[Scikit-learn-general] random forest different misclassification cost

Dear community, Is there a hope that random forest with different misclassification cost will be implemented in scikit-learn? I mean different cost for false positives and false negatives. Like this: http://stats.stackexchange.com/questions/18938/how-to-make-a-randomforest-algorithm-cost-sensitiv

[Scikit-learn-general] Sparse Gradient Boosting & Fully Corrective Gradient Boosting

Is anyone working on making Gradient Boosting Regressor work with sparse matrices? Or is anyone working on adding an option for fully corrective gradient boosting, I.E. all trees in the ensemble are re-weighted at each iteration? These are things I would like to see and may be able to help with i

Re: [Scikit-learn-general] modifying single python file under anaconda

Hi Farhan. That should have an effect. Are you sure this is the folder from which you imported? You can do import sklearn print(sklearn.__path__) to see where it gets imported from. I would recommend to check out the development version in a different folder, do you changes there (in a new br

[Scikit-learn-general] modifying single python file under anaconda

Hi, I am trying to make some changes to gradient_boosting.py for personal experimentation. I am wondering what is the best way to recompile/rebuild the code for the modifications to take effect. I am using scikit-learn 0.15 and using anaconda. I tried directly modifying the python code inside the

Re: [Scikit-learn-general] binarizer with more levels

I don’t know why these emails are so late, I sent them last week!! But thanks anyway, I fixed that issue… From: Joel Nothman [mailto:[email protected]] Sent: Saturday, September 13, 2014 10:05 AM To: scikit-learn-general Subject: Re: [Scikit-learn-general] binarizer with more levels yes, I

Re: [Scikit-learn-general] Backward compat policy in utils

2014-09-15 6:40 GMT-07:00 Mathieu Blondel : > lightning is using the following utils: > > - check_random_state > - safe_sparse_dot > - shuffle > - safe_mask > - sklearn.utils.testing.* > > The latter is not big deal but I like importing assertions from the same > place. > > On a second thought, imp

Re: [Scikit-learn-general] Scikit-learn-general Digest, Vol 56, Issue 42

Hi all, Thanks for the help. 1- With rbf functions I do not receive any error but I am not happy of the obtained result. This is probably just due to my scarce knowledge of SVM and if someone wants to help me we can continue the discussion here http://stats.stackexchange.com/questions/115481/one

Re: [Scikit-learn-general] Bug in one class svm

Coincidentally I implemented and experimented a lot with RBF kernel PCA on various different datasets and gammas. I used the scikit-learn one as reference and comparison and never had any issues with it as long gamma > 0. Maybe it helps if you could post your code and data (if this is okay to sh

Re: [Scikit-learn-general] Bug in one class svm

Actually, you are using nu=.5, which means you are expecting a novelty detection rate up to 50%. You definitely decrease it. With .5 the result will be fairly random . Roberto From: Pagliari, Roberto [mailto:[email protected]] Sent: Monday, September 15, 2014 10:28 AM To: scikit-learn-ge

Re: [Scikit-learn-general] Bug in one class svm

Did you try change the value of nu? Perhaps, it’s too large. From: Pagliari, Roberto [mailto:[email protected]] Sent: Monday, September 15, 2014 10:24 AM To: [email protected] Subject: Re: [Scikit-learn-general] Bug in one class svm I have used it with all kernels a

Re: [Scikit-learn-general] Bug in one class svm

I have used it with all kernels and several values of gamma (including the default) and never had any issue with it, Roberto From: Albert Thomas [mailto:[email protected]] Sent: Monday, September 15, 2014 10:00 AM To: [email protected] Subject: Re: [Scikit-learn-

Re: [Scikit-learn-general] Bug in one class svm

When using the rbf kernel, you should try with a gamma > 0. It seems that you set it to 0. Albert 2014-09-15 15:37 GMT+02:00 Luca Puggini : > Hi, > > there is no segmentation fault in the default settings. > Even if according to the original paper it can make sense to use OCSVM also > with not rb

Re: [Scikit-learn-general] Backward compat policy in utils

lightning is using the following utils: - check_random_state - safe_sparse_dot - shuffle - safe_mask - sklearn.utils.testing.* The latter is not big deal but I like importing assertions from the same place. On a second thought, importing all public utils in __init__.py might quickly become messy

[Scikit-learn-general] Bug in one class svm

Hi, there is no segmentation fault in the default settings. Even if according to the original paper it can make sense to use OCSVM also with not rbf kernel. Maybe there is a bug in the polynomial kernel, I don't know. Despite that also with the RBF kernel I am having some problems with the fron

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

Hi Luca, it segfaults?! Can you confirm that it also segfaults if you use the default arguments? There is no plot so I cannot say anything about the strange decision boundaries. For my part, I've never used something else than a RBF kernel for a one class svm; the RBF kernel has the nice property

[Scikit-learn-general] Bug in OneClassSVM

Hi, I am having some problems with the OneClassSVM function. Here you can see my file and the output. http://justpaste.it/h3pw I am sorry but I can not share the used data. I have experienced also other problems like strange decision boundaries. Can someone tell me if I am doing something wrong