From 5bc7a5c23d7d9bb0a160fa200887d571f48eb061 Mon Sep 17 00:00:00 2001
From: Yannick DAYER <yannick.dayer@idiap.ch>
Date: Wed, 29 May 2024 15:10:31 +0200
Subject: [PATCH] tests: skip tests with k-means++ in dask_ml k_init
With the latests versions of dask_ml, `KMeans` and `k_init` try to use
the `kmeans_plusplus` of sklearn without the `sample_weights` argument.
---
tests/test_gmm.py | 4 ++++
tests/test_kmeans.py | 4 ++++
2 files changed, 8 insertions(+)
diff --git a/tests/test_gmm.py b/tests/test_gmm.py
index af77842..a7e67c2 100644
--- a/tests/test_gmm.py
+++ b/tests/test_gmm.py
@@ -15,6 +15,7 @@ import tempfile
from copy import deepcopy
import numpy as np
+import pytest
from dask.distributed import Client
from h5py import File as HDF5File
@@ -484,6 +485,9 @@ def test_machine_parameters():
np.testing.assert_equal(machine.variances, new_variances)
+@pytest.mark.skip(
+ reason="dask-ml=2024.4.4 not compatible with sklearn>=1.3 ('_kmeans_plusplus' requires the 'sample_weight' argument)"
+)
def test_gmm_kmeans_plusplus_init():
n_gaussians = 3
machine = GMMMachine(
diff --git a/tests/test_kmeans.py b/tests/test_kmeans.py
index 15ccfe1..ecaf2fa 100644
--- a/tests/test_kmeans.py
+++ b/tests/test_kmeans.py
@@ -14,6 +14,7 @@ import copy
import dask.array as da
import numpy as np
+import pytest
import scipy.spatial.distance
from bob.learn.em import KMeansMachine, kmeans
@@ -116,6 +117,9 @@ def test_kmeans_fit_parallel():
np.testing.assert_almost_equal(centroids, expected, decimal=7)
+@pytest.mark.skip(
+ reason="dask-ml=2024.4.4 not compatible with sklearn>=1.3 ('_kmeans_plusplus' requires the 'sample_weight' argument)"
+)
def test_kmeans_fit_init_pp():
np.random.seed(0)
data1 = np.random.normal(loc=1, size=(2000, 3))
--
GitLab