OneAgainstRestClassifier.predict will accept design matrices with the
wrong number of features. Here I train on 1000 features and test on
1500. It quite happily returns predictions instead of raising an error
as SVC would. Same thing happens if I test on 500 features.

import numpy as np
from sklearn.multiclass import OneVsRestClassifier
from sklearn.svm import SVC

rng = np.random.RandomState([1,2,3])

X = rng.randn(1000,1000)
y = rng.randint(0,10,(1000,))

svm = OneVsRestClassifier(SVC(kernel = 'linear', C = 1.0)).fit(X,y)

X2 = rng.randn(1000,1500)

print svm.predict(X2).shape

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

Reply via email to