Sequential class
A Network
model in which every Layer
has one input and one output
tensor.
Constructors
-
Sequential({required double learningRate, List<
Layer> ? layers}) -
Creates a
Sequential
model network.
Properties
- hashCode → int
-
The hash code for this object.
no setterinherited
-
layers
→ List<
Layer> -
The
Layers
part of thisNetwork
.finalinherited - learningRate ↔ double
-
The degree of radicality at which the
Network
will adjust itsNeurons
weights.getter/setter pairinherited - runtimeType → Type
-
A representation of the runtime type of the object.
no setterinherited
- stopwatch → Stopwatch
-
Used for performance analysis as well as general information logging.
finalinherited
Methods
-
addLayer(
Layer layer) → void -
Adds a
Layer
to thisNetwork
.inherited -
addLayers(
List< Layer> layers) → void -
Adds a list of
Layers
to thisNetwork
.inherited -
clear(
) → void -
Clears the
Network
by removing allLayers
, thereby returning it to its initial, empty state.inherited -
noSuchMethod(
Invocation invocation) → dynamic -
Invoked when a nonexistent method or property is accessed.
inherited
-
process(
List< double> inputs) → List<double> -
Processes the
inputs
by propagating them across everyLayer
. and returns the output.inherited -
propagateBackwards(
List< double> input, List<double> expected) → void -
Performs the backpropagation algorithm by comparing the observed values
with the expected values, and propagates each layer using the knowledge of
the network's 'cost', which indicates how bad the network is performing.
inherited
-
toString(
) → String -
A string representation of this object.
inherited
-
train(
{required List< List< inputs, required List<double> >List< expected, required int iterations, bool quiet = false}) → voiddouble> > -
Trains the network using the passed
inputs
, their respectiveexpected
results, as well as the number of iterations to make during training.inherited
Operators
-
operator ==(
Object other) → bool -
The equality operator.
inherited