Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
wangxingrui c7ece9967f | 2 months ago | |
---|---|---|
davinci | 2 months ago | |
pcl_init | 2 months ago | |
.DS_Store | 2 months ago | |
README.MD | 2 months ago | |
cell_wrapper.py | 2 months ago | |
img.png | 2 months ago | |
main.py | 2 months ago | |
mindspore-verification.py | 2 months ago | |
rank_table_2pcs.json | 2 months ago | |
rank_table_8pcs.json | 2 months ago | |
rank_table_16pcs.json | 2 months ago | |
resnet.py | 2 months ago | |
resnet50_distributed_training.py | 2 months ago | |
resnet50_distributed_training_1.py | 2 months ago | |
resnet50_distributed_training_dataset_slice.py | 2 months ago | |
resnet50_distributed_training_gpu.py | 2 months ago | |
resnet50_distributed_training_pipeline.py | 2 months ago | |
run.sh | 2 months ago | |
run_cluster.sh | 2 months ago | |
run_dataset_slice.sh | 2 months ago | |
run_gpu.sh | 2 months ago | |
train.py | 2 months ago |
这个是怎么在云脑二上跑起来的呢
直接跑跑不了,找了陈老师,给了达芬奇脚本不能用,自己在网上找了一个脚本,改成了pcl_init。
1、用的数据是cifar-10
2、镜像yigousuanli/train-mindspore-1.7.0:mindspore_1.7.0-cann_5.1.0-py_3.7-euler_2.8.3-aarch64-d910
3、使用pcl-demo/distributed-training-mindspore/pcl_init/run.py
运行
pcl-demo/distributed-training-mindspore/resnet50_distributed_training.py
# 启动命令
export PCL_TASK_HOSTS=${VC_WORKER_HOSTS};export PCL_TASK_NUM=${VC_WORKER_NUM};export PCL_TASK_INDEX=${VC_TASK_INDEX};python /code/mindspore_distributed_training/pcl_init/run.py python /code/mindspore_distributed_training/resnet50_distributed_training.py --data_url=/data/distributed-cloud2-zhujian/cifar-10 --train_url=/output ; echo 1
这样就应该跑起来了
No Description
Python Shell
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》