PyBrain - 层 (Layers)

层基本上是一组用于网络隐藏层的函数。

本章我们将介绍以下关于层的细节：

理解层
使用 Pybrain 创建层

理解层

我们之前已经看过一些使用层的示例，如下所示：

TanhLayer
SoftmaxLayer

使用 TanhLayer 的示例

下面是一个使用 TanhLayer 构建网络的示例：

testnetwork.py

from pybrain.tools.shortcuts import buildNetwork
from pybrain.structure import TanhLayer
from pybrain.datasets import SupervisedDataSet
from pybrain.supervised.trainers import BackpropTrainer

# Create a network with two inputs, three hidden, and one output
nn = buildNetwork(2, 3, 1, bias=True, hiddenclass=TanhLayer)

# Create a dataset that matches network input and output sizes:
norgate = SupervisedDataSet(2, 1)

# Create a dataset to be used for testing.
nortrain = SupervisedDataSet(2, 1)

# Add input and target values to dataset
# Values for NOR truth table
norgate.addSample((0, 0), (1,))
norgate.addSample((0, 1), (0,))
norgate.addSample((1, 0), (0,))
norgate.addSample((1, 1), (0,))

# Add input and target values to dataset
# Values for NOR truth table
nortrain.addSample((0, 0), (1,))
nortrain.addSample((0, 1), (0,))
nortrain.addSample((1, 0), (0,))
nortrain.addSample((1, 1), (0,))

#Training the network with dataset norgate.
trainer = BackpropTrainer(nn, norgate)

# will run the loop 1000 times to train it.
for epoch in range(1000):
   trainer.train()
trainer.testOnData(dataset=nortrain, verbose = True)

输出

以上代码的输出如下：

python testnetwork.py

C:\pybrain\pybrain\src>python testnetwork.py
Testing on data:
('out: ', '[0.887 ]')
('correct:', '[1 ]')
error: 0.00637334
('out: ', '[0.149 ]')
('correct:', '[0 ]')
error: 0.01110338
('out: ', '[0.102 ]')
('correct:', '[0 ]')
error: 0.00522736
('out: ', '[-0.163]')
('correct:', '[0 ]')
error: 0.01328650
('All errors:', [0.006373344564625953, 0.01110338071737218, 
   0.005227359234093431, 0.01328649974219942])
('Average error:', 0.008997646064572746)
('Max error:', 0.01328649974219942, 'Median error:', 0.01110338071737218)

使用 SoftMaxLayer 的示例

下面是一个使用 SoftmaxLayer 构建网络的示例：

from pybrain.tools.shortcuts import buildNetwork
from pybrain.structure.modules import SoftmaxLayer
from pybrain.datasets import SupervisedDataSet
from pybrain.supervised.trainers import BackpropTrainer

# Create a network with two inputs, three hidden, and one output
nn = buildNetwork(2, 3, 1, bias=True, hiddenclass=SoftmaxLayer)

# Create a dataset that matches network input and output sizes:
norgate = SupervisedDataSet(2, 1)

# Create a dataset to be used for testing.
nortrain = SupervisedDataSet(2, 1)

# Add input and target values to dataset
# Values for NOR truth table
norgate.addSample((0, 0), (1,))
norgate.addSample((0, 1), (0,))
norgate.addSample((1, 0), (0,))
norgate.addSample((1, 1), (0,))

# Add input and target values to dataset
# Values for NOR truth table
nortrain.addSample((0, 0), (1,))
nortrain.addSample((0, 1), (0,))
nortrain.addSample((1, 0), (0,))
nortrain.addSample((1, 1), (0,))

#Training the network with dataset norgate.
trainer = BackpropTrainer(nn, norgate)

# will run the loop 1000 times to train it.
for epoch in range(1000):
trainer.train()
trainer.testOnData(dataset=nortrain, verbose = True)

输出

输出如下：

C:\pybrain\pybrain\src>python example16.py
Testing on data:
('out: ', '[0.918 ]')
('correct:', '[1 ]')
error: 0.00333524
('out: ', '[0.082 ]')
('correct:', '[0 ]')
error: 0.00333484
('out: ', '[0.078 ]')
('correct:', '[0 ]')
error: 0.00303433
('out: ', '[-0.082]')
('correct:', '[0 ]')
error: 0.00340005
('All errors:', [0.0033352368788838365, 0.003334842961037291, 
   0.003034328685718761, 0.0034000458892589056])
('Average error:', 0.0032761136037246985)
('Max error:', 0.0034000458892589056, 'Median error:', 0.0033352368788838365)

在 Pybrain 中创建层

在 Pybrain 中，您可以创建自己的层，如下所示：

要创建一个层，您需要使用NeuronLayer 类作为基类来创建所有类型的层。

示例

from pybrain.structure.modules.neuronlayer import NeuronLayer
class LinearLayer(NeuronLayer):
   def _forwardImplementation(self, inbuf, outbuf):
      outbuf[:] = inbuf
   def _backwardImplementation(self, outerr, inerr, outbuf, inbuf):
      inerr[:] = outer

要创建一个层，我们需要实现两个方法：_forwardImplementation() 和 _backwardImplementation()。

_forwardImplementation() 接收 2 个参数 inbuf 和 outbuf，它们是 Scipy 数组。其大小取决于层的输入和输出维度。

_backwardImplementation() 用于计算给定输出相对于输入的导数。

因此，要在 Pybrain 中实现一个层，这是层类的框架：

from pybrain.structure.modules.neuronlayer import NeuronLayer
class NewLayer(NeuronLayer):
   def _forwardImplementation(self, inbuf, outbuf):
      pass
   def _backwardImplementation(self, outerr, inerr, outbuf, inbuf):
      pass

如果您想将二次多项式函数实现为一个层，我们可以这样做：

假设我们有一个多项式函数：

f(x) = 3x2

上述多项式函数的导数如下：

f(x) = 6 x

上述多项式函数的最终层类如下：

testlayer.py

from pybrain.structure.modules.neuronlayer import NeuronLayer
class PolynomialLayer(NeuronLayer):
   def _forwardImplementation(self, inbuf, outbuf):
      outbuf[:] = 3*inbuf**2
   def _backwardImplementation(self, outerr, inerr, outbuf, inbuf):
      inerr[:] = 6*inbuf*outerr

现在让我们使用如下所示创建的层：

testlayer1.py

from testlayer import PolynomialLayer
from pybrain.tools.shortcuts import buildNetwork
from pybrain.tests.helpers import gradientCheck

n = buildNetwork(2, 3, 1, hiddenclass=PolynomialLayer)
n.randomize()

gradientCheck(n)

GradientCheck() 将测试层是否正常工作。我们需要将使用该层的网络传递给 gradientCheck(n)。如果层工作正常，它将输出“Perfect Gradient”。

输出

C:\pybrain\pybrain\src>python testlayer1.py
Perfect gradient

打印页面