Create app.py

app.py

@@ -0,0 +1,110 @@
+import openai
+import gradio as gr
+from langchain.chains import RetrievalQA
+from langchain.llms import OpenAI
+from langchain.document_loaders import PyPDFLoader
+from langchain.embeddings.openai import OpenAIEmbeddings
+from langchain.vectorstores import FAISS
+from langchain.chat_models import ChatOpenAI
+from PyPDF2 import PdfReader
+
+# Function to load and process multiple PDFs
+def load_pdfs(files):
+    documents = []
+    for file in files:
+        loader = PyPDFLoader(file.name)
+        documents.extend(loader.load())  # Append documents from each file
+    return documents
+
+# Summarization function using GPT-4 for multiple PDFs
+def summarize_pdfs(files, openai_api_key):
+    openai.api_key = openai_api_key  # Set OpenAI API key
+
+    # Load and process the PDFs
+    documents = load_pdfs(files)
+
+    # Create embeddings for the documents
+    embeddings = OpenAIEmbeddings(openai_api_key=openai_api_key)
+
+    # Use Langchain's FAISS Vector Store to store and search the embeddings
+    vector_store = FAISS.from_documents(documents, embeddings)
+
+    # Create a RetrievalQA chain for summarization
+    llm = ChatOpenAI(model='gpt-4o', openai_api_key=openai_api_key)
+    qa_chain = RetrievalQA.from_chain_type(
+        llm=llm,
+        chain_type="stuff",
+        retriever=vector_store.as_retriever()
+    )
+
+    # Query the model for a summary of all PDFs
+    response = qa_chain.run("Summarize the content of the research papers.")
+    return response
+
+# Function to handle user queries for multiple PDFs
+def query_pdfs(files, user_query, openai_api_key):
+    openai.api_key = openai_api_key  # Set OpenAI API key
+
+    # Load and process the PDFs
+    documents = load_pdfs(files)
+
+    # Create embeddings for the documents
+    embeddings = OpenAIEmbeddings(openai_api_key=openai_api_key)
+
+    # Use LangChain's FAISS Vector Store to store and search the embeddings
+    vector_store = FAISS.from_documents(documents, embeddings)
+
+    # Create a RetrievalQA chain for querying the documents
+    llm = ChatOpenAI(model="gpt-4o", openai_api_key=openai_api_key)
+    qa_chain = RetrievalQA.from_chain_type(
+        llm=llm,
+        chain_type="stuff",
+        retriever=vector_store.as_retriever()
+    )
+
+    # Query the model for the user query
+    response = qa_chain.run(user_query)
+    return response
+
+# Define Gradio interface for handling multiple PDFs
+def create_gradio_interface():
+    with gr.Blocks() as demo:
+        gr.Markdown("### Multi-PDF Chat and Research Paper Summarizer using GPT-4 and LangChain")
+
+        # Input field for API Key
+        with gr.Row():
+            openai_api_key_input = gr.Textbox(label="Enter OpenAI API Key", type="password", placeholder="Enter your OpenAI API key here")
+
+        with gr.Tab("Summarize PDFs"):
+            with gr.Row():
+                pdf_files = gr.File(label="Upload PDF Documents", file_types=[".pdf"], multiple=True)
+                summarize_btn = gr.Button("Summarize")
+                summary_output = gr.Textbox(label="Summary", interactive=False)
+                clear_btn_summary = gr.Button("Clear Response")
+
+            # Summarize Button Logic
+            summarize_btn.click(summarize_pdfs, inputs=[pdf_files, openai_api_key_input], outputs=summary_output)
+
+            # Clear Response Button Logic for Summary Tab
+            clear_btn_summary.click(lambda: "", inputs=[], outputs=summary_output)
+
+        with gr.Tab("Ask Questions"):
+            with gr.Row():
+                pdf_files_q = gr.File(label="Upload PDF Documents", file_types=[".pdf"], multiple=True)
+                user_input = gr.Textbox(label="Enter your question")
+                answer_output = gr.Textbox(label="Answer", interactive=False)
+                query_btn = gr.Button("Ask")
+                clear_btn_answer = gr.Button("Clear Response")
+
+            # Submit Question Logic
+            query_btn.click(query_pdfs, inputs=[pdf_files_q, user_input, openai_api_key_input], outputs=answer_output)
+
+            # Clear Response Button Logic for Answer Tab
+            clear_btn_answer.click(lambda: "", inputs=[], outputs=answer_output)
+
+    return demo
+
+# Run Gradio app
+if __name__ == "__main__":
+    demo = create_gradio_interface()
+    demo.launch(debug=True)