m******n 发帖数: 453 | 1 training data一般都需要standardized
建立好模型之后
那么这时候去predict test data
test data也需要一样的标准化处理吗
是不是用原始数据的test data
可是,如果对test data也进行scale的话
predict出来的y,也是标准化的值啊,怎样recover到原来的值去 |
t******g 发帖数: 2253 | 2 我觉得要对test做同样的transformation |
g**********l 发帖数: 9 | 3 after transformation and model fitting, you may apply an inverse
transformation to get a model based on the original model.
Say, if you new x is ax, where x is the original explanatory variable and
you get y=b*(new x) as the fitted model, then equivalently, it is y=(ba)*x
in terms of original variable. You can use this model for test data. |
s*********h 发帖数: 6288 | 4 python sklearn包里有很不错的函数可以用。 |
m******n 发帖数: 453 | 5 建模时候,需要对y也normalization吗,还是只需要对predictor X进行scale就可以了
? |