PyBrain - Conexiones
Una conexión funciona de manera similar a una capa; la única diferencia es que desplaza los datos de un nodo a otro en una red.
En este capítulo, aprenderemos sobre:
- Entender las conexiones
- Creando conexiones
Entender las conexiones
A continuación, se muestra un ejemplo práctico de las conexiones utilizadas al crear una red.
Ejemplo
ffy.py
from pybrain.structure import FeedForwardNetwork
from pybrain.structure import LinearLayer, SigmoidLayer
from pybrain.structure import FullConnection
network = FeedForwardNetwork()
#creating layer for input => 2 , hidden=> 3 and output=>1
inputLayer = LinearLayer(2)
hiddenLayer = SigmoidLayer(3)
outputLayer = LinearLayer(1)
#adding the layer to feedforward network
network.addInputModule(inputLayer)
network.addModule(hiddenLayer)
network.addOutputModule(outputLayer)
#Create connection between input ,hidden and output
input_to_hidden = FullConnection(inputLayer, hiddenLayer)
hidden_to_output = FullConnection(hiddenLayer, outputLayer)
#add connection to the network
network.addConnection(input_to_hidden)
network.addConnection(hidden_to_output)
network.sortModules()
print(network)
Salida
C:\pybrain\pybrain\src>python ffn.py
FeedForwardNetwork-6
Modules:
[<LinearLayer 'LinearLayer-3'>, <SigmoidLayer 'SigmoidLayer-7'>,
<LinearLayer 'LinearLayer-8'>]
Connections:
[<FullConnection 'FullConnection-4': 'SigmoidLayer-7' -> 'LinearLayer-8'>,
<FullConnection 'FullConnection-5': 'LinearLayer-3' -> 'SigmoidLayer-7'>]
Creando conexiones
En Pybrain, podemos crear conexiones usando el módulo de conexión como se muestra a continuación:
Ejemplo
connect.py
from pybrain.structure.connections.connection import Connection
class YourConnection(Connection):
def __init__(self, *args, **kwargs):
Connection.__init__(self, *args, **kwargs)
def _forwardImplementation(self, inbuf, outbuf):
outbuf += inbuf
def _backwardImplementation(self, outerr, inerr, inbuf):
inerr += outer
Para crear una conexión, hay 2 métodos: _forwardImplementation () y _backwardImplementation () .
El _forwardImplementation () se llama con el búfer de salida del módulo entrante que es inbuf , y el búfer de entrada del módulo saliente llamado outbuf . El inbuf se agrega al módulo saliente outbuf .
La _backwardImplementation () se llama con outsider , inerr e inbuf . El error del módulo saliente se agrega al error del módulo entrante en _backwardImplementation () .
Usemos ahora el YourConnection en una red.
testconnection.py
from pybrain.structure import FeedForwardNetwork
from pybrain.structure import LinearLayer, SigmoidLayer
from connect import YourConnection
network = FeedForwardNetwork()
#creating layer for input => 2 , hidden=> 3 and output=>1
inputLayer = LinearLayer(2)
hiddenLayer = SigmoidLayer(3)
outputLayer = LinearLayer(1)
#adding the layer to feedforward network
network.addInputModule(inputLayer)
network.addModule(hiddenLayer)
network.addOutputModule(outputLayer)
#Create connection between input ,hidden and output
input_to_hidden = YourConnection(inputLayer, hiddenLayer)
hidden_to_output = YourConnection(hiddenLayer, outputLayer)
#add connection to the network
network.addConnection(input_to_hidden)
network.addConnection(hidden_to_output)
network.sortModules()
print(network)
Salida
C:\pybrain\pybrain\src>python testconnection.py
FeedForwardNetwork-6
Modules:
[<LinearLayer 'LinearLayer-3'>, <SigmoidLayer 'SigmoidLayer-7'>,
<LinearLayer 'LinearLayer-8'>]
Connections:
[<YourConnection 'YourConnection-4': 'LinearLayer-3' -> 'SigmoidLayer-7'>,
<YourConnection 'YourConnection-5': 'SigmoidLayer-7' -> 'LinearLayer-8'>]