private protected AveragedTrainStateBase(IChannel ch, int numFeatures, LinearPredictor predictor, AveragedLinearTrainer <TTransformer, TModel> parent)
: base(ch, numFeatures, predictor, parent)
{
// Do the other initializations by setting the setters as if user had set them
// Initialize the averaged weights if needed (i.e., do what happens when Averaged is set)
Averaged = parent.Args.Averaged;
if (Averaged)
{
if (parent.Args.AveragedTolerance > 0)
{
VBufferUtils.Densify(ref Weights);
}
Weights.CopyTo(ref TotalWeights);
}
else
{
// It is definitely advantageous to keep weights dense if we aren't adding them
// to another vector with each update.
VBufferUtils.Densify(ref Weights);
}
_resetWeightsAfterXExamples = parent.Args.ResetWeightsAfterXExamples ?? 0;
_args = parent.Args;
_loss = parent.LossFunction;
Gain = 1;
}
protected AveragedLinearTrainer(AveragedLinearArguments args, IHostEnvironment env, string name, SchemaShape.Column label)
: base(args, env, name, label)
{
Contracts.CheckUserArg(args.LearningRate > 0, nameof(args.LearningRate), UserErrorPositive);
Contracts.CheckUserArg(!args.ResetWeightsAfterXExamples.HasValue || args.ResetWeightsAfterXExamples > 0, nameof(args.ResetWeightsAfterXExamples), UserErrorPositive);
// Weights are scaled down by 2 * L2 regularization on each update step, so 0.5 would scale all weights to 0, which is not sensible.
Contracts.CheckUserArg(0 <= args.L2RegularizerWeight && args.L2RegularizerWeight < 0.5, nameof(args.L2RegularizerWeight), "must be in range [0, 0.5)");
Contracts.CheckUserArg(args.RecencyGain >= 0, nameof(args.RecencyGain), UserErrorNonNegative);
Contracts.CheckUserArg(args.AveragedTolerance >= 0, nameof(args.AveragedTolerance), UserErrorNonNegative);
Args = args;
}
