提交 f3b19ac8 authored 作者: Razvan Pascanu's avatar Razvan Pascanu

more info

上级 b9d87caa
......@@ -295,6 +295,8 @@ class PushOutNonSeqScan(gof.Optimizer):
*shape)
# We need to add one extra dimension to the outputs
# because the scan op expects for a tensor3, to which an
# subtensor is apllied that takes only the last element
if replace_with:
fgraph.replace_all_validate_remove(
replace_with.items(),
......
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