From 184f8f4f7ac78f1dd7f653496d2753211a4e3426 Mon Sep 17 00:00:00 2001 From: Rémi Flamary Date: Tue, 1 Jun 2021 10:10:54 +0200 Subject: [MRG] POT numpy/torch/jax backends (#249) * add numpy and torch backends * stat sets on functions * proper import * install recent torch on windows * install recent torch on windows * now testing all functions in backedn * add jax backedn * clenaup windowds * proper convert for jax backedn * pep8 * try again windows tests * test jax conversion * try proper widows tests * emd fuction ses backedn * better test partial OT * proper tests to_numpy and teplate Backend * pep8 * pep8 x2 * feaking sinkhorn works with torch * sinkhorn2 compatible * working ot.emd2 * important detach * it should work * jax autodiff emd * pep8 * no tast same for jax * new independat tests per backedn * freaking pep8 * add tests for gradients * deprecate ot.gpu * worging dist function * working dist * dist done in backedn * not in * remove indexing * change accuacy for jax * first pull backend * projection simplex * projection simplex * projection simplex * projection simplex no ci * projection simplex no ci * projection simplex no ci * pep8 * add backedn discusion to quickstart guide * projection simplex no ci * projection simplex no ci * projection simplex no ci * pep8 + better doc * proper links * corect doctest * big debug documentation * doctest again * doctest again bis * doctest again ter (last one or i kill myself) * backend test + doc proj simplex * correction test_utils * correction test_utils * correction cumsum * correction flip * correction flip v2 * more debug * more debug * more debug + pep8 * pep8 * argh * proj_simplex * backedn works for sort * proj simplex * jax sucks * update doc * Update test/test_utils.py Co-authored-by: Alexandre Gramfort * Update docs/source/quickstart.rst Co-authored-by: Alexandre Gramfort * Update docs/source/quickstart.rst Co-authored-by: Alexandre Gramfort * Update docs/source/quickstart.rst Co-authored-by: Alexandre Gramfort * Update docs/source/readme.rst Co-authored-by: Alexandre Gramfort * Update test/test_utils.py Co-authored-by: Alexandre Gramfort * Update ot/utils.py Co-authored-by: Alexandre Gramfort * Update docs/source/readme.rst Co-authored-by: Alexandre Gramfort * Update ot/lp/__init__.py Co-authored-by: Alexandre Gramfort * begin comment alex * comment alex part 2 * optimize test gromov * proj_simplex on vectors * add awesome gradient decsnt example on the weights * pep98 of course * proof read example by alex * pep8 again * encoding oos in translation * correct legend Co-authored-by: Nicolas Courty Co-authored-by: Alexandre Gramfort --- test/test_bregman.py | 74 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 74 insertions(+) (limited to 'test/test_bregman.py') diff --git a/test/test_bregman.py b/test/test_bregman.py index 1ebd21f..7c5162a 100644 --- a/test/test_bregman.py +++ b/test/test_bregman.py @@ -9,6 +9,10 @@ import numpy as np import pytest import ot +from ot.backend import get_backend_list +from ot.backend import torch + +backend_list = get_backend_list() def test_sinkhorn(): @@ -30,6 +34,76 @@ def test_sinkhorn(): u, G.sum(0), atol=1e-05) # cf convergence sinkhorn +@pytest.mark.parametrize('nx', backend_list) +def test_sinkhorn_backends(nx): + n_samples = 100 + n_features = 2 + rng = np.random.RandomState(0) + + x = rng.randn(n_samples, n_features) + y = rng.randn(n_samples, n_features) + a = ot.utils.unif(n_samples) + + M = ot.dist(x, y) + + G = ot.sinkhorn(a, a, M, 1) + + ab = nx.from_numpy(a) + Mb = nx.from_numpy(M) + + Gb = ot.sinkhorn(ab, ab, Mb, 1) + + np.allclose(G, nx.to_numpy(Gb)) + + +@pytest.mark.parametrize('nx', backend_list) +def test_sinkhorn2_backends(nx): + n_samples = 100 + n_features = 2 + rng = np.random.RandomState(0) + + x = rng.randn(n_samples, n_features) + y = rng.randn(n_samples, n_features) + a = ot.utils.unif(n_samples) + + M = ot.dist(x, y) + + G = ot.sinkhorn(a, a, M, 1) + + ab = nx.from_numpy(a) + Mb = nx.from_numpy(M) + + Gb = ot.sinkhorn2(ab, ab, Mb, 1) + + np.allclose(G, nx.to_numpy(Gb)) + + +def test_sinkhorn2_gradients(): + n_samples = 100 + n_features = 2 + rng = np.random.RandomState(0) + + x = rng.randn(n_samples, n_features) + y = rng.randn(n_samples, n_features) + a = ot.utils.unif(n_samples) + + M = ot.dist(x, y) + + if torch: + + a1 = torch.tensor(a, requires_grad=True) + b1 = torch.tensor(a, requires_grad=True) + M1 = torch.tensor(M, requires_grad=True) + + val = ot.sinkhorn2(a1, b1, M1, 1) + + val.backward() + + assert a1.shape == a1.grad.shape + assert b1.shape == b1.grad.shape + assert M1.shape == M1.grad.shape + + def test_sinkhorn_empty(): # test sinkhorn n = 100 -- cgit v1.2.3