fix template name (#389)

Author: tastelikefeet

README.md

@@ -62,7 +62,7 @@ Users can check the [documentation of SWIFT](docs/source/GetStarted/快速使用
 
 
 ## 🎉 News
-- 🔥2024.02.05: Support qwen1.5 series: like [qwen1.5-0.5b](https://www.modelscope.cn/models/qwen/Qwen1.5-0.5B/summary), [qwen1.5-7b](https://www.modelscope.cn/models/qwen/Qwen1.5-7B/summary),[qwen1.5-14b](https://www.modelscope.cn/models/qwen/Qwen1.5-14B/summary) , etc. To view all supported qwen1.5 models please check [Model List](https://github.com/modelscope/swift/blob/main/docs/source/LLM/%E6%94%AF%E6%8C%81%E7%9A%84%E6%A8%A1%E5%9E%8B%E5%92%8C%E6%95%B0%E6%8D%AE%E9%9B%86.md).
+- 🔥2024.02.05: Support Qwen1.5 series: like [qwen1.5-0.5b](https://www.modelscope.cn/models/qwen/Qwen1.5-0.5B/summary), [qwen1.5-7b](https://www.modelscope.cn/models/qwen/Qwen1.5-7B/summary),[qwen1.5-14b](https://www.modelscope.cn/models/qwen/Qwen1.5-14B/summary) , etc. To view all supported Qwen1.5 models please check [Model List](https://github.com/modelscope/swift/blob/main/docs/source/LLM/%E6%94%AF%E6%8C%81%E7%9A%84%E6%A8%A1%E5%9E%8B%E5%92%8C%E6%95%B0%E6%8D%AE%E9%9B%86.md).
 - 2024.02.01: Support openbmb-minicpm series: [openbmb-minicpm-2b-sft-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/openbmb_minicpm_2b_sft_chat), openbmb-minicpm-2b-chat.
 - 🔥2024.02.01: Support dataset mixture to reduce **Catastrophic Forgetting**. Use `--train_dataset_mix_ratio 2.0` to train! We also provide a common knowledge dataset [ms-bench](https://www.modelscope.cn/datasets/iic/ms_bench/summary).
 - 🔥2024.02.01: Support Agent training! Agent training algorithm comes from this [paper](https://arxiv.org/pdf/2309.00986.pdf). We also introduce the [ms-agent](https://www.modelscope.cn/datasets/iic/ms_agent/summary) dataset. Use [this script](https://github.com/modelscope/swift/blob/main/examples/pytorch/llm/scripts/qwen_7b_chat/lora/sft.sh) to begin an agent training!

README_CN.md

@@ -60,7 +60,7 @@ SWIFT（Scalable lightWeight Infrastructure for Fine-Tuning）是一个可扩展
 用户可以查看 [SWIFT官方文档](docs/source/GetStarted/快速使用.md) 来了解详细信息。
 
