IBM · ppalmes · Aug 21, 2025 · Aug 21, 2025
diff --git a/AutoAD/Project.toml b/AutoAD/Project.toml
@@ -0,0 +1,4 @@
+name = "AutoAD"
+uuid = "a90a1e1f-13cb-41ec-be1f-24b06fd2b474"
+authors = ["Paulito Palmes, PhD <ppalmes@gmail.com>"]
+version = "0.1.0"
diff --git a/AutoAD/src/AutoAD.jl b/AutoAD/src/AutoAD.jl
@@ -0,0 +1,68 @@
+module AutoAD
+
+greet() = print("Hello World!")
+
+using Reexport
+@reexport using AutoMLPipeline
+@reexport using AMLPipelineBase
+
+using CSV
+using DataFrames
+using AMLPipelineBase.AbsTypes
+export fit, fit!, transform, transform!, fit_transform, fit_transform!
+import AMLPipelineBase.AbsTypes: fit!, transform!, fit, transform
+using AMLPipelineBase: AbsTypes, Utils
+
+export get_iris
+
+
+function get_iris()
+  iris = CSV.read(joinpath(Base.@__DIR__, "../../data", "iris.csv"), DataFrame)
+  return iris
+end
+
+# -------------
+include("autoclassification.jl")
+using .AutoClassifications
+export AutoClassification
+
+include("autoregression.jl")
+using .AutoRegressions
+export AutoRegression
+
+include("skanomalydetector.jl")
+using .SKAnomalyDetectors
+export SKAnomalyDetector, skanomalydetectors
+export skaddriver
+
+include("caretanomalydetector.jl")
+using .CaretAnomalyDetectors
+export CaretAnomalyDetectors, CaretAnomalyDetector
+export caretdriver
+
+
+include("carettspredictor.jl")
+using .CaretTSPredictors
+export CaretTSPredictor, carettsdriver
+
+include("autoanomalydetection.jl")
+using .AutoAnomalyDetections
+export AutoAnomalyDetection
+export autoaddriver
+
+include("automlflowclassification.jl")
+using .AutoMLFlowClassifications
+export AutoMLFlowClassification
+export mlfcldriver
+
+include("automlflowregression.jl")
+using .AutoMLFlowRegressions
+export AutoMLFlowRegression
+export mlfregdriver
+
+include("automlflowanomalydetection.jl")
+using .AutoMLFlowAnomalyDetections
+export AutoMLFlowAnomalyDetection
+export mlfaddriver
+
+end # module AutoAD
diff --git a/AutoAI/src/autoanomalydetection.jl → AutoAD/src/autoanomalydetection.jl b/AutoAI/src/autoanomalydetection.jl → AutoAD/src/autoanomalydetection.jl
@@ -1,8 +1,7 @@
 module AutoAnomalyDetections
 using Distributed
-using AutoMLPipeline
+using ..AutoMLPipeline
 using DataFrames: DataFrame, nrow, rename!
-using AutoMLPipeline: score
 using Random
 using Statistics
 using ..AbsTypes

diff --git a/AutoAI/src/automlflowanomalydetection.jl → AutoAD/src/automlflowanomalydetection.jl b/AutoAI/src/automlflowanomalydetection.jl → AutoAD/src/automlflowanomalydetection.jl
@@ -11,7 +11,6 @@ using Random
 using ..AbsTypes
 using ..Utils
 using ..AutoAnomalyDetections
-using ..AutoMLPipeline: getiris
 
 import ..AbsTypes: fit, fit!, transform, transform!
 export fit, fit!, transform, transform!

