diff --git a/src/Microsoft.ML.CpuMath/ProbabilityFunctions.cs b/src/Microsoft.ML.CpuMath/ProbabilityFunctions.cs index db3e304379..3ce463f574 100644 --- a/src/Microsoft.ML.CpuMath/ProbabilityFunctions.cs +++ b/src/Microsoft.ML.CpuMath/ProbabilityFunctions.cs @@ -119,6 +119,8 @@ public static double Erfinv(double x) /// One intepretation is, the value at which the standard normal CDF evaluates to p. public static double Probit(double p) { + Contracts.CheckParam(0 <= p && p <= 1, nameof(p), "Input probability should be in range 0 to 1."); + double q = p - 0.5; double r = 0.0; if (Math.Abs(q) <= 0.425) @@ -135,8 +137,6 @@ public static double Probit(double p) else r = 1 - p; - Contracts.CheckParam(r >= 0, nameof(p), "Illegal input value"); - r = Math.Sqrt(-Math.Log(r)); double retval = 0.0; if (r < 5) diff --git a/src/Microsoft.ML.Data/DataLoadSave/DataOperationsCatalog.cs b/src/Microsoft.ML.Data/DataLoadSave/DataOperationsCatalog.cs index 9d4e1292ca..60cce3a5b7 100644 --- a/src/Microsoft.ML.Data/DataLoadSave/DataOperationsCatalog.cs +++ b/src/Microsoft.ML.Data/DataLoadSave/DataOperationsCatalog.cs @@ -87,7 +87,7 @@ public IDataView FilterByKeyColumnFraction(IDataView input, string columnName, d var type = input.Schema[columnName].Type; if (type.GetKeyCount() == 0) - throw Environment.ExceptSchemaMismatch(nameof(columnName), "filter", columnName, "a known cardinality key", type.ToString()); + throw Environment.ExceptSchemaMismatch(nameof(columnName), "filter", columnName, "KeyType", type.ToString()); return new RangeFilter(Environment, input, columnName, lowerBound, upperBound, false); } } diff --git a/src/Microsoft.ML.Data/DataView/TypedCursor.cs b/src/Microsoft.ML.Data/DataView/TypedCursor.cs index 29437cce5b..2631fb2162 100644 --- a/src/Microsoft.ML.Data/DataView/TypedCursor.cs +++ b/src/Microsoft.ML.Data/DataView/TypedCursor.cs @@ -103,7 +103,7 @@ private TypedCursorable(IHostEnvironment env, IDataView data, bool ignoreMissing { if (ignoreMissingColumns) continue; - throw _host.Except("Column '{0}' not found in the data view", col.ColumnName); + throw _host.ExceptSchemaMismatch(nameof(_data.Schema), "", col.ColumnName); } var realColType = _data.Schema[colIndex].Type; if (!IsCompatibleType(realColType, col.MemberInfo)) diff --git a/src/Microsoft.ML.Data/Evaluators/AnomalyDetectionEvaluator.cs b/src/Microsoft.ML.Data/Evaluators/AnomalyDetectionEvaluator.cs index 143ecb44b5..e5fbc720bf 100644 --- a/src/Microsoft.ML.Data/Evaluators/AnomalyDetectionEvaluator.cs +++ b/src/Microsoft.ML.Data/Evaluators/AnomalyDetectionEvaluator.cs @@ -96,11 +96,11 @@ 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.Except("Score column '{0}' has type '{1}' but must be R4", score, t).MarkSensitive(MessageSensitivity.Schema); + throw Host.ExceptSchemaMismatch(nameof(schema), "score", score.Name, "float", t.ToString()); Host.Check(schema.Label.HasValue, "Could not find the label column"); t = schema.Label.Value.Type; if (t != NumberType.Float && t.GetKeyCount() != 2) - throw Host.Except("Label column '{0}' has type '{1}' but must be R4 or a 2-value key", schema.Label.Value.Name, t).MarkSensitive(MessageSensitivity.Schema); + throw Host.ExceptSchemaMismatch(nameof(schema), "label", schema.Label.Value.Name, "float or a KeyType with cardinality 2", t.ToString()); } private protected override Aggregator GetAggregatorCore(RoleMappedSchema schema, string stratName) @@ -631,13 +631,13 @@ private protected override void PrintFoldResultsCore(IChannel ch, Dictionary>(index); if (!top.Schema.TryGetColumnIndex(AnomalyDetectionEvaluator.TopKResultsColumns.AnomalyScore, out index)) - throw Host.Except("Data view does not contain the 'Anomaly Score' column"); + throw Host.ExceptSchemaMismatch(nameof(top.Schema), "anomaly score", AnomalyDetectionEvaluator.TopKResultsColumns.AnomalyScore); var scoreGetter = cursor.GetGetter(index); if (!top.Schema.TryGetColumnIndex(AnomalyDetectionEvaluator.TopKResultsColumns.Label, out index)) - throw Host.Except("Data view does not contain the 'Label' column"); + throw Host.ExceptSchemaMismatch(nameof(top.Schema), "label", AnomalyDetectionEvaluator.TopKResultsColumns.Label); var labelGetter = cursor.GetGetter(index); bool hasRows = false; @@ -671,7 +671,7 @@ private protected override void PrintFoldResultsCore(IChannel ch, Dictionary= 0) { Host.Assert(!string.IsNullOrEmpty(_probCol)); t = schema[_probIndex].Type; if (t != NumberType.Float) - throw Host.Except("Probability column '{0}' has type '{1}' but must be R4", _probCol, t); + throw Host.ExceptSchemaMismatch(nameof(schema), "probability", _probCol, "float", t.ToString()); } else if (!_useRaw) throw Host.Except("Cannot compute the predicted label from the probability column because it does not exist"); diff --git a/src/Microsoft.ML.Data/Evaluators/ClusteringEvaluator.cs b/src/Microsoft.ML.Data/Evaluators/ClusteringEvaluator.cs index c81fab91d7..0229e422a6 100644 --- a/src/Microsoft.ML.Data/Evaluators/ClusteringEvaluator.cs +++ b/src/Microsoft.ML.Data/Evaluators/ClusteringEvaluator.cs @@ -100,13 +100,13 @@ private protected override void CheckScoreAndLabelTypes(RoleMappedSchema schema) 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()); + "float or KeyType", type.ToString()); } var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score); type = score.Type; if (!(type is VectorType vectorType) || !vectorType.IsKnownSize || vectorType.ItemType != NumberType.Float) - throw Host.ExceptSchemaMismatch(nameof(schema), "score", score.Name, "R4 vector of known size", type.ToString()); + throw Host.ExceptSchemaMismatch(nameof(schema), "score", score.Name, "known-size vector of float", type.ToString()); } private protected override void CheckCustomColumnTypesCore(RoleMappedSchema schema) @@ -754,7 +754,7 @@ private void CheckInputColumnTypes(Schema schema) var type = schema[(int) ScoreIndex].Type; if (!(type is VectorType vectorType) || !vectorType.IsKnownSize || vectorType.ItemType != NumberType.Float) - throw Host.Except("Score column '{0}' has type {1}, but must be a float vector of known-size", ScoreCol, type); + throw Host.ExceptSchemaMismatch(nameof(schema), "score", ScoreCol, "known-size vector of float", type.ToString()); } } diff --git a/src/Microsoft.ML.Data/Evaluators/EvaluatorBase.cs b/src/Microsoft.ML.Data/Evaluators/EvaluatorBase.cs index a757daaab4..c7a547d367 100644 --- a/src/Microsoft.ML.Data/Evaluators/EvaluatorBase.cs +++ b/src/Microsoft.ML.Data/Evaluators/EvaluatorBase.cs @@ -486,9 +486,9 @@ protected PerInstanceEvaluatorBase(IHostEnvironment env, Schema schema, string s LabelCol = labelCol; if (!string.IsNullOrEmpty(LabelCol) && !schema.TryGetColumnIndex(LabelCol, out LabelIndex)) - throw Host.Except("Did not find the label column '{0}'", LabelCol); + throw Host.ExceptSchemaMismatch(nameof(schema), "label", LabelCol); if (!schema.TryGetColumnIndex(ScoreCol, out ScoreIndex)) - throw Host.Except("Did not find column '{0}'", ScoreCol); + throw Host.ExceptSchemaMismatch(nameof(schema), "score", ScoreCol); } protected PerInstanceEvaluatorBase(IHostEnvironment env, ModelLoadContext ctx, Schema schema) @@ -502,9 +502,9 @@ protected PerInstanceEvaluatorBase(IHostEnvironment