 ## 🎉 新闻
-- 🔥2024.02.05: 支持qwen1.5系列模型: [qwen1.5-0.5b](https://www.modelscope.cn/models/qwen/Qwen1.5-0.5B/summary), [qwen1.5-7b](https://www.modelscope.cn/models/qwen/Qwen1.5-7B/summary),[qwen1.5-14b](https://www.modelscope.cn/models/qwen/Qwen1.5-14B/summary)等, 支持的所有qwen1.5系列模型请查看[模型列表](https://github.com/modelscope/swift/blob/main/docs/source/LLM/%E6%94%AF%E6%8C%81%E7%9A%84%E6%A8%A1%E5%9E%8B%E5%92%8C%E6%95%B0%E6%8D%AE%E9%9B%86.md).
+- 🔥2024.02.05: 支持Qwen1.5系列模型: [qwen1.5-0.5b](https://www.modelscope.cn/models/qwen/Qwen1.5-0.5B/summary), [qwen1.5-7b](https://www.modelscope.cn/models/qwen/Qwen1.5-7B/summary),[qwen1.5-14b](https://www.modelscope.cn/models/qwen/Qwen1.5-14B/summary)等, 支持的所有Qwen1.5系列模型请查看[模型列表](https://github.com/modelscope/swift/blob/main/docs/source/LLM/%E6%94%AF%E6%8C%81%E7%9A%84%E6%A8%A1%E5%9E%8B%E5%92%8C%E6%95%B0%E6%8D%AE%E9%9B%86.md).
 - 2024.02.01: 支持openbmb-minicpm系列: [openbmb-minicpm-2b-sft-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/openbmb_minicpm_2b_sft_chat), openbmb-minicpm-2b-chat.
 - 🔥2024.02.01: 支持数据集打混来减少 **灾难性遗忘问题**. 使用`--train_dataset_mix_ratio 2.0`开启训练！同时我们也开源了通用知识数据集 [ms-bench](https://www.modelscope.cn/datasets/iic/ms_bench/summary).
 - 🔥2024.02.01: 支持Agent训练！Agent训练算法源自这篇[论文](https://arxiv.org/pdf/2309.00986.pdf). 我们也增加了[ms-agent](https://www.modelscope.cn/datasets/iic/ms_agent/summary)这个优质的agent数据集. 使用[这个脚本](https://github.com/modelscope/swift/blob/main/examples/pytorch/llm/scripts/qwen_7b_chat/lora/sft.sh)开启Agent训练!

