summaryrefslogtreecommitdiff
path: root/src/python/gudhi
AgeCommit message (Collapse)Author
2020-05-11Merge branch 'master' into improve_dependencies_documentation2ROUVREAU Vincent
2020-05-11Double underscore is not the correct syntaxROUVREAU Vincent
2020-05-11Merge pull request #294 from VincentRouvreau/fix_sphinx_warningsVincent Rouvreau
Fix sphinx warnings
2020-05-11Fixed by @tlacombeVincent Rouvreau
Co-authored-by: Théo Lacombe <lacombe1993@gmail.com>
2020-05-11Fixed by @tlacombeVincent Rouvreau
Co-authored-by: Théo Lacombe <lacombe1993@gmail.com>
2020-05-11Fix #299ROUVREAU Vincent
2020-05-09Merge pull request #308 from mglisse/bottlepy2Marc Glisse
Bottleneck pybind11 interface
2020-05-07Warn for initialize_filtrationMarc Glisse
2020-05-07Merge remote-tracking branch 'origin/master' into bottlepy2Marc Glisse
2020-05-07Merge pull request #302 from mglisse/hera-gilMarc Glisse
Fewer copies and no GIL for hera
2020-05-07docMarc Glisse
2020-05-06Reimplement the bottleneck python wrapper with pybind11Marc Glisse
2020-05-06Refactor the numpy -> C++ range conversionMarc Glisse
If we want to reuse it for bottleneck...
2020-05-06Avoid reading outside of allocated regionMarc Glisse
The result was unused, but better be safe.
2020-05-05fix use of threads and n_jobs in ParallelMarc Glisse
2020-05-04Merge pull request #205 from MathieuCarriere/wasserstein_representationsVincent Rouvreau
Integration of Wasserstein distances in representations module
2020-05-03Fewer copies and no GIL for heraMarc Glisse
Now the input arrays are not copied as long as they use a float64 data type, even if they are not contiguous. That's not important here, but I wanted an example of how to do it. More importantly, no need to hold the GIL. I was too lazy to benchmark to see if that changed anything...
2020-04-30Merge pull request #285 from mglisse/wass-autodiffMarc Glisse
Automatic differentiation for Wasserstein distance
2020-04-29small modifsMathieuCarriere
2020-04-29Merge branch 'master' of https://github.com/GUDHI/gudhi-devel into ↵MathieuCarriere
wasserstein_representations
2020-04-29Merge remote-tracking branch 'origin/master' into gen2Marc Glisse
2020-04-28added call methods + other fixesMathieuCarriere
2020-04-28fix conflictMathieuCarriere
2020-04-26Fix barycenter sphinx warningsROUVREAU Vincent
2020-04-25Fix #259ROUVREAU Vincent
2020-04-23Merge remote-tracking branch 'origin/master' into compute_persistenceMarc Glisse
2020-04-23Fix pasto in the docMarc Glisse
2020-04-22Document incompatibility of matching=True and enable_autodiffMarc Glisse
2020-04-22Merge remote-tracking branch 'origin/master' into wass-autodiffMarc Glisse
2020-04-22Clarify the doc of enable_autodiffMarc Glisse
2020-04-22Rename local variablesMarc Glisse
2020-04-22Document several optional dependencies of knnMarc Glisse
2020-04-20Drop one commentMarc Glisse
2020-04-20Merge remote-tracking branch 'origin/master' into wass-autodiffMarc Glisse
2020-04-20Merge pull request #265 from mglisse/dtmMarc Glisse
DTM
2020-04-20Reintroduce _proj_on_diag, with a unit testMarc Glisse
2020-04-20Add __license__Marc Glisse
2020-04-19Unwrap the resultMarc Glisse
2020-04-18enable_autodiff for POT wasserstein_distanceMarc Glisse
2020-04-17Simplify distance-to-diagonal in WassersteinMarc Glisse
2020-04-17Merge remote-tracking branch 'origin/master' into filtrationMarc Glisse
2020-04-14Merge remote-tracking branch 'origin/master' into dtmMarc Glisse
2020-04-14Doc improvementsMarc Glisse
2020-04-13Tweak to detect fit_transformMarc Glisse
2020-04-13Remove left-over codeMarc Glisse
eagerpy is only used with enable_autodiff
2020-04-13Generalize enable_autodiff to more implementationsMarc Glisse
Still limited to L^p
2020-04-13Fix NaN gradient with pytorchMarc Glisse
2020-04-13Small autodiff tweaksMarc Glisse
2020-04-13enable_autodiff with keopsMarc Glisse
This doesn't seem like the best way to handle it, we may want to handle it like a wrapper that gets the indices from knn (whatever backend) and then computes the distances.
2020-04-12Parallelize the "precomputed" case of knnMarc Glisse
It is supposed to be possible to compile numpy with openmp, but it looks like it isn't done in any of the usual packages. It may be possible to refactor that code so there is less redundancy.
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-28 19:24:44 +0000