diff --git a/AutoAI/src/caretanomalydetector.jl → AutoAD/src/caretanomalydetector.jl b/AutoAI/src/caretanomalydetector.jl → AutoAD/src/caretanomalydetector.jl
diff --git a/AutoAI/src/carettspredictor.jl → AutoAD/src/carettspredictor.jl b/AutoAI/src/carettspredictor.jl → AutoAD/src/carettspredictor.jl
diff --git a/AutoAI/src/skanomalydetector.jl → AutoAD/src/skanomalydetector.jl b/AutoAI/src/skanomalydetector.jl → AutoAD/src/skanomalydetector.jl
diff --git a/AutoAD/test/runtests.jl b/AutoAD/test/runtests.jl
@@ -0,0 +1,11 @@
+module TestAutoAI
+using Distributed
+
+nprocs() == 1 && addprocs()
+@everywhere using AutoAI
+
+@everywhere include("./test_automl.jl")
+@everywhere include("./test_caret_anomalydetector.jl")
+include("./test_skanomalydetector.jl")
+@everywhere include("./test_caret_tspredictor.jl")
+end
diff --git a/AutoAI/test/test_caret_anomalydetector.jl → AutoAD/test_caret_anomalydetector.jl b/AutoAI/test/test_caret_anomalydetector.jl → AutoAD/test_caret_anomalydetector.jl
diff --git a/AutoAI/test/test_caret_tspredictor.jl → AutoAD/test_caret_tspredictor.jl b/AutoAI/test/test_caret_tspredictor.jl → AutoAD/test_caret_tspredictor.jl
diff --git a/AutoAI/test/test_skanomalydetector.jl → AutoAD/test_skanomalydetector.jl b/AutoAI/test/test_skanomalydetector.jl → AutoAD/test_skanomalydetector.jl
diff --git a/AutoAI/Project.toml b/AutoAI/Project.toml
@@ -11,15 +11,11 @@ CondaPkg = "992eb4ea-22a4-4c89-a5bb-47a3300528ab"
 DataFrames = "a93c6f00-e57d-5684-b7b6-d8193f3e46c0"
 Dates = "ade2ca70-3891-5945-98fb-dc099432e06a"
 Distributed = "8ba89e20-285c-5b6f-9357-94700520ee1b"
-GR = "28b8d3ca-fb5f-59d9-8090-bfdbd6d07a71"
-IJulia = "7073ff75-c697-5162-941a-fcdaad2a7d2a"
-K8sClusterManagers = "5aeab163-63d2-4171-9fbf-e22244d80acb"
 PDFmerger = "3beb2ed1-af7d-458f-b727-6e9beb3586c0"
-PackageCompiler = "9b87118b-4619-50d2-8e1e-99f35a4d4d9d"
 Plots = "91a5bcdd-55d7-5caf-9e0b-520d859cae80"
-PrecompileTools = "aea7be01-6a6a-4083-8856-8a6e6704d82a"
 PythonCall = "6099a3de-0909-46bc-b1f4-468b9a2dfc0d"
 Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
+Reexport = "189a3867-3050-52da-a836-e630ba90ab69"
 Serialization = "9e88b42a-f829-5b0c-bbe9-9e923198166b"
 Statistics = "10745b16-79ce-11e8-11f9-7d13ad32a3b2"
 StatsBase = "2913bbd2-ae8a-5f71-8c99-4fb6c76f3a91"
@@ -33,19 +29,21 @@ CondaPkg = "0.2.29"
 DataFrames = "1.7.0"
 Dates = "1.11.0"
 Distributed = "1.11.0"
-GR = "0.73.16"
-IJulia = "1.29.0"
-K8sClusterManagers = "0.1.5"
 PDFmerger = "0.3.3"
-PackageCompiler = "2.2.1"
 Plots = "1.40.18"
-PrecompileTools = "1.2.1"
 PythonCall = "0.9.25"
 Random = "1.11.0"
+Reexport = "1.2.2"
 Serialization = "1.11.0"
 Statistics = "1.11.1"
 StatsBase = "0.34.5"
 Test = "1.11.0"
 
 [extras]
+CondaPkg = "992eb4ea-22a4-4c89-a5bb-47a3300528ab"
+Distributed = "8ba89e20-285c-5b6f-9357-94700520ee1b"
+Statistics = "10745b16-79ce-11e8-11f9-7d13ad32a3b2"
 Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
+
+[targets]
+test = ["Distributed", "Statistics", "CondaPkg", "Test"]
diff --git a/AutoAI/src/AutoAI.jl b/AutoAI/src/AutoAI.jl
@@ -1,13 +1,15 @@
 module AutoAI
 
+using Reexport
+@reexport using AutoMLPipeline
+@reexport using AMLPipelineBase
+
 using CSV
 using DataFrames
-using AMLPipelineBase
 using AMLPipelineBase.AbsTypes
 export fit, fit!, transform, transform!, fit_transform, fit_transform!
 import AMLPipelineBase.AbsTypes: fit!, transform!, fit, transform
 using AMLPipelineBase: AbsTypes, Utils
-using AutoMLPipeline
 
 export get_iris
 
@@ -26,26 +28,6 @@ include("autoregression.jl")
 using .AutoRegressions
 export AutoRegression
 
-include("skanomalydetector.jl")
-using .SKAnomalyDetectors
-export SKAnomalyDetector, skanomalydetectors
-export skaddriver
-
-include("caretanomalydetector.jl")
-using .CaretAnomalyDetectors
-export CaretAnomalyDetectors, CaretAnomalyDetector
-export caretdriver
-
-
-include("carettspredictor.jl")
-using .CaretTSPredictors
-export CaretTSPredictor, carettsdriver
-
-include("autoanomalydetection.jl")
-using .AutoAnomalyDetections
-export AutoAnomalyDetection
-export autoaddriver
-
 include("automlflowclassification.jl")
 using .AutoMLFlowClassifications
 export AutoMLFlowClassification
@@ -56,9 +38,4 @@ using .AutoMLFlowRegressions
 export AutoMLFlowRegression
 export mlfregdriver
 
-include("automlflowanomalydetection.jl")
-using .AutoMLFlowAnomalyDetections
-export AutoMLFlowAnomalyDetection
-export mlfaddriver
-
 end # module AutoAI
diff --git a/AutoAI/src/autoad.jl b/AutoAI/src/autoad.jl
diff --git a/AutoAI/src/autoclassification.jl b/AutoAI/src/autoclassification.jl
@@ -3,10 +3,9 @@ module AutoClassifications
 
 
 using Distributed
-using AutoMLPipeline
 using DataFrames: DataFrame
-using AutoMLPipeline: score
 using Random
+using ..AutoMLPipeline
 using ..AbsTypes
 using ..Utils
 
