Code Changes - Optimized code to check for HF repo

constants.py

@@ -32,6 +32,8 @@ EMBEDDING_MODELS = [
     "all-MiniLM-L6-v2"
 ]
 
+GLOBAL_HUGGINGFACE_REPO_EXISTS = False
+
 default_json = {
     "relevance_explanation": "Documents 1 and 4 contain useful information for answering the question. Document 1 mentions 'Signal Information' and 'Self Diagnosis', which are relevant to finding signal information. Document 4 provides additional context about signal information, such as it being only available for digital channels.",
     "all_relevant_sentence_keys": [

ragbench.py

@@ -36,7 +36,8 @@ from utils import (
 
 from constants import (
     HF_DATASET_REPO_NAME,
-    HF_REPO_TYPE
+    HF_REPO_TYPE,
+    GLOBAL_HUGGINGFACE_REPO_EXISTS
 )
 
 # Define document structure
@@ -86,7 +87,6 @@ class RAGSystem:
         self.embedder = None
         self.hybrid_retriever = None
         self.generator_client = None
-        self.huggingface_repo_exists = False
         
         # Set up API-based generator
         self._init_generator_api()
@@ -115,7 +115,7 @@ class RAGSystem:
 
     def _check_huggingface_repo(self):
 
-        if self.huggingface_repo_exists:
+        if GLOBAL_HUGGINGFACE_REPO_EXISTS:
             print(f"Repository {HF_DATASET_REPO_NAME} already exists, skipping check.")
             return True
         
@@ -136,7 +136,7 @@ class RAGSystem:
                 repo_type=HF_REPO_TYPE,
             )
 
-            self.huggingface_repo_exists = True
+            GLOBAL_HUGGINGFACE_REPO_EXISTS = True
             print(f"Repository {HF_DATASET_REPO_NAME} is accessible. Files:")
             for index, item in enumerate(repo_files):
                 print(f"Index, {index}, File: {item}")