public static CommonOutputs.RegressionOutput TrainRegression(IHostEnvironment env, FastTreeRegressionTrainer.Options input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("TrainFastTree");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
return(TrainerEntryPointsUtils.Train <FastTreeRegressionTrainer.Options, CommonOutputs.RegressionOutput>(host, input,
() => new FastTreeRegressionTrainer(host, input),
() => TrainerEntryPointsUtils.FindColumn(host, input.TrainingData.Schema, input.LabelColumnName),
() => TrainerEntryPointsUtils.FindColumn(host, input.TrainingData.Schema, input.ExampleWeightColumnName),
() => TrainerEntryPointsUtils.FindColumn(host, input.TrainingData.Schema, input.RowGroupColumnName)));
}
public static CommonOutputs.BinaryClassificationOutput TrainBinary(IHostEnvironment env, XGBoostArguments input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("Train" + EntryPointName);
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
return(LearnerEntryPointsUtils.Train <XGBoostArguments,
CommonOutputs.BinaryClassificationOutput>(host, input,
() => new XGBoostBinaryTrainer(host, input),
getLabel: () => LearnerEntryPointsUtils.FindColumn(host, input.TrainingData.Schema, input.LabelColumn),
getWeight: () => LearnerEntryPointsUtils.FindColumn(host, input.TrainingData.Schema, input.WeightColumn)));
}
public static CommonOutputs.TransformOutput Nop(IHostEnvironment env, NopInput input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("Nop");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
var xf = CreateIfNeeded(host, input.Data);
return(new CommonOutputs.TransformOutput {
Model = new TransformModelImpl(env, xf, input.Data), OutputData = xf
});
}
public static CommonOutputs.TransformOutput GetSample(IHostEnvironment env, BootstrapSamplingTransformer.Arguments input)
{
Contracts.CheckValue(env, nameof(env));
env.CheckValue(input, nameof(input));
var h = EntryPointUtils.CheckArgsAndCreateHost(env, "BootstrapSample", input);
var view = new BootstrapSamplingTransformer(h, input, input.Data);
return(new CommonOutputs.TransformOutput()
{
Model = new TransformModel(h, view, input.Data),
OutputData = view
});
}
public static CommonOutputs.RankingOutput TrainRanking(IHostEnvironment env, LightGbmArguments input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("TrainLightGBM");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
return(LearnerEntryPointsUtils.Train <LightGbmArguments, CommonOutputs.RankingOutput>(host, input,
() => new LightGbmRankingTrainer(host, input),
getLabel: () => LearnerEntryPointsUtils.FindColumn(host, input.TrainingData.Schema, input.LabelColumn),
getWeight: () => LearnerEntryPointsUtils.FindColumn(host, input.TrainingData.Schema, input.WeightColumn),
getGroup: () => LearnerEntryPointsUtils.FindColumn(host, input.TrainingData.Schema, input.GroupIdColumn)));
}
public static CommonOutputs.RegressionOutput TrainTweedieRegression(IHostEnvironment env, FastTreeTweedieTrainer.Arguments input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("TrainTweeedie");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
return(LearnerEntryPointsUtils.Train <FastTreeTweedieTrainer.Arguments, CommonOutputs.RegressionOutput>(host, input,
() => new FastTreeTweedieTrainer(host, input),
() => LearnerEntryPointsUtils.FindColumn(host, input.TrainingData.Schema, input.LabelColumn),
() => LearnerEntryPointsUtils.FindColumn(host, input.TrainingData.Schema, input.WeightColumn),
() => LearnerEntryPointsUtils.FindColumn(host, input.TrainingData.Schema, input.GroupIdColumn)));
}
public static CommonOutputs.TransformOutput WordEmbeddings(IHostEnvironment env, WordEmbeddingsExtractingTransformer.Options input)
{
Contracts.CheckValue(env, nameof(env));
env.CheckValue(input, nameof(input));
var h = EntryPointUtils.CheckArgsAndCreateHost(env, "WordEmbeddings", input);
var view = WordEmbeddingsExtractingTransformer.Create(h, input, input.Data);
return(new CommonOutputs.TransformOutput()
{
Model = new TransformModelImpl(h, view, input.Data),
OutputData = view
});
}
public static CommonOutputs.TransformOutput MinMax(IHostEnvironment env, NormalizeTransform.MinMaxArguments input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("MinMax");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
var xf = NormalizeTransform.Create(host, input, input.Data);
