Skip to content

GitLab

Commit 7f8094a3 authored by zhaoying1's avatar zhaoying1
Browse files

added baichuan2

parents
Pipeline #612 failed with stages
in 0 seconds
Expand all Hide whitespace changes
Inline Side-by-side
Showing with 1057 additions and 0 deletions
Dockerfile 0 → 100644
View file @ 7f8094a3
FROM image.sourcefind.cn:5000/dcu/admin/base/pytorch:1.13.1-centos7.6-dtk-23.04-py37-latest
COPY requirements.txt requirements.txt
RUN source /opt/dtk-23.04/env.sh
RUN cp /usr/share/zoneinfo/Asia/Shanghai /etc/localtime && echo 'Asia/Shanghai' >/etc/timezone
ENV LANG C.UTF-8
RUN pip install -r requirements.txt -i http://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com
RUN pip install accelerate --no-dependencies -i http://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com
\ No newline at end of file
LICENSE 0 → 100644
View file @ 7f8094a3
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
APPENDIX: How to apply the Apache License to your work.
To apply the Apache License to your work, attach the following
boilerplate notice, with the fields enclosed by brackets "[]"
replaced with your own identifying information. (Don't include
the brackets!) The text should be enclosed in the appropriate
comment syntax for the file format. We also recommend that a
file or class name and description of purpose be included on the
same "printed page" as the copyright notice for easier
identification within third-party archives.
Copyright [yyyy] [name of copyright owner]
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
README.md 0 → 100644
View file @ 7f8094a3
# Baichuan 2
## 论文
`Baichuan 2: Open Large-scale Language Models`
https://arxiv.org/abs/2309.10305
## 模型结构
Baichuan 2 是百川智能推出的新一代开源大语言模型，采用 2.6 万亿Tokens 的高质量语料训练。
模型具体参数：
| 模型名称 | 隐含层维度 | 层数 | 头数 | 词表大小 | 位置编码 | 最大长 |
| -------- | -------- | -------- | -------- | -------- | -------- | -------- |
| Baichuan 2-7B | 4,096 | 32 | 32 | 125,696 | RoPE | 4096 |
| Baichuan 2-13B | 5,120 | 40 | 40 | 125,696 | ALiBi | 4096 |
<div align="center">
<img src="./media/transformer.jpg" width="400" height="300">
</div>
## 算法原理
Baichuan整体模型基于标准的Transformer结构，采用了和LLaMA一样的模型设计。其中，Baichuan-7B在结构上采用Rotary Embedding位置编码方案、SwiGLU激活函数、基于RMSNorm的Pre-Normalization。Baichuan-13B使用了ALiBi线性偏置技术，相对于Rotary Embedding计算量更小，对推理性能有显著提升。
<div align="center">
<img src="./media/transformer.png" width="450" height="300">
</div>
## 环境配置
### Docker(方式一)
推荐使用docker方式运行，提供拉取的docker镜像：
```
docker pull image.sourcefind.cn:5000/dcu/admin/base/pytorch:1.13.1-centos7.6-dtk-23.04-py37-latest
```
安装docker中没有的依赖:
```
pip install transformers==4.28.0 -i http://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com
pip install accelerate --no-dependencies -i http://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com
pip install datasets peft tokenizers sentencepiece numpy -i http://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com
```
### Dockerfile(方式二)
```
docker build -t baichuan2:latest .
