zWbH(Wyc=;8O`a4eGq=f)*tV2`P!@?`n_9Jow*np#+(hk|48I0!H#9>+h-bRtv--W( zE{_^}mwQ>`+)v89N0ujQdw<&L53dMK-v@6slQ-iin!Iu&1Xdz+O9(9Dru^f=J`5If zRff}9m`pE^gJp&#ADNCZQKGbU|2WZOT6pQ@d3?nuo75!*Co)H?d)9Ya|-6whlodvUO?l6JFac2 z#ePgfW&+=v?|%jyT7xB+0dW;k2XTbv)e_WmR!M=*>BP@2rZgBCYwy$_+8F>B+j~dg z8`=Nw%LnuaL_gD}o4X#Nst4Dnexk>2XbSEfQmVzly9u z`5rkpH@g(xRLDJ`kVT8RLocO?)@c;|3DWFo0Wt_ zm$AK>W@@P0St1zdmAuv-an{{m&&vxXL@qvK4C9fO7fk@3xx3r7eTi1L&kNHLbKkp7Tv&hJ>mk zuldkd$2EOUiJ_zqHZ@Pldk<(|M?Gg|{xeH8tr%;VfbD*v@4uR67a)r-J?wU(Ed>&r zeD+Fam(mp0$jCl=mr=EI@ZlXJ3so%(jb^bXFFl$Xd$bpi??nWfJ~!SlFo75Sd5Fdx zHu?Ha2ym5~9=b=Vcq$S1y_U)~P%Eu|0HvXkeK|o;rSyeUa8wdd%1!xvaPj=9skd&{ zOMsxbm+c}^Y%Za19~I*MjE~tKeQ(Imwrs|Z>gK2;(-%Gba6|Z9xGZ-~rBZ@%?b{*- zm#C{y>?FZcea}x%m14`u!sbqAAixu1ez317YkEesJ~0&AUcEC1zkyo^^IaPpNCKhj zZV9F@MK3dxeT${&hNB{g7p?{l<~Y_$R=pp|Cktz}XE*E@uo(*ysP{;{_M>`2Myz@I%p0!X0~ZTX z1F=80o6H;PspY2?QMGiKytg=8ZNSMJH*8y`&mS{iTVxPYb} z+YgAB56k>g&27Q)^ZZgp^XylNr=}ZI@Q$in9z{7g!%GpPp=m6k66cPmjTOwbW=i$H zqUdINFr+AP!Po7_ngIZan*ShwTz-3`=@WKBBd%j5|M2p&z;zSTf=O5l`GRlVc4}mG z=KbgwYa-nw?Xk!Cg$Khr9^evDUx!O^p??^#HYBM^dA*Z}Lqr-toJzl-t29S0Qye%f z`%smAf90k3t~UGil{Np>yiNSH0;Pn^#>sJZ)Qc{}Rf5H$(>BA42Ck-68uTOXT zReRoGhTJJ)_c{Q-d;v@3njB25!%wY~fLl z6pTen=`m3YRB13;RFH2}6v>><%tU!_XD06=PgW-Gyj6Md+ahGYWu+0YhK-s{%87m1 zyGo}oog&htY*qnOir9h$^Bc3nBA`*;n5S+Vsv1w zKM7~rNy-kz-DEZXu-9oyjog#VztNjpQcN!(a zkWO5fnOM-&7s)cKF^e(LyP_e@$e?k&ljF9NAgP;u$?z#%lGhCL?DGx3S&4qx>c=Zf zcuYisX3T`?2FNLLl4HiAt*CdDKH|$YUynRFJVkp_Lek(yf@b1Kzkdcz5v@3fjH+8w zT<6zqD__>TCAlNm6CDe8;gA%CU$Csm)w~bf966M6N2RPUM%>Q3?hoIOJJf&RAbKhi z{QdgVeWjEJ$phNX^9H3g+q?-o7t~&Pqry3C!NvLaL7LR4?yM+YKDG6QOTn}vFZ?Gf zGkfPjS&e3_LS4n^j@##q_JV!~=%ZoduVkG1udEom=I+IAU$YN7{XA6@F*e!w0{7(R zhDKhEQ+{q-b8cQx%=Dck6<3F+DN2N1?!@mqhf8i;$%Awv$(;tcZU?oGijV~G6%rSo z+QWI^B=+>SNS2b12$wirt21$Puu;F%3#ME@W?eB{r7&ty@kO-z%K~HKz*Oq3`^)5& zREH1Kp6BAPkm!=-@C<3m0-22=dZSAt*@=kn)%=6$)5o_R-tG6|Kb;5o|3H+6z&utH z#lw*fImxF0-*Xd(RqgOrj>q)wWl1Naur@*(7IZ8#JOKo8Z-E)4goeCfTl+CP2i>Q= zu+C;_bh@k+R8BtD5pI^&{1V3N>}koVU$6b_a})rgtfDM+b@n|hrP0taXDUR4<6uvz zpReTA*H5L1yc1hdm(n@aO?k_unWSFXGK}iy+hf8e=jc}koS81H@%%y@=ioT<}Coxe*6o1+(gOWh!$JEO1IarO-XJhJWz z2VdWEHCk?od#hM#V=FHAIfVt6;s||hRC2<8YY|3`L(`);(55jxQ7w(TN?L|GB| z6xy3&?|+3W_jV}&eFYjS-9nN{b|l{iI>HQM+J*_(6S?uDbBQNQHS%M-6G6lKbFTU;Z>PDp@__ZmT5w*?bsmwL$D$Pr>pN^0F&HQA>%!Y~nxZdJ!hpm$A(fh*c z&#pu&;CqGU{h@QMlHHO@hayQ{e*8GKuRh#OCyM4MW?8Ncu_1#<)3-L7A)Be1(%6 zUOe5;ICK$yZ!yUAlxe#f+NK&YnvWH;ulojuf8IUfcTGAmrVn7l5D zX9-BP%D;`3wsaMDj*gH{D@%rjRSNPGA=Z1C*J=B7w4G32+9(no-T0 zI_DSnT#|;%%f|&vXMI(mz26v8M8CIf35Pc|J6TfC$%D6#>oKkVu5B^KiniH?H4{&6 zzX^(&ouw|zSa>4dcTEQHUb?TTCs23pIuup2bhd5_x9ZgCRMz-z&H#1Wch$DX1QoApOM4jMsrZk%r3>1S%HgeWC)` z*%_Q#*5NSjAcfeAw0UB}X5J~vp3L2qx8$L_AbQ~uP;r^!Oi zl!n)+`jco-<4yf74lbdT&T3N#bIWN%V|v5E8Sr$*C_d*9V5Q-7TU9SH9mC33Ol`N*z&A5qsF~4f=L;-+zH`9h3d~`TQ|L zb9gr@a&J|$G2z7bgR=zL&b<#VM+jP`t2v(1JCgmb8j(wrW1(@GcieoOBEZHEYyw&f zjj-O1cfB?7>6xXdn||8(Cy_!BgHw%M+IS{R+?T&B5B6q9F|~B=F%tx|ei$(yExR z(X2Q*1sF}Je@@~UQ@oH;`Ae3;QtwEvqZpP;gKK&V8!17ZUCJ(`CN^B89fe=ewPQ;M zQr#crb{<;@63Sz!2k~b>zvj;)?&^05NwfPl?$t`A@0@J!Qm3f_DQ*bmk@)aY!^v~z zoJyPDtB_4jX$us?p<>`J$Bg4bbN%IV(ER4+C>KAigV zJ91y-`i7AVrc~cVT6D0iM6Jbcs+~OYY_$DiA@jM5Ey)ma+wa9_m0>>+AvS0n7Eqyz ze2E=b48E0dyZ(8yZmD#=iZR z42ip)JT;~JF1&Y^l}pH8;?HX78%y4Dqt2@aZi^cGit3H~RJ-4NR&wgnb!rnoUDMyE zLv6mZ!Uq>W$Pgb+Z9YXnhuq-g=o~NeKf13*-5=|EN9TPDU68VDixKUnmdi9an{yp0aU{ zLBV2Hz4XsURhV>g(sp9O0PEcgsP&5xqTzy)AT`dI1p%Zm@Y-mzy- zi z<>@5ks~Szk18<;_i~R^X5VD~5gf7JNxvgkC^6BSeJa;O^laAizEuDnZ<#o(yBr()lU4RC&eO%PV`lU9_l}RARd!}K zr--Kvr$ois6M{+dWHV0>+kGMu(x`?P@obE}+0#=St6EU8X!2ye%*x~~l0Ga+=7gHk z@pLbI+W-It-dLbvcT^7sIs{I6Kqj;1oUpR3b&WVb!_0emf61P$%tWHNHcYhx$Gc?9 zWsr-C<;IR^aIAxQ4~B>D>Ia+HL{I3j*91gUS>W7&Px)A^&b+H>||%j+W(i9K2| zblwAKHBuC1U^v#2=u@;```e!M<5=vTsz4mj*gT-&vL#7DTrs6r)%)Ei` zBvs7Hr~DsLPUF0qG_fq;5dp>zo5M8;@pkf9SWFWoo_pvancxsmuH*qrn&Sm z75x+faad+BQGgqp#IW%dRf$Wxn#I!6E$-GSd=JmA;0d1y{vVgKOsVE_pZd((_|?Xy1Lx8FTko;VAf6S%YSUzMHwWr zl>2Z_2$qSFr^SW4xYXh|-z2nSwHG??*3zqR6Dwoyhm5CRtv9yBML(h%&Fr_@;7oeA zO4HNN>5*GgYxcO`7;5%Ks9&BvaV80F7h&pJxwjS60!v#!NkJT8r;^6%n5|@hV0jV0#a8^H)Ns#+X_U4Cf_NSOQ#Wf3NlM~5Qka9ET2~^Hx(2zJ&yy7P(X|a>S0%JpA!|&0F-Dc0+Co+7f$2-3qp7I}E2a5=Okyx6fk09>L zK7ASW`|?NsrVUUyMrYgycR$xl0H@b0-c$A`w1>9G^-Q;9mD@BvFRAw= zVZ;`M5Ex>XbjJzh0I#|LWmwwN6M2Nwc-{nYA5;)KAdLnR!%K zYy4bbw&K|HZKDFG<9LsK&~L|&cfXS@iEV7pJoLkE_yKId*Jl9~qO8l%|GdJCmILLj z)bID^f(6Q~2OW{D-sbzeEH>5C-to7EFN#DXjV%mwKN}8sN=V<#30vmGR82}l%4}ut zsb8D2(`A`WaKt1zraK6?Tk|uS+I#)H{Y*Q4?vWB!dCMSe-s=gGGs%f~+NlWqwDx^P zj%_Z)i^`Ipe|VCO2;#&``xp%=)Bt5d6VmG*fql{0@?Vw4mwR@4BFeeF3|1e(aYyN2 zE+wCHzQtQ*XU)n#bWZ1G<QeeF(Mw6O<{c8$XeI`yG@uF&yK^@J z;V8kd=264JnY7?_FY2^RAX6OZ4H+}k`N{JadC-#U4(}5zmFAvUE;I*zUD1+!YD4NP z)8zrI1pd}H=@JIAp0Eyw3P|4sARsUB04pVJd5P^AN2QZ6G=L@U$p58Si#KZ0#E9|n zTpN9}73tsyY1v|H@sUSloNcGdPtr|7F2c(LXH_}QUE{ta_Pq1uvv+ehAOYra&|Ej2E{5u#PB1{nm6N7TZdtf4wY&J`)+4%8)6~EpafG z3{BZDlxtZv(sWAp}rWzYalXsxGD!I_|QZSqUvbg!+Ts|1V7a?|s`fsY4QbL(IaJ z$eBf<=z9;}fA{w<_Lk+g1MED(@sN#D5_ku)r%of-t3wzomGif~y2OV&omOZnjFLJ{zImaGlmZ}Z8OSIiw<^bt|z8M=(bY?iFM?@t#p&vfVB13&pORE z&0({%gwGDNB7+KiKSwou*5kpgF9>q_jjxSq6NeLAmxS=TTH8?Tzhr(W0`UB)4BvtF zH!c5jiNQWyGthbrnH;w!%?y|P!KDgJJB;w7F94JER?hK*cPXB-Tp&hw?InZB6xS2O zv!tym5Vp9D#6B!=IOQ*{6vOJQZSX3=lV9CdvDDum^wlk1BUKO$IogSnmZctkXb-cT zgw0Fo%uX6wP>K5efY_Td!3MeIfpg6DFBu+dO0aigqC5S%hC11Skt9tm4GADcEM!3! z0MUI~H+bsBDZePC8hSSMNU_E^3{!`fxTx@ZpB>394>=wHMnLkQ_9mM7W!(IkNEd35 z<6Qssd;;AodstgCaJ3OPge6cs7m8S`w}l*SgN|~yF=l0!l@Rf~i_%675l2$@&<_BO z!ZtVlI!Ybe+Ads1`!ffAZp{r)X(NW0_UJ-TYZQQ4HUCf%VBC-<#{o-_$N`)ypIiXU zKF6?3q_#o%Mt}vDG(O9ZG!jh6PCv;`IFr%FKj!yhW}(Ky*fgt<4KZXdPGu9n}D2-)12HD>$|7# z;49Dv>~>yDGsqikhumkF?PoygOL;YING@!i4TjSZAd27=IHeeWz_tenC*fe-Tyu;e*^Uz#E8iK1MyW2Nt1X_NK7?Ehak5j&L zDZ`eVG`$EU^c+?{6GfI{q#?N^#($Fq8_0yZK%{{U^CgX__(dihhoVeB_XHAzi$`jU zBuZ`kt@Vmo*1d{@^mwXhU}ZM9=@`76T95Z-XQ)(7rVZ|q;(n{*E^c;po^$-BK1RMC z(lNejv8rN9;6bawvt_1Uk?(KK*UbcUObV6t}Hs zUI*dy<3hG7-T4535-WqzzBbsgmc$LyO)V~tF?FKI=KAAtZtpCwq_W+)SeeFVA< z|BmuhQ=3KO!CQW;=YkUx0wpk(2NSd}95kl@-gs>XIUE4Ds43_{aChWfRgg8-0qVvg zb??kF4M#Mt$NR%pXL$3pGLr{U>1lA`W$3FP}WUCZvIF>0my z^ItL-6;l#e1JweXG*F}K*rIK;i$@Xd8p+!NSSw%tak!^`Vt%=?p^fF0P}kJ^2-xqd z&f`5UcNj=W@e^3|j5-r#`ulSTiQ&l>fE1B$fct9#<;xMnps%P3#0BGA@4G5sX~BD= zQuYp2cR2y=2YorG7*6t11T+UGoY#G%W}KS! zq&w)X93P?+gdmFiH_pCxTDV=0$Q1yMQ_J_7opun`+h`0~O)OXfp7D#@0Qwfz<=0z& zL{H#5e_TVp=#9Pza)1!}-56)7E)oQ`7e5Dp6<*mfL9p?F$PH=ZaU%}zqK|Fh7nKa5 zWu^wU<>e|8Kg&mY8HUv9zQs@cB!QN|YyLoN|2k}|37B9WgvbJPg)fRyze?&YdkTz3 zQC;i+0Mv>PaE(AvY$~@h(kz-X<=*_+_!ZB=whlh8ladJMDKCbcU$u!Fco8C`Ys|F# z?0EnvpqHkbFd9Nyh@T_cWi%3?=lsT2S5 zYyFSG+jd{G7HwWH|A2Oq%>O+(M5X-&c@@F>oNqF(oiLtrzVTmEgAn=VCZVKoO+fDq z|3^Mro7uu>q;QdAwa&!vLlI#+Ra3rHU5#y(f!6jNB1W^rsEaG=yQWLr6t_1#2;S!c zpwrUVkR2WK3dDBDUk&WsOusfg6BDZ`e~`%e-B=~l$={0)OG#{b#cz1 zU&fz6GJX#_?$e(AcS8VE$^%J6y=-GF4z*RW2`3nbqw@;?b5FmBq8ITF1Q06|P%-~c zYMW)x-4hZiCQH%4dKOwR{8$V(1k}5IJ6Ey&Ci>qWsvo} zLVMV8DpauiTpSPyvvN*Siw=+~m-kb>h_&iXF9){VT=^Sgu0B|AE+913<`r%LMrxV( zZ-V05dGR4@6o4`gGp$^0^|$laRVS3^8leRm427rd=$~Ts4e#*=reqNN19ZKL7gCM` zp5Uo{KI)t3i4}YDGx4h2(9Sbwy>~v2?ofH=3!}nTy;Y=QZD);s4sjtM-dv>lJ+mzE z6RJq^edhf^r2jzqroc(@Y|k&d1ToIk9?cQs0iQ~DacNEx<4*o;M|;yxz_Cx8sbk9N|sOuwcivS)H^L9BrTG><)7ohXySfYbEWaMR+66s{Gyz}ajx|& z07nXRBD{S3Ys>G8pDJ(!NoVEDEfg0l(LVUfu=-v72r&qvAXvPP?V$t2S@JYP%cMWY zOZ~+@8L4$dxzn`bKIzBDyeA3)da#zk(dy5~P^~CQEs`j#r4Qg8-l~S3^ui(os{YKU z(wwp|CQZANLO=wa$o$qBN=s@{8gZu!9|d0T2jMH}uKTR-2)RWYXAf*Y$~mxC42)Pk z{k;_72}Y-Wv~rlT=zOEQ6;M7;b8F@X>CedX_4X3$Gu`{*Uk)9XQRSk0B5=atiwhow zq^Q=$4R^@nYfTbs(Md4k!-Tt7izt+7BV>M?f-03t)yb+#?t-r>@OLxY2MQAX#S_3J zNdJ$C*%hDeihtx@Hx|0Tm+a@r4@8T|l@?}j^0azFF#DsK+Grg$V0B%${Dry{Xx6{3bbavnMaOs1#jz+CL* zpBxJUM8>j=b0b2sGWxv#Vx*kTPPB?3zp0)*mwMe(1!GDAl z)AD)M2kA0M?nkXQ^#PE@mMLI^!_S%20RfecIA>fe*O1f@Jd=t_zmy)6*P-BYDH5^{v~q*EDdSG!eJRT z<6p89Ad|;(mVM6nZ^wmPl;0`N0%9Wv%fQ14kXPrI%2xxN#6 zwvK}5NT=WM&O43Uh)0_(A~`^)`9ukpoP03DDi9?g-Qn-CgEHq<*nSBa@%zx79#BrSjkU7XOytbs+MF-`h8{;9_}O&OccI=Jf^3Cw|dk$aDhWJ(ga(D3&()b^U(L z$h=y5gCejDottaw)N4K;lD_ci87y8D+&R;={ui@~u7z(kKoFC|Kb*ZBzrXY<0?uYX zPnYx8UU?hSijRFZg*6^{|6>8(l6LHksd zfTQYGFx84x`rj9&viWkS8Y#tNNtM=U(B2G5%n?>D%68rg0sNk95^F=pBtS*P9rBlK zx4-kBE0mG5M~dnFOLngpLZo;dL=?&O7~u!{K(O;%sDlZ@o_Sx)lXsc26Ij2(Rn@*Dz1%Joh`Yk!e&!E zK(9nXPIu1x`~TAMe@gw~+P7FO#3!Q4nANc1RL_H|=3F~(*i_Ep9shdu7bXSLbPH|g z)z~+)Q!Ye6+V{c<6|&8sk!6lzUihlEi{3uGMhOyRbM6*G&}q`HLeo#}ljwE-u^wj@ zxUp-Z4Qd;Z{$P8A0IU6AS20koe$Ns<8otQ56cqK*8VW+s8fG-(lw74wox;;#XGf>~Y z0jWoB(dAEVLuvuKs^9Y7yWcY9M<2c|%7(PmW@9OMf-W+-cEaWCZb2-IH9Cq~Bd1Tw z(qC5xWnazKIwBjGjNUTjqijPt7cZ4b>Ienp^p%5hOBb%Ltt7H zg&6g!PO-uC67%XvyKpN@UfZ20@js|4*f$~*Yp0WCo^iIjVd69#(erziXy@;H9ETx3 zU+T@ZeR^^+stI3NX?U`=`LfZVm7u^AnQuUM>i!KjC8WVS3`F@~^I7zKARvlFllx2K z{l!p(v0_*xOB=*v0DvMl0$xJ9u4S=4wg6Ha;eS&XR3JyIHZkbWn4LKCqY3CChIldD zsRnrmBMFu%o~sa|WUr;>rkDIv1l^)S_?OVcsiolM|Pa>2DvjQ?F+NC)io`4q^$5g2{W1So^yWP=Mf8q*ZWo|?iGz5&q3 zSQUde04nF)Y72H?u~3Wm!0#uPSJ*X{aCT2gE?p2fp~|_l}K(PWwZ5I zl~~gJnN@Qfp;a&#a(>O0(K(qz&=NR~CIL>hkvub42DZ%OIWHPxqPjz}rKN^1ldxkILhT+}PKE87Gjv>ddAw~hxXXKL;sTn1 z54>B(tZsziqfSn$e-}X`f$j;2+dJ6?#G0F9Zdd;=*&{?P0H@VaNBYT}s}wq670J)X zVO_c&toqr!C~}CxL+$+~z$Hp`NqhKy^>TxLsme?BnxisK6U~OtM~IPPMY#h?2z_rW z-EP5nsiz`Ad+P2XO}%@#NJxh843(N2phB57OqSiLc3hGW1~P<}Oi z*<=~VDj)R2fsA8p<4mTa>fR+&ZmqjxXdMaPbQg6kX`%EI6*Ed@l)r~i zPOAKjS=T$VvA&f7ihR?LbTu-GU3kIGFqn%;Y6xNYk&~rfX?-7`r8;}J3UqjuenyTd z59q`15LEL2l3~07hG;1~U3i3H>+$sJg%ISB3GwkF@syzW?_EG?pRxPv=nL9Q1=z2o zcw1VGvT*Qng^ZVO=Ov!#B9@f3x*61V?k^*4t9)O8UlPph_$g>gUmS?>1&`zSVoicd z{8!E}H_8ry--E4!%C7lChom7+=imK{zYUwua4(v~X17+)*k(LYyxs|RoE2)aW$-Ao z(^FU_qdxZ;GsOX$e3&w5OX=t&H{2+HZsZk( z&zW={bZP0t)`pv6Z?`cee4X3SwPKwA8XXMbodW`e8bB>oOyIkNemiC|kpxkp7?n z5aI(wR`II(7H2Ke>jaBMHrPmy3Po<%$o|MMkI%Ve zF{3_JrKP*8`7Gdofj{RlxwKZ@_8WOgi9!!HtK4|br@xQ#gm>(Wb0Y;?8b_1v7tz@n zaU2pzN3xan{Po|D+PTO-u)q`^q*NL4zv9FSVIz^8BQ{hlNmOyDuW%z>zalTL#sd|J z=D>@ML#0!LwWv<%nmdODhNZ9N6~;2cc2)Ch44~CxE1!6Krj!wIl+C&{Dni_8;!&M! z%42EF1Wkk(-y#C5?1v7Q`m*>e^DAueOL4}}rVK(0`dzITXS7_|wH49>MZ4{7^XcrB zRZx)MHqOh5|L9vDy>}*UJ9EeNU6XVi3ZyL4(&}nICa;~+P-iT&3W7D>KF4WtjFO0z6Ii`E__#$MrM(=k#W{)y8 z)1-03I{L+-bFFO8f&)KB-syI^atadLDr7FrrLS^0W@fY*TnfZ6_}*X{chtI~(|9{I z8{9jwp~Cro(QAXxoa(#n0ZnpwKhQJ*9o8K-|1RB*JeP?fGN>^pe$CsOAq=BUjg1;> z*Y4{&Qb9BsA*USu#v_l8wC7JrV^O$%#N0S1?H7c*A zF$N09XEd0Z0v27##^@N?xzytM-2Wo&yP}%v_I*)Alq%9|6cAJdM5?qXND~36p$3&s z=)DAD1?d6;5}MLm66w823rLj~dY4`jYJiaMuD$m;R;-MJQE4geR>>uM*BZ*EE4>) zLi`lKVi^w@kJ9b}ck`JHUV$H}u|VF3JdDwUe(fs2Jf|@KuL4NrJOJnM{moCDL;=W= z%D)g2A=i>aV>i}%$ZE|tcYq67d~;VmgT@X3|D%E~XouaAuM;h>i;9gJ1cp?>?&6b4 znO!QuNA?NpxUwN!1wdrmVNQK-h4^8?(giCLKB);Ebe)PO$bM%;O zyqhw2|L30_ht!#d?}N$d6|#AOG3Y!DHt3JDxmrR)ceBM5niHbvu#f6@o;%a-Z3qr_ z=Ra!o&3G#yl(#u(!m?ed0A?iCuUS>!)XTLrH?KY4uK^%J8ba1oPTLvybEAN_$$e>FPkw^VvhADr66@cl;52Q;Js9Eb7jm4q1r`hnoxq z1Pl;6V$lx^EZrFCEH|~abt(!cj5u!rb>DBR7L$+rjv zJ>B@q;s`luHZ(&99VT@fB9XT$^Nm@R@fAENLY4NZH_yfZwD{p)?TV{h*=6 zNs8|nbVdW2P5y_%7q+N31taU~Bjx~HZOEWu9q!m=&}eq{+LK`#&@Qb(eV0CCzfEAY zT3mnDVWbIySzajF^&&|>#$`> zxShP@O8jxE34`yg)oSIS0Nuqn^CovUvnTl)e5;fk*vr)qcF5+ZH1SOaNJx z=fU)2VX8e-l{mfRT6RM>)n4U$s>cDZzFZ8A2?J{0>racFJ)f)l(_;@ytl4s7t>ouM zFOJ~=pRnFcWBH_9_u9o$K2W(bvhXI4mS;>X6>y5;fpaz-|Kb;0GTC$V~ zrabo?fHmSRLQh%W@!*ou%+zv&h~R*udB*ln&kuK)Jh?nGMZ$(qe~hmmqNksxyJ25> zT4x$X)g2lQZ~7yx$~m55rP<2z@;V#d*O@HDIXC2x{zQA5_`WV5>*jffTuyo{kdrl6 zcDi`3Y9<>-zC)`iz~~AFt6EBrzkecl&=3zTIM>InkW5Bq(DRRY=VLq{{Y=EnVW3EO zjQ#Q}A+;rR$WpVc*EZXgyg1TjmqjCbhukBF{SwOnX2s z#HlmdA}d`g))}EY z=i(TWIj;h_Sr}MnTN~vvb4=C^ihb4uA0><4TwB`y7bU8ZXACuu^^HqJ3|u3%P)-Wt zGP$9uFC2JFG(n!MzqVp(Lj1XYP4gtR!VKUGFDZTGPI{1s{h~6{XxKaawnAi9&>^?s z-!>`=vjjD0BTjVLoee~{@u-I-Osi!#nhsN$IDzQ|}6>hxVM z!7y%OndUF6wD@5Gq^=o7^!J$q5d_u^dXC>4_1l^*=rO3TfOZMiECFf1{Q2l{DCq|M zzdEo--MRk1Hi$SJTTR<(qhztdwnd)yj=Zs)LifcAIL!f{A7C2kOtZc&?(!G!(4n1i zFV`k}i|L}~g^MTmds*yhI+=B-WWF*uu5zMlZtTD}X({IEwIfU*UT1IWfl82=b3KR40J01Yu&MlR4M5TZ zJLpWb#QA@GqlhJxk~{#Ih2^1$iO0aWW@~(?1m8PX1pGc5!1(rR4d6}VaewGfG{%c4 z3{Vn9_swikI=NsULuU%h5%V;Xkpt@k$FDMfble-|uu?)gO63940-#EXmF38IvvP{Z z9ECI)jryD}k^NsIcw*)b8~yToqt}|$OWEa0PwUwt*cD>$>*R$?(b>Jy! zu;{X42;}eA5IzEq7_sgR#PNJPzpa;AhNEKBPWefNr9(M*pYf7L_Wd4Tz-3+R_H*aGL zZoO%aG-jv;B=qB~F{J^*dSiRH>N97}O+y!yB=;=m?6746M2#hbXVq-=UmWd|E}3!X zfbBude~PWbo*k#lgrsV6mp4IXunbKvzd7@R)+6E^bEv%|J z>k<>@8dWQtPSl-lK(jb`orsh9qpz8+v0LR6syAbu?5UL(vVgfec0Agub_2Ua`Lf9fMF~{P$@0nnI)!<4g1(P zL^NiAknYp(6C~D{^5n9|L}|aln2jGho-+%{6!stHv!#@G}Jx@r*P zkJJHb^8;Ap6jacAqA`THtw#Hw3*x!zq6XoR{5a@mcI5r;|E@^~i>zaUh(`-b!hkC1 ztq(m|JBqWxMzo!A?!}z-oV(s@D6Ms$#EHqb3LvX zNxHa*FskWveR&BWBb^Y3aMqLESem=btc{_}eI1A!Tgz!u6+P!U#-ghVfw^^QCoVxI zhE!$00)AeNKiIManr6sysHf>ql%n;>ky|%L96K9kFU9H6T0IsVEZtSU zrD{{mQAj(eETZBk{hxHk@h0XasfSQF{itYxvE$Ut!`X>w zu7e%786cgmvOIiGeK8G>E;vpcO=$;HiK5>8LlKvDY#P3zcZlh!jVvyT)sMx2t$ObL zPG+ixe}b_sa6cv+SGYA88C!6WYU6`rTg*pe`6Mm(L&8yApoW*-Eut3Fk+xa)6HV`C z4?ZG%N*zC1FMaaHhVxWCXq^4y5twNT81A%&%wEMxi7TS`o)_L?K8?3x{Oo=;jX*PU z<((8y{L$qm?S;7%heF~ABw$HB(5KE7NBP*b8}UVYZ{e@4_jqiI#I%OR$7*-w14wW5KR)OrNYUw^u-$M3|2@*t;V-yeO z6}KfS&%B|b@*_#9^I=ghU+nT<{2)1f-}G45P>IrS`LZfZ|8{GOR8pLHoJq(JFEJ+u zA!yqjP;L{z=e;rD!u}l4I{z;hxJL)VTR3PhyBDYw|a@KX?l8h`@$bv=S z3Eg)5*9CGG3Vtk^UmjFM=|R@!--f=C{-yDWE^7rWmfOTBet8ZA+V`tIwHGKqGfrYbX3&0_~+jo z$YLVDihjrD-IE%;)Mj_qiVm=vJ4!q@L$Uv^<-P-@Srar2 zy`?Q227kR2X1kv(sx|gxBLg^JV$99ICRhcsiy?5&P-hTdZTWlc-q^x)tI7CVTSoZ? ztxJ_cPdo{`7P!^6!pV_6)tRyiTjjo8X8c28%?t?0?3+sj0QCp7^gtr3 zQwiq!=xJnfP9iGgdY57Pj-e^n*A@Yqxq3GdQcDhc;mEmR&1i8U z+vf)Cs3>TCvshL6M+)yqyw&A7m>-R;9o3~%SZ?_C{QhLF@={Wq9wXFfJiJJQAcpOu z(H%*LSbryGD(o~HCKMn~hmQ>V`WsO{8DcD;)I^l=55DAM`_3LRh;g~ zlPl^A@wHzZ@$BzN(t$-!T|Ma!4ZzB8em})xsLu?=@&z&+L!1chZ2~We2yX<84_~%R z*A&EsXm@?X$V(oSFq_nFF~M)E?@N+YU;hFTj{;?5l-Z`7pv=k3PYKW_y8EdKOdjYg z-KWJ_c)eL~{ivnz&yC>Fes7z?$sQhNxpvS&q(j!{w7(VLrzC}9gd0gTlKc{~gM`|j z0lAeQaglK5Sim7ZKVkJZCg!kAWq$5s*}XGJg$^B*l_>kbpT6q#NHhYO4oL~^tBEHz zEE)wB{zGBv2Az#N|M-ryX*GCz-B*@krwf)k#^zGZpyn@Y@{T7iXs~wV=8JUO(ZCdT z?{^hu0<|W>SGbNbVN>DXS{(!lP>Uk-(gz+A{k{kW-%D2rYdBWT0gc*m@fgI44i2%^ zw*2ofcoiUyXD++^|D9M0If;zt3d>|>?LOcT^iTb=dFhJ)ie6c(jy~XMZKcJgM2Z|Ng%Bv#|#%od@0q zHomGL0YV9PHOk&d{ZZ12p(!