v0.2.0

DeconvOptim v0.2.0

Diff since v0.1.5

• Change conv methods to new convention

Closed issues:

• Allow to edit fft_dims (#14)

v0.1.5

DeconvOptim v0.1.5

Diff since v0.1.4

• CUDA 3.0 works
• Some regularizer fixes
• Documentation update

Closed issues:

• Try Different LineSearch Algorithms (#11)

Merged pull requests:

• CompatHelper: bump compat for "Noise" to "0.3" (#13) (@github-actions[bot])

v0.1.4

DeconvOptim v0.1.4

Diff since v0.1.3

Closed issues:

• GR together with padding produces bad results (#2)
• FFTW use measure instead of estimate (#10)

v0.1.3

DeconvOptim v0.1.3

Diff since v0.1.2

Adapt code that it is able to handle CUDA arrays. Furthermore minor fixes in Poisson loss due to numerical issues.

Merged pull requests:

• CompatHelper: bump compat for "Optim" to "1.2" (#5) (@github-actions[bot])
• CompatHelper: bump compat for "Zygote" to "0.6" (#6) (@github-actions[bot])
• CompatHelper: bump compat for "Interpolations" to "0.13" (#7) (@github-actions[bot])

v0.1.2

DeconvOptim v0.1.2

Closed issues:

• Quick Example Artifacts (#3)
• TagBot trigger issue (#4)

Merged pull requests:

• New pipeline (#1) (@roflmaostc)