Package rdkit :: Package ML :: Package Neural :: Module Network :: Class Network
[hide private]
[frames] | no frames]

Class Network

source code

a neural network

  

Instance Methods [hide private]
 
ConstructRandomWeights(self, minWeight=-1, maxWeight=1)
initialize all the weights in the network to random numbers
source code
 
FullyConnectNodes(self)
Fully connects each layer in the network to the one above it
source code
 
ConstructNodes(self, nodeCounts, actFunc, actFuncParms)
build an unconnected network and set node counts
source code
 
GetInputNodeList(self)
returns a list of input node indices...
source code
 
GetOutputNodeList(self)
returns a list of output node indices...
source code
 
GetHiddenLayerNodeList(self, which)
returns a list of hidden nodes in the specified layer...
source code
 
GetNumNodes(self)
returns the total number of nodes...
source code
 
GetNumHidden(self)
returns the number of hidden layers...
source code
 
GetNode(self, which)
returns a particular node...
source code
 
GetAllNodes(self)
returns a list of all nodes...
source code
 
ClassifyExample(self, example, appendExamples=0)
classifies a given example and returns the results of the output layer.
source code
 
GetLastOutputs(self)
returns the complete list of output layer values from the last time this node classified anything
source code
 
__str__(self)
provides a string representation of the network
source code
 
__init__(self, nodeCounts, nodeConnections=None, actFunc=<class 'rdkit.ML.Neural.ActFuncs.Sigmoid'>, actFuncParms=(), weightBounds=1)
Constructor
source code
Method Details [hide private]

ConstructRandomWeights(self, minWeight=-1, maxWeight=1)

source code 
initialize all the weights in the network to random numbers

**Arguments**

  - minWeight: the minimum value a weight can take

  - maxWeight: the maximum value a weight can take
  

FullyConnectNodes(self)

source code 
Fully connects each layer in the network to the one above it


**Note**
  this sets the connections, but does not assign weights
  

ConstructNodes(self, nodeCounts, actFunc, actFuncParms)

source code 
build an unconnected network and set node counts

**Arguments**

  - nodeCounts: a list containing the number of nodes to be in each layer.
     the ordering is:
      (nInput,nHidden1,nHidden2, ... , nHiddenN, nOutput)

GetInputNodeList(self)

source code 
returns a list of input node indices
    

GetOutputNodeList(self)

source code 
returns a list of output node indices
    

GetHiddenLayerNodeList(self, which)

source code 
returns a list of hidden nodes in the specified layer
    

GetNumNodes(self)

source code 
returns the total number of nodes
    

GetNumHidden(self)

source code 
returns the number of hidden layers
    

GetNode(self, which)

source code 
returns a particular node
    

GetAllNodes(self)

source code 
returns a list of all nodes
    

ClassifyExample(self, example, appendExamples=0)

source code 
classifies a given example and returns the results of the output layer.

**Arguments**

  - example: the example to be classified

**NOTE:**

  if the output layer is only one element long,
  a scalar (not a list) will be returned.  This is why a lot of the other
  network code claims to only support single valued outputs.

__init__(self, nodeCounts, nodeConnections=None, actFunc=<class 'rdkit.ML.Neural.ActFuncs.Sigmoid'>, actFuncParms=(), weightBounds=1)
(Constructor)

source code 
Constructor

This constructs and initializes the network based upon the specified
node counts.

A fully connected network with random weights is constructed.

**Arguments**

  - nodeCounts: a list containing the number of nodes to be in each layer.
     the ordering is:
      (nInput,nHidden1,nHidden2, ... , nHiddenN, nOutput)

  - nodeConnections: I don't know why this is here, but it's optional.  ;-)

  - actFunc: the activation function to be used here.  Must support the API
      of _ActFuncs.ActFunc_.

  - actFuncParms: a tuple of extra arguments to be passed to the activation function
      constructor.

  - weightBounds:  a float which provides the boundary on the random initial weights