NeuralNetworkRun

NeuralNetworkRun [/Q/Z] Input=testWave, WeightsWave1=w1, WeightsWave2=w2

warning

This operation is deprecated.

The NeuralNetworkRun operation uses the interconnection weights generated by NeuralNetworkTrain, and saved in the waves M_Weights1 and M_Weights2, to execute the network for a given input. The input can contain a single run represented by a 1D wave or M runs represented by M columns of a 2D wave. The output of the calculation is saved in the wave W_NNResults or M_NNResults depending on the dimensionality of the input wave. The structure of the network is completely specified by the two weights waves and must match the number of rows in the input wave.

Flags


/Q	Suppresses printing information in the History area.
/Z	No error reporting.

Parameters


Input=testWave	Specifies the input to the neural network. testWave must be a single or double precision wave containing entries in the range [0,1] and have the correct number of rows to match the weights. Execute the network for multiple runs by using a 2D input wave where each column corresponds to a single run. For a 2D input, the result will be stored in M_NNResults with a corresponding column structure.
WeightsWave1=w1	Specifies the interconnection weights between the input and the hidden layer.
WeightsWave2=w2	Specifies the interconnection weights between the hidden layer and the output.