swords

swords copied to clipboard

fixed open_fn definition when evaluating on a personalized dataset

caterinaLacerra

Hi, I've found a bug in the following method that is used for BERT-K and BERT-LS encoding sentences. **** In "methods/bert.py/_logits_for_dropout_target" **** with torch.no_grad(): embeddings = self.bert.embeddings(self.list_to_tensor(context_target_enc)) (Omitting Dropout procedure)...

twadada

Thanks for your code and contribution. I've conducted some testing on the legacy metrics, and it appears there might be an issue with the legacy metric-'best' score. When testing with...

Hyfred

The context of the dataset contains unbalanced double quotation marks, meaning there is an opening quote without a matching closing quote (or vice versa), which can affect tokenization, parsing, and...

NserUname