Merge pull request #510 from alan-turing-institute/dev

For a 0.17.1 release

Project.toml

@@ -1,7 +1,7 @@
 name = "MLJBase"
 uuid = "a7f614a8-145f-11e9-1d2a-a57a1082229d"
 authors = ["Anthony D. Blaom <[email protected]>"]
-version = "0.17.0"
+version = "0.17.1"
 
 [deps]
 CategoricalArrays = "324d7699-5711-5eae-9e2f-1d82baa6b597"
@@ -39,7 +39,7 @@ InvertedIndices = "^1"
 JLSO = "^2.1,^2.2"
 JSON = "^0.21"
 LossFunctions = "0.5, 0.6"
-MLJModelInterface = "^0.3.8,^0.4"
+MLJModelInterface = "^0.4"
 MLJScientificTypes = "^0.4.1"
 Missings = "^0.4"
 OrderedCollections = "^1.1"

src/MLJBase.jl

@@ -81,6 +81,7 @@ using StatisticalTraits
 for trait in StatisticalTraits.TRAITS
     eval(:(export $trait))
 end
+export implemented_methods # defined here and not in StatisticalTraits
 
 # MLJ model hierarchy
 export MLJType, Model, Supervised, Unsupervised,
@@ -105,15 +106,6 @@ export fit, update, update_data, transform, inverse_transform,
     predict_mode, predict_mean, predict_median, predict_joint,
     evaluate, clean!
 
-# model traits
-export input_scitype, output_scitype, target_scitype,
-    is_pure_julia, package_name, package_license,
-    load_path, package_uuid, package_url,
-    is_wrapper, supports_weights, supports_online,
-    docstring, name, is_supervised,
-    prediction_type, implemented_methods, hyperparameters,
-    hyperparameter_types, hyperparameter_ranges
-
 # data operations
 export matrix, int, classes, decoder, table,
     nrows, selectrows, selectcols, select

src/resampling.jl

@@ -706,6 +706,9 @@ end
 
 @static if VERSION >= v"1.3.0-DEV.573"
 
+
+_caches_data(::Machine{M, C}) where {M, C} = C # determines if an instantiated machine caches data
+
 function _evaluate!(func, mach, accel::CPUThreads, nfolds, verbosity)
 
     nthreads = Threads.nthreads()
@@ -735,9 +738,12 @@ function _evaluate!(func, mach, accel::CPUThreads, nfolds, verbosity)
             end
         end
         clean!(mach.model)
-        #One tmach for each task:
-        machines = [mach, [machine(mach.model, mach.args...) for
-                           _ in 2:length(partitions)]...]
+        #One tmach for each task:       
+        machines = vcat(mach, [
+           machine(mach.model, mach.args...; cache = _caches_data(mach)) 
+           for _ in 2:length(partitions)
+ 	])
+
         @sync for (i, parts) in enumerate(partitions)
             Threads.@spawn begin
                 results[i] = mapreduce(vcat, parts) do k