public void TestFastTreeTweedieFeaturizationInPipeline()
{
int dataPointCount = 200;
var data = SamplesUtils.DatasetUtils.GenerateFloatLabelFloatFeatureVectorSamples(dataPointCount).ToList();
var dataView = ML.Data.LoadFromEnumerable(data);
dataView = ML.Data.Cache(dataView);
var trainerOptions = new FastTreeTweedieTrainer.Options
{
NumberOfThreads = 1,
NumberOfTrees = 10,
NumberOfLeaves = 4,
MinimumExampleCountPerLeaf = 10,
FeatureColumnName = "Features",
LabelColumnName = "Label"
};
var options = new FastTreeTweedieFeaturizationEstimator.Options()
{
InputColumnName = "Features",
TreesColumnName = "Trees",
LeavesColumnName = "Leaves",
PathsColumnName = "Paths",
TrainerOptions = trainerOptions
};
var pipeline = ML.Transforms.FeaturizeByFastTreeTweedie(options)
.Append(ML.Transforms.Concatenate("CombinedFeatures", "Features", "Trees", "Leaves", "Paths"))
.Append(ML.Regression.Trainers.Sdca("Label", "CombinedFeatures"));
var model = pipeline.Fit(dataView);
var prediction = model.Transform(dataView);
var metrics = ML.Regression.Evaluate(prediction);
Assert.True(metrics.MeanAbsoluteError < 0.25);
Assert.True(metrics.MeanSquaredError < 0.1);
}
/// <summary>
/// Create <see cref="FastTreeTweedieFeaturizationEstimator"/>, which uses <see cref="FastTreeTweedieTrainer"/> to train <see cref="TreeEnsembleModelParameters"/> to create tree-based features.
/// </summary>
/// <param name="catalog">The context <see cref="TransformsCatalog"/> to create <see cref="FastTreeTweedieFeaturizationEstimator"/>.</param>
/// <param name="options">The options to configure <see cref="FastTreeTweedieFeaturizationEstimator"/>. See <see cref="FastTreeTweedieFeaturizationEstimator.Options"/> and
/// <see cref="TreeEnsembleFeaturizationEstimatorBase.OptionsBase"/> for available settings.</param>
/// <example>
/// <format type="text/markdown">
/// <]
/// ]]>
/// </format>
/// </example>
public static FastTreeTweedieFeaturizationEstimator FeaturizeByFastTreeTweedie(this TransformsCatalog catalog,
FastTreeTweedieFeaturizationEstimator.Options options)
{
Contracts.CheckValue(catalog, nameof(catalog));
var env = CatalogUtils.GetEnvironment(catalog);
return(new FastTreeTweedieFeaturizationEstimator(env, options));
}
// This example requires installation of additional NuGet package
// <a href="https://www.nuget.org/packages/Microsoft.ML.FastTree/">Microsoft.ML.FastTree</a>.
public static void Example()
{
// Create a new context for ML.NET operations. It can be used for
// exception tracking and logging, as a catalog of available operations
// and as the source of randomness. Setting the seed to a fixed number
// in this example to make outputs deterministic.
var mlContext = new MLContext(seed: 0);
// Create a list of training data points.
var dataPoints = GenerateRandomDataPoints(100).ToList();
// Convert the list of data points to an IDataView object, which is
// consumable by ML.NET API.
var dataView = mlContext.Data.LoadFromEnumerable(dataPoints);
// ML.NET doesn't cache data set by default. Therefore, if one reads a
// data set from a file and accesses it many times, it can be slow due
// to expensive featurization and disk operations. When the considered
// data can fit into memory, a solution is to cache the data in memory.
// Caching is especially helpful when working with iterative algorithms
// which needs many data passes.
dataView = mlContext.Data.Cache(dataView);
// Define input and output columns of tree-based featurizer.
string labelColumnName = nameof(DataPoint.Label);
string featureColumnName = nameof(DataPoint.Features);
string treesColumnName = nameof(TransformedDataPoint.Trees);
string leavesColumnName = nameof(TransformedDataPoint.Leaves);
string pathsColumnName = nameof(TransformedDataPoint.Paths);
// Define the configuration of the trainer used to train a tree-based
// model.
var trainerOptions = new FastTreeTweedieTrainer.Options
{
// Only use 80% of features to reduce over-fitting.
FeatureFraction = 0.8,
// Create a simpler model by penalizing usage of new features.
FeatureFirstUsePenalty = 0.1,
// Reduce the number of trees to 3.
NumberOfTrees = 3,
// Number of leaves per tree.
NumberOfLeaves = 6,
LabelColumnName = labelColumnName,
FeatureColumnName = featureColumnName
};
// Define the tree-based featurizer's configuration.
var options = new FastTreeTweedieFeaturizationEstimator.Options
{
InputColumnName = featureColumnName,
TreesColumnName = treesColumnName,
LeavesColumnName = leavesColumnName,
PathsColumnName = pathsColumnName,
TrainerOptions = trainerOptions
};
// Define the featurizer.
var pipeline = mlContext.Transforms.FeaturizeByFastTreeTweedie(
options);
// Train the model.
var model = pipeline.Fit(dataView);
// Create testing data. Use different random seed to make it different
// from training data.
var transformed = model.Transform(dataView);
// Convert IDataView object to a list. Each element in the resulted list
// corresponds to a row in the IDataView.
var transformedDataPoints = mlContext.Data.CreateEnumerable <
TransformedDataPoint>(transformed, false).ToList();
// Print out the transformation of the first 3 data points.
for (int i = 0; i < 3; ++i)
{
var dataPoint = dataPoints[i];
var transformedDataPoint = transformedDataPoints[i];
Console.WriteLine("The original feature vector [" + String.Join(",",
dataPoint.Features) + "] is transformed to three different " +
"tree-based feature vectors:");
Console.WriteLine(" Trees' output values: [" + String.Join(",",
transformedDataPoint.Trees) + "].");
Console.WriteLine(" Leave IDs' 0-1 representation: [" + String
.Join(",", transformedDataPoint.Leaves) + "].");
Console.WriteLine(" Paths IDs' 0-1 representation: [" + String
.Join(",", transformedDataPoint.Paths) + "].");
}
// Expected output:
// The original feature vector [1.543569,1.494266,1.284405] is
// transformed to three different tree-based feature vectors:
// Trees' output values: [-0.05652997,-0.02312196,-0.01179363].
// Leave IDs' 0-1 representation: [0,1,0,0,0,0,0,0,0,0,1,0,0,0,1,0,0,0].
// Paths IDs' 0-1 representation: [1,0,0,0,0,1,1,0,1,0,1,1,0,0,0].
// The original feature vector [0.764918,1.11206,0.648211] is
// transformed to three different tree-based feature vectors:
// Trees' output values: [-0.1933938,-0.1042738,-0.2312837].
// Leave IDs' 0-1 representation: [0,0,1,0,0,0,0,1,0,0,0,0,1,0,0,0,0,0].
// Paths IDs' 0-1 representation: [1,1,1,0,0,1,1,0,0,0,1,0,0,0,0].
// The original feature vector [1.251254,1.269456,1.444864] is
// transformed to three different tree-based feature vectors:
// Trees' output values: [-0.05652997,-0.06082304,-0.04528879].
// Leave IDs' 0-1 representation: [0,1,0,0,0,0,0,0,1,0,0,0,0,0,0,1,0,0].
// Paths IDs' 0-1 representation: [1,0,0,0,0,1,1,0,1,0,1,1,1,0,1].
}
