使用sklearn对预测问题进行评价

MSE、RMSE、MAER、R_Square可直接调用sklearn（scikit-learn的简称）进行计算
代码如下：

import numpy as np
# [45,52,5,32,10,39,40,51,34,35,8,45,8,47,8,61,62]
y_predict = np.array([35,8,45,8,47,8,61,62])  # 预测值
y_output = np.array([18,32,55,32,40,39,33,66])  # 真实值

# 利用sklearn实现
from sklearn import metrics
from sklearn.metrics import r2_score

MSE = metrics.mean_squared_error(y_predict, y_output)
RMSE = metrics.mean_squared_error(y_predict, y_output)**0.5
MAE = metrics.mean_absolute_error(y_predict, y_output)
R_Squared = r2_score(y_predict, y_output)

print("mse:", MSE)
print("rmse:", RMSE)
print("mae:", MAE)
print("r2:", R_Squared)

MAPE、SMAPE不能直接调用，需要自己定义函数进行计算
代码如下：

在这def mape(y_predict, y_output):
    return np.mean(np.abs((y_predict-y_output)/y_output))*100
def smape(y_predict, y_output):
    return 2 * np.mean(np.abs(y_predict-y_output)/(np.abs(y_predict)+np.abs(y_output)))*100

print("mape:", mape(y_predict, y_output))
print("smape:", smape(y_predict, y_output))  # 若输出结果为67，即结果是67%里插入代码片