Spaces:
Sleeping
Sleeping
| from dotenv import load_dotenv | |
| import streamlit as st | |
| from PyPDF2 import PdfReader | |
| from langchain.text_splitter import CharacterTextSplitter | |
| from langchain.embeddings.huggingface import HuggingFaceEmbeddings | |
| from langchain.vectorstores import FAISS | |
| from langchain.chains.question_answering import load_qa_chain | |
| import os | |
| from streamlit_chat import message | |
| from langchain import HuggingFaceHub | |
| def LLM_pdf(model_name = 'google/flan-t5-large'): | |
| # st.header("Ask your PDF 💬") | |
| # upload file | |
| pdf = st.file_uploader("Upload your PDF", type="pdf") | |
| # extract the text | |
| if pdf is not None: | |
| pdf_reader = PdfReader(pdf) | |
| text = "" | |
| for page in pdf_reader.pages: | |
| text += page.extract_text() | |
| # split into chunks | |
| text_splitter = CharacterTextSplitter( | |
| separator="\n", | |
| chunk_size=1000, | |
| chunk_overlap=200, | |
| length_function=len | |
| ) | |
| print(text_splitter) | |
| chunks = text_splitter.split_text(text) | |
| # create embeddings | |
| embeddings = HuggingFaceEmbeddings() | |
| knowledge_base = FAISS.from_texts(chunks, embeddings) | |
| if 'generated' not in st.session_state: | |
| st.session_state['generated'] = [] | |
| if 'past' not in st.session_state: | |
| st.session_state['past'] = [] | |
| # print(st.session_state['generated'],st.session_state['past']) | |
| chat_placeholder = st.empty() | |
| # show user input | |
| with st.container(): | |
| input_placeholder = st.empty() | |
| user_question = input_placeholder.text_input("Ask a question about your PDF:") | |
| if user_question: | |
| docs = knowledge_base.similarity_search(user_question) | |
| llm = HuggingFaceHub(repo_id=model_name, model_kwargs={"temperature":5, | |
| "max_length":64}) | |
| chain = load_qa_chain(llm, chain_type="stuff") | |
| response = chain.run(input_documents=docs,question=user_question) | |
| #st.write(response) | |
| # append user_input and output to state | |
| st.session_state.past.append(user_question) | |
| st.session_state.generated.append(response) | |
| with chat_placeholder.container(): | |
| # If responses have been generated by the model | |
| if st.session_state['generated']: | |
| # Reverse iteration through the list | |
| for i in range(len(st.session_state['generated'])-1, -1, -1): | |
| # message from streamlit_chat | |
| message(st.session_state['past'][::-1][i], is_user=True, key=str(i) + '_user') | |
| message(st.session_state['generated'][::-1][i], key=str(i)) | |