return(new CommonOutputs.TransformOutput {
Model = new TransformModel(env, xf, input.Data), OutputData = xf
});
}
public static CommonOutputs.TransformOutput KeyToText(IHostEnvironment env, KeyToValueMappingTransformer.Arguments input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("KeyToValue");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
var xf = KeyToValueMappingTransformer.Create(host, input, input.Data);
return(new CommonOutputs.TransformOutput {
Model = new TransformModelImpl(env, xf, input.Data), OutputData = xf
});
}
public static CommonOutputs.TransformOutput CharTokenize(IHostEnvironment env, TokenizingByCharactersTransformer.Options input)
{
Contracts.CheckValue(env, nameof(env));
env.CheckValue(input, nameof(input));
var h = EntryPointUtils.CheckArgsAndCreateHost(env, "CharTokenize", input);
var view = TokenizingByCharactersTransformer.Create(h, input, input.Data);
return(new CommonOutputs.TransformOutput()
{
Model = new TransformModelImpl(h, view, input.Data),
OutputData = view
});
}
public static CommonOutputs.TransformOutput Scaler(IHostEnvironment env, ScalerTransform_ArgumentsEntryPoint input)
{
Contracts.CheckValue(env, nameof(env));
env.CheckValue(input, nameof(input));
var h = EntryPointUtils.CheckArgsAndCreateHost(env, EP_Scaler.Name, input);
var view = new ScalerTransform(h, input, input.Data);
return(new CommonOutputs.TransformOutput()
{
Model = new TransformModel(h, view, input.Data),
OutputData = view
});
}
public static CommonOutputs.TransformOutput CatTransformHash(IHostEnvironment env, OneHotHashEncodingTransformer.Options input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("CatTransformDict");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
var xf = OneHotHashEncodingTransformer.Create(host, input, input.Data);
return(new CommonOutputs.TransformOutput {
Model = new TransformModelImpl(env, xf, input.Data), OutputData = xf
});
}
public static CommonOutputs.TransformOutput CountSelect(IHostEnvironment env, CountFeatureSelectionTransformer.Arguments input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("CountSelect");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
var xf = CountFeatureSelectionTransformer.Create(host, input, input.Data);
return(new CommonOutputs.TransformOutput {
Model = new TransformModel(env, xf, input.Data), OutputData = xf
});
}
public static CommonOutputs.TransformOutput MutualInformationSelect(IHostEnvironment env, MutualInformationFeatureSelectingEstimator.Arguments input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("MutualInformationSelect");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
var xf = MutualInformationFeatureSelectingEstimator.Create(host, input, input.Data);
return(new CommonOutputs.TransformOutput {
Model = new TransformModelImpl(env, xf, input.Data), OutputData = xf
});
}
public static CommonOutputs.TransformOutput TensorFlowScorer(IHostEnvironment env, Arguments input)
{
Contracts.CheckValue(env, nameof(env));
env.CheckValue(input, nameof(input));
var h = EntryPointUtils.CheckArgsAndCreateHost(env, "TensorFlow", input);
var view = Create(h, input, input.Data);
return(new CommonOutputs.TransformOutput()
{
Model = new TransformModel(h, view, input.Data),
OutputData = view
});
}
public static CommonOutputs.TransformOutput CatTransformHash(IHostEnvironment env, CategoricalHashTransform.Arguments input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("CatTransformDict");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
var xf = CategoricalHashTransform.Create(host, input, input.Data);
return(new CommonOutputs.TransformOutput {
Model = new TransformModel(env, xf, input.Data), OutputData = xf
});
}
internal static CommonOutputs.TransformOutput Create(IHostEnvironment env, CountTargetEncodingEstimator.Options input)
{
Contracts.CheckValue(env, nameof(env));
env.CheckValue(input, nameof(input));
var h = EntryPointUtils.CheckArgsAndCreateHost(env, nameof(CountTargetEncoder), input);
var view = CountTargetEncodingEstimator.Create(h, input, input.Data);
return(new CommonOutputs.TransformOutput()
{
Model = new TransformModelImpl(h, view, input.Data),
OutputData = view
});
}
public static CommonOutputs.TransformOutput Featurizer(IHostEnvironment env, TreeEnsembleFeaturizerTransform.ArgumentsForEntryPoint input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("TreeFeaturizerTransform");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
