v0.2.0

This is a major update that will break the model construction. Input nodes have been removed, we should use instead regular state nodes that are set at the roots of the network. In most cases, this is just one line of code to change/remove. This will provide better consistency, especially when using custom networks with nested branches, and might also improve performance on some models.

What's Changed

• fix link in tutorial by @LegrandNico in #238
• Replace input nodes with state nodes receiving observations by @LegrandNico in #239
• update notebook for the method paper by @LegrandNico in #241
• update README and add new logo by @LegrandNico in #242
• update README: polished figures and descriptions by @LegrandNico in #243
• add new trajectories figure by @LegrandNico in #244
• v0.2.0 by @LegrandNico in #245

Full Changelog: v0.1.7...v0.2.0

v0.1.7

What's Changed

• fix pip install in cpc tutorial by @LegrandNico in #235
• qol fixes on documentation by @LegrandNico in #236
• Update init.py by @LegrandNico in #237

Full Changelog: v0.1.6...v0.1.7

v0.1.6

What's Changed

• Update documenetaion for Zurich CPC by @LegrandNico in #230
• document HGF and Network classes properly by @LegrandNico in #233
• v0.1.6 by @LegrandNico in #234

Full Changelog: v0.1.5...v0.1.6

v0.1.5

What's Changed

• Add support for nonlinear value coupling by @KoraTMontemagno in #215
• remove unittest code from the tests functions by @LegrandNico in #222
• add notebook for the method paper by @LegrandNico in #223
• remove control over verbosity by @LegrandNico in #226
• v0.1.5 by @LegrandNico in #228

New Contributors

• @KoraTMontemagno made their first contribution in #215

Full Changelog: v0.1.4...v0.1.5

v0.1.4

What's Changed

• polish docs and update github action by @LegrandNico in #214
• support group-level model comparison by @LegrandNico in #217
• v0.1.4 by @LegrandNico in #219

Full Changelog: v0.1.3...v0.1.4

v0.1.3

What's Changed

• fix error in posterior updates with multiple children that have unobserved values by @LegrandNico in #208
• add Dirichlet process node by @LegrandNico in #162
• improve code coverage by @LegrandNico in #209
• vectorize the hgf_logp function by @LegrandNico in #210
• add pointwise sampling distribution class for model comparison by @LegrandNico in #212
• v0.1.3 by @LegrandNico in #213

Full Changelog: v0.1.2...v0.1.3

v0.1.2

What's Changed

• v0.1.2 by @LegrandNico in #207

Full Changelog: v0.1.1...v0.1.2

v0.1.0

What's Changed

• remove the overide decorator for compatibility with previous Python versions by @LegrandNico in #197
• use last JAX release by @LegrandNico in #198
• build different version of the documentation for dev and stable by @LegrandNico in #199
• fix conditions for building stable docs by @LegrandNico in #201
• v0.1.0 by @LegrandNico in #202

Full Changelog: v0.0.19...v0.1.0

v0.0.19

What's Changed

• Add a Network class to handle fundamental network operations by @LegrandNico in #188
• rename the network submodule to utils by @LegrandNico in #190
• add generalised Bayesian filter node by @LegrandNico in #191
• v0.0.19 by @LegrandNico in #196

Full Changelog: v0.0.18...v0.0.19

v0.0.18

What's Changed

• Add methods and tutorial for generalized Bayesian filtering by @LegrandNico in #180
• Use a new equation for the prediction step of continuous state nodes by @LegrandNico in #181
• Allow passing attributes as keyword arguments when using .add_nodes() by @LegrandNico in #185
• show confidence interval when plotting the belief trajectories of input nodes by @LegrandNico in #186
• v0.0.18 by @LegrandNico in #189

Full Changelog: v0.0.17...v0.0.18