diff --git a/src/Microsoft.ML.Core/Data/AnnotationUtils.cs b/src/Microsoft.ML.Core/Data/AnnotationUtils.cs index 0bd0ad4dda..0fff1250bb 100644 --- a/src/Microsoft.ML.Core/Data/AnnotationUtils.cs +++ b/src/Microsoft.ML.Core/Data/AnnotationUtils.cs @@ -308,7 +308,7 @@ public static void GetSlotNames(RoleMappedSchema schema, RoleMappedSchema.Column schema.Schema[list[0].Index].Annotations.GetValue(Kinds.SlotNames, ref slotNames); } - public static bool HasKeyValues(this SchemaShape.Column col) + public static bool NeedsSlotNames(this SchemaShape.Column col) { return col.Annotations.TryFindColumn(Kinds.KeyValues, out var metaCol) && metaCol.Kind == SchemaShape.Column.VectorKind.Vector @@ -442,7 +442,7 @@ public static bool TryGetCategoricalFeatureIndices(DataViewSchema schema, int co public static IEnumerable AnnotationsForMulticlassScoreColumn(SchemaShape.Column? labelColumn = null) { var cols = new List(); - if (labelColumn != null && labelColumn.Value.IsKey && HasKeyValues(labelColumn.Value)) + if (labelColumn != null && labelColumn.Value.IsKey && NeedsSlotNames(labelColumn.Value)) cols.Add(new SchemaShape.Column(Kinds.SlotNames, SchemaShape.Column.VectorKind.Vector, TextDataViewType.Instance, false)); cols.AddRange(GetTrainerOutputAnnotation()); return cols; diff --git a/src/Microsoft.ML.Data/Data/Conversion.cs b/src/Microsoft.ML.Data/Data/Conversion.cs index d047472ad1..683a4de230 100644 --- a/src/Microsoft.ML.Data/Data/Conversion.cs +++ b/src/Microsoft.ML.Data/Data/Conversion.cs @@ -111,6 +111,7 @@ private Conversions() AddStd(Convert); AddStd(Convert); AddAux(Convert); + AddStd(Convert); AddStd(Convert); AddStd(Convert); @@ -119,6 +120,7 @@ private Conversions() AddStd(Convert); AddStd(Convert); AddAux(Convert); + AddStd(Convert); AddStd(Convert); AddStd(Convert); @@ -127,6 +129,7 @@ private Conversions() AddStd(Convert); AddStd(Convert); AddAux(Convert); + AddStd(Convert); AddStd(Convert); AddStd(Convert); @@ -135,6 +138,7 @@ private Conversions() AddStd(Convert); AddStd(Convert); AddAux(Convert); + AddStd(Convert); AddStd(Convert); AddStd(Convert); @@ -144,6 +148,7 @@ private Conversions() AddStd(Convert); AddStd(Convert); AddAux(Convert); + AddStd(Convert); AddStd(Convert); AddStd(Convert); @@ -153,6 +158,7 @@ private Conversions() AddStd(Convert); AddStd(Convert); AddAux(Convert); + AddStd(Convert); AddStd(Convert); AddStd(Convert); @@ -162,6 +168,7 @@ private Conversions() AddStd(Convert); AddStd(Convert); AddAux(Convert); + AddStd(Convert); AddStd(Convert); AddStd(Convert); @@ -171,6 +178,7 @@ private Conversions() AddStd(Convert); AddStd(Convert); AddAux(Convert); + AddStd(Convert); AddStd(Convert); AddStd(Convert); @@ -180,11 +188,13 @@ private Conversions() AddAux(Convert); AddStd(Convert); + AddStd(Convert); AddStd(Convert); AddAux(Convert); AddStd(Convert); AddStd(Convert); + AddStd(Convert); AddAux(Convert); AddStd(Convert); @@ -901,6 +911,19 @@ public void Convert(in BL src, ref SB dst) public void Convert(in DZ src, ref SB dst) { ClearDst(ref dst); dst.AppendFormat("{0:o}", src); } #endregion ToStringBuilder + #region ToBL + public void Convert(in R8 src, ref BL dst) => dst = System.Convert.ToBoolean(src); + public void Convert(in R4 src, ref BL dst) => dst = System.Convert.ToBoolean(src); + public void Convert(in I1 src, ref BL dst) => dst = System.Convert.ToBoolean(src); + public void Convert(in I2 src, ref BL dst) => dst = System.Convert.ToBoolean(src); + public void Convert(in I4 src, ref BL dst) => dst = System.Convert.ToBoolean(src); + public void Convert(in I8 src, ref BL dst) => dst = System.Convert.ToBoolean(src); + public void Convert(in U1 src, ref BL dst) => dst = System.Convert.ToBoolean(src); + public void Convert(in U2 src, ref BL dst) => dst = System.Convert.ToBoolean(src); + public void Convert(in U4 src, ref BL dst) => dst = System.Convert.ToBoolean(src); + public void Convert(in U8 src, ref BL dst) => dst = System.Convert.ToBoolean(src); + #endregion + #region FromR4 public void Convert(in R4 src, ref R4 dst) => dst = src; public void Convert(in R4 src, ref R8 dst) => dst = src; @@ -1139,7 +1162,7 @@ private bool TryParseCore(ReadOnlySpan span, out ulong dst) dst = res; return true; - LFail: + LFail: dst = 0; return false; } @@ -1246,7 +1269,7 @@ private bool TryParseNonNegative(ReadOnlySpan span, out long result) result = res; return true; - LFail: + LFail: result = 0; return false; } diff --git a/src/Microsoft.ML.Data/Scorers/MultiClassClassifierScorer.cs b/src/Microsoft.ML.Data/Scorers/MultiClassClassifierScorer.cs index c0a6061529..fb2fd7dd34 100644 --- a/src/Microsoft.ML.Data/Scorers/MultiClassClassifierScorer.cs +++ b/src/Microsoft.ML.Data/Scorers/MultiClassClassifierScorer.cs @@ -429,7 +429,7 @@ internal static bool CanWrap(ISchemaBoundMapper mapper, DataViewType labelNameTy var scoreType = outSchema[scoreIdx].Type; // Check that the type is vector, and is of compatible size with the score output. - return labelNameType is VectorType vectorType && vectorType.Size == scoreType.GetVectorSize(); + return labelNameType is VectorType vectorType && vectorType.Size == scoreType.GetVectorSize() && vectorType.ItemType == TextDataViewType.Instance; } internal static ISchemaBoundMapper WrapCore(IHostEnvironment env, ISchemaBoundMapper mapper, RoleMappedSchema trainSchema) diff --git a/src/Microsoft.ML.Data/Training/TrainerEstimatorBase.cs b/src/Microsoft.ML.Data/Training/TrainerEstimatorBase.cs index 288000e0c9..ba7bc432c6 100644 --- a/src/Microsoft.ML.Data/Training/TrainerEstimatorBase.cs +++ b/src/Microsoft.ML.Data/Training/TrainerEstimatorBase.cs @@ -145,8 +145,15 @@ private protected virtual void CheckLabelCompatible(SchemaShape.Column labelCol) private protected TTransformer TrainTransformer(IDataView trainSet, IDataView validationSet = null, IPredictor initPredictor = null) { + CheckInputSchema(SchemaShape.Create(trainSet.Schema)); var trainRoleMapped = MakeRoles(trainSet); - var validRoleMapped = validationSet == null ? null : MakeRoles(validationSet); + RoleMappedData validRoleMapped = null; + + if (validationSet != null) + { + CheckInputSchema(SchemaShape.Create(validationSet.Schema)); + validRoleMapped = MakeRoles(validationSet); + } var pred = TrainModelCore(new TrainContext(trainRoleMapped, validRoleMapped, null, initPredictor)); return MakeTransformer(pred, trainSet.Schema); diff --git a/src/Microsoft.ML.FastTree/FastTreeRanking.cs b/src/Microsoft.ML.FastTree/FastTreeRanking.cs index d1a9aa426a..a607054181 100644 --- a/src/Microsoft.ML.FastTree/FastTreeRanking.cs +++ b/src/Microsoft.ML.FastTree/FastTreeRanking.cs @@ -105,6 +105,7 @@ private protected override void CheckLabelCompatible(SchemaShape.Column labelCol if (!labelCol.IsKey && labelCol.ItemType != NumberDataViewType.Single) error(); } + private protected override float GetMaxLabel() { return GetLabelGains().Length - 1; diff --git a/src/Microsoft.ML.StandardTrainers/Standard/MultiClass/MetaMulticlassTrainer.cs b/src/Microsoft.ML.StandardTrainers/Standard/MultiClass/MetaMulticlassTrainer.cs index c7e3647ee1..5a36405abf 100644 --- a/src/Microsoft.ML.StandardTrainers/Standard/MultiClass/MetaMulticlassTrainer.cs +++ b/src/Microsoft.ML.StandardTrainers/Standard/MultiClass/MetaMulticlassTrainer.cs @@ -8,10 +8,9 @@ using Microsoft.ML.Calibrators; using Microsoft.ML.CommandLine; using Microsoft.ML.Data; -using Microsoft.ML.Data.Conversion; using Microsoft.ML.Internal.Internallearn; using Microsoft.ML.Runtime; - +using Microsoft.ML.Transforms; namespace Microsoft.ML.Trainers { using TScalarTrainer = ITrainerEstimator>, IPredictorProducing>; @@ -32,7 +31,7 @@ internal abstract class OptionsBase [Argument(ArgumentType.LastOccurenceWins, HelpText = "Number of instances to train the calibrator", SortOrder = 150, ShortName = "numcali")] internal int MaxCalibrationExamples = 1000000000; - [Argument(ArgumentType.Multiple, HelpText = "Whether to treat missing labels as having negative labels, instead of keeping them missing", SortOrder = 150, ShortName = "missNeg")] + [Argument(ArgumentType.Multiple, HelpText = "Whether to treat missing labels as having negative labels, or exclude their rows from dataview.", SortOrder = 150, ShortName = "missNeg")] public bool ImputeMissingLabelsAsNegative; } @@ -98,20 +97,15 @@ private protected IDataView MapLabelsCore(DataViewType type, InPredicate e Host.AssertValue(data); Host.Assert(data.Schema.Label.HasValue); - var lab = data.Schema.Label.Value; + var label = data.Schema.Label.Value; + IDataView dataView = data.Data; + if (!Args.ImputeMissingLabelsAsNegative) + dataView = new NAFilter(Host, data.Data, false, label.Name); - InPredicate isMissing; - if (!Args.ImputeMissingLabelsAsNegative && Conversions.Instance.TryGetIsNAPredicate(type, out isMissing)) - { - return LambdaColumnMapper.Create(Host, "Label mapper", data.Data, - lab.Name, lab.Name, type, NumberDataViewType.Single, - (in T src, ref float dst) => - dst = equalsTarget(in src) ? 1 : (isMissing(in src) ? float.NaN : default(float))); - } return LambdaColumnMapper.Create(Host, "Label mapper", data.Data, - lab.Name, lab.Name, type, NumberDataViewType.Single, - (in T src, ref float dst) => - dst = equalsTarget(in src) ? 1 : default(float)); + label.Name, label.Name, type, BooleanDataViewType.Instance, + (in T src, ref bool dst) => + dst = equalsTarget(in src) ? true : false); } private protected abstract TModel TrainCore(IChannel ch, RoleMappedData data, int count); diff --git a/src/Microsoft.ML.StandardTrainers/Standard/MultiClass/OneVersusAllTrainer.cs b/src/Microsoft.ML.StandardTrainers/Standard/MultiClass/OneVersusAllTrainer.cs index 82a3b259ff..9fb2dd563a 100644 --- a/src/Microsoft.ML.StandardTrainers/Standard/MultiClass/OneVersusAllTrainer.cs +++ b/src/Microsoft.ML.StandardTrainers/Standard/MultiClass/OneVersusAllTrainer.cs @@ -145,28 +145,18 @@ private ISingleFeaturePredictionTransformer TrainOne(IChannel private IDataView MapLabels(RoleMappedData data, int cls) { - var lab = data.Schema.Label.Value; - Host.Assert(!lab.IsHidden); - Host.Assert(lab.Type.GetKeyCount() > 0 || lab.Type == NumberDataViewType.Single || lab.Type == NumberDataViewType.Double); + var label = data.Schema.Label.Value; + Host.Assert(!label.IsHidden); + Host.Assert(label.Type.GetKeyCount() > 0 || label.Type == NumberDataViewType.Single || label.Type == NumberDataViewType.Double); - if (lab.Type.GetKeyCount() > 0) + if (label.Type.GetKeyCount() > 0) { // Key values are 1-based. uint key = (uint)(cls + 1); return MapLabelsCore(NumberDataViewType.UInt32, (in uint val) => key == val, data); } - if (lab.Type == NumberDataViewType.Single) - { - float key = cls; - return MapLabelsCore(NumberDataViewType.Single, (in float val) => key == val, data); - } - if (lab.Type == NumberDataViewType.Double) - { - double key = cls; - return MapLabelsCore(NumberDataViewType.Double, (in double val) => key == val, data); - } - throw Host.ExceptNotSupp($"Label column type is not supported by OneVersusAllTrainer: {lab.Type.RawType}"); + throw Host.ExceptNotSupp($"Label column type is not supported by OneVersusAllTrainer: {label.Type.RawType}"); } /// Trains a model. diff --git a/src/Microsoft.ML.StandardTrainers/Standard/MultiClass/PairwiseCouplingTrainer.cs b/src/Microsoft.ML.StandardTrainers/Standard/MultiClass/PairwiseCouplingTrainer.cs index 53cc39327c..6fb27a5edd 100644 --- a/src/Microsoft.ML.StandardTrainers/Standard/MultiClass/PairwiseCouplingTrainer.cs +++ b/src/Microsoft.ML.StandardTrainers/Standard/MultiClass/PairwiseCouplingTrainer.cs @@ -142,31 +142,19 @@ private ISingleFeaturePredictionTransformer TrainOne(IChannel ch private IDataView MapLabels(RoleMappedData data, int cls1, int cls2) { - var lab = data.Schema.Label.Value; - Host.Assert(!lab.IsHidden); - Host.Assert(lab.Type.GetKeyCount() > 0 || lab.Type == NumberDataViewType.Single || lab.Type == NumberDataViewType.Double); + var label = data.Schema.Label.Value; + Host.Assert(!label.IsHidden); + Host.Assert(label.Type.GetKeyCount() > 0 || label.Type == NumberDataViewType.Single || label.Type == NumberDataViewType.Double); - if (lab.Type.GetKeyCount() > 0) + if (label.Type.GetKeyCount() > 0) { // Key values are 1-based. uint key1 = (uint)(cls1 + 1); uint key2 = (uint)(cls2 + 1); return MapLabelsCore(NumberDataViewType.UInt32, (in uint val) => val == key1 || val == key2, data); } - if (lab.Type == NumberDataViewType.Single) - { - float key1 = cls1; - float key2 = cls2; - return MapLabelsCore(NumberDataViewType.Single, (in float val) => val == key1 || val == key2, data); - } - if (lab.Type == NumberDataViewType.Double) - { - double key1 = cls1; - double key2 = cls2; - return MapLabelsCore(NumberDataViewType.Double, (in double val) => val == key1 || val == key2, data); - } - throw Host.ExceptNotSupp($"Label column type is not supported by nameof(PairwiseCouplingTrainer): {lab.Type.RawType}"); + throw Host.ExceptNotSupp($"Label column type is not supported by nameof(PairwiseCouplingTrainer): {label.Type.RawType}"); } /// diff --git a/src/Microsoft.ML.StandardTrainers/Standard/Online/AveragedPerceptron.cs b/src/Microsoft.ML.StandardTrainers/Standard/Online/AveragedPerceptron.cs index f4cb2096d8..edd6d1f49e 100644 --- a/src/Microsoft.ML.StandardTrainers/Standard/Online/AveragedPerceptron.cs +++ b/src/Microsoft.ML.StandardTrainers/Standard/Online/AveragedPerceptron.cs @@ -174,20 +174,6 @@ private protected override void CheckLabels(RoleMappedData data) data.CheckBinaryLabel(); } - private protected override void CheckLabelCompatible(SchemaShape.Column labelCol) - { - Contracts.Assert(labelCol.IsValid); - - Action error = - () => throw Host.ExceptSchemaMismatch(nameof(labelCol), "label", labelCol.Name, "float, double, bool or KeyType", labelCol.GetTypeString()); - - if (labelCol.Kind != SchemaShape.Column.VectorKind.Scalar) - error(); - - if (!labelCol.IsKey && labelCol.ItemType != NumberDataViewType.Single && labelCol.ItemType != NumberDataViewType.Double && !(labelCol.ItemType is BooleanDataViewType)) - error(); - } - private protected override TrainStateBase MakeState(IChannel ch, int numFeatures, LinearModelParameters predictor) { return new TrainState(ch, numFeatures, predictor, this); diff --git a/src/Microsoft.ML.StandardTrainers/Standard/SdcaBinary.cs b/src/Microsoft.ML.StandardTrainers/Standard/SdcaBinary.cs index 68814b33fb..991c79d246 100644 --- a/src/Microsoft.ML.StandardTrainers/Standard/SdcaBinary.cs +++ b/src/Microsoft.ML.StandardTrainers/Standard/SdcaBinary.cs @@ -1512,20 +1512,6 @@ private protected SdcaBinaryTrainerBase(IHostEnvironment env, BinaryOptionsBase private protected abstract SchemaShape.Column[] ComputeSdcaBinaryClassifierSchemaShape(); - private protected override void CheckLabelCompatible(SchemaShape.Column labelCol) - { - Contracts.Assert(labelCol.IsValid); - - Action error = - () => throw Host.ExceptSchemaMismatch(nameof(labelCol), "label", labelCol.Name, "float, double, bool or KeyType", labelCol.GetTypeString()); - - if (labelCol.Kind != SchemaShape.Column.VectorKind.Scalar) - error(); - - if (!labelCol.IsKey && labelCol.ItemType != NumberDataViewType.Single && labelCol.ItemType != NumberDataViewType.Double && !(labelCol.ItemType is BooleanDataViewType)) - error(); - } - private protected LinearBinaryModelParameters CreateLinearBinaryModelParameters(VBuffer[] weights, float[] bias) { Host.CheckParam(Utils.Size(weights) == 1, nameof(weights)); diff --git a/src/Microsoft.ML.StandardTrainers/Standard/SdcaMultiClass.cs b/src/Microsoft.ML.StandardTrainers/Standard/SdcaMultiClass.cs index ef50751184..9fa3992b52 100644 --- a/src/Microsoft.ML.StandardTrainers/Standard/SdcaMultiClass.cs +++ b/src/Microsoft.ML.StandardTrainers/Standard/SdcaMultiClass.cs @@ -121,19 +121,6 @@ private protected override SchemaShape.Column[] GetOutputColumnsCore(SchemaShape }; } - private protected override void CheckLabelCompatible(SchemaShape.Column labelCol) - { - Contracts.Assert(labelCol.IsValid); - - Action error = - () => throw Host.ExceptSchemaMismatch(nameof(labelCol), "label", labelCol.Name, "float, double or KeyType", labelCol.GetTypeString()); - - if (labelCol.Kind != SchemaShape.Column.VectorKind.Scalar) - error(); - if (!labelCol.IsKey && labelCol.ItemType != NumberDataViewType.Single && labelCol.ItemType != NumberDataViewType.Double) - error(); - } - /// private protected override void TrainWithoutLock(IProgressChannelProvider progress, FloatLabelCursor.Factory cursorFactory, Random rand, IdToIdxLookup idToIdx, int numThreads, DualsTableBase duals, float[] biasReg, float[] invariants, float lambdaNInv, diff --git a/src/Microsoft.ML.Transforms/Text/NgramTransform.cs b/src/Microsoft.ML.Transforms/Text/NgramTransform.cs index feab3ac49b..ce5fb31119 100644 --- a/src/Microsoft.ML.Transforms/Text/NgramTransform.cs +++ b/src/Microsoft.ML.Transforms/Text/NgramTransform.cs @@ -887,7 +887,7 @@ public SchemaShape GetOutputSchema(SchemaShape inputSchema) if (!IsSchemaColumnValid(col)) throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", colInfo.InputColumnName, ExpectedColumnType, col.GetTypeString()); var metadata = new List(); - if (col.HasKeyValues()) + if (col.NeedsSlotNames()) metadata.Add(new SchemaShape.Column(AnnotationUtils.Kinds.SlotNames, SchemaShape.Column.VectorKind.Vector, TextDataViewType.Instance, false)); result[colInfo.Name] = new SchemaShape.Column(colInfo.Name, SchemaShape.Column.VectorKind.Vector, NumberDataViewType.Single, false, new SchemaShape(metadata)); } diff --git a/test/BaselineOutput/Common/FeatureContribution/AveragePerceptronBinary.tsv b/test/BaselineOutput/Common/FeatureContribution/AveragePerceptronBinary.tsv index e2bb234204..d8b7fecf47 100644 --- a/test/BaselineOutput/Common/FeatureContribution/AveragePerceptronBinary.tsv +++ b/test/BaselineOutput/Common/FeatureContribution/AveragePerceptronBinary.tsv @@ -4,14 +4,15 @@ #@ col=X2VBuffer:R4:1-4 #@ col=X3Important:R4:5 #@ col=Label:R4:6 -#@ col=Features:R4:7-12 -#@ col=Features:R4:13-18 -#@ col=FeatureContributions:R4:19-24 -#@ col=FeatureContributions:R4:25-30 -#@ col=FeatureContributions:R4:31-36 -#@ col=FeatureContributions:R4:37-42 +#@ col=Label:BL:7 +#@ col=Features:R4:8-13 +#@ col=Features:R4:14-19 +#@ col=FeatureContributions:R4:20-25 +#@ col=FeatureContributions:R4:26-31 +#@ col=FeatureContributions:R4:32-37 +#@ col=FeatureContributions:R4:38-43 #@ } -950 757 692 720 297 7515 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0 0.1527809 0 0 0 1 -0.6583012 0 -1 -0.517060339 0 0 12 2:-0.13028869 5:1 8:-0.370813 11:2.84608746 -459 961 0 659 274 2147 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 0 0.6788808 0 0 0 0.99999994 -0.6720779 0 0 -1 -0.870772958 0 12 3:-0.215823054 5:0.99999994 9:-0.175488681 11:0.8131137 -672 275 0 65 195 9818 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0 0.04248268 0 0 0 1 -1 0 0 -0.100242466 -0.6298147 0 12 0:-0.04643902 5:1 6:-0.172673345 11:3.71828127 -186 301 0 681 526 1456 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 0 0.313550383 0 0 0 1 -0.162922 0 0 -0.6181894 -1 0 12 4:-0.5319963 5:1 10:-0.293352127 11:0.5514176 +950 757 692 720 297 7515 1 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0 0.1527809 0 0 0 1 -0.6583012 0 -1 -0.517060339 0 0 12 2:-0.13028869 5:1 8:-0.370813 11:2.84608746 +459 961 0 659 274 2147 0 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 0 0.6788808 0 0 0 0.99999994 -0.6720779 0 0 -1 -0.870772958 0 12 3:-0.215823054 5:0.99999994 9:-0.175488681 11:0.8131137 +672 275 0 65 195 9818 1 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0 0.04248268 0 0 0 1 -1 0 0 -0.100242466 -0.6298147 0 12 0:-0.04643902 5:1 6:-0.172673345 11:3.71828127 +186 301 0 681 526 1456 0 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 0 0.313550383 0 0 0 1 -0.162922 0 0 -0.6181894 -1 0 12 4:-0.5319963 5:1 10:-0.293352127 11:0.5514176 diff --git a/test/BaselineOutput/Common/FeatureContribution/FastForestBinary.tsv b/test/BaselineOutput/Common/FeatureContribution/FastForestBinary.tsv index b9b9f02952..592976645d 100644 --- a/test/BaselineOutput/Common/FeatureContribution/FastForestBinary.tsv +++ b/test/BaselineOutput/Common/FeatureContribution/FastForestBinary.tsv @@ -4,14 +4,15 @@ #@ col=X2VBuffer:R4:1-4 #@ col=X3Important:R4:5 #@ col=Label:R4:6 -#@ col=Features:R4:7-12 -#@ col=Features:R4:13-18 -#@ col=FeatureContributions:R4:19-24 -#@ col=FeatureContributions:R4:25-30 -#@ col=FeatureContributions:R4:31-36 -#@ col=FeatureContributions:R4:37-42 +#@ col=Label:BL:7 +#@ col=Features:R4:8-13 +#@ col=Features:R4:14-19 +#@ col=FeatureContributions:R4:20-25 +#@ col=FeatureContributions:R4:26-31 +#@ col=FeatureContributions:R4:32-37 +#@ col=FeatureContributions:R4:38-43 #@ } -950 757 692 720 297 7515 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0.114403568 0.116716474 0 0 0 1 18 11:1 17:125.600014 -459 961 0 659 274 2147 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 24 1:1 6:-0.08706974 10:-0.05425923 11:-1 13:0.124141291 17:-1 19:15.8156023 23:-127.400009 -672 275 0 65 195 9818 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0.03049014 0 0 0 0 1 0 -0.508646131 0 -1 -0.6718745 0 12 3:-0.0728949159 5:1 9:-9.534656 11:130.8 -186 301 0 681 526 1456 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 24 6:-0.0991563 7:-0.155881777 11:-1 17:-1 23:-129.600021 +950 757 692 720 297 7515 1 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0.114403568 0.116716474 0 0 0 1 18 11:1 17:125.600014 +459 961 0 659 274 2147 0 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 24 1:1 6:-0.08706974 10:-0.05425923 11:-1 13:0.124141291 17:-1 19:15.8156023 23:-127.400009 +672 275 0 65 195 9818 1 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0.03049014 0 0 0 0 1 0 -0.508646131 0 -1 -0.6718745 0 12 3:-0.0728949159 5:1 9:-9.534656 11:130.8 +186 301 0 681 526 1456 0 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 24 6:-0.0991563 7:-0.155881777 11:-1 17:-1 23:-129.600021 diff --git a/test/BaselineOutput/Common/FeatureContribution/FastTreeBinary.tsv b/test/BaselineOutput/Common/FeatureContribution/FastTreeBinary.tsv index 6548b45a3a..c5c1c379da 100644 --- a/test/BaselineOutput/Common/FeatureContribution/FastTreeBinary.tsv +++ b/test/BaselineOutput/Common/FeatureContribution/FastTreeBinary.tsv @@ -4,14 +4,15 @@ #@ col=X2VBuffer:R4:1-4 #@ col=X3Important:R4:5 #@ col=Label:R4:6 -#@ col=Features:R4:7-12 -#@ col=Features:R4:13-18 -#@ col=FeatureContributions:R4:19-24 -#@ col=FeatureContributions:R4:25-30 -#@ col=FeatureContributions:R4:31-36 -#@ col=FeatureContributions:R4:37-42 +#@ col=Label:BL:7 +#@ col=Features:R4:8-13 +#@ col=Features:R4:14-19 +#@ col=FeatureContributions:R4:20-25 +#@ col=FeatureContributions:R4:26-31 +#@ col=FeatureContributions:R4:32-37 +#@ col=FeatureContributions:R4:38-43 #@ } -950 757 692 720 297 7515 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0.0747833252 0 0 0.104883604 0 1 18 11:1 17:88.7116241 -459 961 0 659 274 2147 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 24 1:1 6:-0.03860954 9:-0.09425706 11:-1 13:0.0170135442 17:-1 19:1.83154547 23:-107.652206 -672 