Re: [Scikit-learn-general] lasso_path vs lars_path

Btw, let >>> alphas, _, coefs = lars_path(X, y) Wouldn't it be nice to have 0 in alphas ? (with the plain linear regression as its coefs). That would make interpolation implementation more straightforward, and that way, the regularization path would be totally defined. As of now, it is not possib

Re: [Scikit-learn-general] EllipticEnvelope giving error IndexError: 0-d arrays can't be indexed

You can see that your covariance matrix has two eigenvalues that are almost 0, therefore the determinant of the precision (inverse covariance matrix) degenerates to "infinity", which explain that the comparison fails. I will try to add some code to catch this type of error and try to overcome it.

Re: [Scikit-learn-general] lasso_path vs lars_path

+1 look for interpolate in least_angle.py to see where we already use the linear interpolation Alex On Thu, Jun 28, 2012 at 6:42 PM, Olivier Grisel wrote: > 2012/6/28 Charles-Pierre Astolfi : >> Hi scikiters, >> >> I am trying to compute lasso/lars for several values of the >> regularization pa

Re: [Scikit-learn-general] lasso_path vs lars_path

2012/6/28 Charles-Pierre Astolfi : > Hi scikiters, > > I am trying to compute lasso/lars for several values of the > regularization parameter. > > A clean way to do this, is like this: alphas = [some long list of positive values] models = linear_model.lars_path(X, y, alphas=alphas) Nope

[Scikit-learn-general] lasso_path vs lars_path

Hi scikiters, I am trying to compute lasso/lars for several values of the regularization parameter. A clean way to do this, is like this: >>> alphas = [some long list of positive values] >>> models = linear_model.lars_path(X, y, alphas=alphas) Unfortunately, it seems to be very slow, compared to

Re: [Scikit-learn-general] EllipticEnvelope giving error IndexError: 0-d arrays can't be indexed

np.dot(X.T, X) is invertible.None of the features have a 0 variance and I don't think they are correlated. "Would you look at the eigenvalues of this thing please?" I calculated the eigenvalues in the following manner. Sorry, I am not sure if this is what you wanted me to do or what I should look

Re: [Scikit-learn-general] EllipticEnvelope giving error IndexError: 0-d arrays can't be indexed

> > It may come from the fact that np.dot(X, X.T) is not invertible. > I meant, of course, np.dot(X.T, X). Would you look at the eigenvalues of this thing please? On Thu, Jun 28, 2012 at 10:33 AM, Virgile Fritsch wrote: > Hi, > > It may come from the fact that np.dot(X, X.T) is not invertible.

Re: [Scikit-learn-general] nosetests sklearn errors, e-mailing the list as requested

2012/6/28 Michael Anuzis : > Dear Scikit-Learn list, > > Excited to learn Scikit-Learn and am installing the tutorials. I ran > nosetests sklearn as directed in the Tutorial Setup and got multiple errors > & failures. E-mailing this list with system specs as requested: > > Ran 937 tests in 148.572s

Re: [Scikit-learn-general] nosetests sklearn errors, e-mailing the list as requested

On Thu, Jun 28, 2012 at 08:04:24PM +0900, Michael Anuzis wrote: > Dear Scikit-Learn list, > Excited to learn Scikit-Learn and am installing the tutorials. I ran > *nosetests > sklearn *as directed in the Tutorial > Setup

[Scikit-learn-general] nosetests sklearn errors, e-mailing the list as requested

Dear Scikit-Learn list, Excited to learn Scikit-Learn and am installing the tutorials. I ran *nosetests sklearn *as directed in the Tutorial Setup and got multiple errors & failures. E-mailing

Re: [Scikit-learn-general] EllipticEnvelope giving error IndexError: 0-d arrays can't be indexed

Hi, It may come from the fact that np.dot(X, X.T) is not invertible. Is it possible that two (or more) features are correlated? Or that one (or more) features have a 0 variance? If yes, try to remove those features. Let us know if you think this could be the reason why it fails. Virgile On Thu

Re: [Scikit-learn-general] FastICA question

hi, can you post a code snippet in a gist to reproduce the problem? Alex On Wed, Jun 27, 2012 at 11:44 PM, Igor Nikolskiy wrote: > Hi Andy, > > Thanks for the fast response! >> >> Could you please say which version of sklearn you are using? > >  I am using sklearn version 0.11 > >> About the na