Skip to content

Commit 9c83e13

Browse files
Jintao-HuangJintao-Huang
authored
support gptq export (#485)
1 parent 460e242 commit 9c83e13

File tree

10 files changed

+153
-94
lines changed

10 files changed

+153
-94
lines changed

‎README.md‎

Lines changed: 4 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -66,7 +66,7 @@ Users can check the [documentation of SWIFT](docs/source/GetStarted/快速使用
6666
## 🎉 News
6767
- 🔥2024.02.29: Support [LLaMA PRO](https://arxiv.org/pdf/2401.02415.pdf), use [this script](https://github.com/modelscope/swift/blob/main/examples/pytorch/llm/scripts/yi_6b_chat/llamapro/sft.sh) to begin.
6868
- 🔥2024.02.29: Support [LoRA+](https://arxiv.org/pdf/2402.12354.pdf), use [this script](https://github.com/modelscope/swift/blob/main/examples/pytorch/llm/scripts/yi_6b_chat/lorap/sft.sh) to begin.
69-
- 2024.02.25: Support `swift export` to export models for AWQ quantization and push to ModelScope Hub. For more details, please refer to the document: [LLM Quantization Document](https://github.com/modelscope/swift/blob/main/docs/source/LLM/LLM%E9%87%8F%E5%8C%96%E6%96%87%E6%A1%A3.md).
69+
- 2024.02.25: Support `swift export` to export models for AWQ/GPTQ quantization and push to ModelScope Hub. For more details, please refer to the document: [LLM Quantization Document](https://github.com/modelscope/swift/blob/main/docs/source/LLM/LLM%E9%87%8F%E5%8C%96%E6%96%87%E6%A1%A3.md).
7070
- 2024.02.22: Support gemma series: gemma-2b, [gemma-2b-instruct](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/gemma_2b_instruct), gemma-7b, gemma-7b-instruct.
7171
- 2024.02.16: Support deepseek-math series: deepseek-math-7b, deepseek-math-7b-instruct, deepseek-math-7b-chat.
7272
- 🔥2024.02.05: Support **Qwen1.5** series, 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#%E6%A8%A1%E5%9E%8B). The [qwen1half-7b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen1half_7b_chat), [qwen1half-7b-chat-int8](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen1half_7b_chat_int8) fine-tuned scripts are provided.
@@ -159,7 +159,7 @@ Here is a simple introduction of web-ui:
159159
- Rapidly **fine-tune** and perform inference on LLM, and build a Web-UI, see the [LLM Fine-tuning Documentation](https://github.com/modelscope/swift/blob/main/docs/source/LLM/LLM微调文档.md).
160160
- Using **interface** to fine-tuning and perform inference, see the [WEB-UI Documentation](https://github.com/modelscope/swift/blob/main/docs/source/GetStarted/%E7%95%8C%E9%9D%A2%E8%AE%AD%E7%BB%83%E6%8E%A8%E7%90%86.md).
161161
- **DPO training** supported, see the [DPO Documentation](https://github.com/modelscope/swift/blob/main/docs/source/LLM/LLM%E4%BA%BA%E7%B1%BB%E5%AF%B9%E9%BD%90%E8%AE%AD%E7%BB%83%E6%96%87%E6%A1%A3.md).
162-
- Export fine-tuned models, including: merge-lora, AWQ quantization, and push to ModelScope Hub. For more details, please refer to the [LLM Quantization Documentation](https://github.com/modelscope/swift/blob/main/docs/source/LLM/LLM%E9%87%8F%E5%8C%96%E6%A8%A1%E5%9E%8B.md).
162+
- Export fine-tuned models, including: merge-lora, AWQ/GPTQ quantization, and push to ModelScope Hub. For more details, please refer to the [LLM Quantization Documentation](https://github.com/modelscope/swift/blob/main/docs/source/LLM/LLM%E9%87%8F%E5%8C%96%E6%A8%A1%E5%9E%8B.md).
163163
- Utilize VLLM for **inference acceleration** and **deployment(OpenAI API)**. Please refer to [VLLM Inference Acceleration and Deployment](https://github.com/modelscope/swift/blob/main/docs/source/LLM/VLLM推理加速与部署.md) for more information.
164164
- View the models and datasets supported by Swift. You can check [supported models and datasets](https://github.com/modelscope/swift/blob/main/docs/source/LLM/支持的模型和数据集.md).
165165
- Expand and customize models, datasets, and dialogue templates in Swift, see [Customization and Expansion](https://github.com/modelscope/swift/blob/main/docs/source/LLM/自定义与拓展.md).
@@ -209,10 +209,10 @@ app_ui_main(infer_args)
209209
### Training Scripts
210210
You can refer to the following scripts to customize your own training script.
211211

