private static void PostProcessStatements(StatementBlock block, RandomGenerator random)
{
MulToShiftTransform.Run(block);
NormalizeBinOpTransform.Run(block);
ExpansionTransform.Run(block);
ShuffleTransform.Run(block, random);
ConvertVariables.Run(block);
}
/// <summary>
/// This method ensures that the data meets the requirements of this trainer and its
/// subclasses, injects necessary transforms, and throws if it couldn't meet them.
/// </summary>
/// <param name="ch">The channel</param>
/// <param name="examples">The training examples</param>
/// <param name="weightSetCount">Gets the length of weights and bias array. For binary classification and regression,
/// this is 1. For multi-class classification, this equals the number of classes on the label.</param>
/// <returns>A potentially modified version of <paramref name="examples"/></returns>
private RoleMappedData PrepareDataFromTrainingExamples(IChannel ch, RoleMappedData examples, out int weightSetCount)
{
ch.AssertValue(examples);
CheckLabel(examples, out weightSetCount);
examples.CheckFeatureFloatVector();
var idvToShuffle = examples.Data;
IDataView idvToFeedTrain;
if (idvToShuffle.CanShuffle)
{
idvToFeedTrain = idvToShuffle;
}
else
{
var shuffleArgs = new ShuffleTransform.Arguments
{
PoolOnly = false,
ForceShuffle = _args.Shuffle
};
idvToFeedTrain = new ShuffleTransform(Host, shuffleArgs, idvToShuffle);
}
ch.Assert(idvToFeedTrain.CanShuffle);
var roles = examples.Schema.GetColumnRoleNames();
var examplesToFeedTrain = new RoleMappedData(idvToFeedTrain, roles);
ch.AssertValue(examplesToFeedTrain.Schema.Label);
ch.AssertValue(examplesToFeedTrain.Schema.Feature);
if (examples.Schema.Weight != null)
{
ch.AssertValue(examplesToFeedTrain.Schema.Weight);
}
int numFeatures = examplesToFeedTrain.Schema.Feature.Type.VectorSize;
ch.Check(numFeatures > 0, "Training set has no features, aborting training.");
return(examplesToFeedTrain);
}