275 0 65 195 9818 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0 0.0271605756 0 0.004087467 0 1 18 0:-1 4:-0.000762519543 6:-0.08071669 11:1 12:-6.67459 17:82.6915741 -186 301 0 681 526 1456 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 24 6:-0.05714011 9:-0.107010774 11:-1 17:-1 23:-94.6040039 +950 757 692 720 297 7515 1 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0.0747833252 0 0 0.104883604 0 1 18 11:1 17:88.7116241 +459 961 0 659 274 2147 0 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 24 1:1 6:-0.03860954 9:-0.09425706 11:-1 13:0.0170135442 17:-1 19:1.83154547 23:-107.652206 +672 275 0 65 195 9818 1 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0 0.0271605756 0 0.004087467 0 1 18 0:-1 4:-0.000762519543 6:-0.08071669 11:1 12:-6.67459 17:82.6915741 +186 301 0 681 526 1456 0 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 24 6:-0.05714011 9:-0.107010774 11:-1 17:-1 23:-94.6040039 diff --git a/test/BaselineOutput/Common/FeatureContribution/GAMBinary.tsv b/test/BaselineOutput/Common/FeatureContribution/GAMBinary.tsv index 9bd321e1e6..d708288b64 100644 --- a/test/BaselineOutput/Common/FeatureContribution/GAMBinary.tsv +++ b/test/BaselineOutput/Common/FeatureContribution/GAMBinary.tsv @@ -4,14 +4,15 @@ #@ col=X2VBuffer:R4:1-4 #@ col=X3Important:R4:5 #@ col=Label:R4:6 -#@ col=Features:R4:7-12 -#@ col=Features:R4:13-18 -#@ col=FeatureContributions:R4:19-24 -#@ col=FeatureContributions:R4:25-30 -#@ col=FeatureContributions:R4:31-36 -#@ col=FeatureContributions:R4:37-42 +#@ col=Label:BL:7 +#@ col=Features:R4:8-13 +#@ col=Features:R4:14-19 +#@ col=FeatureContributions:R4:20-25 +#@ col=FeatureContributions:R4:26-31 +#@ col=FeatureContributions:R4:32-37 +#@ col=FeatureContributions:R4:38-43 #@ } -950 757 692 720 297 7515 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0.184881687 0.1251098 0 0 0 0.99999994 18 3:-0.349488348 4:-1 10:-0.0656147 11:0.99999994 16:-0.176268816 17:2.68642259 -459 961 0 659 274 2147 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 24 1:1 6:-0.0243161488 10:-0.06347539 11:-1 13:0.121030726 17:-1 19:0.336097836 23:-2.776963 -672 275 0 65 195 9818 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0.016381301 0.0037277115 0 0 0 0.99999994 18 3:-0.627149343 4:-1 10:-0.0343219377 11:0.99999994 16:-0.09220323 17:2.68642259 -186 301 0 681 526 1456 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 24 1:0.99999994 6:-0.07755054 9:-0.02218391 11:-1 13:0.003606173 17:-1 19:0.010014209 23:-2.776963 +950 757 692 720 297 7515 1 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0.184881687 0.1251098 0 0 0 0.99999994 18 3:-0.349488348 4:-1 10:-0.0656147 11:0.99999994 16:-0.176268816 17:2.68642259 +459 961 0 659 274 2147 0 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 24 1:1 6:-0.0243161488 10:-0.06347539 11:-1 13:0.121030726 17:-1 19:0.336097836 23:-2.776963 +672 275 0 65 195 9818 1 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0.016381301 0.0037277115 0 0 0 0.99999994 18 3:-0.627149343 4:-1 10:-0.0343219377 11:0.99999994 16:-0.09220323 17:2.68642259 +186 301 0 681 526 1456 0 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 24 1:0.99999994 6:-0.07755054 9:-0.02218391 11:-1 13:0.003606173 17:-1 19:0.010014209 23:-2.776963 diff --git a/test/BaselineOutput/Common/FeatureContribution/LightGbmBinary.tsv b/test/BaselineOutput/Common/FeatureContribution/LightGbmBinary.tsv index b4247eb756..9ae6814bb4 100644 --- a/test/BaselineOutput/Common/FeatureContribution/LightGbmBinary.tsv +++ b/test/BaselineOutput/Common/FeatureContribution/LightGbmBinary.tsv @@ -4,14 +4,15 @@ #@ col=X2VBuffer:R4:1-4 #@ col=X3Important:R4:5 #@ col=Label:R4:6 -#@ col=Features:R4:7-12 -#@ col=Features:R4:13-18 -#@ col=FeatureContributions:R4:19-24 -#@ col=FeatureContributions:R4:25-30 -#@ col=FeatureContributions:R4:31-36 -#@ col=FeatureContributions:R4:37-42 +#@ col=Label:BL:7 +#@ col=Features:R4:8-13 +#@ col=Features:R4:14-19 +#@ col=FeatureContributions:R4:20-25 +#@ col=FeatureContributions:R4:26-31 +#@ col=FeatureContributions:R4:32-37 +#@ col=FeatureContributions:R4:38-43 #@ } -950 757 692 720 297 7515 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0.0871070847 0.116535939 0 0 0 1 18 4:-1 10:-0.0126099735 11:1 16:-0.262451649 17:20.81302 -459 961 0 659 274 2147 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 24 1:1 6:-0.0739851445 9:-0.0561627634 11:-1 13:0.103110082 17:-1 19:2.238687 23:-21.7116222 -672 275 0 65 195 9818 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 24 0:0.09341861 5:1 7:-0.99999994 10:-0.267225862 13:-0.0506075844 17:1 19:-0.9821341 23:19.4068565 -186 301 0 681 526 1456 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 24 4:1 6:-0.0786664 9:-0.05971634 11:-1 16:0.0128529193 17:-1 22:0.262451649 23:-20.4196148 +950 757 692 720 297 7515 1 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0.0871070847 0.116535939 0 0 0 1 18 4:-1 10:-0.0126099735 11:1 16:-0.262451649 17:20.81302 +459 961 0 659 274 2147 0 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 24 1:1 6:-0.0739851445 9:-0.0561627634 11:-1 13:0.103110082 17:-1 19:2.238687 23:-21.7116222 +672 275 0 65 195 9818 1 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 24 0:0.09341861 5:1 7:-0.99999994 10:-0.267225862 13:-0.0506075844 17:1 19:-0.9821341 23:19.4068565 +186 301 0 681 526 1456 0 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 24 4:1 6:-0.0786664 9:-0.05971634 11:-1 16:0.0128529193 17:-1 22:0.262451649 23:-20.4196148 diff --git a/test/BaselineOutput/Common/FeatureContribution/LightGbmRanking.tsv b/test/BaselineOutput/Common/FeatureContribution/LightGbmRanking.tsv index 0072bd847e..5b9fe7dca0 100644 --- a/test/BaselineOutput/Common/FeatureContribution/LightGbmRanking.tsv +++ b/test/BaselineOutput/Common/FeatureContribution/LightGbmRanking.tsv @@ -13,7 +13,7 @@ #@ col=FeatureContributions:R4:33-38 #@ col=FeatureContributions:R4:39-44 #@ } -950 757 692 720 297 7515 4 1 0 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0.678461254 0 0 0.453011334 0 1 18 4:-1 10:-0.09636103 11:1 16:-0.08202158 17:0.851190269 -459 961 0 659 274 2147 1 1 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 0.47573778 0.736526 0 1 0 0 18 5:-1 9:0.0479629859 11:-1 15:0.430636823 17:-8.978524 -672 275 0 65 195 9818 4 1 0 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0.161843508 0 0 0 0 1 0 -0.99999994 0 -0.844456732 -0.413860559 0 12 1:-0.105168335 5:1 7:-0.4455722 11:4.23675251 -186 301 0 681 526 1456 0 1 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 0.7739479 0 0 1 0 0 0 -0.02896308 0 0 -0.0137862992 -1 12 3:0.0473209359 5:-1 9:0.430636823 11:-9.100345 +950 757 692 720 297 7515 4 1 0 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0.67846173 0 0 0.453011572 0 1 18 4:-1 10:-0.09636109 11:1 16:-0.08202157 17:0.8511897 +459 961 0 659 274 2147 1 1 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 0.475736767 0.736526251 0 1 0 0 18 5:-1 9:0.0479629934 11:-1 15:0.430636883 17:-8.978524 +672 275 0 65 195 9818 4 1 0 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0.161843479 0 0 0 0 1 0 -0.99999994 0 -0.8444566 -0.413860559 0 12 1:-0.10516832 5:1 7:-0.4455722 11:4.236753 +186 301 0 681 526 1456 0 1 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 0.7739468 0 0 1 0 0 0 -0.0289630964 0 0 -0.0137862926 -1 12 3:0.04732094 5:-1 9:0.430636883 11:-9.100346 diff --git a/test/BaselineOutput/Common/FeatureContribution/LogisticRegressionBinary.tsv b/test/BaselineOutput/Common/FeatureContribution/LogisticRegressionBinary.tsv index 5c25c00253..a3be66c30f 100644 --- a/test/BaselineOutput/Common/FeatureContribution/LogisticRegressionBinary.tsv +++ b/test/BaselineOutput/Common/FeatureContribution/LogisticRegressionBinary.tsv @@ -4,14 +4,15 @@ #@ col=X2VBuffer:R4:1-4 #@ col=X3Important:R4:5 #@ col=Label:R4:6 -#@ col=Features:R4:7-12 -#@ col=Features:R4:13-18 -#@ col=FeatureContributions:R4:19-24 -#@ col=FeatureContributions:R4:25-30 -#@ col=FeatureContributions:R4:31-36 -#@ col=FeatureContributions:R4:37-42 +#@ col=Label:BL:7 +#@ col=Features:R4:8-13 +#@ col=Features:R4:14-19 +#@ col=FeatureContributions:R4:20-25 +#@ col=FeatureContributions:R4:26-31 +#@ col=FeatureContributions:R4:32-37 +#@ col=FeatureContributions:R4:38-43 #@ } -950 757 692 720 297 7515 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 24 1:0.0202900227 5:1 17:1 23:3.28757548 -459 961 0 659 274 2147 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 24 1:0.09015856 5:0.99999994 17:0.99999994 23:0.9392448 -672 275 0 65 195 9818 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 24 1:0.00564189861 5:1 17:1 23:4.2950654 -186 301 0 681 526 1456 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 24 1:0.04164096 5:1 17:1 23:0.6369541 +950 757 692 720 297 7515 1 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 24 1:0.0202900227 5:1 17:1 23:3.28757548 +459 961 0 659 274 2147 0 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 24 1:0.09015856 5:0.99999994 17:0.99999994 23:0.9392448 +672 275 0 65 195 9818 1 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 24 1:0.00564189861 5:1 17:1 23:4.2950654 +186 301 0 681 526 1456 0 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 24 1:0.04164096 5:1 17:1 23:0.6369541 diff --git a/test/BaselineOutput/Common/FeatureContribution/SDCABinary.tsv b/test/BaselineOutput/Common/FeatureContribution/SDCABinary.tsv index d04a85014d..773de450c6 100644 --- a/test/BaselineOutput/Common/FeatureContribution/SDCABinary.tsv +++ b/test/BaselineOutput/Common/FeatureContribution/SDCABinary.tsv @@ -4,14 +4,15 @@ #@ col=X2VBuffer:R4:1-4 #@ col=X3Important:R4:5 #@ col=Label:R4:6 -#@ col=Features:R4:7-12 -#@ col=Features:R4:13-18 -#@ col=FeatureContributions:R4:19-24 -#@ col=FeatureContributions:R4:25-30 -#@ col=FeatureContributions:R4:31-36 -#@ col=FeatureContributions:R4:37-42 +#@ col=Label:BL:7 +#@ col=Features:R4:8-13 +#@ col=Features:R4:14-19 +#@ col=FeatureContributions:R4:20-25 +#@ col=FeatureContributions:R4:26-31 +#@ col=FeatureContributions:R4:32-37 +#@ col=FeatureContributions:R4:38-43 #@ } -950 757 692 720 297 7515 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0.2054445 0 0 0 0.003720135 1 0 -1 -0.8797944 -0.645997 0 0 12 1:-0.0406201333 5:1 7:-0.9716351 11:23.9200363 -459 961 0 659 274 2147 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 0.347440571 0 0 0 0.012012952 0.99999994 18 1:-1 3:-0.465753257 7:-0.180495262 11:0.99999994 13:-1.233476 17:6.83384132 -672 275 0 65 195 9818 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0.111236207 0 0 0 0.00186957465 1 18 1:-1 3:-0.160536781 7:-0.0112949461 11:1 13:-0.352971822 17:31.2504215 -186 301 0 681 526 1456 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 0.2076115 0 0 0 0.0340060033 1 18 1:-0.650766969 3:-1 9:-0.128101453 11:1 15:-0.5936744 17:4.634408 +950 757 692 720 297 7515 1 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0.2054445 0 0 0 0.003720135 1 0 -1 -0.8797944 -0.645997 0 0 12 1:-0.0406201333 5:1 7:-0.9716351 11:23.9200363 +459 961 0 659 274 2147 0 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 0.347440571 0 0 0 0.012012952 0.99999994 18 1:-1 3:-0.465753257 7:-0.180495262 11:0.99999994 13:-1.233476 17:6.83384132 +672 275 0 65 195 9818 1 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0.111236207 0 0 0 0.00186957465 1 18 1:-1 3:-0.160536781 7:-0.0112949461 11:1 13:-0.352971822 17:31.2504215 +186 301 0 681 526 1456 0 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 0.2076115 0 0 0 0.0340060033 1 18 1:-0.650766969 