debian-python-pot - Debian package for POT.

Age	Commit message (Collapse)	Author
2017-06-09	add doc and correct encoding	Rémi Flamary

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

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	small normalization bug	Rémi Flamary

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	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.
2017-04-13	update bregman file	Leo gautheron
	- change commented prints to python3 compatible syntax - Correct issue that could cause the sinkhorn algo to stop with u and v containing nan/infinite numbers: - Assign uprev and vprev before changing u and v. - Then update u and v. - Then check if u and v contain nan, but ALSO infinite values. - if there are issues, then display error (with 2 r, not 3 :p) along with the iteration number (there may have errors at iteration 0)
2017-04-07	doc update	Rémi Flamary

2017-04-07	V0.2	Rémi Flamary

2017-04-07	doc wda	Rémi Flamary

2017-04-07	better dicumentation	Rémi Flamary

2017-04-07	add WDA	Rémi Flamary

2017-04-07	avoid filename conclict in windows	Rémi Flamary

2017-03-21	cleanupt cpp wrapper name	Rémi Flamary

2017-03-14	add emd2+ multiproc	Rémi Flamary

2017-03-10	runs but not quicker	Rémi Flamary

2017-03-10	first try	Rémi Flamary

2017-02-07	log return for sinkhorn	Rémi Flamary

2017-01-05	version 0.1.12	Rémi Flamary

2017-01-05	add tic() and toc() functions for easy coarse timing	Rémi Flamary

2017-01-05	more efficient sinkhorn	Rémi Flamary

2016-12-02	small update	Rémi Flamary

2016-12-02	cleanup pyflakes issue #23	Rémi Flamary

2016-12-01	update barycenter demo	Rémi Flamary

2016-11-09	da with GL	Nicolas Courty

2016-11-08	da with GL	Nicolas Courty

2016-11-08	travis CI	Rémi Flamary

2016-11-07	gcg	Nicolas Courty

2016-11-07	v0.1.11	Rémi Flamary

2016-11-07	v0.1.11	Rémi Flamary

2016-11-07	v0.1.10	Rémi Flamary

2016-11-07	add log and epsilon scaling stabilizations	Rémi Flamary

2016-11-04	working	Rémi Flamary

2016-11-04	etter doc for classes	Rémi Flamary

2016-11-04	doc nonlinear mapping estimation	Rémi Flamary

2016-11-04	doc linear mapping estimation	Rémi Flamary

2016-11-04	V0.1.9	Rémi Flamary

2016-11-04	V0.1.8	Rémi Flamary

2016-11-04	add demo mapping	Rémi Flamary

2016-11-04	add mapping estimation with kernels works!	Rémi Flamary

2016-11-03	add mapping estimation with kernels (smaller bugs)	Rémi Flamary

2016-11-03	add mapping estimation with kernels (still debugging)	Rémi Flamary

2016-11-03	add mapping estimation with kernels (still debugging)	Rémi Flamary

2016-11-03	add mapping estimation (still debugging)	Rémi Flamary