I want to develop an Learn++ Algorithm same as Adaboost but small variation..I tried to modify the code of example.m but it was not possible.
1 次查看(过去 30 天)
显示 更早的评论
Learn++ Algorithm.For that i will attach an image ..... What i basically need is that an adaboost example should take an user defined inputs in txt file as an datasets and plot result accordingly...i have attached an text file of datasets....this file it should take as an intake.and produce results as same as adaboost.
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/158505/image.jpeg)
0 个评论
回答(1 个)
Ilya
2014-3-14
I don't know what example.m is to which you are referring.
Sounds like you need the AdaBoost algorithm. Step 2 makes me think you want to boost by resampling. Instead of sampling from Dt in step 2, you could use the entire training set and pass weights to the next weak learner. Then you would boost by reweighting.
Both kinds of AdaBoost, by resampling and reweighting, are available from the fitensemble function in the Statistics Toolbox. Use the 'Resample' parameter to turn resampling on or off.
2 个评论
Ilya
2014-3-15
I have pointed you to the code for AdaBoost, and there are various utilities in MATLAB for reading ascii, xls and csv files. Why not start by reading the MATLAB documentation for all these functions?
另请参阅
类别
在 Help Center 和 File Exchange 中查找有关 Classification 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!