Update app.py

app.py

@@ -2,6 +2,7 @@ import os
 import json
 import re
 from sentence_transformers import SentenceTransformer, CrossEncoder
+from huggingface_hub import hf_hub_download
 from openai import OpenAI
 import hnswlib
 import numpy as np
@@ -19,11 +20,12 @@ MAX_INPUT_TOKEN_LENGTH = 4000
 EMBED_DIM = 1024
 K = 10
 EF = 100
-SEARCH_INDEX = "search_index.bin"
-EMBEDDINGS_FILE = "embeddings.npy"
-DOCUMENT_DATASET = "chunked_data.parquet"
 COSINE_THRESHOLD = 0.7
 
+SEARCH_INDEX = hf_hub_download(repo_id="sayakpaul/diffusers-qa-chatbot-artifacts", filename="search_index.bin", repo_type="dataset")
+EMBEDDINGS_FILE = hf_hub_download(repo_id="sayakpaul/diffusers-qa-chatbot-artifacts", filename="embeddings.npy", repo_type="dataset")
+DOCUMENT_DATASET = hf_hub_download(repo_id="sayakpaul/diffusers-qa-chatbot-artifacts", filename="chunked_data.parquet", repo_type="dataset")
+
 torch_device = "cuda" if torch.cuda.is_available() else "cpu"
 print("Running on device:", torch_device)
 print("CPU threads:", torch.get_num_threads())
@@ -294,7 +296,7 @@ def check_input_token_length(message: str, chat_history: list[tuple[str, str]],
         )
 
 
-search_index = create_hnsw_index(EMBEDDINGS_FILE)  # load_hnsw_index(SEARCH_INDEX)
+search_index = load_hnsw_index(SEARCH_INDEX) # create_hnsw_index(EMBEDDINGS_FILE) 
 data_df = pd.read_parquet(DOCUMENT_DATASET).reset_index()
 with gr.Blocks(css="style.css") as demo:
     gr.Markdown(DESCRIPTION)