docker run -dit --network=host --name=baichuan2 --privileged --device=/dev/kfd --device=/dev/dri --ipc=host --shm-size=16G --group-add video --cap-add=SYS_PTRACE --security-opt seccomp=unconfined -u root --ulimit stack=-1:-1 --ulimit memlock=-1:-1 baichuan2:latest
docker exec -it baichuan2 /bin/bash
```
### Conda(方式三)
1. 创建conda虚拟环境：
```
conda create -n chatglm python=3.7
```
2. 关于本项目DCU显卡所需的工具包、深度学习库等均可从[光合](https://developer.hpccube.com/tool/)开发者社区下载安装。
- [DTK 23.04](https://cancon.hpccube.com:65024/1/main/DTK-23.04.1)
- [Pytorch 1.13.1](https://cancon.hpccube.com:65024/4/main/pytorch/dtk23.04)
- [Deepspeed 0.9.2](https://cancon.hpccube.com:65024/4/main/deepspeed/dtk23.04)
Tips：以上dtk驱动、python、deepspeed等工具版本需要严格一一对应。
3. 其它依赖库参照requirements.txt安装：
```
pip install -r requirements.txt -i http://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com
pip install accelerate --no-dependencies -i http://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com
```
## 数据集
输入数据为放置在项目[fine-tune/data](./fine-tune/data)目录下的 json 文件，`fine-tune/data/belle_chat_ramdon_10k.json`，该样例数据是从 [multiturn_chat_0.8M](https://huggingface.co/datasets/BelleGroup/multiturn_chat_0.8M) 采样出 1 万条，并且做了格式转换。主要是展示多轮数据怎么训练，不保证效果。json 文件示例格式如下：
```
[
{
"id": "77771",
"conversations": [
{
"from": "human",
"value": "写一篇关于气候变化对海洋生态的影响的文章。\n"
},
{
"from": "gpt",
"value": "好的，以下是你要求的文章：\n气候变化对海洋生态的影响\n气候变化对海洋生态产生了广泛的影响。全球变暖导致海洋温度上升，这可能会对许多水生生物的生存、繁殖和迁移造成巨大的影响。另外，海洋酸化也是一个问题，因为大气中二氧化碳的增加导致海水中的pH值下降，使得许多生物的外骨骼变得脆弱，从而影响它们的生存和繁殖。\n此外，气候变化还会引起海平面上升，这不仅对海岸线和沿海居民产生影响，还会对许多海洋生物的栖息地造成影响。一些岛屿上的陆地生物面临几乎完全消失的风险，而另一些生物范围则会扩大到更广泛的领域，这可能会导致新的竞争和生态平衡的改变。\n总之，气候变化并不只是一个地球气温的问题，它对海洋生态系统产生了广泛和深远的影响。\n"
},
{
"from": "human",
"value": "从刚刚的文章中提取出气候变化对海洋生物繁殖的影响。\n"
},
{
"from": "gpt",
"value": "文章已经提到过气候变化会对海洋生物的生存、繁殖和迁移造成影响。例如，海洋酸化会使得许多生物的外骨骼变得脆弱，进而影响它们的生存和繁殖。此外，海洋温度上升也可能会对很多生物的生存和繁殖产生负面影响。"
}
]
}
....
]
```
## 模型下载
Hugging Face模型下载地址：
[Baichuan2-7B-Base](https://huggingface.co/baichuan-inc/Baichuan2-7B-Base)
[Baichuan2-7B-Chat](https://huggingface.co/baichuan-inc/Baichuan2-7B-Chat)
[Baichuan2-13B-Chat](https://huggingface.co/baichuan-inc/Baichuan2-13B-Chat)
[Baichuan2-13B-Base](https://huggingface.co/baichuan-inc/Baichuan2-13B-Base)
## 训练
### 全参数微调训练
1. 单机训练
```
cd fine-tune
bash ft_train.sh
```
2. 多机训练
```
cd fine-tune/multi_node
```
进入节点1，根据环境修改hostfile，保证两节点文件路径一致，配置相同，按需修改run-13b-sft.sh中--mca btl_tcp_if_include enp97s0f1，enp97s0f1改为ip a命令后对应节点ip的网卡名，numa可以根据当前节点拓扑更改绑定，微调命令：
```
bash run_ft.sh
```
### LoRA微调训练
1. 单机训练
```
cd fine-tune
bash run_lora.sh
```
2. 多机训练
```
cd fine-tune/multi_node
```
进入节点1，根据环境修改hostfile，保证两节点文件路径一致，配置相同，按需修改run-13b-sft.sh中--mca btl_tcp_if_include enp97s0f1，enp97s0f1改为ip a命令后对应节点ip的网卡名，numa可以根据当前节点拓扑更改绑定，微调命令：
```
bash run_lora.sh
```
## 推理
### 命令行测试
```bash
python cli_demo.py
```
请根据情况修改其中的模型加载路径。
## Result
- 以下为我们基于baichuan2-7b-base模型进行全参数指令微调实验后的推理效果：
<div align="center">
<img src="./media/baichuan2-test.png" width="500" height="230">
</div>
## 精度
- 以下为我们基于baichuan2-7b-base模型进行全参数指令微调实验的loss收敛情况：
<div align="center">
<img src="./media/baichuan2_7bbase_ft_96c_bs1_acum1_fp16_lr2e-5.jpg" width="300" height="250">
</div>
## 应用场景
### 算法类别
`对话问答`
### 热点应用行业
`医疗,教育,科研,金融`
## 源码仓库及问题反馈
- https://developer.hpccube.com/codes/modelzoo/baichuan2_pytorch
## 参考
- [https://github.com/baichuan-inc/Baichuan2/tree/main](https://github.com/baichuan-inc/Baichuan2/tree/main)
\ No newline at end of file
cli_demo.py 0 → 100644
View file @ 7f8094a3
import os
import torch
import platform
import subprocess
from colorama import Fore, Style
from tempfile import NamedTemporaryFile
from transformers import AutoModelForCausalLM, AutoTokenizer
from transformers.generation.utils import GenerationConfig
def init_model():
print("init model ...")
