allenai / allenai/longformer

ValueError: hidden size is called d_model

Offen
#174 0 Kommentare 0 Reaktionen 0 zugewiesene Personen Auf GitHub ansehen
Vorherrschende Sprache
Python
Sterne
2.2k
Forks
286
PR-Merge-Kennzahlen
Keine gemergten PRs in 30 T.

Beschreibung

I tried to run the summarization.py, and use
parser.add_argument("--model_path", type=str, default='longformer-base-4096', help="Path to the checkpoint directory or model name") instead of "facebook/bart-base" then it gave me error like this.

the full traceback is as following:
Traceback (most recent call last):
File "/home/featurize/work/heterformer/Heter Former/summarization.py", line 439, in
main(args)
File "/home/featurize/work/heterformer/Heter Former/summarization.py", line 382, in main
model = Summarizer(args)
File "/home/featurize/work/heterformer/Heter Former/summarization.py", line 161, in __init__
config = LongformerEncoderDecoderConfig.from_pretrained(self.args.model_path)
File "/home/featurize/tmp/lib/python3.7/site-packages/transformers/configuration_utils.py", line 312, in from_pretrained
return cls.from_dict(config_dict, **kwargs)
File "/home/featurize/tmp/lib/python3.7/site-packages/transformers/configuration_utils.py", line 397, in from_dict
config = cls(**config_dict)
File "/cloud/heterformer/Heter Former/longformer/longformer_encoder_decoder.py", line 34, in __init__
super().__init__(**kwargs)
File "/home/featurize/tmp/lib/python3.7/site-packages/transformers/configuration_bart.py", line 155, in __init__
raise ValueError("hidden size is called d_model")
ValueError: hidden size is called d_model

Beitragsleitfaden

Für dieses Repository ist kein Beitragsleitfaden indexiert

Bewertung

Dieses Issue wurde noch nicht bewertet.

Neue Issues direkt in Ihr Postfach

Eine kurze Übersicht über anfängerfreundliche GitHub-Issues.