Confirm valid implementation References: > Loss? Loss is: > Total span extraction loss is the sum of a Cross-Entropy for the start and end positions. https://huggingface.co/transformers/v4.9.2/model_doc/bert.html?highlight=bertforquestionanswering https://github.com/huggingface/transformers/blob/v4.44.2/src/transformers/models/bert/modeling_bert.py#L1996 google: https://github.com/google-research/bert/blob/master/run_squad.py#L646
Confirm valid implementation
References:
https://huggingface.co/transformers/v4.9.2/model_doc/bert.html?highlight=bertforquestionanswering
https://github.com/huggingface/transformers/blob/v4.44.2/src/transformers/models/bert/modeling_bert.py#L1996
google:
https://github.com/google-research/bert/blob/master/run_squad.py#L646