model = AutoModelForCausalLM.from_pretrained(
"/public/home/zhaoying1/work/Baichuan2-main/fine-tune/slurm_script/output/checkpoint-420",
torch_dtype=torch.float16,
device_map="auto",
trust_remote_code=True
)
model.generation_config = GenerationConfig.from_pretrained(
"/public/home/zhaoying1/work/Baichuan2-main/fine-tune/slurm_script/output/checkpoint-420"
)
print(model.generation_config)
tokenizer = AutoTokenizer.from_pretrained(
"/public/home/zhaoying1/work/Baichuan2-main/fine-tune/slurm_script/output/checkpoint-420",
use_fast=False,
trust_remote_code=True
)
return model, tokenizer
def clear_screen():
if platform.system() == "Windows":
os.system("cls")
else:
os.system("clear")
print(Fore.YELLOW + Style.BRIGHT + "欢迎使用百川大模型，输入进行对话，vim 多行输入，clear 清空历史，CTRL+C 中断生成，stream 开关流式生成，exit 结束。")
return []
def vim_input():
with NamedTemporaryFile() as tempfile:
tempfile.close()
subprocess.call(['vim', '+star', tempfile.name])
text = open(tempfile.name).read()
return text
def main(stream=True):
model, tokenizer = init_model()
messages = clear_screen()
while True:
prompt = input(Fore.GREEN + Style.BRIGHT + "\n用户：" + Style.NORMAL)
if prompt.strip() == "exit":
break
if prompt.strip() == "clear":
messages = clear_screen()
continue
if prompt.strip() == 'vim':
prompt = vim_input()
print(prompt)
print(Fore.CYAN + Style.BRIGHT + "\nBaichuan 2：" + Style.NORMAL, end='')
if prompt.strip() == "stream":
stream = not stream
print(Fore.YELLOW + "({}流式生成)\n".format("开启" if stream else "关闭"), end='')
continue
messages.append({"role": "user", "content": prompt})
if stream:
position = 0
try:
for response in model.chat(tokenizer, messages, stream=True):
print(response[position:], end='', flush=True)
position = len(response)
if torch.backends.mps.is_available():
torch.mps.empty_cache()
except KeyboardInterrupt:
pass
print()
else:
response = model.chat(tokenizer, messages)
print(response)
if torch.backends.mps.is_available():
torch.mps.empty_cache()
messages.append({"role": "assistant", "content": response})
print(Style.RESET_ALL)
if __name__ == "__main__":
main()
fine-tune/data/belle_chat_ramdon_10k.json 0 → 100644
View file @ 7f8094a3
This diff is collapsed.
fine-tune/ds_config.json 0 → 100644
View file @ 7f8094a3
{
"train_micro_batch_size_per_gpu": "auto",
"zero_allow_untested_optimizer": true,
"fp16": {
"enabled": "auto",
"loss_scale": 0,
"initial_scale_power": 16,
"loss_scale_window": 1000,
"hysteresis": 2,
"min_loss_scale": 1
},
"zero_force_ds_cpu_optimizer": false,
"zero_optimization": {
"stage": 3,
"offload_param": {
"device": "cpu",
"pin_memory": true
},
"offload_optimizer": {
"device": "cpu",
"pin_memory": true
},
"stage3_gather_16bit_weights_on_model_save": true,
"allgather_partitions": true,
"allgather_bucket_size": 5e8,
"overlap_comm": false,
"reduce_scatter": true,
"reduce_bucket_size": 5e8,
"contiguous_gradients" : true
}
}
fine-tune/ds_config_zero2.json 0 → 100644
View file @ 7f8094a3
{
"train_micro_batch_size_per_gpu": "auto",
"zero_allow_untested_optimizer": true,
"fp16": {
"enabled": "auto",
"loss_scale": 0,
"initial_scale_power": 16,
"loss_scale_window": 1000,
"hysteresis": 2,
"min_loss_scale": 1
},
"zero_force_ds_cpu_optimizer": false,
"zero_optimization": {
"stage": 2,
"stage3_gather_16bit_weights_on_model_save": true,
"allgather_partitions": true,
"allgather_bucket_size": 5e8,
"overlap_comm": false,
"reduce_scatter": true,
"reduce_bucket_size": 5e8,
"contiguous_gradients" : true
}
}
fine-tune/fine-tune.py 0 → 100644
View file @ 7f8094a3
import os
import math
import pathlib
from typing import Optional, Dict
from dataclasses import dataclass, field
import json
import torch
from torch.utils.data import Dataset
import transformers
from transformers.training_args import TrainingArguments
@dataclass
class ModelArguments:
model_name_or_path: Optional[str] = field(default="baichuan-inc/Baichuan2-7B-Base")
@dataclass
class DataArguments:
data_path: str = field(
default=None, metadata={"help": "Path to the training data."}
)
@dataclass
class TrainingArguments(transformers.TrainingArguments):
cache_dir: Optional[str] = field(default=None)
optim: str = field(default="adamw_torch")
model_max_length: int = field(
default=512,
metadata={
"help": "Maximum sequence length. Sequences will be right padded (and possibly truncated)."
},
)
use_lora: bool = field(default=False)
class SupervisedDataset(Dataset):
"""Dataset for supervised fine-tuning."""
def __init__(
self,
data_path,
tokenizer,
model_max_length,
user_tokens=[195],
assistant_tokens=[196],
):
super(SupervisedDataset, self).__init__()
self.data = json.load(open(data_path))
self.tokenizer = tokenizer
self.model_max_length = model_max_length
self.user_tokens = user_tokens
self.assistant_tokens = assistant_tokens
self.ignore_index = -100
item = self.preprocessing(self.data[0])
print("input:", self.tokenizer.decode(item["input_ids"]))
labels = []
for id_ in item["labels"]:
if id_ == -100:
continue
labels.append(id_)
print("label:", self.tokenizer.decode(labels))
def __len__(self):
return len(self.data)
def preprocessing(self, example):
input_ids = []
labels = []
for message in example["conversations"]:
from_ = message["from"]
value = message["value"]
value_ids = self.tokenizer.encode(value)
if from_ == "human":
input_ids += self.user_tokens + value_ids
labels += [self.tokenizer.eos_token_id] + [self.ignore_index] * len(
value_ids
)
# print("human_input_ids",input_ids)
# print("human_input_ids",labels)
else:
input_ids += self.assistant_tokens + value_ids
labels += [self.ignore_index] + value_ids
# print("gpt_input_ids",input_ids)
# print("gpt_labels",labels)
input_ids.append(self.tokenizer.eos_token_id)
labels.append(self.tokenizer.eos_token_id)
# print("input_ids!!!!",input_ids)
# print("labels!!!",labels)
input_ids = input_ids[: self.model_max_length]
labels = labels[: self.model_max_length]
input_ids += [self.tokenizer.pad_token_id] * (
self.model_max_length - len(input_ids)
)
labels += [self.ignore_index] * (self.model_max_length - len(labels))
input_ids = torch.LongTensor(input_ids)
labels = torch.LongTensor(labels)
attention_mask = input_ids.ne(self.tokenizer.pad_token_id)
return {
"input_ids": input_ids,
"labels": labels,
"attention_mask": attention_mask,
}
def __getitem__(self, idx) -> Dict[str, torch.Tensor]:
return self.preprocessing(self.data[idx])
def train():
parser = transformers.HfArgumentParser(
(ModelArguments, DataArguments, TrainingArguments)
)
model_args, data_args, training_args = parser.parse_args_into_dataclasses()
model = transformers.AutoModelForCausalLM.from_pretrained(
model_args.model_name_or_path,
trust_remote_code=True,
cache_dir=training_args.cache_dir,
)
tokenizer = transformers.AutoTokenizer.from_pretrained(
model_args.model_name_or_path,
use_fast=False,
trust_remote_code=True,
model_max_length=training_args.model_max_length,
cache_dir=training_args.cache_dir,
)
if training_args.use_lora:
from peft import LoraConfig, TaskType, get_peft_model
peft_config = LoraConfig(
task_type=TaskType.CAUSAL_LM,
target_modules=["W_pack"],
inference_mode=False,
r=1,
lora_alpha=32,
lora_dropout=0.1,
)
model.enable_input_require_grads()
model = get_peft_model(model, peft_config)
model.print_trainable_parameters()
dataset = SupervisedDataset(
data_args.data_path, tokenizer, training_args.model_max_length
)
trainer = transformers.Trainer(
model=model, args=training_args, train_dataset=dataset, tokenizer=tokenizer
)
trainer.train()
trainer.save_state()
trainer.save_model(output_dir=training_args.output_dir)
if __name__ == "__main__":
train()
fine-tune/ft_train.sh 0 → 100644
View file @ 7f8094a3
hostfile=""
HIP_VISIBLE_DEVICES=0,1,2,3 deepspeed --hostfile=$hostfile fine-tune.py \
--report_to "none" \
--data_path "data/belle_chat_ramdon_10k.json" \
--model_name_or_path "../baichuan2-7b-base" \
--output_dir "output" \
--model_max_length 512 \
--num_train_epochs 4 \
--per_device_train_batch_size 2 \
--gradient_accumulation_steps 1 \
--save_strategy epoch \
--learning_rate 2e-5 \
--lr_scheduler_type constant \
--adam_beta1 0.9 \
--adam_beta2 0.98 \
--adam_epsilon 1e-8 \
--max_grad_norm 1.0 \
--weight_decay 1e-4 \
--warmup_ratio 0.0 \
--logging_steps 1 \
--gradient_checkpointing True \
--deepspeed ds_config.json \
--fp16
fine-tune/lora_train.sh 0 → 100644
View file @ 7f8094a3
hostfile=""
HIP_VISIBLE_DEVICES=0,1,2,3 deepspeed --hostfile=$hostfile fine-tune.py \
--report_to "none" \
--data_path "data/belle_chat_ramdon_10k.json" \
--model_name_or_path "../baichuan2-7b-base" \
--output_dir "output" \
--model_max_length 512 \
--num_train_epochs 4 \
--per_device_train_batch_size 2 \
--gradient_accumulation_steps 1 \
--save_strategy epoch \
--learning_rate 2e-5 \
--lr_scheduler_type constant \
--adam_beta1 0.9 \
--adam_beta2 0.98 \
--adam_epsilon 1e-8 \
--max_grad_norm 1.0 \
--weight_decay 1e-4 \
--warmup_ratio 0.0 \
--logging_steps 1 \
--gradient_checkpointing True \
--deepspeed ds_config_zero2.json \
--fp16 \
--use_lora True
fine-tune/multi-node/ds_config.json 0 → 100644
View file @ 7f8094a3
{
"train_micro_batch_size_per_gpu": "auto",
"zero_allow_untested_optimizer": true,
"fp16": {
"enabled": "auto",
"loss_scale": 0,
"initial_scale_power": 16,
"loss_scale_window": 1000,
"hysteresis": 2,
"min_loss_scale": 1
},
"zero_force_ds_cpu_optimizer": false,
"zero_optimization": {
"stage": 3,
"offload_param": {
"device": "cpu",
"pin_memory": true
},
"offload_optimizer": {
"device": "cpu",
"pin_memory": true
},
"stage3_gather_16bit_weights_on_model_save": true,
"allgather_partitions": true,
"allgather_bucket_size": 5e8,
"overlap_comm": false,
"reduce_scatter": true,
"reduce_bucket_size": 5e8,
"contiguous_gradients" : true
}
}
fine-tune/multi-node/ds_config_zero2.json 0 → 100644
View file @ 7f8094a3
{
"train_micro_batch_size_per_gpu": "auto",
"zero_allow_untested_optimizer": true,
"fp16": {
"enabled": "auto",
"loss_scale": 0,
"initial_scale_power": 16,
"loss_scale_window": 1000,
"hysteresis": 2,
"min_loss_scale": 1
},
"zero_force_ds_cpu_optimizer": false,
"zero_optimization": {
"stage": 2,
"stage3_gather_16bit_weights_on_model_save": true,
"allgather_partitions": true,
"allgather_bucket_size": 5e8,
"overlap_comm": false,
"reduce_scatter": true,
"reduce_bucket_size": 5e8,
"contiguous_gradients" : true
}
}
fine-tune/multi-node/env.sh 0 → 100644
View file @ 7f8094a3
#!/bin/bash
