提交 cb476fd2 authored 作者: Frederic Bastien's avatar Frederic Bastien

small fix, use the type may_share_memory fct and not the numpy one.

上级 3ed7d6da
......@@ -522,8 +522,8 @@ def _check_inputs(node, storage_map, r_vals, dr_vals, active_nodes, clobber_dr_v
if warn_input_not_reused:
vmap=getattr(node.op,'view_map',{})
for oo,ii in vmap.iteritems():
if hasattr(node.outputs[00].type,"may_share_memory"):
if not numpy.may_share_memory(storage_map[node.outputs[oo]][0],storage_map[node.inputs[ii[0]]][0]):
if hasattr(node.outputs[0].type,"may_share_memory"):
if not node.outputs[0].type.may_share_memory(storage_map[node.outputs[oo]][0],storage_map[node.inputs[ii[0]]][0]):
#when a subtensor return a tensor ofndim==0, numpy seam to return a copy.
#when have an empty ndarray(happen with output guard) it is not the same. why?
if storage_map[node.outputs[oo]][0].ndim>0 and storage_map[node.outputs[oo]][0].size>0:
......
Markdown 格式
0%
您添加了 0 到此讨论。请谨慎行事。
请先完成此评论的编辑!
注册 或者 后发表评论