Skip to content

Releases: JuliaDiff/FiniteDifferences.jl

v0.12.2

23 Feb 14:30
5d823e1
Compare
Choose a tag to compare

FiniteDifferences v0.12.2

Diff since v0.12.1

Merged pull requests:

v0.12.1

18 Jan 13:56
d39d622
Compare
Choose a tag to compare

FiniteDifferences v0.12.1

Diff since v0.12.0

Closed issues:

  • Make History object not part of Method object (#62)

Merged pull requests:

v0.12.0

10 Jan 18:11
c2147a0
Compare
Choose a tag to compare

FiniteDifferences v0.12.0

Diff since v0.11.7

Closed issues:

  • gradient of non-smooth functions (#129)
  • Bug when using Matrix as differential for Adjoint (#133)
  • Drop CI for Julia Nightly (#135)

Merged pull requests:

v0.11.7

06 Jan 01:06
e262eb8
Compare
Choose a tag to compare

FiniteDifferences v0.11.7

Diff since v0.11.6

Merged pull requests:

v0.11.6

03 Jan 17:56
e6e2573
Compare
Choose a tag to compare

FiniteDifferences v0.11.6

Diff since v0.11.5

Closed issues:

  • Accuracy is a bit brittle (#124)

Merged pull requests:

v0.11.5

03 Dec 18:17
cfef6f8
Compare
Choose a tag to compare

FiniteDifferences v0.11.5

Diff since v0.11.4

Merged pull requests:

  • WIP: Remove add_tiny and add estimate_magitude (#125) (@wesselb)

v0.11.4

02 Dec 18:15
461fb8a
Compare
Choose a tag to compare

FiniteDifferences v0.11.4

Diff since v0.11.3

Closed issues:

  • to_vec for non-real vectors is type-unstable (#117)

Merged pull requests:

v0.11.3

20 Nov 16:23
673d6fa
Compare
Choose a tag to compare

FiniteDifferences v0.11.3

Diff since v0.11.2

Closed issues:

  • High order, high adapt test failed in nightly (#113)

Merged pull requests:

v0.11.2

13 Oct 12:24
b71d2fc
Compare
Choose a tag to compare

FiniteDifferences v0.11.2

Diff since v0.11.1

Merged pull requests:

  • Defined to_vec for ChainRulesCore differential types (#111) (@oxinabox)

v0.11.1

08 Oct 21:17
0f812fa
Compare
Choose a tag to compare

FiniteDifferences v0.11.1

Diff since v0.11.0

Merged pull requests: