Re: [Scikit-learn-general] Joblib compressed file error?

2013-01-29 Thread Afik Cohen
Ronnie Ghose writes: > > > do you think it isn't saving correctly or it isn't loading correctly? > > On Thu, Jan 24, 2013 at 8:14 PM, Ark wrote: > Gael Varoquaux ...> writes: > > > > On Wed, Jan 23, 2013 at 12:16:32AM +, Afik Cohen wrote: &g

Re: [Scikit-learn-general] Joblib compressed file error?

2013-01-22 Thread Afik Cohen
Gael Varoquaux writes: > > On Tue, Jan 22, 2013 at 10:30:01PM +, Ark wrote: > > /home/n7/newenv/lib/python2.6/site- > > packages/sklearn/externals/joblib/numpy_pickle.pyc in read_zfile(file_handle) > > 69 assert len(data) == length, ( > > 70 "Incorrect data length whil

Re: [Scikit-learn-general] Implementing n_jobsfor OneVsRestClassifier

2012-12-04 Thread Afik Cohen
> > Will this let us run SGDClassifier and show us per-class probability outputs? > > Again, that's the only reason we've been using OneVsRestClassifier. Let me > > explain what I mean by per-class probability, just in case it isn't clear: > > > > SGDClassifier's predict_proba() returns probabilit

Re: [Scikit-learn-general] Implementing n_jobsfor OneVsRestClassifier

2012-12-03 Thread Afik Cohen
Andreas Mueller writes: > > On 12/03/2012 09:39 PM, Afik Cohen wrote: > > No, we aren't doing multi-label classification, just multiclass. He was saying > > we could just use SGDClassifier directly, which is true, but AFAIK there is no > > way to get good predic

Re: [Scikit-learn-general] Implementing n_jobsfor OneVsRestClassifier

2012-12-03 Thread Afik Cohen
Andreas Mueller writes: > > Am 29.11.2012 23:45, schrieb Afik Cohen: > > > > > > Hey Mathieu! > > > > Pretty much the only reason we wrap SGDClassifier in a OneVsRestClassifier is so > > we can get predict_proba results on a per class basis. This

Re: [Scikit-learn-general] Implementing n_jobsfor OneVsRestClassifier

2012-11-29 Thread Afik Cohen
Mathieu Blondel writes: > > > On Thu, Nov 29, 2012 at 10:39 AM, Afik Cohen wrote: > > It's easy to see how with some slight modifications (wrapping that in a joblib > Parallel() call) we could enable n_jobs for OneVsRestClassifier. This almost > seems too simpl

[Scikit-learn-general] Implementing n_jobs for OneVsRestClassifier

2012-11-28 Thread Afik Cohen
, so there must be a good reason why this isn't done; could you give your opinions on this? Thanks, Afik Cohen -- Keep yourself connected to Go Parallel: VERIFY Te

Re: [Scikit-learn-general] Prediction Probabilities in LinearSVC with scikit-learn >0.12

2012-10-30 Thread Afik Cohen
* Woops, my previous reply got munged up, so I'm resubmitting it. Please ignore my previous messed up email. > 2012/10/30 Afik Cohen : > >> Do you know what they are doing? I would expect they just do a soft-max. > > I don't. :) But according to the LIBLINEAR FAQ:

Re: [Scikit-learn-general] Prediction Probabilities in LinearSVC with scikit-learn >0.12

2012-10-30 Thread Afik Cohen
Hi Lars, Thanks for your reply. > > 2012/10/30 Afik Cohen : > >> Do you know what they are doing? I would expect they just do a soft-max. > > I don't. :) But according to the LIBLINEAR FAQ: "If you really would like to > > have probability outputs for SVM

Re: [Scikit-learn-general] Prediction Probabilities in LinearSVC with scikit-learn >0.12

2012-10-30 Thread Afik Cohen
or something, not classifying inputs to discrete classes, right? We're classifying emails into ~1200 distinct classes, so Logistic Regression is meaningless for us (in fact, when we tried it, it achieved a hilarious 48% cross-validated k=3 accuracy. LinearSVC achieves 95% accuracy.) Thanks again.

[Scikit-learn-general] Prediction Probabilities in LinearSVC with scikit-learn >0.12

2012-10-30 Thread Afik Cohen
o continue doing so with current and future versions of scikit! Thanks, Afik Cohen Abhijeet Kolhe LinearSVC prediction probability patch follows: diff --git a/sklearn/svm/classes.py b/sklearn/svm/classes.py index 79cb76d..d432792 100644 --- a/sklearn/svm/classes.py +++ b/sklearn/svm/classes.py @@ -1