-
Notifications
You must be signed in to change notification settings - Fork 25
ONNX release notes #17
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
Conversation
Signed-off-by: Justin Chu <justinchuby@users.noreply.github.com>
|
I will have another commit to move it to done. |
jbschlosser
left a comment
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.
Awesome, thanks!
|
|
||
| #### `onnxscript>=0.3.1` is required for the `dynamo=True` option ([#157017](https://github.com/pytorch/pytorch/pull/157017)) | ||
|
|
||
| You must upgrade `onnxscript` to version 0.3.1 or higher for it to be compatible with PyTorch 2.8. |
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.
Should we specify the reason?
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.
I think it's fine? We have a new api in this version that pytorch uses
|
cc @justinchuby @xadupre for final review. |
|
@jbschlosser could you grant write access to me and @titaiwangms ? |
sure thing, one sec |
|
@titaiwangms @justinchuby I just added you to the repo; you should have received an invite email |
|
Looks like the file was directly merged to done. @jbschlosser let us know if there are any changes you would like us to make |
|
|
||
| The default will be `dynamo=True` starting from PyTorch 2.9. You are encouraged to migrate to use the `dynamo=True` option in `torch.onnx.export`. This flag makes `torch.export.export` the default export path, replacing `TorchScript`, as `TorchScript` is nearing end-of-life" | ||
|
|
||
| To maintain the old behavior, set `dynamo=True`. You are encouraged to also experiment with the `fallback=True` option that will make the exporter fall back to the `dynamo=False` path if there are errors. |
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.
dynamo=False?
|
|
||
| #### Support for symbolic operators in the `dynamo=True` export path ([#148905](https://github.com/pytorch/pytorch/pull/148905), [#149678](https://github.com/pytorch/pytorch/pull/149678), [#150038](https://github.com/pytorch/pytorch/pull/150038)) | ||
|
|
||
| Two operators `torch.onnx.ops.symbolic` and `torch.onnx.ops.symbolic_multi_out` are defined to allow you to create symbolic ONNX operators directly in your PyTorch models. You can use them in a forward function: |
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.
You can use them in a function:
* dataloader release notes updated * address comments
Signed-off-by: Justin Chu justinchuby@users.noreply.github.com
cc @titaiwangms @xadupre