-
Notifications
You must be signed in to change notification settings - Fork 5.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[PHI]Standardise some C++ API #47385
Conversation
你的PR提交成功,感谢你对开源项目的贡献! |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
@@ -2056,7 +2056,7 @@ | |||
backward : psroi_pool_grad | |||
|
|||
- op : put_along_axis | |||
args : (Tensor arr, Tensor index, Tensor value, int axis, str reduce) | |||
args : (Tensor arr, Tensor indices, Tensor values, int axis, str reduce) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
反向中的参数名也需要调整下吧?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done,3q very much
paddle/phi/api/yaml/ops.yaml
Outdated
@@ -145,15 +145,15 @@ | |||
func : fft_r2c | |||
backward : fft_r2c_grad | |||
|
|||
- op : graph_send_uv | |||
- op : send_uv |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
这里生成的静态图 op 名也会从 graph_send_uv
变为了 send_uv
, 得考虑是否有不兼容的问题
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done , 3q your revew
paddle/phi/infermeta/multiary.h
Outdated
void SendUERecvInferMeta(const MetaTensor& x, | ||
const MetaTensor& y, | ||
const MetaTensor& src_index, | ||
const MetaTensor& dst_index, | ||
const std::string& message_op, | ||
const std::string& reduce_op, | ||
const IntArray& out_size, | ||
MetaTensor* out, | ||
MetaTensor* dst_count); | ||
|
||
void SendUVInferMeta(const MetaTensor& x, | ||
const MetaTensor& y, | ||
const MetaTensor& src_index, | ||
const MetaTensor& dst_index, | ||
const std::string& message_op, | ||
MetaTensor* out); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
infermeta函数按字典序再调整下
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done
const IntArray& out_size, | ||
MetaTensor* out, | ||
MetaTensor* dst_count); | ||
void SendURecvInferMeta(const MetaTensor& x, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
同上
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done
const IntArray& out_size, | ||
MetaTensor* out, | ||
MetaTensor* dst_count) { | ||
void SendURecvInferMeta(const MetaTensor& x, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
这个函数位置好像没调整
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
下个pr修复一下
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM for CI-OP-Benchmark
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
PR types
Others
PR changes
Others
Describe
标准化如下C++ API接口,使其与Python API对齐:
addmm
put_along_axis
graph_send_uv
brelu
crop_tensor
gaussian_random
graph_send_recv
graph_send_ue_recv
后续还会继续对接口标准化,此PR是系列工作的第一个PR