Re: [Scikit-learn-general] Complex Valued Data and PCA

easiest would be to associate each one of your complex features to an R^2 real feature. You can do this with np.hstack([np.real(X), np.imag(X)]). This has more degrees of freedom in a sense, since you are not constrained by complex multiplication anymore. If however this is essential to you I'd sug

[Scikit-learn-general] Complex Valued Data and PCA

Hello all, I have a complex valued array that I would like to do PCA on, but I am running into an issue where sklearn is casting the complex128 array as a float64. X = np.random.random((30, 50)) + 1j * np.random.random((30, 50)) pca = RandomizedPCA() pca.fit(X) ComplexWarning: C

Re: [Scikit-learn-general] pomegranate v0.4.0 released

Oh, that looks like an awesome package, thanks for sharing! PS: Just noticed that there's a little problem on the readthedocs page, the Edit on GitHub button links to https://github.com/jmschrei/pomegranate/blob/master/docs/source/index.rst which doesn't exist. > On Mar 30, 2016, at 12:52 PM,

[Scikit-learn-general] pomegranate v0.4.0 released

Hello all! I've just released a new version of pomegranate, which is a probabilistic modelling package for Python with a speedy cython implementation. It currently supports the following: * a wide range of probability distributions * general mixture models * hidden markov models * naive bayes * m

[Scikit-learn-general] Announcing scikit-learn-contrib

Dear scikit-learners, The scikit-learn team is happy to announce the creation of scikit-learn-contrib, a github organization for gathering high-quality scikit-learn compatible projects. https://github.com/scikit-learn-contrib scikit-learn-contrib currently includes two projects: - lightning: ht

[Scikit-learn-general] Feature selection using Information Gain and Informatin Gain Ratio

Dear all, I've created a pull request with new functions to perform feature selection using IG and GR: https://github.com/scikit-learn/scikit-learn/pull/6534 These are two popular feature selection methods, it would be great if scikit-learn implemented those. The implementation closely follows th

Re: [Scikit-learn-general] Confidence-weighted learning

Hi Mathieu, thanks for the response and the feedback. It is correct that there are other more recent algorithms available, on the other hand CW learning only requires minor extensions from passive-aggressive learning (PA) which is already available in scikit-learn and it achieves very competitive