{ "cells": [ { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "import xgboost as xgb\n", "import pandas as pd\n", "import numpy as np\n", "import matplotlib.pyplot as plt" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "train_data = pd.read_csv('./data/train.csv')\n", "test_data = pd.read_csv('./data/test.csv')" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "train_data.drop(train_data[(train_data[\"GrLivArea\"]>4000)&(train_data[\"SalePrice\"]<300000)].index,inplace=True)#pandas 里面的条件索引" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "train_data" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(2917, 81)" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "all_data = pd.concat([train_data, test_data]).reset_index(drop=True)\n", "all_data.shape" ] }, { "cell_type": "code", "execution_count": 50, "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", "
特征名称缺失率
0PoolQC0.995885
1MiscFeature0.962963
2Alley0.937586
3Fence0.807270
4FireplaceQu0.473251
5LotFrontage0.177641
6GarageYrBlt0.055556
7GarageCond0.055556
8GarageType0.055556
9GarageFinish0.055556
10GarageQual0.055556
11BsmtFinType20.026063
12BsmtExposure0.026063
13BsmtQual0.025377
14BsmtCond0.025377
15BsmtFinType10.025377
16MasVnrArea0.005487
17MasVnrType0.005487
18Electrical0.000686
\n", "
" ], "text/plain": [ " 特征名称 缺失率\n", "0 PoolQC 0.995885\n", "1 MiscFeature 0.962963\n", "2 Alley 0.937586\n", "3 Fence 0.807270\n", "4 FireplaceQu 0.473251\n", "5 LotFrontage 0.177641\n", "6 GarageYrBlt 0.055556\n", "7 GarageCond 0.055556\n", "8 GarageType 0.055556\n", "9 GarageFinish 0.055556\n", "10 GarageQual 0.055556\n", "11 BsmtFinType2 0.026063\n", "12 BsmtExposure 0.026063\n", "13 BsmtQual 0.025377\n", "14 BsmtCond 0.025377\n", "15 BsmtFinType1 0.025377\n", "16 MasVnrArea 0.005487\n", "17 MasVnrType 0.005487\n", "18 Electrical 0.000686" ] }, "execution_count": 50, "metadata": {}, "output_type": "execute_result" } ], "source": [ "miss_value = train_data.isnull().sum().sort_values(ascending=False).to_frame().reset_index()\n", "miss_value.columns = ['feature', 'miss_per']\n", "miss_value = miss_value[miss_value.miss_per > 0]\n", "miss_value.miss_per = miss_value.miss_per / train_data.shape[0]\n", "miss_value.columns = ['特征名称', '缺失率']\n", "miss_value" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Id 0\n", "Foundation 0\n", "Heating 0\n", "SaleCondition 0\n", "CentralAir 0\n", " ... \n", "SalePrice 1459\n", "Fence 2346\n", "Alley 2719\n", "MiscFeature 2812\n", "PoolQC 2908\n", "Length: 81, dtype: int64" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "miss = all_data.isnull().sum().sort_values(ascending=True)\n", "miss" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "删除缺失比例过高的列" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Alley\n", "FireplaceQu\n", "PoolQC\n", "Fence\n", "MiscFeature\n" ] }, { "data": { "text/plain": [ "(2917, 76)" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "all_cols = [x for x in all_data.columns if x != 'Id' and x != 'SalePrice']\n", "for col in all_cols:\n", " if miss[col] > 1000:\n", " print(col)\n", " all_data.drop(columns=[col], inplace=True)\n", "all_data.shape" ] }, { "cell_type": "code", "execution_count": 58, "metadata": {}, "outputs": [], "source": [ "import seaborn as sns\n", "from scipy.stats import norm\n", "from scipy import stats" ] }, { "cell_type": "code", "execution_count": 59, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/home/zhaojh/miniconda3/envs/py37/lib/python3.7/site-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n", " warnings.warn(msg, FutureWarning)\n" ] }, { "data": { "text/plain": [ "" ] }, "execution_count": 59, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "image/svg+xml": "\n\n\n \n \n \n \n 2022-07-27T14:10:04.575558\n image/svg+xml\n \n \n Matplotlib