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
PaddlePaddle
Paddle
提交
b4a1f43f
P
Paddle
项目概览
PaddlePaddle
/
Paddle
大约 1 年 前同步成功
通知
2298
Star
20931
Fork
5422
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
1423
列表
看板
标记
里程碑
合并请求
543
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
1,423
Issue
1,423
列表
看板
标记
里程碑
合并请求
543
合并请求
543
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
b4a1f43f
编写于
10月 17, 2022
作者:
W
Weilong Wu
提交者:
GitHub
10月 17, 2022
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
[Eager] use CastPyArg2Double to parse python float obj (#47029)
上级
73196e5a
变更
1
隐藏空白更改
内联
并排
Showing
1 changed file
with
72 addition
and
72 deletion
+72
-72
paddle/fluid/pybind/eager_math_op_patch.cc
paddle/fluid/pybind/eager_math_op_patch.cc
+72
-72
未找到文件。
paddle/fluid/pybind/eager_math_op_patch.cc
浏览文件 @
b4a1f43f
...
...
@@ -129,7 +129,7 @@ void SetDevice(paddle::platform::Place place) {
// this function will update gradually.
paddle
::
experimental
::
Tensor
CallScalarFuction
(
const
paddle
::
experimental
::
Tensor
&
self_tensor
,
float
other
,
double
other
,
std
::
string
op_type
)
{
paddle
::
experimental
::
Tensor
ret
;
if
(
op_type
==
"add"
||
op_type
==
"radd"
)
{
...
...
@@ -169,16 +169,16 @@ static PyObject* tensor__add__method(TensorObject* self,
// 1. scalar exists cases
if
(
PyFloat_Check
(
other_obj
)
||
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
float
other
=
0.0
;
double
other
=
0.0
;
if
(
PyFloat_Check
(
other_obj
))
{
other
=
CastPyArg2
AttrFloat
(
other_obj
,
0
);
other
=
CastPyArg2
Double
(
other_obj
,
"__add__"
,
0
);
if
(
_supported_int_dtype_
.
find
(
self_tensor
.
dtype
())
!=
_supported_int_dtype_
.
end
())
{
eager_gil_scoped_release
guard
;
self_tensor
=
cast_ad_func
(
self_tensor
,
DataType
::
FLOAT32
);
}
}
else
if
(
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
other
=
CastPyArg2
AttrFloat
(
other_obj
,
0
);
other
=
CastPyArg2
Double
(
other_obj
,
"__add__"
,
0
);
}
{
...
...
@@ -267,16 +267,16 @@ static PyObject* tensor__sub__method(TensorObject* self,
// 1. scalar exists cases
if
(
PyFloat_Check
(
other_obj
)
||
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
float
other
=
0.0
;
double
other
=
0.0
;
if
(
PyFloat_Check
(
other_obj
))
{
other
=
CastPyArg2
AttrFloat
(
other_obj
,
0
);
other
=
CastPyArg2
Double
(
other_obj
,
"__sub__"
,
0
);
if
(
_supported_int_dtype_
.
find
(
self_tensor
.
dtype
())
!=
_supported_int_dtype_
.
end
())
{
eager_gil_scoped_release
guard
;
self_tensor
=
cast_ad_func
(
self_tensor
,
DataType
::
FLOAT32
);
}
}
else
if
(
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
other
=
CastPyArg2
AttrFloat
(
other_obj
,
0
);
other
=
CastPyArg2
Double
(
other_obj
,
"__sub__"
,
0
);
}
{
eager_gil_scoped_release
guard
;
...
...
@@ -360,16 +360,16 @@ static PyObject* tensor__rsub__method(TensorObject* self,
// 1. scalar exists cases
if
(
PyFloat_Check
(
other_obj
)
||
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
float
other
=
0.0
;
double
other
=
0.0
;
if
(
PyFloat_Check
(
other_obj
))
{
other
=
CastPyArg2
AttrFloat
(
other_obj
,
0
);
other
=
CastPyArg2
Double
(
other_obj
,
"__rsub__"
,
0
);
if
(
_supported_int_dtype_
.
find
(
self_tensor
.
dtype
())
!=
_supported_int_dtype_
.
end
())
{
eager_gil_scoped_release
guard
;
self_tensor
=
cast_ad_func
(
self_tensor
,
DataType
::
FLOAT32
);
}
}
else
if
(
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
other
=
CastPyArg2
AttrFloat
(
other_obj
,
0
);
other
=
CastPyArg2
Double
(
other_obj
,
"__rsub__"
,
0
);
}
{
eager_gil_scoped_release
guard
;
...
...
