Developer Guide and Reference

  • 2021.6
  • 04/11/2022
  • Public Content
Contents

Batch Processing

Training

For a description of the input and output, refer to Recommendation Systems Usage Model.
At the training stage, the implicit ALS recommender has the following parameters:
Training Parameters for Implicit Alternating Least Squares Computaion (Batch Processing)
Parameter
Default Value
Description
algorithmFPType
float
The floating-point type that the algorithm uses for intermediate computations. Can be
float
or
double
.
method
defaultDense
Available computation methods:
  • defaultDense
    - performance-oriented method
  • fastCSR
    - performance-oriented method for CSR numeric tables
nFactors
LaTex Math image.
The total number of factors.
maxIterations
LaTex Math image.
The number of iterations.
alpha
LaTex Math image.
The rate of confidence.
lambda
LaTex Math image.
The parameter of the regularization.
preferenceThreshold
LaTex Math image.
Threshold used to define preference values. LaTex Math image. is the only threshold supported so far.

Prediction

For a description of the input and output, refer to Recommendation Systems Usage Model.
At the prediction stage, the implicit ALS recommender has the following parameters:
Prediction Parameters for Implicit Alternating Least Squares Computaion (Batch Processing)
Parameter
Default Value
Description
algorithmFPType
float
The floating-point type that the algorithm uses for intermediate computations. Can be
float
or
double
.
method
defaultDense
Performance-oriented computation method, the only method supported by the algorithm.

Product and Performance Information

1

Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.