single Gpu report Bug

[fanabcd]

Traceback (most recent call last):
File "train.py", line 184, in
main()
File "train.py", line 180, in main
train(args, cfg)
File "train.py", line 166, in train
trainer.train(trn_loader, st_step, cfg.max_iter)
File "/202102014/Fy/mxfont/trainer/fact_trainer.py", line 145, in train
self.add_ac_losses_and_update_stats(
File "/202102014/Fy/mxfont/trainer/fact_trainer.py", line 281, in add_ac_losses_and_update_stats
ac_loss_c, cross_ac_loss_c, acc_c = self.infer_comp_ac(char_facts, comp_ids)
File "/202102014/Fy/mxfont/trainer/fact_trainer.py", line 251, in infer_comp_ac
acc = T_probs[cids, eids].sum() / n_experts
RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu)

[fany4311]

same issue