new scan_permutation op

theano/sandbox/scan_module/scan_utils.py

@@ -340,3 +340,53 @@ def allocate_memory(T, y_info, y):
             ins_shapes.append(in_shape)
         shape = infer_shape([y], inputs, ins_shapes)[0]
         return tensor.zeros([T, ] + shape, dtype=y.dtype)
+
+
+class ScanPermutation(gof.Op):
+    def __init__(self, inplace=False):
+        self.inplace = inplace
+        if inplace:
+            self.destroy_map = {0: [0]}
+
+    def __eq__(self, other):
+        return type(self) == type(other) and self.inplace == other.inplace
+
+    def __hash__(self):
+        return hash(type(self)) ^ hash(self.inplace)
+
+    def __str__(self):
+        if self.inplace:
+            return "scan_permutation{inplace}"
+        else:
+            return "scan_permutation"
+
+    def make_node(self, membuffer, index):
+        # index has to be a scalar
+        assert index.ndim = 0
+        # we neeed at least one dimension
+        assert membuffer.ndim > 0
+        return gof.Apply(self, [membuffer, index], [membuffer.type()])
+
+    def perform(self, node, inputs, outputs):
+        membuffer = inputs[0]
+        index = inputs[0]
+        if index <= membuffer.shape[0] or index % membuffer.shape[0] == 0:
+            if self.inplace:
+                outputs[0] = membuffer
+            else:
+                outputs[0][:] = membuffer
+        else:
+            pos = index % membuffer.shape[0]
+            if outputs[0] is membuffer:
+                membuffer = membuffer.copy()
+            outputs[0][:membuffer.shape[0] - p] = membuffer[p:]
+            outputs[0][membuffer.shape[0] - p:] = membuffer[:p]
+
+    def R_op(self, inputs, eval_points):
+        if eval_points[0] is None:
+            return [None]
+        return self.make_node(eval_points[0], inputs[1]).outputs
+
+    def grad(self, inputs, grads):
+        pos = inputs[0].shape[0] - (inputs[1] % inputs[0].shape[0])
+        return self.make_node(grads[0], pos).outputs