Open masakuri opened 6 years ago
OK. please let me know following information
ls -lr ./save_jpmodel_dir/
head ./save_jpmodel_dir/bilstm-cnn-crf_adam_jp.vocab_char
Please tell me your .vocab_char
file contains '\t' ?
I just fix the code now. https://github.com/aonotas/deep-crf/commit/0fec504da3636230d4e8a9adf670e953cbbf527c
So please try recent version.
It worked! Thank you very much.
Thank you.
I fix code again.
please try to use recent version again.
Maybe you char_file contains :
CHAR
\t
CHAR
\t Index
.
Recent version is also worked. Thank you very much.
I checked my char_file (bilstm-cnn-crf_adam_jp.vocab_char
) but I could not find CHAR[tab]CHAR [tab]Index
.
I found an following unusual example but I could not find CHAR[tab]CHAR[tab]Index
ones.
<U+0080>[tab]438
~Maybe this is the cause.~ I don't know.
When I predicted with English test file, it worked. But when I predicted with Japanese test file (and set pre-trained Japanese word embeddings file), I got the following error.
My command is like:
deep-crf predict input_test_jp.txt --delimiter=" " --model_filename ./save_jpmodel_dir/bilstm-cnn-crf_adam_jp_epoch41.model --save_dir save_jpmodel_dir --save_name bilstm-cnn-crf_adam_jp --word_emb_file jp_word_emb300.txt --n_word_emb 300 --word_emb_vocab_type replace_only --predicted_output predicted41_jp.txt --gpu 0
Any ideas? Thank you.