Md Imam Ahasan 1fe550b7cd
Iris-Species
2021-11-05 02:15:02 +06:00

643 lines
222 KiB
Python
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"accelerator": "GPU",
"colab": {
"name": "linear-clf.ipynb",
"provenance": [],
"collapsed_sections": [],
"toc_visible": true
},
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.6.3"
}
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "ri6UiGU5T5aj"
},
"source": [
"# **সাইকিট-লার্ন দিয়ে একটা সহজ লিনিয়ার ক্লাসিফিকেশন **\n",
"\n",
"চারটার জায়গায় দুটো ফিচার, তিনটার জায়গায় দুটো টার্গেট ভ্যারিয়েবল"
]
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "3sKNXPKNBqwO"
},
"source": [
"চারটার জায়গায় দুটো। প্রস্তাব - এটাকে দুটো দিয়ে দেখান না কেন? বুঝলাম - জিনিসটাকে আরো পানির মতো করতে হবে। আমাকে অনেকে বলেন, আইরিস ডেটাসেটে চারটা অ্যাট্রিবিউট। ফলে ডেটা ভিজ্যুয়ালাইজেশনে একটার ভেতরে আরেকটা চলে যায়। খালি চোখে ডেটার মধ্যে ফারাক বের করা তো দুস্কর। প্রস্তাবটা ভালো। এটা একটা বড় সমস্যাকে আরো রিফাইন করে আনবে আমাদের ভালোভাবে বুঝতে। \n",
"\n",
" সত্যি বলতে সেই আইডিয়াটা নিয়ে লিখেছেন বেশ কয়েকজন লেখক। তবে, এখানে আইডিয়াটা এলো আমার একটা প্রিয় বই থেকে, ২০১৩তে লেখা। লার্নিং সাইকিট-লার্ন:: মেশিন লার্নিং ইন পাইথন, রাউল গ্যারেটার। \"কী করবো সামনে?\" চ্যাপ্টারে দ্রষ্টব্য। \n",
"\n",
"আচ্ছা, তিনটা প্রজাতি না বের করে, একটা প্রজাতি বের করা যায় না? আরো, ভালো! তাহলে তো একটা প্রজাতি ভার্সেস ওই প্রজাতি নয়। মানে, প্রেডিক্ট করতে হবে - ধরুন, ফুলটা \"সেটোসা\" অথবা \"সেটোসা নয়\"! তাহলে তো জিনিসটা একদম পানি হয়ে যাবে। "
]
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "gfRxmu0QT5a5"
},
"source": [
"## লোড করে নেই আইরিস ডেটাসেট "
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "c32_FG83T5a7",
"colab": {}
},
"source": [
"import sklearn\n",
"from sklearn import datasets\n",
"\n",
"iris = datasets.load_iris()\n",
"X_temp = iris.data\n",
"y_temp = iris.target"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "TrVDNHPBT5bL"
},
"source": [
"### ভাগ করে ফেলি টেস্ট এবং ট্রেনিং ডেটাসেট (ফিচার স্কেলিং সহ)"
]
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "zTvelXXOT5bM"
},
"source": [
"এখানে আমাদের কাজ হচ্ছে ডেটাসেটকে দুভাগে ভাগ করে ফেলা। ৭৫% ব্যবহার হবে আমাদের ক্লাসিফায়ারকে ট্রেনিং করাতে। ২৫% যাবে ইভ্যালুয়েট করতে। ৪টা ফিচারের জায়গায় আমরা ব্যবহার করবো ২টা মাত্র। সিপাল দৈর্ঘ্য এবং প্রস্থ। শুধুমাত্র সিপাল অংশ। \n",
"\n",
"এর পাশাপাশি আমরা ফিচার স্কেলিং করবো আমাদের ফিচারগুলোর ডেটা রেঞ্জ স্ট্যান্ডার্ডাইজ করার জন্য। প্রতিটা ফিচারের জন্য এটা সব ভ্যালুকে গড় করে সেটাকে বিয়োগ দেয় ওই ফিচার ভ্যালু থেকে। এরপর তার উত্তরকে ভাগ দেয় সেটার স্ট্যান্ডার্ড ডেভিয়েশন দিয়ে। আমাদের এই স্কেলিং এর পর প্রতিটা ফিচারের গড় হবে শূন্য। পাশাপাশি স্ট্যান্ডার্ড ডেভিয়েশন হচ্ছে ১। \n",
"\n",
"এর ফলে ভ্যালুগুলোর স্ট্যান্ডার্ডাইজেশন হয়ে আসে। এটা খানিকটা স্ট্যান্ডার্ড প্র্যাক্টিস হয়ে গেছে ইন্ডিপেন্ডেন্ট ফিচার/ভ্যারিয়েবলগুলোর রেঞ্জকে একটা স্কেলের মধ্যে নিয়ে আসা। এটাকে আমরা ডেটা নর্মালাইজেশন বলতে পারি। এটা আমরা করি ডেটা প্রি-প্রসেসিং এর সময়। \n",
"\n",
"ডেটার রেঞ্জ নিয়ে আমাদের যেহেতু কোন ফিল্টার নেই, সেকারণে একটা ডেটাসেটে বিক্ষিপ্ত ডেটা মেশিন লার্নিংকে বিপদে ফেলতে পারে। বড় বড় ভ্যালুগুলো ফাইনাল আউটকামে সমস্যা করে। আর সেকারণে মেশিন লার্নিং অ্যালগরিদমকে ভালোভাবে কাজ করানোর জন্য এই স্কেলিং দরকার পড়ে অনেক সময়। তবে, \"\"গ্রাডিয়েন্ট ডিসেন্ট\"\" কনভার্জেন্স ভালো কাজ করে স্কেলিং দিয়ে। মজার কথা হচ্ছে এক্স ভ্যালুগুলোকে প্লট করলে আগে এবং পরে একই জিনিস পাওয়া যায়। "
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "RknPx8MPT5bN",
"colab": {}
},
"source": [
"from sklearn.model_selection import train_test_split\n",
"from sklearn import preprocessing\n",
"\n",
"# শুধুমাত্র প্রথম দুটো অ্যাট্রিবিউট নিয়ে আমাদের ডেটাসেট \n",
"X, y = X_temp[:, [0,1]], y_temp\n",
"# আমাদের টেস্টসেট হবে ২৫%, দৈবচয়নের ভিত্তিতে \n",
"X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=33)\n",
" \n",
"# ফিচারগুলোকে স্ট্যান্ডার্ডাইজ করছি এখানে \n",
"scaler = preprocessing.StandardScaler().fit(X_train)\n",
"X_train = scaler.transform(X_train)\n",
"X_test = scaler.transform(X_test)"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "IRRoCZxmT5bQ"
},
"source": [
"চলুন, দেখি ফিচার স্কেলিং এর পর কি অবস্থা? এখানে গড় হচ্ছে \"\", স্ট্যান্ডার্ড ডেভিয়েশন হচ্ছে \"\"। ট্রেনিংসেটে ঠিকমতো হবে সবকিছু, তবে টেস্টসেটে ব্যাপারটা কাছাকাছি হবে। "
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "Xon7VkoGT5bR",
"outputId": "2b912fb5-aed9-4975-8443-308c01074409",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 50
}
},
"source": [
"import numpy as np\n",
"print ('Training set mean:{:.2f} and standard deviation:{:.2f}'.format(np.average(X_train),np.std(X_train)))\n",
"print ('Testing set mean:{:.2f} and standard deviation:{:.2f}'.format(np.average(X_test),np.std(X_test)))\n"
],
"execution_count": 3,
"outputs": [
{
"output_type": "stream",
"text": [
"Training set mean:0.00 and standard deviation:1.00\n",
"Testing set mean:0.13 and standard deviation:0.71\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "jFf9lDTeT5bV"
},
"source": [
"ফিচার স্কেলিং এর পর ট্রেনিং ডেটাকে প্লটিং করি। একই জিনিস। "
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "DPCGNVefT5bW",
"outputId": "9885b09b-f504-4c82-8d26-ea0dc7e37370",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 295
}
},
"source": [
"# প্লটিং লাইব্রেরি লোড করে নেই \n",
"import matplotlib.pyplot as plt\n",
"\n",
"# তিন প্রজাতির তিনটা আলাদা রং, মার্কার সহ \n",
"colour_mk = [ ['red','s'], ['green','o'], ['blue','x']]\n",
"plt.figure('Training Data')\n",
"\n",
"# লুপে ফেলে দিলাম, x এবং y এক্সিসে \n",
"for i in range(len(colour_mk)):\n",
" xs = X_train[:, 0][y_train == i]\n",
" ys = X_train[:, 1][y_train == i]\n",
" plt.scatter(xs, ys, c=colour_mk[i][0], marker=colour_mk[i][1])\n",
"\n",
"# সাদা ব্যাকগ্রাউন্ড দরকার আমার, গুগল কোলাবে বাড়তি শেড দরকার নেই \n",
"# plt.rcParams['axes.facecolor'] = 'white'\n",
"plt.style.use('default')\n",
"plt.grid(c='grey')\n",
"\n",
"# প্লটিং প্যারামিটার \n",
"plt.title('Training instances, after scaling')\n",
"plt.legend(iris.target_names)\n",
"plt.xlabel('Sepal length')\n",
"plt.ylabel('Sepal width')\n",
"plt.show()\n"
],
"execution_count": 4,
"outputs": [
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "-fYy0VkkT5bb"
},
"source": [
"### একটা লিনিয়ার বাইনারি ক্লাসিফিকেশন "
]
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "pfEKD6K5T5bc"
},
"source": [
"মানুষের মাথা প্যাটার্ন বুঝতে ওস্তাদ। এই প্লট থেকে কী বুঝতে পারছেন? ঠিক ধরেছেন। খালি চোখে সেটোসা প্রজাতিকে বোঝা যাচ্ছে একদম আলাদা করে। কেমন হয়, কমপ্লেক্সিটি এড়াতে আমরা যদি বের করতে চাই শুধুমাত্র সেটোসা প্রজাতি বের করতে চাই। মানে, প্রেডিক্ট করতে হবে হয় \"সেটোসা\" অথবা \"সেটোসা না\"? এখন আমাদের দুটো টার্গেট ভ্যারিয়েবল। সেকারণে এটাকে আমরা কনভার্ট করছি বাইনারি ক্লাসিফিকেশন টাস্কে। আমাদের দুটো টার্গেট। হয় \"\" অথবা \"\", তাহলে কী করতে হবে? \"\" নম্বর এবং \"\" নম্বর ক্লাসকে আমরা \"\" বানিয়ে ফেলেছি। \n",
"\n"
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "yLoZKBnCT5bd",
"outputId": "7240f679-1330-47c9-80f4-426d5c4bf9c6",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 101
}
},
"source": [
"import copy \n",
"y_train_setosa = copy.copy(y_train) \n",
"# আমাদের ট্রেনিংসেটের ১ এবং ২ ক্লাসকে ১ বানিয়ে ফেলছি \n",
"y_train_setosa[y_train_setosa > 0]=1\n",
"y_test_setosa = copy.copy(y_test)\n",
"y_test_setosa[y_test_setosa > 0]=1\n",
"# এখন দেখি ট্রেনিং টার্গেট ক্লাসগুলো কী কী?\n",
"print ('New training target classes:\\n{0}'.format(y_train_setosa))\n"
],
"execution_count": 5,
"outputs": [
{
"output_type": "stream",
"text": [
"New training target classes:\n",
"[1 0 1 1 1 0 0 1 0 1 0 0 1 1 0 1 1 1 1 1 0 0 1 0 0 1 1 1 1 1 1 1 0 0 1 1 0\n",
" 1 1 1 1 0 1 0 1 0 1 1 0 1 1 0 0 1 0 0 0 1 1 0 1 0 1 0 1 1 1 1 1 0 1 0 1 1\n",
" 0 0 0 0 1 1 0 1 1 1 1 0 0 1 1 1 0 1 1 0 1 1 1 1 1 0 1 0 0 0 1 1 1 1 1 1 1\n",
" 0]\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "mj_02w5mT5bi"
},
"source": [
"ছবিটা দেখে কী মনে হচ্ছে? একটা প্রজাতি একেবারে আলাদা। এটা আমাদের জন্য ভালো। \n",
"\n",
"আমরা যদি ভালোমতো করে ছবিটা দেখি - তাহলে \"সেটোসা\" প্রজাতিতে আমরা একেবারে আলাদা হাইপারপ্লেন এ দেখতে পাচ্ছি। অর্থাৎ একটা লাইন টেনে দুটো প্রজাতিকে আলাদা করতে পারছি। ব্যাপারটা কমন মেশিন লার্নিং কনসেপ্টে। আমাদের প্রশ্ন হচ্ছে নতুন মাপজোক দিলে সেটা থেকে বের করতে হবে নতুন জিনিসটা কোন প্রজাতির? এখন বাকি প্রজাতিগুলো যেহেতু একটা আরেকটার ভেতরে ঢুকে গেছে, সেকারনে ওই দুটোকে একটা প্রজাতি হিসেবে দেখাচ্ছি। \n",
"\n",
"যেহেতু, আমরা \"সেটোসা\"কে একেবারে একটা লাইন টেনে আলাদা করতে পারছি, সেকারণে এই জিনিসটাকে একটা লিনিয়ার ক্লাসিফিকেশন মডেলে পাঠাতে পারি। মানে, একটা সোজা লাইন টেনে দুটো টার্গেট ক্লাসকে আলাদা করবো এখানে। এটাকে আমরা বলতে পারি ফিচার স্পেসে একটা হাইপারপ্লেন। দুটো ফিচার স্পেসের মধ্যে লাইনটা ডিসিশন বাউন্ডারি। কে কোন প্রজাতির, সেটা নির্ভর করবে কে ওই লাইনটার কোন দিকে আছে। \n",
"\n",
"মনে আছে, আমাদের ওই এরর কমানোর কথা? লিস্ট স্কয়ার রিগ্রেশন, লস ফাংশন? যেটা আসলে বের করে আমাদের প্রতিটা ইনস্ট্যান্স থেকে ডিসিশন বাউন্ডারি কতো দুরে। এখানে আমাদের এই অ্যালগরিদম হাইপারপ্লেনের \"কোএফিসিয়েন্ট\" জানবে লসকে কমিয়ে। এখানে আমরা ইন্টারসেপ্টও জানবো সামনে। \n",
"\n",
"এ কারণে আমরা সাইকিট লার্ন থেকে `SGDClassifier` ব্যবহার করবো এই লিনিয়ার মডেল তৈরি করতে। আমাদের সাইকিট লার্নে \"SGDClassifier\" মডেল হিসেবে থাকলেও এটা আসলে ক্লাসিফায়ার নয়। বরং এটা একটা লিনিয়ার তবে এটাকে অপ্টিমাইজড করা হয়েছে স্টোকাস্টিক গ্র্যাডিয়েন্ট ডিসেন্ট দিয়ে। \n"
]
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "5KCIy6waT5bm"
},
"source": [
"এই কাজ আমরা আগেও করেছি। \"linear_model\" কে ইম্পোর্ট করে নিয়ে আসছি sklearn থেকে। একটা ক্লাসিফায়ারের ইনস্ট্যান্স তৈরি করে হাইপারপ্যারামিটারকে বলছি \"লগ\" লস ফাংশন ব্যবহার করতে। এখানে ক্লাসিফায়ার হচ্ছে \"linear_model.SGDClassifier\"। এমুহুর্তে ব্যবহার করবো সব ডিফল্ট ভ্যালু। বেশি ঝামেলায় যাবো না। \n",
"\n"
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "0jrNibUVT5bo",
"colab": {}
},
"source": [
"from sklearn import linear_model \n",
"clf = linear_model.SGDClassifier(loss='log', random_state=42)"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "djVcNsMYT5bv"
},
"source": [
"এখন কি বাকি? ট্রেনিং করানো। ফিট মেথড কল করছি আমাদের ক্লাসিফায়ারকে ট্রেনিং করানোর জন্য। এখানে আমাদের ট্রেনিং ডেটা হচ্ছে \"সেটোসা\" সেট। \n"
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "h9Tf4RvNT5bx",
"outputId": "05109615-5878-442d-ffce-dbf0f07e5968",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 118
}
},
"source": [
"clf.fit(X_train, y_train_setosa)"
],
"execution_count": 7,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"SGDClassifier(alpha=0.0001, average=False, class_weight=None,\n",
" early_stopping=False, epsilon=0.1, eta0=0.0, fit_intercept=True,\n",
" l1_ratio=0.15, learning_rate='optimal', loss='log', max_iter=1000,\n",
" n_iter_no_change=5, n_jobs=None, penalty='l2', power_t=0.5,\n",
" random_state=42, shuffle=True, tol=0.001, validation_fraction=0.1,\n",
" verbose=0, warm_start=False)"
]
},
"metadata": {
"tags": []
},
"execution_count": 7
}
]
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "dVVYycW3T5b0"
},
"source": [
"লিনিয়ার মডেল। মনে আছে \"y = mx + b\" এর কথা? নাহ, অংক পিছু ছাড়ছেই না, আমাদেরকে m এবং b পেতে হবে। মানে, clf.coef_ এবং clf.intercept_ ছাড়া আমাদের গতি নেই। \n",
"\n",
"\n",
"এখন এই সমীকরণকে y = mx + b ধারণায় লিখলে কেমন দেখা যাবে?"
