scikit-learn Cookbook , Second Edition - Second Edition

How it works...

Now that we've walked through how, is performed in scikit-learn, let's look at how we can use only SciPy, and learn a bit in the process.

First, we need to use SciPy's linalg to perform SVD:

from scipy.linalg import svdimport numpy as npD = np.array([[1, 2], [1, 3], [1, 4]])Darray([[1, 2],[1, 3],[1, 4]])U, S, V = svd(D, full_matrices=False)U.shape, S.shape, V.shape((3L, 2L), (2L,), (2L, 2L))

We can reconstruct the original matrix D to confirm U, S, and V as a decomposition:

np.dot(U.dot(np.diag(S)), V)array([[1, 2],[1, 3],[1, 4]])

The matrix that is actually returned by truncated SVD is the dot product of the U and S matrices. If we want to simulate the truncation, we will drop the smallest singular values and the corresponding ...