Merge pull request #908 from lamblin/fix_test_neib

theano/sandbox/cuda/neighbours.py

@@ -404,6 +404,7 @@ def gpu_images2neibs(ten4, neib_shape, neib_step=None, mode='valid'):
 @local_optimizer()
 def use_gpu_images2neibs(node):
     if (type(node.op) is Images2Neibs and
+        node.inputs[0].dtype == 'float32' and
         node.op.mode in ['valid', 'wrap_centered']):
         return [host_from_gpu(gpu_images2neibs(gpu_from_host(node.inputs[0]),
                                                node.inputs[1], node.inputs[2],

theano/sandbox/cuda/nvcc_compiler.py

@@ -139,6 +139,7 @@ class NVCC_compiler(object):
                                         default_to_move_computation_to_gpu=False,
                                         move_shared_float32_to_gpu=False,
                                         enable_cuda=False)
+                n = theano.sandbox.cuda.use.device_number
 
             p = theano.sandbox.cuda.device_properties(n)
             flags.append('-arch=sm_' + str(p['major']) + str(p['minor']))

theano/sandbox/cuda/tests/test_neighbours.py

@@ -20,6 +20,7 @@ else:
 class T_GpuImages2Neibs(theano.sandbox.test_neighbours.T_Images2Neibs):
     mode = mode_with_gpu
     op = GpuImages2Neibs
+    dtypes = ['float32']
 
 if __name__ == '__main__':
     unittest.main()

theano/sandbox/test_neighbours.py

@@ -18,9 +18,11 @@ else:
 if not theano.config.cxx:
     raise SkipTest("G++ not available, so we need to skip this test.")
 
+
 class T_Images2Neibs(unittest_tools.InferShapeTester):
     mode = mode_without_gpu
     op = Images2Neibs
+    dtypes = ['int64', 'float32', 'float64']
 
     def test_neibs(self):
         for shape, pshape in [((100, 40, 18, 18), (2, 2)),
@@ -29,17 +31,23 @@ class T_Images2Neibs(unittest_tools.InferShapeTester):
                               ((10, 40, 68, 66), (34, 33))
                                   ]:
             for border in ['valid', 'ignore_borders']:
-                images = shared(numpy.arange(numpy.prod(shape)).reshape(shape))
+                for dtype in self.dtypes:
+                    images = shared(
+                            numpy.arange(numpy.prod(shape), dtype=dtype
+                            ).reshape(shape))
                     neib_shape = T.as_tensor_variable(pshape)
 
-                f = function([], images2neibs(images, neib_shape, mode=border),
+                    f = function([],
+                                 images2neibs(images, neib_shape, mode=border),
                                  mode=self.mode)
 
                     #print images.get_value(borrow=True)
                     neibs = f()
                     #print neibs
-                g = function([], neibs2images(neibs, neib_shape, images.shape),
+                    g = function([],
+                                 neibs2images(neibs, neib_shape, images.shape),
                                  mode=self.mode)
+                    if border in ['valid']:
                         assert any([isinstance(node.op, self.op)
                                     for node in f.maker.fgraph.toposort()])
 
@@ -48,7 +56,10 @@ class T_Images2Neibs(unittest_tools.InferShapeTester):
 
     def test_neibs_manual(self):
         shape = (2, 3, 4, 4)
-        images = shared(numpy.arange(numpy.prod(shape)).reshape(shape))
+        for dtype in self.dtypes:
+            images = shared(
+                    numpy.arange(numpy.prod(shape), dtype=dtype
+                    ).reshape(shape))
             neib_shape = T.as_tensor_variable((2, 2))
 
             for border in ['valid', 'ignore_borders']:
@@ -89,15 +100,19 @@ class T_Images2Neibs(unittest_tools.InferShapeTester):
 
     def test_neibs_manual_step(self):
         shape = (2, 3, 5, 5)
+        for dtype in self.dtypes:
             images = shared(numpy.asarray(numpy.arange(numpy.prod(
-                        shape)).reshape(shape), dtype='float32'))
+                            shape)).reshape(shape), dtype=dtype))
             neib_shape = T.as_tensor_variable((3, 3))
             neib_step = T.as_tensor_variable((2, 2))
             for border in ['valid', 'ignore_borders']:
-            f = function([], images2neibs(images, neib_shape, neib_step, mode=border),
+                f = function([],
+                             images2neibs(images, neib_shape, neib_step,
+                                          mode=border),
                              mode=self.mode)
 
                 neibs = f()
+                if border in ['valid']:
                     assert self.op in [type(node.op)
                                        for node in f.maker.fgraph.toposort()]
 
