Trying to weave the current GMM tutorial fails with:
ERROR: AssertionError: Difference between estimated mean of μ ([-3.1465091632143567, 0.6286046354218603]) and data-generating μ ([-3.5, 0.5]) unexpectedly large!
Due to this check:
let
# Verify that the output of the chain is as expected.
for i in MCMCChains.chains(chains)
# μ[1] and μ[2] can switch places, so we sort the values first.
chain = Array(chains[:, ["μ[1]", "μ[2]"], i])
μ_mean = vec(mean(chain; dims=1))
@assert isapprox(sort(μ_mean), μ; rtol=0.1) "Difference between estimated mean of μ ($(sort(μ_mean))) and data-generating μ ($μ) unexpectedly large!"
end
end
The notebook only draws 100 samples, and the data (only 60 samples) will also vary from the true mean, so I think this is bound should be loosened.
I am already working on extending tutorial as discussed in the slack, and so I can change this too, if desired.
Trying to weave the current GMM tutorial fails with:
Due to this check:
The notebook only draws 100 samples, and the data (only 60 samples) will also vary from the true mean, so I think this is bound should be loosened.
I am already working on extending tutorial as discussed in the slack, and so I can change this too, if desired.