Create app.py

app.py

@@ -0,0 +1,154 @@
+# app.py
+
+import gradio as gr
+from transformers import pipeline
+from sentence_transformers import SentenceTransformer
+import faiss
+import numpy as np
+
+# --- Backend Logic ---
+
+# Step 1: Load the necessary models
+# UPGRADED: The generator model is now 'google/flan-t5-large' for better responses.
+print("Loading models... This may take a moment, especially the first time.")
+generator = pipeline("text2text-generation", model="google/flan-t5-large")
+embedder = SentenceTransformer("sentence-transformers/all-MiniLM-L6-v2")
+print("Models loaded successfully!")
+
+def chunk_text(text, chunk_size=256, overlap=32):
+    """Splits text into overlapping chunks."""
+    words = text.split()
+    chunks = []
+    for i in range(0, len(words), chunk_size - overlap):
+        chunks.append(" ".join(words[i:i + chunk_size]))
+    return chunks
+
+def process_chat_request(user_question, chat_history, state_data):
+    """
+    The main function that handles the chat logic using the RAG pipeline.
+    """
+    index = state_data.get("index")
+    chunks = state_data.get("chunks")
+
+    if not all([index, chunks]):
+        raise gr.Error("File index is missing. Please restart by uploading a file.")
+    if not user_question:
+        raise gr.Error("Please enter a question.")
+
+    try:
+        # 1. RETRIEVE: Find the most relevant chunks
+        question_embedding = embedder.encode([user_question])
+        _, top_k_indices = index.search(question_embedding, k=3) # Retrieve top 3 chunks
+        
+        context = " ".join([chunks[i] for i in top_k_indices[0]])
+
+        # 2. GENERATE: Create a prompt and get an answer
+        prompt = f"""
+        Based on the following context, provide a detailed answer to the user's question.
+
+        CONTEXT:
+        ---
+        {context}
+        ---
+
+        QUESTION: {user_question}
+
+        ANSWER:
+        """
+
+        result = generator(
+            prompt,
+            max_length=512,
+            num_beams=4,
+            temperature=0.1
+        )
+        bot_response = result[0]['generated_text']
+
+    except Exception as e:
+        raise gr.Error(f"An error occurred during processing: {e}")
+
+    chat_history.append((user_question, bot_response))
+    return "", chat_history
+
+# --- Gradio UI Definition ---
+
+with gr.Blocks(theme=gr.themes.Soft(primary_hue="teal", secondary_hue="teal"), title="Text File Analyzer") as demo:
+    app_state = gr.State({})
+
+    with gr.Column(visible=True) as welcome_page:
+        gr.Markdown(
+            """
+            <div style='text-align: center; font-family: "Garamond", serif; padding-top: 30px;'>
+                <h1 style='font-size: 3.5em;'>Efficient Text File Analyzer</h1>
+                <p style='font-size: 1.5em; color: #555;'>Chat with any .txt document using an efficient RAG pipeline.</p>
+            </div>
+            """
+        )
+        gr.HTML(
+            """
+            <div style='text-align: center; padding: 20px;'>
+                <img src='https://media.giphy.com/media/v1.Y2lkPTc5MGI3NjExd2Vjb3M2eGZzN2FkNWZpZzZ0bWl0c2JqZzZlMHVwZ2l4b2t0eXFpcyZlcD12MV9pbnRlcm5hbF9naWZfYnlfaWQmY3Q9Zw/YWjDA4k2n6d5Ew42zC/giphy.gif'
+                     style='max-width: 350px; margin: auto; border-radius: 20px; box-shadow: 0 8px 16px rgba(0,0,0,0.1);' />
+            </div>
+            """
+        )
+        with gr.Column(horizontal_alignment="center"):
+            gr.Markdown("### Upload Your Text File")
+            chat_file_upload = gr.File(label="Upload any .txt file", file_types=[".txt"])
+            lets_chat_button = gr.Button("💬 Process File and Start Chatting 💬", variant="primary")
+
+    with gr.Column(visible=False) as chat_page:
+        gr.Markdown("<h1 style='text-align: center;'>Chat with your Document</h1>")
+        chatbot_ui = gr.Chatbot(height=600, bubble_full_width=False)
+        with gr.Row():
+            user_input_box = gr.Textbox(placeholder="Ask a question about your file...", scale=5)
+            submit_button = gr.Button("Send", variant="primary", scale=1)
+
+    def go_to_chat(current_state, chat_file, progress=gr.Progress()):
+        if chat_file is None:
+            raise gr.Error("A file must be uploaded.")
+        
+        progress(0, desc="Reading file...")
+        with open(chat_file.name, 'r', encoding='utf-8') as f:
+            content = f.read()
+        
+        progress(0.2, desc="Chunking text...")
+        chunks = chunk_text(content)
+        
+        progress(0.5, desc="Creating embeddings... (This might take a moment)")
+        embeddings = embedder.encode(chunks, show_progress_bar=True)
+        
+        progress(0.8, desc="Building search index...")
+        index = faiss.IndexFlatL2(embeddings.shape[1])
+        index.add(np.array(embeddings).astype('float32'))
+        
+        new_state = {
+            "index": index,
+            "chunks": chunks
+        }
+        
+        progress(1, desc="Done!")
+        return (
+            new_state,
+            gr.Column(visible=False),
+            gr.Column(visible=True)
+        )
+
+    lets_chat_button.click(
+        fn=go_to_chat,
+        inputs=[app_state, chat_file_upload],
+        outputs=[app_state, welcome_page, chat_page]
+    )
+    submit_button.click(
+        fn=process_chat_request,
+        inputs=[user_input_box, chatbot_ui, app_state],
+        outputs=[user_input_box, chatbot_ui]
+    )
+    user_input_box.submit(
+        fn=process_chat_request,
+        inputs=[user_input_box, chatbot_ui, app_state],
+        outputs=[user_input_box, chatbot_ui]
+    )
+
+if __name__ == "__main__":
+    demo.launch()