v3.0.0

RxInfer v3.0.0

Diff since v2.17.1

Merged pull requests:

• fix Bayesian Linear Regression Tutorial (#243) (@bvdmitri)
• Update MountainCart example with new Graph PPL (#264) (@Sepideh-Adamiat)
• Initialization macro (#265) (@wouterwln)
• update to coin toss model for GraphPPL v4 (#267) (@raganapa)
• Dev 4.0.0 examples chance constraints (#269) (@Raphael-Tresor)
• update to example (#270) (@raganapa)
• Fixed examples (#271) (@bvdmitri)
• New big release 3.0.0 (#272) (@bvdmitri)
• change "init" to "initialization" (#276) (@wouterwln)
• Migration guide initial commit (#278) (@wouterwln)
• update most examples (#280) (@wouterwln)
• Add error message for trailing commas (#281) (@wouterwln)
• update bayesian Linear Regression example to new initialization (#282) (@HoangMHNguyen)
• Documentation check (#283) (@wouterwln)
• Proofread documentation (#284) (@albertpod)
• Fix show for initialization (#285) (@wouterwln)
• New section in the documentation for static inference (#286) (@bvdmitri)
• Use setup-julia version 2 (#289) (@bvdmitri)
• Update README.md (#290) (@albertpod)

Closed issues:

• Splatting does not work (with non-linear node) (#122)
• Unable to target single variables in form constraints (#147)
• Document the functional form constraints and the @constraints macro (#150)
• Integrate new GraphPPL.jl (#213)
• 'FullFactorisation' is a confusing name for a constraint that impose no factorisation. (#231)
• Examples need adjustments for new GraphPPL (#242)
• Documentation need adjustments for new GraphPPL.jl (#257)
• Use initialization instead of init for the initialization (#274)
• We need a migration guide for v3 (#275)
• Adjust examples to new @initialization macro (#277)
• Consufing error message with the @initialization macro (#279)