Hi, thank you for sharing such impressive work. I'm new for this topic. I met a problem when I try to run the code. I wrote the test code as you guided, showing as follows:
from transformers import BertTokenizer, EncoderDecoderModel
from SentenceFusion import sentence_fusion as sf
primary_sentence = "cables on a table"
secondary_sentence = "a laptop on a table"
fuser.fuse(primary_sentence, secondary_sentence, delta=.5)
But when I run it, it reported, RuntimeError: probability tensor contains either inf, nan or element < 0, would you mind helping me fix it? Looking forward to your reply. Thank you.
Hi, thank you for sharing such impressive work. I'm new for this topic. I met a problem when I try to run the code. I wrote the test code as you guided, showing as follows:
from transformers import BertTokenizer, EncoderDecoderModel from SentenceFusion import sentence_fusion as sf
bert2bert = EncoderDecoderModel.from_pretrained('ralcanta/do_nothing_bert') tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
fuser = sf.SentenceFusion(tokenizer, bert2bert)
primary_sentence = "cables on a table" secondary_sentence = "a laptop on a table" fuser.fuse(primary_sentence, secondary_sentence, delta=.5)
But when I run it, it reported, RuntimeError: probability tensor contains either
inf
,nan
or element < 0, would you mind helping me fix it? Looking forward to your reply. Thank you.