v2.3.0

NeuralPDE v2.3.0

Diff since v2.2.0

Merged pull requests:

• CompatHelper: bump compat for "Optim" to "1.0" (#142) (@github-actions[bot])
• CompatHelper: bump compat for "GalacticOptim" to "0.2" (#143) (@github-actions[bot])
• seed tests (#144) (@ChrisRackauckas)

v2.2.0

NeuralPDE v2.2.0

Diff since v2.1.0

Merged pull requests:

• Generalize parser and add access to the inner functions (#133) (@KirillZubov)
• Stochastic Loss Function PINNS (#135) (@KirillZubov)
• Support gpu (#137) (@KirillZubov)

v2.1.0

NeuralPDE v2.1.0

Diff since v2.0.0

Closed issues:

• Slight refactoring of PINNs (#118)
• Array error in 2D poisson example (#128)
• third derivative (#129)

Merged pull requests:

• Slight refactoring of PINNs (#130) (@KirillZubov)
• add PINNs examples in docs (#132) (@KirillZubov)

v2.0.0

NeuralPDE v2.0.0

Diff since v1.6.0

Closed issues:

• Kolmogorov equations through neural nets (#22)
• Optimal stopping problems through neural networks (#23)
• Deep primal-dual method for semilinear PDEs (#26)
• General PDE solving with NNs (#53)
• Run solve on vectors in NNODE_tests.jl (#80)
• Make a real documentation (#85)
• Rename? (#117)

Merged pull requests:

• General PDE solving with NNs (#69) (@KirillZubov)
• Initialised RODE from the beginning according to new changes in ODE solver. (#75) (@ashutosh-b-b)
• Add Primal-Dual Algorithm for BSDEs (#96) (@ashutosh-b-b)
• CompatHelper: bump compat for "Optim" to "0.22" (#104) (@github-actions[bot])
• Revert "General PDE solving with NNs" (#105) (@ChrisRackauckas)
• General PDE solving with NNs fix revert (#106) (@KirillZubov)
• Documentation for Optimal Stopping Time (#107) (@ashutosh-b-b)
• documentation for PINNs solver (#108) (@KirillZubov)
• [WIP] Adding Docstrings to the code. (#109) (@ashutosh-b-b)
• CompatHelper: bump compat for "Flux" to "0.11" (#110) (@github-actions[bot])
• Add Upper and Lower Bounds for the NNPDENS. (#111) (@ashutosh-b-b)
• CompatHelper: add new compat entry for "DiffEqNoiseProcess" at version "5.1" (#114) (@github-actions[bot])
• PINNs solver, high-order numerical derivative (#115) (@KirillZubov)
• [WIP] Add tutorial for NNRODE and finish remaining docstrings. (#116) (@ashutosh-b-b)
• system of pde (#119) (@KirillZubov)
• Create Documentation.yml (#120) (@kanav99)
• Rename the package to NeuralPDE.jl (#121) (@ChrisRackauckas)
• improve boundary condition (#124) (@KirillZubov)

v1.6.0

NeuralNetDiffEq v1.6.0

Diff since v1.5.0

Closed issues:

• Learning Symbolic Physics with Graph Networks (#45)
• Method error (#88)
• Failed to precompile Zygote (#91)

Merged pull requests:

• Add trajectories to Kolmogorov Solver and fix bugs. (#93) (@ashutosh-b-b)
• CompatHelper: bump compat for "Adapt" to "2.0" (#94) (@github-actions[bot])
• Increase trajectories and iterations to stopping_test.jl (#95) (@ashutosh-b-b)
• Add GPU support Kolmogorov Solver (#97) (@ashutosh-b-b)
• CompatHelper: add new compat entry for "CuArrays" at version "2.2" (#98) (@github-actions[bot])
• Set up GitLab CI (#99) (@christopher-dG)
• CompatHelper: bump compat for "Zygote" to "0.5" (#101) (@github-actions[bot])

v1.5.0

NeuralNetDiffEq v1.5.0

Diff since v1.4.0

Merged pull requests:

• Intitialised Kolmogorov Equations (#61) (@ashutosh-b-b)
• Added Documentation for the Kolmogorov PDE in readme. (#62) (@ashutosh-b-b)
• CompatHelper: add new compat entry for "Adapt" at version "1.0" (#72) (@github-actions[bot])
• Added SDEProblem to Kolmogorov Solver. (#76) (@ashutosh-b-b)
• CompatHelper: add new compat entry for "Distributions" at version "0.23" (#78) (@github-actions[bot])
• Added noise_rate_prototype for Non Diagonal Noise. (#79) (@ashutosh-b-b)
• Revert 3babce713d7… (#81) (@ChrisRackauckas)
• Improving the error message for DimensionMismatch (#82) (@ashutosh-b-b)
• CompatHelper: bump compat for "Optim" to "0.21" (#83) (@github-actions[bot])
• Initialising the Optimal Stopping Time Problem (#84) (@ashutosh-b-b)
• convert the README to a Documenter documentation (#86) (@TeAmP0is0N)

v1.4.0

NeuralNetDiffEq v1.4.0

Diff since v1.3.0

Merged pull requests:

• initial parameter choices and GPU compatibility (#70) (@ChrisRackauckas)
• CompatHelper: add new compat entry for "Adapt" at version "1.0" (#72) (@github-actions[bot])

v1.3.0

NeuralNetDiffEq v1.3.0

Diff since v1.2.0

Closed issues:

• Error while trying to run code from Readme (#63)

Merged pull requests:

• NNODE Forward Mode AD (#55) (@ChrisRackauckas)
• Install TagBot as a GitHub Action (#60) (@JuliaTagBot)
• CompatHelper: add new compat entry for "Optim" at version "0.20" (#64) (@github-actions[bot])
• CompatHelper: add new compat entry for "ForwardDiff" at version "0.10" (#65) (@github-actions[bot])
• CompatHelper: add new compat entry for "DiffEqFlux" at version "1.5" (#66) (@github-actions[bot])
• Update Project.toml (#67) (@ChrisRackauckas)

v1.2.0

v1.2.0 (2020-02-07)

Diff since v1.1.0

Closed issues:

• Test Errors? (#57)
• Chaotic Attractor Replication (#34)
• A more robust solution method (#31)

Merged pull requests:

• fix tests (#59) (KirillZubov)
• rename μ -> μ_f σ -> σ_f (#58) (KirillZubov)
• CompatHelper: add new compat entry for "Zygote" at version "0.4" (#51) (github-actions[bot])
• CompatHelper: add new compat entry for "Tracker" at version "0.2" (#50) (github-actions[bot])
• fix ode_solver (#48) (KirillZubov)
• update to the new Flux based on Zygote (#33) (KirillZubov)

v1.1.0

v1.1.0 (2019-12-14)

Diff since v1.0.2

Closed issues:

• Tests sometimes fail (#41)

Merged pull requests:

• CompatHelper: bump compat for "DiffEqFlux" to "0.10" (#43) (github-actions[bot])