[BIT-104] Max-upscale weights #99
Merged
+114
−20
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
* Add vec_max_upscale_to_u16 * Do max-upscale pruning_scores * Add epoch test log details for u16 pruning_scores * Norm when no max * Added max-upscaling unnormalized handling * Update spec_version to 116
shibshib
approved these changes
Apr 5, 2023
Eugene-hu
approved these changes
Apr 6, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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Outdated suggestions cannot be applied.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.
[BIT-104] Max-upscale weights
Now stores max-upscaled u16 weights, instead of normalized u16 weights, to maximize u16 utilization. With max-upscaled u16 weights, we have
max_weight=u16::MAX. Weight normalization is eventually done in epoch, but weight storage can be unnormalized.