add sugeno inference system (#5)

* rename FuzzyInferenceSystem to MamdaniInferenceSystem

* initial Sugeno inference system implementation

* add plotting for sugeno membership functions

* Update sugeno system

- better API docs
- differentiate between membership functions and sugeno outputs

* add plotting tests and sugeno tutorial

* fix typos

Author: lucaferranti

README.md

@@ -32,7 +32,7 @@ using FuzzyLogic
 ## Quickstart example
 
 ```julia
-fis = @fis function tipper(service, food)::tip
+fis = @mamfis function tipper(service, food)::tip
     service := begin
       domain = 0:10
       poor = GaussianMF(0.0, 1.5)

docs/make.jl

@@ -101,7 +101,7 @@ generate_memberships()
 makedocs(;
          modules = [FuzzyLogic], authors = "Luca Ferranti",
          sitename = "FuzzyLogic.jl",
-         doctest = false,
+         doctest = false, checkdocs = :exports, strict = true,
          format = Documenter.HTML(;
                                   assets = [DocThemeIndigo.install(FuzzyLogic)],
                                   prettyurls = IS_CI, collapselevel = 1,
@@ -110,10 +110,17 @@ makedocs(;
              "Home" => "index.md",
              "Tutorials" => [
                  "Build a Mamdani inference system" => "tutorials/mamdani.md",
+                 "Build a Sugeno inference system" => "tutorials/sugeno.md",
              ],
              "API" => [
-                 "Inference system" => "api/fis.md",
+                 "Inference system API" => [
+                     "Types" => "api/fis.md",
+                     "Logical connectives" => "api/logical.md",
+                     "Aggregation methods" => "api/aggregation.md",
+                     "Defuzzification methods" => "api/defuzzification.md",
+                 ],
                  "Membership functions" => "api/memberships.md",
+                 "Reading/Writing" => "api/readwrite.md",
              ],
              "Contributor's Guide" => "contributing.md",
              "Release notes" => "changelog.md",

docs/src/api/aggregation.md

@@ -0,0 +1,6 @@
+# Aggregation methods
+
+```@autodocs
+Modules = [FuzzyLogic]
+Filter = t -> typeof(t) === DataType && t <: FuzzyLogic.AbstractAggregator
+```

docs/src/api/defuzzification.md

@@ -0,0 +1,6 @@
+# Defuzzification methods
+
+```@autodocs
+Modules = [FuzzyLogic]
+Filter = t -> typeof(t) === DataType && t <: FuzzyLogic.AbstractDefuzzifier
+```

docs/src/api/fis.md

@@ -1,46 +1,15 @@
 # Inference Systems
 
-## Inference systems
+## Mamdani inference system
 
 ```@docs
-FuzzyInferenceSystem
-@fis
+MamdaniFuzzySystem
 ```
 
-
-## Inference system settings
-
-### Conjuction methods
-
-```@docs
-MinAnd
-ProdAnd
-```
-
-### Disjunction methods
-
-```@docs
-MaxOr
-ProbSumOr
-```
-
-### Implication methods
-
-```@docs
-MinImplication
-ProdImplication
-```
-
-### Aggregation methods
-
-```@docs
-MaxAggregator
-ProbSumAggregator
-```
-
-### Defuzzification methods
+## Sugeno inference system
 
