From 982b1a38a4c80e4e090d0fc55a1adb54d4e0aa20 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?R=C3=A9mi=20Flamary?= Date: Fri, 19 Sep 2025 11:59:01 +0200 Subject: [PATCH 1/3] Release 0.9.6 --- RELEASES.md | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/RELEASES.md b/RELEASES.md index 529c3ac7b..4bef2dcd4 100644 --- a/RELEASES.md +++ b/RELEASES.md @@ -2,6 +2,12 @@ ## 0.9.6 +*September 2025* + +This new release contains several new features and bug fixes. Among the new features we have a new submodule `ot.batch` that contains batch parallel solvers for several OT problems including [Sinkhorn, Gromov-Wasserstein and Fused Gromov-Wasserstein](https://pythonot.github.io/master/auto_examples/backends/plot_ot_batch.html). This new submodule can be used to solve many independent OT problems in parallel on CPU or GPU. We also implemented a new Nystrom kernel approximation for the Sinkhorn solver that can be used to speed up the computation of the Sinkhorn divergence on large datasets. We also added new 1D solvers for [Linear circular OT](https://pythonot.github.io/master/auto_examples/sliced-wasserstein/plot_compute_wasserstein_circle.html) and new solvers for free support [OT barycenters with generic cost functions](https://pythonot.github.io/master/auto_examples/barycenters/plot_free_support_barycenter_generic_cost.html) and for [barycenters between Gaussian Mixture Models (GMMs)](https://pythonot.github.io/master/auto_examples/barycenters/plot_gmm_barycenter.html). + +Finally we have updated the documentation to reflect the new generic API and reorganized the [examples gallery](https://pythonot.github.io/auto_examples/index.html). + #### New features - Implement CG solvers for partial FGW (PR #687) - Added feature `grad=last_step` for `ot.solvers.solve` (PR #693) @@ -44,6 +50,7 @@ - Avoid raising unnecessary warnings in `ot.lp.solver_1d.binary_search_circle` (Issue #738) - Avoid deprecation warning in `ot.lp.solver_1d.wasserstein_1d` (Issue #760, PR #761) + ## 0.9.5 *November 2024* From ef7b6a7d3b85e478b8d359890ff12ee5f4bae9a5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?C=C3=A9dric=20Vincent-Cuaz?= Date: Fri, 19 Sep 2025 13:39:22 +0200 Subject: [PATCH 2/3] Update RELEASES.md --- RELEASES.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/RELEASES.md b/RELEASES.md index 4bef2dcd4..ddfcc415c 100644 --- a/RELEASES.md +++ b/RELEASES.md @@ -4,7 +4,7 @@ *September 2025* -This new release contains several new features and bug fixes. Among the new features we have a new submodule `ot.batch` that contains batch parallel solvers for several OT problems including [Sinkhorn, Gromov-Wasserstein and Fused Gromov-Wasserstein](https://pythonot.github.io/master/auto_examples/backends/plot_ot_batch.html). This new submodule can be used to solve many independent OT problems in parallel on CPU or GPU. We also implemented a new Nystrom kernel approximation for the Sinkhorn solver that can be used to speed up the computation of the Sinkhorn divergence on large datasets. We also added new 1D solvers for [Linear circular OT](https://pythonot.github.io/master/auto_examples/sliced-wasserstein/plot_compute_wasserstein_circle.html) and new solvers for free support [OT barycenters with generic cost functions](https://pythonot.github.io/master/auto_examples/barycenters/plot_free_support_barycenter_generic_cost.html) and for [barycenters between Gaussian Mixture Models (GMMs)](https://pythonot.github.io/master/auto_examples/barycenters/plot_gmm_barycenter.html). +This new release contains several new features and bug fixes. Among the new features we have a new submodule `ot.batch` that contains batch parallel solvers for several OT problems including [Sinkhorn, Gromov-Wasserstein and Fused Gromov-Wasserstein](https://pythonot.github.io/master/auto_examples/backends/plot_ot_batch.html). This new submodule can be used to solve many independent OT problems in parallel on CPU or GPU with shared source and target support sizes. We also implemented a new Nystrom kernel approximation for the Sinkhorn solver that can be used to speed up the computation of the Sinkhorn divergence on large datasets. We also added new 1D solvers for [Linear circular OT](https://pythonot.github.io/master/auto_examples/sliced-wasserstein/plot_compute_wasserstein_circle.html) and