Update app.py

app.py

@@ -179,7 +179,8 @@ class BaseGenerator(ABC):
         self.cache = ResponseCache(cache_size)
         self.batch_processor = BatchProcessor(max_batch_size)
         self.health_check = HealthCheck()
-        
+        self.tokenizer = self.model_manager.tokenizers[model_name]
+        #self.tokenizer = self.load_tokenizer(llama_model_name)  # Add this line to initialize the tokenizer
         self.default_config = default_generation_config or GenerationConfig()
         self.model_config = model_config or ModelConfig()
         
@@ -411,7 +412,9 @@ class LlamaGenerator(BaseGenerator):
         default_generation_config: Optional[GenerationConfig] = None,
         model_config: Optional[ModelConfig] = None,
         cache_size: int = 1000,
-        max_batch_size: int = 32
+        max_batch_size: int = 32,
+#        self.tokenizer = self.load_tokenizer(llama_model_name)  # Add this line to initialize the tokenizer
+
     ):
         super().__init__(
             llama_model_name,