Issues
- 3
- 0
- 3
Fine-tuning SPECTER?
#20 opened by jacklxc - 1
json.decoder.JSONDecodeError: Expecting property name enclosed in double quotes: line 1 column 2 (char 1)
#28 opened by rrsun - 1
concat_title_abstract must be True in scripts/pytorch_lightning_training_script/train.py#L73
#49 opened by piegu - 0
- 5
How to create the vocab files? (tokens.txt, non_padded_namespaces.txt, venue.txt)
#45 opened by piegu - 1
Does the model use Bert's vocabulary or scibert's vocabulary, or do they all use it? Where is it used
#43 opened by humdingers - 4
- 0
- 1
where can I get the dataset used in the paper?
#41 opened by BinbinJin - 0
Using trained model: which tokenizer?
#38 opened by mle-els - 1
- 2
Positive paper sampling
#36 opened by malteos - 6
- 1
- 0
Use custom dataset without hard negatives
#34 opened by ryangawei - 6
Create preprocessed training files: metadata.json is missing ids in the train.txt, test.txt and val.txt
#13 opened by shauryr - 0
How to train my own model with other checkpoints?
#33 opened by ryangawei - 1
How to change the sequence length?
#31 opened by nshinton - 0
How to deal with this bug?
#32 opened by dandandandan77 - 5
Training Data Set for Spectre
#29 opened by ringkasannet - 12
- 1
Error when training: "None" is not a <Class 'allennlp.data.fields.field.Field'>
#30 opened by ringkasannet - 2
Model name 'data/scibert_scivocab_uncased/vocab.txt' was not found in model name list
#25 opened by nshinton - 2
- 0
Nan loss in training specter with specter/scripts/pytorch_lightning_training_script/train.py
#26 opened by AlexRossi6 - 3
Huggingface model produces different embeddings
#24 opened by malteos - 2
Confused about the Huggingface Usage (using ' ' instead of '[SEP]' for concatenating)
#22 opened by Punchwes - 4
- 1
- 2
Streaming data for inferrence
#11 opened by ckald - 3
- 2
Label namespace warning
#12 opened by DWFlanagan - 4
Help required for training data preparation
#3 opened by victor714 - 3
Question regarding train / val / test split
#9 opened by victor714 - 1
- 1
How to install this model in the docker
#5 opened by freshnemo - 4
- 2