AIAS1/2_nlp_sdks/bert_qa_sdk
2024-11-25 10:43:10 +08:00
..
models no message 2024-11-25 10:43:10 +08:00
src no message 2024-11-25 10:43:10 +08:00
bert_qa_sdk.iml no message 2024-11-25 10:43:10 +08:00
bert-qa-sdk.iml no message 2024-11-25 10:43:10 +08:00
pom.xml no message 2024-11-25 10:43:10 +08:00
README_CN.md no message 2024-11-25 10:43:10 +08:00
README_EN.md no message 2024-11-25 10:43:10 +08:00

Download the model and place it in the models directory

Bert Q&A SDK

Based on the BERT QA model, input a question and a text paragraph containing the answer (maximum length of 384), the model can find the best answer from the text paragraph.

image

Running Example - BertQaInferenceExample

  • Question:
When did Radio International start broadcasting?
  • Paragraph containing the answer (maximum length of 384):
Radio International was a general entertainment Channel.
Which operated between December 1983 and April 2001.

After a successful run, the command line should display the following information:

...
[INFO ] - Paragraph: Radio International was a general entertainment Channel.
          Which operated between December 1983 and April 2001.
          
[INFO ] - Question: When did Radio International start broadcasting?

[INFO ] - Answer: [december, 1983]