[ENH] Consolidate stats unit tests and fix linting errors in tests (#421

)

* Consolidates stats unit tests
* Changes to simpler assert for stats errors
* Oof linting

Co-Authored-By: Elizabeth DuPre <[email protected]>

setup.cfg

@@ -8,7 +8,7 @@ parentdir_prefix =
 
 [flake8]
 max-line-length = 99
-exclude=*build/,tests
+exclude=*build/
 ignore = E126,E402,W504
 per-file-ignores =
     */__init__.py:F401

tedana/tests/test_decay.py

@@ -14,7 +14,7 @@
 @pytest.fixture(scope='module')
 def testdata1():
     tes = np.array([14.5, 38.5, 62.5])
-    in_files = [op.join(get_test_data_path(), 'echo{0}.nii.gz'.format(i+1))
+    in_files = [op.join(get_test_data_path(), 'echo{0}.nii.gz'.format(i + 1))
                 for i in range(3)]
     data, _ = io.load_data(in_files, n_echos=len(tes))
     mask, mask_sum = utils.make_adaptive_mask(data, getsum=True)
@@ -64,22 +64,21 @@ def test_fit_decay_ts(testdata1):
 
 def test_smoke_fit_decay():
     """
-    test_smoke_fit_decay tests that the function fit_decay returns reasonable 
-    objects with random inputs in the correct format 
+    test_smoke_fit_decay tests that the function fit_decay returns reasonable
+    objects with random inputs in the correct format
     """
     n_samples = 100
     n_echos = 5
     n_times = 20
     data = np.random.random((n_samples, n_echos, n_times))
     tes = np.random.random((n_echos)).tolist()
-    mask = np.random.randint(2, size=n_samples) # generate binary mask of random 0s and 1s
+    mask = np.random.randint(2, size=n_samples)  # generate binary mask of random 0s and 1s
     masksum = np.random.random((n_samples))
-    fittype='loglin'
-    t2s_limited, s0_limited, t2ss, s0vs, t2s_full, s0_full = me.fit_decay(data, 
-                                                                        tes,
-                                                                        mask,
-                                                                        masksum,
-                                                                        fittype)
+    fittype = 'loglin'
+    t2s_limited, s0_limited, t2ss, s0vs, t2s_full, s0_full = me.fit_decay(
+                                                            data, tes,
+                                                            mask, masksum,
+                                                            fittype)
     assert t2s_limited is not None
     assert s0_limited is not None
     assert t2ss is not None
@@ -90,23 +89,23 @@ def test_smoke_fit_decay():
 
 def test_smoke_fit_decay_curvefit():
     """
-    test_smoke_fit_decay tests that the function fit_decay returns reasonable 
+    test_smoke_fit_decay tests that the function fit_decay returns reasonable
     objects with random inputs in the correct format when using the direct
-    monoexponetial approach 
+    monoexponetial approach
     """
     n_samples = 100
     n_echos = 5
     n_times = 20
     data = np.random.random((n_samples, n_echos, n_times))
     tes = np.random.random((n_echos)).tolist()
-    mask = np.random.randint(2, size=n_samples) # generate binary mask of random 0s and 1s
+    mask = np.random.randint(2, size=n_samples)  # generate binary mask of random 0s and 1s
     masksum = np.random.random((n_samples))
-    fittype='curvefit'
-    t2s_limited, s0_limited, t2ss, s0vs, t2s_full, s0_full = me.fit_decay(data, 
-                                                                        tes,
-                                                                        mask,
-                                                                        masksum,
-                                                                        fittype)
+    fittype = 'curvefit'
+    t2s_limited, s0_limited, t2ss, s0vs, t2s_full, s0_full = me.fit_decay(
+                                                                data, tes,
+                                                                mask,
+                                                                masksum,
+                                                                fittype)
     assert t2s_limited is not None
     assert s0_limited is not None
     assert t2ss is not None
@@ -117,22 +116,22 @@ def test_smoke_fit_decay_curvefit():
 