v3.5.2, https://matplotlib.org/\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 \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 \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 \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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "sns.distplot(train_data.SalePrice, fit=norm)" ] }, { "cell_type": "code", "execution_count": 61, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "image/svg+xml": "\n\n\n \n \n \n \n 2022-07-27T14:10:13.679452\n image/svg+xml\n \n \n Matplotlib v3.5.2, https://matplotlib.org/\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 \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 \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 \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 \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 \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 \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 \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 \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 \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 \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 \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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "rest = stats.probplot(train_data.SalePrice, plot=plt)" ] }, { "cell_type": "code", "execution_count": 63, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/home/zhaojh/miniconda3/envs/py37/lib/python3.7/site-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n", " warnings.warn(msg, FutureWarning)\n" ] }, { "data": { "text/plain": [ "" ] }, "execution_count": 63, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "image/svg+xml": "\n\n\n \n \n \n \n 2022-07-27T14:17:35.675100\n image/svg+xml\n \n \n Matplotlib v3.5.2, https://matplotlib.org/\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 \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 \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 \n \n \n\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "sns.distplot(np.log1p(train_data.SalePrice), fit=norm)" ] }, { "cell_type": "code", "execution_count": 64, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "image/svg+xml": "\n\n\n \n \n \n \n 2022-07-27T14:17:39.713631\n image/svg+xml\n \n \n Matplotlib v3.5.2, https://matplotlib.org/\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 \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 \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 \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 \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 \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 \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 \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 \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 \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 \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 \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 \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "rest = stats.probplot(np.log1p(train_data.SalePrice), plot=plt)" ] }, { "cell_type": "code", "execution_count": 56, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(0, 76)" ] }, "execution_count": 56, "metadata": {}, "output_type": "execute_result" } ], "source": [ "all_data[all_data['GarageYrBlt'].isna()].shape" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "na_index = all_data[all_data['GarageYrBlt'] > 2022].index\n", "all_data.loc[na_index, 'GarageYrBlt'] = None" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(160, 76)" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "all_data[all_data['GarageYrBlt'].isna()].shape" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(2917, 76)" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "all_data.GarageYrBlt.fillna(all_data.YearBuilt, inplace=True)\n", "year_cols = ['YearBuilt', 'YearRemodAdd', 'GarageYrBlt']\n", "for col in year_cols:\n", " all_data[col] = 2022 - all_data[col]\n", "all_data.shape" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(2917, 76)" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "cols1 = [\"GarageQual\", \"GarageCond\", \"GarageFinish\", \"GarageType\", \"BsmtExposure\", \"BsmtCond\", \"BsmtQual\", \"BsmtFinType2\", \"BsmtFinType1\", \"MasVnrType\"]\n", "for col in cols1:\n", " all_data[col].fillna(\"None\",inplace=True)\n", "all_data.shape" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(2917, 76)" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "cols2=[\"MasVnrArea\", \"BsmtUnfSF\", \"TotalBsmtSF\", \"GarageCars\", \"BsmtFinSF2\", \"BsmtFinSF1\", \"GarageArea\"]\n", "for col in cols2:\n", " all_data[col].fillna(0, inplace=True)\n", "all_data.shape" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(2917, 76)" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "all_data[\"LotFrontage\"].fillna(np.mean(all_data[\"LotFrontage\"]),inplace=True)\n", "cols3 = [\"MSZoning\", \"BsmtFullBath\", \"BsmtHalfBath\", \"Utilities\", \"Functional\", \"Electrical\", \"KitchenQual\", \"SaleType\",\"Exterior1st\", \"Exterior2nd\"]\n", "for col in cols3:\n", " all_data[col].fillna(all_data[col].mode()[0], inplace=True)\n", "all_data.shape" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [], "source": [ "numeric_cols = [x for x in all_data.select_dtypes(exclude=['object']).columns.tolist() if x != 'Id' and x != 'SalePrice']\n", "object_cols = [x for x in all_data.select_dtypes(include=['object']).columns.tolist()]" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [], "source": [ "for col in numeric_cols:\n", " all_data[col] = np.log1p(all_data[col])\n", " all_data[col] = (all_data[col] - all_data[col].min()) / (all_data[col].max() - all_data[col].min())" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [], "source": [ "dataset = pd.get_dummies(all_data, columns=object_cols)" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [], "source": [ "dataset.SalePrice = np.log1p(dataset.SalePrice)" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(1458, 280)" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "train = dataset[~dataset.SalePrice.isna()].copy()\n", "train.shape" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(1459, 280)" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "test = dataset[dataset.SalePrice.isna()].copy()\n", "test.shape" ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [], "source": [ "feature_cols = [x for x in dataset.columns if x != 'Id' and x != 'SalePrice']" ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [], "source": [ "from sklearn.model_selection import train_test_split" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [], "source": [ "train, valid = train_test_split(train, test_size=0.12, shuffle=True, random_state=42)" ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [], "source": [ "X_train, Y_train = train[feature_cols], train['SalePrice']\n", "X_valid, Y_valid = valid[feature_cols], valid['SalePrice']\n", "X_test, Y_test = test[feature_cols], test['SalePrice']" ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [], "source": [ "dtrain = xgb.DMatrix(X_train, Y_train)\n", "dvalid = xgb.DMatrix(X_valid, Y_valid)\n", "watchlist = [(dtrain, 'train'), (dvalid, 'eval')]" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [], "source": [ "params = {'objective': 'reg:squarederror', \n", " 'booster': 'gbtree', \n", " 'eta': 0.05,\n", " 'max_depth': 15, \n", " 'subsample': 0.7, \n", " 'colsample_bytree': 0.7,\n", " 'eval_metric':['rmse'],\n", " 'silent': 1, \n", " 'seed': 10} \n" ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[10:33:47] WARNING: ../src/learner.cc:627: \n", "Parameters: { \"silent\" } might not be used.