Merge pull request #100 from delallea/prep_tests_new_grad_type

theano/tensor/tensor_grad.py

@@ -632,7 +632,8 @@ def verify_grad(fun, pt, n_tests=2, rng=None, eps=None, abs_tol=None, rel_tol=No
         g_cost = cast(g_cost, o_output.dtype)
 
     symbolic_grad = grad(cost, tensor_pt, g_cost,
-                         disconnected_inputs='ignore')
+                         disconnected_inputs='ignore',
+                         keep_wrt_type=True)
     #if o_output.dtype in ['float32','float64']:
     #    assert all([x.dtype == o_output.dtype for x in symbolic_grad]),("Expected grad of type %s, got %s "%( symbolic_grad.dtype, o_output.dtyp))
 
@@ -644,8 +645,8 @@ def verify_grad(fun, pt, n_tests=2, rng=None, eps=None, abs_tol=None, rel_tol=No
 
         analytic_grad = grad_fn(*[p.copy() for p in pt])
 
-        if not isinstance(analytic_grad, (list, tuple)):
-            analytic_grad = [analytic_grad]
+        # Since `tensor_pt` is a list, `analytic_grad` should be one too.
+        assert isinstance(analytic_grad, list)
 
         max_arg, max_err_pos, max_abs_err, max_rel_err =\
                 num_grad.max_err(analytic_grad, abs_tol, rel_tol)

theano/tensor/tests/test_blas.py

@@ -808,13 +808,13 @@ def test_dot_w_self():
     # This can trigger problems in the optimization because what would normally be a gemm must
     # not be because the output is aliased to one of the inputs.
 
-    A = shared(value = numpy.ones((2,2)))
+    A = shared(value=numpy.ones((2,2)))
     B = T.matrix()
 
     p = T.dot(A,A)*B
 
-    grad = T.grad(T.mean(p),[A])
-    f = theano.function([B], p, updates = { A : A - grad[0]} )
+    grad = T.grad(T.mean(p), A)
+    f = theano.function([B], p, updates={A : A - grad})
 
     # tests correctness in debugmode
     f(numpy.asarray([[0,1], [2,3]], dtype=config.floatX))