{ "cells": [ { "cell_type": "markdown", "id": "d25254ef-77e8-414a-a4c8-f367e48a4eff", "metadata": {}, "source": [ "# Анализ данных FDA Drug Adverse Events\n", "\n", "В данной работе проводится анализ данных о побочных эффектах лекарственных препаратов. \n", "Цель работы — загрузить датасет, изучить его структуру, провести базовый анализ и построить визуализации с использованием библиотек `pandas`, `numpy`, `matplotlib`, `seaborn` и `tqdm`." ] }, { "cell_type": "code", "execution_count": 1, "id": "f9f3cdba-7699-400d-941d-b6aa79357639", "metadata": { "jp-MarkdownHeadingCollapsed": true }, "outputs": [], "source": [ "import pandas as pd\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "import seaborn as sns\n", "from tqdm import tqdm" ] }, { "cell_type": "code", "execution_count": 2, "id": "1665af01-f42a-48f3-aa29-2e9c8b823f96", "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", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "

	report_id	country	occur_country	report_type	serious	death	life_threatening	hospitalization	disabling	receive_date	...	reaction_pneumonia	reaction_bleeding	reaction_complexity	year	month	day	weekday	report_type_risk	age_drug_risk	gender_drug_risk
0	26016170	US	US	1	2.0	0	0	0	0	2025-11-10	...	0	0	2	2025	11	10	0	0.080736	0.651584	0.000000
1	26017452	NP	NP	1	1.0	1	0	1	0	2025-11-10	...	0	0	4	2025	11	10	0	0.080736	8.846939	0.000000
2	26017586	US	US	1	2.0	0	0	0	0	2025-11-10	...	0	0	3	2025	11	10	0	0.080736	0.000000	0.000000
3	26018910	SG	SG	1	1.0	0	0	0	0	2025-11-10	...	0	0	2	2025	11	10	0	0.080736	4.428571	0.071429
4	26018949	EU	EU	3	1.0	0	0	0	0	2025-11-10	...	0	0	2	2025	11	10	0	0.178571	0.000000	0.000000

\n", "

5 rows × 40 columns

\n", "

" ], "text/plain": [ " report_id country occur_country report_type serious death \\\n", "0 26016170 US US 1 2.0 0 \n", "1 26017452 NP NP 1 1.0 1 \n", "2 26017586 US US 1 2.0 0 \n", "3 26018910 SG SG 1 1.0 0 \n", "4 26018949 EU EU 3 1.0 0 \n", "\n", " life_threatening hospitalization disabling receive_date ... \\\n", "0 0 0 0 2025-11-10 ... \n", "1 0 1 0 2025-11-10 ... \n", "2 0 0 0 2025-11-10 ... \n", "3 0 0 0 2025-11-10 ... \n", "4 0 0 0 2025-11-10 ... \n", "\n", " reaction_pneumonia reaction_bleeding reaction_complexity year month day \\\n", "0 0 0 2 2025 11 10 \n", "1 0 0 4 2025 11 10 \n", "2 0 0 3 2025 11 10 \n", "3 0 0 2 2025 11 10 \n", "4 0 0 2 2025 11 10 \n", "\n", " weekday report_type_risk age_drug_risk gender_drug_risk \n", "0 0 0.080736 0.651584 0.000000 \n", "1 0 0.080736 8.846939 0.000000 \n", "2 0 0.080736 0.000000 0.000000 \n", "3 0 0.080736 4.428571 0.071429 \n", "4 0 0.178571 0.000000 0.000000 \n", "\n", "[5 rows x 40 columns]" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df = pd.read_csv(\"data/gm_feature_engineered_fda.csv\")\n", "\n", "df.head()" ] }, { "cell_type": "code", "execution_count": 3, "id": "b2c52280-a63e-415f-9174-3ddeeb0e9092", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Index(['report_id', 'country', 'occur_country', 'report_type', 'serious',\n", " 'death', 'life_threatening', 'hospitalization', 'disabling',\n", " 'receive_date', 'age', 'age_unit', 'gender', 'weight', 'drug',\n", " 'reaction', 'is_serious', 'severity_score', 'severity_norm',\n", " 'age_group', 'country_group', 'drug_risk', 'drug_freq',\n", " 'high_risk_drug', 'reaction_length', 'reaction_pain',\n", " 'reaction_infection', 'reaction_death', 'reaction_cancer',\n", " 'reaction_failure', 'reaction_pneumonia', 'reaction_bleeding',\n", " 'reaction_complexity', 'year', 'month', 'day', 'weekday',\n", " 'report_type_risk', 'age_drug_risk', 'gender_drug_risk'],\n", " dtype='str')" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.columns" ] }, { "cell_type": "markdown", "id": "c7e87bc6-bbbb-45c1-a51c-f1483111eb56", "metadata": {}, "source": [ "## 1. Первичный анализ данных\n", "\n", "На этом этапе рассматривается структура таблицы: количество строк и столбцов, типы данных, наличие пропусков, а также основные статистические характеристики числовых признаков." ] }, { "cell_type": "code", "execution_count": 4, "id": "a17b1b49-7d0d-4f70-a1c2-57dfc056a1d9", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "RangeIndex: 1287 entries, 0 to 1286\n", "Data columns (total 40 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 report_id 1287 non-null int64 \n", " 1 country 1287 non-null str \n", " 2 occur_country 1287 non-null str \n", " 3 report_type 1287 non-null int64 \n", " 4 serious 1287 non-null float64\n", " 5 death 1287 non-null int64 \n", " 6 life_threatening 1287 non-null int64 \n", " 7 hospitalization 1287 non-null int64 \n", " 8 disabling 1287 non-null int64 \n", " 9 receive_date 1287 non-null str \n", " 10 age 1287 non-null float64\n", " 11 age_unit 1287 non-null float64\n", " 12 gender 1287 non-null str \n", " 13 weight 1287 non-null float64\n", " 14 drug 1287 non-null str \n", " 15 reaction 1287 non-null str \n", " 16 is_serious 1287 non-null int64 \n", " 17 severity_score 1287 non-null int64 \n", " 18 severity_norm 1287 non-null float64\n", " 19 age_group 1286 non-null str \n", " 20 country_group 1287 non-null str \n", " 21 drug_risk 1287 non-null float64\n", " 22 drug_freq 1287 non-null int64 \n", " 23 high_risk_drug 1287 non-null int64 \n", " 24 reaction_length 1287 non-null int64 \n", " 25 reaction_pain 1287 non-null int64 \n", " 26 reaction_infection 1287 non-null int64 \n", " 27 reaction_death 1287 non-null int64 \n", " 28 reaction_cancer 1287 non-null int64 \n", " 29 reaction_failure 1287 non-null int64 \n", " 30 reaction_pneumonia 1287 non-null int64 \n", " 31 reaction_bleeding 1287 non-null int64 \n", " 32 reaction_complexity 1287 non-null int64 \n", " 33 year 1287 non-null int64 \n", " 34 month 1287 non-null int64 \n", " 35 day 1287 non-null int64 \n", " 36 weekday 1287 non-null int64 \n", " 37 report_type_risk 1287 non-null float64\n", " 38 age_drug_risk 1287 non-null float64\n", " 39 gender_drug_risk 1287 non-null float64\n", "dtypes: float64(9), int64(23), str(8)\n", "memory usage: 402.3 KB\n" ] }, { "data": { "text/plain": [ "report_id 0\n", "country 0\n", "occur_country 0\n", "report_type 0\n", "serious 0\n", "death 0\n", "life_threatening 0\n", "hospitalization 0\n", "disabling 0\n", "receive_date 0\n", "age 0\n", "age_unit 0\n", "gender 0\n", "weight 0\n", "drug 0\n", "reaction 0\n", "is_serious 0\n", "severity_score 0\n", "severity_norm 0\n", "age_group 1\n", "country_group 0\n", "drug_risk 0\n", "drug_freq 0\n", "high_risk_drug 0\n", "reaction_length 0\n", "reaction_pain 0\n", "reaction_infection 0\n", "reaction_death 0\n", "reaction_cancer 0\n", "reaction_failure 0\n", "reaction_pneumonia 0\n", "reaction_bleeding 0\n", "reaction_complexity 0\n", "year 0\n", "month 0\n", "day 0\n", "weekday 0\n", "report_type_risk 0\n", "age_drug_risk 0\n", "gender_drug_risk 0\n", "dtype: int64" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.info()\n", "df.describe()\n", "df.isnull().sum()" ] }, { "cell_type": "markdown", "id": "676c57bf-d2c3-4cba-b333-bf9f2990135a", "metadata": {}, "source": [ "## 2. Группировка и анализ признаков\n", "\n", "Для более подробного анализа используются группировки данных. \n", "Это позволяет сравнить показатели по полу, странам и лекарственным препаратам." ] }, { "cell_type": "code", "execution_count": 5, "id": "124f939b-cc53-4b0d-a7b4-5101cbda0645", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "serious\n", "1.0 808\n", "2.0 479\n", "Name: count, dtype: int64" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[\"serious\"].value_counts()" ] }, { "cell_type": "code", "execution_count": 6, "id": "a1fb32cc-b1ff-4585-b7b1-661824bcea9d", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "death\n", "0 1215\n", "1 72\n", "Name: count, dtype: int64" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[\"death\"].value_counts()" ] }, { "cell_type": "code", "execution_count": 7, "id": "20a5a103-b2e5-416a-8f90-0e4fc05d8d8b", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "np.float64(53.92307692307692)" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[\"age\"].mean()" ] }, { "cell_type": "code", "execution_count": 8, "id": "8965b8d8-db80-4d18-a86c-70446756e7c7", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "gender\n", "Female 54.161939\n", "Male 53.464853\n", "Name: age, dtype: float64" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.groupby(\"gender\")[\"age\"].mean()" ] }, { "cell_type": "code", "execution_count": 9, "id": "7c4ae568-ebae-4675-90e0-6f6bea52e4a5", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "country\n", "US 1252.0\n", "EU 184.0\n", "CA 110.0\n", "JP 62.0\n", "GB 55.0\n", "CN 34.0\n", "AU 20.0\n", "BR 7.0\n", "IN 6.0\n", "PR 3.0\n", "Name: serious, dtype: float64" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.groupby(\"country\")[\"serious\"].sum().sort_values(ascending=False).head(10)" ] }, { "cell_type": "code", "execution_count": 10, "id": "ab8f72ba-d2d4-4c50-a0a0-d233fc0ab9b9", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "drug\n", "APREMILAST 14.0\n", "ELRANATAMAB 12.0\n", "SILDENAFIL 9.0\n", "DOLUTEGRAVIR 8.0\n", "DEFITELIO 8.0\n", "MELPHALAN 8.0\n", "GLIVEC 8.0\n", "HEMLIBRA 8.0\n", "YUTREPIA 8.0\n", "TUKYSA 7.0\n", "Name: severity_score, dtype: float64" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.groupby(\"drug\")[\"severity_score\"].mean().sort_values(ascending=False).head(10)" ] }, { "cell_type": "markdown", "id": "f4994af1-aaa7-4c2b-90a7-70c16fca4b12", "metadata": {}, "source": [ "## 3. Визуализация данных\n", "\n", "В этом разделе построены графики для наглядного анализа данных: распределение возраста пациентов, сравнение серьёзности случаев по полу и количество случаев по странам." ] }, { "cell_type": "code", "execution_count": 11, "id": "122f15cd-3374-4148-a201-9c105329e160", "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjsAAAHHCAYAAABZbpmkAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjksIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvJkbTWQAAAAlwSFlzAAAPYQAAD2EBqD+naQAAPLNJREFUeJzt3Qd8FHX+//FPEkJCj4A06YLSRUH6WeiIKIJ3FkQEzsIBUhQRpSOCeFJUkNND0ENEsaAU6U0FBEFEqqAoKEIUhVAkhGT+j8/38Zv9724ChrCbzXx5PR+PJezM7Mzsd2d33/stM1GO4zgCAABgqehI7wAAAEA4EXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIO8hWM2bMkKioKN8tPj5errrqKunVq5ccPnyYVwMAEHK5Qr9K4K+NHDlSKlSoIKdPn5bPPvtMXnnlFVm4cKFs27ZN8ubNSxECAEKGsIOIaNOmjdStW9f8/5///KcUKVJExo8fLx999JHcc889vCoAgJChGQs5QtOmTc3fffv2mb+///67PP7441KzZk3Jnz+/FCxY0ASkr7/+Ot1jtXZo+PDhpjlMm8VKliwpHTp0kO+++87M/+GHHwKazoJvN910k29dq1atMtPeeecdeeqpp6REiRKSL18+ue222+TAgQPptv3FF19I69atpVChQqZG6sYbb5TPP/88w+eo28lo+7rvwWbOnCl16tSRPHnySOHCheXuu+/OcPvne27+0tLSZOLEiVK9enVTRsWLF5eHH35Y/vjjj4DlypcvL7feemu67WgzY/A6M9r3559/Pl2ZquTkZBk2bJhUqlRJ4uLipEyZMvLEE0+Y6X8luNyKFi0qbdu2NbWA/s6ePSujRo2SK6+80mxDn4u+hsHbuP322808LYdixYqZ1/abb75J99z0Ob/11lty9dVXm2X19VizZk3Acj/++KP861//Msvoa6Wh/e9//7t5XYIdPXpU+vXrZ7at+1e6dGm5//775bfffvMdd+e7uWV9IdsMdiHvhTNnzsjQoUPN89bjW98Hf/vb32TlypUZrvPf//53uu3VqFEj3bGg94Onbdy4McPjNrPvBV2fbiuY7pOu0y0bLfvzPX+d7zp58qQ89thj5ljV10vLW9fnOE7ANvwfHxMTI1dccYU89NBD5vVGzkHNDnIEN5joB7f6/vvvZe7cueZDXJu7tD/Pf/7zHxMmduzYIaVKlTLLpaammi/n5cuXmw/BPn36yPHjx2Xp0qXmy1C/+FxaY3TLLbcEbHfQoEEZ7s/o0aPNh9fAgQMlMTHRBIXmzZvLli1bzIeuWrFihQlg+kGsX+TR0dEyffp0E9w+/fRTqVevXrr16hfcmDFjzP9PnDghPXr0yHDbQ4YMkX/84x+m1uvXX3+Vl156SW644Qb56quvJCEhId1j9MNVv4jUBx98IB9++GHAfA022l+qa9eu8uijj5pQ+fLLL5v1aTiLjY2Vi6Uf7u5zCw5aGii0uVL3s2rVqiZcTJgwQb799lvzOv+VKlWqyNNPP22+aPRY0VpAfS3379/vW0bL6o033pA777zTfElpENX92blzZ7ry0P3QIHvw4EFTDvraapn4N6GuXr3ahF4tL/2ymzJligm2GzZs8H2x6pf02rVrzbGnr61+qWqTrH756nHqrk9fa319dF+6desm1113nQk5H3/8sfz000+mTP73v//5tv3qq6+aZbWMXLVq1bqgbZ5PZt4LSUlJ8t///tcs++CDD5r31bRp06RVq1amDGrXri2hou+zjGTlvXA++j7W10Jp+T777LMmEGv5K/1hpfQ402NWg1337t3Nc128eLEMGDBAfv7554DXRd1xxx3mB5YG7nXr1pnX788//wx4TRFhDpCNpk+frj+LnGXLljm//vqrc+DAAWf27NlOkSJFnDx58jg//fSTWe706dNOampqwGP37dvnxMXFOSNHjvRNe/311836xo8fn25baWlpvsfpMs8//3y6ZapXr+7ceOONvvsrV640y15xxRVOUlKSb/q7775rpk+aNMm37sqVKzutWrXybUedOnXKqVChgtOiRYt022rUqJFTo0YN3319/rrOYcOG+ab98MMPTkxMjDN69OiAx37zzTdOrly50k3fs2ePWccbb7zhm6br839rf/rpp+b+W2+9FfDYRYsWpZterlw5p23btun2vWfPngHrVMH7/sQTTzjFihVz6tSpE1Cm//vf/5zo6GizH/6mTp1q1vH5558756Pr8l+feuqpp8xjExMTzf0tW7aY+//85z8Dlnv88cfN9BUrVpxz/e5r++WXXwY8t+BpP/74oxMfH+/ccccdAa93sHXr1pnHvvnmm75pQ4cONdM++OCDdMv7Hz+uLl26mNciI5ndZkYu5L1w9uxZJzk5OWCZP/74wylevLjTrVu3LK0zo9dz4cKF5vGtW7cOOMYu5L2g69NtBdN90nXqPgZz3+v6N9jcuXPNvGeeeSZg+p133ulERUU5e/fuPef7wH2vV6tWLd16ETk0YyEi9Jf05ZdfbqqI9Req/qLSX99aBaz0l7TWlLi1N0eOHDHLaFXy5s2bfet5//33TbNG7969020joyrxzNLmhQIFCvjua22BNo9pJ2qlNTx79uyRe++91+yb/krXm1Z9N2vWzDR3aI1GcHObNoecj9bK6OP0l6y7Tr1pLUTlypXTNSFoU4NbXucyZ84c0wzRokWLgHVqjZSWafA6U1JSApbTm+77+eivXf3Frb/C3V/H/tvXX85aO+O/TrfpMnj7GXH3SX/Z6y9nPVa0pkNfe+W+Lv379w94nNbwqAULFgRMP3XqlFmfvo6vvfaaadbTZlB/DRs2NGXkKlu2rGkC01/4ekwqt5bP3Uc9FrSpTmscgo/Ta665xtQAXOxxmtltXixtksmdO7f5vx6T2rSsNRfa1y5U29GsoDVKHTt2lPr161/Ue0Ffk+DjVl/nrNDjSZ+/1uoFH0+6z5988kmGx9OhQ4fMa63N7fo5gJyDZixExOTJk82XS65cucwXjYYYN9wo/ZCbNGmSaTrQ5gX3y8W/qUtpk4Y+VtcTSvphGvyFpF8obtu/Bh3VpUuXc67j2LFjctlll/nu64dh8HqD6Xr1w/RcywU3N7n9AoIDRvA6dV+0f0pGtJnO35IlS0wQvRDajKdNi9pc9t5776XbvjYZnGudwdvPiDbb+D9ey0ebv9ygoP1Y9PjR18iffjFqCND5waMBn3vuOd+6tM+Mf7h1pwfTY1a/2DR06bq1qUKbyrT5UgOff38OLXP/41S/0EMhs9sMBW0WfOGFF2TXrl0mWLm0aTkUtE/U9u3b5d1335VZs2Zd1HtB9/FCj9tz0eNFj+fgY8Jt7go+nrSvmt5c2tzpHl/IGQg7iAjtz+KOxsqItqVrLYH2b9BOp9oxUb/M+vbtm67GJBLcfdAPuHP1XfAPIFoD88svv5jalb9ar36B6y9H/WV5vnUq/SWp9Iv3fOvUoKNfLBkJ/oLQX9jPPPNMwDTt16Ij5TKiQUb7A2lH0oz6/uj2taO59rPJiNbu/RWtxdEvXaVB48UXXzR9VLSGwf+5Z7aWRPt/6C9v7S+j/S80iGig0hqwC6E1iho69LjUmiB9vO6D1laG6zjNrm3q6/nAAw9I+/btTV8VPYb0mNSg5faxuxj6ntD3uPaJCa5Vy8p7QTsXay1dcK2i9p8Jt86dO5vaYN1n7W+on1nal3DZsmUXVcOM0CHsIEfS2oGbb77ZdIgMrslwmy6UdkDWjqj6qzMUnWxdbs2NS39h7t2719dJ1O34rKPEtEnur2i1tu7j+QKeu17dlv5yzugLIJh2SNUPU63dOt869UO3cePGAU0g56LlG/yczteJWJshNPDddddd59y+W62f1Q9+rSHz3ycNOvrLW7/0dfvlypUzXzT6urm/vpV2bNdjRuf70xogtxZI16tNVFqz4N9hPPgYUNqhWjsAuwFRj1Ot3XODmNImv+CROFoGwaPHsiqz2wzFdipWrGiak/xfN63FCwWttdVavYxGI2blvaCjxYKPW22mzAo9XvQ9o52y/Wt3tPbIne9Py8l/2xpAtYl7/fr1JpAi8uizgxxJf8kFD/HUX2labe9Pf5Fr85DWPAQLfvyFePPNN80Hnf8Hv9bM6OgrpX059MNYh6K6ozv8ae1D8L7rc8poWLc/HdGhy40YMSLd/ut97Z/h0v4T2j9Aa8nO14ylfR60GVB/bQbTdVzMl6T2n9Ean7Fjx54zyOj29XUL/tXtNsloP6cLpY9T7rByd2SRjrbx59Ym6VD1c9Hjx39d/s/Nv2+KDnfW59qyZUtfTUNGx6n2XfJvdnWPUw18waPCsnKcZnabF8t9jv7b0h8WWi4XS99bOtJKh+Kfq1byQt4LoabHk5Zn8OeK1gLqce5+DmT2+ETkUbODHElDgfar0KHSjRo1MkOVtRlGf0H506pjDSbaMVWHw+rwXv3y1F9lei4S7VCaFdps1qRJE7N9rR3QL1GtCdAhuEqb1HRYrn7o6blrdDntXK1f6tpxUmt85s2bZ/ZF+ydps4v+OtW+IS43JG3dutV8gegvQA1Q2oSktRXaP0ibEPSXpfZb0i9KHTKt5x/S56dNAPpY3c756HB97UujzQ/6S1e/rLUWTGsuNIRp3yjtgJ0V2r9Hm+bOV7ulVfzaJ+ORRx4xZaM1TPpFor+Sdbp2+P2rGi99DbRZxQ0nehoC7aflhkft/Ku1HdpkoeFNn7MeD9rnRMtQawndjqf6uukxpa+xNjloCNNageDOwzq8XIdZ+w89V/rl69Lt6/Bi/SVfrVo18zrqa+Pfr0xpM5AGZj2VgjbNaljWDr869Hzq1Klm/zMrs9u8WLodrdXRctGwqMeg7qtuM6OAv3v3blm0aFHANF1O3ys6XfuxuDREag2inmvpXDL7XgiHdu3amWNGT3eg29bXR491DbvafOh/Sgul70M9Pt1TI+j7XU8L8FfHNbJRBEeC4RIeer5x48bzLqdDzx977DGnZMmSZkh648aNzfDajIYh61Dcp59+2gz5jo2NdUqUKGGGiH733XdZHnr+9ttvO4MGDTJDqXX7Ohxbhx4H++qrr5wOHTqYofM6LF6HC//jH/9wli9fHrDtv7rpUGN/77//vtOkSRMnX7585lalShUz/Hv37t1mfu/evZ0bbrjBDB8PFjz03PXqq6+aYeH6fAoUKODUrFnTDBc/ePBgloee6zDcTZs2BUzP6DU6c+aM89xzz5ny1nK67LLLzL6MGDHCOXbsWLrtBa/Pv6wSEhLM8aBDlv2lpKSY9bnHQZkyZcxrqMeSa9u2bU7Lli3N65U7d26zzN133+1s3bo13XPT5zxz5kxzigHd52uvvTbdMGUdit21a1enaNGiTv78+c2pCHbt2mXKMfg1PXLkiNOrVy9zWgPddunSpc0yv/322wUNPb+QbQa7kPeCDol/9tlnzXrd5z9//vx0+5bZYzz49ZwwYUKmjtu/ei+EY+i5On78uNOvXz+nVKlS5njS40DXF3yqAP/nqO8H/fzRz4SdO3dmuF5ERpT+k53hCsjJtOZFf9FpjUdWazv86a9C7XOgv0b9z87qT/ss6HLayRc5gzZV9OzZM8PmUWTtPcVXDSKJPjsAAMBq9NkBwkg7Dnfq1Om8HYh1hJd7+QvANto3Svs+AZFE2AHCSDthuh1rz0VHnQC20jAf3HEZyG702QEAAFajzw4AALAaYQcAAFiNPjv/dw2WgwcPmhNWcR0TAAC8QU9poGfk1kEe/heTDkbYETFBJzMXIwQAADmPXs5Fz1p9LoQdEd+F3rSw9DT/oaIXftRTjLun50d4UM7Zh7KmnG3C8ez9ck5KSjKVFf4XbM0IYef/zpaqNOiEOuzoFZJ1nYSd8KGcsw9lTTnbhOPZnnL+qy4odFAGAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWC1XpHcAAMKt/JMLPFXIcTGOjKsX6b0A7EHNDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWC2iYWf48OESFRUVcKtSpYpv/unTp6Vnz55SpEgRyZ8/v3Ts2FEOHz4csI79+/dL27ZtJW/evFKsWDEZMGCAnD17NgLPBgAA5ES5Ir0D1atXl2XLlvnu58r1/3epX79+smDBApkzZ44UKlRIevXqJR06dJDPP//czE9NTTVBp0SJErJ27Vr55Zdf5P7775fY2Fh59tlnI/J8AABAzhLxsKPhRsNKsGPHjsm0adNk1qxZ0rRpUzNt+vTpUrVqVVm/fr00aNBAlixZIjt27DBhqXjx4lK7dm0ZNWqUDBw40NQa5c6dOwLPCAAA5CQR77OzZ88eKVWqlFSsWFE6depkmqXUpk2bJCUlRZo3b+5bVpu4ypYtK+vWrTP39W/NmjVN0HG1atVKkpKSZPv27RF4NgAAIKeJaM1O/fr1ZcaMGXL11VebJqgRI0bI3/72N9m2bZscOnTI1MwkJCQEPEaDjc5T+tc/6Ljz3XnnkpycbG4uDUdKw5XeQsVdVyjXCco5krx6TMfFOOIlcdGOJ8vZa7x6PHtNShjLObPrjGjYadOmje//tWrVMuGnXLly8u6770qePHnCtt0xY8aYYBVMm8W0o3OoLV26NOTrBOUcSV47psfVE0/yWjl7FeXs3XI+deqUN/rs+NNanKuuukr27t0rLVq0kDNnzsjRo0cDand0NJbbx0f/btiwIWAd7mitjPoBuQYNGiT9+/cPqNkpU6aMtGzZUgoWLBjSxKkvrj4X7TSN8KCcs49Xy7rG8MXitZqdUXXTPFfOXuPV49lrUsJYzm7LjKfCzokTJ+S7776Tzp07S506dUyhLF++3Aw5V7t37zZ9eho2bGju69/Ro0dLYmKiGXautEA1sFSrVu2c24mLizO3YLq9cBzw4VovKOdI8doxnZwaJV7ktXL2KsrZu+Wc2fVFNOw8/vjj0q5dO9N0dfDgQRk2bJjExMTIPffcY4aad+/e3dTAFC5c2ASY3r17m4CjI7GU1sRoqNFwNG7cONNPZ/DgwebcPBmFGQAAcOmJaNj56aefTLA5cuSIXH755dKkSRMzrFz/ryZMmCDR0dGmZkc7FOtIqylTpvger8Fo/vz50qNHDxOC8uXLJ126dJGRI0dG8FkBAICcJKJhZ/bs2eedHx8fL5MnTza3c9FaoYULF4Zh7wAAgA0ifp4dAACAcCLsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgtRwTdsaOHStRUVHSt29f37TTp09Lz549pUiRIpI/f37p2LGjHD58OOBx+/fvl7Zt20revHmlWLFiMmDAADl79mwEngEAAMiJckTY2bhxo/znP/+RWrVqBUzv16+fzJs3T+bMmSOrV6+WgwcPSocOHXzzU1NTTdA5c+aMrF27Vt544w2ZMWOGDB06NALPAgAA5EQRDzsnTpyQTp06yWuvvSaXXXaZb/qxY8dk2rRpMn78eGnatKnUqVNHpk+fbkLN+vXrzTJLliyRHTt2yMyZM6V27drSpk0bGTVqlEyePNkEIAAAgIiHHW2m0tqZ5s2bB0zftGmTpKSkBEyvUqWKlC1bVtatW2fu69+aNWtK8eLFfcu0atVKkpKSZPv27dn4LAAAQE6VK5Ibnz17tmzevNk0YwU7dOiQ5M6dWxISEgKma7DRee4y/kHHne/OO5fk5GRzc2k4Uhqu9BYq7rpCuU5QzpHk1WM6LsYRL4mLdjxZzl7j1ePZa1LCWM6ZXWfEws6BAwekT58+snTpUomPj8/WbY8ZM0ZGjBiRbro2i2lH51DT54jwo5yzj9fKelw98SSvlbNXUc7eLedTp07l7LCjzVSJiYly3XXXBXQ4XrNmjbz88suyePFi0+/m6NGjAbU7OhqrRIkS5v/6d8OGDQHrdUdructkZNCgQdK/f/+Amp0yZcpIy5YtpWDBgiFNnPritmjRQmJjY0O2XlDOkeLVY7rG8MXitZqdUXXTPFfOXuPV49lrUsJYzm7LTI4NO82aNZNvvvkmYFrXrl1Nv5yBAwea8KGFsnz5cjPkXO3evdsMNW/YsKG5r39Hjx5tQpMOO1daoBpYqlWrds5tx8XFmVsw3V44DvhwrReUc6R47ZhOTo0SL/JaOXsV5ezdcs7s+iIWdgoUKCA1atQImJYvXz5zTh13evfu3U0NTOHChU2A6d27twk4DRo0MPO1JkZDTefOnWXcuHGmn87gwYNNp+eMwgwAALj0RLSD8l+ZMGGCREdHm5od7VCsI62mTJnimx8TEyPz58+XHj16mBCkYalLly4ycuTIiO43AADIOXJU2Fm1alXAfe24rOfM0du5lCtXThYuXJgNewcAALwo4ufZAQAACCfCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsFqurD4wNTVV5s6dKzt37jT3q1evLrfddpvExMSEcv8AAACyP+zs3btX2rZtKz/99JNcffXVZtqYMWOkTJkysmDBArnyyisvbq8AAAAi2Yz16KOPSsWKFeXAgQOyefNmc9u/f79UqFDBzAMAAPB0zc7q1atl/fr1UrhwYd+0IkWKyNixY6Vx48ah3D8AAIDsr9mJi4uT48ePp5t+4sQJyZ0798XtEQAAQKTDzq233ioPPfSQfPHFF+I4jrlpTc8jjzxiOikDAAB4Ouy8+OKLphNyw4YNJT4+3ty0+apSpUoyadKk0O8lAABAdvbZSUhIkI8++kj27Nkju3btMtOqVq1qwg4AAIAV59lRlStXNjf3vDsAAABWNGPt27dP7rnnHunRo4f88ccfpp+OdlrWc+5s3bo19HsJAACQnWHn4YcfNmdO3rZtmzRt2lTOnDljmrWqVasmffv2zeq+AAAA5IxmLB2F9emnn0q5cuXMuXY2btwo1113nemzU79+/dDvJQAAQHbW7Og5dkqWLCmFChWSvHnzmg7LSv9mdP4dAAAAz3VQXrRokQk7aWlpsnz5ctOkdfTo0dDuHQAAQKTCTpcuXQL68LiioqIudp8AAAAiG3a0NgcAAMDaPjtvvvmmJCcnh35vAAAAckLY6dq1qxw7dizU+wIAAJAzwo5e+BMAAMDqDsrvvvuuFCxYMMN5999//8XsEwAAQOTDzrhx4yQmJibddB2NRdgBAACeDztffvmlFCtWLLR7AwAAkBP67AAAAFgddvSaWBk1YQEAAFjRjLVv377Q7wkAAEBOqdl59NFH5cUXX0w3/eWXX5a+ffuGYr8AAAAiF3bef/99ady4cbrpjRo1kvfeey8U+wUAABC5sHPkyBFzxfNget6d3377LdPreeWVV6RWrVrmcXpr2LChfPLJJ775p0+flp49e0qRIkUkf/780rFjRzl8+HDAOvbv3y9t27aVvHnzmtFhAwYMkLNnz2blaQEAAAtlKexUqlRJFi1alG66BpWKFStmej2lS5eWsWPHyqZNm8xQ9qZNm8rtt98u27dvN/P79esn8+bNkzlz5sjq1avl4MGD0qFDB9/jU1NTTdA5c+aMrF27Vt544w2ZMWOGDB06NCtPCwAAWChLHZT79+8vvXr1kl9//dUEFLV8+XJ54YUXZOLEiZleT7t27QLujx492tT2rF+/3gShadOmyaxZs3zbmD59ulStWtXMb9CggSxZskR27Nghy5Ytk+LFi0vt2rVl1KhRMnDgQBk+fLjkzp07K08PAABc6jU73bp1M8FGw8jNN99sbjNnzjRB5cEHH8zSjmgtzezZs+XkyZOmOUtre1JSUqR58+a+ZapUqSJly5aVdevWmfv6t2bNmibouFq1aiVJSUm+2iEAAHBpy/IZlHv06GFuWruTJ08e06cmK7755hsTbrR/jq7jww8/lGrVqsmWLVtMzUxCQkLA8hpsDh06ZP6vf/2DjjvfnXcuycnJ5ubScKQ0XOktVNx1hXKdoJwjyavHdFyMty5eHBfteLKcvcarx7PXpISxnDO7ziyHHe0EvGrVKvnuu+/k3nvvNdO0T412NL6Q4HP11VebYHPs2DEzkqtLly6mf044jRkzRkaMGJFuujaLaUfnUFu6dGnI1wnKOZK8dkyPqyee5LVy9irK2bvlfOrUqfCFnR9//FFat25tRkJpDUmLFi2kQIEC8txzz5n7U6dOzfS6tPZGOzyrOnXqyMaNG2XSpEly1113mY7HR48eDajd0dFYJUqUMP/Xvxs2bAhYnztay10mI4MGDTL9jvxrdsqUKSMtW7Y855Xcs5o49cXV8omNjQ3ZekE5R4pXj+kawxeL12p2RtVN81w5e41Xj2evSQljObstM2EJO3369JG6devK119/bYaFu+64444s99lxpaWlmcCkwUcLRTs+65BztXv3bhOwtNlL6V/t1JyYmOi7KKkWqAYWbQo7l7i4OHMLptsLxwEfrvWCco4Urx3TyalR4kVeK2evopy9W86ZXV+Wws6nn35qhnoHj3YqX768/Pzzz5lej