[MAINT] Update some missed tests and examples to use new standardize strategy to silence warnings

examples/03_connectivity/plot_atlas_comparison.py

@@ -72,6 +72,7 @@
 masker = MultiNiftiLabelsMasker(
     labels_img=yeo["thick_17"],
     standardize="zscore_sample",
+    standardize_confounds="zscore_sample",
     memory="nilearn_cache",
     n_jobs=2,
 )
@@ -156,6 +157,7 @@ def lag_correlation(time_series, lag):
 masker = MultiNiftiMapsMasker(
     maps_img=difumo.maps,
     standardize="zscore_sample",
+    standardize_confounds="zscore_sample",
     memory="nilearn_cache",
     n_jobs=2,
 )

examples/03_connectivity/plot_group_level_connectivity.py

@@ -52,6 +52,7 @@
     memory="nilearn_cache",
     memory_level=1,
     standardize="zscore_sample",
+    standardize_confounds="zscore_sample",
 ).fit()
 
 ###############################################################################

examples/03_connectivity/plot_inverse_covariance_connectome.py

@@ -48,6 +48,7 @@
 masker = NiftiMapsMasker(
     maps_img=atlas_filename,
     standardize="zscore_sample",
+    standardize_confounds="zscore_sample",
     memory="nilearn_cache",
     verbose=5,
 )

examples/03_connectivity/plot_multi_subject_connectome.py

@@ -72,6 +72,7 @@ def plot_matrices(cov, prec, title, labels):
     high_pass=0.01,
     t_r=2,
     standardize="zscore_sample",
+    standardize_confounds="zscore_sample",
     memory="nilearn_cache",
     memory_level=1,
     verbose=2,

examples/03_connectivity/plot_seed_to_voxel_correlation.py

@@ -67,6 +67,7 @@
     radius=8,
     detrend=True,
     standardize="zscore_sample",
+    standardize_confounds="zscore_sample",
     low_pass=0.1,
     high_pass=0.01,
     t_r=2,
@@ -93,6 +94,7 @@
     smoothing_fwhm=6,
     detrend=True,
     standardize="zscore_sample",
+    standardize_confounds="zscore_sample",
     low_pass=0.1,
     high_pass=0.01,
     t_r=2,

examples/03_connectivity/plot_signal_extraction.py

@@ -56,6 +56,7 @@
 masker = NiftiLabelsMasker(
     labels_img=atlas_filename,
     standardize="zscore_sample",
+    standardize_confounds="zscore_sample",
     memory="nilearn_cache",
     verbose=5,
 )

examples/03_connectivity/plot_sphere_based_connectome.py

@@ -68,6 +68,7 @@
     radius=8,
     detrend=True,
     standardize="zscore_sample",
+    standardize_confounds="zscore_sample",
     low_pass=0.1,
     high_pass=0.01,
     t_r=2,
@@ -203,6 +204,7 @@
     radius=5.0,
     detrend=True,
     standardize="zscore_sample",
+    standardize_confounds="zscore_sample",
     low_pass=0.1,
     high_pass=0.01,
     t_r=2,
@@ -345,6 +347,7 @@
     radius=4.5,
     detrend=True,
     standardize="zscore_sample",
+    standardize_confounds="zscore_sample",
     low_pass=0.1,
     high_pass=0.01,
     t_r=2,

examples/06_manipulating_images/plot_nifti_labels_simple.py

@@ -42,7 +42,9 @@
 
 # Instantiate the masker with label image and label values
 masker = NiftiLabelsMasker(
-    atlas.maps, labels=atlas.labels, standardize="zscore_sample"
+    atlas.maps,
+    labels=atlas.labels,
+    standardize="zscore_sample",
 )
 
 # Visualize the atlas

examples/07_advanced/plot_age_group_prediction_cross_val.py

@@ -41,6 +41,7 @@
     memory="nilearn_cache",
     memory_level=1,
     standardize="zscore_sample",
+    standardize_confounds="zscore_sample",
 ).fit()
 
 masked_data = [

nilearn/decoding/tests/test_decoder.py

@@ -566,7 +566,9 @@ def test_decoder_binary_classification_cross_validation(
     # check cross-validation scheme and fit attribute with groups enabled
     rand_local = np.random.RandomState(42)
 
-    model = Decoder(estimator="svc", mask=mask, standardize=True, cv=cv)
+    model = Decoder(
+        estimator="svc", mask=mask, standardize="zscore_sample", cv=cv
+    )
     groups = None
     if isinstance(cv, LeaveOneGroupOut):
         groups = rand_local.binomial(2, 0.3, size=len(y))
@@ -888,7 +890,9 @@ def test_decoder_multiclass_classification_cross_validation(
     # check cross-validation scheme and fit attribute with groups enabled
     rand_local = np.random.RandomState(42)
 
-    model = Decoder(estimator="svc", mask=mask, standardize=True, cv=cv)
+    model = Decoder(
+        estimator="svc", mask=mask, standardize="zscore_sample", cv=cv
+    )
     groups = None
     if isinstance(cv, LeaveOneGroupOut):
         groups = rand_local.binomial(2, 0.3, size=len(y))

nilearn/maskers/tests/test_nifti_maps_masker.py

@@ -527,11 +527,11 @@ def test_standardization():
     unstandarized_label_signals = masker.fit_transform(img)
 
     # z-score
-    masker = NiftiMapsMasker(maps, standardize="zscore")
+    masker = NiftiMapsMasker(maps, standardize="zscore_sample")
     trans_signals = masker.fit_transform(img)
 
     np.testing.assert_almost_equal(trans_signals.mean(0), 0)
-    np.testing.assert_almost_equal(trans_signals.std(0), 1)
+    np.testing.assert_almost_equal(trans_signals.std(0), 1, decimal=3)
 
     # psc
     masker = NiftiMapsMasker(maps, standardize="psc")

nilearn/maskers/tests/test_nifti_spheres_masker.py

@@ -189,12 +189,12 @@ def test_standardization():
     img = nibabel.Nifti1Image(data, np.eye(4))
 
     # test zscore
-    masker = NiftiSpheresMasker([(1, 1, 1)], standardize="zscore")
+    masker = NiftiSpheresMasker([(1, 1, 1)], standardize="zscore_sample")
     # Test the fit
     s = masker.fit_transform(img)
 
     np.testing.assert_almost_equal(s.mean(), 0)
-    np.testing.assert_almost_equal(s.std(), 1)
+    np.testing.assert_almost_equal(s.std(), 1, decimal=1)
 
     # test psc
     masker = NiftiSpheresMasker([(1, 1, 1)], standardize="psc")

nilearn/tests/test_masking.py

@@ -57,7 +57,9 @@ def _cov_conf(tseries, conf):
     return cov_mat
 
 
-def _confounds_regression(standardize_signal=True, standardize_confounds=True):
+def _confounds_regression(
+        standardize_signal="zscore_sample", standardize_confounds=True
+):
     img, mask, conf = _simu_img()
     masker = NiftiMasker(standardize=standardize_signal,
                          standardize_confounds=standardize_confounds,
@@ -73,11 +75,11 @@ def _confounds_regression(standardize_signal=True, standardize_confounds=True):
 def test_high_variance_confounds():
     img, mask, conf = _simu_img()
     hv_confounds = high_variance_confounds(img)
-    masker1 = NiftiMasker(standardize=True, detrend=False,
+    masker1 = NiftiMasker(standardize="zscore_sample", detrend=False,
                           high_variance_confounds=False,
                           mask_img=mask).fit()
     tseries1 = masker1.transform(img, confounds=[hv_confounds, conf])
-    masker2 = NiftiMasker(standardize=True, detrend=False,
+    masker2 = NiftiMasker(standardize="zscore_sample", detrend=False,
                           high_variance_confounds=True,
                           mask_img=mask).fit()
     tseries2 = masker2.transform(img, confounds=conf)
@@ -98,12 +100,7 @@ def test_confounds_standardization():
 
     # Signal is z-scored with string arg
     # Explicit standardization of confounds
-    assert (_confounds_regression(standardize_signal='zscore',
-                                  standardize_confounds=True) < eps)
-
-    # Signal is z-scored with boolean arg
-    # Explicit standardization of confounds
-    assert (_confounds_regression(standardize_signal=True,
+    assert (_confounds_regression(standardize_signal='zscore_sample',
                                   standardize_confounds=True) < eps)
 
     # Signal is psc standardized
@@ -120,7 +117,7 @@ def test_confounds_standardization():
     # Signal is z-scored with string arg
     # Confounds are not standardized
     # In this case, the regression should fail...
-    assert (_confounds_regression(standardize_signal='zscore',
+    assert (_confounds_regression(standardize_signal='zscore_sample',
                                   standardize_confounds=False) > 100)
 
     # Signal is psc standardized

nilearn/tests/test_signal.py

@@ -257,12 +257,6 @@ def test_standardize():
 
     # transpose array to fit _standardize input.
     # Without trend removal
-    b = nisignal._standardize(a, standardize='zscore')
-    stds = np.std(b)
-    np.testing.assert_almost_equal(stds, np.ones(n_features))
-    np.testing.assert_almost_equal(b.sum(axis=0), np.zeros(n_features))
-
-    # Repeating test above but for new correct strategy
     b = nisignal._standardize(a, standardize='zscore_sample')
     stds = np.std(b)
     np.testing.assert_almost_equal(stds, np.ones(n_features), decimal=1)
@@ -276,16 +270,11 @@ def test_standardize():
     b = nisignal._standardize(a, detrend=True, standardize="zscore_sample")
     np.testing.assert_almost_equal(b, np.zeros(b.shape))
 
-    length_1_signal = np.atleast_2d(np.linspace(0, 2., n_features))
-    np.testing.assert_array_equal(length_1_signal,
-                                  nisignal._standardize(length_1_signal,
-                                                        standardize='zscore'))
-
-    # Repeating test above but for new correct strategy
     length_1_signal = np.atleast_2d(np.linspace(0, 2., n_features))
     np.testing.assert_array_equal(
-        length_1_signal,
-        nisignal._standardize(length_1_signal, standardize="zscore_sample")
+        length_1_signal, nisignal._standardize(
+            length_1_signal, standardize='zscore_sample'
+        )
     )
 
 

pyproject.toml

@@ -187,6 +187,8 @@ minversion = "6.0"
 addopts = "--doctest-modules -s -vv --durations=0"
 doctest_optionflags = "NORMALIZE_WHITESPACE ELLIPSIS"
 junit_family = "xunit2"
+# TODO: Remove filter in release 0.13
+filterwarnings = "ignore:.*Please use 'zscore_sample' instead.:FutureWarning"
 
 [tool.codespell]
 skip = "./.git,plotly-gl3d-latest.min.js,jquery.min.js,localizer_behavioural.tsv,.mypy_cache,env,venv,./doc/auto_examples,*/tmp,./doc/modules/generated/*"