text2text模型不支持多卡推理
Open
- Dominant language
- Python
- Stars
- 3.1k
- Forks
- 520
- PR merge metrics
- No merged PRs in 30d
Description
多卡模式下，调用model.encoder和model.decoder时报错，如下：
Traceback (most recent call last):
File "finetune/run_text2text.py", line 292, in
main()
File "finetune/run_text2text.py", line 276, in main
result = evaluate(args, read_dataset(args, args.dev_path))
File "finetune/run_text2text.py", line 149, in evaluate
memory_bank = args.model.encode(src_batch, seg_batch)
File "/home/yirenchen/miniconda3/envs/py36/lib/python3.6/site-packages/torch/nn/modules/module.py", line 948, in getattr
type(self).name, name))
AttributeError: 'DataParallel' object has no attribute 'encode'
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.