new solvers for free support [OT barycenters with generic cost functions](https://pythonot.github.io/master/auto_examples/barycenters/plot_free_support_barycenter_generic_cost.html) and for [barycenters between Gaussian Mixture Models (GMMs)](https://pythonot.github.io/master/auto_examples/barycenters/plot_gmm_barycenter.html). We also implemented two solvers for partial Fused Gromov-Wasserstein problems based on conditional gradient and projected gradient descents. Finally we have updated the documentation to reflect the new generic API and reorganized the [examples gallery](https://pythonot.github.io/auto_examples/index.html). From 0652267d4dba294a2ed8163605480185606ace8a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?C=C3=A9dric=20Vincent-Cuaz?= Date: Fri, 19 Sep 2025 14:16:08 +0200 Subject: [PATCH 3/3] Update RELEASES.md --- RELEASES.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/RELEASES.md b/RELEASES.md index ddfcc415c..e76fd575a 100644 --- a/RELEASES.md +++ b/RELEASES.md @@ -4,9 +4,9 @@ *September 2025* -This new release contains several new features and bug fixes. Among the new features we have a new submodule `ot.batch` that contains batch parallel solvers for several OT problems including [Sinkhorn, Gromov-Wasserstein and Fused Gromov-Wasserstein](https://pythonot.github.io/master/auto_examples/backends/plot_ot_batch.html). This new submodule can be used to solve many independent OT problems in parallel on CPU or GPU with shared source and target support sizes. We also implemented a new Nystrom kernel approximation for the Sinkhorn solver that can be used to speed up the computation of the Sinkhorn divergence on large datasets. We also added new 1D solvers for [Linear circular OT](https://pythonot.github.io/master/auto_examples/sliced-wasserstein/plot_compute_wasserstein_circle.html) and new solvers for free support [OT barycenters with generic cost functions](https://pythonot.github.io/master/auto_examples/barycenters/plot_free_support_barycenter_generic_cost.html) and for [barycenters between Gaussian Mixture Models (GMMs)](https://pythonot.github.io/master/auto_examples/barycenters/plot_gmm_barycenter.html). We also implemented two solvers for partial Fused Gromov-Wasserstein problems based on conditional gradient and projected gradient descents. +This new release contains several new features and bug fixes. Among the new features we have a new submodule `ot.batch` that contains batch parallel solvers for several OT problems including [Sinkhorn, Gromov-Wasserstein and Fused Gromov-Wasserstein](https://pythonot.github.io/master/auto_examples/backends/plot_ot_batch.html). This new submodule can be used to solve many independent OT problems in parallel on CPU or GPU with shared source and target support sizes. We also implemented a new Nystrom kernel approximation for the Sinkhorn solver that can be used to speed up the computation of the Sinkhorn divergence on large datasets. We also added new 1D solvers for [Linear circular OT](https://pythonot.github.io/master/auto_examples/sliced-wasserstein/plot_compute_wasserstein_circle.html) and new solvers for free support [OT barycenters with generic cost functions](https://pythonot.github.io/master/auto_examples/barycenters/plot_free_support_barycenter_generic_cost.html) and for [barycenters between Gaussian Mixture Models (GMMs)](https://pythonot.github.io/master/auto_examples/barycenters/plot_gmm_barycenter.html). We also implemented two solvers for partial Fused Gromov-Wasserstein problems based on [conditional gradient](https://pythonot.github.io/master/gen_modules/ot.gromov.html#ot.gromov.partial_fused_gromov_wasserstein) and [projected gradient](https://pythonot.github.io/master/gen_modules/ot.gromov.html#ot.gromov.entropic_partial_fused_gromov_wasserstein) descents. -Finally we have updated the documentation to reflect the new generic API and reorganized the [examples gallery](https://pythonot.github.io/auto_examples/index.html). +Finally we have updated the documentation to reflect the new generic API and reorganized the [examples gallery](https://pythonot.github.io/auto_examples/index.html). #### New features - Implement CG solvers for partial FGW (PR #687)