Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/ssbuild/chatglm3_finetuning
https://github.com/ssbuild/chatglm3_finetuning
chatglm3 qlora
Last synced: 4 days ago
JSON representation
- Host: GitHub
- URL: https://github.com/ssbuild/chatglm3_finetuning
- Owner: ssbuild
- License: apache-2.0
- Created: 2023-10-27T09:15:00.000Z (about 1 year ago)
- Default Branch: dev
- Last Pushed: 2024-04-21T16:58:15.000Z (7 months ago)
- Last Synced: 2024-04-28T04:59:05.309Z (7 months ago)
- Topics: chatglm3, qlora
- Language: Python
- Homepage:
- Size: 318 KB
- Stars: 33
- Watchers: 1
- Forks: 2
- Open Issues: 1
-
Metadata Files:
- Readme: README.MD
- License: LICENSE
Awesome Lists containing this project
README
### 天妒英才!人民的好总理,一路走好 !
```text
2024-04-22 简化
2023-10-29 train and test pass
2023-10-28 initial chatglm3
```
## statement
- [deep_training](https://github.com/ssbuild/deep_training)## 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
dev 通过一下方式安装
pip install -U git+https://github.com/ssbuild/deep_training.git
pip install -U transformers>=4.30 deepspeed xformers bitsandbytes>=0.39 accelerate>=0.20
```## weight
- [chatglm3-6b](https://huggingface.co/THUDM/chatglm3-6b) 支持四种微调方式
- [chatglm3-6b-int4](https://huggingface.co/THUDM/chatglm3-6b-int4) 支持四种微调方式
- [chatglm3-6b-32k](https://huggingface.co/THUDM/chatglm3-6b-32k) 支持四种微调方式
- [chatglm3-6b-32k-int4](https://huggingface.co/ssbuild/chatglm3-6b-32k-int4/tree/main) 支持四种微调方式## data sample
- open_data 不定时开放新数据 https://github.com/ssbuild/open_data
- [prompt](assets/PROMPT.md)
- [tools prompt](assets/TOOLS.md)```text
支持两种数据格式示例如下
例子依次分别是 对话,工具,代码执行,对话,对话数组组成
role: 可选字段(str) 标志 q字段 角色, one of user system, observation ,system 标识是否为 system prompt , system prompt a 空
q: 问题
a: 回答注意事项:
a字段:对于工具,代码执行 格式为{metadata}\n{content} , metadata 一般为函数名或者 interpreter,工具函数例如: 函数名\n{content} 或者 代码执行: interpreter\n{content}
详细参考(assets/PROMPT.md),(assets/TOOLS.md)
a字段:对于普通聊天,a即为回答。
``````json
{"id": 1, "paragraph": [{"role": "system", "q": "You are ChatGLM3, a large language model trained by Zhipu.AI. Follow the user's instructions carefully. Respond using markdown.", "a": ""}, {"role": "user", "q": "Hello", "a": "Hello, I'm ChatGLM3. What can I assist you today?"}]}
{"id": 2, "paragraph": [{"role": "system", "q": "Answer the following questions as best as you can. You have access to the following tools:\n[\n {\n \"name\": \"get_current_weather\",\n \"description\": \"Get the current weather in a given location\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"location\": {\n \"type\": \"string\",\n \"description\": \"The city and state, e.g. San Francisco, CA\",\n },\n \"unit\": {\"type\": \"string\"},\n },\n \"required\": [\"location\"],\n },\n }\n]\n", "a": ""}, {"role": "user", "q": "今天北京的天气怎么样?", "a": "好的,让我们来查看今天的天气\n<|assistant|>get_current_weather\n```python\ntool_call(location=\"beijing\", unit=\"celsius\")\n```\n"}, {"role": "observation", "q": "{\"temperature\": 22}", "a": "根据查询结果,今天北京的气温为 22 摄氏度。"}]}
