fix docs

docs/src/guide/models/recurrence.md

@@ -86,7 +86,7 @@ Chain(
   ),
   Dense(5 => 1),                        # 6 parameters
 )         # Total: 6 trainable arrays, 51 parameters,
-          # plus 1 non-trainable, 5 parameters, summarysize 580 bytes.   
+          # plus 1 non-trainable, 5 parameters, summarysize 540 bytes.   
 ```
 In this example, each output has only one component.
 

docs/src/guide/saving.md

@@ -62,7 +62,7 @@ julia> m = Chain(Dense(10 => 5, relu), Dense(5 => 2))
 Chain(
   Dense(10 => 5, relu),                 # 55 parameters
   Dense(5 => 2),                        # 12 parameters
-)                   # Total: 4 arrays, 67 parameters, 524 bytes.
+)                   # Total: 4 arrays, 67 parameters, 476 bytes.
 
 julia> for epoch in 1:10
           # ... train model ...
@@ -131,7 +131,7 @@ julia> model
 Chain(
   Dense(10 => 5, relu),                 # 55 parameters
   Dense(5 => 2),                        # 12 parameters
-)                   # Total: 4 arrays, 67 parameters, 524 bytes.
+)                   # Total: 4 arrays, 67 parameters, 476 bytes.
 ```
 !!! warning
     Saving models this way could lead to compatibility issues across julia versions

src/layers/basic.jl

@@ -298,7 +298,7 @@ Maxout(
   Dense(5 => 7, tanh),                  # 42 parameters
   Dense(5 => 7, tanh),                  # 42 parameters
   Dense(5 => 7, tanh),                  # 42 parameters
-)                   # Total: 6 arrays, 126 parameters, 888 bytes.
+)                   # Total: 6 arrays, 126 parameters, 816 bytes.
 
 julia> Flux.outputsize(m3, (5, 11))
 (7, 11)

src/layers/macro.jl

@@ -43,7 +43,7 @@ Trio(
   Dense(2 => 1, tanh),                  # 3 parameters
   Dense(1 => 1; bias=false),            # 1 parameters
   Dropout(0.4),
-)                   # Total: 3 arrays, 4 parameters, 224 bytes.
+)                   # Total: 3 arrays, 4 parameters, 240 bytes.
 ```
 
 """

src/layers/recurrent.jl

@@ -232,7 +232,7 @@ julia> r = RNN(3 => 5)
 Recur(
   RNNCell(3 => 5, tanh),                # 50 parameters
 )         # Total: 4 trainable arrays, 50 parameters,
-          # plus 1 non-trainable, 5 parameters, summarysize 432 bytes.
+          # plus 1 non-trainable, 5 parameters, summarysize 424 bytes.
 
 julia> r(rand(Float32, 3)) |> size
 (5,)
@@ -341,7 +341,7 @@ julia> l = LSTM(3 => 5)
 Recur(
   LSTMCell(3 => 5),                     # 190 parameters
 )         # Total: 5 trainable arrays, 190 parameters,
-          # plus 2 non-trainable, 10 parameters, summarysize 1.062 KiB.
+          # plus 2 non-trainable, 10 parameters, summarysize 1.023 KiB.
 
 julia> l(rand(Float32, 3)) |> size
 (5,)
@@ -415,7 +415,7 @@ julia> g = GRU(3 => 5)
 Recur(
   GRUCell(3 => 5),                      # 140 parameters
 )         # Total: 4 trainable arrays, 140 parameters,
-          # plus 1 non-trainable, 5 parameters, summarysize 792 bytes.
+          # plus 1 non-trainable, 5 parameters, summarysize 784 bytes.
 
 julia> g(rand(Float32, 3)) |> size
 (5,)
@@ -485,7 +485,7 @@ julia> g = GRUv3(3 => 5)
 Recur(
   GRUv3Cell(3 => 5),                    # 140 parameters
 )         # Total: 5 trainable arrays, 140 parameters,
-          # plus 1 non-trainable, 5 parameters, summarysize 848 bytes.
+          # plus 1 non-trainable, 5 parameters, summarysize 840 bytes.
 
 julia> g(rand(Float32, 3)) |> size
 (5,)

src/utils.jl

@@ -186,7 +186,7 @@ julia> round(std(Flux.kaiming_normal(10, 1000)), digits=3)
 0.044f0
 
 julia> round(std(Flux.kaiming_normal(1000, 10)), digits=3)
-0.449f0
+0.45f0
 
 julia> round(std(Flux.kaiming_normal(1000, 1000)), digits=3)
 0.045f0
@@ -590,7 +590,7 @@ Chain(
   ),
   Dense(64 => 10),                      # 650 parameters
 )         # Total: 6 trainable arrays, 51_018 parameters,
-          # plus 2 non-trainable, 128 parameters, summarysize 200.312 KiB.
+          # plus 2 non-trainable, 128 parameters, summarysize 200.211 KiB.
 
 julia> Flux.modules(m2)
 7-element Vector{Any}: