{ "cells": [ { "cell_type": "markdown", "id": "0716d800", "metadata": {}, "source": [ "## Прогнозирование заболеваемости COVID\n", "Для решения данной задачи будет проведено обучение рекурентной LSTM нейронной сети. Данная архитектура сети лучше всего позволяет прогнозировать временные ряды.\n", "В качестве тестовых и трейновых данных будет использоваться датасет о заболеваемости короновирусом в России - [COVID.csv](https://github.com/Koldim2001/time_series_theory/blob/main/COVID.csv)" ] }, { "cell_type": "code", "execution_count": 843, "id": "85987ebc", "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "from sklearn.preprocessing import MinMaxScaler\n", "from keras.models import Sequential\n", "from keras.layers import Dense, LSTM, Dropout, Activation\n", "import math\n", "from sklearn.metrics import mean_squared_error\n", "from matplotlib.pyplot import figure\n", "import datetime\n", "from sklearn.metrics import mean_absolute_error" ] }, { "cell_type": "markdown", "id": "3b7935ca", "metadata": {}, "source": [ "Будем рассматривать данные по числу заражений в день" ] }, { "cell_type": "code", "execution_count": 844, "id": "0d30c9d5", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Всего дней в датасете 976\n" ] }, { "data": { "text/html": [ "
| \n", " | DateTime | \n", "Заражений за день | \n", "Выздоровлений за день | \n", "Смертей за день | \n", "
|---|---|---|---|---|
| 0 | \n", "2020-03-12 00:00:00 | \n", "34 | \n", "0 | \n", "0 | \n", "
| 1 | \n", "2020-03-13 00:00:00 | \n", "11 | \n", "0 | \n", "0 | \n", "
| 2 | \n", "2020-03-14 00:00:00 | \n", "14 | \n", "0 | \n", "0 | \n", "
| 3 | \n", "2020-03-15 00:00:00 | \n", "4 | \n", "3 | \n", "0 | \n", "
| 4 | \n", "2020-03-16 00:00:00 | \n", "28 | \n", "2 | \n", "0 | \n", "
| \n", " | Заражений за день | \n", "
|---|---|
| DateTime | \n", "\n", " |
| 2020-03-12 | \n", "34 | \n", "
| 2020-03-13 | \n", "11 | \n", "
| 2020-03-14 | \n", "14 | \n", "
| 2020-03-15 | \n", "4 | \n", "
| 2020-03-16 | \n", "28 | \n", "
"
]
}
],
"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": "3.9.7"
}
},
"nbformat": 4,
"nbformat_minor": 5
}