# How to solve the error while loading saved chain as JLD2 file or JLS file?

**URL:** <https://discourse.julialang.org/t/how-to-solve-the-error-while-loading-saved-chain-as-jld2-file-or-jls-file/36031>\
**Category:** New to Julia\
**Tags:** jld2, turing\
**Created:** [March 16, 2020, 12:56am UTC](https://discourse.julialang.org/t/how-to-solve-the-error-while-loading-saved-chain-as-jld2-file-or-jls-file/36031 "2020-03-16T00:56:38Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![Manu\_Francis](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/manu_francis/32/10010_2.png) [@Manu\_Francis](https://discourse.julialang.org/u/Manu_Francis)\
**Post date:** [March 16, 2020, 12:56am UTC](https://discourse.julialang.org/t/how-to-solve-the-error-while-loading-saved-chain-as-jld2-file-or-jls-file/36031/1 "2020-03-16T00:56:39Z")

</div>

Hi,

I have saved the chain that generated using Turing package as below:

```julia
using JLD2, FileIO
using Turing

# Set the true probability of heads in a coin.
    p_true = 0.5

    # Iterate from having seen 0 observations to 100 observations.
    Ns = 0:100;

    # Draw data from a Bernoulli distribution, i.e. draw heads or tails.
    Random.seed!(12)
    data = rand(Bernoulli(p_true), last(Ns))
    # g_data = CuArray(data)

    # Declare our Turing model.
    @model coinflip(y) = begin
        # Our prior belief about the probability of heads in a coin.
        p ~ Beta(1, 1)

        # The number of observations.
        N = length(y)
        for n in 1:N
            # Heads or tails of a coin are drawn from a Bernoulli distribution.
            y[n] ~ Bernoulli(p)
        end
    end;

    # Settings of the Hamiltonian Monte Carlo (HMC) sampler.
    # iterations = 1000
    ϵ = 0.05
    τ = 10
    iterations = 1000
    # Start sampling.
    chain = sample(coinflip(data), HMC(ϵ, τ), iterations);
    @save "data\\test.jld2" chain

```

After that I cleared the chain and tried to load as below:

```julia
chain = 0
@load "data\\test.jld2" chain

```

But I got error as below:

```julia
ERROR: UndefRefError: access to undefined reference
Stacktrace:
 [1] getindex at .\array.jl:744 [inlined]
 [2] #86 at .\none:0 [inlined]
 [3] iterate at .\generator.jl:47 [inlined]
 [4] collect(::Base.Generator{UnitRange{Int64},MCMCChains.var"#86#87"{Array{ChainDataFrame,N} where N,Array{ChainDataFrame,1}}}) at .\array.jl:622
 [5] convert at C:\Users\z5168736\.julia\packages\MCMCChains\YabhG\src\summarize.jl:92 [inlined]
 [6] construct_array(::JLD2.MmapIO, ::Type{ChainDataFrame}, ::Int64) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\datasets.jl:291
 [7] read_array(::JLD2.JLDFile{JLD2.MmapIO}, ::JLD2.ReadDataspace, ::JLD2.ReadRepresentation{ChainDataFrame,JLD2.RelOffset}, ::Int64, ::UInt16, ::JLD2.RelOffset, ::Array{JLD2.ReadAttribute,1}) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\datasets.jl:316
 [8] read_data(::JLD2.JLDFile{JLD2.MmapIO}, ::JLD2.ReadRepresentation{Any,JLD2.RelOffset}, ::Array{JLD2.ReadAttribute,1}) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\datasets.jl:198
 [9] macro expansion at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\datasets.jl:156 [inlined]
 [10] macro expansion at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\datatypes.jl:76 [inlined]
 [11] read_data(::JLD2.JLDFile{JLD2.MmapIO}, ::JLD2.ReadDataspace, ::UInt8, ::Int64, ::Int64, ::Int64, ::UInt16, ::JLD2.RelOffset, ::Array{JLD2.ReadAttribute,1}) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\datasets.jl:152
 [12] load_dataset(::JLD2.JLDFile{JLD2.MmapIO}, ::JLD2.RelOffset) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\datasets.jl:92
 [13] jlconvert at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\data.jl:674 [inlined]
 [14] macro expansion at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\data.jl:1322 [inlined]
 [15] jlconvert(::JLD2.ReadRepresentation{Tuple{UInt64,Array{ChainDataFrame,1}},JLD2.OnDiskRepresentation{(0, 8),Tuple{UInt64,Array{ChainDataFrame,1}},Tuple{UInt64,JLD2.RelOffset}}()}, ::JLD2.JLDFile{JLD2.MmapIO}, ::Ptr{Nothing}, ::JLD2.RelOffset) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\data.jl:1269
 [16] read_data(::JLD2.JLDFile{JLD2.MmapIO}, ::JLD2.ReadRepresentation{Tuple{UInt64,Array{ChainDataFrame,1}},JLD2.OnDiskRepresentation{(0, 8),Tuple{UInt64,Array{ChainDataFrame,1}},Tuple{UInt64,JLD2.RelOffset}}()}, ::Array{JLD2.ReadAttribute,1}) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\dataio.jl:37
 [17] read_data(::JLD2.JLDFile{JLD2.MmapIO}, ::JLD2.ReadDataspace, ::UInt8, ::Int64, ::Int64, ::Int64, ::UInt16, ::JLD2.RelOffset, ::Array{JLD2.ReadAttribute,1}) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\datasets.jl:149
 [18] load_dataset(::JLD2.JLDFile{JLD2.MmapIO}, ::JLD2.RelOffset) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\datasets.jl:92
 [19] jlconvert at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\data.jl:674 [inlined]
 [20] macro expansion at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\data.jl:1322 [inlined]
 [21] jlconvert(::JLD2.ReadRepresentation{Base.RefValue{Tuple{UInt64,Array{ChainDataFrame,1}}},JLD2.OnDiskRepresentation{(0,),Tuple{Tuple{UInt64,Array{ChainDataFrame,1}}},Tuple{JLD2.RelOffset}}()}, ::JLD2.JLDFile{JLD2.MmapIO}, ::Ptr{Nothing}, ::JLD2.RelOffset) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\data.jl:1269
 [22] read_data(::JLD2.JLDFile{JLD2.MmapIO}, ::JLD2.ReadRepresentation{Base.RefValue{Tuple{UInt64,Array{ChainDataFrame,1}}},JLD2.OnDiskRepresentation{(0,),Tuple{Tuple{UInt64,Array{ChainDataFrame,1}}},Tuple{JLD2.RelOffset}}()}, ::Array{JLD2.ReadAttribute,1}) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\dataio.jl:37
 [23] read_data(::JLD2.JLDFile{JLD2.MmapIO}, ::JLD2.ReadDataspace, ::UInt8, ::Int64, ::Int64, ::Int64, ::UInt16, ::JLD2.RelOffset, ::Array{JLD2.ReadAttribute,1}) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\datasets.jl:149
 [24] load_dataset(::JLD2.JLDFile{JLD2.MmapIO}, ::JLD2.RelOffset) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\datasets.jl:92
 [25] jlconvert at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\data.jl:674 [inlined]
 [26] macro expansion at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\data.jl:1322 [inlined]
 [27] jlconvert(::JLD2.ReadRepresentation{NamedTuple{(:range, :model, :spl, :vi, :hashedsummary),Tuple{Random._GLOBAL_RNG,DynamicPPL.Model{var"##inner_function#608#24",NamedTuple{(:y,),Tuple{Array{Int64,1}}},DynamicPPL.ModelGen{(:y,),var"###coinflip#622",NamedTuple{(),Tuple{}}},Val{()}},DynamicPPL.Sampler{HMC{Turing.Core.ZygoteAD,(),AdvancedHMC.Adaptation.UnitEuclideanMetric},Turing.Inference.HMCState{DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},AdvancedHMC.StaticTrajectory{AdvancedHMC.EndPointTS,AdvancedHMC.Leapfrog{Float64}},AdvancedHMC.Adaptation.NoAdaptation,AdvancedHMC.PhasePoint{Array{Float64,1},AdvancedHMC.DualValue{Float64,Array{Float64,1}}}}},DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},Base.RefValue{Tuple{UInt64,Array{ChainDataFrame,1}}}}},JLD2.OnDiskRepresentation{(0, 0, 8, 16, 96),Tuple{Random._GLOBAL_RNG,DynamicPPL.Model{var"##inner_function#608#24",NamedTuple{(:y,),Tuple{Array{Int64,1}}},DynamicPPL.ModelGen{(:y,),var"###coinflip#622",NamedTuple{(),Tuple{}}},Val{()}},Any,DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},Any},Tuple{nothing,JLD2.OnDiskRepresentation{(0, 0, 8, 8),Tuple{var"##inner_function#608#24",NamedTuple{(:y,),Tuple{Array{Int64,1}}},DynamicPPL.ModelGen{(:y,),var"###coinflip#622",NamedTuple{(),Tuple{}}},Val{()}},Tuple{nothing,JLD2.OnDiskRepresentation{(0,),Tuple{Array{Int64,1}},Tuple{JLD2.RelOffset}}(),nothing,nothing}}(),JLD2.RelOffset,JLD2.OnDiskRepresentation{(0, 64, 72),Tuple{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Any,Any},Tuple{JLD2.OnDiskRepresentation{(0,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}},Tuple{JLD2.OnDiskRepresentation{(0, 8, 16, 24, 32, 40, 48, 56),Tuple{Dict{DynamicPPL.VarName{:p},Int64},Any,Any,Any,Any,Any,Any,Dict{String,BitArray{1}}},Tuple{JLD2.CustomSerialization{Array,JLD2.RelOffset},JLD2.RelOffset,JLD2.RelOffset,JLD2.RelOffset,JLD2.RelOffset,JLD2.RelOffset,JLD2.RelOffset,JLD2.CustomSerialization{Array,JLD2.RelOffset}}}()}}(),JLD2.RelOffset,JLD2.RelOffset}}(),JLD2.RelOffset}}()}, ::JLD2.JLDFile{JLD2.MmapIO}, ::Ptr{Nothing}, ::JLD2.RelOffset) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\data.jl:1269
 [28] jlconvert(::JLD2.ReadRepresentation{Chains{Real,Missing,NamedTuple{(:internals, :parameters),Tuple{Array{String,1},Array{String,1}}},NamedTuple{(:range, :model, :spl, :vi, :hashedsummary),Tuple{Random._GLOBAL_RNG,DynamicPPL.Model{var"##inner_function#608#24",NamedTuple{(:y,),Tuple{Array{Int64,1}}},DynamicPPL.ModelGen{(:y,),var"###coinflip#622",NamedTuple{(),Tuple{}}},Val{()}},DynamicPPL.Sampler{HMC{Turing.Core.ZygoteAD,(),AdvancedHMC.Adaptation.UnitEuclideanMetric},Turing.Inference.HMCState{DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},AdvancedHMC.StaticTrajectory{AdvancedHMC.EndPointTS,AdvancedHMC.Leapfrog{Float64}},AdvancedHMC.Adaptation.NoAdaptation,AdvancedHMC.PhasePoint{Array{Float64,1},AdvancedHMC.DualValue{Float64,Array{Float64,1}}}}},DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},Base.RefValue{Tuple{UInt64,Array{ChainDataFrame,1}}}}}},JLD2.OnDiskRepresentation{(0, 