Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
taoht 8998064478 | 1 year ago | |
---|---|---|
.idea | 1 year ago | |
docs | 1 year ago | |
scripts | 1 year ago | |
serving_increment | 1 year ago | |
src | 1 year ago | |
tokenizer | 1 year ago | |
LICENSE | 2 years ago | |
README.md | 1 year ago | |
ckpt_strategy_exp4.ckpt | 1 year ago | |
hostfile | 1 year ago | |
hostfile_1gpus | 1 year ago | |
hostfile_2gpus | 1 year ago | |
hostfile_8gpus | 1 year ago | |
predict.py | 1 year ago | |
requirements.txt | 1 year ago | |
train.py | 1 year ago | |
train_client.py | 1 year ago | |
train_ft.py | 1 year ago |
MindSpore 1.6.1
docker pull hengtao2/mindspore:gpu_ms_1.6.1
或者使用MindSpore官方在Docker Hub 上托管的Docker镜像。
./src/utils.py 文件中配置模型路径:
path_to_ckpt=args_opt.load_ckpt_path + args_opt.load_ckpt_name
1、脚本启动
bash scripts/run_distribute_inference.sh #启动脚本
8 #使用卡数
hostfile_8gpus #hostfile文件
2.6B #模型规模
'8,9,10,11,12,13,14,15' #使用的卡id
bash scripts/run_distribute_inference.sh 8 /tmp/hostfile_8gpus 2.6B '8,9,10,11,12,13,14,15'
2、命令行启动
mpirun --allow-run-as-root
-x PATH
-x LD_LIBRARY_PATH
-x PYTHONPATH
-x NCCL_DEBUG
-x GLOG_v
-n 8 #使用卡数
--hostfile hostfile_8gpus
--output-filename log_output
--merge-stderr-to-stdout python -s predict-flo50-zh2en2langs_GPU.py
--mode 2.6B
--run_type predict
运行如下命令开始训练,2.6B 单机16卡GPU运行
bash scripts/run_distribute_train_gpu.sh
16 #使用卡数
hostfile #hostfile文件
dataset/test/ #训练集
8 #batchsize
2.6B #模型规模
bash scripts/run_distribute_train_gpu.sh 16 /tmp/hostfile dataset/test/ 8 2.6B
鹏城实验室-智能部-开源所-基础技术研究室
mPanGu-α-53是首个以中文为中心的多语言&机器翻译模型,在一带一路沿线66个国家53种语种上进行预训练和单双语混合增量训练,单模型支持一带一路53个语种任两语种间的互译,对比WMT2021多语言任务赛道No.1在”中外“100个方向上平均BLEU值提升0.354,支持在NPU/GPU上基于MindSpore分布式训练(最少8卡)、推理(全精度/FP16,1卡)和多语言任务的迁移学习。
Text Python
Apache-2.0
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》