 ```@docs
-CentroidDefuzzifier
-BisectorDefuzzifier
+SugenoFuzzySystem
+ConstantSugenoOutput
+LinearSugenoOutput
 ```

docs/src/api/logical.md

@@ -0,0 +1,22 @@
+# Logical connectives
+
+## Conjuction methods
+
+```@autodocs
+Modules = [FuzzyLogic]
+Filter = t -> typeof(t) === DataType && t <: FuzzyLogic.AbstractAnd
+```
+
+## Disjunction methods
+
+```@autodocs
+Modules = [FuzzyLogic]
+Filter = t -> typeof(t) === DataType && t <: FuzzyLogic.AbstractOr
+```
+
+## Implication methods
+
+```@autodocs
+Modules = [FuzzyLogic]
+Filter = t -> typeof(t) === DataType && t <: FuzzyLogic.AbstractImplication
+```

docs/src/api/readwrite.md

@@ -0,0 +1,8 @@
+# Reading / writing functionalities
+
+## Parse Julia code
+
+```@docs
+@mamfis
+@sugfis
+```

docs/src/changelog.md

@@ -15,6 +15,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
   - min and prod implication
   - max and probabilistic sum aggregation method
   - centroid and bisector defuzzifier
+  - linear and constant output for Sugeno
 - initial plotting functionalities
   - plotting variables and membership functions
   - plotting rules of fuzzy inference system

docs/src/index.md

@@ -32,7 +32,7 @@ using FuzzyLogic
 ## Quickstart example
 
 ```julia
-fis = @fis function tipper(service in 0:10, food in 0:10)::{tip in 0:30}
+fis = @mamfis function tipper(service in 0:10, food in 0:10)::{tip in 0:30}
     poor = GaussianMF(0.0, 1.5)
     good = GaussianMF(5.0, 1.5)
     excellent = GaussianMF(10.0, 1.5)

docs/src/literate/tutorials/mamdani.jl

@@ -24,10 +24,10 @@ First, we need to load the library.
 
 using FuzzyLogic
 
-# The Mamdani inference system can be constructed with the [`@fis`](@ref) macro.
+# The Mamdani inference system can be constructed with the [`@mamfis`](@ref) macro.
 # We will first give a full example and then explain every step.
 
-fis = @fis function tipper(service, food)::tip
+fis = @mamfis function tipper(service, food)::tip
     service := begin
         domain = 0:10
         poor = GaussianMF(0.0, 1.5)
@@ -61,7 +61,7 @@ fis = @fis function tipper(service, food)::tip
 end
 
 #=
-As you can see, defining a fuzzy inference system with `@fis` looks a lot like writing
+As you can see, defining a fuzzy inference system with `@mamfis` looks a lot like writing
 Julia code. Let us now take a closer look at the components. The first line
 
 ```julia
@@ -133,14 +133,13 @@ and = ProdAnd
 are used to set the settings of the inference system. For a Mamdani inference system,
 the following settings are available
 
-- `and`: algorithm to evaluate `&&`. Must be one of the available [Conjuction methods](@ref).
-- `or`: algorithm to evaluate `||`. Must be one of the available [Disjunction methods](@ref).
-- `implication`: algorithm to evalute `-->`. Must be one of the available [Implication methods](@ref).
-- `aggregato`: algorithm to perform outputs aggregation. Must be one of the available [Aggregation methods](@ref).
-- `defuzzifier`: algorithm to perform defuzzification. Must be one of the available [Defuzzification methods](@ref).
+- `and`: algorithm to evaluate `&&`. Must be one of the available [Conjuction methods](@ref). Default [`MinAnd`](@ref).
+- `or`: algorithm to evaluate `||`. Must be one of the available [Disjunction methods](@ref). Default [`MaxOr`](@ref)
+- `implication`: algorithm to evalute `-->`. Must be one of the available [Implication methods](@ref). Default [`MinImplication`](@ref).
+- `aggregato`: algorithm to perform outputs aggregation. Must be one of the available [Aggregation methods](@ref). Default [`MaxAggregator`](@ref).
+- `defuzzifier`: algorithm to perform defuzzification. Must be one of the available [Defuzzification methods](@ref). Default [`CentroidDefuzzifier`](@ref).
 
-Each of the previous settings has a default value that will be used if the setting is not specified.
-The default values are listed in [`FuzzyInferenceSystem`](@ref) documentation.
+If one of the above settings is not specified, the corresponding default value is used.
 
 Some of the above settings may have internal parameters.
 For example, [`CentroidDefuzzifier`](@ref) has an integer parameter `N`, the number of points used to perform numerical integration.