8, 8, 24),Tuple{Any,Missing,NamedTuple{(:internals, :parameters),Tuple{Array{String,1},Array{String,1}}},NamedTuple{(:range, :model, :spl, :vi, :hashedsummary),Tuple{Random._GLOBAL_RNG,DynamicPPL.Model{var"##inner_function#608#24",NamedTuple{(:y,),Tuple{Array{Int64,1}}},DynamicPPL.ModelGen{(:y,),var"###coinflip#622",NamedTuple{(),Tuple{}}},Val{()}},DynamicPPL.Sampler{HMC{Turing.Core.ZygoteAD,(),AdvancedHMC.Adaptation.UnitEuclideanMetric},Turing.Inference.HMCState{DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},AdvancedHMC.StaticTrajectory{AdvancedHMC.EndPointTS,AdvancedHMC.Leapfrog{Float64}},AdvancedHMC.Adaptation.NoAdaptation,AdvancedHMC.PhasePoint{Array{Float64,1},AdvancedHMC.DualValue{Float64,Array{Float64,1}}}}},DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},Base.RefValue{Tuple{UInt64,Array{ChainDataFrame,1}}}}}},Tuple{JLD2.RelOffset,nothing,JLD2.OnDiskRepresentation{(0, 8),Tuple{Array{String,1},Array{String,1}},Tuple{JLD2.RelOffset,JLD2.RelOffset}}(),JLD2.OnDiskRepresentation{(0, 0, 8, 16, 96),Tuple{Random._GLOBAL_RNG,DynamicPPL.Model{var"##inner_function#608#24",NamedTuple{(:y,),Tuple{Array{Int64,1}}},DynamicPPL.ModelGen{(:y,),var"###coinflip#622",NamedTuple{(),Tuple{}}},Val{()}},Any,DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},Any},Tuple{nothing,JLD2.OnDiskRepresentation{(0, 0, 8, 8),Tuple{var"##inner_function#608#24",NamedTuple{(:y,),Tuple{Array{Int64,1}}},DynamicPPL.ModelGen{(:y,),var"###coinflip#622",NamedTuple{(),Tuple{}}},Val{()}},Tuple{nothing,JLD2.OnDiskRepresentation{(0,),Tuple{Array{Int64,1}},Tuple{JLD2.RelOffset}}(),nothing,nothing}}(),JLD2.RelOffset,JLD2.OnDiskRepresentation{(0, 64, 72),Tuple{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Any,Any},Tuple{JLD2.OnDiskRepresentation{(0,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}},Tuple{JLD2.OnDiskRepresentation{(0, 8, 16, 24, 32, 40, 48, 56),Tuple{Dict{DynamicPPL.VarName{:p},Int64},Any,Any,Any,Any,Any,Any,Dict{String,BitArray{1}}},Tuple{JLD2.CustomSerialization{Array,JLD2.RelOffset},JLD2.RelOffset,JLD2.RelOffset,JLD2.RelOffset,JLD2.RelOffset,JLD2.RelOffset,JLD2.RelOffset,JLD2.CustomSerialization{Array,JLD2.RelOffset}}}()}}(),JLD2.RelOffset,JLD2.RelOffset}}(),JLD2.RelOffset}}()}}()}, ::JLD2.JLDFile{JLD2.MmapIO}, ::Ptr{Nothing}, ::JLD2.RelOffset) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\data.jl:1322
 [29] read_data(::JLD2.JLDFile{JLD2.MmapIO}, ::JLD2.ReadRepresentation{Chains{Real,Missing,NamedTuple{(:internals, :parameters),Tuple{Array{String,1},Array{String,1}}},NamedTuple{(:range, :model, :spl, :vi, :hashedsummary),Tuple{Random._GLOBAL_RNG,DynamicPPL.Model{var"##inner_function#608#24",NamedTuple{(:y,),Tuple{Array{Int64,1}}},DynamicPPL.ModelGen{(:y,),var"###coinflip#622",NamedTuple{(),Tuple{}}},Val{()}},DynamicPPL.Sampler{HMC{Turing.Core.ZygoteAD,(),AdvancedHMC.Adaptation.UnitEuclideanMetric},Turing.Inference.HMCState{DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},AdvancedHMC.StaticTrajectory{AdvancedHMC.EndPointTS,AdvancedHMC.Leapfrog{Float64}},AdvancedHMC.Adaptation.NoAdaptation,AdvancedHMC.PhasePoint{Array{Float64,1},AdvancedHMC.DualValue{Float64,Array{Float64,1}}}}},DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},Base.RefValue{Tuple{UInt64,Array{ChainDataFrame,1}}}}}},JLD2.OnDiskRepresentation{(0, 