212-
- full: [qwen1half-7b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen1half_7b_chat/full) (A100)
212+
- full: [qwen1half-7b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen1half_7b_chat/full) (A100), [qwen-7b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_7b_chat/full_mp) (2\*A100)
213213
- full+ddp+zero2: [qwen-7b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_7b_chat/full_ddp_zero2) (4\*A100)
214214
- full+ddp+zero3: [qwen-14b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_14b_chat/full_ddp_zero3) (4\*A100)
215-
- lora: [chatglm3-6b](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/chatglm3_6b/lora) (3090), [yi-34b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/yi_34b_chat/lora) (A100)
215+
- lora: [chatglm3-6b](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/chatglm3_6b/lora) (3090), [yi-34b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/yi_34b_chat/lora) (A100), [qwen-72b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_72b_chat/lora_mp) (2\*A100)
216216
- lora+ddp: [chatglm3-6b](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/chatglm3_6b/lora_ddp) (2\*3090)
217217
- lora+ddp+zero3: [qwen-14b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_14b_chat/lora_ddp_zero3) (4\*3090), [qwen-72b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_72b_chat/lora_ddp_zero3) (4\*A100)
218218
- qlora(gptq-int4): [qwen-7b-chat-int4](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_7b_chat_int4/qlora) (3090)

‎README_CN.md‎

Lines changed: 4 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -64,7 +64,7 @@ SWIFT（Scalable lightWeight Infrastructure for Fine-Tuning）是一个可扩展
6464
## 🎉 新闻
6565
- 🔥2024.02.29: 支持[LLaMA PRO](https://arxiv.org/pdf/2401.02415.pdf), 使用[这个脚本](https://github.com/modelscope/swift/blob/main/examples/pytorch/llm/scripts/yi_6b_chat/llamapro/sft.sh)即可开始训练.
6666
- 🔥2024.02.29: 支持[LoRA+](https://arxiv.org/pdf/2402.12354.pdf), 使用[这个脚本](https://github.com/modelscope/swift/blob/main/examples/pytorch/llm/scripts/yi_6b_chat/lorap/sft.sh)即可开始训练.
67-
- 2024.02.25: 支持`swift export`, 对模型进行AWQ量化导出, 以及推送ModelScope Hub. 具体可以查看文档: [LLM量化文档](https://github.com/modelscope/swift/blob/main/docs/source/LLM/LLM%E9%87%8F%E5%8C%96%E6%96%87%E6%A1%A3.md).
67+
- 2024.02.25: 支持`swift export`, 对模型进行AWQ/GPTQ量化导出, 以及推送ModelScope Hub. 具体可以查看文档: [LLM量化文档](https://github.com/modelscope/swift/blob/main/docs/source/LLM/LLM%E9%87%8F%E5%8C%96%E6%96%87%E6%A1%A3.md).
6868
- 2024.02.22: 支持gemma系列: gemma-2b, [gemma-2b-instruct](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/gemma_2b_instruct), gemma-7b, gemma-7b-instruct.
6969
- 2024.02.16: 支持deepseek-math系列: deepseek-math-7b, deepseek-math-7b-instruct, deepseek-math-7b-chat.
7070
- 🔥2024.02.05: 支持**Qwen1.5**系列模型, 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#%E6%A8%A1%E5%9E%8B). 提供了[qwen1half-7b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen1half_7b_chat), [qwen1half-7b-chat-int8](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen1half_7b_chat_int8)微调的脚本.
@@ -159,7 +159,7 @@ swift web-ui
159159
- 快速对LLM进行**微调**, 推理并搭建Web-UI, 可以查看[LLM微调文档](https://github.com/modelscope/swift/blob/main/docs/source/LLM/LLM微调文档.md).
160160
- 使用**界面**方式进行微调和推理, 可以查看[WEB-UI文档](https://github.com/modelscope/swift/blob/main/docs/source/GetStarted/%E7%95%8C%E9%9D%A2%E8%AE%AD%E7%BB%83%E6%8E%A8%E7%90%86.md).
161161
- 支持**DPO训练**, 可以查看[DPO文档](https://github.com/modelscope/swift/blob/main/docs/source/LLM/LLM%E4%BA%BA%E7%B1%BB%E5%AF%B9%E9%BD%90%E8%AE%AD%E7%BB%83%E6%96%87%E6%A1%A3.md).
162-
- 对微调的模型进行导出, 包括: merge-lora, AWQ量化, 推送ModelScope Hub, 可以查看[LLM量化文档](https://github.com/modelscope/swift/blob/main/docs/source/LLM/LLM%E9%87%8F%E5%8C%96%E6%A8%A1%E5%9E%8B.md).
162+
- 对微调的模型进行导出, 包括: merge-lora, AWQ/GPTQ量化, 推送ModelScope Hub, 可以查看[LLM量化文档](https://github.com/modelscope/swift/blob/main/docs/source/LLM/LLM%E9%87%8F%E5%8C%96%E6%A8%A1%E5%9E%8B.md).
163163
- 使用VLLM进行**推理加速**和**部署(OpenAI API)**. 可以查看[VLLM推理加速与部署](https://github.com/modelscope/swift/blob/main/docs/source/LLM/VLLM推理加速与部署.md).
164164
- 查看swift支持的模型和数据集. 可以查看[支持的模型和数据集](https://github.com/modelscope/swift/blob/main/docs/source/LLM/支持的模型和数据集.md).
165165
- 对swift中的模型, 数据集, 对话模板进行**拓展**, 可以查看[自定义与拓展](https://github.com/modelscope/swift/blob/main/docs/source/LLM/自定义与拓展.md).
@@ -209,10 +209,10 @@ app_ui_main(infer_args)
209209
### 训练脚本
210210
你可以参考以下脚本来自定义属于你的训练脚本.
211211

