Closed mgmverburg closed 3 years ago
This looks like a bug in Distributions that was fixed in >= 0.24.10: https://github.com/JuliaStats/Distributions.jl/pull/1244 Updating your old version of Distributions should fix the problem.
Yep that indeed fixes it, thanks!
Hi,
So I was trying out Bijectors and for some reason when I call the transformed method, I get an error:
I noticed now that it seems like this error only causes a problem with the TransformedDistribution not being able to be printed, because apparently, despite the error I can still actually use the result
z
to sample from for example. E.g. evenrand(transformed(Beta()))
actually works without the error. So I guess this is potentially a minor thing, but still weird that error gets thrown. Additionally, I can just arbitrarily do something likelength(::Logit{0, Float64}) = 1
which also fixes the problem, which leads me to believe that maybe there is another package that defines this function, which I simply don't have installed in my Julia environment?I am using Julia 1.5.2 and the output of "] st" is as follows:
] st output
``` [fbe9abb3] AWS v1.23.1 [4f1ea46c] AWSCore v0.6.17 [1c724243] AWSS3 v0.8.1 [0d499d91] AWSSDK v0.5.0 [99985d1d] AbstractGPs v0.2.16 [c7e460c6] ArgParse v1.1.1 [76274a88] Bijectors v0.8.14 [336ed68f] CSV v0.8.2 [a93c6f00] DataFrames v0.22.2 [b4f34e82] Distances v0.10.0 [31c24e10] Distributions v0.24.8 [634d3b9d] DrWatson v1.16.6 [366bfd00] DynamicPPL v0.10.5 [1a297f60] FillArrays v0.10.1 [682c06a0] JSON v0.21.1 [ec8451be] KernelFunctions v0.8.11 [929cbde3] LLVM v3.5.2 [c7f686f2] MCMCChains v4.4.0 [c03570c3] Memoize v0.4.4 [90014a1f] PDMats v0.10.1 [91a5bcdd] Plots v1.9.1 [7f904dfe] PlutoUI v0.6.10 [37e2e3b7] ReverseDiff v1.4.5 [21efa798] SIMDPirates v0.8.26 [2d09df54] StatisticalRethinking v2.2.9 [4c63d2b9] StatsFuns v0.9.6 [f3b207a7] StatsPlots v0.14.17 [5d786b92] TerminalLoggers v0.1.2 [fce5fe82] Turing v0.15.5 [56ddb016] Logging ```