#}eP*JVYpxEqxFZ%kjh&=ZRbDxxi4$~m+y}audh2nO z=l770F|Q)ZPPC(OgUYuEyDiHogQu^Cr_IAUJli~;97#i7;*%R}9FFVg5;fa}y}hlN ze$0kTu7}7vl~_R4l?Odpn#k%`06R&~0dq1Y?g9L1k>iQB_mzEBL5Ed?$*avDi7)~e zBb)iDWR@6#x{q2Nv@&jdcOOOnW;f_^X6=?~doM9$)5EH9#H_@d zzi!i+?R>OC%(h=)Pt@Rv`n(SgYhCHnHor`@(s%vZm&dI=i%-c($9&2*hyNlahPanm zhvUB(Py`I(BZqrvQQ__9Dgmyckjh5-cLRgYP3$}Li(;~QOCdHwX$&#WM$GQCbc_p} z8zxIb+tYF7s>Ci4@u#s9~@I>t8qEXikD|+SPno`LZOmega8JPV{;Lp5z6VKqnqpZp;CbFSa>T ze|kxp1r=nu+eW?jrzu&;Vz&s8?RYq6b^Q*_G`x^}u=W#`p!Re#BqD#q+tRh7;&W9- z`Q?GDEJ4Tvz0X`nP2;)a@f;H!o^=c`P?t4pqw12Vk`$1yjXz4fYp>Ip3;)mqjtMLc z<*J2+`QfNj(=!(0pGzbvQA_jccKWP3U3#4{cD2ggDaSl8<ouIJ>q_}K02GKT$!!Fk8c8Oz6ESFbVB zzSmVwN-K08A2Cnu6VvH}HJIvRGxx3yf0WI-f zO~g6zVdC&A+Xoc4FogOc*Wq1z=C%a!Sz@^^P1lZexV z*<&(%2$sL*`p^C2AJm&&YDvg6*GCdW$7S}uuH{v zr=3>}eDby8maAxURO3!}Th3@9)K>Gc^L}xR7yi~4R3#rvo&mGdz0dz+?0Q*p#z+d6 z+B-Xq`-ei6wF@L424n?{n|UB zQX!HC@7zS~D^uhk|6AnZPm9&NmE4m$i!l^uDP0{Bp_ z?@g$K6qb%l8D0N7Q@p?abIC&7gmg6tIyQLm1aKw|M+Ket4oq8U*bJaZ+DIoQnX_fK zSnp`MajzLecQNpvtPMlK7iI^Du01Ag2a(r5!56-5Brw5nPN&!H*76YwF?FdW@bitk;&mTm{ZQg}9=C0)vN6D<7tsjfp5! z>YXq4XeRZ7Wa0;tgk!?ApE~STcmL=g0ZxvxU8kxS6JB*&(1!Deq^F6k|4?wkl>+u% zs*u(dI=gYP;D;jJkS3=Mwv+hUQPJ!-)o$e}+t(h0(D(0@RXz6Vkvyzes%g^!cR&?K zj(YeZF*i=*e-7wz>#aDskZGNQmQMM**>SHdZ?E^hD=jtAN8QxY1os_98S-%NMjGgXN@TuKcsV}T;=X3kJ=OyUY(^klgRrmg z_c6mb?36;}p5b4o3@K;t1A)UFquPre%K1}VCX5-RxHevYVW9J~=JW{7te>_9Hb37i zc?MTwEm21&c>E}K4K@IDy|5yNE3o*ZVspjs#cR7K=-@8MpSc;)TALLtmal4c>3*=h zo8}SzP$}=37Jgu#yC16En{?SMO8{)1cW(8#t`fhZ$vb36hGsAlN6D8eYDX3IFIzmz zXQ32m3ejVT%5$DFpSIp8>;Fj8<_nB2?GnT>9t$t9Lz^K2@)9hGx{Oyi^GEqo_HtmV zNsx1KqM`MBvj_ZEVvAvrUXB;%I6!R6M83HNraz*jB8do6Us!b@e4D7plN8&et>R-| zNM+Z>GUV9NF$5Wu0vBVKbX3p5#4s~y9St=%8>54l=a;FEFmBOCZzjJY zLpOF}mhPf49KEkS>}LW||c`rJe z3x7xYPSHe>a4?_mxjkIDVknfSv#@&d2&G}bZ1v&;4fe_o-%M!IvbSX_%5`9dHR3v7 zI;4<13%+r?VH_^GSoKCTW|uPrIMU)Upqjc^ImP~o1OA9=K;R3T`awHfw`jvIqM7hU zvxlSK!7^k-w{|RN=BbBrRbpUC=K!zXaIN!`r6_y$h%Bb2K%pvML2rp29c{zT-DnXI8}73q362 ze)c=_Z%ZU&LpyK_M~7L!b~}V9pC`p%{5M0@-@+uGysTFJ#5*bp$h(FNgUa_v3H1NX zP_36c?!PuTa7c?H-`i$o8Eq&km+gh_rFQ7opXb-bm^GW6;k13`@1dc0LU&a4g><2i>;3W-OYq-20Rzx^d+cfkJJk-&WsS3#XJM2K&(vM_xOXAqyjE42| z@hXpNtt{TS37bUNgWj2dnS6N7q`577{pn(^WJa^!MJPgIcGIkyBgCmEcUWtrY>Kcp zJJwKg4_Ri!2f_S+-JQp;!TueMxjX5E};Y;*;E{tJm5=X68m*qs;H7K4*vo8n>CH>C$7r|3-1{GveNrKz5G zW;=(q?ioDvD?U^_DrJnzVy&EiPiFrkmBA$H?qC*3_xtY?`r35w^ra7p^Cima(eqOv zEG-HyQHx(qEw-54@_!2ItsS_j97s0q3mpiR!ms_B5C50Welut$C}uDHL{vJPH?qHn z6N7BDBV5)4{&`_FfwFsU411#Lte#NM0%En6`|Pl~Wi?AtC$)2HCrWQLi($)LFE()G zh}&zYJJzJV+Bw8fyLM#dx+ig@Hlbz#qLx{V^p{mg=;sdNYKq#dyufU*@f8ZnIr3~|fVLr>V9ie+Di$Wz`K496XGxYsML3aEHl5 za`X*YM7p7FOlPHfvWo{)WH?$o!VS zgg@>k+_*8_eaEQFo9z8u`GE0b+QDS*_{y3e-RNcuk#fUTCxLls zaoSD!%&4m+E)fr|vh=;f=6M`A%Cf-}_vff)Nu2#bCh@t>2>K(r9j=psKX#8_{x7Ju9nMT%8+E=lxr z_(eOUA7qUjp)VAf-KN$uX|S&ge&_$q_<6lsjs&YGtbIhmZHv%7-n^v-<=AUA-0$yl zaGoK|*D{5!&ukW?j3$iY|BhuFpsSz8H9B|{inAW|8d6VwV|z>=4Qy{Az*#cvuw+6l zsWoQx^N!i5O{g5Xit)2DA2z?@;S)%qbaa_#QmOBQdXK0&LiLYJRZU3#=~SNgu%AS0 zuh7{*>XV!UFhBKCDS~yPlxwxPY1qD%1mv-9*3ay=EC``)5-sMzhgV z%W_snp3a#~r*Xsrr9m<-7A#w{&nzD^f9a_^mD<#s@0MiG8O~9f|7DqFeb+DDc&d%L z%V2MK33Lc>7owJSKAD1PeLtfSBTo>jma}FL07J+XR>V`lXtnw;s8#}iYTpzZUr>`( zk^UEOVA%Q*ITc3`*o+ndJkp_Y`(VA4-MH^$l}PsgT*zO;+v1itNbX~qsB zZn7gM9fcJXOgoQ1nu4~b$dse8k!szqf+0Qg6&6&)R8GN5$6DS%A+3K4C)n}IPs#O6 zu2ksO*V-1m-K>9(4s-Bm5TB7~?K^jOOb5SqD|UUt5s#wQ{yaus&0d z5cTU!?>))vPG(|gjl{#bX2fM6U!8QP*>nkK3qyZkVGY zwW*z&*Wyhe**OP99+W)iiD*iBk{u~hx&(gkargnB-nSPPH7>L9S(b((UdwH=ep{i+ ze;iNP8jvub_?p5iR~6C9grhyvS1X<)L+Q8!ri)|&LtLIj(bn+)HO^$WypB+8i83!6 z0556ClD}HuW>CIZOI(|EnaDP=q&VyBJ#ezlS-@1~)=ryafuWgzjPgxhP2OJ+EqPOx zyhm@K?E}3@_1MvYcN;!WSeBGz+JyCz`L4z;C&nS0YL+r%#gQsO$o&hM0Z}2t0R~uV z4{y9~S3%4XNUSwxzWqXEW&U$YjP&wB7i^u85B3^0K^Y%;n$szp8M6I>b7j12UZtvT z5KG&0%b3$E^yO78<)ji+fQ)NspLvQBKJsxBGu?H!7`>eq)r%LDn6|uo3dP8T>L9NiH7-HOqS5TBy0!V6a9)S(iH#I< zOEJ;LB(qhwnA@D>usYh1H=8&u~#kyK{`KKE?TDVY$;Javp+xes9^iUVl}RrZ3aCo;kC($Xbc2>m)<}vD7_h2%%S>y zuT?1Z{KiJ8-beUTJAOn?R4eu9WpGz>iZxBijg~+K$eri1V|-)x%Omv_9W?L_PF&ve z&ca?@kF*YX0NjmqdXcSsN1MQmtx0&xHas@Pp;1W9i5b-cIUSFmiFQ^dTW^=MHwKU$4IW zu(alPUv~j_mNAZXGSTNz zt(mlOFYz5rbJNd~Yd9|7P;LuKbh=tzao&fW=O)EW}-vY z@FY&Ycb?FQA}1556n zX+;7z@ndy(Jw4zzj(Zug#yzm>O@t&7KLDjW#9H1nTp>wdlsf!{+Q36ZVQ~Soe1b{2 z^5C#C=Y-6c(!xtv=~ZPV{EEDxR?XY0ngX~(BsXhxPc+Q|V+7)6c5|3YxtlP+c0oTH zSY-O3J)nTjb<(St_^UHuH4l!PQ!LLopHUqoT{CWBCg!cAAM9DKruiRZZkYCGs`Y6A zo!6|a$jT7iQbB8}CdM%(1EQ1*b<2Tt7qq`FPHr`#Do$D=bK-)^8`<1gAM_P0*+Ofr z4{fx$GP(E~U!I}OwTGcse)M)fKssC|hj;W;_)d%$r1@5Y7ANn%GcJ5tmMf;v?_#?9 z5+)97&-B;%!Fgkg_=>;dS@42&mQ_|{e64`UpBr?Wvfe0!K*IbcT8641*F)A^YU zPgN9NyH1_wD`}-&K&5P%;2rnnCJdkM>JBqE2pRly*HAyQ%fD23i=1*g z6RXKL#f;hSnh?1h)^g)l7I>loBBPkU*^m(x1Z|?v&sxp(^L+a~HbY6|H%YB9&=cNd z`Yv`p0J6cdQ6j4M)je5*p76x=4;nQ73Y-xLZqH)Pq9dt{plh~c@*no>mTG{$v2KA` zt|w}QF)k&zki!}oJ7mC<=B1x>d7pQ4&!d8`7mkz*tKIS5zEGL7_$+1@qs!_~=MP%J z#ROz^$q$5k*150++bjOu6e%WZ5NHsr0aiBkr*6yEBL08_G2Mg|MDw8U&dXo1y6S51 z(Zzm4!;Ov6iz)SjHcoWa#&-VCVAH?JYAWa_U1}jsc$~^;DazV+Hu^=2h^x!{C&!qC zW}bDKo6a;hS)=GrayV9-vx$hs%H$0;E$yxu*IB5Lgh*u&3%g9sePS=r652#q&04gk zu0l=P9fwsbomEG|J}TV5TJM(U!f-$tLEZd#aMKyV zyJ2IF*V*|k5}0KhsdjArmFaY;xjZ`1$t+78S|By;5HTN6Mrvb8_buH}G`_cI>@GDf z*7=TS?7GxkQ^}p@;@--((6kPq5T~}-0M<8qU@SB;XoD;s&*Q3zxpw<8{~>bvVQZm< z*%u$V;%2SACt+1UYJy6LTJxA-V~l`JtH;ctq%$P$ z{1Ku@V5H^WlrSeIyETtXOQ5GyVN}W2JVNl)nbld)J-O+s7Yj!4RL5(oGOfa5>LZXa zg#4HmW#Up1k6W4HfJlJ)AD-b<(K3>zY>oXom!^sEkoi1ckGDb5k$(@F_EEkjO|YkY z^yivzq0y?-WddKE3(lP`)T&S7y#;0Mg~%^B$lV_l`QXD;@AX){v_6%lIP0Jby2_f_ z4^xE!?QPX5bG$7gPS-PR#7fKFPM!@JgCiu@(t_WXGI)gC@9$q-@1}mtQMyikQS=g> z8^+R;F{YPtQe=5nELr(H!JO?&g5Sy!YL@06Y9&@~mrE2&^<`nHE>LhRb#S+9t_w3E zgU)-7O&(<7rBTFYI~!~QP1&OP{Y5i+HjU5uSN!I|GE zpA7u^<(Iebt((}tnGH_JyzZcsPh=TlTHBdoa=r1i3a4M#mG_rQQL0_C$c*}^o|W2o zufBZ=g$VTEqIlbROt5-=x>~aV=bIM}tGAXK-3{2~mCL*fpg(Hr-mAxIn!b9^D8+eR zb0Cv5meU+-dTr`CvLC;j(2rKnB}Gn$uUTodxnG?ospG1|8L=MK$e5$DMZ4DgaB68Dxl*5TB`BA}(@)y%@way~< zMOt6icOPL8grr9EwNIThrpl%V#!WwAzp4tmvk)hBMarQNC5^?{?@wa%3WyN@W$RPs zs*!&vbeVxgSdHPqQyr*ELA#oQI}HNosK@7VdVnoOt;$d`Y8<=ZzP<0c{9jJ3*yeC2 zSuqh{Jl^_`d+Xoh@yA&e*=H_fiNivuFUOz7}*lI$Ze9wb#ToX4stDYW^af1(_XRNncXrb1!An66IdO`#?&8G z?oQGsuYH*_>v8Aw>JatbZneBjZ8+Y*`~1N5>rZ*x_;^dj)R^nw7bZ3JEH<#r61=|w zTM#xG5y#(jJR9Y+5qTQ5IEzaxo~k*21sP%6-!*pMqgg89b#pS8S4$DF>6TiWmz8GG z4inW)6df*?3yQPn2 z37$yU2X^yMc>%pA<8qqIiaRW{McFlRrNlL}dZSsYV$oVy{(a%5iXB2cvyvhIQ z`sKxd)-D9s3S5{6Y5j4RE_sA|tLP+odII-q>Y_S)G5A})u|=4{6Z>D-(@t-%nI@IF z<-Ta9k%A_3vu5*ymg~gDBP+f0#|y^XUU>{!D-f>}rncYZn(^Q#JPV^rlm`3#dy$B6 z87ty@Ga)01P6o+a-J2qc&}Uk-f^Pp%%o-+|kR+u%`a}00&OLvNezIdaIx)p(KmR$u z@|^P8_@%~Ar$#}z3|M??z0!;qfA5`96vq@$Ha8Sdzt>w1iB)@)O8vW(;p+9lb&>Ef z>Fe-bkw8wXjGCWk1T_VbpDB&-bJWIO-X?^tufb2E^|+{#J3?w!W&Bw314|^r9Pn^O z&rwNW+3H**xr%c)6>ORUr8`{RCQXzrV%nOnjAq;58?s!@=l-E^b2=95uYb&M0KUJm zdplh1Puuh@SOg?iG7>-ku346AJe)60;U9`T3wQ~6C%ANAXYn+T+J|tDa(b@O{I`Bm z9OlCAQx8^-`IJN_=8(^t z6^s7AKYR!0oIQU@dSi?iPyd2W`Do5>qqmk2lVGFj;ZS~ONh?bYS8gYRTTUEzg$F(^ zA4(q?2F$Asl0Lb51I6l}VC+BE#`V%So5%W};}mkFvEnh6Z`#R1)mhPPr8O}=kFxgJ z%Bjf>G=>wuuh0Bdz76F#iQkGF!bLbEqc|<7c>w$KRtJ~e;%nZE#thdd%G;>I>;BkVMOXkX>=!dbAhd3D+r&#Aqy{%Co_QSd=_p7k+8)|D-v*z7p zm>*q0uJ_7(gc$X)Yg2ku_3pSvtntZ}k?IsV32OP4pyri}cQzK`qr0kZj>gK?9gueI zJ!=DUJKsYD1F_er?8{E%(rDr-ipid(S>Q;rt^Qqo{UyDAkEgGOCVX$+2l9&K@g|I!o>q`)lTs0>3Kl0VzO4Qj+DGOk2kT~?-9UF8XfxZM;bd2$cta=UD0Dal z(-Ak&FEkohNCVsAO(KDyAgo_QsP{)u@Wwk!hNe=fsgL{2Z5a{3p2D9h8!IZhJxMhq zm_*54w=NQ&Vp(1LrNZjjb?{ti*+fHI_tB#nmcY|AY4eTT6c6-=c4Wn`LW`d|e=Tl7 z1l)_vY{0OOAb)g2tNeJlJ8D`hB*dlc>PqEzhAM(-apG62_+yrh`(88Waf@zE(%4V8 z;+O7w&JWj{5U467)V~P$&v=M@z&ms@Qci8ecPMJ)sec>5)m!3pfRq*M)~2Hirf@H} zgyyRO=H-d5q0|_zpW0skP=u4z9EzRgbFcL(CU+WX#ihiLY?(sl*sB52I#uzWX0@d9 zz_vnY)v)8SJVTjM)$B+89@Z^~{%UwmYzlBpS5>Y@_!UBqdxxw!%e7;{qkZU^e`qN821zY zVy-?cvv*9}2G6?2Css}VQ&CnMVb<5N$M+FZo}KYXICtQ~BCXP=@^ar<=w9BeSPt!V z-8(x_N&KL;Uo6M*;xySDikEC^Guww4^Mc%W zP^I7q#i#DsX1}{|I&r}OSTYMP!dw|cC5%tEMR@K%_?R^`{6qaXQ4VIJuOsk>BqlhT z0!@EcrLSo?<7vhI$=4_$Am-*{;~re*6UTyO5cR#=^>EPxaM(;fCba^N`50>&k_Fzq z3lhpffM(o@{o&Rp5Fg&-&4(BluhAQ*e|T7zSs+`b4xbZ>(6b!7r>M`SD~IOa$#n=iISx=bKt8rO8-p&#vHpy z1+OFkrq`o6#05h$4L*%+G~IuH0KE6Z+9z~Co}89Xx;L`e&|DMdF7KYKMFEE6jOFgb zl6Xe46w$$an=Y{`=_U};ohji9DpAB9zHq*C>K-uLF>70&lSE$MaE`U`^_F$3+UR`iIXO|FeN)q+A9PI{F+;# z(AB+fY_KVrO;^Idc9h*`^qQ+=sw)T`p0ddW{jt!(rOgr) zx95YqpJGy~zNF5;sC$Gd*h-jDwzqfK?epq5L$3K=|LgRv+h5}GgP@U8JCR#L@k6Vq zqD^zL+>#ohCd&5ETM4wE`}C?RIsF?Bq5vIBr4Kti+^ACJ4>XrNXa%S zPtm}~qg6G1{c#}~zO_UrdAd#0bps0+7ClJhg9( zWeX^L$7k+8AOEiwjgZ=W=(@}LVWKUxV>%~a8@evEGhcBob^=RPmXThLmAQJ-rQ{fV zbJGJ?ynm(!{t_6~?ULzn8W-%CoS)RzLJ@58T|M~vABmq=rqnAP98J& z{@#!ouHy0JKSV``D8sZ?7tWq|nRT0=xA{&=6$3*im(Av%P%An{Xn6PW=J1@NEM9!B z=p6iTDg2j4s-GM?y1;|~(}A$tI>r0=UYfP0UXLPsw25@d-4(gkmigk(&HWs|@};m- zi4x8ki5`|~hKaXn$d2*QaxV!gkFXyt!KEUVGwYRg(&MY@nS-qBzC2&L@o`;=!@|b$ z_D3yg3$BVuPmTIy#4+|w*D`4*cbNJz^_%R4b$LtLyfZG~`#o@8b+HTSck!IG;k&Q> z6N2gw`NjZWjlXHJ-(5lp3pK(T=&HvPcSpbOa2jTNGovYiatbw3>AG#)nz+NQCnCvE)b0Ql;kdB7N6-1SQggKw4Fapz|G zgf_w?3hTx$f`q=9qv7~cJCWP{6Wn+239w*O3Swv7r&R%{ox~3Dc7-L>CCtf0VIW76nwHHUluf?k2%uZDR^qSSZTysCHeix=%{ECLt&GyPZwB6} zL$wYR`PbBG(*5IKGex9c=%@LN$t6Ai8L&7kx)QeCxBFuFsXuqGbGgi*v2nSD^7Q@E zn`3|W&TDe)6W+TY)Hn63G)~}vCS9XEslvVu5f{z zVOHRO6rD~1U~)KQnM8;Jw59d__oDxM7Yol(iP<8MYuzcZnIxJ^o=mCw|ALWE#5kUNS5DO$r(CM6Ia`<07G!y%US1+v4;+1LpB;6LnO_&}E zwY?aA{#*I}I7+I0Tuj7LEtx9whCW0}^f%#sQG2FRDKEyo*@vV|Anp?&b@+Z4V>)&_1<@y|W_ zuk0FTp!eSGq+fJ6R%Fe?KR8Q<7AcQPKs2u;34$Jm{2KNEVH;Q8Q`i`cF=}oF4uyfH zrTRrudOBQ+n;bO84Gw<^vg&O;v7=gu?Sunzz(wd;Md`OLZu%}YZ?OxPY4lkI{S)Uk z`3N^pRPn<&+XjyIIPSH10B5iBCLQPXWFp(1o@e$pmEIM#*QNY+&g-J>5;~x6F-AOk zStn8Hxqfg{rA_99ZGFm`P}9eUrtjic9vc0P;xe4GFrRk!wv@rJmV6J$FrM5m2(_e^ ziAgz{*5(_3sXq7D3TJ2h3~~6jXoAd!Pg22B76$di`IeM@Zz2&@=tWG8!p|t*_rC3D zUR1ypO#7I{xoJ3Ng-)Hir^}?>$9OhPNE`@&Kd)?vLiT3vuBe|`TTu zw%a|GpAtI73$z6!NEZ~vePhzV2PG>6i@F$zTirUn{HM7+5L2yoLd}dw{^L<7L@wFbLt(G*K*YxYdNsofl|%@M};AAav~}n8;*YF z!iArRW!9F1@b^UW=n;RH8VkV zlJwOPqY7Ti@bhj$NPuRm0@irmEWy*QVm>hQABCGNc%+?BGqw1k$7RE6zPzIntsAlZ z^T)=nO8zh{;L|rN4lafC{MrZ~s!PUPwY-zk^J4RIb}3QKH_4soX^#7@Akn~eR>fwK zd&aYX%+-Y7HB1;s!D91p?vnkx7WPpz4YRy3$dNA1LDr7=%RtzVcAo8@t2bcwELp~i z@@ZnF+=$4{bj20q74)EF3BIM@`U8#c=~+|=uUp`C{-?6TA&Z13PBq(5w#^1r)OWX$ z;G=>fd3fLtVG(6Wl;*01^dtl;o$mm>8#|z`?D8@G0sAiA7?$|BjY@S&^v&d(ubT3( zlwVCAcY_;1T|t*-nN0JzrYauCu%Zo<5B>KU1H;~gQ!Tw|Vt4gKzSpd@Vp>*3SGw$b zUSsWxZkV5cP*Zr++xJ-^nKbP@W%cRe7v7`eB29$5>*56Y$ZWRfM!Hml-pBeN z5Bf&>r;sqp8N@0zJWKWt8;>2SXmz2B>NRhpT6E%ScmLH3`xaH~e(RFq8RS9wZ81^% z$lLdgr2B<$buy}1a6gGQD>a9`K&pr+2+W=jq@;OxM~{AX6npM> zc`=qZjOFx|V9V>nxTjv`HK*Yks7nS+?==EZy9{l6*Pcz1jRk8hR#3;xY| zOqP@_Y#p^a6JsqiOpD8pv^tBaN?i>aF2ENUj8Pwa7^?5w`G&RPeE@OG%^R=E`RA{0vZxW-~#g9oBAL(Pz0HPKXRyUZpvX0l$OK!H$N68zmDKZnP(q-wSBkn%VF718KT&gh zeFmLExZOB)woohcs_>+Wt=Ez|yUW<}vf1ln{(i@XD~U{e`4erMJ*con?N{)I_(9w| z+?c56GBvS$__6F?Hfo!)hPlxQk&UnMn?wHGV1sj$@$tlHOqq7(!t}Upuo(UFyjq;t z+}sm8P>3!4riF4s`BRTo&RoU~aIHmsE%V<{wt>|Anbzf%g;ac$&QxXG!SM|rky&@z zq%jlEZ#!4elvC25RhY~;QvTGmuc2jj&rQ7J~_{2f%c$g-T=Szka5(5g6M%zKpm^bXW~urrAM7X8wFOb5<`y z>!Bz@q${BUncKJh?C+k7OjQ@0O4}$#J`IPjg3qsO1FYgC@+bI-?)#UD0uH;G;1fwJ zg#cahGe&@c!wW!bMMC*8UGrj*dyp-KXT{I-#U(a9J2tD8MvLz6#eOne|A7O_)_EWd zIeWED8dG;&MRklv2xAlZJ3J4(GJGtF);-JhL# zI1@XTA8Bq-*|DxNKe#F=tW;4k`a;6@&w7$ z$!Bk8Sg$5ntajHTC-42S1>`lRu01N-Igbj6P7TKgp0ckQ}S{h8TGF;!JW#M!oXO zin|G%dpJzvmltlW^tD3tFHExVEnUOt-2_qf1{Z6aV$?R-Ov~!tgAVr*ia+z)s4NEL zQiw42F)At6HQiL~pw2?or|$^43tDg;WWGzV=rhvobJ>=l0WMKUex&CGn^7t7VyRvm^{ z8Y4F-!&3t>4d#=Is`tN?%q}|D49k5$p&wj7o_+R9R&Vq=C>5}q245rUK^-dGCSZ}3 z#Y?7U{l(FAM;W&kH103vw~!OdWY_ym4iS#p9!$ra z%g$KU>$U7vu9XiO_w4XR-z>Ij9sS!mrbHj}6K>QdD!oznzFp52e>Ejgm`~WlJcshP zxlPuW`l8|sE_gEUANK1AaRYQ2#Wn75_Di;`Hw=y8~v>L6YH6tGr#a!;*sV$RDy|J z*r=+glj;h%bHQ~B%uH8Q>3D{ivuCg=%YEJDE)$sj4IY2r@ANVfjFGo5d*K(cl0E$V zg?60x;|2+H)2{jrE>^Z%3jNVgWU$+3w6vWSLhlUL7kZu7#@z3O zYE+6*<6F09{8@{~G~EK~-z^lcHWfrZToZ(aSEs(KZ1pCz`j?yV+^I{|K0?Q0 zcFo$Z>em&#^Ff05rZ_%EVm_K%W8^xG?ZLZt!Rq6}iOo!iLU0UCU)7#CF?!u?q?w`2c0ozzUx*;yZ9P1Zt{T|Xw^TQ;9)9XCmzmIh7iX+FbIgxg^ru|E*W&R>S;9>?V6gK{pZIyH zpN}D)?)GQTbAVv&LYm2P74hbu3wdu*TTQd-V!Dbh7_tx!)qFX$QP=YR=cOu`b}rfc z@>5w(moEUX%(KEGH1ATM>}W|_aK^`zSw$(S9PhGWrjV3K^P^y7w$1bxH)=8!hD18< zyc7S!K_O?w1PQtZQR3NQQkr~3arvh*1atfFUiD{>t%UNO)|mzw}_8mx@X>l;V!bC!!J7FJ?L3 z2SJ<3y2y9Qgj&)}IDs#U3=u!RL_zaj>2}_9Z}i5yI`>dk-Mzap23?JS(bh!^*T^s2 zqk-zz-^Dwm;a#5&GLrxuH<~q;@g1HU6L&F|xfQ0nla|ZQOz;YbQ?;##{JGs9uP)H( z?~{FV`~9JpdpV>%y18cE7Igvag_JQ)9a0&WhwaFa+K-yl9Ai5jme;? z1Gt%mASZhk*(YHY{!05-KVu*41T~LXkwOI&pgUZS?sXO~JZBDioF&g0S0nr#;+Lkt zz(JoT{^}bYbI8J*9kHIl+ipaKb-W$4?(tWVJ2JE)wKr5))h6v(=PU}m;fvh){^@Lr z)9hSyMJtQ`+Ec0#GufV$3n{&qRZP+YO+ubPTiG>m*S=+ULh6F9&m`u)YkuktLgBb$ z=Xj|mg{#0f58c|qJJLb$!~&j-k{Fw8%`8Xq@YT|Xb@hofYQ^_zVx-+UJ>PpK#|}2* z3tX4!4iv@Q3N)Sgrx-dPZ~H{vx6RSsygtLiA5E$0#HkRJnC5dC5gx|HNMT*nW#0R;q)k*;#nsK zGj$I;7xyl0ayy#(tAJ+V?ajZR)R^jVO53Ks9DWWNyH<1|+TRum{hPVh8Kq|!zK(?X zp<>nFYV=OUmXrptb+O6}+bAamvh3+EKIhY6r{S5_%*VcXo=oHs* zRMrK>-d?APye9GZP3q@00YxlCP(kc$uhh(>f+`3u9!dIO$s9;f24RzLlj@S zoRUj7><+^Mi}_|#sS+2ZtYP-%?4Jhsqw=6B3`+4m<=>d!l{3}z< z$A}MqOzJXFDx1EmCeNYFj5qUX@LXCPVe=R&xBQ%un(+36Bag3rBMk_ zCvR?@gd_9cwQBx+Igvd=&`y6dk5|FS{4Hd%X+MpO$&Lu$&3j&tdf$caPW>g1j4O9j z)|CrsT&Wd1t_V$ixogA5H`9{qqVaM<<;8K473eP%D)iSQVVyn( zsgff)>#dpK>JvsixpxX^8Sp$*} z-AcO4y6$~ojaL18hc9BaPjNjpuegz)eV{pL9;2LGY8g=v0KD25YQcn8RthZ{8ZQs4nq9T~x zVBcZRoA&-=DTpD*tmfLNAbrXgjc1N_CJtmKPnqiiY(FS`Vqd=cJuNPs?^0@=mrzAB zshC4MiGHy-PM9soc7{Hn#@tMp&BckQkG%h5;P!h7h&>g^|2ubE;np5?hhA z0C~|4nu;v&ba!8C2#6M;+mATjMhNBWQ>w2(DJ%j8yEr1V`Cuj7%1bUSa2LOHUyW$G zQpTeL3A*vA%fTi(D@h5`VJwqA^JWOT{!r0;m&dGUua<=+)ee1e2?N7dv(+S`9Ir-# z#o4JS@M|xo$O5~{Ler1;Ux|=Td%iPsh!)rTvmW(7{Op$BAC?0!4K-bHB~vesm!s2h zettpmtZexxY0qELRjsLXr~DfCB?s~XsU}*x;K86C7y!oT@ciil*@6?;<~gJvu^n@L zW(esAIWF_=FYQMRN_tw#5Dk48oJZ6WRqbT&U@fCOe_G7p?XR%UwXQduR59t;8S7KY zZM1N<8C6wUf2!JAgY-CW&1C?(~5?t-er23x!8G1En(pr~1EKuY86?fM(NdlPc0esYvea8$=6J1T^I1DQ z{G)|VZ^B+x0mF0g!0VeiZ(ExtT`!t_@4Dn{cC~z{Px*6u!QwWkm^KXRa+7yj^h+3e z;CGns>1K}@*Sur1>TdRXq{Lm7OSS#CVTr{yeDOnGrsZRQI$r7ID`XkQ7coZ$Pj2W` zo)!litiKsAmQNHKSG+7AgiZ9+v4jgwUTMYXMi(~f)-3qyrFpnhomO=!o(>0)iaJp* zQhLwQHrmuRr%VigC9VeWSx2=w@HxhvYMmqe=E{ZVO{8*trB|k}?ApX)6?ykP`^~b1 zS1VI_z9ge<6`mU|ai+D^(^p>kD^HDFbT1&mwwx0njsGb05mwtqE*g6dA|0_0%J=}v zDaHT4BZZwmAtq*yliLtoqA6x8%cKF_FM3M91lF<=@0nw9V%O4<7@7Y^alR`+A{KR@ z|D#C%1zjHS@wY~L1o`I#8+jG0OWr#Y%dSXVHYp@u?