3:-1 9:-0.128101453 11:1 15:-0.5936744 17:4.634408 diff --git a/test/BaselineOutput/Common/FeatureContribution/SGDBinary.tsv b/test/BaselineOutput/Common/FeatureContribution/SGDBinary.tsv index bb00bf43f3..581029695f 100644 --- a/test/BaselineOutput/Common/FeatureContribution/SGDBinary.tsv +++ b/test/BaselineOutput/Common/FeatureContribution/SGDBinary.tsv @@ -4,14 +4,15 @@ #@ col=X2VBuffer:R4:1-4 #@ col=X3Important:R4:5 #@ col=Label:R4:6 -#@ col=Features:R4:7-12 -#@ col=Features:R4:13-18 -#@ col=FeatureContributions:R4:19-24 -#@ col=FeatureContributions:R4:25-30 -#@ col=FeatureContributions:R4:31-36 -#@ col=FeatureContributions:R4:37-42 +#@ col=Label:BL:7 +#@ col=Features:R4:8-13 +#@ col=Features:R4:14-19 +#@ col=FeatureContributions:R4:20-25 +#@ col=FeatureContributions:R4:26-31 +#@ col=FeatureContributions:R4:32-37 +#@ col=FeatureContributions:R4:38-43 #@ } -950 757 692 720 297 7515 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0 0.04723223 0 0 0 1 -0.7672315 0 0 -0.6374357 -1 0 12 4:-0.06509055 5:1 10:-0.09172161 11:1.40913868 -459 961 0 659 274 2147 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 0 0.209876046 0 0 0 0.99999994 -0.401810616 0 0 -0.6324048 -1 0 12 4:-0.210188508 5:0.99999994 10:-0.08461859 11:0.402584285 -672 275 0 65 195 9818 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0 0.0131335221 0 0 0 0.99999994 -0.826597333 0 0 -0.0876474 -1 0 12 4:-0.0327116176 5:0.99999994 10:-0.06022126 11:1.84097457 -186 301 0 681 526 1456 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 0 0.09693411 0 0 0 1 -0.08481769 0 0 -0.340425164 -0.99999994 0 12 4:-0.594997048 5:1 10:-0.162443 11:0.273014784 +950 757 692 720 297 7515 1 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0 0.04723223 0 0 0 1 -0.7672315 0 0 -0.6374357 -1 0 12 4:-0.06509055 5:1 10:-0.09172161 11:1.40913868 +459 961 0 659 274 2147 0 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 0 0.209876046 0 0 0 0.99999994 -0.401810616 0 0 -0.6324048 -1 0 12 4:-0.210188508 5:0.99999994 10:-0.08461859 11:0.402584285 +672 275 0 65 195 9818 1 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0 0.0131335221 0 0 0 0.99999994 -0.826597333 0 0 -0.0876474 -1 0 12 4:-0.0327116176 5:0.99999994 10:-0.06022126 11:1.84097457 +186 301 0 681 526 1456 0 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 0 0.09693411 0 0 0 1 -0.08481769 0 0 -0.340425164 -0.99999994 0 12 4:-0.594997048 5:1 10:-0.162443 11:0.273014784 diff --git a/test/BaselineOutput/Common/FeatureContribution/SSGDBinary.tsv b/test/BaselineOutput/Common/FeatureContribution/SSGDBinary.tsv index 978425888e..52594221f6 100644 --- a/test/BaselineOutput/Common/FeatureContribution/SSGDBinary.tsv +++ b/test/BaselineOutput/Common/FeatureContribution/SSGDBinary.tsv @@ -4,14 +4,15 @@ #@ col=X2VBuffer:R4:1-4 #@ col=X3Important:R4:5 #@ col=Label:R4:6 -#@ col=Features:R4:7-12 -#@ col=Features:R4:13-18 -#@ col=FeatureContributions:R4:19-24 -#@ col=FeatureContributions:R4:25-30 -#@ col=FeatureContributions:R4:31-36 -#@ col=FeatureContributions:R4:37-42 +#@ col=Label:BL:7 +#@ col=Features:R4:8-13 +#@ col=Features:R4:14-19 +#@ col=FeatureContributions:R4:20-25 +#@ col=FeatureContributions:R4:26-31 +#@ col=FeatureContributions:R4:32-37 +#@ col=FeatureContributions:R4:38-43 #@ } -950 757 692 720 297 7515 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0.0389229059 0 0.185620338 0 0 1 18 1:-1 4:-0.8814666 7:-0.06705473 11:1 13:-4.40001869 17:65.61832 -459 961 0 659 274 2147 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 0.06582507 0 0 0.0127044721 0 1 18 1:-1 4:-0.6405787 7:-0.2979572 11:1 13:-5.585757 17:18.7468414 -672 275 0 65 195 9818 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0.0210744832 0 0 0.000274027116 0 1 18 1:-0.6277009 4:-1 10:-0.02970431 11:1 16:-2.54647 17:85.72729 -186 301 0 681 526 1456 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 0.0393334627 0 0 0.0193592682 0 1 18 1:-0.2547038 4:-1 10:-0.5402966 11:1 16:-6.86894 17:12.7132759 +950 757 692 720 297 7515 1 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0.0389229059 0 0.185620338 0 0 1 18 1:-1 4:-0.8814666 7:-0.06705473 11:1 13:-4.40001869 17:65.61832 +459 961 0 659 274 2147 0 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 0.06582507 0 0 0.0127044721 0 1 18 1:-1 4:-0.6405787 7:-0.2979572 11:1 13:-5.585757 17:18.7468414 +672 275 0 65 195 9818 1 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0.0210744832 0 0 0.000274027116 0 1 18 1:-0.6277009 4:-1 10:-0.02970431 11:1 16:-2.54647 17:85.72729 +186 301 0 681 526 1456 0 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 0.0393334627 0 0 0.0193592682 0 1 18 1:-0.2547038 4:-1 10:-0.5402966 11:1 16:-6.86894 17:12.7132759 diff --git a/test/BaselineOutput/Common/FeatureContribution/SVMBinary.tsv b/test/BaselineOutput/Common/FeatureContribution/SVMBinary.tsv index 197d199376..ed82732086 100644 --- a/test/BaselineOutput/Common/FeatureContribution/SVMBinary.tsv +++ b/test/BaselineOutput/Common/FeatureContribution/SVMBinary.tsv @@ -4,14 +4,15 @@ #@ col=X2VBuffer:R4:1-4 #@ col=X3Important:R4:5 #@ col=Label:R4:6 -#@ col=Features:R4:7-12 -#@ col=Features:R4:13-18 -#@ col=FeatureContributions:R4:19-24 -#@ col=FeatureContributions:R4:25-30 -#@ col=FeatureContributions:R4:31-36 -#@ col=FeatureContributions:R4:37-42 +#@ col=Label:BL:7 +#@ col=Features:R4:8-13 +#@ col=Features:R4:14-19 +#@ col=FeatureContributions:R4:20-25 +#@ col=FeatureContributions:R4:26-31 +#@ col=FeatureContributions:R4:32-37 +#@ col=FeatureContributions:R4:38-43 #@ } -950 757 692 720 297 7515 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0 0.05771441 0 0 0.0166732743 1 18 0:-0.99999994 2:-0.838546038 6:-0.111203589 11:1 12:-0.688422263 17:6.190648 -459 961 0 659 274 2147 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 0 0.256453544 0 0 0.053840857 1 18 0:-1 6:-0.188063636 11:1 12:-0.332616657 17:1.76863885 -672 275 0 65 195 9818 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0 0.0160482265 0 0 0.008379247 1 18 0:-0.99999994 6:-0.060210254 11:1 12:-0.4869682 17:8.087795 -186 301 0 681 526 1456 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 0 0.118446559 0 0 0.15241152 1 18 0:-1 6:-0.112376556 11:1 12:-0.134785831 17:1.19941235 +950 757 692 720 297 7515 1 1 950 757 692 720 297 7515 0.956696868 0.760804 0.7872582 0.754716933 0.297893673 0.7578661 0 0.05771441 0 0 0.0166732743 1 18 0:-0.99999994 2:-0.838546038 6:-0.111203589 11:1 12:-0.688422263 17:6.190648 +459 961 0 659 274 2147 0 0 459 961 0 659 274 2147 0.462235659 0.965829134 0 0.690775633 0.27482447 0.21651876 0 0.256453544 0 0 0.053840857 1 18 0:-1 6:-0.188063636 11:1 12:-0.332616657 17:1.76863885 +672 275 0 65 195 9818 1 1 672 275 0 65 195 9818 0.6767372 0.2763819 0 0.06813417 0.195586756 0.990116954 0 0.0160482265 0 0 0.008379247 1 18 0:-0.99999994 6:-0.060210254 11:1 12:-0.4869682 17:8.087795 +186 301 0 681 526 1456 0 0 186 301 0 681 526 1456 0.187311172 0.302512556 0 0.713836432 0.527582765 0.1468334 0 0.118446559 0 0 0.15241152 1 18 0:-1 6:-0.112376556 11:1 12:-0.134785831 17:1.19941235 diff --git a/test/BaselineOutput/Common/OVA/OVA-CV-iris-out.txt b/test/BaselineOutput/Common/OVA/OVA-CV-iris-out.txt index e81b3a9f75..deb50892c8 100644 --- a/test/BaselineOutput/Common/OVA/OVA-CV-iris-out.txt +++ b/test/BaselineOutput/Common/OVA/OVA-CV-iris-out.txt @@ -1,4 +1,4 @@ -maml.exe CV tr=OVA{p=AvgPer{ lr=0.8 }} threads=- norm=No dout=%Output% data=%Data% seed=1 +maml.exe CV tr=OVA{p=AvgPer{ lr=0.8 }} threads=- norm=No dout=%Output% data=%Data% seed=1 xf=Term{col=Label} Not adding a normalizer. Training learner 0 Training calibrator. diff --git a/test/BaselineOutput/Common/OVA/OVA-CV-iris-rp.txt b/test/BaselineOutput/Common/OVA/OVA-CV-iris-rp.txt index 23b93e5bc7..aaa1e35d12 100644 --- a/test/BaselineOutput/Common/OVA/OVA-CV-iris-rp.txt +++ b/test/BaselineOutput/Common/OVA/OVA-CV-iris-rp.txt @@ -1,4 +1,4 @@ OVA Accuracy(micro-avg) Accuracy(macro-avg) Log-loss Log-loss reduction /p Learner Name Train Dataset Test Dataset Results File Run Time Physical Memory Virtual Memory Command Line Settings -0.929667 0.936111 0.346785 68.08855 AvgPer{lr=0.8} OVA %Data% %Output% 99 0 0 maml.exe CV tr=OVA{p=AvgPer{ lr=0.8 }} threads=- norm=No dout=%Output% data=%Data% seed=1 /p:AvgPer{lr=0.8} +0.929667 0.936111 0.346785 68.08855 AvgPer{lr=0.8} OVA %Data% %Output% 99 0 0 maml.exe CV tr=OVA{p=AvgPer{ lr=0.8 }} threads=- norm=No dout=%Output% data=%Data% seed=1 xf=Term{col=Label} /p:AvgPer{lr=0.8} diff --git a/test/BaselineOutput/Common/OVA/OVA-FastForest-CV-iris-out.txt b/test/BaselineOutput/Common/OVA/OVA-FastForest-CV-iris-out.txt index 1a5da994b3..a51313142c 100644 --- a/test/BaselineOutput/Common/OVA/OVA-FastForest-CV-iris-out.txt +++ b/test/BaselineOutput/Common/OVA/OVA-FastForest-CV-iris-out.txt @@ -1,4 +1,4 @@ -maml.exe CV tr=OVA{p=FastForest{ }} threads=- norm=No dout=%Output% data=%Data% seed=1 +maml.exe CV tr=OVA{p=FastForest{ }} threads=- norm=No dout=%Output% data=%Data% seed=1 xf=Term{col=Label} Not adding a normalizer. Training learner 0 Making per-feature arrays diff --git a/test/BaselineOutput/Common/OVA/OVA-FastForest-CV-iris-rp.txt b/test/BaselineOutput/Common/OVA/OVA-FastForest-CV-iris-rp.txt index 56ab2aa973..6c2a435b51 100644 --- a/test/BaselineOutput/Common/OVA/OVA-FastForest-CV-iris-rp.txt +++ b/test/BaselineOutput/Common/OVA/OVA-FastForest-CV-iris-rp.txt @@ -1,4 +1,4 @@ OVA Accuracy(micro-avg) Accuracy(macro-avg) Log-loss Log-loss reduction /p Learner Name Train Dataset Test Dataset Results File Run Time Physical Memory Virtual Memory Command Line Settings -0.940899 0.942785 0.150571 86.14639 FastForest{} OVA %Data% %Output% 99 0 0 maml.exe CV tr=OVA{p=FastForest{ }} threads=- norm=No dout=%Output% data=%Data% seed=1 /p:FastForest{} +0.940899 0.942785 0.150571 86.14639 FastForest{} OVA %Data% %Output% 99 0 0 maml.exe CV tr=OVA{p=FastForest{ }} threads=- norm=No dout=%Output% data=%Data% seed=1 xf=Term{col=Label} /p:FastForest{} diff --git a/test/BaselineOutput/Common/OVA/OVA-FastForest-TrainTest-iris-out.txt b/test/BaselineOutput/Common/OVA/OVA-FastForest-TrainTest-iris-out.txt index 685dbd1c35..269de78db3 100644 --- a/test/BaselineOutput/Common/OVA/OVA-FastForest-TrainTest-iris-out.txt +++ b/test/BaselineOutput/Common/OVA/OVA-FastForest-TrainTest-iris-out.txt @@ -1,4 +1,4 @@ -maml.exe TrainTest test=%Data% tr=OVA{p=FastForest{ }} norm=No dout=%Output% data=%Data% out=%Output% seed=1 +maml.exe TrainTest test=%Data% tr=OVA{p=FastForest{ }} norm=No dout=%Output% data=%Data% out=%Output% seed=1 xf=Term{col=Label} Not adding a normalizer. Training learner 0 Making per-feature arrays diff --git a/test/BaselineOutput/Common/OVA/OVA-FastForest-TrainTest-iris-rp.txt b/test/BaselineOutput/Common/OVA/OVA-FastForest-TrainTest-iris-rp.txt index ccedd71525..9ecf267068 100644 --- a/test/BaselineOutput/Common/OVA/OVA-FastForest-TrainTest-iris-rp.txt +++ b/test/BaselineOutput/Common/OVA/OVA-FastForest-TrainTest-iris-rp.txt @@ -1,4 +1,4 @@ OVA Accuracy(micro-avg) Accuracy(macro-avg) Log-loss Log-loss reduction /p Learner Name Train Dataset Test Dataset Results File Run Time Physical Memory Virtual Memory Command Line Settings -0.973333 0.973333 0.088201 91.97161 FastForest{} OVA %Data% %Data% %Output% 99 0 0 maml.exe TrainTest test=%Data% tr=OVA{p=FastForest{ }} norm=No dout=%Output% data=%Data% out=%Output% seed=1 /p:FastForest{} +0.973333 0.973333 0.088201 91.97161 FastForest{} OVA %Data% %Data% %Output% 99 0 0 maml.exe TrainTest test=%Data% tr=OVA{p=FastForest{ }} norm=No