-
### 🐛 Describe the bug
When running `torch.nn.InstanceNorm2d` with a `float64` tensor and `track_running_stats=True`, an error is thrown on CPU:
`RuntimeError: mixed dtype (CPU): all inputs must sha…
-
### 🐛 Describe the bug
Trying to average pool over integers work on CPU but fails on CUDA [colab](https://colab.research.google.com/drive/1UufNjNlX6dwPFH0D_ooBKURVPhERziHw?usp=sharing)
Minimal rep…
-
Say the user wants to retry a previously failed function call...
We'll need to modify the optree to insert a GOTO+jump-target right before the currently active line.
http://search.cpan.org/~rurban/B…
-
Devel::GoFaster provide a prototype to optimize the optree for some common patterns
```
//= 1
```
@arc was also suggesting in https://github.com/p5h/p5summit-2019/issues/18
to consider
```
E…
-
### 🐛 Describe the bug
In deepspeed workloads, Zero parameter offload is implemented via module hooks. We find under torch.compile scenrio, if there any graph breaks happen in the pre/post hook of a …
-
Hello,
Per https://github.com/Perl/perl5/issues/20384, this module is currently broken on development perls.
At issue are two tests: `t/20-optree-deparse.t` and `t/21-deparse-trace-anon-sub.t`. …
-
### 🐛 Describe the bug
`torch.dot` can be calculated with tensors of `int64` dtype on CPU, but shows as not implemented for GPU ([colab](https://colab.research.google.com/drive/1oFgSRK7T6alyBJgHbWahL…
-
I'm getting this message at the end of every run using hdb.
Here's a trivial reproduction. Same on 5.10.1 and 5.36.0.
```
$ perl -d:hdb -e1
Debugger pid 22503 listening on http://127.0.0.1:808…
-
### 🐛 Describe the bug
When running the command make triton , the following error is generated,
`export USE_XPU=1
make triton`
Building wheels for collected packages: triton
Building wheel …
-
### 🐛 Describe the bug
Under certain circumstances `nn.Linear` will have memory leaks on MPS. The exact failure mode and condition that leads to leakage is unclear at this moment. I'll give an upda…