8, 8, 24),Tuple{Any,Missing,NamedTuple{(:internals, :parameters),Tuple{Array{String,1},Array{String,1}}},NamedTuple{(:range, :model, :spl, :vi, :hashedsummary),Tuple{Random._GLOBAL_RNG,DynamicPPL.Model{var"##inner_function#608#24",NamedTuple{(:y,),Tuple{Array{Int64,1}}},DynamicPPL.ModelGen{(:y,),var"###coinflip#622",NamedTuple{(),Tuple{}}},Val{()}},DynamicPPL.Sampler{HMC{Turing.Core.ZygoteAD,(),AdvancedHMC.Adaptation.UnitEuclideanMetric},Turing.Inference.HMCState{DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},AdvancedHMC.StaticTrajectory{AdvancedHMC.EndPointTS,AdvancedHMC.Leapfrog{Float64}},AdvancedHMC.Adaptation.NoAdaptation,AdvancedHMC.PhasePoint{Array{Float64,1},AdvancedHMC.DualValue{Float64,Array{Float64,1}}}}},DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},Base.RefValue{Tuple{UInt64,Array{ChainDataFrame,1}}}}}},Tuple{JLD2.RelOffset,nothing,JLD2.OnDiskRepresentation{(0, 8),Tuple{Array{String,1},Array{String,1}},Tuple{JLD2.RelOffset,JLD2.RelOffset}}(),JLD2.OnDiskRepresentation{(0, 0, 8, 16, 96),Tuple{Random._GLOBAL_RNG,DynamicPPL.Model{var"##inner_function#608#24",NamedTuple{(:y,),Tuple{Array{Int64,1}}},DynamicPPL.ModelGen{(:y,),var"###coinflip#622",NamedTuple{(),Tuple{}}},Val{()}},Any,DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},Any},Tuple{nothing,JLD2.OnDiskRepresentation{(0, 0, 8, 8),Tuple{var"##inner_function#608#24",NamedTuple{(:y,),Tuple{Array{Int64,1}}},DynamicPPL.ModelGen{(:y,),var"###coinflip#622",NamedTuple{(),Tuple{}}},Val{()}},Tuple{nothing,JLD2.OnDiskRepresentation{(0,),Tuple{Array{Int64,1}},Tuple{JLD2.RelOffset}}(),nothing,nothing}}(),JLD2.RelOffset,JLD2.OnDiskRepresentation{(0, 64, 72),Tuple{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Any,Any},Tuple{JLD2.OnDiskRepresentation{(0,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}},Tuple{JLD2.OnDiskRepresentation{(0, 8, 16, 24, 32, 40, 48, 56),Tuple{Dict{DynamicPPL.VarName{:p},Int64},Any,Any,Any,Any,Any,Any,Dict{String,BitArray{1}}},Tuple{JLD2.CustomSerialization{Array,JLD2.RelOffset},JLD2.RelOffset,JLD2.RelOffset,JLD2.RelOffset,JLD2.RelOffset,JLD2.RelOffset,JLD2.RelOffset,JLD2.CustomSerialization{Array,JLD2.RelOffset}}}()}}(),JLD2.RelOffset,JLD2.RelOffset}}(),JLD2.RelOffset}}()}}()}, ::Array{JLD2.ReadAttribute,1}) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\dataio.jl:37
 [30] read_data(::JLD2.JLDFile{JLD2.MmapIO}, ::JLD2.ReadDataspace, ::UInt8, ::Int64, ::Int64, ::Int64, ::UInt16, ::JLD2.RelOffset, ::Array{JLD2.ReadAttribute,1}) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\datasets.jl:149
 [31] load_dataset(::JLD2.JLDFile{JLD2.MmapIO}, ::JLD2.RelOffset) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\datasets.jl:92
 [32] getindex(::JLD2.Group{JLD2.JLDFile{JLD2.MmapIO}}, ::String) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\groups.jl:108
 [33] read(::JLD2.JLDFile{JLD2.MmapIO}, ::String) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\JLD2.jl:326
 [34] #27 at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\loadsave.jl:112 [inlined]
 [35] #jldopen#31(::Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}, ::typeof(jldopen), ::var"#27#28", ::String) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\loadsave.jl:4
 [36] jldopen(::Function, ::String) at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\loadsave.jl:2
 [37] top-level scope at C:\Users\z5168736\.julia\packages\JLD2\hB4ya\src\loadsave.jl:111

```

Moreover, I tried to write and read the file using commands mentioned in MCMCChains.j documentation as below:  
`write("chain_i.jls", chain)`  
When I tried to read using below commands,  
`chn2 = read("chain_i.jls", chain)`, I got error like below,

```julia
ERROR: MethodError: no method matching read(::IOStream, ::Chains{Real,Missing,NamedTuple{(:internals, :parameters),Tuple{Array{String,1},Array{String,1}}},NamedTuple{(:range, :model, :spl, :vi, :hashedsummary),Tuple{Random._GLOBAL_RNG,DynamicPPL.Model{var"##inner_function#608#24",NamedTuple{(:y,),Tuple{Array{Int64,1}}},DynamicPPL.ModelGen{(:y,),var"###coinflip#622",NamedTuple{(),Tuple{}}},Val{()}},DynamicPPL.Sampler{HMC{Turing.Core.ZygoteAD,(),AdvancedHMC.Adaptation.UnitEuclideanMetric},Turing.Inference.HMCState{DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},AdvancedHMC.StaticTrajectory{AdvancedHMC.EndPointTS,AdvancedHMC.Leapfrog{Float64}},AdvancedHMC.Adaptation.NoAdaptation,AdvancedHMC.PhasePoint{Array{Float64,1},AdvancedHMC.DualValue{Float64,Array{Float64,1}}}}},DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},Base.RefValue{Tuple{UInt64,Array{ChainDataFrame,1}}}}}})
Closest candidates are:
  read(::IOStream) at iostream.jl:453
  read(::IOStream, ::Type{UInt8}) at iostream.jl:348
  read(::IOStream, ::Union{Type{Int16}, Type{Int32}, Type{Int64}, Type{UInt16}, Type{UInt32}, Type{UInt64}}) at iostream.jl:357
  ...
Stacktrace:
 [1] (::Base.var"#278#279"{Tuple{Chains{Real,Missing,NamedTuple{(:internals, :parameters),Tuple{Array{String,1},Array{String,1}}},NamedTuple{(:range, :model, :spl, :vi, :hashedsummary),Tuple{Random._GLOBAL_RNG,DynamicPPL.Model{var"##inner_function#608#24",NamedTuple{(:y,),Tuple{Array{Int64,1}}},DynamicPPL.ModelGen{(:y,),var"###coinflip#622",NamedTuple{(),Tuple{}}},Val{()}},DynamicPPL.Sampler{HMC{Turing.Core.ZygoteAD,(),AdvancedHMC.Adaptation.UnitEuclideanMetric},Turing.Inference.HMCState{DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},AdvancedHMC.StaticTrajectory{AdvancedHMC.EndPointTS,AdvancedHMC.Leapfrog{Float64}},AdvancedHMC.Adaptation.NoAdaptation,AdvancedHMC.PhasePoint{Array{Float64,1},AdvancedHMC.DualValue{Float64,Array{Float64,1}}}}},DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},Base.RefValue{Tuple{UInt64,Array{ChainDataFrame,1}}}}}}}})(::IOStream) at .\io.jl:379
 [2] #open#271(::Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}, ::typeof(open), ::Base.var"#278#279"{Tuple{Chains{Real,Missing,NamedTuple{(:internals, :parameters),Tuple{Array{String,1},Array{String,1}}},NamedTuple{(:range, :model, :spl, :vi, :hashedsummary),Tuple{Random._GLOBAL_RNG,DynamicPPL.Model{var"##inner_function#608#24",NamedTuple{(:y,),Tuple{Array{Int64,1}}},DynamicPPL.ModelGen{(:y,),var"###coinflip#622",NamedTuple{(),Tuple{}}},Val{()}},DynamicPPL.Sampler{HMC{Turing.Core.ZygoteAD,(),AdvancedHMC.Adaptation.UnitEuclideanMetric},Turing.Inference.HMCState{DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},AdvancedHMC.StaticTrajectory{AdvancedHMC.EndPointTS,AdvancedHMC.Leapfrog{Float64}},AdvancedHMC.Adaptation.NoAdaptation,AdvancedHMC.PhasePoint{Array{Float64,1},AdvancedHMC.DualValue{Float64,Array{Float64,1}}}}},DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},Base.RefValue{Tuple{UInt64,Array{ChainDataFrame,1}}}}}}}}, ::String) at .\io.jl:298
 [3] open(::Function, ::String) at .\io.jl:296
 [4] read(::String, ::Chains{Real,Missing,NamedTuple{(:internals, :parameters),Tuple{Array{String,1},Array{String,1}}},NamedTuple{(:range, :model, :spl, :vi, :hashedsummary),Tuple{Random._GLOBAL_RNG,DynamicPPL.Model{var"##inner_function#608#24",NamedTuple{(:y,),Tuple{Array{Int64,1}}},DynamicPPL.ModelGen{(:y,),var"###coinflip#622",NamedTuple{(),Tuple{}}},Val{()}},DynamicPPL.Sampler{HMC{Turing.Core.ZygoteAD,(),AdvancedHMC.Adaptation.UnitEuclideanMetric},Turing.Inference.HMCState{DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},AdvancedHMC.StaticTrajectory{AdvancedHMC.EndPointTS,AdvancedHMC.Leapfrog{Float64}},AdvancedHMC.Adaptation.NoAdaptation,AdvancedHMC.PhasePoint{Array{Float64,1},AdvancedHMC.DualValue{Float64,Array{Float64,1}}}}},DynamicPPL.VarInfo{NamedTuple{(:p,),Tuple{DynamicPPL.Metadata{Dict{DynamicPPL.VarName{:p},Int64},Array{Beta{Float64},1},Array{DynamicPPL.VarName{:p},1},Array{Float64,1},Array{Set{DynamicPPL.Selector},1}}}},Float64},Base.RefValue{Tuple{UInt64,Array{ChainDataFrame,1}}}}}}) at .\io.jl:379
 [5] top-level scope at REPL[146]:1