export ROCM_PATH=/opt/dtk-23.04
export ROCM_SOURCE_DIR=${ROCM_PATH}
echo $ROCM_PATH
export HIP_PATH=${ROCM_PATH}/hip
export AMDGPU_TARGETS="gfx900;gfx906"
export PATH=${ROCM_PATH}/bin:${ROCM_PATH}/llvm/bin:${ROCM_PATH}/hcc/bin:${ROCM_PATH}/hip/bin:$PATH
export LD_LIBRARY_PATH=${ROCM_PATH}/lib:${ROCM_PATH}/lib64:$LD_LIBRARY_PATH
export LD_LIBRARY_PATH=${ROCM_PATH}/hip/lib:${ROCM_PATH}/llvm/lib:${ROCM_PATH}/opencl/lib/x86_64:$LD_LIBRARY_PATH
export C_INCLUDE_PATH=${ROCM_PATH}/include:${ROCM_PATH}/hip/include/hip:${ROCM_PATH}/llvm/include:/opencl/include:${ROCM_PATH}/include/rocrand:${ROCM_PATH}/include/hiprand
export CPLUS_INCLUDE_PATH=${ROCM_PATH}/include:${ROCM_PATH}/hip/include/hip:${ROCM_PATH}/llvm/include:/opencl/include:${ROCM_PATH}/include/rocrand:${ROCM_PATH}/include/hiprand
export PATH=${ROCM_PATH}/miopen/bin:${ROCM_PATH}/rocblas/bin:${ROCM_PATH}/hipsparse/bin:$PATH
export LD_LIBRARY_PATH=${ROCM_PATH}/miopen/lib:${ROCM_PATH}/rocblas/lib:$LD_LIBRARY_PATH
export MIOPEN_SYSTEM_DB_PATH=${ROCM_PATH}/miopen/share/miopen/db/
export LD_LIBRARY_PATH=/usr/lib64:$LD_LIBRARY_PATH
export LIBRARY_PATH=/usr/lib64:$LIBRARY_PATH
export RCCL_PATH=$ROCM_PATH/rccl
export NCCL_PATH=$ROCM_PATH/rccl
export LD_LIBRARY_PATH=$RCCL_PATH/lib:$LD_LIBRARY_PATH
export MIOPEN_FIND_MODE=3
export HSA_FORCE_FINE_GRAIN_PCIE=1
export NCCL_P2P_LEVEL=5
export NCCL_GDR_FLUSH_DISABLE=1
export NCCL_NET_GDR_LEVEL=SYS
export RCCL_NCHANNELS=2
export NCCL_IB_HCA=mlx5
export NCCL_SOCKET_IFNAME=ib0
export NCCL_DEBUG=INFO
export MIOPEN_FIND_MODE=3
export HSA_FORCE_FINE_GRAIN_PCIE=1
export MIOPEN_COMPILE_PARALLEL_LEVEL=1
export NCCL_PLUGIN_P2P=ucx
export HIP_CLANG_PATH=/opt/dtk-23.04/llvm/bin
export HSA_PATH=/opt/dtk-23.04/hsa
export AOMP=/opt/dtk-23.04/llvm
export LD_LIBRARY_PATH=/opt/dtk-23.04/rccl/lib:/usr/lib64:/opt/dtk-23.04/miopen/lib:/opt/dtk-23.04/rocblas/lib:/opt/dtk-23.04/hip/lib:/opt/dtk-23.04/llvm/lib:/opt/dtk-23.04/opencl/lib/x86_64:/opt/dtk-23.04/lib:/opt/dtk-23.04/lib64:/opt/dtk-23.04/rccl/lib:/usr/lib64:/opt/dtk-23.04/miopen/lib:/opt/dtk-23.04/rocblas/lib:/opt/dtk-23.04/hip/lib:/opt/dtk-23.04/llvm/lib:/opt/dtk-23.04/opencl/lib/x86_64:/opt/dtk-23.04/lib:/opt/dtk-23.04/lib64:/opt/dtk-23.04/roctracer/lib:/opt/dtk-23.04/rocthrust/lib:/opt/dtk-23.04/rocsparse/lib:/opt/dtk-23.04/rocsolver/lib:/opt/dtk-23.04/rocrand/lib:/opt/dtk-23.04/rocprofiler/lib:/opt/dtk-23.04/rocprim/lib:/opt/dtk-23.04/dtk-23.04_smi/lib:/opt/dtk-23.04/rocfft/lib:/opt/dtk-23.04/rocblas/lib:/opt/dtk-23.04/rocalution/lib:/opt/dtk-23.04/rccl/lib:/opt/dtk-23.04/opencl/lib:/opt/dtk-23.04/oam/lib:/opt/dtk-23.04/migraphx/lib:/opt/dtk-23.04/miopengemm/lib:/opt/dtk-23.04/miopen/lib:/opt/dtk-23.04/llvm/lib-debug/src/openmp/libomptarget/plugins/remote/lib:/opt/dtk-23.04/llvm/lib/clang/14.0.0/lib:/opt/dtk-23.04/llvm/lib:/opt/dtk-23.04/hsa/lib:/opt/dtk-23.04/hipsparse/lib:/opt/dtk-23.04/hipsolver/lib:/opt/dtk-23.04/hiprand/lib:/opt/dtk-23.04/hipfft/lib:/opt/dtk-23.04/hipcub/lib:/opt/dtk-23.04/hipblas-clients/lib:/opt/dtk-23.04/hipblas/lib:/opt/dtk-23.04/hip/lib:/opt/dtk-23.04/lib:/opt/dtk-23.04/lib64:/opt/mpi/lib:/usr/local/lib/:/usr/local/lib64/:/usr/lib64/