@@ -455,16 +455,16 @@ static PyObject* tensor__mul__method(TensorObject* self,
// 1. scalar exists cases
if
(
PyFloat_Check
(
other_obj
)
||
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
float
other
=
0.0
;
double
other
=
0.0
;
if
(
PyFloat_Check
(
other_obj
))
{
other
=
CastPyArg2
AttrFloat
(
other_obj
,
0
);
other
=
CastPyArg2
Double
(
other_obj
,
"__mul__"
,
0
);
if
(
_supported_int_dtype_
.
find
(
self_tensor
.
dtype
())
!=
_supported_int_dtype_
.
end
())
{
eager_gil_scoped_release
guard
;
self_tensor
=
cast_ad_func
(
self_tensor
,
DataType
::
FLOAT32
);
}
}
else
if
(
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
other
=
CastPyArg2
AttrFloat
(
other_obj
,
0
);
other
=
CastPyArg2
Double
(
other_obj
,
"__mul__"
,
0
);
}
{
eager_gil_scoped_release
guard
;
...
...
@@ -557,11 +557,11 @@ static PyObject* tensor__div__method(TensorObject* self,
// 1. scalar exists cases
if
(
PyFloat_Check
(
other_obj
)
||
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
float
other
=
0.0
;
double
other
=
0.0
;
if
(
PyFloat_Check
(
other_obj
))
{
other
=
CastPyArg2
AttrFloat
(
other_obj
,
0
);
other
=
CastPyArg2
Double
(
other_obj
,
"__div__"
,
0
);
}
else
if
(
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
other
=
CastPyArg2
AttrFloat
(
other_obj
,
0
);
other
=
CastPyArg2
Double
(
other_obj
,
"__div__"
,
0
);
}
if
(
_supported_int_dtype_
.
find
(
self_tensor
.
dtype
())
!=
_supported_int_dtype_
.
end
())
{
...
...
@@ -667,16 +667,16 @@ static PyObject* tensor__rdiv__method(TensorObject* self,
// 1. scalar exists cases
// there is no scalar_div function for __rdiv__ and __rtruediv__
float
other_float
=
0.0
;
bool
has_other_
float
=
false
;
double
other_double
=
0.0
;
bool
has_other_
double
=
false
;
if
(
PyFloat_Check
(
other_obj
)
||
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
if
(
PyFloat_Check
(
other_obj
))
{
other_
float
=
CastPyArg2AttrFloat
(
other_obj
,
0
);
has_other_
float
=
true
;
other_
double
=
CastPyArg2Double
(
other_obj
,
"__rdiv__"
,
0
);
has_other_
double
=
true
;
}
else
if
(
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
other_
float
=
CastPyArg2AttrFloat
(
other_obj
,
0
);
has_other_
float
=
true
;
other_
double
=
CastPyArg2Double
(
other_obj
,
"__rdiv__"
,
0
);
has_other_
double
=
true
;
}
if
(
_supported_int_dtype_
.
find
(
self_tensor
.
dtype
())
!=
_supported_int_dtype_
.
end
())
{
...
...
@@ -687,10 +687,10 @@ static PyObject* tensor__rdiv__method(TensorObject* self,
// 2. create or get tensor for other_obj
paddle
::
experimental
::
Tensor
other_tensor
;
if
(
has_other_
float
)
{
if
(
has_other_
double
)
{
eager_gil_scoped_release
guard
;
other_tensor
=
full_ad_func
(
self_tensor
.
shape
(),
phi
::
Scalar
(
other_
float
),
phi
::
Scalar
(
other_
double
),
self_tensor
.
dtype
(),
place
);
}
else
if
(
!
PyCheckTensor
(
other_obj
))
{
...
...
@@ -781,30 +781,30 @@ static PyObject* tensor__gt__method(TensorObject* self,
// 1. scalar exists cases
// there is no scalar function for __gt__ now
float
other_float
=
0.0
;
bool
has_other_
float
=
false
;
double
other_double
=
0.0
;
bool
has_other_
double
=
false
;
if
(
PyFloat_Check
(
other_obj
)
||
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
if
(
PyFloat_Check
(
other_obj
))
{
other_
float
=
CastPyArg2AttrFloat
(
other_obj
,
0
);
has_other_
float
=
true
;
other_
double
=
CastPyArg2Double
(
other_obj
,
"__gt__"
,
0
);
has_other_
double
=
true
;
if
(
_supported_int_dtype_
.
find
(
self_tensor
.
dtype
())
!=
_supported_int_dtype_
.
end
())
{
eager_gil_scoped_release
guard
;
self_tensor
=
cast_ad_func
(
self_tensor
,
DataType
::
FLOAT32
);
}
}
else
if
(
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
other_
float
=
CastPyArg2AttrFloat
(
other_obj
,
0
);
has_other_
float
=
true
;
other_
double
=
CastPyArg2Double
(
other_obj
,
"__gt__"
,
0
);
has_other_
double
=
true
;
}
}
// 2. create or get tensor for other_obj
paddle
::
experimental
::
Tensor
other_tensor
;
if
(
has_other_
float
)
{
if
(
has_other_
double
)
{
eager_gil_scoped_release
guard
;
other_tensor
=
full_ad_func
(
self_tensor
.
shape
(),
phi
::
Scalar
(
other_
float
),
phi
::
Scalar
(
other_
double
),
self_tensor
.
dtype
(),
place
);
}
else
if
(
!
