protected override IRowMapper CreatePerInstanceRowMapper(RoleMappedSchema schema)
{
var scoreInfo = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
int numClusters = scoreInfo.Type.VectorSize;
return(new ClusteringPerInstanceEvaluator(Host, schema.Schema, scoreInfo.Name, numClusters));
}
private protected override IRowMapper CreatePerInstanceRowMapper(RoleMappedSchema schema)
{
Contracts.CheckParam(schema.Label.HasValue, nameof(schema), "Could not find the label column");
var scoreInfo = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
return(new RegressionPerInstanceEvaluator(Host, schema.Schema, scoreInfo.Name, schema.Label.Value.Name));
}
public override void InitializeNextPass(IRow row, RoleMappedSchema schema)
{
Contracts.Assert(PassNum < 1);
Contracts.AssertValue(schema.Label);
Contracts.AssertValue(schema.Group);
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
_labelGetter = RowCursorUtils.GetLabelGetter(row, schema.Label.Index);
_scoreGetter = row.GetGetter <Single>(score.Index);
_newGroupDel = RowCursorUtils.GetIsNewGroupDelegate(row, schema.Group.Index);
if (schema.Weight != null)
{
_weightGetter = row.GetGetter <Single>(schema.Weight.Index);
}
if (UnweightedCounters.GroupSummary)
{
ValueGetter <StringBuilder> groupIdBuilder = RowCursorUtils.GetGetterAsStringBuilder(row, schema.Group.Index);
_groupSbUpdate = () => groupIdBuilder(ref _groupSb);
}
else
{
_groupSbUpdate = () => { }
};
}
protected override Aggregator GetAggregatorCore(RoleMappedSchema schema, string stratName)
{
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
Host.Assert(score.Type.VectorSize > 0);
return(new Aggregator(Host, LossFunction, score.Type.VectorSize, schema.Weight != null, stratName));
}
public override void InitializeNextPass(Row row, RoleMappedSchema schema)
{
Host.Assert(!_streaming && PassNum < 2 || PassNum < 1);
Host.AssertValue(schema.Label);
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
_labelGetter = RowCursorUtils.GetLabelGetter(row, schema.Label.Index);
_scoreGetter = row.GetGetter <float>(score.Index);
Host.AssertValue(_labelGetter);
Host.AssertValue(_scoreGetter);
if (IsMainPass())
{
Host.Assert(_topExamples.Count == 0);
if (_nameIndex < 0)
{
int rowCounter = 0;
_nameGetter = (ref ReadOnlyMemory <char> dst) => dst = (rowCounter++).ToString().AsMemory();
}
else
{
_nameGetter = row.GetGetter <ReadOnlyMemory <char> >(_nameIndex);
}
}
}
/// <summary>
/// Used in the Evaluate() method, to get the predicate for cursoring over the data.
/// The base class implementation activates the score column, the label column if it exists, the weight column if it exists
/// and the stratification columns.
/// Override if other input columns need to be activated.
/// </summary>
protected virtual Func <int, bool> GetActiveColsCore(RoleMappedSchema schema)
{
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
var label = schema.Label == null ? -1 : schema.Label.Index;
var weight = schema.Weight == null ? -1 : schema.Weight.Index;
return(i => i == score.Index || i == label || i == weight);
}
private protected virtual Func <int, bool> GetActiveColsCore(RoleMappedSchema schema)
{
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
int label = schema.Label?.Index ?? -1;
int weight = schema.Weight?.Index ?? -1;
return(i => i == score.Index || i == label || i == weight);
}
protected override IRowMapper CreatePerInstanceRowMapper(RoleMappedSchema schema)
{
Host.CheckParam(schema.Label != null, nameof(schema), "Could not find the label column");
var scoreInfo = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
Host.AssertValue(scoreInfo);
return(new MultiOutputRegressionPerInstanceEvaluator(Host, schema.Schema, scoreInfo.Name, schema.Label.Name));
}
protected override Aggregator GetAggregatorCore(RoleMappedSchema schema, string stratName)
{
Host.AssertValue(schema);
Host.Assert(!_calculateDbi || (schema.Feature != null && schema.Feature.Type.IsKnownSizeVector));
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
Host.Assert(score.Type.VectorSize > 0);
int numClusters = score.Type.VectorSize;
return(new Aggregator(Host, schema.Feature, numClusters, _calculateDbi, schema.Weight != null, stratName));
}
protected override Aggregator GetAggregatorCore(RoleMappedSchema schema, string stratName)
{
var scoreInfo = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
var t = scoreInfo.Type;
Host.Assert(t.VectorSize > 0 && (t.ItemType == NumberType.R4 || t.ItemType == NumberType.R8));
var slotNames = default(VBuffer <ReadOnlyMemory <char> >);
t = schema.Schema.GetMetadataTypeOrNull(MetadataUtils.Kinds.SlotNames, scoreInfo.Index);
if (t != null && t.VectorSize == scoreInfo.Type.VectorSize && t.ItemType.IsText)
{
schema.Schema.GetMetadata(MetadataUtils.Kinds.SlotNames, scoreInfo.Index, ref slotNames);
}
return(new Aggregator(Host, LossFunction, schema.Weight != null, scoreInfo.Type.VectorSize, in slotNames, stratName));
}
protected override IRowMapper CreatePerInstanceRowMapper(RoleMappedSchema schema)
{
Host.CheckParam(schema.Label != null, nameof(schema), "Schema must contain a label column");
var scoreInfo = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
int scoreSize = scoreInfo.Type.VectorSize;
var type = schema.Schema.GetMetadataTypeOrNull(MetadataUtils.Kinds.SlotNames, scoreInfo.Index);
Host.Check(type != null && type.IsKnownSizeVector && type.ItemType.IsText, "Quantile regression score column must have slot names");
var quantiles = default(VBuffer <ReadOnlyMemory <char> >);
schema.Schema.GetMetadata(MetadataUtils.Kinds.SlotNames, scoreInfo.Index, ref quantiles);
Host.Assert(quantiles.IsDense && quantiles.Length == scoreSize);
return(new QuantileRegressionPerInstanceEvaluator(Host, schema.Schema, scoreInfo.Name, schema.Label.Name, scoreSize, quantiles.Values));
}
private protected override void CheckScoreAndLabelTypes(RoleMappedSchema schema)
{
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
var t = score.Type;
if (t.VectorSize == 0 || t.ItemType != NumberType.Float)
{
throw Host.ExceptSchemaMismatch(nameof(schema), "score", score.Name, "known size vector of R4", t.ToString());
}
Host.Check(schema.Label.HasValue, "Could not find the label column");
t = schema.Label.Value.Type;
if (!t.IsKnownSizeVector || (t.ItemType != NumberType.R4 && t.ItemType != NumberType.R8))
{
throw Host.ExceptSchemaMismatch(nameof(schema), "label", schema.Label.Value.Name, "known size vector of R4 or R8", t.ToString());
}
}
protected override void CheckScoreAndLabelTypes(RoleMappedSchema schema)
{
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
var t = score.Type;
if (t != NumberType.Float)
{
throw Host.Except("Score column '{0}' has type '{1}' but must be R4", score, t).MarkSensitive(MessageSensitivity.Schema);
}
Host.Check(schema.Label != null, "Could not find the label column");
t = schema.Label.Type;
if (t != NumberType.Float && t.KeyCount != 2)
{
throw Host.Except("Label column '{0}' has type '{1}' but must be R4 or a 2-value key", schema.Label.Name, t).MarkSensitive(MessageSensitivity.Schema);
}
}
protected override void CheckScoreAndLabelTypes(RoleMappedSchema schema)
{
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
var t = score.Type;
if (t.VectorSize == 0 || t.ItemType != NumberType.Float)
{
throw Host.Except("Score column '{0}' has type '{1}' but must be a known length vector of type R4", score.Name, t);
}
Host.Check(schema.Label != null, "Could not find the label column");
t = schema.Label.Type;
if (!t.IsKnownSizeVector || (t.ItemType != NumberType.R4 && t.ItemType != NumberType.R8))
{
throw Host.Except("Label column '{0}' has type '{1}' but must be a known-size vector of R4 or R8", schema.Label.Name, t);
}
}
private protected override void CheckScoreAndLabelTypes(RoleMappedSchema schema)
{
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
var t = score.Type;
if (t != NumberType.Float)
{
throw Host.ExceptSchemaMismatch(nameof(schema), "score", score.Name, "R4", t.ToString());
}
Host.CheckParam(schema.Label.HasValue, nameof(schema), "Could not find the label column");
t = schema.Label.Value.Type;
if (t != NumberType.R4)
{
throw Host.ExceptSchemaMismatch(nameof(schema), "label", schema.Label.Value.Name, "R4", t.ToString());
}
}
private protected override void CheckScoreAndLabelTypes(RoleMappedSchema schema)
{
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
var t = score.Type;
if (t.VectorSize == 0 || (t.ItemType != NumberType.R4 && t.ItemType != NumberType.R8))
{
throw Host.ExceptSchemaMismatch(nameof(schema), "score", score.Name, "vector of type R4 or R8", t.ToString());
}
Host.CheckParam(schema.Label.HasValue, nameof(schema), "Must contain a label column");
t = schema.Label.Value.Type;
if (t != NumberType.R4)
{
throw Host.ExceptSchemaMismatch(nameof(schema), "label", schema.Label.Value.Name, "R4", t.ToString());
}
