Introduce ManagedDeviceMesh to integrate DeviceMesh with TorchFT #175
Annotations
3 errors
unittest (linux.2xlarge, cpu) / linux-job
Process completed with exit code 1.
|
unittest (linux.4xlarge.nvidia.gpu, cuda, 12.1) / linux-job
FailFast: cancelling since parallel instance has failed
|
unittest (linux.4xlarge.nvidia.gpu, cuda, 12.1) / linux-job
The operation was canceled.
|