env, ModelLoadContext ctx, ScoreCol = ctx.LoadNonEmptyString(); LabelCol = ctx.LoadStringOrNull(); if (!string.IsNullOrEmpty(LabelCol) && !schema.TryGetColumnIndex(LabelCol, out LabelIndex)) - throw Host.Except($"Did not find the label column '{LabelCol}'"); + throw Host.ExceptSchemaMismatch(nameof(schema), "label", LabelCol); if (!schema.TryGetColumnIndex(ScoreCol, out ScoreIndex)) - throw Host.Except($"Did not find column '{ScoreCol}'"); + throw Host.ExceptSchemaMismatch(nameof(schema), "score", ScoreCol); } public virtual void Save(ModelSaveContext ctx) diff --git a/src/Microsoft.ML.Data/Evaluators/MultiClassClassifierEvaluator.cs b/src/Microsoft.ML.Data/Evaluators/MultiClassClassifierEvaluator.cs index c6447281a4..448ee88e53 100644 --- a/src/Microsoft.ML.Data/Evaluators/MultiClassClassifierEvaluator.cs +++ b/src/Microsoft.ML.Data/Evaluators/MultiClassClassifierEvaluator.cs @@ -77,11 +77,11 @@ private protected override void CheckScoreAndLabelTypes(RoleMappedSchema schema) var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score); var scoreType = score.Type as VectorType; if (scoreType == null || scoreType.Size < 2 || scoreType.ItemType != NumberType.Float) - throw Host.ExceptSchemaMismatch(nameof(schema), "score", score.Name, "vector of two or more items of type R4", scoreType.ToString()); + throw Host.ExceptSchemaMismatch(nameof(schema), "score", score.Name, "vector of two or more items of type float", scoreType.ToString()); Host.CheckParam(schema.Label.HasValue, nameof(schema), "Could not find the label column"); var labelType = schema.Label.Value.Type; if (labelType != NumberType.Float && labelType.GetKeyCount() <= 0) - throw Host.ExceptSchemaMismatch(nameof(schema), "label", schema.Label.Value.Name, "float or a known-cardinality key", labelType.ToString()); + throw Host.ExceptSchemaMismatch(nameof(schema), "label", schema.Label.Value.Name, "float or KeyType", labelType.ToString()); } private protected override Aggregator GetAggregatorCore(RoleMappedSchema schema, string stratName) @@ -815,10 +815,10 @@ private void CheckInputColumnTypes(Schema schema) var scoreType = schema[ScoreIndex].Type as VectorType; if (scoreType == null || scoreType.Size < 2 || scoreType.ItemType != NumberType.Float) - throw Host.Except("Score column '{0}' has type '{1}' but must be a vector of two or more items of type R4", ScoreCol, scoreType); + throw Host.ExceptSchemaMismatch(nameof(schema), "score", ScoreCol, "vector of two or more items of type float", scoreType.ToString()); var labelType = schema[LabelIndex].Type; if (labelType != NumberType.Float && labelType.GetKeyCount() <= 0) - throw Host.Except("Label column '{0}' has type '{1}' but must be a float or a known-cardinality key", LabelCol, labelType); + throw Host.ExceptSchemaMismatch(nameof(schema), "label", LabelCol, "float or KeyType", labelType.ToString()); } } @@ -1000,7 +1000,7 @@ private protected override IDataView GetPerInstanceMetricsCore(IDataView perInst // text file, since if there are different key counts the columns cannot be appended. string labelName = schema.Label.Value.Name; if (!perInst.Schema.TryGetColumnIndex(labelName, out int labelCol)) - throw Host.Except("Could not find column '{0}'", labelName); + throw Host.ExceptSchemaMismatch(nameof(schema), "label", labelName); var labelType = perInst.Schema[labelCol].Type; if (labelType is KeyType keyType && (!perInst.Schema[labelCol].HasKeyValues(keyType) || labelType.RawType != typeof(uint))) { diff --git a/src/Microsoft.ML.Data/Evaluators/MultiOutputRegressionEvaluator.cs b/src/Microsoft.ML.Data/Evaluators/MultiOutputRegressionEvaluator.cs index 7be7cdae8c..5de03bcf8a 100644 --- a/src/Microsoft.ML.Data/Evaluators/MultiOutputRegressionEvaluator.cs +++ b/src/Microsoft.ML.Data/Evaluators/MultiOutputRegressionEvaluator.cs @@ -60,11 +60,11 @@ private protected override void CheckScoreAndLabelTypes(RoleMappedSchema schema) var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score); var t = score.Type as VectorType; if (t == null || !t.IsKnownSize || t.ItemType != NumberType.Float) - throw Host.ExceptSchemaMismatch(nameof(schema), "score", score.Name, "known size vector of R4", t.ToString()); + throw Host.ExceptSchemaMismatch(nameof(schema), "score", score.Name, "known-size vector of float", t.ToString()); Host.Check(schema.Label.HasValue, "Could not find the label column"); t = schema.Label.Value.Type as VectorType; if (t == null || !t.IsKnownSize || (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()); + throw Host.ExceptSchemaMismatch(nameof(schema), "label", schema.Label.Value.Name, "known-size vector of float or double", t.ToString()); } private protected override Aggregator GetAggregatorCore(RoleMappedSchema schema, string stratName) @@ -548,7 +548,7 @@ private void CheckInputColumnTypes(Schema schema, out VectorType labelType, out var t = schema[LabelIndex].Type as VectorType; if (t == null || !t.IsKnownSize || (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", LabelCol, t); + throw Host.ExceptSchemaMismatch(nameof(schema), "label", LabelCol, "known-size vector of float or double", t.ToString()); labelType = new VectorType((PrimitiveType)t.ItemType, t.Size); var slotNamesType = new VectorType(TextType.Instance, t.Size); var builder = new MetadataBuilder(); @@ -557,7 +557,7 @@ private void CheckInputColumnTypes(Schema schema, out VectorType labelType, out t = schema[ScoreIndex].Type as VectorType; if (t == null || !t.IsKnownSize || t.ItemType != NumberType.Float) - throw Host.Except("Score column '{0}' has type '{1}' but must be a known length vector of type R4", ScoreCol, t); + throw Host.ExceptSchemaMismatch(nameof(schema), "score", ScoreCol, "known-size vector of float", t.ToString()); scoreType = new VectorType((PrimitiveType)t.ItemType, t.Size); builder = new MetadataBuilder(); builder.AddSlotNames(t.Size, CreateSlotNamesGetter(schema, ScoreIndex, scoreType.Size, "Predicted")); diff --git a/src/Microsoft.ML.Data/Evaluators/QuantileRegressionEvaluator.cs b/src/Microsoft.ML.Data/Evaluators/QuantileRegressionEvaluator.cs index 32e2f75d3a..01e1f2f70a 100644 --- a/src/Microsoft.ML.Data/Evaluators/QuantileRegressionEvaluator.cs +++ b/src/Microsoft.ML.Data/Evaluators/QuantileRegressionEvaluator.cs @@ -58,11 +58,11 @@ private protected override void CheckScoreAndLabelTypes(RoleMappedSchema schema) var score = schema.GetUniqueColumn(MetadataUtils.Const.ScoreValueKind.Score); var t = score.Type as VectorType; if (t == null || t.Size == 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()); + throw Host.ExceptSchemaMismatch(nameof(schema), "score", score.Name, "vector of float or double", t.ToString()); Host.CheckParam(schema.Label.HasValue, nameof(schema), "Must contain a label column"); var labelType = schema.Label.Value.Type; if (labelType != NumberType.R4) - throw Host.ExceptSchemaMismatch(nameof(schema), "label", schema.Label.Value.Name, "R4", t.ToString()); + throw Host.ExceptSchemaMismatch(nameof(schema), "label", schema.Label.Value.Name, "float", t.ToString()); } private protected override Aggregator GetAggregatorCore(RoleMappedSchema schema, string stratName) @@ -446,13 +446,13 @@ private void CheckInputColumnTypes(Schema