public override IPredictor CreatePredictor()
{
Contracts.Assert(WeightArraySize == 1);
Contracts.Assert(Utils.Size(Weights) == 1);
Contracts.Assert(Utils.Size(Bias) == 1);
Host.Check(Weights[0].Length > 0);
VBuffer <Float> maybeSparseWeights = VBufferUtils.CreateEmpty <Float>(Weights[0].Length);
VBufferUtils.CreateMaybeSparseCopy(ref Weights[0], ref maybeSparseWeights, Conversions.Instance.GetIsDefaultPredicate <Float>(NumberType.Float));
return(new LinearRegressionPredictor(Host, ref maybeSparseWeights, Bias[0]));
}
private TPredictor CreatePredictor(VBuffer <Float> weights, Float bias)
{
Host.CheckParam(weights.Length > 0, nameof(weights));
VBuffer <Float> maybeSparseWeights = default;
VBufferUtils.CreateMaybeSparseCopy(ref weights, ref maybeSparseWeights,
Conversions.Instance.GetIsDefaultPredicate <Float>(NumberType.Float));
var predictor = new LinearBinaryPredictor(Host, ref maybeSparseWeights, bias);
return(new ParameterMixingCalibratedPredictor(Host, predictor, new PlattCalibrator(Host, -1, 0)));
}
protected override LinearRegressionPredictor CreatePredictor(VBuffer <Float>[] weights, Float[] bias)
{
Host.CheckParam(Utils.Size(weights) == 1, nameof(weights));
Host.CheckParam(Utils.Size(bias) == 1, nameof(bias));
Host.CheckParam(weights[0].Length > 0, nameof(weights));
VBuffer <Float> maybeSparseWeights = default;
VBufferUtils.CreateMaybeSparseCopy(ref weights[0], ref maybeSparseWeights,
Conversions.Instance.GetIsDefaultPredicate <Float>(NumberType.Float));
return(new LinearRegressionPredictor(Host, ref maybeSparseWeights, bias[0]));
}
protected override LinearRegressionPredictor CreatePredictor(VBuffer <Float>[] weights, Float[] bias)
{
Host.CheckParam(Utils.Size(weights) == 1, nameof(weights));
Host.CheckParam(Utils.Size(bias) == 1, nameof(bias));
Host.CheckParam(weights[0].Length > 0, nameof(weights));
VBuffer <Float> maybeSparseWeights = default;
// below should be `in weights[0]`, but can't because of https://github.com/dotnet/roslyn/issues/29371
VBufferUtils.CreateMaybeSparseCopy(weights[0], ref maybeSparseWeights,
Conversions.Instance.GetIsDefaultPredicate <Float>(NumberType.Float));
return(new LinearRegressionPredictor(Host, in maybeSparseWeights, bias[0]));