@@ -44,7 +43,7 @@ mutable struct AutoClassification <: Workflow
   function AutoClassification(args=Dict())
     default_args = Dict(
       :name => "autoclass",
-      :complexity => "high",
+      :complexity => "low",
       :prediction_type => "classification",
       :nfolds => 3,
       :metric => "balanced_accuracy_score",

diff --git a/AutoAI/src/automlflowclassification.jl b/AutoAI/src/automlflowclassification.jl
@@ -110,26 +110,26 @@ function mlfcldriver()
   X = df[:, 1:end-1]
   Y = df[:, end] |> collect
 
-  #mlfclass = AutoMLFlowClassification(Dict(:url => url))
-  #Yc = fit_transform!(mlfclass, X, Y)
-  #println("accuracy = ", mean(Y .== Yc))
+  mlfclass = AutoMLFlowClassification(Dict(:url => url))
+  Yc = fit_transform!(mlfclass, X, Y)
+  println("accuracy = ", mean(Y .== Yc))
 
-  #  newmfclass = AutoMLFlowClassification(Dict(:url => url, :impl_args => Dict(:nfolds => 2)))
-  #  Yc = fit_transform!(newmfclass, X, Y)
-  #  println("accuracy = ", mean(Y .== Yc))
+  newmfclass = AutoMLFlowClassification(Dict(:url => url, :impl_args => Dict(:nfolds => 2)))
+  Yc = fit_transform!(newmfclass, X, Y)
+  println("accuracy = ", mean(Y .== Yc))
 
   nclass = AutoMLFlowClassification(Dict(:url => url))
   nclass.model[:automodel](; nfolds=2)
   Yc = fit_transform!(nclass, X, Y)
   println("accuracy = ", mean(Y .== Yc))
 
-  ## test prediction using exisiting trained model from artifacts
-  #run_id = mlfclass.model[:run_id]
-  #newmfclass = AutoMLFlowClassification(Dict(:run_id => run_id, :url => url))
-  #newmfclass = AutoMLFlowClassification(Dict(:url => url))
-  #newmfclass(; run_id=run_id)
-  #Yn = transform!(newmfclass, X)
-  #println("accuracy = ", mean(Yn .== Y))
+  # test prediction using exisiting trained model from artifacts
+  run_id = mlfclass.model[:run_id]
+  newmfclass = AutoMLFlowClassification(Dict(:run_id => run_id, :url => url))
+  newmfclass = AutoMLFlowClassification(Dict(:url => url))
+  newmfclass(; run_id=run_id)
+  Yn = transform!(newmfclass, X)
+  println("accuracy = ", mean(Yn .== Y))
 
   return nothing
 end

diff --git a/AutoAI/src/autoregression.jl b/AutoAI/src/autoregression.jl
@@ -1,12 +1,9 @@
 module AutoRegressions
 # classification search blocks
-
-
 using Distributed
-using AutoMLPipeline
 using DataFrames: DataFrame
-using AutoMLPipeline: score
 using Random
+using ..AutoMLPipeline
 using ..AbsTypes
 using ..Utils
 
@@ -44,7 +41,7 @@ mutable struct AutoRegression <: Workflow
   function AutoRegression(args=Dict())
     default_args = Dict(
       :name => "autoreg",
-      :complexity => "high",
+      :complexity => "low",
       :prediction_type => "regression",
       :nfolds => 3,
       :metric => "mean_squared_error",

diff --git a/AutoAI/test/Project.toml b/AutoAI/test/Project.toml
diff --git a/AutoAI/test/runtests.jl b/AutoAI/test/runtests.jl
@@ -1,11 +1,22 @@
 module TestAutoAI
+using CondaPkg
 using Distributed
 
 nprocs() == 1 && addprocs()
+
 @everywhere using AutoAI
 
-@everywhere include("./test_automl.jl")
-@everywhere include("./test_caret_anomalydetector.jl")
-include("./test_skanomalydetector.jl")
-@everywhere include("./test_caret_tspredictor.jl")
+include("./test_automl.jl")
+
+const tmpdir = tempdir()
+const mlflowpath = joinpath(CondaPkg.envdir(), "bin", "mlflow")
+const backendpath = joinpath("sqlite:///", tmpdir, "mlflow.db")
+const artifactpath = joinpath(tmpdir, "mlruns")
+cmd = "$mlflowpath server --host 127.0.0.1 --port 8080 --backend-store-uri $backendpath --default-artifact-root $artifactpath "
+mlflowprocess = run(Cmd(`sh -c "$cmd"`), wait=false)
+
+include("./test_automlflow.jl")
+
+kill(mlflowprocess)
+
 end
diff --git a/AutoAI/test/test_automl.jl b/AutoAI/test/test_automl.jl
@@ -1,10 +1,6 @@
 module TestAutoML
 using Test
-using CSV
 using AutoAI
-using DataFrames: DataFrame
-using Serialization
-using Distributed
 using Statistics
 
 const df = get_iris()