/// <summary>
/// Initializes a new instance of <see cref="SymSgdClassificationTrainer"/>
/// </summary>
internal SymSgdClassificationTrainer(IHostEnvironment env, Options options)
: base(Contracts.CheckRef(env, nameof(env)).Register(LoadNameValue), TrainerUtils.MakeR4VecFeature(options.FeatureColumn),
TrainerUtils.MakeBoolScalarLabel(options.LabelColumn))
{
Host.CheckValue(options, nameof(options));
options.Check(Host);
_options = options;
Info = new TrainerInfo(supportIncrementalTrain: true);
}
internal LbfgsTrainerBase(IHostEnvironment env, TArgs args, SchemaShape.Column labelColumn,
float?l1Weight = null,
float?l2Weight = null,
float?optimizationTolerance = null,
int?memorySize = null,
bool?enforceNoNegativity = null)
: base(Contracts.CheckRef(env, nameof(env)).Register(RegisterName), TrainerUtils.MakeR4VecFeature(args.FeatureColumn),
labelColumn, TrainerUtils.MakeR4ScalarWeightColumn(args.WeightColumn))
{
Host.CheckValue(args, nameof(args));
Args = args;
Host.CheckUserArg(!Args.UseThreads || Args.NumThreads > 0 || Args.NumThreads == null,
nameof(Args.NumThreads), "numThreads must be positive (or empty for default)");
Host.CheckUserArg(Args.L2Weight >= 0, nameof(Args.L2Weight), "Must be non-negative");
Host.CheckUserArg(Args.L1Weight >= 0, nameof(Args.L1Weight), "Must be non-negative");
Host.CheckUserArg(Args.OptTol > 0, nameof(Args.OptTol), "Must be positive");
Host.CheckUserArg(Args.MemorySize > 0, nameof(Args.MemorySize), "Must be positive");
Host.CheckUserArg(Args.MaxIterations > 0, nameof(Args.MaxIterations), "Must be positive");
Host.CheckUserArg(Args.SgdInitializationTolerance >= 0, nameof(Args.SgdInitializationTolerance), "Must be non-negative");
Host.CheckUserArg(Args.NumThreads == null || Args.NumThreads.Value >= 0, nameof(Args.NumThreads), "Must be non-negative");
Host.CheckParam(!(l2Weight < 0), nameof(l2Weight), "Must be non-negative, if provided.");
Host.CheckParam(!(l1Weight < 0), nameof(l1Weight), "Must be non-negative, if provided");
Host.CheckParam(!(optimizationTolerance <= 0), nameof(optimizationTolerance), "Must be positive, if provided.");
Host.CheckParam(!(memorySize <= 0), nameof(memorySize), "Must be positive, if provided.");
// Review: Warn about the overriding behavior
L2Weight = l2Weight ?? Args.L2Weight;
L1Weight = l1Weight ?? Args.L1Weight;
OptTol = optimizationTolerance ?? Args.OptTol;
MemorySize = memorySize ?? Args.MemorySize;
MaxIterations = Args.MaxIterations;
SgdInitializationTolerance = Args.SgdInitializationTolerance;
Quiet = Args.Quiet;
InitWtsDiameter = Args.InitWtsDiameter;
UseThreads = Args.UseThreads;
NumThreads = Args.NumThreads;
DenseOptimizer = Args.DenseOptimizer;
EnforceNonNegativity = enforceNoNegativity ?? Args.EnforceNonNegativity;
if (EnforceNonNegativity && ShowTrainingStats)
{
ShowTrainingStats = false;
using (var ch = Host.Start("Initialization"))
{
ch.Warning("The training statistics cannot be computed with non-negativity constraint.");
ch.Done();
}
}
ShowTrainingStats = false;
_srcPredictor = default;
}
protected OnlineLinearTrainer(OnlineLinearArguments args, IHostEnvironment env, string name, SchemaShape.Column label)
: base(Contracts.CheckRef(env, nameof(env)).Register(name), TrainerUtils.MakeR4VecFeature(args.FeatureColumn), label, TrainerUtils.MakeR4ScalarWeightColumn(args.InitialWeights))
{
Contracts.CheckValue(args, nameof(args));
Contracts.CheckUserArg(args.NumIterations > 0, nameof(args.NumIterations), UserErrorPositive);
Contracts.CheckUserArg(args.InitWtsDiameter >= 0, nameof(args.InitWtsDiameter), UserErrorNonNegative);
Contracts.CheckUserArg(args.StreamingCacheSize > 0, nameof(args.StreamingCacheSize), UserErrorPositive);
Args = args;
Name = name;
// REVIEW: Caching could be false for one iteration, if we got around the whole shuffling issue.
