Skip to content

Conversation

@rddefauw
Copy link

Issue #, if available: n/a

Description of changes:

  1. Updated to MLFlow 2.0.1
  2. Switch to using private NLB for ECS service
  3. Create session manager jump host so you can access the MLFlow UI via a session manager port forwarding tunnel

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

"outputs": [],
"source": [
"!pip install -q mlflow==1.28.0"
"!pip install -q mlflow==1.30.0"

Choose a reason for hiding this comment

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

Here you havent upgrade to 2.0.1. I have

@@ -1 +1 @@
mlflow==1.28.0
mlflow==1.30.0

Choose a reason for hiding this comment

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

Same as above there is version mismatch between the client SDK and the tracking server

Choose a reason for hiding this comment

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

side note: mlflow 2.0.1 requires python >= 3.8. The SKLearn image from SageMaker runs on python 3.7 which will make the install fail :(

Choose a reason for hiding this comment

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

you can update the SKLearn framework to framework_version='1.0-1', and it should work. I am testing it currently

Choose a reason for hiding this comment

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

ok, is a little more complicated, since there has been a major change in how MLFlow now deploys to SageMaker, so upgrading the SDK client requires more work

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.

2 participants