-
import pandas as pd
-
from sklearn.linear_model import LinearRegression
-
from sklearn.model_selection import train_test_split
-
def data_astype(x):
-
"""数据类型转换"""
-
x['hour'] = x['hour'].astype('float32')
-
x['DEWP'] = x['DEWP'].astype('float32')
-
x['TEMP'] = x['TEMP'].astype('float32')
-
x['PRES'] = x['PRES'].astype('float32')
-
x['Iws'] = x['Iws'].astype('float32')
-
x['Is'] = x['Is'].astype('float32')
-
x['Ir'] = x['Ir'].astype('float32')
-
x['cbwd_NE'] = x['cbwd_NE'].astype('float32')
-
x['cbwd_NW'] = x['cbwd_NW'].astype('float32')
-
x['cbwd_SE'] = x['cbwd_SE'].astype('float32')
-
x['cbwd_cv'] = x['cbwd_cv'].astype('float32')
-
x['year'] = x['year'].astype('float32')
-
x['month'] = x['month'].astype('float32')
-
x['day'] = x['day'].astype('float32')
-
x['week'] = x['week'].astype('float32')
-
if __name__ == "__main__":
-
# 加载数据
-
x = pd.read_csv('src/step1/data/pm25_data.csv', encoding='utf-8')
-
data_astype(x)
-
# 加载标签
-
y = pd.read_csv('src/step1/data/result.csv')['pm2.5']
-
# 对数据进行线性回归预测
-
#####Begin#####
-
reg = LinearRegression().fit(x, y)
-
#####End#####
-
# 加载未来18天的数据
-
x_val = pd.read_csv('src/step1/data/test_1.csv')
-
# 预测
-
y_val_pre = reg.predict(x_val)
-
print('未来18天的pm25预测值：', y_val_pre)
版权声明：本文为qq_44398212原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。