Thank you Merve and Lewis for the chat. It was nice to watch just because of how much fun you guys have with these topics. Look forward to future workshops :)
42:45 I guess there can be added a sanity checking by first generating the questions from the documents and then running QA on the documents with the generated questions.
I have a very important question. For the dataset containing only question and answer features. How should I approach? For eg. If the user input question, model must generate answer.
Hi Lewis and Merve, is there any pre-trained model which handles abstractive Q&A ? After giving the context and question, the model frames the answers from the given contexts. ? In this case - No start token is provided in answer but just the text as answer in training data.
Stu****d question: there's a lot of examples of calling model(...) and assigning the result to 'output' .... what to do with the 'ouput' such that it will render the same out as there is when using the pipeline?
why question answers generate one line answers, suppose we have a document i want answers more than one line sentences,its doesn;t work........question-answering piplines .....generate only one lines answers,
like we have a document hugging face, i ask the question using question-answering piplines ........explain about hugging face, so, answers should be more than one lines...........if its reply answers in one sentences, i want more line to reply answers, becoz i asked about explain hugging face, how to solve this problem how to generate answers more than one lines ,please help me............thanks in advance
Thank you Merve and Lewis for the chat. It was nice to watch just because of how much fun you guys have with these topics. Look forward to future workshops :)
42:45 I guess there can be added a sanity checking by first generating the questions from the documents and then running QA on the documents with the generated questions.
Awesome talk. Amazing information. Thanks a lot for great talk.
Why haven't you put the links for the slides?
I have a very important question. For the dataset containing only question and answer features. How should I approach? For eg. If the user input question, model must generate answer.
can you kindly share the QAuestion answer impementation code?
Thank you! Excellent content. I enjoyed it .
Hi Lewis and Merve, is there any pre-trained model which handles abstractive Q&A ? After giving the context and question, the model frames the answers from the given contexts. ? In this case - No start token is provided in answer but just the text as answer in training data.
what is the text limit for text contest and can we use it offline for commercial use by downloading it?
I have 1mb of text can it analyze it?
Stu****d question:
there's a lot of examples of calling model(...) and assigning the result to 'output' .... what to do with the 'ouput' such that it will render the same out as there is when using the pipeline?
why question answers generate one line answers, suppose we have a document i want answers more than one line sentences,its doesn;t work........question-answering piplines .....generate only one lines answers,
like we have a document hugging face, i ask the question using question-answering piplines ........explain about hugging face, so, answers should be more than one lines...........if its reply answers in one sentences, i want more line to reply answers, becoz i asked about explain hugging face, how to solve this problem how to generate answers more than one lines ,please help me............thanks in advance
Great time
前排