dout=%Output% data=%Data% out=%Output% seed=1 xf=Term{col=Label} /p:FastForest{} diff --git a/test/BaselineOutput/Common/OVA/OVA-TrainTest-iris-out.txt b/test/BaselineOutput/Common/OVA/OVA-TrainTest-iris-out.txt index f0bb6cd1b8..f5d83dd965 100644 --- a/test/BaselineOutput/Common/OVA/OVA-TrainTest-iris-out.txt +++ b/test/BaselineOutput/Common/OVA/OVA-TrainTest-iris-out.txt @@ -1,4 +1,4 @@ -maml.exe TrainTest test=%Data% tr=OVA{p=AvgPer{ lr=0.8 }} norm=No dout=%Output% data=%Data% out=%Output% seed=1 +maml.exe TrainTest test=%Data% tr=OVA{p=AvgPer{ lr=0.8 }} norm=No dout=%Output% data=%Data% out=%Output% seed=1 xf=Term{col=Label} Not adding a normalizer. Training learner 0 Training calibrator. diff --git a/test/BaselineOutput/Common/OVA/OVA-TrainTest-iris-rp.txt b/test/BaselineOutput/Common/OVA/OVA-TrainTest-iris-rp.txt index 12797db1fb..8c7cc8e672 100644 --- a/test/BaselineOutput/Common/OVA/OVA-TrainTest-iris-rp.txt +++ b/test/BaselineOutput/Common/OVA/OVA-TrainTest-iris-rp.txt @@ -1,4 +1,4 @@ OVA Accuracy(micro-avg) Accuracy(macro-avg) Log-loss Log-loss reduction /p Learner Name Train Dataset Test Dataset Results File Run Time Physical Memory Virtual Memory Command Line Settings -0.973333 0.973333 0.291849 73.43472 AvgPer{lr=0.8} OVA %Data% %Data% %Output% 99 0 0 maml.exe TrainTest test=%Data% tr=OVA{p=AvgPer{ lr=0.8 }} norm=No dout=%Output% data=%Data% out=%Output% seed=1 /p:AvgPer{lr=0.8} +0.973333 0.973333 0.291849 73.43472 AvgPer{lr=0.8} OVA %Data% %Data% %Output% 99 0 0 maml.exe TrainTest test=%Data% tr=OVA{p=AvgPer{ lr=0.8 }} norm=No dout=%Output% data=%Data% out=%Output% seed=1 xf=Term{col=Label} /p:AvgPer{lr=0.8} diff --git a/test/BaselineOutput/Common/PKPD/PKPD-CV-iris-out.txt b/test/BaselineOutput/Common/PKPD/PKPD-CV-iris-out.txt index 60bf133c7c..e9406d195c 100644 --- a/test/BaselineOutput/Common/PKPD/PKPD-CV-iris-out.txt +++ b/test/BaselineOutput/Common/PKPD/PKPD-CV-iris-out.txt @@ -1,4 +1,4 @@ -maml.exe CV tr=PKPD{p=AvgPer { lr=0.8 }} threads=- norm=No dout=%Output% data=%Data% seed=1 +maml.exe CV tr=PKPD{p=AvgPer { lr=0.8 }} threads=- norm=No dout=%Output% data=%Data% seed=1 xf=Term{col=Label} Not adding a normalizer. Training learner (0,0) Training calibrator. diff --git a/test/BaselineOutput/Common/PKPD/PKPD-CV-iris-rp.txt b/test/BaselineOutput/Common/PKPD/PKPD-CV-iris-rp.txt index d2d984758b..2eae61966b 100644 --- a/test/BaselineOutput/Common/PKPD/PKPD-CV-iris-rp.txt +++ b/test/BaselineOutput/Common/PKPD/PKPD-CV-iris-rp.txt @@ -1,4 +1,4 @@ PKPD Accuracy(micro-avg) Accuracy(macro-avg) Log-loss Log-loss reduction /p Learner Name Train Dataset Test Dataset Results File Run Time Physical Memory Virtual Memory Command Line Settings -0.966928 0.966667 0.351395 67.66164 AvgPer{lr=0.8} PKPD %Data% %Output% 99 0 0 maml.exe CV tr=PKPD{p=AvgPer { lr=0.8 }} threads=- norm=No dout=%Output% data=%Data% seed=1 /p:AvgPer{lr=0.8} +0.966928 0.966667 0.351395 67.66164 AvgPer{lr=0.8} PKPD %Data% %Output% 99 0 0 maml.exe CV tr=PKPD{p=AvgPer { lr=0.8 }} threads=- norm=No dout=%Output% data=%Data% seed=1 xf=Term{col=Label} /p:AvgPer{lr=0.8} diff --git a/test/BaselineOutput/Common/PKPD/PKPD-TrainTest-iris-out.txt b/test/BaselineOutput/Common/PKPD/PKPD-TrainTest-iris-out.txt index e4ceccca5e..7537ead92f 100644 --- a/test/BaselineOutput/Common/PKPD/PKPD-TrainTest-iris-out.txt +++ b/test/BaselineOutput/Common/PKPD/PKPD-TrainTest-iris-out.txt @@ -1,4 +1,4 @@ -maml.exe TrainTest test=%Data% tr=PKPD{p=AvgPer { lr=0.8 }} norm=No dout=%Output% data=%Data% out=%Output% seed=1 +maml.exe TrainTest test=%Data% tr=PKPD{p=AvgPer { lr=0.8 }} norm=No dout=%Output% data=%Data% out=%Output% seed=1 xf=Term{col=Label} Not adding a normalizer. Training learner (0,0) Training calibrator. diff --git a/test/BaselineOutput/Common/PKPD/PKPD-TrainTest-iris-rp.txt b/test/BaselineOutput/Common/PKPD/PKPD-TrainTest-iris-rp.txt index 05d350bd96..b87f2ebc53 100644 --- a/test/BaselineOutput/Common/PKPD/PKPD-TrainTest-iris-rp.txt +++ b/test/BaselineOutput/Common/PKPD/PKPD-TrainTest-iris-rp.txt @@ -1,4 +1,4 @@ PKPD Accuracy(micro-avg) Accuracy(macro-avg) Log-loss Log-loss reduction /p Learner Name Train Dataset Test Dataset Results File Run Time Physical Memory Virtual Memory Command Line Settings -0.98 0.98 0.272667 75.18077 AvgPer{lr=0.8} PKPD %Data% %Data% %Output% 99 0 0 maml.exe TrainTest test=%Data% tr=PKPD{p=AvgPer { lr=0.8 }} norm=No dout=%Output% data=%Data% out=%Output% seed=1 /p:AvgPer{lr=0.8} +0.98 0.98 0.272667 75.18077 AvgPer{lr=0.8} PKPD %Data% %Data% %Output% 99 0 0 maml.exe TrainTest test=%Data% tr=PKPD{p=AvgPer { lr=0.8 }} norm=No dout=%Output% data=%Data% out=%Output% seed=1 xf=Term{col=Label} /p:AvgPer{lr=0.8} diff --git a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-out.txt b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-out.txt index 2dd90d6312..85a6761d32 100644 --- a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-out.txt +++ b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-out.txt @@ -1,4 +1,4 @@ -maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} threads=- norm=No dout=%Output% data=%Data% seed=1 +maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} threads=- norm=No dout=%Output% data=%Data% seed=1 xf=Term{col=Label} Not adding a normalizer. Beginning optimization num vars: 15 @@ -53,50 +53,56 @@ Virtual memory usage(MB): %Number% %DateTime% Time elapsed(s): %Number% --- Progress log --- -[1] 'LBFGS data prep' started. -[1] 'LBFGS data prep' finished in %Time%. -[2] 'LBFGS Optimizer' started. -[2] (%Time%) 0 iterations Loss: 1.0986123085022 -[2] (%Time%) 1 iterations Loss: 1.00646448135376 Improvement: 0.09215 -[2] (%Time%) 2 iterations Loss: 0.909583747386932 Improvement: 0.09593 -[2] (%Time%) 3 iterations Loss: 0.525106191635132 Improvement: 0.3158 -[2] (%Time%) 4 iterations Loss: 0.400520384311676 Improvement: 0.1718 -[2] (%Time%) 5 iterations Loss: 0.332601189613342 Improvement: 0.09382 -[2] (%Time%) 6 iterations Loss: 0.281388521194458 Improvement: 0.06186 -[2] (%Time%) 7 iterations Loss: 0.237996473908424 Improvement: 0.04801 -[2] (%Time%) 8 iterations Loss: 0.212298363447189 Improvement: 0.03128 -[2] (%Time%) 9 iterations Loss: 0.199792444705963 Improvement: 0.0172 -[2] (%Time%) 10 iterations Loss: 0.194789424538612 Improvement: 0.008052 -[2] (%Time%) 11 iterations Loss: 0.193230450153351 Improvement: 0.003182 -[2] (%Time%) 12 iterations Loss: 0.192447692155838 Improvement: 0.001383 -[2] (%Time%) 13 iterations Loss: 0.189304739236832 Improvement: 0.002703 -[2] (%Time%) 14 iterations Loss: 0.187662661075592 Improvement: 0.001907 -[2] (%Time%) 15 iterations Loss: 0.185374572873116 Improvement: 0.002193 -[2] (%Time%) 16 iterations Loss: 0.18364554643631 Improvement: 0.001845 -[2] (%Time%) 17 iterations Loss: 0.180794909596443 Improvement: 0.002599 -[2] (%Time%) 18 iterations Loss: 0.178908497095108 Improvement: 0.002065 -[2] (%Time%) 19 iterations Loss: 0.175620675086975 Improvement: 0.002982 -[2] (%Time%) 20 iterations Loss: 0.174758642911911 Improvement: 0.001392 -[2] (%Time%) 21 iterations Loss: 0.173962101340294 Improvement: 0.0009454 -[2] 'LBFGS Optimizer' finished in %Time%. -[3] 'LBFGS data prep #2' started. -[3] 'LBFGS data prep #2' finished in %Time%. -[4] 'LBFGS Optimizer #2' started. -[4] (%Time%) 0 iterations Loss: 1.0986123085022 -[4] (%Time%) 1 iterations Loss: 1.05856335163116 Improvement: 0.04005 -[4] (%Time%) 2 iterations Loss: 1.00281620025635 Improvement: 0.05261 -[4] (%Time%) 3 iterations Loss: 0.97780430316925 Improvement: 0.03158 -[4] (%Time%) 4 iterations Loss: 0.752716302871704 Improvement: 0.1773 -[4] (%Time%) 5 iterations Loss: 0.542387366294861 Improvement: 0.2021 -[4] (%Time%) 6 iterations Loss: 0.443084180355072 Improvement: 0.125 -[4] (%Time%) 7 iterations Loss: 0.343867212533951 Improvement: 0.1057 -[4] (%Time%) 8 iterations Loss: 0.284590691328049 Improvement: 0.07087 -[4] (%Time%) 9 iterations Loss: 0.254261910915375 Improvement: 0.04046 -[4] (%Time%) 10 iterations Loss: 0.224356189370155 Improvement: 0.03255 -[4] (%Time%) 11 iterations Loss: 0.215291574597359 Improvement: 0.01493 -[4] (%Time%) 12 iterations Loss: 0.212821274995804 Improvement: 0.005586 -[4] (%Time%) 13 iterations Loss: 0.212086588144302 Improvement: 0.001948 -[4] (%Time%) 14 iterations Loss: 0.21061946451664 Improvement: 0.001587 -[4] (%Time%) 15 iterations Loss: 0.209799557924271 Improvement: 0.001012 -[4] (%Time%) 16 iterations Loss: 0.209267094731331 Improvement: 0.0006523 -[4] 'LBFGS Optimizer #2' finished in %Time%. +[1] 'Building term dictionary' started. +[1] (%Time%) 71 examples Total Terms: 3 +[1] 'Building term dictionary' finished in %Time%. +[2] 'LBFGS data prep' started. +[2] 'LBFGS data prep' finished in %Time%. +[3] 'LBFGS Optimizer' started. +[3] (%Time%) 0 iterations Loss: 1.0986123085022 +[3] (%Time%) 1 iterations Loss: 1.00646448135376 Improvement: 0.09215 +[3] (%Time%) 2 iterations Loss: 0.909583747386932 Improvement: 0.09593 +[3] (%Time%) 3 iterations Loss: 0.525106191635132 Improvement: 0.3158 +[3] (%Time%) 4 iterations Loss: 0.400520384311676 Improvement: 0.1718 +[3] (%Time%) 5 iterations Loss: 0.332601189613342 Improvement: 0.09382 +[3] (%Time%) 6 iterations Loss: 0.281388521194458 Improvement: 0.06186 +[3] (%Time%) 7 iterations Loss: 0.237996473908424 Improvement: 0.04801 +[3] (%Time%) 8 iterations Loss: 0.212298363447189 Improvement: 0.03128 +[3] (%Time%) 9 iterations Loss: 0.199792444705963 Improvement: 0.0172 +[3] (%Time%) 10 iterations Loss: 0.194789424538612 Improvement: 0.008052 +[3] (%Time%) 11 iterations Loss: 0.193230450153351 Improvement: 0.003182 +[3] (%Time%) 12 iterations Loss: 0.192447692155838 Improvement: 0.001383 +[3] (%Time%) 13 iterations Loss: 0.189304739236832 Improvement: 0.002703 +[3] (%Time%) 14 iterations Loss: 0.187662661075592 Improvement: 0.001907 +[3] (%Time%) 15 iterations Loss: 0.185374572873116 Improvement: 0.002193 +[3] (%Time%) 16 iterations Loss: 0.18364554643631 Improvement: 0.001845 +[3] (%Time%) 17 iterations Loss: 0.180794909596443 Improvement: 0.002599 +[3] (%Time%) 18 iterations Loss: 0.178908497095108 Improvement: 0.002065 +[3] (%Time%) 19 iterations Loss: 0.175620675086975 Improvement: 0.002982 +[3] (%Time%) 20 iterations Loss: 0.174758642911911 Improvement: 0.001392 +[3] (%Time%) 21 iterations Loss: 0.173962101340294 Improvement: 0.0009454 +[3] 'LBFGS Optimizer' finished in %Time%. +[4] 'Building term dictionary #2' started. +[4] (%Time%) 79 examples Total Terms: 3 +[4] 'Building term dictionary #2' finished in %Time%. +[5] 'LBFGS data prep #2' started. +[5] 'LBFGS data prep #2' finished in %Time%. +[6] 'LBFGS Optimizer #2' started. +[6] (%Time%) 0 iterations Loss: 1.0986123085022 +[6] (%Time%) 1 iterations Loss: 1.05856335163116 Improvement: 0.04005 +[6] (%Time%) 2 iterations Loss: 1.00281620025635 Improvement: 0.05261 +[6] (%Time%) 3 iterations Loss: 0.97780430316925 Improvement: 0.03158 +[6] (%Time%) 4 iterations Loss: 0.752716302871704 Improvement: 0.1773 +[6] (%Time%) 5 iterations Loss: 0.542387366294861 Improvement: 0.2021 +[6] (%Time%) 6 iterations Loss: 0.443084180355072 Improvement: 0.125 +[6] (%Time%) 7 iterations Loss: 0.343867212533951 Improvement: 0.1057 +[6] (%Time%) 8 iterations Loss: 0.284590691328049 Improvement: 0.07087 +[6] (%Time%) 9 iterations Loss: 0.254261910915375 Improvement: 0.04046 +[6] (%Time%) 10 iterations Loss: 0.224356189370155 Improvement: 0.03255 +[6] (%Time%) 11 iterations Loss: 0.215291574597359 Improvement: 0.01493 +[6] (%Time%) 12 iterations Loss: 