调整训练参数

AITrain/train_npc_dialogue_lora.py

@@ -39,7 +39,7 @@ def process_func(example, tokenizer):
     instruction = tokenizer(
         f"<s><|im_start|>system\n{system_prompt}<|im_end|>\n"
         f"<|im_start|>user\n{instruction + user_input}<|im_end|>\n"
-        f"<|im_start|>assistant\n",
+        f"<|im_start|>assistant\n<think>\n\n</think>\n\n",
         add_special_tokens=False
     )
     
@@ -142,12 +142,12 @@ def train_lora_model(model_path, data_path, output_dir):
     model.config.use_cache = False  # 关闭缓存以节省显存
     
     # 5. 准备数据集
-    train_dataset = prepare_dataset(data_path, tokenizer)
+    train_preparedataset = prepare_dataset(data_path, tokenizer)
     
     # 6. 配置训练参数 - 针对3080显卡优化
     training_args = TrainingArguments(
         output_dir=output_dir,
-        per_device_train_batch_size=1,  # 减小batch size
+        per_device_train_batch_size=2,  # 减小batch size
         gradient_accumulation_steps=4,   # 增加梯度累积
         logging_steps=10,
         num_train_epochs=3,  # 增加训练轮数以充分学习角色特征
@@ -161,6 +161,7 @@ def train_lora_model(model_path, data_path, output_dir):
         dataloader_pin_memory=False,  # 减少内存使用
         remove_unused_columns=False,
         report_to="none",
+        #bf16=True,
         #fp16=True,  # 使用混合精度训练
         save_total_limit=3,  # 只保留最新的3个检查点
     )
@@ -175,7 +176,7 @@ def train_lora_model(model_path, data_path, output_dir):
     trainer = Trainer(
         model=model,
         args=training_args,
-        train_dataset=train_dataset,
+        train_dataset=train_preparedataset,
         data_collator=DataCollatorForSeq2Seq(tokenizer=tokenizer, padding=True),
          callbacks=[swanlab_callback] # 传入之前的swanlab_callback
     )
@@ -270,13 +271,13 @@ def test_trained_model(model_path, lora_path):
 
 def main():
     # 配置路径
-    model_path = '/mnt/g/Project02/AITrain/Qwen/Qwen3-8B-AWQ'  # 基础模型路径
+    model_path = '/mnt/e/AI/Project02/AITrain/Qwen/Qwen3-8B-AWQ'  # 基础模型路径
     data_path = './npc_dialogue_dataset.json'  # 训练数据路径
     output_dir = './output/NPC_Dialogue_LoRA'  # 输出目录
     
-    #####test
-    final_model_path = os.path.join(output_dir, "final_model")
-    test_trained_model(model_path, final_model_path)
+    # #####test
+    # final_model_path = os.path.join(output_dir, "final_model")
+    # test_trained_model(model_path, final_model_path)
     # 确保数据文件存在
     if not os.path.exists(data_path):
         print(f"数据文件不存在: {data_path}")