Update README.md

README.md

@@ -117,6 +117,47 @@ pip install -r requirements.txt -i http://mirrors.aliyun.com/pypi/simple
 --image_folder
 --output_dir
 
+同时微调代码还需要修改[finetune_multitask_dialouge_doc.py](TextMonkey/Monkey/finetune_multitask_dialouge_doc.py)中的路径为本地模型路径;
+
+```
+class ModelArguments:
+    model_name_or_path: Optional[str] = field(default="../TextMonkey/TextMonkey_base")
+```
+
+```
+    # Set RoPE scaling factor
+    config = MonkeyConfig.from_pretrained(
+        "../TextMonkey/TextMonkey_base",
+        cache_dir=training_args.cache_dir,
+        trust_remote_code=True,
+    )
+```
+
+```
+    model = TextMonkeyLMHeadModel.from_pretrained(
+        model_args.model_name_or_path,
+        config=config,
+        cache_dir=training_args.cache_dir,
+        device_map=device_map,
+        trust_remote_code=True,
+        quantization_config=GPTQConfig(
+            bits=4, disable_exllama=True
+        )
+        if training_args.use_lora and lora_args.q_lora
+        else None,
+        ignore_mismatched_sizes=True
+    )
+
+    tokenizer = QWenTokenizer.from_pretrained(
+        model_args.model_name_or_path,
+        cache_dir=training_args.cache_dir,
+        model_max_length=training_args.model_max_length,
+        padding_side="right",
+        use_fast=False,
+        trust_remote_code=True,
+    )
+```
+
 ### 单机多卡
 
 训练需要8卡 A800 80G