Re: [Scikit-learn-general] Bug in BernoulliRBM

2014-02-17 Thread Lars Buitinck
2014-02-10 17:42 GMT+01:00 Pedro Cardoso : > I believe there is another one at line 188 > > return - np.dot(v, self.intercept_visible_) - ... I think that's fixed in master too. -- Android apps run on BlackBerry 10 Intro

Re: [Scikit-learn-general] Bug in BernoulliRBM

2014-02-17 Thread Pedro Cardoso
I believe there is another one at line 188 return - np.dot(v, self.intercept_visible_) - ... ValueError: setting an array element with a sequence In my case, v is a sparse matrix . 2014-02-10 17:17 GMT+01:00 Lars Buitinck : > 2014-02-10 12:56 GMT+01:00 Pedro Cardoso : > > I am using the vers

Re: [Scikit-learn-general] Bug in BernoulliRBM

2014-02-10 Thread Lars Buitinck
2014-02-10 12:56 GMT+01:00 Pedro Cardoso : > I am using the version 0.14 > > I believe that there is a bug in defining the slices for the baches. ex: on > a matrix qith 1078 rows, the last batch is from 1070 to 1080. > > Created with : > batch_slices = list(gen_even_slices(n_batches * self.batch_si

[Scikit-learn-general] Bug in BernoulliRBM

2014-02-10 Thread Pedro Cardoso
I am using the version 0.14 I believe that there is a bug in defining the slices for the baches. ex: on a matrix qith 1078 rows, the last batch is from 1070 to 1080. Created with : batch_slices = list(gen_even_slices(n_batches * self.batch_size, n_batch