Update app.py

app.py

@@ -1,64 +1,202 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+from langchain_community.document_loaders import PyPDFLoader
+from langchain_community.embeddings import HuggingFaceEmbeddings
+from langchain_community.vectorstores import FAISS
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain.chains import RetrievalQA
+from langchain.prompts import PromptTemplate
+from langchain_community.chat_models import ChatOpenAI
+import os
+from tempfile import NamedTemporaryFile
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+# Load API Key
+def load_api_key():
+    if "OPENROUTER_API_KEY" in os.environ:
+        return os.getenv("OPENROUTER_API_KEY")
+    raise ValueError("API key not found in environment variables")
 
+# Process PDF files
+def process_pdfs(files):
+    all_chunks = []
+    for file_info in files:  # file_info is a Gradio File object
+        with NamedTemporaryFile(delete=False, suffix=".pdf") as tmp_file:
+            # Copy file content to temporary file
+            with open(file_info.name, "rb") as f:
+                tmp_file.write(f.read())
+            tmp_file_path = tmp_file.name
+        
+        try:
+            loader = PyPDFLoader(tmp_file_path)
+            pages = loader.load()
+            
+            text_splitter = RecursiveCharacterTextSplitter(
+                chunk_size=1000,
+                chunk_overlap=200,
+                length_function=len
+            )
+            chunks = text_splitter.split_documents(pages)
+            all_chunks.extend(chunks)
+        finally:
+            os.unlink(tmp_file_path)
+    
+    if not all_chunks:
+        raise ValueError("No content was loaded from the files")
+    
+    embeddings = HuggingFaceEmbeddings(
+        model_name="sentence-transformers/paraphrase-multilingual-mpnet-base-v2"
+    )
+    vectorstore = FAISS.from_documents(all_chunks, embeddings)
+    return vectorstore.as_retriever(search_kwargs={"k": 3})
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+# Initialize language model
+def load_model():
+    return ChatOpenAI(
+        base_url="https://openrouter.ai/api/v1",
+        api_key="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",
+        model="mistralai/mistral-7b-instruct",
+        temperature=0.3
+    )
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+# Prompt template
+template = """
+You are an intelligent assistant specialized in document analysis.
+Use the following information from PDF files to answer the question:
 
-    messages.append({"role": "user", "content": message})
+{context}
 
-    response = ""
+Question: {question}
+Answer (in detail and in clear language):
+"""
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+prompt = PromptTemplate(
+    input_variables=["context", "question"],
+    template=template
+)
 
-        response += token
-        yield response
+# Application state variables
+qa_chain = None
+chat_history = []
 
+# Function to process messages and respond
+def respond(message, chat_history):
+    global qa_chain
+    
+    if qa_chain is None:
+        return chat_history + [(message, "Please upload PDF files first")]
+    
+    try:
+        result = qa_chain({"query": message})
+        response = result["result"]
+        return chat_history + [(message, response)]
+    except Exception as e:
+        return chat_history + [(message, f"An error occurred: {str(e)}")]
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
+# Handle file uploads
+def handle_upload(files):
+    global qa_chain
+    
+    try:
+        file_contents = [file.read() for file in files]
+        retriever = process_pdfs(file_contents)
+        llm = load_model()
+        
+        qa_chain = RetrievalQA.from_chain_type(
+            llm=llm,
+            retriever=retriever,
+            chain_type="stuff",
+            chain_type_kwargs={"prompt": prompt},
+            return_source_documents=False
+        )
+        return "Files uploaded successfully! You can now ask questions."
+    except Exception as e:
+        return f"Error uploading files: {str(e)}"
 
+# UI
+with gr.Blocks(title="Smart Document Assistant", theme=gr.themes.Default()) as demo:
+    gr.Markdown("# 📄 Smart Document Assistant")
+    gr.Markdown("Upload PDF files then start chatting")
+    
+    # Chat section
+    chatbot = gr.Chatbot(height=500)
+    
+    # Input section
+    with gr.Row():
+        msg = gr.Textbox(
+            placeholder="Type your question here...",
+            show_label=False,
+            scale=4
+        )
+        submit_btn = gr.Button("Send", scale=1)
+    
+    # File section
+    with gr.Row():
+        file_upload = gr.Files(
+            label="Upload PDF files",
+            file_types=[".pdf"],
+            file_count="multiple"
+        )
+        upload_status = gr.Textbox(label="Upload Status", interactive=False)
+    
+    clear_btn = gr.Button("Clear Chat")
+    
+    # Event handling
+    def handle_upload(files):
+        global qa_chain
+        try:
+            retriever = process_pdfs(files)
+            llm = load_model()
+            
+            qa_chain = RetrievalQA.from_chain_type(
+                llm=llm,
+                retriever=retriever,
+                chain_type="stuff",
+                chain_type_kwargs={
+                    "prompt": PromptTemplate(
+                        template=template,
+                        input_variables=["context", "question"]
+                    )
+                },
+                return_source_documents=False
+            )
+            return "Files uploaded successfully!"
+        except Exception as e:
+            return f"Error uploading files: {str(e)}"
+    
+    file_upload.change(
+        handle_upload,
+        inputs=file_upload,
+        outputs=upload_status
+    )
+    
+    submit_btn.click(
+        respond,
+        inputs=[msg, chatbot],
+        outputs=[chatbot]
+    ).then(
+        lambda: "",
+        None,
+        [msg]
+    )
+    
+    msg.submit(
+        respond,
+        inputs=[msg, chatbot],
+        outputs=[chatbot]
+    ).then(
+        lambda: "",
+        None,
+        [msg]
+    )
+    
+    clear_btn.click(
+        lambda: [],
+        None,
+        [chatbot]
+    )
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch(
+        server_name="0.0.0.0",
+        server_port=7860,
+        share=True  # To get a public link
+    )