Update app.py

app.py

@@ -17,6 +17,7 @@ import torch
 import scipy.io.wavfile
 import tempfile
 
+from huggingface_hub import hf_hub_download
 from sentence_transformers import SentenceTransformer
 from huggingface_hub import InferenceClient
 from transformers import AutoTokenizer, pipeline
@@ -33,9 +34,15 @@ torch_dtype = torch.float16 if device == "cuda" else torch.float32
 print(f"Running on: {device}")
 
 # ── RAG: FAISS index ──────────────────────────────────────────
-print("Loading FAISS index...")
-index = faiss.read_index("alzheimers_index.faiss")
-with open("chunks.pkl", "rb") as f:
+print("Loading FAISS index and chunks from HF...")
+
+HF_RAG_REPO = "AgentsWithoutBorders/alzheimers-faiss"
+
+faiss_file = hf_hub_download(HF_RAG_REPO, "alzheimers_index_233.faiss")
+chunks_file = hf_hub_download(HF_RAG_REPO, "chunks_233.pkl")
+
+index = faiss.read_index(faiss_file)
+with open(chunks_file, "rb") as f:
     chunks = pickle.load(f)
 
 embed_model = SentenceTransformer("all-MiniLM-L6-v2")