Mateusz Pieniak comments

Repositories
Issues
Comments

Results 4 comments of


                                            Mateusz Pieniak

Recommended GPU size when training BERT-base

@BigBadBurrow What batch size & float precision did you end up on Titan X (12GB)?

Error while running Named Entity Recognition

``` from transformers import BertTokenizer, BertForTokenClassification tokenizer = BertTokenizer.from_pretrained('allenai/scibert_scivocab_uncased') model = BertForTokenClassification.from_pretrained('allenai/scibert_scivocab_uncased') nlp = pipeline('ner', model=model, tokenizer=tokenizer) text = 'Clinical features of culture-proven Mycoplasma pneumoniae infections at King Abdulaziz University...

Add std.parseFloat

Thank you for the workaround!

Cuda enabling and Multi-label example

About the multi-label approach, equation 26 already gives you a scalar, so you don't need to pass it to any NLL function. Just remember to scale your `q` (multi-label target)...