python - Scikit-learn (sklearn) PCA throws Type Error on sparse matrix -
from documentation of sklearn randomizedpca, sparse matrices accepted input. when called sparse matrix, got typeerror :
> sklearn.__version__ '0.16.1' > pca = randomizedpca(n_components=2) > pca.fit(my_sparce_mat) typeerror: sparse matrix passed, dense data required. use x.toarray() convert dense numpy array. i obtained same error using fit_transform.
any suggestion on how have work?
the answer is not possible have randomizedpca work sparse matrix version 0.16.1 of scikit-learn (current stable version). documentation referring previous version of scikit-learn , alternative functions should used current stable version.
a possible alternative truncatedsvd
Comments
Post a Comment