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
OPTHREE
Paddle
提交
fee42441
P
Paddle
项目概览
OPTHREE
/
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
体验新版 GitCode,发现更多精彩内容 >>
未验证
提交
fee42441
编写于
1月 11, 2021
作者:
W
wawltor
提交者:
GitHub
1月 11, 2021
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
just add the op error message for the matmul xpu (#30246)
add the op error message for the matmul xpu
上级
6bfdef72
变更
1
隐藏空白更改
内联
并排
Showing
1 changed file
with
26 addition
and
10 deletion
+26
-10
paddle/fluid/operators/matmul_op_xpu.cc
paddle/fluid/operators/matmul_op_xpu.cc
+26
-10
未找到文件。
paddle/fluid/operators/matmul_op_xpu.cc
浏览文件 @
fee42441
...
...
@@ -127,10 +127,18 @@ class MatMulXPUKernel : public framework::OpKernel<T> {
PADDLE_ENFORCE_EQ
(
mat_dim_a
.
width_
,
mat_dim_b
.
height_
,
platform
::
errors
::
InvalidArgument
(
"Shape mistake in matmul_op"
));
PADDLE_ENFORCE_EQ
(
mat_dim_a
.
batch_size_
,
mat_dim_b
.
batch_size_
,
platform
::
errors
::
InvalidArgument
(
"Shape mistake in matmul_op"
));
platform
::
errors
::
InvalidArgument
(
"Shape mistake in matmul_op, the "
"first tensor width must be same as "
"second tensor height, but received "
"width:%d, height:%d"
,
mat_dim_a
.
width_
,
mat_dim_b
.
height_
));
PADDLE_ENFORCE_EQ
(
mat_dim_a
.
batch_size_
,
mat_dim_b
.
batch_size_
,
platform
::
errors
::
InvalidArgument
(
"Shape mistake in matmul_op, the two input"
"tensor batch_size must be same, but received first "
"tensor batch_size:%d, second "
"tensor batch_size:%d"
,
mat_dim_a
.
batch_size_
,
mat_dim_b
.
batch_size_
));
T
alpha
=
static_cast
<
T
>
(
context
.
Attr
<
float
>
(
"alpha"
));
auto
&
dev_ctx
=
context
.
template
device_context
<
DeviceContext
>();
...
...
@@ -251,12 +259,20 @@ class MatMulGradXPUKernel : public framework::OpKernel<T> {
}
}
PADDLE_ENFORCE_EQ
(
mat_dim_a
.
width_
,
mat_dim_b
.
height_
,
platform
::
errors
::
InvalidArgument
(
"Shape mistake in matmul_grad_op"
));
PADDLE_ENFORCE_EQ
(
mat_dim_a
.
batch_size_
,
mat_dim_b
.
batch_size_
,
platform
::
errors
::
InvalidArgument
(
"Shape mistake in matmul_grad_op"
));
PADDLE_ENFORCE_EQ
(
mat_dim_a
.
width_
,
mat_dim_b
.
height_
,
platform
::
errors
::
InvalidArgument
(
"Shape mistake in matmul_grad_op, the "
"first tensor width must be same as second tensor "
"height, but received "
"width:%d, height:%d"
,
mat_dim_a
.
width_
,
mat_dim_b
.
height_
));
PADDLE_ENFORCE_EQ
(
mat_dim_a
.
batch_size_
,
mat_dim_b
.
batch_size_
,
platform
::
errors
::
InvalidArgument
(
"Shape mistake in matmul_grad_op, the two input"
"tensor batch_size must be same, but received first "
"tensor batch_size:%d, second "
"tensor batch_size:%d"
,
mat_dim_a
.
batch_size_
,
mat_dim_b
.
batch_size_
));
T
alpha
=
static_cast
<
T
>
(
context
.
Attr
<
float
>
(
"alpha"
));
...
...
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