Let filter_variable fix broadcast between GPU vars

theano/gpuarray/tests/test_type.py

@@ -91,6 +91,22 @@ def test_filter_float():
         del theano.compile.sharedvalue.shared.constructors[-1]
 
 
+def test_filter_variable():
+    # Test that filter_variable accepts more restrictive broadcast
+    gpu_row = GpuArrayType(dtype=theano.config.floatX,
+                           broadcastable=(True, False))
+    gpu_matrix = GpuArrayType(dtype=theano.config.floatX,
+                              broadcastable=(False, False))
+    r = gpu_row()
+    m = gpu_matrix.filter_variable(r)
+    assert m.type == gpu_matrix
+
+    # On CPU as well
+    r = theano.tensor.row()
+    m = gpu_matrix.filter_variable(r)
+    assert m.type == gpu_matrix
+
+
 def test_gpuarray_shared_scalar():
     # By default, we don't put scalar as shared variable on the GPU
     nose.tools.assert_raises(

theano/gpuarray/type.py

@@ -303,8 +303,6 @@ class GpuArrayType(Type):
         return data
 
     def filter_variable(self, other, allow_convert=True):
-        from theano.gpuarray.basic_ops import GpuFromHost
-
         if hasattr(other, '_as_GpuArrayVariable'):
             other = other._as_GpuArrayVariable(self.context_name)
 
@@ -314,7 +312,7 @@ class GpuArrayType(Type):
         if other.type == self:
             return other
 
-        if not isinstance(other.type, tensor.TensorType):
+        if not isinstance(other.type, (TensorType, GpuArrayType)):
             raise TypeError('Incompatible type', (self, other.type))
         if (other.type.dtype != self.dtype):
             raise TypeError('Incompatible dtype', (self.dtype,
@@ -335,7 +333,7 @@ class GpuArrayType(Type):
                                  str(self.broadcastable)))
             other = other2
 
-        return GpuFromHost(self.context_name)(other)
+        return other.transfer(self.context_name)
 
     @staticmethod
     def values_eq(a, b, force_same_dtype=True):