Cannot load pre-trained model

[flipz357]

Hi Deng,

thanks for the awesome work. I am interested in parsing arbitrary sentence. I took your advice with appending the sentences with dummy-AMRs to the test. The preprocessing worked fine so far, however, when I want to apply the pre-trained model to predict using work.sh (with adapted file path for pre-trained model), I get the following error:

File "/xxxxx/AMR-gs/parser/work.py", line 94, in
model_args = torch.load(fname)['args']
....
File "/zzzzzr/lib/python3.6/site-packages/torch/serialization.py", line 563, in _load
magic_number = pickle_module.load(f, **pickle_load_args)
_pickle.UnpicklingError: invalid load key, '\x00'

The path in work.sh looks as follows:

--load_path amr2.0.bert.gr \

Previously I had downloaded your pre-trained model in the main directory and used tar -xvzf to extract. Did I make any error in the extraction process? My amr2.0.bert.gr contains a file "ckpt.pt" and a directory "vocabs".

[jcyk]

the --load_path should only point to the ckpt.pt file, not the whole directory. The vocab files should be placed to the directory specified by model_args, see

AMR-gs/parser/work.py

Lines 101 to 109 in b9c0043

	vocabs['tok'] = Vocab(model_args.tok_vocab, 5, [CLS])
	vocabs['lem'] = Vocab(model_args.lem_vocab, 5, [CLS])
	vocabs['pos'] = Vocab(model_args.pos_vocab, 5, [CLS])
	vocabs['ner'] = Vocab(model_args.ner_vocab, 5, [CLS])
	vocabs['predictable_concept'] = Vocab(model_args.predictable_concept_vocab, 5, [DUM, END])
	vocabs['concept'] = Vocab(model_args.concept_vocab, 5, [DUM, END])
	vocabs['rel'] = Vocab(model_args.rel_vocab, 50, [NIL])
	vocabs['word_char'] = Vocab(model_args.word_char_vocab, 100, [CLS, END])
	vocabs['concept_char'] = Vocab(model_args.concept_char_vocab, 100, [CLS, END])

I am sorry for this inconvenience.

[flipz357]

Thanks, Deng, it works now!

[jcyk]

It seems to me that the result is good, except for that you did not do the post-processing step so the result is incomplete. please let me know your point so I can help --). @fliegenpilz357

[flipz357]

Oh, stupid me! This is completely right, I did not run the post processing. Thank you for your patience and the quick answers!