回归树

训练集给出如下数据：

 测试集提供其中的部分列：

要求预测以下列的数据：

['Tdewpoint', 'Visibility', 'Windspeed', 'RH_out', 'Press_mm_hg', 'RH_9', 'T_out', 'RH_4']

使用回归树进行预测：

import pandas as pd
from sklearn.tree import DecisionTreeRegressor
from sklearn.multioutput import MultiOutputRegressor

# 读入训练集和测试集数据
train_data = pd.read_csv('train_dataset.csv')
test_data = pd.read_csv('test_dataset.csv')

li=train_data.columns.to_list()[2::]
goal=['Tdewpoint', 'Visibility', 'Windspeed', 'RH_out', 'Press_mm_hg', 'RH_9', 'T_out', 'RH_4']
feature=list(set(li)-set(goal))
print(li)
print(feature)

# 从训练集中分离出目标变量和特征变量
#X_train = train_data.drop(goal, axis=1)
X_train = train_data[feature]
y_train = train_data[goal]

# 创建决策树回归模型并拟合训练集
model = MultiOutputRegressor(DecisionTreeRegressor())
model.fit(X_train, y_train)

# 使用模型对测试集进行预测
X_test = test_data[feature]
y_pred = model.predict(X_test)

# 将预测结果保存为CSV文件
submission = pd.DataFrame(y_pred, columns=goal)
submission.to_csv('test_result.csv', index=False)