Merge branch 'master' into torfjelde/logdensitymodel

Author: cpfiffer

.github/workflows/CompatHelper.yml

@@ -2,18 +2,32 @@ name: CompatHelper
 
 on:
   schedule:
-    - cron: '00 00 * * *'
+    - cron: 0 0 * * *
+  workflow_dispatch:
 
 jobs:
   CompatHelper:
     runs-on: ubuntu-latest
     steps:
-      - uses: julia-actions/setup-julia@latest
-        with:
-          version: 1.3
-      - name: Pkg.add("CompatHelper")
-        run: julia -e 'using Pkg; Pkg.add("CompatHelper")'
-      - name: CompatHelper.main()
+      - name: "Add the General registry via Git"
+        run: |
+          import Pkg
+          ENV["JULIA_PKG_SERVER"] = ""
+          Pkg.Registry.add("General")
+        shell: julia --color=yes {0}
+      - name: "Install CompatHelper"
+        run: |
+          import Pkg
+          name = "CompatHelper"
+          uuid = "aa819f21-2bde-4658-8897-bab36330d9b7"
+          version = "3"
+          Pkg.add(; name, uuid, version)
+        shell: julia --color=yes {0}
+      - name: "Run CompatHelper"
+        run: |
+          import CompatHelper
+          CompatHelper.main(; subdirs = ["", "test"], bump_version=true)
+        shell: julia --color=yes {0}
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-        run: julia -e 'using CompatHelper; CompatHelper.main()'
+          COMPATHELPER_PRIV: ${{ secrets.DOCUMENTER_KEY }}

README.md

@@ -129,14 +129,19 @@ rw_prop = RandomWalkProposal(Normal(0,1))
 Different methods are easily composeable. One parameter can be static and another can be a random walk,
 each of which may be drawn from separate distributions.
 
-## Multithreaded sampling
+## Multiple chains
 
-AdvancedMH.jl implements the interface of [AbstractMCMC](https://github.com/TuringLang/AbstractMCMC.jl/), which means you get multiple chain sampling
-in parallel for free:
+AdvancedMH.jl implements the interface of [AbstractMCMC](https://github.com/TuringLang/AbstractMCMC.jl/) which means sampling of multiple chains is supported for free:
 
 ```julia
-# Sample 4 chains from the posterior.
-chain = psample(model, RWMH(init_params), 100000, 4; param_names=["μ","σ"], chain_type=Chains)
+# Sample 4 chains from the posterior serially, without thread or process parallelism.
+chain = sample(model, RWMH(init_params), MCMCSerial(), 100000, 4; param_names=["μ","σ"], chain_type=Chains)
+
+# Sample 4 chains from the posterior using multiple threads.
+chain = sample(model, RWMH(init_params), MCMCThreads(), 100000, 4; param_names=["μ","σ"], chain_type=Chains)
+
+# Sample 4 chains from the posterior using multiple processes.
+chain = sample(model, RWMH(init_params), MCMCDistributed(), 100000, 4; param_names=["μ","σ"], chain_type=Chains)
 ```
 
 ## Metropolis-adjusted Langevin algorithm (MALA)

src/AdvancedMH.jl

@@ -24,7 +24,7 @@ export
     MALA
 
 # Reexports
-export sample, MCMCThreads, MCMCDistributed
+export sample, MCMCThreads, MCMCDistributed, MCMCSerial
 
 # Abstract type for MH-style samplers. Needs better name? 
 abstract type MHSampler <: AbstractMCMC.AbstractSampler end
@@ -35,26 +35,26 @@ abstract type AbstractTransition end
 # Define a model type. Stores the log density function and the data to 
 # evaluate the log density on.
 """
-    DensityModel{F<:Function} <: AbstractModel
+    DensityModel{F} <: AbstractModel
 
 `DensityModel` wraps around a self-contained log-liklihood function `logdensity`.
 
 Example:
 
 ```julia
-l
-DensityModel
+l(x) = logpdf(Normal(), x)
+DensityModel(l)
 ```
 """
-struct DensityModel{F<:Function} <: AbstractMCMC.AbstractModel
+struct DensityModel{F} <: AbstractMCMC.AbstractModel
     logdensity :: F
 end
 
 const DensityModelOrLogDensityModel = Union{<:DensityModel,<:AbstractMCMC.LogDensityModel}
 
 # Create a very basic Transition type, only stores the
 # parameter draws and the log probability of the draw.
-struct Transition{T<:Union{Vector, Real, NamedTuple}, L<:Real} <: AbstractTransition
+struct Transition{T,L<:Real} <: AbstractTransition
     params :: T
     lp :: L
 end