-
### š Describe the bug
# Description
I tried to export a PyTorch model using `torch.nn.GroupNorm` to Core ML with ExecuTorch, but encountered a `ValueError: Unsupported fx node aten_native_group_norā¦
-
### š Describe the bug
When running a simple model including torch.nn.LayerNorm using deepspeed zero3 with torch.compile and [compiled_autograd](https://github.com/pytorch/tutorials/blob/main/intermeā¦
-
### What happened?
`GGML_CUDA_ENABLE_UNIFIED_MEMORY` is documented as automatically swapping out VRAM under pressure automatically, letting you run any model as long as it fits within available RAMā¦
-
### Describe the bug
The Graph/RecordReplay/usm_fill.cpp test has been observed to timeout in CUDA CI for unrelated changes. For example, see https://github.com/intel/llvm/pull/14985.
```
TIMEOUTā¦
-
I am having a lot of trouble with speaker diarization across lots of different platforms and models.
```toml
[target.'cfg(any(windows, target_os = "linux"))'.dependencies]
sherpa-rs = { version =ā¦
-
### Your current environment
The output of `python collect_env.py`
```text
# For security purposes, please feel free to check the contents of collect_env.py before running it.
python collect_eā¦
-
### Checklist
- [X] 1. I have searched related issues but cannot get the expected help.
- [X] 2. The bug has not been fixed in the latest version.
- [X] 3. Please note that if the bug-related issā¦
-
The test `WaitAnyHostAndDeviceSemaphoresAndDeviceSignals` from `cuda_graph_semaphore_submission_test` seems to sometimes fail
with
```
12/412 Test #66: iree/hal/drivers/cuda/cts/cuda_graph_semaā¦
-
### š Describe the bug
As mentioned in this [blog](https://dev-discuss.pytorch.org/t/higher-order-operators-2023-10/1565), HigherOrderOperator does not support graph break inside the input/output funā¦
-
### Name and Version
```
.\llama-cli.exe --version
ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 2 CUDA devices:
Device 0: NVIDIAā¦