Repository navigation
Conversation
Matrix#eigensystem hung in an infinite loop for a non-symmetric matrix of size 3 or more when the QR iteration did not converge, for example with a NaN entry or with entries that overflow or underflow. The QR iteration now raises Matrix::ErrNotConverged after 30 * size steps. Co-authored-by: Claude Mythos 5.2 <noreply@anthropic.com>
wtn
marked this pull request as draft
November 16, 2026 20:10
wtn
marked this pull request as ready for review
November 16, 2026 20:25
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Matrix#eigensystemhangs in an infinite loop for a non-symmetric matrix of size 3 or more when the QR iteration inhessenberg_to_real_schurdoes not converge. This happens with aNaNentry, with most placements of anInfinityentry, and with finite entries above about1e153, where the product of two entries overflows.The QR iteration now counts steps and raises the new
Matrix::ErrNotConvergedafter30 * sizesteps, the limit EISPACKhqr2uses.Matrix#**with a non-Integer exponent callseigensystem, som ** 0.5hangs for the same matrices before this change and raisesMatrix::ErrNotConvergedafter. Symmetric matrices use a separate loop, which is unchanged.There is one more case: a matrix with very small numbers, below about
1e-163. Multiplying two numbers of this size gives 0 in floating point, so the loop takes many more steps than30 * size.Matrix[[1e-300, 2e-300, 3e-300], [4e-300, 5e-300, 6e-300], [7e-300, 8e-300, 1e-299]].eigensystemreturns a wrong answer before this change (largest eigenvalue7.9e-300, correct value1.67e-299) and raisesMatrix::ErrNotConvergedafter.