Skip to content
This repository has been archived by the owner on Sep 18, 2024. It is now read-only.

Remove unneccessary dependencies in pipeline #1597

Merged
merged 3 commits into from
Oct 10, 2019
Merged

Conversation

chicm-ms
Copy link
Contributor

No description provided.

@chicm-ms chicm-ms changed the title Dev refactor pipeline Remove unneccessary dependencies in pipeline Oct 10, 2019
python3 -m pip install torch==0.4.1 --user
python3 -m pip install torchvision==0.2.1 --user
python3 -m pip install keras==2.1.6 --user
python3 -m pip install tensorflow==1.12.0 --user
Copy link
Contributor

Choose a reason for hiding this comment

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

these packages are not necessary any more or installed at somewhere else?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

They are not necessary at all, the test trial jobs are running on remote kubeflow training service. These dependencies are installed on local kubeflow client (nnimanager).

Copy link
Contributor

@liuzhe-lz liuzhe-lz Oct 10, 2019

Choose a reason for hiding this comment

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

Seems this installs dependencies to host environment but the trials are running in container?
If that's the case, then nnictl package install are also unnecessary? I got it.

@chicm-ms chicm-ms merged commit f60bf1d into master Oct 10, 2019
@chicm-ms chicm-ms deleted the dev-refactor-pipeline branch October 10, 2019 08:12
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants