-
Notifications
You must be signed in to change notification settings - Fork 186
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
deduplicate code for some torchao q/dq ops #173
Merged
Merged
Changes from all commits
Commits
Show all changes
2 commits
Select commit
Hold shift + click to select a range
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains 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
This file contains 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
This file contains 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
This file contains 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
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -218,8 +218,11 @@ def dequantize(self, dtype=None): | |
""" | ||
Obtain the dequantized version of the quantized tensor subclass | ||
""" | ||
zero_points = torch.zeros(self.q_scales.shape, device=self.q_scales.device, dtype=self.q_scales.dtype) | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. I'm surprised this didn't cause a regression. Seems like a big change. |
||
# zero_points = 0 | ||
# TODO: fix dtype here? `to(self.dtype)` is not overwritten by `dtype` arg? | ||
dq_t = dequantize_per_channel( | ||
self.int_data.t(), self.q_scales, 0, self.dtype if dtype is None else dtype | ||
self.int_data.t(), self.q_scales, zero_points, self.dtype if dtype is None else dtype | ||
).to(self.dtype) | ||
# data was transposed to dequantize so make sure shape is correct | ||
return dq_t if not self.transposed else dq_t.t() | ||
|
@@ -292,6 +295,7 @@ def from_float(cls, input_float, qmin=-128, qmax=127): | |
Int8DynamicallyQuantizedLinearWeight.from_float(model.lin_mod.weight) | ||
) | ||
""" | ||
# because we call transpose in dequantization | ||
w_int_repr, w_scales, _ = dynamically_quantize_per_channel( | ||
input_float, qmin, qmax, torch.int8 | ||
) | ||
|
Oops, something went wrong.
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.
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.
Very, very nit: Hm, I wondering if
MappingType
is the right name... - We can definitely do this in a follow up.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.
so MappingType means how we map from floating point to quantized values. I'm open to other suggestions as well. although we may remove this and just split the function into two in the future, so we could discuss this a little bit later (after we verified this with executorch)