search

aws / sagemaker-mxnet-training-toolkit

Toolkit for running MXNet training scripts on SageMaker. Dockerfiles used for building SageMaker MXNet Containers are at https://github.com/aws/deep-learning-containers.
Apache License 2.0
60 stars 55 forks source link

add serving dockerfile for 1.4.0 #78

Closed ChoiByungWook closed 5 years ago

ChoiByungWook commented 5 years ago

The build arg now requires 3.6 and 2.7 instead of the respective 2 and 3. As the base image we build from has that in the name and doing if statement conditionals within the Dockerfile can't fulfill that purpose.

I've removed some of the arguments that existed in the 1.3.0 Dockerfile.

docker build -t 520713654638.dkr.ecr.us-west-2.amazonaws.com/sagemaker-mxnet-serving:1.4-cpu-py3 -f docker/1.4.0/final/Dockerfile.cpu.serving --build-arg py_version=3.6 .
Successfully built 39aff94ef040
Successfully tagged 520713654638.dkr.ecr.us-west-2.amazonaws.com/sagemaker-mxnet-serving:1.4-cpu-py3

docker build -t 520713654638.dkr.ecr.us-west-2.amazonaws.com/sagemaker-mxnet-serving:1.4-gpu-py3 -f docker/1.4.0/final/Dockerfile.gpu.serving --build-arg py_version=3.6 .
Successfully built 810823b4cf77
Successfully tagged 520713654638.dkr.ecr.us-west-2.amazonaws.com/sagemaker-mxnet-serving:1.4-gpu-py3

I've added a few TODOs, as names and such are subject to change for the official release.

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.