var xf = TreeEnsembleFeaturizerTransform.CreateForEntryPoint(env, input, input.Data);
return(new CommonOutputs.TransformOutput {
Model = new TransformModelImpl(env, xf, input.Data), OutputData = xf
});
}
public static CommonOutputs.TransformOutput LabelIndicator(IHostEnvironment env, Options input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("LabelIndictator");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
var xf = Create(host, input, input.Data);
return(new CommonOutputs.TransformOutput {
Model = new TransformModelImpl(env, xf, input.Data), OutputData = xf
});
}
public static CommonOutputs.TransformOutput Ungroup(IHostEnvironment env, UngroupTransform.Arguments input)
{
Contracts.CheckValue(env, nameof(env));
env.CheckValue(input, nameof(input));
var h = EntryPointUtils.CheckArgsAndCreateHost(env, "UngroupTransform", input);
var view = new UngroupTransform(h, input, input.Data);
return(new CommonOutputs.TransformOutput()
{
Model = new TransformModelImpl(h, view, input.Data),
OutputData = view
});
}
public static CommonOutputs.TransformOutput Convert(IHostEnvironment env, TypeConvertingTransformer.Arguments input)
{
Contracts.CheckValue(env, nameof(env));
env.CheckValue(input, nameof(input));
var h = EntryPointUtils.CheckArgsAndCreateHost(env, "Convert", input);
var view = TypeConvertingTransformer.Create(h, input, input.Data);
return(new CommonOutputs.TransformOutput()
{
Model = new TransformModel(h, view, input.Data),
OutputData = view
});
}
public static CommonOutputs.TransformOutput TextToKey(IHostEnvironment env, TermTransform.Arguments input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("Term");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
var xf = new TermTransform(host, input, input.Data);
return(new CommonOutputs.TransformOutput {
Model = new TransformModel(env, xf, input.Data), OutputData = xf
});
}
public static CommonOutputs.BinaryClassificationOutput TrainBinary(IHostEnvironment env, FastForestBinaryTrainer.Options input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("TrainFastForest");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
return(TrainerEntryPointsUtils.Train <FastForestBinaryTrainer.Options, CommonOutputs.BinaryClassificationOutput>(host, input,
() => new FastForestBinaryTrainer(host, input),
() => TrainerEntryPointsUtils.FindColumn(host, input.TrainingData.Schema, input.LabelColumnName),
() => TrainerEntryPointsUtils.FindColumn(host, input.TrainingData.Schema, input.ExampleWeightColumnName),
() => TrainerEntryPointsUtils.FindColumn(host, input.TrainingData.Schema, input.RowGroupColumnName),
calibrator: input.Calibrator, maxCalibrationExamples: input.MaxCalibrationExamples));
}
public static CommonOutputs.TransformOutput LightLda(IHostEnvironment env, LatentDirichletAllocationTransformer.Arguments input)
{
Contracts.CheckValue(env, nameof(env));
env.CheckValue(input, nameof(input));
var h = EntryPointUtils.CheckArgsAndCreateHost(env, "LightLda", input);
var cols = input.Column.Select(colPair => new LatentDirichletAllocationTransformer.ColumnInfo(colPair, input)).ToArray();
var est = new LatentDirichletAllocationEstimator(h, cols);
var view = est.Fit(input.Data).Transform(input.Data);
return(new CommonOutputs.TransformOutput()
{
Model = new TransformModelImpl(h, view, input.Data),
OutputData = view
});
}
public static PermutationFeatureImportanceOutput PermutationFeatureImportance(IHostEnvironment env, PermutationFeatureImportanceArguments input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("Pfi");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
input.PredictorModel.PrepareData(env, input.Data, out RoleMappedData roleMappedData, out IPredictor predictor);
Contracts.Assert(predictor != null, "No predictor found in model");
IDataView result = PermutationFeatureImportanceUtils.GetMetrics(env, predictor, roleMappedData, input);
return(new PermutationFeatureImportanceOutput {
Metrics = result
});
}
public static CommonOutputs.TransformOutput ConcatColumns(IHostEnvironment env, ColumnCopyingTransformer.Options input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("PrefixConcatColumns");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
// Get all column names with preserving order.
