omegaconf.errors.ValidationError: Invalid value 'simple',

🐛 Bug

To Reproduce

Steps to reproduce the behavior (always include the command you ran):

Run cmd 'fairseq-generate $binarized_data_path --batch-size 32 --path /data2/liqiang/418M_last_checkpoint.pt --fixed-dictionary model_dict.128k.txt -s de -t fr --remove-bpe 'sentencepiece' --beam 5 --task translation_multi_simple_epoch --lang-pairs language_pairs_small_models.txt --decoder-langtok --encoder-langtok src --gen-subset test '
See error ERROR | fairseq.dataclass.utils | Error when composing. Overrides:[…] … omegaconf.errors.ValidationError: Invalid value 'simple', expected one of [c10d, fully_sharded, legacy_ddp, no_c10d, pytorch_ddp,_mo] full_key: distributed_training.ddp_backend reference_type=DistributedTrainingConfig object_type=DistributedTrainingConfig

Expected behavior

translation from de to fr

Environment

fairseq Version ( 1.0 or master):
PyTorch Version ( 1.0)
OS (Linux):
How you installed fairseq (source):
Build command you used (if compiling from source): pip install --editable ./
Python version: 3.8
CUDA/cuDNN version:
GPU models and configuration:
Any other relevant information:

Additional context

Traceback (most recent call last): File "/data1/home/liqiang/anaconda3/envs/fairseq/lib/python3.8/site-packages/hydra/_internal/config_loader_impl.py", line 513, pply_overrides_to_config OmegaConf.update(cfg, key, value, merge=True) File "/data1/home/liqiang/anaconda3/envs/fairseq/lib/python3.8/site-packages/omegaconf/omegaconf.py", line 613, in update root.setattr(last_key, value) File "/data1/home/liqiang/anaconda3/envs/fairseq/lib/python3.8/site-packages/omegaconf/dictconfig.py", line 286, in setattr self._format_and_raise(key=key, value=value, cause=e) File "/data1/home/liqiang/anaconda3/envs/fairseq/lib/python3.8/site-packages/omegaconf/base.py", line 95, in _format_and_raise format_and_raise( File "/data1/home/liqiang/anaconda3/envs/fairseq/lib/python3.8/site-packages/omegaconf/_utils.py", line 694, in format_and_rais _raise(ex, cause) File "/data1/home/liqiang/anaconda3/envs/fairseq/lib/python3.8/site-packages/omegaconf/_utils.py", line 610, in _raise raise ex # set end OC_CAUSE=1 for full backtrace omegaconf.errors.ValidationError: Invalid value 'simple', expected one of [c10d, fully_sharded, legacy_ddp, no_c10d, pytorch_ddp,_mo] full_key: distributed_training.ddp_backend reference_type=DistributedTrainingConfig object_type=DistributedTrainingConfig

The above exception was the direct cause of the following exception:

Traceback (most recent call last): File "/data1/home/liqiang/anaconda3/envs/fairseq/bin/fairseq-generate", line 33, in sys.exit(load_entry_point('fairseq', 'console_scripts', 'fairseq-generate')()) File "/data1/home/liqiang/fairseq/fairseq_cli/generate.py", line 404, in cli_main main(args) File "/data1/home/liqiang/fairseq/fairseq_cli/generate.py", line 49, in main return _main(cfg, sys.stdout) File "/data1/home/liqiang/fairseq/fairseq_cli/generate.py", line 96, in _main models, saved_cfg = checkpoint_utils.load_model_ensemble( File "/data1/home/liqiang/fairseq/fairseq/checkpoint_utils.py", line 319, in load_model_ensemble ensemble, args, _task = load_model_ensemble_and_task( File "/data1/home/liqiang/fairseq/fairseq/checkpoint_utils.py", line 361, in load_model_ensemble_and_task state = load_checkpoint_to_cpu(filename, arg_overrides) File "/data1/home/liqiang/fairseq/fairseq/checkpoint_utils.py", line 295, in load_checkpoint_to_cpu state = _upgrade_state_dict(state) File "/data1/home/liqiang/fairseq/fairseq/checkpoint_utils.py", line 537, in _upgrade_state_dict state["cfg"] = convert_namespace_to_omegaconf(state["args"]) File "/data1/home/liqiang/fairseq/fairseq/dataclass/utils.py", line 357, in convert_namespace_to_omegaconf composed_cfg = compose("config", overrides=overrides, strict=False) File "/data1/home/liqiang/anaconda3/envs/fairseq/lib/python3.8/site-packages/hydra/experimental/compose.py", line 31, in compos cfg = gh.hydra.compose_config( File "/data1/home/liqiang/anaconda3/envs/fairseq/lib/python3.8/site-packages/hydra/_internal/hydra.py", line 507, in compose_co cfg = self.config_loader.load_configuration( File "/data1/home/liqiang/anaconda3/envs/fairseq/lib/python3.8/site-packages/hydra/_internal/config_loader_impl.py", line 151, ad_configuration return self._load_configuration( File "/data1/home/liqiang/anaconda3/envs/fairseq/lib/python3.8/site-packages/hydra/_internal/config_loader_impl.py", line 277, oad_configuration ConfigLoaderImpl._apply_overrides_to_config(config_overrides, cfg) File "/data1/home/liqiang/anaconda3/envs/fairseq/lib/python3.8/site-packages/hydra/_internal/config_loader_impl.py", line 520, pply_overrides_to_config raise ConfigCompositionException( hydra.errors.ConfigCompositionException: Error merging override distributed_training.ddp_backend='simple'

facebookresearch / fairseq