*|X9VbfOB+cp9SC zee@qi!%GWkteMc~(86vWJg0aDJ|6rHa@MLYHi5p5N4*y!gAx1S3tL@ij=;3?t+uQ z41K0au;0>LX~75b0%fAMy!%q*R3AE9slUKQ8pd2 z6mysuve2f%Z0eUnXz6dVFKc-qF4&4ci~yKw<z5of>;^MZcqmK8W=yW zT!|Li%91-Z+g@|8>Gv6;U(qrC2zTmCC}5FR(JG8L8i#%LtsE?`-=@#E49(xG00(C~Klriw9ZGsx}uo+EjFf zg{!<2IdTDtM_>-1C|+{JRurjcAA#QoVGa-|F+>!)mlP#}02-RZ-m~_Zxajt=HxgpG z4SvIOu0A7R2Hh`Ik`8Nf|2d72$GM4MHZuKxua-iraLHqH_rxvCRzkijKD}i%p;Zdlvt$!gpYtk<5+|oKx^Co?*sWle zqP1+s9}Ectau+6FEj9nv3sIMhEwQD5B@0PufU~^QFFt||-75SPs4J-%(c^6piDCAD z62uw(BBK7sC^;TBan`tNgjy3(mul~a-2+SO?)hU!j43O)>vQ6F^?0D{4*`*kUNE=i z8=#fA$G6?+BAy7Qf=$?E&lHdjPoM`#9KC{$U>karytIU>htH>Y z#`Zh$o_I~GpPxndjN@-KT+3)77B=ut5lcK~J#bN(aoo!4UfxacElrXgAT!o^jXV}V zvOISa@-%X!!6OBHxF1%lZftGPBiNL(r89TV??Jdk9`hR{B(T_wT4XJ@>ylMU_W`)A z6-w}iO7-?-8LB%8L6pEZh4^x@$#F7lo}hN^!N&bPVp2UR3hXrRMpmsH3LY_Yh+gF) zGa7v%MM0hY{@S#SlH!4mE_Q(A-SA`~3vo03H}}I*rqc7vU~2@bJp8W}!!lXWG3hv^ zdIJ~*j^we1$$kL7>gI0u`9Ms%)_ExO$QFj0-S2#GTCUBrbXp>L1ys*EBF zOVNDhvE`?7k`jvHem(2q^#XF+-GNsnEUy2f@aF}lLG68^(@8QlpbT<4w$?JQe9botmDIbYCP=H@=)n+E11tf9OF3NtN=}M${nhgu%~| z1ofW%HCp2UI}NP`$y$vj&fsyWfvOKY$5Z%ML{6+1J^tuPXjLlFc^LKw%&{b{QSl|I zZNcU|>nyvdwV^Sg_)U%GEuUA7H1ujbi%@ywIET)F`nA?gOOu-UJ(K61B2c@-7=M?q zbz8GM(Dl2(r#n|RK_19w`SI6hYjVN)6%f0_YKVNl21-*ja4`KE48iun?0s z=B=Cg;R9IK4Cjr{8{$4WbJ))Y@Ym>{NjNAz-+o`%&yA+=O_OPufRF0PR)H#u?Tb$-oj#Be?EZ~!Pu-Jt+2 zg!41M_nd+E5H51pp!0)ztPNGqyt2L=$^2}#COOr;qQn!O(6egyuyhq4$9<^)lV{CV zC-mSayNoW5x}U!z)!UQ9Z=Ve76+#=AwJ0Vq^zVHSqqDkD{!x|P!zb@JqfF9N{H_@Z z2&)C!5qHOJNU3DO;WM9~zyf7`i|heeu8}{$R|4aEmcxn!q6}TN0bOn`oiH$Qpmm-o z5#l{2klESLlhC32iqKgtda`H|-qtFPx_`n3Z1LZpi021t;?f|4cyw>*Zz-Upzh0UmEpI78t487q)amZR{ex-upMoyXMC-65gh<5Y%fo( z$NQt>nhFyajaLLHQ?NJ5fiR63;Z`dwJ%r#8E#;JZo0#QW18%i?DKr5zodmj2AIdGO zC+b$9{_bBF82S8MbBmS54wa~5a}H!t69K8{5Pd~DK7#=@h3vJRuOYq>|wZ2T&W1)5n^ zyo~e2$9Z2K<*v9dElY!ne|`=R4yz<6&i1zJD4V+R?-`X3g%}kPRq?)@nl$^+5(l~G z>427Kd0DQ;ptsy)&WkOuomH$!PH#!4So6+1V31ab?WJPfm9z9O?Uz0U+;EZVm<(J> z*2pRk4x7`X;dtB$Isou5S_Jx408AsIe8(~0t@blxBV!LIzwShf;T@dp!Y0$I4MBit z!uPT0;l++PaMkY~a|-D61D~G&*v0xf7~f2yOS$!gw@ANHzfDNjhB376Onq}LYyiq^ z!?mrKM2PjZffG-6Zz=>oJg3`yzRY#JKY{7k6QU@wIHb`Y7g!jT6EbbSu=*!@S=GMN<^wmlKxt_6DIhfbjxxN!j=%l;m<*7Z}rBs)RZ3(%A|MX#Fm$K}NF!y}h?$ZX%;GA)c!s<`AV1ONZux zBF3w+UHv{>InZWr^N#@=wr@)eTi2?ScW0n3`?dhZaG*zK0MeTi>aux9Mv?>d`n4*@ zvw{=_Y3w=xih2Z4S}#tP@wz~W`>kjbeQ2(b`+HJ-k?4}*^CK^EJ9m^QFbp`V0U}WN zQ$9${3{{?eV1sA#^Hj^A>Dd4W^3xi?+8CPaEoE=ou&y(IFuty1g6N2SqIYz^pf3vs zEdL=fKacL0PrahN22Xz)sp`2kJsBc{mqF9V$**64BkN|x#8w+z8WJ^A?(iBL35Y(~ z+3mG-xz}C@Y=(Y+a7hV4C}7KJ1NPniZ0Ae`p#+{FL^Yxtv5HyDU;SxDPs*$8B~TJ= zKih@KxB%`ILLR^*)tlEj`CYnJn>RC~)&SVAHQxNR@&@CQjj9F<14^$S+9hFNOd=$S z5Hk;4ARRO-xu%lx7ECwq-V-qGh5!bU8m7H9@BTEbE@nSnXkKN>z|SkQ2dZd{h-#&$ zi$g068=6a2b_Iu|j&UVIb!UDU*^SeJzHAe&5`B9ISMg79)0MsCQC+K9v9*r8#}9fZ zUjyJ?^zP|oAJ5mvC-tP zME~zOtJ?aRusL)I0aS)AH40`#C*w0=&_Bvyk^v6Sr=;Ql+qF4W8*2~#sUy(>*KR%L z8d>_h>ElD(yUknierwoWJn|6M+ouA$)(WgFZ#WN00x1ZpfHjsn?xCU4z2%Fy53%z) zpn(F=L5|Ud&{JTqCmax8Iw_9=>V9pi6b_v10DZX=kns+{zwXEy|M6b3Fk;QT`uDK9 zi3ehu{;cc^#s61JX+72dkMlgv-6{KZJ|*ci-!rI|YJH=J>dW5yzLVVkvr#rnDxR@n zBI6X_O&3~5>O->cJxloPxof#D8$nuo)6>APhJg&Y0(fCeACj=v*2&aB^(6MHQ~7NGKqGpzy{DoW2s2G-{ekZM$rL z{Pj8|vLj*{%eeORFA02uJ%Nw=(sNjuI$h|rEa(KR-nbGw+_x3MdL|3ZWVY>g_Rqq% z7AzL&;!AG4euZB%Y0G=o9FodTzoAbm0x( z)HD6!fL;jc0luGbKR~^wIV7b5i0szRygMjRfjyz+$4H=98U+U?WV;~wHxi;9rv}=} zi2b->_1t9WMQK%3*67{FMkjjxpy@czzQW`bwS|h?Al1vcw>cZjUT!cZBnX8SY(xA= zW3er{k%6gX5e}q5xDOa8vkF&Xqlw%{NRdSLMZith& zNFDQue0)>0Iw@=(y4C$NuUCgYn*2%Ks`Dc2-KeuguI~BaOX}uY1vOHT5A1WhQix$Y z-XBZyl@&dvja6mpfF``Qg^?95OlGxjFJx{vLgk5*q)I6eUZd5iio#Q1p2o9!&wv+P z>4h%yddgFLISF2BANDCgx7&|L7IRt*FCO5BH^2G~9hFQpIVKy0tsRi1&QA$Ehehk6~kSe5OB4Svm_S1_sB|5^23Ve*H|o-748RJa3bL zFGY7_z?~!ybB`7Ws;>e4T<{mtE)gl&3ABhM!SBK@*TO^;C57Z?aES(Uai!GwQ>w=1$oR z4)A0#HDK)CEsdHPnhr!b(kkd|ScKWGi?OO~3KWH?_m1-kKx7w3-P&CrHG}j{9j;0CD`#}<7j9@r0`&iBdu+uAC@+L)W~5h^v%F~|JXE3pY$MpO3xqxhN!-5=gLQI>D=k}O}F-%AvtTcX@z zlM2W3_Pz>{!4NqL(`iHa8*hD`?s&suFRZyA0nRn4RnxB=Xeu1liRwYJgRlU*sNcAf z+Jobz<_IjG;Ol$@jsj+7>%+CJaaWl22W+z$Ab1kIwYWP`P^&Y0giA&?61Rh;yffbx zD7|$v4>%?$Pj5EPsJ!cuc|;J(n{N1lZ#J9vOlAvMe9i2A9tVR zw(1s2*pIQN+Ts*{W~%78Y~h_=~? ze9V&_7N*Lp+(l?hh>a5JGWK+OK1 z=Oz>OOSe@QnE)oa-T!jIIhKF{kX){|PtK+y$3w`G2Kuq1+V8I%2fjt*iP7}0<5M-E z2a6q0EY_-#HUAcG3QEc| zE{;5I&0AcF5$~;mB`7~MztL(??f}A`3h?Tr|%ser_&@`vb0S~o@9gVMGJ zAo(f*tg=Wa8%sp-qWb!Kx&YW{FcigPHg@0NDfV~e(N7A8F%x(QC+Qn6R;|>@?PoLE ze)f@&Ez^Z@-40zbt6!^bsY6qOiSWF7=^bOGnMy$15B05YJxSai*6m;)>v`2JHN`Av zLecX%Pv6@0g6i8WliuOI-$yK*Uoj6^y3QSPJR$`d^Dh!3A_wIL&iF0?$fK$s=aJ&q zgy^>0PucGEZd&aN5B;OyY<*$Mt{f0{Mhfn>w6Xk7(6J%2WyAvEs}k^4P6U(>?iO}H z!~sS4|Eb?T1Rcmsq$iWtLLXcKmYTO}^H?h*@aZ+6prQW9DsV1pmWu}49l}qb>G>66 zHr?q%?#c$9nny0~?9$kIQSe@o+8Q$nUq^$ZJ@b^Ov(PBIvlW)ZfdeI23d1LT%*s{0M8?lXg2ma_h!=*48t|Wg6nGst z=W-0kkSnH__j&QN#mnp5%O$(!Ut_C;o);%56b40gg2Sr=Jal4%`&idrr;8uU?eYZ9 zlS75Z?_kQhB-mJ%`fr-?Qj`_?c3YcUMSQLdIn^3bN2sN!-z^RB_g<%SVchL3GR!gg z{wrpu_`4g|E7l$zub!8i-v^TJw~pazxlBM6+YTEv;cEhvuN`)tvgJ z>9jG*&oyqpu_mIklJCUfTB{vBp3m7tV%dhqfASU>1#B?%He#6FJ-(GozSa45()}i2 zgXlDS}eIWOhIf(-nFM+BC1g?%P;cMMBp(L4@9Oz*A-}OTfn$w5kgH*#0*g{_{wn zfK`B3vKeB+EW47jx|937&xyY53H$P)D*fy!o#)WHVshzD^C+Yn2t1CUg$`8Ps*_GJ1=AH#Y9 ziKulcdYZQdT}$`8xXx2vWqm?_B60($A4glrq=RH6GUEE1xMcD*lcxf~dJ>@t4-<^`c73 zZgsSjDmK?#6;sSMfFYT;-B3%M#1h_tHI)!;d7PW8FV!9d)yYaH@2GU?MVh`K9E(Rr&H2kH?58=uMv{4+s6K zcE5Xr3Xe9g97%PHj6zbzF0^&0^Okf9b39K9Ak#e0I2BDs!URd>shxX#A6Ln3fsR2I zG~uU}^wi`rR%O}AJTz(sJp@~7Xa-44XGx9`b;=ZeZ#S+1D#JDe!vMeeULH{Ehn0uT z^|Zpm@_(PgjIo>h0;|h3o%gH;B{jUhyRC`NUC>Rhgx$kWuZ%LAE~_E^lNU8P&t1IT z$KyW-=aHwRwUkvR3MC-#H_u}Is>$#YZf|4%Ya(hpkI$pBx@wY|q7#-Z|MP0_L)(@z=%}&|2Z+sW6UW9Oyl1>o} zOX^T`B)M-w{+!|<-E)h|^>~<<&V;uognmUNxEGn^^vx;NSzll&0D0!LH;kXk6UMx3 zTQNJV*+se58}P&jKHsAaOWbajPQ~o1M`@1LZS70{yg~DPWBV-UVw%>;^|0>Ts$095 z(*ohORt2PE0{TPS7x3xP;V=b2WA6WzX-wraw(}6BX535iJ1!=+GuSs+J^wO|>V_bP zE{fAHz};;UGnzjdZQI~opX&{i#Pv5f$I1c%4KgIb-~Ixm;7SD#xndZ!pMkIxWHg-3(IDKMHanq#CFtRwHH* zVnm;gP>t_|fku)y$)$6s=tq?Iw!~La2j1aRAyXcs*1FVzwt9AxJ@t|`tH7{caGeT| zB8F(mG`+#FlU}AjS4h6|DD48-B>{<^>-Chq#3?m_c}GXUebIhTU(c!G7qKiGkmtyFAi=&T*lz*4r}( zieIK_y*$=;tq}Qpp>si^AO|^k(Us&|=`7kyS;glu(`KsdXpXI}%awzu-l*G}iy1a1 z_a|hhn&j~=Owm&F;92?yM(re|Cae#fU!VMBJ}nleZ~2&NpiaG+Wo)nJrouvd2v4?W?SY1qaa%S670 zb}+3bS?$e&fl~>Ivalcp-WBH%*9kaCY|T=KU9&NYtfwPMNFNSS%WR+3j!5hD?NS6rC_5Qt(BGidn^oC83`f;_*@P z3Cv*z>X1(#(EJCT$9(+yT}QhKzJ?&YYk8?LkD#gERQtT;EzL^MOx^B8DPLBt;-0aQ zR_#k|!%&YBqVE{gQ7i_+Q|jb!Klx|BBJ0M81I>_kV>RyF<;4C0uGRr|W&1bmcYZ{e z@^m|Qz@X0b_oTk53O>dT0ZSP=IpQu(8)r!>e0+$a?%3xQoWp+@e$ob~f<0{wwoMv| zZ4A>565YbdRa$ZW23aq+p&sn0Ez?G9YXF z#~^kK!56wLfnS1j0iY z(y=g&?Imj7|GzD()O+L$wj^@OO% z$E0fl;HJFu7g6JChr6-=7iaGo)MOjI`=Y3*D2R%5qSBNuy+=g4G?Ctv4x#rFh=}wO zihxo@dI_P0UZjR1gdU2t&^w`qmUlnzfA4qZ%=vQ8?3rXTlTS&WC-;4?YyH-B9lwB4 ztfiP47Q{o-+ zUW$1>i8yqK$Ha3-1;I=b?bN{$s9v^Drp6IhTvvnMwLD@su$ZM>JHBtO6tTQ3R)$r+ z`nJ4}ix7f6Fp3W5x59 zET;7w&C0U)=fN4i9cUmv`V`>&)fRs^7Qi^=jY;(t(&}dh9>)UNW}O<8>!J_eyo)6e znUhw~IIk$^Z&{D*i=r(Pg|Nle`-GOo<+0?|1M|RbCz@m7X3u@awg;1fii@gy)y%r` z4Q4AQ@dcLdeZOonF#*ikP9qA7K}ZI|;sOfK5w_x?brOEf3akOxx%q>EG?F==z*(zN zj|;*tsjpVtUGVbkczD5rsAcjk_ms3Ocs) zWmuM5&5??5El~d9>j44A6~ELH*Ltbn{znKpeIagr=yrC{iPVI)%0WA0tT&a3*QSep zj^u-d%rqDKjLryn0sxyiU5O5{C6c`oChGn3!C;dNe!}@&6*mj zyp`tP@9_%PrsZ7GEnGYEVebR7(0+4S8J`Q|_;5hPlE_tMH5k0ulJ_!f;DMA|)<;Hn zQ_zHsc!B%;dG2Em@GB{ni|3bWc&{(Qko%M(BgpSjQojWBK>k^W1HRbECQy+NhklyM z_`(zHaKN&5U7+Vh#DX+CMw8&0XFC0KMz~Udl0YeZmUj_Ac;&$t;WeGWK3~)A9RTY= zYEwoxol!n-p_4ZOJ^#xKlh|ki*5jw^=nIAVnf&RsEfbEgSP!6jniHNfl`vk(_h|i} z(#v->`@czgGh+Zc5U3~Mk+4aWE8tV9i#iYvfc!&pV<@qouv3KXtw~<@|c@UYn4Apw= zEg}l&-700GBKP?Wo1c_d#8Pm<(>@%`kRBHA6iJ1Vw+Rm=!GY_aJJF>6Cx+36Tj8dS ztsi4U2>QFPHCWJ3*X=t`E6kg-i~eA6HpFBE>d-KeZTBfx?&M$Z11ZdD-nqYSB(KSB zM;FeYe2;eyqJ@8up4_g%~IaG>Hv~Ut1)2=Y!@SpbNr;fekdFQtxnP z{_=bRKN$x^)PMeu5<~_N3t`P4ig!-Wvx&#=e_oG#@&D^d)rc>0LZm z9%~gjKXzPl?s99HbiD76T zPjtTlQ$uxIg0^1D`|rj4-Hw&fGj1Tps zoUbd#PsD8})|o>QE_M4Cp%rQdjCon1mcFWU6}4wB4$ZQ)Q2c_@_OyGr>>-_T@*m|xapJEe(>N?39D=oR>nJj?FNc|Sll%?F8n>C)k}H) zwr*X2gh5xD;=Ws7PBT0s`MHWw$|YTX8^S(?$~d9GdPU}|$AG`acFJPw-Aje2C5reZ zLgalo#tn&)=7P!xLQo7xg68;_3lgF;S+@q#G^b{l0qiZIIiz zz4AJ@dBVcwmQQcaY{#5yqcF?{blU)MOc|*8US{)k@Xf9`C9N3QnchZp*1h8=`{#B< z29Z?bxcsq9%lp;mL0f?8hy?oXlk##<&!C2oYA=%WP<5E$mEtJWz?t%4PvHj&La9~# zu2!RBU!Vj*1J9`%QUjx!-{eY9D5O?CL0S=n_d^ermE;GuBLrvSC3}9xb$+~Czr+{& zByKAr*Vd&ayAamzrE`-9TMzz@*x^_r`zjQ2;4GzOCmB?KLKQdLW$o!^3k|@{llp{W zy6D>tHCj~t!%TLpEiqCqT)i?+?EvI~tS@ zC=tL?&t0IL43mP=v^< zw?k+BNwspL%$G2UbFn!|E?*cIU^~=hVJD#<+BY}H-~U*tBHy3nl}^(qoT?d_ZK8=P z{|!FvOBrCnbQ6bIRHi=^2S=JcuEvB&NS%s$zc}4(Oropoyy9Af(L-P3$vV^ef1uRf z*zRyxzmj91o^FPp9TyR~ps!PKVKJ-?kGLm9%6b_n_h9K*iB`^?oz%K6L2h}NRYi|o zAVcGY&TQ)pTP;`O;{@@u=Hev8Sc7xq(bURy1D=AA$;7xBW6*}lpCPo6mAVV3LGvMmg>g3kDPB}4cqE(7cv?7XL0P6)gM0Op}p-K|rOHZ?YB5QTm z)Ce()lRM6Isk}@|fjaolu;-A~y+}~iYU5&KgtN!u72@L`GY_yYSsjvrNg;PSekv&1 zWutp7A{;)oGVn=Z^-d;D+tO$M=*W#S3xEAN@oTZ~4odmm4H_IHO{41b%u;XmWkDnN zJ>OUC9M$bSX_uB+>nQ#VKl+g`6HL`gz}{i%*I=m;x8{b?AE~{O`{eo&=(NBcMSTAu z5y<(6#2Wc@)H5UU+W-9Ve&*3^k7ERr#A`gxyhuJq`pwj8rxrLDIcd7OQWeb`V@$ko(9sH@^)R%F9sH5H~Q-U3)Wo ze>|n7pQ^j{1qb8t5pqCd280u-1uP^DeQ4V8ssWJv)~Qd`@ZIWJCio~A5tvxoWs!uw ztcuCq$P5OLvi+rUy;|wrrV^>@g<3bsEhi2 zYqjQ^S`2(N^VdFgDt65@?KE&GC>lA@yt+%cMMo<};^%5$=#X_x%*K>YEMT{o37&2r=YAzY5pP8jQik=pZ`=^GlnF0Lk<-yF03GP1&s^9+WZ z(DvJ&Jg_7Rh} z>{`QZ&lvw9@!0{z7Xck!c&um^xo3&sqJb_}$1Ygj?I%+cBlf$G_enk>7>5wRwohkf z+qpV=yErX7v)*C#{CElajkOXRK=&LJYp2avtnU%HW}W~gu<$5oKY7Lj82STRbxHrb zK3<1iKBhT|*gx&YOnG;RoV%nl>CSZ4U|gLA7Gi2r?9m&^po~Bt13=imgv94({8$qj ztzrkVk#GOEX;17Y?BY!$!5in4SG1^$Ue^LAxHM$ny=}=m%Hy34*eMwB#x}g#8$(@* zs(v>Z@c6Q>VAjvVL1JB%+114uSdhP>zr=`bORcGR0>?LO`i{@N*ju%}^TmZ>%Pi2u zCrquSx-OZnH&*2CA3ajnHR^r#6jq6_9a^oLUIw#2zI02^>k#!*_vvLa_st@Cq-0MT z`;{2MlT06sNpjo;mYRLW|E6B}oXnBWJ|E!zagr9!!1@t8bdo=#fy&5<(b8oHb1u~( zg-81B{l>daGQSuWnkfKK zxlNyreilcQAHBV&8w9Ks&2FyyM3h>8#d21aP4F?8ns7d^Ra0wY^0H-edtu2%Tnk;) z6@UTq>B(X^YVS=uHaBIu!0$!RvrlkD$A*D2WEZ4G7f(CH_AEwF(jpU*y-gdWbL?(n zB+PZR!nMUv-F**EolC{MLSU5u+1k%ynEHRrrgwD``IFa6gk0rnpCTI8z)^aVC5pJ! zx(}gVRLN)ama13oM$;+X&bPWUT3(LyoipT851Il#8_#;2AK6{N8PyS&MFy3=@%TOT zi?U~gqB6uxO5H34ZnF2P&ku;xZtX_+^>@xcLN#lsXYlgy(z7c+SY~_^e6B}jCabZi z7$)iq^i?=}0E@EEcNH&Uj`IFDC3*^;^sKow9B)1{2qP-iA}})=x+WIJBMw3r*(Xtt zW$Y7U-Hq3_^%{v!k}GQWVCMbsi+@O*O?B|-&J=gR9g{4LF49;@UP*S^BU)SLEigo=v-n1Z|lKMkKzsV_s!iw}H&lNGX>Y^fDEM zA+uA8e>~@GzVFKsCS|s?erX))D6xuh_`%D{@apIo38}?nBaxOm$x6@3Vv1-7lqo|| zCT^$xPx%;}d7hDvj?_Qjs~@L>n1>?pHU{-o4LqJy&DF0%Qg3o(xtsA504VJVVrf8m zu?_&bRe~aS-;$#jxHg<0`OsW&bcmV@gQsb@5Bn2S&p7m}>-I4bz?jQ-(tBZ#xV%oU zxGWPyTpewd7a6XMl*dFqTUo=5sZ7&~Y+IzTIrsCW?vNN058hFokIB#ul|l{7s}w%; zpjrA^>DO5Z&TI|oe~l?mbhXu2KWxomuOT}dIzK+bteFbq zXB3?uK@k4Af<_YP)#oJ^Yh<70{gW_ZxDnL|0kOeV*2EoCR~(nmsxK=kSNfwJ1Nn>Y zk*J!2wc1Y1LzAscfy_;RSh-18cy5fM0+Ys)!MWpTM~TsjK8&}VI9U4C3w|YA9 zVW=qa1};nXIbwV$Zr6#rK}WJRfnRn&m6u$ir@1`8xVpHvJlZ@j`qv0?--(X;)VqU0 zIBT)x`{R!R8OoO>iyYBClV*lj*9q!TppIEON`Ftxim8nFMY44C!z!uyHruBFmTWzL zFw*rmkRoeI?;jF7z8h#9ILvQ0Rb+~-%1tIGYeNUcM~EX1`w8PRDlibXRxFzlh_K^5 z|A04VOvtEk%Ez`QD|^7_#mNds#fn*FO?Ar)0@>oKRKLxGsI3BEr`yCZZ+7B1AU!9l4pZsp+Pa4_X_z8tswb_Q=H-|7|B+TOa~^ z!!BW?jPg8%1JLh)4y!~!HH~C{z zXslxgW3pae^3+zq(G%VK)^o3iv%VvZy^(=g$qO?JY`S z`u^nW`Cl)Gm6I@q!g~_2IWQdG{f+4sIE*1@!Puxb-3g?Q5{p!8%$&Pk`oY(9t=#S$ zubk5`Ce*L*!&n;h*Q0D};ZNKKn4UK#w}esV>Cou=1nTWA$)A{o{1(rVqYQt;x*(K^ z0!@v^cl7;AboI4Smt|}NvG+s`rx8KFnqTixj#Aa1ndAqc`wv$GTKTVDRtJdt`m*o( zpuyiu$O{#%Y}R@Ysd_*mip9Ts7WhanIYY`aop@#3aj;0_!**-cPIFQ#uIeKp3s^J)P!pjm8SNl%j+$=OqU9fhyHjiQA< zmTD<~ksyI<&m-g8}qj2rq;W!FP0FF>M!jvl2itp z@f(34f=xh)L64NLfl2d(K|2r3o!P$FzPFY(Vr6LB={1a$n!3L+VNDk+MZVaWj~eIZ zbJ(Sh6#CTf#@Q+JlHlku{APbs%Xe8Jw}>Rcb)rcS9k(fw63u5Fx&Ll}h5xK7I5TNu z?)LsfQL(woYYsFhYRtSG44eW2=kvJ;?|Ffgkt zcXs*3&#@J@^W`!{I3AHv?TTPM5z+Y43UN;I>jGRM)%)&*88nT^6Ilbx1|`)O=s#BP zVM&1)AG3D593Owh3P5>h*6hv-e+M(1y%6g@FwE2&&wFpWab^M9PAno^2Un;SY~E$N z*Ot%e+jfMEh1qJ{$xp7~_3%XFrpJ=OOo z8t%dHOAqWUj7sR@0wH45%)aJO!n`tl2V@9HYNojj@6Ihx)G9a84n@Z#sTVd2W}R3( zI;^^$FVKLOh2g-p5#^$sB`-HCAoJu>LE{=BQq@theYz&c-!u6_^{uYBFC(cw;02mp z(Gu!t{_in7O6qrl)GwWw01MB&j$3Kunp*0d ze#~D3*4%MmvaRHESopL6+lBE!OzG{XYoq%JvnXcPvE>oopLJeTH`+tX=GzT#W#kcJ zPt8)>-ts%Svv!b4>4jc?Bz%^#DO2HT&-+zBU2^uJdnX;W{HxS0leWRgM7j1Kl76KV zc$sWj{;lQMvoa&uD)xP)+tWExSkxscfuj`aiZ7KANsUhC{5>~+^LLQsXIOL36{JQ$ zPbxyb>Kn@Hu2?_wb%2AMZhZr`1k}p?V1klQn!#6e>835c*i5Nr`U5Z{WnZ&k8@RHz z0~|V$+-fnj)ZF^@rQ@6ZU0+7n1M11G)GI_%{hOZfVB48)vb=ZsR0T!aAMiRY5symZ z;{+oKB_~M2is*?|SM@_!X`jVb>Ipo`6g9pe&xnU-`JZzwyu!Y#9RP)_47@pbWV4j)D~?xLINNu zsSM6bPRc-T^7p{Um*Nvgt<(^alw?s6YmsU5B&EsUv7s%46Ed%Sl248i&s#I?1gjMt z?yv}l6cRML4GYfYk)g=^q#ho=Z+fmW`3=uks4X^%R6Isb;BtIz0`{3qnhCp>WRb^U z3P_CqGq#hpGDx?lbNG9LmvSkJ>!suFM;YzXsA&G3R3rqS(B9qcADZ+vQlPLIlc(a6 z9Fvin=i%qvMH%d;*va*>%z=qWW38g5b)a`raQf=xhrZ<2xKjuS&b}YfLje7FsUyP| zKKGEw7kYBTUjqOhr?&c@kVk-53Rd3dq59H)s3{84l*jD){WYV4(hr0Z9L>qDpvnN_ zI`O0#Pu@m+dlATFpx~9xpV3RkybXys*mVsO{s`BaQz^md9W&Mb-?)vGM^7(4)cg#S3_1(^PSiP@g2=Gdeuo}J zecncwc~i6;Z_cBfBg&j*1bpIojenSrS!sf&Zos!pm4Ei0T$%Tc``Xk;!|P6_tF~Qx_2`9Lz$=4H&LL0M*l`|fBcD!N zRxgsvM!x_^_pvmkiP@uf8-;06&-${pw8L`q{Ik>sSd%ch)$>%c#YSiGA}dmB%~p^= ziQ1vGhbk+In!)UaP?lk*{1GeBlF<7HjOO*Ah*X`;&4=M>_C-t4Yts!DFKYWj_bi%+ zi9&XIbxKP=dX5c`0?2S8X@z!*al;rRaZp)jSsSLDSTmTVxhyr zz18!VyzEaub}vZ-9?E9KumOi+EYdH>_382)xug8pT1$P9Ac!{)P)s2o1a+%in9CgrR0hQ5Z%UkMSbcf zZI)%t!7IrnmV032qFGKLL(OoL@BXnT;`-t=c^)@wtz(|Xsm9rk+2hNX00hRPGJ!V3 z%a2J|r=KNWu)pc|7vMY9vnBCgZRsJSTUTc$dmxXjdNDlFAgsq86pY zX=wPJOX9#G%i~T?bjt}k#)${G7E63e&|Y@z=7E^;8vy}`l(am1%&UZC1fwB=5=cS# zheV|BT--?~&9g*trOHw4Oyj8ZrIaLn#o4`#O0?KX8Cq^1(S5-m-mR$hSz5U)4JV|v zzGr7y-gbflwOGw$k}RBJgtvF$*Qtfa4q)42SfYNAW=%YO0yNu%A_p;=jH$RvjUDkeWEeXqVcERV_XZzkg%56$Gi@$Q*b8)3sQ1Bdy9V9}YJ zwGGpjs`YocHcYE77&j}|VUL9U3F`Fo3Q2>~H}>@c%Ag>#$T_}Ntc{g+@!v&*#+uvL zn@gqDjiuoNB`96*`F*7eBYzBy&#$=!#=nLy-b*qT2%nQKfkPXTS}ZFXV)duq#Ir(i zEwn&==11TK0Gw*moE!{C3TJ7&!?t1{LbU#KTcD(oNE%kx~%Z z{U$>PJj48)i0@n40k7U)8&A&Tn!gKIQ$4*JWxRKeeTMY8csr`AI_z@C)awHKnQ(FP z1sH^jU@6yq0LK{~T;(?~W6Pst5ED|OIX=P4{{70-zdp*Yi6JA2-1>w!04HHeKy&tK zwR$O0k3&VTa)lkGjZ{-q0O~Q_`~`+?^IuA?pu6><2_f>_xx!BM6T>UV(lA%y0!g75 z8F}kn!H)CfjKx-3yM5c~&-In&QX>j;KtmWc+H^D9r%^3-RdOo9MTWs)u zSk&w4A_N}mjR!s=%w}I|kJ!|tIEeOT*IZB%>!}Y5&^$ET5~a)6sX`YC$Fc@XGL(c1BNQa{#y;&hY9yEV zs>*H7PqNP@I6pRqQ9J~1U2$zG#I3Q6w76V_dsGx$3Xh{0KOvo8gXC}M4YdunL$cE` zx{s8+(kImVljD_+OA^E##`!pnqP|87&$32S>y3^0aaArctcRB|NS$_B_RNcyp*xvL zCypR63EXTFov3wPNp$h`wjib$_t7Hygi2nIJZdlby3!wesaIqKy@Kj_%4$1?y?~6{ zNR0kz)_0FVjTN(8*v;kGrA5=&C4!@7VA85J6|}&M9$F)%fioGg1A6Av?%>%3fV~Ds z))Mp@x8^|l4#vNB#lSt-KV~S3BI)#G(VFB+4dJ{?x0|w=U!#erqa=8AN@pfjq|ef; zzse+ib?AZGJRdToTU*_I@{yWWO^=nYjm}xBR&r=TsPzeJQBf0FmF;t%XPrk{I7VT%&V+Xk+>UGl${x`ai>C^bspm!ulo2u`SdoV!�ol$&E2C z!JX_`$mqBd`92Q}xiZ%WfUN4`0@Y0M3)ak@&CpW|zjy*tpF8k)`6p~%XD9RiSiD-x zQqTdmp+<@e;$BSH0}`^qIM7v81Du%9Ylbm{9d4NorJxn=?~CH@Fj0mH--9i1CtaQZ z0*VZ3(CI%`{@>UCdtUxp15Xb-XEEBz__Lty>d5wp{B_)u9!w9ea<6pmTsVft^*)fU zmT`Z3+$zO8C7Swi>CCMihSWb3Q8qPmwizkKIoBb-2$0;#EoOeeLrs@d!+aIUXgag* z46G|N#i#}M3v_;YhL+w0QC0I#it{Q<^G}vtxOL)}&(^Ys2sF450BvrKNS;RIqI;?{ zvRw*fEG5eo*SU`LaQPc86#4Zx8xb+bg{gPE7FCis=#ah5ul_{wQvSKdl4W65M3k9u zq9NYN*(ch>Cn%=A5Zr8J6T5n;V7kTMxrmi=3Un@_EHSPJ`8g0#$LRI|o@ZC7M`t>2 z*akf9@pKt5IP}8;w3i2cb?c2P2oaFI$l&YIItBl2{^3A=F5BJpk>BSsFygJFe};}> z^m}XcHpYSp-9;tY{oHzNNv)t&`-|2_WJk6)ZT#B67CrJln)@^}Jv8PsOtrg3)@ zG#y`*xyqy^waz z3@`rGcp_eA3fm&-LoORKuoqt1M?q)dxOPB&R(oM$pGDy@Ps*Wl{DT>w^9r=UFJFePOarACh}f-Ti5Bs6+VorAAdKL+vdo`h7ks{hm3Co05WE z9*50TUt%PUZMu=MG&0;4Rs}|RG4U%&3U{`M!nnBZQKjo&I@8@pg*TZ=30%0nXv2#c zA9tn9gU10#kG?q-w`*q&-Wz8Am-nweUkWccS8U2~k|Q=N%;iI54=kL-#Woc+x*N^6 z;kTO8ccCd462#t2enMMdRdaQ!Vpvp=NVZF(Dqlcmt0CYWZ?K=!(RvfgMiH@iq4k*S zPs;t%HK%O|qAGM=4fS9~o~enGzliqQ>!a)Cm&!;v8>f|drZrgbf$zzo9My4C$Q#t{U?5kJ!F5)wBL7Luch{*du76(?>k;W zx%gyMBNWTwbaL#33b4Lgcf(?&`0U$CqVRBE{Bpf=WAZyRqb_xE6g)ijts9Ajf4`@d zbT_^WVqM9tF7w=WiVoxikk7+W+B{UlVb0%)U#eU0ZqS1-{p==&1l{y`jl_jftY7A6 z;=k|2s4mgcGEO-7L8>RE#23+5NwwiA7wVLj)Qk85v#wI!8PU*m9sK| zQWeEgXzv%&<`9K=<9#X0sP|)X8r9v^?fnPLgD}yEwHAcgbdBAn#*7_6D3zf0h7Jxc zcN-lq6$#isZT05DGMlfoIxb_J?{x~znv;ijPv}w~WyMNKWId93@LdZwTD=y#$$mnX zHI#ZbF}ahWax`nE4h1$ywf*g(#Li(USSqfi1|`g?YXkUx1$sc0zW=ptVNLQmJVueO zSJWW|QB{H0#8;aom)O$hXpJ&N7NPYs22E})*;X!;L}}Ac&d`}q3a*YKW)rHdPse%^ zl9IZUKNGl2C3W8}B)HC%^Eju!XhYm=iT`|7!X5?DR+sdiIb`~kdq5!;Hz7HvjKL_? z9yoxdRbD%AW-OcmMtE%Obmhw#VQ25#d}dWw3pcURy-&$93P4F8bPF|fECatnucZtU zZY~U7e1smx;!wx)k;I4lQfjaMV;)Rku{*rJ1^vh9NSb24M zNJ!`KLuN0F%zy^q%v;d;-gr{n*HqW*X;r=b$wt@YS>$&bUv+1>`YRL)bO&Gn#^#>w z$Ce7KN^3!TlRn_+$ZiA*FK69Lvc{i$L7b{B!_FU+`sUJ5eMU0{l}t1B*`1%Hc#_PV z_~uwU8}lP9^B(obCMs?^mE;@E86~WdT!Z)MaZgiZ-0xnZPsV5)Jn%W)NqwiiJE&;` zk~K*7%UiATUn8v%8+`5bZRyFcSxi~gD6o?-8G~^4j zsQvnJ*}AD&N>(J%P1jELjgRw&Nq*SUjUT@>#P0vq(~e3QGu;HRO}K{k{?!WEcPl-j z#W}WjA9IrSjTJxFGD+B3RFmswzS&eY?8rXG$9nfIf04D5>lI`bLwGDiKse@q3VY3~pHf_ltz=s? ze$HVin(=2f_4Ja21GBi^`+`I=jJeNChB2e1=GPaw4AZ0nK{PJ6GzCl6%@~>2?(K2I&&gIG?O()y%C!v9pvq*!DTw!mI*Jirv+1AQI4 zdWudpEwOsph^n)A9F_G$wOkBnLUQ(wH5q5!BMPWBof&#w6dPxqgwtV~eTnr;(t4_R zrZEYV%L(q0DsvG!+W2cM(qev*Ck(I$^AKI_eCAGWN0wJ?e+#E4PkFG@^~LXXo{tUK zPMo-@fj;OHtozOUe>^%(t~#jS1@|I(2^-n#bD*Jx+J;O?a_#54@og-O_-I(VZp!vN z{+>O(P^v;UziL0*w49quOYk}uF#Ku@4?lpao5&&%L`=@)yW!4(ewDu4Jdp?dM^^L& z^^r(Q^Hd%}u;zYk$WgL{N=Gr|QK71@h4xCBQX8XdH>4j$Ngne?zAj)?^`(DQRH!un zYr3*Z2pvx#Lwit7FZI|{G1IS!p8>0|aSu?Xt(O+Ff%#i~kR+iUWO$ z7-pf+xk1UafG?{CENu0^-yKkHn3BOnlUNFuuPeBxo~xo8xD0wWTsj6VnzN%gAb#~5 znP9`St!_h`$-P|HtT`xQXTbaSYVvA+DA78R)StfDU=A2HiW9EABnM1lgmZMC{#*M0 zzsvuRSMViOc=+@{(^qbH+bV@W{?@kQWm#RI$0%fws5iZOo-I}caF`H%LX*3|ObIF~ z(>{?CBP11FYzV~G0bZIz*zrKtedaK{KsOH%@AAl~Z|kA*a87Tx?4`z!zVT!{dWb z*oKIYE_L_EGRY^Ls`EyFMea1M_QuzV0WeVl;6`D z0(*+~SoZsg0C>2t0t6wllHcHNpF!e~7^Dehs2GuQ#*O}*`cpY8uuIiDMj>eSVQc=@ zkt!vX)V_gYMJ%-3r;JbUD(&ck$&c?b!CS%_L3KzC1dL-3kW>nT=wWTQg!;|QI|NzE z(4*7ZPzImVt2L^_biGWd0fB>NN`5)v=U*?Hr!S1xQr8TzxQ5Lrj~346=b{|Fx6_QI z>zT*lcrIByI97-u?f#pCqmEbiRQqFpU`X`dy;q1Y&`UFclMovB2hklOUegiH)nios zpXEsV{*_e0l#WhK-VhdfbDpJWYLn&*nBfMBFPPGHSW&17^VqjtDkMyJ+w4A+!+G(F z%JNn2T}fut`>`VVw}%>E=cp#C%^nQ+O4rv^P&`wcRqpwhdbGMwpOu6BMD@HMEVm$e zXhecHVf%+9YTJ4uVb)64(Nd}#U5I4Hr^Gx1RK~3UVRc9IKQXl}61vrPvpE>V`@Wvb zZam*KujjxUoWJjL0b|LyK1J;=C#WF7t`w?%KcP?h=14f(s;`p!v+iTTP) zHGirPSY#lo%QZRjFaAojcCQ{P@C!`p+HJYX`~%FxzOViWs1Y8rcsAro9upiQ zjzqw$K###uEmUpJ?)7C3Z2s1>>k@i*@LR8Ay)&dL2Q?S?49^I?!$p7mwNSQV(wNws zGlv=M4ZU^b_TlZcMW=+X$F+b0k%1kS^1gLTEzRr`)7Fvu!n?Q)7)5nB>&L9R_jk_q zOrP6VYpuOWFPeRp`D*_yjp{Soq=;oF<8kmcV4^_=;>`f9IrI6ARBh4s)SDI-8(a@uOucp%*?X zykxeoP40=hLL9G{)Lq<+_*M`UOi{P^FRgRBHEBnv7!U%d87FfcKXk>Yauv+MGhyhJVm zvgCK}e?m6a9km3DpjlqpCf;g|3y|+ZysM7T((|%e$8^ufFf*Hl(TCKAHOYP|w69f9 zCaUFbvHYr06_Ge0#Uls7F%v0%-<(~F8vE4-;Oc%CcthY{v%`n1zSw)f`r1y@SkZb! z@W9H&jP{(=gjfWzE3y-f?|s0z-{5hA{NDNDn?1Y#q`h*P_TpOjJ0o#YadD*HKHrx) zq0n7lxm#*IC8M{RPBtHh`et{FNIql!jm?Eh;Sr^VJ{8UJ=E=M)+^*<6Kbhe{4itB> z&_XdwVcMHm+c_FKV?8E>kNSOHCVYiM>Ta2E|0_CY32<>bmI?=ZBnO>StG`E+NQBWh zP_y|n#?dOges-z&4~ZdwywlMC;Rm`859@YcBdj(g{z$f3t|0Dok9#P9J!`z~K>6nhieKb8D zpmP^AQ2YkSOqJLF547=r4Qzl>>?8VW*E%YuhL|M)NT7qucjj~6^^GV~sJv3y0CfgL3cg&vJwNOKR z<72va3M%#N*rL4|UPs*i6q=x&@(;=9F$Gm?JKeM*p+R%Bh2m49jQXzs71w07bGYf# zh7pg9yPHb`L)B?E5OVkEjCmiTb$JsDBzps9FQ6~Ha4==CmYBrq`FMw~>(tOOW{d@r ztGmc^`aB~`Ou9aIde?eMyyZi+bpIkEDL*B5MhIJS()(Cx0$*d`KKi^dMuqrN$rU8l zE!&2?y`L4+f_COvo18+Hh4YAiq+||e;`$YZbDBS%s2ZGAK736~I-{`q(yh~}q9S^h zH8jO;^|@m)n~clAOJZhgLi2YODVBmOHBmQKnV#LM^hz5ij5>lHP+GA}XlfjkDI9ig zVzvI1Q=iqJZ4DMATQE$3Nm>r1iv=3CT7M1or<7f`U8reT{jp6@F#>6^aC6ff>km!jOXoXn>@m9YngS;hpzfMy?NEpp6ra_IF}jZU#oFXEmhPg6|>>59}o$H-Q%exZ2O5 zUWW3~^q!>_k(hmM+DEAVJD#%LK&RCQkdCzPvx)?=Ah%0O&tQ(8tHEs}Y3d=JUi*us!M z_*!^v$#tl$tv7SkAX(=RXR)iHOmjHm$QOl>Mcp^o^_@wIKB3!E$fGwH_}tnkrJIgr z5WcNJLfy^9VVJg)yz}BBdS#_p@kha}yG@2wIGWE_U z1bSeUDq9rU4Z`79vvp-o=TjC*HAjPxt;#>WCEA67W?9SkdvW|1o^sTO?{5{GTF-GR zE&qjVjn$*T{OP`y=5VQwoNUF8^*(GNJ7g8ux zG;YCF6H!qa#phdb?@>dQXN>ZvFRuloKi~!fa+5WG+9XCUNof_Mnb*7idq+sRJ31^M_a$!I1U`6(9ieKF~1JXEvU!ihMJN4A;m=tIq%Z+_(wO|OrqsRR~| z{zGK3lxSJn_dyD@HVWHH9dl*^_X<(BWgn(qi{VAATGkDPElF>g`$mvW4GPg8`OjWs*pFpoyC`V`6 zFM2df;%=@C85z0Se!Xn5430c`nVo2m#G*5UG29KLC){#=u+{5C_C2mB{%~}KZnbLF zyHadZrDyXW5*zqbB6^N}zqyNuTFM_eeV{Ap(Qm;dxM8ZjL`gnN0O3MTBK(B;sG|5s zl-U}xUWFB%texqN&q)#T3ZbC#siuxLKV}Mb4gF8|08yiC?C(pwk)#L_4YGVxvm8{mE{>VlbC~vEaT)_N6> z2Yvp}5rM7s5-LwT2mAz0rwV3f(9XUr4=^Y1uSr*Km~uGByWRoKo2{89vBt3eq}*ke zc9AXfD>7jc^2>Tp>=t~5_iFTHLT1*SNTzD+USD*PKD*+oCUki|G8in=pT61?U`dWD zQw) zUXq@Xn(_p2pNsqy+F<7sZpNfhu|E8XhmSPp;PSpT_}ez*vOLnCX+IEgjq6~qPO-RI z!qPJ4sd2|Afyun>PULpD<*S6(ZeGzr^Q$GG$JKRB;01qFOfS{ruTL}^^V|&O;O!0b z7|=YAUk^03fyS z;~&E>#kl~#@=!@|81{Y72PGs#;%2vtNILN?Hb-Oe#F0`XnM~)Q90M18!I(7uxXZV; zOUVhgXgDLTY{KjCY5psl-VO6+o5ityTvo^&r~&lKNbAvy8`t|(n7=tFYX5F|>EqQe zH#(a~25~Fy4^67uBVhZ$4;+kVBc|@2g)zVygMTAin|v3IjB4#Z2XT<2lt1QzfKIm> z0kLrEj0#B*wS;CAu!xrxyNz`##@R+ysni-$6`>8l#@^XpAEGx^;V-pZrx!z5w4fZLB`g|61!6d^tWQAZZ#;txZh8o0rZ-7 zNf$k0EATr_adBu~mFCp-{zLNB?_s^G6_Q2}O9^T3520k(~WW8%8|En>m9^ zNp}9dy*bPC?ri+SmE;qpzfS*)v-b>Z;*Z(}u^=K%M0$x*1q1{HL`oE;H>sgV=@5$a zP839xDj*;rpwdeSy%VH2X`%Pt2_)12A@2C!eRkjdvb)!Q;9@c`GnrpGh-5CsU+ax#h`4!(VXvxH07FDRR2@|u~^bXa|C;(U9k`4Ku&HZ?)l z1f>1&d%~-Vj|myG6N5k~qKb4a)Meb0TlB%%TFsY;Bkh|6bL>VmSJc*NZ+!Qa1NPD} z%t&N@Q@daMsrl4gcPm}PXll`0<8RgOq7fkhn86C^TTQAx`6kf07>@+wh|ACP5yG0! zMgrSUH~9kr*sThZ5Y|%QUvz6aW9@=B@D-F4f1t^+=Uneysz5TA!_-2k3HCbSi;%tV zTJ+-JWgd_~zs@EVKQ-GW)Zm=XBo?F$MmMVxZ}=_+-wzZ=SMfUi;i6S19^fn#{q2GT z21HWjXrh5)44RKV@}#)yK3IiLW;4)Ny8>jcJPW>121L1eq8qk8r)p@%;bxv370AcY zl>f4(y2#Bb{OyXI;1`_a*M~0jOCKJ(MUfhGlKP>Y?fr2&@#~UR0b7Wffr_{+_TO6( zo3dQXN*VLq>F!1=@Fu_SPnT-L9k@BZ6=l$kp$79+uSs(9xnII~f~M%XysTydf7A)i zu4ij~|32=W;AvZBH~^uyDUnQN`)WSCanSTRC))h2KMO${NXSVwMbu!tkcCGVRI$WQ zCYL$!7xsfq!GLaTQMbNO#<-TS*8rK@VqV?);`dM08!U=uvI~L->-M8v(r3Z?R%f9o9HgltTOKh3JZ9iQ7Sp_`Z2! z^g&Y$+Wpo6Y+ZzUXF4X%Gj|&!e9#if$jsR!*0ucA%@o7%Fo-Z9xZ&pOAMelsQdOr_ z6D_xkBRnhDWb8DFZiH{I!KL9W8`BnN9iAXsVd{ouQQs>cm~b`H`)0TbnP&^ByLnow zJXTu|*7fBwA7)boJbE(N2f+xRSn=iy{nR_H(0a4FSk}!)KfxqX*K-^}9lRn3U6Nal zy`$81(L4$n&mbQi^FlnK&Rqsy-h_05$z8z9K3uW&h2U?<0u*Swr7Buta!J>dd-405 zFpr*T$Bo#i)4kwag(8X68&?v)SU>8PW*1zhnNC}hJALn^KIh*$XN6rGv2u_5X8qdz zhd*e`V!fIEj?tfLpSQEs-k&U|%dM{r)m=MSZ(KNu!YA^5lE1sd0ct=WQ?M53*r_5P z^&LENxw!(q^gp(G9N3Pe19E(qZAay)2$1*R(lc1o%eK-S-UAt?2Hy{%-UJ5R*H?Iv z<*>|@y1hUV&zpP+)sh=PvOn^}g%#m6w>zpXy&WhBW%!?1$N&G722ucWfXmbC;v206t+pBfw7m>TaqrZ1%C%cSt zNAFox)!Kc*Xx4ULKV>g!Kodn^sjlH`L-%kDKVjKV>nT)LUq>K;EMlu!gZ9t7J^)iF zyVyjVln=u^KyW>Ne|Tja)qCzT1Lt3TIkQ1dE^^ThKQ{FPKYDTldWS%Yu6#A&;hTD7 z;xK-yUF?K5$gPUMQ5pR!jP&=NOtE9eb^Fe{mx5_yD0xSpREw0n$?b{M;7lu5*IUNL zVP79VdGOo$xMgSb+H~y>D1w7?#PwHG^~CYb63ckMmqq-1flDvjfx2H5iPMF@Uc!)0 z#O#SSWpMvE%VPa^-+C29HcT{6x;#WG-NiZ8TI^x6n!&qQlI^^%T33pNOqWdUGb!KT zR7VF-BGm1ZP0UK$kt}8%++VtHrewE04B}Rly?`)V;fgx^gZc?re8wgBC_7K>mpbpn z#9JbKxp%)ZZ8D9Tp(A#}WJJH65lJ&8#gc!2K9?wjE3M`{^J~-Y~exWlT-0$P z?wBk|`+}R})q$1jYJ?y4GMBLiO_*%h+vP@6B10}}o-VcxG*c7#qcay=y4Ol7#w__o zT7a9UY?~}c-3MoCGFswAvC|arDua0L`gGZk$JURjb0}W>(!)J#Dq2W}p4D3pAAf(J5`i-+vew%4TU7J=fZ;bX&P0L_Er;2pv3d@^&j&urUD|ziD4{T1~6WUhK%na6mdc zlS@f)V}PFRHROsH?tn_cG|OvQa0@kc$8@mq$swmQW&V82SE;(IX;NKsPZ!lUU{V_B z%7F`j?n|#;;W@i;4E)yEt2}2~8b-tE?(^mot0Ct%ST=Mv1#8vnyJ~8cD{cdFm#D-& z+81mBf~&}o<|;^ACMjM4Mx|#|TT=(C`6e)>(dKZAOc6*hY!F1wm@WnI)52;fgXCMl zK$8tkJ#{N-!$d^~@l91PCd|-bM9Y{gl)YFiaB3OlH44O^v`GdV`o{KHhk_#B5IWpU zPBEHG0UexFTHbr&PRmsm*5*bhe@U1ERgJQo`l6^44>My?e422sF1kJ5TIk(C`|`1r zb|0#KerjDpkMzEfq_BC_v@1R-l(Sb+0T4PzKzu+~__r_F(30u)NJ%!IRQ7Y$_-%`I* z#G5p!ED0zZFnVm*r7YR_UKs7HN5-bNM+^ixyh+L8%5*HP&#CbN$4RIyv$B5XSiz9m z%!HZeq7tqc7DIVdHiMnhUt?!Ws#$)ymfJLA`U|fqcxNsbth`^reN1ILw!%-TeXVMdTwEW|EYd+h5R=J0~?G zZ1>T3#ED_(alw6Y*|LIU(R=`94OR~!k(E!rP~GHUPhG2Qa$ldou)YivT3DYnb#$rY4xFlvyZlV;XBOJff7?@g81s~^I=Jd@`#PWxKw_2-h;hc%9R@LFb>q(FJ>}Eg4wqgUTcGmYpcg>M^;D2$62oq zDZ`lE-9*2Ly0JO+qAbX-_;`f2&IAvcrw1p(1r2 zBUHqpYr%|x?C5=Asy|Oh8T_5sf9A9T5)~?hmibgN_bpc35p(gxJg-1NorK!AVQ`w( z>&E+iubGDAng|su+vpynImstrgYI+NianGJasA0+bBZBD*sy`=ntf=o@6j~RR~lOh z5tz85h?tz-QqW~{WbROLo^YF8qGYyZO}wILQMf?))1}a>w;DpoToUsbe*>M2p1Mf+ ztNC2trloH5k!W!Q-*!GmSOzKUWdvLlP;YT3eygB<=Qe2-i|})e5IIhV>1U__38-eo zEe0nN!SoT`+5MboPcCc!=JtMhXqdADM^S){heS8G@*DSx*VG$ig!RTQ9ej@Dt;u#S z(1J(5d%1n;S4^@lW*dDCDDC>-{g0bfw|l5dNs1O8f+zIt)-lbO62B)rAB7v4AwyWa2q}6MFJ6N7rLqBgd(?g5UPRm4T({ zN-uv~q0zq0-80ddFwl8cj`V)C!=vBt>v)P@glq01u zplL^)NDCy-o~<^YPG-xqqd*5e@cp0a>BrYb;kMAb)4uSoXwscxSO8;ue9g1=0?-eM z{S4uuk|Q|`DmU)Wp4Z;p-pX%1!U-FO`BLabx=QGl3cT=xfDq zOb?X5oMzN8I5TLm&p2`V-2C^^!R&;m4Ta;N@{AtLV;g#Ho1xtF3@OmKTWcU!?=H9% zOIeDCVS`UQO1DC`r==M&wXe%>PTn@ML(0IP;{Q>)Gd0rvoL3l`?_X0=R`-$E0?~tA zgt$!bE<_saVS|a;BPlvrt?c(1(gd(TUn>8e$7lhwpxHi3Eb!Hk{C&dXk(k5+e$+@# zhV+jf>QDV~nP`;en1uby?w9S2?|)e}$kGz;LFn^*d(lF=%LjIG8u4~(LcJFkkSzPZ zB)5Y(=8Vjr7jrR3&|QyHX#a>wh)Jw@EK)D6zj95y(_}X_Z}&LQ*i0H0En}{dd^tk7 zaenJCosKB0e9_s&xZE5Q$hb`N@jN*5N0O!~W?0(;6u*^dw5B`7#cm9LdPpUu8-gC? zj(=0`drhLvoCoZ=;;!qXIH$LV9Iq%0nA4rNeifrS5WFXlIm1Of_V|<(5s8Td$g^Xs zl`pqR{_meE$jUsf#49Y0Qxb?5r9~M_5&Wk#K z_ZPDopFXr5e^94OriL##uDwx3x~gdhj!R=#AtGpo`dJMfy<6NN{VVuA z7?~=5Y)J^*dG1nD{~_c_H{BJ-_QS3>M&R23EM{snhS+=D&}lDpl)o5xN==hmYYEmf zcJC|Qi!BTkM`)qpwn|eRFLf%N|GZ?nA`|X*FcE+~)ijb7adUvdTwSkAxt616NeyhT ze7@%vqeU7B&n6B9vsKkaob7K^H`wb-$MBJI>)v!7itTJtUxH(t7oDYEjg=4-myO#Y zD({6Rw7#UT6x}JE7 zv<6BEv!s~0z)v~Gjxt+QN)Sf~r3cQ$fEuQ>E6=0#SE(1>ueb*KAd?$~ie6+dXg-{r z9KK5u;&I5|1;pq5<;66j<$Hb!aw|}MoKktoPVX+SGjdS8qw7_0kQ-|%(1HHF9MojG zm78G)U(tMCQu2^DF{J(8RSCU=AM95zu9@3J+Sg3XGh*8_KvhP^RNV90MW&U%))}U_ zQfioQVKh@-Lzr$fD(m5hiTS=3z1QA$66ew$aK3)ze;B1()7fvUO&LwCO$)6~ilcC| zuhSRZkfD|q{LaQ9M(Tt9z(vZ6pt)0JJX*ltW3hYuAdvCboXKWPpl1Hm?1GVnqxg(^ z);)XY#p?K1+ls@p+@j?_?u*?BczwQd>N}n0-x{^TRb5jV->HVE10%=!u#sVtQv2)&7S6ApgmO4!4ZsNniL%ZlWDk z(d-VRy4ThAII;HC!j9Mn)K7(nV}Irtjz8CI*bWGkXU1N3M*cJhU@VL2#%ZVeFx&<8 zE@{1Q8PnknZ8kfQ$wpQG)dl4d6YMsJS0(ahrEQc*Ml?<8pSQt=3cbbB(r^0xb^-&3 zvqx4&I_tbn?m{STfByYympxo7{FWvJC(^l3x#R7x`9aD@a@r-)b#=ycfWwgWd!HcP z`yL;;ddpgi>2!ykpWCT5p@500h+P{i<{hATay|8ExQx3!u*Y!GM^~4hYn#M=g?Sx- zKTm zS)OxRjgjy&zE533TQ=%q;N+j5C)HJy6S6%oa@SGaHSpjujvDkg-Eb8`QsgR(usY=oS&`&ypS0V?Dp>fOJxvN+`yah<-9OfG94xv z$)5C6GB0PxFl3#%M2NKbhaZxbhRKKBvRkI!8!42aF3&n4T)gZJ<%!_B-||~xa%sP;p2XOQVwl@4*FxczwMNpazQ79wlf81zgXb`Lp%~BS((R>E0g6GqD&q zv~W7a#U+F-dHftEa^XYU4(`>Ud}3vbyEvOf)GmK8sEoPpYArVJaLLkHT$4xG?}~E4 z7kG2S&8uQby4#3XVLLEop@PC9P!k86g#g&*vTHaA=rB{+lP!Q#zvVS|Dfm2C$eucF zdi|9QF7*s0LTAKhY)U+npvBQA9R!Yq5_^zhzE0Ps3orV|oD{1!=@>QNcodVO|Z*RNEuPl8n17Q>#vIE@(EON!7#$)zy~6<~`RlTQ{?B!^*KW zU4jVVWpOWWyJ&CXp_S`O?J1IxF|2JGo1e0&Q>5ph)U`i;$Nqx;O$*t*2`PZGrNe{b zwL5GX`VNHE6A4e%{-~#p8?W$kmZ+s5Z8hrWw*j>(@JpYbAR!6FTd)hc@-KnL{_s5- zput|}&Ke)75baK@3+0Z4^9^;@k0fqJ(c;s8*xvr0U@ry|tiW~+?amnw8f{O}!r{b` z5=Q;Y2GmV6aVA++&*y(h@(ln^uYdHN!d*UeW^0Q0`$$!03EDOC(uZI}BAH;qLV zHZKey&o^LsqALi}KF?eqFvsD0FGe~iY+3qc3)7o@)1{>cefl=Ks+~P4&9_!zaJH>v zg#`@1W50O8?^jp9GYmS#nbUM4V7ftW#e3E5r#d_xL(nAUZCdMVKXlTICfPEoHca3# zY3dc~FrUwyhaZsZFVc5KE5Z3@xc!J&Qyfs0trJhgtVq$Z{~7+xMYcdOJC5q;Ezdva z_~Ul8xmMo#9Odws&^Ty}7sOtu;F`k-5U>1>lOoL1gBA?p{+FDNlTBm_f|KtZn(51d z@=YaoMO&jlsVdydc@AxW4#Q?=Ua)Q;UfVUntuDcqHul2Zh z#FMCcs$(8$uAOcoC_Z$o{SF%$tsuwP!N5{bj-Jt(@$#~9?OiaP`5G6${ZjFbJF^jL zrDBf~=c;j;=7N-Pa}aJ|ulg#rpEn1k@bXe6Z7S+4D*GL`3VT<04N#IxNP%&ppyT|# z_&sxV(}sFjq?^M7Z*b!$m3R#JWSBO#J_5W1(7I^2MSEkpAyC&CxeeE8dfCbZ(EQ}P zASN`w>CbuJ$i=zpaBNIlX91$ee?rWJzv{M9nRB{JjLwHY8xB=AGQV_^7QcI-GRZxW z%mf)cTsV=ubPUZbBCF`l{7oXRu^w7Q8+GN*%q}0-dZHha1%-d_k#F2%=TSJTZFBEE zrAvDjPks4|IAUw#Hvq&Og=0&ru2(S(ZBM*y7XoREW|bFI`-rKMAL|Zi$+@ZPc*-rb zgRgsn@A@w2C%i}}2xdz}rw=4tD|XLDSErE9w4107A4nCk7cT7hk7lsrAxm>!p4#R3 zOdXy4%|piniSZQYTkW48f7zYm+5UO9;$r;8YQE7H6rI5{az9=WOt<~^<>u#m5w^^< zU2CK#9J&IRF*n|xp4dIU z+EZXQk1w3i{wP!6baW<6S_!r1r*sd8i!HrHY;YDS@LgX)=!TxSzb}y$dE>S`^K$gq zH0>r|^$RBqn~%;u)y<1naS=rioOf=Yh5vcVK}KY>$X}Q?ZBz~(-qtYL6&ELGb!qFu zrZ>7x#?5Ef;D-ZTU2{y4pO7E?)s;?O?pa+^PEYd~>`8b(!y7;Hvsp&nqp|8T-DzLe zIB#nK8{b`TaNN|OSjY-{7U?F7e)*-$bdWp#p5=92nYZqF@v>a)I8?HaM0N~|y5H*O zZ&yNEQdF5_?d$_Ya6Pd8mGlVPNnD59x_MB2`a z2jXdjjl;4F4P*|K(RDdw((DF;9S9nga-0)`|5`sH8@U`g__$%tFPb#EneW#7w*$&0 zZWGfDMb1jS=l#nKDj42`eN`*jz5$Q9PZ9@C3!p}cpIiK{Sy~5w3Y9b)C_(#lJ7x@d zxWZHYwEi@cdg@tAx!*woOXOw;wr8=0!J@v}s&L9}h)>yL2!