assert that convolution performed in the test is as specified

theano/tensor/nnet/tests/test_abstract_conv.py

@@ -1736,6 +1736,8 @@ class Grouped_conv_noOptim(unittest.TestCase):
             else:
                 grouped_conv_output = grouped_abstractconv_op(img_sym, kern_sym)
             grouped_func = theano.function([img_sym, kern_sym], grouped_conv_output, mode=self.mode)
+            assert any([isinstance(node.op, self.conv2d)
+                       for node in grouped_func.maker.fgraph.toposort()])
             grouped_output = grouped_func(img, kern)
 
             normal_conv_op = conv2d_corr(img_sym,
@@ -1772,6 +1774,8 @@ class Grouped_conv_noOptim(unittest.TestCase):
             if self.flip_filter:
                 grouped_conv_output = grouped_conv_output[:, :, ::-1, ::-1]
             grouped_func = theano.function([img_sym, top_sym], grouped_conv_output, mode=self.mode)
+            assert any([isinstance(node.op, self.conv2d_gradw)
+                       for node in grouped_func.maker.fgraph.toposort()])
             grouped_output = grouped_func(img, top)
 
             normal_conv_op = conv2d_corr_gw(img_sym,
@@ -1814,6 +1818,8 @@ class Grouped_conv_noOptim(unittest.TestCase):
             else:
                 grouped_conv_output = grouped_abstractconvgrad_op(kern_sym, top_sym, imshp[-2:])
             grouped_func = theano.function([kern_sym, top_sym], grouped_conv_output, mode=self.mode)
+            assert any([isinstance(node.op, self.conv2d_gradi)
+                       for node in grouped_func.maker.fgraph.toposort()])
             grouped_output = grouped_func(kern, top)
 
             normal_conv_op = conv2d_corr_gi(kern_sym,