Публикация решения для taskId 6a1d75c5fd30e81cf3126ae7: update src/utils.py

src/utils.py

@@ -9,6 +9,9 @@ from langchain_ollama import OllamaEmbeddings
 # Chroma collection name
 COLLECTION_NAME = "faq"
 
+# Persistence directory from config
+from src.config import CHROMA_PERSIST_DIR
+
 
 def load_faq_to_chroma(md_dir: str = "data") -> None:
     """Load all .md files from md_dir into a Chroma vector store.
@@ -22,21 +25,23 @@ def load_faq_to_chroma(md_dir: str = "data") -> None:
         all_docs.extend(docs)
     splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
     docs = splitter.split_documents(all_docs)
-    embeddings = OllamaEmbeddings(model="nomic-embed-text")
+    embeddings = OllamaEmbeddings("nomic-embed-text")
-    # Create or update Chroma collection
+    # Create or update Chroma collection with persistence
     Chroma.from_documents(
         docs,
         embeddings,
         collection_name=COLLECTION_NAME,
+        persist_directory=CHROMA_PERSIST_DIR,
     )
 
 
 def search_course_docs(query: str, k: int = 3) -> List[str]:
     """Return top k document snippets from the persisted Chroma store."""
-    embeddings = OllamaEmbeddings(model="nomic-embed-text")
+    embeddings = OllamaEmbeddings("nomic-embed-text")
     store = Chroma(
         collection_name=COLLECTION_NAME,
         embedding_function=embeddings,
+        persist_directory=CHROMA_PERSIST_DIR,
     )
     results = store.similarity_search(query, k=k)
     return [doc.page_content for doc in results]