(@m&G*$N>`@HNe zUC^9B7V4$*%&uy|@uP+5tM;Dk+DfA(tElq&swDb5H+su-udFu)VBGe=qTP#^@a~c= znXC+2hiy$o#SNyJ@ny6DI;#&kK)YSUUWp3K^h&cL10HtF=Jl`_X_7 zAKG4a1ANdM@EUFeP;|W<0eR#jZSfWTZ^?lmv6^ahImk}Q4!D1tF}lyTZo*GdQb z+#CP}dfe@sKqPm?_9@qib}hph9+83MItu=h%=lk2q8VM%ZUPv2W9V|ua8>P>*cZlf zp?WJt0HGbYjWq_;J}&_E${2-XZjET;c?MgPplKZz5C4+)-Dfy#Pz?U>Uq6_|()$7pgl0PO{*rKwIe-_x!Eae@EoU^iTA-^Sv#^!x)d!MEO_fH|YopwS)7#MVV}~wcU@;a%RK125tWFK@g-ZGB|Pf0f{ocK(I`0 z#%jaEohCW%$N0H< zQQY(|2`*v?isyy4<-US1ap9CNiT^$M2pT3rHCA7CR5CCpwH)KX+?(By2?9X9>-~5i z7Hi+g4$$d20HZ}2@@x?3ihfIOP%P8Xalj*BaH5#to`aAFNH${KOI>uGLuQO$rEgl( zOa4(Bk$h+It3n^rH}-wz@eo{PXH6s)bn9q849<{;%}w|6aI;~VDNWt_+H>(+eCXEY z`w`E_Dbid2sL%gs&*YAp1dlOdV=^>jiIFC?H6?k9)AvR6&5z2vOb*TPmj9_WvzE`0umDUlM2A{LqL7`XbeX_Z`|*neJmOK{|)M#I9fl z8xZdCI6!A>&IJs){uJ0MlPU(HHwLT*UFFzI2EnZvbIMXmWqQl)P8p;0vR;I~<6se) zQhKE_z}iOzegmMMzQfj1sRMP?eW7G~JfA$t#>LHn9J3Y~c@SdaYvA)`T`5d=>H;v( zkSYRcrB&W@-X{SmsSSPJV%xa9%f}_t?yep^-*ppXsR6*oC{WTtD1mqfN`E6&4o_za z$3^^PB+$_;RTTfmCv+$-2Yv7VZ%tx|H-6qwQZU&#yVcnME#FbB_*60yy4@B9e>zb4 z-P<}p>k!CHJGqLbN0k+AZI$4^^jKeE%?0R&&*)=8yMl{bq7B9q;p~B%;(tkg?QFs6 zX9x2LhP6J>BupT(7ujZWc%_C7lJ*qnZi`T0m^_OIPl~j-zx8SgAOif>z_!@_123fs z$^Ol<@Ci(g02)uu1*yZV2t``s-7fS3O&06AQ4C$ZL;qwV4WCd6{u6u#SXxq$EMWD! z;u)R~G%ddXrsm(3!$^dXQx|;=%vLhyxYBWK&(1!R1Nh_h5iIvC+|~Ds4=_Wnz%vEl zcyH)?8p0nAZlDwIIDj`>qDHGL%_yS`Mk^KRkyV+7im~A!B6-T9qFL;1->-mB03Ir; zZwEZo8P5qmi}n2y>-X=O-N=UZGTl1*6u&&os)uq_^4l`(*JeNll6JOJxwgNAPyG9~ z`m4Z&!M;MQpqb$N1lxZb)9d7NJp!0e@d_m&~C&UE&AGDTMjx%vYB` zjt*r>nWbd!Scz6BVAuRN3y#^^c)o?UlPAPE9O!?SUj|3f8R_nmLieKJm2M{E)!uHa zgTYx-T8=Z1*6sTGfXO|^2hYM?@Cn#|yCa$<4!j6JmNYZKZFva1|B1lkKuD?t>Jjp- z^3M)apb1kKF-uz^$a=l>YbuMoT@TPcu0^VmLOd{a;q$`Gz<-xy0Rw%#6&p5UDFPPc=N zJt=W+#?Q&ZTSSNqfOzd1jFx>Vmz6OUOZ*J(MSZbNvgK7}22f_F5tfIe_2qkzilD(% zVPMk!l8_+)`1u2};}E2n1gNqmJ2QC8QzFBai*%okjmw=#Gx(NVklGM?qrSrIXn?T~ zzn)zW8rUHy4&>YfK@BoAYf1-fN5z*5SZ$gs@FAaez$z*Ex6U0=)sHNHCBAet{;2O& zkZ4Em-Or_Q#=Al!-(O_n682i4{Y$d211CzD^-t_VuCBz%J{@=hD?86xv z-%i^UN3#E}VSOv+mg#sFywf=ye$u9Px@R?)Ue}EjWB#xq$m^*V5#{2|m|42f*7rDi zp`GL2bX!#azx7*?)LgpXwliQgaeVlH?A$qPeB`p?{-?^dDB1Uxzp1HZ45h+@xg7n4 zmSrI^ZD29cZi&rdr1ajv)6a@#d*G$eK-wBEvkAcX`|q_}ih3@{@w~DtCKH2~)uF&T zm?#AttT@-@1|I6bx-Nr~-q%Lgy2k74a%MlNi38=`+LJylaR(CVKDeZ70E6>(ab}rz zC{Wo;wTDeYh+Q}>23JXNjTSchgCZORwv^%=$RMYn&7 zs2?AB;xlxFklbFTxYV4#`|w)5)sYTV$s%Za-K)FtLJ&X_8l%8344Bn6A0mF@sy4!x zYr<04`>5qDsa$-uf1kTFQDQvY++*UzV?|tFi4O$m`$yR_+kH4zvLQqpV`H*TB)@t; z*!jnmbFy-GC4fV{;7RWUO$e|#066rsyg>bkgkWT(uVqC*Vu0n(zK5TFu!V?z1m^@e z2KNx@Jwgk2u5IG}eb@Dc@;)bgcs|f{L|6m9uvxTSyD{zeq?n7Tsvv_idqsxVkA>Jx zyz6=vu>(Fm5FtuwL#Jq3qg%$jis2I~&Bb236n-($>L+bGsd-FUOk+i}2NIXD1VF7m zceV$eZ~YtaJUGAsB!dF#3m3WpBh`Phi8J})1%2Au>{ggswBsNkBR+z$KF&}RAiB2* zOo3C0wg(i!KNH&1JrFvQy3-pfG9cw$k~-4CP;MMq{f#q53r6-yS?$-Lc-n~~ zHY!NM+zuBH+G2n4mxMBTNq)Md6pqP;6I`kqb#Uf;66B->2Z$jnAA~!wHyI@=>%o>$ ze?p5?-iFTqTox)T{4o`*J*C_lci^RIg^LFhrC!zSB9^qKON#%yjiR?a6+Q*F><|{V z(*ny^ZGEhnXnO?c>o89#J#AMmrJau!G&%aPS@*Un4lRK1aUm?3Xqj4e-UEz0t%3Dr z3PHFL?y$XZdH^PJHtDL@X$6e&=YHL|H}TTZtY@Tb4d~X0Vu{VoDWu)!h7%O^6HbD$ z2S=k4oZy96pm+zVQJbXwV!XnI?+eGrYE3UOH@ho9w3DXo;@w4X&R>$C4#4d&%!%Dy zT8d24;ms*jVy%rA*-ji{ouV&%Q>Y1ZJxPfDy5ufC2;DNQ(?AHXT8LaG2GTC=->61X zS0Gf~Z{TG;K5tUkr|>`a*^xCitCs|&Q38h&r@V}FeSW2 z7{y3X$NOaa#_Y5cg+2yso;G@)GW@(kcrB6sYdZ3h&*D?_c^@y+=?qkl*cyC~IEEK; zwl|`)Bx)Zbw)J8wG}+=CZAWDNnipdi$ftlFn#L>7_k^9nXDq0zm&Ni|HZ!ll?P0Ou z=j@o{z4*AP=4la|=sXU0ez2GF0dl^c00pdrFVPFtmw+nDeF`N;sI8c8OFREWxLw@F zOUlEmB~VK6@fYt@27PyUS573bQQRw#@z`+{c(&JZ62hihtc!45BK7JO(|hxb0l8Qq z!^Bg0o94@}jUx8s=6XipeOX}9RXD?Ko5?HFc*Z2&>~R3oaKunR_N_!PY)r@|LSbdC zL--{OV_qYj2A>lDeej~Ls>zkx^UgmTZ6_WF+1*)6r}3FOmebYnJme7B3~69j?{wdC zd>%|j-|2{wVM$;|zsllh4^TaQLXKkED(ik$SO0`~D1z75#BOLo&wPQv11cO3G2mHN zO=y-bOi*nBFrFm2kA>QG1Mt?Skp?p}K>55N2pFN*w*It9F+#BnP%J_}ZKK0+Gs2~c ziL3LuOK_sFXj!El@2&?+OrO^eq{yM{{UPPCk)n5TVp z+cFc{`MIMPQFgFTcq+JLxEE-!LK0LW1V7_Q<0IS=S$d@aUlPE*?>hbO9BSJnWx?aJrU>OUAhdZP!rrft;3cZ>kW2AGvh8@ekWgR)VA=8@T^6SE~O8x&2V(=0m?O35%-ZNZy)i z2p9hMC6%?IuA%MGT0d*pT^(J=|4DWIfBGE+2L|;w(3|df`3`*3s1|f4rKG_1FPrW#lgJNJM(omF*iL zMNH6@5qazybI(z>CBaL$`f4xcrxRd5zFru>zQWYrib^<*)*4wo3o}F|h-j5L7gM>9 zK92H|VHePz7YcA~{Y!E@41xnCdD*C~yO0KgAZu>GnWq8L_=x@1Ur(eUWJrSL*dDK#iJ$k| zjo4(~@`)AO3Whtz z4R9KKgy54gRg7u{BO^Pr?M{nhNrubUGr6J<<|-DB&q74MPpFwNaqMF+eWp7svht?)=r=%A zmWM)1NL%zSQH_h*O1gow!icz$3riY9C^WcPL`sxQs+NG-X)$p{!KO?eX1yU;laAlW zKG&Jl6%4e>W@iOdUkW;-5q4eYyrkBB>(l=vo z;`Q)x=1P$l81IF-L!$)>OhNEjm+T?Nz1a3xL6VO0MNX^VuFQKdN{UCBpSHpXpK7I0+uTQ%W z>9LTN@>rH8vf(tN{VM;P5AN_tiE;dU@1XL9i})LamJ;uq!LI@)5_EQ~ehDhT0s<*d z@(~2pJfQeRl_88ZrRv6$`&Nr@0S;D-UVHVl^0&0pdcZhrhHjYwDcdGOR?C$p2^G!` zrrsUhAb$F)r_N0psd9N*@iL!Y#vRmkX?V=n2qoXMt%XzKV}`4wkE+vO?r@d!K2qD; zIiGNK`c3I62g~V7R*=@THxr0Rx4uy^d=fR%J{o38T&gZof$&Mtq-!`lptxcdbpOR_ETKKYty>KE~J4ROj-Itk|;~ZBelfn?hvoRGzc{M;|{saut zk8tcM;6z3IS^#iXmmR<#w3bs_776jU*6D7MAC^9YFGv8e5mp~C9~rK-Eu4r0Vs}eu z&QE(c8ekN((f}GYef~>w^z0TXp=EN;Lork?AVxf==#mOZ|`I);C}rRq{!l%G~+WHxuT?I0q>>daAP zWjWVqn^_Ue=Ox~~lR9LN&$o_U<9aekpH^xzol=4MY?KfMVPA1&GWB&;jK|ICk)(Ik z+$apR$MyPx6(c0R1Qo7TzYDmsa0cmx-j)r5uS|W6qqPeQzKBWLiXCs1^-@b@$x$!Z zdb8EF9M?_I#&@SS9rx6HG_6aijUz6y9i47j1mw(Yr>@36h+6zUQS7ilfl&BZYd*Md zu9dvh;QUmstep(q@V3@yI5q`2wq{}oI0~R}3B@X|06zNYTm5#y)dzxiwC9mOKEz+d5HUbHF4w-RHN{%*jdWU)F0SIRzwJ&WA*_D;5+7_vr* zxJ8e5>J)g{2DYuJi&`x^YjI86Uiy{zGgCE=h}gcch{WY7261(u_jlZGCV;enCnCln z4xl-%BcyF?(={MM6SoN}uV!85&5Fy!C-wlTKR=O$iI9tT6TJ)NNHDDg2)bJUKJt_} zb}!jJw4QcCmz#Af>JyhEyzhD%fYkiEv}ni0^1DCi(4OmhY#A~<9BE8l`Ojmg{q0BFN6ir=lW`mGYy^Hd?ec8hmW$Cy;qu)_08!CHJC z)}pN|mGT7LO4|EVRE3eun_7Toazm+(fn`6;OB`h%_vQY6i0C;aNY*5EEW;-7dcpVW zap_(5_Y9=7f?-1h_@W;{1YWD8nEhvlbtrV?jA+oV$-E)4J(4EN-(uEU8(r6LW1)P2 zbu%X&4bUGqKob&21_m5n{Q z)xNI%3;oJg#0ePBH;_AC9n3ZSaV}14fBy?`_Z9d}eO0fuxG!y>93P;NJN;}pSjsm` zgWD4Fg(Ua~Kp}F(P&qAI{L)!6XkiN)1;y&*hy~9k!!9?Eegdpn=w>@XTrxnS+X8oi z<;rEDV~I}^0Ym-uQKWn)E;@PA#|4nPJ4l|ZO_Id zj;ig0&A?M|<8V-(YJ+Wg6-+;#Y&kQJZteTVlL9Z$F<%tUxQS}`aIpQ@p2xiro?sy& zSpcw$WE}6o?>ciRCB%TwI~(bAYV2*DqCAYHel|)<+{axLc%K0?oG>+3B44jHw7tIk zE9Pt^xVE@1y3WO4U%WW#rgX{5(($i3M_qaGwa!1_l4u;81F4xX)RFGObyG%0O;?|8GAa2)soKYS%^O`(7yQcT~#!pFcv&+#b!y^Ca^>K;!Ze9%%>%AL+1`)}+!3Vh&a#VJjBKzb; zhVCw$b-729D%(efm)n=H1$ny61?6MnFI3R5_I?uTJzm9477E=Kj7~SFDezptV*wA~ zU@-D+3Ulu{q!8F7xD^_7W&_T9{b}ks%L| zylBr95mOAKlX*`!g60!TBrI62$@`*6)ZqMD1Ul^c+xHl=_+fkhN{3v-di3i=(TI&> z8j9A%^6xBLEYA=V^sHhicUYp9xwdMnkkMgER_3!Ng(+G9?s4TPc$_q>lK1g$KSV`S zcW!yFEf%4;$Z9Ndz22rvOlcObHZ6wQNdELAtB_SVM>T8EiqE0l?}`!#SFh$A3vKuLvG%sk=9o^#>ahXKCXuOlIK?fT_DlZg$B$ z6-K;&oSU3?IT8gdgfOpq+q4EY%Yv=y%m?yqXSX(YlkKlIdRN&^n`Jx+tn5(cJT7Vv ztTK77Ur8q}oV@m=gsyY9P1B2bQJC325#UEe6urk~5%!vH&c@G)^ei3>(PH zerF%#WvK#4&mpXtEtJmm!s9$5^!Zt!h!6h@+IU4)*8JAg5quDcps!Q&Lt?&Kw!pnIiIC`Wkt5jH$HZS6_$*um-N0!X1%ZljB}>8^;rcu=@O`Z z>0zyW;@Y@>)-Oi8j5fP~d^0ndtV-R4Bur}P;>T~}^RgOWLNW1#JmPghZT(>{$f0FT zgZ+=U?qp`3B>`%9FOS^94$@s*VVU$?h{t$C)~ARFQZ3E&q!jkujHk&Azl~2EgYM&((3Mn^V`_u z92T4~6v5rNIqZ^Je5{XFs$PYhPp_*QQhds@PqX?o^8~DhQswdXI4%pa?&p^rdO?gSrgbBWD7QYj41+vQ>5#jFrQoUdEwoGlY2o1iZ%dC;YnX?qc zufz)Vaf;1_A;*Mw$2+C;Rb-owW4h=qdfgy&S~8g|@D}Y!z(ESdV!=oM<0~ZZh6BC= zn-Z62doy@yY;Qfi;_L~|FA+Qez}|SlTp(5r6q=s1d!g$; z=8E~)HJC;TIhq)nBVaQ}^m)%D5CMkm6g5}N5DkV^b!UoJdX{l{u84XG~Wf<5~o_rB(uAVo{a}MvO5JzUl{wg2EAA| zOSNXau4u>7X~mHZky}jXf4Dnt$mGK|&evY_X)joA=gV9W&@cvy@Wv_$4THI9Yx#XrNb;{VD0#^X>>HZZOji5}x6dkhu%Bj2@4Ldc(L4^|EtQpV(XS(D%OfO;`fh z83TlF&tu zIUq`OaSdm?qr~3aNCLvuv=kPgT)ebi7o!hacyZ<05g6U%vz)Uva%9g9_M8i z%xN6|H^SaCoX!6YAJ(Z=swirvtyL5)VvCldC~C&mQZu#pOlj>^lomx>dqiqPY_(z) zwGw;9jukV4e4qRGjJMDKMUEUtk~c{{_kCU0d7Xm@3g2!rErn5->)htStyA00S!`E> z0E578zp0?f87((4#`bLvu{wZIB}6{!VlTi%5xv@gTl3%F78^98AvZkUqR{F>V-?%AqtA)_Q8rq-WjVU-l@4s?iB~jDGz?poLv|^ba-bz%dg|9Wj%~m9jtT?TDdTGg2t~>8V5&4tp9v3IW8PFTa|ny}D;*~DOf!?({-e5f-xjpp z8r(s8GcrU`*1X70Kw}US3pl{T?fh;=EoIs)2H0J&|8IA3;A>BBAEJ$PE<6idlL6r^ zxPQKnH*Fk0J)&N|O%3WX7TI323P7s9Jf&Eg3x`)89u>Z7SDg};TRK=Hrq2z-8oSY+JEkhc8!QywDsw*U|%rqb~I6g#pr8QPBdY6y+dxGu6{!Jc& z=~ssHZsuz5CTx({Usy&q!KOL?F|WSdYIEb_b8&xZO9M*vfJLZUQ~3CEyX}FXuWhpm z8jDJd(qFU7^OVBMm))+7kpbe_{9=&aQViJpHLeOnM2|f|HYsQnn>nt3p6@AJr7d0K zM(z2XGeu~fziR#Hbq`B@s#{^)vP0@x)Bq3B42WbG*Nr_;9;U_rsBrKaiBJDabOSjB z8ZAn*d`&geijPXvBu1kU1~13&sh0H%t=&tm&P3wHVv4j->C=Xzq6qK}!?Mzp=Ux7a zt+5KKc=%nu+L)pl$@mH5&v&Ss?N1ZyigPVUL8h(aS>oB-nP{Ycgo0e-YDgaQ#5ps` zw9J%bJc|MwHzr>{rc0qO@K4%~;H1!Jg6Tenb+2i7Jr-Tz+Ci6owKk0LDu!m)cVA% zjLd2RD=i?%a6wokY=7xZx6KVK!6V~ns^7;k>WKZzYnDwwbK`dw>>HBX_)2!HukWJ$17C7g*@jGzad|OS z_1jht_F~uGL>rhiq?fy3=q!Trr(jt)h)bt; zY01ua!`0QTK{ycjH|vY+XhTw|2XUF7cvLU^69+I?Fs7!y4K z+wJr-SyV~CC=3~h9j;CdGZpSPBACk2tfY7An{&BF36Pa>!=M}WIbw)lJTRyubP*wJ zdvO%VeKJn)-!Y_gw3|$8t%(o4uf}q#h$sK?B;cf(x#hh}B1T-58S`c0OS5+`?6p zk*W4fIM#FQM9{!jc0lZP%1R{v0TW8MY<9r8mFGPKRZz51d)&JD;Yxvo(eK!8U#0tF zy`7s0QudAw@EF8gg&9@y8KqaDU$n&lbNuTB>hWYQqwHDrD~OT)-NWUZjxc zFfe!u%qfkz+0QF!B(4*KlJ%6&L3=rY6buf}B<=56jJIx0ni_sqabY_zgLi~-q_2Fy zz#MZW6xp%@3I!%(CKH8+D$0$(5b8j(6#h>YBIdp?k_&eZ!Dg0RwWYJH8SWDB8ly0G zP~uoenj3Y|oo{fT&xO$!oyyHESOuXA;h*dAw$!pFo*nyG(Z>9^Z@pdKhS+$Sw(>8D zz5n!$)xQn=Kr6g~E+e-$T;E>&t{4NP$BOd*Q9U!es7Pt!KzjfQr~pDD;ab~T+qHiw zzM~Ly{wa<2k--=KEC?nl>0rnCqS3p2okAr4M`d8fLTF@OQ5-5YXOD*az1G%OYc$Tw ztPYAFX!?)J7}M0aq*lO^la7=`UbXAXFerV;(6~1cn6O5!MMB1Fcish)<>T~{EHIGpW?50M(`pWb;t}itntJ>8NU)im5A%v-@p4zPDbw5q7 zl3FNrH^dD9-c={v>ZG7Nvvo7W61QUOcNTV4XTE?FWbA$#SU+w7wF~rZEm!I? zF>R^(V_)9ZUQ7%XVVjs+od;~3w6|9debY_es6%bR77=XLolHS5`6;vhk)q=&~qlFLdBWfs2ZcJ`9 z+k1ZlwNPS85nw`HkzwK{DU9*T)hW$X} zIH%pw2cal6ytaMM!E0#5BXIq2s7s=4_uP{><-OUXsb`|xSRT>Q(J9$e`I7a!%~1Lf9W0qOoh4y=ncD9RwT}yz%)Q|c?@VwhnVRf9j-23DesX(v z4cq*eH&0Uzv#O9%e3+J~2Y;W$KHbY(B9bK;C2~|g3p@TzjBIIirG;4Qvre_-jYICN z$313`%#F)yf@gj?&(}rh@`crt3zCijvnJpoEw*W>=Ev@accj|e@Z_kJh2e<#HT)_JSYCt{P+VA z+ln9hl<9JGQBY9(4g3u&#v>nhqp*819#M&OVro9`% zG7TQrN3=cpK#w?2SwV}YN>vZK+Y}`zjWh6dp9C#UC@Rs2+mhZF}Bw1#Qs=n4!pfaK^PB%v#1T&-Hm?Yv3i-#L+*2Ph0DE0i(SP0J%m<(@W7#> zx@D4i`ji&c8(!HjWcJ;~&u^0d3_jI+kaefWe=11aX>aR};r&laaq6Knbkqj1b@2Ejm!p$TC zo=&XGigoXzW%ikKe_#%LitUvpW&Mim1{s&(f7U+rg>~RzrMb%_uTS^g;$A_sH6K~U zPIP9FZ@-l=8K?4Vvwg8WlD|n;chrXibVv6wPRzq*x~XouxZcK>4ZGf?!xvT%IHvKm7%J{KRjHJ`Y)&serI!@_*0=QAFHYf_op6qbbl3x{ zqKCx;3)6NB3FGU5Uu*=qJi_VT&WPU44Bl64aMv|RbfcY8oA)Y!46dDB*WjLdwCt;# z_3mo848YJDZ#e4)rv_i|I4ZXjunpCgRM^FEg8AIt& zG5h@eNI9Ea_R*d$uYTL%C}nhtgx=q**SUigM&?P)R_~L{l2_o|KT~us_48igmsL3& zdW9R3Q|i>crnlZhBF~=}bRnFieGRVduf~^%OwYBv+jEE* zNUSea8h4%KBe5H!Rr?uI{Yj3 z$=iLA)B#6zCuXWgW4}MHHdmgq&H%^j$o4gnr4nyR>mO$-p;Gut16>;tTUxv` zu?y7gDrgh}bq$A-?_t2VWTb;XU-?G%J7OMn3cQ^LD8!8RL`#IsCM;g!*bS@J`ja_! z??ZkhlW~$O8k+wSi`*)^WJ(Ov;}7OqU=NJ<^%y7szBa4ZG zEHjc4)89SS@1NN!B?@6v_oRkYo1A`B#;)*!;C<9b6(4m*w-scIL&EzWjhZ$HeKyBH zTwl(-RGrsi1AMI}VT0ph^qJyuR7iF?&B3-+vseVT&O=bs`-{tazQ4e3G#h?lXf~IBsU#@A z==^S((%#fmq0-v)D}d6XVO^ek*@vk#Sag_QJRqv~R^vXhx0NJl?s94GWYRNovrF}OC2DaWBUMnupk;^y>Y8a#f$bZ_w+qT*L ziKwHbwk&83i6clNzP!2?&SUy_D`xY5!m*I&??Z(xp3Mh3W}PBJvmW<8{5-+)bJjZ5 zx=N>}bzsEMQ{AdWgVn*L7qT6jHn%Mn3<^}Md zRU|iDnm>B@OTT_nCv`LAX^~p(adPJms^+rAUDBf~kB4qpHz|>ij%$-S_a^c^(OE*p z*SeKqOS;qgi@UD82^gAO;?FIB`krwx-)70(ZEPNA-fve&>v0guJm#NwY&Q(HnRlRb zT7`QrH5Dx)6NVzr_3mU#xbSA(3gtN$CV;W{hDr^*OK&`Kq)@z4Jm%o)e050F$HH`| zp0s;c6^!0+bbFiUGWZ7Tx$tywNQH)f)ujz9z9rN1IF1-yBQ>Q0s8o?cp6$wRG}PJ= zR}n2sGf|<*vWoI4OG3@|Ke%5l0m4@Ovd(NEb82UIe|QHADDsivdrFSo!H?Tn9rc}q zN3~afR9?#N1EK8bzHSf_kvme67Ss%sRAU)vhudC*OS;V3i3PkG&2#utoB$|=RGd5s3JoBIwx zt!Ni2Pps?B=1wbp^5)aQhSl0$l!~S1z^=eqoKO{@nf(YXD3CechTkwL7NWcaYqMa` z2&I1^h-=oae`IBBxzlJRBC*8f>tE_I9DD(GFHTU=Yp!lmB)nGDCzSsC^M@=x&%fIx z3jbn&cE2KPqPQzV25-|)T0TJnqMZRb>a6+JKb z!K&G-*k*B>Veyd+<>ppECvTiV9CL2&@hk>QYzz}{l~HpNjF)mT%^*V}qm%UIr?BXQ zZt(1S3m2+#Z-270KqOi7!}aUJednfoGs;yKlUAp12dTvU$cFVL8Z%@XNsIfL>V+g| z*S`TbHVY(D<`2?n_&=&e5GiJnu+3GmkiGN3F^B6vDrue89N|F)>b-8GsKv-?k$jIC zm@X{Kb2$fJ)GU7-uxA}x4mK4P9bphp;ZOJ%t|mJb;+2F&6($cGi@T2v0P28OEnK96 zw*^|hiwpH%wI6Xj_yTb1tV2927wfK2JDmq02e-u*=A$VLl;J7A81QTB=Q;LuvHN0m zY_skO6s*K4``Hqq^NC6C<=MG>kkZ{Ok?QR3=fk?kC--UefQw)4@d7A@~?jLN<8{m=C4Fy zq+`{IJTJcl_a3XHF;ajh4BZv)56E!n@F>Jc&!mvZQd5j&W&$K1kY<;RDjQG{HuORj3pWM> z0KWe-5(nqL<1G*O2>w*rYaa178D-~+Azp^kHv2M~7bC+f5k7Nw|5>ZVbfS~|fZIOP zqbtm-rL1X_4J8GMeRC@FvF2SlM1?O9GimkG-z_ujA3LD~ZuypSm1;LT1Tn%IW%m;* ze0j7)A)C0^EM?aX}lbZU0Z{{$Ax}40308S)|eVDp-u0`MxvzgtG4IylTC?=)#jcW zjSuVHIGp~YnlhuC;dxN&YRvVVi_XKiXlSPrlb^S@<1e;NS1_?ITqNmtU58t` z(tXN-r+ITf97BM$^y=(5FDH)LsRP+xr*N(tC=>p_q~!>Gjv;x|8w8RlQ(!`(Y)M@Ndr?td+~Z;o{9K$ zfO`K{2M$cgFHbV_{^DZ0UZ$ulze`p80R61)6tLkF@&3wsZMdoAp(kzsyw?B&phaQt6e7EBb0#^c?LANR zCov$=OLzrDJ@=$B=H{9bwk35h`!sUaB8Mlw2YxA&(4oQ2?Wb*Og7?vluAM?h`!vD= zg1b?5umJU+`Y_$+bz1#CpTD!8O0>1erMShk2)4~~jc*-Lbdw0y>i$Y5iCTuArP)p2{GM7X}^V~kmh+Iq^(MLnLs)*1i6N@36Nx9(&1OY{!oT)N6CE=d8|XWOTQBq^o^G^Jfc%&ixql)K z{uSUjlL8e~p7@97C(NCsyyHK_YY{8xW1VCN>Qu&$yrmkhPhhK-LHemiSh3v#RSnld zr{xtr1#*Dq{U2%SX!R0L#NQF(4Li9`PxrYUvlki)0gmFj_AL=?KeD`2fd zTjOz)Iw;fB*648~yy@wMgcXvgKQ`xY)&H8d{9Zy!olrMfkENtnyl}|UC1WYDKFQo$7E-x%s4@M?SZ7j{&NrwYSg~Gf-w-~2wW-QGZ{oHoY9oGA4MQ}4D zsNmWysUT?EfLlj{)i)7F$EByi-=7Q$ox~xV=<&MpubLVYH|=Ci#)rPq7oV&)p*tYd zd)N#~hcn%iAIlY+M=&MQ3zwgXCYgUNe(j%yI%B-2vmAGmB!A`c{lsigSVko^mSQQp z(2u$qwHVd5`+=8XQwkjUB)|4HQSV}7r<2RByq-I7_63wyvv8+zl__8Q%X@Dw2OSrX zV6fup+1gxR4F*jH>uM3Qwgp|Pz6T%&(ACf$miG;SB*{<=GFqwGUqx*$gJz#0r|_&$Xx z2?&u=nbqd~t9Xm^z*_Zserix|4mM3w?^t1ufW`1Fr^Wg3dztw~onlqTy+x9QTsKz) zl%AH#MUhjF%G^#$vllx>EZt?uNh5Hq>Jjq!u;<*}=AqdG_)nz#zvbD7h(OUe4(SL$ zpxnR<>cOc@9PS;jY3D@j@!y%Mq4EPfwSARPEx8E-T3;Ky&I!pR&c6-WjgPLD2wYui zRzDsX!RuZj>Fmf%+JAqV`aLnS**Q>6;H&9enPTg)%D7TZD8q!9{~qcz8R%+7ZCY#7 zOW_+h?f-kG|CRfDwq0Wn>rHEl)yq5wM&a;pwiFI^ z3&2-1kGVY+iQ_gQ(=Q8-GMMoXainjBdDX3=yhH7+5q9ye2s+Vcso}y}AC2V@&Xx`4 zYolanzY+sKU2%^hVMF(hrR(qnyxLkc_*oW(|F-ix3mq%nqjy+dfcJte&hq9pR-C-p zEbg_Jp^-?3gxB*1K58sVR9`#wvUc`&)6xblxF%eUvYk=0hn!Vgx$UVRD~>mrJGdZ8 ziOWuzrX7eZLB={fH6-`K`2-$k=FKh@SFmwRsGWNFyXfo(``VX`S#4+kxsL$=&R)^{ zwxsOTrf-FR2tTwAJDgK|?vr%W)R`Qdwb)Rr@g?y)QV9`iPTF|!(7(!|?^>bjGEzbH)3$k3xx`!VPiYoj^TV&|Sa>F}_KIKb5ZSS;8Qsj)FRLXoE-K z+u4P*KD*kn+3#8j;@#+*(Om*8rGQ`u#p!c3rxgf7FeTaI={W#x&DWf}#E>Rw7^kv} zykU$xT93G&da;=MZ(P+ft4~{m{*8+b7X)jVaa?C$eI;}p`cZ$pP?%qZ6D~)_EwRP} zE;?573FMG})?fYc)L^16hQQp?{G}sQG&@R3bHp;<2Gt1hYe>9TNG0~SwQFP7Dq#T` z!