diff --git a/latest/backends.html b/latest/backends.html new file mode 100644 index 00000000..2a89208c --- /dev/null +++ b/latest/backends.html @@ -0,0 +1,160 @@ + + + + + + +Backends · Flux + + + + + + + + + + + + + + +
+
+ +
+
+

+ +Batching + +

+

+[WIP] +

+ +
+ + diff --git a/latest/batching.html b/latest/batching.html new file mode 100644 index 00000000..1cefb019 --- /dev/null +++ b/latest/batching.html @@ -0,0 +1,160 @@ + + + + + + +Batching · Flux + + + + + + + + + + + + + + +
+
+ +
+
+

+ +Batching + +

+

+[WIP] +

+ +
+ + diff --git a/latest/contributing.html b/latest/contributing.html index a54f7317..8bb83a8b 100644 --- a/latest/contributing.html +++ b/latest/contributing.html @@ -74,6 +74,16 @@ Debugging +
  • + +Batching + +
    • +
  • + +Backends + +
  • In Action @@ -109,7 +119,7 @@ Contributing & Help
    • - + diff --git a/latest/examples/logreg.html b/latest/examples/logreg.html index b44a5c82..d7e73593 100644 --- a/latest/examples/logreg.html +++ b/latest/examples/logreg.html @@ -74,6 +74,16 @@ Debugging +
  • + +Batching + +
    • +
  • + +Backends + +
  • In Action @@ -112,7 +122,7 @@ Logistic Regression
    • - + @@ -131,12 +141,12 @@ Logistic Regression with MNIST

    diff --git a/latest/models/recurrent.html b/latest/models/recurrent.html index fefeef68..0bcbe3ee 100644 --- a/latest/models/recurrent.html +++ b/latest/models/recurrent.html @@ -75,6 +75,16 @@ Debugging +
  • + +Batching + +
    • +
  • + +Backends + +
  • In Action @@ -112,7 +122,7 @@ Recurrence
    • - + diff --git a/latest/models/templates.html b/latest/models/templates.html index dfe25821..4e728678 100644 --- a/latest/models/templates.html +++ b/latest/models/templates.html @@ -91,6 +91,16 @@ Debugging +
  • + +Batching + +
    • +
  • + +Backends + +
  • In Action @@ -128,7 +138,7 @@ Model Templates
    • - + @@ -183,16 +193,20 @@ This is much better: we can now make as many affine layers as we want. This is a 
    @net type MyAffine
   W
   b
-  x -> W * x + b
+  x -> x * W + b
 end

    The function provided, -x -> W * x + b +x -> x * W + b , will be used when MyAffine is used as a model; it's just a shorter way of defining the (::MyAffine)(x) - method above. + method above. (You may notice that +W + and +x + have swapped order in the model; this is due to the way batching works, which will be covered in more detail later on.)

    However, @@ -244,7 +258,7 @@ Clearly, the x . The Affine - layer fits the bill so we can instantiate + layer fits the bill, so we can instantiate TLP with two of them:

    @@ -290,12 +304,13 @@ However, for convenience and to avoid errors, we'd probably rather specify t

    This is easy to implement using the usual Julia syntax for constructors:

    -
    Affine(in::Integer, out::Integer) = Affine(randn(in, out), randn(1, out))
    +
    Affine(in::Integer, out::Integer) =
+  Affine(randn(in, out), randn(1, out))

    In practice, these constructors tend to take the parameter initialisation function as an argument so that it's more easily customisable, and use Flux.initn by default (which is equivalent to -randn()/100 +randn(...)/100 ). So Affine 's constructor really looks like this: @@ -309,10 +324,10 @@ Supported syntax

    The syntax used to define a forward pass like -x -> W*x + b +x -> x*W + b behaves exactly like Julia code for the most part. However, it's important to remember that it's defining a dataflow graph, not a general Julia expression. In practice this means that anything side-effectful, or things like control flow and println -s, won't work as expected. In future we'll continue expand support for Julia syntax and features. +s, won't work as expected. In future we'll continue to expand support for Julia syntax and features.