0.212821274995804 Improvement: 0.005586 +[6] (%Time%) 13 iterations Loss: 0.212086588144302 Improvement: 0.001948 +[6] (%Time%) 14 iterations Loss: 0.21061946451664 Improvement: 0.001587 +[6] (%Time%) 15 iterations Loss: 0.209799557924271 Improvement: 0.001012 +[6] (%Time%) 16 iterations Loss: 0.209267094731331 Improvement: 0.0006523 +[6] 'LBFGS Optimizer #2' finished in %Time%. diff --git a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-rp.txt b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-rp.txt index 185123a3b5..2e823ace9e 100644 --- a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-rp.txt +++ b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-rp.txt @@ -1,4 +1,4 @@ MulticlassLogisticRegression Accuracy(micro-avg) Accuracy(macro-avg) Log-loss Log-loss reduction /l2 /l1 /ot /nt /nn Learner Name Train Dataset Test Dataset Results File Run Time Physical Memory Virtual Memory Command Line Settings -0.966928 0.965873 0.12771 88.24678 0.1 0.001 0.001 1 + MulticlassLogisticRegression %Data% %Output% 99 0 0 maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} threads=- norm=No dout=%Output% data=%Data% seed=1 /l2:0.1;/l1:0.001;/ot:0.001;/nt:1;/nn:+ +0.966928 0.965873 0.12771 88.24678 0.1 0.001 0.001 1 + MulticlassLogisticRegression %Data% %Output% 99 0 0 maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} threads=- norm=No dout=%Output% data=%Data% seed=1 xf=Term{col=Label} /l2:0.1;/l1:0.001;/ot:0.001;/nt:1;/nn:+ diff --git a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-out.txt b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-out.txt index 8dbe1d90e5..9f83acd9f2 100644 --- a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-out.txt +++ b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-out.txt @@ -1,4 +1,4 @@ -maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} norm=No dout=%Output% data=%Data% out=%Output% seed=1 +maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} norm=No dout=%Output% data=%Data% out=%Output% seed=1 xf=Term{col=Label} Not adding a normalizer. Beginning optimization num vars: 15 @@ -33,33 +33,36 @@ Virtual memory usage(MB): %Number% %DateTime% Time elapsed(s): %Number% --- Progress log --- -[1] 'LBFGS data prep' started. -[1] 'LBFGS data prep' finished in %Time%. -[2] 'LBFGS Optimizer' started. -[2] (%Time%) 0 iterations Loss: 1.0986123085022 -[2] (%Time%) 1 iterations Loss: 1.06389963626862 Improvement: 0.03471 -[2] (%Time%) 2 iterations Loss: 1.01654124259949 Improvement: 0.04483 -[2] (%Time%) 3 iterations Loss: 0.944314062595367 Improvement: 0.0657 -[2] (%Time%) 4 iterations Loss: 0.668209552764893 Improvement: 0.2241 -[2] (%Time%) 5 iterations Loss: 0.553279459476471 Improvement: 0.1421 -[2] (%Time%) 6 iterations Loss: 0.427209556102753 Improvement: 0.1301 -[2] (%Time%) 7 iterations Loss: 0.33543187379837 Improvement: 0.1014 -[2] (%Time%) 8 iterations Loss: 0.271388441324234 Improvement: 0.07337 -[2] (%Time%) 9 iterations Loss: 0.218755051493645 Improvement: 0.05782 -[2] (%Time%) 10 iterations Loss: 0.192830204963684 Improvement: 0.0339 -[2] (%Time%) 11 iterations Loss: 0.184821993112564 Improvement: 0.01448 -[2] (%Time%) 12 iterations Loss: 0.182577073574066 Improvement: 0.005304 -[2] (%Time%) 13 iterations Loss: 0.180941790342331 Improvement: 0.002552 -[2] (%Time%) 14 iterations Loss: 0.178911954164505 Improvement: 0.00216 -[2] (%Time%) 15 iterations Loss: 0.171350136399269 Improvement: 0.006211 -[2] (%Time%) 16 iterations Loss: 0.157612159848213 Improvement: 0.01186 -[2] (%Time%) 17 iterations Loss: 0.15358293056488 Improvement: 0.005986 -[2] (%Time%) 18 iterations Loss: 0.151476576924324 Improvement: 0.003076 -[2] (%Time%) 19 iterations Loss: 0.146950766444206 Improvement: 0.004163 -[2] (%Time%) 20 iterations Loss: 0.143808200955391 Improvement: 0.003398 -[2] (%Time%) 21 iterations Loss: 0.141508430242538 Improvement: 0.002574 -[2] (%Time%) 22 iterations Loss: 0.140696823596954 Improvement: 0.001252 -[2] (%Time%) 23 iterations Loss: 0.140071913599968 Improvement: 0.0007818 -[2] 'LBFGS Optimizer' finished in %Time%. -[3] 'Saving model' started. -[3] 'Saving model' finished in %Time%. +[1] 'Building term dictionary' started. +[1] (%Time%) 150 examples Total Terms: 3 +[1] 'Building term dictionary' finished in %Time%. +[2] 'LBFGS data prep' started. +[2] 'LBFGS data prep' finished in %Time%. +[3] 'LBFGS Optimizer' started. +[3] (%Time%) 0 iterations Loss: 1.0986123085022 +[3] (%Time%) 1 iterations Loss: 1.06389963626862 Improvement: 0.03471 +[3] (%Time%) 2 iterations Loss: 1.01654124259949 Improvement: 0.04483 +[3] (%Time%) 3 iterations Loss: 0.944314062595367 Improvement: 0.0657 +[3] (%Time%) 4 iterations Loss: 0.668209552764893 Improvement: 0.2241 +[3] (%Time%) 5 iterations Loss: 0.553279459476471 Improvement: 0.1421 +[3] (%Time%) 6 iterations Loss: 0.427209556102753 Improvement: 0.1301 +[3] (%Time%) 7 iterations Loss: 0.33543187379837 Improvement: 0.1014 +[3] (%Time%) 8 iterations Loss: 0.271388441324234 Improvement: 0.07337 +[3] (%Time%) 9 iterations Loss: 0.218755051493645 Improvement: 0.05782 +[3] (%Time%) 10 iterations Loss: 0.192830204963684 Improvement: 0.0339 +[3] (%Time%) 11 iterations Loss: 0.184821993112564 Improvement: 0.01448 +[3] (%Time%) 12 iterations Loss: 0.182577073574066 Improvement: 0.005304 +[3] (%Time%) 13 iterations Loss: 0.180941790342331 Improvement: 0.002552 +[3] (%Time%) 14 iterations Loss: 0.178911954164505 Improvement: 0.00216 +[3] (%Time%) 15 iterations Loss: 0.171350136399269 Improvement: 0.006211 +[3] (%Time%) 16 iterations Loss: 0.157612159848213 Improvement: 0.01186 +[3] (%Time%) 17 iterations Loss: 0.15358293056488 Improvement: 0.005986 +[3] (%Time%) 18 iterations Loss: 0.151476576924324 Improvement: 0.003076 +[3] (%Time%) 19 iterations Loss: 0.146950766444206 Improvement: 0.004163 +[3] (%Time%) 20 iterations Loss: 0.143808200955391 Improvement: 0.003398 +[3] (%Time%) 21 iterations Loss: 0.141508430242538 Improvement: 0.002574 +[3] (%Time%) 22 iterations Loss: 0.140696823596954 Improvement: 0.001252 +[3] (%Time%) 23 iterations Loss: 0.140071913599968 Improvement: 0.0007818 +[3] 'LBFGS Optimizer' finished in %Time%. +[4] 'Saving model' started. +[4] 'Saving model' finished in %Time%. diff --git a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-rp.txt b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-rp.txt index b6f5c0b23a..3632e10049 100644 --- a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-rp.txt +++ b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-rp.txt @@ -1,4 +1,4 @@ MulticlassLogisticRegression Accuracy(micro-avg) Accuracy(macro-avg) Log-loss Log-loss reduction /l2 /l1 /ot /nt /nn Learner Name Train Dataset Test Dataset Results File Run Time Physical Memory Virtual Memory Command Line Settings -0.98 0.98 0.095534 91.30415 0.1 0.001 0.001 1 + MulticlassLogisticRegression %Data% %Data% %Output% 99 0 0 maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} norm=No dout=%Output% data=%Data% out=%Output% seed=1 /l2:0.1;/l1:0.001;/ot:0.001;/nt:1;/nn:+ +0.98 0.98 0.095534 91.30415 0.1 0.001 0.001 1 + MulticlassLogisticRegression %Data% %Data% %Output% 99 0 0 maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} norm=No dout=%Output% data=%Data% out=%Output% seed=1 xf=Term{col=Label} /l2:0.1;/l1:0.001;/ot:0.001;/nt:1;/nn:+ diff --git a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-out.txt b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-out.txt index 9a7765db33..c4b420c01e 100644 --- a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-out.txt +++ b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-out.txt @@ -1,4 +1,4 @@ -maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} threads=- norm=No dout=%Output% data=%Data% seed=1 +maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} threads=- norm=No dout=%Output% data=%Data% seed=1 xf=Term{col=Label} Not adding a normalizer. Beginning optimization num vars: 15 @@ -53,53 +53,59 @@ Virtual memory usage(MB): %Number% %DateTime% Time elapsed(s): %Number% --- Progress log --- -[1] 'LBFGS data prep' started. -[1] 'LBFGS data prep' finished in %Time%. -[2] 'LBFGS Optimizer' started. -[2] (%Time%) 0 iterations Loss: 1.0986123085022 -[2] (%Time%) 1 iterations Loss: 0.975501239299774 Improvement: 0.1231 -[2] (%Time%) 2 iterations Loss: 0.828468441963196 Improvement: 0.1422 -[2] (%Time%) 3 iterations Loss: 0.49238583445549 Improvement: 0.2899 -[2] (%Time%) 4 iterations Loss: 0.410263001918793 Improvement: 0.1335 -[2] (%Time%) 5 iterations Loss: 0.373202115297318 Improvement: 0.06109 -[2] (%Time%) 6 iterations Loss: 0.326229214668274 Improvement: 0.0505 -[2] (%Time%) 7 iterations Loss: 0.30860298871994 Improvement: 0.02584 -[2] (%Time%) 8 iterations Loss: 0.249911725521088 Improvement: 0.05048 -[2] (%Time%) 9 iterations Loss: 0.197030156850815 Improvement: 0.05228 -[2] (%Time%) 10 iterations Loss: 0.183768630027771 Improvement: 0.02302 -[2] (%Time%) 11 iterations Loss: 0.174268662929535 Improvement: 0.01288 -[2] (%Time%) 12 iterations Loss: 0.1489098072052 Improvement: 0.02224 -[2] (%Time%) 13 iterations Loss: 0.146679118275642 Improvement: 0.007233 -[2] (%Time%) 14 iterations Loss: 0.127629071474075 Improvement: 0.0161 -[2] (%Time%) 15 iterations Loss: 0.127402290701866 Improvement: 0.004194 -[2] (%Time%) 16 iterations Loss: 0.127095967531204 Improvement: 0.001278 -[2] (%Time%) 17 iterations Loss: 0.1268040984869 Improvement: 0.0005385 -[2] 'LBFGS Optimizer' finished in %Time%. -[3] 'LBFGS data prep #2' started. -[3] 'LBFGS data prep #2' finished in %Time%. -[4] 'LBFGS Optimizer #2' started. -[4] (%Time%) 0 iterations Loss: 1.0986123085022 -[4] (%Time%) 1 iterations Loss: 1.03655636310577 Improvement: 0.06206 -[4] (%Time%) 2 iterations Loss: 1.00361847877502 Improvement: 0.03876 -[4] (%Time%) 3 iterations Loss: 0.937079250812531 Improvement: 0.05993 -[4] (%Time%) 4 iterations Loss: 0.819244384765625 Improvement: 0.1035 -[4] (%Time%) 5 iterations Loss: 0.728321373462677 Improvement: 0.09406 -[4] (%Time%) 6 iterations Loss: 0.581992864608765 Improvement: 0.1333 -[4] (%Time%) 7 iterations Loss: 0.440624892711639 Improvement: 0.1393 -[4] (%Time%) 8 iterations Loss: 0.368180394172668 Improvement: 0.08917 -[4] (%Time%) 9 iterations Loss: 0.287548065185547 Improvement: 0.08277 -[4] (%Time%) 10 iterations Loss: 0.239883854985237 Improvement: 0.05644 -[4] (%Time%) 11 iterations Loss: 0.217700272798538 Improvement: 0.03075 -[4] (%Time%) 12 iterations Loss: 0.206228733062744 Improvement: 0.01629 -[4] (%Time%) 13 iterations Loss: 0.192829161882401 Improvement: 0.01412 -[4] (%Time%) 14 iterations Loss: 0.185032933950424 Improvement: 0.009378 -[4] (%Time%) 15 iterations Loss: 0.181731522083282 Improvement: 0.00482 -[4] (%Time%) 16 iterations Loss: 0.168401405215263 Improvement: 0.0112 -[4] (%Time%) 17 iterations Loss: 0.159209698438644 Improvement: 0.009694 -[4] (%Time%) 18 iterations Loss: 0.150576055049896 Improvement: 0.008899 -[4] (%Time%) 19 iterations Loss: 0.14181961119175 Improvement: 0.008792 -[4] (%Time%) 20 iterations Loss: 0.135607719421387 Improvement: 0.006857 -[4] (%Time%) 21 iterations Loss: 0.134872287511826 Improvement: 0.002266 -[4] (%Time%) 22 iterations Loss: 0.133358553051949 Improvement: 0.001702 -[4] (%Time%) 23 iterations Loss: 0.132842555642128 Improvement: 0.0008124 -[4] 'LBFGS Optimizer #2' finished in %Time%. +[1] 'Building term dictionary' started. +[1] (%Time%) 71 examples Total Terms: 3 +[1] 'Building term dictionary' finished in %Time%. +[2] 'LBFGS data prep' started. +[2] 