{p>koSd!CoFL|E9=Ne;I4oT7YfWQLzXX?DtR9xm-NY3T`HHi)Fq0OG>ctiIUf z>0Y}5pIbYJ1qOZ}=Z;GDs!Q?gZZ@g5Jh)%_q{Ou)8|I}^dtmu&d%*X7l57rck6uG7 z_wH2w#|L4Qw4-aSVIq?{cW{HbSMZb}qW!mSQlc98 zQ4#ovdgiT@XIgCFE&{d@&0xgVIQdP7-7)xpSkWaCEb-p(%|)d;MdoN}d?%Ot3o>$B z{Q`odgz@=XI-W;qm_1Ovt1m_K=+}q;(?fCSSShuSlvc`#r|>Z*ohQ(R$Rn-Znt#Xj zD^&&OSa)gzl+6E^GR3?kJo^LpQ+22tiHBG~R%ZFn zQ|E+gw=&`$YS7R|Q{JeU115&8HmybJ(zZS+U-nVuI*DBVTVnm=x`t*h}*9f(J!^Jt2sjzpU+92+!MPumV-UQZ`PmM)DpxFE#{mH$T!8Hn3WnXk-+U zMBVDONVb?Hg}pD!FzLE=~ zuI3DYslarZoyeA`gjh3kmCM(xXU{unPDjm%%!8;vhwQ}pMcwDGQ1!lVt9NjAa>iyk zMuD+;RgkZB7v&0WAvfc#)u};&k4`Im>0<*2D}lVDO|<^2xfEvRWtNmzgR&THSC#iO zPl9DuYf(vxUECXst?}%XDE9=n(wU0U?`*edp@sg|OCA1GXFbhR=;rTmuHd5E2=@gw zgWK1op?V7!ur0Z?rV)w9!?i2jXmGyY(~=uCp)u!A>x7DLv@JDb2heJ*ntQqt(2`O| zvd=l(bCu6CXtjA-Td7gEz}t|dEBB?p_LEEK<yF2=E`@27Wzr+9%iO4+(}gb-%TNfml|vC}Vu&!4SshEQgysP1u3gK=RCAnw zGcK9(kA0>=y<_`D`?Te#!&>G+QDw8tfd(i;sTSR73gJ1|PxQC;%`SYE-0>gPrT4ij z>XEIEFTsFUgIO2nX7RpIZxjBLvJ%1**a+py&0D4G9(~qsg{*-On$PvkE$c1v<2~e?C@63M0Z)vBsj5xBWOTk!?Aae!9%fV zQ8yl3w#{RlE`L4q-ZtlUQfYRRP1czNX&-OgFb>uCgXmX7@Zq04pO=KP`io8*j^w-D zw(8AumiJ{-Xzv}e5*~o?!068u&RK~qD7c(-^ilm_wr$#!$@kRz+}{jNaj$HgqRSVx z54huqNqD=olXg1KIn(iix_aHpimkLalb|2YCoW1IQ+`0~I?0A$@U13!s_U^ouhe$S zd_t(KBSZjuuH^xk&Q?#gPup5e4;h>jj&hHUhz00a2x!toQ~fwHC!8O&l?c#?TU^zW zQW5m!1}@wKb{x~)T)%5CW?nLzyUUVK2CQOPdasL=CF)2R|NL~@+5R&pm4}%*&}KUe zb*IB}xLUMUrP=>VNkuBZ?)kV+b)*Q(Bu)C!N1cNg8|!^wFbzhAnPo6NQ?Mvhx} z62-;RVVpUosM)`rUOFjQ7thHJ&vwt{gmA65@a}O{RwHhz zq{CEF5?F6av$*RA>yiAHhOSsn!rL`qZl!H(DZE)HLD}k5Oh@45bfvpB3Cdkx4RJ2p zg=!pZydZBsL;&P6>Ak09zQ6_Eu!zL8IKFl!`jnJv0hzuT)TKzTGKocH;9c-y%ub>F zI7r8ma1cTimd97!>n?sbHItwqk5?&H_hF4eW%=qq4?{;jY;dB`tv^{~_jaak5e5SY zkv&^&mI?ZHo=*sEbstm58NCC=f$`6{!D@Y9=p=~YT(*KT2;taU{aTDTx~OQq*0H@6 zKDVN~*{^$%3SW-ffCJhpN|c5Aw)G+cq(*Wo7$|y9>sgCnWK2k(UU>!o#ylC?r;A>g z13+KPH9)^Y=rs5=yHF$gI!(G?amX#AOlO>UYAmU5)MJu76nuU*MagFbeGm{PFuBlJB}nHO@_! zV1b@AesBL|W*^o7W>ReMVKx_J+OSec*SP9FQ4P=pSv3i^l&LYPSHD%6Tg}$IgRVtx zyo|~`YNgpobq-^QWJHt3lSj=wetk(cZBBn3}JYfNn zYS{6h3{Klq*V1FTha1uy*QJ@}T6`~RQ1z?^do3yD?Xz!E^ot4?o((N)yXc7e~NVCcVa_EY-ylj|zBG_A7_s*jL} zU}kZts0 zR9(im=L0tqyGuqe$Ozj51@!_XV$7O+4~b1+#Py>-z()c4z(3VcU4KqtQn%`>aLoP# z0Pl;v2p%bid$1`gUn0}+CMhxHCxe?53@K1%PY$_cL_5M${+Su+p?Kx_409#`5P8yWpJ{I+K`*rz7W9Of} zI5SuvY{SnHGRl-5pHoB?ABtLl-&MG^1ZYHlx=A)x`xL}@fzw0ZNc4i%nuGXTZ>>*- zJF_4vP0r_za+g+9J70GNxk?Bnsw0^fi~G{lC(bmP#~sUNiuQJmNM2lk^Vsp2U-;Ht z$4|fOdE*_Dry7SC{Hy;+v(jC}EDASvP9fGV8)*nVC+u1oy{YIkeMkpvIBwA{HQ(w$ zqnReknV$Vr=W*I*56J6_MxwvY3m!FLDFYROPbNl%UB_4kZM zy}Pl837p@~Matrr+3w=Lnq;GIQBIDK&`~D82pzE-j|Njl1>b`Ri0ED)>-Dk>}~Zo5*km7IhKxprpZM4)d=qR*@C;0#kD z0wyUmhrXY5MSoqKS|8Jmp&iKu8M0Gap?T-9g?W--q|#yy3Qj&Ag&%4{ar2ee5f|A@ z!g0Q`8ajVMTPdZdS12@Z_bF5CCELfAlJI5TiooHO?7u>;{c$Ji|KJZ^cAgYtb-&QR zJ&)j6S5J1)06Z2m8*QF_hKd!#Y(`$Oe^n=5p3w+$ps6E~+Er4I=R>r?5dUGy={2pxNcC4t{23o_`+g$wIoc?mr*7#2Z zHN@yd?kZ-Sc41T(I1@-FLkqNurZzvnuhXle6%O>kwkiR?L=GIKeth&r>2?#;^V)VX zCfoml_jyw4(S`6KSItSvxEZTna-GYrx%#Y~pg^8OTCjfQn4~sjw?7R0&Of-$*7MRI zWNIj%b7n+NE?d=}Q{s9BcJ5F7=Ky3ylY0ZI^zYCm|1UmQzz1TT3HlpNT`96u3SsdE zp0CzdV$C;$EtisLoI`oBB86Ci)B7T|3@2*Ze%v?yEtx`T7$1?@o0nj+6=KgB#1ps1 zDmV2U$*(FeMG6mi^=^9g*M||~x-0#4RqC>M1}BW?2?T5A`C08^>wXW}*~baZN!Vs3 zygf<^RUEwUnm~W+6qo+!w5&Gv-+{1}z9Ys|z1b&+33Xuq*{{NNPH`)bSTXeHX+UOX zcl`4Bkjjm-?%yQl)~X{HuG!k{_W*H9(Kax+xxWv6AJ;E8TVFZ6oo_{xB0<7BM2GSb zAZ()x!X*~P)(z3(!a0D_O|J@q2~mCZjvEzGYRU7H`JgBqB14$U_vjaNS{S7NRtMhv z?O!Eg*2VwKFsF39Dh9KHNRqqw0Wjp-rJdE!DMggHlqSaJ4t&ACy1juPpaXKze^h|) z6{Lpqe(jY>6*H28qIUW|DaU&9|E%+=W6` zeP1jJCIeX6cfHuF-T1elx5~8PdA)a`tn(X>vsNP^zt?_1mJNP1U~MbUoTP6#^6f)L z7@At2(0-X6z3)rL#+JXUh!@gNldT&t%_%9W*ld+>G21r@)-Cz9VAXR?*uTuMn>8g> z@2|44qL2o}IbBwr4Jnk`8TehtsG`95uxOf)IWYfCWTh^vUxnus*65s=PrjX-t(6?y z-Q>vHSAI!|IvDr1DEKuud$5`MYF{&31AX9WYkL)`u?xyL?I=agsr29swe2dJmfR`Z zRQ~uuG0@9YpOgxXbmB>0^-i{iJY(>ke z<&{gAMQtPAkT!TjIxxi(_Gha}w*v9$q}R{m*l{KCCVzJyx2mryWLa$u2$cXC6Qv4# z!uXaGVL3Ijvk8u4ug7!deUK_8*Ue>#qT<)QTJFbNPFAON_B`pHYx3oCrOes+@8m;| z5HEcO94s2b2LW|`LqoC)J|VDLcIrncuQ7L5kr$FcgM9?^Q`!iB;t@8WN>`Uk4H84V zv$D_tk4mJAYVe{h*1l6d?v;@(JCkJe%^k1ef{l{f1ZxjPW+!QXaCzx>!&d4M0*=9T z^Zq*56)#Fg!g%F-CIYWfYIf}T;^hnV=FoZZm_O1MSiA4<9pZZ9dw?xf^=L3hNU`V| z#1?*f_L^3mu8r!ff@F&PcUm(rwd9O`fMXox@+M(D&eN9{Z z1Ca3dRFj!`%T7P*lj~4vtHY!e>5A&m3H<@{p;b!TPqR~CkzK>EQEQz3x#D3g*a(8! zH4}D4>lOrxTWu_ke%+<3)Oy+Kn_j2ut%mFZJWyHBa{&9+3wNNt9qL23uXrQ7hDd^1 z;*=?JmY`6-Tj7(p6Vad74)gmG*U!Y6yOq`uRtbJ$U?Sx!Q5X}Lb|{rTwcV5AFj0U1 z(LL_D@v>6?c??-t2Mys(eqk+CCG%kD$Gr_J!(a6Xul0@5GW|HjJ?U$CA@mi9F_)5i zZ=e_~h&k9RM5^D#nG&EZ8S&<|xwD|J!tD=&0uzRyy}BkllqM;B_iEsyd)tW;NyM__ z$IIkSt6B(xDDnPxz06P#%zOb9Q`kZ{{qL88VB!~k6#N;{7LWqCz1&Aub}IdLr(n+7p{YgYBeTS<8*Da*Zko{amk zakE-GYQs8Ho7ms&`SKnq^F(&gA2AJ|y0ygXUW$wp3nzP6Dim=I{X0vHB6Rr=Z9j12 z-xI;t|9Ne4LB-!<=mvZebV2vE^W=ZYzidPUJSb^bK|Y9%Qdy$?KDOjxPwu1nzDSwP zKb%nB4XMOZBc7Z5LyW|lIG{%WbQa^5`zI%7%<#!e^%vKIPs?qz24+bcRz*uOu1!N* zjmzm0c04bz2NhU#eWR-8M6Xkv1b>1+j#Jy-?JX_^;J< zE>%TE!2mR>&>~(nc0c@;=d;t0vSmen?7U86q6yB{As+hEloY;gsOEZmDr4AXhR}h} z=7WJ;&Lv66MQb`oY4lVoB3`P;bVbpBOK||u^S(TPTX*rZ$^4>%0daVX+VQc7x^g|{ zV3#7AQaZM2xg^HY$_d!oqK0j;4qCUy0M|X$V49kmAZ!nmV-p}Z= zt6hH)TE|70f!sdV%lgw$m|EH`kX|je`*y1jNW6NFQ&8o!nfo(PdGG4EQV&wQ0i|s- zuNpb1U1d6TaPa2c!oew2@x|n@YH=9|VRen@@-;z09#(2n=WlP3ydEO_WB&x&3;(?v zBg&(#{-Q4ApLmR|#hTNC$M!&od+Vd_z^z{wqwSq-23eKdBXUOcj86N`nUAsS;@ww0 zuQ~1B!d9QFhk%o!j&8*nChR6{ufgdyvRe%)5&_Yt76nL*80+Tn5WByvPIHe?|LxKbxj5OfISxRlTV|3VXB}l=HqmpM zZd+Wo&82Y#{OO6eGZ^HEBp_$JwV3(-VX%^PW=pvK#B0zD=@-gS?9}l44Pd}7*!f+X54i!3kUnChrt@ehX)r(56h2U1lA*)3Ni`o$f8iC?oOwbxG*-$!3OJ5e!7rKiI+fU$SXOPA36E)}MmJkx6->yAbuN&lThKpB4M_q=6*mV& zV%5H$KRo&RxXiuP-WY@tcg`MfiO4%g`|bC=Ug*9@l|#A^{#r%)h_tknJ@b6|_i2MF z8~0L!3kyb2vM_P&{m4s<@#lEEXmE$rC>SaWVhxn&rLH?Jknn}7;u+hZ;`mF;KnP z`*-u#sgI!Pj2f?U-gjJ?zqN%{&=ZExV4EhJ!5yhAH`yUg!r_44&m?DI>BY>Gg~?&7 z&?!ztPKjM8fBgY3JZq%I`T}WwX3&Wh@06B&x_vG@_4r5-EB?3dGs zr1!!5jC&51OKGtE9HAm7ZlwiT?zu(Y;16=KXbhZYx8F?Izv zDlxufc{wFx;$p9V3EXStwC4}kK>dymUR2u;ObHxyaSe`+9+ZccY39X2WePs&t3@R! z+0BYAwZs*fZ)En+b+OqH~h?M9#Z6~9odh^Jufqj zCH;W{K<~rUk7$^gcy9!V63BA$64Z(uVg(qwqBDW;tX!9QZ0U&8|JnnZM z4R$9W<@j4rOZ53yIiHap`t(qP4=;f)*|bNgEOLsI9lPXmjN!X-M*2fR9Yec@CW>@Z zRU3)A9~OcL*F=tIIdkVtdnD4HUkoQ!>9idOgD-Ng8RSk0Q;#w5meU!8a{kV@TlvHB(Cs;&ongpwetGfcaRod?kR}xVW z>|K07^;jY+b39_}H|1b%E~>&Gb`0NhjpM4AzdgkfP1SiO{wEP&=f3cLs6f|rsrz2{ z59AlO76Hf1b3kffunfIC^i1P!2T`kA3wE52b!!<|z1Ot=(SA*I{MnLOrafb5%m zWAU(Q9gCv6zo=aAKeu&ck=Z zc=y&PT!sP7@}Wa>WVeu$RL0?IN;>pf#>>+NoOynS15JD;X;q1XQ2XhI&&^me&GGJ) zd6?PIq}lrC#JD=NG4q;HdxP^9IiA~&eXviS`|K^U7FKZ^Nd$TF(AhJjYdR>^EVSnJn340Of8oWoQ!lkPd6PlBh>Wf_sv;hon@TvI45+x0gx7#zqhx-pVK~hUTPq1cEZDi=Zt05_k@WqHjERZ zma(d1S-`Nv#DB;$8*2+$ksKN40GX|aSIbU6sBu9v06g9c7`7(hY7X=G*nd=PX8X5% zx!t)=Vvj=TqRt;mw;qi0FpnI|uR9MxFR~-&8BF-a**<^@LWIwBCsDW#4#$FPdB&ow z^DugXH8^x|lAC`-7yiV)1#wZZFu&JqLQYzSHMe>S5q%IXDmGp85dL$}688Yiged1*A4U%JSrSJu(Lv>GRk#H|J4l7n>fdPGlGb zqHJu`Ad!POo%EIeSwru(F1Y%(2A8|;xk+9vffbp~Jk7M9ifg3uTboY8)&@V%BUrJ* zYMRF%T)xaG6{|c4<1YOH_h!mmPIFY|{ z%iO|N?>vH}hS%8-qt0ZLf}um9V&DFwa+G#Hl%Kz2t$j(o(MD;if(aTa9258Q|0?XO zgwZ_ zq)YmFLh;dh%WCSWB@4cy!}3n0l-RtbaeX~iFG1r6XTS|2BIyfMxUc-cLv}HR8tjLN z_shlp54QM+C*!~~i2N_B)+D#DAhB3PcpJj*kgq%J28sKo1-lPMiw!DXHd@KLjnGpu zS{kTonj>S&Mm=J3FEO{m7>4b!ib5EvTqeql49rH##oG%h%)d2yz3aoIB|DwjY6Mha zT5eTZu`SpirS6F%V0w-9qMCl3b<1ldo;ArPDVUVL(t*XkOUOX5wBAH!doeRx%Wp`J zJCeiBP`E*v%QdiE=m2hV6^G};R{LAEAI#YAaCk49SI}aZkSL?$Ffn<1KDu=)M{+E3 zn0C6b%X!PhA+=A?dbWnXP*OOv|2((TLZZ6X;8_|cJAZ34m%Yw&07X2_&Oa%xMWs^w z+gdA0sqqPt6^jb2y1_g~n#{(M%;4dcyS_;GoU5*zh^A_-`EjFg*AknHMHOa|cFt7& zsEB@9L98a8?)y?Szo5BdpYl5RtD%yEVzS)2c+a#+vi-)ThRwsle9nt;lMmm$+@|O^ zlnm68b=*V}qlz>l4IxWRPqBBd%Tv2;+b%k4%e$Eyws%r4OVOQ!Hud;yUt3vC2i&&{ zPz@%&7HmD>ESdV!t3Kn)ogdC4#DSs~Xe{mXs75e^4${*1S<=s%1JWtJt9J3|;zj-Q z#u2v6m++k!ll6ir@ra@WDs0FPDEWJein1dUt9FOqkslgbLcD~QshOQD$jcL>CoNjf zW9&B@X#w_w)T+ArU?nzeL*=cG~WfW$}}#T@Q}M5EEWEP-<#AB$!z0@WWcOO35R&BmuP-`%q%5XSXTGf&x(L05$i(8pR4S(W zT>6BHck3L_?LBqNQtg-{M1?8?rB(lY6n+rmBl=C4O?xCn#L*IpNkUCTTQlq~!I?%v`(6 zy6FN)$|P65(ZzvHtOJ>LW2TTbnVEAHgx3Nf3+ur(SBJ2!M5tc?Z1tb73(oKOCl*?)xL_BMv}U_6t#B91K1 z#?@U~=cB`An1$@8U!;kJO3W!e$rRki&h2tj1Mq`P%k?YGf3z30+_!fy3tQy5wA*0U z%7QY~2h^b{sRd1ErIstl*PEKgoP+{8gdTV=KcBaj6OMsIHU3&eF(Wn^f8W|XEh^N? zG21ZppsP(iI7>}W2>Wos+VP4l5dOjGn?s+1PbU;WYEaB?{za4|v3Lr27%c7q3eCYa zHw1<*NdPGhMpYLRxUm|-PEY9T@$Lf%vil@Wa@lb$ zo-bq;y?)7=x~zKsaq?(h-J?)$G-qBhO7ZsF_tR^?B(j=!F(O=2=6!L8dGOm+K zA_db*oP%}>V)4IEnGP|WS7Z8bd4|soNDeCNgZrCV6WRqtO=j4+AQvJJ=GJ!1&fq*V zUf;Wg+LSz@vyd-X>ZSzmZ37l6*~$djG#y8#g8rB~c$!VYUqoqBRX(?{D!tWcwx>V| zBE(7spIx-!r6DG4Dr@D)Wq?BdIR4=^&&c|5NKHonYU}~b=9T%Yr;nbNR;V1X_A_WI z;=B$G&kU1b?MmKX*self@l06(+=3o5gi((BIx#y}K!9de8jPBA!bujl@epT~Mre!=Ab!`UM+0yrd=6+PqGNK4&Td8>b(;J4CyHGNlbQSx1u6Mf zSKn6&F-p1obA$SH32H!N`M$%k8txLmiV`??~JLf0!ym6 zG_{x^2_fO;4gfe_D*20u?)c<7n95tFb(z+|Bj$=1Y5v+Q9qG2dz+n&Vv%)beY!n)< zC;@^#Z<-$i+-pJp1FSJcdjTvpetsRzb{ur-FA7GIKX5V;Z^Bzy(z+tyvy?1J@1wcuQPd(Ec@s!hlXtk~W@sb@B zdMWGRd2W{M()iR&Ze;hnd`XAT)>oznb2oza1=_YFCro8)1&-%!UN%H1R@K&z@4SoO z>gpvkLVG)sws&+ZY`8eZ=eg~so+(&Hnl#pujds6T4~Aij%b=c<%etgdA+)Z*KG>=E zbbhE@oVSv~Lj(M)G=ibDnkvn)Fl^UEKZWSSpW<6qw_`7-b$WQ5O$IG@G+<;Mdk&>l zX~oQ6!bJ+^ld!!t+QjuFU2|WaqL|-Wav@PsF0TUG3qg3mF2#CS5m;C)`74b!+at;T zti-q+>Bsp$_NT`#V zV#0fMsxEdaFly~ej_f_oPBW{C!m{0==p%Dhr4(WsfWnKsQ%_$vCao<))H$uw-@s&w zuZn~Tbor2bYvbo`9hs5`0r?o@9krjG+2(d$7P6l4d$LlQYv#i1DJvd{WQTRrM|^{i zq#U7lsXNQP5psg(Uy`OKk9yDt1vj!2vow5EhrSI4m$F7X=X!}BfpOoOeviWJ)HnF1 zMUdm?qQ;YQ^k{P#_mR8p=h6B3c)OjXH4URZUEHO8HF}NW-Xjeq$D1~J`eU<@78-E5 z$98qa$lgASjbDU|i<4sn`Lxv|f%a%&zI2Rm`sZj-OKzB*#vh^O7c=RI9tym^#CJCj zh3~^oa+FtabQG~cf?}p0yOo;cY4-!ejQ1=-3qhi9%2f+`MNC)cigzt!u~sT0jrRk} zYyk{+Y*YB0X0=iNg;?G%VV^kmXrh~n?4o9BxX8@Focel5vCjiX`YxJBmhWDNn$35X znEsh7>l)1?TyM}@zMjAvtrYCJ6m$9`sE*SlKJIN%tAM~-xYPJ8TbDQ6<^LH&;P+<&0w#x!kJ|&ix*f6%wc}ML* zdZus4DLq~YQaU%&9i2v@7t5NBbjM3xIC*a`6~ri{r#<7&UVz<{jXEd}dH`czqJo(IIKN-ym@NjjJog@c>khXO|Bno%$z3QbWf{^yO!(Z$z(LrD49? zVlA_7T_&#u=zJwYmJYm7e9H&ELva>3-t)9q*yM+fOc; z)5TFa5ZeZ_X!pp20}8+@RH{QCF_ z6pSuy+T8==cRhM@?FZ(*S_rPy>;}IrnzuktsR_NXxJSJ@+M9CtVd>f;5^QqR zGJUEYsy^Px>gOg^RZ}zZAXPo3y2QRBW$|yH6}IKdpOF#BXiq!cGVjH- z7atUit2~VAsvI@ZVmpwb2vnyVaJj2`*ZX$QYJs(*%VDGqNF0SN?zpvCv57_08=J}W zJf>{qFApSh_F>}w2?|ezbM8!LEv+~>y5W!WH^lrG56+kS4lJU9!z~FkSx?qHlV2;f z1Hoi~08$=<(1O`%l>*ycI^J81>J_Nb7fJiLdLUuY7zg{XdE1d`E=*8uOysp#m)}ry zvpi2IQb?e3o5DlrSHX6QRgcjb@mZWHoOYjK6OC0Jj0O&~OI2&_TA^N|8P|Jk7HGvQj0;>BR% zm(AW6-5PbrgS8K1bW+#$)^{Zxr+q{-GeWCOC#twp-mS+7z!ISuLc0POX-4Y8%`Z4v zt|3CG?!d(>?Xm()s&sI1eXrPgm}bD&EEq3C~aTr zDu0V*6LGxRcwNgLpG5n&`B$xHs#a*HByC9jeFbm|{3l{)2$LLq2mPOSE9W0*oo}+W zjBJ-jsZemc7-a9>Tb{dOnP`0Lx8y{xS!@^8Oc^P#3b!L7+y~6IDJ1Jvz5;_X#))?R z^7#RK`~cfX`8Ep_cGweGvmaXHgr|r0z%Cr{W4f`Tn=(vJ=j)ORi`$Q2RD^jBse=|= zxUrqdy+8tmKwz*0BCHSfhxuCz=ewogqIYjpqR5DknZ~SLWEF-tPaV2EGlJtO0e4Em zDT<}BED^%b_GWz*riFBD33g0>{V|5Gs7~TkVZ1&;eYp9anQ5OBI>CuHTw#c`2~q#@ zPOH`fv&ZOAv+vdGnOvRPqaFe@+pXup6Z44l)c#Dn3@=5e)taQ7yEm7m*gy&CpHt5m zv|{QnXA&olW{jE!`lGn|0gGL!dKp!U#LgKNk8H{cU>W6NRNhD)E#Kj2&e?_tXVHE?`xK zv-Z6A38oQHZ||yt;}6FGfE0`%7)OwShzKflzQq^+ZPRhQ7dr>(ZJs*_EdNR?(&HXR!^s%NIc{FI`~lyew-6yk<%m~ z;6eaa3rlC`vDz~(M6p8ZRkpw-|GLrwpsi@Yl7Wu75Nn-9P-xvtAnzp;{me7evzxlB z0GgNG^b9etGfHMcMw(&_a=EWnRk7g=fG2HL0{C9#4io>{5Po*CpHW%m&9LJ~OC9Jf zeRa;+Rx4X{nupLS$f+b`ojTo<{usV9<>GJ^`s2@BEOdnv?h<~PS#wn^XJ}t*$}UR4 zCjj0YLe_Ft8D)^J=v2^E+a=KpPM8v)60jFpE58_0rcZfWc59`Nzp`J)r>5^w<-(LIE;t1ks zPoJ@ihRxiCUlo}e0EyDYFaw4)x{8rhp}j`EUzd%&CsSg_-?Y*a@-on}Y3IR>lU}r> z9acxwfQ%8vzDGh6UuFtoG%E^7MY&?PBAHXp60 z;Gm_e$}f8RYJc0J30-Y0#~EN+3i#bUYMAQ62Ir~@@+;I|M3~~-s;k^f6`O({>(qdh z)fVhWD8NqQ(NG<#|L7+fa7#^ zOi+#w(e2@;{d4^bF7c0PS%8TvX_kMFp z2fXyKgk-AmYkWH|RI44W3Yk2Ss5pp}DBB*M^+^pbS7iM2^jkNlwnY3><->p@&eeT0 z$66=|fG|2N{s{?j|1jnVExTuhfY~|1I6;TQg3xAo!*u%P1Ud?P_jP<{3HDZ%2!+Y5 zSCXHe%KKAXZUb}-#xweERMO;}oZ!jiEXeRhMMtV+ASafze3M);WZG$eFS{maCy)=( z3VLr~JCS?ymCIUDK=aAq!j_C+4EBZTfW#KNtVSIB+xn=W{L1(ldP6EAA~pWKZsI*> z{?#-X;DO&JVMvPqXr5JGqcvE(1*fYrH&~Fe4xAKMZdfq{g*PavBuA6Z}hJ zO_{V+JZq?Kch3jDor^fSLw8Kp{5odYpp3r6nNoxMwHcW;F!yC*2#gG+eFGUWmU}Ju z@t$OA=41b;t}qY8=5T-NqZzx@XL@;V$|OpMD_@h0@)Adq!D0hP#n)EfP{lqI_LkJB z#Wx|;@}D8mVTE-U{H7VbzZlPMaM=(q9&cJo3}E$izu$3q+{X6z>SNg_uo5O0xbap9 z?p&+v4X?dlmE5xUJnI>k?wea~D5RU&VZ^tO4$MD#*R`;+=z*y(9%>P*b>(q#-Hvot zt0?7t6Ck~O=>h(7sm^wnU%WV)i#>+E&$|XUiY1RgEU?1Ky$x|=Mn3_@u8s_NzlBm{tZ!JP-og}byVsq7E)JF@yqbh4>z?KK4sz*3Hwh8w#l`-^nrgsR8AZ=z7 z3AVSq?L9f(=AUc(voIY+mqjS|8^yT_&2+41AAn{#{fgAO+nvuR<<41*>$IiFkTH=g zKW86P7XJB3is}Whf}IBRQ;~vur^HT<#IDYTm-h1M`!0UZzY#hg3?ANFmO{i_+tlm9 z*zO5OwP5(4c&EB>w2?M1z}+V@c)DtQ#eP2*Ko~{$BVyKHZ}#N)=V|46G;|Sql$K*_ zfnpU?wIFSED88{npoJszq-z0cJ1KNeSUx|z2;lhw_TAzcsL}Y&uEoQO^829+)~>MQ z`^Kt~-%2~ynLqlI9@yxNjde~Y33vbVj-2yhJ2vajv8C^j(g{GQp$oKwt6j%iaKVLmZCT=6KQymRMAazdjO`Jr0 z<8K>3{n$ILw_{SeQT>qEqN0a<=B#fw>z!9)byVo%M*r`Uma0pQarko>eETUW{Vu#> zac7aItJqviyM6-DXR6$0Cb>D>!4xiN@Zf!4H{H$b$yj&vr3Y4XD|2DCMQ?vJHhX-45`Z(2(qP;{Z{7US^9&jn8g5l93qL7g?v9*1f85%#Xt=|x zxD$L+gF6#U`f#(J)=BibCfv$LHnZZUOzgsgE`c(`*8%b|qJO(ol@48Ee2+oFe-T}W z8e?5q>yBBCRPu{I`lS`gRnGbHDl};=*%A-^Kn4Jd@*zGd2)+~Gy#}n%r;q^x50o7X z{w+IeB&Tc}mp_^mh@l|vy70aSG9kj)=7badXIku^zoVA}UwIQPtygip^^KGo2J)jLuAB$S!Vk#A)68stI`ll_vQdX_J$BG_uGeKK81FUoDkp76V`l9iF8)5w@uvYy9a(Ac1VYR-Y)`d;`cb%8jFpY|DFuau_QY2#N zwuy=xMF;NOBXKaBz1T}VKV0Nz`ycGWC%Me4Zf2*(%@l(?p>$=JgbM8H_#i=8|A%zxS@o=b5Z@0xY~&t;%Le^<30%Yd*eFv^PG z+ft;nH?{3q=E4yp(dLGi{18F?OE+|I0!ZX+ES{BVmx4Uj50^#rAbb|97Az>eT#7a<;J@?#1=|DF{-39V0aiZafBzBq_kI%>O}B#MInK|$ zJRaIS_h7dr-go-5tOUE^?bIP7-q~^2?CR&Nwj~cfJ9UOf_0or|Q3%&GoOy)dZox=7 z3M}+erkV4_1zc$2-tETVUQ_uS>BBxNaMw**g$xKgvm?M+Hayh(Q}UP+~S_HX#zbb zI{PlfhJ|bS?n8wSPL(I+Wa;W-!gt&UoOJH6D-0^uKe?NL0}>@kek~!jk9e;cn@BwR zBM?t3-A2wd@Q_aXnhJ0|Uv#P1u(P9mrN~*!JU|qXMi^Xz{Z2Rt0Izq#QjkiiOndj^ z`Eu{HKlm0e!Aw;cNLx*T92yD+SG9yGFySpgwAC3l2xF6YfdxBe4sk$a=EV*icRCg6 zc=z{=uL2Eosre^`JTH{x?@zK?G?y|5r=j2vFt)8~6L(0Fg^9h(fcq<>N17s!-{7=w zx$N*`l%(Yq5@+pT%yNdg8-)9#h!Zjy&)SEK2Ck(W6M52)gUZ&=rtY>bQC%AYB_C&9 zkM>4Q%|_eie3BI+#w#mLeq3+H(uTboc3{wfkl5zq#lZWHb#3 zDesu9E$qy8kDN`qI)%K1vE1&O+>)36`TCz@^7pv>w~GuFgnT;&HP4=(BYIQQ9B52T z6;=Eo0T9Jpz$WiXqvcJ^A*~WPnG%+P>Evg@nIyOJkQa%vLDoW?RZu#a=xe(JzOXNs znpyaW4i4rC@!h8z=UvGzAFZ2{L=F~}?+PwH;ps0YYH*v=CeOD69IO3k1)2>OTye>^ zga-Pn`)q~Jdx+nCV-3?ryYINAj1@0r%GdYk(0(cTC-TJ`@EKEFMhqtZ$H96iXVPN( zHcy(|k>FJQy%k<|C0dyM+4$9RI`MGKP~rN&Y$g9xrX#v{>-nJcc95kx0mwbmHL5Sc zLJ|4!&~rk~dp3ldjZdLzM!fYp#F{ElRfHC8$!j|m2Y{fWCOSa&ffUPeyZ9?i2nllY z+;H9iUOm+wToo2DP;81;y2@}MzA;F2kL@^%hIPpc1)vl8yEvrNP@a0;dS(;I=0DV( zxs1=d0rhX^JnWg_AczsFmWOY_7=HE0V3^uT)tqljV#0lni)3h9PY1BBg&spQeXYJN z9YZt_;n^)6vM3sr3-JxV!L(2z>aGmFi9Nwx#}W_vu`iyu+m)}%TFV{I=1GN5*>^1L)T)wXc~pGY*e+_^wvNA6}KF zQ01jK?pssPJw)`o0?z3oyy@ZEhK*jKCxzc`_W(kY}|wjg}9T451hu$yvrHom7* zaT|{>kR`i=21}$;tbT0E@R!5nn=$g_dZW!ISIt~n_-9alj@2XT%M!+Bxn`J|VYqh% zt{xv&>K(YX|8f#sCbn+1VysN{k@F=sz9${iIMoHn^D1L+p7Rgg+NzPs!wPoyuYckA zM8iuAM_5vB*kR;dXSL!9Q!hhwU)Fe)lZ10=8-HIpYL#Li&H$-Mc%_mx zaMa~5klh^R4PWDVNoM?RP}TKj=F$}ZznCk!$OXD@+2My|TI0ID#ka-gSWMxC@6@x$ z^mj4Fr3~yNC{B8VHzL;*Qm!}pp`%VaB=K<=N`EKq@^P_kp9j%fQx6%&p!ZSnexHXg z;$Dlz%$&A4?)GNU&n59AXJ` z{P1D$4sSD7&eit~%NlPMHH>~k*xSau_E(Tj^#mdT((Yr6j(_$xR_oNropapa?+r_B z%hzf$oR@e);ra{;hrDaa)vQ0?^hLXq_hJ~)+5beH_xL`S$6rLBS+1BzDotsrHziU( zq&Trx!#@T}Wk-Cjyf%JbgS^l`vs*e zQxZNFHJ7~hQaNw5kX3#3HD30Jyi(eWL~LlOXyh_^mG6F+P-O8kSHsxO-kvyjOA7aR z733>4*X-9;#SDDTiKd=NN&W&RocWG8B(ecsk=B%*4%wIOWCzkYW1eqh>g9lb~!xy zpZm!*wJEzD7z+?@Yq^C?&>k%jV-MJ(X6REZ3*`Oi$RRaB7z}Z|29h{*+Ydi->jn>T z1yt=29^;G9Ew`cI2xaOYstERPe1&Pe&G$c(^`a2UnV^l(4!%AG}qZ_DR$mg6-GytLb848qa@?U&Q`5a%C+_%Gt z_5gf-JP07`vZ4c>Yl=IgvX4>XBi@!MLjq%YU0L)(yY|=~mvLU@xWku`u-CWPh(|ts zSf889zvQ?o$j58u_7$q2**PxdBL5jZ|dc$`D&^zI% zV^ovzfz#Kqb`HkzJx4*O>}dXV0n*IE_jFCKU8sgDLy=kM=U6axqgf_7wmP|_@8%F` zW9WcE*n8O<_hJ#_pWMvh?Tf^hQdr5aRYoU34YhjSv`V#x;RR#74dB{K-}gKaj`qc6 zQ3-ubRyzGnpaB@HxeQCGz6%6-@hor>gB8u#{k*z{Zd`(gq0D(a!Hu4Thjr!OXHg$! zS*-;vl%{XeQdh!Si#uX!kO(54LVTS0)E7C*Y>0`-oIQcRa>U6gBYSMOnQPXB!ZQp)L$!fWv=-L0LLHDdVQa#mbO4ci{ex7rOm_M;s zhhYL$zDj<|q77g)c(NB3FG^lpbZodo;qwn3s1oSzr`Neiw zr`Kdib>dtMcmJF@{cmlvfTo>{oyevkZje-WNybMBsCT{SOzw1*Mq#GMTC)Dvy3u5jrTf)pVmH3vZmI@+nj~ajaBMqoi)`YW z@BJDVjPE^PL9vKAa@X)88JG_8YG>LbiMKTp?yjE#6>|LB%y&kI{~~HNJOhaUFmKju z;0qF5BwUREUz%em5Y=daU7T8~(@ryX=vtV~2o~DbNR-(f&WGr&w$|L(>mQ`K*~KY$ z+I^DWcopwa%+6!clx8uubU{6^BT?2LtuQH|PBXpnDfUJ5vFkf8uM41pFR)`pGoK#F zVg`Ghfy4%u{BPP$(`4MV!RNkwa&8#oou(o~R^?{Rb#zCxj_IUrJ@s@EJ#_D!%_WxB z2@pVL!~+ByoF1r)u*>2Wfnr^KGqci|#O0SQ6dJ@gkBffvZdyD7fK+rkmd|FIQ5aYw z?B*~KZ^`;M{y4Buw%Pg&aMou}Hgn+9+UsAkWEN#@F)-X+QgM-^`Z`VWIic%$4@c{i zjO@z#*@((LhH&2OPyY-R(LV>mYI*| z2V~)V;VY@9p&nq$MOJx}E^~#QAF=IE+Hd*ZH6GVdVtEVD+_3;%3{(o|&3FFWy#^fX zm!xJ1pPW&jPY@c5H@8{bwv@JC)(2gPsX9(faXe38rBFBnf=az@7Bpa*dDw5j_f}M) z(<1?bIS+ssBi)Vv3^&Oi#jG1lsWc_La8MMmhCHk_)pMy>H8x_nef3=LOh@I(-Pfag zm6}SJ-#7{Z$1KXBmnggegOP}1! z9|MuQB-nHVQaatgDGliV9$mX&P>9wWaxUEUaK6t|TksoDyX;NZEiVG=xPUwUmIl|M z%Hbonw=va5hWGc=Gc%s>{#UavuLD!&Nxr4Ud>44=4CqSKyk%>^{R2V-q{QAtej+K& zWA&?J{!6v{){koy^2h)wN|yf!-+sG2jLifNWbsNQ#kS03vd@ocBXtS5JuTcLQZ?_o zOD#52#vj6Y!XE85E2i_b6d1?*$6}QrYWvJbjeeRBy8Kl)_4=Qf3U1!qBT8f^91K-! z*%$gi1fs?#$8u0&B4Uo=(=%#-ovu%(P5F*x>R{-L2b_7aDJitj%-+kT>w_&K!HtB3 z@149zYrunWzW7e8WY9?q>AbQ-sa>*+*^w%DH2=4m7xAiUt47BJWx(U{dKgIKkB-4(uRF$D1ti0YTTC7J9(SyEQZ{HS%0#3een ziO#V1L+0-LhjkZ=*T+3XrpMw(1qeJ@0~gOORnoMp>J@lvqcz;)#f);I9$Z%p?Y&mH z9_@?O48?uLo||jAF5EkEpA5zC!hOMR=xVIC~s`(0drlK%h+f=Eo;CShZvSI z25vs)krqLz;je@_eOYCTfmMVN3*MLwz0VYxsk*E}FM`mWTzh+!9Q19%P~Bp{p`gH^ z=|bgF1uK!jodUxczb}tfEh#fmyo-^b`EI*R;_!%;yh>-HS=xKutUd`c@2g&rH0VbT z<9^&IN_uo{Imsa+-Uk!TCo=(>d9Qr!i>Yq-*#nAZ5${0&9NSs78Jy94ZO#`Ak|NQErh<&P4)VY}ALL$ehwoHQvj%U4AtB{K-z8US;NmZBFj&E8Itz?(YJ` z3mG5IFSCcsOv}v7o2|@4zF48JH7@h`YHY4-AgPjnYfd;$B!9mS1b3M*Jdtrn6pBdy TUgUCHT{18+kF(0 literal 0 HcmV?d00001 From 8797e1cd8626e833cebbca456a0b37e582672894 Mon Sep 17 00:00:00 2001 From: MariaLukina Date: Mon, 12 Feb 2024 18:50:02 +0700 Subject: [PATCH 5/6] Add distances calculations to knn.py --- knn.py | 30 ++++++++++++++---------------- 1 file changed, 14 insertions(+), 16 deletions(-) diff --git a/knn.py b/knn.py index c73ba6e..f88d7d2 100644 --- a/knn.py +++ b/knn.py @@ -54,11 +54,11 @@ def compute_distances_two_loops(self, X): distances, np array (num_test_samples, num_train_samples) - array with distances between each test and each train sample """ - - """ - YOUR CODE IS HERE - """ - pass + distances = np.zeros((X.shape[0], self.train_X.shape[0])) + for i in range(distances.shape[0]): + for j in range(distances.shape[1]): + distances[i, j] = np.sum(np.abs(self.train_X[j] - X[i])) + return distances def compute_distances_one_loop(self, X): @@ -73,11 +73,10 @@ def compute_distances_one_loop(self, X): distances, np array (num_test_samples, num_train_samples) - array with distances between each test and each train sample """ - - """ - YOUR CODE IS HERE - """ - pass + distances = np.zeros((X.shape[0], self.train_X.shape[0])) + for i in range(distances.shape[0]): + distances[i] = np.sum(np.abs(self.train_X - X[i]), axis = 1) + return distances def compute_distances_no_loops(self, X): @@ -92,11 +91,9 @@ def compute_distances_no_loops(self, X): distances, np array (num_test_samples, num_train_samples) - array with distances between each test and each train sample """ - - """ - YOUR CODE IS HERE - """ - pass + + distances = np.abs(X[:, np.newaxis, :] - self.train_X).sum(axis=2) + return distances def predict_labels_binary(self, distances): @@ -114,7 +111,8 @@ def predict_labels_binary(self, distances): n_train = distances.shape[1] n_test = distances.shape[0] prediction = np.zeros(n_test) - + #for i in range(n_test): + # neib_idx = np.argsort(distances[i])[0:self.k] """ YOUR CODE IS HERE """ From 7f4a6713b5101b7d909d68a9d77a13fe29894bba Mon Sep 17 00:00:00 2001 From: MariaLukina Date: Mon, 12 Feb 2024 18:51:17 +0700 Subject: [PATCH 6/6] Update KNN.ipynb --- KNN.ipynb | 468 ++++++++++++++++++------------------------------------ 1 file changed, 151 insertions(+), 317 deletions(-) diff --git a/KNN.ipynb b/KNN.ipynb index a4f5dd8..f2bedbc 100644 --- a/KNN.ipynb +++ b/KNN.ipynb @@ -23,7 +23,7 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 21, "id": "9638c464-806f-41b5-9dfe-1ea2048a1fa1", "metadata": {}, "outputs": [], @@ -44,7 +44,7 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 22, "id": "43bd8dc9-c430-4313-a6a1-4d3e5a7e9c47", "metadata": {}, "outputs": [], @@ -75,7 +75,7 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 23, "id": "54fa4253-ea6a-4ec4-b914-f7cb2346b195", "metadata": {}, "outputs": [], @@ -85,7 +85,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 24, "id": "3a188c83-6bf3-485d-9995-9f71d0868d30", "metadata": {}, "outputs": [], @@ -105,7 +105,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 25, "id": "277e132c-b89f-4dbb-8efd-cbcea015876d", "metadata": {}, "outputs": [ @@ -146,7 +146,7 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 26, "id": "362137fb-4577-4a21-8088-98cba79206f2", "metadata": {}, "outputs": [ @@ -191,7 +191,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 27, "id": "74595ef7-06ab-4700-b9d9-42db3fdd36d5", "metadata": {}, "outputs": [ @@ -248,28 +248,18 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 28, "id": "1932bd43-16d6-4201-8950-7dbe720a9fa1", "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "870\n", - "870\n" - ] - } - ], + "outputs": [], "source": [ "train_X, test_X, train_y, test_y = train_test_split(X,\n", " y,\n", " test_size = 0.13,\n", " random_state=SEED)\n", - "train_y\n", - "test_y\n", - "print(len(train_X))\n", - "print(train_X.shape[0])" + "#train_y\n", + "#test_y\n", + "#print(train_X.shape[0])" ] }, { @@ -290,7 +280,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 29, "id": "f40beae7-54a4-4323-b467-3173737dfd84", "metadata": {}, "outputs": [], @@ -323,12 +313,12 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 30, "id": "44d468e9-2a00-4268-bfdc-fbae3857ce90", "metadata": {}, "outputs": [], "source": [ - "knn_classifier = KNNClassifier(k=5)\n", + "knn_classifier = KNNClassifier(k=1)\n", "knn_classifier.fit(binary_train_X, binary_train_y)" ] }, @@ -367,43 +357,37 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 32, "id": "01b1ef27-4284-4d6c-978b-25f0fefd39be", "metadata": {}, - "outputs": [ - { - "ename": "TypeError", - "evalue": "'NoneType' object is not subscriptable", - "output_type": "error", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mTypeError\u001b[0m Traceback (most recent call last)", - "Cell \u001b[0;32mIn[13], line 3\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[38;5;66;03m# TODO: compute_distances_two_loops\u001b[39;00m\n\u001b[1;32m 2\u001b[0m dists \u001b[38;5;241m=\u001b[39m knn_classifier\u001b[38;5;241m.\u001b[39mcompute_distances_two_loops(binary_test_X)\n\u001b[0;32m----> 3\u001b[0m \u001b[38;5;28;01massert\u001b[39;00m np\u001b[38;5;241m.\u001b[39misclose(\u001b[43mdists\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;241;43m0\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m100\u001b[39;49m\u001b[43m]\u001b[49m, np\u001b[38;5;241m.\u001b[39msum(np\u001b[38;5;241m.\u001b[39mabs(binary_test_X[\u001b[38;5;241m0\u001b[39m] \u001b[38;5;241m-\u001b[39m binary_train_X[\u001b[38;5;241m100\u001b[39m])))\n", - "\u001b[0;31mTypeError\u001b[0m: 'NoneType' object is not subscriptable" - ] - } - ], + "outputs": [], "source": [ "# TODO: compute_distances_two_loops\n", + " \n", "dists = knn_classifier.compute_distances_two_loops(binary_test_X)\n", - "assert np.isclose(dists[0, 100], np.sum(np.abs(binary_test_X[0] - binary_train_X[100])))" + "\n", + "assert np.isclose(dists[0, 100], np.sum(np.abs(binary_test_X[0] - binary_train_X[100])))\n", + "#print(dists)\n", + "#dists.shape" ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 33, "id": "505e2c4b-1cfe-4e4a-8002-d9ce089d100e", "metadata": {}, "outputs": [], "source": [ "# TODO: compute_distances_one_loop\n", "dists = knn_classifier.compute_distances_one_loop(binary_test_X)\n", - "assert np.isclose(dists[0, 100], np.sum(np.abs(binary_test_X[0] - binary_train_X[100])))" + "assert np.isclose(dists[0, 100], np.sum(np.abs(binary_test_X[0] - binary_train_X[100])))\n", + "#print(dists)\n", + "#dists.shape" ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 34, "id": "dd81b766-5de2-4f62-82cf-8b6bd52002db", "metadata": {}, "outputs": [], @@ -423,10 +407,20 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 35, "id": "8ed08354-d0ef-497a-9d9c-b17c92b7bef9", "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "35.2 ms ± 4.73 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)\n", + "6.61 ms ± 358 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)\n", + "17 ms ± 1.02 ms per loop (mean ± std. dev. of 7 runs, 100 loops each)\n" + ] + } + ], "source": [ "%timeit knn_classifier.compute_distances_two_loops(binary_test_X)\n", "%timeit knn_classifier.compute_distances_one_loop(binary_test_X)\n", @@ -443,7 +437,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 36, "id": "9ca2679e-c731-467b-94b5-91a3a0641d7e", "metadata": {}, "outputs": [], @@ -470,7 +464,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 37, "id": "94268969-5334-4672-a939-65733068a89a", "metadata": {}, "outputs": [], @@ -480,7 +474,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 38, "id": "7842978b-a328-4035-b9c4-eb3711cb58c4", "metadata": {}, "outputs": [], @@ -506,7 +500,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 39, "id": "5caeb94f-6464-4adf-b3b6-8e12bf4a50b4", "metadata": {}, "outputs": [], @@ -534,7 +528,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 40, "id": "374d6fcf-21f2-433c-b011-76019201ce52", "metadata": {}, "outputs": [], @@ -562,12 +556,24 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 41, "id": "a2418dd8-93f1-4a11-8488-a8bce98e7d5f", "metadata": { "tags": [] }, - "outputs": [], + "outputs": [ + { + "ename": "TypeError", + "evalue": "cannot unpack non-iterable NoneType object", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mTypeError\u001b[0m Traceback (most recent call last)", + "Cell \u001b[0;32mIn[41], line 2\u001b[0m\n\u001b[1;32m 1\u001b[0m params \u001b[38;5;241m=\u001b[39m [\u001b[38;5;241m1\u001b[39m, \u001b[38;5;241m2\u001b[39m, \u001b[38;5;241m4\u001b[39m, \u001b[38;5;241m5\u001b[39m, \u001b[38;5;241m8\u001b[39m, \u001b[38;5;241m10\u001b[39m, \u001b[38;5;241m30\u001b[39m]\n\u001b[0;32m----> 2\u001b[0m train_metrics, test_metrics \u001b[38;5;241m=\u001b[39m find_best_k(binary_train_X, binary_train_y, binary_test_X, binary_test_y, params, accuracy_score)\n", + "\u001b[0;31mTypeError\u001b[0m: cannot unpack non-iterable NoneType object" + ] + } + ], "source": [ "params = [1, 2, 4, 5, 8, 10, 30]\n", "train_metrics, test_metrics = find_best_k(binary_train_X, binary_train_y, binary_test_X, binary_test_y, params, accuracy_score)" @@ -575,10 +581,22 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 42, "id": "7053ce06-854c-412b-8559-833595b1d6c0", "metadata": {}, - "outputs": [], + "outputs": [ + { + "ename": "NameError", + "evalue": "name 'train_metrics' is not defined", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", + "Cell \u001b[0;32mIn[42], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m plt\u001b[38;5;241m.\u001b[39mplot(params, \u001b[43mtrain_metrics\u001b[49m, label\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtrain\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m 2\u001b[0m plt\u001b[38;5;241m.\u001b[39mplot(params, test_metrics, label\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtest\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m 3\u001b[0m plt\u001b[38;5;241m.\u001b[39mlegend()\n", + "\u001b[0;31mNameError\u001b[0m: name 'train_metrics' is not defined" + ] + } + ], "source": [ "plt.plot(params, train_metrics, label=\"train\")\n", "plt.plot(params, test_metrics, label=\"test\")\n", @@ -613,10 +631,22 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 43, "id": "88b37ef1-fa15-4cc5-8558-0254890c899d", "metadata": {}, - "outputs": [], + "outputs": [ + { + "ename": "NameError", + "evalue": "name 'X_train' is not defined", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", + "Cell \u001b[0;32mIn[43], line 3\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[38;5;66;03m# TODO: predict_labels_multiclass in knn.py\u001b[39;00m\n\u001b[1;32m 2\u001b[0m knn_classifier \u001b[38;5;241m=\u001b[39m KNNClassifier(k\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m1\u001b[39m)\n\u001b[0;32m----> 3\u001b[0m knn_classifier\u001b[38;5;241m.