Merge pull request #21 from JuliaAI/dev

For a 0.3.0 release
JuliaAI · Apr 23, 2021 · 5943532 · 5943532
2 parents c488ecd + af9c484
commit 5943532
Show file tree

Hide file tree

Showing 3 changed files with 25 additions and 20 deletions.
diff --git a/Project.toml b/Project.toml
@@ -1,15 +1,15 @@
 name = "MLJIteration"
 uuid = "614be32b-d00c-4edb-bd02-1eb411ab5e55"
 authors = ["Anthony D. Blaom <[email protected]>"]
-version = "0.2.3"
+version = "0.3.0"
 
 [deps]
 IterationControl = "b3c1a2ee-3fec-4384-bf48-272ea71de57c"
 MLJBase = "a7f614a8-145f-11e9-1d2a-a57a1082229d"
 Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
 
 [compat]
-IterationControl = "0.3.3"
+IterationControl = "0.4"
 MLJBase = "0.17.7, 0.18"
 julia = "1"
 

diff --git a/src/controls.jl b/src/controls.jl
@@ -29,7 +29,9 @@ IterationControl.@create_docs(
 
 function IterationControl.update!(c::WithIterationsDo,
                                   ic_model,
-                                  verbosity, state...)
+                                  verbosity,
+                                  n,
+                                  state...)
     i = ic_model.n_iterations
     r = c.f(i)
     done = (c.stop_if_true && r isa Bool && r) ? true : false
@@ -83,7 +85,9 @@ IterationControl.@create_docs(
 
 function IterationControl.update!(c::WithEvaluationDo,
                                   ic_model,
-                                  verbosity, state...)
+                                  verbosity,
+                                  n,
+                                  state...)
     e = ic_model.evaluation
     r = c.f(e)
     done = (c.stop_if_true && r isa Bool && r) ? true : false
@@ -156,6 +160,7 @@ end
 function IterationControl.update!(control::CycleLearningRate,
                                   wrapper,
                                   verbosity,
+                                  ncycles,
                                   state = (n = 0,))
     n = state.n
     rates = n == 0 ? one_cycle(control) : state.learning_rates

diff --git a/test/controls.jl b/test/controls.jl
@@ -15,11 +15,11 @@ X, y = make_dummy(N=8);
     c = WithIterationsDo(f)
     m = MLJIteration.ICModel(machine(DummyIterativeModel(n=0), X, y), :n, 0)
     IC.train!(m, 2)
-    state = IC.update!(c, m, 0)
+    state = IC.update!(c, m, 0, 1)
     @test !state.done
     @test v == [2, ]
     IC.train!(m, 2)
-    state = IC.update!(c, m, 0, state)
+    state = IC.update!(c, m, 0, 2, state)
     @test !state.done
     @test v == [2, 4]
     @test IC.takedown(c, 0, state) == (done = false, log="")
@@ -29,11 +29,11 @@ X, y = make_dummy(N=8);
     c = WithIterationsDo(f2, stop_if_true=true)
     m = MLJIteration.ICModel(machine(DummyIterativeModel(n=0), X, y), :n, 0)
     IC.train!(m, 2)
-    state = IC.update!(c, m, 0)
+    state = IC.update!(c, m, 0, 1)
     @test !state.done
     @test v == [2, ]
     IC.train!(m, 2)
-    state = IC.update!(c, m, 0, state)
+    state = IC.update!(c, m, 0, 2, state)
     @test state.done
     @test v == [2, 4]
     @test IC.takedown(c, 0, state) ==
@@ -45,11 +45,11 @@ X, y = make_dummy(N=8);
     c = WithIterationsDo(f3, stop_if_true=true, stop_message="foo")
     m = MLJIteration.ICModel(machine(DummyIterativeModel(n=0), X, y), :n, 0)
     IC.train!(m, 2)
-    state = IC.update!(c, m, 0)
+    state = IC.update!(c, m, 0, 1)
     @test !state.done
     @test v == [2, ]
     IC.train!(m, 2)
-    state = IC.update!(c, m, 0, state)
+    state = IC.update!(c, m, 0, 2, state)
     @test state.done
     @test v == [2, 4]
     @test IC.takedown(c, 0, state) ==
@@ -68,11 +68,11 @@ resampler = MLJBase.Resampler(model=DummyIterativeModel(n=0),
     resampling_machine = machine(deepcopy(resampler), X, y)
     m = MLJIteration.ICModel(resampling_machine, :n, 0)
     IC.train!(m, 2)
-    state = IC.update!(c, m, 0)
+    state = IC.update!(c, m, 0, 1)
     @test !state.done
     @test length(v) == 1
     IC.train!(m, 2)
-    state = IC.update!(c, m, 0, state)
+    state = IC.update!(c, m, 0, 2, state)
     @test !state.done
     @test length(v) == 2
     @test IC.takedown(c, 0, state) == (done = false, log="")
@@ -83,11 +83,11 @@ resampler = MLJBase.Resampler(model=DummyIterativeModel(n=0),
     resampling_machine = machine(deepcopy(resampler), X, y)
     m = MLJIteration.ICModel(resampling_machine, :n, 0)
     IC.train!(m, 2)
-    state = IC.update!(c, m, 0)
+    state = IC.update!(c, m, 0, 1)
     @test !state.done
     @test length(v) == 1
     IC.train!(m, 2)
-    state = IC.update!(c, m, 0, state)
+    state = IC.update!(c, m, 0, 2, state)
     @test state.done
     @test length(v) == 2
     @test IC.takedown(c, 0, state) ==
@@ -100,11 +100,11 @@ resampler = MLJBase.Resampler(model=DummyIterativeModel(n=0),
     resampling_machine = machine(deepcopy(resampler), X, y)
     m = MLJIteration.ICModel(resampling_machine, :n, 0)
     IC.train!(m, 2)
-    state = IC.update!(c, m, 0)
+    state = IC.update!(c, m, 0, 1)
     @test !state.done
     @test length(v) == 1
     IC.train!(m, 2)
-    state = IC.update!(c, m, 0, state)
+    state = IC.update!(c, m, 0, 2, state)
     @test state.done
     @test length(v) == 2
     @test IC.takedown(c, 0, state) ==
@@ -125,13 +125,13 @@ end
                           upper = 1.5)
     model = DummyIterativeModel(n=0, learning_rate=42)
     m = MLJIteration.ICModel(machine(model, X, y), :n, 0)
-    state = @test_logs (:info, r"learning rate") IC.update!(c, m, 2)
-    state = @test_logs IC.update!(c, m, 1)
+    state = @test_logs (:info, r"learning rate") IC.update!(c, m, 2, 1)
+    state = @test_logs IC.update!(c, m, 1, 1)
     @test state == (n = 1, learning_rates = [0.5, 1.5])
     @test model.learning_rate == 0.5
-    state = IC.update!(c, m, 0, state)
+    state = IC.update!(c, m, 0, 2, state)
     @test model.learning_rate == 1.5
-    state = IC.update!(c, m, 0, state)
+    state = IC.update!(c, m, 0, 3, state)
     @test model.learning_rate == 0.5
 end