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.

Sign up for GitHub

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

Jump to bottom

Support different modes for float to bfloat conversion #4237

Closed
gramalingam opened this issue Jun 1, 2022 · 1 comment
Closed

Support different modes for float to bfloat conversion #4237

gramalingam opened this issue Jun 1, 2022 · 1 comment
Assignees
@manbearian
Labels
enhancement Request for new feature or operator

Comments

@gramalingam
Copy link
Contributor

Feature Request

As discussed in the recent PR #4193 , it would be good to extend the utility function for converting float32 to bfloat16 to support different modes (e.g., truncation vs rounding; the treatment of NaN could potentially lead to distinguishing pure-truncation vs. truncation with correct NaN handling).
@gramalingam gramalingam added the enhancement Request for new feature or operator label Jun 1, 2022
@manbearian manbearian mentioned this issue Jun 1, 2022
Merged
@jcwchen
Copy link
Member

jcwchen commented Jun 9, 2022

I guess this has been handled by Ian's fresh PR: #4238? If yes, feel free to close it. Thanks!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@manbearian manbearian

Labels
enhancement Request for new feature or operator
Projects
ONNX triaged work items
Status: Done
Milestone
No milestone
Development

No branches or pull requests
3 participants
@gramalingam @jcwchen @manbearian