From 399d655b8ab35d09585404adb6a1b0aec13a6eca Mon Sep 17 00:00:00 2001
From: Hossein Pourbozorg <prbzrg@gmail.com>
Date: Fri, 13 Jan 2023 23:41:37 +0330
Subject: [PATCH] move example to a file (#110)

---
 README.md                  | 42 +-------------------------------------
 examples/simple_example.jl | 37 +++++++++++++++++++++++++++++++++
 2 files changed, 38 insertions(+), 41 deletions(-)
 create mode 100644 examples/simple_example.jl

diff --git a/README.md b/README.md
index 4c41be28..8ce1e8d8 100644
--- a/README.md
+++ b/README.md
@@ -29,44 +29,4 @@ using Pkg
 Pkg.add(url="https://github.com/impICNF/ICNF.jl")
 ```
 
-To use this package, here is an example:
-
-```julia
-using ICNF
-using Distributions, Flux
-using DifferentialEquations, SciMLSensitivity
-
-# Parameters
-nvars = 1
-n = 1024
-
-# Data
-data_dist = Beta(2.0, 4.0)
-r = rand(data_dist, nvars, n)
-
-# Model
-nn = Chain(
-    Dense(nvars => 4*nvars, tanh),
-    Dense(4*nvars => nvars, tanh),
-) |> f64
-icnf = RNODE{Float64, Array}(nn, nvars; tspan=(0.0, 4.0))
-
-# Training
-using DataFrames, MLJBase
-df = DataFrame(transpose(r), :auto)
-model = ICNFModel(icnf; opt_app=SciMLOptApp())
-mach = machine(model, df)
-fit!(mach)
-
-# Use It
-d = ICNFDist(icnf)
-actual_pdf = pdf.(data_dist, vec(r))
-estimated_pdf = pdf(d, r)
-new_data = rand(d, n)
-
-# Evaluation
-using LinearAlgebra, Distances
-n1 = norm(estimated_pdf - actual_pdf, 1) / n
-n2 = norm(estimated_pdf - actual_pdf, 2) / n
-tv_dis = totalvariation(estimated_pdf, actual_pdf) / n
-```
+See [`simple_example.jl`](examples/simple_example.jl) for a simple example of using this package.
diff --git a/examples/simple_example.jl b/examples/simple_example.jl
new file mode 100644
index 00000000..b1a3ebec
--- /dev/null
+++ b/examples/simple_example.jl
@@ -0,0 +1,37 @@
+using ICNF
+using Distributions, Flux
+using DifferentialEquations, SciMLSensitivity
+
+# Parameters
+nvars = 1
+n = 1024
+
+# Data
+data_dist = Beta(2.0, 4.0)
+r = rand(data_dist, nvars, n)
+
+# Model
+nn = Chain(
+    Dense(nvars => 4*nvars, tanh),
+    Dense(4*nvars => nvars, tanh),
+) |> f64
+icnf = RNODE{Float64, Array}(nn, nvars; tspan=(0.0, 4.0))
+
+# Training
+using DataFrames, MLJBase
+df = DataFrame(transpose(r), :auto)
+model = ICNFModel(icnf; opt_app=SciMLOptApp())
+mach = machine(model, df)
+fit!(mach)
+
+# Use It
+d = ICNFDist(icnf)
+actual_pdf = pdf.(data_dist, vec(r))
+estimated_pdf = pdf(d, r)
+new_data = rand(d, n)
+
+# Evaluation
+using LinearAlgebra, Distances
+n1 = norm(estimated_pdf - actual_pdf, 1) / n
+n2 = norm(estimated_pdf - actual_pdf, 2) / n
+tv_dis = totalvariation(estimated_pdf, actual_pdf) / n