summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2017-07-07éadd macosxéRémi Flamary
2017-07-07add sinkhorbn2 +v3Rémi Flamary
2017-07-06add mailmapRémi Flamary
2017-07-06update gitattributesRémi Flamary
2017-07-06rename fileRémi Flamary
2017-07-06add .gitattributeRémi Flamary
2017-07-06update readmeRémi Flamary
2017-07-06move notebooksRémi Flamary
2017-07-06update notebookRémi Flamary
2017-07-06update wda exampleRémi Flamary
2017-07-05add FDA for comparisonRémi Flamary
2017-07-04delete useless imagesRémi Flamary
2017-07-04update thumb imageRémi Flamary
2017-07-04add autograd mock againRémi Flamary
2017-07-04add autograd mockRémi Flamary
2017-07-04add autograd mockRémi Flamary
2017-07-03add notebooks to reamdeRémi Flamary
2017-07-03add notebooksRémi Flamary
2017-07-03doc updateRémi Flamary
2017-07-03add mock in doc conf fileRémi Flamary
2017-07-03doc drRémi Flamary
2017-07-03update docRémi Flamary
2017-07-03add example normRémi Flamary
2017-07-03debug bregman stabilizedRémi Flamary
2017-06-23add apt update tarvisRémi Flamary
2017-06-23xRémi Flamary
Merge branch 'master' of github.com:rflamary/POT
2017-06-23test travisRémi Flamary
2017-06-23links 2/2Rémi Flamary
2017-06-23add links 1/2Rémi Flamary
2017-06-20add clean zeros function for sparse distributionsRémi Flamary
2017-06-13example compute emdRémi Flamary
2017-06-13implement paralell sinkhorn stabilizedRémi Flamary
2017-06-13implement paralell sinkhornRémi Flamary
2017-06-13add em2 computation exampleRémi Flamary
2017-06-13make sinkhorn more general with method selectionRémi Flamary
2017-06-09update docRémi Flamary
2017-06-09update .gitignore with sphinwgallery temop filesRémi Flamary
2017-06-09add doc and correct encodingRémi Flamary
2017-06-09add doc ot.gpu.bregmanRémi Flamary
2017-04-27Merge pull request #10 from aje/masterncourty
performance improvement sinkhorn lpl1
2017-04-24Remove unnecessary parameterLeo gautheron
2017-04-24add GPU implementation sinkhorn lpl1Leo gautheron
2017-04-21performance improvement sinkhorn lpl1Leo 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.
2017-04-21Merge branch 'master' of github.com:rflamary/POTRémi Flamary
2017-04-21small normalization bugRémi Flamary
2017-04-20Update README.mdRémi Flamary
2017-04-20Update README.mdRémi Flamary
2017-04-20update readmeRémi Flamary
2017-04-20add test gpuRémi Flamary
2017-04-20update docRémi Flamary