Package samer.models

  • Interface Summary 
    Interface Description
    Model  
    Model.Trainer
    This represents a training algorithm for a Model Trainer is responsible for counting calls to accumulate() between flushes
  • Class Summary 
    Class Description
    AlignedGaussian
    Differential scaler: scales and offsets each element of a vector independently aiming to match a given prior model.
    BatchedTrainer
    Manages a Model.Trainer to do batched learning.
    Covariance
    Collect covariance statistics assuming input is zero mean.
    DiffScaler
    Differential scaler: scales and offsets each element of a vector independently aiming to match a given prior model.
    GaussianStats
    Collect statistics about a vector: accumulates sums and sums of products, then computes mean and covariance.
    GaussianStatsOnline
    Similar to GaussianStats: Collect statistics about a vector: accumulates sums and sums of products, then computes mean and covariance.
    GeneralisedExponential  
    ICA  
    ICAScalerSync
    This is a task which subsumes a post-scaling into an ICA weight matrix
    ICAWithScaler  
    IIDPrior
    Non adaptive model of vector in which each component is independent of the others and all are identically distributed according to a given prior
    JointHistogramBase  
    MatrixTrainer
    Handles batched delta updates to a matrix.
    Mixture  
    MOGModel  
    MOGVector  
    NoisyICA  
    Scaler
    Automatic gain control for a given input vector.
    SignalHistogram  
    SmoothGeneralisedExponential
    Non-adaptive generalised exponential factorial prior: the pointy bit of the usual GeneralisedExponential has been smoothed out by blending with a quadratic.
    SparseICA  
    VarianceICA