 def test_smoke_fit_decay_ts():
     """
-    test_smoke_fit_decay_ts tests that the function fit_decay_ts returns reasonable 
-    objects with random inputs in the correct format 
+    test_smoke_fit_decay_ts tests that the function fit_decay_ts returns reasonable
+    objects with random inputs in the correct format
     """
     n_samples = 100
     n_echos = 5
     n_times = 20
     data = np.random.random((n_samples, n_echos, n_times))
     tes = np.random.random((n_echos)).tolist()
-    mask = np.random.randint(2, size=n_samples) # generate binary mask of random 0s and 1s
+    mask = np.random.randint(2, size=n_samples)  # generate binary mask of random 0s and 1s
     masksum = np.random.random((n_samples))
-    fittype='loglin'
-    t2s_limited_ts, s0_limited_ts, t2s_full_ts, s0_full_ts = me.fit_decay_ts(data, 
-                                                                        tes,
-                                                                        mask,
-                                                                        masksum,
-                                                                        fittype)
+    fittype = 'loglin'
+    t2s_limited_ts, s0_limited_ts, t2s_full_ts, s0_full_ts = me.fit_decay_ts(
+                                                                data, tes,
+                                                                mask,
+                                                                masksum,
+                                                                fittype)
     assert t2s_limited_ts is not None
     assert s0_limited_ts is not None
     assert t2s_full_ts is not None
@@ -141,25 +140,25 @@ def test_smoke_fit_decay_ts():
 
 def test_smoke_fit_decay_curvefit_ts():
     """
-    test_smoke_fit_decay_ts tests that the function fit_decay_ts returns reasonable 
+    test_smoke_fit_decay_ts tests that the function fit_decay_ts returns reasonable
     objects with random inputs in the correct format when using the direct
-    monoexponetial approach 
+    monoexponetial approach
     """
     n_samples = 100
     n_echos = 5
     n_times = 20
     data = np.random.random((n_samples, n_echos, n_times))
     tes = np.random.random((n_echos)).tolist()
-    mask = np.random.randint(2, size=n_samples) # generate binary mask of random 0s and 1s
+    mask = np.random.randint(2, size=n_samples)  # generate binary mask of random 0s and 1s
     masksum = np.random.random((n_samples))
-    fittype='curvefit'
-    t2s_limited_ts, s0_limited_ts, t2s_full_ts, s0_full_ts = me.fit_decay_ts(data, 
-                                                                        tes,
-                                                                        mask,
-                                                                        masksum,
-                                                                        fittype)
+    fittype = 'curvefit'
+    t2s_limited_ts, s0_limited_ts, t2s_full_ts, s0_full_ts = me.fit_decay_ts(
+                                                                data, tes,
+                                                                mask,
+                                                                masksum,
+                                                                fittype)
     assert t2s_limited_ts is not None
     assert s0_limited_ts is not None
     assert t2s_full_ts is not None
     assert s0_full_ts is not None
-# TODO: BREAK AND UNIT TESTS
+# TODO: BREAK AND UNIT TESTS

tedana/tests/test_gscontrol.py

@@ -18,22 +18,22 @@ def test_break_gscontrol_raw():
     optcom = np.empty((n_samples, n_vols))
     ref_img = ''
 
-    catd = np.empty((n_samples+1, n_echos, n_vols))
+    catd = np.empty((n_samples + 1, n_echos, n_vols))
     with pytest.raises(ValueError) as e_info:
         gsc.gscontrol_raw(catd=catd, optcom=optcom, n_echos=n_echos,
                           ref_img=ref_img, dtrank=4)
     assert str(e_info.value) == ('First dimensions of catd ({0}) and optcom ({1}) do not '
                                  'match'.format(catd.shape[0], optcom.shape[0]))
 
