File size: 4,992 Bytes
8e29341 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 |
import os
import streamlit as st
from langchain_openai import ChatOpenAI
from langchain.chains import load_summarize_chain
from langchain_community.document_loaders import PyPDFLoader, TextLoader, UnstructuredWordDocumentLoader
from langchain.prompts import PromptTemplate
from modules import app_page_definitions, app_logger,app_constants,file_utils
# Use the logger from app_config
app_logger = app_logger.app_logger
# Configurable batch size (4 pages per batch)
batch_size = app_constants.SUMMARIZER_BATCH
WORKSPACE_DIRECTORY = app_constants.WORKSPACE_DIRECTORY
def process_file(file_path, file_type):
if file_type == "text/plain":
loader = TextLoader(file_path)
elif file_type == "application/pdf":
loader = PyPDFLoader(file_path)
elif file_type == "application/vnd.openxmlformats-officedocument.wordprocessingml.document":
loader = UnstructuredWordDocumentLoader(file_path)
else:
raise ValueError(f"Unsupported file type: {file_type}")
app_logger.info(f"Processing file {file_path} of type {file_type}")
return loader.load_and_split()
def app():
app_logger.info("Starting Streamlit app - Summarizer Tool page")
# Fetch page configuration from app_page_definitions
page_config = app_page_definitions.PAGE_CONFIG.get("nav_summarize")
st.title(page_config["title"])
st.caption(page_config["caption"])
st.session_state.current_page = "nav_summarize"
uploaded_file = st.file_uploader("Upload your document here:", type=['txt', 'pdf', 'docx'], key="file_uploader")
if uploaded_file is not None:
file_path = file_utils.save_uploaded_file(uploaded_file,uploads_path=WORKSPACE_DIRECTORY + "/tmp")
docs = process_file(file_path, uploaded_file.type)
total_docs = len(docs)
app_logger.info(f"Total documents processed: {total_docs}")
if total_docs > 1:
doc_range = st.slider("Select document range for summarization", 1, total_docs, (1, total_docs))
else:
doc_range = (1, 1)
progress_bar = st.progress(0)
if st.button("Summarize"):
with st.spinner('Processing... Please wait'):
llm = ChatOpenAI(
model_name=app_constants.MODEL_NAME,
openai_api_key=app_constants.openai_api_key,
base_url=app_constants.local_model_uri,
streaming=True
)
prompt_template = """Write a concise summary of the following:
{text}
CONCISE SUMMARY:"""
prompt = PromptTemplate.from_template(prompt_template)
refine_template = (
"You are a content writer and your job is to produce a summary of input\n"
"We have provided an existing summary up to a certain point: {existing_answer}\n"
"Start and end properly and refine the existing summary "
"with some more context below.\n"
"------------\n"
"{text}\n"
"------------\n"
"Given the new context, refine the original summary. "
"If the context isn't useful, return the original summary."
)
refine_prompt = PromptTemplate.from_template(refine_template)
chain = load_summarize_chain(
llm=llm,
chain_type="refine",
question_prompt=prompt,
refine_prompt=refine_prompt,
return_intermediate_steps=True,
input_key="input_documents",
output_key="output_text",
)
start_doc, end_doc = doc_range
for i in range(start_doc - 1, min(end_doc, total_docs), batch_size):
batch_docs = docs[i:min(i + batch_size, total_docs)]
progress_value = (i + len(batch_docs)) / total_docs
progress_bar.progress(progress_value)
with st.expander(f"Processing Documents {i + 1} - {i + len(batch_docs)}", expanded=False):
intermediate_summary = chain.invoke({"input_documents": batch_docs}, return_only_outputs=True)
st.write(intermediate_summary)
selected_docs = docs[start_doc - 1:end_doc]
final_summary_response = chain.invoke({"input_documents": selected_docs}, return_only_outputs=True)
final_summary = final_summary_response['output_text'] if 'output_text' in final_summary_response else "No summary generated."
st.text_area("Final Summary", final_summary, height=300)
st.success("Summarization Completed!")
progress_bar.empty()
else:
st.warning("Please upload a document to summarize.")
app_logger.warning("No document uploaded for summarization") |