Re: [Scikit-learn-general] Convolutive NMF

2015-04-09 Thread Dan Stowell
FWIW I didn't have a very good experience when working with > convolutional (shouldn't it be that?) NMF. Why no use an > autoencoder approach? Thanks for the suggestion. Exploring a few different paradigms at the moment - Best Dan > Andy > > > On 04/09/2015 08:20 AM,

[Scikit-learn-general] Convolutive NMF

2015-04-09 Thread Dan Stowell
.ie/1375/1/getPDF2.pdf Thanks Dan -- Dan Stowell EPSRC Research Fellow Centre for Digital Music Queen Mary, University of London Mile End Road, London E1 4NS http://www.mcld.co.uk/research/ -- BPM Camp - Free Virtual W

[Scikit-learn-general] Matching Pursuit Toolkit (MPTK) 0.7 released

2013-05-15 Thread Dan Stowell
signals directly from within Python. Best Dan -- Dan Stowell Postdoctoral Research Assistant Centre for Digital Music Queen Mary, University of London Mile End Road, London E1 4NS http://www.elec.qmul.ac.uk/digitalmusic/people/dans.htm http://www.mcld.co.uk

[Scikit-learn-general] Fwd: IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events - deadline 31st March

2013-01-29 Thread Dan Stowell
info/aasp-challenge>. Best wishes, The organisers -- Dimitrios Giannoulis (QMUL), Emmanouil Benetos (CityU/QMUL), Dan Stowell (QMUL), Mathias Rossignol (IRCAM), Mathieu Lagrange (IRCAM) and Mark D. Plumbley (QMUL) -- M

Re: [Scikit-learn-general] mixture.GMM and data ranges

2012-10-31 Thread Dan Stowell
On 31/10/12 16:09, bthirion wrote: > On 10/31/2012 04:50 PM, Dan Stowell wrote: >> Hi all, >> >> I'm still getting odd results using mixture.GMM depending on data >> scaling. In the following code example, I change the overall scaling but >> I do NOT change

Re: [Scikit-learn-general] mixture.GMM and data ranges

2012-10-31 Thread Dan Stowell
27;, n_components=10) 7094.87886779 GMM(cvtype='diag', n_components=10) -14681.566456 GMM(cvtype='diag', n_components=10) -37576.4496656 In principle, I don't think the overall data scaling should matter, but maybe there's an implementation issue I'm overlooking? Th

Re: [Scikit-learn-general] mixture.GMM and data ranges

2012-10-02 Thread Dan Stowell
On 02/10/12 13:58, Alexandre Passos wrote: > On Tue, Oct 2, 2012 at 7:48 AM, Dan Stowell > wrote: >> >> Hi all, >> >> I'm using the GMM class as part of a larger system, and something is >> misbehaving. Can someone confirm please: the results of u

[Scikit-learn-general] mixture.GMM and data ranges

2012-10-02 Thread Dan Stowell
as a range 0-1, that difference shouldn't have much bearing? Thanks Dan -- Dan Stowell Postdoctoral Research Assistant Centre for Digital Music Queen Mary, University of London Mile End Road, London E1 4NS http://www.elec.qmul.ac.uk/digitalmusic/people/dans.htm http:/

[Scikit-learn-general] Probability hypothesis density (PHD) filter

2012-03-22 Thread Dan Stowell
Hi - I'm looking for Python implementations of the PHD filter (Probability Hypothesis Density filter). The acronym makes web search quite hard! It's a technique for multi-object tracking in noisy observations. Aware of any python code for PHD filtering, please? Thanks Dan -- D

[Scikit-learn-general] linear_model.BayesianRidge: retrieving the MAP probability

2011-11-09 Thread Dan Stowell
mplied by the prior on the coefficients. * Given the above, is it correct to include log(clf.alpha) in the probability calculation? Thanks for any advice you can give - Dan -- Dan Stowell Postdoctoral Research Assistant Centre for Digital Music Queen Mary, University of London Mile End Road,