Create test_up_down_state.py

tests/test_up_down_state.py

@@ -0,0 +1,43 @@
+import nelpy as nel
+import numpy as np
+
+from neuro_py.detectors.up_down_state import detect_up_down_states
+
+
+def test_detect_up_down_states_detection():
+
+    # Mock data generation
+    down_state_times = [4, 7, 15]
+
+    spikes_upstate = []
+    for i in range(10):
+        up_spikes = np.random.poisson(0.25, 20_000)
+        spikes = np.where(up_spikes > 0)[0] * 0.001
+        # silence around down states
+        for down_state_time in down_state_times:
+            spikes = spikes[
+                ~((spikes > down_state_time - 0.1) & (spikes < down_state_time + 0.1))
+            ]
+        spikes_upstate.append(spikes)
+
+    nrem_epochs = nel.EpochArray([np.array([[0, 20]])])
+    st = nel.SpikeTrainArray(timestamps=spikes_upstate, fs=1000, support=nrem_epochs)
+
+    # Run the function with mock data
+    down_state_epochs, up_state_epochs = detect_up_down_states(
+        st=st,
+        nrem_epochs=nrem_epochs,
+        save_mat=False,  # Disable saving
+    )
+
+    # Assertions to check detection accuracy
+    assert down_state_epochs is not None, "Expected detected DOWN states, got None."
+    assert up_state_epochs is not None, "Expected detected UP states, got None."
+
+    # Check DOWN states detected in low activity periods
+    down_state_intervals = down_state_epochs.time
+    for down_state_time in down_state_times:
+        assert any(
+            (down_state_time >= down_state_intervals[:, 0])
+            & (down_state_time <= down_state_intervals[:, 1])
+        ), f"DOWN state at {down_state_time} not detected."