v0.11.1

ProbNumDiffEq v0.11.1

Diff since v0.11.0

Merged pull requests:

• Correctly compute the Jacobian at the predicted state (#232) (@nathanaelbosch)

v0.11.0

ProbNumDiffEq v0.11.0

Diff since v0.10.2

Merged pull requests:

• Split up discretize_sqrt to make profiling easier; and speed up some stuff (#227) (@nathanaelbosch)
• Fix a bug in the sampling (#229) (@nathanaelbosch)
• README updates (#230) (@nathanaelbosch)
• Backward Kernels (#231) (@nathanaelbosch)

v0.10.2

ProbNumDiffEq v0.10.2

Diff since v0.10.1

Merged pull requests:

• CompatHelper: bump compat for TaylorSeries to 0.14, (keep existing compat) (#224) (@github-actions[bot])
• Add dependabot (#225) (@nathanaelbosch)
• Simplify triangularize! a bit (#226) (@nathanaelbosch)
• Small speed-ups for solvers with FixedDiffusion() (#228) (@nathanaelbosch)

v0.10.1

ProbNumDiffEq v0.10.1

Diff since v0.10.0

Merged pull requests:

• Split the "solver" page in the documentation to make it easier to see where to find what (#220) (@nathanaelbosch)
• Change the keyword argument from init_on_du to init_on_ddu (#222) (@nathanaelbosch)

v0.10.0

ProbNumDiffEq v0.10.0

Diff since v0.9.3

Merged pull requests:

• Better benchmarks with Weave.jl (#215) (@nathanaelbosch)
• Clean up the src/ProbNumDiffEq.jl file a bit (#217) (@nathanaelbosch)
• Use SimpleUnPack (#218) (@devmotion)
• Add a DAE benchmark: ROBER (#219) (@nathanaelbosch)
• CompatHelper: bump compat for ToeplitzMatrices to 0.8, (keep existing compat) (#221) (@github-actions[bot])
• Rename destats to stats (#223) (@nathanaelbosch)

v0.9.3

ProbNumDiffEq v0.9.3

Diff since v0.9.2

Merged pull requests:

• Add Matern priors and refactor priors a bit more (#214) (@nathanaelbosch)
• Bugfix: Add StaticArrayInterface and update the prepare_alg function (#216) (@nathanaelbosch)

v0.9.2

ProbNumDiffEq v0.9.2

Diff since v0.9.1

Merged pull requests:

• QR is now triangularize + remove some more preallocations (#211) (@nathanaelbosch)
• Some code optimizations with Octavian.jl and FastBroadcast.jl (#212) (@nathanaelbosch)
• Fix bug with fixed steps where transitions were not recomputed (#213) (@nathanaelbosch)

v0.9.1

ProbNumDiffEq v0.9.1

Diff since v0.9.0

Closed issues:

• std doesn't work (#209)

Merged pull requests:

• Document the new prior option, at least a bit (#206) (@nathanaelbosch)
• CompatHelper: bump compat for TaylorSeries to 0.13, (keep existing compat) (#207) (@github-actions[bot])
• CompatHelper: bump compat for TaylorIntegration to 0.11, (keep existing compat) (#208) (@github-actions[bot])
• Implement var and std for Gaussians (#210) (@nathanaelbosch)

v0.9.0

ProbNumDiffEq v0.9.0

Diff since v0.8.6

Merged pull requests:

• Add downstream test to make sure I don't break Fenrir.jl (#193) (@nathanaelbosch)
• CompatHelper: bump compat for SpecialMatrices to 3 (#198) (@github-actions[bot])
• Change return codes from symbol to type (#199) (@nathanaelbosch)
• Make the prior a kwarg and refactor the transition matrix parts (#200) (@nathanaelbosch)
• Remove a transpose that should be a bug (#201) (@nathanaelbosch)
• Build a cache inside the ODE solution so make interpolation easier (#202) (@nathanaelbosch)
• A New Prior: Integrated Ornstein Uhlenbeck Processes (#203) (@nathanaelbosch)
• Add another overloaded function to the dev docs (#204) (@nathanaelbosch)
• Remove the EK1FDB (#205) (@nathanaelbosch)

v0.8.6

ProbNumDiffEq v0.8.6

Diff since v0.8.5

Merged pull requests:

• Introduce "measurement models" to simplify some code (#185) (@nathanaelbosch)
• Add a make_new_transitions check before building transitions (#190) (@nathanaelbosch)
• Introduce xprev, mimicking OrdinaryDiffEq.jl's use of uprev (#191) (@nathanaelbosch)
• Refactor prior (#192) (@nathanaelbosch)
• CompatHelper: bump compat for TaylorIntegration to 0.10, (keep existing compat) (#195) (@github-actions[bot])

EDIT: This list is not correct! The patch release has been made from an older branch.