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This repository has been archived by the owner on Sep 18, 2024.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

Integration test #3088

Merged
merged 5 commits into from
Dec 5, 2020
Merged

Integration test #3088

merged 5 commits into from
Dec 5, 2020

Conversation

liuzhe-lz
Copy link
Contributor

@liuzhe-lz liuzhe-lz commented Nov 13, 2020

Pending tasks: (in future PR)

  • port tf1 test cases to pytorch or tf2
  • windows for non-local training services
  • kubeflow
  • framework controller

@liuzhe-lz liuzhe-lz mentioned this pull request Nov 13, 2020
77 tasks
@liuzhe-lz liuzhe-lz changed the title Integration test for local Integration test Nov 18, 2020
@liuzhe-lz liuzhe-lz closed this Nov 23, 2020
@liuzhe-lz liuzhe-lz reopened this Nov 23, 2020
@@ -73,15 +75,15 @@ RUN python3 -m pip --no-cache-dir install pandas==0.23.4 lightgbm==2.2.2
#
# Install NNI
#
RUN python3 -m pip --no-cache-dir install nni
COPY dist/nni-${NNI_RELEASE}-py3-none-manylinux1_x86_64.whl .
Copy link
Contributor

Choose a reason for hiding this comment

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

why change the install method?

Copy link
Contributor Author

@liuzhe-lz liuzhe-lz Nov 23, 2020

Choose a reason for hiding this comment

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

The Dockerfile might be used by two category of users:

  1. Customers
  2. NNI devs

For customers, if they want NNI image of latest release version, they can simply pull from docker hub. So if a customer uses Dockerfile, they are most likely using a modified NNI package, in which case they won't want latest NNI from pypi.

For us, we use pipelines to build and upload docker images. The release pipeline always builds docker image together with wheel, so there is no benefit to push then pull. It can use the same wheel file for pypi and docker.

@QuanluZhang QuanluZhang merged commit 580ce0a into master Dec 5, 2020
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.

4 participants