{"id": 3, "paragraph": [{"role": "system", "q": "你是一位智能AI助手,你叫ChatGLM3,你连接着一台电脑,但请注意不能联网。在使用Python解决任务时,你可以运行代码并得到结果,如果运行结果有错误,你需要尽可能对代码进行改进。你可以处理用户上传到电脑上的文件,文件默认存储路径是/mnt/data/", "a": ""}, {"role": "user", "q": "#File: /mnt/data/metadata.jsonl\n#Size: 35380\n#File uploaded\n文件中是否存在缺失值或异常值?", "a": "首先,我将读取您提供的文件,并查看其内容以确定是否存在缺失值或异常值。\n\n我们从加载文件开始:\n<|assistant|>interpreter\n```python\nimport json\n\n# Load the uploaded file\ndata = []\nwith open(\"/mnt/data/metadata.jsonl\", 'r') as file:\n for line in file:\n data.append(json.loads(line))\n\n# Display the first few records to get a sense of the data\ndata[:5]\n```\n"}, {"role": "observation", "q": "\n[{'file_name': 'aceinthehole.png',\n 'name': 'Ace in the Hole',\n 'type': 'survivor',\n 'description': 'Lady Luck always seems to be throwing something good your way.'},\n {'file_name': 'adrenaline.png',\n 'name': 'Adrenaline',\n 'type': 'survivor',\n 'description': 'You are fuelled by unexpected energy when on the verge of escape.'},\n {'file_name': 'aftercare.png',\n 'name': 'Aftercare',\n 'type': 'survivor',\n 'description': \"Unlocks potential in one's aura reading ability.\"},\n {'file_name': 'agitation.png',\n 'name': 'Agitation',\n 'type': 'killer',\n 'description': 'You get excited in anticipation of hooking your prey.'},\n {'file_name': 'alert.png',\n 'name': 'Alert',\n 'type': 'survivor',\n 'description': 'Your acute senses are on high alert.'}]\n ", "a": "\n该文件看起来包含有关某些条目的元数据,每个条目有以下字段:\n\n- `file_name`: 文件名称\n- `name`: 名称\n- `type`: 类型(例如 \"survivor\" 或 \"killer\")\n- `description`: 描述\n\n接下来,我将检查这些字段是否存在缺失值。\n<|assistant|>interpreter\n```python\n# Check for missing values in each column\nmissing_values = {key: sum(1 for item in data if key not in item) for key in data[0].keys()}\n\nmissing_values\n```\n"}, {"role": "observation", "q": "{'file_name': 0, 'name': 0, 'type': 0, 'description': 0}", "a": "\n每个字段都没有缺失值。这意味着每个条目都完整。\n\n接下来,我们可以检查是否存在任何潜在的异常值。这通常需要更深入的分析和对数据的了解,但我可以从一些基本的统计和分布开始进行检查。为了这样做,我们可以查看“type”字段的值分布,看看是否有任何意外的类别。\n<|assistant|>interpreter\n```python\n# Check the distribution of values in the 'type' column\ntype_distribution = {key: sum(1 for item in data if item['type'] == key) for key in set(item['type'] for item in data)}\n\ntype_distribution\n "}, {"role": "observation", "q": "{'survivor': 116, 'killer': 103}", "a": "\n“type”字段有两个值:“survivor”和“killer”。其中,“survivor”有116个条目,而“killer”有103个条目。\n\n从这个初步的检查中,我们没有发现任何明显的异常值。\n\n如果您有任何特定的标准或其他字段需要检查,我可以为您提供更详细的分析。您还希望我检查什么呢?\n "}, {"role": "user", "q": "帮我画一个爱心", "a": "interpreter\n```python\nimport numpy as np\nimport matplotlib.pyplot as plt\n\n# 爱心形状的参数方程\ndef heart(t):\n x = 16 * np.sin(t) ** 3\n y = 13 * np.cos(t) - 5 * np.cos(2 * t) - 2 * np.cos(3 * t) - np.cos(4 * t)\n return x, y\n\nt = np.linspace(0, 2 * np.pi, 1000)\nx, y = heart(t)\n\nplt.figure(figsize=(6, 6))\nplt.plot(x, y, color='red')\nplt.axis('equal')\nplt.axis('off')\nplt.show()\n```\n"}, {"role": "observation", "q": "【image】", "a": "这是一个爱心形状。我使用了参数方程来描述这个形状,并使用matplotlib进行了绘制。如果您有任何其他需求或问题,请随时告诉我。"}]}
{"id": 4, "paragraph": [{"role": "user", "q": "写一个诗歌,关于冬天", "a": "冬夜寂静冷,\n云在天边飘,\n冰封白雪上, \n寒冷像一场雪。\n \n雪花融化成冰,\n像那雪花飘洒,\n在寒冷的冬天,\n感受春天的喜悦。\n 冬日里,\n风雪渐消,\n一片寂静,\n把快乐和温暖带回家。"}]}
