Hi:
Is there a lower bound for n_neighbors when using the the Local
Tangent Space Alignment method with LLE?
I get this warning
/usr/local/lib/python2.6/dist-packages/scipy/linalg/decomp_lu.py:59:
RuntimeWarning: Diagonal number 198 is exactly zero. Singular matrix.
RuntimeWarning)
when I run the following code
###############################
import numpy as np
from sklearn import manifold
n = 1000;
m = 3;
X = np.random.rand(n,m)
out_dim = 2
n_neighbors = 6
print 'Using random data and LTSA'
Y = manifold.LocallyLinearEmbedding(n_neighbors, out_dim,
method='ltsa').fit_transform(X)
###############################
It works OK with a bigger n_neighbors. Are the results useful if I get
this warning?
Alejandro.
------------------------------------------------------------------------------
RSA(R) Conference 2012
Save $700 by Nov 18
Register now
http://p.sf.net/sfu/rsa-sfdev2dev1
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