\n", "\n", " This could be a false alarm, with some parameters getting used by language bindings but\n", " then being mistakenly passed down to XGBoost core, or some parameter actually being used\n", " but getting flagged wrongly here. Please open an issue if you find any such cases.\n", "\n", "\n", "[0]\ttrain-rmse:10.95491\teval-rmse:10.96235\n", "[1]\ttrain-rmse:10.40916\teval-rmse:10.41661\n", "[2]\ttrain-rmse:9.89034\teval-rmse:9.89780\n", "[3]\ttrain-rmse:9.39722\teval-rmse:9.40469\n", "[4]\ttrain-rmse:8.92885\teval-rmse:8.93633\n", "[5]\ttrain-rmse:8.48375\teval-rmse:8.49124\n", "[6]\ttrain-rmse:8.06123\teval-rmse:8.06873\n", "[7]\ttrain-rmse:7.66021\teval-rmse:7.66773\n", "[8]\ttrain-rmse:7.27851\teval-rmse:7.28504\n", "[9]\ttrain-rmse:6.91608\teval-rmse:6.92262\n", "[10]\ttrain-rmse:6.57212\teval-rmse:6.57776\n", "[11]\ttrain-rmse:6.24453\teval-rmse:6.24978\n", "[12]\ttrain-rmse:5.93355\teval-rmse:5.93791\n", "[13]\ttrain-rmse:5.63820\teval-rmse:5.64171\n", "[14]\ttrain-rmse:5.35791\teval-rmse:5.36060\n", "[15]\ttrain-rmse:5.09149\teval-rmse:5.09384\n", "[16]\ttrain-rmse:4.83796\teval-rmse:4.84034\n", "[17]\ttrain-rmse:4.59742\teval-rmse:4.59968\n", "[18]\ttrain-rmse:4.36846\teval-rmse:4.37007\n", "[19]\ttrain-rmse:4.15155\teval-rmse:4.15304\n", "[20]\ttrain-rmse:3.94554\teval-rmse:3.94632\n", "[21]\ttrain-rmse:3.74977\teval-rmse:3.74999\n", "[22]\ttrain-rmse:3.56360\teval-rmse:3.56321\n", "[23]\ttrain-rmse:3.38712\teval-rmse:3.38680\n", "[24]\ttrain-rmse:3.21874\teval-rmse:3.21847\n", "[25]\ttrain-rmse:3.05978\teval-rmse:3.05902\n", "[26]\ttrain-rmse:2.90867\teval-rmse:2.90743\n", "[27]\ttrain-rmse:2.76500\teval-rmse:2.76388\n", "[28]\ttrain-rmse:2.62812\teval-rmse:2.62685\n", "[29]\ttrain-rmse:2.49820\teval-rmse:2.49628\n", "[30]\ttrain-rmse:2.37453\teval-rmse:2.37196\n", "[31]\ttrain-rmse:2.25692\teval-rmse:2.25370\n", "[32]\ttrain-rmse:2.14536\teval-rmse:2.14147\n", "[33]\ttrain-rmse:2.03937\teval-rmse:2.03521\n", "[34]\ttrain-rmse:1.93883\teval-rmse:1.93448\n", "[35]\ttrain-rmse:1.84381\teval-rmse:1.83979\n", "[36]\ttrain-rmse:1.75285\teval-rmse:1.74887\n", "[37]\ttrain-rmse:1.66676\teval-rmse:1.66205\n", "[38]\ttrain-rmse:1.58492\teval-rmse:1.57965\n", "[39]\ttrain-rmse:1.50715\teval-rmse:1.50159\n", "[40]\ttrain-rmse:1.43321\teval-rmse:1.42713\n", "[41]\ttrain-rmse:1.36283\teval-rmse:1.35596\n", "[42]\ttrain-rmse:1.29620\teval-rmse:1.28879\n", "[43]\ttrain-rmse:1.23316\teval-rmse:1.22663\n", "[44]\ttrain-rmse:1.17272\teval-rmse:1.16596\n", "[45]\ttrain-rmse:1.11549\teval-rmse:1.10860\n", "[46]\ttrain-rmse:1.06120\teval-rmse:1.05444\n", "[47]\ttrain-rmse:1.00958\teval-rmse:1.00254\n", "[48]\ttrain-rmse:0.96067\teval-rmse:0.95273\n", "[49]\ttrain-rmse:0.91434\teval-rmse:0.90591\n", "[50]\ttrain-rmse:0.87015\teval-rmse:0.86133\n", "[51]\ttrain-rmse:0.82834\teval-rmse:0.81927\n", "[52]\ttrain-rmse:0.78870\teval-rmse:0.77968\n", "[53]\ttrain-rmse:0.75082\teval-rmse:0.74161\n", "[54]\ttrain-rmse:0.71492\teval-rmse:0.70547\n", "[55]\ttrain-rmse:0.68106\teval-rmse:0.67230\n", "[56]\ttrain-rmse:0.64849\teval-rmse:0.63960\n", "[57]\ttrain-rmse:0.61769\teval-rmse:0.60831\n", "[58]\ttrain-rmse:0.58868\teval-rmse:0.57939\n", "[59]\ttrain-rmse:0.56057\teval-rmse:0.55152\n", "[60]\ttrain-rmse:0.53451\teval-rmse:0.52523\n", "[61]\ttrain-rmse:0.50950\teval-rmse:0.50035\n", "[62]\ttrain-rmse:0.48564\teval-rmse:0.47651\n", "[63]\ttrain-rmse:0.46293\teval-rmse:0.45377\n", "[64]\ttrain-rmse:0.44159\teval-rmse:0.43343\n", "[65]\ttrain-rmse:0.42131\teval-rmse:0.41326\n", "[66]\ttrain-rmse:0.40179\teval-rmse:0.39410\n", "[67]\ttrain-rmse:0.38364\teval-rmse:0.37700\n", "[68]\ttrain-rmse:0.36614\teval-rmse:0.36009\n", "[69]\ttrain-rmse:0.34965\teval-rmse:0.34418\n", "[70]\ttrain-rmse:0.33389\teval-rmse:0.32955\n", "[71]\ttrain-rmse:0.31898\teval-rmse:0.31511\n", "[72]\ttrain-rmse:0.30487\teval-rmse:0.30192\n", "[73]\ttrain-rmse:0.29146\teval-rmse:0.28948\n", "[74]\ttrain-rmse:0.27854\teval-rmse:0.27745\n", "[75]\ttrain-rmse:0.26624\teval-rmse:0.26603\n", "[76]\ttrain-rmse:0.25467\teval-rmse:0.25535\n", "[77]\ttrain-rmse:0.24384\teval-rmse:0.24510\n", "[78]\ttrain-rmse:0.23341\teval-rmse:0.23538\n", "[79]\ttrain-rmse:0.22357\teval-rmse:0.22674\n", "[80]\ttrain-rmse:0.21429\teval-rmse:0.21868\n", "[81]\ttrain-rmse:0.20526\teval-rmse:0.21073\n", "[82]\ttrain-rmse:0.19662\teval-rmse:0.20326\n", "[83]\ttrain-rmse:0.18837\teval-rmse:0.19614\n", "[84]\ttrain-rmse:0.18054\teval-rmse:0.18948\n", "[85]\ttrain-rmse:0.17345\teval-rmse:0.18387\n", "[86]\ttrain-rmse:0.16646\teval-rmse:0.17787\n", "[87]\ttrain-rmse:0.15977\teval-rmse:0.17240\n", "[88]\ttrain-rmse:0.15350\teval-rmse:0.16762\n", "[89]\ttrain-rmse:0.14754\teval-rmse:0.16333\n", "[90]\ttrain-rmse:0.14182\teval-rmse:0.15882\n", "[91]\ttrain-rmse:0.13632\teval-rmse:0.15475\n", "[92]\ttrain-rmse:0.13127\teval-rmse:0.15126\n", "[93]\ttrain-rmse:0.12620\teval-rmse:0.14789\n", "[94]\ttrain-rmse:0.12159\teval-rmse:0.14519\n", "[95]\ttrain-rmse:0.11702\teval-rmse:0.14218\n", "[96]\ttrain-rmse:0.11266\teval-rmse:0.13953\n", "[97]\ttrain-rmse:0.10853\teval-rmse:0.13714\n", "[98]\ttrain-rmse:0.10450\teval-rmse:0.13514\n", "[99]\ttrain-rmse:0.10078\teval-rmse:0.13347\n", "[100]\ttrain-rmse:0.09716\teval-rmse:0.13144\n", "[101]\ttrain-rmse:0.09377\teval-rmse:0.12970\n", "[102]\ttrain-rmse:0.09061\teval-rmse:0.12809\n", "[103]\ttrain-rmse:0.08744\teval-rmse:0.12667\n", "[104]\ttrain-rmse:0.08450\teval-rmse:0.12523\n", "[105]\ttrain-rmse:0.08152\teval-rmse:0.12383\n", "[106]\ttrain-rmse:0.07869\teval-rmse:0.12271\n", "[107]\ttrain-rmse:0.07611\teval-rmse:0.12161\n", "[108]\ttrain-rmse:0.07358\teval-rmse:0.12084\n", "[109]\ttrain-rmse:0.07116\teval-rmse:0.11998\n", "[110]\ttrain-rmse:0.06895\teval-rmse:0.11904\n", "[111]\ttrain-rmse:0.06676\teval-rmse:0.11830\n", "[112]\ttrain-rmse:0.06457\teval-rmse:0.11761\n", "[113]\ttrain-rmse:0.06251\teval-rmse:0.11679\n", "[114]\ttrain-rmse:0.06071\teval-rmse:0.11642\n", "[115]\ttrain-rmse:0.05873\teval-rmse:0.11584\n", "[116]\ttrain-rmse:0.05691\teval-rmse:0.11509\n", "[117]\ttrain-rmse:0.05539\teval-rmse:0.11460\n", "[118]\ttrain-rmse:0.05374\teval-rmse:0.11408\n", "[119]\ttrain-rmse:0.05229\teval-rmse:0.11369\n", "[120]\ttrain-rmse:0.05087\teval-rmse:0.11348\n", "[121]\ttrain-rmse:0.04938\teval-rmse:0.11326\n", "[122]\ttrain-rmse:0.04790\teval-rmse:0.11283\n", "[123]\ttrain-rmse:0.04652\teval-rmse:0.11271\n", "[124]\ttrain-rmse:0.04506\teval-rmse:0.11234\n", "[125]\ttrain-rmse:0.04385\teval-rmse:0.11213\n", "[126]\ttrain-rmse:0.04264\teval-rmse:0.11208\n", "[127]\ttrain-rmse:0.04140\teval-rmse:0.11193\n", "[128]\ttrain-rmse:0.04036\teval-rmse:0.11187\n", "[129]\ttrain-rmse:0.03931\teval-rmse:0.11160\n", "[130]\ttrain-rmse:0.03824\teval-rmse:0.11150\n", "[131]\ttrain-rmse:0.03722\teval-rmse:0.11131\n", "[132]\ttrain-rmse:0.03628\teval-rmse:0.11130\n", "[133]\ttrain-rmse:0.03530\teval-rmse:0.11123\n", "[134]\ttrain-rmse:0.03441\teval-rmse:0.11112\n", "[135]\ttrain-rmse:0.03345\teval-rmse:0.11104\n", "[136]\ttrain-rmse:0.03262\teval-rmse:0.11096\n", "[137]\ttrain-rmse:0.03188\teval-rmse:0.11098\n", "[138]\ttrain-rmse:0.03105\teval-rmse:0.11097\n", "[139]\ttrain-rmse:0.03025\teval-rmse:0.11102\n", "[140]\ttrain-rmse:0.02952\teval-rmse:0.11110\n", "[141]\ttrain-rmse:0.02890\teval-rmse:0.11103\n", "[142]\ttrain-rmse:0.02824\teval-rmse:0.11104\n", "[143]\ttrain-rmse:0.02761\teval-rmse:0.11102\n", "[144]\ttrain-rmse:0.02702\teval-rmse:0.11100\n", "[145]\ttrain-rmse:0.02634\teval-rmse:0.11108\n", "[146]\ttrain-rmse:0.02584\teval-rmse:0.11106\n", "[147]\ttrain-rmse:0.02540\teval-rmse