var colNames = new List <string>(input.Data.Schema.Count);
for (int i = 0; i < input.Data.Schema.Count; i++)
{
colNames.Add(input.Data.Schema[i].Name);
}
// Iterate through input options, find matching source columns, create new input options
var inputOptions = new ColumnConcatenatingTransformer.Options()
{
Data = input.Data
};
var columns = new List <ColumnConcatenatingTransformer.Column>(input.Columns.Length);
foreach (var col in input.Columns)
{
var newCol = new ColumnConcatenatingTransformer.Column();
newCol.Name = col.Name;
var prefix = col.Source;
newCol.Source = colNames.Where(x => x.StartsWith(prefix, StringComparison.InvariantCulture)).ToArray();
if (newCol.Source.Length == 0)
{
throw new ArgumentOutOfRangeException("No matching columns found for prefix: " + prefix);
}
columns.Add(newCol);
}
inputOptions.Columns = columns.ToArray();
var xf = ColumnConcatenatingTransformer.Create(env, inputOptions, inputOptions.Data);
return(new CommonOutputs.TransformOutput {
Model = new TransformModelImpl(env, xf, inputOptions.Data), OutputData = xf
});
}
public static SummaryOutput Summarize(IHostEnvironment env, SummarizePredictor.Input input)
{
Contracts.CheckValue(env, nameof(env));
var host = env.Register("PipelineEnsemblePredictor");
host.CheckValue(input, nameof(input));
EntryPointUtils.CheckInputArgs(host, input);
input.PredictorModel.PrepareData(host,
new EmptyDataView(host, input.PredictorModel.TransformModel.InputSchema),
out RoleMappedData rmd, out IPredictor predictor
);
var calibrated = predictor as CalibratedPredictorBase;
while (calibrated != null)
{
predictor = calibrated.SubPredictor;
calibrated = predictor as CalibratedPredictorBase;
}
var ensemble = predictor as SchemaBindablePipelineEnsembleBase;
host.CheckUserArg(ensemble != null, nameof(input.PredictorModel.Predictor), "Predictor is not a pipeline ensemble predictor");
var summaries = new IDataView[ensemble.PredictorModels.Length];
var stats = new IDataView[ensemble.PredictorModels.Length];
for (int i = 0; i < ensemble.PredictorModels.Length; i++)
{
var pm = ensemble.PredictorModels[i];
pm.PrepareData(host, new EmptyDataView(host, pm.TransformModel.InputSchema), out rmd, out IPredictor pred);
summaries[i] = SummarizePredictor.GetSummaryAndStats(host, pred, rmd.Schema, out stats[i]);
}
return(new SummaryOutput()
{
Summaries = summaries, Stats = stats
});
}
public static OnnxTransformOutput ApplyOnnxModel(IHostEnvironment env, OnnxTransformInput input)
{
var host = EntryPointUtils.CheckArgsAndCreateHost(env, "OnnxTransform", input);
var inputColumns = input.InputColumns ?? (Array.Empty <string>());
var outputColumns = input.OutputColumns ?? (Array.Empty <string>());
var transformsCatalog = new TransformsCatalog(host);
var onnxScoringEstimator = OnnxCatalog.ApplyOnnxModel(transformsCatalog,
outputColumns,
inputColumns,
input.ModelFile,
input.GpuDeviceId,
input.FallbackToCpu);
var view = onnxScoringEstimator.Fit(input.Data).Transform(input.Data);
return(new OnnxTransformOutput()
{
Model = new TransformModelImpl(host, view, input.Data),
OutputData = view
});
}
