GCN-D MMdistributedDataparallel train
joelulu opened this issue · comments
你好，我想用MMdistributedDataparallel训练GCN-D
更改了--distributed 为True
报错：
Default process group has not been initialized, please make sure to call init_process_group
the same question, the codes seems not supportable for distributed gpus
Yes. It currently does not support distributed training. We welcome any related PR!