Make sure that the updates of sitsot shared variables are initially tensor variables

theano/scan_module/scan.py

@@ -870,7 +870,8 @@ def scan(fn,
                         tensor.unbroadcast(
                             tensor.shape_padleft(input.variable), 0),
                         actual_n_steps))
-                sit_sot_inner_outputs.append(input.update)
+                tensor_update = tensor.as_tensor_variable(input.update)
+                sit_sot_inner_outputs.append(tensor_update)
                 # Not that pos is not a negative index. The sign of pos is used
                 # as a flag to indicate if this output should be part of the
                 # update rules or part of the standard outputs of scan.

theano/scan_module/tests/test_scan.py

@@ -1363,6 +1363,19 @@ class T_Scan(unittest.TestCase):
 
         assert_raises(TypeError, cPickle.load, open(path, "r"))
 
+    def test_consistent_inner_fct(self):
+        # Test that scan does not falsely detect inconsistencies in a valid
+        # inner graph
+
+        # The pickled scan op used in this test requires the use of a gpu
+        from theano.sandbox import cuda
+        if not cuda.cuda_available:
+            raise SkipTest('Optional package cuda disabled')
+
+        rs = theano.sandbox.rng_mrg.MRG_RandomStreams()
+        output, _ = theano.scan(lambda : rs.uniform((3,)), n_steps=3)
+        cPickle.loads(cPickle.dumps(output))
+
     def test_cuda_gibbs_chain(self):
         from theano.sandbox import cuda
         if not cuda.cuda_available: