Set cache dir

main.py

@@ -67,7 +67,7 @@ def generate(item: Item):
     device = "cuda" if torch.cuda.is_available() else "cpu"
 
     model_path = "ibm-granite/granite-34b-code-instruct-8k"
-    tokenizer = AutoTokenizer.from_pretrained(model_path)
+    tokenizer = AutoTokenizer.from_pretrained(model_path, cache_dir="/code/huggingface/transformers")
     # drop device_map if running on CPU
     model = AutoModelForCausalLM.from_pretrained(model_path, device_map=device)
     model.eval()