35 lines
922 B
ObjectPascal
35 lines
922 B
ObjectPascal
|
// Полный пример задачи регрессии:
|
|||
|
|
// загружаем датасет, делим его на выборки,
|
|||
|
|
// обучаем pipeline и оцениваем качество.
|
|||
|
|
|
|||
|
|
uses MLABC;
|
|||
|
|
|
|||
|
|
begin
|
|||
|
|
var ds := Datasets.MoscowHousing;
|
|||
|
|
var df := ds.Data;
|
|||
|
|
|
|||
|
|
var features := ['rooms', 'area', 'kitchen_area', 'floor', 'floors_total', 'metro_minutes', 'renovation'];
|
|||
|
|
var target := 'price';
|
|||
|
|
|
|||
|
|
var (trainDf, testDf) := df.TrainTestSplit(0.2, seed := 42);
|
|||
|
|
|
|||
|
|
var pipe := DataPipeline.Build(
|
|||
|
|
TaskKind.tkRegression,
|
|||
|
|
target,
|
|||
|
|
features,
|
|||
|
|
new OneHotEncoder('renovation'),
|
|||
|
|
new StandardScaler,
|
|||
|
|
new LinearRegression
|
|||
|
|
);
|
|||
|
|
|
|||
|
|
pipe.Fit(trainDf);
|
|||
|
|
|
|||
|
|
var pred := pipe.Predict(testDf);
|
|||
|
|
var yTest := testDf.ToVector(target);
|
|||
|
|
var r2 := Metrics.R2(yTest, pred);
|
|||
|
|
|
|||
|
|
Println('Прогноз цен на жильё: полный пример');
|
|||
|
|
Println;
|
|||
|
|
Println($'R² = {r2:F3}');
|
|||
|
|
end.
|