Hands-on Question Answering Systems with BERT by Navin Sabharwal & Amit Agrawal

Hands-on Question Answering Systems with BERT by Navin Sabharwal & Amit Agrawal

Author:Navin Sabharwal & Amit Agrawal
Language: eng
Format: epub
ISBN: 9781484266649
Publisher: Apress


Please note that the value for the max_seq_length parameters should be the same as what was used during the training process.

For this book, we will demonstrate implementation of a question classification dataset where questions will be classified into their respective categories. There are mainly two types of questions, factoid (nondescriptive) and non-factoid questions. As an example, “What is the temperature in Delhi?” is a factoid question, as it is looking for an answer based on some facts. “What is temperature?” is a non-factoid question, as it is looking for text snippets about temperature. For this implementation, please refer to the dataset at https://cogcomp.seas.upenn.edu/Data/QA/QC/.

Now we will see how a question classification system can be implemented using BERT.1.

For this implementation, we will download the BERT base-cased model from GitHub as described previously.



Download



Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.