schema) var t = schema[(int)LabelIndex].Type; if (t != NumberType.R4) - throw Host.Except("Label column '{0}' has type '{1}' but must be R4", LabelCol, t); + throw Host.ExceptSchemaMismatch(nameof(schema), "label", LabelCol, "float", t.ToString()); VectorType scoreType = schema[ScoreIndex].Type as VectorType; if (scoreType == null || scoreType.Size == 0 || (scoreType.ItemType != NumberType.R4 && scoreType.ItemType != NumberType.R8)) { - throw Host.Except( - "Score column '{0}' has type '{1}' but must be a known length vector of type R4 or R8", ScoreCol, t); + throw Host.ExceptSchemaMismatch(nameof(schema), "score", ScoreCol, "known-size vector of float or double", t.ToString()); + } } } diff --git a/src/Microsoft.ML.Data/Evaluators/RegressionEvaluator.cs b/src/Microsoft.ML.Data/Evaluators/RegressionEvaluator.cs index b04634b2c3..bbf5ed84d1 100644 --- a/src/Microsoft.ML.Data/Evaluators/RegressionEvaluator.cs +++ b/src/Microsoft.ML.Data/Evaluators/RegressionEvaluator.cs @@ -57,11 +57,11 @@ 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()); + throw Host.ExceptSchemaMismatch(nameof(schema), "score", score.Name, "float", 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()); + throw Host.ExceptSchemaMismatch(nameof(schema), "label", schema.Label.Value.Name, "float", t.ToString()); } private protected override Aggregator GetAggregatorCore(RoleMappedSchema schema, string stratName) @@ -319,11 +319,11 @@ private void CheckInputColumnTypes(Schema schema) var t = schema[(int) LabelIndex].Type; if (t != NumberType.R4) - throw Host.Except("Label column '{0}' has type '{1}' but must be R4", LabelCol, t); + throw Host.ExceptSchemaMismatch(nameof(schema), "label", LabelCol, "float", t.ToString()); t = schema[ScoreIndex].Type; if (t != NumberType.Float) - throw Host.Except("Score column '{0}' has type '{1}' but must be R4", ScoreCol, t); + throw Host.ExceptSchemaMismatch(nameof(schema), "score", ScoreCol, "float", t.ToString()); } } diff --git a/src/Microsoft.ML.Data/Prediction/CalibratorCatalog.cs b/src/Microsoft.ML.Data/Prediction/CalibratorCatalog.cs index f4b3a74e61..ca384c0e4e 100644 --- a/src/Microsoft.ML.Data/Prediction/CalibratorCatalog.cs +++ b/src/Microsoft.ML.Data/Prediction/CalibratorCatalog.cs @@ -94,14 +94,14 @@ protected CalibratorEstimatorBase(IHostEnvironment env, /// The input . SchemaShape IEstimator>.GetOutputSchema(SchemaShape inputSchema) { - Action checkColumnValid = (SchemaShape.Column column, string expected) => + Action checkColumnValid = (SchemaShape.Column column, string columnRole) => { if (column.IsValid) { if (!inputSchema.TryFindColumn(column.Name, out var outCol)) - throw Host.Except($"{expected} column '{column.Name}' is not found"); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), columnRole, column.Name); if (!column.IsCompatibleWith(outCol)) - throw Host.Except($"{expected} column '{column.Name}' is not compatible"); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), columnRole, column.Name, column.GetTypeString(), outCol.GetTypeString()); } }; diff --git a/src/Microsoft.ML.Data/Scorers/PredictionTransformer.cs b/src/Microsoft.ML.Data/Scorers/PredictionTransformer.cs index d7b3eda05d..78c17c857c 100644 --- a/src/Microsoft.ML.Data/Scorers/PredictionTransformer.cs +++ b/src/Microsoft.ML.Data/Scorers/PredictionTransformer.cs @@ -177,7 +177,7 @@ public SingleFeaturePredictionTransformerBase(IHost host, TModel model, Schema t if (featureColumn == null) FeatureColumnType = null; else if (!trainSchema.TryGetColumnIndex(featureColumn, out int col)) - throw Host.ExceptSchemaMismatch(nameof(featureColumn), RoleMappedSchema.ColumnRole.Feature.Value, featureColumn); + throw Host.ExceptSchemaMismatch(nameof(featureColumn), "feature", featureColumn); else FeatureColumnType = trainSchema[col].Type; @@ -192,7 +192,7 @@ internal SingleFeaturePredictionTransformerBase(IHost host, ModelLoadContext ctx if (FeatureColumn == null) FeatureColumnType = null; else if (!TrainSchema.TryGetColumnIndex(FeatureColumn, out int col)) - throw Host.ExceptSchemaMismatch(nameof(FeatureColumn), RoleMappedSchema.ColumnRole.Feature.Value, FeatureColumn); + throw Host.ExceptSchemaMismatch(nameof(FeatureColumn), "feature", FeatureColumn); else FeatureColumnType = TrainSchema[col].Type; @@ -206,9 +206,9 @@ public override Schema GetOutputSchema(Schema inputSchema) if (FeatureColumn != null) { if (!inputSchema.TryGetColumnIndex(FeatureColumn, out int col)) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), RoleMappedSchema.ColumnRole.Feature.Value, FeatureColumn, FeatureColumnType.ToString(), null); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "feature", FeatureColumn); if (!inputSchema[col].Type.Equals(FeatureColumnType)) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), RoleMappedSchema.ColumnRole.Feature.Value, FeatureColumn, FeatureColumnType.ToString(), inputSchema[col].Type.ToString()); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "feature", FeatureColumn, FeatureColumnType.ToString(), inputSchema[col].Type.ToString()); } return Transform(new EmptyDataView(Host, inputSchema)).Schema; diff --git a/src/Microsoft.ML.Data/Training/TrainerEstimatorBase.cs b/src/Microsoft.ML.Data/Training/TrainerEstimatorBase.cs index eb6af5f57f..7db8d2e47d 100644 --- a/src/Microsoft.ML.Data/Training/TrainerEstimatorBase.cs +++ b/src/Microsoft.ML.Data/Training/TrainerEstimatorBase.cs @@ -95,16 +95,18 @@ private void CheckInputSchema(SchemaShape inputSchema) { // Verify that all required input columns are present, and are of the same type. if (!inputSchema.TryFindColumn(FeatureColumn.Name, out var featureCol)) - throw Host.Except($"Feature column '{FeatureColumn.Name}' is not found"); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "feature", FeatureColumn.Name); if (!FeatureColumn.IsCompatibleWith(featureCol)) - throw Host.Except($"Feature column '{FeatureColumn.Name}' is not compatible"); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "feature", FeatureColumn.Name, + FeatureColumn.GetTypeString(), featureCol.GetTypeString()); if (WeightColumn.IsValid) { if (!inputSchema.TryFindColumn(WeightColumn.Name, out var weightCol)) - throw Host.Except($"Weight column '{WeightColumn.Name}' is not found"); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "weight", WeightColumn.Name); if (!WeightColumn.IsCompatibleWith(weightCol)) - throw Host.Except($"Weight column '{WeightColumn.Name}' is not compatible"); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "weight", WeightColumn.Name, + WeightColumn.GetTypeString(), weightCol.GetTypeString()); } // Special treatment for label column: we allow different types of labels, so the trainers @@ -112,7 +114,7 @@ private void CheckInputSchema(SchemaShape inputSchema) if (LabelColumn.IsValid) { if (!inputSchema.TryFindColumn(LabelColumn.Name, out var labelCol)) - throw Host.Except($"Label column '{LabelColumn.Name}' is not found"); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "label", LabelColumn.Name); CheckLabelCompatible(labelCol); } } @@ -123,7 +125,8 @@ protected virtual void CheckLabelCompatible(SchemaShape.Column labelCol) Host.Assert(LabelColumn.IsValid); if (!LabelColumn.IsCompatibleWith(labelCol)) - throw Host.Except($"Label column '{LabelColumn.Name}' is not compatible"); + throw Host.ExceptSchemaMismatch(nameof(labelCol), "label", WeightColumn.Name, + LabelColumn.GetTypeString(), labelCol.GetTypeString()); } protected TTransformer TrainTransformer(IDataView trainSet, diff --git a/src/Microsoft.ML.Data/Transforms/FeatureContributionCalculationTransform.cs b/src/Microsoft.ML.Data/Transforms/FeatureContributionCalculationTransform.cs index 66f3f92aca..dd49ed484c 100644 --- a/src/Microsoft.ML.Data/Transforms/FeatureContributionCalculationTransform.cs +++ b/src/Microsoft.ML.Data/Transforms/FeatureContributionCalculationTransform.cs @@ -225,7 +225,7 @@ public Mapper(FeatureContributionCalculatingTransformer parent, Schema schema) throw Host.ExceptSchemaMismatch(nameof(schema), "input", _parent.ColumnPairs[0].inputColumnName); _featureColumnType = schema[_featureColumnIndex].Type as VectorType; if (_featureColumnType == null || _featureColumnType.ItemType != NumberType.R4) - throw Host.ExceptSchemaMismatch(nameof(schema), "feature column", _parent.ColumnPairs[0].inputColumnName, "Expected type is vector of float.", _featureColumnType.ItemType.ToString()); + throw Host.ExceptSchemaMismatch(nameof(schema), "feature", _parent.ColumnPairs[0].inputColumnName, "vector of float.", _featureColumnType.ItemType.ToString()); if (InputSchema[_featureColumnIndex].HasSlotNames(_featureColumnType.Size)) InputSchema[_featureColumnIndex].Metadata.GetValue(MetadataUtils.Kinds.SlotNames, ref _slotNames); @@ -320,7 +320,7 @@ public override SchemaShape GetOutputSchema(SchemaShape inputSchema) throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", _featureColumn); // Check that the feature column is of the correct type: a vector of float. if (col.ItemType != NumberType.R4 || col.Kind != SchemaShape.Column.VectorKind.Vector) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), "feature column", _featureColumn, "Expected type is vector of float.", col.GetTypeString()); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "column", _featureColumn, "vector of float.", col.GetTypeString()); // Build output schemaShape. var result = inputSchema.ToDictionary(x => x.Name); diff --git a/src/Microsoft.ML.Data/Transforms/KeyToValue.cs b/src/Microsoft.ML.Data/Transforms/KeyToValue.cs index f0bba8d894..0fb1ca6e75 100644 --- a/src/Microsoft.ML.Data/Transforms/KeyToValue.cs +++ b/src/Microsoft.ML.Data/Transforms/KeyToValue.cs @@ -524,7 +524,7 @@ public override SchemaShape GetOutputSchema(SchemaShape inputSchema) if (!inputSchema.TryFindColumn(colInfo.inputColumnName, out var col)) throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", colInfo.inputColumnName); if (!col.IsKey) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", colInfo.inputColumnName, "key type", col.GetTypeString()); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", colInfo.inputColumnName, "KeyType", col.GetTypeString()); if (!col.Metadata.TryFindColumn(MetadataUtils.Kinds.KeyValues, out var keyMetaCol)) throw Host.ExceptParam(nameof(inputSchema), $"Input column '{colInfo.inputColumnName}' doesn't contain key values metadata"); diff --git a/src/Microsoft.ML.Data/Transforms/NAFilter.cs b/src/Microsoft.ML.Data/Transforms/NAFilter.cs index d44dd36e4c..f682a8da6d 100644 --- a/src/Microsoft.ML.Data/Transforms/NAFilter.cs +++ b/src/Microsoft.ML.Data/Transforms/NAFilter.cs @@ -146,13 +146,13 @@ public NAFilter(IHost host, ModelLoadContext ctx, IDataView input) string src = ctx.LoadNonEmptyString(); int index; if (!schema.TryGetColumnIndex(src, out index)) - throw Host.Except("Source column '{0}' not found", src); + throw Host.ExceptSchemaMismatch(nameof(schema), "source", src); if (_srcIndexToInfoIndex.ContainsKey(index)) throw Host.Except("Source column '{0}' specified multiple times", src); var type = schema[index].Type; if (!TestType(type)) - throw Host.Except($"Column '{src}' has type {type} which does not support missing values, so we cannot filter on them", src); + throw Host.ExceptSchemaMismatch(nameof(schema), "source", src, "scalar or vector of float, double or KeyType", type.ToString()); _infos[i] = new ColInfo(index, type); _srcIndexToInfoIndex.Add(index, i); diff --git a/src/Microsoft.ML.Data/Transforms/Normalizer.cs b/src/Microsoft.ML.Data/Transforms/Normalizer.cs index e5091f30cf..76626fe7e9 100644 --- a/src/Microsoft.ML.Data/Transforms/Normalizer.cs +++ b/src/Microsoft.ML.Data/Transforms/Normalizer.cs @@ -255,10 +255,10 @@ public SchemaShape GetOutputSchema(SchemaShape inputSchema) if (!inputSchema.TryFindColumn(colInfo.InputColumnName, out var col)) throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", colInfo.InputColumnName); if (col.Kind == SchemaShape.Column.VectorKind.VariableVector) - throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", colInfo.InputColumnName, "fixed-size vector or scalar", col.GetTypeString()); + throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", colInfo.InputColumnName, "known-size vector or scalar", col.GetTypeString()); if (!col.ItemType.Equals(NumberType.R4) && !col.ItemType.Equals(NumberType.R8)) - throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", colInfo.InputColumnName, "vector or scalar of R4 or R8", col.GetTypeString()); + throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", colInfo.InputColumnName, "vector or scalar of float or double", col.GetTypeString()); var isNormalizedMeta = new SchemaShape.Column(MetadataUtils.Kinds.IsNormalized, SchemaShape.Column.VectorKind.Scalar, BoolType.Instance, false); diff --git a/src/Microsoft.ML.Data/Transforms/RangeFilter.cs b/src/Microsoft.ML.Data/Transforms/RangeFilter.cs index ce3e21bd70..8191de206e 100644 --- a/src/Microsoft.ML.Data/Transforms/RangeFilter.cs +++ b/src/Microsoft.ML.Data/Transforms/RangeFilter.cs @@ -152,11 +152,11 @@ private RangeFilter(IHost host, ModelLoadContext ctx, IDataView input) var column = ctx.LoadNonEmptyString(); var schema = Source.Schema; if (!schema.TryGetColumnIndex(column, out _index)) - throw Host.Except("column", "Source column '{0}' not found", column); + throw Host.ExceptSchemaMismatch(nameof(schema), "source", column); _type = schema[_index].Type; if (_type != NumberType.R4 && _type != NumberType.R8 && _type.GetKeyCount() == 0) - throw Host.Except("column", "Column '{0}' does not have compatible type", column); + throw Host.ExceptSchemaMismatch(nameof(schema), "source", column, "float, double or KeyType", _type.ToString()); _min = ctx.Reader.ReadDouble(); _max = ctx.Reader.ReadDouble(); diff --git a/src/Microsoft.ML.Data/Transforms/TransformBase.cs b/src/Microsoft.ML.Data/Transforms/TransformBase.cs index 8a34204b1e..ec8dec1af9 100644 --- a/src/Microsoft.ML.Data/Transforms/TransformBase.cs +++ b/src/Microsoft.ML.Data/Transforms/TransformBase.cs @@ -373,7 +373,7 @@ public static Bindings Create(OneToOneTransformBase parent, ModelLoadContext ctx int colSrc; if (!inputSchema.TryGetColumnIndex(src, out colSrc)) - throw host.Except("Source column '{0}' is required but not found", src); + throw host.ExceptSchemaMismatch(nameof(inputSchema), "source", src); var type = inputSchema[colSrc].Type; if (testType != null) { diff --git a/src/Microsoft.ML.EntryPoints/FeatureCombiner.cs b/src/Microsoft.ML.EntryPoints/FeatureCombiner.cs index 10a724c327..006f990b7f 100644 --- a/src/Microsoft.ML.EntryPoints/FeatureCombiner.cs +++ b/src/Microsoft.ML.EntryPoints/FeatureCombiner.cs @@ -234,7 +234,7 @@ public static CommonOutputs.TransformOutput PrepareClassificationLabel(IHostEnvi var labelCol = input.Data.Schema.GetColumnOrNull(input.LabelColumn); if (!labelCol.HasValue) - throw host.ExceptSchemaMismatch(nameof(input), "Label", input.LabelColumn); + throw host.ExceptSchemaMismatch(nameof(input), "predicted label", input.LabelColumn); var labelType = labelCol.Value.Type; if (labelType is KeyType || labelType is BoolType) @@ -270,7 +270,7 @@ public static CommonOutputs.TransformOutput ConvertPredictedLabel(IHostEnvironme var predictedLabelCol = input.Data.Schema.GetColumnOrNull(input.PredictedLabelColumn); if (!predictedLabelCol.HasValue) - throw host.ExceptSchemaMismatch(nameof(input), "PredictedLabel",input.PredictedLabelColumn); + throw host.ExceptSchemaMismatch(nameof(input), "label", input.PredictedLabelColumn); var predictedLabelType = predictedLabelCol.Value.Type; if (predictedLabelType is NumberType || predictedLabelType is BoolType) { diff --git a/src/Microsoft.ML.FastTree/FastTreeRanking.cs b/src/Microsoft.ML.FastTree/FastTreeRanking.cs index 591b16d326..2cca0b0add 100644 --- a/src/Microsoft.ML.FastTree/FastTreeRanking.cs +++ b/src/Microsoft.ML.FastTree/FastTreeRanking.cs @@ -101,7 +101,7 @@ protected override void CheckLabelCompatible(SchemaShape.Column labelCol) Contracts.Assert(labelCol.IsValid); Action error = - () => throw Host.ExceptSchemaMismatch(nameof(labelCol), RoleMappedSchema.ColumnRole.Label.Value, labelCol.Name, "R4 or a Key", labelCol.GetTypeString()); + () => throw Host.ExceptSchemaMismatch(nameof(labelCol), "label", labelCol.Name, "float or KeyType", labelCol.GetTypeString()); if (labelCol.Kind != SchemaShape.Column.VectorKind.Scalar) error(); diff --git a/src/Microsoft.ML.FastTree/TreeEnsembleFeaturizer.cs b/src/Microsoft.ML.FastTree/TreeEnsembleFeaturizer.cs index b1cb260163..2a400d034d 100644 --- a/src/Microsoft.ML.FastTree/TreeEnsembleFeaturizer.cs +++ b/src/Microsoft.ML.FastTree/TreeEnsembleFeaturizer.cs @@ -720,7 +720,7 @@ private static IDataView AppendLabelTransform(IHostEnvironment env, IChannel ch, var col = input.Schema.GetColumnOrNull(labelName); if (!col.HasValue) - throw ch.ExceptSchemaMismatch(nameof(input), "Label", labelName); + throw ch.ExceptSchemaMismatch(nameof(input), "label", labelName); ColumnType labelType = col.Value.Type; if (!(labelType is KeyType)) diff --git a/src/Microsoft.ML.LightGBM/LightGbmRankingTrainer.cs b/src/Microsoft.ML.LightGBM/LightGbmRankingTrainer.cs index a1860d569f..6c36cf390f 100644 --- a/src/Microsoft.ML.LightGBM/LightGbmRankingTrainer.cs +++ b/src/Microsoft.ML.LightGBM/LightGbmRankingTrainer.cs @@ -141,7 +141,7 @@ protected override void CheckLabelCompatible(SchemaShape.Column labelCol) Contracts.Assert(labelCol.IsValid); Action error = - () => throw Host.ExceptSchemaMismatch(nameof(labelCol), RoleMappedSchema.ColumnRole.Label.Value, labelCol.Name, "R4 or a Key", labelCol.GetTypeString()); + () => throw Host.ExceptSchemaMismatch(nameof(labelCol), "label", labelCol.Name, "float or KeyType", labelCol.GetTypeString()); if (labelCol.Kind != SchemaShape.Column.VectorKind.Scalar) error(); diff --git a/src/Microsoft.ML.OnnxTransform/OnnxTransform.cs b/src/Microsoft.ML.OnnxTransform/OnnxTransform.cs index a46e7c963e..79d7894db3 100644 --- a/src/Microsoft.ML.OnnxTransform/OnnxTransform.cs +++ b/src/Microsoft.ML.OnnxTransform/OnnxTransform.cs @@ -559,7 +559,7 @@ public override SchemaShape GetOutputSchema(SchemaShape inputSchema) if (!inputSchema.TryFindColumn(input, out var col)) throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", input); if (!(col.Kind == SchemaShape.Column.VectorKind.VariableVector || col.Kind == SchemaShape.Column.VectorKind.Vector)) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", input, nameof(VectorType), col.GetTypeString()); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", input, "vector", col.GetTypeString()); var inputsInfo = Transformer.Model.ModelInfo.InputsInfo; var idx = Transformer.Model.InputNames.IndexOf(input); diff --git a/src/Microsoft.ML.PCA/PcaTransformer.cs b/src/Microsoft.ML.PCA/PcaTransformer.cs index ec8704c5c7..0fbce95ffd 100644 --- a/src/Microsoft.ML.PCA/PcaTransformer.cs +++ b/src/Microsoft.ML.PCA/PcaTransformer.cs @@ -551,7 +551,7 @@ internal static void ValidatePcaInput(IExceptionContext ectx, string name, Colum string inputSchema; // just used for the excpections if (!(type is VectorType vectorType && vectorType.Size > 1 && vectorType.ItemType.Equals(NumberType.R4))) - throw ectx.ExceptSchemaMismatch(nameof(inputSchema), "input", name, "vector of floats with fixed size greater than 1", type.ToString()); + throw ectx.ExceptSchemaMismatch(nameof(inputSchema), "input", name, "known-size vector of float of two or more items", type.ToString()); } private sealed class Mapper : OneToOneMapperBase diff --git a/src/Microsoft.ML.Recommender/MatrixFactorizationPredictor.cs b/src/Microsoft.ML.Recommender/MatrixFactorizationPredictor.cs index a967692747..e40740e2e0 100644 --- a/src/Microsoft.ML.Recommender/MatrixFactorizationPredictor.cs +++ b/src/Microsoft.ML.Recommender/MatrixFactorizationPredictor.cs @@ -407,10 +407,10 @@ public MatrixFactorizationPredictionTransformer(IHostEnvironment env, MatrixFact MatrixRowIndexColumnName = matrixRowIndexColumnName; if (!trainSchema.TryGetColumnIndex(MatrixColumnIndexColumnName, out int xCol)) - throw Host.ExceptSchemaMismatch(nameof(MatrixColumnIndexColumnName), RecommenderUtils.MatrixColumnIndexKind.Value, MatrixColumnIndexColumnName); + throw Host.ExceptSchemaMismatch(nameof(MatrixColumnIndexColumnName), "matrixColumnIndex", MatrixColumnIndexColumnName); MatrixColumnIndexColumnType = trainSchema[xCol].Type; if (!trainSchema.TryGetColumnIndex(MatrixRowIndexColumnName, out int yCol)) - throw Host.ExceptSchemaMismatch(nameof(yCol), RecommenderUtils.MatrixRowIndexKind.Value, MatrixRowIndexColumnName); + throw Host.ExceptSchemaMismatch(nameof(yCol), "matrixRowIndex", MatrixRowIndexColumnName); MatrixRowIndexColumnType = trainSchema[yCol].Type; BindableMapper = ScoreUtils.GetSchemaBindableMapper(Host, model); @@ -445,11 +445,11 @@ public MatrixFactorizationPredictionTransformer(IHostEnvironment host, ModelLoad MatrixRowIndexColumnName = ctx.LoadString(); if (!TrainSchema.TryGetColumnIndex(MatrixColumnIndexColumnName, out int xCol)) - throw Host.ExceptSchemaMismatch(nameof(MatrixColumnIndexColumnName), RecommenderUtils.MatrixColumnIndexKind.Value, MatrixColumnIndexColumnName); + throw Host.ExceptSchemaMismatch(nameof(MatrixColumnIndexColumnName), "matrixColumnIndex", MatrixColumnIndexColumnName); MatrixColumnIndexColumnType = TrainSchema[xCol].Type; if (!TrainSchema.TryGetColumnIndex(MatrixRowIndexColumnName, out int yCol)) - throw Host.ExceptSchemaMismatch(nameof(MatrixRowIndexColumnName), RecommenderUtils.MatrixRowIndexKind.Value, MatrixRowIndexColumnName); + throw