Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
guomy d7e1c3eac8 | 1 year ago | |
---|---|---|
.idea | 1 year ago | |
.ipynb_checkpoints | 1 year ago | |
ascend310_infer | 1 year ago | |
scripts | 1 year ago | |
src | 1 year ago | |
.gitignore | 1 year ago | |
README.md | 1 year ago | |
README_CN.md | 1 year ago | |
__init__.py | 1 year ago | |
eval.py | 1 year ago | |
export.py | 1 year ago | |
postprocess.py | 1 year ago | |
preprocess.py | 1 year ago | |
requriments.txt | 1 year ago | |
train.py | 1 year ago |
https://git.openi.org.cn/zeizei/OpenI_Learning
https://git.openi.org.cn/OpenIOSSG
官方pytorch版本代码:
https://github.com/microsoft/Swin-Transformer/blob/main/models/swin_transformer_v2.py
参考mindspore版本代码:
https://git.openi.org.cn/a-861414389/Swin_v2;
https://gitee.com/mindspore/models/tree/r1.6/research/cv/swin_transformer
Swin-B 跑32卡NPU并行(4节点,每节点8卡),训练时长90h左右(当时资源回收影响,手动kill)
镜像:tensorflow1.15-mindspore1.7.0-cann5.1.0-euler2.8-aarch64
数据集不能太大,选择imagenet small或tiny版本
debug过程设置run_modelarts=False,运行脚本在./scripts/里,具体运行看README_CN.md,其中RANK_TABLE_FILE在该环境下直接用/user/config/nbstart_hccl.json;
镜像:MindSpore-1.6.1-python3.7-aarch64
数据集:imagenet-1K
train过程设置run_modelarts=True,目前args.py
配置可直接运行
启动文件train.py
即可,选择4节点,每节点8卡
镜像:MindSpore-1.7-c81-python3.7-euleros2.8-aarch64
数据集:imagenet-1K
test过程设置run_modelarts=True,目前args.py配置可直接运行
往模型
中导入训练好的模型,开启推理任务,选择相应的模型,启动文件eval.py
即可,选择单节点1卡
数据读写要使用绝对路径!结果要上传到指定的obs路!debug模式可以直接查看当前全路径,train和test模式可先获取当前目录,然后拼接成全路径
No Description
Python C++ Shell Markdown Text other
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》