Release of v0.6.3

Happy 2024 🎆 🥳

• Add support for selecting a particular device when there are multiple GPU devices.
• Fix an bug when loading a solver whose .nets have shared instance(s) of torch.nn.Module.
• Miscellaneous documentation improvement.
• PR CI is up and running again

Release of v0.6.2

1. Allow update best_net without running validation every epoch
2. New 1-D Dirichlet boundary condition for bundles
3. Some API changes for bundle solvers and bundle conditions
4. Allow saving and loading custom solvers inheriting Sovler1D, Solver2D, or BundleSolver1D
5. Suppress error related to removing colorbar caused by newer versions of matplotlib

Release of v0.6.1

Hot fix: solve a fatal compatibility issue with torch v1.13

Release of v0.6.0

• Bug fix: deduplicate parameters when networks share common layers/params
• Removed deprecated APIs
• Forward compatibility with newer versions of PyTorch

Release of V0.5.2

1. Better progress bar
2. Allow user to conveniently set torch and numpy seeds with neurodiffeq.utils.set_seed
3. Implement solvers.get_residuals
4. Misc bug fixes and better error messages

Release of v0.5.1

• Support Chebyshev points of the first and second kine for Generator1D, Generator2D, Generator3D, GeneratorND
• Modify the usage of log-spaced and log-spaced-noisy for Generator1D to make it consistent with other sampling methods
• Support tqdm progress bar with solver.fit(). Default is writing progress bar to stderr.

Release of v0.5.0

1. Integration with NeuroDiffHub (in beta)
2. Extensive operator support
3. Better monitors
4. Clearer documentation
5. Miscellaneous improvement

Release of v0.4.0

• Misc bug fixes
• Add Generic Solver (Solver for arbitrary dimensions)
• Custom loss metrics
• More Efficient grad operator
• Bundle Solver for ODE

Release of v0.3.5

Fix an issue where images are not properly rendered on PyPI index page.

Release of v0.3.4

Update the README page & misc changes.