更改建造的随机森林模型
我对R.进行模型培训的新手很陌生。使用它来预测与经过训练的数据略有不同的新数据？
具体而言，这些新数据缺少一个或两个无关紧要的标识符列/变量（例如文件名，用户ID），而我想了解的只是我是否可以从构建模型中获取这些变量，以便我可以获得一些预测没有再次训练模型（因为培训模型需要很长时间）？
I am pretty new to the model training with R. Recently, I have built a random forest model and now I wonder if there's any way that I could modify the already-built model(which was saved as a RDS file), so I can use it to predict on new data which is slightly different from the ones it was trained on?
Specifically, these new data missing one or two insignificant identifier column/variable (e.g. filename, user ID), and all I am trying to understand is if there's any way I can take out these variables from the built model so I can get some prediction without training the model again (as training models takes long time)?
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论