[ENH]: Add test for JuniferNiftiSpheresMasker against nilearn's version

docs/changes/newsfragments/136.enh

@@ -0,0 +1 @@
+Add test to be sure that :class:`.JuniferNiftiSpheresMasker` with mean aggregation function behaves exactly as :class:`nilearn.maskers.NiftiSpheresMasker` by `Synchon Mandal`_

junifer/external/nilearn/__init__.py

@@ -4,3 +4,6 @@
 # License: AGPL
 
 from .junifer_nifti_spheres_masker import JuniferNiftiSpheresMasker
+
+
+__all__ = ["JuniferNiftiSpheresMasker"]

junifer/external/nilearn/junifer_nifti_spheres_masker.py

@@ -29,9 +29,12 @@
     from pandas import DataFrame
 
 
+__all__ = ["JuniferNiftiSpheresMasker"]
+
+
 # New BSD License
 
-# Copyright (c) 2007 - 2022 The nilearn developers.
+# Copyright (c) The nilearn developers.
 # All rights reserved.
 
 
@@ -99,6 +102,17 @@ def _apply_mask_and_get_affinity(
         Contains the boolean indices for each sphere.
         shape: (number of seeds, number of voxels)
 
+    Raises
+    ------
+    ValueError
+        If ``niimg`` and ``mask_img`` are both provided or
+        if overlap is detected between spheres.
+
+    Warns
+    -----
+    RuntimeWarning
+        If the provided images contain NaN, they will be converted to zeroes.
+
     """
     seeds = list(seeds)
 
@@ -210,7 +224,7 @@ def _iter_signals_from_spheres(
     X, A = _apply_mask_and_get_affinity(
         seeds, niimg, radius, allow_overlap, mask_img=mask_img
     )
-    for _, row in enumerate(A.rows):
+    for row in A.rows:
         yield X[:, row]
 
 

junifer/external/nilearn/tests/test_junifer_nifti_spheres_masker.py

@@ -4,20 +4,22 @@
 # License: AGPL
 
 import warnings
+from typing import List, Tuple
 
 import nibabel
 import numpy as np
 import pytest
 from nilearn._utils import data_gen
 from nilearn.image import get_data
+from nilearn.maskers import NiftiSpheresMasker
 from numpy.testing import assert_array_equal
 
 from junifer.external.nilearn import JuniferNiftiSpheresMasker
 
 
 # New BSD License
 
-# Copyright (c) 2007 - 2022 The nilearn developers.
+# Copyright (c) The nilearn developers.
 # All rights reserved.
 
 
@@ -331,3 +333,76 @@ def test_nifti_spheres_masker_io_shapes() -> None:
         )
         test_data = masker.transform(img_4d)
         assert test_data.shape == (n_volumes, n_regions)
+
+
+@pytest.mark.parametrize(
+    "shape",
+    [
+        (10, 11, 12),
+        (10, 11, 12, 5),
+    ],
+)
+@pytest.mark.parametrize(
+    "radius, allow_overlap",
+    [
+        (2.0, True),
+        (2.0, False),
+        (3.0, True),
+        (4.0, True),
+        (5.0, True),
+    ],
+)
+@pytest.mark.parametrize(
+    "coords",
+    [
+        [(1, 1, 1)],
+        [(1, 1, 1), (4, 4, 4)],
+        [(1, 1, 1), (4, 4, 4), (10, 10, 10)],
+    ],
+)
+def test_junifer_and_nilearn_mean_agg_are_equal(
+    shape: Tuple[int, ...],
+    radius: float,
+    allow_overlap: bool,
+    coords: List[Tuple[int, int, int]],
+) -> None:
+    """Test junifer's masker behaves same as nilearn's when agg is mean.
+
+    Parameters
+    ----------
+    shape : tuple of int
+        The parametrized shape of the input image.
+    radius : float
+        The parametrized radius of the spheres.
+    allow_overlap : bool
+        The parametrized option to overlap spheres or not.
+    coords : list of tuple of int, int and int
+        The parametrized seeds.
+
+    """
+    # Set affine
+    affine = np.eye(4)
+    # Generate random image
+    input_img, mask_img = data_gen.generate_random_img(
+        shape=shape,
+        affine=affine,
+    )
+    # Compute junifer's version
+    junifer_masker = JuniferNiftiSpheresMasker(
+        seeds=coords,
+        radius=radius,
+        allow_overlap=allow_overlap,
+        mask_img=mask_img,
+    )
+    junifer_output = junifer_masker.fit_transform(input_img)
+    # Compute nilearn's version
+    nilearn_masker = NiftiSpheresMasker(
+        seeds=coords,
+        radius=radius,
+        allow_overlap=allow_overlap,
+        mask_img=mask_img,
+    )
+    nilearn_output = nilearn_masker.fit_transform(input_img)
+    # Checks
+    assert junifer_output.shape == nilearn_output.shape
+    np.testing.assert_almost_equal(junifer_output, nilearn_output)