python - how PCA fit transform on test set -

i using from sklearn.decomposition import pca library, incrementalpca reduce dimensionality of problem follows:

training_data = [...]  training_target = [...] test_data = [...] test_target = [...]  ipca = incrementalpca(n_components, batch_size) new_training_data = ipca.fit_transform(training_data) 

to run tests given classifier, need fit test set information obtained in training set (something eigenvalues , eigenvectors), reduce same size of new training set. how can this library (or other), since ipca.fit_transform(data) not return me eg eigpairs or value resize dimension of test set?

the transform internal incrementalpca object after call fit or fit_transform. when call icpa.fit_transform, telling determine principal components transform given data , apply transform data. transform data set, use transform method of trained incrementalpca object:

new_test_data = icpa.transform(test_data)