def train(train_data_dir, labels, sample_num):
f_index, y, X = FeatureFactor.getFeatureSpace(train_data_dir, labels,
sample_num)
saveTheFSpace(f_index)
model = Libsvm.train(y, X)
Libsvm.saveModel(model)
def train( train_data_dir, labels, sample_num ):
f_index, y, X = FeatureFactor.getFeatureSpace( train_data_dir, labels, sample_num )
saveTheFSpace( f_index )
model = Libsvm.train( y, X )
Libsvm.saveModel( model )
def recommend(allWeibos):
model = Libsvm.loadModel()
f_index = loadTheFSpace()
f_vector = FeatureFactor.getFeature(allWeibos, f_index)
label = Libsvm.predict(f_vector, model)
return label
def recommend( allWeibos ):
model = Libsvm.loadModel()
f_index = loadTheFSpace()
f_vector = FeatureFactor.getFeature( allWeibos, f_index )
label = Libsvm.predict( f_vector, model )
return label
