Don't put scalar in shared var on the GPU by default.

theano/gpuarray/opt.py

@@ -29,7 +29,7 @@ from theano.tensor.nnet.abstract_conv import (AbstractConv2d,
 from theano.tests.breakpoint import PdbBreakpoint
 
 from .type import (GpuArrayType, GpuArrayConstant, get_context,
-                   ContextNotDefined)
+                   ContextNotDefined, move_to_gpu)
 from .basic_ops import (as_gpuarray_variable, infer_context_name,
                         host_from_gpu, GpuToGpu,
                         HostFromGpu, GpuFromHost,
@@ -242,9 +242,8 @@ class InputToGpuOptimizer(Optimizer):
             target = getattr(input.tag, 'target', None)
             if target == 'cpu':
                 continue
-            # Do not move *int* scalar to the GPU.
             if (isinstance(input.type, tensor.TensorType) and
-                    input.ndim == 0 and 'int' in input.dtype):
+                    not move_to_gpu(input)):
                 continue
 
             try:
@@ -301,9 +300,7 @@ class GraphToGPU(Optimizer):
         target = infer_context_name(*fgraph.inputs)
         for i in fgraph.inputs:
             # Do not move *int* scalar to the GPU.
-            if (isinstance(i.type, tensor.TensorType) and
-                    (i.ndim > 0 or 'int' not in i.dtype) and
-                    "complex" not in i.dtype):
+            if isinstance(i.type, tensor.TensorType) and move_to_gpu(i):
                 mapping[i] = i.transfer(getattr(i.tag, 'target', target))
             else:
                 mapping[i] = i

theano/gpuarray/type.py

@@ -22,6 +22,26 @@ except ImportError:
 _context_reg = {}
 
 
+def move_to_gpu(data):
+    """
+    Do we want to move this computation to the GPU?
+
+    Currently, we don't move complex and scalar int.
+
+    Parameters
+    ----------
+    data : numpy.ndarray or TensorVariable
+           (it must have dtype and ndim parameter)
+    """
+    # We don't support complex on the GPU
+    if str(data.dtype) in tensor.basic.complex_dtypes:
+        return False
+    # We don't want scalar int on the GPU.
+    if data.ndim == 0 and str(data.dtype) in tensor.basic.discrete_dtypes:
+        return False
+    return True
+
+
 class ContextNotDefined(ValueError):
     pass
 
@@ -572,6 +592,8 @@ def gpuarray_shared_constructor(value, name=None, strict=False,
     See :func:`theano.shared`.
 
     """
+    if target is None and not move_to_gpu(value):
+        raise TypeError('We do not move that data by deault to the GPU')
     if target == 'gpu' or target == 'cpu':
         raise TypeError('not for me')