Skip to content

fix: allow output_path without trailing slash in Local Mode training jobs #1439

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Apr 24, 2020

Conversation

laurenyu
Copy link
Contributor

Issue #, if available:
#1354

Description of changes:
Local Mode's logic for constructing the final output path assumes that the estimator's output_path ended with a slash ("/"), which deviates from SageMaker's behavior (see #1354).

Testing done:

  • unit tests
  • Chainer integ test (see diff)

Merge Checklist

Put an x in the boxes that apply. You can also fill these out after creating the PR. If you're unsure about any of them, don't hesitate to ask. We're here to help! This is simply a reminder of what we are going to look for before merging your pull request.

General

  • I have read the CONTRIBUTING doc
  • I used the commit message format described in CONTRIBUTING
  • I have passed the region in to any/all clients that I've initialized as part of this change.
  • I have updated any necessary documentation, including READMEs and API docs (if appropriate)

Tests

  • I have added tests that prove my fix is effective or that my feature works (if appropriate)
  • I have checked that my tests are not configured for a specific region or account (if appropriate)
  • I have used unique_name_from_base to create resource names in integ tests (if appropriate)

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

@sagemaker-bot
Copy link
Collaborator

AWS CodeBuild CI Report

  • Result: FAILED
  • Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

@sagemaker-bot
Copy link
Collaborator

AWS CodeBuild CI Report

  • Result: SUCCEEDED
  • Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

@sagemaker-bot
Copy link
Collaborator

AWS CodeBuild CI Report

  • Result: SUCCEEDED
  • Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

@sagemaker-bot
Copy link
Collaborator

AWS CodeBuild CI Report

  • Result: SUCCEEDED
  • Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

@sagemaker-bot
Copy link
Collaborator

AWS CodeBuild CI Report

  • Result: SUCCEEDED
  • Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

@sagemaker-bot
Copy link
Collaborator

AWS CodeBuild CI Report

  • Result: SUCCEEDED
  • Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

@sagemaker-bot
Copy link
Collaborator

AWS CodeBuild CI Report

  • Result: SUCCEEDED
  • Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

str: an S3 prefix of the form ``"path/job_name"``
"""
path = path.strip("/")
return job_name if path == "" else "/".join((path, job_name))

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Just curious to know instead of doing string manipulation, wouldn't os.path.join or pathlib.Path a better options for correctly making unified paths for local and non-local modes?

Copy link
Contributor Author

@laurenyu laurenyu Apr 24, 2020

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

it would, except that S3 always uses "/" as a separator, so using os.path.join would fail for Windows users.

as far as I can tell, there's not an easy way of overriding the separator used in os.path.join (I wrote a little about this in #1435). The other option would be to use something like urlunparse, but I think it might be a little heavy-handed for this use-case.

edit: looking into pathlib

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

PurePosixPath does actually look to solve the problem, except that it was introduced in Python 3.4 and never backported to Python 2.7. since this SDK still currently supports Python 2.7, we can't use it here either. however, I'll make a note in our internal backlog to use PurePosixPath after we do get around to dropping Python 2.7 support. Thanks for sharing that! We've had quite a few conversations on our team around finding a module that would work no matter the OS, but hadn't yet found pathlib.

@sagemaker-bot
Copy link
Collaborator

AWS CodeBuild CI Report

  • Result: SUCCEEDED
  • Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

@sagemaker-bot
Copy link
Collaborator

AWS CodeBuild CI Report

  • Result: SUCCEEDED
  • Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

@sagemaker-bot
Copy link
Collaborator

AWS CodeBuild CI Report

  • Result: SUCCEEDED
  • Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

@sagemaker-bot
Copy link
Collaborator

AWS CodeBuild CI Report

  • Result: SUCCEEDED
  • Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants