sagemaker-studio-image-build-cli
sagemaker-studio-image-build-cli copied to clipboard
CLI for building Docker images in SageMaker Studio using AWS CodeBuild.
*Issue #, if available:* *Description of changes:* Can we modify the container repositories created by CodeBuild to enable ScanOnPush by default? This seems like a best practice that we would...
As far as I can tell from the source, it seems like the user has no input to control the prefix of the uploaded code. The s3.upload_fileobj() is used but...
*Issue #, if available:* Issue https://github.com/aws-samples/sagemaker-studio-image-build-cli/issues/10 *Description of changes:* Adds `--environment` option to set `LINUX_GPU_CONTAINER` when build involves compiling Cuda operations. By submitting this pull request, I confirm that you...
[Enhancement] - Ability to specify custom buildspec template (or additional parameters in buildspec)
a use case in an environment where custom libraries are being pulled in from JFrog or even Docker hub, we need to be able to pass in credentials. The typical...
If a Docker file installs libraries from a private PyPi repo (a repo. within a VPC), the docker build fails from the codebuild project. The reason being, the code build...
Added the eu-west-1 ECR registry so it's possible to run sm-docker with eu-west-1 as AWS_DEFAULT_REGION *Issue #, if available:* *Description of changes:* Added the eu-west-1 registry for sagemaker studio images...
**Issue #, if available:** Stop-gap for missing containers mentioned in #13, but doesn't provide the requested CLI input feature **Description of changes:** Conditionally log in to the appropriate ECR accounts...
Currently build environment is hardcoded to `LINUX_CONTAINER`. This will cause builds that involve compiling Cuda operations to fail (for example, Docker images that include custom PyTorch or Tensorflow operations). Adding...
Thanks for providing the useful tool, I tried it in two regions, it works for me in us-west-2, but failed in cn-northwest-1 and the first problem seems incorrect URI, for...
When building a conda environment in the Dockerfile I am getting the following error message: ``` CondaMemoryError: The conda process ran out of memory. Increase system memory and/or try again....