Remove Model and rewrite turing_model (#45)

* Remove `Model` and rewrite `turing_model`

* Apply suggestions from code review

Co-authored-by: Jose Storopoli <[email protected]>

Co-authored-by: Jose Storopoli <[email protected]>

Author: devmotion

Project.toml

@@ -1,7 +1,7 @@
 name = "TuringGLM"
 uuid = "0004c1f4-53c5-4d43-a221-a1dac6cf6b74"
 authors = ["Jose Storopoli <[email protected]>, Rik Huijzer <[email protected]>, and contributors"]
-version = "1.0.0"
+version = "2.0.0"
 
 [deps]
 Distributions = "31c24e10-a181-5473-b8eb-7969acd0382f"

docs/src/index.md

@@ -48,13 +48,13 @@ The most popular ones are `DataFrame`s and `NamedTuple`s.
 TuringGLM supports non-hiearchical and hierarchical models.
 For hierarchical models, only single random-intercept hierarchical models are supported.
 
-For likelihoods, `TuringGLM.jl` supports:
+Currently, for likelihoods `TuringGLM.jl` supports:
 
-* `Gaussian()` (the default if not specified): linear regression
-* `Student()`: robust linear regression
-* `Logistic()`: logistic regression
-* `Pois()`: Poisson count data regression
-* `NegBin()`: negative binomial robust count data regression
+* `Normal` (the default if not specified): linear regression
+* `TDist`: robust linear regression
+* `Bernoulli`: logistic regression
+* `Poisson`: Poisson count data regression
+* `NegativeBinomial`: negative binomial robust count data regression
 
 ## Tutorials
 

docs/src/tutorials/custom_priors.jl

@@ -64,7 +64,7 @@ priors = CustomPrior(Normal(0, 2.5), Normal(10, 20), nothing);
 
 # ╔═╡ 56498ac7-3476-42eb-9c12-078562fff51d
 md"""
-We instantiate our model with `turing_model` without specifying any model, thus the default model will be used: `Gaussian()`.
+We instantiate our model with `turing_model` without specifying any model, thus the default model will be used (`model=Normal`).
 Notice that we are specifying the `priors` keyword argument:
 """
 

docs/src/tutorials/hierarchical_models.jl

@@ -59,7 +59,7 @@ fm = @formula(y ~ (1 | cheese) + background)
 
 # ╔═╡ 934b7ec9-d4b3-435c-876f-49215dca8809
 md"""
-We instantiate our model with `turing_model` without specifying any model, thus the default model will be used: `Gaussian()`
+We instantiate our model with `turing_model` without specifying any model, thus the default model will be used (`model=Normal`):
 """
 
 # ╔═╡ 5fc736a2-068e-4deb-9a77-bd21d93f6f32

docs/src/tutorials/linear_regression.jl

@@ -57,7 +57,7 @@ fm = @formula(kid_score ~ mom_hs * mom_iq)
 
 # ╔═╡ 43d63761-adf5-4a52-b996-4ad3adfb35d0
 md"""
-Next, we instantiate our model with `turing_model` without specifying any model, thus the default model will be used: `Gaussian()`:
+Next, we instantiate our model with `turing_model` without specifying any model, thus the default model will be used (`model=Normal`):
 """
 
 # ╔═╡ 55b91963-001e-4753-93a6-2fa64190f353

docs/src/tutorials/logistic_regression.jl

@@ -58,11 +58,11 @@ fm = @formula(switch ~ dist + arsenic + assoc + educ)
 
 # ╔═╡ 2ebfa422-f8a5-44d3-8f2c-a34d7832d3f2
 md"""
-Now we instantiate our model with `turing_model` passing a third argument `Logistic()` to indicate that the model is a logistic regression:
+Now we instantiate our model with `turing_model` passing a keyword argument `model=Bernoulli` to indicate that the model is a logistic regression:
 """
 
 # ╔═╡ 1f1158ce-73f0-49fd-a48a-af3b36376030
-model = turing_model(fm, wells, Logistic());
+model = turing_model(fm, wells; model=Bernoulli);
 
 # ╔═╡ ebe074cb-7bad-4b52-9c9d-b9752af4bedd
 chn = sample(model, NUTS(), 2_000);

docs/src/tutorials/negative_binomial_regression.jl

@@ -57,11 +57,11 @@ fm = @formula(y ~ roach1 + treatment + senior)
 
 # ╔═╡ 124aeb01-7661-4402-a2fa-77d2771b686c
 md"""
-We instantiate our model with `turing_model` passing a third argument `NegBin()` to indicate that the model is a negative binomial regression:
+We instantiate our model with `turing_model` passing a keyword argument `model=NegativeBinomial` to indicate that the model is a negative binomial regression:
 """
 
 # ╔═╡ 7ea3a50d-3b0d-4cfd-b311-806d6ae59c1a
-model = turing_model(fm, roaches, NegBin());
+model = turing_model(fm, roaches; model=NegativeBinomial);
 
 # ╔═╡ 597ce5d7-df3e-44e3-a154-e06e64894854
 chn = sample(model, NUTS(), 2_000);

docs/src/tutorials/poisson_regression.jl

@@ -57,12 +57,11 @@ fm = @formula(y ~ roach1 + treatment + senior)
 
 # ╔═╡ 65f4f379-e9a5-4571-bc80-77c024f3f560
 md"""
-We instantiate our model with `turing_model` passing a third argument `Pois()` to
-indicate that the model is a Poisson Regression
+We instantiate our model with `turing_model` passing a keyword argument `model=Poisson` to indicate that the model is a Poisson Regression:
 """
 
 # ╔═╡ 9147ed9e-a047-42ca-aa36-f522fad8388b
-model = turing_model(fm, roaches, Pois());
+model = turing_model(fm, roaches; model=Poisson);
 
 # ╔═╡ aeeac527-2c32-407f-84a1-912cc74f51b7
 md"""

docs/src/tutorials/robust_regression.jl

@@ -55,12 +55,11 @@ fm = @formula(kid_score ~ mom_hs * mom_iq)
 
 # ╔═╡ 4f9ff3fd-3f04-49a5-924e-aa23702e75a0
 md"""
-We instantiate our model with `turing_model` passing a third argument `Student()` to
-indicate that the model is a robust regression with the Student's t-distribution:
+We instantiate our model with `turing_model` passing a keyword argument `model=TDist` to indicate that the model is a robust regression with the Student's t-distribution:
 """
 
 # ╔═╡ 3f4241d4-1c76-4d7c-99c1-aaa2111385f9
-model = turing_model(fm, kidiq, Student());
+model = turing_model(fm, kidiq; model=TDist);
 
 # ╔═╡ 7bbe4fe4-bcaf-4699-88e4-0dff92250d30
 chn = sample(model, NUTS(), 2_000);

src/TuringGLM.jl

@@ -41,11 +41,9 @@ end
 include("utils.jl")
 include("data_constructors.jl")
 include("priors.jl")
-include("model.jl")
 include("turing_model.jl")
 
 export turing_model
 export CustomPrior, DefaultPrior
-export Gaussian, Student, Logistic, Pois, NegBin
 
 end # module