[opensearch] Index dataset

[KushalGaywala]

Retrieval Augmented Generation

• Retrieve the data from the documents, then use the model to formulate the answer or to answer relevant to the question.

• Index on the paragraph, and the tables with their respective headings

• that we can fetch the exact documents that is relative to the question and not the whole text file document

• Then push into opensearch

[sid-code14]

Use of chromadb, code by @mmshress cloud host was not possible for very large number for documents