We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent 6b69723 commit 4794230Copy full SHA for 4794230
torch/csrc/README.md
@@ -30,6 +30,11 @@ One important case where this assumption is important is when tracking
30
the CUDA device a tensor is stored in: this information is stored
31
solely in the storage, so if a storage is NULL, we lose this information.
32
33
+Although storage is never NULL, the data field of THStorage may be NULL. This
34
+mostly occurs when we want to pre-allocate an output tensor struct, but then
35
+have it be resized and filled with data by some operator: there's no point in
36
+allocating data for it in this case!
37
+
38
## Files
39
40
### `Exceptions.h`
0 commit comments