若API未标明支持情况，则代表该API的支持情况待验证。
API名称 |
是否支持 |
限制与说明 |
|---|---|---|
torch.distributed.GradBucket |
||
torch.distributed.GradBucket.index |
||
torch.distributed.GradBucket.buffer |
||
torch.distributed.GradBucket.gradients |
||
torch.distributed.GradBucket.is_last |
||
torch.distributed.GradBucket.set_buffer |
||
torch.distributed.GradBucket.parameters |
||
torch.distributed.algorithms.ddp_comm_hooks.default_hooks.allreduce_hook |
否 |
|
torch.distributed.algorithms.ddp_comm_hooks.default_hooks.fp16_compress_hook |
否 |
|
torch.distributed.algorithms.ddp_comm_hooks.default_hooks.bf16_compress_hook |
否 |
|
torch.distributed.algorithms.ddp_comm_hooks.default_hooks.fp16_compress_wrapper |
否 |
|
torch.distributed.algorithms.ddp_comm_hooks.default_hooks.bf16_compress_wrapper |
否 |
|
torch.distributed.algorithms.ddp_comm_hooks.powerSGD_hook.PowerSGDState |
否 |
|
torch.distributed.algorithms.ddp_comm_hooks.powerSGD_hook.powerSGD_hook |
否 |
|
torch.distributed.algorithms.ddp_comm_hooks.powerSGD_hook.batched_powerSGD_hook |
否 |
|
torch.distributed.algorithms.ddp_comm_hooks.debugging_hooks.noop_hook |
||
torch.distributed.algorithms.ddp_comm_hooks.powerSGD_hook.PowerSGDState |
否 |