'LBFGS data prep' finished in %Time%. +[3] 'LBFGS Optimizer' started. +[3] (%Time%) 0 iterations Loss: 1.0986123085022 +[3] (%Time%) 1 iterations Loss: 0.975501239299774 Improvement: 0.1231 +[3] (%Time%) 2 iterations Loss: 0.828468441963196 Improvement: 0.1422 +[3] (%Time%) 3 iterations Loss: 0.49238583445549 Improvement: 0.2899 +[3] (%Time%) 4 iterations Loss: 0.410263001918793 Improvement: 0.1335 +[3] (%Time%) 5 iterations Loss: 0.373202115297318 Improvement: 0.06109 +[3] (%Time%) 6 iterations Loss: 0.326229214668274 Improvement: 0.0505 +[3] (%Time%) 7 iterations Loss: 0.30860298871994 Improvement: 0.02584 +[3] (%Time%) 8 iterations Loss: 0.249911725521088 Improvement: 0.05048 +[3] (%Time%) 9 iterations Loss: 0.197030156850815 Improvement: 0.05228 +[3] (%Time%) 10 iterations Loss: 0.183768630027771 Improvement: 0.02302 +[3] (%Time%) 11 iterations Loss: 0.174268662929535 Improvement: 0.01288 +[3] (%Time%) 12 iterations Loss: 0.1489098072052 Improvement: 0.02224 +[3] (%Time%) 13 iterations Loss: 0.146679118275642 Improvement: 0.007233 +[3] (%Time%) 14 iterations Loss: 0.127629071474075 Improvement: 0.0161 +[3] (%Time%) 15 iterations Loss: 0.127402290701866 Improvement: 0.004194 +[3] (%Time%) 16 iterations Loss: 0.127095967531204 Improvement: 0.001278 +[3] (%Time%) 17 iterations Loss: 0.1268040984869 Improvement: 0.0005385 +[3] 'LBFGS Optimizer' finished in %Time%. +[4] 'Building term dictionary #2' started. +[4] (%Time%) 79 examples Total Terms: 3 +[4] 'Building term dictionary #2' finished in %Time%. +[5] 'LBFGS data prep #2' started. +[5] 'LBFGS data prep #2' finished in %Time%. +[6] 'LBFGS Optimizer #2' started. +[6] (%Time%) 0 iterations Loss: 1.0986123085022 +[6] (%Time%) 1 iterations Loss: 1.03655636310577 Improvement: 0.06206 +[6] (%Time%) 2 iterations Loss: 1.00361847877502 Improvement: 0.03876 +[6] (%Time%) 3 iterations Loss: 0.937079250812531 Improvement: 0.05993 +[6] (%Time%) 4 iterations Loss: 0.819244384765625 Improvement: 0.1035 +[6] (%Time%) 5 iterations Loss: 0.728321373462677 Improvement: 0.09406 +[6] (%Time%) 6 iterations Loss: 0.581992864608765 Improvement: 0.1333 +[6] (%Time%) 7 iterations Loss: 0.440624892711639 Improvement: 0.1393 +[6] (%Time%) 8 iterations Loss: 0.368180394172668 Improvement: 0.08917 +[6] (%Time%) 9 iterations Loss: 0.287548065185547 Improvement: 0.08277 +[6] (%Time%) 10 iterations Loss: 0.239883854985237 Improvement: 0.05644 +[6] (%Time%) 11 iterations Loss: 0.217700272798538 Improvement: 0.03075 +[6] (%Time%) 12 iterations Loss: 0.206228733062744 Improvement: 0.01629 +[6] (%Time%) 13 iterations Loss: 0.192829161882401 Improvement: 0.01412 +[6] (%Time%) 14 iterations Loss: 0.185032933950424 Improvement: 0.009378 +[6] (%Time%) 15 iterations Loss: 0.181731522083282 Improvement: 0.00482 +[6] (%Time%) 16 iterations Loss: 0.168401405215263 Improvement: 0.0112 +[6] (%Time%) 17 iterations Loss: 0.159209698438644 Improvement: 0.009694 +[6] (%Time%) 18 iterations Loss: 0.150576055049896 Improvement: 0.008899 +[6] (%Time%) 19 iterations Loss: 0.14181961119175 Improvement: 0.008792 +[6] (%Time%) 20 iterations Loss: 0.135607719421387 Improvement: 0.006857 +[6] (%Time%) 21 iterations Loss: 0.134872287511826 Improvement: 0.002266 +[6] (%Time%) 22 iterations Loss: 0.133358553051949 Improvement: 0.001702 +[6] (%Time%) 23 iterations Loss: 0.132842555642128 Improvement: 0.0008124 +[6] 'LBFGS Optimizer #2' finished in %Time%. diff --git a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-rp.txt b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-rp.txt index a4dc51d9a7..fa88103c95 100644 --- a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-rp.txt +++ b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-rp.txt @@ -1,4 +1,4 @@ MulticlassLogisticRegression Accuracy(micro-avg) Accuracy(macro-avg) Log-loss Log-loss reduction /l2 /l1 /ot /nt Learner Name Train Dataset Test Dataset Results File Run Time Physical Memory Virtual Memory Command Line Settings -0.97397 0.974206 0.088839 91.82503 0.1 0.001 0.001 1 MulticlassLogisticRegression %Data% %Output% 99 0 0 maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} threads=- norm=No dout=%Output% data=%Data% seed=1 /l2:0.1;/l1:0.001;/ot:0.001;/nt:1 +0.97397 0.974206 0.088839 91.82503 0.1 0.001 0.001 1 MulticlassLogisticRegression %Data% %Output% 99 0 0 maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} threads=- norm=No dout=%Output% data=%Data% seed=1 xf=Term{col=Label} /l2:0.1;/l1:0.001;/ot:0.001;/nt:1 diff --git a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-out.txt b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-out.txt index 0e64ac225a..a48f89b011 100644 --- a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-out.txt +++ b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-out.txt @@ -1,4 +1,4 @@ -maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} norm=No dout=%Output% data=%Data% out=%Output% seed=1 +maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} norm=No dout=%Output% data=%Data% out=%Output% seed=1 xf=Term{col=Label} Not adding a normalizer. Beginning optimization num vars: 15 @@ -33,34 +33,37 @@ Virtual memory usage(MB): %Number% %DateTime% Time elapsed(s): %Number% --- Progress log --- -[1] 'LBFGS data prep' started. -[1] 'LBFGS data prep' finished in %Time%. -[2] 'LBFGS Optimizer' started. -[2] (%Time%) 0 iterations Loss: 1.0986123085022 -[2] (%Time%) 1 iterations Loss: 1.09053671360016 Improvement: 0.008076 -[2] (%Time%) 2 iterations Loss: 0.964357972145081 Improvement: 0.1026 -[2] (%Time%) 3 iterations Loss: 0.874466478824615 Improvement: 0.09291 -[2] (%Time%) 4 iterations Loss: 0.53207802772522 Improvement: 0.2808 -[2] (%Time%) 5 iterations Loss: 0.460592895746231 Improvement: 0.1236 -[2] (%Time%) 6 iterations Loss: 0.381620526313782 Improvement: 0.09013 -[2] (%Time%) 7 iterations Loss: 0.301508545875549 Improvement: 0.08262 -[2] (%Time%) 8 iterations Loss: 0.230116382241249 Improvement: 0.0742 -[2] (%Time%) 9 iterations Loss: 0.170902773737907 Improvement: 0.06296 -[2] (%Time%) 10 iterations Loss: 0.143164187669754 Improvement: 0.03654 -[2] (%Time%) 11 iterations Loss: 0.135387286543846 Improvement: 0.01497 -[2] (%Time%) 12 iterations Loss: 0.133318409323692 Improvement: 0.005294 -[2] (%Time%) 13 iterations Loss: 0.132491216063499 Improvement: 0.001944 -[2] (%Time%) 14 iterations Loss: 0.124604761600494 Improvement: 0.006401 -[2] (%Time%) 15 iterations Loss: 0.120595537126064 Improvement: 0.004607 -[2] (%Time%) 16 iterations Loss: 0.119206272065639 Improvement: 0.002194 -[2] (%Time%) 17 iterations Loss: 0.117203310132027 Improvement: 0.002051 -[2] (%Time%) 18 iterations Loss: 0.116163291037083 Improvement: 0.001293 -[2] (%Time%) 19 iterations Loss: 0.109811097383499 Improvement: 0.005087 -[2] (%Time%) 20 iterations Loss: 0.106156274676323 Improvement: 0.004013 -[2] (%Time%) 21 iterations Loss: 0.104246392846107 Improvement: 0.002436 -[2] (%Time%) 22 iterations Loss: 0.10310410708189 Improvement: 0.001466 -[2] (%Time%) 23 iterations Loss: 0.102218925952911 Improvement: 0.00103 -[2] (%Time%) 24 iterations Loss: 0.101610459387302 Improvement: 0.0007139 -[2] 'LBFGS Optimizer' finished in %Time%. -[3] 'Saving model' started. -[3] 'Saving model' finished in %Time%. +[1] 'Building term dictionary' started. +[1] (%Time%) 150 examples Total Terms: 3 +[1] 'Building term dictionary' finished in %Time%. +[2] 'LBFGS data prep' started. +[2] 'LBFGS data prep' finished in %Time%. +[3] 'LBFGS Optimizer' started. +[3] (%Time%) 0 iterations Loss: 1.0986123085022 +[3] (%Time%) 1 iterations Loss: 1.09053671360016 Improvement: 0.008076 +[3] (%Time%) 2 iterations Loss: 0.964357972145081 Improvement: 0.1026 +[3] (%Time%) 3 iterations Loss: 0.874466478824615 Improvement: 0.09291 +[3] (%Time%) 4 iterations Loss: 0.53207802772522 Improvement: 0.2808 +[3] (%Time%) 5 iterations Loss: 0.460592895746231 Improvement: 0.1236 +[3] (%Time%) 6 iterations Loss: 0.381620526313782 Improvement: 0.09013 +[3] (%Time%) 7 iterations Loss: 0.301508545875549 Improvement: 0.08262 +[3] (%Time%) 8 iterations Loss: 0.230116382241249 Improvement: 0.0742 +[3] (%Time%) 9 iterations Loss: 0.170902773737907 Improvement: 0.06296 +[3] (%Time%) 10 iterations Loss: 0.143164187669754 Improvement: 0.03654 +[3] (%Time%) 11 iterations Loss: 0.135387286543846 Improvement: 0.01497 +[3] (%Time%) 12 iterations Loss: 0.133318409323692 Improvement: 0.005294 +[3] (%Time%) 13 iterations Loss: 0.132491216063499 Improvement: 0.001944 +[3] (%Time%) 14 iterations Loss: 0.124604761600494 Improvement: 0.006401 +[3] (%Time%) 15 iterations Loss: 0.120595537126064 Improvement: 0.004607 +[3] (%Time%) 16 iterations Loss: 0.119206272065639 Improvement: 0.002194 +[3] (%Time%) 17 iterations Loss: 0.117203310132027 Improvement: 0.002051 +[3] (%Time%) 18 iterations Loss: 0.116163291037083 Improvement: 0.001293 +[3] (%Time%) 19 iterations Loss: 0.109811097383499 Improvement: 0.005087 +[3] (%Time%) 20 iterations Loss: 0.106156274676323 Improvement: 0.004013 +[3] (%Time%) 21 iterations Loss: 0.104246392846107 Improvement: 0.002436 +[3] (%Time%) 22 iterations Loss: 0.10310410708189 Improvement: 0.001466 +[3] (%Time%) 23 iterations Loss: 0.102218925952911 Improvement: 0.00103 +[3] (%Time%) 24 iterations Loss: 0.101610459387302 Improvement: 0.0007139 +[3] 'LBFGS Optimizer' finished in %Time%. +[4] 'Saving model' started. +[4] 'Saving model' finished in %Time%. diff --git a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-rp.txt b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-rp.txt index 02b121c96d..817c1490f6 100644 --- a/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-rp.txt +++ b/test/BaselineOutput/SingleDebug/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-rp.txt @@ -1,4 +1,4 @@ MulticlassLogisticRegression Accuracy(micro-avg) Accuracy(macro-avg) Log-loss Log-loss reduction /l2 /l1 /ot /nt Learner Name Train Dataset Test Dataset Results File Run Time Physical Memory Virtual Memory Command Line Settings -0.98 0.98 0.072218 93.42639 0.1 0.001 0.001 1 MulticlassLogisticRegression %Data% %Data% %Output% 99 0 0 maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} norm=No dout=%Output% data=%Data% out=%Output% seed=1 /l2:0.1;/l1:0.001;/ot:0.001;/nt:1 +0.98 0.98 0.072218 93.42639 0.1 0.001 0.001 1 MulticlassLogisticRegression %Data% %Data% %Output% 99 0 0 maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} norm=No dout=%Output% data=%Data% out=%Output% seed=1 xf=Term{col=Label} /l2:0.1;/l1:0.001;/ot:0.001;/nt:1 diff --git a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-out.txt b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-out.txt index 2dd90d6312..85a6761d32 100644 --- a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-out.txt +++ b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-out.txt @@ -1,4 +1,4 @@ -maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} threads=- norm=No dout=%Output% data=%Data% seed=1 +maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} threads=- norm=No dout=%Output% data=%Data% seed=1 xf=Term{col=Label} Not adding a normalizer. Beginning optimization num vars: 15 @@ -53,50 +53,56 @@ Virtual memory usage(MB): %Number% %DateTime% Time elapsed(s): %Number% --- Progress log --- -[1] 'LBFGS data prep' started. -[1] 'LBFGS data prep' finished in %Time%. -[2] 'LBFGS Optimizer' started. -[2] (%Time%) 0 iterations Loss: 1.0986123085022 -[2] (%Time%) 1 iterations Loss: 1.00646448135376 Improvement: 0.09215 -[2] (%Time%) 2 iterations Loss: 0.909583747386932 Improvement: 