Field-aware Factorization Machine

The documentation is generated based on the sources available at dotnet/machinelearning and released under MIT License.

Type: binaryclassifiertrainer Aliases: FieldAwareFactorizationMachine, ffm Namespace: Microsoft.ML.Runtime.FactorizationMachine Assembly: Microsoft.ML.StandardLearners.dll Microsoft Documentation: Field-aware Factorization Machine

Description

Train a field-aware factorization machine for binary classification

Parameters

Name Short name Default Description
iters iter 5 Number of training iterations
lambdaLatent   0.0001 Regularization coefficient of latent weights
lambdaLinear   0.0001 Regularization coefficient of linear weights
latentDim d 20 Latent space dimension
learningRate lr 0.1 Initial learning rate
norm   True Whether to normalize the input vectors so that the concatenation of all fields’ feature vectors is unit-length
radius rad 0.5 Radius of initial latent factors
shuffle shuf True Whether to shuffle for each training iteration
verbose   True Report traning progress or not