🌐 Mailman 3 February 2018 - scikit-learn - python.org

One-hot encoding
by Sarah Wait Zaranek Aug. 3, 2018

Aug. 3, 2018

Hello - I was just wondering if there was a way to improve performance on the one-hot encoder. Or, is there any plans to do so in the future? I am working with a matrix that will ultimately have 20 million categorical variables, and my bottleneck is the one-hot encoder. Let me know if this isn't the place to inquire. My code is very simple when using the encoder, but I cut and pasted it here for completeness. enc = OneHotEncoder(sparse=True) Xtrain = enc.fit_transform(tiledata) Thanks, Sarah

3 20

transfer-learning for random forests
by peignier sergio March 5, 2018

March 5, 2018

Hello, I recently begun a research project on Transfer Learning with some colleagues. We would like to contribute to scikit-learn incorporating Transfer Learning functions for Random Forests as described in this recent paper: ** *https://arxiv.org/abs/1511.01258* * *Before starting we would like to ensure that no existing project is ongoing. Thanks! BR, Sergio Peignier

2 1

custom loss function in RandomForestRegressor
by Thomas Evangelidis March 1, 2018

March 1, 2018

Greetings, The feature importance calculated by the RandomForest implementation is a very useful feature. I personally use it to select the best features because it is simple and fast, and then I train MLPRegressors. The limitation of this approach is that although I can control the loss function of the MLPRegressor (I have modified scikit-learn's implementation to accept an arbitrary loss function), I cannot do the same with RandomForestRegressor, and hence I have to rely on 'mse' which is not in accordance with the loss functions I use in MLPs. Today I was looking at the _criterion.pyx file: https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/tree/_crit… However, the code is in Cython and I find it hard to follow. I know that for Regression the relevant class are Criterion(), RegressionCriterion(Criterion), and MSE(RegressionCriterion). My question is: is it possible to write a class that takes an arbitrary function "loss(predictions, targets)" to calculate the loss and impurity of the nodes? thanks, Thomas -- ====================================================================== Dr Thomas Evangelidis Post-doctoral Researcher CEITEC - Central European Institute of Technology Masaryk University Kamenice 5/A35/2S049, 62500 Brno, Czech Republic email: tevang(a)pharm.uoa.gr tevang3(a)gmail.com website: https://sites.google.com/site/thomasevangelidishomepage/

5 15

Re: [scikit-learn] New Transformer (Guillaume Lema?tre)
by David Burns Feb. 28, 2018

Feb. 28, 2018

Thanks everyone for your suggested. I will have a look at PipeGraph - which might be a suitable option for us as Guillaume suggested. If it works out, I will share it Thanks David On 02/28/2018 08:29 AM, scikit-learn-request(a)python.org wrote: > Send scikit-learn mailing list submissions to > scikit-learn(a)python.org > > To subscribe or unsubscribe via the World Wide Web, visit > https://mail.python.org/mailman/listinfo/scikit-learn > or, via email, send a message with subject or body 'help' to > scikit-learn-request(a)python.org > > You can reach the person managing the list at > scikit-learn-owner(a)python.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of scikit-learn digest..." > > > Today's Topics: > > 1. New Transformer (David Burns) > 2. Re: New Transformer (Guillaume Lema?tre) > 3. Re: New Transformer (Manuel Castej?n Limas) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Tue, 27 Feb 2018 12:02:27 -0500 > From: David Burns <david.mo.burns(a)gmail.com> > To: scikit-learn(a)python.org > Subject: [scikit-learn] New Transformer > Message-ID: <726f2e70-63eb-783f-b470-5ea45af930e5(a)gmail.com> > Content-Type: text/plain; charset="utf-8"; Format="flowed" > > First post on this mailing list. > > I have been working with time series data for a project, and thought I > could contribute a new transformer to segment time series data using a > sliding window, with variable overlap. I have attached demonstration of > how this would fit in the existing framework. The only challenge for me > here is that the transformer needs to transform both the X and y > variable in order to perform the segmentation. I am not sure from the > documentation how to implement this in the framework. > > Overlapping segments is a great way to boost performance for time series > classifiers, so this may be a worthwhile contribution for some in this > area of ML. Ultimately, model_selection.TimeSeries.Split would need to > be modified to support overlapping segments, or a new class created to > enable validation for this. > > Please let me know if this would be a worthwhile contribution, and if so > how to go about transforming the target vector y in the framework / > pipeline? > > Thanks! > > David Burns > > > >

1 0

New Transformer
by David Burns Feb. 28, 2018

Feb. 28, 2018

First post on this mailing list. I have been working with time series data for a project, and thought I could contribute a new transformer to segment time series data using a sliding window, with variable overlap. I have attached demonstration of how this would fit in the existing framework. The only challenge for me here is that the transformer needs to transform both the X and y variable in order to perform the segmentation. I am not sure from the documentation how to implement this in the framework. Overlapping segments is a great way to boost performance for time series classifiers, so this may be a worthwhile contribution for some in this area of ML. Ultimately, model_selection.TimeSeries.Split would need to be modified to support overlapping segments, or a new class created to enable validation for this. Please let me know if this would be a worthwhile contribution, and if so how to go about transforming the target vector y in the framework / pipeline? Thanks! David Burns

3 2

Google Summer of Code Project- Implementing R Package of SAGA algorithm
by Sirin Baireddy Feb. 23, 2018

Feb. 23, 2018

Hi, I'm attempting to work on a project for google summer of code to create an R package for the SAGA algorithm this summer. I was hoping I could get one of the developers of the algorithm to mentor me this summer while I work on it. Thanks, Sirin Baireddy

2 2

Implement Catalyst-SVRG optimizer for linear models?
by Krishna Pillutla Feb. 22, 2018

Feb. 22, 2018

Hello all, *TL;DR*: I'd like to implement Catalyst-SVRG <https://arxiv.org/pdf/1712.05654.pdf>, an accelerated optimization algorithm for sklearn (or scikit-learn-contrib/lightning, if it is more appropriate). Any feedback? *Long version*: I've been playing around with Catalyst-SVRG <https://arxiv.org/pdf/1712.05654.pdf>, an accelerated stochastic variance reduced optimization algorithm for my research. I've found in my experience and in the experiments section of the attached paper that this algorithm does lead to faster optimization than vanilla (un-accerelated) SVRG <https://papers.nips.cc/paper/4937-accelerating-stochastic-gradient-descent-…>, which itself is much faster than SGD and on roughly the same footing as SAG/SAGA. Moreover, the per-iteration computational complexity of this algorithm practically matches that of SVRG. I was wondering whether it would be beneficial to the community if I implemented this algorithm in sklearn/linear_model or perhaps in scikit-learn-contrib/lightning. I would love to hear your thoughts on this. Cheers, Krishna

1 0

Getting the indexes of the data points after clustering using Kmeans
by prince gosavi Feb. 21, 2018

Feb. 21, 2018

Hi, I have applied Kmeans clustering using the scikit library from kmeans=KMeans(max_iter=4,n_clusters=10,n_init=10).fit(euclidean_dist) After applying the algorithm.I would like to get the data points in the clusters so as to further use them to apply a model. Example: kmeans.cluster_centers_[1] gives me distance array of all the data points. Is there any way around this available in scikit so as to get the data points id/index. Regards

2 2

Clustering with sparse matrix
by Luigi Lomasto Feb. 21, 2018

Feb. 21, 2018

Hi all, I have a sparse matrix where each row (item) has 160 features. For each of them only three or four features are different by 0. Can I do clustering with this data? I’m thinking to use PCA to reduce dimensionality. Thanks for any answer. Luigi

1 0

KMeans cluster
by Shiheng Duan Feb. 20, 2018

Feb. 20, 2018

Hello all, In KMeans cluster, there is a parameter n_init. It shows that the algorithm will run n_init times and output the best. I wonder how to compare the output of each run. Can we get the score for each run? Thanks.

3 3