summary-search.py

from langchain.embeddings.openai import OpenAIEmbeddings
from langchain.text_splitter import CharacterTextSplitter
from langchain.vectorstores.faiss import FAISS
from langchain.docstore.document import Document
from langchain.prompts import PromptTemplate

from langchain.chains.question_answering import load_qa_chain
from langchain.llms import OpenAI

def test_document(filename, query):

    with open(filename) as f:
        article = f.read()
    text_splitter = CharacterTextSplitter(chunk_size=2000, chunk_overlap=0)
    texts = text_splitter.split_text(article)

    embeddings = OpenAIEmbeddings()
    docsearch = FAISS.from_texts(texts, embeddings)

    chain = load_qa_chain(OpenAI(temperature=0), chain_type="stuff")

    docs = docsearch.similarity_search(query)

    out = chain.run(input_documents=docs, question=query)
    print(out)



document = "Summary-of-Benefits-original.txt"
print("document", document)
query = "What is my copay?"
query = "What is my copay for a mamography?"
test_document(document, query)

query = "What is my copay for a mamography?"
document = "summary-of-benefits-paragraphs.txt"
print("document", document)
test_document(document, query)