-    catd = np.empty((n_samples, n_echos+1, n_vols))
+    catd = np.empty((n_samples, n_echos + 1, n_vols))
     with pytest.raises(ValueError) as e_info:
         gsc.gscontrol_raw(catd=catd, optcom=optcom, n_echos=n_echos,
                           ref_img=ref_img, dtrank=4)
     assert str(e_info.value) == ('Second dimension of catd ({0}) does not match '
                                  'n_echos ({1})'.format(catd.shape[1], n_echos))
 
     catd = np.empty((n_samples, n_echos, n_vols))
-    optcom = np.empty((n_samples, n_vols+1))
+    optcom = np.empty((n_samples, n_vols + 1))
     with pytest.raises(ValueError) as e_info:
         gsc.gscontrol_raw(catd=catd, optcom=optcom, n_echos=n_echos,
                           ref_img=ref_img, dtrank=4)

tedana/tests/test_integration.py

@@ -83,7 +83,7 @@ def test_integration_five_echo(skip_integration):
                        os.path.dirname(out_dir))
     prepend = '/tmp/data/five-echo/p06.SBJ01_S09_Task11_e'
     suffix = '.sm.nii.gz'
-    datalist = [prepend + str(i+1) + suffix for i in range(5)]
+    datalist = [prepend + str(i + 1) + suffix for i in range(5)]
     tedana_workflow(
         data=datalist,
         tes=[15.4, 29.7, 44.0, 58.3, 72.6],

tedana/tests/test_io.py

@@ -16,6 +16,7 @@
 
 data_dir = get_test_data_path()
 
+
 def test_new_nii_like():
     data, ref = me.load_data(fnames, n_echos=len(tes))
     nimg = me.new_nii_like(ref, data)
@@ -61,15 +62,16 @@ def test_smoke_split_ts():
     Ensures that split_ts returns output when fed in with random inputs
     Note: classification is ["accepted", "rejected", "ignored"]
     """
-    np.random.seed(0) # seeded because comptable MUST have accepted components
+    np.random.seed(0)  # seeded because comptable MUST have accepted components
     n_samples = 100
     n_times = 20
     n_components = 6
     data = np.random.random((n_samples, n_times))
     mmix = np.random.random((n_times, n_components))
     mask = np.random.randint(2, size=n_samples)
 
-    # creating the component table with component as random floats, a "metric," and random classification
+    # creating the component table with component as random floats,
+    # a "metric," and random classification
     component = np.random.random((n_components))
     metric = np.random.random((n_components))
     classification = np.random.choice(["accepted", "rejected", "ignored"], n_components)
@@ -86,14 +88,15 @@ def test_smoke_write_split_ts():
     """
     Ensures that write_split_ts writes out the expected files with random input and tear them down
     """
-    np.random.seed(0) # at least one accepted and one rejected, thus all files are generated
+    np.random.seed(0)  # at least one accepted and one rejected, thus all files are generated
     n_samples, n_times, n_components = 64350, 10, 6
     data = np.random.random((n_samples, n_times))
     mmix = np.random.random((n_times, n_components))
     mask = np.random.randint(2, size=n_samples)
-    ref_img = os.path.join(data_dir, 'mask.nii.gz')  # ref_img has shape of (39, 50, 33) so data is 64350 (39*33*50) x 10
-
-    # creating the component table with component as random floats, a "metric," and random classification
+    ref_img = os.path.join(data_dir, 'mask.nii.gz')
+    # ref_img has shape of (39, 50, 33) so data is 64350 (39*33*50) x 10
+    # creating the component table with component as random floats,
+    # a "metric," and random classification
     component = np.random.random((n_components))
     metric = np.random.random((n_components))
     classification = np.random.choice(["accepted", "rejected", "ignored"], n_components)
@@ -103,7 +106,8 @@ def test_smoke_write_split_ts():
     assert me.write_split_ts(data, mmix, mask, comptable, ref_img) is not None
 