Host.ExceptSchemaMismatch(nameof(MatrixRowIndexColumnName), "matrixRowIndex", MatrixRowIndexColumnName); MatrixRowIndexColumnType = TrainSchema[yCol].Type; BindableMapper = ScoreUtils.GetSchemaBindableMapper(Host, Model); @@ -462,9 +462,9 @@ public MatrixFactorizationPredictionTransformer(IHostEnvironment host, ModelLoad public override Schema GetOutputSchema(Schema inputSchema) { if (!inputSchema.TryGetColumnIndex(MatrixColumnIndexColumnName, out int xCol)) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), RecommenderUtils.MatrixColumnIndexKind.Value, MatrixColumnIndexColumnName); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "matrixColumnIndex", MatrixColumnIndexColumnName); if (!inputSchema.TryGetColumnIndex(MatrixRowIndexColumnName, out int yCol)) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), RecommenderUtils.MatrixRowIndexKind.Value, MatrixRowIndexColumnName); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "matrixRowIndex", MatrixRowIndexColumnName); return Transform(new EmptyDataView(Host, inputSchema)).Schema; } diff --git a/src/Microsoft.ML.Recommender/MatrixFactorizationTrainer.cs b/src/Microsoft.ML.Recommender/MatrixFactorizationTrainer.cs index f7a47cd31c..c3696d22bf 100644 --- a/src/Microsoft.ML.Recommender/MatrixFactorizationTrainer.cs +++ b/src/Microsoft.ML.Recommender/MatrixFactorizationTrainer.cs @@ -431,24 +431,25 @@ public SchemaShape GetOutputSchema(SchemaShape inputSchema) { Host.CheckValue(inputSchema, nameof(inputSchema)); - void CheckColumnsCompatible(SchemaShape.Column cachedColumn, string expectedColumnName) + void CheckColumnsCompatible(SchemaShape.Column cachedColumn, string columnRole) { if (!inputSchema.TryFindColumn(cachedColumn.Name, out var col)) - throw Host.ExceptSchemaMismatch(nameof(col), expectedColumnName, expectedColumnName); + throw Host.ExceptSchemaMismatch(nameof(col), columnRole, cachedColumn.Name); if (!cachedColumn.IsCompatibleWith(col)) - throw Host.Except($"{expectedColumnName} column '{cachedColumn.Name}' is not compatible"); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), columnRole, cachedColumn.Name, + cachedColumn.GetTypeString(), col.GetTypeString()); } // Check if label column is good. var labelColumn = new SchemaShape.Column(LabelName, SchemaShape.Column.VectorKind.Scalar, NumberType.R4, false); - CheckColumnsCompatible(labelColumn, LabelName); + CheckColumnsCompatible(labelColumn, "label"); // Check if columns of matrix's row and column indexes are good. Note that column of IDataView and column of matrix are two different things. var matrixColumnIndexColumn = new SchemaShape.Column(MatrixColumnIndexName, SchemaShape.Column.VectorKind.Scalar, NumberType.U4, true); var matrixRowIndexColumn = new SchemaShape.Column(MatrixRowIndexName, SchemaShape.Column.VectorKind.Scalar, NumberType.U4, true); - CheckColumnsCompatible(matrixColumnIndexColumn, MatrixColumnIndexName); - CheckColumnsCompatible(matrixRowIndexColumn, MatrixRowIndexName); + CheckColumnsCompatible(matrixColumnIndexColumn, "matrixColumnIndex"); + CheckColumnsCompatible(matrixRowIndexColumn, "matrixRowIndex"); // Input columns just pass through so that output column dictionary contains all input columns. var outColumns = inputSchema.ToDictionary(x => x.Name); diff --git a/src/Microsoft.ML.StandardLearners/FactorizationMachine/FactorizationMachineTrainer.cs b/src/Microsoft.ML.StandardLearners/FactorizationMachine/FactorizationMachineTrainer.cs index dfd474dc78..aa6a122329 100644 --- a/src/Microsoft.ML.StandardLearners/FactorizationMachine/FactorizationMachineTrainer.cs +++ b/src/Microsoft.ML.StandardLearners/FactorizationMachine/FactorizationMachineTrainer.cs @@ -518,25 +518,26 @@ public SchemaShape GetOutputSchema(SchemaShape inputSchema) Host.CheckValue(inputSchema, nameof(inputSchema)); - void CheckColumnsCompatible(SchemaShape.Column column, string defaultName) + void CheckColumnsCompatible(SchemaShape.Column column, string columnRole) { if (!inputSchema.TryFindColumn(column.Name, out var col)) - throw Host.ExceptSchemaMismatch(nameof(col), defaultName, defaultName); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), columnRole, column.Name); if (!column.IsCompatibleWith(col)) - throw Host.Except($"{defaultName} column '{column.Name}' is not compatible"); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), columnRole, column.Name, + column.GetTypeString(), col.GetTypeString()); } - CheckColumnsCompatible(LabelColumn, DefaultColumnNames.Label); + CheckColumnsCompatible(LabelColumn, "label"); foreach (var feat in FeatureColumns) { - CheckColumnsCompatible(feat, DefaultColumnNames.Features); + CheckColumnsCompatible(feat, "feature"); } if (WeightColumn.IsValid) - CheckColumnsCompatible(WeightColumn, DefaultColumnNames.Weight); + CheckColumnsCompatible(WeightColumn, "weight"); var outColumns = inputSchema.ToDictionary(x => x.Name); foreach (var col in GetOutputColumnsCore(inputSchema)) diff --git a/src/Microsoft.ML.StandardLearners/FactorizationMachine/FieldAwareFactorizationMachineModelParameters.cs b/src/Microsoft.ML.StandardLearners/FactorizationMachine/FieldAwareFactorizationMachineModelParameters.cs index 43006d4b24..675c3d2799 100644 --- a/src/Microsoft.ML.StandardLearners/FactorizationMachine/FieldAwareFactorizationMachineModelParameters.cs +++ b/src/Microsoft.ML.StandardLearners/FactorizationMachine/FieldAwareFactorizationMachineModelParameters.cs @@ -321,7 +321,7 @@ public FieldAwareFactorizationMachinePredictionTransformer(IHostEnvironment host foreach (var feat in featureColumns) { if (!trainSchema.TryGetColumnIndex(feat, out int col)) - throw Host.ExceptSchemaMismatch(nameof(featureColumns), RoleMappedSchema.ColumnRole.Feature.Value, feat); + throw Host.ExceptSchemaMismatch(nameof(featureColumns), "feature", feat); FeatureColumnTypes[i++] = trainSchema[col].Type; } @@ -351,7 +351,7 @@ public FieldAwareFactorizationMachinePredictionTransformer(IHostEnvironment host { FeatureColumns[i] = ctx.LoadString(); if (!TrainSchema.TryGetColumnIndex(FeatureColumns[i], out int col)) - throw Host.ExceptSchemaMismatch(nameof(FeatureColumns), RoleMappedSchema.ColumnRole.Feature.Value, FeatureColumns[i]); + throw Host.ExceptSchemaMismatch(nameof(FeatureColumns), "feature", FeatureColumns[i]); FeatureColumnTypes[i] = TrainSchema[col].Type; } @@ -376,10 +376,10 @@ public override Schema GetOutputSchema(Schema inputSchema) { var feat = FeatureColumns[i]; if (!inputSchema.TryGetColumnIndex(feat, out int col)) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), RoleMappedSchema.ColumnRole.Feature.Value, feat, FeatureColumnTypes[i].ToString(), null); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "feature", feat, FeatureColumnTypes[i].ToString(), null); if (!inputSchema[col].Type.Equals(FeatureColumnTypes[i])) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), RoleMappedSchema.ColumnRole.Feature.Value, feat, FeatureColumnTypes[i].ToString(), inputSchema[col].Type.ToString()); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "feature", feat, FeatureColumnTypes[i].ToString(), inputSchema[col].Type.ToString()); } return Transform(new EmptyDataView(Host, inputSchema)).Schema; diff --git a/src/Microsoft.ML.StandardLearners/Standard/MultiClass/MetaMulticlassTrainer.cs