-
Notifications
You must be signed in to change notification settings - Fork 4.2k
[export] update dynamic shapes section #3183
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/tutorials/3183
Note: Links to docs will display an error until the docs builds have been completed. ❌ 2 New FailuresAs of commit e42792e with merge base 7038ce7 ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
…b.com/pytorch/tutorials into pianpwk/export_dynamic_shapes_12_2024
# example inputs given to the initial ``torch.export.export()`` call. | ||
# If we try to run the ``ExportedProgram`` in the example below with a tensor | ||
# with a different shape, we get an error: | ||
# This section covers dynamic behavior and representation of exported programs. Dynamic behavior is |
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.
In line 14, it says "This tutorial provides a snapshot of torch.export
usage as of PyTorch 2.3."
Should we update that to PyTorch 2.5 or 2.6?
def forward(self, x, y): | ||
return torch.nn.functional.relu(self.lin(x + y), inplace=True) | ||
###################################################################### | ||
# Before we look at the program that's produced, let's understand what specifying ``dynamic_shapes`` entails, |
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.
nit: can we make this the regular text instead of a comment? It's a little hard to read because of the color and line breaks.
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.
similarly for the other chunks of long texts below
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.
Oh hmm I thought this was fixed, my bad
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.
Actually I think they just need to be regenerated.. not sure how
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.
Actually I think they just need to be regenerated.. not sure how
I thought the preview here should be the regenerated one? They are still comments here. https://docs-preview.pytorch.org/pytorch/tutorials/3183/intermediate/torch_export_tutorial.html#constraints-dynamic-shapes
Not sure if it has anything to do with the CI failure
https://github.com/pytorch/tutorials/actions/runs/12287425965/job/34289430334?pr=3183
Description
Updates the export dynamic shapes tutorial to be more up-to-date with our new APIs.
Checklist
cc @avikchaudhuri @gmagogsfm @zhxchen17 @tugsbayasgalan @angelayi @suo @ydwu4