}
protected override void CheckScoreAndLabelTypes(RoleMappedSchema schema)
{
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
var t = score.Type;
if (t.IsVector || t.ItemType != NumberType.Float)
{
throw Host.Except("Score column '{0}' has type '{1}' but must be R4", score, t);
}
Host.Check(schema.Label != null, "Could not find the label column");
t = schema.Label.Type;
if (t != NumberType.R4)
{
throw Host.Except("Label column '{0}' has type '{1}' but must be R4", schema.Label.Name, t);
}
}
private protected override void CheckScoreAndLabelTypes(RoleMappedSchema schema)
{
var t = schema.Label.Value.Type;
if (t != NumberType.Float && !(t is KeyType))
{
throw Host.ExceptSchemaMismatch(nameof(RankerMamlEvaluator.Arguments.LabelColumn),
"label", schema.Label.Value.Name, "R4 or a key", t.ToString());
}
var scoreCol = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
if (scoreCol.Type != NumberType.Float)
{
throw Host.ExceptSchemaMismatch(nameof(RankerMamlEvaluator.Arguments.ScoreColumn),
"score", scoreCol.Name, "R4", t.ToString());
}
}
public override void InitializeNextPass(IRow row, RoleMappedSchema schema)
{
Contracts.Assert(PassNum < 1);
Contracts.AssertValue(schema.Label);
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
_labelGetter = RowCursorUtils.GetVecGetterAs <Float>(NumberType.Float, row, schema.Label.Index);
_scoreGetter = row.GetGetter <VBuffer <Float> >(score.Index);
Contracts.AssertValue(_labelGetter);
Contracts.AssertValue(_scoreGetter);
if (schema.Weight != null)
{
_weightGetter = row.GetGetter <Float>(schema.Weight.Index);
}
}
protected override void CheckScoreAndLabelTypes(RoleMappedSchema schema)
{
var t = schema.Label.Type;
if (t != NumberType.Float && !t.IsKey)
{
throw Host.ExceptUserArg(nameof(RankerMamlEvaluator.Arguments.LabelColumn), "Label column '{0}' has type '{1}' but must be R4 or a key",
schema.Label.Name, t);
}
var scoreInfo = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
if (scoreInfo.Type != NumberType.Float)
{
throw Host.ExceptUserArg(nameof(RankerMamlEvaluator.Arguments.ScoreColumn), "Score column '{0}' has type '{1}' but must be R4",
scoreInfo.Name, t);
}
}
private protected override void CheckScoreAndLabelTypes(RoleMappedSchema schema)
{
ColumnType type = schema.Label?.Type;
if (type != null && type != NumberType.Float && !(type is KeyType keyType && keyType.Count > 0))
{
throw Host.ExceptSchemaMismatch(nameof(schema), "label", schema.Label.Value.Name,
"R4 or key of known cardinality", type.ToString());
}
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
type = score.Type;
if (!type.IsKnownSizeVector || type.ItemType != NumberType.Float)
{
throw Host.ExceptSchemaMismatch(nameof(schema), "score", score.Name, "R4 vector of known size", type.ToString());
}
}
public override void InitializeNextPass(IRow row, RoleMappedSchema schema)
{
AssertValid(assertGetters: false);
Host.AssertValue(row);
Host.AssertValue(schema);
if (_calculateDbi)
{
Host.AssertValue(schema.Feature);
_featGetter = row.GetGetter <VBuffer <Single> >(schema.Feature.Index);
}
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
Host.Assert(score.Type.VectorSize == _scoresArr.Length);
_scoreGetter = row.GetGetter <VBuffer <Single> >(score.Index);
if (PassNum == 0)
{
if (schema.Label != null)
{
_labelGetter = RowCursorUtils.GetLabelGetter(row, schema.Label.Index);
}
else
{
_labelGetter = (ref Single value) => value = Single.NaN;
}
if (schema.Weight != null)
{
_weightGetter = row.GetGetter <Single>(schema.Weight.Index);
}
}
else
{
Host.Assert(PassNum == 1 && _calculateDbi);
UnweightedCounters.InitializeSecondPass(_clusterCentroids);
if (WeightedCounters != null)
{
WeightedCounters.InitializeSecondPass(_clusterCentroids);
}
}
AssertValid(assertGetters: true);
}
protected override void CheckScoreAndLabelTypes(RoleMappedSchema schema)
{
ColumnType type;
if (schema.Label != null && (type = schema.Label.Type) != NumberType.Float && type.KeyCount == 0)
{
throw Host.Except("Clustering evaluator: label column '{0}' type must be {1} or Key of known cardinality." +
" Provide a correct label column, or none: it is optional.",
schema.Label.Name, NumberType.Float);
}
var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score);
type = score.Type;
if (!type.IsKnownSizeVector || type.ItemType != NumberType.Float)
{
throw Host.Except("Scores column '{0}' type must be a float vector of known size", score.Name);
}
}