Skip to content

renatoviolin/Question-Answering-Albert-Electra

Repository files navigation

Question-Answering-Albert-Electra

Question Answering using Albert and Electra using wikipedia text as context.

Description

This repository implements a pipeline to answer questions using wikipedia text. Bellow is the pipeline:

  1. Using the input query, search on google filtering the wikipedia pages.
  2. Read the body content of the wikipedia, preprocess text and split the corpus in paragraphs.
  3. Use BM25 algorithm to rank the best candidate passages, using the top K paragraphs.
  4. Selected paragraphs are used as input to Albert and Electra models.
  5. Both models try to find the answer given the candidate paragraphs.

Running

To predict with Electra, you need to download the pre-trained model from here. Extract the folder and adjust the DATA_MODEL_DIR (line 26) in qa_predict.py to point to the root folder.

Question 1 Question 2 Question 3 BM Scores

About

Question Answering using Albert and Electra

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages