Get non normalized eigenvectors in scipy -

scipy , numpy returns eigenvectors normalized. trying use vectors physical application , need them not normalized.

for example a = np.matrix('-3, 2; -1, 0') w,v = spl.eig(a)

scipy returns eigenvalues (w) of [-2,-1] , modal matrix (v) (eigenvalues columns) [[ 0.89442719 0.70710678][ 0.4472136 0.70710678]]

i need original modal matrix [[2 1][1 1]]

according various related threads (1) (2) (3), there no such thing "non normalized" eigenvector.

indeed, eigenvector v corresponding eigenvalue l of matrix a defined by,

a*v = l*v 

and can therefore multiplied scalar , remain valid.

while depending on algorithm, computed eigenvector can have norm different 1, not hold particular meaning (physical or otherwise), , should not relied on. customary return normalized eigenvector in numerical libraries (scipy, r, matlab, etc).