summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2018-09-24update testsRémi Flamary
2018-09-24pep8Rémi Flamary
2018-09-24convert ot.gpu to cupyRémi Flamary
2018-09-24Merge pull request #64 from rflamary/convolutionRémi Flamary
[MRG] Wasserstein convolutional barycenter This PR closes Issue #51
2018-09-14speedup test stochasticRémi Flamary
2018-09-14correct path function in exampleRémi Flamary
2018-09-14correct path function in exampleRémi Flamary
2018-09-14Merge pull request #65 from kilianFatras/stochastic_OTRémi Flamary
better implementation on stocjastic gradient updates
2018-09-13better implementation on gradient updatesKilian Fatras
2018-09-07test wasserstein barycenterNicolas Courty
2018-09-07whitetrail pep8Nicolas Courty
2018-09-07stabThr and pep8Nicolas Courty
2018-09-07Merge branch 'master' into convolutionNicolas Courty
2018-09-07pep8 fixed (contd)Nicolas Courty
2018-09-07pep8 normalizationNicolas Courty
2018-09-07Wasserstein convolutional barycenterNicolas Courty
2018-08-30Merge pull request #62 from kilianFatras/stochastic_OTRémi Flamary
Debug and speedup SGD stochastic OT
2018-08-29Merge branch 'master' into stochastic_OTKilian
2018-08-29fixed conflictsKilian Fatras
2018-08-29replaced marginal testsKilian Fatras
2018-08-29Merge pull request #48 from rflamary/remove_otda_v05Rémi Flamary
Remove deprecated OTDA Classes
2018-08-29Merge pull request #58 from rflamary/speedupRémi Flamary
Speedup Sinkhorn with einsum + bench
2018-08-28fixed test errorKilian Fatras
2018-08-28fixed test gaussKilian Fatras
2018-08-28fixed argument functionsKilian Fatras
2018-08-28updated README and fixed pep8Kilian Fatras
2018-08-28added gaussian testKilian Fatras
2018-08-28fixed pep8Kilian Fatras
2018-08-28debug sgd dualKilian Fatras
2018-08-28fixed bug in sgd dualKilian Fatras
2018-07-24ensum tets marginals sinkhornRémi Flamary
2018-07-24final makefile benchRémi Flamary
2018-07-24final makefile benchRémi Flamary
2018-07-24cancel einsumRémi Flamary
2018-07-24pb indexRémi Flamary
2018-07-24correction sizeRémi Flamary
2018-07-24test eisum instead of dotRémi Flamary
2018-07-24last update makefileRémi Flamary
2018-07-24pep8 all the wayRémi Flamary
2018-07-24last debug makefileRémi Flamary
2018-07-24upate makefileRémi Flamary
2018-07-24update makefileRémi Flamary
2018-07-24update makefileRémi Flamary
2018-07-24add bench to makefileRémi Flamary
2018-07-24speedup einsum constraint violationRémi Flamary
2018-07-18Merge pull request #57 from LeoGautheron/masterRémi Flamary
Speed-up Sinkhorn
2018-07-16Add comment & fix flake8 errorLeoGautheron
2018-07-16Remove dependency sklearnLeoGautheron
2018-07-12Merge pull request #56 from vivienseguy/vivien-barycentersRémi Flamary
Free support barycenters
2018-07-11Speed-up SinkhornLeoGautheron
Speed-up in 3 places: - the computation of pairwise distance is faster with sklearn.metrics.pairwise.euclidean_distances - faster computation of K = np.exp(-M / reg) - faster computation of the error every 10 iterations Example with this little script: import time import numpy as np import ot rng = np.random.RandomState(0) transport = ot.da.SinkhornTransport() time1 = time.time() Xs, ys, Xt = rng.randn(10000, 100), rng.randint(0, 2, size=10000), rng.randn(10000, 100) transport.fit(Xs=Xs, Xt=Xt) time2 = time.time() print("OT Computation Time {:6.2f} sec".format(time2-time1)) transport = ot.da.SinkhornLpl1Transport() transport.fit(Xs=Xs, ys=ys, Xt=Xt) time3 = time.time() print("OT LpL1 Computation Time {:6.2f} sec".format(time3-time2)) Before OT Computation Time 19.93 sec OT LpL1 Computation Time 133.43 sec After OT Computation Time 7.55 sec OT LpL1 Computation Time 82.25 sec
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-02 10:47:47 +0000