Skip to content
体验新版
项目
组织
正在加载...
登录
切换导航
打开侧边栏
Crayon鑫
Paddle
提交
9e29d3eb
P
Paddle
项目概览
Crayon鑫
/
Paddle
与 Fork 源项目一致
Fork自
PaddlePaddle / Paddle
通知
1
Star
1
Fork
0
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
1
列表
看板
标记
里程碑
合并请求
0
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
P
Paddle
项目概览
项目概览
详情
发布
仓库
仓库
文件
提交
分支
标签
贡献者
分支图
比较
Issue
1
Issue
1
列表
看板
标记
里程碑
合并请求
0
合并请求
0
Pages
分析
分析
仓库分析
DevOps
Wiki
0
Wiki
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
创建新Issue
提交
Issue看板
未验证
提交
9e29d3eb
编写于
2月 11, 2020
作者:
H
huzhiqiang
提交者:
GitHub
2月 11, 2020
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
【OpPorting Example】DEMO OF FIX COMPILE&RUNTIME LOD_EQUALITY (#22460)
上级
00c110f3
变更
3
隐藏空白更改
内联
并排
Showing
3 changed file
with
33 addition
and
7 deletion
+33
-7
paddle/fluid/operators/sequence_ops/sequence_erase_op.cc
paddle/fluid/operators/sequence_ops/sequence_erase_op.cc
+9
-0
paddle/fluid/operators/sequence_ops/sequence_reshape_op.cc
paddle/fluid/operators/sequence_ops/sequence_reshape_op.cc
+3
-0
python/paddle/fluid/tests/unittests/white_list/compile_vs_runtime_white_list.py
...sts/unittests/white_list/compile_vs_runtime_white_list.py
+21
-7
未找到文件。
paddle/fluid/operators/sequence_ops/sequence_erase_op.cc
浏览文件 @
9e29d3eb
...
...
@@ -32,6 +32,15 @@ class SequenceEraseOp : public framework::OperatorWithKernel {
"Input(X) of SequenceEraseOp should be a 2-D LoDTensor "
"with the 2nd dimension equal to 1."
);
ctx
->
SetOutputDim
(
"Out"
,
x_dims
);
// The output LoDTensor's lod_level should be input X's lod_level.
// For compile-time, we call SetLoDLevel to set output's lod_level.
// For runtime, output LoDTensor's lod is determined by input X's lod and
// the level specified by input RandTable.
// We cannot get X's detail lod and RankTable's level in this function, so
// leave this work to the detail kernel implementation.
if
(
!
ctx
->
IsRuntime
())
{
ctx
->
SetLoDLevel
(
"Out"
,
ctx
->
GetLoDLevel
(
"X"
));
}
}
};
...
...
paddle/fluid/operators/sequence_ops/sequence_reshape_op.cc
浏览文件 @
9e29d3eb
...
...
@@ -37,6 +37,9 @@ class SequenceReshapeOp : public framework::OperatorWithKernel {
}
else
{
// when compiling, the batch size is undetermined, just set to -1
ctx
->
SetOutputDim
(
"Out"
,
{
-
1
,
static_cast
<
int64_t
>
(
new_dim
)});
// when compiling, the LodLevel of Out is set to be 1, which is consistent
// with that in running time.
ctx
->
SetLoDLevel
(
"Out"
,
1
);
}
}
};
...
...
python/paddle/fluid/tests/unittests/white_list/compile_vs_runtime_white_list.py
浏览文件 @
9e29d3eb
...
...
@@ -19,11 +19,25 @@
# reasons for skipping compile_vs_runtime test or be fixed later.
COMPILE_RUN_OP_WHITE_LIST
=
[
'lod_reset'
,
'sequence_pool'
,
'sequence_slice'
,
'generate_mask_labels'
,
'sequence_reshape'
,
'generate_proposals'
,
'mine_hard_examples'
,
'retinanet_detection_output'
,
'ctc_align'
,
'fusion_seqpool_cvm_concat'
,
'gru'
,
'sequence_erase'
,
'rpn_target_assign'
,
'retinanet_target_assign'
,
'filter_by_instag'
,
'multiclass_nms'
,
'multiclass_nms2'
,
'im2sequence'
,
'generate_proposal_labels'
,
'distribute_fpn_proposals'
,
'detection_map'
,
'locality_aware_nms'
,
'var_conv_2d'
'lod_reset'
,
\
'sequence_pool'
,
\
'sequence_slice'
,
\
'generate_mask_labels'
,
\
'generate_proposals'
,
\
'mine_hard_examples'
,
\
'retinanet_detection_output'
,
\
'ctc_align'
,
\
'fusion_seqpool_cvm_concat'
,
\
'gru'
,
\
'rpn_target_assign'
,
\
'retinanet_target_assign'
,
\
'filter_by_instag'
,
\
'multiclass_nms'
,
\
'multiclass_nms2'
,
\
'im2sequence'
,
\
'generate_proposal_labels'
,
\
'distribute_fpn_proposals'
,
\
'detection_map'
,
\
'locality_aware_nms'
,
\
'var_conv_2d'
]
编辑
预览
Markdown
is supported
0%
请重试
或
添加新附件
.
添加附件
取消
You are about to add
0
people
to the discussion. Proceed with caution.
先完成此消息的编辑!
取消
想要评论请
注册
或
登录