frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-11-03 23:45:05 +08:00
Code Issues Packages Projects Releases Wiki Activity

Make integer parameters and buffers immune to float(), double() and half() (#3820)

Browse Source 
* Avoid casting integer params and buffers to float(), double() and half()

* Add test for immune integer buffers

* Fix documentation for float(), double() and half()

* Fix test
This commit is contained in:
Luca Antiga
2017-11-23 00:34:53 +01:00
committed by Soumith Chintala
parent b5ad8c8d16
commit af58bfbb1b
7 changed files with 41 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
torch/cuda/sparse.py
View File

@ -86,3 +86,9 @@ _sparse_tensor_classes.add(CharTensor)
_sparse_tensor_classes.add(ByteTensor)
_sparse_tensor_classes.add(HalfTensor)
torch._tensor_classes.update(_sparse_tensor_classes)
torch._integer_tensor_classes.add(LongTensor)
torch._integer_tensor_classes.add(IntTensor)
torch._integer_tensor_classes.add(ShortTensor)
torch._integer_tensor_classes.add(CharTensor)
torch._integer_tensor_classes.add(ByteTensor)