Failure Analysis of Static Analysis Software Module Based on Big Data Tendency Prediction
Algorithm 1
Specific algorithm of training encoder.
Input: training data x, the number of input layer nodes inputSize, the number of hidden layer nodes hiddenSize, the weight attenuation coefficient , the sparse regularization constant , the maximum number of iterations of the optimization loss cost function maxIter
Output: optimal network parameters ,b
(1)
Initialize the encoder’s network parameters: weight matrix , bias vector
(2)
Initialize iteration number, overall ;
(3)
Noise processing is carried out on training data x to obtain noise data ;
(4)
optimizes the network parameters of the encoder by iterative methods
(5)
By coding the noise data , the feature y of the hidden layer is obtained.
(6)
In the decoding operation, the feature y of the hidden layer is decoded to obtain the reconstructed data z;
(7)
Calculate the overall cost of the encoder;
(8)
Calculate the partial derivative of the loss cost function
