Slice epsilons with dead channels to match gains during agc.

src/spikeinterface/preprocessing/highpass_spatial_filter.py

@@ -131,7 +131,7 @@ def __init__(
                 rms_values = recording.get_property("noise_level_rms_raw")
             else:
                 random_slice_kwargs = {} if random_slice_kwargs is None else random_slice_kwargs
-                rms_values = get_noise_levels(recording, method="rms", return_scaled=False, **random_slice_kwargs)
+                rms_values = get_noise_levels(recording, method="rms", return_in_uV=False, **random_slice_kwargs)
 
         # Pre-compute spatial filtering parameters
         butter_kwargs = dict(btype="highpass", N=highpass_butter_order, Wn=highpass_butter_wn)
@@ -308,7 +308,9 @@ def agc(traces, window, epsilons):
 
     dead_channels = np.sum(gain, axis=0) == 0
 
-    traces[:, ~dead_channels] = traces[:, ~dead_channels] / np.maximum(epsilons, gain[:, ~dead_channels])
+    traces[:, ~dead_channels] = traces[:, ~dead_channels] / np.maximum(
+        epsilons[~dead_channels], gain[:, ~dead_channels]
+    )
 
     return traces, gain
 

src/spikeinterface/preprocessing/tests/test_highpass_spatial_filter.py

@@ -6,7 +6,7 @@
 import spikeinterface.core as si
 import spikeinterface.preprocessing as spre
 import spikeinterface.extractors as se
-from spikeinterface.core import generate_recording
+from spikeinterface.core import generate_recording, NumpyRecording
 import importlib.util
 
 ON_GITHUB = bool(os.getenv("GITHUB_ACTIONS"))
@@ -103,6 +103,27 @@ def test_highpass_spatial_filter_synthetic_data(num_channels, ntr_pad, ntr_tap,
             assert raw_traces.shape == si_filtered.shape
 
 
+def test_highpass_spatial_filter_with_dead_channels():
+    """Regression test: AGC must handle dead (all-zero) channels without broadcast error.
+
+    PR #4286 changed epsilon from a scalar to a per-channel array, but the agc()
+    function indexed gain with ~dead_channels without applying the same mask to
+    epsilons, causing a broadcast error when any channels had zero signal.
+    """
+    num_channels = 32
+    rec = generate_recording(num_channels=num_channels, durations=[0.5])
+    # Materialize traces and zero out 3 channels to make them "dead"
+    traces = rec.get_traces().copy()
+    traces[:, [0, 15, 31]] = 0.0
+    rec_with_dead = NumpyRecording(
+        traces_list=[traces], sampling_frequency=rec.sampling_frequency, channel_ids=rec.channel_ids
+    )
+    rec_with_dead.set_probe(rec.get_probe(), in_place=True)
+    filtered = spre.highpass_spatial_filter(rec_with_dead, n_channel_pad=2)
+    result = filtered.get_traces()
+    assert result.shape == traces.shape
+
+
 @pytest.mark.parametrize("dtype", [np.int16, np.float32, np.float64])
 def test_dtype_stability(dtype):
     """