Changelog

This documents notable changes in CurveFit.jl. The format is based on Keep a Changelog.

v1.11.3 - 2026-07-29

Changed

• CurveFit no longer reexports CommonSolve or StatsAPI functions (#128). Import solver functions from CommonSolve.jl and statistical functions from StatsAPI.jl instead.
• Added compatibility with FastRationals.jl v0.4 (#129).

v1.11.2 - 2026-07-12

Patch release for internal tests changes.

v1.11.1 - 2026-07-12

Patch release for internal docs/tests changes.

v1.11.0 - 2026-07-08

Added

• Implemented CurveFitAliasSpecifier so that it's possible to specify which arrays are aliased or copied when creating a CurveFitProblem (#117).

Fixed

• Changed the solver for NonlinearCurveFitProblem to (by default) make copies of the input arrays into its internal cache so that the original input arrays won't be overwritten if reinit!(cache; x, y, sigma) is called (#117). This can be controlled with CurveFitAliasSpecifier

v1.10.1 - 2026-07-01

Fixed

• Fixed the residuals space of ModifiedKingCurveFitAlgorithm, previously they were in E² space but now they're calculated in velocity space such that they're consistent with the standard definition of residuals = y - ŷ (#116).
• The reinit!() method for NonlinearCurveFitProblem's will now explicitly throw an error when invalid inputs are given (e.g. arrays of the wrong length) to prevent internal corruption (#116).
• Fixed the residuals and statistics calculations for polynomial fits, previously they were computed from corrupted data (#116).
• Solving with ExpSumFitAlgorithm will no longer modify the input x/y arrays, and will return a vector of residuals instead of a (n, 1) matrix (#116).
• Fixed various statistics methods when using residual-only functions with NonlinearCurveFitProblem, as well as adding support for sigma with such such functions (#116)

v1.10.0 - 2026-06-28

Added

• Added support for passing arrays to solutions of KingCurveFitAlgorithm, ModifiedKingCurveFitAlgorithm, and RationalPolynomialFitAlgorithm (#115).

Fixed

• Fixed the statistics functions for ExpSumFitAlgorithm to handle withconst=true correctly (#115).
• Fixed sol(x::Number) of ExpSumFitAlgorithm to return scalars for consistency with the other solutions (#115).
• Previously the original CurveFitProblem from a nonlinear fit was always copied into the solution, even after calling reinit!(cache, ...). This meant that the statistics functions like margin_error() etc would incorrectly return values for the original problem rather than the one actually solved. Now sol.prob is reconstructed using the correct inputs (#115).

v1.9.4 - 2026-06-28

Changed

• Previously nonlinear fits would compute the residuals as ŷ − y, they are now computed as y − ŷ to be consistent with the linear fits (#114).

Fixed

• Corrected margin_error() to use the residual degrees of freedom rather than the degrees of freedom of the model (#114).
• Fixed the covariance calculation in StatsAPI.vcov to correctly handle the uncertainties produced by linear fits of a transformed nonlinear model (e.g. from PowerCurveFitAlgorithm, ExpCurveFitAlgorithm, and KingCurveFitAlgorithm) by using the delta method (#114).

v1.9.3 - 2026-06-26

Fixed

• LinearCurveFitAlgorithm will now automatically invert the intercept when yfun is given to ensure that the returned parameters match the values being fitted. Also affects ExpCurveFitAlgorithm and PowerCurveFitAlgorithm. This is considered a bugfix rather than a breaking change.
• Fixed the parameter handling and Jacobian of KingCurveFitAlgorithm (#112).
• Fixed the statistics methods for LinearCurveFitAlgorithm when a transform is applied with yfun (#112). Previously the Jacobian for a linear function would be computed on the residuals stored in the original y-space.

v1.9.2 - 2026-06-24

Changed

• Various improvements to CI.

v1.9.1 - 2026-04-25

Fixed

• Fixed support for reinit!()'ing nonlinear fit caches when using the new AutoSpecializeCallable wrapper from NonlinearSolve (#98).

v1.9.0 - 2026-04-24

Added

• Added an absolute_sigma argument to StatsAPI.vcov and related functions to control whether the covariance matrix is rescaled by reduced χ² (#97).
• Added a weighted argument to StatsAPI.residuals and related functions to control whether the returned residuals are scaled by the problem weights, if any (#97).

Changed

• The resid field of CurveFitSolution now consistently stores the weighted residuals for both linear and nonlinear fits (#97). Previously the unweighted residuals were stored for linear fits.

Fixed

• StatsAPI.vcov and related functions previously ignored the problem weights, they are now taken into account when present (#97).

v1.8.1 - 2026-04-13

Changed

• Added support for SciMLBase v3 (#95).

v1.8.0 - 2026-04-07

Changed

• Added support for RecursiveArrayTools 4.0 (#94).

v1.7.0 - 2026-03-02

Added

• Added a precompilation workload to reduce TTFX (#90).

v1.6.0 - 2026-02-28

Added

• Implemented support for bounds for some algorithms (#87).

v1.5.1 - 2026-02-16

Changed

• Fixed compatibility with NonlinearSolveFirstOrder.jl v2 (#86).

v1.5.0 - 2026-02-14

Changed

• CurveFit now depends only on NonlinearSolveFirstOrder.jl to reduce dependencies (#85). The default algorithm remains the same.

v1.4.0 - 2026-01-31

Added

• Implemented margin_error() (#81).
• Added support for standard deviation weights for linear fits (#80).

Changed

• ScalarModel()'s will now operate in-place for improved performance (#82).

v1.3.0 - 2026-01-26

Added

• Added support for standard deviation weights for nonlinear fits (#79).

Changed

• Breaking: reinit!(::GenericNonlinearCurveFitCache) now takes in u0 as a keyword argument rather than a positional argument for consistency with NonlinearSolve.jl (#79).

Fixed

• Fixed reinit!(::GenericNonlinearCurveFitCache) to allow passing a new x/y as well as u0 (#79).

v1.2.0 - 2026-01-21

Added

• Implemented ScalarModel to allow using scalar functions as models (#75).
• Implemented SciMLBase.successful_retcode() for CurveFitSolution (#78).