Update app.py

app.py

@@ -11,13 +11,13 @@ from groq import Groq
 def load_data():
     dataset = load_dataset("FreedomIntelligence/RAG-Instruct", split="train")
     df = pd.DataFrame(dataset)
-    return df[["instruction", "response"]]
+    return df[["question", "answer"]]
 
 # Generate embeddings and index
 @st.cache_resource
 def setup_faiss(data):
     model = SentenceTransformer("sentence-transformers/all-MiniLM-L6-v2")
-    embeddings = model.encode(data["instruction"].tolist())
+    embeddings = model.encode(data["question"].tolist())
     index = faiss.IndexFlatL2(embeddings.shape[1])
     index.add(np.array(embeddings))
     return model, index, embeddings
@@ -26,7 +26,7 @@ def setup_faiss(data):
 def retrieve_context(query, model, index, data, top_k=1):
     query_vec = model.encode([query])
     distances, indices = index.search(np.array(query_vec), top_k)
-    results = [data.iloc[i]["instruction"] + "\n\n" + data.iloc[i]["response"] for i in indices[0]]
+    results = [data.iloc[i]["question"] + "\n\n" + data.iloc[i]["answer"] for i in indices[0]]
     return "\n\n".join(results)
 
 # Call Groq LLM
@@ -46,15 +46,15 @@ st.title("🧠 RAG App using Groq API + RAG-Instruct Dataset")
 data = load_data()
 model, index, _ = setup_faiss(data)
 
-st.markdown("Ask a question based on the instruction-response knowledge base.")
+st.markdown("Ask a question based on the QA knowledge base.")
 
 # Optional queries
 optional_queries = [
-    "How to use a specific API function?",
-    "Explain how to fine-tune a model.",
-    "What is the difference between pretraining and finetuning?",
-    "How does retrieval-augmented generation work?",
-    "Explain self-supervised learning."
+    "What is retrieval-augmented generation?",
+    "How can I fine-tune a language model?",
+    "What are the components of a RAG system?",
+    "Explain prompt engineering basics.",
+    "How does FAISS indexing help in RAG?"
 ]
 
 query = st.text_input("Enter your question:", value=optional_queries[0])