export PATH=/opt/dtk-23.04/miopen/bin:/opt/dtk-23.04/rocblas/bin:/opt/dtk-23.04/hipsparse/bin:/opt/dtk-23.04/bin:/opt/dtk-23.04/llvm/bin:/opt/dtk-23.04/hcc/bin:/opt/dtk-23.04/hip/bin:/opt/dtk-23.04/miopen/bin:/opt/dtk-23.04/rocblas/bin:/opt/dtk-23.04/hipsparse/bin:/opt/dtk-23.04/bin:/opt/dtk-23.04/llvm/bin:/opt/dtk-23.04/hcc/bin:/opt/dtk-23.04/hip/bin:/opt/dtk-23.04/libexec/rocprofiler:/opt/dtk-23.04/libexec/dtk-23.04_smi:/opt/dtk-23.04/rocprofiler/bin:/opt/dtk-23.04/opencl/bin:/opt/dtk-23.04/miopen/bin:/opt/dtk-23.04/llvm/lib/clang/14.0.0/bin:/opt/dtk-23.04/llvm/bin:/opt/dtk-23.04/hip/bin:/opt/dtk-23.04/bin:/opt/mpi/bin:/root/anaconda3/bin:/root/anaconda3/condabin:/usr/lib64/qt-3.3/bin:/root/perl5/bin:/opt/dtk-23.04/bin:/opt/dtk-23.04/hip/bin:/opt/dtk-23.04/llvm/bin:/opt/dtk-23.04/llvm/lib/clang/14.0.0/bin:/opt/dtk-23.04/miopen/bin:/opt/dtk-23.04/opencl/bin:/opt/dtk-23.04/rocprofiler/bin:/opt/dtk-23.04/libexec/dtk-23.04_smi:/opt/dtk-23.04/libexec/rocprofiler:/opt/rh/devtoolset-7/root/usr/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/root/bin
export ROCM_ROOT=/opt/dtk-23.04
export ROCBLAS_TENSILE_LIBPATH=/opt/dtk-23.04/lib/rocblas/library
export HIP_ROCCLR_HOME=/opt/dtk-23.04/hip
export HIP_LIB_PATH=/opt/dtk-23.04/hip/lib
export DEVICE_LIB_PATH=/opt/dtk-23.04/amdgcn/bitcode
\ No newline at end of file
fine-tune/multi-node/hostfile 0 → 100644
View file @ 7f8094a3
10.0.21.163 slots=8
10.0.21.116 slots=8
fine-tune/multi-node/run_ft.sh 0 → 100644
View file @ 7f8094a3
source env.sh
echo "START TIME: $(date)"
hostfile=./hostfile
np=$(cat $hostfile|sort|uniq |wc -l)
np=$(($np*8))
which mpirun
mpirun -np $np --allow-run-as-root --hostfile hostfile --bind-to none --mca btl_tcp_if_include enp97s0f1 run_ft_single.sh 8
echo "END TIME: $(date)"
fine-tune/multi-node/run_ft_single.sh 0 → 100644
View file @ 7f8094a3
#!/bin/bash
source env.sh
GPUS=$1
string=""
for ((i=0; i<$GPUS; i++)); do
string="$string$i,"
done
string=${string%","}
export HIP_VISIBLE_DEVICES=$string
lrank=$OMPI_COMM_WORLD_LOCAL_RANK
RANK=$OMPI_COMM_WORLD_RANK
WORLD_SIZE=$OMPI_COMM_WORLD_SIZE
APP="python3 ../fine-tune.py \
--deepspeed ../ds_config.json \
--report_to "none" \
--data_path "../data/belle_chat_ramdon_10k.json" \
--model_name_or_path "../../baichuan2-7b-base" \
--output_dir "output" \
--model_max_length 64 \
--num_train_epochs 4 \
--per_device_train_batch_size 1 \
--gradient_accumulation_steps 1 \
--save_strategy epoch \
--learning_rate 2e-5 \
--lr_scheduler_type constant \
--adam_beta1 0.9 \
--adam_beta2 0.98 \
--adam_epsilon 1e-8 \
--max_grad_norm 1.0 \
--weight_decay 1e-4 \
--warmup_ratio 0.0 \
--logging_steps 1 \
--gradient_checkpointing False \
--fp16 \
--local_rank $lrank "
case ${lrank} in
[0])
export HIP_VISIBLE_DEVICES=0,1,2,3
export UCX_NET_DEVICES=mlx5_0:1
export UCX_IB_PCI_BW=mlx5_0:50Gbs
numactl --cpunodebind=0 --membind=0 ${APP}
;;
[1])