```

How can I solve this error?

Thanks in advance  
Manu

---

<div class="post-metadata">

**Author:** ![mohamed82008](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/mohamed82008/32/18171_2.png) [@mohamed82008](https://discourse.julialang.org/u/mohamed82008)\
**Post date:** [March 16, 2020, 5:55am UTC](https://discourse.julialang.org/t/how-to-solve-the-error-while-loading-saved-chain-as-jld2-file-or-jls-file/36031/2 "2020-03-16T05:55:20Z")

</div>

CC: @cpfiffer @devmotion

---

<div class="post-metadata">

**Author:** ![cpfiffer](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/cpfiffer/32/208747_2.png) [@cpfiffer](https://discourse.julialang.org/u/cpfiffer)\
**Post date:** [March 16, 2020, 1:13pm UTC](https://discourse.julialang.org/t/how-to-solve-the-error-while-loading-saved-chain-as-jld2-file-or-jls-file/36031/3 "2020-03-16T13:13:42Z")

</div>

JLD doesn’t appear to work out of the box, but I’m not sure why. Try using chain serialization (fixed below) instead, though we should probably get JLD working at some point because it’s a much more stable storage format that serialization.

> [@Manu\_Francis](#):
>
> chn2 = read(“chain\_i.jls”, chain)

You need to specify the type you want to receive in `read`. Currently you are calling `read` with a `Chains` object you already have. You need to do this instead:

```julia
chn2 = read("chain_i.jls", Chains)

```

Here’s a full copy-pasteable working example:

```julia
using Random
using Turing

# Set the true probability of heads in a coin.
p_true = 0.5

# Iterate from having seen 0 observations to 100 observations.
Ns = 0:100;

# Draw data from a Bernoulli distribution, i.e. draw heads or tails.
Random.seed!(12)
data = rand(Bernoulli(p_true), last(Ns))
# g_data = CuArray(data)

# Declare our Turing model.
@model coinflip(y) = begin
    # Our prior belief about the probability of heads in a coin.
    p ~ Beta(1, 1)

    # The number of observations.
    N = length(y)
    for n in 1:N
        # Heads or tails of a coin are drawn from a Bernoulli distribution.
        y[n] ~ Bernoulli(p)
    end
end;

# Settings of the Hamiltonian Monte Carlo (HMC) sampler.
# iterations = 1000
ϵ = 0.05
τ = 10
iterations = 1000
# Start sampling.
chain = sample(coinflip(data), HMC(ϵ, τ), iterations);

write("chain.jls", chain)

chn2 = read("chain.jls", Chains)