     # TODO: midk_ts.nii is never generated?
-    for filename in ["hik_ts_.nii.gz", "lowk_ts_.nii.gz", "dn_ts_.nii.gz"]: # remove all files generated
+    for filename in ["hik_ts_.nii.gz", "lowk_ts_.nii.gz", "dn_ts_.nii.gz"]:
+        # remove all files generated
         try:
             os.remove(filename)
         except OSError:
@@ -113,7 +117,8 @@ def test_smoke_write_split_ts():
 
 def test_smoke_writefeats():
     """
-    Ensures that writefeats writes out the expected feature with random input, since there is no suffix, remove feats_.nii
+    Ensures that writefeats writes out the expected feature with random
+    input, since there is no suffix, remove feats_.nii
     """
     n_samples, n_times, n_components = 64350, 10, 6
     data = np.random.random((n_samples, n_times))
@@ -133,9 +138,10 @@ def test_smoke_writefeats():
 
 def test_smoke_filewrite():
     """
-    Ensures that filewrite writes out a neuroimage with random input, since there is no name, remove the image named .nii
+    Ensures that filewrite writes out a neuroimage with random input,
+    since there is no name, remove the image named .nii
     """
-    n_samples, n_times, n_components = 64350, 10, 6
+    n_samples, n_times, _ = 64350, 10, 6
     data_1d = np.random.random((n_samples))
     data_2d = np.random.random((n_samples, n_times))
     filename = ""

tedana/tests/test_model_fit_dependence_metrics.py

@@ -18,80 +18,57 @@ def test_break_dependence_metrics():
     tsoc = np.empty((n_samples, n_vols))
     mmix = np.empty((n_vols, n_comps))
     t2s = np.empty((n_samples, n_vols))
-    t2s_full = np.empty((n_samples, n_vols))
     tes = np.empty((n_echos))
-    combmode = 't2s'
     ref_img = ''
 
     # Shape of catd is wrong
-    catd = np.empty((n_samples+1, n_echos, n_vols))
-    with pytest.raises(ValueError) as e_info:
+    catd = np.empty((n_samples + 1, n_echos, n_vols))
+    with pytest.raises(ValueError):
         kundu_fit.dependence_metrics(catd=catd, tsoc=tsoc, mmix=mmix,
-                               t2s=t2s, tes=tes, ref_img=ref_img,
-                               reindex=False, mmixN=None, algorithm='kundu_v3')
-    assert str(e_info.value) == ('First dimensions (number of samples) of '
-                                 'catd ({0}), tsoc ({1}), '
-                                 'and t2s ({2}) do not match'.format(
-                                    catd.shape[0], tsoc.shape[0],
-                                    t2s.shape[0]))
+                                     t2s=t2s, tes=tes, ref_img=ref_img,
+                                     reindex=False, mmixN=None,
+                                     algorithm='kundu_v3')
 
     # Shape of t2s is wrong
     catd = np.empty((n_samples, n_echos, n_vols))
-    t2s = np.empty((n_samples+1, n_vols))
-    with pytest.raises(ValueError) as e_info:
+    t2s = np.empty((n_samples + 1, n_vols))
+    with pytest.raises(ValueError):
         kundu_fit.dependence_metrics(catd=catd, tsoc=tsoc, mmix=mmix,
-                               t2s=t2s, tes=tes, ref_img=ref_img,
-                               reindex=False, mmixN=None, algorithm='kundu_v3')
-    assert str(e_info.value) == ('First dimensions (number of samples) of '
-                                 'catd ({0}), tsoc ({1}), '
-                                 'and t2s ({2}) do not match'.format(
-                                    catd.shape[0], tsoc.shape[0],
-                                    t2s.shape[0]))
+                                     t2s=t2s, tes=tes, ref_img=ref_img,
+                                     reindex=False, mmixN=None,
+                                     algorithm='kundu_v3')
 
