Remove the legacy loader axes: QuantPolicy, StagingPolicy, WeightOrientation - #1162
Merged
Merged
Conversation
…icy, WeightOrientation With the resolver arc landed, WeightForm (plus per-tensor weightFormFor) is the loader's whole configuration surface; the three deprecated axes were pure duplication held together by a precedence rule. All three had carried deprecation notices since #1109/#1115 naming their WeightForm counterparts. StreamingGgufParametersLoader loses the three parameters and the legacy→form synthesis; the default is WeightForm.AS_STORED_ON_HEAP, the loader's historical behaviour, so a caller who passed nothing sees no change. AndroidGguf.loader takes a WeightForm (default mapped residency, the Android behaviour). The RAW_BYTES eager rejection goes with the enum that carried it. The flags-vs-form parity matrix retires with the flags; the behavioural coverage it carried — mapped ≡ heap values, streaming-dequant parity, shape orientation — is re-expressed directly in WeightForm terms. Remaining prose references now name WeightResidency/EncodingRequest/ WeightShapeOrientation; the historical 'why one type, not three flags' note in WeightForm's KDoc stays as history. Closes #1159. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
|
📖 Documentation Preview The documentation has been built successfully for this PR. Generated Files:
Artifacts:
This comment will be updated automatically when the PR is updated. |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes #1159 (parent #932). First slice of façade removal.
With the resolver arc landed,
WeightForm(plus per-tensorweightFormFor) is the loader's whole configuration surface — the three deprecated axes were pure duplication held together by a precedence rule.QuantPolicy,StagingPolicy,WeightOrientationdeleted fromskainet-io-core(all carried deprecation notices since Declared in-memory weight form, resolved from the target and the file (repack and dequant as stated intent) #1109/The GGUF loader takes a WeightForm; the three policy enums become views of it #1115 naming theirWeightFormcounterparts).StreamingGgufParametersLoaderloses the three parameters and the legacy→form synthesis; defaultWeightForm.AS_STORED_ON_HEAPpreserves the historical behaviour byte-for-byte. TheRAW_BYTESeager rejection goes with the enum that carried it.AndroidGguf.loadertakes aWeightForm(default mapped residency — the Android behaviour).WeightFormterms.WeightResidency/EncodingRequest/WeightShapeOrientation; the historical "why one type, not three flags" note inWeightFormstays as history.grep -r 'QuantPolicy|StagingPolicy|WeightOrientation'now hits only that historical note. Full pr-gate green (rebased on currentdevelop).🤖 Generated with Claude Code