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

Support export >2G model when using optimizer.py only #6014

Merged
merged 2 commits into from
Dec 8, 2020
Merged

Conversation

wangyems
Copy link
Contributor

@wangyems wangyems commented Dec 3, 2020

Description: Describe your changes.

Motivation and Context

  • Why is this change required? What problem does it solve?
  • If it fixes an open issue, please link to the issue here.

@wangyems wangyems requested a review from a team as a code owner December 3, 2020 02:27
@wangyems wangyems requested a review from hanbitmyths December 3, 2020 02:27
@@ -321,6 +324,11 @@ def main():

_setup_logger(args.verbose)

if os.path.realpath(args.input) == os.path.realpath(args.output):
Copy link
Contributor

@hanbitmyths hanbitmyths Dec 4, 2020

Choose a reason for hiding this comment

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

Also need to add checking dirname is the same when args.use_external_data_format is true? If input and output uses the same path, data will be overwritten anyway. And what do you think if exiting instead of warning? Do you think we have any case of using the same input and output name?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

hmm..I think this line is enough for check paths for whatever the args.use_external_data_format is if we were to use this file separately?
I am not quite following the second question is that saying we should like return the program if input and output share the same path? As warning user may loss the original model...
From our side, no, we will output it to a different path.

@wangyems
Copy link
Contributor Author

wangyems commented Dec 7, 2020

/azp run orttraining-linux-gpu-ci-pipeline

@wangyems wangyems merged commit fa06be2 into master Dec 8, 2020
@wangyems wangyems deleted the wangye/raw branch December 8, 2020 01:18
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.

3 participants