Summary of common commands submitted in docker competition

Login account

export DOCKER_REGISTRY=registry.cn-hangzhou.aliyuncs.com/cqy_test
docker login $DOCKER_REGISTRY --username= Xijiao Chen Qingyuan 

Create your own image

Pull the base image

docker pull registry.cn-shanghai.aliyuncs.com/tcc-public/python:3
# Base mirror address 
#https://tianchi.aliyun.com/forum/postDetail?spm=5176.12586973.0.0.176a4127dBw2Xr&postId=67720

Get the mirroring toolkit and modify dockerfile according to your own needs

wget http://tianchi-media.oss-cn-beijing.aliyuncs.com/docker-tool/tianchi_docker.zip
unzip tianchi_docker.zip

Or enter the mirror modification

docker run --gpus all -it registry.cn-shanghai.aliyuncs.com/tcc-public/pytorch:1.4-cuda10.1-py3 /bin/bash
# If you need a dataset for testing, you can map the local path to the image 
docker run --gpus all -it -v /tmp:/tcdata registry.cn-shanghai.aliyuncs.com/tcc-public/pytorch:1.4-cuda10.1-py3 /bin/bash
pip intall ***
# Newly opened 1 Terminals 
# View the current container (running image) 
docker ps 
# Package to generate a new base Mirror image 
docker commit  Container ID registry.cn-shenzhen.aliyuncs.com/test_for_tianchi/test_for_tianchi_submit:base

Finally, execute docker_build. sh or step by step:

# Build 
docker build -t registry.cn-shenzhen.aliyuncs.com/test_for_tianchi/test_for_tianchi_submit:1.0 .
# Local testing 
nvidia-docker run your_image sh run.sh
# Upload 
docker push registry.cn-shenzhen.aliyuncs.com/test_for_tianchi/test_for_tianchi_submit:1.0