随手记录------＞备忘录

Docker Image

Using pre-built images

You can also pull a pre-built docker image from Docker Hub and run with docker v19.03+

docker run --gpus all --rm -ti --ipc=host pytorch/pytorch:latest

Please note that PyTorch uses shared memory to share data between processes, so if torch multiprocessing is used (e.g. for multithreaded data loaders) the default shared memory segment size that container runs with is not enough, and you should increase shared memory size either with --ipc=host or --shm-size command line options to nvidia-docker run.

Building the image yourself

NOTE: Must be built with a docker version > 18.06

The Dockerfile is supplied to build images with CUDA 11.1 support and cuDNN v8. You can pass PYTHON_VERSION=x.y make variable to specify which Python version is to be used by Miniconda, or leave it unset to use the default.

make -f docker.Makefile # images are tagged as docker.io/${your
_docker_username}/pytorch

cmake升级：

 

 Kitware APT Repository