Tweets
- Tweets
- Tweets & replies
- Photos & videos
@glouppe is blocked
Are you sure you want to view these Tweets? Viewing Tweets won't unblock @glouppe.
A manual for tree huggers: My PhD thesis "Understanding Random Forests: From Theory to Practice" http://orbi.ulg.ac.be/handle/2268/170309 …
51 retweets 97 favoritesLooks like physicists are already using Scikit-Learn for real HEP analysis! http://www-public.slac.stanford.edu/scidoc/docMeta.aspx?slacPubNumber=SLAC-PUB-15983 … CC: @betatim @balazskegl @particleist
Interview with @ylecun on ConvNets in general and the Kaggle CIFAR-10 challenge http://blog.kaggle.com/2014/12/22/convolutional-nets-and-cifar-10-an-interview-with-yan-lecun/ …
The scikit-learn paper recently reached 1000+ citations. Not bad considering we published it only 4 years ago!
9 retweets 14 favoritesCause And Effect: The Revolutionary New Statistical Test That Can Tease Them Apart https://medium.com/the-physics-arxiv-blog/cause-and-effect-the-revolutionary-new-statistical-test-that-can-tease-them-apart-ed84a988e?source=tw-78dde0e7942-1419098713587 …
16 retweets 24 favoritesApproximate Nearest Neighbors search with LSH Forest has landed in sklearn master: http://scikit-learn.org/dev/modules/neighbors.html#approximate-nearest-neighbors …
28 retweets 38 favorites"Further to new unacceptable contractual conditions, EPFL decided not to renew its subscription to Science Online" http://library.epfl.ch/en/news/?id=441
6 retweets 2 favoritesLasagne, a simple library to use deep neural nets in #Theano.
Promising... #machinelearning
https://github.com/benanne/Lasagne
PS: Love the name ;-)
Wrote a tutorial: Using convolutional neural nets to detect facial keypoints. http://danielnouri.org/notes/2014/12/17/using-convolutional-neural-nets-to-detect-facial-keypoints-tutorial/ …
#DeepLearning #Lasagne #Theano #Python
You can now tame your favorite SLURM or SGE supercomputer with clusterlib. The 0.1 is out. https://pypi.python.org/pypi/clusterlib/0.1 …
7 retweets 8 favoritesOverheard at #NIPS2014 "Why should I care about neuroscience, when I know an engineered solution can beat the brain on any given problem?"
Deeppy, a "keep it simple" approach to doing deep learning in Python https://github.com/andersbll/deeppy … Early stages but looks pretty sweet :)
46 retweets 87 favoritesRT @IEA: #CO2 emissions from #electricity and heat almost doubled between 1990 and 2012 http://bit.ly/1wxWzUh pic.twitter.com/1u7mKgiFvk
New fast online clustering in scikit-learn: BIRCH
http://scikit-learn.org/dev/modules/clustering.html#birch …
http://scikit-learn.org/dev/auto_examples/cluster/plot_birch_vs_minibatchkmeans.html …
Thanks @swagKumar
How to fool a computer. http://arxiv.org/pdf/1412.1897v1.pdf … pic.twitter.com/QpD8vDwFRY
326 retweets 294 favoriteshttp://contentmine.org/ Mining scientific facts from the literature, with @petermurrayrust
My slides of yesterday Paris ML meetup on Bias-variance decomposition in Random Forests http://www.slideshare.net/glouppe/slides2-42551950 …
14 retweets 17 favoritesCode for Mondrian Forests https://github.com/balajiln/mondrianforest … #NIPS2014
particle physics, collaborative statistical modeling, open science, family man
Spoilers: The best classifiers in the world are Random Forests & SVMs w/ gaussian kernel. Answer to the title is no. http://jmlr.csail.mit.edu/papers/volume15/delgado14a/delgado14a.pdf …
10 retweets 6 favoritesTwitter may be over capacity or experiencing a momentary hiccup.
Visit Twitter Status for more information.