9awtGnTxnQ6Pn78uBl5pU1jixcvNufx6d69u6mBKVy4sAkwvXv3NgFHR2IprYnRUNO5c2cZN26c6aczePBgc26ejMIMAAC49OTKau2Ljp4K9tNPP5nmrMzSGpn7779ffvnlFxNu9ASDGnS0qktNmDBBoqOjTc2O1vboSKspU6b4Hq8XI50/f77pKK0hKF++fKbPz8iRI7PytAAAgIWyFHa0RkXPp/Pqq6+a+1FRUXLixAkZNmyY3HLLLZlejw5dP5/4+HiZPHmyuZ3vCuwLFy68gL0HAACXkiyFHT3HjtayaBOSDhnX0Vh79uyRokWLyttvvx36vQQAAMjOsKNnN9bOyXoSwK1bt5paHe1f06lTJ3POHQAAgJwiy+fZyZUrl9x3332h3RsAAICcEHY+/vjj886/7bbbsro/AAAAkQ877du3D7ivHZQdx/H9P6ORWgAAAJ65EKgOPfe/6SUW9u7de84h6QAAAJ4KO8G0NgcAAMDKsPPDDz/IyZMnL+hkggAAADm6z06HDh3M3z///FPWr18vzZo1k8svvzzU+wYAABCZsKOXdnCvLN6uXTvp1q3bxe8JAABATgk706dPD/2eAAAA5JSwk5SUdN75eoVyAAAAz4adhISEDEdg6bl2OM8OAADwfNipWLGiJCYmypNPPimNGzcO/V4BAABEMuzs3LlTXnrpJRk9erR89dVXMm7cOKlQoUKo9gkAACCy59mJjY2V/v37y549e+SKK66QWrVqyWOPPSZHjx4N3Z4BAABE+qSChQsXlokTJ5raHT25YKVKlcx9AAAATzdjXXvttek6KGvn5OTkZFPD07dv31DtHwAAQOSveg4AAGBV2Bk2bFjo9wQAACAMOKkgAACwGicVBAAAVstS2FHvvfeeGY0FAABgZdjRMycXK1YstHsDAACQU8LOjh075MiRI5IvXz4pUaKE5M6dO7R7BgAAEMmTCjZr1kyqV69uLhOhgadmzZoyYcKEUOwTAABAZGt29u3bZ04imJKSIklJSXLw4EHZsGGDDBkyRM6ePSsDBgwI3R4CAABkd9gpV65cwP06depIu3bt5KqrrpKRI0cSdgAAgPf77GTk7rvvNk1bAAAAVoSdTZs2yc6dO83/q1WrJtddd525AQAAeDrsJCYmmlqcVatWSUJCgpl29OhRufnmm2X27Nly+eWXh3o/AQAAsm80Vu/eveX48eOyfft2+f33381t27ZtprPyo48+mrU9AQAAyCk1O4sWLZJly5ZJ1apVfdO0GWvy5MnSsmXLUO4fAABA9tfspKWlSWxsbLrpOk3nAQAAeDrsNG3aVPr06WPOr+P6+eefpV+/fuZkgwAAAJ4OOy+//LLpn1O+fHm58sorzU3PpKzTXnrppdDvJQAAQHb02dFOyQUKFJAyZcrI5s2bTb+dXbt2mXnaf6d58+ayceNGKV26dFb3BwAAIHJhRzsfL126VPLnzy9RUVHSokULc1N6mQi9XMRzzz0nZ86cCe1eAgAAZEczltbsaO2NNlf502Hn119/vbz++usyd+7crO4LAABAZMPOypUr5eTJk6Y2RwOPXgxUa3Lq1q1rmrE09Nxyyy2h30sAAIDsaMbSMyOvWLHC1O7oiKy4uDjZs2ePzJw5U+68886s7gMAAEDOOamgBp7ly5ebwKM1OVu2bJEqVaqEZ+8AAAAiMfS8aNGipoZHz5p87733yh9//HGx+wEAABD5mp0OHToE3C9YsKCsWbNG6tWrJzVr1vRN/+CDD0K3hwAAANkVdgoVKpTuvp5MEAAAwIqwM3369PDtCQAAQE7pswMAAOAVhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKtFNOyMGTNGrr/+eilQoIAUK1ZM2rdvL7t37w5Y5vTp09KzZ08pUqSI5M+fXzp27CiHDx8OWGb//v3Stm1byZs3r1nPgAED5OzZs9n8bAAAQE4U0bCzevVqE2TWr18vS5culZSUFGnZsqWcPHnSt0y/fv1k3rx5MmfOHLP8wYMHAy5bkZqaaoLOmTNnZO3atfLGG2/IjBkzZOjQoRF6VgAAwNNXPQ+lRYsWBdzXkKI1M5s2bZIbbrhBjh07JtOmTZNZs2ZJ06ZNfWdxrlq1qglIDRo0kCVLlsiOHTtk2bJlUrx4caldu7aMGjVKBg4cKMOHD5fcuXNH6NkBAICcIEf12dFwowoXLmz+aujR2p7mzZv7lqlSpYqULVtW1q1bZ+7rX70IqQYdV6tWrSQpKUm2b9+e7c8BAADkLBGt2fGXlpYmffv2lcaNG0uNGjXMtEOHDpmamYSEhIBlNdjoPHcZ/6DjznfnZSQ5OdncXBqMlAYrvYWKu65QrhOUcyR59ZiOi3HES+KiHU+Ws9d49Xj2mpQwlnNm15ljwo723dm2bZt89tln2dIxesSIEemma5OYdnIONe2PhPCjnLOP18p6XD3xJK+Vs1dRzt4t51OnTnkn7PTq1Uvmz58va9askdKlS/umlyhRwnQ8Pnr0aEDtjo7G0nnuMhs2bAhYnztay10m2KBBg6R///4BNTtlypQxnaMLFiwY0sSpL26LFi0kNjY2ZOsF5RwpXj2mawxfLF6r2RlVN81z5ew1Xj2evSYljOXstszk6LDjOI707t1bPvzwQ1m1apVUqFAhYH6dOnVMwSxfvtwMOVc6NF2Hmjds2NDc17+jR4+WxMRE07lZaaFqaKlWrVqG242LizO3YLqtcBzw4VovKOdI8doxnZwaJV7ktXL2KsrZu+Wc2fXlinTTlY60+uijj8y5dtw+NoUKFZI8efKYv927dze1MNppWQOMhiMNODoSS2ltjIaazp07y7hx48w6Bg8ebNadUaABAACXloiGnVdeecX8vemmmwKm6/DyBx54wPx/woQJEh0dbWp2tFOxjrSaMmWKb9mYmBjTBNajRw8TgvLlyyddunSRkSNHZvOzAQAAOVHEm7H+Snx8vEyePNnczqVcuXKycOHCEO8dAACwQY46zw4AAECoEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYLWIXggUAHBuNYYvluTUKM8U0Q9j20Z6F4AMUbMDAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNkwoCAEKi/JMLPFWScTGOjKsX6b1AdqBmBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgtYiGnTVr1ki7du2kVKlSEhUVJXPnzg2Y7ziODB06VEqWLCl58uSR5s2by549ewKW+f3336VTp05SsGBBSUhIkO7du8uJEyey+ZkAAICcKqJh5+TJk3LNNdfI5MmTM5w/btw4efHFF2Xq1KnyxRdfSL58+aRVq1Zy+vRp3zIadLZv3y5Lly6V+fPnmwD10EMPZeOzAAAAOVmuSG68TZs25pYRrdWZOHGiDB48WG6//XYz7c0335TixYubGqC7775bdu7cKYsWLZKNGzdK3bp1zTIvvfSS3HLLLfLvf//b1BgBAIBLW47ts7Nv3z45dOiQabpyFSpUSOrXry/r1q0z9/WvNl25QUfp8tHR0aYmCAAAIKI1O+ejQUdpTY4/ve/O07/FihULmJ8rVy4pXLiwb5mMJCcnm5srKSnJ/E1JSTG3UHHXFcp1gnKOJK8e03ExjnhJXLQT8BfhLWevHc9ekxLGz43MrjPHhp1wGjNmjIwYMSLd9CVLlkjevHlDvj3tT4Two5yzj9fKelw98aRRddMivQuXBK8dz161NAzlfOrUKW+HnRIlSpi/hw8fNqOxXHq/du3avmUSExMDHnf27FkzQst9fEYGDRok/fv3D6jZKVOmjLRs2dKM6gpl4tQXt0WLFhIbGxuy9YJyjhT3mB7yZbQkp0VxKIaxxkGDDuWcPeXMZ3R4hfO70G2Z8WzYqVChggksy5cv94UbfVLaF6dHjx7mfsOGDeXo0aOyadMmqVOnjpm2YsUKSUtLM317ziUuLs7cgumLEI5QEq71gnKOFA06yamEHcrZDnxGe7ecM7u+iIYdPR/O3r17Azolb9myxfS5KVu2rPTt21eeeeYZqVy5sgk/Q4YMMSOs2rdvb5avWrWqtG7dWh588EEzPF3TY69evcxILUZiAQCAiIedL7/8Um6++WbffbdpqUuXLjJjxgx54oknzLl49Lw5WoPTpEkTM9Q8Pj7e95i33nrLBJxmzZqZUVgdO3Y05+YBAACIeNi56aabzPl0zkXPqjxy5EhzOxetBZo1a1aY9hAAAHhdjj3PDgAAQCjk2A7KgO3KP7lAvEbPV+PVYdwALl3U7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACslivSOwCESo3hiyU5NYoCBQAEoGYHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVGI2FdMo/ucBTpRIX48i4epHeCwBATkXYAQBc0rx22oofxraN9C54Ds1YAADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqnEE5G3jt7JwAANiEmh0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAarkivQMAACDzyj+5wFPFFRfjyLh6kd0HanYAAIDVrAk7kydPlvLly0t8fLzUr19fNmzYEOldAgAAOYAVYeedd96R/v37y7Bhw2Tz5s1yzTXXSKtWrSQxMTHSuwYAACLMirAzfvx4efDBB6Vr165SrVo1mTp1quTNm1def/31SO8aAACIMM+HnTNnzsimTZukefPmvmnR0dHm/rp16yK6bwAAIPI8Pxrrt99+k9TUVClevHjAdL2/a9euDB+TnJxsbq5jx46Zv7///rukpKSEbN90XadOnZJcKdGSmhYVsvUiUK40R06dSqOcswFlnT0oZ8rZxuP5yJEjEhsbG9J1Hz9+3Px1HOf8+yCXoDFjxsiIESPSTa9QoUJE9gcX714KMdtQ1pSzTTie7ShnDT2FChWyN+wULVpUYmJi5PDhwwHT9X6JEiUyfMygQYNMh2ZXWlqaqdUpUqSIREWFrgYmKSlJypQpIwcOHJCCBQuGbL2gnCOFY5pytgnHs/fLWWt0NOiUKlXqvMt5Puzkzp1b6tSpI8uXL5f27dv7wove79WrV4aPiYuLMzd/CQkJYdtHfXEJO+FHOWcfyppytgnHs7fL+Xw1OtaEHaW1NF26dJG6detKvXr1ZOLEiXLy5EkzOgsAAFzarAg7d911l/z6668ydOhQOXTokNSuXVsWLVqUrtMyAAC49FgRdpQ2WZ2r2SpStKlMT3QY3GQGytmrOKYpZ5twPF865Rzl/NV4LQAAAA/z/EkFAQAAzoewAwAArEbYAQAAViPsAAAAqxF2wmjy5MlSvnx5iY+Pl/r168uGDRvCublL4jIf119/vRQoUECKFStmTiK5e/fugGVOnz4tPXv2NGfDzp8/v3Ts2DHd2bWReWPHjjVnFe/bty9lHAY///yz3HfffeZ4zZMnj9SsWVO+/PJL33wdP6Kn1ChZsqSZrxc43rNnTzh2xVp67cQhQ4aYywFpGV555ZUyatSogGspUc4Xbs2aNdKuXTtz5mL9jJg7d27A/MyUqV65oFOnTuZEg3pi3+7du8uJEyckLHQ0FkJv9uzZTu7cuZ3XX3/d2b59u/Pggw86CQkJzuHDhynuLGrVqpUzffp0Z9u2bc6WLVucW265xSlbtqxz4sQJ3zKPPPKIU6ZMGWf58uXOl19+6TRo0MBp1KgRZZ4FGzZscMqXL+/UqlXL6dOnD2UcYr///rtTrlw554EHHnC++OIL5/vvv3cWL17s7N2717fM2LFjnUKFCjlz5851vv76a+e2225zKlSo4Pz555+h3h1rjR492ilSpIgzf/58Z9++fc6cOXOc/PnzO5MmTfItQzlfuIULFzpPP/2088EHH2hqdD788MOA+Zkp09atWzvXXHONs379eufTTz91KlWq5Nxzzz1OOBB2wqRevXpOz549ffdTU1OdUqVKOWPGjAnXJi85iYmJ5k22evVqc//o0aNObGys+TBz7dy50yyzbt26CO6p9xw/ftypXLmys3TpUufGG2/0hR3KOHQGDhzoNGnS5Jzz09LSnBIlSjjPP/+8b5qWf1xcnPP222+HcE/s1rZtW6dbt24B0zp06OB06tTJ/J9yvnjBYSczZbpjxw7zuI0bN/qW+eSTT5yoqCjn559/dkKNZqwwOHPmjGzatMlU27mio6PN/XXr1oVjk5ekY8eOmb+FCxc2f7XMU1JSAsq9SpUqUrZsWcr9AmlTYNu2bQPKkjIOrY8//thc4ubvf/+7aZa99tpr5bXXXvPN37dvnzkjvP9roNcA0iZxPkcyr1GjRuZaid9++625//XXX8tnn30mbdq0oZzDJDPHrv7Vpit9D7h0ef2u/OKLL0K+T9acQTkn+e2330w7cfDlKvT+rl27IrZfNtGLvWo/ksaNG0uNGjXMNH1z6YVhgy/qquWu85A5s2fPls2bN8vGjRvTzaOMQ+f777+XV155xVzb76mnnjLl/eijj5pjWK/15x6zGX2OcDxn3pNPPmmuuq0/fGJiYsxn8+jRo01fEfeYppxDKzNlqn815PvLlSuX+fEajuObsAPP1jxs27bN/EJD6Bw4cED69OkjS5cuNR3rEd7Arr9qn332WXNfa3b0mJ46daoJOwiNd999