多项式回归 scikit
In a typical regression or a classification scenario, we have a set of the independent variable and one or more dependent variables. Sometimes it is not possible to predict all the dependent variable values together. Due to the correlations among the targets, we need to consider the predicted value of the other dependent variable in the sequence to predict the next target value. Hold your breath, and it will become clear with a detailed explanation and example discussed in this article. All variables are also known as features and the dependent variables as the targets.
在典型的回归或分类方案中,我们有一组自变量和一个或多个因变量。 有时不可能一起预测所有因变量值。 由于目标之间的相关性,我们需要考虑序列中其他因变量的预测值以预测下一个目标值。 屏住呼吸,本文将提供详细的解释和示例,这将使您变得清晰。 所有变量也称为特征,因变量为目标。
To make predictions with machine learning in such scenarios, we can put together a sequential regression model with Regressorchain in the Scikit-Learn
为了在这种情况下通过机器学习进行预测,我们可以在Scikit-Learn 中将顺序回归模型与Regressorchain组合在一起
The first model is trained based on the independent features and the first dependent variable in the sequence of target variables. Other dependent variable values are not taken into consideration while training the first model.
基于独立特征和目标变量序列中的第一因变量来训练第一模型。 训练第一个模型时不考虑其他因变量值。