```

---

<div class="post-metadata">

**Author:** ![Manu\_Francis](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/manu_francis/32/10010_2.png) [@Manu\_Francis](https://discourse.julialang.org/u/Manu_Francis)\
**Post date:** [March 16, 2020, 11:16pm UTC](https://discourse.julialang.org/t/how-to-solve-the-error-while-loading-saved-chain-as-jld2-file-or-jls-file/36031/4 "2020-03-16T23:16:05Z")

</div>

@cpfiffer: Sorry to disturb you. I misunderstood `Chains` in `chn2 = read("chain.jls", Chains)` as the chain name… But now it is clear that `Chains` is the object type.  
Thanks you  
Manu

---

<div class="post-metadata">

**Author:** ![ALIAN\_LI\_MARTIN](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/alian_li_martin/32/38653_2.png) [@ALIAN\_LI\_MARTIN](https://discourse.julialang.org/u/ALIAN_LI_MARTIN)\
**Post date:** [May 3, 2023, 12:51am UTC](https://discourse.julialang.org/t/how-to-solve-the-error-while-loading-saved-chain-as-jld2-file-or-jls-file/36031/5 "2023-05-03T00:51:15Z")

</div>

Hello, I have the same error when saving the MCMC strings. Could someone help me with this problem. Thank you so much.

```julia
MethodError: no method matching write(::IOStream, ::Chains{Float64, AxisArrays.AxisArray{Float64, 3, Array{Float64, 3}, Tuple{AxisArrays.Axis{:iter, StepRange{Int64, Int64}}, AxisArrays.Axis{:var, Vector{Symbol}}, AxisArrays.Axis{:chain, UnitRange{Int64}}}}, Missing, NamedTuple{(:parameters, :internals), Tuple{Vector{Symbol}, Vector{Symbol}}}, NamedTuple{(:start_time, :stop_time), Tuple{Vector{Float64}, Vector{Float64}}}})
Closest candidates are:
  write(::IO, ::Any) at io.jl:672
  write(::IO, ::Any, ::Any...) at io.jl:673
  write(::FilePathsBase.AbstractPath, ::Any) at C:\Users\alian\.julia\packages\FilePathsBase\9kSEl\src\path.jl:771

```

---

<div class="post-metadata">

**Author:** ![RMF](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/rmf/32/45530_2.png) [@RMF](https://discourse.julialang.org/u/RMF)\
**Post date:** [July 1, 2023, 6:19am UTC](https://discourse.julialang.org/t/how-to-solve-the-error-while-loading-saved-chain-as-jld2-file-or-jls-file/36031/6 "2023-07-01T06:19:53Z")

</div>

Hello ALIAN, I have found an alternative to save chains in Turing.  
We have the chains from the package [example](https://turing.ml/v0.22/tutorials/10-bayesian-differential-equations/).

> chain = sample(model, NUTS(0.65), MCMCSerial(), 1000, 3; progress=false

We use the JLD2 package:

> using Pkg  
> Pkg.add(“JLD2”)

> file = “C:/Users/Pc/Documents/CODE/Bayesian Statistics/2. EDO/Lotka-Volterra/ChainS.jld2” # Specify the path and name of the file  
> data = Dict(“chain” =\> chain) # Create a dictionary to store the chains  
> save(file, data) # Save the chains to the file

Then to read the chains:

> data = load(file)  
> chainguarded = data[“chain”]

---

<div class="post-metadata">

**Author:** ![ALIAN\_LI\_MARTIN](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/alian_li_martin/32/38653_2.png) [@ALIAN\_LI\_MARTIN](https://discourse.julialang.org/u/ALIAN_LI_MARTIN)\
**Post date:** [August 31, 2023, 1:52pm UTC](https://discourse.julialang.org/t/how-to-solve-the-error-while-loading-saved-chain-as-jld2-file-or-jls-file/36031/7 "2023-08-31T13:52:39Z")

</div>

Muchas Gracias Rafa, a ver si lo pruebo entonces. Saludos.