b/src/Microsoft.ML.StandardLearners/Standard/MultiClass/MetaMulticlassTrainer.cs index 42e6de63d7..eac950e68a 100644 --- a/src/Microsoft.ML.StandardLearners/Standard/MultiClass/MetaMulticlassTrainer.cs +++ b/src/Microsoft.ML.StandardLearners/Standard/MultiClass/MetaMulticlassTrainer.cs @@ -153,10 +153,10 @@ public SchemaShape GetOutputSchema(SchemaShape inputSchema) if (LabelColumn.IsValid) { if (!inputSchema.TryFindColumn(LabelColumn.Name, out var labelCol)) - throw Host.ExceptSchemaMismatch(nameof(labelCol), DefaultColumnNames.PredictedLabel, DefaultColumnNames.PredictedLabel); + throw Host.ExceptSchemaMismatch(nameof(labelCol), "label", LabelColumn.Name); if (!LabelColumn.IsCompatibleWith(labelCol)) - throw Host.Except($"Label column '{LabelColumn.Name}' is not compatible"); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "label", LabelColumn.Name, LabelColumn.GetTypeString(), labelCol.GetTypeString()); } var outColumns = inputSchema.ToDictionary(x => x.Name); diff --git a/src/Microsoft.ML.StandardLearners/Standard/Online/AveragedPerceptron.cs b/src/Microsoft.ML.StandardLearners/Standard/Online/AveragedPerceptron.cs index 9c9275d06f..f79c7c1e10 100644 --- a/src/Microsoft.ML.StandardLearners/Standard/Online/AveragedPerceptron.cs +++ b/src/Microsoft.ML.StandardLearners/Standard/Online/AveragedPerceptron.cs @@ -164,7 +164,7 @@ protected override void CheckLabelCompatible(SchemaShape.Column labelCol) Contracts.Assert(labelCol.IsValid); Action error = - () => throw Host.ExceptSchemaMismatch(nameof(labelCol), RoleMappedSchema.ColumnRole.Label.Value, labelCol.Name, "BL, R8, R4 or a Key", labelCol.GetTypeString()); + () => throw Host.ExceptSchemaMismatch(nameof(labelCol), "label", labelCol.Name, "float, double, bool or KeyType", labelCol.GetTypeString()); if (labelCol.Kind != SchemaShape.Column.VectorKind.Scalar) error(); diff --git a/src/Microsoft.ML.StandardLearners/Standard/SdcaBinary.cs b/src/Microsoft.ML.StandardLearners/Standard/SdcaBinary.cs index 3b4d23c022..4b1d2515bf 100644 --- a/src/Microsoft.ML.StandardLearners/Standard/SdcaBinary.cs +++ b/src/Microsoft.ML.StandardLearners/Standard/SdcaBinary.cs @@ -1538,7 +1538,7 @@ protected override void CheckLabelCompatible(SchemaShape.Column labelCol) Contracts.Assert(labelCol.IsValid); Action error = - () => throw Host.ExceptSchemaMismatch(nameof(labelCol), RoleMappedSchema.ColumnRole.Label.Value, labelCol.Name, "BL, R8, R4 or a Key", labelCol.GetTypeString()); + () => throw Host.ExceptSchemaMismatch(nameof(labelCol), "label", labelCol.Name, "float, double, bool or KeyType", labelCol.GetTypeString()); if (labelCol.Kind != SchemaShape.Column.VectorKind.Scalar) error(); diff --git a/src/Microsoft.ML.StandardLearners/Standard/SdcaMultiClass.cs b/src/Microsoft.ML.StandardLearners/Standard/SdcaMultiClass.cs index 8d187923a2..ce0490b973 100644 --- a/src/Microsoft.ML.StandardLearners/Standard/SdcaMultiClass.cs +++ b/src/Microsoft.ML.StandardLearners/Standard/SdcaMultiClass.cs @@ -110,7 +110,7 @@ protected override void CheckLabelCompatible(SchemaShape.Column labelCol) Contracts.Assert(labelCol.IsValid); Action error = - () => throw Host.ExceptSchemaMismatch(nameof(labelCol), RoleMappedSchema.ColumnRole.Label.Value, labelCol.Name, "R8, R4 or a Key", labelCol.GetTypeString()); + () => throw Host.ExceptSchemaMismatch(nameof(labelCol), "label", labelCol.Name, "float, double or KeyType", labelCol.GetTypeString()); if (labelCol.Kind != SchemaShape.Column.VectorKind.Scalar) error(); diff --git a/src/Microsoft.ML.TensorFlow/TensorflowTransform.cs b/src/Microsoft.ML.TensorFlow/TensorflowTransform.cs index d804375d52..5c3949ebb8 100644 --- a/src/Microsoft.ML.TensorFlow/TensorflowTransform.cs +++ b/src/Microsoft.ML.TensorFlow/TensorflowTransform.cs @@ -818,7 +818,7 @@ public Mapper(TensorFlowTransformer parent, Schema inputSchema) : for (int i = 0; i < _parent.Inputs.Length; i++) { if (!inputSchema.TryGetColumnIndex(_parent.Inputs[i], out _inputColIndices[i])) - throw Host.Except($"Column {_parent.Inputs[i]} doesn't exist"); + throw Host.ExceptSchemaMismatch(nameof(InputSchema), "source", _parent.Inputs[i]); var type = inputSchema[_inputColIndices[i]].Type; if (type is VectorType vecType && vecType.Size == 0) @@ -1135,7 +1135,7 @@ public SchemaShape GetOutputSchema(SchemaShape inputSchema) if (!inputSchema.TryFindColumn(input, out var col)) throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", input); if (!(col.Kind == SchemaShape.Column.VectorKind.Vector)) - throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", input, nameof(VectorType), col.GetTypeString()); + throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", input, "vector", col.GetTypeString()); var expectedType = TensorFlowUtils.Tf2MlNetType(_tfInputTypes[i]); if (col.ItemType != expectedType) throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", input, expectedType.ToString(), col.ItemType.ToString()); diff --git a/src/Microsoft.ML.TimeSeries/AdaptiveSingularSpectrumSequenceModeler.cs b/src/Microsoft.ML.TimeSeries/AdaptiveSingularSpectrumSequenceModeler.cs index 60b906a5f0..4a3593ca14 100644 --- a/src/Microsoft.ML.TimeSeries/AdaptiveSingularSpectrumSequenceModeler.cs +++ b/src/Microsoft.ML.TimeSeries/AdaptiveSingularSpectrumSequenceModeler.cs @@ -1230,7 +1230,7 @@ internal override void Train(RoleMappedData data) _host.CheckParam(data.Schema.Feature.HasValue, nameof(data), "Must have features column."); var featureCol = data.Schema.Feature.Value; if (featureCol.Type != NumberType.Float) - throw _host.ExceptSchemaMismatch(nameof(data), "feature", featureCol.Name, "R4", featureCol.Type.ToString()); + throw _host.ExceptSchemaMismatch(nameof(data), "feature", featureCol.Name, "float", featureCol.Type.ToString()); Single[] dataArray = new Single[_trainSize]; diff --git a/src/Microsoft.ML.TimeSeries/IidAnomalyDetectionBase.cs b/src/Microsoft.ML.TimeSeries/IidAnomalyDetectionBase.cs index 7599b1dcaa..eeff5db3d2 100644 --- a/src/Microsoft.ML.TimeSeries/IidAnomalyDetectionBase.cs +++ b/src/Microsoft.ML.TimeSeries/IidAnomalyDetectionBase.cs @@ -43,7 +43,7 @@ public override Schema GetOutputSchema(Schema inputSchema) var colType = inputSchema[col].Type; if (colType != NumberType.R4) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", InputColumnName, NumberType.R4.ToString(), colType.ToString()); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", InputColumnName, "float", colType.ToString()); return Transform(new EmptyDataView(Host, inputSchema)).Schema; } diff --git a/src/Microsoft.ML.TimeSeries/IidChangePointDetector.cs b/src/Microsoft.ML.TimeSeries/IidChangePointDetector.cs index 1735d6a4dd..cb0874e2ce 100644 --- a/src/Microsoft.ML.TimeSeries/IidChangePointDetector.cs +++ b/src/Microsoft.ML.TimeSeries/IidChangePointDetector.cs @@ -247,7 +247,7 @@ public override SchemaShape GetOutputSchema(SchemaShape inputSchema) if (!inputSchema.TryFindColumn(Transformer.InputColumnName, out var col)) throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", Transformer.InputColumnName); if (col.ItemType != NumberType.R4) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", Transformer.InputColumnName, NumberType.R4.ToString(), col.GetTypeString()); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", Transformer.InputColumnName, "float", col.GetTypeString()); var metadata = new List() { new SchemaShape.Column(MetadataUtils.Kinds.SlotNames, SchemaShape.Column.VectorKind.Vector, TextType.Instance, false) diff --git a/src/Microsoft.ML.TimeSeries/IidSpikeDetector.cs b/src/Microsoft.ML.TimeSeries/IidSpikeDetector.cs index dabbf45d22..20cecfc285 100644 --- a/src/Microsoft.ML.TimeSeries/IidSpikeDetector.cs +++ b/src/Microsoft.ML.TimeSeries/IidSpikeDetector.cs @@ -222,7 +222,7 @@ public override SchemaShape GetOutputSchema(SchemaShape inputSchema) if (!inputSchema.TryFindColumn(Transformer.InputColumnName, out var col)) throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", Transformer.InputColumnName); if (col.ItemType != NumberType.R4) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", Transformer.InputColumnName, NumberType.R4.ToString(), col.GetTypeString()); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", Transformer.InputColumnName, "float", col.GetTypeString()); var metadata = new List() { new SchemaShape.Column(MetadataUtils.Kinds.SlotNames, SchemaShape.Column.VectorKind.Vector, TextType.Instance, false) diff --git a/src/Microsoft.ML.TimeSeries/SequentialAnomalyDetectionTransformBase.cs b/src/Microsoft.ML.TimeSeries/SequentialAnomalyDetectionTransformBase.cs index 7e82cc889f..91a554b96d 100644 --- a/src/Microsoft.ML.TimeSeries/SequentialAnomalyDetectionTransformBase.cs +++ b/src/Microsoft.ML.TimeSeries/SequentialAnomalyDetectionTransformBase.cs @@ -594,7 +594,7 @@ public Mapper(IHostEnvironment env, SequentialAnomalyDetectionTransformBase() { new SchemaShape.Column(MetadataUtils.Kinds.SlotNames, SchemaShape.Column.VectorKind.Vector, TextType.Instance, false) diff --git a/src/Microsoft.ML.TimeSeries/SsaSpikeDetector.cs b/src/Microsoft.ML.TimeSeries/SsaSpikeDetector.cs index 976a223a5a..02b81458b3 100644 --- a/src/Microsoft.ML.TimeSeries/SsaSpikeDetector.cs +++ b/src/Microsoft.ML.TimeSeries/SsaSpikeDetector.cs @@ -262,7 +262,7 @@ public SchemaShape GetOutputSchema(SchemaShape inputSchema) if (!inputSchema.TryFindColumn(_args.Source, out var col)) throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", _args.Source); if (col.ItemType != NumberType.R4) - throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", _args.Source, NumberType.R4.ToString(), col.GetTypeString()); + throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", _args.Source, "float", col.GetTypeString()); var metadata = new List() { new SchemaShape.Column(MetadataUtils.Kinds.SlotNames, SchemaShape.Column.VectorKind.Vector, TextType.Instance, false) diff --git a/src/Microsoft.ML.Transforms/MissingValueDroppingTransformer.cs b/src/Microsoft.ML.Transforms/MissingValueDroppingTransformer.cs index ddc4b69a22..1ce692a343 100644 --- a/src/Microsoft.ML.Transforms/MissingValueDroppingTransformer.cs +++ b/src/Microsoft.ML.Transforms/MissingValueDroppingTransformer.cs @@ -105,7 +105,7 @@ protected override void CheckInputColumn(Schema inputSchema, int col, int srcCol { var inType = inputSchema[srcCol].Type; if (!(inType is VectorType)) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", inputSchema[srcCol].Name, "Vector", inType.ToString()); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", inputSchema[srcCol].Name, "vector", inType.ToString()); } // Factory method for SignatureLoadModel @@ -377,7 +377,7 @@ public override SchemaShape GetOutputSchema(SchemaShape inputSchema) if (!inputSchema.TryFindColumn(colPair.inputColumnName, out var col) || !Data.Conversion.Conversions.Instance.TryGetIsNAPredicate(col.ItemType, out Delegate del)) throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", colPair.inputColumnName); if (!(col.Kind == SchemaShape.Column.VectorKind.Vector || col.Kind == SchemaShape.Column.VectorKind.VariableVector)) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", colPair.inputColumnName, "Vector", col.GetTypeString()); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", colPair.inputColumnName, "known-size vector", col.GetTypeString()); var metadata = new List(); if (col.Metadata.TryFindColumn(MetadataUtils.Kinds.KeyValues, out var keyMeta)) metadata.Add(keyMeta); diff --git a/src/Microsoft.ML.Transforms/Text/LdaTransform.cs b/src/Microsoft.ML.Transforms/Text/LdaTransform.cs index bd312b55fe..8d70d1b252 100644 --- a/src/Microsoft.ML.Transforms/Text/LdaTransform.cs +++ b/src/Microsoft.ML.Transforms/Text/LdaTransform.cs @@ -717,7 +717,7 @@ public Mapper(LatentDirichletAllocationTransformer parent, Schema inputSchema) var srcCol = inputSchema[_srcCols[i]]; var srcType = srcCol.Type as VectorType; if (srcType == null || !srcType.IsKnownSize || !(srcType.ItemType is NumberType)) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", _parent.ColumnPairs[i].inputColumnName, "a fixed vector of floats", srcCol.Type.ToString()); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "input", _parent.ColumnPairs[i].inputColumnName, "known-size vector of float", srcCol.Type.ToString()); } } @@ -950,7 +950,7 @@ private static List>> Train(IHostEnvironment env, I var srcColType = inputSchema[srcCol].Type as VectorType; if (srcColType == null || !srcColType.IsKnownSize || !(srcColType.ItemType is NumberType)) - throw env.ExceptSchemaMismatch(nameof(inputSchema), "input", columns[i].InputColumnName, "a fixed vector of floats", srcColType.ToString()); + throw env.ExceptSchemaMismatch(nameof(inputSchema), "input", columns[i].InputColumnName, "known-size vector of float", srcColType.ToString()); srcCols[i] = srcCol; activeColumns.Add(inputData.Schema[srcCol]); @@ -1161,7 +1161,7 @@ public SchemaShape GetOutputSchema(SchemaShape inputSchema) if (!inputSchema.TryFindColumn(colInfo.InputColumnName, out var col)) throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", colInfo.InputColumnName); if (col.ItemType.RawType != typeof(float) || col.Kind == SchemaShape.Column.VectorKind.Scalar) - throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", colInfo.InputColumnName, "a vector of floats", col.GetTypeString()); + throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", colInfo.InputColumnName, "vector of float", col.GetTypeString()); result[colInfo.Name] = new SchemaShape.Column(colInfo.Name, SchemaShape.Column.VectorKind.Vector, NumberType.R4, false); } diff --git a/src/Microsoft.ML.Transforms/Text/StopWordsRemovingTransformer.cs b/src/Microsoft.ML.Transforms/Text/StopWordsRemovingTransformer.cs index e8690253a2..6c12f5a865 100644 --- a/src/Microsoft.ML.Transforms/Text/StopWordsRemovingTransformer.cs +++ b/src/Microsoft.ML.Transforms/Text/StopWordsRemovingTransformer.cs @@ -406,7 +406,7 @@ public Mapper(StopWordsRemovingTransformer parent, Schema inputSchema) if (!string.IsNullOrEmpty(_parent._columns[i].LanguageColumn)) { if (!inputSchema.TryGetColumnIndex(_parent._columns[i].LanguageColumn, out int langCol)) - throw Host.ExceptSchemaMismatch(nameof(inputSchema), "language column", _parent._columns[i].LanguageColumn); + throw Host.ExceptSchemaMismatch(nameof(inputSchema), "language", _parent._columns[i].LanguageColumn); _languageColumns[i] = langCol; } else