[Github]LIBSVM.jl

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

LIBSVM.jl

This is a Julia interface for LIBSVM.

Features:

• Supports all LIBSVM models: classification C-SVC, nu-SVC, regression: epsilon-SVR, nu-SVR and distribution estimation: one-class SVM
• Model objects are represented by Julia type SVM which gives you easy access to model features and can be saved e.g. as JLD file
• Supports ScikitLearn.jl API
  

[url=https://github.com/mpastell/LIBSVM.jl#usage][/url]

本帖隐藏的内容

https://github.com/mpastell/LIBSVM.jl

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

1. LIBSVM API
   
2. 
   
3. This provides a lower level API similar to LIBSVM C-interface. See ?svmtrain for options.
   
4. 
   
5. using RDatasets, LIBSVM
   
6. 
   
7. # Load Fisher's classic iris data
   
8. iris = dataset("datasets", "iris")
   
9. 
   
10. # LIBSVM handles multi-class data automatically using a one-against-one strategy
    
11. labels = convert(Vector, iris[:Species])
    
12. 
    
13. # First dimension of input data is features; second is instances
    
14. instances = convert(Array, iris[:, 1:4])'
    
15. 
    
16. # Train SVM on half of the data using default parameters. See documentation
    
17. # of svmtrain for options
    
18. model = svmtrain(instances[:, 1:2:end], labels[1:2:end]);
    
19. 
    
20. # Test model on the other half of the data.
    
21. (predicted_labels, decision_values) = svmpredict(model, instances[:, 2:2:end]);
    
22. 
    
23. # Compute accuracy
    
24. @printf "Accuracy: %.2f%%\n" mean((predicted_labels .== labels[2:2:end]))*100

复制代码

1. ScikitLearn API
   
2. 
   
3. You can alternatively use ScikitLearn.jl API with same options as svmtrain:
   
4. 
   
5. using LIBSVM
   
6. import RDatasets
   
7. 
   
8. #Classification C-SVM
   
9. iris = dataset("datasets", "iris")
   
10. labels = iris[:, 5]
    
11. instances = convert(Matrix{Float64}, iris[:, 1:4]')
    
12. model = fit!(SVC(), instances[:,1:2:end], labels[1:2:end])
    
13. yp = predict(model, instances[:, 2:2:end])
    
14. 
    
15. #epsilon-regression
    
16. whiteside = RDatasets.dataset("MASS", "whiteside")
    
17. X = Array(whiteside[:Gas]')
    
18. y = Array(whiteside[:Temp])
    
19. svrmod = fit!(EpsilonSVR(cost = 10., gamma = 1.), X, y)
    
20. yp = predict(svrmod, X)

复制代码

顶一顶

谢谢楼主分享！

谢谢分享