Info = new TrainerInfo(calibration: NeedCalibration, supportIncrementalTrain: true);
}
/// <summary>
/// Initializes a new instance of <see cref="SymSgdClassificationTrainer"/>
/// </summary>
/// <param name="env">The private instance of <see cref="IHostEnvironment"/>.</param>
/// <param name="labelColumn">The name of the label column.</param>
/// <param name="featureColumn">The name of the feature column.</param>
/// <param name="advancedSettings">A delegate to apply all the advanced arguments to the algorithm.</param>
public SymSgdClassificationTrainer(IHostEnvironment env, string featureColumn, string labelColumn, Action <Arguments> advancedSettings = null)
: base(Contracts.CheckRef(env, nameof(env)).Register(LoadNameValue), TrainerUtils.MakeR4VecFeature(featureColumn),
TrainerUtils.MakeBoolScalarLabel(labelColumn))
{
_args = new Arguments();
// Apply the advanced args, if the user supplied any.
_args.Check(Host);
advancedSettings?.Invoke(_args);
_args.FeatureColumn = featureColumn;
_args.LabelColumn = labelColumn;
Info = new TrainerInfo();
}
private protected LightGbmTrainerBase(IHostEnvironment env,
string name,
SchemaShape.Column label,
string featureColumn,
string weightColumn,
string groupIdColumn,
int?numLeaves,
int?minDataPerLeaf,
double?learningRate,
int numBoostRound,
Action <LightGbmArguments> advancedSettings)
: base(Contracts.CheckRef(env, nameof(env)).Register(name), TrainerUtils.MakeR4VecFeature(featureColumn), label, TrainerUtils.MakeR4ScalarWeightColumn(weightColumn), TrainerUtils.MakeU4ScalarColumn(groupIdColumn))
{
Args = new LightGbmArguments();
Args.NumLeaves = numLeaves;
Args.MinDataPerLeaf = minDataPerLeaf;
Args.LearningRate = learningRate;
Args.NumBoostRound = numBoostRound;
//apply the advanced args, if the user supplied any
advancedSettings?.Invoke(Args);
Args.LabelColumn = label.Name;
Args.FeatureColumn = featureColumn;
if (weightColumn != null)
{
Args.WeightColumn = Optional <string> .Explicit(weightColumn);
}
if (groupIdColumn != null)
{
Args.GroupIdColumn = Optional <string> .Explicit(groupIdColumn);
}
InitParallelTraining();
}
private protected GamTrainerBase(IHostEnvironment env, TOptions options, string name, SchemaShape.Column label)
: base(Contracts.CheckRef(env, nameof(env)).Register(name), TrainerUtils.MakeR4VecFeature(options.FeatureColumnName),
label, TrainerUtils.MakeR4ScalarWeightColumn(options.ExampleWeightColumnName))
{
Contracts.CheckValue(env, nameof(env));
Host.CheckValue(options, nameof(options));
Host.CheckParam(options.LearningRate > 0, nameof(options.LearningRate), "Must be positive.");
Host.CheckParam(options.NumberOfThreads == null || options.NumberOfThreads > 0, nameof(options.NumberOfThreads), "Must be positive.");
Host.CheckParam(0 <= options.EntropyCoefficient && options.EntropyCoefficient <= 1, nameof(options.EntropyCoefficient), "Must be in [0, 1].");
Host.CheckParam(0 <= options.GainConfidenceLevel && options.GainConfidenceLevel < 1, nameof(options.GainConfidenceLevel), "Must be in [0, 1).");
Host.CheckParam(0 < options.MaximumBinCountPerFeature, nameof(options.MaximumBinCountPerFeature), "Must be positive.");
Host.CheckParam(0 < options.NumberOfIterations, nameof(options.NumberOfIterations), "Must be positive.");
Host.CheckParam(0 < options.MinimumExampleCountPerLeaf, nameof(options.MinimumExampleCountPerLeaf), "Must be positive.");
GamTrainerOptions = options;
