pass shape as tensorvariables and remove condition

theano/tensor/nnet/tests/test_abstract_conv.py

@@ -1717,7 +1717,7 @@ class Grouped_conv_noOptim(unittest.TestCase):
         self.top_shape = [(5, 6, 3, 3), (4, 6, 3, 3), (3, 4, 3, 1), (2, 4, 5, 3)]
         self.filter_dilation = (1, 1)
         self.ref_mode = 'FAST_RUN'
-        if theano.config.cxx == "" or not theano.config.blas.ldflags:
+        if theano.config.cxx == "":
             raise SkipTest("CorrMM needs cxx and blas")
 
     def test_fwd(self):
@@ -1772,7 +1772,7 @@ class Grouped_conv_noOptim(unittest.TestCase):
                                                     subsample=self.subsample,
                                                     filter_dilation=self.filter_dilation,
                                                     num_groups=groups)
-            grouped_conv_output = grouped_convgrad_op(img_sym, top_sym, kshp[-2:])
+            grouped_conv_output = grouped_convgrad_op(img_sym, top_sym, tensor.as_tensor_variable(kshp[-2:]))
             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)
@@ -1795,7 +1795,7 @@ class Grouped_conv_noOptim(unittest.TestCase):
             utt.assert_allclose(grouped_output, ref_concat_output)
 
             def conv_gradweight(inputs_val, output_val):
-                return grouped_convgrad_op(inputs_val, output_val, kshp[-2:])
+                return grouped_convgrad_op(inputs_val, output_val, tensor.as_tensor_variable(kshp[-2:]))
 
             utt.verify_grad(conv_gradweight,
                             [img, top],
@@ -1815,9 +1815,9 @@ class Grouped_conv_noOptim(unittest.TestCase):
                                                     filter_dilation=self.filter_dilation,
                                                     num_groups=groups)
             if self.flip_filter:
-                grouped_conv_output = grouped_convgrad_op(kern_sym[:, :, ::-1, ::-1], top_sym, imshp[-2:])
+                grouped_conv_output = grouped_convgrad_op(kern_sym[:, :, ::-1, ::-1], top_sym, tensor.as_tensor_variable(imshp[-2:]))
             else:
-                grouped_conv_output = grouped_convgrad_op(kern_sym, top_sym, imshp[-2:])
+                grouped_conv_output = grouped_convgrad_op(kern_sym, top_sym, tensor.as_tensor_variable(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()])
@@ -1838,7 +1838,7 @@ class Grouped_conv_noOptim(unittest.TestCase):
             utt.assert_allclose(grouped_output, ref_concat_output)
 
             def conv_gradinputs(filters_val, output_val):
-                return grouped_convgrad_op(filters_val, output_val, imshp[2:])
+                return grouped_convgrad_op(filters_val, output_val, tensor.as_tensor_variable(imshp[-2:]))
 
             utt.verify_grad(conv_gradinputs,
                             [kern, top],