deploy docs

README.md

@@ -2,4 +2,9 @@
 
 [![CI](https://github.com/szcf-weiya/MonotoneDecomposition.jl/actions/workflows/ci.yml/badge.svg)](https://github.com/szcf-weiya/MonotoneDecomposition.jl/actions/workflows/ci.yml) [![codecov](https://codecov.io/gh/szcf-weiya/MonotoneDecomposition.jl/branch/master/graph/badge.svg?token=VFznaxvMS5)](https://codecov.io/gh/szcf-weiya/MonotoneDecomposition.jl)
 
-*MonotoneDecomposition.jl* is a Julia package for monotone decomposition.
+*MonotoneDecomposition.jl* is a Julia package for monotone decomposition.
+
+Check the following paper for more details.
+
+> Lijun Wang, Xiaodan Fan, Hongyu Zhao, and Jun S. Liu. "Decomposition with Monotone B-splines: Fitting and Testing"
+>

docs/make.jl

@@ -4,7 +4,7 @@ using Documenter, MonotoneDecomposition
 using Literate
 indir = joinpath(@__DIR__, "..", "examples")
 outdir = joinpath(@__DIR__, "src", "examples")
-for file in ["gp.jl", "md_SE.jl"]
+for file in ["gp.jl", "md_SE.jl", "benchmark.jl", "benchmark_parallel.jl"]
     Literate.markdown(joinpath(indir, file), outdir; credit = false)
 end
 # using Pkg
@@ -16,7 +16,13 @@ makedocs(sitename="MonotoneDecomposition.jl",
             "Examples" => [
                 "Gaussian Process" => "examples/gp.md",
                 "Monotone Decomposition" => "examples/md_SE.md",
+                "Benchmarking" => "examples/benchmark.md",
+                "Benchmarking in Parallel" => "examples/benchmark_parallel.md"
             ],
             "API" => "api.md"
         ]
+)
+
+deploydocs(
+    repo = "github.com/szcf-weiya/MonotoneDecomposition.jl"
 )

examples/benchmark_parallel.jl

@@ -52,7 +52,7 @@ pmap(
         show_progress = f == "x^3" # keep one progressbar
     ),
     fs[idxf]
-)
+);
 
 # !!! tip "run from command line"
 #     ```

examples/md_SE.jl

@@ -12,10 +12,10 @@ x, y, x0, y0 = gen_data(100, 0.5, "SE_1")
 # ### `fixJ = true`
 
 # Pefrom Monotone Decomposition with Cubic B-splines, where the number of basis functions is chosen by cross-validation for cubic splines.
-ss = 10.0 .^ (-6:0.1:0)
-D, μmin, μs, errs, σerrs, yhat, yhatnew = 
-    cv_mono_decomp_cs(x, y, ss = ss, fixJ = true, x0 = x0);
-
+μs = 10.0 .^ (-6:0.1:0)
+D, μmin, errs, σerrs = 
+    cv_mono_decomp_cs(x, y, ss = μs, fixJ = true, x0 = x0);
+yhat, yhatnew = cubic_spline(D.workspace.J)(x, y, x0)
 # Plot it
 plot([x, y], [x0, y0], D, yhatnew, prefix_title = "SE (ℓ = 1, σ = 0.5): ")
 
@@ -25,13 +25,13 @@ cvplot(errs, σerrs, 1.0 * D.workspace.J:D.workspace.J, μs)
 # ### `fixJ = false`
 
 Js = 4:50
-D, μmin, μs, errs, σerrs = 
-    cv_mono_decomp_cs(x, y, ss = ss, fixJ = false, x0 = x0, Js = Js);
+D, μmin, errs, σerrs = 
+    cv_mono_decomp_cs(x, y, ss = μs, fixJ = false, x0 = x0, Js = Js);
 J, yhat, yhatnew = cv_cubic_spline(x, y, x0, Js = Js)
 plot([x, y], [x0, y0], D, yhatnew, prefix_title = "SE (ℓ = 1, σ = 0.5): ")
 
 # the heatmap of CV-error along the two parameter (J, μ) is as follows,
-cvplot(errs, σerrs, Js * 1.0, ss)
+cvplot(errs, σerrs, Js * 1.0, μs)
 
 # ## With Smoothing Splines