@@ -136,15 +151,21 @@ class T_Images2Neibs(unittest_tools.InferShapeTester):
 
     def test_neibs_bad_shape(self):
         shape = (2, 3, 10, 10)
-        images = shared(numpy.arange(numpy.prod(shape)).reshape(shape))
+        for dtype in self.dtypes:
+            images = shared(numpy.arange(
+                numpy.prod(shape), dtype=dtype
+                ).reshape(shape))
 
             for neib_shape in [(3, 2), (2, 3)]:
                 neib_shape = T.as_tensor_variable(neib_shape)
-            f = function([], images2neibs(images, neib_shape), mode=self.mode)
+                f = function([], images2neibs(images, neib_shape),
+                             mode=self.mode)
                 self.assertRaises(TypeError, f)
 
                 #Test that ignore border work in that case.
-            f = function([], images2neibs(images, neib_shape, mode='ignore_borders'),
+                f = function([],
+                             images2neibs(images, neib_shape,
+                                          mode='ignore_borders'),
                              mode=self.mode)
                 f()
 
@@ -197,8 +218,10 @@ class T_Images2Neibs(unittest_tools.InferShapeTester):
             [(1, 1, 1045, 5), (3, 3), (3, 3), None],
             ]):
 
+            for dtype in self.dtypes:
+
                 images = shared(numpy.asarray(numpy.arange(numpy.prod(
-                            shape)).reshape(shape), dtype='float32'))
+                                shape)).reshape(shape), dtype=dtype))
                 neib_shape = T.as_tensor_variable(neib_shape)
                 neib_step = T.as_tensor_variable(neib_step)
                 expected = numpy.asarray(expected)
@@ -210,9 +233,10 @@ class T_Images2Neibs(unittest_tools.InferShapeTester):
 
                 if expected.size > 1:
                     for i in range(shape[0] * shape[1]):
-                    assert numpy.allclose(neibs[i * expected.shape[0]:
+                        assert numpy.allclose(
+                                neibs[i * expected.shape[0]:
                                       (i + 1) * expected.shape[0], :],
-                                          expected + 25 * i), mode_idx
+                                expected + 25 * i), "wrap_centered"
 
                 assert self.op in [type(node.op)
                                    for node in f.maker.fgraph.toposort()]
@@ -223,7 +247,11 @@ class T_Images2Neibs(unittest_tools.InferShapeTester):
 
     def test_neibs_bad_shape_wrap_centered(self):
         shape = (2, 3, 10, 10)
-        images = shared(numpy.arange(numpy.prod(shape)).reshape(shape))
+
+        for dtype in self.dtypes:
+            images = shared(numpy.arange(
+                numpy.prod(shape), dtype=dtype
+                ).reshape(shape))
 
             for neib_shape in [(3, 2), (2, 3)]:
                 neib_shape = T.as_tensor_variable(neib_shape)

theano/tests/unittest_tools.py

@@ -166,9 +166,11 @@ class T_OpContractMixin(object):
 class InferShapeTester(unittest.TestCase):
     def setUp(self):
         seed_rng()
+        # Take into account any mode that may be defined in a child class
+        mode = getattr(self, 'mode', theano.compile.get_default_mode())
         # This mode seems to be the minimal one including the shape_i
         # optimizations, if we don't want to enumerate them explicitly.
-        self.mode = theano.compile.get_default_mode().including("canonicalize")
+        self.mode = mode.including("canonicalize")
 
     def _compile_and_check(self, inputs, outputs, numeric_inputs, cls,
                            excluding=None):