\u001b[39mfit(\u001b[43mX_train\u001b[49m, y_train)\n\u001b[1;32m 4\u001b[0m predictions \u001b[38;5;241m=\u001b[39m knn_classifier\u001b[38;5;241m.\u001b[39mpredict(X_test)\n", + "\u001b[0;31mNameError\u001b[0m: name 'X_train' is not defined" + ] + } + ], "source": [ "# TODO: predict_labels_multiclass in knn.py\n", "knn_classifier = KNNClassifier(k=1)\n", @@ -634,10 +664,22 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 44, "id": "65887922-a799-4042-902e-f63f69508beb", "metadata": {}, - "outputs": [], + "outputs": [ + { + "ename": "NameError", + "evalue": "name 'predictions' is not defined", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", + "Cell \u001b[0;32mIn[44], line 2\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[38;5;66;03m# TODO: multiclass_accuracy in metrics.py\u001b[39;00m\n\u001b[0;32m----> 2\u001b[0m multiclass_accuracy(\u001b[43mpredictions\u001b[49m, y_test)\n", + "\u001b[0;31mNameError\u001b[0m: name 'predictions' is not defined" + ] + } + ], "source": [ "# TODO: multiclass_accuracy in metrics.py\n", "multiclass_accuracy(predictions, y_test)" @@ -669,7 +711,7 @@ }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 45, "id": "8ab4c84e-6036-4fe2-b81f-8115bf0a4774", "metadata": {}, "outputs": [], @@ -682,7 +724,7 @@ }, { "cell_type": "code", - "execution_count": 17, + "execution_count": 46, "id": "4136b3bb-e482-4102-ad5f-af5b3b1a030a", "metadata": {}, "outputs": [], @@ -711,7 +753,7 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": 47, "id": "a4aa413d-d830-4355-a063-9c5d6d5a6c9a", "metadata": {}, "outputs": [ @@ -1067,10 +1109,10 @@ "id": "35da5f1f-c18c-4164-ade1-5ee2793ce2bd", "metadata": {}, "source": [ - "Мы видим, что датасет хорошего качества, нет пропущенных значений, а данные уже стандартизованы ( все колонки, в том числе возраст имеют числовой тип,по всем колонкам одинаковое стандартное отклонение, а среднее значение равно нулю), некоторые переменные сильно коррелирую между собой.Наверное (хотя я не уверена вообще) в данном случае это может быть не очень хорошо, в частности коллерируют содержание холестерина и каких-то липопротеинов, вроде, это довльно ожидаемо, и как-то будто увеличивает размерность, не давая новой информации. Но я дилетант, и интересно было бы про такое поговорить.\n", + "Мы видим, что датасет хорошего качества, нет пропущенных и дублирующих значений, а данные уже стандартизованы ( все колонки, в том числе возраст имеют числовой тип,по всем колонкам одинаковое стандартное отклонение, а среднее значение равно нулю), некоторые переменные сильно коррелирую между собой.Наверное (хотя я не уверена вообще) в данном случае это может быть не очень хорошо, в частности коллерируют содержание холестерина и каких-то липопротеинов, вроде, это довльно ожидаемо, и как-то будто увеличивает размерность, не давая новой информации. Но я дилетант, и интересно было бы про такое поговорить.\n", "\n", "\n", - "Работа с исходным датасетом и составление pipeline будет в конце ноутбука" + "Работа с исходным датасетом и составление pipeline будет в конце ноутбука (в следущий раз сделаю более порядочно)" ] }, { @@ -1091,7 +1133,7 @@ }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 48, "id": "75690c55-b60b-435f-a6e0-a3a39975863e", "metadata": {}, "outputs": [], @@ -1141,7 +1183,7 @@ }, { "cell_type": "code", - "execution_count": 45, + "execution_count": 49, "id": "580f604f-e9c5-4109-8b9f-235369836057", "metadata": {}, "outputs": [], @@ -1169,7 +1211,7 @@ }, { "cell_type": "code", - "execution_count": 46, + "execution_count": 50, "id": "138a7969-8079-4a38-b92e-8eae5d246325", "metadata": {}, "outputs": [ @@ -1184,12 +1226,6 @@ "source": [ "from metrics import r_squared, mse, mae\n", "\n", - "#a = np.mean((y_test - y_pred)**2)\n", - "#mse(y_pred, y_test)\n", - "#mae = mae(y_pred, y_test)\n", - "#r2 = r_squared(y_pred, y_test)\n", - "#y_mean = np.mean(y_test)\n", - "#e = 1 - ((np.square(np.subtract(y_pred, y_test))).sum())/np.square(y_test - y_mean).sum()\n", "print('MSE, MAE, R_squared for k =1 ', \n", " mse(y_pred, y_test), mae(y_pred, y_test), r_squared(y_pred, y_test))\n", "\n", @@ -1199,7 +1235,7 @@ }, { "cell_type": "code", - "execution_count": 53, + "execution_count": 51, "id": "2d5d2005-c7fd-4562-8b7b-789565d3c6ca", "metadata": {}, "outputs": [ @@ -1261,7 +1297,7 @@ "metadata": {}, "source": [ "Выберем оптимальное k исходя из максимального значения r_squared и минимального mae, видим, что это происходит при k = 10 - 11, \n", - "Также можно еще посмотреть на разницу между метриками, посчитанными для train и выборками, она становится минимальной для k = 20, \n", + "Также можно еще посмотреть на разницу между метриками, посчитанными для train и test выборками, она становится минимальной для k = 20, \n", "Но после k = 10, скорость изменений очень снижается\n", "Поэтому я выбрала бы 11\n" ] @@ -1284,6 +1320,29 @@ "Так как у нас теперь большая группа, то было бы здорово всем познакомиться получше (так как выпускной не за горами). Соберитесь с одногруппниками в зуме, познакомьтесь, а сюда прикрепите скриншот с камерами всех участников." ] }, + { + "cell_type": "code", + "execution_count": 52, + "id": "5f92ad2e-2218-4a5a-8441-5c75f34e2cd3", + "metadata": {}, + "outputs": [ + { + "data": { + "image/jpeg": "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", + "text/plain": [ + "" + ] + }, + "execution_count": 52, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "from IPython.display import Image\n", + "Image(filename='socialization.jpg')" + ] + }, { "cell_type": "markdown", "id": "e116a42f-fae8-499c-a985-dc09e66a29b0", @@ -1309,12 +1368,13 @@ ] }, { - "cell_type": "code", - "execution_count": null, - "id": "9b873437-f42d-43fc-8222-327eef80fc8a", + "cell_type": "raw", + "id": "3dc63de8-8cb5-4792-8a22-6336c4f90be0", "metadata": {}, - "outputs": [], - "source": [] + "source": [ + "Я в шоке. Эта домашка заняла абсолютно все мои вечера и я все равно не смогла ее сделать, второй пункт был легче, так как мы делали это на консультации. Первый пункт ранил, ранил и убил. Очень хочется в академ, потому что даже если я буду успевать что-то делать, я не буду успевать делать в нормальном оформлении и, и осмыслять тем более.\n", + "Очень хочется разбор ноутбука для первой части, просто: что за чем, и зачем именно так на консультации." + ] }, { "cell_type": "raw", @@ -1326,235 +1386,10 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": null, "id": "f76ed0e4-9817-4ab0-89a0-016df7d47628", "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
agesexbmibps1s2s3s4s5s6
count442.000000442.000000442.000000442.000000442.000000442.000000442.000000442.000000442.000000442.000000
mean48.5181001.46832626.37579294.647014189.140271115.43914049.7884624.0702494.64141191.260181
std13.1090280.4995614.41812213.83128334.60805230.41308112.9342021.2904500.52239111.496335
min19.0000001.00000018.00000062.00000097.00000041.60000022.0000002.0000003.25810058.000000
25%38.2500001.00000023.20000084.000000164.25000096.05000040.2500003.0000004.27670083.250000
50%50.0000001.00000025.70000093.000000186.000000113.00000048.0000004.0000004.62005091.000000
75%59.0000002.00000029.275000105.000000209.750000134.50000057.7500005.0000004.99720098.000000
max79.0000002.00000042.200000133.000000301.000000242.40000099.0000009.0900006.107000124.000000
\n", - "
" - ], - "text/plain": [ - " age sex bmi bp s1 s2 \\\n", - "count 442.000000 442.000000 442.000000 442.000000 442.000000 442.000000 \n", - "mean 48.518100 1.468326 26.375792 94.647014 189.140271 115.439140 \n", - "std 13.109028 0.499561 4.418122 13.831283 34.608052 30.413081 \n", - "min 19.000000 1.000000 18.000000 62.000000 97.000000 41.600000 \n", - "25% 38.250000 1.000000 23.200000 84.000000 164.250000 96.050000 \n", - "50% 50.000000 1.000000 25.700000 93.000000 186.000000 113.000000 \n", - "75% 59.000000 2.000000 29.275000 105.000000 209.750000 134.500000 \n", - "max 79.000000 2.000000 42.200000 133.000000 301.000000 242.400000 \n", - "\n", - " s3 s4 s5 s6 \n", - "count 442.000000 442.000000 442.000000 442.000000 \n", - "mean 49.788462 4.070249 4.641411 91.260181 \n", - "std 12.934202 1.290450 0.522391 11.496335 \n", - "min 22.000000 2.000000 3.258100 58.000000 \n", - "25% 40.250000 3.000000 4.276700 83.250000 \n", - "50% 48.000000 4.000000 4.620050 91.000000 \n", - "75% 57.750000 5.000000 4.997200 98.000000 \n", - "max 99.000000 9.090000 6.107000 124.000000 " - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\n", - "RangeIndex: 442 entries, 0 to 441\n", - "Data columns (total 10 columns):\n", - " # Column Non-Null Count Dtype \n", - "--- ------ -------------- ----- \n", - " 0 age 442 non-null float64\n", - " 1 sex 442 non-null float64\n", - " 2 bmi 442 non-null float64\n", - " 3 bp 442 non-null float64\n", - " 4 s1 442 non-null float64\n", - " 5 s2 442 non-null float64\n", - " 6 s3 442 non-null float64\n", - " 7 s4 442 non-null float64\n", - " 8 s5 442 non-null float64\n", - " 9 s6 442 non-null float64\n", - "dtypes: float64(10)\n", - "memory usage: 34.7 KB\n", - "None\n", - "age 48.518100\n", - "sex 1.468326\n", - "bmi 26.375792\n", - "bp 94.647014\n", - "s1 189.140271\n", - "s2 115.439140\n", - "s3 49.788462\n", - "s4 4.070249\n", - "s5 4.641411\n", - "s6 91.260181\n", - "dtype: float64\n", - "age 13.109028\n", - "sex 0.499561\n", - "bmi 4.418122\n", - "bp 13.831283\n", - "s1 34.608052\n", - "s2 30.413081\n", - "s3 12.934202\n", - "s4 1.290450\n", - "s5 0.522391\n", - "s6 11.496335\n", - "dtype: float64\n" - ] - }, - { - "ename": "NameError", - "evalue": "name 'train_test_split' is not defined", - "output_type": "error", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", - "Cell \u001b[0;32mIn[2], line 19\u001b[0m\n\u001b[1;32m 14\u001b[0m category \u001b[38;5;241m=\u001b[39m [\u001b[38;5;124m'\u001b[39m\u001b[38;5;124msex\u001b[39m\u001b[38;5;124m'\u001b[39m]\n\u001b[1;32m 15\u001b[0m preprocessor \u001b[38;5;241m=\u001b[39m ColumnTransformer(transformers\u001b[38;5;241m=\u001b[39m[\n\u001b[1;32m 16\u001b[0m (\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mscaler\u001b[39m\u001b[38;5;124m\"\u001b[39m, StandardScaler(), numeric),\n\u001b[1;32m 17\u001b[0m (\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mohe\u001b[39m\u001b[38;5;124m\"\u001b[39m, OneHotEncoder(drop\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mfirst\u001b[39m\u001b[38;5;124m\"\u001b[39m), category)\n\u001b[1;32m 18\u001b[0m ])\n\u001b[0;32m---> 19\u001b[0m X_train, X_test,y_train, y_test \u001b[38;5;241m=\u001b[39m \u001b[43mtrain_test_split\u001b[49m(X,\n\u001b[1;32m 20\u001b[0m y,\n\u001b[1;32m 21\u001b[0m test_size \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0.15\u001b[39m,\n\u001b[1;32m 22\u001b[0m random_state\u001b[38;5;241m=\u001b[39mSEED)\n\u001b[1;32m 23\u001b[0m knn \u001b[38;5;241m=\u001b[39m KNeighborsRegressor(n_neighbors\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m1\u001b[39m)\n\u001b[1;32m 24\u001b[0m knn_pipeline \u001b[38;5;241m=\u001b[39m Pipeline(steps\u001b[38;5;241m=\u001b[39m[\n\u001b[1;32m 25\u001b[0m (\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mpreprocessor\u001b[39m\u001b[38;5;124m\"\u001b[39m, preprocessor),\n\u001b[1;32m 26\u001b[0m (\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mknn\u001b[39m\u001b[38;5;124m\"\u001b[39m, KNeighborsRegressor(n_neighbors\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m8\u001b[39m, n_jobs\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m8\u001b[39m))\n\u001b[1;32m 27\u001b[0m ])\n", - "\u001b[0;31mNameError\u001b[0m: name 'train_test_split' is not defined" - ] - } - ], + "outputs": [], "source": [ "from sklearn.preprocessing import (StandardScaler,\n", " OneHotEncoder,\n", @@ -1565,9 +1400,6 @@ "from sklearn.datasets import load_diabetes\n", "X, y = load_diabetes(as_frame = True, return_X_y = True, scaled = False)\n", "display(X.describe())\n", - "print(X.info())\n", - "print(X.mean())\n", - "print(X.std())\n", "numeric = ['age', 'bmi', 'bp', 's1', 's2', 's3', 's4', 's5', 's6']\n", "category = ['sex']\n", "preprocessor = ColumnTransformer(transformers=[\n", @@ -1578,24 +1410,26 @@ " y,\n", " test_size = 0.15,\n", " random_state=SEED)\n", - "knn = KNeighborsRegressor(n_neighbors=1)\n", + "#knn = KNeighborsRegressor(n_neighbors=1)\n", "knn_pipeline = Pipeline(steps=[\n", " (\"preprocessor\", preprocessor),\n", - " (\"knn\", KNeighborsRegressor(n_neighbors=8, n_jobs=8))\n", + " (\"knn\", KNeighborsRegressor(n_neighbors=1, n_jobs=8))\n", "])\n", "knn_pipeline.fit(X_train, y_train)\n", "knn_pipeline.predict(X_test)\n", "y_pred = knn_pipeline.predict(X_test)\n", - "#print(mse(y_pred, y_test))" + "# проверка, увидим, что данные аналогичны тем, что получилось при работе с уже стандартизованным датасетом\n", + "print(mse(y_pred, y_test), mae(y_pred, y_test), r_squared(y_pred, y_test))" ] }, { - "cell_type": "code", - "execution_count": null, - "id": "2351cb18-78a3-416c-aebc-ab269d835602", + "cell_type": "raw", + "id": "6590eed0-03d8-47dd-8a1e-0263510d715d", "metadata": {}, - "outputs": [], - "source": [] + "source": [ + "Эти значения чуть-чуть отличаются от аналогичных для работы с уже стандартизованным датасетом, это может быть связано с тем, что мы используем mean и std,\n", + "полученные на тренировочной выборки" + ] } ], "metadata": {