Make quantize_pt2 return an ExportedProgram instead of a GraphModule #10644
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/10644
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: ✅ No FailuresAs of commit f3ed763 with merge base 4cc6532 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D73722640 |
zonglinpeng
approved these changes
May 2, 2025
…10644) Summary: This will help differentiating the fp32 models from the quantized models, and prevent people from using the wrong APIs. For fp32 cases, we have a `torch.nn.Module`, which we trace and then lower. For quantized cases, we trace, quantize, and lower. After this diff, `export_to_<edge, executorch>` will ONLY handle non-quantized cases, and importantly, the sequence of `quantize_pt2` and then `export_to_<edge, executorch>` will not work anymore. Those cases should use the (existing) `lower_ep_to_<edge, executorch>` instead. Note that in subsequent diffs, both `quantize_pt2` and `lower_ep_to<edge, executorch` should be inner calls, not exposed to users Reviewed By: Vysarat, zonglinpeng Differential Revision: D73722640
facebook-github-bot
force-pushed
the
export-D73722640
branch
from
d26ff69 to
f3ed763
Compare
|
This pull request was exported from Phabricator. Differential Revision: D73722640 |
jhelsby
pushed a commit
to jhelsby/executorch
that referenced
this pull request
May 9, 2025
Differential Revision: D73722640 Pull Request resolved: pytorch#10644
# for free
to join this conversation on GitHub.
Already have an account?
# to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
This will help differentiating the fp32 models from the quantized models, and prevent people from using the wrong APIs.
For fp32 cases, we have a
torch.nn.Module, which we trace and then lower. For quantized cases, we trace, quantize, and lower.After this diff,
export_to_<edge, executorch>will ONLY handle non-quantized cases, and importantly, the sequence ofquantize_pt2and thenexport_to_<edge, executorch>will not work anymore. Those cases should use the (existing)lower_ep_to_<edge, executorch>instead.Differential Revision: D73722640