export HIP_VISIBLE_DEVICES=0,1,2,3
export UCX_NET_DEVICES=mlx5_1:1
export UCX_IB_PCI_BW=mlx5_1:50Gbs
numactl --cpunodebind=1 --membind=1 ${APP}
;;
[2])
export HIP_VISIBLE_DEVICES=0,1,2,3
export UCX_NET_DEVICES=mlx5_2:1
export UCX_IB_PCI_BW=mlx5_2:50Gbs
numactl --cpunodebind=2 --membind=2 ${APP}
;;
[3])
export HIP_VISIBLE_DEVICES=0,1,2,3
export UCX_NET_DEVICES=mlx5_3:1
export UCX_IB_PCI_BW=mlx5_3:50Gbs
numactl --cpunodebind=3 --membind=3 ${APP}
;;
esac
fine-tune/multi-node/run_lora.sh 0 → 100644
View file @ 7f8094a3
source env.sh
echo "START TIME: $(date)"
hostfile=./hostfile
np=$(cat $hostfile|sort|uniq |wc -l)
np=$(($np*8))
which mpirun
mpirun -np $np --allow-run-as-root --hostfile hostfile --bind-to none --mca btl_tcp_if_include enp97s0f1 run_lora_single.sh 8
echo "END TIME: $(date)"
fine-tune/multi-node/run_lora_single.sh 0 → 100644
View file @ 7f8094a3
#!/bin/bash
export HSA_FORCE_FINE_GRAIN_PCIE=1
export MIOPEN_FIND_MODE=3
#!/bin/bash
source env.sh
GPUS=$1
string=""
for ((i=0; i<$GPUS; i++)); do
string="$string$i,"
done
string=${string%","}
export HIP_VISIBLE_DEVICES=$string
lrank=$OMPI_COMM_WORLD_LOCAL_RANK
echo "LRANK===============================$lrank"
RANK=$OMPI_COMM_WORLD_RANK
WORLD_SIZE=$OMPI_COMM_WORLD_SIZE
echo "WORLD_SIZE*************$WORLD_SIZE"
APP="python3 ../fine-tune.py \
--deepspeed ../ds_config_zero2.json \
--report_to "none" \
--data_path "../data/belle_chat_ramdon_10k.json" \
--model_name_or_path "../../baichuan2-7b-base" \
--output_dir "output-lora" \
--model_max_length 64 \
--num_train_epochs 4 \
--per_device_train_batch_size 1 \
--gradient_accumulation_steps 1 \
--save_strategy epoch \
--learning_rate 2e-5 \
--lr_scheduler_type constant \
--adam_beta1 0.9 \
--adam_beta2 0.98 \
--adam_epsilon 1e-8 \
--max_grad_norm 1.0 \
--weight_decay 1e-4 \
--warmup_ratio 0.0 \
--logging_steps 1 \
--gradient_checkpointing False \
--fp16 \
--use_lora True \
--local_rank $lrank "
case ${lrank} in
[0])
export HIP_VISIBLE_DEVICES=0,1,2,3
export UCX_NET_DEVICES=mlx5_0:1
export UCX_IB_PCI_BW=mlx5_0:50Gbs
numactl --cpunodebind=0 --membind=0 ${APP}
;;
[1])
export HIP_VISIBLE_DEVICES=0,1,2,3
export UCX_NET_DEVICES=mlx5_1:1
export UCX_IB_PCI_BW=mlx5_1:50Gbs
numactl --cpunodebind=1 --membind=1 ${APP}
;;
[2])
export HIP_VISIBLE_DEVICES=0,1,2,3
export UCX_NET_DEVICES=mlx5_2:1
export UCX_IB_PCI_BW=mlx5_2:50Gbs
numactl --cpunodebind=2 --membind=2 ${APP}
;;
[3])
export HIP_VISIBLE_DEVICES=0,1,2,3
export UCX_NET_DEVICES=mlx5_3:1
export UCX_IB_PCI_BW=mlx5_3:50Gbs
numactl --cpunodebind=3 --membind=3 ${APP}
;;
esac
fine-tune/requirements.txt 0 → 100644
View file @ 7f8094a3
numpy
transformers==4.28.0
sentencepiece
tokenizers
accelerate
fine-tune/slurm_script/hostfile/46547085 0 → 100644
View file @ 7f8094a3
f14r1n19
f14r2n00
f14r2n01
f14r2n02
f14r2n03
f14r2n04
f14r2n05
f14r2n06
f14r2n07
f14r2n08
f14r2n09
f14r2n10
f14r2n11
f14r2n12
f14r2n13
f14r2n14
f14r2n15
f14r2n16
f14r2n17
f14r2n18
f14r2n19
f14r3n00
f14r3n01
f14r3n02
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment