v0.42.5

Turing v0.42.5

Diff since v0.42.4

SMC and PG can now be used for models with keyword arguments, albeit with one requirement: the user must mark the model function as being able to produce.
For example, if the model is

@model foo(x; y) = a ~ Normal(x, y)

then before sampling from this with SMC or PG, you will have to run

using Turing

@might_produce(foo)

Merged pull requests:

• Enable keyword arguments for particle methods (#2660) (@penelopeysm)

Closed issues:

• SMC samplers no longer support usage of keyword arguments in models (#2007)
• PDMat on vector from arraydist fails with reversediff (#2101)
• Consistent inferface for specifying model parameter spaces (#2139)
• Modernise the tracing data structure for DynamicPPL (#2423)
• VonMises failed to find valid initial parameters in 1000 tries. (#2584)
• Meta-Bayesian inference with Turing.jl (#2615)