PermutationFeatureImportance(
this RankingCatalog catalog,
ITransformer model,
IDataView data,
string labelColumnName = DefaultColumnNames.Label,
string rowGroupColumnName = DefaultColumnNames.GroupId,
bool useFeatureWeightFilter = false,
int?numberOfExamplesToUse = null,
int permutationCount = 1)
{
Contracts.CheckValue(catalog, nameof(catalog));
var env = catalog.GetEnvironment();
Contracts.CheckValue(env, nameof(env));
env.CheckValue(data, nameof(data));
env.CheckValue(model, nameof(model));
RankingMetricsStatistics resultInitializer() => new();
RankingMetrics evaluationFunc(IDataView idv) => catalog.Evaluate(idv, labelColumnName, rowGroupColumnName);
return(PermutationFeatureImportance(
env,
model,
data,
resultInitializer,
evaluationFunc,
RankingDelta,
permutationCount,
useFeatureWeightFilter,
numberOfExamplesToUse
));
}
/// <summary>
/// Create the ML context.
/// </summary>
/// <param name="seed">Random seed. Set to <c>null</c> for a non-deterministic environment.</param>
/// <param name="conc">Concurrency level. Set to 1 to run single-threaded. Set to 0 to pick automatically.</param>
public MLContext(int?seed = null, int conc = 0)
{
_env = new LocalEnvironment(seed, conc, MakeCompositionContainer);
_env.AddListener(ProcessMessage);
BinaryClassification = new BinaryClassificationCatalog(_env);
MulticlassClassification = new MulticlassClassificationCatalog(_env);
Regression = new RegressionCatalog(_env);
Clustering = new ClusteringCatalog(_env);
Ranking = new RankingCatalog(_env);
Transforms = new TransformsCatalog(_env);
Model = new ModelOperationsCatalog(_env);
Data = new DataOperationsCatalog(_env);
}
/// <summary>
/// Create the ML context.
/// </summary>
/// <param name="seed">Random seed. Set to <c>null</c> for a non-deterministic environment.</param>
public MLContext(int?seed = null)
{
_env = new LocalEnvironment(seed);
_env.AddListener(ProcessMessage);
BinaryClassification = new BinaryClassificationCatalog(_env);
MulticlassClassification = new MulticlassClassificationCatalog(_env);
Regression = new RegressionCatalog(_env);
Clustering = new ClusteringCatalog(_env);
Ranking = new RankingCatalog(_env);
AnomalyDetection = new AnomalyDetectionCatalog(_env);
Transforms = new TransformsCatalog(_env);
Model = new ModelOperationsCatalog(_env);
Data = new DataOperationsCatalog(_env);
}
PermutationFeatureImportance <TModel>(
this RankingCatalog catalog,
IPredictionTransformer <TModel> model,
IDataView data,
string label = DefaultColumnNames.Label,
string groupId = DefaultColumnNames.GroupId,
string features = DefaultColumnNames.Features,
bool useFeatureWeightFilter = false,
int?topExamples = null,
int permutationCount = 1)
{
return(PermutationFeatureImportance <TModel, RankingMetrics, RankingMetricsStatistics> .GetImportanceMetricsMatrix(
catalog.GetEnvironment(),
model,
data,
idv => catalog.Evaluate(idv, label, groupId),
RankingDelta,
features,
permutationCount,
useFeatureWeightFilter,
topExamples));
}
PermutationFeatureImportance <TModel>(
this RankingCatalog catalog,
ISingleFeaturePredictionTransformer <TModel> predictionTransformer,
IDataView data,
string labelColumnName = DefaultColumnNames.Label,
string rowGroupColumnName = DefaultColumnNames.GroupId,
bool useFeatureWeightFilter = false,
int?numberOfExamplesToUse = null,
int permutationCount = 1) where TModel : class
{
return(PermutationFeatureImportance <TModel, RankingMetrics, RankingMetricsStatistics> .GetImportanceMetricsMatrix(
catalog.GetEnvironment(),
predictionTransformer,
data,
() => new RankingMetricsStatistics(),
idv => catalog.Evaluate(idv, labelColumnName, rowGroupColumnName),
RankingDelta,
predictionTransformer.FeatureColumnName,
permutationCount,
useFeatureWeightFilter,
numberOfExamplesToUse));
}
internal RankingTrainers(RankingCatalog catalog)
: base(catalog)
{
}
