Adding schema to the interface (#40)

* schema interafce
Co-authored-by: Shuhei Kadowaki <[email protected]>

Author: tlienart

.travis.yml

@@ -3,9 +3,7 @@ language: julia
 os:
   - linux
 julia:
-  - 1.1
-  - 1.2
-  - 1.3
+  - 1.0
   - 1.4
   - nightly
 matrix:

Project.toml

@@ -1,7 +1,7 @@
 name = "MLJModelInterface"
 uuid = "e80e1ace-859a-464e-9ed9-23947d8ae3ea"
 authors = ["Thibaut Lienart and Anthony Blaom"]
-version = "0.2.2"
+version = "0.2.3"
 
 [deps]
 Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
@@ -16,9 +16,10 @@ CategoricalArrays = "324d7699-5711-5eae-9e2f-1d82baa6b597"
 DataFrames = "a93c6f00-e57d-5684-b7b6-d8193f3e46c0"
 Distances = "b4f34e82-e78d-54a5-968a-f98e89d6e8f7"
 InteractiveUtils = "b77e0a4c-d291-57a0-90e8-8db25a27a240"
+MLJBase = "a7f614a8-145f-11e9-1d2a-a57a1082229d"
 MLJScientificTypes = "2e2323e0-db8b-457b-ae0d-bdfb3bc63afd"
 Tables = "bd369af6-aec1-5ad0-b16a-f7cc5008161c"
 Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
 
 [targets]
-test = ["Test", "Tables", "Distances", "CategoricalArrays", "InteractiveUtils", "DataFrames", "MLJScientificTypes"]
+test = ["Test", "Tables", "Distances", "CategoricalArrays", "InteractiveUtils", "DataFrames", "MLJScientificTypes", "MLJBase"]

src/data_utils.jl

@@ -123,6 +123,23 @@ classes(x) = classes(get_interface_mode(), x)
 
 classes(::LightInterface, x) = errlight("classes")
 
+# ------------------------------------------------------------------------
+# schema
+
+"""
+    schema(X)
+
+If `X` is tabular, inspect the column types and scitypes, otherwise return
+`nothing`.
+"""
+schema(X; kw...) = schema(get_interface_mode(), vtrait(X), X; kw...)
+
+schema(::Mode, ::Val{:other}, X; kw...) = nothing
+
+schema(::LightInterface, ::Val{:other}, X; kw...) = errlight("schema")
+
+schema(::LightInterface, ::Val{:table}, X; kw...) = errlight("schema")
+
 # ------------------------------------------------------------------------
 # decoder
 

test/data_utils.jl

@@ -58,6 +58,28 @@ end
     @test classes(x[1]) == ['a', 'b']
 end
 # ------------------------------------------------------------------------
+@testset "schema-light" begin
+    # throw error for any input anyway
+    setlight()
+    ary = rand(10, 3)
+    @test_throws M.InterfaceError M.schema(ary)
+    df = DataFrame(rand(10, 3))
+    @test_throws M.InterfaceError M.schema(df)
+end
+@testset "schema-full" begin
+    setfull()
+    ary = rand(10, 3)
+    @test M.schema(ary) === nothing
+    M.schema(::FI, ::Val{:table}, X; kw...) = MLJBase.schema(X; kw...) # this would be defined in MLJBase.jl
+    df = DataFrame(A = rand(10), B = categorical(rand('a':'c', 10)))
+    sch = M.schema(df)
+    @test sch.names == (:A, :B)
+    @test sch.types[1] <: Float64
+    @test sch.types[2] <: CategoricalValue
+    @test sch.scitypes[1] <: Continuous
+    @test sch.scitypes[2] <: Multiclass
+end
+# ------------------------------------------------------------------------
 @testset "decoder-light" begin
     setlight()
     x = 5

test/runtests.jl

@@ -2,6 +2,7 @@ using Test, MLJModelInterface
 using ScientificTypes, MLJScientificTypes
 using Tables, Distances, CategoricalArrays, InteractiveUtils
 import DataFrames: DataFrame
+import MLJBase
 
 const M  = MLJModelInterface
 const FI = M.FullInterface