MASS-Summarization get error in Inference

[jimmyhsia]

When i use the script to implement Inference and get error like Traceback (most recent call last): File "/home/user/.local/bin/fairseq-generate", line 11, in <module> load_entry_point('fairseq==0.8.0', 'console_scripts', 'fairseq-generate')() File "/home/user/.local/lib/python3.7/site-packages/fairseq_cli/generate.py", line 186, in cli_main main(args) File "/home/user/.local/lib/python3.7/site-packages/fairseq_cli/generate.py", line 32, in main task = tasks.setup_task(args) File "/home/user/.local/lib/python3.7/site-packages/fairseq/tasks/__init__.py", line 17, in setup_task return TASK_REGISTRY[args.task].setup_task(args, **kwargs) File "/home/user/.local/lib/python3.7/site-packages/fairseq/tasks/translation.py", line 155, in setup_task raise Exception('Could not infer language pair, please provide it explicitly') Exception: Could not infer language pair, please provide it explicitly what i miss?? or it need input language type anyone help ,thank

[StillKeepTry]

Can you give more details about your running scripts? @jimmyhsia

[steremma]

I have another issue with inference in summarization. I have downloaded the binarised data files into cnndm/processed and the FT model base. I also cloned MASS. I am launching inference as:

DATADIR=cnndm/processed/
MODEL=mass-base-uncased.pt
fairseq-generate $DATADIR --path $MODEL --user-dir MASS/MASS-summarization/mass/ --task translation_mass  --batch-size 64 --beam 5 --min-len 50 --no-repeat-ngram-size 3 --lenpen 1.0 

I get this error from fair-seq:

$ fairseq-generate $DATADIR --path $MODEL     --user-dir MASS/MASS-summarization/mass --task translation_mass     --batch-size 64 --beam 5 --min-len 50 --no-repeat-ngram-size 3     --lenpen 1.0
Namespace(beam=5, bpe=None, cpu=False, criterion='cross_entropy', data='cnndm', dataset_impl=None, diverse_beam_groups=-1, diverse_beam_strength=0.5, force_anneal=None, fp16=False, fp16_init_scale=128, fp16_scale_tolerance=0.0, fp16_scale_window=None, gen_subset='test', lazy_load=False, left_pad_source='True', left_pad_target='False', lenpen=1.0, log_format=None, log_interval=1000, lr_scheduler='fixed', lr_shrink=0.1, match_source_len=False, max_len_a=0, max_len_b=200, max_sentences=64, max_source_positions=1024, max_target_positions=1024, max_tokens=None, memory_efficient_fp16=False, min_len=50.0, min_loss_scale=0.0001, model_overrides='{}', momentum=0.99, nbest=1, no_beamable_mm=False, no_early_stop=False, no_progress_bar=False, no_repeat_ngram_size=3, num_shards=1, num_workers=1, optimizer='nag', path='mass-base-uncased.pt', prefix_size=0, print_alignment=False, quiet=False, raw_text=False, remove_bpe=None, replace_unk=None, required_batch_size_multiple=8, results_path=None, sacrebleu=False, sampling=False, sampling_topk=-1, sampling_topp=-1.0, score_reference=False, seed=1, shard_id=0, skip_invalid_size_inputs_valid_test=False, source_lang=None, target_lang=None, task='translation_mass', tbmf_wrapper=False, temperature=1.0, tensorboard_logdir='', threshold_loss_scale=None, tokenizer=None, unkpen=0, unnormalized=False, upsample_primary=1, user_dir='MASS/MASS-summarization/mass', warmup_updates=0, weight_decay=0.0)
| [src] dictionary: 30522 types
| [tgt] dictionary: 30522 types
| loaded 11490 examples from: cnndm/test.src-tgt.src
| loaded 11490 examples from: cnndm/test.src-tgt.tgt
| cnndm test src-tgt 11490 examples
| loading model(s) from mass-base-uncased.pt
Traceback (most recent call last):
  File "/mnt/work/home/estergiadis/mass_env/bin/fairseq-generate", line 8, in <module>
    sys.exit(cli_main())
  File "/mnt/work/home/estergiadis/mass_env/lib/python3.6/site-packages/fairseq_cli/generate.py", line 186, in cli_main
    main(args)
  File "/mnt/work/home/estergiadis/mass_env/lib/python3.6/site-packages/fairseq_cli/generate.py", line 47, in main
    task=task,
  File "/mnt/work/home/estergiadis/mass_env/lib/python3.6/site-packages/fairseq/checkpoint_utils.py", line 155, in load_model_ensemble
    ensemble, args, _task = load_model_ensemble_and_task(filenames, arg_overrides, task)
  File "/mnt/work/home/estergiadis/mass_env/lib/python3.6/site-packages/fairseq/checkpoint_utils.py", line 166, in load_model_ensemble_and_task
    state = load_checkpoint_to_cpu(filename, arg_overrides)
  File "/mnt/work/home/estergiadis/mass_env/lib/python3.6/site-packages/fairseq/checkpoint_utils.py", line 138, in load_checkpoint_to_cpu
    args = state['args']
KeyError: 'args'

[A-Rain]

@jimmyhsia have solve it? I have met the same problem.