Mailman 3 May 2018 - scikit-learn

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

Scikit Multi learn error.
by aijaz qazi June 26, 2018

June 26, 2018

Dear developer , I am working on web page categorization with http://scikit.ml/ . *Question*: I am not able to execute MLkNN code on the link http://scikit.ml/api/classify.html. I have installed py 3.6. I found scipy versions not compatible with scikit.ml 0.0.5. Which version of scipy would work with scikit.ml 0.0.5. Kindly let me know. I will be grateful. *Regards,* *Aijaz A.Qazi *

3 2

scikit-learn-contrib: building Cython, cpp files
by Dylan Fernando June 11, 2018

June 11, 2018

Hi, I would like to publish this: https://github.com/dil12321/scikit-learn/tree/aode https://github.com/scikit-learn/scikit-learn/pull/11093 as a scikit-learn-contrib project. However, I'm not sure how to write the setup.py file so that aode_helper.cpp and _aode.pyx get included in the package, and run correctly. How should I write setup.py? Regards, Dylan

2 2

PyCM: Multiclass confusion matrix library in Python
by Sepand Haghighi June 6, 2018

June 6, 2018

PyCM is a multi-class confusion matrix library written in Python that supports both input data vectors and direct matrix, and a proper tool for post-classification model evaluation that supports most classes and overall statistics parameters. PyCM is the swiss-army knife of confusion matrices, targeted mainly at data scientists that need a broad array of metrics for predictive models and an accurate evaluation of large variety of classifiers. Github Repo : https://github.com/sepandhaghighi/pycm Webpage : http://pycm.shaghighi.ir/ JOSS Paper : https://doi.org/10.21105/joss.00729

5 7

Possible Feature Suggestions.
by Lekan Wahab May 31, 2018

May 31, 2018

Hello, new contributor here. I've been meaning to contribute to the library for a while now but I haven't found anything easy or clearenough for me to. While going through the source today, I noticed some possible features I could implement and would love to run it by the team here to see which is feasible and which is not. 1. More dataset benchmark. I noticed the benchmarksfolder only has benchmark on one of the datasets. MNIST. My plan is to add benchmarks for more datasets like iris, 'wine' and boston datasets. 2. Implement a Batch Gradient Descent Regressor and a Mini Batch Gradient Regressor just like the Stochastic Gradient Regressor available in the linear_model module. This is really my first attempt at contributing to the package so if there's anything i'm missing about either feature suggestions, please, do let me know.

1 0

Why doesn't sklearn have support for a Batch Gradient Descent Regressor
by Lekan Wahab May 30, 2018

May 30, 2018

I have a feeling this question might have been asked before or there's some sort of resource somewhere on it but so far I haven't found any. I would appreciate any response I can get for this. -- <https://about.me/olamyy?promo=email_sig&utm_source=product&utm_medium=email…> Olamilekan Wahab about.me/olamyy <https://about.me/olamyy?promo=email_sig&utm_source=product&utm_medium=email…>

2 1

Scikit-image 0.14.0 release [cross-listed]
by Egor Panfilov May 29, 2018

May 29, 2018

Hello list, On behalf of `scikit-image` team I'm happy to announce the release of `scikit-image` version 0.14.0 ! This release brings a lot of exciting additions (such as new segmentation tools and algorithms, new datasets and data generating routines), various enhancements (nD support for image moments and regionprops, rescale, resize and pyramid_* function, multichannel support for HOG, multiple performance improvements, and many more) and a number of bugfixes to the code and to the documentation. I am also pleased to inform you that 0.14.x is the last major release with official support for Python 2.7. In order to make the transition to Python 3.x smooth for all the users we're making this release a long-term support one (it will receive important bugfixes and backports for approx. 2 years, following the Python 2.7 end-of-life cycle). For more details on the changes and additions, please, refer to the release notes [1]. I'd like to thank all the developers of the library for their hard work, and all the users for staying with us! Regards, Egor Panfilov, scikit-image core team .. [1] https://github.com/scikit-image/scikit-image/blob/v0.14.x/doc/release/relea… .. [2] https://pypi.org/project/scikit-image/

1 0

DBScan freezes my computer !!!
by Mauricio Reis May 29, 2018

May 29, 2018

The DBScan "fit" method (in scikit-learn v0.19.1) is freezing my computer without any warning message! I am using WinPython 3.6.5 64 bit. The method works normally with the original data, but freezes when I use the normalized data (between 0 and 1). What should I do? Att., Mauricio Reis

9 15

Should we standardize data before PCA?
by Shiheng Duan May 27, 2018

May 27, 2018

Hello all, I wonder is it necessary or correct to do z score transformation before PCA? I didn't see any preprocessing for face image in the example of Faces recognition example using eigenfaces and SVMs, link: http://scikit-learn.org/stable/auto_examples/applications/plot_face_recogni… I am doing on a similar dataset and got a weird result if I standardized data before PCA. The components figure will have a strong gradient and it doesn't make any sense. Any ideas about the reason? Thanks.

3 6

(no subject)
by aijaz qazi May 24, 2018

May 24, 2018

scikit- multi learn is misleading. *Regards,* *Aijaz A.Qazi *

3 2