:0.11111\n", "[148]\ttrain-rmse:0.02489\teval-rmse:0.11130\n", "[149]\ttrain-rmse:0.02439\teval-rmse:0.11131\n", "[150]\ttrain-rmse:0.02382\teval-rmse:0.11130\n", "[151]\ttrain-rmse:0.02333\teval-rmse:0.11134\n", "[152]\ttrain-rmse:0.02277\teval-rmse:0.11133\n", "[153]\ttrain-rmse:0.02238\teval-rmse:0.11135\n", "[154]\ttrain-rmse:0.02189\teval-rmse:0.11143\n", "[155]\ttrain-rmse:0.02146\teval-rmse:0.11156\n", "[156]\ttrain-rmse:0.02101\teval-rmse:0.11152\n", "[157]\ttrain-rmse:0.02058\teval-rmse:0.11150\n", "[158]\ttrain-rmse:0.02017\teval-rmse:0.11143\n", "[159]\ttrain-rmse:0.01975\teval-rmse:0.11141\n", "[160]\ttrain-rmse:0.01932\teval-rmse:0.11136\n", "[161]\ttrain-rmse:0.01901\teval-rmse:0.11136\n", "[162]\ttrain-rmse:0.01860\teval-rmse:0.11142\n", "[163]\ttrain-rmse:0.01820\teval-rmse:0.11150\n", "[164]\ttrain-rmse:0.01792\teval-rmse:0.11156\n", "[165]\ttrain-rmse:0.01758\teval-rmse:0.11161\n", "[166]\ttrain-rmse:0.01725\teval-rmse:0.11173\n", "[167]\ttrain-rmse:0.01694\teval-rmse:0.11173\n", "[168]\ttrain-rmse:0.01661\teval-rmse:0.11172\n", "[169]\ttrain-rmse:0.01629\teval-rmse:0.11181\n", "[170]\ttrain-rmse:0.01602\teval-rmse:0.11185\n", "[171]\ttrain-rmse:0.01574\teval-rmse:0.11181\n", "[172]\ttrain-rmse:0.01544\teval-rmse:0.11183\n", "[173]\ttrain-rmse:0.01520\teval-rmse:0.11179\n", "[174]\ttrain-rmse:0.01489\teval-rmse:0.11181\n", "[175]\ttrain-rmse:0.01463\teval-rmse:0.11181\n", "[176]\ttrain-rmse:0.01435\teval-rmse:0.11179\n", "[177]\ttrain-rmse:0.01409\teval-rmse:0.11177\n", "[178]\ttrain-rmse:0.01373\teval-rmse:0.11180\n", "[179]\ttrain-rmse:0.01350\teval-rmse:0.11181\n", "[180]\ttrain-rmse:0.01327\teval-rmse:0.11180\n", "[181]\ttrain-rmse:0.01304\teval-rmse:0.11185\n", "[182]\ttrain-rmse:0.01279\teval-rmse:0.11187\n", "[183]\ttrain-rmse:0.01256\teval-rmse:0.11186\n", "[184]\ttrain-rmse:0.01232\teval-rmse:0.11188\n", "[185]\ttrain-rmse:0.01211\teval-rmse:0.11191\n", "[186]\ttrain-rmse:0.01186\teval-rmse:0.11187\n", "[187]\ttrain-rmse:0.01172\teval-rmse:0.11188\n", "[188]\ttrain-rmse:0.01150\teval-rmse:0.11201\n", "[189]\ttrain-rmse:0.01133\teval-rmse:0.11203\n", "[190]\ttrain-rmse:0.01110\teval-rmse:0.11207\n", "[191]\ttrain-rmse:0.01092\teval-rmse:0.11210\n", "[192]\ttrain-rmse:0.01075\teval-rmse:0.11209\n", "[193]\ttrain-rmse:0.01057\teval-rmse:0.11205\n", "[194]\ttrain-rmse:0.01042\teval-rmse:0.11211\n", "[195]\ttrain-rmse:0.01025\teval-rmse:0.11215\n", "[196]\ttrain-rmse:0.01008\teval-rmse:0.11213\n", "[197]\ttrain-rmse:0.00993\teval-rmse:0.11216\n", "[198]\ttrain-rmse:0.00973\teval-rmse:0.11215\n", "[199]\ttrain-rmse:0.00959\teval-rmse:0.11218\n", "[200]\ttrain-rmse:0.00946\teval-rmse:0.11218\n", "[201]\ttrain-rmse:0.00929\teval-rmse:0.11218\n", "[202]\ttrain-rmse:0.00911\teval-rmse:0.11218\n", "[203]\ttrain-rmse:0.00896\teval-rmse:0.11220\n", "[204]\ttrain-rmse:0.00884\teval-rmse:0.11217\n", "[205]\ttrain-rmse:0.00872\teval-rmse:0.11216\n", "[206]\ttrain-rmse:0.00861\teval-rmse:0.11219\n", "[207]\ttrain-rmse:0.00844\teval-rmse:0.11218\n", "[208]\ttrain-rmse:0.00830\teval-rmse:0.11227\n", "[209]\ttrain-rmse:0.00819\teval-rmse:0.11229\n", "[210]\ttrain-rmse:0.00809\teval-rmse:0.11230\n", "[211]\ttrain-rmse:0.00800\teval-rmse:0.11231\n", "[212]\ttrain-rmse:0.00783\teval-rmse:0.11234\n", "[213]\ttrain-rmse:0.00772\teval-rmse:0.11234\n", "[214]\ttrain-rmse:0.00762\teval-rmse:0.11232\n", "[215]\ttrain-rmse:0.00747\teval-rmse:0.11235\n", "[216]\ttrain-rmse:0.00734\teval-rmse:0.11236\n", "[217]\ttrain-rmse:0.00723\teval-rmse:0.11240\n", "[218]\ttrain-rmse:0.00709\teval-rmse:0.11241\n", "[219]\ttrain-rmse:0.00697\teval-rmse:0.11240\n", "[220]\ttrain-rmse:0.00687\teval-rmse:0.11242\n", "[221]\ttrain-rmse:0.00680\teval-rmse:0.11245\n", "[222]\ttrain-rmse:0.00667\teval-rmse:0.11250\n", "[223]\ttrain-rmse:0.00658\teval-rmse:0.11254\n", "[224]\ttrain-rmse:0.00647\teval-rmse:0.11255\n", "[225]\ttrain-rmse:0.00639\teval-rmse:0.11258\n", "[226]\ttrain-rmse:0.00627\teval-rmse:0.11257\n", "[227]\ttrain-rmse:0.00616\teval-rmse:0.11256\n", "[228]\ttrain-rmse:0.00605\teval-rmse:0.11257\n", "[229]\ttrain-rmse:0.00595\teval-rmse:0.11261\n", "[230]\ttrain-rmse:0.00583\teval-rmse:0.11262\n", "[231]\ttrain-rmse:0.00577\teval-rmse:0.11264\n", "[232]\ttrain-rmse:0.00566\teval-rmse:0.11263\n", "[233]\ttrain-rmse:0.00558\teval-rmse:0.11263\n", "[234]\ttrain-rmse:0.00552\teval-rmse:0.11264\n", "[235]\ttrain-rmse:0.00543\teval-rmse:0.11264\n", "[236]\ttrain-rmse:0.00536\teval-rmse:0.11265\n", "[237]\ttrain-rmse:0.00530\teval-rmse:0.11266\n", "[238]\ttrain-rmse:0.00524\teval-rmse:0.11267\n", "[239]\ttrain-rmse:0.00513\teval-rmse:0.11265\n", "[240]\ttrain-rmse:0.00505\teval-rmse:0.11265\n", "[241]\ttrain-rmse:0.00497\teval-rmse:0.11265\n", "[242]\ttrain-rmse:0.00488\teval-rmse:0.11264\n", "[243]\ttrain-rmse:0.00481\teval-rmse:0.11265\n", "[244]\ttrain-rmse:0.00472\teval-rmse:0.11266\n", "[245]\ttrain-rmse:0.00465\teval-rmse:0.11267\n", "[246]\ttrain-rmse:0.00461\teval-rmse:0.11266\n", "[247]\ttrain-rmse:0.00453\teval-rmse:0.11265\n", "[248]\ttrain-rmse:0.00445\teval-rmse:0.11265\n", "[249]\ttrain-rmse:0.00439\teval-rmse:0.11266\n", "[250]\ttrain-rmse:0.00431\teval-rmse:0.11266\n", "[251]\ttrain-rmse:0.00425\teval-rmse:0.11267\n", "[252]\ttrain-rmse:0.00417\teval-rmse:0.11268\n", "[253]\ttrain-rmse:0.00411\teval-rmse:0.11269\n", "[254]\ttrain-rmse:0.00404\teval-rmse:0.11268\n", "[255]\ttrain-rmse:0.00399\teval-rmse:0.11269\n", "[256]\ttrain-rmse:0.00391\teval-rmse:0.11270\n", "[257]\ttrain-rmse:0.00385\teval-rmse:0.11270\n", "[258]\ttrain-rmse:0.00379\teval-rmse:0.11272\n", "[259]\ttrain-rmse:0.00372\teval-rmse:0.11272\n", "[260]\ttrain-rmse:0.00367\teval-rmse:0.11271\n", "[261]\ttrain-rmse:0.00360\teval-rmse:0.11271\n", "[262]\ttrain-rmse:0.00355\teval-rmse:0.11272\n", "[263]\ttrain-rmse:0.00349\teval-rmse:0.11272\n", "[264]\ttrain-rmse:0.00342\teval-rmse:0.11273\n", "[265]\ttrain-rmse:0.00337\teval-rmse:0.11272\n", "[266]\ttrain-rmse:0.00333\teval-rmse:0.11272\n", "[267]\ttrain-rmse:0.00328\teval-rmse:0.11273\n", "[268]\ttrain-rmse:0.00324\teval-rmse:0.11274\n", "[269]\ttrain-rmse:0.00319\teval-rmse:0.11272\n", "[270]\ttrain-rmse:0.00313\teval-rmse:0.11272\n", "[271]\ttrain-rmse:0.00308\teval-rmse:0.11272\n", "[272]\ttrain-rmse:0.00303\teval-rmse:0.11273\n", "[273]\ttrain-rmse:0.00300\teval-rmse:0.11273\n", "[274]\ttrain-rmse:0.00297\teval-rmse:0.11273\n", "[275]\ttrain-rmse:0.00293\teval-rmse:0.11273\n", "[276]\ttrain-rmse:0.00288\teval-rmse:0.11273\n", "[277]\ttrain-rmse:0.00283\teval-rmse:0.11274\n", "[278]\ttrain-rmse:0.00278\teval-rmse:0.11273\n", "[279]\ttrain-rmse:0.00273\teval-rmse:0.11274\n", "[280]\ttrain-rmse:0.00268\teval-rmse:0.11273\n", "[281]\ttrain-rmse:0.00264\teval-rmse:0.11274\n", "[282]\ttrain-rmse:0.00259\teval-rmse:0.11273\n", "[283]\ttrain-rmse:0.00255\teval-rmse:0.11273\n", "[284]\ttrain-rmse:0.00251\teval-rmse:0.11273\n", "[285]\ttrain-rmse:0.00248\teval-rmse:0.11272\n", "[286]\ttrain-rmse:0.00243\teval-rmse:0.11272\n", "[287]\ttrain-rmse:0.00240\teval-rmse:0.11272\n", "[288]\ttrain-rmse:0.00236\teval-rmse:0.11272\n", "[289]\ttrain-rmse:0.00233\teval-rmse:0.11272\n", "[290]\ttrain-rmse:0.00230\teval-rmse:0.11272\n", "[291]\ttrain-rmse:0.00228\teval-rmse:0.11272\n", "[292]\ttrain-rmse:0.00224\teval-rmse:0.11271\n", "[293]\ttrain-rmse:0.00220\teval-rmse:0.11271\n", "[294]\ttrain-rmse:0.00217\teval-rmse:0.11271\n", "[295]\ttrain-rmse:0.00214\teval-rmse:0.11271\n", "[296]\ttrain-rmse:0.00211\teval-rmse:0.11271\n", "[297]\ttrain-rmse:0.00208\teval-rmse:0.11271\n", "[298]\ttrain-rmse:0.00205\teval-rmse:0.11270\n", "[299]\ttrain-rmse:0.00202\teval-rmse:0.11270\n", "[300]\ttrain-rmse:0.00199\teval