phiyodr/roberta-large-finetuned-squad2

The model is based on roberta-large and was finetuned on SQuAD2 . The corresponding papers you can found here (model) and here (data) The corresponding model and data can be found in the original version of this article . The training procedure is followed by a training procedure called a question-answering pipeline . The data is then evaluated using an evaluation function called evaluate-v2.0.0 (original script from here) The results are published in the next version of the above article . For more information about the training procedure, visit the original script and testnet . Back to the page you came from the original article: evaluate.py.py,