swift/llm/utils/model.py

@@ -724,7 +724,7 @@ def cross_entropy_forward(self, inputs: Tensor,
     ModelType.qwen1half_0_5b_chat,
     'qwen/Qwen1.5-0.5B-Chat',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     support_flash_attn=True,
     support_vllm=True,
     requires=['transformers>=4.37'])
@@ -740,7 +740,7 @@ def cross_entropy_forward(self, inputs: Tensor,
     ModelType.qwen1half_1_8b_chat,
     'qwen/Qwen1.5-1.8B-Chat',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     support_flash_attn=True,
     support_vllm=True,
     requires=['transformers>=4.37'])
@@ -756,7 +756,7 @@ def cross_entropy_forward(self, inputs: Tensor,
     ModelType.qwen1half_4b_chat,
     'qwen/Qwen1.5-4B-Chat',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     support_flash_attn=True,
     support_vllm=True,
     requires=['transformers>=4.37'])
@@ -772,7 +772,7 @@ def cross_entropy_forward(self, inputs: Tensor,
     ModelType.qwen1half_7b_chat,
     'qwen/Qwen1.5-7B-Chat',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     support_flash_attn=True,
     support_vllm=True,
     requires=['transformers>=4.37'])
@@ -788,7 +788,7 @@ def cross_entropy_forward(self, inputs: Tensor,
     ModelType.qwen1half_14b_chat,
     'qwen/Qwen1.5-14B-Chat',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     support_flash_attn=True,
     support_vllm=True,
     requires=['transformers>=4.37'])
@@ -804,7 +804,7 @@ def cross_entropy_forward(self, inputs: Tensor,
     ModelType.qwen1half_72b_chat,
     'qwen/Qwen1.5-72B-Chat',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     support_flash_attn=True,
     support_vllm=True,
     requires=['transformers>=4.37'])
@@ -1068,7 +1068,7 @@ def get_model_tokenizer_with_flash_attn(model_dir: str,
     ModelType.qwen1half_0_5b_chat_int4,
     'qwen/Qwen1.5-0.5B-Chat-GPTQ-Int4',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     requires=['auto_gptq>=0.5', 'transformers>=4.37'],
     torch_dtype=torch.float16,
     function_kwargs={'bits': 4},
@@ -1078,7 +1078,7 @@ def get_model_tokenizer_with_flash_attn(model_dir: str,
     ModelType.qwen1half_0_5b_chat_int8,
     'qwen/Qwen1.5-0.5B-Chat-GPTQ-Int8',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     requires=['auto_gptq>=0.5', 'transformers>=4.37'],
     torch_dtype=torch.float16,
     function_kwargs={'bits': 8},
@@ -1088,7 +1088,7 @@ def get_model_tokenizer_with_flash_attn(model_dir: str,
     ModelType.qwen1half_1_8b_chat_int4,
     'qwen/Qwen1.5-1.8B-Chat-GPTQ-Int4',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     requires=['auto_gptq>=0.5', 'transformers>=4.37'],
     torch_dtype=torch.float16,
     function_kwargs={'bits': 4},
@@ -1098,7 +1098,7 @@ def get_model_tokenizer_with_flash_attn(model_dir: str,
     ModelType.qwen1half_1_8b_chat_int8,
     'qwen/Qwen1.5-1.8B-Chat-GPTQ-Int8',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     requires=['auto_gptq>=0.5', 'transformers>=4.37'],
     torch_dtype=torch.float16,
     function_kwargs={'bits': 8},
@@ -1108,7 +1108,7 @@ def get_model_tokenizer_with_flash_attn(model_dir: str,
     ModelType.qwen1half_4b_chat_int4,
     'qwen/Qwen1.5-4B-Chat-GPTQ-Int4',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     requires=['auto_gptq>=0.5', 'transformers>=4.37'],
     torch_dtype=torch.float16,
     function_kwargs={'bits': 4},
@@ -1118,7 +1118,7 @@ def get_model_tokenizer_with_flash_attn(model_dir: str,
     ModelType.qwen1half_4b_chat_int8,
     'qwen/Qwen1.5-4B-Chat-GPTQ-Int8',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     requires=['auto_gptq>=0.5', 'transformers>=4.37'],
     torch_dtype=torch.float16,
     function_kwargs={'bits': 8},
@@ -1128,7 +1128,7 @@ def get_model_tokenizer_with_flash_attn(model_dir: str,
     ModelType.qwen1half_7b_chat_int4,
     'qwen/Qwen1.5-7B-Chat-GPTQ-Int4',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     requires=['auto_gptq>=0.5', 'transformers>=4.37'],
     torch_dtype=torch.float16,
     function_kwargs={'bits': 4},
@@ -1138,7 +1138,7 @@ def get_model_tokenizer_with_flash_attn(model_dir: str,
     ModelType.qwen1half_7b_chat_int8,
     'qwen/Qwen1.5-7B-Chat-GPTQ-Int8',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     requires=['auto_gptq>=0.5', 'transformers>=4.37'],
     torch_dtype=torch.float16,
     function_kwargs={'bits': 8},
@@ -1148,7 +1148,7 @@ def get_model_tokenizer_with_flash_attn(model_dir: str,
     ModelType.qwen1half_14b_chat_int4,
     'qwen/Qwen1.5-14B-Chat-GPTQ-Int4',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     requires=['auto_gptq>=0.5', 'transformers>=4.37'],
     torch_dtype=torch.float16,
     function_kwargs={'bits': 4},
@@ -1158,7 +1158,7 @@ def get_model_tokenizer_with_flash_attn(model_dir: str,
     ModelType.qwen1half_14b_chat_int8,
     'qwen/Qwen1.5-14B-Chat-GPTQ-Int8',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     requires=['auto_gptq>=0.5', 'transformers>=4.37'],
     torch_dtype=torch.float16,
     function_kwargs={'bits': 8},
@@ -1168,7 +1168,7 @@ def get_model_tokenizer_with_flash_attn(model_dir: str,
     ModelType.qwen1half_72b_chat_int4,
     'qwen/Qwen1.5-72B-Chat-GPTQ-Int4',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     requires=['auto_gptq>=0.5', 'transformers>=4.37'],
     torch_dtype=torch.float16,
     function_kwargs={'bits': 4},
@@ -1178,7 +1178,7 @@ def get_model_tokenizer_with_flash_attn(model_dir: str,
     ModelType.qwen1half_72b_chat_int8,
     'qwen/Qwen1.5-72B-Chat-GPTQ-Int8',
     LoRATM.qwen1half,
-    TemplateType.chatml,
+    TemplateType.qwen,
     requires=['auto_gptq>=0.5', 'transformers>=4.37'],
     torch_dtype=torch.float16,
     function_kwargs={'bits': 8},