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "94ScDMvbT5b1",
"outputId": "4a52e7ed-2c14-4ad4-c105-959ce00bbfbe",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 34
}
},
"source": [
"print (clf.coef_,clf.intercept_)\n"
],
"execution_count": 8,
"outputs": [
{
"output_type": "stream",
"text": [
"[[ 21.76180378 -10.51985219]] [13.90763026]\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "bg1L0kSLT5b5"
},
"source": [
"এখন তো ডিসিশন বাউন্ডারি আঁকাই যায়, কি বলুন? কোড দিলাম না ইচ্ছে করে। মেইন লিংকে পাওয়া যাবে। \n",
"\n",
"<img src=\"https://github.com/raqueeb/ml-python/blob/master/assets/line.png?raw=1\">"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "OhODP9vIRYxk",
"colab_type": "text"
},
"source": [
"## এখানে কোড\n",
"\n",
"ধন্যবাদ \"A Gentle Introduction to Machine Learning with Python and Scikit-learn\" বইটাকে। ষষ্ঠ অধ্যায় দেখুন।"
]
},
{
"cell_type": "code",
"metadata": {
"id": "5_JvEW9JReec",
"colab_type": "code",
"outputId": "0ef65ae8-0904-4bda-c2a9-1dcd4eb9bb1c",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 472
}
},
"source": [
"x_min, x_max = X_train[:, 0].min() - .5, X_train[:, 0].max() + .5\n",
"y_min, y_max = X_train[:, 1].min() - .5, X_train[:, 1].max() + .5\n",
"xs = np.arange(x_min, x_max, 0.5)\n",
"\n",
"fig,axes = plt.subplots()\n",
"\n",
"axes.set_aspect('equal')\n",
"axes.set_title('Setosa species classification')\n",
"axes.set_xlabel('Sepal length')\n",
"axes.set_ylabel('Sepal width')\n",
"axes.set_xlim(x_min, x_max)\n",
"axes.set_ylim(y_min, y_max)\n",
"\n",
"plt.sca(axes)\n",
"\n",
"plt.scatter(X_train[:, 0][y_train == 0], X_train[:, 1][y_train == 0], c='red', marker='s')\n",
"plt.scatter(X_train[:, 0][y_train == 1], X_train[:, 1][y_train == 1], c='black', marker='x')\n",
"\n",
"ys = (-clf.intercept_[0]- xs * clf.coef_[0, 0]) / clf.coef_[0, 1]\n",
"\n",
"plt.plot(xs, ys)\n",
"plt.show()"
],
"execution_count": 9,
"outputs": [
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {
"tags": []
}
}
]
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "ViC_b6UrT5b-"
},
"source": [
"ঠিক ধরেছেন। এই নীল/কালো রেখাটাই হচ্ছে আমাদের ডিসিশন বাউন্ডারি। প্রতিবার ৩০. x \"সিপাল দৈর্ঘ্য\" - ১৭.৭৮ x \"সিপাল প্রস্থ্য\" - ১৭.৩১ এর আউটপুট যখন শূন্য থেকে বড় হবে তখন সেটা হবে আইরিস সেটোসা, মানে ক্লাস ০। \n",
"\n",
"চলুন, একটা প্রেডিক্ট করি। ফুলটা কি সেটোসা কি না? যদি একটা ফুলের পেটাল প্রস্থ্য .৬ এবং পেটাল দৈর্ঘ্য ৩.২ হয়, তাহলে প্রজাতিটা কি সেটোসা হবে কি হবে না? "
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "J3_a65KZ8Si6",
"outputId": "1d41c074-4803-42be-9d00-4d8487f79fc3",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 34
}
},
"source": [
"print ('If the flower has 4.6 petal width and 3.2 petal length is a {}'.format(\n",
" iris.target_names[clf.predict(scaler.transform([[4.6, 3.2]]))]))"
],
"execution_count": 10,
"outputs": [
{
"output_type": "stream",
"text": [
"If the flower has 4.6 petal width and 3.2 petal length is a ['setosa']\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "gVKeWGJq9gZ4"
},
"source": [
"উত্তর: সেটোসা!"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "9dr7-Y2WWeHk",
"colab_type": "text"
},
"source": [
"## মজার বোনাস কাজ\n",
"\n",
"কোড বোঝার দরকার নেই শুরুতে"
]
},
{
"cell_type": "code",
"metadata": {
"id": "Q_JYuiBdWqrP",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 861
},
"outputId": "bf47d8ba-4ca8-40e9-caff-984dfb3c3bb5"
},
"source": [
"from sklearn.tree import DecisionTreeClassifier, plot_tree\n",
"\n",
"# কিছু প্যারামিটার\n",
"n_classes = 3\n",
"plot_colors = \"ryb\"\n",
"plot_step = 0.02\n",
"\n",
"for pairidx, pair in enumerate([[0, 1], [0, 2], [0, 3],\n",
" [1, 2], [1, 3], [2, 3]]):\n",
" # দুটো করেসপন্ডিং ফিচার\n",
" X = iris.data[:, pair]\n",
" y = iris.target\n",
"\n",
" # ট্রেনিং\n",
" clf = DecisionTreeClassifier().fit(X, y)\n",
"\n",
" # ডিসিশন বাউন্ডারি প্লট\n",
" plt.subplot(2, 3, pairidx + 1)\n",
"\n",
" x_min, x_max = X[:, 0].min() - 1, X[:, 0].max() + 1\n",
" y_min, y_max = X[:, 1].min() - 1, X[:, 1].max() + 1\n",
" xx, yy = np.meshgrid(np.arange(x_min, x_max, plot_step),\n",
" np.arange(y_min, y_max, plot_step))\n",
" plt.tight_layout(h_pad=0.5, w_pad=0.5, pad=2.5)\n",
"\n",
" Z = clf.predict(np.c_[xx.ravel(), yy.ravel()])\n",
" Z = Z.reshape(xx.shape)\n",
" cs = plt.contourf(xx, yy, Z, cmap=plt.cm.RdYlBu)\n",
"\n",
" plt.xlabel(iris.feature_names[pair[0]])\n",
" plt.ylabel(iris.feature_names[pair[1]])\n",
"\n",
" # ট্রেনিং পয়েন্ট প্লট\n",
" for i, color in zip(range(n_classes), plot_colors):\n",
" idx = np.where(y == i)\n",
" plt.scatter(X[idx, 0], X[idx, 1], c=color, label=iris.target_names[i],\n",
" cmap=plt.cm.RdYlBu, edgecolor='black', s=15)\n",
"\n",
"plt.suptitle(\"Decision surface of a decision tree using paired features\")\n",
"plt.legend(loc='lower right', borderpad=0, handletextpad=0)\n",
"plt.axis(\"tight\")\n",
"\n",
"plt.figure()\n",
"clf = DecisionTreeClassifier().fit(iris.data, iris.target)\n",
"plot_tree(clf, filled=True)\n",
"plt.show()"
],
"execution_count": 11,
"outputs": [
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 640x480 with 6 Axes>"
]
},
"metadata": {
"tags": []
}
},
{
"output_type": "display_data",
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAgMAAAGFCAYAAABg2vAPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOzdd1yV5fvA8c85bESGIqgIDhyQiqaWK1fmKLeSWs7MlFK/5gJnuUvcM9yWOTJz6zez31ctdypuUHABgjJlwwHO+f1BHj0CCoocxvV+vXzFecb9XM95xTkX93Nf963QaDQahBBCCFFiKfUdgBBCCCH0S5IBIYQQooSTZEAIIYQo4SQZEEIIIUo4SQaEEEKIEk6SASGEEKKEk2RACCGEKOEkGRBCCCFKOEkGhBBCiBJOkgEhhBCihJNkQAghhCjhJBkQQgghSjhJBoQQQogSTpIBIYQQooSTZEAIIYQo4SQZEEIIIUo4SQaEEEKIEk6SASGEEKKEk2RACCGEKOEkGRBCCCFKOEkGhBBCiBLOUN8BCFGSBAUFERkZqe8wigxbW1ucnJz0HYYQxZ4kA0IUkKCgIFxdXUlKStJ3KEWGubk5fn5+khAI8YZJMiBEAYmMjCQpKYmff/4ZV1dXfYdT6Pn5+dG/f38iIyMlGRDiDZNkQIgC5urqSoMGDfQdhhBCaMkAQiGEEKKEk2RAiCLm5MmTTJ06FYBRo0bh5+fH4MGDWbBgAQDe3t4MHTqUbt26ER0djb+/P61bt+batWs5tpmenk56evpLr52YmMigQYP44osv2LJli86+Q4cO0bt3b3r37s0ff/zxGncohChokgwIUcQ0b94ctVrNN998Q7Vq1bTjDzw8PADw9PRk3bp1tG7dmsDAQFxcXGjdunW2bV25cgUvLy+GDRtGQkLCS6+9a9cu3N3dWbt2Lfv27dPZd/ToURYtWsTSpUs5dOjQ692kEKJAyZgBIYqgAQMG0KRJE0JDQ7PsU6lUjBkzhvv37/PZZ59le35gYCBDhw6lc+fOjB07Fnt7ewDOnTvH1q1bdY5dsmSJ9ueQkBDq1q0LgIGBgc5xffv25eOPP0aj0bBmzZrXuj8hRMGSngEhihi1Ws23337Lb7/9xvTp07PsNzY2ZuXKlQwcOJC9e/dm24aTkxMjR44kICCAxYsX4+vrC4BGo9E+Msju0UGlSpUICQnRxvGs7777juPHj/PXX3/x/fff58OdCiEKivQMCFHELFq0iIEDB/LBBx9w+vRpjh8/rrPfy8uLpKQkYmJiWLhwYbZtGBsb4+7ujru7O+Hh4WzZsgVHR0caN25M48aNc7x2z549GTlyJAcPHqRLly5AZi/F5s2b6dmzJ0OHDgWgY8eO+XS3QoiCoNBoNBp9ByFESXDx4kUaNmzIhQsX8r20cPDgwaxYsQILC4ts90+fPh13d3fq1KmTr9d9k97k+yWE0CWPCYQoBlxcXFi7dm22+/z9/bl37x6lSpUq4KiEEEWFPCYQooibMGEC8+fPz7Ldx8eHdu3a4eLiwqZNm3LV1sSJE0lKSsLc3Fznub9arWbatGnExcXRqFEjBg0axIkTJ9i+fTsGBgZMnDiRwMBAtmzZQnp6Ojdu3ODUqVP5dYtCiDdMegaEKEL8/Pzo27cv33zzjfa5/N27dwGoW7cuCxcu5NNPPyU5OZmHDx+SnJyc67aDgoJIS0tj2bJlZGRkEBwcrN23d+9eQkJCMDIyolKlSkBmlUGpUqUoVaoUZcqUoUWLFvj4+NC5c2cGDRqUj3cthHjTpGdAiCJk3bp1zJs3DwcHBzp06KCzr1KlSowbN47ly5dz6dKlLOdGR0czc+ZMnW0jR46kevXqADx48ABHR0cgs9ogJCRE+/rmzZs0a9aM4cOH4+7uTtu2bbl8+TLbt2/n8OHDbNmyhSFDhgCwdetW1q9fn+/3LoR4c6RnQIgiSKFQoFAodLY9GRNgZGREampqtuc9Xzb47PhhBwcHbdlgcHCwtgcAMhMNGxsb4On8Aq6urhgaGmJjY0N8fDyQ2btgZWVF6dKl8+lOhRAFQXoGhChChg4dysSJE6lZs2aeBwSWKVOGFStW5LjfyckJIyMjxo4di4mJCY6OjixatIg2bdrQs2dPRo0axd9//03Lli0B6N+/P19++SWJiYnaEsb169fnONGREKLwktJCIQpIfpTKRUdHs2TJEqKiomjbti09e/bM5ygLDyktFKLgSM+AEEVImTJlsjz3F0KI1yVjBoQoITZt2sSBAwfytc3hw4dr1yqAzMmN+vTpg4eHB6GhoWg0GoYPH86IESNynA1RCKF/0jMgRCG2detWjh07RunSpZk7dy6//vorvr6+xMfHs3LlSubMmUNMTAwxMTG4ubkRFRVFSEgImzdvpnPnzrRp04bAwEC+/PJLbZv3799n4cKFaDQanJ2d6dChAzNmzMDJyYmBAwfmaZbC1atX4+7urn1taGiIsbExRkZGWFtbc+LECerUqcOoUaMYMGAAKpUKY2PjfH2PhBCvT5IBIQqx27dv4+bmRrdu3TAxMQEyqwUePHigXVyoT58+1K5dm88++4xdu3YxbNgwYmJiyMjIYPTo0cTHxzNx4kSaNm0KwKpVqzAzM8PMzIyrV6/y9ttvY2Njw6effqqTCLysFDE7kydPRqlUsm/fPtatW0e5cuW05Yl2dnZERUVRoUKFfH2PhBCvT5IBIQqxadOmcfnyZSZMmMCsWbPYsWMH+/btY8aMGSQlJQFgaWmJiYkJlpaWQOYiRKmpqajVajIyMkhLS9NpU61WM2DAANzc3LTbqlWrxvr167l8+bLOhEHPr1r4svHGSmXmk0c7OzuuXbvG22+/zeXLlwGIiIigbNmyr/hOCCHeJEkGhCjE1qxZQ0BAAEqlkrJly1KhQgW8vb05d+4crVq1euG5xsbGzJ49m4CAACZPnszFixeBzL/uJ0+eTIUKFShdujQtW7Zk//79REdH065dO+35LytFBJgyZQq+vr54eHiwdOlSFi5cSHBwMJGRkSxbtozy5cuzdetWRo8eTb169eQRgRCFlJQWClFACrpUzt3dnZ07d77x67wpUlooRMGRagIhiqminAgIIQqWJANCCCFECSfJgBCF2LNle/mlWbNm7Nu3DwBnZ2c8PDxYs2YNAEePHmXQoEH069eP0NDQbM8fPHgwQ4cOxcPDg9TUVBITExk0aBBffPEFW7ZsyfacY8eO0aJFCzw8PDh27BgACxcuZOTIkQwfPhyNRsOZM2eoX78+CQkJ+X7PQogXk2RACD3x8PAgKioKtVpN3759CQ0NZcqUKXh4eLBnzx6dY58kBT4+Phw7dgxfX19Gjx7NiBEj+Pnnn/N03YoVK9K1a1cALCwsSE5O1pb/+fj4sHHjRiZNmpTjyoNmZmYoFAqsra0xMjJi165duLu7s3btWm2S8TyFQoGFhQUpKSlUqlQJlUrFxYsXWbFiBXXr1uXEiRM0adKE+vXr5+lehBD5Q6oJhNCT3r17s2PHDmrUqMH777+PoaEhqamp2Nvbs2XLFrp3757juYsWLcLZ2RkAX19f+vfvr903c+ZMoqOjta87duxIx44ds23H19cXjUZDp06d+PDDD9FoNCiVSipXrqxdwfB5K1euRKlUsmzZMg4cOEBISIh2FsInKxo+r0WLFrRq1YpHjx4xduxYFixYQLly5QBeeC0hRMGQZEAIPWndujVr1qzhypUrzJ07lw0bNtC1a1caN25Mt27ddI59Ur+fmJgIgEqlYvTo0dplhZ+VkZGhMz+AWq3OMYYn7ZqamqJWq1EqlajVaoKCgnSWMM7uHDs7OxISEqhUqRIhISHUr18/x2s9OcfGxobU1FTKli1LZGQkkLns8bNzHgghCp4kA0LoyZO/wENDQ7GxsaFZs2b4+Phw8uTJLPX4Dg4OLFiwgBMnTtCwYUO8vLwYNWoU9vb2VKlShVGjRmmPnTFjRq6uf/PmTebNmwdkJiZKpZJhw4YxdOhQ0tLStPvmzp3L5MmTteeNGzeO5ORkYmJiWLduHZA5d8HBgwfp0qUL8HQ8wJNZE3ft2sXhw4d5/PgxI0eOxNjYmAYNGjB69GhSU1P56quvXvFdFELkB5lnQIgCUljq5vMy/4C/vz+nTp1iyJAhebqGp6cn3t7eeY5t8ODBrFixAgsLi0LzfglREsgAQiFKGBsbmxwH+j3PxcUlz4kA8EqJwJkzZ0hJSclx3IEQ4s2RxwRCFDA/Pz+9Xv/JCoZPpicuLIyNjfH09NS+P/p+n4QoSSQZEKKA2NraYm5urjPyX7yYubk5tra2+g5DiGJPxgwIUYCCgoK0o+jflICAAPr168fw4cP5/PPP87XttWvXsnbtWrZu3frCpYzzi62tLU5OTm/8OkKUdJIMCFGMqNVqWrRoQXR0NJcuXdKO5s8vqamp1KtXj7Jly/L3339rSwaFEEWb/CYLUYxs2LCBU6dO4ePjk++JAICJiQk+Pj6cOnWKDRs25Hv7Qgj9kJ4BIYqJ8PBwXFxc6NatGxs3bnyj1xo0aBD79+/H398fOzu7N3otIcSbJ8mAEMXEwIEDOXjwIDdv3nzjg+4iIiKoVasWXbp04ccff3yj1xJCvHnymECIYuDo0aNs3ryZ+fPnF8jo+3LlyuHt7c1PP/3E0aNH3/j1hBBvlvQMCFHEpaam4ubmhp2dHcePHy+wQX1qtZqWLVsSGRnJ5cuX38gYBSFEwZCeASGKOG9vb+7cuYOPj0+Bju5XKpX4+Phw+/btV5pxUAhReEjPgBBFWEBAAHXr1mXMmDF89913eolh4sSJLFmyhGvXrhXI3ANCiPwnyYAQRZRGo6FDhw4EBgZy7do1zM3N9RJHUlIStWvXpkaNGhw+fBiFQqGXOIQQr04eEwhRRG3fvp0jR46wYsUKvSUCkDll8IoVKzhy5Ai//PKL3uIQQrw66RkQogh6/PgxLi4utGjRgl9//VXf4QCZSyOfOHECf39/rK2t9R2OECIPpGdAiCJo8uTJJCUlsWTJEn2HorV06VISExOZPHmyvkMRQuSRJANCFDFnz57Fx8eH2bNn4+DgoO9wtBwcHJg9ezY+Pj6cPXtW3+EIIfJAHhMIUYSkp6fTqFEjDA0NOXv2LAYGBvoOSUd6ejqNGzdGrVbzzz//YGgoq6QLURRIz4AQRciyZcu4evUqPj4+hS4RADA0NMTHx4fLly+zfPlyfYcjhMgl6RkQoogIDg7G1dWVIUOGsGzZMn2H80KjRo1i48aN+Pn54ejoqO9whBAvIcmAEEVEjx49OHv2LH5+flhZWek7nBeKjY3F1dWVJk2asGvXLn2HI4R4CXlMIEQRsG/fPvbs2cPSpUsLfSIAYGVlxZIlS9i9ezf79+/n9u3bpKWl6TssIUQOpGdAiEIuMTGRt956i9q1a3Pw4MEiM8OfRqPho48+4saNGyQkJLBw4UIGDx6s77CEENmQob5CFHLTp08nPDyco0ePFplEAMDPzw+Ahw8fYmRkxN27d/UckRAiJ5IMCFGIXblyhcWLFzNr1iyqVaum73DypEyZMoSFhZGWloZKpdImB0KIwkceEwhRSKnVapo3b05cXBy+vr4YGxvrO6Q8S09PZ9GiRUyaNAkXFxeuX7+u75CEENmQZECIQiY2NhZvb28cHR358ssvOX78OC1bttR3WK/l3r17GBkZFaoZE4UQT0kyIEQhs2fPHnr06IGlpSVdu3Zlw4YNGBkZ6TssIUQxJqWFQhQyjx49AiAtLY2dO3eyZcsWPUckhCjuZAChEIXMmTNnAFCpVIwbN47evXsX2LWDgoKIjIwssOsVNba2tjg5Oek7DCHynSQDQhQyFStWxNHRkQMHDuDm5lZg1w0KCsLV1ZWkpKQCu2ZRY25ujp+fnyQEotiRMQNCCAAuXrxIw4YN+fnnn3F1ddV3OIWOn58f/fv358KFCzRo0EDf4QiRr6RnQAihw9XVVb7shChhJBkQJYI8C9clz76FEM+SZEAUe0FBQbi61CIpOUXfoRQa5mam+PnfzJeE4OTJk/z3v/9l9uzZjBo1iq+++op58+ZRp04dxo8fj7e3N7du3SIiIoKNGzcSHh6Oh4cHK1asoE6dOtm2mZ6eDoCh4Ys/ou7cucOcOXOIjY1l586dOvvWr1/PP//8Q3BwMG5ubnz33Xc4OzvTrl07GjRowLBhw1773oUoLiQZEMVeZGQkSckpLO9VnRq2ZvoOR+8CIpMZ9VsgkZGR+ZIMNG/enIMHD/LNN99QrVo17XgDDw8PADw9PQFYvHgxgYGBvPvuu7Ru3Trbtq5cucKWLVuIiIhg0aJFWFtbv/Da1apVY/369bi7u2fZ9/nnn/P5558zZswY7QJJFhYWJCcn4+jo+Ip3K0TxJMmAKDFq2JpRt6KFvsMolgYMGECTJk0IDQ3Nsk+lUjFmzBju37/PZ599lu35gYGBDB06lM6dOzN27Fjs7e0BOHfuHFu3btU5dsmSJbmOKyUlhbt371KrVi0AfH190Wg0dOrUiQ8//DDX7QhR3MmkQ0K8hlmH72W7/ad/HnIvOm+PJeYeuc+0Q3eZe+S+zvYkVQajdwUyYe9tdl2JeNVQ3xi1Ws23337Lb7/9xvTp07PsNzY2ZuXKlQwcOJC9e/dm24aTkxMjR44kICCAxYsX4+vrC2Qug5yenq7zLy927txJz549ta+VSiUGBgaYmpqiVqvz1JYQxZn0DAiRSwERSSw6FkK1sqZcepDAlgFvEfQ4FYC2Ky/xcX07roYlsKCbMxEJaaSk5f7L5sHjVNIyNMz6qCqzDt/jQWwqDlYmABzyi6ZT7TK0r1UGjx236OlW7o3c36tatGgRAwcO5IMPPuD06dMcP35cZ7+XlxdJSUnExMSwcOHCbNswNjbG3d0dd3d3wsPD2bJlC46OjjRu3JjGjRvneO2oqCimTJmCr68v3333HZMmTWLAgAFs3rwZgF9//ZXt27cDcPPmTebNmwdA69atUSrlbyEhnpBkQIhc2nohnCntKlO+tDGfbr6hs6+CpQkezSuy4UwY1x9mnbQnJimNxcdDdLZ99m55qpbNHMMQFq+i4r9f/g5WJoTFqbTJQFisChe7zGfnBoXw+2v8+PHan6dNmwbAxo0btduefAHnlp2dHWPGjMnVsWXLlsXHx0dn25NEANDpiahVqxYbNmzIUyxClBSF8KNFiMJNoQCFQqGzzcw481fJ0ECBKj37HoEMtUbn37OzfVUobUxYXGYvQ2icigqWT5crrmBlTFicCgB1EZkizMXFhbVr12a7z9/fn3v37lGqVKkCjkoIkRPpGRAilz5taMfcI/epVtYMc6O85dE25kbM6VQtx/0O1iYYKRVM//0eJgYKHKxMWH0qlOZVrfjItQxTDt7l/27F0K6WzeveRoGYOHFittsnTJjA/Pnz2bRpk852Hx8f2rVrh7Ozc56ukZSUhLm5Od9//712+7Fjx5g2bRq1a9emb9++OVYuCCGekmRAiFyyLWVE1bKmRCel8XH9zOf2a/vU0vnvwHfKA9CsqlWe25/UrrLO6+HNKmp/Xtyj+ivFrE9+fn7MmDGDmjVrcu7cOX7//Xfu3r0LQN26dRk8eDAXLlxg/fr1PHz4kOTk5Fy3HRQURFpaGsuWLWPChAkEBwdrywUVCgUWFhakpKRQqVKlN3JvQhQ3kgwIkUs25kZMeF9m7cutdevWMW/ePBwcHOjQoYPOvkqVKjFu3DiWL1/OpUuXspwbHR3NzJkzdbaNHDmS6tUzk6IHDx5ov/ydnJwICQnRvm7RogWtWrXi0aNHjB07VpaAFiIXZMyAEG/IL77hHLkZk69tNltyEa/9d/j5/CMATt6NZfSuQEbuDODhv+MKChuFQpFljMWT8QJGRkakpqZme97zJYXPrqnm4OBASEjmgMzg4GCdHoAnVQI2NjY5ti2E0CU9A0I8Y/eVCE7fi6OUsQETP3DiwPUoroUlkqjKYE6nqiz76wGxyek8TknnLftSxCSlERqnYnmvGgzc4kezKlbci05h4Dv22jZDHqey+lQoGg1ULmNK6+pWLDoWgoOVCe71yuFib57r+MyNDUhJU1PRKnOA4eZ/HrHKvQa3IpLZdjGcMa0LT7f40KFDmThxIjVr1szzYMEyZcqwYsWKHPc7OTlhZGTE2LFjMTExwdHRkUWLFtGmTRvu3r3L4cOHefz4MSNHjnzd2xCiRJBkQIhn3ItOxdXenPYuZTAxzPwL08hAQVicimthiQB0qWNLLTszxu65zbq+tfDcd5vHyemo1TC0SQUSVBnMPXKfho6lAfjx3ENMDZWYGinxf5RInQrmWJka0r2urU4i8LLyQ4A/PNzQAAO3+PF+DRs0gFKpoJK1ibYaobCwt7enRo0aREZGMmjQIADt+gFP/vtkyuJXGeT33Xff6bweO3YsAG+//bbORENCiJeTZECIZ4xpXYnrDxOZ/cd9JrzvyP7rUWz61IVFR4NJ/ncSodImBhgbKLEwMQAykwVVuhq1RkOGRkN6hm79n1qjoVe9crxV/ulfx042pmy/GM6NR4n0rm+n3Z7xXO3g85WESmVmd7uJoRK1WoNSAWq1hgexqVSwNMmvtyFflClTJstzfyFE4STJgBDP+Pn8I+5GJaNUgI2ZIXYWRqw68QDfBwk0qWL5wnONDJQsPR7C3egURrVw4Oq/PQmfNa7A938GYVfaCAtjAxpXseTIzRgeJ6fT0vlp1cHLyg8DI5NZdeIBAE2rWKFUKujX0J7x+26TlqFhynPVCEXFpk2bsLW1pXPnzvnW5vDhwzl16hRXr14F4Nq1a9qehEmTJlGnTp0cSxOFKIkkGRDiGf0b2eu89u6aWff+1XsOgG7J4JJ/y/2efIEbGyrwbPu02uDZnoAV7jV02m3+CqWH1W3NWNRdt8TwvWpWvFct7229rq1bt3Ls2DFKly7N3Llz+fXXX/H19SU+Pp6VK1cyZ84cYmJiiImJwc3NjaioKEJCQti8eTOdO3emTZs2BAYG8uWXX2rbvH//PgsXLkSj0eDs7EyHDh2YMWMGTk5ODBw4MMfljrOzevVqnZUMly5dysqVK1EoFHh6ejJlypQcSxOFKIkkGRAinzyZa6AkuH37Nm5ubnTr1g0Tk8zHE0ZGRjx48EC7yFCfPn2oXbs2n332Gbt27WLYsGHExMSQkZHB6NGjiY+PZ+LEiTRt2hSAVatWYWZmhpmZGVevXuXtt9/GxsaGTz/9VCcReFnZYXZiY2O1yyHHx8e/sDRRiJJIkgEhRJ5NmzaNy5cvM2HCBGbNmsWOHTvYt28fM2bMICkpc20GS0tLTExMsLTMfLxibGxMamoqarWajIwM0tLSdNpUq9UMGDAANzc37bZq1aqxfv16Ll++rB2ECGRZvfDZssPsWFlZERsbi0KhoHTp0llKE7t37/7qb4YQxYAkA0Jk44tfbub7X/pd111l5HsOtHcpQ7MlF2nhbE3dCqXo38iek3dj2eEbQYZaw9T2lSn/zNoET7zKMUoFTDxwh08a2OfrVMZr1qwhICAApVJJ2bJlqVChAt7e3pw7d45WrVq98FxjY2Nmz55NQEAAkydP5uLFi0DmX/eTJ0+mQoUKlC5dmpYtW7J//36io6Np166d9vyXlR0C2pUMPTw8WLp0KaNHj2bUqFEAeHp6ZluaKERJJsmAKHG89t/Bq60j1qaGjPgtgG87VGHTuYfEJKfTpro1HV3LaI99khT89M9DqtuaYWlqyA7fcNLVGho6lqZXvdwvJ2xf2pj2Lpltv8p8Aa96TAeXMlmOe13Dhg3Teb169Wog84sWdEsFn6xD8OQL3MTEhFmzZmn3P9sT8PxsgW3atHml+ObMmcOcOXO0r+vUqcNPP/2kc8zzpYlClGSSDIgSp0vtsuy/FkXVsqY0r2qFoVKBKkNNuVJG7LoSoZMMPG/1qVCqlDEF4FpYok4ysPhYMDHJT7uv21S3pk2N7P8af5X5AvLrGH17MseAEKLwkGRAlDjNqliy5cIj/B4lMfEDJ7ZfDKd9rTK8XcmCIdv8dY5V/juNbpIqc46BtAwNnzepgLVZ1l+dDLXuPAEvWm74VeYLyK9jhBDieZIMiBJHqVRQycqEh/EqrM0MaeRUms3/POKfoHiMDHSX6yhf2hifk6GcC4rDrWIpRrxXkWmH7mJbyghHaxOGNKmgPXb8+7l77pzb+QKW/RXCf1o+fRTwqse8Se7u7vn+l36zZs2YOHEiXbt2zdVcAM7OzrRr144GDRowbNgwjh49yqZNm0hPT2f+/PlUrFgxyzmDBw/G0NAQQ0NDli5dSnp6Ol999RXGxsa0bt2afv364ePjw59//ik9GaJEkGRAlEhT2j/9omzkWJpG/04d/MSTwYMzPqwCgEfzp18oy3vpzhmQV7mZLyAwIhk7C+PXPuZ1eHh4MGfOHG1536JFi1i5ciVRUVF07NhRZwT+k6TAx8cHFxcXrKystF/ITZs2pX///rm+bsWKFenatesLlyl+loWFBcnJydp9Pj4+bNu2jRs3brB+/XqmTZuW5RwzMzPS09OxtrbGyMiIHTt24O7uTpcuXejTpw/9+vXDw8ODP//88xXeOSGKHkkGhCgg1qaG/OEfrR1E+CLVy5lRvZzZax8THq/iYkgCH9fP/UDHJ3r37s2OHTuoUaMG77//PoaGhqSmpmJvb8+WLVteWI63aNEinJ0zJ2zy9fXVSQZmzpxJdHS09nXHjh3p2LFjljZyOxeAr68vGo2GTp068eGHH6LRaFAqlVSuXFlbPvi8lStXolQqWbZsGQcOHCAkJIS6desCYGBgkIt3R4jiRZIBUWIERCbr9fpPVjK8GppQoNft39BO57q5fR9at27NmjVruHLlCnPnzmXDhg107dqVxo0b061bN51jnywbnJiYOQWzSqVi9OjR2NhkHUCZkZGhM0+AWq3O9vq5nQvgybVNTU1Rq9UolUrUajVBQUE6Sxtnd46dnR0JCQlUqlSJkJAQ6tevn2M8QhRnkgyIYs/W1hZzM1NG/Rao71AKDXMzU2xtbV94zJO/rkNDQ7GxsaFZs2b4+Phw8uRJjI11H084ODiwYMECTpw4QcOGDcuWO+IAACAASURBVPHy8mLUqFHY29tTpUoVbY0/wIwZM3IVY3ZzARw5cgRra2veeecdAG7evMm8efOAzORFqVQybNgwhg4dSlpamnbf3LlzmTx5srbtcePGkZycTExMDOvWrQMy5zk4ePAgXbp0yVV8QhQnCs3Lpu4SohgICgoiMjJS32EUGra2tjg5Oelsu3jxIg0bNuTChQs0aNBAT5G9eFDi3LlzGTFiBFZWuR8X4e/vz6lTpxgyZMhrxVJY3h8h3gTpGRAlgpOTU5YvP6ErKipK3yEAYGNjw759++jatWuWfc/+dZ9bLi4uuLi45Pk8Hx8fateunefzhCiKpGdAiBLu1q1bLFy4kI0bN5KWlsbPP/+Mq6urvsMqdPz8/Ojfvz/vvPMOs2bNon379ij+nYdCiKJOkgEhSqizZ8/i7e3N7t27sbOzY/DgwSxfvly70JDIytTUlBo1anD16lXc3Nzw9PSkd+/eGBkZ6Ts0IV6LJANClCAajYb//ve/eHt7c/z4cWrWrMmECRPo378/pqamMrbiJWxtbXF0dOTYsWN4e3vz+++/4+TkxNixY/n888+xsLDQd4hCvBJJBoQoAVQqFdu3b2f+/Plcu3aNJk2a4OXlRdeuXbVldiLvLl++zIIFC9i2bRuWlpaMHDmSkSNHYmdnp+/QhMgTSQaEKMbi4+NZu3YtixcvJiQkhM6dO+Pl5UXz5s3leXc+un//PkuWLGHt2rVkZGTw2WefMW7cOO3ES0IUdpIMCFEMPXz4kGXLlrFq1SqSkpLo168f48ePl9Hxb1h0dDSrVq1i2bJlREVF0atXLzw9PWnUqJG+QxPihSQZEKIYuXXrFgsWLODHH3/ExMSE4cOHM3r06Bxn4hNvRnJyMj/++CMLFizg9u3btGnTBk9PTzp06CA9MqJQkmRAiGLg7NmzzJs3jz179mBvb8/XX3/N8OHDsba21ndoJVpGRga7d+9m3rx5nD9/XioQRKElI4eEKKLUajUHDx6kVatWNGnShBs3brB27Vru3buHl5eXJAKFgIGBAe7u7pw7d46jR4/i4OBA//79qV69OkuXLiUhoWDXqRAiJ5IMCFHEqFQqfvzxR9zc3OjcuTMqlYrdu3dz48YNPv/8c0xMTPQdoniOQqGgdevWHDp0iMuXL9OqVSvGjx+Pk5MT06ZNIzw8XN8hihJOHhMIUUTExcVpKwMePHhAly5d8PT0lMqAIiooKIjFixdrKxAGDx7MuHHjqF69ur5DEyWQJANCFHJhYWEsW7aMH374gaSkJPr378/48eN566239B2ayAfR0dH88MMPLFu2jIiICG0FwpOVGYUoCJIMCFFI3bx5kwULFvDTTz9hYmKCh4cHo0ePxsHBQd+hiTcgOTmZn376iQULFhAYGEjr1q3x8vKSCgRRIGTMgBCFzJkzZ+jZsyeurq4cOHCAmTNnEhwcjLe3tyQCxZiZmRnDhw/H39+fnTt3kpiYyIcffki9evX4+eefSUtL03eIohiTZECIQkCtVnPgwAFatmxJ06ZNs1QGWFlZ6TtEUUAMDAzo1asXZ8+e5dixYzg6OjJgwACcnZ1ZsmSJVCCIN0KSASH0SKVSsWnTJurWrUuXLl1IT09nz549UhkgUCgUtGrVioMHD3LlyhXatGnDhAkTcHJyYurUqTx69EjfIYpiRMYMCKEHcXFxrFmzhiVLlvDgwQO6du2qrQwQIidBQUEsWbKENWvWkJ6erq1AqFGjhr5DE0WcJANCFKCwsDCWLl3KDz/8QHJyMgMGDGD8+PG4urrqOzRRhMTExPDDDz+wdOlSIiIi6NmzJ56enrz77rv6Dk0UUZIMCFEAbt68yfz589m8ebNUBoh8k5KSoq1ACAgIoFWrVnh5edGxY0epQBB5ImMGhHiDTp8+TY8ePXB1deXQoUPMmjVLKgNEvjE1NWXYsGH4+fnx22+/kZyczEcffYSbmxubN2+WCgSRa5IMCJHP1Go1+/fvp0WLFjRr1gx/f3/WrVvH3bt38fT0lMoAke8MDAzo2bMnZ86c4fjx41SuXJmBAwfi7OzM4sWLiY+P13eIopCTZECIfJKamsrGjRupU6cOXbt2Ra1Ws3fvXq5fv86QIUOkMkC8cQqFgpYtW3LgwAGuXr2qXTrZycmJKVOmSAWCyJGMGRDiNcXGxmorA0JDQ6UyQBQqwcHB2gqEtLQ0Bg0axLhx46hZs6a+QxOFiCQDQryi0NBQli5dio+Pj1QGiEIvJiYGHx8fli5dSnh4OD169MDT05PGjRvrOzRRCMhjAiFy4fLly6hUKgD8/f0ZOnQoVatWxcfHBw8PD+7du8f69eslERCFlo2NDZMmTeLevXusXr2aq1ev0qRJE+3Syk/+Lrx06ZIMPCyBJBkQ4iX27t1L/fr1WblyJd27d9dWBsyePZugoCDmzZtHxYoV9R2mELliamrKF198oVOB0KlTJ9zc3Ni0aRPNmzfnq6++QjqNSxZ5TCDEC9y4cYNGjRphbm5OVFQULi4ueHp68umnn8qAQFEsaDQa/v77b7y9vTl48CA2NjbExMSwePFivv76a32HJwqIJAOCoKAgIiMj9R1GoWBra4uTk5P2tY2NDY8fP8bU1BQHBweWLl1Kp06d9BihEG/Go0eP6NatG7du3SImJgaAixcv8vbbb2uPkc+Kp57/rCjqDPUdgNCvoKAgarm6kJKUrO9QCgVTczNu+vlrf8m/+OILQkNDsbW1JTU1lfLly+s5QiHeDFNTU5o3b867776LSqUiIiKCatWqafcHBQXhUsuV5JQkPUZZeJiZmuN/06/YJATSM1DCXbx4kYYNG/L2yn5Y1LDXdzh6lRDwCN8RW7hw4QINGjTQdzhCFCpPPivca63Azry6vsPRq/CkQHbeHFmsPiukZ0AAYFHDHmu3SvoOQwhRyNmZV6eihZu+wxD5TKoJhBBCiBJOkgGRa9Hn7uL//SEArk7eRfytR/j+Zxu3Vx0F4NLobVyesINLo7ehyVATc+Eex9suID0xNcc2M1JyX8+c8iiO/3t3NnF+YTrb/b8/xOXxO7jiuZP0JBWRJwM52W05Vzx/JfJk4CvcqRDidd2PPceRe/MAOBA4hfCkAH67+TUnQn4A4LdbX7MnwJPfbn2NWpNBcNwFVlz8gNSMxBzbTFOn5Pr68apHLPynCY8S/XW2Hw1awq5bY9hyYwixqaHceXyKtZe7szfAizuPT73CnRYPkgyIXCvzblU0ag3+8/6LeeWylK6ZOcag8qBmANRf+gn15vfGsLQpKQ9jsWlYBcva2a/MF33+Htem7ubalF25vn7gyv9RoUu9LNvj/B9Sb0FvbJtX5+GhK6AAA3MTMlLSMasoiwIJoQ+Vrd5Fg5o/73ljY1YZO/MaALxTYSAAvWouoXsNb0wNShOneoijZUMqlKqTbVtBcec5eHsaB25PzfX1/w5eRR3bzlm2hyfdpGfNxdSx7cKD+MsoUGBsUIp0dSpWJhVe4U6LBxkzIPKkknsjTnRaSrtL32a7Pz7gEWpVOmYONtnujz5/D7/ZB6jUqyG1JnTEyMoMgPD/+RF+9GkGb2xtTs1xHbSvg7ado2LnekQcv5mlzQof1eXq5MykwrSCFQ49G2DbrDqpEfFc/3YvDVb1f+X7FUK8uvp27qy+1Bmvxr7Z7o9ICiBdrcLaJPs/GoLizvPH3bnUs+vJ+5XHY2aYmdzfij5KQMxR7XFmhta8X3ms9vWFh9upbduJwMd/ZWmzmlVz1l/5GA1qBrz1I0YG5gyybkqCKoJDd6bT22Xl69xykSU9AyLXNGo1N+f/TqP1g7m14HCW/XF+Ydz54Rh1ZvfIsQ1L1wo4fdKYx75B3Fn7Fwl3Iv5tW4MmXf30X4Zukctj3yBC918i/Kg/9zef1tnn2Psd6s7tiVWdilhUt0OhzPzf2sjKDHVq+uvethDiFag1av53fwGfuK7lf/cXZNn/KNGfEyE+dHKelWMb5Uu9RYPyfQmJ9+X0g3VEJt8BQIMatSZd+09Dhs55D+IvcS3yAAExx/gnbLPOPv/oP/jc7VfaVfbiwqNtKBWZnxemhlZkqFWve9tFlvQMiFy743Mcx48bUa5lTWLO3yPy1NPn8Rq1mjN9fLB735VrU3dT4+t2mFW0ztKGYSkTHPu8g2Ofd0gKjiZk5wVqjmmH/QdvYf/BWzle283bHYCb83+nQufMRwUXR26hwYp+3FlznITbESgMlNSZ3Z2wg1cIP+ZPemwKVYa8l8/vghAiN049WE19O3eq27QkOP4Cdx8/TeLVGjUbr/alZpk2HLw9jdZOo7EyyTqlt7GBOQ3se9PAvjcxKSFcCt9JG6evqVWmLbXKtM3x2l1rfA/A/91foH1U8OvNUXxcazl25rXYG+BFYloUbZzGcD3yEAExx0hJj6Vxxc/y+V0oOiQZELnm/FUb7c81x7YHIHj7PwAolEraX5mRp/bMHctQc0y7PJ1Ta0JH7c8NVvQDoNqwVjrHVOjkRoVOUvokhD69V+lL7c9tnMYAcPHRLwAoFUomNrmUp/ZsTCvRxilv0yO3rTxe+/PHtZYD0KHqFJ1jKljUprbtR3lqtziSxwTitVjUsOP+z2ey3Rdz4R7q1DQUSkUBRyWEKIxszZ05/3BLtvuC4y6Qrk5BKV9LeiE9A+K11BiVc1edTcMqNFxdpeCCEUIUaq0cR+W4z9GyIX0sGxZgNOJZkgyIfHFjxj7e+rZrlu33fjxFuVY1KVXFNtdt+c0+QEayCgMzY1ynPi0NSrwfRcCSI6THpdBo/WAAos7eIXSPLwoDJdVHtUWdlsG1qbsxtjanlHO5FyYrQoiC9/udmXSs9k2W7efCfsLZuiVlzarkuq3Dd+eQpk7GSGmm0/2vykhiX+AkDJRGVLVqRn27ngBceLgN30e/MrTeLu7FnuVK+G7iVGE0sO/LW7Yfvva9FWXSHyPyLP7WIy4M/wn/ef/lzCerAUgKigbgWGtvbv9wjItfbiYjWUVqeBwZybmfWCgpJAZ1WgZ15vREk6Em+UGMdl+pymWpv7ivzvF31/yFgbkxBubGGFmbE+8XRsXObtRf0pe4qw/y4W6FEK8qPCmAX/w8+POeNz9e+xSAmJRgAJZfeJ8TIT7s8P+KtIxk4lXhpOdhUqHHKSGoNel0dp6NRpPB49Snv+/XIw9Rx7YTPWoswD8qs/IpOvk+SWnRlDIqC0AVq8Z0rfE9vWou5XY2JYgljfQMiDwL2nIG16mdMa1gxdm+a3T2mVawxvnL1txd9zex10OznKuKSeTWoj90tlUd8h6lqpYDIOVhLGYOmVUIZg42JIfF5jhnAUDcjVAarB5AxNGbPPjtAuU71uH80E0EbTtHJfdGr3urQojXcOHhVtpXnYqlSXl+vPqpzj5Lkwq8V8mD0w/WE5Z4Pcu5SWkxHA1arLOtScXPKGtWFYA41UNtBYKViQNxqWHa+QriVGGUL+UCgFJhgFqj5uQDHz6sNoNf/Udo27v46Bd8H+3M88DE4kh6BsQrUygU8NzYQENz48x9RgY51vjrzCeQrubZdTNNy1uRHBoLQHLoY8wqvHgGQYsa9igNDTCyNiM9MZXg7eeoNaEjzX77ivA/b7z6zQkh8o2CrJ8VxkpzAAyURqTnUN//7FwCmfMJPP2wsDQuT2xq5h8csapQLJ+ZPdDSuAKxqZnTlms0amJS7pOYFs3hu7N5mHiDm9H/B0AD+z4MrrOV06Hr8u1eiyrpGRB55tSvCX5zDmJRrRwG/37555axTSnqftcrx/3mlWxQGim5/u1elMYGmDnYcNvnGLbNq2PmYIP/94eIvfaAgGV/UuM/H1CpV0OueO4kI0nFW9O7khoRz60Fh3mw+yJmjmVe91aFEK+hYflP+ePeHGzNqmm//HPL3MiGLtXn5rjf2rQSBgojDt2ZjqHCGGsTB06GrKaqdTNq237E/tuTuRX9f9Qq256yZlXp65r5SDMuNYxaZdpyPfIQdx+fRKVOpl65nD+TSgqFRqPRvPwwUVw9WaO8xR9jc72EsSomkTtr/iItJgnbFjWKTU3/4ysh/N1+UbFao1yI/PLks+Krt3/P9RLGSWkxnHqwluT0GKpZtyg29fyhCVdY5duxWH1WSM+AyDNjm1K4eL3eyNvg7ecwLlMK+/a18ykq+L/GcyjXqiZWdStReUBTAMKP+nN10m+0PTPlJWcLIfKbuZENH1TxfK02Lj76BXPDMriUzdsEZS+zN8ALYwNzPqz2LeFJAZwN3YBCYcC75QdiV6pmvl6rKJBkQORayK4LRJ26jWEpE1wmdyJs/yVirz4gPTGVut/1ImDJEdJik0l7nIzlWxVQxSSSHBZLgxX9ONt/HbbNq5N4N5Iqg5tp20wKjuaOzzE0GihVpSzlWrtwa+FhzBxsqPRxIyxdc7+KmGEpYzKS0zD9dxrktNhkok4FYpXDyolCiDfjcvgu7saexsTAgnZVJnIt8gBhCddIzUigi/NcjgUvJSU9luT0WMqXciUpLYZYVRgf11rO5usDqWrVjOjke7xbcaC2zZiUEE4+8AGNhjJmVahu05qj9xdiZerA23YfY//vgMHcuBZxAIfS9YhICgDgRMgPWBiXI0OtwsK4XL6/H0WBJAMi15LuRWHpWoHyHetgYJL5v47SyICUsFhir4YAULFbfUrXKs+lr7fzzobPuDx+B6rHSWgy1FT9ogXpCan4zTlAmYZVALi36SRKUyMMTI2I8wvDso4DRlbmOPRsoJMIvKwKAaDln+NAA2f7r8W+rSsBS45Q/T9tuTL+1zf8zgghnhWdcp/ypVxxKdsBQ6UJAEqFIXGpDwlLvAZA3XJdsTOvxa5bY/j0rfXsCZhActpj1JoMmjoMRZWewOF7c3H6dyKic2GbMFKaYqg05WGiHxVK1cHM0Jp65XroJAIvq0JIUEUQlniNRuX7aZOB0IQrDHXbRWxqGKcerH3t3oyiSJIBkWs1x7Yn9voDbszcTy2vDwndd5l3f/qcmwsOa+cSMLQwRWlsiKGFKZCZLKhV6aDRoMnQoEnTXV0MtSazB+Ctp4uUlKpclqCtZ4m7Hopjn3e02zXpap1Tnx/t8mS1QqWJEemJqSTei+TWoj+IuxFKyM7zUmooRAFp4zSGsITrHL4zi7ZVPLkWsY/+tX/kf/cXospIBsDEoDQGSmNMDEoDYKAwIl2jQqNRo9FkkKHRrUbSaNTUt3enfKmnC5rZmDlx4eE2whJv0MC+t3a7+vlzn6lCuBt7moS0SI4GLeJh4g0ik25TxrQyRgbmmBlaocpIzPf3oyiQZEDk2v3Np0m8E4FCqcDYxhwT+9IErvgfj32DKNvU+YXnKo0MCFhyhMQ7EdQY/QGxVzJ7EqoMeQ//7w5hYmeJoYUJZZtU49GRG6hikijXqpb2/JdVISQEhhO44n8A2DZzxrCUCe9sHALA+dBYSQSEKED/hP1MVPIdFAol5oY2WBjb83fwSkLiL1HFqukLzzVQGnEsaClRyXdo5fQfQhOuAtC44hCO3PuO0sZ2mBhYUMWqCf7RR0hOe0x165ba819WhVC3XFfqlutKTEowZ0I3YGvuTFOHL9gb4EmGJo3WjqPz500oYqSaoIR7lWqCV3H+803aKYQLK6kmECJnr1JN8Cq23fiCT95a+8bazw/FsZpAJh0SBaKwJwJCiMKhsCcCxZUkA0IIIUQJJ8mAyLPzn2/K9zZPdF7Gw8OZo4z9Zh/g2pRd+M0+8MJzrnj+yvXp+wC4s/YvLo3ZzuWxv5ASHpft8b7/2cblsb9wxfNXMlLTiQ94xKkeK4nzC8vfmxFCAJld/vlt9aUu+P27+NDhu3M4cHsqh+/OyfH4PQGeLL/w/kvb3RvgxX/vzADgt5tfs/vWOPYGeJGuTiUiKYB1V3rxKNE/f26iEJJkQOi44vkrquhENGo1F4b/RMrDWPy+O8QVz18J++9VnWOfJAX3fjxF5MlAYq+GcG3qbq5O/I2QnefzdF1Te0vKd6jzwlULnxW6/zLW9R21r6NOBVJ/cV+c+jchaMvZbM8xMDUCBRhZmqE0UlK6hj1lm7144KMQInt7A7xISotGrVHzi58HcakPOXLve/YGeHEj8r86xz5JCs6F/cSdx6cITbjKwdvT2B84iUvhv+XpupbG5XEt2+GFqxY+q3sNb2zNXvx7/mTegSeMlKagUGBqaIlSYUQ58xpUfcnAx6JOqgmEjopd6xO67xKlqpXD9r0aKAyUqFPTMbEtzYPfLlLhw7o5nnt79XFKVc5cHjT26gOdEfy3Fh5G9ThJ+9qujQt277tmaSM3qxamRsQTe+0Blfs1If5WOABO/ZtyxWsnxjbmqKKzLw2q+31PFEold9b9xaMjNyjfoU4u3xUhxPPqlOvC1Yj9lDWrSjXr91AqDElXp2JhbMvliN28ZZvzLKUnH6yhjGllAMISrlHf7mml0P/uLyI5/bH2dQ2bNtQs0yZLGy9atTAvspt3oHP1uSgVSk4/WMfN6CO4lu2Q53aLGkkGhI6yzZy5v/k0cTfCcJn8EcHbzlG+Q22sG1Tmn8HrdQ9WZi5DlpGUueKYRpVB1S9aYmyddUESTYZGZ54AjTr7IpbnVy0s