V9566y2ZNWuWVK9eXbZs2WJ+KGnHWsr50kEzVhgULVrU/IIIHgWk90uUKBGOTV5S9Bpo8+fPl5UrV0rp0qV907VstQnx6NGjActT7pmnTYGJiYly3XXXmV9Zelu9erW8+OKL5v/6y4wyDg0dpVKtWrWAaVWrVpX9+/eb/7ufFXyOXJwBAwaY2p27777bjHbr3Lmz9OvXz4zupJzDIzPHrv7Vzxp/Z8+eNSO0wvE9SdgJA62GrlOnjmkn9v8Vp/cbNmwYjk1eErQfnAadDz/8UFasWGGGkvrTMo+NjQ0odx2arl8elHvmNGvWTL755hvz69e9ae2DVvm7/6eMQ0ObYINPnaD9SsqVK2f+r8e3fuj7H8/aHKP9GTieM+/UqVOmH4g//TGqn8mUc3hk5tjVv/rDVH9gufRzXV8X7dsTciHv8gzf0HPteT5jxgzT6/yhhx4yQ88PHTpECWVRjx49zFDGVatWOb/88ovvdurUqYCh5zocfcWKFWboecOGDc0NWec/GosyDu3Q/ly5cpmh0Xv27HHeeustJ2/evM7MmTMDhu/q58ZHH33kbN261bn99tsZen6BunTp4lxxxRW+oec6VLpo0aLOE088QTlf5IjNr776ytw0SowfP978/8cff8z0satDz6+99lpz6oXPPvvMjABl6LkHvfTSS+aLV8+3o0PR9VwCyDp9Q2V003PvuPSN9K9//cu57LLLzBfHHXfcYQIRQhd2KOPQmTdvnlOjRg3zw6hKlSrOq6++GjBfh/AOGTLEKV68uFmmWbNmzu7du0O4B/ZLSkoyx69+FsfHxzsVK1Y054dJTk72LUM5X7iVK1dm+Hms4TKzZXrkyBETbvS8RwULFnS6du1qQlQ4ROk/oa8vAgAAyBnoswMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAyDbPfDAAxIVFeW7FSlSRFq3bi1bt27l1QAQcoQdABGh4eaXX34xN71goF5V/dZbb+XVABByhB0AEREXF2eujKy32rVry5NPPikHDhyQX3/91czXq683bdpU8uTJY2p+HnroITlx4oTv8bp8qVKlJHfu3HLFFVfIwIEDfVeyXrVqlakxWrBggdSqVUvi4+OlQYMGsm3bNt/jjxw5Ivfcc495bN68eaVmzZry9ttvB+yjrm/cuHFSqVIls79ly5aV0aNHm3n+NVPBN90+gJyDsAMg4jTEzJw504QKDTYnT56UVq1ayWWXXSYbN26UOXPmyLJly6RXr16+x7Rs2VLmz58ve/fulf/+97/y6quvmnX4GzBggLzwwgtmHZdffrm0a9dOUlJSzLzTp09LnTp1TCDSEKRhqnPnzrJhwwbf4wcNGiRjx46VIUOGyI4dO2TWrFlSvHhxM8+tldKbev/99333GzVqlE0lByBTwnJ5UQA4D70yckxMjJMvXz5z04+ikiVLOps2bTLz9erfeuX6EydO+B6zYMECJzo62jl06FC69X3//ffm8a+//nrAFZlnz54dcIXlPHnyOO+8884596tt27bOY4895rtatl6t+bXXXvvL11K3pdsEkDNRswMgIm6++WbZsmWLuWltitbktGnTRn788UfZuXOnXHPNNZIvXz7f8o0bNzbNSrt37/ZNe/bZZ00TVMWKFaVjx45y//33B2yjYcOGvv8XLlxYrr76arNulZqaKqNGjTLNVzovf/78snjxYtm/f7+Zr8slJydLs2bNsqE0AIRTrrCuHQDOQYOMNlu5tCmqUKFC8tprr2W6zB555BHp0KGDbNq0Sfr27Wv+ryEqM55//nmZNGmSTJw40QQe3R9dx5kzZ8x87SsEwA7U7ADIEbRjb3R0tPz5559StWpV+frrr03fHdfnn39u5mvtjEtrZKpUqSKdOnWSJk2amH4z/tavX+/7/x9//CHffvutWbe7vttvv13uu+8+U4uktUM631W5cmUTeHSkGABvI+wAiAhtIjp06JC5aZNR7969TUdl7USs4UVHUHXp0sV0Hl65cqWZrx2I3Q7CU6ZMke3bt8sPP/xgOiYvXbpUrr322oBtjBw50oQVXYee26do0aLSvn17X5jRx6xdu9Zs/+GHH5bDhw/7Hqvb1xFeTzzxhLz55pvy3XffmfA0bdq0bC4pABeLZiwAEbFo0SIpWbKk+X+BAgVMDY2OurrpppvMNO0/06dPH7n++utNvxztkzN+/Hjf43UU1bBhw+T48eNSpkwZeeqpp6Rbt24B29CRVLqOPXv2mOHt8+bNM0PV1eDBg+X77783fYV0/ToaS4PQsWPHfI/XUVh6/p+hQ4fKwYMHzf5q0xkAb4nSXsqR3gkACCU9z4323dGmq4SEBAoXuMTRjAUAAKxG2AEAAFajGQsAAFiNmh0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAIDb7f0g9G9FzdQ6RAAAAAElFTkSuQmCC", "text/plain": [ "

" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "df[\"age\"].hist()\n", "plt.title(\"Распределение возраста пациентов\")\n", "plt.xlabel(\"Возраст\")\n", "plt.ylabel(\"Количество\")\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 12, "id": "a313c60d-5e66-485b-8d4b-bf0ade4de2da", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "

" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "sns.boxplot(x=\"gender\", y=\"severity_score\", data=df)\n", "plt.title(\"Серьёзность случаев по полу\")\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 13, "id": "be1b82e2-6630-4332-9292-1fba6c6bef27", "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAigAAAHNCAYAAAA0bIApAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjksIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvJkbTWQAAAAlwSFlzAAAPYQAAD2EBqD+naQAAPmxJREFUeJzt3QucTfX+//HPjMu4ZUSYnIhKIaQodDldyCQ6Cd2OUEQJhUI6onQhFV3RFYUuTqlwUpo6VORaHbmlIgpDaUx0jNv+P97f/2/ts/eYYQbTfPfM6/l47Meevdfae6/Lnr3e63tbcaFQKGQAAAAeic/vBQAAAMiMgAIAALxDQAEAAN4hoAAAAO8QUAAAgHcIKAAAwDsEFAAA4B0CCgAA8A4BBQAAeIeAAgAAvENAwVETFxeXo9u///1vtjoA4KCKHnwykHOvvvpq1ONXXnnFZs+efcDztWvXZrMCAA4qjosFIq/06tXLnn32WeN6lACA3KKKB/lmy5Yt1rVrV6tcubKVKFHCzjjjDJs4cWLUPOvWrTtoddFFF12Uo8+aNGmSnXPOOVaqVCk79thj7a9//at9+OGHblr16tUP+hmaHrksjz32mI0ePdpOPPFEK1mypF144YX2zTffRH3ef/7zH7vxxhvtpJNOcuuWlJRkXbp0sV9//TVqvvvuu8+95y+//BL1/OLFi93zEyZMOOh6afrBll3vH+nLL7+0li1bWtmyZa1MmTLWrFkz++KLLw65/YJ1j1ye33//3Ro2bGg1atSwTZs2hZ/fuXOn3XnnnVa1alVLSEiw0047zW2zrIJqdssfuV+DebQMgf3791v9+vUPWCbtK233SKpSzKpqccGCBXbZZZdZYmKi+15oP37++ecHLOPPP//svqdVqlRx66P17dGjh+3evfuQ2z9y+bRckc/re6j1/PTTT8Of1blzZzvuuONsz549ByxHixYt3LY8FK3X5Zdf7t6/dOnSbjs9+eSTB8yX3fIG21nbQ/+TWdFyJCcnZ7t/ROsWuR+1vYYMGeK+M9rmWrYLLrjAPvnkkwPeX/v3iSeesNNPP939/+g34pZbbrHffvstar7I/934+Hj3f3bttdfa+vXrD7md4D+qeJAv/vvf/7ofr++++86VtOhHf+rUqe5HPC0tze64446o+a+//nr3oxtp0KBBOfqs+++/3x2ozz33XBs2bJgVL17c/Yh//PHH7kdfP4Q7duxw865cudIefvhhu+eee8JVUTqQZ6660sG5Z8+etmvXLvfjf8kll9iyZcvcD6moauuHH36wm266yf1oLl++3J5//nl3r0CgH9SjSeulbRjQ+uggGkmfrQOCwsmAAQOsWLFi9txzz7n9MGfOHGvcuHGOP08H0Hbt2rkDgQ7qxx9/vHteIeRvf/ubO+jooN6gQQP74IMPrH///u5Ar2CXFT2vA7M89NBDh/x8VRtqex8u7XsFNR0shw4d6g5u48ePd/tRgUFhVjZu3Oj+1neye/fuVqtWLbce//znP+2PP/5wQTeyCjNY9n/84x/h5/S9C2gdg23w008/ue+OvtcbNmywcuXKWceOHd33S9usdevW4ddt3rzZLbOW9WD0vdPrtD/0P6Tvnr7TM2bMOOB/Sq666ipr27at+1vrre9oQMvSrVs3F77r1q0bfn7RokX27bff2uDBg3O1zdPT0+3FF190/8t6X/0PvfTSSy7oLFy40H1XAgojCj76/7n99ttt7dq19swzz7iAre+bvrsBfae1bxRqtKz6f9Z+iwx+iFGq4gHyQs+ePXXKnOW0J554wk2bNGlS+Lndu3eHmjZtGipTpkwoPT3dPbd27Vo336OPPnrAe5x++umhCy+88KDLsGbNmlB8fHzoqquuCu3bty9q2v79+w+Y/5NPPnGfp/vMgmUpWbJk6Keffgo/v2DBAvd83759w8/98ccfB7z+tddec/PNnTs3/NzQoUPdc1u3bo2ad9GiRe758ePHH3T9NF3zaf5Iej89r/cPtGnTJlS8ePHQ999/H35u48aNoWOOOSb017/+9aCfE6y7Pk/brUOHDqFSpUq5dY/0zjvvuPkefPDBqOfbt28fiouLC3333XdRz7/wwgtu/h9//DH8nPZp5H4N1lHLILt27QpVq1Yt1LJlywO2UY0aNUKdOnU66D7V8tesWTOUnJwc9R3QPtPrL7300vBzei99fzJv3+B9Msu87JE6d+4cOvHEE6Oee/75592yLVy40D3Wd/SEE04IXXvttVHzjRo1ym2/H374IZSdvXv3uuXXZ/z2228HXdY9e/a4z73//vuz3c5paWmhEiVKhAYOHBj12ttvvz1UunTp0I4dO9zjiRMnutdlXrbM20LLl5GRETWPlrNy5cqhLl26hJ/79NNP3ftNnjw5at5Zs2Yd8LzWVds10t///nf33UTso4oH+eJf//qXO7vT2VRAZ0U6W9LZv87oj4Z33nnHnVmpaFlnyZEOtxSjTZs29pe//CX8WGfYKn3QOgVU9RNQKYuqcJo0aeIeL1269ID33LZtm5snuG3fvt2Opn379rkqLS27qp0COtP++9//bp999pk7w80JlYZMnjzZ3nzzzXBJQ0DboEiRIm4/RlKVj0pX3n///ajnVewvqjrJKbVrUlVZVqUJlSpVciUTB/PVV1/ZmjVr3HrrfYJtrqopVXnNnTvXfWd00/fniiuusEaNGh3wPofz/dF7Bp+n5VBpifZBUFqn72iHDh3svffecyUMAW1vlcRElpJlptIFlTT06dPHlcYcbFlzst1VDXPllVfaa6+9Fq6e0/fojTfecN8jVdEE21wOtd31vVDpZbAd9J3fu3ev27aR/xMqSdVnX3rppVH/EyrtUmlm5iqhjIwMN11VxipBUkmT9iNiHwEF+eLHH3+0mjVrHhAagh9qTT8avv/+e/cZderUsaNFy53ZqaeeGlUHrx9fFamrykdhpWLFiuGDS1bhQ3X6mie4NW/e3I6mrVu3uiqJrNowaJvrgKFqhkNRldDjjz/u/s7cHiDYb2qrccwxxxzwGcH0SKo6yaoaLTvadqqC69evX7g6LZIO4gq3r7/+ujtgZRX2FE6C9h6R21w3VUHogKfXaJsptEVWbxwpbePgs84880z3/Xzrrbei1r9Tp06uCnTatGnu8erVq23JkiWuyuVg9F6Sk+XN6XbXsqgaL6gu+eijjyw1NTVqWbQeaieiqlRt2yBQZNWORm3M1CZG81eoUMFth5kzZ0btI72HHiv4ZN4/OnnRfo2kfa1p+j6oylZtn7QfEftogwLkgWuuucbmzZvnShtUt64DgUKAGmXqPjMdpNQ2JKA6frVx8Y3az6idhdoh9O3b161P0HbkcKhthbZNcDZ+KI888ogLnNqumRsci9oOqY1CZMlcZsH2f/TRR6PaPUTSMilkHm06iKrBtugg/PLLL7ttqBKsevXquecVplVaoPkUEHSvkgd9p44WbXdRKebBqH1IsMxqb6N7vSYyQGv6008/7b6vCuqR1NA2oNeqjZlKX7T/FEBUqjJ8+PBwuAr2j6ap1CgrCiORFEr0fkEpjr4jF198sWtoHlmSidhDQEG+UA8Y9XTRj1FkKcqqVavC04+Gk08+2X3GihUrsj0Y5VZwBh5JgSLo7aOShZSUFHdGqaqlg70uoB//yAN95iL6I6UfdfVU0dl4Ztrm2gc68zwU9URSCFAjRB1IFVIiG4lqv+ksW9UTkaUo2e1X7Zecjoujz1SjUh3Q9N5ZBRRtw/nz57v3DQ7CX3/9td11111R3wlRIDxYSZW2mebJ3EPrSKjkIPIz1aC4fPnyrgGoSqcCCiYqJVLvqClTplirVq1cr5yDCdZLy3uoEjhtHznUtleAUFWYGqzqwK8qLzVw1fORbr75ZtfYVp8dVB+pWi+SGharevHtt9+OqnLKXFWn9dB36LzzzstRwFAVWeT6qpRQJWla1oMFVfiPKh7kC/Vc0AFE9dkB1UfrTExnr5FnXkdCZ2s6+KqXS+aSi8Mdn0U/fOrJEVAPBPUKUq8QCX68M7+/ehfkFy2TzjTffffdqKooFdfrAHj++edHleBkRz0mRNU4OmDprDjorh3sV7VT0AE3knqu6KAUbKOgukOlHeo5kxMKfDpbv/XWWw86n/a3qjl00NJNpRGR9FgHQXV9DnpvRVLVTvA++v5Mnz7dnY1ndjTG99HBXN97VStF0oFV20vVhOoNdsMNNxzyvc466yxXjajvWVCFk92y6v8usu3Lwag6R6FbPWu0vbJbFgUtBe1gu2cOVFn9X+j/RoEykkqK9B164IEHDvgMbavM65aZqsck8zZF7KEEBflC3QJ1xqgiX9Wvq/RBZ1g6YOkHNnMbhsN1yimnuC6f+rHTwVVneWoYqCoKHWR1Nn4476kDurrx6kdQy6v6dHXdFR3o9UM9cuRIVw+vBrU6iKsBY3568MEHXSNCLfttt91mRYsWdftA66BlPZx9qHCjwKAzZ5XQqEGpite1zRWENI6G1l3BSI03g7P8sWPHum2v12RuUJsdvY+K/YOGlodLwUNtFBSWNM6GurJqHyl0qgGm9p9Ciai9iz5XgVnrqwO6SjXUkFPVMrkt6VJD3MgqHpU+qRG1uvtmLr1R1Y8+R5+hEpScrJe2q/aBSgu1XgohKr1SF3N1XVbQuvfee23WrFk2bty4HDX0VRsTBT4ti9ZfQehwqPuzSk+0rlof/T9oGVQSFxkUta0VhvT9UENiBWs1oFcJpJZBpWjt27cPz68AF2xT7UOFY+1DGsoWAPndjQiFs5uxpKamhm666abQcccd57q/1qtX74ButUfazTjw8ssvh84888xQQkJC6Nhjj3Wvmz179mF1M9ayPP7446GqVau697vgggtCX3/9ddS86oasrs3lypULJSYmhq6++mrXpTdz198/s5uxLF261HWvVVdudcW8+OKLQ/PmzTvE1ovuZhxp9erVritqZBfr33//3T2uUqVKqFixYq5Lr7ZZZFfXc845x22TVatWHfBZ2XUzbtCgQdR7ZLdMOd2nX375Zaht27ahChUquP2oLqvXXHNNKCUlJWo+dYFWd+OKFSu6+U466ST33c7cZTarZY+k7rBajuCmfXDWWWeFXn311Sznf/PNN9183bt3D+XGZ5995rpKq/u4ugPXr18/9PTTT7tpjzzySOjss88+oAtvVt2MI40cOdJNe/jhh3O8HJm3hfadXq/trO2o/8cZM2Zk2f066ILdsGFD161f66LfhwEDBrj/o4BeF7lN9VvSokWL0Pz583O8nPAXQ90DOaQSARWhq3FlZJsGIC+o1ElVTOr2HFSt5ReVWqi9kf4HqlWrlq/LgsKDNigA4KEXXnjBNSpVlVx+UpsRjfiqqhfCCf5MtEEBAI9oXA/1cNP4ICq5ONqXRchNexkNGKd2ObqsgEp0gD8TAQUAPKIePOrJpmsZqTFzflFvJnUxViNddS1Xl2jgz0QbFAAA4B3aoAAAAO8QUAAAgHdisg2KRgTVsNcazCu/GpABAIDc9wrTpTA0UGbmi8UWiICicJKT64YAAAD/6FIXJ5xwQsELKMEw6FrBnFw/BAAA5L/09HRXwJCTy5nEZEAJqnUUTggoAADElpw0z6CRLAAA8A4BBQAAxHZA2bdvn7tUty6YVrJkSXfpdF3GXq1yA/p7yJAh7jLfmqd58+buMtmRtm3bZh06dHDVMxqlUCMmRl5uGwAAFG65CiiPPPKIjR071p555hlbuXKlezxy5Eh7+umnw/Po8VNPPWXjxo2zBQsWWOnSpS05Odl27doVnkfhZPny5TZ79mybMWOGu1pn9+7dj+6aAQCAwjHUfevWra1y5cruypaBdu3auZKSSZMmudIT9W2+8847w5ej3759u3vNhAkT7LrrrnPBpk6dOrZo0SJr1KiRm2fWrFl2+eWX208//eRen5NWwImJie69aSQLAEBsyM3xO1clKOeee66lpKTYt99+6x5//fXX9tlnn1nLli3d47Vr19rmzZtdtU5AC9K4cWObP3++e6x7VesE4UQ0vwZsUYlLVjIyMtxKRd4AAEDBlatuxnfffbcLB7Vq1bIiRYq4NikPPfSQq7IRhRNRiUkkPQ6m6b5SpUrRC1G0qJUvXz48T2bDhw+3+++/P3drBgAAYlauSlDefPNNmzx5sk2ZMsWWLl1qEydOtMcee8zd56VBgwa54qDgpgHaAABAwZWrEpT+/fu7UhS1JZF69erZjz/+6Eo4OnfubElJSe751NRU14snoMcNGjRwf2ueLVu2RL3v3r17Xc+e4PWZJSQkuBsAACgcclWC8scffxxwcR9V9ejifaLuxwoZaqcSUJWQ2pY0bdrUPdZ9WlqaLVmyJDzPxx9/7N5DbVUAAAByVYJyxRVXuDYn1apVs9NPP92+/PJLGzVqlHXp0iU8dG2fPn3swQcftJo1a7rAonFT1DOnTZs2bp7atWvbZZddZt26dXNdkffs2WO9evVypTI56cEDAAAKvlwFFI13osBx2223uWoaBYpbbrnFDcwWGDBggO3cudONa6KSkvPPP991Iy5RokR4HrVjUShp1qyZK5FRV2WNnQIAAJDrcVB8wTgoAADEnjwbBwUAAMC7Kp5YVv3umXn+GetGtMrzzwAAoDCgBAUAAHiHgAIAALxDQAEAAN4hoAAAAO8QUAAAgHcIKAAAwDsEFAAA4B0CCgAA8A4BBQAAeIeAAgAAvENAAQAA3iGgAAAA7xBQAACAdwgoAADAOwQUAADgHQIKAADwDgEFAAB4h4ACAAC8Q0ABAADeIaAAAADvEFAAAIB3CCgAAMA7BBQAAOAdAgoAAPAOAQUAAHiHgAIAALxDQAEAAN4hoAAAAO8QUAAAgHcIKAAAwDsEFAAA4B0CCgAAiO2AUr16dYuLizvg1rNnTzd9165d7u8KFSpYmTJlrF27dpaamhr1HuvXr7dWrVpZqVKlrFKlSta/f3/bu3fv0V0rAABQeALKokWLbNOmTeHb7Nmz3fNXX321u+/bt69Nnz7dpk6danPmzLGNGzda27Ztw6/ft2+fCye7d++2efPm2cSJE23ChAk2ZMiQo71eAAAghsWFQqHQ4b64T58+NmPGDFuzZo2lp6dbxYoVbcqUKda+fXs3fdWqVVa7dm2bP3++NWnSxN5//31r3bq1Cy6VK1d284wbN84GDhxoW7duteLFi+foc/VZiYmJtn37ditbtmyOXlP97pmW19aNaJXnnwEAQKzKzfH7sNugqBRk0qRJ1qVLF1fNs2TJEtuzZ481b948PE+tWrWsWrVqLqCI7uvVqxcOJ5KcnOwWePny5dl+VkZGhpsn8gYAAAquww4o77zzjqWlpdmNN97oHm/evNmVgJQrVy5qPoURTQvmiQwnwfRgWnaGDx/uEldwq1q16uEuNgAAKMgB5aWXXrKWLVtalSpVLK8NGjTIFQcFtw0bNuT5ZwIAgPxT9HBe9OOPP9pHH31kb7/9dvi5pKQkV+2jUpXIUhT14tG0YJ6FCxdGvVfQyyeYJysJCQnuBgAACofDKkEZP3686yKsHjmBhg0bWrFixSwlJSX83OrVq1234qZNm7rHul+2bJlt2bIlPI96AqmhTJ06dY5sTQAAQOEtQdm/f78LKJ07d7aiRf/3crUN6dq1q/Xr18/Kly/vQkfv3r1dKFEPHmnRooULIh07drSRI0e6dieDBw92Y6dQQgIAAA47oKhqR6Ui6r2T2ejRoy0+Pt4N0KaeN+qhM2bMmPD0IkWKuG7JPXr0cMGldOnSLugMGzYst4sBAAAKsCMaByW/MA4KAACx508ZBwUAACCvEFAAAIB3CCgAAMA7BBQAAOAdAgoAAPAOAQUAAHiHgAIAALxDQAEAAN4hoAAAAO8QUAAAgHcIKAAAwDsEFAAA4B0CCgAA8A4BBQAAeIeAAgAAvENAAQAA3iGgAAAA7xBQAACAdwgoAADAOwQUAADgHQIKAADwDgEFAAB4h4ACAAC8Q0ABAADeIaAAAADvEFAAAIB3CCgAAMA7BBQAAOAdAgoAAPAOAQUAAHiHgAIAALxDQAEAAN4hoAAAAO8QUAAAQOwHlJ9//tluuOEGq1ChgpUsWdLq1atnixcvDk8PhUI2ZMgQO/7449305s2b25o1a6LeY9u2bdahQwcrW7aslStXzrp27Wo7duw4OmsEAAAKV0D57bff7LzzzrNixYrZ+++/bytWrLDHH3/cjj322PA8I0eOtKeeesrGjRtnCxYssNKlS1tycrLt2rUrPI/CyfLly2327Nk2Y8YMmzt3rnXv3v3orhkAAIhZcSEVeeTQ3XffbZ9//rl9+umnWU7XW1WpUsXuvPNOu+uuu9xz27dvt8qVK9uECRPsuuuus5UrV1qdOnVs0aJF1qhRIzfPrFmz7PLLL7effvrJvf5Q0tPTLTEx0b23SmFyovrdMy2vrRvRKs8/AwCAWJWb43euSlDee+89Fyquvvpqq1Spkp155pn2wgsvhKevXbvWNm/e7Kp1AlqQxo0b2/z5891j3ataJwgnovnj4+NdiUtWMjIy3EpF3gAAQMGVq4Dyww8/2NixY61mzZr2wQcfWI8ePez222+3iRMnuukKJ6ISk0h6HEzTvcJNpKJFi1r58uXD82Q2fPhwF3SCW9WqVXO3lgAAoOAGlP3799tZZ51lDz/8sCs9UbuRbt26ufYmeWnQoEGuOCi4bdiwIU8/DwAAxFBAUc8ctR+JVLt2bVu/fr37Oykpyd2npqZGzaPHwTTdb9myJWr63r17Xc+eYJ7MEhISXF1V5A0AABRcuQoo6sGzevXqqOe+/fZbO/HEE93fNWrUcCEjJSUlPF3tRdS2pGnTpu6x7tPS0mzJkiXheT7++GNXOqO2KgAAAEVzswn69u1r5557rqviueaaa2zhwoX2/PPPu5vExcVZnz597MEHH3TtVBRY7r33Xtczp02bNuESl8suuyxcNbRnzx7r1auX6+GTkx48AACg4MtVQDn77LNt2rRprk3IsGHDXAB54okn3LgmgQEDBtjOnTtd+xSVlJx//vmuG3GJEiXC80yePNmFkmbNmrneO+3atXNjpwAAAOR6HBRfMA4KAACxJ8/GQQEAAPgzEFAAAIB3CCgAAMA7BBQAAOAdAgoAAPAOAQUAAHiHgAIAALxDQAEAAN4hoAAAAO8QUAAAgHcIKAAAwDsEFAAA4B0CCgAA8A4BBQAAeIeAAgAAvENAAQAA3iGgAAAA7xBQAACAdwgoAADAOwQUAADgHQIKAADwDgEFAAB4h4ACAAC8Q0ABAADeIaAAAADvEFAAAIB3CCgAAMA7BBQAAOAdAgoAAPAOAQUAAHiHgAIAALxDQAEAAN4hoAAAAO8QUAAAQGwHlPvuu8/i4uKibrVq1QpP37Vrl/Xs2dMqVKhgZcqUsXbt2llqamrUe6xfv95atWplpUqVskqVKln//v1t7969R2+NAABAzCua2xecfvrp9tFHH/3vDYr+7y369u1rM2fOtKlTp1piYqL16tXL2rZta59//rmbvm/fPhdOkpKSbN68ebZp0ybr1KmTFStWzB5++OGjtU4AAKCwBRQFEgWMzLZv324vvfSSTZkyxS655BL33Pjx46127dr2xRdfWJMmTezDDz+0FStWuIBTuXJla9CggT3wwAM2cOBAVzpTvHjxo7NWAACgcLVBWbNmjVWpUsVOOukk69Chg6uykSVLltiePXusefPm4XlV/VOtWjWbP3++e6z7evXquXASSE5OtvT0dFu+fHm2n5mRkeHmibwBAICCK1cBpXHjxjZhwgSbNWuWjR071tauXWsXXHCB/f7777Z582ZXAlKuXLmo1yiMaJroPjKcBNODadkZPny4qzIKblWrVs3NYgMAgIJcxdOyZcvw3/Xr13eB5cQTT7Q333zTSpYsaXll0KBB1q9fv/BjlaAQUgAAKLiOqJuxSktOPfVU++6771y7lN27d1taWlrUPOrFE7RZ0X3mXj3B46zatQQSEhKsbNmyUTcAAFBwHVFA2bFjh33//fd2/PHHW8OGDV1vnJSUlPD01atXuzYqTZs2dY91v2zZMtuyZUt4ntmzZ7vAUadOnSNZFAAAUFireO666y674oorXLXOxo0bbejQoVakSBG7/vrrXduQrl27uqqY8uXLu9DRu3dvF0rUg0datGjhgkjHjh1t5MiRrt3J4MGD3dgpKiUBAADIdUD56aefXBj59ddfrWLFinb++ee7LsT6W0aPHm3x8fFugDb1vFEPnTFjxoRfrzAzY8YM69GjhwsupUuXts6dO9uwYcPYGwAAICwuFAqFLMaokaxKbDT2Sk7bo1S/e2aeL9e6Ea3y/DMAAIhVuTl+cy0eAADgHQIKAADwDgEFAAB4h4ACAAC8Q0ABAADeIaAAAADvEFAAAIB3CCgAAMA7BBQAAOAdAgoAAPAOAQUAAHiHgAIAALxDQAEAAN4hoAAAAO8QUAAAgHcIKAAAwDsEFAAA4B0CCgAA8A4BBQAAeIeAAgAAvENAAQAA3iGgAAAA7xBQAACAdwgoAADAOwQUAADgHQIKAADwDgEFAAB4h4ACAAC8Q0ABAADeIaAAAADvEFAAAIB3CCgAAMA7BBQAAOAdAgoAAChYAWXEiBEWFxdnffr0CT+3a9cu69mzp1WoUMHKlClj7dq1s9TU1KjXrV+/3lq1amWlSpWySpUqWf/+/W3v3r1HsigAAKAAOeyAsmjRInvuueesfv36Uc/37dvXpk+fblOnTrU5c+bYxo0brW3btuHp+/btc+Fk9+7dNm/ePJs4caJNmDDBhgwZcmRrAgAACndA2bFjh3Xo0MFeeOEFO/bYY8PPb9++3V566SUbNWqUXXLJJdawYUMbP368CyJffPGFm+fDDz+0FStW2KRJk6xBgwbWsmVLe+CBB+zZZ591oQUAAOCwAoqqcFQK0rx586jnlyxZYnv27Il6vlatWlatWjWbP3++e6z7evXqWeXKlcPzJCcnW3p6ui1fvjzLz8vIyHDTI28AAKDgKprbF7z++uu2dOlSV8WT2ebNm6148eJWrly5qOcVRjQtmCcynATTg2lZGT58uN1///25XVQAAFAYSlA2bNhgd9xxh02ePNlKlChhf5ZBgwa56qPgpuUAAAAFV64CiqpwtmzZYmeddZYVLVrU3dQQ9qmnnnJ/qyRE7UjS0tKiXqdePElJSe5v3Wfu1RM8DubJLCEhwcqWLRt1AwAABVeuAkqzZs1s2bJl9tVXX4VvjRo1cg1mg7+LFStmKSkp4desXr3adStu2rSpe6x7vYeCTmD27NkudNSpU+dorhsAACgMbVCOOeYYq1u3btRzpUuXdmOeBM937drV+vXrZ+XLl3eho3fv3i6UNGnSxE1v0aKFCyIdO3a0kSNHunYngwcPdg1vVVICAACQ60ayhzJ69GiLj493A7Sp94166IwZMyY8vUiRIjZjxgzr0aOHCy4KOJ07d7Zhw4axNwAAgBMXCoVCFmPUzTgxMdE1mM1pe5Tqd8/M8+VaN6JVnn8GAACxKjfHb67FAwAAvENAAQAA3iGgAAAA7xBQAACAdwgoAADAOwQUAADgHQIKAADwDgEFAAB4h4ACAAC8Q0ABAADeIaAAAADvEFAAAIB3CCgAAMA7BBQAAOAdAgoAAPAOAQUAAHiHgAIAALxDQAEAAN4hoAAAAO8QUAAAgHcIKAAAwDsEFAAA4B0CCgAA8A4BBQAAeIeAAgAAvENAAQAA3iGgAAAA7xBQAACAdwgoAADAOwQUAADgHQIKAADwDgEFAAB4h4ACAAC8Q0ABAACxHVDGjh1r9evXt7Jly7pb06ZN7f333w9P37Vrl/Xs2dMqVKhgZcqUsXbt2llqamrUe6xfv95atWplpUqVskqVKln//v1t7969R2+NAABA4QooJ5xwgo0YMcKWLFliixcvtksuucSuvPJKW758uZvet29fmz59uk2dOtXmzJljGzdutLZt24Zfv2/fPhdOdu/ebfPmzbOJEyfahAkTbMiQIUd/zQAAQMyKC4VCoSN5g/Lly9ujjz5q7du3t4oVK9qUKVPc37Jq1SqrXbu2zZ8/35o0aeJKW1q3bu2CS+XKld0848aNs4EDB9rWrVutePHiOfrM9PR0S0xMtO3bt7uSnJyofvdMy2vrRrTK888AACBW5eb4fdhtUFQa8vrrr9vOnTtdVY9KVfbs2WPNmzcPz1OrVi2rVq2aCyii+3r16oXDiSQnJ7sFDkphspKRkeHmibwBAICCK9cBZdmyZa59SUJCgt166602bdo0q1Onjm3evNmVgJQrVy5qfoURTRPdR4aTYHowLTvDhw93iSu4Va1aNbeLDQAACnJAOe200+yrr76yBQsWWI8ePaxz5862YsUKy0uDBg1yxUHBbcOGDXn6eQAAIH8Vze0LVEpyyimnuL8bNmxoixYtsieffNKuvfZa1/g1LS0tqhRFvXiSkpLc37pfuHBh1PsFvXyCebKi0hrdAABA4XDE46Ds37/ftRFRWClWrJilpKSEp61evdp1K1YbFdG9qoi2bNkSnmf27NmuoYyqiQAAAHJdgqKqlpYtW7qGr7///rvrsfPvf//bPvjgA9c2pGvXrtavXz/Xs0eho3fv3i6UqAePtGjRwgWRjh072siRI127k8GDB7uxUyghAQAAhxVQVPLRqVMn27RpkwskGrRN4eTSSy9100ePHm3x8fFugDaVqqiHzpgxY8KvL1KkiM2YMcO1XVFwKV26tGvDMmzYsNwsBgAAKOCOeByU/MA4KAAAxJ4/ZRwUAACAvEJAAQAA3iGgAAAA7xBQAACAdwgoAADAOwQUAADgHQIKAADwDgEFAAB4h4ACAAC8Q0ABAADeIaAAAADvEFAAAIB3CCgAAMA7BBQAAOAdAgoAAPAOAQUAAHiHgAIAALxDQAEAAN4hoAAAAO8QUAAAgHcIKAAAwDsEFAAA4B0CCgAA8A4BBQAAeIeAAgAAvENAAQAA3iGgAAAA7xBQAACAdwgoAADAOwQUAADgHQIKAADwDgEFAAB4h4ACAAC8Q0ABAACxHVCGDx9uZ599th1zzDFWqVIla9Omja1evTpqnl27dlnPnj2tQoUKVqZMGWvXrp2lpqZGzbN+/Xpr1aqVlSpVyr1P//79be/evUdnjQAAQOEKKHPmzHHh44svvrDZs2fbnj17rEWLFrZz587wPH379rXp06fb1KlT3fwbN260tm3bhqfv27fPhZPdu3fbvHnzbOLEiTZhwgQbMmTI0V0zAAAQs+JCoVDocF+8detWVwKiIPLXv/7Vtm/fbhUrVrQpU6ZY+/bt3TyrVq2y2rVr2/z5861Jkyb2/vvvW+vWrV1wqVy5sptn3LhxNnDgQPd+xYsXP+TnpqenW2Jiovu8smXL5mhZq9890/LauhGt8vwzAACIVbk5fh9RGxR9gJQvX97dL1myxJWqNG/ePDxPrVq1rFq1ai6giO7r1asXDieSnJzsFnr58uVZfk5GRoabHnkDAAAF12EHlP3791ufPn3svPPOs7p167rnNm/e7EpAypUrFzWvwoimBfNEhpNgejAtu7YvSlzBrWrVqoe72AAAoCAHFLVF+eabb+z111+3vDZo0CBXWhPcNmzYkOefCQAA8k/Rw3lRr169bMaMGTZ37lw74YQTws8nJSW5xq9paWlRpSjqxaNpwTwLFy6Mer+gl08wT2YJCQnuBgAACodclaCoPa3CybRp0+zjjz+2GjVqRE1v2LChFStWzFJSUsLPqRuyuhU3bdrUPdb9smXLbMuWLeF51CNIjWXq1Klz5GsEAAAKVwmKqnXUQ+fdd991Y6EEbUbULqRkyZLuvmvXrtavXz/XcFaho3fv3i6UqAePqFuygkjHjh1t5MiR7j0GDx7s3ptSEgAAkOuAMnbsWHd/0UUXRT0/fvx4u/HGG93fo0ePtvj4eDdAm3rfqIfOmDFjwvMWKVLEVQ/16NHDBZfSpUtb586dbdiwYewRAABw5OOg5BfGQQEAIPb8aeOgAAAAeNOLB/mHEXEBAIUBJSgAAMA7BBQAAOAdAgoAAPAOAQUAAHiHgAIAALxDQAEAAN4hoAAAAO8QUAAAgHcIKAAAwDsEFAAA4B0CCgAA8A4BBQAAeIeAAgAAvENAAQAA3iGgAAAA7xBQAACAdwgoAADAOwQUAADgHQIKAADwDgEFAAB4h4ACAAC8Q0ABAADeIaAAAADvEFAAAIB3CCgAAMA7BBQAAOAdAgoAAPAOAQUAAHiHgAIAALxDQAEAAN4hoAAAAO8QUAAAgHeK5vYFc+fOtUcffdSWLFlimzZtsmnTplmbNm3C00OhkA0dOtReeOEFS0tLs/POO8/Gjh1rNWvWDM+zbds26927t02fPt3i4+OtXbt29uSTT1qZMmWO3prBa9Xvnpmn779uRKs8fX8AgGclKDt37rQzzjjDnn322Synjxw50p566ikbN26cLViwwEqXLm3Jycm2a9eu8DwdOnSw5cuX2+zZs23GjBku9HTv3v3I1gQAABTeEpSWLVu6W1ZUevLEE0/Y4MGD7corr3TPvfLKK1a5cmV755137LrrrrOVK1farFmzbNGiRdaoUSM3z9NPP22XX365PfbYY1alSpUjXScAABDjjmoblLVr19rmzZutefPm4ecSExOtcePGNn/+fPdY9+XKlQuHE9H8qupRiUtWMjIyLD09PeoGAAAKrqMaUBRORCUmkfQ4mKb7SpUqRU0vWrSolS9fPjxPZsOHD3dBJ7hVrVr1aC42AADwTEz04hk0aJBt3749fNuwYUN+LxIAAIiVgJKUlOTuU1NTo57X42Ca7rds2RI1fe/eva5nTzBPZgkJCVa2bNmoGwAAKLiOakCpUaOGCxkpKSnh59ReRG1LmjZt6h7rXt2P1U058PHHH9v+/ftdWxUAAIBc9+LZsWOHfffdd1ENY7/66ivXhqRatWrWp08fe/DBB924Jwos9957r+uZE4yVUrt2bbvsssusW7durivynj17rFevXq6HDz14AADAYQWUxYsX28UXXxx+3K9fP3ffuXNnmzBhgg0YMMCNlaJxTVRScv7557tuxSVKlAi/ZvLkyS6UNGvWLDxQm8ZOAQAAOKyActFFF7nxTrITFxdnw4YNc7fsqLRlypQp7AEAABC7vXgAAEDhQkABAADeIaAAAADvEFAAAIB3CCgAAMA7BBQAAOAdAgoAAPAOAQUAAHiHgAIAALxDQAEAAN4hoAAAAO8QUAAAQOxfLBDA/1f97pl5vinWjWjF5gZQKFGCAgAAvEMJClDIURIEwEeUoAAAAO8QUAAAgHcIKAAAwDsEFAAA4B0CCgAA8A4BBQAAeIeAAgAAvENAAQAA3iGgAAAA7zCSLICYx2i4QMFDCQoAAPAOAQUAAHiHgAIAALxDQAEAAN4hoAAAAO/QiwcAPEFvJOB/KEEBAADeIaAAAADvEFAAAIB38jWgPPvss1a9enUrUaKENW7c2BYuXJifiwMAAAp7I9k33njD+vXrZ+PGjXPh5IknnrDk5GRbvXq1VapUKb8WCwDgeWPfdSNa5en7o5AHlFGjRlm3bt3spptuco8VVGbOnGkvv/yy3X333fm1WAAA0KOqsAaU3bt325IlS2zQoEHh5+Lj46158+Y2f/78A+bPyMhwt8D27dvdfXp6eo4/c3/GH5bXcrM8h4v1yBn2ReH6ThWEdRDWI2fYFzlXd+gHlte+uT851/suFAodeuZQPvj555+1ZKF58+ZFPd+/f//QOeecc8D8Q4cOdfNzYxvwHeA7wHeA7wDfAYv5bbBhw4ZDZoWYGKhNJS1qrxLYv3+/bdu2zSpUqGBxcXF58plKeVWrVrUNGzZY2bJlLRYVhHUQ1sMf7At/sC/8UhD2R/qfsA4qOfn999+tSpUqh5w3XwLKcccdZ0WKFLHU1NSo5/U4KSnpgPkTEhLcLVK5cuXsz6CdFKtftoK0DsJ6+IN94Q/2hV8Kwv4om8frkJiY6G834+LFi1vDhg0tJSUlqlREj5s2bZofiwQAADySb1U8qrLp3LmzNWrUyM455xzXzXjnzp3hXj0AAKDwyreAcu2119rWrVttyJAhtnnzZmvQoIHNmjXLKleubD5QldLQoUMPqFqKJQVhHYT18Af7wh/sC78UhP2R4Nk6xKmlbH4vBAAAQCSuxQMAALxDQAEAAN4hoAAAAO8QUAAAgHcIKAAAwDsEFMScffv2ma804OAjjzxi5513np199tnuytz//e9/83uxACDmxMS1eAD59ttv7cUXX7RXX33VNm3a5OVGeeihh+y+++5zV+YuWbKkPfnkk7ZlyxZ7+eWXrSDQqASffPKJC13nnnuuHXvssfm9SIXCU089le2Q4aeeeiojcOOo/59rXLKXXnrJ/vnPf1p+YRyUg9i7d6/t2rXLypQp8+ftEUT5448/7I033nAH+Pnz57uRh9u1a2f9+/f3ckvVrFnT7rrrLrvlllvc448++shatWrlDujx8bFVYJmWlmZ33HGHLV261Jo0aWKPP/64XX755TZv3jw3vVKlSvbhhx9a/fr1zWcXX3zxIS8qqumRl97wTY0aNbLdR9u3b3dh8b333rPy5ctbLNOJh0L+M888k9+LUiitXbvW/dZOmDDBDaSqE60ZM2bk2/IQUMxs+vTp9uuvv9qNN94Y3jD6J3nggQdcSLnkkkvcQdL3s8W2bdse9Czr5ptvtooVK1os+OKLL1xpydSpU61atWq2cuVKd+Z+wQUXmM80AuN3333nrggaKFGihHvuhBNOsFii78vcuXPdJSn0P6KApTMrXZZCfw8YMMCFd03zWd++fbOdpquqTpkyxTIyMryuOjyYH374wW644QY3GveYMWPMd8uXL3f/y7om2zXXXOMu/PrLL7+439xx48bZSSed5Obxmb7/OQm9On74LiMjw5WSqLTks88+c/8Hjz32mHXt2jXfL3pIQPm/M6z27dtbz5493UbRGaIOhMOGDbPatWvbP/7xD2vZsqWNGjXKfJbddYx0lvX111+7ex1w6tata77SWboSvM4Kr7/+evfDe8YZZ1ixYsXcOtSpU8d8pqt069INkUHwmGOOsf/85z/ZngX76i9/+Ys7eF944YX2888/u9D18ccf20UXXeSmL1y40P72t7+59Y01OnA8++yz7qCoAK+Tkeuuu85ilf6vu3Tp4oKwz1TKo9/a4MCtMPLCCy+4oKILyPbp08cuu+wy8927776b7TSV9KpKTu3RVALvqyVLlrhQ8tprr9kpp5xiHTt2dJeg0YmUN7+1Guq+sKtYsWJo6dKl4cd9+/YNJScnhx/PnDkzdMopp4Ri2b59+0JdunQJtW7dOuSzIkWKhO65557Q3r17o54vWrRoaPny5SHfxcXFhS6//PLQVVddFb5p2Vu0aBH1XCzQvti4cWP4ccmSJUPfffdd+PGmTZtC8fHxoVgzadKk0EknnRQ6/vjjQ88++2xoz549oVi3du3aUOnSpUO+O/vss0N9+vQJ/f7776HRo0e7/5e6deuGFi5cGIp1q1atCrVp08b933Tq1Cm0bt26kM+KFCni9oWW29ff2tiqFM8jKuatUKFC+LGKuZo1axZ+fPrpp9vGjRstlqlI8vbbb3ep2Wc6k1W1jkobBg4caN98843Fkk6dOrm2GTorD24qBapSpUrUc7FAZ4AqEQro78hi7UMVcftGjf5UDXLbbbe56tw1a9a4v4sWjf2+AsuWLbMTTzzRfLd69WpXUq2qwd69e7vfpdGjR7seb7FKx4Zu3bpZvXr1XMnQV199ZRMnTvR+fzRr1syVoKimQP8bPl6WL/b/M49SUbbaOKitw44dO1zxlv5pAmqfUqpUKYt1pUuXdo1OfTZo0CB3mzNnjqvqady4sSt+1D/Pb7/9Zr5T47KCRO2Agkbi+vHV+h133HHhYB8LVBWlsKt2TbfeeqtruBysQ6xIT0/P8nlVheqk484773RthXyn70zQrkGBVz3dVM0Ti7TtH374YXv66add8FUja9/byEX64IMPbMOGDTZ+/Hjr0aOHa8ivKh6fTj5og/J/B8V33nnH7rnnHvvXv/7l2qCo4Vlw9vj888/bK6+84kpWYtnYsWPdl1E/2LEiaMSosKIf4nPOOcfVYffr189iqaFyJJ2xJyUl2aWXXmpXXHGF+ap69eo5+qFSy3+f6SxdB8Lu3bsftB2QShhjsVGmnleDZrV7UMNTn2k9VLoQlCKqnZkaXVeuXDlqPrVt8tnIkSPdeEf6P1ZIufLKKy3WzZ492x0fpk2b5tqb6XdWPSbVNii/EFDMXHJUt1D1RtAXToEkMgmrEa0abukszPcGaAc7y9LZsG6x2hhQ1T0qkpw8ebIbWySWGipnrjrR8quUSF2SVcSK/A1amq6TEl/pu5IVlUaoa7tKufT/4XMDeMlJV3vtC997VAWhV91wI6tBM3v77bct1vz22282adIkd1KoqsP87IlEQClAsvvnVy+S0047zZU6+B5O1EukV69erjg+cxc3Ba2mTZu6M0X9MMQ6jS+gNhDr1683H6kHgqpDWrduHS5pVJfEyJIghSt1o0b+lTCqF4aC++LFi70/sBcUasOUk9JFlUj43AalZ8+e2Zb6ahyUk08+OdvqxT9FfrfS9UG5cuVCxx577AG36tWru94XH374YX4vYqFxxRVXhEaNGpXt9CeffNK1lC8IfvvtN6979IwdOzaq11eZMmVCjRs3Dl100UXulpSUFHr88cdDvktJSQnVrl07tH379gOmpaWlherUqROaO3duKJbMmTPH9RRRz52aNWuGBg4cGFM9YX755Zfw3+vXrw/de++9of79+8fcfohlcXFxrifPkCFDspy+efPmfO+lRwmKmasTzYrGDVHViAZp00A2PrcXEI3yqbOpoH53xIgRrlGgBkIKGvuq6mrFihXmK7V8V4tyjT+TlVWrVlmLFi28LXUoSPRd0WBswfdeJXFqQB40alQxsMYS0bgPPlN7BlXTZjdgm0rkNHCY6t59pvFm1EhZpSU6q9XYIRrYzJsxK3JAVQb6PqlxpqqmXn/9dVd9vnPnTlcCrHv91rZp08Z8lpO2Zipheeutt8xX8fHx9txzz7lqZg1Gqv9ndaQIpKamut6H+VkqR0DJAQ3Qpn+aYIhvn79w+hFTN1dRFYm6vAUHFB++cIei6gLVpavnTlY0EJW683EBvrx3/PHHu/ChNhyiwecWLVoUfqxrI6l7qKrefFYQQq8O6hqMTZdN6NChgzuoq+1DrAxgGNCAl6oa1EU0dU0tVXMmJye7wdpEXY91UqgqXp/lpK2Z71U88f93vNCJqxr5ahRsDUDn1fEiX8tvYsTq1atdlU8sFNmlpqZGFcl///33XhXZHYoG0Jo2bVq20996661QjRo1/tRlKqxKlChxwCBOkVauXBlKSEgI+U7LuGbNmmyna5rW1WcqitcAkt9++623g2rlRIUKFUJff/21+1uDtek3a/HixVHfqcTExHxcwsIjLuJ4oarOli1bhsqXLx+aPXu2N8cLBmrLATUM9L37XkGhaqp77703yyGiVWoydOjQcKNN5C0NeX2wgfI0fH8sXF9I4xwdaj1UWuQzDXGgBrHq8qmxgXQxPV2/JtZs27bN9ZQU9TxSlULkNc70d6yMr1OQJCYm2syZM92Ac/oNjhwHLD8RUHJAdb4aiMd3qvPM3LLclwF3cmrw4MHuR0wXN9RYAypy1E1jDqgnkqbp2kjIe/qhGjJkSLZh8f7773dVDr4rCKFXV5NWNYiu9qshEdR2Q8Xv6rKu8Sti6aAe679RBUVcFvtB7RY15pf+XzS2Tn6jDYpZtoN+qW5dl5pXXbvqf/NzwJqc1imqjld1iaJxXdT4KWj4pJIg1cX73AZFfvzxRzeyoUY6DIZf1j+P6qrVKDPWLroXq1QHrWCu0kN1/VZoDIYr1xm8xkf48ssvDxhky8f1OOuss1ybDa2Hgm7Q9kTfJ/0/6P/c9/XITPtBJ09qy6EG/Rr4L7uxkHxRUH6jCoL4TG0WI6ntohoqqzEzjWTzmVr4Z0WNTPVjpoNlLBwUC0LDrcwDBqlRrEKKWvxHFgXjz6FRYvX911l6ZFjUwXDMmDExM0x5QQ69OoDoQK+BtXwPKAXtNyqWzZkzx84777xsr0WlxrOq9tH1xfILJSgADklVawqLoh5W5cuXj8mtRugFYgcBBQAAeIdGsgAAwDsEFAAA4B0CCgAA8A4BBQAAeIeAAgAAvENAARBT1q1b58Yw0WBSAAouAgqAAmn37t35vQgAjgABBUCu6Povuk6SBmzTkOXVqlWzhx56yE1btmyZG7q8ZMmSVqFCBevevbvt2LEj/NqLLrrI+vTpE/V+GlL7xhtvDD+uXr26Pfzww9alSxc75phj3Ps///zz4enBqK9nnnmmK0nRe4reQ++lZdF1ajQK9LBhw6xu3boHrIOG8Nf1RgD4i4ACIFcGDRrkLiqmA/yKFStsypQp7jo2O3fudEPH65IEixYtsqlTp9pHH33krn+TW48//rg1atTIXevntttuc8PU69ozsnDhQnev99bF895+++3w61JSUtx8Gpp/xowZLuSsXLnSLU9A76krGOd02HUA+SPrQfgBIAu6au6TTz7pLhbYuXNn99zJJ59s559/vrvarq4YrKuhBhd/03xXXHGFuxp1bi7GpysQK5jIwIED3eXfP/nkE1cqUrFiRfe8SmiSkpKiXqfPffHFF90FDgMKTbq2y9lnn+0e6+8LL7wwZq4jBBRWlKAAyDGVRuiKs82aNcty2hlnnBEOJ6KLkalKKCj9yKn69euH/1Y1joLIli1bDvm6evXqRYUT6datm7322msuPKldikp8VLICwG+UoADIMbUtOdJLvAdXEw7s2bPngPmKFSsW9VghRUHnUCLDUUAlOGorM23aNBde9Hnt27c/rOUH8OehBAVAjtWsWdOFFLX1yKx27dr29ddfu7Yogc8//9yFElXNiKpn1G4ksG/fPvvmm29ytQeCEhK9Nid0OXlVR6lqR7frrrvuiIMWgLxHCQqAHCtRooRrEzJgwAAXFFSFs3XrVlu+fLl16NDBhg4d6sLAfffd557v3bu3dezYMdz+RD18+vXrZzNnznRtV0aNGmVpaWm52gOVKlVyAWPWrFl2wgknuGVKTEw86GtuvvlmF6CC0ATAf5SgAMgV9d658847bciQIe6gf+2117r2IaVKlbIPPvjAtm3b5hqkqhpFbVXUUDagth8KMJ06dQo3VL344otz9fkqEXnqqafsueeec92Jr7zyyhyV/Jx77rlWq1Yta9y4MXsciAFxocwVwgBQwOhnTiFFPYNUggPAf1TxACjQVNX0+uuv2+bNmxn7BIghBBQABZrarBx33HFuNFoNIgcgNhBQABRo1GIDsYlGsgAAwDsEFAAA4B0CCgAA8A4BBQAAeIeAAgAAvENAAQAA3iGgAAAA7xBQAACA+eb/ASe/aQ3jRAOcAAAAAElFTkSuQmCC", "text/plain": [ "

" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "top_countries = df[\"country\"].value_counts().head(10)\n", "\n", "top_countries.plot(kind=\"bar\")\n", "plt.title(\"Топ стран по количеству случаев\")\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "2abe77de-739a-4768-8b64-4381480bd2e1", "metadata": {}, "source": [ "## 4. Использование tqdm\n", "\n", "Библиотека `tqdm` используется для отображения прогресс-бара при обработке данных. \n", "В данной работе она применена для демонстрации прохода по строкам датасета." ] }, { "cell_type": "code", "execution_count": 14, "id": "8aa34e5c-26c8-4b35-b920-f427700dd8bd", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1287/1287 [00:00<00:00, 1789217.52it/s]\n" ] } ], "source": [ "from tqdm import tqdm\n", "\n", "count = 0\n", "for i in tqdm(range(len(df))):\n", " count += 1" ] }, { "cell_type": "markdown", "id": "9cb10000-fa7b-4809-9f95-017e28d20f49", "metadata": {}, "source": [ "## 5. Создание нового признака\n", "\n", "Для дополнительного анализа был создан новый столбец `risk_flag`. \n", "Он показывает, относится ли случай к группе повышенного риска на основе значения `severity_score`." ] }, { "cell_type": "code", "execution_count": 15, "id": "d675d16a-2e41-4051-94f4-37a01ff4bb2f", "metadata": {}, "outputs": [], "source": [ "df[\"risk_flag\"] = df[\"severity_score\"] > 7" ] }, { "cell_type": "code", "execution_count": 16, "id": "f8ac6ccd-fd69-4472-8840-aeeb2daa1e26", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "risk_flag\n", "False 1258\n", "True 29\n", "Name: count, dtype: int64" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[\"risk_flag\"].value_counts()" ] }, { "cell_type": "markdown", "id": "f1d00383-7ac0-4548-bea7-b961c53e7506", "metadata": {}, "source": [ "## Вывод\n", "\n", "В ходе работы был загружен и проанализирован датасет FDA Drug Adverse Events. \n", "Были изучены основные признаки, выполнена проверка структуры данных и построены визуализации.\n", "\n", "В работе были использованы:\n", "- `pandas` для загрузки и обработки данных;\n", "- `numpy` для работы с числовыми значениями;\n", "- `matplotlib` и `seaborn` для построения графиков;\n", "- `tqdm` для отображения прогресса обработки.\n", "\n", "Полученные графики помогают наглядно оценить распределение возраста пациентов, различия по полу и страны с наибольшим количеством зарегистрированных случаев." ] } ], "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": "3.14.4" } }, "nbformat": 4, "nbformat_minor": 5 }