Info = new TrainerInfo(normalization: false, calibration: NeedCalibration, caching: false, supportValid: true);
_gainConfidenceInSquaredStandardDeviations = Math.Pow(ProbabilityFunctions.Probit(1 - (1 - GamTrainerOptions.GainConfidenceLevel) * 0.5), 2);
_entropyCoefficient = GamTrainerOptions.EntropyCoefficient * 1e-6;
InitializeThreads();
}
private protected GamTrainerBase(IHostEnvironment env, TArgs args, string name, SchemaShape.Column label)
: base(Contracts.CheckRef(env, nameof(env)).Register(name), TrainerUtils.MakeR4VecFeature(args.FeatureColumn),
label, TrainerUtils.MakeR4ScalarWeightColumn(args.WeightColumn))
{
Contracts.CheckValue(env, nameof(env));
Host.CheckValue(args, nameof(args));
Host.CheckParam(args.LearningRates > 0, nameof(args.LearningRates), "Must be positive.");
Host.CheckParam(args.NumThreads == null || args.NumThreads > 0, nameof(args.NumThreads), "Must be positive.");
Host.CheckParam(0 <= args.EntropyCoefficient && args.EntropyCoefficient <= 1, nameof(args.EntropyCoefficient), "Must be in [0, 1].");
Host.CheckParam(0 <= args.GainConfidenceLevel && args.GainConfidenceLevel < 1, nameof(args.GainConfidenceLevel), "Must be in [0, 1).");
Host.CheckParam(0 < args.MaxBins, nameof(args.MaxBins), "Must be posittive.");
Host.CheckParam(0 < args.NumIterations, nameof(args.NumIterations), "Must be positive.");
Host.CheckParam(0 < args.MinDocuments, nameof(args.MinDocuments), "Must be positive.");
Args = args;
Info = new TrainerInfo(normalization: false, calibration: NeedCalibration, caching: false, supportValid: true);
_gainConfidenceInSquaredStandardDeviations = Math.Pow(ProbabilityFunctions.Probit(1 - (1 - Args.GainConfidenceLevel) * 0.5), 2);
_entropyCoefficient = Args.EntropyCoefficient * 1e-6;
InitializeThreads();
}
private protected LightGbmTrainerBase(IHostEnvironment env, string name, SchemaShape.Column label, string featureColumn,
string weightColumn = null, string groupIdColumn = null, Action <LightGbmArguments> advancedSettings = null)
: base(Contracts.CheckRef(env, nameof(env)).Register(name), TrainerUtils.MakeR4VecFeature(featureColumn), label, TrainerUtils.MakeR4ScalarWeightColumn(weightColumn))
{
Args = new LightGbmArguments();
//apply the advanced args, if the user supplied any
advancedSettings?.Invoke(Args);
Args.LabelColumn = label.Name;
Args.FeatureColumn = featureColumn;
if (weightColumn != null)
{
Args.WeightColumn = weightColumn;
}
if (groupIdColumn != null)
{
Args.GroupIdColumn = groupIdColumn;
}
InitParallelTraining();
}
/// <summary>
/// Initializes a new instance of <see cref="MultiClassNaiveBayesTrainer"/>
/// </summary>
internal MultiClassNaiveBayesTrainer(IHostEnvironment env, Arguments args)
: base(Contracts.CheckRef(env, nameof(env)).Register(LoadName), TrainerUtils.MakeR4VecFeature(args.FeatureColumn),
TrainerUtils.MakeU4ScalarLabel(args.LabelColumn))
{
Host.CheckValue(args, nameof(args));
}
public SdcaRegressionTrainer(IHostEnvironment env, Arguments args, string featureColumn, string labelColumn, string weightColumn = null)
: base(Contracts.CheckRef(env, nameof(env)).Register(LoadNameValue), args, TrainerUtils.MakeR4VecFeature(featureColumn),
TrainerUtils.MakeR4ScalarLabel(labelColumn), TrainerUtils.MakeR4ScalarWeightColumn(weightColumn))
{
Host.CheckValue(labelColumn, nameof(labelColumn));
Host.CheckValue(featureColumn, nameof(featureColumn));
_loss = args.LossFunction.CreateComponent(env);
Loss = _loss;
_args = args;
}
