【Kaggle笔记】House Prices: Advanced Regression Techniques

比赛题目

• 房价预测

---

代码

# -*- coding: utf-8 -*-"""Created on Mon May 22 11:32:29 2017"""import pandas as pd train = pd.read_csv('train.csv')test = pd.read_csv('test.csv')# 挑选特征值selected_features = ['Foundation', 'Heating', 'Electrical', 'SaleType', 'SaleCondition', 'GarageArea','YearRemodAdd','YearBuilt','1stFlrSF', '2ndFlrSF', 'TotalBsmtSF', 'BsmtUnfSF', 'CentralAir']X_train = train[selected_features]X_test = test[selected_features]y_train = train['SalePrice']# 补充特征缺失值X_train['Electrical'].fillna('SBrkr', inplace=True)X_train['SaleType'].fillna('WD', inplace=True)X_train['GarageArea'].fillna(X_train['GarageArea'].mean(), inplace=True)X_train['TotalBsmtSF'].fillna(X_train['TotalBsmtSF'].mean(), inplace=True)X_train['BsmtUnfSF'].fillna(X_train['BsmtUnfSF'].mean(), inplace=True)X_test['Electrical'].fillna('SBrkr', inplace=True)X_test['SaleType'].fillna('WD', inplace=True)X_test['GarageArea'].fillna(X_test['GarageArea'].mean(), inplace=True)X_test['TotalBsmtSF'].fillna(X_test['TotalBsmtSF'].mean(), inplace=True)X_test['BsmtUnfSF'].fillna(X_test['BsmtUnfSF'].mean(), inplace=True)print X_train.info()print X_test.info()# 采用DictVectorizer进行特征向量化from sklearn.feature_extraction import DictVectorizerdict_vec = DictVectorizer(sparse=False)X_train = dict_vec.fit_transform(X_train.to_dict(orient='record'))X_test = dict_vec.transform(X_test.to_dict(orient='record'))# 使用随机森林回归模型进行 回归预测#from sklearn.ensemble import RandomForestRegressorfrom sklearn.ensemble import GradientBoostingRegressor#rfr = RandomForestRegressor()rfr = GradientBoostingRegressor()rfr.fit(X_train, y_train)rfr_y_predict = rfr.predict(X_test)# 输出结果rfr_submission = pd.DataFrame({'Id': test['Id'], 'SalePrice': rfr_y_predict})rfr_submission.to_csv('rfr_submission.csv', index=False)