add device 'cpu' to embed model

app.py

@@ -17,7 +17,7 @@ st.title("Aplikacja z LlamaIndex")
 db = chromadb.PersistentClient(path="./abc")
 chroma_collection = db.get_or_create_collection("pomoc_ukrainie")
 vector_store = ChromaVectorStore(chroma_collection=chroma_collection)
-embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-small-en-v1.5")
+embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-small-en-v1.5", device="cpu")
 
 # Utwórz pipeline do przetwarzania dokumentów
 pipeline = IngestionPipeline(
@@ -32,18 +32,13 @@ pipeline = IngestionPipeline(
 index = VectorStoreIndex.from_vector_store(vector_store, embed_model=embed_model)
 
 # Utwórz silnik zapytań
-# huggingface
 from transformers import AutoTokenizer
 
-# Settings.tokenizer = AutoTokenizer.from_pretrained(
-#     "Qwen/Qwen2-7B-Instruct"
-# )
-
 # Load the correct tokenizer for Qwen/Qwen2-7B-Instruct
 tokeni = AutoTokenizer.from_pretrained("Qwen/Qwen2-0.5B")
 
 llm = HuggingFaceLLM(model_name="Qwen/Qwen2-0.5B", tokenizer=tokeni)
-# print(llm._tokenizer)
+
 query_engine = index.as_query_engine(
     llm=llm,
     response_mode='compact')