facebookresearch / av_hubert

A self-supervised learning framework for audio-visual speech
Other
852 stars 136 forks source link

AssertionError #100

Open mabaisen opened 1 year ago

mabaisen commented 1 year ago

AssertionError: Could not infer task type from {'_name': 'av_hubert_pretraining', 'is_s2s': True, 'data': '/checkpoint/bshi/data/lrs3//exp/ls-hubert/tune-modality/all_tsv/', 'label_dir': '/checkpoint/bshi/data/lrs3//exp/ls-hubert/tune-modality/all_bpe/unigram1000/', 'normalize': True, 'labels': ['wrd'], 'single_target': True, 'stack_order_audio': 4, 'tokenizer_bpe_name': 'sentencepiece', 'max_sample_size': 500, 'modalities': ['video'], 'image_aug': True, 'pad_audio': True, 'random_crop': False, 'tokenizer_bpe_model': '/checkpoint/bshi/data/lrs3//lang/spm/spm_unigram1000.model', 'fine_tuning': True}. Available tasks: dict_keys(['translation', 'translation_lev', 'translation_from_pretrained_bart', 'legacy_masked_lm', 'hubert_pretraining', 'multilingual_translation', 'sentence_ranking', 'multilingual_masked_lm', 'speech_to_text', 'denoising', 'multilingual_denoising', 'sentence_prediction', 'online_backtranslation', 'semisupervised_translation', 'masked_lm', 'language_modeling', 'cross_lingual_lm', 'translation_from_pretrained_xlm', 'translation_multi_simple_epoch', 'simul_speech_to_text', 'simul_text_to_text', 'audio_pretraining', 'dummy_lm', 'dummy_masked_lm', 'dummy_mt'])

Nisarg-MARZ commented 10 months ago

I was able to resolve error by doing fairseq.utils.import_user_module(Namespace(user_dir="/path/to/repo/av_hubert/avhubert")).