PyCheckTensor
(
other_obj
))
{
...
...
@@ -867,30 +867,30 @@ static PyObject* tensor__ge__method(TensorObject* self,
// 1. scalar exists cases
// there is no scalar function for __ge__ now
float
other_float
=
0.0
;
bool
has_other_
float
=
false
;
double
other_double
=
0.0
;
bool
has_other_
double
=
false
;
if
(
PyFloat_Check
(
other_obj
)
||
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
if
(
PyFloat_Check
(
other_obj
))
{
other_
float
=
CastPyArg2AttrFloat
(
other_obj
,
0
);
has_other_
float
=
true
;
other_
double
=
CastPyArg2Double
(
other_obj
,
"__ge__"
,
0
);
has_other_
double
=
true
;
if
(
_supported_int_dtype_
.
find
(
self_tensor
.
dtype
())
!=
_supported_int_dtype_
.
end
())
{
eager_gil_scoped_release
guard
;
self_tensor
=
cast_ad_func
(
self_tensor
,
DataType
::
FLOAT32
);
}
}
else
if
(
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
other_
float
=
CastPyArg2AttrFloat
(
other_obj
,
0
);
has_other_
float
=
true
;
other_
double
=
CastPyArg2Double
(
other_obj
,
"__ge__"
,
0
);
has_other_
double
=
true
;
}
}
// 2. create or get tensor for other_obj
paddle
::
experimental
::
Tensor
other_tensor
;
if
(
has_other_
float
)
{
if
(
has_other_
double
)
{
eager_gil_scoped_release
guard
;
other_tensor
=
full_ad_func
(
self_tensor
.
shape
(),
phi
::
Scalar
(
other_
float
),
phi
::
Scalar
(
other_
double
),
self_tensor
.
dtype
(),
place
);
}
else
if
(
!
PyCheckTensor
(
other_obj
))
{
...
...
@@ -954,30 +954,30 @@ static PyObject* tensor__mod__method(TensorObject* self,
// 1. scalar exists cases
// there is no scalar_mod function for __mod__ now
float
other_
float
=
0.0
;
bool
has_other_
float
=
false
;
float
other_
double
=
0.0
;
bool
has_other_
double
=
false
;
if
(
PyFloat_Check
(
other_obj
)
||
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
if
(
PyFloat_Check
(
other_obj
))
{
other_
float
=
CastPyArg2AttrFloat
(
other_obj
,
0
);
has_other_
float
=
true
;
other_
double
=
CastPyArg2Double
(
other_obj
,
"__mod__"
,
0
);
has_other_
double
=
true
;
if
(
_supported_int_dtype_
.
find
(
self_tensor
.
dtype
())
!=
_supported_int_dtype_
.
end
())
{
eager_gil_scoped_release
guard
;
self_tensor
=
cast_ad_func
(
self_tensor
,
DataType
::
FLOAT32
);
}
}
else
if
(
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
other_
float
=
CastPyArg2AttrFloat
(
other_obj
,
0
);
has_other_
float
=
true
;
other_
double
=
CastPyArg2Double
(
other_obj
,
"__mod__"
,
0
);
has_other_
double
=
true
;
}
}
// 2. create or get tensor for other_obj
paddle
::
experimental
::
Tensor
other_tensor
;
if
(
has_other_
float
)
{
if
(
has_other_
double
)
{
eager_gil_scoped_release
guard
;
other_tensor
=
full_ad_func
(
self_tensor
.
shape
(),
phi
::
Scalar
(
other_
float
),
phi
::
Scalar
(
other_
double
),
self_tensor
.
dtype
(),
self_tensor
.
place
());
}
else
if
(
!
PyCheckTensor
(
other_obj
))
{
...
...
@@ -1040,30 +1040,30 @@ static PyObject* tensor__matmul__method(TensorObject* self,
// 1. scalar exists cases
// there is no scalar_matmul function for __matmul__ now
float
other_
float
=
0.0
;
bool
has_other_
float
=
false
;
float
other_
double
=
0.0
;
bool
has_other_
double
=
false
;
if
(
PyFloat_Check
(
other_obj
)
||
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
if
(
PyFloat_Check
(
other_obj
))
{
other_
float
=
CastPyArg2AttrFloat
(
other_obj
,
0
);
has_other_
float
=
true
;
other_
double
=
CastPyArg2Double
(
other_obj
,
"__matmul__"
,
0
);
has_other_
double
=
true
;
if
(
_supported_int_dtype_
.
