Remove Sampler and its interface

HISTORY.md

@@ -42,6 +42,19 @@ The main change that this is likely to create is for those who are implementing
 The exact way in which this happens will be detailed in the Turing.jl changelog when a new release is made.
 Broadly speaking, though, `SamplingContext(MySampler())` will be removed so if your sampler needs custom behaviour with the tilde-pipeline you will likely have to define your own context.
 
+### Removal of `DynamicPPL.Sampler`
+
+`DynamicPPL.Sampler` and **all associated interface functions** have also been removed entirely.
+If you were using these, the corresponding replacements are:
+
+  - `DynamicPPL.Sampler(S)`: just don't wrap `S`; but make sure `S` subtypes `AbstractMCMC.AbstractSampler`
+  - `DynamicPPL.initialstep`: directly implement `AbstractMCMC.step` and `AbstractMCMC.step_warmup` as per the AbstractMCMC interface
+  - `DynamicPPL.loadstate`: `Turing.loadstate` (will be introduced in the next version)
+  - `DynamicPPL.default_chain_type`: removed, just use the `chain_type` keyword argument directly
+  - `DynamicPPL.initialsampler`: `Turing.Inference.init_strategy` (will be introduced in the next version; note that this function must return an `AbstractInitStrategy`, see above for explanation)
+  - `DynamicPPL.default_varinfo`: `Turing.Inference.default_varinfo` (will be introduced in the next version)
+  - `DynamicPPL.TestUtils.test_sampler` and related methods: removed, please implement your own testing utilities as needed
+
 ### Simplification of the tilde-pipeline
 
 There are now only two functions in the tilde-pipeline that need to be overloaded to change the behaviour of tilde-statements, namely, `tilde_assume!!` and `tilde_observe!!`.
@@ -58,8 +71,8 @@ The only flag other than `"del"` that `Metadata` ever used was `"trans"`. Thus t
 
 ### Removal of `resume_from`
 
-The `resume_from=chn` keyword argument to `sample` has been removed; please use `initial_state=DynamicPPL.loadstate(chn)` instead.
-`loadstate` is exported from DynamicPPL.
+The `resume_from=chn` keyword argument to `sample` has been removed; please use the `initial_state` argument instead.
+`loadstate` will be exported from Turing in the next release of Turing.
 
 ### Change of output type for `pointwise_logdensities`
 

docs/src/api.md

@@ -243,14 +243,7 @@ DynamicPPL.TestUtils.AD.ADIncorrectException
 
 ## Demo models
 
-DynamicPPL provides several demo models and helpers for testing samplers in the `DynamicPPL.TestUtils` submodule.
-
-```@docs
-DynamicPPL.TestUtils.test_sampler
-DynamicPPL.TestUtils.test_sampler_on_demo_models
-DynamicPPL.TestUtils.test_sampler_continuous
-DynamicPPL.TestUtils.marginal_mean_of_samples
-```
+DynamicPPL provides several demo models in the `DynamicPPL.TestUtils` submodule.
 
 ```@docs
 DynamicPPL.TestUtils.DEMO_MODELS
@@ -504,27 +497,7 @@ DynamicPPL.AbstractInitStrategy
 DynamicPPL.init
 ```
 
-### Samplers
-
-In DynamicPPL a generic sampler for inference is implemented.
-
-```@docs
-Sampler
-```
-
-The default implementation of [`Sampler`](@ref) uses the following unexported functions.
-
-```@docs
-DynamicPPL.initialstep
-DynamicPPL.loadstate
-DynamicPPL.init_strategy
-```
-
-Finally, to specify which varinfo type a [`Sampler`](@ref) should use for a given [`Model`](@ref), this is specified by [`DynamicPPL.default_varinfo`](@ref) and can thus be overloaded for each  `model`-`sampler` combination. This can be useful in cases where one has explicit knowledge that one type of varinfo will be more performant for the given `model` and `sampler`.
-
-```@docs
-DynamicPPL.default_varinfo
-```
+### Choosing a suitable VarInfo
 
 There is also the _experimental_ [`DynamicPPL.Experimental.determine_suitable_varinfo`](@ref), which uses static checking via [JET.jl](https://github.com/aviatesk/JET.jl) to determine whether one should use [`DynamicPPL.typed_varinfo`](@ref) or [`DynamicPPL.untyped_varinfo`](@ref), depending on which supports the model:
 

ext/DynamicPPLMCMCChainsExt.jl

@@ -3,18 +3,6 @@ module DynamicPPLMCMCChainsExt
 using DynamicPPL: DynamicPPL, AbstractPPL
 using MCMCChains: MCMCChains
 
-# Load state from a `Chains`: By convention, it is stored in `:samplerstate` metadata
-function DynamicPPL.loadstate(chain::MCMCChains.Chains)
-    if !haskey(chain.info, :samplerstate)
-        throw(
-            ArgumentError(
-                "The chain object does not contain the final state of the sampler: Metadata `:samplerstate` missing.",
-            ),
-        )
-    end
-    return chain.info[:samplerstate]
-end
-
 _has_varname_to_symbol(info::NamedTuple{names}) where {names} = :varname_to_symbol in names
 
 function DynamicPPL.supports_varname_indexing(chain::MCMCChains.Chains)

src/DynamicPPL.jl

@@ -92,8 +92,6 @@ export AbstractVarInfo,
     getargnames,
     extract_priors,
     values_as_in_model,
-    # Samplers
-    Sampler,
     # LogDensityFunction
     LogDensityFunction,
     # Contexts
@@ -128,8 +126,6 @@ export AbstractVarInfo,
     prefix,
     returned,
     to_submodel,
-    # Chain save/resume
-    loadstate,
     # Convenience macros
     @addlogprob!,
     value_iterator_from_chain,
@@ -181,7 +177,6 @@ include("contexts/transformation.jl")
 include("contexts/prefix.jl")
 include("contexts/conditionfix.jl")  # Must come after contexts/prefix.jl
 include("model.jl")
-include("sampler.jl")
 include("varname.jl")
 include("distribution_wrappers.jl")
 include("submodel.jl")

src/test_utils.jl

@@ -17,7 +17,6 @@ include("test_utils/model_interface.jl")
 include("test_utils/models.jl")
 include("test_utils/contexts.jl")
 include("test_utils/varinfo.jl")
-include("test_utils/sampler.jl")
 include("test_utils/ad.jl")
 
 end

test/runtests.jl

@@ -61,7 +61,6 @@ include("test_util.jl")
         include("varinfo.jl")
         include("simple_varinfo.jl")
         include("model.jl")
-        include("sampler.jl")
         include("distribution_wrappers.jl")
         include("logdensityfunction.jl")
         include("linking.jl")