Add JupyterLab analysis notebook

02_jupyter_analysis/week2_analysis.ipynb

@@ -0,0 +1,144 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "e33cfdf4",
+   "metadata": {
+    "vscode": {
+     "languageId": "plaintext"
+    }
+   },
+   "outputs": [],
+   "source": [
+    "# Ячейка 1: Импорт библиотек\n",
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "import seaborn as sns\n",
+    "from tqdm import tqdm\n",
+    "import time"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "bc6236ff",
+   "metadata": {
+    "vscode": {
+     "languageId": "plaintext"
+    }
+   },
+   "outputs": [],
+   "source": [
+    "# Ячейка 2: Загрузка данных\n",
+    "df = pd.read_csv(\"kinopoisk-top250.csv\", index_col=0)\n",
+    "print(f'Размер таблицы: {df.shape}')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "e616e786",
+   "metadata": {
+    "vscode": {
+     "languageId": "plaintext"
+    }
+   },
+   "outputs": [],
+   "source": [
+    "# Ячейка 3: Анализ\n",
+    "df.info()\n",
+    "df.describe()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "d9df3843",
+   "metadata": {
+    "vscode": {
+     "languageId": "plaintext"
+    }
+   },
+   "outputs": [],
+   "source": [
+    "# Ячейка 4: Гистограмма\n",
+    "plt.figure(figsize=(8, 5))\n",
+    "sns.histplot(df['year'], kde=True, bins=20, color='blue')\n",
+    "plt.title('Распределение фильмов по годам')\n",
+    "plt.xlabel('Год')\n",
+    "plt.ylabel('Количество фильмов')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "e82506e0",
+   "metadata": {
+    "vscode": {
+     "languageId": "plaintext"
+    }
+   },
+   "outputs": [],
+   "source": [
+    "# Ячейка 5: Scatter plot\n",
+    "df_top30 = df.head(30).copy()\n",
+    "plt.figure(figsize=(10, 6))\n",
+    "sns.scatterplot(x='director', y='year', hue='rating_ball', \n",
+    "                data=df_top30, palette='viridis')\n",
+    "plt.title('Режиссеры, год выпуска и рейтинг')\n",
+    "plt.xlabel('Режиссер')\n",
+    "plt.ylabel('Год выпуска')\n",
+    "plt.xticks(rotation=90)\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "f51c0053",
+   "metadata": {
+    "vscode": {
+     "languageId": "plaintext"
+    }
+   },
+   "outputs": [],
+   "source": [
+    "# Ячейка 6: Boxplot\n",
+    "filtered_df = df[df['year'] > 2000].copy()\n",
+    "plt.figure(figsize=(10, 6))\n",
+    "sns.boxplot(x='year', y='rating_ball', data=filtered_df, palette='pastel')\n",
+    "plt.title('Распределение рейтингов (>2000)')\n",
+    "plt.xlabel('Год')\n",
+    "plt.ylabel('Рейтинг')\n",
+    "plt.xticks(rotation=90)\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "bc5173ad",
+   "metadata": {
+    "vscode": {
+     "languageId": "plaintext"
+    }
+   },
+   "outputs": [],
+   "source": [
+    "# Ячейка 7: tqdm\n",
+    "for i in tqdm(range(100), desc=\"Обработка\"):\n",
+    "    time.sleep(0.01)\n",
+    "print(\"Готово\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "language_info": {
+   "name": "python"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}