Merge branch 'dev'

docs/make.jl

@@ -7,18 +7,5 @@ using StatsBase
 using AbstractMCMC
 
 makedocs(
-    sitename="MultilevelChainSampler", remotes=nothing,
-
-    pages = [
-        "Package" => "index.md",
-        "Usage" => [
-            "usage/proposals.md",
-            "usage/sampling.md"
-        ],
-        "References" => "references.md"
-    ]
+    sitename="MultilevelChainSamplers", remotes=nothing,
 )
-
-deploydocs(
-    #...
-)
\ No newline at end of file

docs/src/index.md

-# MultilevelChainSamplers.jl Package
+# MultilevelChainSamplers.jl
 
-## About
+```@raw html
+<h3> Table of contents </h3>
+```
+
+```@contents
+Depth=5
+```
+
+```@raw html
+<h3> Table of contents </h3>
+```
 
 MCMC Sampler for multilevel delayed acceptance.
 
 
-## Overview 
+## Usage
 
-Documentation 
+### Proposal generators
 
-```@contents
-Pages = ["usage/proposals.md", "usage/sampling.md"] 
-Depth = 3
+#### List of built-in proposals
+
+```@docs
+StaticProposal
+```
+
+```@docs
+RandomWalk
+```
+
+```@docs
+TransformedProposal
+```
+
+```@docs
+StackedProposal
+```
+
+```@docs
+ComponentwiseProposal
+```
+
+#### Custom Proposal generator
+
+To define a custom proposal inherit from the abstract type 
+```julia
+struct MyProposal <: AbstractProposal{issymmetric::Bool}
+    # ... 
+end 
 ```
 
+```@docs
+propose
+```
+
+Implement the sampling methods 
+- initialization x₀ ~ p₀(⋅) by `x₀ = propose(p)`  
+- and transition y ~ q(⋅|x) by `y  = propose(p, x)`
 
-## Index
+```julia
+propose(rng::AbstractRNG, p::MyProposal) = ...
+propose(rng::AbstractRNG, p::MyProposal, x) = ...
+```
+
+For symmetric proposals `q(x|y) = q(y|x)` and the likelihood ratio can be neglected. 
+For asymmetric proposals also implement method for calculating  q(y|x)  
+```julia
+logpdf(p::AbstractProposal{false},x,y) = ...
+```
+
+
+### Sampling 
+Implements the `AbstractMCMC` sampling interface. 
+
+```@raw html 
+More information can be found on the 
+<a href="https://turinglang.org/AbstractMCMC.jl"> AbstractMCMC.jl </a>
+documentation page.
+```
+
+In a nutshell the sampling function takes the form
+
+```julia
+sample(t::LogDensity, mh::Sampler, N_or_isdone, options...)
+```
+
+The target `t` is a logarithmic unnormalized probability density 
+```@docs
+LogDensity
+```
+
+In particular the options includes thread and process parallel sampling.
+
+
+
+#### Metropolis Hastings (MH)
+
+```@docs
+MH
+```
+
+Published in [[1]](#1).
+
+
+
+#### Multilevel Delayed Acceptance (MLDA)
+
+
+Construct a multilevel proposal as vector of simple proposals.
+
+```@docs
+MLDA
+```
+
+Published in [[1]](#2).
+
+
+## Reference
+
+### Index 
 
 ```@index
 ```
 
+### Citations 
+
+```@raw html
+<a id="1">[1]</a> 
+Hastings, W.K. (1970). 
+"Monte Carlo Sampling Methods Using Markov Chains and Their Applications". 
+Biometrika, Volume 57, Issue 1
+
+<br />
+
+<a id="2">[2]</a>
+Mikkel B. Lykkegaard, T. Dodwell, C. Fox, Grigorios Mingas, Robert Scheichl (2022). 
+"Multilevel Delayed Acceptance MCMC"
+SIAM/ASA J. Uncertain. Quantification
+```
+
+
+
+

docs/src/usage/proposals.md

-# Proposal generators
-
-## List of built-in proposals
-
-```@docs
-StaticProposal
-```
-
-```@docs
-RandomWalk
-```
-
-```@docs
-TransformedProposal
-```
-
-```@docs
-StackedProposal
-```
-
-```@docs
-ComponentwiseProposal
-```
-
-## Custom Proposal generator
-
-To define a custom proposal inherit from the abstract type 
-```julia
-struct MyProposal <: AbstractProposal{issymmetric::Bool}
-    # ... 
-end 
-```
-
-```@docs
-propose
-```
-
-Implement the sampling methods 
-- initialization x₀ ~ p₀(⋅) by `x₀ = propose(p)`  
-- and transition y ~ q(⋅|x) by `y  = propose(p, x)`
-
-```julia
-propose(rng::AbstractRNG, p::MyProposal) = ...
-propose(rng::AbstractRNG, p::MyProposal, x) = ...
-```
-
-For symmetric proposals `q(x|y) = q(y|x)` and the likelihood ratio can be neglected. 
-For asymmetric proposals also implement method for calculating  q(y|x)  
-```julia
-logpdf(p::AbstractProposal{false},x,y) = ...
-```

docs/src/usage/sampling.md

-
-# Sampling 
-Implements the `AbstractMCMC` sampling interface. 
-
-```@raw html 
-More information can be found on the 
-<a href="https://turinglang.org/AbstractMCMC.jl"> AbstractMCMC.jl </a>
-documentation page.
-```
-
-In a nutshell the sampling function takes the form
-
-```julia
-sample(t::LogDensity, mh::Sampler, N_or_isdone, options...)
-```
-
-The target `t` is a logarithmic unnormalized probability density 
-```@docs
-LogDensity
-```
-
-In particular the options includes thread and process parallel sampling.
-
-
-
-## Metropolis Hastings (MH)
-
-```@docs
-MH
-```
-
-```@raw html
-<a id="1">[1]</a> 
-Hastings, W.K. (1970). 
-"Monte Carlo Sampling Methods Using Markov Chains and Their Applications". 
-Biometrika, Volume 57, Issue 1
-```
-
-
-## Multilevel Delayed Acceptance (MLDA)
-
-
-Construct a multilevel proposal as vector of simple proposals.
-
-```@docs
-MLDA
-```
-
-```@raw html
-<a id="2">[2]</a>
-Mikkel B. Lykkegaard, T. Dodwell, C. Fox, Grigorios Mingas, Robert Scheichl (2022). 
-"Multilevel Delayed Acceptance MCMC"
-SIAM/ASA J. Uncertain. Quantification
-```
-