     # Shape of tsoc is wrong
     t2s = np.empty((n_samples, n_vols))
-    tsoc = np.empty((n_samples+1, n_vols))
-    with pytest.raises(ValueError) as e_info:
+    tsoc = np.empty((n_samples + 1, n_vols))
+    with pytest.raises(ValueError):
         kundu_fit.dependence_metrics(catd=catd, tsoc=tsoc, mmix=mmix,
-                               t2s=t2s, tes=tes, ref_img=ref_img,
-                               reindex=False, mmixN=None, algorithm='kundu_v3')
-    assert str(e_info.value) == ('First dimensions (number of samples) of '
-                                 'catd ({0}), tsoc ({1}), '
-                                 'and t2s ({2}) do not match'.format(
-                                    catd.shape[0], tsoc.shape[0],
-                                    t2s.shape[0]))
+                                     t2s=t2s, tes=tes, ref_img=ref_img,
+                                     reindex=False, mmixN=None,
+                                     algorithm='kundu_v3')
 
     # Shape of catd is wrong
-    catd = np.empty((n_samples, n_echos+1, n_vols))
+    catd = np.empty((n_samples, n_echos + 1, n_vols))
     tsoc = np.empty((n_samples, n_vols))
-    with pytest.raises(ValueError) as e_info:
+    with pytest.raises(ValueError):
         kundu_fit.dependence_metrics(catd=catd, tsoc=tsoc, mmix=mmix,
-                               t2s=t2s, tes=tes, ref_img=ref_img,
-                               reindex=False, mmixN=None, algorithm='kundu_v3')
-    assert str(e_info.value) == ('Second dimension of catd ({0}) does not '
-                                 'match number '
-                                 'of echoes provided (tes; '
-                                 '{1})'.format(catd.shape[1], len(tes)))
+                                     t2s=t2s, tes=tes, ref_img=ref_img,
+                                     reindex=False, mmixN=None,
+                                     algorithm='kundu_v3')
 
     # Shape of catd is wrong
-    catd = np.empty((n_samples, n_echos, n_vols+1))
-    with pytest.raises(ValueError) as e_info:
+    catd = np.empty((n_samples, n_echos, n_vols + 1))
+    with pytest.raises(ValueError):
         kundu_fit.dependence_metrics(catd=catd, tsoc=tsoc, mmix=mmix,
-                               t2s=t2s, tes=tes, ref_img=ref_img,
-                               reindex=False, mmixN=None, algorithm='kundu_v3')
-    assert str(e_info.value) == ('Number of volumes in catd '
-                                 '({0}), tsoc ({1}), and '
-                                 'mmix ({2}) do not '
-                                 'match.'.format(catd.shape[2], tsoc.shape[1],
-                                                 mmix.shape[0]))
+                                     t2s=t2s, tes=tes, ref_img=ref_img,
+                                     reindex=False, mmixN=None,
+                                     algorithm='kundu_v3')
 
     # Shape of t2s is wrong
     catd = np.empty((n_samples, n_echos, n_vols))
-    t2s = np.empty((n_samples, n_vols+1))
-    with pytest.raises(ValueError) as e_info:
+    t2s = np.empty((n_samples, n_vols + 1))
+    with pytest.raises(ValueError):
         kundu_fit.dependence_metrics(catd=catd, tsoc=tsoc, mmix=mmix,
-                               t2s=t2s, tes=tes, ref_img=ref_img,
-                               reindex=False, mmixN=None, algorithm='kundu_v3')
-    assert str(e_info.value) == ('Number of volumes in catd ({0}) '
-                                 'does not match number of volumes in '
-                                 't2s ({1})'.format(catd.shape[2], t2s.shape[1]))
+                                     t2s=t2s, tes=tes, ref_img=ref_img,
+                                     reindex=False, mmixN=None,
+                                     algorithm='kundu_v3')

tedana/tests/test_model_kundu_metrics.py

@@ -2,7 +2,6 @@
 Tests for tedana.model.fit
 """
 
-import pytest
 import numpy as np
 import pandas as pd
 

tedana/tests/test_selection.py

@@ -2,9 +2,6 @@
 Tests for tedana.selection
 """
 
-import os.path as op
-
-import pytest
 import numpy as np
 import pandas as pd