FIX: Better fix

Author: larsoner

mne/cov.py

@@ -1161,6 +1161,7 @@ def _unpack_epochs(epochs):
 
     epochs = np.hstack(epochs)
     n_samples_tot = epochs.shape[-1]
+    _check_n_samples(n_samples_tot, len(picks_meeg), on_few_samples)
 
     epochs = epochs.T  # sklearn | C-order
     cov_data = _compute_covariance_auto(

mne/preprocessing/tests/test_maxwell.py

@@ -1883,13 +1883,13 @@ def test_compute_maxwell_basis(regularize, n, int_order):
     raw = read_raw_fif(raw_small_fname).crop(0, 2)
     assert raw.info["bads"] == []
     raw.del_proj()
-    rank = compute_rank(raw, on_few_samples="ignore")["meg"]
+    rank = compute_rank(raw)["meg"]
     assert rank == 306
     raw.info["bads"] = ["MEG 2443"]
     kwargs = dict(regularize=regularize, int_order=int_order, verbose=True)
     raw_sss = maxwell_filter(raw, **kwargs)
     want = raw_sss.get_data("meg")
-    rank = compute_rank(raw_sss, on_few_samples="ignore")["meg"]
+    rank = compute_rank(raw_sss)["meg"]
     assert rank == n
     S, pS, reg_moments, n_use_in = compute_maxwell_basis(raw.info, **kwargs)
     assert n_use_in == n

mne/rank.py

@@ -200,11 +200,15 @@ def _estimate_rank_meeg_signals(
         If return_singular is True, the singular values that were
         thresholded to determine the rank are also returned.
     """
-    from .cov import _check_n_samples
-
     picks_list = _picks_by_type(info)
     assert data.ndim == 2, data.shape
-    _check_n_samples(*data.shape[::-1], on_few_samples=on_few_samples)
+    n_channels, n_samples = data.shape
+    if n_samples < n_channels:
+        msg = (
+            f"Too few samples ({n_samples=} is less than {n_channels=}), "
+            "rank estimate may be unreliable"
+        )
+        _on_missing(on_few_samples, msg, "on_few_samples")
     with _scaled_array(data, picks_list, scalings):
         out = estimate_rank(
             data,

mne/tests/test_cov.py

@@ -4,7 +4,6 @@
 
 import itertools as itt
 import sys
-from contextlib import nullcontext
 from inspect import signature
 from pathlib import Path
 
@@ -802,11 +801,7 @@ def test_low_rank_methods(rank, raw_epochs_events):
             empirical=(-15000, -5000), diagonal_fixed=(-700, -600), oas=(-700, -600)
         ),
     }
-    if rank is None:
-        ctx = pytest.warns(RuntimeWarning, match="Too few samples")
-    else:
-        ctx = nullcontext()
-    with ctx:
+    with pytest.warns(RuntimeWarning, match="Too few samples"):
         covs = compute_covariance(
             epochs, method=methods, return_estimators=True, rank=rank, verbose=True
         )
@@ -858,7 +853,8 @@ def test_low_rank_cov(raw_epochs_events):
         epochs_meg, method="oas", rank="full", verbose="error"
     )
     assert _cov_rank(cov_full, epochs_meg.info) == 306
-    cov_dict = compute_covariance(epochs_meg, method="oas", rank=dict(meg=306))
+    with pytest.warns(RuntimeWarning, match="few samples"):
+        cov_dict = compute_covariance(epochs_meg, method="oas", rank=dict(meg=306))
     assert _cov_rank(cov_dict, epochs_meg.info) == 306
     assert_allclose(cov_full["data"], cov_dict["data"])
     cov_dict = compute_covariance(