frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-21 05:34:18 +08:00
Code Issues Packages Projects Releases Wiki Activity

Cleanup arg{min, max} (#17103)

Browse Source 
Summary:
Why do we need this workaround? `PythonArgParser` handles these two cases well.

The discussion started at https://github.com/pytorch/pytorch/pull/6201#issuecomment-378724406. The conclusion at that time by goldsborough was:

> Because we wanted to allow `dim=None` in Python and route to a different function. Essentially the problem was wanting to wrap the C++ function in Python. AFAIK there is no way of translating `dim=None` behavior into C++? So Richard and I came up with this strategy

Maybe at that time `PythonArgParser` was not powerful enough to handle the routing of two function with same name but different C++ signature.

Will keep an eye on the CI.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/17103

Differential Revision: D14523503

Pulled By: VitalyFedyunin

fbshipit-source-id: cae3e2678062da2eccd93b51d4050578c7a9ab80
This commit is contained in:
Gao, Xiang
2019-03-20 16:18:49 -07:00
committed by Facebook Github Bot
parent ebc9f75895
commit 7e6220393f
13 changed files with 156 additions and 140 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
torch/_tensor_docs.py
View File

@ -1487,6 +1487,13 @@ max(dim=None, keepdim=False) -> Tensor or (Tensor, Tensor)
See :func:`torch.max`
""")
add_docstr_all('argmax',
r"""
argmax(dim=None, keepdim=False) -> LongTensor
See :func:`torch.argmax`
""")
add_docstr_all('mean',
r"""
mean(dim=None, keepdim=False) -> Tensor or (Tensor, Tensor)
@ -1508,6 +1515,13 @@ min(dim=None, keepdim=False) -> Tensor or (Tensor, Tensor)
See :func:`torch.min`
""")
add_docstr_all('argmin',
r"""
argmin(dim=None, keepdim=False) -> LongTensor
See :func:`torch.argmin`
""")
add_docstr_all('mm',
r"""
mm(mat2) -> Tensor