public void BuildAndTrain(int kClusters = 3)
{
ConsoleWriteHeader("Build and Train using Static API");
Console.Out.WriteLine($"Input file: {pivotLocation}");
ConsoleWriteHeader("Reading file ...");
var reader = new TextLoader(env,
new TextLoader.Arguments
{
Column = new[] {
new TextLoader.Column("Features", DataKind.R4, new[] { new TextLoader.Range(0, 31) }),
new TextLoader.Column("LastName", DataKind.Text, 32)
},
HasHeader = true,
Separator = ","
});
var estrimator = new PcaEstimator(env, "Features", "PCAFeatures", rank: 2, advancedSettings: (p) => p.Seed = 42)
.Append(new CategoricalEstimator(env, new[] { new CategoricalEstimator.ColumnInfo("LastName", "LastNameKey", CategoricalTransform.OutputKind.Ind) }))
.Append(new KMeansPlusPlusTrainer(env, "Features", clustersCount: kClusters));
ConsoleWriteHeader("Training model for customer clustering");
var dataSource = reader.Read(new MultiFileSource(pivotLocation));
var model = estrimator.Fit(dataSource);
var data = model.Transform(dataSource);
// inspect data
var columnNames = data.Schema.GetColumnNames().ToArray();
var trainDataAsEnumerable = data.AsEnumerable <PivotPipelineData>(env, false).Take(10).ToArray();
ConsoleWriteHeader("Evaluate model");
var clustering = new ClusteringContext(env);
var metrics = clustering.Evaluate(data, score: "Score", features: "Features");
Console.WriteLine($"AvgMinScore is: {metrics.AvgMinScore}");
Console.WriteLine($"Dbi is: {metrics.Dbi}");
ConsoleWriteHeader("Save model to local file");
ModelHelpers.DeleteAssets(modelLocation);
using (var f = new FileStream(modelLocation, FileMode.Create))
model.SaveTo(env, f);
Console.WriteLine($"Model saved: {modelLocation}");
}
public void TestPcaEstimator()
{
var data = TextLoader.CreateReader(_env,
c => (label: c.LoadFloat(11), features: c.LoadFloat(0, 10)),
separator: ';', hasHeader: true)
.Read(_dataSource);
var est = new PcaEstimator(_env, "features", "pca", rank: 5, seed: 1);
var outputPath = GetOutputPath("PCA", "pca.tsv");
using (var ch = _env.Start("save"))
{
IDataView savedData = TakeFilter.Create(_env, est.Fit(data.AsDynamic).Transform(data.AsDynamic), 4);
savedData = new ChooseColumnsTransform(_env, savedData, "pca");
using (var fs = File.Create(outputPath))
DataSaverUtils.SaveDataView(ch, _saver, savedData, fs, keepHidden: true);
}
CheckEquality("PCA", "pca.tsv", digitsOfPrecision: 4);
Done();
}
public void PcaWorkout()
{
var data = TextLoader.CreateReader(_env,
c => (label: c.LoadFloat(11), weight: c.LoadFloat(0), features: c.LoadFloat(1, 10)),
separator: ';', hasHeader: true)
.Read(_dataSource);
var invalidData = TextLoader.CreateReader(_env,
c => (label: c.LoadFloat(11), weight: c.LoadFloat(0), features: c.LoadText(1, 10)),
separator: ';', hasHeader: true)
.Read(_dataSource);
var est = new PcaEstimator(_env, "features", "pca", rank: 4, seed: 10);
TestEstimatorCore(est, data.AsDynamic, invalidInput: invalidData.AsDynamic);
var estNonDefaultArgs = new PcaEstimator(_env, "features", "pca", rank: 3, weightColumn: "weight", overSampling: 2, center: false);
TestEstimatorCore(estNonDefaultArgs, data.AsDynamic, invalidInput: invalidData.AsDynamic);
Done();
}
public void PcaWorkout()
{
var env = new ConsoleEnvironment(seed: 1, conc: 1);
string dataSource = GetDataPath("generated_regression_dataset.csv");
var data = TextLoader.CreateReader(env,
c => (label: c.LoadFloat(11), features: c.LoadFloat(0, 10)),
separator: ';', hasHeader: true)
.Read(new MultiFileSource(dataSource));
var invalidData = TextLoader.CreateReader(env,
c => (label: c.LoadFloat(11), features: c.LoadText(0, 10)),
separator: ';', hasHeader: true)
.Read(new MultiFileSource(dataSource));
var est = new PcaEstimator(env, "features", "pca", rank: 5, advancedSettings: s => {
s.Seed = 1;
});
// The following call fails because of the following issue
// https://github.com/dotnet/machinelearning/issues/969
// TestEstimatorCore(est, data.AsDynamic, invalidInput: invalidData.AsDynamic);
var outputPath = GetOutputPath("PCA", "pca.tsv");
using (var ch = env.Start("save"))
{
var saver = new TextSaver(env, new TextSaver.Arguments {
Silent = true, OutputHeader = false
});
IDataView savedData = TakeFilter.Create(env, est.Fit(data.AsDynamic).Transform(data.AsDynamic), 4);
savedData = new ChooseColumnsTransform(env, savedData, "pca");
using (var fs = File.Create(outputPath))
DataSaverUtils.SaveDataView(ch, saver, savedData, fs, keepHidden: true);
}
CheckEquality("PCA", "pca.tsv");
Done();
}