find
(
self_tensor
.
dtype
())
!=
_supported_int_dtype_
.
end
())
{
eager_gil_scoped_release
guard
;
self_tensor
=
cast_ad_func
(
self_tensor
,
DataType
::
FLOAT32
);
}
}
else
if
(
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
other_
float
=
CastPyArg2AttrFloat
(
other_obj
,
0
);
has_other_
float
=
true
;
other_
double
=
CastPyArg2Double
(
other_obj
,
"__matmul__"
,
0
);
has_other_
double
=
true
;
}
}
// 2. create or get tensor for other_obj
paddle
::
experimental
::
Tensor
other_tensor
;
if
(
has_other_
float
)
{
if
(
has_other_
double
)
{
eager_gil_scoped_release
guard
;
other_tensor
=
full_ad_func
({
1
},
phi
::
Scalar
(
other_
float
),
phi
::
Scalar
(
other_
double
),
self_tensor
.
dtype
(),
self_tensor
.
place
());
}
else
if
(
!
PyCheckTensor
(
other_obj
))
{
...
...
@@ -1144,30 +1144,30 @@ static PyObject* tensor__lt__method(TensorObject* self,
// 1. scalar exists cases
// there is no scalar function for __lt__ now
float
other_
float
=
0.0
;
bool
has_other_
float
=
false
;
float
other_
double
=
0.0
;
bool
has_other_
double
=
false
;
if
(
PyFloat_Check
(
other_obj
)
||
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
if
(
PyFloat_Check
(
other_obj
))
{
other_
float
=
CastPyArg2AttrFloat
(
other_obj
,
0
);
has_other_
float
=
true
;
other_
double
=
CastPyArg2Double
(
other_obj
,
"__lt__"
,
0
);
has_other_
double
=
true
;
if
(
_supported_int_dtype_
.
find
(
self_tensor
.
dtype
())
!=
_supported_int_dtype_
.
end
())
{
eager_gil_scoped_release
guard
;
self_tensor
=
cast_ad_func
(
self_tensor
,
DataType
::
FLOAT32
);
}
}
else
if
(
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
other_
float
=
CastPyArg2AttrFloat
(
other_obj
,
0
);
has_other_
float
=
true
;
other_
double
=
CastPyArg2Double
(
other_obj
,
"__lt__"
,
0
);
has_other_
double
=
true
;
}
}
// 2. create or get tensor for other_obj
paddle
::
experimental
::
Tensor
other_tensor
;
if
(
has_other_
float
)
{
if
(
has_other_
double
)
{
eager_gil_scoped_release
guard
;
other_tensor
=
full_ad_func
(
self_tensor
.
shape
(),
phi
::
Scalar
(
other_
float
),
phi
::
Scalar
(
other_
double
),
self_tensor
.
dtype
(),
self_tensor
.
place
());
}
else
if
(
!
PyCheckTensor
(
other_obj
))
{
...
...
@@ -1230,30 +1230,30 @@ static PyObject* tensor__le__method(TensorObject* self,
// 1. scalar exists cases
// there is no scalar function for __le__ now
float
other_
float
=
0.0
;
bool
has_other_
float
=
false
;
float
other_
double
=
0.0
;
bool
has_other_
double
=
false
;
if
(
PyFloat_Check
(
other_obj
)
||
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
if
(
PyFloat_Check
(
other_obj
))
{
other_
float
=
CastPyArg2AttrFloat
(
other_obj
,
0
);
has_other_
float
=
true
;
other_
double
=
CastPyArg2Double
(
other_obj
,
"__le__"
,
0
);
has_other_
double
=
true
;
if
(
_supported_int_dtype_
.
find
(
self_tensor
.
dtype
())
!=
_supported_int_dtype_
.
end
())
{
eager_gil_scoped_release
guard
;
self_tensor
=
cast_ad_func
(
self_tensor
,
DataType
::
FLOAT32
);
}
}
else
if
(
PyCheckInteger
(
other_obj
)
||
IsNumpyType
(
other_obj
))
{
other_
float
=
CastPyArg2AttrFloat
(
other_obj
,
0
);
has_other_
float
=
true
;
other_
double
=
CastPyArg2Double
(
other_obj
,
"__le__"
,
0
);
has_other_
double
=
true
;
}
}
// 2. create or get tensor for other_obj
paddle
::
experimental
::
Tensor
other_tensor
;
if
(
has_other_
float
)
{
if
(
has_other_
double
)
{
eager_gil_scoped_release
guard
;
other_tensor
=
full_ad_func
(
self_tensor
.
shape
(),
phi
::
Scalar
(
other_
float
),
phi
::
Scalar
(
other_
double
),
self_tensor
.
dtype
(),
self_tensor
.
place
());
}
else
if
(
!
PyCheckTensor
(
other_obj
))
{
...
...
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