Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/ssbuild/llm_finetuning
Large language Model fintuning bloom , opt , gpt, gpt2 ,llama,llama-2,cpmant and so on
https://github.com/ssbuild/llm_finetuning
adalora bloom cpmant gpt gpt2 llama llama2 lora mistral opt qlora
Last synced: 2 months ago
JSON representation
Large language Model fintuning bloom , opt , gpt, gpt2 ,llama,llama-2,cpmant and so on
- Host: GitHub
- URL: https://github.com/ssbuild/llm_finetuning
- Owner: ssbuild
- Created: 2023-04-02T16:44:14.000Z (almost 2 years ago)
- Default Branch: dev
- Last Pushed: 2024-04-24T16:02:26.000Z (9 months ago)
- Last Synced: 2024-11-08T21:40:45.434Z (2 months ago)
- Topics: adalora, bloom, cpmant, gpt, gpt2, llama, llama2, lora, mistral, opt, qlora
- Language: Python
- Homepage:
- Size: 325 KB
- Stars: 97
- Watchers: 2
- Forks: 12
- Open Issues: 4
-
Metadata Files:
- Readme: README.MD
Awesome Lists containing this project
README
## update information
- [deep_training](https://github.com/ssbuild/deep_training)
```text
2024-04-23 support qwen2
2024-04-22 简化配置
2023-11-27 yi modle_type change to llama
2023-11-15 support load custom model , only modify config/constant_map.py
2023-10-09 support accelerator trainer
2023-10-07 support colossalai trainer
2023-09-26 support transformers trainer
2023-08-16 推理可选使用 Rope NtkScale , 不训练扩展推理长度
2023-08-02 增加 muti lora infer 例子, 手动升级 aigc_zoo , pip install -U git+https://github.com/ssbuild/deep_training.zoo.git --force-reinstall --no-deps
2023-06-13 fix llama resize_token_embeddings
2023-06-01 support deepspeed training for lora adalora prompt,0.1.9 和 0.1.10合并
2023-05-27 add qlora transformers>=4.30
2023-05-24 fix p-tuning-v2 load weight bugs
2023-05-12 fix lora int8 多卡训练 , ppo training move to https://github.com/ssbuild/rlhf_llm
2023-05-02 增加p-tuning-v2
2023-04-28 deep_training 0.1.3 pytorch-lightning 改名 ligntning ，旧版本 deep_training <= 0.1.2
2023-04-23 增加lora merge权重（修改infer_lora_finetuning.py enable_merge_weight 选项）
2023-04-11 升级 lora , 增加adalora
```
## install
- pip install -U -r requirements.txt
- 如果无法安装， 可以切换官方源 pip install -i https://pypi.org/simple -U -r requirements.txt
```text
# flash-attention对显卡算例要求算力7.5 以上 ， 下面可选安装 ，如果卡不支持可以不安装。
git clone -b https://github.com/Dao-AILab/flash-attention
cd flash-attention && pip install .
pip install csrc/layer_norm
pip install csrc/rotary
```
## weigtht select one is suitable for you
支持且不限于以下权重
- [Qwen1.5-1.8B-Chat](https://huggingface.co/Qwen/Qwen1.5-1.8B-Chat)
- [Qwen1.5-7B-Chat](https://huggingface.co/Qwen/Qwen1.5-7B-Chat)
- [Qwen1.5-14B-Chat](https://huggingface.co/Qwen/Qwen1.5-14B-Chat)
- [Qwen1.5-32B-Chat](https://huggingface.co/Qwen/Qwen1.5-32B-Chat)
- [zephyr-7b-beta](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta)
- [mistral-7b-sft-beta](https://huggingface.co/HuggingFaceH4/mistral-7b-sft-beta)
- [Yi-6B](https://huggingface.co/01-ai/Yi-6B)
- [Yi-6B-200K](https://huggingface.co/01-ai/Yi-6B-200K)
- [Yi-34B](https://huggingface.co/01-ai/Yi-34B)
- [Yi-34B-200K](https://huggingface.co/01-ai/Yi-34B-200K)
- [Yi-34B-Chat](https://huggingface.co/01-ai/Yi-34B-Chat)
- [LingoWhale-8B](https://www.modelscope.cn/models/DeepLang/LingoWhale-8B)
- [CausalLM-14B](https://huggingface.co/CausalLM/14B)
- [CausalLM-7B](https://huggingface.co/CausalLM/7B)
- [BlueLM-7B-Chat](https://huggingface.co/vivo-ai/BlueLM-7B-Chat)
- [BlueLM-7B-Chat-32K](https://huggingface.co/vivo-ai/BlueLM-7B-Chat-32K)
- [BlueLM-7B-Base](https://huggingface.co/vivo-ai/BlueLM-7B-Base)
- [BlueLM-7B-Base-32K](https://huggingface.co/vivo-ai/BlueLM-7B-Base-32K)
- [XVERSE-13B-Chat](https://huggingface.co/xverse/XVERSE-13B-Chat)
- [xverse-13b-chat-int4](https://huggingface.co/ssbuild/xverse-13b-chat-int4)
- [XVERSE-13B](https://huggingface.co/xverse/XVERSE-13B)
- [xverse-13b-int4](https://huggingface.co/ssbuild/xverse-13b-int4)
- [Skywork-13B-base](https://huggingface.co/Skywork/Skywork-13B-base)
- [internlm-chat-20b](https://huggingface.co/internlm/internlm-chat-20b)
- [internlm-20b](https://huggingface.co/internlm/internlm-20b)
- [internlm-chat-7b](https://huggingface.co/internlm/internlm-chat-7b)
- [internlm-chat-7b-8k](https://huggingface.co/internlm/internlm-chat-7b-8k)
- [internlm-7b](https://huggingface.co/internlm/internlm-7b)
- [internlm-chat-7b-int4](https://huggingface.co/ssbuild/internlm-chat-7b-int4)
- [bloom预训练模型](https://huggingface.co/bigscience)
- [bloom第三方中文训练模型](https://huggingface.co/Langboat/bloom-6b4-zh) # 注意 需要修改tokenizer_config.json BloomTokenizer -> BloomTokenizerFast
- [tigerbot](https://huggingface.co/TigerResearch)
- [opt预训练模型](https://huggingface.co/facebook)
- [llama 官方权重转换](https://huggingface.co/decapoda-research) # llama 词典等下载地址 https://huggingface.co/hf-internal-testing/llama-tokenizer
- [llama vicuna-7B第三方权重1](https://huggingface.co/TheBloke/vicuna-7B-1.1-HF)
- [llama vicuna-7B第三方权重2](https://huggingface.co/Tribbiani/vicuna-7b)
- [cpm-ant-10b](https://huggingface.co/openbmb/cpm-ant-10b)
- [rwkv](https://huggingface.co/BlinkDL) 需要转换权重
- [rwkv](https://huggingface.co/RWKV/rwkv-4-169m-pile) hf 权重
- [Llama2-Chinese-7b-Chat](https://huggingface.co/FlagAlpha/Llama2-Chinese-7b-Chat)
- [Llama2-Chinese-13b-Chat](https://huggingface.co/FlagAlpha/Llama2-Chinese-13b-Chat)
- [TransGPT](https://huggingface.co/DUOMO-Lab/TransGPT-v0) Llama2中文权重
- [tigerbot-13b-chat](https://huggingface.co/TigerResearch/tigerbot-13b-chat) Llama2中文权重
## data sample
- [open_data 不定时更新数据集](https://github.com/ssbuild/open_data)
单条数据示例
```text
role one of user system function assistant
q question optional
a answer must
```
```json
{"id": 1, "paragraph": [{"q": "从南京到上海的路线", "a": ["你好，南京到上海的路线如下：", "1. 南京到上海，可以乘坐南京地铁1号线，在南京站乘坐轨道交通1号线。", "2. 南京到浦东机场，可以搭乘上海地铁1号，在陆家嘴站乘坐地铁1线，在浦东国际机场站乘坐机场快线，前往上海浦东国际机场。", "3. 上海到南京，可以换乘上海地铁2号线，从南京站换乘地铁2线，再从南京南站换乘地铁1路，然后到达上海站"]}]}
{"id": 2, "paragraph": [{"q": "写一个诗歌，关于冬天", "a": ["冬夜寂静冷，", "云在天边飘，", "冰封白雪上， ", "寒冷像一场雪。", " ", "雪花融化成冰，", "像那雪花飘洒，", "在寒冷的冬天，", "感受春天的喜悦。", " 冬日里，", "风雪渐消，", "一片寂静，", "把快乐和温暖带回家。"]}]}
{"id": 3, "paragraph": [{"q": "晚上睡不着应该怎么办", "a": ["晚上睡不着可能会让你感到焦虑或不舒服,但以下是一些可以帮助你入睡的方法:", "", "1. 制定规律的睡眠时间表:保持规律的睡眠时间表可以帮助你建立健康的睡眠习惯,使你更容易入睡。尽量在每天的相同时间上床,并在同一时间起床。", "2. 创造一个舒适的睡眠环境:确保睡眠环境舒适,安静,黑暗且温度适宜。可以使用舒适的床上用品,并保持房间通风。", "3. 放松身心:在睡前做些放松的活动,例如泡个热水澡,听些轻柔的音乐,阅读一些有趣的书籍等,有助于缓解紧张和焦虑,使你更容易入睡。", "4. 避免饮用含有咖啡因的饮料:咖啡因是一种刺激性物质,会影响你的睡眠质量。尽量避免在睡前饮用含有咖啡因的饮料,例如咖啡,茶和可乐。", "5. 避免在床上做与睡眠无关的事情:在床上做些与睡眠无关的事情,例如看电影,玩游戏或工作等,可能会干扰你的睡眠。", "6. 尝试呼吸技巧:深呼吸是一种放松技巧,可以帮助你缓解紧张和焦虑,使你更容易入睡。试着慢慢吸气,保持几秒钟,然后缓慢呼气。", "", "如果这些方法无法帮助你入睡,你可以考虑咨询医生或睡眠专家,寻求进一步的建议。"]}]}
