From 8f5b4923c8caca1f352581eb6f2fda583517b1a6 Mon Sep 17 00:00:00 2001
From: zhangsibo1129 <134488188+zhangsibo1129@users.noreply.github.com>
Date: Sat, 23 Dec 2023 17:16:27 +0800
Subject: [PATCH] reformatted (#1128)

---
 examples/scripts/ppo_multi_adapter.py | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/examples/scripts/ppo_multi_adapter.py b/examples/scripts/ppo_multi_adapter.py
index 044a2419de..2bd489dfbd 100644
--- a/examples/scripts/ppo_multi_adapter.py
+++ b/examples/scripts/ppo_multi_adapter.py
@@ -21,8 +21,9 @@
 from tqdm import tqdm
 from transformers import AutoTokenizer, BitsAndBytesConfig, HfArgumentParser
 
-from trl import AutoModelForCausalLMWithValueHead, PPOConfig, PPOTrainer, is_xpu_available
+from trl import AutoModelForCausalLMWithValueHead, PPOConfig, PPOTrainer
 from trl.core import LengthSampler
+from trl.import_utils import is_npu_available, is_xpu_available
 
 
 input_min_text_length = 6
@@ -82,7 +83,7 @@ def tokenize(example):
 )
 model = AutoModelForCausalLMWithValueHead.from_pretrained(
     script_args.model_name,
-    device_map={"": "xpu:0"} if is_xpu_available() else {"": 0},
+    device_map={"": "xpu:0"} if is_xpu_available() else {"": "npu:0"} if is_npu_available else {"": 0},
     peft_config=lora_config,
     quantization_config=nf4_config,
     reward_adapter=script_args.rm_adapter,