Split tests for seizure_detect.py

README.rst

@@ -31,6 +31,9 @@ Cite
 """""""""""""""""""""""""""
 When use whole, parts, or are inspired by, we appreciate you acknowledge and refer these journal papers:
 
+
+
+
 | Mivalt F, Kremen V, Sladky V, Balzekas I, Nejedly P, Gregg N, Lundstrom BN, Lepkova K, Pridalova T, Brinkmann BH, et al. Electrical brain stimulation and continuous behavioral state tracking in ambulatory humans. J Neural Eng (2022) Available at: http://iopscience.iop.org/article/10.1088/1741-2552/ac4bfd
 
 

best/seizure_detection/seizure_detect.py

@@ -72,7 +72,6 @@ def preprocess_input(x, fs):
     else:
         raise ValueError("too many NaN values in input")
 
-
 def infer_seizure_probability(x, model, use_cuda=False, cuda_number=0):
     """
     infers seizure probability for a given input x; recommended signal len is 300 seconds.

best/test/test_best_deep_learning.py

@@ -8,43 +8,11 @@
 import unittest
 from unittest import TestCase
 
-import torch.cuda
-
-from best.seizure_detection.seizure_detect import load_trained_model, preprocess_input, infer_seizure_probability
-from numpy.random import rand
-from best.deep_learning.models_generic import *
-
-
 class Test_deep_learning(TestCase):
     def test_import(self):
         print("Testing import 'from best.deep_learning'" )
 
 
-class Test_seizure_detect(TestCase):
-    def test_load_trained_model(self):
-        modelA = load_trained_model('modelA')
-        modelB = load_trained_model('modelB')
-
-    def test_preprocess_input(self):
-        modelA = load_trained_model('modelA')
-        fs = 500
-        x_len = 300
-        channels = 1
-        x_input = rand(channels, fs * x_len)
-        x = preprocess_input(x_input, fs)
-        y = infer_seizure_probability(x, modelA)
-        self.assertEqual(599, y.shape[1])
-        channels = 3
-        x_input = rand(channels, fs * x_len)
-        x = preprocess_input(x_input, fs)
-        y = infer_seizure_probability(x, modelA)
-        self.assertEqual(599, y.shape[1])
-        if torch.cuda.is_available():
-            modelA.cuda(0)
-            y = infer_seizure_probability(x, modelA, True, 0)
-            self.assertEquals(599, y.shape[1])
-
-
 
 if __name__ == '__main__':
     unittest.main()

best/test/test_best_seizure_detection.py

@@ -0,0 +1,46 @@
+# Copyright 2020-present, Mayo Clinic Department of Neurology
+# All rights reserved.
+#
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+
+import unittest
+from unittest import TestCase
+
+import torch.cuda
+
+from best.seizure_detection.seizure_detect import load_trained_model, preprocess_input, infer_seizure_probability
+from numpy.random import rand
+from best.deep_learning.models_generic import *
+
+
+
+class Test_seizure_detect(TestCase):
+    def test_load_trained_model(self):
+        modelA = load_trained_model('modelA')
+        modelB = load_trained_model('modelB')
+
+    def test_preprocess_input(self):
+        modelA = load_trained_model('modelA')
+        fs = 500
+        x_len = 300
+        channels = 1
+        x_input = rand(channels, fs * x_len)
+        x = preprocess_input(x_input, fs)
+        y = infer_seizure_probability(x, modelA)
+        self.assertEqual(599, y.shape[1])
+        channels = 3
+        x_input = rand(channels, fs * x_len)
+        x = preprocess_input(x_input, fs)
+        y = infer_seizure_probability(x, modelA)
+        self.assertEqual(599, y.shape[1])
+        if torch.cuda.is_available():
+            modelA.cuda(0)
+            y = infer_seizure_probability(x, modelA, True, 0)
+            self.assertEquals(599, y.shape[1])
+
+
+
+if __name__ == '__main__':
+    unittest.main()