🌘w🌖

app.py

@@ -9,6 +9,7 @@ import spaces
 from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
 import torch
 from threading import Thread
+from huggingface_hub import hf_hub_download
 
 token = os.environ["HF_TOKEN"]
 model = AutoModelForCausalLM.from_pretrained("google/gemma-7b-it",
@@ -25,9 +26,12 @@ title_text_dataset = load_dataset(
 ).select_columns(["title", "text"])
 
 # Load the int8 and binary indices. Int8 is loaded as a view to save memory, as we never actually perform search with it.
-int8_view = Index.restore("https://huggingface.co/spaces/sentence-transformers/quantized-retrieval/resolve/main/wikipedia_int8_usearch_1m.index", view=True)
+path_int8_view = hf_hub_download(repo_id="sentence-transformers/quantized-retrieval",repo_type="space", filename="wikipedia_ubinary_faiss_1m.index")
+int8_view = Index.restore(path_int8_view, view=True)
+
+path_binary_index = hf_hub_download(repo_id="sentence-transformers/quantized-retrieval",repo_type="space", filename="wikipedia_ubinary_faiss_1m.index")
 binary_index: faiss.IndexBinaryFlat = faiss.read_index_binary(
-    "https://huggingface.co/spaces/sentence-transformers/quantized-retrieval/resolve/main/wikipedia_ubinary_faiss_1m.index"
+    path_binary_index
 )
 
 # Load the SentenceTransformer model for embedding the queries
@@ -154,5 +158,6 @@ demo = gr.ChatInterface(fn=talk,
                         chatbot=gr.Chatbot(show_label=True, show_share_button=True, show_copy_button=True, likeable=True, layout="bubble", bubble_full_width=False),
                         theme="Soft",
                         examples=[["what is machine learning"]],
-                        title="Text Streaming")
+                        title=TITLE,
+                        description=DESCRIPTION)
 demo.launch()