Merge pull request #39 from TuringLang/yebai-patch-1

replace `filldist` with `product_distribution`

Author: penelopeysm

Project.toml

@@ -9,7 +9,6 @@ DynamicPPL = "366bfd00-2699-11ea-058f-f148b4cae6d8"
 Enzyme = "7da242da-08ed-463a-9acd-ee780be4f1d9"
 FiniteDifferences = "26cc04aa-876d-5657-8c51-4c34ba976000"
 ForwardDiff = "f6369f11-7733-5829-9624-2563aa707210"
-LazyArrays = "5078a376-72f3-5289-bfd5-ec5146d43c02"
 LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
 LogDensityProblems = "6fdf6af0-433a-55f7-b3ed-c6c6e0b8df7c"
 MLDatasets = "eb30cadb-4394-5ae3-aed4-317e484a6458"

models/dppl_gauss_unknown.jl

@@ -6,7 +6,7 @@ y = randn() .+ s * randn(n)
     N = length(y)
     m ~ Normal(0, 1)
     s ~ truncated(Cauchy(0, 5); lower=0)
-    y ~ filldist(Normal(m, s), N)
+    y ~ product_distribution(fill(Normal(m, s), N))
 end
 
 model = dppl_gauss_unknown(y)

models/dppl_hier_poisson.jl

@@ -1,4 +1,3 @@
-using LazyArrays
 using Turing: LogPoisson
 
 nd, ns = 5, 10
@@ -13,15 +12,13 @@ y = mapreduce(λi -> rand(Poisson(λi), nd), vcat, λ)
 x = repeat(logpop, inner=nd)
 idx = repeat(collect(1:ns), inner=nd)
 
-lazyarray(f, x) = LazyArray(Base.broadcasted(f, x))
-
 @model function dppl_hier_poisson(y, x, idx, ns)
     a0 ~ Normal(0, 10)
     a1 ~ Normal(0, 1)
     a0_sig ~ truncated(Cauchy(0, 1); lower=0)
-    a0s ~ filldist(Normal(0, a0_sig), ns)
+    a0s ~ product_distribution(fill(Normal(0, a0_sig), ns))
     alpha = a0 .+ a0s[idx] .+ a1 * x
-    y ~ arraydist(lazyarray(LogPoisson, alpha))
+    y ~ product_distribution(LogPoisson.(alpha))
 end
 
 model = dppl_hier_poisson(y, x, idx, ns)

models/dppl_high_dim_gauss.jl

@@ -1,5 +1,5 @@
 @model function dppl_high_dim_gauss(D)
-    m ~ filldist(Normal(0, 1), D)
+    m ~ product_distribution(fill(Normal(0, 1), D))
 end
 
 model = dppl_high_dim_gauss(10_000)

models/dppl_hmm_semisup.jl

@@ -28,8 +28,8 @@ for t in 2:T_unsup
 end
 
 @model function dppl_hmm_semisup(K, T, T_unsup, w, z, u, alpha, beta)
-    theta ~ filldist(Dirichlet(alpha), K)
-    phi ~ filldist(Dirichlet(beta), K)
+    theta ~ product_distribution(fill(Dirichlet(alpha), K))
+    phi ~ product_distribution(fill(Dirichlet(beta), K))
     for t in 1:T
         w[t] ~ Categorical(phi[:, z[t]]);
     end

models/dppl_lda.jl

@@ -21,8 +21,8 @@ for i in 1:m
 end
 
 @model function dppl_lda(k, m, w, doc, alpha, beta)
-    theta ~ filldist(Dirichlet(alpha), m)
-    phi ~ filldist(Dirichlet(beta), k)
+    theta ~ product_distribution(fill(Dirichlet(alpha), m))
+    phi ~ product_distribution(fill(Dirichlet(beta), k))
     log_phi_dot_theta = log.(phi * theta)
     @addlogprob! sum(log_phi_dot_theta[CartesianIndex.(w, doc)])
 end

models/dppl_logistic_regression.jl

@@ -1,21 +1,19 @@
 using StatsFuns: logistic
-using LazyArrays
+
+function safelogistic(x::T) where {T}
+    logistic(x) * (1 - 2 * eps(T)) + eps(T)
+end
 
 d, n = 100, 10_000
 X = randn(d, n)
 w = randn(d)
 y = Int.(logistic.(X' * w) .> 0.5)
 
-function safelogistic(x::T) where {T}
-    logistic(x) * (1 - 2 * eps(T)) + eps(T)
-end
-
-lazyarray(f, x) = LazyArray(Base.broadcasted(f, x))
 
 @model function dppl_logistic_regression(Xt, y)
     N, D = size(Xt)
-    w ~ filldist(Normal(), D)
-    y ~ arraydist(lazyarray(x -> Bernoulli(safelogistic(x)), Xt * w))
+    w ~ product_distribution(Normal.(zeros(D)))
+    y ~ product_distribution(Bernoulli.(safelogistic.(Xt * w)))
 end
 
 model = dppl_logistic_regression(X', y)

models/dppl_naive_bayes.jl

@@ -16,12 +16,12 @@ image = transform(pca, image_raw)
 # Take only the first 1000 images and vectorise
 N = 1000
 image_subset = image[:, 1:N]'
-image_vec = vec(image_subset[:, :])
+image_vec = image_subset[:, :]
 labels = labels[1:N]
 
 @model function dppl_naive_bayes(image_vec, labels, C, D)
-    m ~ filldist(Normal(0, 10), C, D)
-    image_vec ~ MvNormal(vec(m[labels, :]), I)
+    m ~ product_distribution(fill(Normal(0, 10), C, D))
+    image_vec ~ product_distribution(Normal.(m[labels, :]))
 end
 
 model = dppl_naive_bayes(image_vec, labels, C, D)