功能介绍

回归评估是对回归算法的预测结果进行效果评估，支持下列评估指标。

SST 总平方和(Sum of Squared for Total)

SSE 误差平方和(Sum of Squares for Error)

SSR 回归平方和(Sum of Squares for Regression)

R^2 判定系数(Coefficient of Determination)

R 多重相关系数(Multiple Correlation Coeffient)

MSE 均方误差(Mean Squared Error)

RMSE 均方根误差(Root Mean Squared Error)

SAE/SAD 绝对误差(Sum of Absolute Error/Difference)

MAE/MAD 平均绝对误差(Mean Absolute Error/Difference)

MAPE 平均绝对百分误差(Mean Absolute Percentage Error)

count 行数

explained variance 解释方差

参数说明

脚本示例

脚本代码

import numpy as np
import pandas as pd
data = np.array([
    [0, 0],
    [8, 8],
    [1, 2],
    [9, 10],
    [3, 1],
    [10, 7]
])
df = pd.DataFrame({"pred": data[:, 0], "label": data[:, 1]})
inOp = BatchOperator.fromDataframe(df, schemaStr='pred int, label int')
metrics = EvalRegressionBatchOp().setPredictionCol("pred").setLabelCol("label").linkFrom(inOp).collectMetrics()
print("Total Samples Number:", metrics.getCount())
print("SSE:", metrics.getSse())
print("SAE:", metrics.getSae())
print("RMSE:", metrics.getRmse())
print("R2:", metrics.getR2())

脚本运行结果

Total Samples Number: 6.0
SSE: 15.0
SAE: 7.0
RMSE: 1.5811388300841898
R2: 0.8282442748091603