revert: "test: enable enzyme testing for recurrent models"

This reverts commit fe967bd.

test/layers/recurrent_tests.jl

@@ -26,7 +26,7 @@
  @test_throws ErrorException ps.train_state
 
  @test_gradients(loss_loop_rnncell, ps; atol=1.0f-3, rtol=1.0f-3,
- soft_fail=[AutoFiniteDiff()])
+ soft_fail=[AutoFiniteDiff()], broken_backends=[AutoEnzyme()])
  end
  end
 
@@ -84,7 +84,7 @@ end
  end
 
  @test_gradients(loss_loop_lstmcell, ps; atol=1.0f-3, rtol=1.0f-3,
- soft_fail=[AutoFiniteDiff()])
+ soft_fail=[AutoFiniteDiff()], broken_backends=[AutoEnzyme()])
 
  @test_throws ErrorException ps.train_state
  @test_throws ErrorException ps.train_memory
@@ -195,7 +195,7 @@ end
  end
 
  @test_gradients(loss_loop_grucell, ps; atol=1e-3, rtol=1e-3,
- soft_fail=[AutoFiniteDiff()])
+ soft_fail=[AutoFiniteDiff()], broken_backends=[AutoEnzyme()])
 
  @test_throws ErrorException ps.train_state
  end
@@ -282,7 +282,7 @@ end
  end
 
  @test_gradients(loss_loop_rnn, ps; atol=1e-3, rtol=1e-3,
- soft_fail=[AutoFiniteDiff()])
+ broken_backends=[AutoEnzyme()], soft_fail=[AutoFiniteDiff()])
  end
  end
  end
@@ -327,11 +327,11 @@ end
 
  __f = p -> sum(first(rnn(x, p, st)))
  @test_gradients(__f, ps; atol=1e-3, rtol=1e-3,
- soft_fail=[AutoFiniteDiff()])
+ skip_backends=[AutoEnzyme()], soft_fail=[AutoFiniteDiff()])
 
  __f = p -> sum(Base.Fix1(sum, abs2), first(rnn_seq(x, p, st)))
  @test_gradients(__f, ps; atol=1e-3, rtol=1e-3,
- soft_fail=[AutoFiniteDiff()])
+ skip_backends=[AutoEnzyme()], soft_fail=[AutoFiniteDiff()])
  end
 
  # Batched Time Series without data batches
@@ -362,11 +362,11 @@ end
 
  __f = p -> sum(first(rnn(x, p, st)))
  @test_gradients(__f, ps; atol=1e-3, rtol=1e-3,
- soft_fail=[AutoFiniteDiff()])
+ skip_backends=[AutoEnzyme()], soft_fail=[AutoFiniteDiff()])
 
  __f = p -> sum(Base.Fix1(sum, abs2), first(rnn_seq(x, p, st)))
  @test_gradients(__f, ps; atol=1e-3, rtol=1e-3,
- soft_fail=[AutoFiniteDiff()])
+ skip_backends=[AutoEnzyme()], soft_fail=[AutoFiniteDiff()])
  end
  end
  end
@@ -415,13 +415,17 @@ end
  @test all(x -> size(x) == (5, 2), y_[1])
 
  __f = p -> sum(Base.Fix1(sum, abs2), first(bi_rnn(x, p, st)))
- @test_gradients(__f, ps; atol=1e-3, rtol=1e-3)
+ @test_gradients(__f, ps; atol=1e-3,
+ rtol=1e-3,
+ broken_backends=Sys.isapple() ? [AutoEnzyme()] : [])
 
  __f = p -> begin
  (y1, y2), st_ = bi_rnn_no_merge(x, p, st)
  return sum(Base.Fix1(sum, abs2), y1) + sum(Base.Fix1(sum, abs2), y2)
  end
- @test_gradients(__f, ps; atol=1e-3, rtol=1e-3)
+ @test_gradients(__f, ps; atol=1e-3,
+ rtol=1e-3,
+ broken_backends=Sys.isapple() ? [AutoEnzyme()] : [])
 
  @testset "backward_cell: $_backward_cell" for _backward_cell in (
  RNNCell, LSTMCell, GRUCell)
@@ -449,13 +453,17 @@ end
  @test all(x -> size(x) == (5, 2), y_[1])
 
  __f = p -> sum(Base.Fix1(sum, abs2), first(bi_rnn(x, p, st)))
- @test_gradients(__f, ps; atol=1e-3, rtol=1e-3)
+ @test_gradients(__f, ps; atol=1e-3,
+ rtol=1e-3,
+ broken_backends=Sys.isapple() ? [AutoEnzyme()] : [])
 
  __f = p -> begin
  (y1, y2), st_ = bi_rnn_no_merge(x, p, st)
  return sum(Base.Fix1(sum, abs2), y1) + sum(Base.Fix1(sum, abs2), y2)
  end
- @test_gradients(__f, ps; atol=1e-3, rtol=1e-3)
+ @test_gradients(__f, ps; atol=1e-3,
+ rtol=1e-3,
+ broken_backends=Sys.isapple() ? [AutoEnzyme()] : [])
  end
  end
  end