Age | Commit message (Collapse) | Author | |
---|---|---|---|
2018-05-09 | Merge pull request #42 from rflamary/linear_mapping | Rémi Flamary | |
Linear mapping + tests | |||
2018-05-09 | doc typos in linear map function | Rémi Flamary | |
2018-04-18 | BUG: correct typo problems | Tardy Benjamin | |
2018-04-18 | BUG: typo error lo->log | Tardy Benjamin | |
2018-04-18 | BUG: Parameter log unusable in sinkhorn classes | Tardy Benjamin | |
2018-04-18 | ENH: Change the parameter type to bool in class EMDTransport documentation | Tardy Benjamin | |
2018-04-18 | BUG: EMDTransport parameter log unusable | Tardy Benjamin | |
2018-03-21 | pep cleanup | Rémi Flamary | |
2018-03-21 | update readme + doc | Rémi Flamary | |
2018-03-20 | passing tests | Rémi Flamary | |
2018-03-20 | add class LinearTransport | Rémi Flamary | |
2018-03-20 | add linear mapping function | Rémi Flamary | |
2018-02-16 | no more pep8 problem, need cleanup plots in test and examples | Rémi Flamary | |
2018-02-16 | better pep8 but not solved | Rémi Flamary | |
2017-09-04 | doc string + example | Slasnista | |
2017-09-04 | correction semi supervised case | Slasnista | |
2017-08-30 | Move norm out of fit to init for deprecated OTDA | aje | |
2017-08-30 | Move normalize function in utils.py | aje | |
2017-08-30 | Fix name error | aje | |
2017-08-30 | Changes: | aje | |
- Rename numItermax to max_iter - Default value to 100000 instead of 10000 - Add max_iter to class SinkhornTransport(BaseTransport) - Add norm to all BaseTransport | |||
2017-08-29 | Changes to LP solver: | aje | |
- Allow to modify the maximal number of iterations - Display an error message in the python console if the solver encountered an issue | |||
2017-08-28 | check input parameters with helper functions | Slasnista | |
2017-08-28 | handling input arguments in fit, transform... methods + remove old examples | Slasnista | |
2017-08-28 | move no da objects into utils.py | Slasnista | |
2017-08-25 | solving log issues to avoid errors and adding further tests | Slasnista | |
2017-08-23 | added deprecation warning on old classes | Slasnista | |
2017-08-23 | test functions for MappingTransport Class | Slasnista | |
2017-08-23 | out of samples transform and inverse transform by batch | Slasnista | |
2017-08-23 | make doc strings compliant with numpy / modif according to AG review | Slasnista | |
2017-08-04 | added new class MappingTransport to support linear and kernel mapping, not ↵ | Slasnista | |
yet tested | |||
2017-08-04 | out of samples by Ferradans supported for transform and inverse_transform | Slasnista | |
2017-08-04 | reformat doc strings + remove useless log / verbose parameters for emd | Slasnista | |
2017-08-04 | correction of semi supervised mode | Slasnista | |
2017-08-04 | semi supervised mode supported | Slasnista | |
2017-08-04 | added new class SinkhornL1l2Transport() + dedicated test | Slasnista | |
2017-08-04 | added new class SinkhornLpl1Transport() + dedicated test | Slasnista | |
2017-08-04 | added test for fit_transform + correction of fit_transform bug (missing ↵ | Slasnista | |
return self) | |||
2017-08-01 | added EMDTransport Class from NG's code + added dedicated test | Slasnista | |
2017-08-01 | update SinkhornTransport class + added test for class | Slasnista | |
2017-07-31 | own BaseEstimator class written + rflamary comments addressed | Slasnista | |
2017-07-28 | integrate AG comments | Slasnista | |
2017-07-28 | small modifs according to NG proposals | Slasnista | |
2017-07-28 | first proposal for OT wrappers | Slasnista | |
2017-07-28 | remove linewidth error message | Slasnista | |
2017-07-26 | pep8 | Rémi Flamary | |
2017-07-26 | add license and authors on all modules | Rémi Flamary | |
2017-07-20 | pep8 | Alexandre Gramfort | |
2017-07-11 | add documentation | Rémi Flamary | |
2017-04-24 | Remove unnecessary parameter | 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. |