212-
- full: [qwen1half-7b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen1half_7b_chat/full) (A100)
212+
- full: [qwen1half-7b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen1half_7b_chat/full) (A100), [qwen-7b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_7b_chat/full_mp) (2\*A100)
213213
- full+ddp+zero2: [qwen-7b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_7b_chat/full_ddp_zero2) (4\*A100)
214214
- full+ddp+zero3: [qwen-14b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_14b_chat/full_ddp_zero3) (4\*A100)
215-
- lora: [chatglm3-6b](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/chatglm3_6b/lora) (3090), [yi-34b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/yi_34b_chat/lora) (A100)
215+
- lora: [chatglm3-6b](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/chatglm3_6b/lora) (3090), [yi-34b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/yi_34b_chat/lora) (A100), [qwen-72b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_72b_chat/lora_mp) (2\*A100)
216216
- lora+ddp: [chatglm3-6b](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/chatglm3_6b/lora_ddp) (2\*3090)
217217
- lora+ddp+zero3: [qwen-14b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_14b_chat/lora_ddp_zero3) (4\*3090), [qwen-72b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_72b_chat/lora_ddp_zero3) (4\*A100)
218218
- qlora(gptq-int4): [qwen-7b-chat-int4](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_7b_chat_int4/qlora) (3090)

‎docs/source/LLM/LLM微调文档.md‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -154,10 +154,10 @@ cd examples/pytorch/llm
154154

155155
如果你想要**自定义脚本**, 可以参考以下脚本进行修改: (以下脚本会**定期维护**)
156156

157-
- full: [qwen1half-7b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen1half_7b_chat/full) (A100)
157+
- full: [qwen1half-7b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen1half_7b_chat/full) (A100), [qwen-7b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_7b_chat/full_mp) (2\*A100)
158158
- full+ddp+zero2: [qwen-7b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_7b_chat/full_ddp_zero2) (4\*A100)
159159
- full+ddp+zero3: [qwen-14b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_14b_chat/full_ddp_zero3) (4\*A100)
160-
- lora: [chatglm3-6b](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/chatglm3_6b/lora) (3090), [yi-34b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/yi_34b_chat/lora) (A100)
160+
- lora: [chatglm3-6b](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/chatglm3_6b/lora) (3090), [yi-34b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/yi_34b_chat/lora) (A100), [qwen-72b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_72b_chat/lora_mp) (2\*A100)
161161
- lora+ddp: [chatglm3-6b](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/chatglm3_6b/lora_ddp) (2\*3090)
162162
- lora+ddp+zero3: [qwen-14b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_14b_chat/lora_ddp_zero3) (4\*3090), [qwen-72b-chat](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_72b_chat/lora_ddp_zero3) (4\*A100)
163163
- qlora(gptq-int4): [qwen-7b-chat-int4](https://github.com/modelscope/swift/tree/main/examples/pytorch/llm/scripts/qwen_7b_chat_int4/qlora) (3090)

0 commit comments

Comments
 (0)