/2HWL+yoU7dRRcZr5xBIuB2OfVtX7Nu6EvH3LeJz6PZ/Mg+BiW1p0hNTX/JOCCFepKpVM86H/cyjRD/aVZnIhUfbcSnbAcfSb7Plxmc6xyoUmb97qozMPwgy1CqaVRyKmZF1lnY1ZOjME6BBneUYyLpqoatJx1e6j+zmHbA1z+xJKGVkW2LmHZBkQOhQKJWYVbIh5VEcxtbm2LxThfs/niL6n7sojXT/dzGtYMXtVUeJOncXK7dKOI98n2tTdmFiWxpzxzJUHdpCe2wtz9z9oma3amHE8ZsYWZph/bYTkDnLYcVu9UkKiubuhhNYONsR/Ot5Ys7fQ61Kp87sHgAELP2TGqM/0LZ9/du9ZKSkkRabTL2FvbO9vhAid5QKJdamlYhTPcLMyBony0acC9tMUNw/GCiMdI61NC7PiZAfuB93jooWbrR0HMGB21OxMC6HtUklmjp8rj22beUJubp+dqsWBsYcx9TQikql62uPO3Lve0ITr7E3wItOzjO5F3tG55js5h34750ZpKlTSE5/TI8aC/Ph3Sr8JBkQWbw1rYv25zKNqlCmURWd/U/KBOvM7A6A81dPu/AarOz/2td3ndJZ5/XjS8FU+ax5luPMncpQe3pXABw/boTjx08fS8QHPMKkXGmd42vP6PbasQkhnupQdar2ZyfLRjhZ6k7u9aRM8CPnzIF57/Gldt/HLite+/rtq07WeR0Sf5kmFQfrbGtXZSLtqkx84TEANqaOfFjtWwDtf0sSGUAoCgUjazNtNcHzaoz+ACNLszy1V7qGPU6fNn7hMfEBj0gKjsHQ3DhPbQsh9MfUyEpbTfC81k7/wdTQ8oXn5+aY50UkBfA4JQQjg6yPQIsL6RkQACQEPNLr9SsPyvzL//G/0xQXlKpDW6CKTUZ1JUTv74EQRUF4UqBer9+4QuZKhqEJVwr0uk0dhpCS/pjQhMd6fw/eBEkGSjhbW1tMzc3wHbFF36EUCqbmZtja2uo7DCEKHVtbW8xMzdl5c6S+QykUzEzNi9VnhaxNIAgKCiIyMvKFx4SFhTF+/Hju3r3L9OnTad++fQFF9+r8/PwYP348qampzJ8/n7fffvul59ja2uLk5FQA0QlR9OTmsyKv/vjjDyZNmsTixYtp2bLly0/IpeTkZNzd3alatSrLly9HoVDkW9tQ/D4rJBkQL/XXX3/h7u6Oubk5e/bsoX79+i8/qZAIDw+nd+/enDx5kuXLl+Ph4aHvkIQQ/4qNjcXFxYWmTZuya9eufG//wIEDdOnShV9++YXevaWC6EVkAKHIkUajYeXKlbRt25Y6depw/vz5IpUIANjZ2XHkyBE8PDz48ssvGT58OCqVSt9hCSGAKVOmkJCQwLJly95I+507d6ZHjx6MHj2a2NjYN3KN4kKSAZGt1NRUhg0bxsiRIxkxYgSHDx8uss/HjIyMWL58OevXr2fTpk20adOGhw8f6jssIUq0c+fOsWrVKmbOnEmlSm9u+fRly5aRkJDA1KlTX35wCSaPCUQWYWFh9OrViwsXLrB69WoGDx6s75DyzZkzZ+jZsydKpZLdu3fzzjvv6DskIUqc9PR03n33XTQaDf/88w+Ghm92LPuSJUsYO3YsZ8+eld/5HEgyIHScPXuWHj16oFAo2L17N++++66+Q8p3oaGh9OrVC19fX1avXs2gQYP0HZIQJcqTL+czZ84UyGdMeno677zzDgqFgnPnzr3x5KMokscEQmvjxo20bNmSKlWqcP78+WKZCABUrFiRY8eO0a9fPwYPHszXX39Nenr6y08UQry2kJAQpk2bxpdffllgnzGGhoasXr2aS5cusWLF6898WBxJz4AgLS2NcePGsXz5cr744guWL1+OiYmJvsN64zQaDatWreLrr7+mZcuW/PLLL0V2XIQQRUWvXr04deoUfn5+WFtnXajoTRoxYgQ//fQTfn5+b3ScQlEkyUAJFxERQe/evTlx4gTLli3Dw8Mj3+txC7vjx4/j7u6OhYUFe/bsoV69ei8/SQiRZ09K/bZt20bfvn0L/PpPShmbN2/Ozp07C/z6hZkkAyWYr68vPXr0ICkpiZ07d+brhB9Fzf379+nRowc3b95k48aNUpMsRD5LTEykdu3a1KpVi99//11vf3Rs376dTz75hAMHDtCpUye9xFAYyZiBEmrbtm00b94cW1tbLly4UKITAYDKlStz4sQJunXrRp8+fZg0aRIZGRn6DkuIYmPmzJk8fPiQlStX6rX3sU+fPrRv354RI0aQmJiotzgKG0kGSpiMjAy8vLz49NNP6dWrF3///TeOjo76DqtQMDc3Z8uWLcyfPx9vb2+6dOnC48eP9R2WEEXecU5C4wAAIABJREFU1atXWbRoEVOnTqV69ep6jUWhULBq1SoePnzIrFmz9BpLYSKPCUqQmJgYPvnkE44cOcL8+fMZM2ZMiRsfkFuHDx+mb9++lCtXjr179+Lq6qrvkIQoktRqNS1atCAqKorLly8XmsHJs2fPZsaMGfj6+lKnTh19h6N3kgyUENevX6d79+5ERUWxY8cOPvjgA32HVOgFBgbSvXt3goKC+Pnnn+natau+QxKiyFm3bh1ffPEFR48epXXr1voORys1NZX69etTtmxZ/vrrL5TKkt1RXrLvvoTYs2cPTZo0wdTUlPPnz0sikEvVq1fn9OnTfPDBB3Tr1o1Zs2ahVqv1HZYQRUZ4eDienp4MHDiwUCUCACYmJvzwww+cPHmSDRs26DscvZNkoBhTq9VMnz6dHj160KFDB06fPk21atX0HVaRUrp0aXbu3MmMGTP45ptv+Pjjj4mPj9d3WEIUCRMmTEChULBgwQJ9h5Kt1q1bM2jQIDw9PQkPD9d3OHoljwmKqfj4eAYOHMjevXuZNWsWkydPlvEBr2nv3r3079+fypUrs3fvXpydnfUdkhCF1tGjR3n//fdZt24dn3/+ub7DyVFERAQuLi507tyZH3/8Ud/h6I0kA8VQYGAg3bp1Izg4mC1bttClSxd9h1Rs3Lhxg+7duxMZGckvv/zC/7N3nuFRFV0AfrPZTdn0ZNMLCQQSehAEBERqUFCkKoKiKIhI+URRQClKFRQQERGxYaRjoYgC0qT3EiCBACE9pJOeTbL3+7FmYUNIY5Obct/nyZPsnbkz505mZs+dmXNO7969xRZJQqLGkZeXR6tWrXB0dKwV+/Hff/89o0ePZv/+/XTv3l1scUShZv+HJCrM33//zeOPP05BQQGnTp2SFAED06xZM06dOkWHDh14+umn+fzzz5H0aQkJLUVjYfHixdy6dYtvvvmmxisCAKNGjaJz586MGzeOvLw8scURhZr/X5IoF4IgsGjRIvr27Uvnzp05deoU/v7+YotVJ7G1tWXnzp188MEHvP/++7z88stkZ2eLLZaEhOiMGTOG1157jfnz5/Pee+/VGpM9mUzG6tWruXnzJosXLxZbHFGQlIFaTGFhIcOHD+fff/9l+PDhTJs2jQ8//JBt27ZhY2Mjtnh1GmNjYxYuXMimTZv4448/6NKlC8HBwfTr14/Y2FixxZOQEIWLFy/yzz//YG9vj52dXa1aNWvevDlTpkxh/vz5hIWFiS1OtSOdGajF/P777wwaNIjGjRsTExPD2rVrGTJkiNhi1TsuXrzI888/T1ZWFrm5uYwdO7bGnp6WkKhKVCoVycnJyGQy2rVrx7FjxzA2NhZbrHKTnZ1N8+bNady4Mbt3765Xh66llYFazMyZM5HJZCQnJzNgwAAp6IZItG7dmldffRVLS0uysrJYuXKl5MZYol6SkpKCsbExixYt4siRI7VKEQCtS/KVK1eyd+9eNm7cKLY41YqkDNRSjhw5wpUrV9BoNOTk5JCcnEx+fr7YYtVbUlJSyMjIQBAEcnNzmTVrltgiSUhUO59++innz59nypQpKBQKscWpFH379mXw4MFMnjyZW7dusW/fPrFFqhakbYJaSl5eHvPmzeP5558nICAAuVwutkj1HkEQuHnzJhs2bGD48OGSHwIJiVpKTEwM/v7+tG3blhMnTpCdnV0rrCIeBUkZkJCQkJCQuI8xY8Zw4sQJLl++DEBSUhIODg4iS1W11MnXycjISJKSksQWo0agUqnw8vISW4w6SX3uZ1K/qh6kPiZOH3vppZfYvXu37nNMTIykDNQ2IiMj8WvqT252jtii1AjMlOZcCwmVJm4DExkZSVP/pmTn1E//AkpzJSGhIVK/qkKkPiZeH+vRowdXr15l3LhxrFu3jvT09GqXobqpc8pAUlISudk5+H01BKWvk9jiiEr2jQSuTdhKUlKSNGkbmKSkJLJzsvlo9BoauPqJLU61EhF3jfnfjZH6VRUj9TFx+5ilpSVBQUGsXbu2zp8XgDqoDBSh9HXCspWb2GJI1HEauPrRpEGA2GJI1GGkPiYu9UERAMm0UEJCQkJCot5TZ1cGysPdUxGkHriO99Te3PhoJ26vtSdq5WEs/JzxGNeFqJWHybmVRH5yFk2WDSI/KYsb07bRaP5zWPg7l1imUFAIgJG8dGcbOREpRC0/SEFGHs3WvKSXpr6TQdSKQwgCOD7fEpv2DTj9xBJsu/pi2dIN15cfN0wDSFQ5wWEnOHl5D6MHzmL5+ikM6DaG9X8vw8e9GcP6TGLDX18QdecGdzOTmPra16RmJLE06B3+N+JzGro3K7HMgsICAOTGpQ/f2MRwfvnzczJz0pkzLkgv7cdtC4iMu46lhS2vPjeV/AI1M1YOp1nDx2nXrDtPtX3eMA0gUaWU1b/+ObmZ86GHyS/IY/LLy0hIiTFY/zofepgfts3D260pPR4fTBv/J3VpW//5mlvRVwAj3hg4E3V+bpX2r/p80LIkKnP4sl4rAzbtG5C67xq3F/+DeQM7lI21ZwxcR2q/bD3Hazt3zLdHyb2dglUbD2ye8CmxrKyr8ST8dpH85CwafvwMchvzUus2b2BPk6WDuDpmwwNp0d8cwdjSlMIsNaZu1gAYW5iiyc3H1E2KOVCbaNm4IyeCd/PDH/NwU3nTwE2799v/qdcBeOmZdwDYsucrYhJu0bRhOwL8upRY1s2oy/xzcjNpGUm8/eICrJS2pdbt5ujDB6+tZNaqVx5IMzaWI5crkBvLsTS3ITUjEXNTC/LU2TjbezzKI0tUI2X1r8PndvLJuJ85dvEvDp/bTuATLxmsfxkZGWFuaoE6PxdHO/0t2QvXjjBv/HpCbp3hz8Nr6d3xxSrrX5GRkfj7NSUnt34etCwJczMlodcqdviyXisDAE5DArjw7Go6nJ/6QJpGXcCt2X+RF5OG84uPlXh/TngyYVN+x76XP+5jO2PiaAlAxvloEn6/qJe30ZzyuQvOvpZAwzl9UagsCZ/7N02WDKTNnrdBgCuvBGHfo0kFn1JCTAI7DuPthT3Z+vm1B9LyC9Ss3DSN+OQonu7ycon3R9+5yWdrJ/JE66cZ2nsC9jZapTXk1hn+ObVFL+/EYYvKJdPLfacgk8k4emEXfx5ey6Ceb/HVtD3k5mUz59tRLJi4qYJPKSEWpfWvIt/6Lg5eHI+5WuL9le1frRp3IsCvCyl3E/h683RmjPlel/Zc19dY9stkrCzsuJuZgouDV5X1r6SkJHJysxnsuwKVeWODlVtbScoJ49cbEyt8+LJeKwOCRkPE5/tpuuYlIj7fT8NZT+uly0zk+C58jsTtwSTvDsH5hQcVAlN3G1xHdSTt3xvEfHsUx/4tsWzphiAICAWaSsll4mqN3NYcYwsTNHnaJTuj/w6xyEzlCBqN7rNEzUaj0fDj9gV8Mi6In7YvZNzQeXrpCrkJ74xYyoHTv3H0/J883XnEA2U4O3gysMebnL16gC17V9Kj/SAae7VGQKDwvyXdilJ0KMrWSkV4zFXdl4aZqbJS5UmIQ1n9q4g7KVEPvL0XUdn+VdSHrCxsUReo9dI6tAykQ8tAzoYc5Fb0lWrpXyrzxrhZtjRombtvz6WP98wHrp+O/5lGtl2xN/Mud1l7IxaQr8lBITOnd4MPddfVhdnsDJ+OsZEJPtZP0MpxkCFErzD1WhmIWX0MpyEB2HX1JeNsFGnHw/XSw+ftpjAnn4K7OTSc/UyJZchM5Dg+2wLHZ1ugTsok8beLmLrZYP2YJ9aPeT607vyUbG4v2kvW5ViiVhzCc+JTXJu4Bb8VQ3Ef25nwebsxAlxHtif7RiLRKw8DYNPJR1IEahGb935F4BPDaNesO1dvnuLCtSN66au3ziJXnU1GVhpvv7CgxDIUchO6tRtAt3YDSE1P5J8Tm3C086BZw8dp1vDh50fuZibz3e9zuRF5kXW7ljCi73vM/24MH41ewy9/fk5CSgx3M5OZ9NJiLl0/xu7jG8hTZ9Ozw1CDtoFE1VFW/+rSph9LgyaTl5/DOyOWlFhGZfvXv2e3c/rKPjKz7zKw+5sAuv615/gGrtw8RX6Bmom1pH8lZodxMHopDmYNicm8wCvN1pGWFwnAygs9CXAaSlxmMM83+pzM/ETyNbnlLjstL4ZCIZ++PnPZfXsud/NisDF1ByAkZRfN7fvhZx/I5utvScqAGHiMu7d35jW5OwB3Np3TXfOZ0adC5ZmoLHF/s3O58irslTRepH+Ixm+FdpBYNHHC74vBemlNlonTQSQejWF9Jun+Hvmcdivqr6O/6K6NHTKnQuXZWTsyNHBCufLaWDrw3itf6F37aPQaAF7uN0XvuoOtC62adKqQLBLiU1b/6tXhBXp1eKHc5VWkf3Vt25+ubfvrXSvqX4FPvETgE/cORrdq0qnG96+zCevp3eAjrExcCLo6XC/N2tSVzm5vcSLuB+Kzrzxwb3Z+Koeil+lda+8yCgdz7RmzDHUcNqbalRkbU3fS1XE6ZeCuOg4nW38AZIgX5VF6xSyGspGK+HVnSkzLDkskNzoNY2XtjMYlUTPwcmnCn//+VGJaRNx14pMjMTeRluslKofUvx4NI4x02xpFmMi0B8KNjeQUaNQl3YZGKNT7gXthf6xMXEnPiwMgXR2LtYmrLs3GxJV0tTZNoHJby4ZAUgaK4TnxKdzf7MytOX8/kKZs7Ij1Y54ImorFdgqfv5ubM3YSPn+33vXCbDXX/reVsCm/k/DbBd31+A1nuTjoO0B7EDFk7EZuffJXJZ5GoiYyou+7JN2NLzHtwrXDjHz2A1wdvctd3uqts/ly/fus3jr7gbScvCzenPsUxy7e6z/rdi3VszC4FX2FAZMbkp2bWf6HkKixjOj7LkMDJ7Bqy4wH0hq4NqFZw8fRCBX70qkPfayt03D2RizgYPQyFLKKKUtKhR39Gs7X+3Ewb6hLtzV1R2ak4O/bH2OEMTam7hyLXU1c1mWa2vflavIudtyahp9db0M/Vrmp19sERWSHJRCx5ADmDR3IvBBDi/WvkhuVCsDZHitwfqENmZdiabxkIOqEDDS55T+0lRudhlCgodG8Z7k152/yYtIwddea7CTtuoKqXwscAv0JGbsRp0EB5ESkkJ+SjcLBAgCrNh74zOhD7A8nDP/gEtVCROw1ftqxEE9nX0LCz/LZ5N+JS4oAYNTsjjzdaQTXI87zwWsrSbl7h7z88u9F3kmOorAwn0nDP2PVlhkkpETjdJ/p1oa/ltG93UDd5ys3T+Jgc89HRkFBPn8eXkuHFuJNQhKPjtTHHh0LhQoHMx+y81MIcNRu2b7ot0bv9+MuIwHwsan4lkfvBtP1PndyG6v7e6DvsuLZqx1JGQDi15/FZ0Ygpi7WBA9fq5dm6mqNx1tdiPn+OFlX4h64Nz81m8hlB/SuuY3qiLmPNsKVOj5d5xvA1N2GvLh0nTKgjkvHwt8FACNjGYJGQ8w3R2n4yTOEjtc36ZGovfx5eC1jh8xBZevGB8sG6qU52rnxYp+J/LbvG25EXXrg3vTMFNbu1DcXHNj9TTycGwGQlBaLo71279HZ3oPE1BjdRH3myn4auPmjzs8DIE+dw76TW5k0/DOOXdKufG3a8yWDer7FzzsXG/ahJaoVqY89OkqFHT283hdbDNGQlIH7MTLCqNglmdJE+1thjEZd8orAAyaEwr1tBBMXa/Ji7wKgjr2L6dNN76W5WpMXdxfLFq4IGoHciFTyU7IIn7ebrKvxpOy7hn3P+hWgpC5jZGQExfYizUy1K0DGxgrU+SXvRT5o3nWvf6ls3UhMjQUgISWGLm2e1aVduHaEHHUWEbGhmJqYozSzIjPnLis2TuVmVDBXbp7iRmQwKekJhIafZcehH3mxz0QDPKmEWEh9rPo4n7AJpcLe4Ev7O25NxURmQR/vWZy9s4GYzPOkq+Po6zO3QqaMFUVSBgCX4W25PX8P5g1Vui//8qKwU+K74LmHppt52GKkMObWx7swMpFj6m5L9Oqj2HbyQdW3OTc/3EHKvus4BPph7uNA09XDAMiLS8e+px/ZN5OIXLqf7GsJxP1yWnJFXAvp9+SrfPvrbDydfTGvoJ21taX9Q03CQGsjLjdWsHLTdBRyU5zsPdi85yva+D/J6EGzAPjr6DpsLO0J8Oui8z6XmBpL80btad6oPQALf3iL554aVcknlBAbqY+Vj0uJv3M7/Tgmxhb08prGleSdxGVdRl2YRT+f+fwb8yU5BXfJLUjD2aIZ2fmppKtjGdx4BetCRuJt04mU3Ns87jxSV2ZaXjTHYlcjIGBv1gBfm24cjF6Kjak7rR2H4Kz0L7d8V5J34mYRQFJOGABtnV+irfNLhCT/RVxWsKQMVDUKlSVmPg7kp2Tj/EIbAF28gKLfriO1Hdq2U8OSCykFnw8D9T57jL1nftikmAlhEUX1Khup8F9ZftMgiZqHnbUjHk6NuJuZTJ8ntCZLRbECin4/3+0NAD3/7uXlzcEf631+oZhp2DMlODIqHqtg+uvfVLheiZqD1MfKR0rebZyVTfG3D0QuMwXA2EhBujqOuKzLALRQPYeTuR9/3HyXYX7fsf3mB+QUpKFBQ0fX0agLM9kbsQBPq7YAnIpfi1xmhkJmxp2sUFyVLTCT29BSNUBPESjL/DBTnUhc1mXaOo3QKQMAeyMWEpt1gQGNqvZcgaQMoH279/6gl9hiSNRRrC3teX3Agye7JSQMhdTHykc3j8nEZ11hT8Q8eni+z5XkHQz3/4kDUUvJ1+QAYGpshbHMBFNjrWt5YyMFBRo1gqBBEAopFPS3VARBQ2vHwbhY3Av8ZGfmxbmEjdzJukqA072XOa3Jod7dur9upx8nKz+ZQ9HLiM++SlLOTVTmjejdYDrRGec5l7CR7p7vGrhF7iEpAwbgzqZzyO2VOPQu/3JQeQibug1jpQkNZz9D/PozpB0LR5OTj9fk7li2cC27AIk6R9FybKfWJXvErAy7jgQRFnkRC3NrRg+cZbByJWonhu5jGo2GZb9ovSCamigfcIRVnZy58wvJOeEYIcNcboelwokjMV8Tk3keb+uOpd5rLFNwKHo5KbnhPOk+kbisYAA6uI7in8hPsVI4YWJsibd1B66l7iWnII1GNl119xeZHz6MFqr+tFD1JzU3ilPxP6Iyb8SRmK+5q44lt+AuXT3+Z5hGeAj1VhlI+O0id4+HY2xpive03iTtvEzm5TgKM/NotOA5bXjhu7kU3M3BoqkL+anZqOPu4rdiKFdGBmHTyYec2ym4/bd9AJAbnUrMN0cRBDD3tseumy8RSw5g5m6D09A2Dw17XBKJOy9j1dqd7LBEAO6euE2TZYPIvBjD3ePhkjJQi/jn5GYuXDuC0tSS0YNmc/DM79yIvER2XibvDF9C0J+fkZmdRkZ2Gg09WpCemUJiagwfjV7DtC+H0savKzEJt3TLvADxyZFs3r0CAQE3Rx/aN+/FTzsW4mzvSWCnlx4anrY4qemJHDj9K76erXGwcamqJpCoYmpyH5PJZLw3cjkA89aMRqPR6OIaVDftnPWDgfVvpLVw6OL+NqBvMjjQV6u0FH2By41M6On1gS79/pWAIY2/0ivXx6Z8nmhLws7Mkz7es/Tkqg7qrTKQG5GCRVMXHPr4IzPVNoORXEZefDpZl7UmhI79W6L0c+L65N9o9v1wwt7/g/y0HIRCDe6jn6AgU83tBbuxbquNDBX30ylkZgpkZnKyQuKxaOGK3NYcx4Gt9RSBMs0REzPJuhyHy4h2OmXA8fmWBA/5HqFQQ9M1+q4yJWo2sQnhNPRoTpeAfpgo/tunNFaQlBpLWKQ2smX3xwfj7ebPoh/fZu74dXz+8yQyslLRaAoZ3GscObkZfPvrxzT77zDWtgPfYWpijonCjPCYqzT2ao2V0o6eHYbqTdJlmY3FJoZjZWHH2CGf8M2WmcQk3MLdqeLnYiTEpSb3MYDbsaFs+GsZlkob0RSBR6XI10Bdpd4qA16Tu5N5JY5bc3fj/UFPErdfpvnal4lYsp/CHK35jbGVKTITY4yttIPLSGGMoC5A0AgIhQJCgf7+j6ARcB4SgEWze29Y5l52xG84S9bVOL2oh6WZI949Hk5+UiaRSw+QdTWe7JtJxP18mtbb3iQvLp3oVYfLHQ5ZQnxGPjeVG1HBrNoygzcGzODgmd9ZMHETP21fSK5au0+pNLdCoTBFaW4FgNxYgbpAu0+p0RRSUMz0SyNoCOw4jEaeLXTX3By9+fPwz9yMCtaLflia2ZijnRvWFnYAWCptyMnLMuSjS1QTNbmPAXi7+TP9jdUs+2Uy8cmRuDiUP7SuRPVQb5WBuF9Ok3MrGSOZEXI7JSbOlkStPEzGhWhsnvAu9V6ZwpjI5QfJuZWM16SnyAzW2uC6vd6B2wv3YuJkhbGlKTYdvUnZG0p+Wg62XX1195dljujYvyWO/VuSG5VK7A8nUDZSYfOEN2FTfqcgIw+X4e0M0gYS1cOOQz8SnXATmZEMawt7HGxc2PDXF4SGn6V1ky6l3is3NiHoz8VE37nJy32ncP2/t7yBPd7ku98+wd7GBaWZJa2bdObYxb9Iz0qlXbMeuvvLMhtzsvfAysKOlZumU1CYj6+nYUPASlQPNbmPJaXFsW7XEgRBg7FMjpOdx0PzVhebro0x+Jv+d8H96eI+AX/7wIeGK76f8uSB0v0OKGRKdt6axmPOLz2yv4N6qwwUt9dvvHgAAJ7jtWY395sQFkUQLPoCNzKV61kf3L8SUNwM0LZz5ZdczTztdKGTPd4qfUBL1FyK21YX7Z++9Mw7gL6pV5H5VdHkqlCY8saAe/HU739LmzHme71y2/h3pTLcX75E7aQm9zGVrSv/G/55he+rLDtuTaWn51TM5Lb8GjaePt6zORX/EzkFqfjadqep/dO6vEVKwen4n1GZ+2Imt+Z8wmY0QgGeVm1p7Viy6XdJWJk4428fWGq44iLKkwfK9jvQ3OE5/O0rFl33YdRbZeBRKPIBICFR1RS31ZaQMDR1rY81d3iOy8k7cDDzwcemMzIjOYUaNRYKRy4l/qanDBTnWOxqnWOfuKzLesrAwahl5BSk6j772nansV33B8ooLVxxRfJUt9+B2nmSQ0JCQkJCogS8rTsRkX6Cqym7aO7wLBcTt+JnH0hX94moC/XPxBgZab8C1ZpsAAqFfDq6vkF3z/d42ls/QqMG/RDFDws3XFq44orkud/vQHj6MZJybgLagEc9PadxLmFjRZqlTOr9ysDVMRsM/qZ/4bnVeE7oikOfpoTP340mJx+ZuQKfjx6+nHO/T4GolYfJjUwh+3oCToNa4/pK+wfy39lynvj1Z/B4+0kcevujTsjgxtTtOA9va3B/BxKGYdaqVwz+FjZ+YS+GP/MunQP6snrrbPLU2ZiaKBk75JMH8kbGh7Hhr2UIgoCXaxOGPzO5xDI1Gg0frniB9i17M6jHWLYd/J6zIQfr3BtkXUTsPgaw5Of/cfnmSX78pORIqyX1w32ntvLn4Z9Z+t72R5ZXZiTDxtSDDHU85nJbvKzacfpOEJEZpzGWKfTyWpm4cDT2GyLTT+Fm0YoubuPZFT4TC4UKW1NPOrq+rsvbw3NKueq/P1yxsZEpNqbu3Ez7FzO5Ne6WAeXOU91+B+q0MhA2dRveU3sjtzUj9O0tNPz4GeJ+Okl+ajZ23RqjeuaeeUyRUhD38ynMfVXIrcy4s/k8QoEG63aeOA0OKHe9Ji7WOPRpWmr44vsp7lOg6NxC6LhNOPYv+UCX89A2oLkvIJKTFQ73BUGSqF6WBL3D6IEzsVLaMXfN64x/cSF/7F/D3awU2rfoxZP3BXcpmrC3HfweL5cmWCqt+fvoOgo1BTRr2J7AJ4aVu14HW1c6B/QtM8wsgJdLY6aO+hqAmSsfdB9bxO/7V9OxVR9dzPvnu73B2ZCDFWgNiaqgNvQx0J5XmLXqlYeWV1I/7Nl+CIfObiu3TGUR2OAj3d+eVu3wtNI/dF10ePAZb61C09ntLV3a4MYrHrn+4uGKYzIv0N7ltQrngerzO1CnlQHH51qQuCMYcx8HbLs0xEguQ5NXgInKksTfL+opA8WJ+fYoZg3sAci8HKenDEQs3U9BWo7us133xth3b/JAGaWFL9blKcGnAEBefDoyMwVyG/PKPbxEtdK93UAOnv4dd+dGPOb/FMYyOeqCPOytHfnnxGa9ibo4m/d8hZuj1j/5jchLehP12h2fkp51b5+yfYteJcaFLy3MbHH2ndpKu+Y9SkwLjwlBoymkoUcLwmOvlv3gEtVGbepj5aG0flgbMZPbEpqyB3/7wAfSunpMKvP+8uQpToY6gejMcwQ4Dq3wvcWp08qATScf4n45Q1bIHbyn9ebOxrM49PHHqo0nV0et08trJNOG/SzM1voY0KgLcRvdCYVtCV/GhYK+nwCN8GAeSg9fXERJPgWUjVTc2XgW5xfbVOaxJUQgwO9Jdvz7IzejrzBm0Cx2HfmFzgF9aerTjo++0n8Lk/23T5mbp92nLCjIZ0ivcVj9Z+9/PxpNoZ4Nt6ApeZ+ytDCz97Pv1FbuJEc9dIvgbMgBYhLDCb55kvTMFHo8PgRbK4cynl6iOqgtfaw8lNUPK8P9h+zEoCiSYWxmcLXW29bpZb16K9sOdVoZMJLJMPOwRX0nHYWtOdbtvIgLOkX66UiMFMZ6eU1crIledYT0UxFYtnLDc3xXbs7YiYmjJaYetri/8YQub4P3e5ar/pLCF6ceuoHcxgyrAK1GXZJPAUEQSD8Vidc7906qRn55EK9J3XSfk/eGcmfLeWRmCowtTCoVTVHCcMhkMlwcvEhKi8PKwo4Wvu3ZfvAHLt84gUKuHxZbZevKxt1fEhx2nCYNAnjpmXdYvv597K2dcFF5MajnvSXLUc9/VLyqEikpzOyZK/uxVNrg76ONrhYWeZFVW2bwRKunWblpOuNfXMie4xsdv0QcAAAgAElEQVRp3aQzzg6eAAzppV2GPB96mPDYq5IiUIOoDX0M4Lvf5nAj8iJLgt5h4rBFXLp+tMx++CioVCrMzZT8emPiI5VTlzA3U6JSqSp0T51WBgB8Ztw7tGfdzgvrdvqer4oODzb6pK/2wrh79vz+Xz360kvx8MUZF6Nxe+3BgBj3+xQwMjKixfpXdWnZYYmYOFrp5Xfo7S8dFKxhjB0yR/d380YdaN6og1560cGuCcM+BWBYn3vLgjPGfPfI9RcPMxt6+xwDuo/RfW7s1Zqtn4Xq5bkdG0rPDg/28zb+T1Yq1K1E1VLT+xjA6EGzGD1o1kPzlNQPHwUvLy9Cr4WQlJRksDJrOyqVCi+vinl5rPPKgBgobMxI3h2CQ58HtwXuf7svL8rGjigbO5aaR52QQfq5KO3BQol6g5XSlqMXdtE5oO8DaS/3K/v0c/HJvSS2HfweHzdJ8ayvPGofK0+efae24mhb+eBrXl5eFf7yk9CnzioD2TcSRKvbdaRWW8+8FFu99Y54XK9eMdugvhARd03U+ouizF2PuFBldTT9b3m3qA6xn7m+IXZ7V0cf83T2xdPZV+pjIlLnlAGVSoWZ0pxrE7aKLUqNwExpXuG9I4myUalUKM2VzP9uTNmZ6yBK84rvSUpUDKmPSX2sOjESBKHko/C1mMjISIPuH2VkZDBo0CDatGnD4sWLDVYuwJ07dxg8eDDPPvss06ZNM2jZULm9I4nyYch+tmvXLmbOnMmKFSvo1KlT2TdUgHXr1rF06VLWrl1LixYtyr6hHEj9qnowVB8TBIFx48YRGxvL5s2bMTMzM4B0WgoKCnjllVeQyWSsXbsWudww75hSH6tmBIkyeeuttwQrKyshJiamSsr/4osvBCMjI+HkyZNVUr5EzSYlJUVwcnISXnjhhSopPz8/X2jTpo3Qpk0bIT8/v0rqkKjZBAUFCYDw999/V0n5J0+eFIyMjIQvvviiSsqXqHrq5MqAITlx4gSdOnVi+fLlTJxYNaYrBQUFdOjQAY1Gw+nTpw2mWUvUDt566y02bNhASEgIbm5uVVLHqVOn6NixI0uXLuWdd96pkjokaiapqan4+fnRo0cPNm40rD/7+xk/fjw///wzoaGhuLs/GIFPomYjKQOlUFBQQLt27ZDL5Zw8eRJjY+Oyb6okp0+fpkOHDixZsoTJkw3niEOiZnP8+HE6derEihUrmDBhQpXWNWHCBNauXUtISAgeHuLHlJeoHsaOHcvGjRsJDQ3F1bXyJ/bLIi0tjaZNm9K5c2e2bpXObNU6xF2YqNksWbJEkMlkwpkzZ6qlvgkTJggWFhZCZGRktdQnIS5qtVpo2bKl0K5dO6GgoKDK60tLSxNcXFyEQYMGVXldEjWDo0ePCoDw1VdfVUt9GzZsEABh586d1VKfhOGQVgYeQlRUFE2bNuX111/nyy+/rJY67969S9OmTenYsSO//fZbtdQpIR6ff/45U6dO5fTp0zz22GPVUuemTZsYNmwYO3bs4NlnK+9OVqLmk5+fT7t27TAxMeHEiRNVurJZhCAI9OnTh7CwMK5cuYJSqazyOiUMhMjKSI1lwIABgqurq3D37t1qrXfTpk0CIGzfvr1a65WoXiIiIgSlUilMmjSpWuvVaDRCnz59hAYNGgiZmZnVWrdE9fLZZ58JMplMOHv2bLXWGxYWJpiamgrTpk2r1nolHg1JGSiBbdu2CYCwefPmaq+7aLL28vKSJus6TP/+/QU3N7dqVzYFQRBu3LghmJmZCVOnTq32uiWqh9u3bwtKpVL43//+J0r9c+bMEeRyuRAcHCxK/RIVR9omKEZmZibNmjWjRYsW/PnnnxgZGVW7DLdu3aJ58+ZMmjSJRYsWVXv9ElXLtm3bGDBgAFu3bmXw4MGiyDB//nw+/vhjzp8/bzDfAxI1h+eff56zZ88SEhKClZVV2TcYmLy8PFq3bo2joyOHDh1CJpNVuwwSFURsbaSmMWXKFMHMzEy4deuWqHLMnz9fkMvlwqVLl0SVQ8KwZGRkCJ6enkK/fv0EjUYjmhy5ubmCv7+/0LlzZ6GwsFA0OSQMz++//y4Awq+//iqqHAcOHBAA4bvvvhNVDonyIa0M3MfFixdp27Ytc+fOZfr06aLKolarCQgIwNbWliNHjkiadR3hvffeY9WqVVy9ehVvb29RZTl06BDdunVjzZo1jB49WlRZJAxDRkYGzZo1o3Xr1uzYsUOUlc37efXVV9m5cyehoaE4OpYebE1CZMTWRmoKhYWFQseOHYVmzZoJeXl5YosjCIIgHDp0SACEb7/9VmxRJAzA+fPnBWNjY+HTTz8VWxQdr776qmBnZyfcuXNHbFEkDMC7774rmJubC+Hh4WKLIgiCICQkJAh2dnbCq6++KrYoEmUgKQP/8c033wiA8O+//4otih6vvfaaNFnXAQoKCoQOHToIzZs3F9Rqtdji6EhISBDs7e2FkSNHii2KxCNSE5VNQRCENWvWCIBw4MABsUWRKAVpmwBtsCB/f38GDRrE999/L7Y4eiQlJeHn50e/fv34+eefxRZHopJ88803jBs3jsOHD9OlSxexxdHj+++/Z/To0ezfv5/u3buLLY5EJSgsLKRTp05kZWVx/vx5FAqF2CLp0Gg0dO3alaSkJC5evIipqanYIkmUhNjaSE1gxIgRgoODg5CYmCi2KCXy/fffC4Cwb98+sUWRqARxcXGCjY2N8MYbb4gtSokUFhYKXbp0Efz8/ITc3FyxxZGoBF9//bUACEeOHBFblBIJDg4W5HK5MHfuXLFFkXgI9V4Z2Lt3rwAIP/74o9iiPJTCwkLhySefFJo0aSJN1rWQ4cOHCyqVSkhKShJblIdy+fJlQS6XC3PmzBFbFIkKUqRsjh49WmxRSmXq1KmCqampEBYWJrYoEiVQr7cJcnNzadWqFa6urhw8eFD0k7elceXKFQICApg5cyazZs0SWxyJcrJ3714CAwP56aefePXVV8UWp1SmT5/OsmXLuHz5Mr6+vmKLI1FOhg8fzt69ewkNDcXBwUFscR5KVlYWzZs3x8/Pj7///rtGz7f1ErG1ETGZPXu2oFAohKtXr4otSrmYPn26YGpqKly/fl1sUSTKQU5OjuDr6yt069ZNVJ8C5SUrK0vw9vYWevfuXSvklRCEPXv2CICwdu1asUUpFzt37hQAYePGjWKLIlGMersycO3aNVq1asX777/PvHnzxBanXGRnZ9OiRQsaNWrEnj17JM26hjN79mwWLlzIpUuX8Pf3F1uccrFr1y769evHhg0bkMvl9OvXD3Nzc7HFkiiB3NxcWrZsiYeHB/v3768188GQIUM4evQoISEh2Nraii2ORBFiayNioNFohB49eggNGzYUsrOzxRanQuzatUsAhHXr1oktikQphIaGCiYmJsKMGTPEFqXCDBkyRHBychIAYf369WKLI/EQZs6cKSgUCiEkJERsUSpEVFSUYGlpKbz99ttiiyJxH/XSrd26devYv38/K1eurHVvPc888wxDhw5l8uTJpKamii2ORAkIgsC4cePw9PTkww8/FFucCrFr1y7OnDlDeno6crmc+Ph4sUWSKIHQ0FA+/fRTpk2bVmtWnYrw8PBg3rx5rFq1ilOnToktjsR/1LttgtTUVPz8/OjevTubNm0SW5xKERsbi7+/PyNGjGDVqlViiyNRjKCgIEaOHMnu3bsJDAwUW5wKkZqaypgxY/j1118BePnllwkKChJZKon7EQSBHj16EBUVRXBwcK17oQEoKCigffv2CILA6dOnkcvlYotU76l3KwPTpk0jLy+PZcuWiS1KpXFzc2P+/PmsXr2aEydOiC2OxH2kpKTw3nvvMWzYsFqnCADY2dmxdetWtm3bhrm5OWlpaWKLJFGMX375hYMHD/L111/XSkUAQC6Xs3r1ai5evMhXX30ltjgS1KOVgZEjR9KlSxfGjh3LihUrmDBhgtgiPRKFhYV06NCBgoIC+vbtS+vWrXnxxRfFFqvecvXqVT788EMcHR3ZvHkzoaGhuLq6ii3WIyEIQq05lFYfWLx4MQqFgoULF9KzZ082bNggtkiPzIQJE1i7di2LFi0iJiaG+fPniy1SvaVerM3k5eURFBTEgQMHaNmyJd26dRNbpEdGJpMxZswYxo0bR3p6OlFRUZIyICKHDh1i586dFBYWMnnyZCwtLcUW6ZGRFIGaxZYtW8jIyCAvL48RI0aILY5BGDJkCFu2bGH58uXk5+dLyoCI1IttgoSEBACio6O5du0ac+bMEVmiRyc5OZlJkyZhY2PD7du3iYiIEFukek1sbCyCIGBtbc2yZcs4duyY2CJJ1DEiIyO5du0agiDw1ltvodFoxBbpkZk0aRJZWVlcv36d2NhYscWp19SLlYHr168D2rfp9957jxkzZogs0aOjUqk4d+4cY8aM4fjx4wQHB4stUr3m0KFDaDQarKys2LBhA3369BFVnsjISJKSkkSVQSxUKhVeXl5ii2FQBEEgMTERgD59+vDFF18gk9X+d7l9+/bxwQcf8NNPP5GXl0dmZmadWFWrjdQLZaBZs2a0b9+eFStW0L59e7HFMRjNmzfnyJEjzJo1i8jISLHFqdcMHDgQZ2dn1q5di1KpFFWWyMhI/Js2JSc7W1Q5xMJcqSQ0JKROKQRGRkb06tWLoUOHMmbMGLHFMRiOjo78+OOP9O3bl+XLl4s+duoz9eYAoYREfeHcuXO0bduW0Ws+wtWvgdjiVCtx1yL4bsx8zp49y2OPPSa2OBIStYZ6sTIgIVEfcfVrQIOAJmKLISEhUQuotDJQn/ckS6Ii+5T1ue0qs58rtVfdWe6uqUh9TBqT5aWujslKKQORkZH4NfUnNzvH0PLUWsyU5lwLCS2zk0RGRuLv15Sc3Hq6n2umJPRa+fdzpfaqWHuVl7ATwVzec5KBs0azfspyuo0ZwN/L1uPezIc+k4ZxcvM/hB4+T0FePi8vm0xKTAJB7yxlxOf/w71ZwxLLLCwoAMC4DG9yGo2GP+b9QG56Fg3a+NF5xNO6tFO/7id49wmMFXICJ72Im18DprcaTrMebfFq3YSnRj1nuEb4j/o+n5V37ipCGpNVMybFplLKQFJSErnZOfiuGIx5Y5WhZap15IQlcWPiryQlJZXZQZKSksjJzWaw7wpU5o2rScKaQVJOGL/emFiudtLdI7VXhdqrvDTu2JLg3Sf4Y94PqLzdcPvvbMFTr/cH4NzOw4z7+RMu/nWMc9sP88RLgfh1CSixrKjLNzm5+R8yktJ4ccHbKG2tSq37wp9HSY1JxNLeGnt3R720c9v/ZfR3M8hOy+C3T9bw2lcfYGppjjpHjb27kwGe/EHq83xWkbmrCGlMVs2YFJtHOjNg3liFZUs3Q8lSr1CZN8bNsqXYYtQapPYyPB2HBbKw59t8fm3rA2lFDoccvFyIuXq8xPvv3Ixm7cTPaP30E/SeMBQbJ3sAbp0J4dSWf/TyDls0Ufd3fFgkvh2a89Tr/Vn1yiyadmurSwuc+CIb3l+OtZM92WmZAMw6sgYEgeVDptEysMOjPXQpSPNZxZDGZN2ixhmq3p67u8Tr8T+fJvd2SoXKiliwl/CZu4hYsFfvemG2mhv/+42b728j8bdLlZa1JrD79twSr5+O/5mU3NsVKmtvxAJ2hc9kb8QCvevqwmx+u/E/tt18n0uJv1VW1BqD1GbapfrtC35kXNAnbF/400PzpUTdwc7NscQ0B09nerw5kDs3Y9i7cguRF8O0CYJAYUGh3s/92Lk56lYPjIyN9dIatmvKK1+8R8cXe+tWDWQyGTJjYxRmJrXC0Y40h1UMaTzWDERVBrLDErk+bguRn+3n6ghtZLS8SG1glAs9VxL7zVGuj99KYU4++YmZaHLzy112XkwaQn4hPnP7IhRqyIu5q0tL2RWCfb/mNPrseVL3hBr2oaqQxOwwtlwfx/7Izwi6qnVHmpan9S+w8kJPjsZ+w9br48kvzCEzP5F8TW65y07Li6FQyKevz1w0QiF382J0aSEpu2hu34/nG31GaOoewz5UFSO1Wcns/WozTwwLpFn3dphZWXDtyAW99Db9uhA0eSln/jjIY/2fLLEMuYmCdgO6MXL5ewROeIHQwxfISE6j4ePNGLHkHb2f+3msf1eu7DvN+inLadKpFQDfjdG6ob20+wRBk5ey49O1PPPucOLDIvnx7UX8+PYi/LoE1DhHO9IcVjGk8VhzEdW0MGH9WRp81BsTFyuuDtcPk2rqao3bW52J++EE2VcejKmen5pN9LJDetdcRrXH3McBAHVcBqZuNtqy3G1Qx6Vj6m7zX9pdbP3/2380rlmTS2mcTVhP7wYfYWXiQtDV4Xpp1qaudHZ7ixNxPxCffeWBe7PzUzkUrR+psb3LKBzMfQDIUMdhY6pdIrUxdSddHYeNqTsAd9VxONlqY6bL0H+Tq+lIbVYyfSYN0/393NSRABz95S/dtQ4v9KLDC73KXZ61ox2BE4aWK6+p0ozXVn6gd230mo8AaNWnI636dNRdt3F2YNTXU8stR3UjzWEVQxqPNZea0YuMjB4IiiIzN9EmyY3RqAtKvE0o1Oj9cJ/7JBNXK/Li0gFQx6Zj4mp9X5p2YAGgqX0+l4x4sL1MZNpQpsZGcgo06hLv0wiFej/3N5iViSvpeXEApKtjsTa5F3HPxsSVdLU2TaDmL9OWhNRmZePSxIt/f/qzxLS46xEkR8ZjoqydIXOrHGkOqxDSeKx5iLoy4DS8LREL9mLe0AGZUlGhexV2ShrO7/fQdFN3W4wUMm5//DdGpsaYutsQu/oYNp19sO/blPCP/iR133Xsevs96mNUG22dhrM3YgEO5g1RyCrmtlOpsKNfw4dHBLM1dUdmpODv2x9jbGSKjak7x2JX42PTmab2ffkz/COup+7Dz673oz5GtSK1Wfnp+6522XbLjFUMnTdOL821SQMaPt4MoYJ79ltnr0adnYeJ0pQhn4zVS8vLymFx3//Rf9prtH6mEwC7lq4j4vw1xgVpg4lFX7nFkufeZeGl9ZhZ1jxXtdIcVjGk8VhzEVUZUKgsMPNxID8lG8ehWrMlvzUv6v12Gfk4ADadfCpcfoPp+v90t7GddH/7LhtYKZnFxEKhwsHMh+z8FAIctUuyL/qt0fv9uIt2ydfHplPJhZRC7wbT9T53crs3eQ/0XVY8e61AarPSib0WwY6FP+Hs60n42RAm//4ZSRHaN6jZHUfRacTTRJy/zmsrP+DunRTyc/PKXXZy1B0K8wsZ/tkktsxYRUp0AvYe98wD/1q2gXYDu+s+3zx5BRtnB93ngvwCDq/9kxa9q86C4FGR5rCKIY3Hmou4yoCdEq/3e4gpQq1CqbCjh9f7YotRq5DarHQOr/2TIXPGYuumYtlA/X18OzdH+kx8kX3f/EbUpRsP3JuZks7ORWv1rnV/cyDOjTwASItN0lkE2Hs4kxqTqFMGruw/g5t/A/LztMvB6pw8Tm7dx/DPJnHpb2345z1fbqLnW4PYufhnwz60AZHmsIohjceaS62KTZCw6TwKe6XBl8VuTd2BzMIE71l9iPn6CHkRqWSHJeI4sCXOrzxu0LrE4nzCJpQKe4MukW2/+QFRGWcZH7DPYGXWFAzdXhpBw85b0yjQ5KIwNue5hosMUq6hMDIyotgWLqYWZgAYK4zJV5e8h1vcbPD+PW9bNxWpsdqwuykxCbR5tosu7dqRC6izcogNjcDE3BQzKyU5dzPZOHUFUcE3uXnqCpHBN0hPSCH8bCiHftxBn4kvPvqD1iCqYj67+cF2Ms5GEbBvvMHKrCkYekwm5dzgSMzXCAiozH150r3utVlFqHJlIPH3S6Qfv42xhQle03qRvPMKWZfjKMxS4zO/HzFf/kvB3RwK0nKxaOZMfmo26th0Gq8YTMjIddh08ib3dgrOI+99KedFpxG7+hgIAmYN7LHp5kv00oOYutvgOKQ1Sn/ncsuXvPMKFgFu5IRp/Wy7v62dsK6/vQWH/i0M2hYV5VLi79xOP46JsQW9vKZxJXkncVmXURdm0c9nPv/GfElOwV1yC9JwtmhGdn4q6epYBjdewbqQkXjbdCIl9zaPO4/UlZmWF82x2NUICNibNcDXphsHo5diY+pOa8chOCv9yy1f/0aL2XSt5oRTrcntJTOS0b/RYgB+DZuIRtAgMxL//O6Tr/bj19nf4uzriWkFDwda2ls/YDZ4Pw6ezhgr5GyavhK5qQJ7Dyf2fLUZ/yfbMGjWaACOrvsLS3sb/LoE6DwcpsYm0qh9cxq1bw7AD28trBI3xJWhps9njRb359qYTVXw5JWjJo9JlbkvA3yXArDx2ugqef7aRJUrA3m3U1A2dcY+0B+ZqbY6I4Ux6rh0si5r9yZVz7XA3M+Jm+/+gd93w7j5wXYK0nJAo8F1dEcKM9VELNiLVVtPAOLXnkJmJkdmpiAr9A7KFq7IbcxQDWipN3DKNN1JzCTrchxOI9rqlAEAdXw6MjMFchtxT06n5N3GWdkUf/tA5DJTAIyNFKSr44jLugxAC9VzOJn78cfNdxnm9x3bb35ATkEaGjR0dB2NujCTvREL8LTSenk7Fb8WucwMhcyMO1mhuCpbYCa3oaVqgN4gKsuMpyZS09srIfs6R2JXYmZsXSMUAdCaBDo18iAz+S5PDO8DoDu8V/S72xvPA+D/ZJsKlz/44zf1PgdOeEHvc+cRzzxwT1G9Rbz+zfQH8ohFTZ7PaiI1fUwCBCdto5FN16pshlpBlSsDHpO7kXUlnoh5e/B8vwfJO67g/9NwopYeQJOjdcBhbGWKzMQYY0ttZzFSaE1xBI2AUCggFFuKFDQCjoNbY9HMRXfNzMuOhI3nyLp6B6cX7vlQFwqLnX6+bxkz/fht8pOziF52iOyr8eTcTMK8kYqEjef1yhCLbh6Tic+6wp6IefTwfJ8ryTsY7v8TB6KWkq/RBlUxNbbCWGaCqbEloB1oBRo1gqBBEAopFPRNmgRBQ2vHwbhYNNNdszPz4lzCRu5kXSXA6d5krTXd0bu7ah7UQNT09nJSNmGQ73J23ppOWl40tqYeBnz6ymFpb82AGa+LLUatoSbPZzWRmj4mg5O2kZYXXe+3CKAalIE7v5whJzwZZEbI7cxROFkS8/URMs/HYN3Ru9R7ZQpjopcfIjc8BfeJT5IVrNW8XUd1IPLTf1A4WWFsaYJ1B29S916jIC0Hm66NdPeXZbqj6t8CVf8W5EalEv/jKcwbqRAEgfTTkXi885RBnv9ROHPnF5JzwjFChrncDkuFE0diviYm8zze1h1LvddYpuBQ9HJScsN50n0icVnBAHRwHcU/kZ9ipXDCxNgSb+sOXEvdS05Bmp52XJYZD8A/kZ8Sl3WFHbem8oz3HJ3mLxY1ub3S1fEcjvkKQdAgM5JjbVK3fOAXLfcXmQg+KhqNhl8mLyM/R2uW+MoX7xmk3EelJs9nAJGf/kPWlThuTd2B95xndKsXYlGTx2Rc1mX2RMyhiV0v/r79MU97f2yQZ66tGAmCUGHd8ty5c7Rt25aWf4+t0sAe18Zs0pnn1GQyg2MJfno1Z8+e5bHHHis1b1HbjW35d5UG+dh0bYzOVKemEJsZzOrgp8vVTkVI7VWx9oJ7bTbz329pENDkoflObv6Ha0cuYGqpZNDs0Zz5/SCRl26Ql5nN8CXv8OdnQWSnZZKdloFHi4ZkpqSTGpPI6DUf8eXQafh1bUPCrRi6vfE8kRevY2lvg0fzhuxesRkEAUcfN5r3as+OhT9h7+lMp5cCHxr+uDTWjJ7HG99+WC5XxBEXrjO365uV6mP1cT6ryNxVhDQmKz4mawM12pqgpg2c2kRNG0Q1nfrYXgnhsXg0b0hAvy4oTLXe8owVxqTGJumCDj0+uDtu/t78+PYixq+by8+TPicrNQNNoYZe4waTm5HDrx9/S6P22iXbA99tw8TcFIWZCTFXw/Fq3RilnRUdhvbUUwTKMksEiA29zV/LNqC0saxxMQkqgzSfVYz6OCbFpPaPMAkJiUrx3NSRNO7Uii0zVnHnRjRnfj/IkE/G4tPWH3WONkCMuZUShakCcyuttzhjhZwCtRpBI6Ap1FBYUGw/V6Ohwwu9eP7DUby28gP8ugTQ772XOb/zCEfX/a2X94HIhsXWKN38vXlj9XQEjUBy5IO+/SUkJAxHta0MVMUSWXD/73Cf0AX7QH8iFuxFk5OPzFxBgw8fbodaUZ8Cd4+Gk7j5PEKhhgYzAkFmxK1pO3F+6bFqcwNaFctl3wX3p4v7BPztA9kbsYB8TQ4KmTm9G3xYYv7y5CnudyAs9QD7oxYzttVfJeavKmpDe5Xkd6C62+vQjztIuBmNkUyGhb01Ni4O/PXFBsLPhtKkS+tS7zU2kfPn4iDu3Iym75SXibx4HYAebw7kt0++w8bFHjNLJU06t+biX8fISk2nWY92uvvLMktMi0ti15J1CBoBmdwYu/s8F9YEasJ8Vh6fAsXLST0QRtTi/bT6a+xD76kKasOYLMnvQHDSNs4lrOfVZjXHXLOqMIgycGvqDjyn9kRua0bY+F/xnt2H+J9OUZCag213X+yfbqrLWzSI4n8+jbmvCrm1GQmbzyMUaLBq64nj4NInofsxcbbCPtBfL9Tn7bm7yYu5q4vudT+V8SlwJ+g0jb8eQs71RBI2nMNjcjfs+5TfFr8sdtyaSk/PqZjJbfk1bDx9vGdzKv4ncgpS8bXtTlP7p3V5iwbU6fifUZn7Yia35nzCZjRCAZ5WbWntOLjc9VqZOONvH6gX9nP37bnczYvRRfoqojx54EG/A43tunMuYX0lWuXh1JX2KsnvQFW0V2kUt90fuVx7SO+Zd14C9E0Ji8z7ir7AFaYKBsx8Q5fu2eLeQbcx38/QK9e/a8VNEm1dVQz//H8Vvs8Q1Jb5rCyfAiWVY9e9MQnrz1WsQcqgrozJkvwOtFQ9z453+Z0AABlzSURBVNXknZVtmlqFQZQBh+eak7zjMmY+Dth09sFILkOjLkThaEHib5f0Bk9xYlcfw8zbHoCsy3F6gydq2UEKUnN0n227+2LXvfEDZZQW6lOXp7I+BQQwkskw9bDVRRAzJM0dnuNy8g4czHzwsemMzEhOoUaNhcKRS4m/6Q2k4hyLXY29mTegPRl7/0A6GLWMnIJU3Wdf2+40tutevIhSw35WJE91UZfaqyb6HSgvxX0B1CVqw3xWHgxVTlnUpTEJ9dfvgEFmIOtO3qSfiCBl11Ucnm1O4taL2Af64T6xK4VZ+m5MjWRan6eabO11Ib8Q1zc64vled7xnF+s0xcN7PiRUZ2mhPou436dA+rFwcm5qFYIyfQrIjBA0Gq12XkK5j4q3dSci0k9wNWUXzR2e5WLiVvzsA+nqPhF1YZZeXqP/vjDUmmwACoV8Orq+QXfP93jae7ZeXg36oT4fFraztLCfFclTXdSl9iryOyCgIS0vugKtYFhWvTLL4GUu7DWeC7uOAtrIhevf/5Kts1eXmDc+LJIf