debian-python-pot - Debian package for POT.

Age	Commit message (Collapse)	Author
2017-07-05	add FDA for comparison	Rémi Flamary

2017-07-04	delete useless images	Rémi Flamary

2017-07-04	update thumb image	Rémi Flamary

2017-07-04	add autograd mock again	Rémi Flamary

2017-07-04	add autograd mock	Rémi Flamary

2017-07-04	add autograd mock	Rémi Flamary

2017-07-03	add notebooks to reamde	Rémi Flamary

2017-07-03	add notebooks	Rémi Flamary

2017-07-03	doc update	Rémi Flamary

2017-07-03	add mock in doc conf file	Rémi Flamary

2017-07-03	doc dr	Rémi Flamary

2017-07-03	update doc	Rémi Flamary

2017-07-03	add example norm	Rémi Flamary

2017-07-03	debug bregman stabilized	Rémi Flamary

2017-06-23	add apt update tarvis	Rémi Flamary

2017-06-23	x	Rémi Flamary
	Merge branch 'master' of github.com:rflamary/POT
2017-06-23	test travis	Rémi Flamary

2017-06-23	links 2/2	Rémi Flamary

2017-06-23	add links 1/2	Rémi Flamary

2017-06-20	add clean zeros function for sparse distributions	Rémi Flamary

2017-06-13	example compute emd	Rémi Flamary

2017-06-13	implement paralell sinkhorn stabilized	Rémi Flamary

2017-06-13	implement paralell sinkhorn	Rémi Flamary

2017-06-13	add em2 computation example	Rémi Flamary

2017-06-13	make sinkhorn more general with method selection	Rémi Flamary

2017-06-09	update doc	Rémi Flamary

2017-06-09	update .gitignore with sphinwgallery temop files	Rémi Flamary

2017-06-09	add doc and correct encoding	Rémi Flamary

2017-06-09	add doc ot.gpu.bregman	Rémi Flamary

2017-04-27	Merge pull request #10 from aje/master	ncourty
	performance improvement sinkhorn lpl1
2017-04-24	Remove unnecessary parameter	Leo gautheron

2017-04-24	add GPU implementation sinkhorn lpl1	Leo gautheron

2017-04-21	performance improvement sinkhorn lpl1	Leo gautheron
	- instead of updating individually for each target examples, update for all target examples at once using numpy functions. This allows for a faster computation (for me, divided by 4 on 3000*100 random matricies and random labels in [0,1]). - if I understoud correctly, a value of -1 in the array labels_a meant that we didn't have a label for this example. But in machine learning, we often encounter the binary case where we say we have the positive class (+1) and negative class (-1); thus with a dataset like this, the algorithm wouldn't work as expected. I replaced the default value for 'no label' to '-99' instead of '-1', and I added a parameter to modify it.
2017-04-21	Merge branch 'master' of github.com:rflamary/POT	Rémi Flamary

2017-04-21	small normalization bug	Rémi Flamary

2017-04-20	Update README.md	Rémi Flamary

2017-04-20	Update README.md	Rémi Flamary

2017-04-20	update readme	Rémi Flamary

2017-04-20	add test gpu	Rémi Flamary

2017-04-20	update doc	Rémi Flamary

2017-04-20	add dependencies discussion in readme	Rémi Flamary

2017-04-20	Merge pull request #8 from aje/master	Rémi Flamary
	sinkhorn GPU implementation
2017-04-20	more changes from feeback	Leo gautheron
	in addition add the posibility to normalize the cost matrix through the function fit
2017-04-20	missing argument function sinkhorn	Leo gautheron

2017-04-20	changes from feedback	Leo gautheron

2017-04-20	sinkhorn GPU implementation	Leo gautheron

2017-04-18	Update README.md	Rémi Flamary
	Add Leo as contributor
2017-04-18	Merge pull request #7 from aje/master	Rémi Flamary
	Performance improvement sinkhorn
2017-04-18	little cleanup sinkhorn	Leo gautheron

2017-04-18	Performance improvement sinkhorn	Leo gautheron
	Doing the computation this way is equivalent and allows to reduce the space complexity required from O(max(a, b)^2) to O(a*b) (especially usefull to transport a small number of sources example to a lot of target) This also allows to decrease the computation time.