{"id": 5, "paragraph": [{"role": "user", "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:\n\n1. 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。\n2. 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。\n3. 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。\n4. 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。\n5. 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。\n6. 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。\n\n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": "system", "value": "You are ChatGLM3, a large language model trained by Zhipu.AI. Follow the user's instructions carefully. Respond using markdown."}, {"from": "user", "value": "Hello"}, {"from": "assistant", "value": "Hello, I'm ChatGLM3. What can I assist you today?"}]}
{"id": 2, "conversations": [{"from": "system", "value": "Answer the following questions as best as you can. You have access to the following tools:\n[\n {\n \"name\": \"get_current_weather\",\n \"description\": \"Get the current weather in a given location\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"location\": {\n \"type\": \"string\",\n \"description\": \"The city and state, e.g. San Francisco, CA\",\n },\n \"unit\": {\"type\": \"string\"},\n },\n \"required\": [\"location\"],\n },\n }\n]\n"}, {"from": "user", "value": "今天北京的天气怎么样?"}, {"from": "assistant", "value": "好的,让我们来查看今天的天气\n<|assistant|>get_current_weather\n```python\ntool_call(location=\"beijing\", unit=\"celsius\")\n```\n"}, {"from": "observation", "value": "{\"temperature\": 22}"}, {"from": "assistant", "value": "根据查询结果,今天北京的气温为 22 摄氏度。"}]}
{"id": 3, "conversations": [{"from": "system", "value": "你是一位智能AI助手,你叫ChatGLM3,你连接着一台电脑,但请注意不能联网。在使用Python解决任务时,你可以运行代码并得到结果,如果运行结果有错误,你需要尽可能对代码进行改进。你可以处理用户上传到电脑上的文件,文件默认存储路径是/mnt/data/"}, {"from": "user", "value": "#File: /mnt/data/metadata.jsonl\n#Size: 35380\n#File uploaded\n文件中是否存在缺失值或异常值?"}, {"from": "assistant", "value": "首先,我将读取您提供的文件,并查看其内容以确定是否存在缺失值或异常值。\n\n我们从加载文件开始:\n<|assistant|>interpreter\n```python\nimport json\n\n# Load the uploaded file\ndata = []\nwith open(\"/mnt/data/metadata.jsonl\", 'r') as file:\n for line in file:\n data.append(json.loads(line))\n\n# Display the first few records to get a sense of the data\ndata[:5]\n```\n"}, {"from": "observation", "value": "\n[{'file_name': 'aceinthehole.png',\n 'name': 'Ace in the Hole',\n 'type': 'survivor',\n 'description': 'Lady Luck always seems to be throwing something good your way.'},\n {'file_name': 'adrenaline.png',\n 'name': 'Adrenaline',\n 'type': 'survivor',\n 'description': 'You are fuelled by unexpected energy when on the verge of escape.'},\n {'file_name': 'aftercare.png',\n 'name': 'Aftercare',\n 'type': 'survivor',\n 'description': \"Unlocks potential in one's aura reading ability.\"},\n {'file_name': 'agitation.png',\n 'name': 'Agitation',\n 'type': 'killer',\n 'description': 'You get excited in anticipation of hooking your prey.'},\n {'file_name': 'alert.png',\n 'name': 'Alert',\n 'type': 'survivor',\n 'description': 'Your acute senses are on high alert.'}]\n "}, {"from": "assistant", "value": "\n该文件看起来包含有关某些条目的元数据,每个条目有以下字段:\n\n- `file_name`: 文件名称\n- `name`: 名称\n- `type`: 类型(例如 \"survivor\" 或 \"killer\")\n- `description`: 描述\n\n接下来,我将检查这些字段是否存在缺失值。\n<|assistant|>interpreter\n```python\n# Check for missing values in each column\nmissing_values = {key: sum(1 for item in data if key not in item) for key in data[0].keys()}\n\nmissing_values\n```\n"}, {"from": "observation", "value": "{'file_name': 0, 'name': 0, 'type': 0, 'description': 0}"}, {"from": "assistant", "value": "\n每个字段都没有缺失值。这意味着每个条目都完整。\n\n接下来,我们可以检查是否存在任何潜在的异常值。这通常需要更深入的分析和对数据的了解,但我可以从一些基本的统计和分布开始进行检查。为了这样做,我们可以查看“type”字段的值分布,看看是否有任何意外的类别。\n<|assistant|>interpreter\n```python\n# Check the distribution of values in the 'type' column\ntype_distribution = {key: sum(1 for item in data if item['type'] == key) for key in set(item['type'] for item in data)}\n\ntype_distribution\n "}, {"from": "observation", "value": "{'survivor': 116, 'killer': 103}"}, {"from": "assistant", "value": "\n“type”字段有两个值:“survivor”和“killer”。其中,“survivor”有116个条目,而“killer”有103个条目。\n\n从这个初步的检查中,我们没有发现任何明显的异常值。\n\n如果您有任何特定的标准或其他字段需要检查,我可以为您提供更详细的分析。您还希望我检查什么呢?\n "}, {"from": "user", "value": "帮我画一个爱心"}, {"from": "assistant", "value": "interpreter\n```python\nimport numpy as np\nimport matplotlib.pyplot as plt\n\n# 爱心形状的参数方程\ndef heart(t):\n x = 16 * np.sin(t) ** 3\n y = 13 * np.cos(t) - 5 * np.cos(2 * t) - 2 * np.cos(3 * t) - np.cos(4 * t)\n return x, y\n\nt = np.linspace(0, 2 * np.pi, 1000)\nx, y = heart(t)\n\nplt.figure(figsize=(6, 6))\nplt.plot(x, y, color='red')\nplt.axis('equal')\nplt.axis('off')\nplt.show()\n```\n"}, {"from": "observation", "value": "【image】"}, {"from": "assistant", "value": "这是一个爱心形状。我使用了参数方程来描述这个形状,并使用matplotlib进行了绘制。如果您有任何其他需求或问题,请随时告诉我。"}]}
{"id": 4, "conversations": [{"from": "user", "value": "写一个诗歌,关于冬天"}, {"from": "assistant", "value": "冬夜寂静冷,\n云在天边飘,\n冰封白雪上, \n寒冷像一场雪。\n \n雪花融化成冰,\n像那雪花飘洒,\n在寒冷的冬天,\n感受春天的喜悦。\n 冬日里,\n风雪渐消,\n一片寂静,\n把快乐和温暖带回家。"}]}
{"id": 5, "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:\n\n1. 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。\n2. 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。\n3. 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。\n4. 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。\n5. 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。\n6. 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。\n\n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。"}]}
```## aigc-serving
部署chatglm3之后 , 可以通过
- [test_chatglm3_tools.py](https://github.com/ssbuild/aigc_serving/blob/main/tests/test_chatglm3_tools.py)
- [test_chatglm3_code.py](https://github.com/ssbuild/aigc_serving/blob/main/tests/test_chatglm3_code.py)体验tools 和 code 功能
## infer
# infer.py 推理预训练模型
# infer_finetuning.py 推理微调模型
# infer_lora_finetuning.py 推理lora微调模型
python infer.py| **量化等级** | **最低 GPU 显存** |
| -------------- | ----------------- |
| FP16(无量化) | 13 GB |
| INT8 | 10 GB |
| INT4 | 6 GB |## 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](assets/args.MD)## 友情链接
- [pytorch-task-example](https://github.com/ssbuild/pytorch-task-example)
- [moss_finetuning](https://github.com/ssbuild/chatmoss_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)
- [aigc_evals](https://github.com/ssbuild/aigc_evals)
- [aigc_serving](https://github.com/ssbuild/aigc_serving)##
纯粹而干净的代码
## Reference
https://huggingface.co/THUDM/chatglm3-6b## Star History
[![Star History Chart](https://api.star-history.com/svg?repos=ssbuild/chatglm3_finetuning&type=Date)](https://star-history.com/#ssbuild/chatglm3_finetuning&Date)