317ED+M+5a9lJa+MZN/N5Me3F/FZv3tbC6e27mNJ/3cNLnttmM/Kg6HKKYu6NCaL/A4UBUuqTxhkZUD75myDOj4Dua05Vu28uBN0mozTkcgUxnp5TVysiP3mKOmnIrFo5Ybb+C6Ez9yFQmWBqactrq/fsz31nFK+ACAlhfpM+/cmcmszLAO02l95fQpEf/kvHpPuaYXOI9pyc8p2hPxCGnxk+NCXMiMZNqYeZKjjMZfb4mXVjtN3gojMOI2xTD8kqpWJC0djvyEy/RRuFq3o4jaeXeEzsVCosDX1pKPrPecxPTynlKv+ksJ+3kz7FzO5Ne6WAeXOA9Xjd6CutFd1+R0IemcJA2eORmlnxZrX5/LiwvHsX/MHWSl3adGrPW2efVKXd9UrsxgXNIeD32/DpYkXSmtLjq77G01BIQ3bN+OJYYHlrtfW1YGAvp3LjFwI4NLYi1FfTwVg5YiZJZantLFk1NdT9RSX9kN6cnbboRLzPwq1YT6DB30KpJ+M0MtTUjlVQV0Zk/Xd74DBDhA2+OjeRGHVzhOrdp566UWHbbw/0bofdXursy6t8Yry7xM9tP5ioT4zL8Tg8lr7B/KZedrhPUvrdtXIyIhm617RpeXcSMTEyVIvv02Xhth0qbhtdEUIbPCR7m9Pq3Z4WrXTSy86ePOM9ycAdHZ7S5c2uPGKR66/eNjPmMwLtHd5rcJ5enlNo5fXtEeWpyzqQntZm7jQz2feI8tSFu0Gduf07wdxbuSO/1OPIZMbU5CnxtrRnhOb/9FTBoqz56vNOPpolZTISzf0lIEdn64lK/XetlmLXu1LDDVcWuTC4pzauo/mPdqVmFbd1Ib5zGtaL7ym9So1T/Fyqoq6MCZdLVrwXtuzjyxLbaV2bVQWQ25rRsqe0BLTPCZ1RW5tVqHyzH0dcRpWuiMJdUIGmeeikdspK1R2TcNMbktoyp4S07p6TMJMXvqSYnnyhKUeQGFcu9upiNraXn5PBnD9yAXObv+XdgO6cXzjHgL6dqbvlBHkZmTr5TX6z5Y/L1trVliQX0CvcUN4/sNRvLjgbb28WrPCe2aBmocseRePXGj3n2JQnFNb95EcdUcXB6E+8qjzWXnypB4Iw1ipKDVPbaE6xmRw0jasRNwWrU4eaWXg/oN4YlAU+SszOLZa63V6ua1evZVph6ScMIPKVFGKooDFZgZXWR0WChUdXUbr6niUZ5baq3LIZDIcvFxIi0vCws4K3/YtOPjDdm6cuIzcRP9LwdZVxe4vNxJ2PJgGAU145p2XWP/+cqyd7FF5udDzrUG6vM9/NKpc9ZcUufDK/jMobSzxaau1yom8GMaWGato9fQTbJq+khcXjuf4xj006dwaB897gXqC3llC5MUbbP7oa16Y//bDqqw09WE+U6gscBnd8ZHmriLqw5h0MGuIg1lDg47JmkqllAGVSoWZ0pwbE381tDy1FjOlOSqVqsx8KpUKczMlv96YWA1S1TzMzZTlaqcipPaqWHuVxJA592zKG3VoTqMOzfXSiywDhn06AYA+k4bp0sZ8p28mWBmKRy68fS6U7mMG6D57tW7MZ6Fb9fLEht6mw9CeeteqKj5BfZ/Pyjt3FSGNyUcfkzWRSikDXl5eXAsJJSlJXE26JqFSqfDy8iozn5eXF6HXQupt25W3nYqQ2qti7VVTUNpacWHXUQL6dn4grd+Ul8u8v7gCURKntu7D1rXkbYeKUN/nM2lMVozaOibLolKBiiQkJGou5Q1UVBepTKAiCQmJGh6oSEJCovLEXYsQW4Rqpz4+s4SEIZBWBiQk6hiRkZH4N21KTnZ22ZnrIOZKJaEhIXVyKVdCoqqQlAEJiTpIZGSkQfZ0g4KCWL58OUFBQTRt+nA3vJUhMTGRwYMHExgYyIwZj35QsYi6uqcrIVGVSMqAhIREiURGRtKsWTPeeOMNli9fXiV1fP3114wfP56jR4/SqVOnKqlDQkKibCRlQEJCokQGDBjA6dOnCQkJwdq6avzaFxYW8sQTT5CTk8O5c+dQKOqGQxwJidpGrfZAKCEhUTVs27aN/7d358FR1/cfx5+7m70SNhcL5IAcHD9CEESxSqF1BAGt/jwqre3gjFNbqVrHzuhPEXQ8Wmtk1AJqBVJtp5Viw1irAmWQUBGtyMgRJAlJuBIWckFIQm42ye7vjy2rCwEScmySfT1mMux+j8/3/f3OkH3ls9/P9/PRRx/x2muv9VoQADCZTGRmZrJ//36WL1/ea8cRkYtTz4CIBGhoaCA9PZ1JkyaxYcMGDAZDrx/zscce84eC5OTkXj+eiARSGBCRAI8//jgrVqwgPz+f1NTUPjlmfX096enpTJkyhXXr1vVJABGRb+hrAhHx+/rrr1m+fDnPPvtsnwUBAIfDweuvv86GDRv48MMP++y4IuKjngERAcDj8TBjxgzq6urIycnBYrH06fG9Xi+33347OTk5FBQU4HA4+vT4IqFMPQMiAsBbb73Fjh07WLVqVZ8HAQCDwcAf/vAHampqeO655/r8+CKhTD0DIkJlZSVpaWncdddd/OlPfwpqLS+//DKLFy9m165dXHXVVUGtRSRUKAyICPfccw8ff/wxRUVFDB06NKi1tLa2cvXVV2O32/nyyy8xmUxBrUckFOhrApEQt2XLFt59911effXVoAcBALPZTGZmJjt37iQzMzPY5YiEBPUMiISwlpYWJk+eTEJCAlu3bu1XQ/p++ctfsnbtWoqKioiLiwt2OSKDmnoGRELYkiVLKCkpYeXKlf0qCICvNqvVyqOPPhrsUkQGPYUBkRBVVFTESy+9xMKFC3t8RsKeEBsby+9//3uysrLYvHlzsMsRGdT0NYFICPJ6vcyePZuSkhLy8vKw2+3BLqlDXq+XG2+8EZfLRW5ubr+tU2SgU8+ASAhas2YNn3zyCStWrOjXH7AGg4GVK1dy7NgxMjIygl2OyKClngGREFNdXU1aWhqzZs0iKysr2OV0ynPPPcdLL73Evn37SEtLC3Y5IoOOwoBIiHnggQfIysqisLCQ+Pj4YJfTKS0tLUyaNInExMR+N+pBZDDQ1wQiIWT79u388Y9/JCMjY8AEAQCbzcaKFSvYtm0bq1evDnY5IoOOegZEQsTZJ/vZbDZ27NgxIJ/sN3/+fLKzsyksLOwXD0gSGSzUMyASIpYvX87+/fvJzMwckEEAYOnSpbS2trJo0aJglyIyqKhnQGSQKy4uxmg0kp6ezoIFC1i+fHmwS+qWlStX8qtf/YrPP/+ckSNHkpycrHsIRLpJYUBkECsoKCA9PZ2ZM2dy4MABCgoKcDgcwS6rWzweD9OnT6e+vp4jR46QlZXFHXfcEeyyRAY0fU0gMogVFxcDsHXrViZOnMjWrVuDXFH37d27F6fTSWFhIR6PhyNHjgS7JJEBT2FAZBArKSkBwGQykZOTMyhuuouOjqawsBAAt9tNUVFRkCsSGfgUBkQGsV27dgFw9913U1hYyIwZM4JcUfeNHj2a3NxcFi5cCPh6CkSke3TPgMggVl9fT35+PtOmTQt2Kb0iLy+PmJgYEhMTg12KyICmMCAiIhLi9DWBiIhIiAsLdgEiA5XL5aKqqirYZQSF0+kkKSmpy/vpmnX9mon0BYUBkcvgcrkYPyGNlqbmYJcSFLZwO0UFhV36cHO5XKSNn0BzS1MvVtZ/2W3hFBYVKBBIv6QwIHIZqqqqaGlqZuwb87CPcwa7nD7VfLCKQ4+8T1VVVZc+2KqqqmhuaWLe2Ddw2sf1YoX9T1XzQd4/9EiXr5lIX1EYEOkG+zgnQyYl9EhbJS98TMozN523vOKdnURfPwZbSmyn2zqakY2nuRWj3UzyU3P8y9ub3BQv3oDBYiLyu6kMu2tyj9TeFU77OBKGTOp2Ox+XvMBNKc+ct3xnxTuMib6eWFtKp9vKPppBq6cZs9HOnOSn/Mvd7U1sKF6MyWAhNfK7TB52V7frFumPdAOhSBA0HTzJgYfew/XKJ+y/xzcl7xlXLQB7b3yTslVfcODhf9De3ErryQY8La2dbvtMaS3e1nZSX7gFb7uHM6Wn/euqNxYQe+tExrxyBzWbC3v2pHrRyaaDvHfgIT5xvcLq/fcAUHvGBcCbe2/ki7JV/OPAw7S2N9PQepJWT0un2649U0q7t5VbUl/A423n9JlS/7qC6o1MjL2VO8a8QmHN5p49KZF+RD0DIkFw4t3dJD89B0ucg/3zVwess8ZHkvDgDMr/vIOm/Irz9m2taeL4sm0By+LuuxZ7qu/pgu7yeqwJUb62EqNwl9dhTYz677rTRKcN9+1kGjh/C+w+8S5zkp/GYYlj9f75AesirfHMSHiQHeV/pqIp/7x9m1pr2HZ8WcCya+PuY6g9FYB6dzlRVl/vTpQ1kTp3OVFW33MLTrvLGR6dBoCRgTnTo0hnDJzfBiKDkcFw3ox7RrvFtyrMhMfd1uFu3nZPwA/felqIJd7BmfI6ANxldVjiI7+1zhcOAPAMvEeMGDj/elmMdgBMhjDaPO4O9/N42wN+vn3BHJZ46s6UA1DnLiPSEu9fF2WJp87tW+fF05OnItKvqGdAJAiGz5/K0Yxs7KOHYgw3d2lfc0w4o1+89YLrrYnRGMxGSp7fhMFqwpoYRVnmdqJmpBJ7ywSKn/4XNf8+QMyc8d09jT4zdfh8so9mMNQ+GrMxvEv7hptjuHX0ixdcH21NxGgws6nkeUwGK1HWRLaXZZIaNYMJsbfwr+KnOVDzb8bHzLlgGyIDncKASBCYnRHYUofSWt3EsB9PAWD8Wz8J+Dfu3u8AEDU9tcvtJy8O/OBKeGC6//XYZT+8rJqDKcLsZKgtlabWaqYM+zEAPxn/VsC/34m7F4DUqOkdN3IRc5IXB7yfnvCA//UPxy47d3ORQUdhQCQIzDHhJD0xK9hlDBjh5hhmJT0R7DJEBi2FAZEB5MTaHMyx4T3Wxd98qIrSFf8Brxf7WCeJD3+/R9rtL3JOrCXcHNujXfzrDi/kWP1uHp7y7x5rUyTYFAZEetnJD/ZR92UJpggLSYtmc2pDPo155bQ3ukl98VZKX/+MttPNtNW2EJE+gtaaJtxldYx7Yx4F964hanoKLSXVjPjv1wYAZ47XUpa5HbxebMmxRN0wluNLP8WaGMWwH11JeNqITtVmH+tk7NI7ASi6P6s3Tr/L9p38gJK6L7GYIpidtIj8Uxsob8zD3d7Irakv8lnp6zS3naalrZYREek0tdZQ5y5j3rg3WFNwLylR06luKeE7I+71t1l75jjbyzLx4iXWlszYqBv49PhSoqyJXDnsR4wIT+t0fbePeZm1RQt649RFgkZhQKSXnSmpJnzCCGLnpmG0+v7LGcwm3OV1NOb57lR33nYF9vHDOfzYh4x/+6ccXriOttpm8HiIv38a7Q1ujmZk45g6CoCKv36F0RaG0WamsbCS8CviCYuy4bxzUkAQuNQwxLOqPsol6voxvXkZOq36TAkjwieQFjuXMKMVAJPBTJ27nPLGPACucN7GcPt4Pjz8GD8d/zbrDi+kua0WDx6mxd+Pu72B7KMZjHJMBeCrir8SZrRhNtqobCwkPvwKbGFRTHLeGRAELjUMUWSwUhgQ6WUjH72BxvwKjv5uM6OemMWp9fmk/WU+x5ZuxdPse5iQyWHFaDFhGuL78DOYfcMKvR4v3nYv3rb2gDa9Hi/D5l1JRHqcf5ktKYYTWXto3F/J8LunfLNt+zlD4s4ZUVj1US5njtf2m68Ibhj5KBWN+Ww++jtmjXqC/FPrmZ/2F7YeW0qrxzcXhNXkwGS0YDUNAXxhoc3jxuv14PW20+4NHJLp9Xq4ctg84iLS/ctibEnsOZFFZeN+pgy/27/cN/QwYO/eOVGRfkRhQKSXVf5tF83Fp8BoICzGjnn4EEpX/IeGnFIip6VcdF+j2cTx17bRUlxN4iPfpzHX15MQf991uJZswTzcgWmIhcjrUqjJLqKttjngL/xLDUNszCvn6G83EzP7fyh5fhMpz9/cI+fcHbsq/8ap5mIMGLGHxTDEPJz/lK6gtCGHlMhpF93XZDSz7fhrVLcU8/3ERyhvzAXguvj72OJagsM8HItpCCmR11FUk01zWy1joq7373+pYYgAW1xLKG/MZ/2RJ/lBym/9vRciA5nB6/Uq9op00Z49e5g6dSqTNj3QY3MTdKRowVr/UMP+oiG3jNybM9m9ezdXX311p/c7e80emLSpR+Ym6MjaogX+oYb9SVlDLpm5N3f5mon0FT2BUKQf629BoL/rj0FAZCBQGBAREQlxCgMifaRowdoebzP39rep/u/sg0czsil+ZiNHM7I73Lb5UBWHHvuQQ49+QOmbn3e4jdfr5fDCdRx5agNlq74AfDcY7v/JX3u89kvpjeF7b+feTmG1b/bB7KMZbCx+huyjGRfcvjPbrDu8kDf33uh/f7BmK5n7ftBzRYv0AYUBkR5w5Mn1tFY34fV4OPDQe7gr6nAt2cKRJ9dTvakgYNuzoaDinZ2c3l5MY145xc9u5MhTGzj5/tddOq5lhIPYuWkXnbb4rLPPFBi77Ic05JR20BrUf+UiPG04ozP+l8b8CjzuNpx3TMIUaetSXZey/siTNLVW4/F6eO/AQ9S5K9jiWsL6I09SUL0pYNuzoWBnxTsUn95OeWMeG4ufZcORp/j65PtdOq7DMoK02LkXnbb4rM5sA77nDjjto/3vx8XMJNo6skt1iQSbRhOI9ICht03k1Po8bKlDiZqRiiHMiMfdjnlYBCf/uY/YmydccN+yzO3YUmIB3939w+Zd6V93bNmntNU0+99HzxxLzMxx57VxsWmLz3WxZwq4y+v87ZidEbTVNGMZ4bjE2XfdxKG3kXdqPUNtqaRGzcBoCKPd4ybCPIx9J//JhNgLj2rYXpZJrC0FgPLGPK4cNs+/7tNjy2huq/G/Hxs9k3ExM89r42LTFndlG5HBQmFApAdETk+hcs1umgoqSVo0mxNZe4idO54hV42k8Od/D9jWYPRNwetp8k23621tJ/4X0wiLtp/f8Nkpis+6wLTDlngHZzZ+a9rimzoOH5d6poAlPpLG/RUAtJ5qIiymg5p6QErkdHZXrqGyqYDZSYvYcyKL8bFzGTnkKv5e+POAbQ0GXwem29MEQLu3lWnxv8AeFn1eux7aA54TcKFph33TFm8EfNMWT7DcdFnbiAwWCgMiPcBgNGIdGYW7op6waDuOa5KoXL2T+p0ujGZTwLaWOAdlq76g7isXEZMTSHj4exQ/sxGzMwLrqGjif/7NWPpRj3duMqOOpi2u/ewwYZE2hkzx/TXb0TMFTr7/NZHTkrEm+j5YHdcmUfXBPoqf3UhE+giMlt75FWE0GImyjqTeXYE9LJokxzXsrFyNq34nJmPglM4OSxxflK3CVfcVCRGT+V7Cw2wsfoYIs5No6yimxX8THmaNerxTx+9o2uLDtZ9hC4skcciUTm8Deu6ADA4KAyI9JPnpuf7XjmtG4bhmVMD6s8MEU37ju7ks4cEZ/nXj3phHd507bXHD3lLifnat/33EFfFM3f1/Ads0HTiB885vxvwbDAZGL7mt27V0xtzkp/2vRzmuYZTjmoD1Z4cJ/iDlNwDMSHjQv27euDe6ffxzpy0ubdjLtXE/6/I2s5MWMTtpUbfrEQkm3UAoMoCFRdv8ownONfLX1xN2iRv/khfPwWC6+K+Bqo9yscT3/H0DwWALi/aPJjjX9SN/jS0s8qL7d2abgzVbMZvCL7tGkWBQz4BINzQfrArq8c/OZNiQW9Zrx7CNHopt9FD/Mbp7zlXNB3uirMtydibDsobcXjtGhNnJtLj7A44RzHMW6QyFAZHL4HQ6sYXbOfRI14a2DRa2cDtOp7NL+zidTuy2cN4/9EgvVdW/2W3hXb5mIn1FcxOIXCaXy0VVVXB7BoLF6XSSlJTU5f10zbp+zUT6gsKAiIhIiNMNhCIiIiFOYUBERCTEKQyIiIiEOIUBERGREKcwICIiEuIUBkREREKcwoCIiEiIUxgQEREJcQoDIiIiIU5hQEREJMQpDIiIiIQ4hQEREZEQpzAgIiIS4hQGREREQpzCgIiISIhTGBAREQlxCgMiIiIhTmFAREQkxCkMiIiIhDiFARERkRCnMCAiIhLiFAZERERCnMKAiIhIiFMYEBERCXEKAyIiIiHu/wHY/tP1LnocawAAAABJRU5ErkJggg==\n",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {
"tags": []
}
}
]
}
]
}