numenta/nupic.core-legacy#1380: Fix SP tests with correct dtype values

lscheinkman · Jan 11, 2018 · 100166e · 100166e
1 parent 90eea66
commit 100166e
Show file tree

Hide file tree

Showing 4 changed files with 25 additions and 7 deletions.
diff --git a/tests/unit/nupic/algorithms/sp_overlap_test.py b/tests/unit/nupic/algorithms/sp_overlap_test.py
@@ -130,10 +130,10 @@ def frequency(self,
                                  maxval=maxVal, periodic=False, forced=True) # forced: it's strongly recommended to use w>=21, in the example we force skip the check for readibility
       for y in xrange(numColors):
         temp = enc.encode(rnd.random()*maxVal)
-        colors.append(numpy.array(temp, dtype=realDType))
+        colors.append(numpy.array(temp, dtype=numpy.uint32))
     else:
       for y in xrange(numColors):
-        sdr = numpy.zeros(n, dtype=realDType)
+        sdr = numpy.zeros(n, dtype=numpy.uint32)
         # Randomly setting w out of n bits to 1
         sdr[rnd.sample(xrange(n), w)] = 1
         colors.append(sdr)
@@ -144,7 +144,7 @@ def frequency(self,
     for i in xrange(numColors):
       # TODO: See https://github.com/numenta/nupic/issues/2072
       spInput = colors[i]
-      onCells = numpy.zeros(columnDimensions)
+      onCells = numpy.zeros(columnDimensions, dtype=numpy.uint32)
       spImpl.compute(spInput, True, onCells)
       spOutput.append(onCells.tolist())
       activeCoincIndices = set(onCells.nonzero()[0])

diff --git a/tests/unit/nupic/algorithms/spatial_pooler_cpp_unit_test.py b/tests/unit/nupic/algorithms/spatial_pooler_cpp_unit_test.py
@@ -185,12 +185,30 @@ def testSameOutputForSameInputNoLearningNoBoosting(self):
 
     out1 = np.zeros(columns, dtype=uintDType)
     out2 = np.zeros(columns, dtype=uintDType)
-    inputVector = np.array([1, 1, 0, 0, 1, 1, 0, 0, 1, 1], dtype=realDType)
+    inputVector = np.array([1, 1, 0, 0, 1, 1, 0, 0, 1, 1], dtype=uintDType)
 
     sp.compute(inputVector, False, out1);
     sp.compute(inputVector, False, out2);
     self.assertEqual(list(out1), list(out2))
 
+  def testComputeParametersValidation(self):
+    sp = SpatialPooler(inputDimensions=[5], columnDimensions=[5])
+    inputGood = np.ones(5, dtype=uintDType)
+    outGood = np.zeros(5, dtype=uintDType)
+    inputBad = np.ones(5, dtype=realDType)
+    outBad = np.zeros(5, dtype=realDType)
+
+    # Validate good parameters
+    sp.compute(inputGood, False, outGood)
+
+    # Validate bad input
+    with self.assertRaises(RuntimeError):
+      sp.compute(inputBad, False, outGood)
+
+    # Validate bad output
+    with self.assertRaises(RuntimeError):
+      sp.compute(inputGood, False, outBad)
+
 
 
 if __name__ == "__main__":

diff --git a/tests/unit/nupic/algorithms/spatial_pooler_py_api_test.py b/tests/unit/nupic/algorithms/spatial_pooler_py_api_test.py
@@ -42,8 +42,8 @@ def setUp(self):
 
   def testCompute(self):
     # Check that there are no errors in call to compute
-    inputVector = numpy.ones(5)
-    activeArray = numpy.zeros(5)
+    inputVector = numpy.ones(5, dtype=uintType)
+    activeArray = numpy.zeros(5, dtype=uintType)
     self.sp.compute(inputVector, True, activeArray)
 
 

diff --git a/tests/unit/nupic/algorithms/spatial_pooler_unit_test.py b/tests/unit/nupic/algorithms/spatial_pooler_unit_test.py
@@ -1579,7 +1579,7 @@ def testSameOutputForSameInputNoLearningNoBoosting(self):
 
     out1 = numpy.zeros(columns, dtype=uintDType)
     out2 = numpy.zeros(columns, dtype=uintDType)
-    inputVector = numpy.array([1, 1, 0, 0, 1, 1, 0, 0, 1, 1], dtype=realDType)
+    inputVector = numpy.array([1, 1, 0, 0, 1, 1, 0, 0, 1, 1], dtype=uintDType)
 
     sp.compute(inputVector, False, out1);
     sp.compute(inputVector, False, out2);