Use normalize MSE in mixed precision sensitivity evaluation #1082
Conversation
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
ofirgo
reviewed
May 21, 2024
| @@ -89,10 +89,12 @@ def __init__(self, | |||
| fw_impl.count_node_for_mixed_precision_interest_points, | |||
| quant_config.num_interest_points_factor) | |||
|
|
|||
| self.ips_distance_fns, self.ips_axis = self._init_metric_points_lists(self.interest_points) | |||
| self.ips_distance_fns, self.ips_axis = self._init_metric_points_lists(self.interest_points, | |||
There was a problem hiding this comment.
add a comment here explaining the default behavior (that we run normalized MSE when not running hessian-based, you can also explain why in a sentence), otherwise we won't be able to figure out why we did it if we'll look at this one year from now
ofirgo
reviewed
May 21, 2024
| @@ -145,7 +145,7 @@ def test_softmax_interest_point(self): | |||
| if axis is None: | |||
| axis = sn.op_call_kwargs.get(AXIS) | |||
|
|
|||
| distance_fn = KerasImplementation().get_node_distance_fn(layer_class=sn.layer_class, | |||
| distance_fn = KerasImplementation().get_mp_node_distance_fn(layer_class=sn.layer_class, | |||
There was a problem hiding this comment.
this change makes the lines not aligned
ofirgo
approved these changes
May 21, 2024
There was a problem hiding this comment.
See minor comments.
In addition:
- Update @elad-c directly regarding adding this to the release notes because he already edited them for the release.
- Did you make sure that both cases are covered in our MP tests? if not, maybe we should add a dedicated test
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in 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Outdated suggestions cannot be applied.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.
Pull Request Description:
Use normalized MSE distance function in mixed precision sensitivity evaluation. For hessian-based sensitivity, simple MSE (not normalized) is used.
Checklist before requesting a review: