push

app.py

@@ -1,14 +1,16 @@
 import gradio as gr
-from langchain.document_loaders import PyPDFLoader
+from langchain_community.document_loaders import PyPDFLoader
 from langchain.chains.summarize import load_summarize_chain
-from langchain.llms import OpenAI
+from langchain_openai import OpenAI
 from dotenv import load_dotenv
 import os
+from langchain.chat_models import ChatOpenAI
+from langchain.chains.question_answering import load_qa_chain
 
 # Load environment variables from .env file
 load_dotenv()
 
-def extract_text_and_summary_from_pdf(pdf_file):
+def extract_text_summary_and_guidance_from_pdf(pdf_file):
     # Load the PDF file using PyPDFLoader
     loader = PyPDFLoader(pdf_file.name)
     documents = loader.load()
@@ -19,25 +21,47 @@ def extract_text_and_summary_from_pdf(pdf_file):
         text += document.page_content
     
     # Initialize the OpenAI model with the API key from environment variables
-    openai_api_key = os.getenv("OPENAI_API_KEY")
-    llm = OpenAI(model="gpt-3.5-turbo", temperature=0, api_key=openai_api_key)
+    llm = ChatOpenAI(model_name="gpt-3.5-turbo", temperature=0)
     
     # Load the summarization chain
     summarize_chain = load_summarize_chain(llm)
-    
+    qa_chain = load_qa_chain(llm)
+
     # Get the summary of the text
-    summary = summarize_chain.run(text)
+    summary = summarize_chain.invoke(documents)
+    
+    # Get the QA chain answer
+    question = """
+    Context Setup:
+    You are given an earnings transcript PDF from a company's quarterly earnings call.
+    The document contains detailed discussions about the company's financial performance, future outlook, and guidance from executives.
+
+    Task:
+    Your task is to extract and summarize the key guidance provided by the company's executives during the earnings call.
+    
+    Instructions:
+    Focus on extracting specific forward-looking statements and guidance provided by the executives.
+    Include information about revenue projections, earnings forecasts, strategic initiatives, and any notable remarks about future performance.
+    Ignore general commentary, introductions, and routine financial data unless it directly pertains to future guidance.
+    Present the extracted information in bullet points.
+
+    Output Format:
+    Provide the extracted key guidance in bullet points.
+
+    """
+
+    answer = qa_chain.run(input_documents=documents, question=question)
     
-    return text, summary
+    return text, summary['output_text'], answer
 
 # Create a Gradio interface
 interface = gr.Interface(
-    fn=extract_text_and_summary_from_pdf,
+    fn=extract_text_summary_and_guidance_from_pdf,
     inputs=gr.File(label="Upload PDF"),
-    outputs=[gr.Textbox(label="Extracted Text"), gr.Textbox(label="Summary")],
-    title="PDF Text Extractor and Summarizer",
-    description="Upload a PDF file to extract and display its text content and summary."
+    outputs=[gr.Textbox(label="Extracted Text"), gr.Textbox(label="Summary"), gr.Textbox(label="Guidance")],
+    title="PDF Text Extractor, Summarizer, and QA Guidance",
+    description="Upload a PDF file to extract and display its text content, summary, and QA guidance."
 )
 
 # Launch the interface
-interface.launch(share=True)
+interface.launch(share=True)

requirements.txt

@@ -69,4 +69,5 @@ yarl==1.9.2
 pypdf==3.10.0
 pypdf2
 python-dotenv
-openai
+openai
+langchain-community