-rmse:0.11270\n", "[301]\ttrain-rmse:0.00196\teval-rmse:0.11271\n", "[302]\ttrain-rmse:0.00192\teval-rmse:0.11271\n", "[303]\ttrain-rmse:0.00190\teval-rmse:0.11271\n", "[304]\ttrain-rmse:0.00189\teval-rmse:0.11271\n", "[305]\ttrain-rmse:0.00185\teval-rmse:0.11272\n", "[306]\ttrain-rmse:0.00182\teval-rmse:0.11272\n", "[307]\ttrain-rmse:0.00179\teval-rmse:0.11273\n", "[308]\ttrain-rmse:0.00176\teval-rmse:0.11273\n", "[309]\ttrain-rmse:0.00175\teval-rmse:0.11273\n", "[310]\ttrain-rmse:0.00173\teval-rmse:0.11273\n", "[311]\ttrain-rmse:0.00170\teval-rmse:0.11274\n", "[312]\ttrain-rmse:0.00168\teval-rmse:0.11274\n", "[313]\ttrain-rmse:0.00165\teval-rmse:0.11274\n", "[314]\ttrain-rmse:0.00163\teval-rmse:0.11274\n", "[315]\ttrain-rmse:0.00160\teval-rmse:0.11275\n", "[316]\ttrain-rmse:0.00158\teval-rmse:0.11275\n", "[317]\ttrain-rmse:0.00155\teval-rmse:0.11275\n", "[318]\ttrain-rmse:0.00154\teval-rmse:0.11275\n", "[319]\ttrain-rmse:0.00152\teval-rmse:0.11275\n", "[320]\ttrain-rmse:0.00150\teval-rmse:0.11275\n", "[321]\ttrain-rmse:0.00148\teval-rmse:0.11275\n", "[322]\ttrain-rmse:0.00145\teval-rmse:0.11276\n", "[323]\ttrain-rmse:0.00143\teval-rmse:0.11275\n", "[324]\ttrain-rmse:0.00141\teval-rmse:0.11275\n", "[325]\ttrain-rmse:0.00138\teval-rmse:0.11275\n", "[326]\ttrain-rmse:0.00136\teval-rmse:0.11276\n", "[327]\ttrain-rmse:0.00134\teval-rmse:0.11275\n", "[328]\ttrain-rmse:0.00132\teval-rmse:0.11276\n", "[329]\ttrain-rmse:0.00130\teval-rmse:0.11276\n", "[330]\ttrain-rmse:0.00128\teval-rmse:0.11276\n", "[331]\ttrain-rmse:0.00127\teval-rmse:0.11275\n", "[332]\ttrain-rmse:0.00125\teval-rmse:0.11275\n", "[333]\ttrain-rmse:0.00123\teval-rmse:0.11275\n", "[334]\ttrain-rmse:0.00121\teval-rmse:0.11275\n", "[335]\ttrain-rmse:0.00119\teval-rmse:0.11276\n" ] } ], "source": [ "gbm = xgb.train(params, dtrain, evals=watchlist, num_boost_round=5000,\n", " early_stopping_rounds=200, verbose_eval=True)" ] }, { "cell_type": "code", "execution_count": 36, "metadata": {}, "outputs": [], "source": [ "x_pred = gbm.predict(xgb.DMatrix(X_test))" ] }, { "cell_type": "code", "execution_count": 38, "metadata": {}, "outputs": [], "source": [ "test['SalePrice'] = np.expm1(x_pred)" ] }, { "cell_type": "code", "execution_count": 41, "metadata": {}, "outputs": [], "source": [ "test[['Id', 'SalePrice']].to_csv('house_pred2.csv', index=False, encoding='utf-8')" ] }, { "cell_type": "code", "execution_count": 30, "metadata": {}, "outputs": [], "source": [ "gbm.save_model('./pretrain_models/house_price_eta0.05_round280.json')" ] }, { "cell_type": "code", "execution_count": 31, "metadata": {}, "outputs": [], "source": [ "gg = xgb.XGBRegressor()" ] }, { "cell_type": "code", "execution_count": 32, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/home/zhaojh/miniconda3/envs/py37/lib/python3.7/site-packages/xgboost/sklearn.py:742: UserWarning: Loading a native XGBoost model with Scikit-Learn interface.\n", " 'Loading a native XGBoost model with Scikit-Learn interface.'\n" ] } ], "source": [ "gg.load_model('./pretrain_models/house_price_eta0.05_round280.json')" ] }, { "cell_type": "code", "execution_count": 40, "metadata": {}, "outputs": [], "source": [ "test['SalePrice'] = np.expm1(gg.predict(X_test))" ] }, { "cell_type": "code", "execution_count": 37, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([11.706002, 12.04607 , 12.116972, ..., 11.978775, 11.649101,\n", " 12.330935], dtype=float32)" ] }, "execution_count": 37, "metadata": {}, "output_type": "execute_result" } ], "source": [ "x_pred" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3.7.13 ('py37')", "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": "3.7.13" }, "orig_nbformat": 4, "vscode": { "interpreter": { "hash": "993bd31d5df1020fab369d79a34ff0a2a159e1798f3e25d3ad4b7751d38184c9" } } }, "nbformat": 4, "nbformat_minor": 2 }