```
或者
```json
{"id": 1, "conversations": [{"from": "user", "value": "从南京到上海的路线"}, {"from": "assistant", "value": ["你好，南京到上海的路线如下：", "1. 南京到上海，可以乘坐南京地铁1号线，在南京站乘坐轨道交通1号线。", "2. 南京到浦东机场，可以搭乘上海地铁1号，在陆家嘴站乘坐地铁1线，在浦东国际机场站乘坐机场快线，前往上海浦东国际机场。", "3. 上海到南京，可以换乘上海地铁2号线，从南京站换乘地铁2线，再从南京南站换乘地铁1路，然后到达上海站"]}]}
{"id": 2, "conversations": [{"from": "user", "value": "写一个诗歌，关于冬天"}, {"from": "assistant", "value": ["冬夜寂静冷，", "云在天边飘，", "冰封白雪上， ", "寒冷像一场雪。", " ", "雪花融化成冰，", "像那雪花飘洒，", "在寒冷的冬天，", "感受春天的喜悦。", " 冬日里，", "风雪渐消，", "一片寂静，", "把快乐和温暖带回家。"]}]}
{"id": 3, "conversations": [{"from": "user", "value": "晚上睡不着应该怎么办"}, {"from": "assistant", "value": ["晚上睡不着可能会让你感到焦虑或不舒服,但以下是一些可以帮助你入睡的方法:", "", "1. 制定规律的睡眠时间表:保持规律的睡眠时间表可以帮助你建立健康的睡眠习惯,使你更容易入睡。尽量在每天的相同时间上床,并在同一时间起床。", "2. 创造一个舒适的睡眠环境:确保睡眠环境舒适,安静,黑暗且温度适宜。可以使用舒适的床上用品,并保持房间通风。", "3. 放松身心:在睡前做些放松的活动,例如泡个热水澡,听些轻柔的音乐,阅读一些有趣的书籍等,有助于缓解紧张和焦虑,使你更容易入睡。", "4. 避免饮用含有咖啡因的饮料:咖啡因是一种刺激性物质,会影响你的睡眠质量。尽量避免在睡前饮用含有咖啡因的饮料,例如咖啡,茶和可乐。", "5. 避免在床上做与睡眠无关的事情:在床上做些与睡眠无关的事情,例如看电影,玩游戏或工作等,可能会干扰你的睡眠。", "6. 尝试呼吸技巧:深呼吸是一种放松技巧,可以帮助你缓解紧张和焦虑,使你更容易入睡。试着慢慢吸气,保持几秒钟,然后缓慢呼气。", "", "如果这些方法无法帮助你入睡,你可以考虑咨询医生或睡眠专家,寻求进一步的建议。"]}]}
```
## infer
# infer_finetuning.py 推理微调模型
# infer_lora_finetuning.py 推理微调模型
# infer_ptuning.py 推理p-tuning-v2微调模型
python infer_finetuning.py
## training
```text
# 制作数据
cd scripts
bash train_full.sh -m dataset
or
bash train_lora.sh -m dataset
or
bash train_ptv2.sh -m dataset
注: num_process_worker 为多进程制作数据 ， 如果数据量较大 ， 适当调大至cpu数量
dataHelper.make_dataset_with_args(data_args.train_file,mixed_data=False, shuffle=True,mode='train',num_process_worker=0)
# 全参数训练
bash train_full.sh -m train
# lora adalora ia3
bash train_lora.sh -m train
# ptv2
bash train_ptv2.sh -m train
```
## 训练参数
[训练参数](args.MD)
## 友情链接
- [pytorch-task-example](https://github.com/ssbuild/pytorch-task-example)
- [moss_finetuning](https://github.com/ssbuild/moss_finetuning)
- [chatglm_finetuning](https://github.com/ssbuild/chatglm_finetuning)
- [chatglm2_finetuning](https://github.com/ssbuild/chatglm2_finetuning)
- [chatglm3_finetuning](https://github.com/ssbuild/chatglm3_finetuning)
- [t5_finetuning](https://github.com/ssbuild/t5_finetuning)
- [llm_finetuning](https://github.com/ssbuild/llm_finetuning)
- [llm_rlhf](https://github.com/ssbuild/llm_rlhf)
- [chatglm_rlhf](https://github.com/ssbuild/chatglm_rlhf)
- [t5_rlhf](https://github.com/ssbuild/t5_rlhf)
- [rwkv_finetuning](https://github.com/ssbuild/rwkv_finetuning)
- [baichuan_finetuning](https://github.com/ssbuild/baichuan_finetuning)
- [xverse_finetuning](https://github.com/ssbuild/xverse_finetuning)
- [internlm_finetuning](https://github.com/ssbuild/internlm_finetuning)
- [qwen_finetuning](https://github.com/ssbuild/qwen_finetuning)
- [skywork_finetuning](https://github.com/ssbuild/skywork_finetuning)
- [bluelm_finetuning](https://github.com/ssbuild/bluelm_finetuning)
- [yi_finetuning](https://github.com/ssbuild/yi_finetuning)
##
纯粹而干净的代码
## Star History
[](https://star-history.com/#ssbuild/llm_finetuning&Date)