Create app.py

app.py

@@ -0,0 +1,49 @@
+import gradio as gr
+import torch
+from transformers import RagTokenizer, RagRetriever, RagSequenceForGeneration
+
+# Load RAG model
+model_name = "facebook/rag-sequence-nq"
+tokenizer = RagTokenizer.from_pretrained(model_name)
+retriever = RagRetriever.from_pretrained(model_name, index_name="exact", use_dummy_dataset=True)
+model = RagSequenceForGeneration.from_pretrained(model_name, retriever=retriever)
+
+# Function to process uploaded document
+def process_file(file):
+    if file is None:
+        return "Please upload a document."
+    
+    file_text = file.decode("utf-8")
+    return file_text
+
+# Function to answer questions using RAG
+def answer_question(document, question):
+    if not document.strip():
+        return "Please provide document content."
+
+    inputs = tokenizer(question, document, return_tensors="pt", truncation=True)
+    with torch.no_grad():
+        generated = model.generate(**inputs)
+    answer = tokenizer.batch_decode(generated, skip_special_tokens=True)[0]
+    
+    return answer
+
+# Gradio UI
+with gr.Blocks() as app:
+    gr.Markdown("# 📄 Advanced RAG NLP Document Editor")
+
+    # File Uploader
+    file_input = gr.File(label="Upload Document (TXT only)", type="binary")
+    file_output = gr.Textbox(label="Extracted Text", lines=10)
+
+    file_input.change(process_file, inputs=file_input, outputs=file_output)
+
+    # Question Answering
+    question_input = gr.Textbox(label="Ask a Question")
+    answer_output = gr.Textbox(label="Answer", lines=2)
+    
+    submit_btn = gr.Button("Get Answer")
+    submit_btn.click(answer_question, inputs=[file_output, question_input], outputs=answer_output)
+
+# Launch in Hugging Face Spaces
+app.launch()