Repository navigation
Add bounded typed reductions for native array references - #9
Merged
Merged
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Structured native models can now compute MEAN and unary array MIN/MAX alongside SUM over declared references. Scalar MIN/MAX clipping and unary literal-cell behavior retain their current validation, and the existing compiler still preserves editable source documents, module scope, source maps, declared order and resource budgets.
The bounded typed adapter rejects unsupported array operands rather than guessing semantics. Tests compare Euler/RK4/Radau trajectories and all 33 samples from the pinned numerical reference. The static public validation receipt binds the exact new engine to a fresh successful audit of all 993 catalog defaults; model definitions are unchanged. Numerical evidence does not establish scientific validity or general XMILE interoperability.
Validation: 3,086 passed, 18 subtests passed, 3 expected failures (357.09 s); actual local HTTP/SSE/stdio MCP baseline checks passed with 993-model inventory. Additional real SDK calls compiled and simulated all four reductions over 33 samples on each transport, matched independent formulas and rejected unsupported operands. No model saves, deployment or dependency changes.