{ "cells": [ { "cell_type": "markdown", "id": "9e30b402-4ed9-4e45-aa7c-a77a7d68df17", "metadata": {}, "source": [ "# AdaBoost回归" ] }, { "cell_type": "code", "execution_count": 1, "id": "1a11e27d-9c4e-47eb-958b-cf9ed1c4c603", "metadata": {}, "outputs": [], "source": [ "import numpy as np\n", "import pandas as pd\n", "from sklearn.ensemble import RandomForestRegressor, AdaBoostRegressor\n", "from sklearn.metrics import mean_squared_error\n", "from sklearn.preprocessing import StandardScaler\n", "from sklearn.model_selection import train_test_split" ] }, { "cell_type": "markdown", "id": "7a704744-26f2-40f1-9bc1-0e25608dd40f", "metadata": {}, "source": [ "读入数据，由于样本量很大，直接删除有缺失值的样本。" ] }, { "cell_type": "code", "execution_count": 2, "id": "9f2f693d-44fd-432f-b732-7a2117106150", "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", "
host_response_ratehost_acceptance_rateaccommodatespricenumber_of_reviewsreview_scores_rating
01.000.332.0120.090.04.50
11.000.982.090.0351.04.58
21.000.982.066.067.04.52
31.000.981.033.0297.04.70
51.001.002.045.042.04.98
.....................
2032521.000.934.0152.01.04.00
2032531.000.972.045.01.03.00
2032541.000.972.040.01.01.00
2032760.990.992.043.01.05.00
2033081.001.003.0110.01.05.00
\n", "

134835 rows × 6 columns

\n", "
" ], "text/plain": [ " host_response_rate host_acceptance_rate accommodates price \\\n", "0 1.00 0.33 2.0 120.0 \n", "1 1.00 0.98 2.0 90.0 \n", "2 1.00 0.98 2.0 66.0 \n", "3 1.00 0.98 1.0 33.0 \n", "5 1.00 1.00 2.0 45.0 \n", "... ... ... ... ... \n", "203252 1.00 0.93 4.0 152.0 \n", "203253 1.00 0.97 2.0 45.0 \n", "203254 1.00 0.97 2.0 40.0 \n", "203276 0.99 0.99 2.0 43.0 \n", "203308 1.00 1.00 3.0 110.0 \n", "\n", " number_of_reviews review_scores_rating \n", "0 90.0 4.50 \n", "1 351.0 4.58 \n", "2 67.0 4.52 \n", "3 297.0 4.70 \n", "5 42.0 4.98 \n", "... ... ... \n", "203252 1.0 4.00 \n", "203253 1.0 3.00 \n", "203254 1.0 1.00 \n", "203276 1.0 5.00 \n", "203308 1.0 5.00 \n", "\n", "[134835 rows x 6 columns]" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "variables = ['number_of_reviews', 'price', 'accommodates',\n", " 'host_response_rate', 'host_acceptance_rate', 'review_scores_rating']\n", "df = pd.read_csv('../data/2022-01(US_25).csv', usecols=variables)\n", "df['price'] = df['price'].replace('\\$', '', regex=True)\n", "df['price'] = df['price'].replace('\\,', '', regex=True).astype(float)\n", "df[['host_response_rate', 'host_acceptance_rate']] = df[['host_response_rate',\n", " 'host_acceptance_rate']].replace('\\%', '', regex=True).astype(float)*0.01\n", "df[['number_of_reviews']] = df[['number_of_reviews']].astype(float)\n", "for col in variables:\n", " df[col] = df[col].astype(np.float32)\n", " df = df[np.isnan(df[col]) != 1]\n", "df" ] }, { "cell_type": "markdown", "id": "16658a50-232f-49c7-ad4c-c93a6a43e118", "metadata": {}, "source": [ "划分测试集和训练集，测试集大小为0.3，并对数据标准化" ] }, { "cell_type": "code", "execution_count": 3, "id": "00158193-2ea9-4fb8-91f1-cb5ea4b9fb96", "metadata": {}, "outputs": [], "source": [ "# 固定划分训练集和测试集\n", "info = df.iloc[:, :-1].values\n", "target = df.iloc[:, -1].values\n", "# 标准化\n", "stdscaler = StandardScaler()\n", "info_train, info_test, target_train, target_test = train_test_split(\n", " info, target, test_size=0.3,shuffle=True, random_state=420)\n", "info_train = stdscaler.fit_transform(info_train)\n", "info_test = stdscaler.transform(info_test)" ] }, { "cell_type": "code", "execution_count": 4, "id": "91622729-b48b-45bd-9cd6-b33f8d839d7a", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([[ 0.2976891 , 0.44020966, -0.10153121, -0.14794447, -0.62968445],\n", " [ 0.2976891 , 0.44020966, -0.10153121, -0.27286685, 0.06075969],\n", " [ 0.2976891 , 0.39323488, -0.8076375 , -0.25845274, -0.62968445],\n", " ...,\n", " [ 0.2976891 , 0.5341586 , 0.6045751 , -0.22241743, 0.03610097],\n", " [ 0.2976891 , 0.25231144, -0.8076375 , -0.3016951 , 1.1457433 ],\n", " [ 0.2976891 , -0.6871791 , -0.8076375 , -0.3185116 , 0.72654516]],\n", " dtype=float32)" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "info_train" ] }, { "cell_type": "code", "execution_count": 5, "id": "3b292dc8-9c79-45e4-8a4a-6bda192a2b64", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([3. , 4.93, 4. , ..., 4.69, 4.85, 4.83], dtype=float32)" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "target_train" ] }, { "cell_type": "markdown", "id": "47bf67ca-2ede-4bf3-9f77-1f49efd68bc3", "metadata": {}, "source": [ "拟合并预测，计算在测试集上的均方误差" ] }, { "cell_type": "code", "execution_count": 6, "id": "81f63b17-e62c-4775-a525-edb8430a0910", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "MSE:0.4843\n" ] } ], "source": [ "# AdaBoost回归\n", "rf = RandomForestRegressor(n_estimators=10, random_state=0, max_depth=2)\n", "ada = AdaBoostRegressor(estimator=rf,n_estimators=100, loss='square', random_state=0)\n", "ada.fit(info_train, target_train)\n", "target_pred = ada.predict(info_test)\n", "print(\"MSE:%.4f\" % mean_squared_error(target_test, target_pred))" ] }, { "cell_type": "code", "execution_count": null, "id": "e4e758bb-8599-41c0-a6db-ef367d744505", "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.10.12" } }, "nbformat": 4, "nbformat_minor": 5 }