-
Notifications
You must be signed in to change notification settings - Fork 6.8k
[v1.x] Move majority of ONNX model tests to nightly, only test a few models in PR pipeline #19848
[v1.x] Move majority of ONNX model tests to nightly, only test a few models in PR pipeline #19848
Conversation
Hey @josephevans , Thanks for submitting the PR
CI supported jobs: [clang, sanity, centos-cpu, miscellaneous, website, windows-cpu, unix-gpu, edge, unix-cpu, windows-gpu, centos-gpu] Note: |
3218ebb
to
c8ee99a
Compare
c8ee99a
to
ec3bdaf
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, thanks!
…CV models for active pipeline due to resource and time limits.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
Description
This PR moves the running of all ONNX export model end-to-end tests to the nightly build pipeline. There are just too many models to test for the normal CI pipeline that gets triggered for each PR. This tests one (or a few) models of each category. We can still manually run the tests and nightly pipeline will also run them.
Added onnx end-to-end tests for all currently supported models.
Also replace deprecated ConstantFill operator with Identity operator.