-
Notifications
You must be signed in to change notification settings - Fork 295
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
aten.full_like.default #3843
aten.full_like.default #3843
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/3843
Note: Links to docs will display an error until the docs builds have been completed. This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This pull request was exported from Phabricator. Differential Revision: D58121891 |
Summary: Implement aten.full_like.default, which is required in OCR full model. Reuse the implementation of aten.full.default ``` func: full(SymInt[] size, Scalar fill_value, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor func: full_like(Tensor self, Scalar fill_value, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor ``` The major difference between full and full_like is the first argument, which full is an integer list and full_like is an input tensor. We can reuse lots of code here. And to support dynamic reshape, just add a condition in resize_full_node to determine the out_sizes. Reviewed By: yipjustin Differential Revision: D58121891
478f4f0
to
e6daf87
Compare
This pull request was exported from Phabricator. Differential Revision: D58121891 |
Summary: Implement aten.full_like.default, which is required in OCR full model. Reuse the implementation of aten.full.default ``` func: full(SymInt[] size, Scalar fill_value, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor func: full_like(Tensor self, Scalar fill_value, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor ``` The major difference between full and full_like is the first argument, which full is an integer list and full_like is an input tensor. We can reuse lots of code here. And to support dynamic reshape, just add a condition in resize_full_node to determine the out_sizes. Reviewed By: yipjustin Differential Revision: D58121891
e6daf87
to
303ee66
Compare
This pull request was exported from Phabricator. Differential Revision: D58121891 |
Summary: Implement aten.full_like.default, which is required in OCR full model. Reuse the implementation of aten.full.default ``` func: full(SymInt[] size, Scalar fill_value, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor func: full_like(Tensor self, Scalar fill_value, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor ``` The major difference between full and full_like is the first argument, which full is an integer list and full_like is an input tensor. We can reuse lots of code here. And to support dynamic reshape, just add a condition in resize_full_node to determine the out_sizes. Reviewed By: yipjustin Differential Revision: D58121891
303ee66
to
4f13094
Compare
This pull request was exported from Phabricator. Differential Revision: D58121891 |
Summary: Implement aten.full_like.default, which is required in OCR full model. Reuse the implementation of aten.full.default ``` func: full(SymInt[] size, Scalar fill_value, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor func: full_like(Tensor self, Scalar fill_value, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor ``` The major difference between full and full_like is the first argument, which full is an integer list and full_like is an input tensor. We can reuse lots of code here. And to support dynamic reshape, just add a condition in resize_full_node to determine the out_sizes. Reviewed By: yipjustin Differential Revision: D58121891
4f13094
to
31c995a
Compare
This pull request was exported from Phabricator. Differential Revision: D58121891 |
This pull request has been merged in 9d58de1. |
Summary:
Implement aten.full_like.default, which is required in OCR full model. Reuse the implementation of aten.full.default
The major difference between full and full_like is the first argument, which full is an integer list and full_like is an input tensor. We can reuse lots of code here. And to support dynamic reshape, just add a condition in resize_full_node to determine the out_sizes.
Reviewed By: yipjustin
Differential Revision: D58121891