jetpack4.3_docker

instructions for building basic environments like jetpack4.3 based on nvidia CUDA docker

Environment after installation

python 3.6.9

CUDA 10.0

cuDNN 7.6.5

TensorRT 6.0.1

torch 1.2.0

torchvision 0.4.0

opencv 4.1.1

etc.

Download corresponding resources

• Base docker with CUDA 10.0 from nvidia

  docker pull nvidia/cuda:10.0-devel-ubuntu18.04

• TensorRT

  TensorRT 6.0.1.5 GA for Ubuntu 1804 and CUDA 10.0 DEB local repo packages

  Need Nvidia Developer account for access.

• Torch & Torchvision

  Torch 1.2.0

  Torchvision 0.4.0

Start making the docker

1. start the base docker with:

   docker run --gpus 0 --name tensorrt_test -it nvidia/cuda:10.0-devel-ubuntu18.04 bash

2. copy the needed file to docker in another terminal:

   docker cp nv-tensorrt-repo-ubuntu1804-cuda10.0-trt6.0.1.5-ga-20190913_1-1_amd64.deb tensorrt_test:/root

   docker cp torch-1.2.0-cp36-cp36m-manylinux1_x86_64.whl tensorrt_test:/root

   docker cp torchvision-0.4.0-cp36-cp36m-manylinux1_x86_64.whl tensorrt_test:/root

   docker cp create_tensorrt_pt_docker.sh tensorrt_test:/root

3. go to the docker terminal and execute the bash script:

   bash create_tensorrt_pt_docker.sh

TODO

• opencv