Skip to content
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

Problem loading image_classifier handler? #42

Closed
fbbradheintz opened this issue Feb 18, 2020 · 2 comments
Closed

Problem loading image_classifier handler? #42

fbbradheintz opened this issue Feb 18, 2020 · 2 comments

Comments

@fbbradheintz
Copy link
Contributor

When I do an image classification task, I get the following error:

2020-02-18 23:29:48,871 [INFO ] W-9001-densenet161_1.0-stdout org.pytorch.serve.wlm.WorkerLifeCycle - No module named 'image_classifier'

This is consistent on Mac (latest MacOS) and Linux (Ubuntu 16.04 DL AMI v26). Latest master from the TorchServe repo. The thing still works and I get inference results, but that error message was a surprise.

If things are actually working correctly (which they mostly appear to be, we shouldn't be getting an error message. Conversely, if the error message is describing a legitimate issue, it should be fixed, and we should understand why things are working anyway.

@harshbafna
Copy link
Contributor

This is a misleading error message due to a debug print statement which was not removed in the code. Will create a PR to remove the unwanted print statement.

mycpuorg pushed a commit that referenced this issue Feb 20, 2020
Issue #43 (or #42++) exactly invokes the question
we have all been pondering. It was all happening behind
the safe cocoons of comments, but the issue reporter
now brings this up.

This commit washes our dirty laundry in public. Seriously, after
thinking about it, customers would want to perform A/B testing
without disrupting status quo in production. So leaving
default version untouched makes sense since we already
have an API for setting default explicitly.
@mycpuorg
Copy link
Collaborator

Sorry, #45 is the correct PR for this

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

No branches or pull requests

3 participants