change :auto to stationary, bump version to 1.0.0

Project.toml

@@ -1,7 +1,7 @@
 name = "PCCAPlus"
 uuid = "f48fc343-7e38-490c-be15-e66d68689cd5"
 authors = ["Alexander Sikorski"]
-version = "0.2.0"
+version = "1.0.0"
 
 [deps]
 Arpack = "7d9fca2a-8960-54d3-9f78-7d1dccf2cb97"

README.md

@@ -26,7 +26,7 @@ P = P ./ sum(P, dims=2) # sparse row stochastic matrix
 
 # solve the PCCA+ problem weighted with the stationary density 
 # and optimize for crispness, using the KrylovKit.jl eigensolver
-chi = pcca(P, 2; pi=:auto, optimize=true, solver=KrylovSolver())
+chi = pcca(P, 2; pi=:stationary, optimize=true, solver=KrylovSolver())
 ```
 
 For sparse matrix support, add either the `ArnoldiMethod.jl` or `KrylovKit.jl` and pass the corresponding `ArnoldiSolver()` or `KrylovSolver()` as a solver.

ext/KrylovExt.jl

@@ -1,10 +1,8 @@
 module KrylovExt
+
 import PCCAPlus
 import KrylovKit
 
-
-
-
 function PCCAPlus.schurvecs(T, n, israte, ::PCCAPlus.KrylovSolver)
     which = israte ? :LR : :LM
     R, Qs, = KrylovKit.schursolve(T, rand(size(T, 1)), n, which, KrylovKit.Arnoldi())

src/pccap.jl

@@ -7,17 +7,17 @@ using LinearAlgebra
 performs the pcca clustering on the transition matrix `T` with `n` clusters.
 Here `T` can be either a stochastic propagator with row sum 1 or a rate matrix with row sum 0.
 
-- `pi` is a density for weighting the result. `pi=:auto` uses the stationary density
+- `pi` is a density for weighting the result. `pi=:stationary` uses the stationary density
 - `optimize` uses the optimization from Roeblitz (2013) to improve crispness
 - `solver` is the solver to use for computing the schur decomposition. 
    The BaseSolver() is built in. ArnoldiSolver() and KrylovSolver() require ArnoldiMethod.jl and KrylovKit.jl respectively and provide support for sparse matrices.
 
-Returns the membership matrix `chi`.
+Returns the membership matrix `chi` assigning to each state a membership to one of the `n` clusters.
 """
 
 function pcca(T::AbstractMatrix, n::Integer; pi=nothing, optimize=false, solver=BaseSolver())
     israte = isratematrix(T)
-    if pi == :auto
+    if pi == :stationary
         pi = stationarydensity(T, israte)
     end
     X = schurvectors(T, pi, n, israte, solver)