0.09593 -[2] (%Time%) 3 iterations Loss: 0.525106191635132 Improvement: 0.3158 -[2] (%Time%) 4 iterations Loss: 0.400520384311676 Improvement: 0.1718 -[2] (%Time%) 5 iterations Loss: 0.332601189613342 Improvement: 0.09382 -[2] (%Time%) 6 iterations Loss: 0.281388521194458 Improvement: 0.06186 -[2] (%Time%) 7 iterations Loss: 0.237996473908424 Improvement: 0.04801 -[2] (%Time%) 8 iterations Loss: 0.212298363447189 Improvement: 0.03128 -[2] (%Time%) 9 iterations Loss: 0.199792444705963 Improvement: 0.0172 -[2] (%Time%) 10 iterations Loss: 0.194789424538612 Improvement: 0.008052 -[2] (%Time%) 11 iterations Loss: 0.193230450153351 Improvement: 0.003182 -[2] (%Time%) 12 iterations Loss: 0.192447692155838 Improvement: 0.001383 -[2] (%Time%) 13 iterations Loss: 0.189304739236832 Improvement: 0.002703 -[2] (%Time%) 14 iterations Loss: 0.187662661075592 Improvement: 0.001907 -[2] (%Time%) 15 iterations Loss: 0.185374572873116 Improvement: 0.002193 -[2] (%Time%) 16 iterations Loss: 0.18364554643631 Improvement: 0.001845 -[2] (%Time%) 17 iterations Loss: 0.180794909596443 Improvement: 0.002599 -[2] (%Time%) 18 iterations Loss: 0.178908497095108 Improvement: 0.002065 -[2] (%Time%) 19 iterations Loss: 0.175620675086975 Improvement: 0.002982 -[2] (%Time%) 20 iterations Loss: 0.174758642911911 Improvement: 0.001392 -[2] (%Time%) 21 iterations Loss: 0.173962101340294 Improvement: 0.0009454 -[2] 'LBFGS Optimizer' finished in %Time%. -[3] 'LBFGS data prep #2' started. -[3] 'LBFGS data prep #2' finished in %Time%. -[4] 'LBFGS Optimizer #2' started. -[4] (%Time%) 0 iterations Loss: 1.0986123085022 -[4] (%Time%) 1 iterations Loss: 1.05856335163116 Improvement: 0.04005 -[4] (%Time%) 2 iterations Loss: 1.00281620025635 Improvement: 0.05261 -[4] (%Time%) 3 iterations Loss: 0.97780430316925 Improvement: 0.03158 -[4] (%Time%) 4 iterations Loss: 0.752716302871704 Improvement: 0.1773 -[4] (%Time%) 5 iterations Loss: 0.542387366294861 Improvement: 0.2021 -[4] (%Time%) 6 iterations Loss: 0.443084180355072 Improvement: 0.125 -[4] (%Time%) 7 iterations Loss: 0.343867212533951 Improvement: 0.1057 -[4] (%Time%) 8 iterations Loss: 0.284590691328049 Improvement: 0.07087 -[4] (%Time%) 9 iterations Loss: 0.254261910915375 Improvement: 0.04046 -[4] (%Time%) 10 iterations Loss: 0.224356189370155 Improvement: 0.03255 -[4] (%Time%) 11 iterations Loss: 0.215291574597359 Improvement: 0.01493 -[4] (%Time%) 12 iterations Loss: 0.212821274995804 Improvement: 0.005586 -[4] (%Time%) 13 iterations Loss: 0.212086588144302 Improvement: 0.001948 -[4] (%Time%) 14 iterations Loss: 0.21061946451664 Improvement: 0.001587 -[4] (%Time%) 15 iterations Loss: 0.209799557924271 Improvement: 0.001012 -[4] (%Time%) 16 iterations Loss: 0.209267094731331 Improvement: 0.0006523 -[4] 'LBFGS Optimizer #2' finished in %Time%. +[1] 'Building term dictionary' started. +[1] (%Time%) 71 examples Total Terms: 3 +[1] 'Building term dictionary' finished in %Time%. +[2] 'LBFGS data prep' started. +[2] 'LBFGS data prep' finished in %Time%. +[3] 'LBFGS Optimizer' started. +[3] (%Time%) 0 iterations Loss: 1.0986123085022 +[3] (%Time%) 1 iterations Loss: 1.00646448135376 Improvement: 0.09215 +[3] (%Time%) 2 iterations Loss: 0.909583747386932 Improvement: 0.09593 +[3] (%Time%) 3 iterations Loss: 0.525106191635132 Improvement: 0.3158 +[3] (%Time%) 4 iterations Loss: 0.400520384311676 Improvement: 0.1718 +[3] (%Time%) 5 iterations Loss: 0.332601189613342 Improvement: 0.09382 +[3] (%Time%) 6 iterations Loss: 0.281388521194458 Improvement: 0.06186 +[3] (%Time%) 7 iterations Loss: 0.237996473908424 Improvement: 0.04801 +[3] (%Time%) 8 iterations Loss: 0.212298363447189 Improvement: 0.03128 +[3] (%Time%) 9 iterations Loss: 0.199792444705963 Improvement: 0.0172 +[3] (%Time%) 10 iterations Loss: 0.194789424538612 Improvement: 0.008052 +[3] (%Time%) 11 iterations Loss: 0.193230450153351 Improvement: 0.003182 +[3] (%Time%) 12 iterations Loss: 0.192447692155838 Improvement: 0.001383 +[3] (%Time%) 13 iterations Loss: 0.189304739236832 Improvement: 0.002703 +[3] (%Time%) 14 iterations Loss: 0.187662661075592 Improvement: 0.001907 +[3] (%Time%) 15 iterations Loss: 0.185374572873116 Improvement: 0.002193 +[3] (%Time%) 16 iterations Loss: 0.18364554643631 Improvement: 0.001845 +[3] (%Time%) 17 iterations Loss: 0.180794909596443 Improvement: 0.002599 +[3] (%Time%) 18 iterations Loss: 0.178908497095108 Improvement: 0.002065 +[3] (%Time%) 19 iterations Loss: 0.175620675086975 Improvement: 0.002982 +[3] (%Time%) 20 iterations Loss: 0.174758642911911 Improvement: 0.001392 +[3] (%Time%) 21 iterations Loss: 0.173962101340294 Improvement: 0.0009454 +[3] 'LBFGS Optimizer' finished in %Time%. +[4] 'Building term dictionary #2' started. +[4] (%Time%) 79 examples Total Terms: 3 +[4] 'Building term dictionary #2' finished in %Time%. +[5] 'LBFGS data prep #2' started. +[5] 'LBFGS data prep #2' finished in %Time%. +[6] 'LBFGS Optimizer #2' started. +[6] (%Time%) 0 iterations Loss: 1.0986123085022 +[6] (%Time%) 1 iterations Loss: 1.05856335163116 Improvement: 0.04005 +[6] (%Time%) 2 iterations Loss: 1.00281620025635 Improvement: 0.05261 +[6] (%Time%) 3 iterations Loss: 0.97780430316925 Improvement: 0.03158 +[6] (%Time%) 4 iterations Loss: 0.752716302871704 Improvement: 0.1773 +[6] (%Time%) 5 iterations Loss: 0.542387366294861 Improvement: 0.2021 +[6] (%Time%) 6 iterations Loss: 0.443084180355072 Improvement: 0.125 +[6] (%Time%) 7 iterations Loss: 0.343867212533951 Improvement: 0.1057 +[6] (%Time%) 8 iterations Loss: 0.284590691328049 Improvement: 0.07087 +[6] (%Time%) 9 iterations Loss: 0.254261910915375 Improvement: 0.04046 +[6] (%Time%) 10 iterations Loss: 0.224356189370155 Improvement: 0.03255 +[6] (%Time%) 11 iterations Loss: 0.215291574597359 Improvement: 0.01493 +[6] (%Time%) 12 iterations Loss: 0.212821274995804 Improvement: 0.005586 +[6] (%Time%) 13 iterations Loss: 0.212086588144302 Improvement: 0.001948 +[6] (%Time%) 14 iterations Loss: 0.21061946451664 Improvement: 0.001587 +[6] (%Time%) 15 iterations Loss: 0.209799557924271 Improvement: 0.001012 +[6] (%Time%) 16 iterations Loss: 0.209267094731331 Improvement: 0.0006523 +[6] 'LBFGS Optimizer #2' finished in %Time%. diff --git a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-rp.txt b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-rp.txt index 185123a3b5..2e823ace9e 100644 --- a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-rp.txt +++ b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-CV-iris-rp.txt @@ -1,4 +1,4 @@ MulticlassLogisticRegression Accuracy(micro-avg) Accuracy(macro-avg) Log-loss Log-loss reduction /l2 /l1 /ot /nt /nn Learner Name Train Dataset Test Dataset Results File Run Time Physical Memory Virtual Memory Command Line Settings -0.966928 0.965873 0.12771 88.24678 0.1 0.001 0.001 1 + MulticlassLogisticRegression %Data% %Output% 99 0 0 maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} threads=- norm=No dout=%Output% data=%Data% seed=1 /l2:0.1;/l1:0.001;/ot:0.001;/nt:1;/nn:+ +0.966928 0.965873 0.12771 88.24678 0.1 0.001 0.001 1 + MulticlassLogisticRegression %Data% %Output% 99 0 0 maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} threads=- norm=No dout=%Output% data=%Data% seed=1 xf=Term{col=Label} /l2:0.1;/l1:0.001;/ot:0.001;/nt:1;/nn:+ diff --git a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-out.txt b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-out.txt index 8dbe1d90e5..9f83acd9f2 100644 --- a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-out.txt +++ b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-out.txt @@ -1,4 +1,4 @@ -maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} norm=No dout=%Output% data=%Data% out=%Output% seed=1 +maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} norm=No dout=%Output% data=%Data% out=%Output% seed=1 xf=Term{col=Label} Not adding a normalizer. Beginning optimization num vars: 15 @@ -33,33 +33,36 @@ Virtual memory usage(MB): %Number% %DateTime% Time elapsed(s): %Number% --- Progress log --- -[1] 'LBFGS data prep' started. -[1] 'LBFGS data prep' finished in %Time%. -[2] 'LBFGS Optimizer' started. -[2] (%Time%) 0 iterations Loss: 1.0986123085022 -[2] (%Time%) 1 iterations Loss: 1.06389963626862 Improvement: 0.03471 -[2] (%Time%) 2 iterations Loss: 1.01654124259949 Improvement: 0.04483 -[2] (%Time%) 3 iterations Loss: 0.944314062595367 Improvement: 0.0657 -[2] (%Time%) 4 iterations Loss: 0.668209552764893 Improvement: 0.2241 -[2] (%Time%) 5 iterations Loss: 0.553279459476471 Improvement: 0.1421 -[2] (%Time%) 6 iterations Loss: 0.427209556102753 Improvement: 0.1301 -[2] (%Time%) 7 iterations Loss: 0.33543187379837 Improvement: 0.1014 -[2] (%Time%) 8 iterations Loss: 0.271388441324234 Improvement: 0.07337 -[2] (%Time%) 9 iterations Loss: 0.218755051493645 Improvement: 0.05782 -[2] (%Time%) 10 iterations Loss: 0.192830204963684 Improvement: 0.0339 -[2] (%Time%) 11 iterations Loss: 0.184821993112564 Improvement: 0.01448 -[2] (%Time%) 12 iterations Loss: 0.182577073574066 Improvement: 0.005304 -[2] (%Time%) 13 iterations Loss: 0.180941790342331 Improvement: 0.002552 -[2] (%Time%) 14 iterations Loss: 0.178911954164505 Improvement: 0.00216 -[2] (%Time%) 15 iterations Loss: 0.171350136399269 Improvement: 0.006211 -[2] (%Time%) 16 iterations Loss: 0.157612159848213 Improvement: 0.01186 -[2] (%Time%) 17 iterations Loss: 0.15358293056488 Improvement: 0.005986 -[2] (%Time%) 18 iterations Loss: 0.151476576924324 Improvement: 0.003076 -[2] (%Time%) 19 iterations Loss: 0.146950766444206 Improvement: 0.004163 -[2] (%Time%) 20 iterations Loss: 0.143808200955391 Improvement: 0.003398 -[2] (%Time%) 21 iterations Loss: 0.141508430242538 Improvement: 0.002574 -[2] (%Time%) 22 iterations Loss: 0.140696823596954 Improvement: 0.001252 -[2] (%Time%) 23 iterations Loss: 0.140071913599968 Improvement: 0.0007818 -[2] 'LBFGS Optimizer' finished in %Time%. -[3] 'Saving model' started. -[3] 'Saving model' finished in %Time%. +[1] 'Building term dictionary' started. +[1] (%Time%) 150 examples Total Terms: 3 +[1] 'Building term dictionary' finished in %Time%. +[2] 'LBFGS data prep' started. +[2] 'LBFGS data prep' finished in %Time%. +[3] 'LBFGS Optimizer' started. +[3] (%Time%) 0 iterations Loss: 1.0986123085022 +[3] (%Time%) 1 iterations Loss: 1.06389963626862 Improvement: 0.03471 +[3] (%Time%) 2 iterations Loss: 1.01654124259949 Improvement: 0.04483 +[3] (%Time%) 3 iterations Loss: 0.944314062595367 Improvement: 0.0657 +[3] (%Time%) 4 iterations Loss: 0.668209552764893 Improvement: 0.2241 +[3] (%Time%) 5 iterations Loss: 0.553279459476471 Improvement: 0.1421 +[3] (%Time%) 6 iterations Loss: 0.427209556102753 Improvement: 0.1301 +[3] (%Time%) 7 iterations Loss: 0.33543187379837 Improvement: 0.1014 +[3] (%Time%) 8 iterations Loss: 0.271388441324234 Improvement: 0.07337 +[3] (%Time%) 9 iterations Loss: 0.218755051493645 Improvement: 0.05782 +[3] (%Time%) 10 iterations Loss: 0.192830204963684 Improvement: 0.0339 +[3] (%Time%) 11 iterations Loss: 0.184821993112564 Improvement: 0.01448 +[3] (%Time%) 12 iterations Loss: 0.182577073574066 Improvement: 0.005304 +[3] (%Time%) 13 iterations Loss: 0.180941790342331 Improvement: 0.002552 +[3] (%Time%) 14 iterations Loss: 0.178911954164505 Improvement: 0.00216 +[3] (%Time%) 15 iterations Loss: 0.171350136399269 Improvement: 0.006211 +[3] (%Time%) 16 iterations Loss: 0.157612159848213 Improvement: 0.01186 +[3] (%Time%) 17 iterations Loss: 0.15358293056488 Improvement: 0.005986 +[3] (%Time%) 18 iterations Loss: 0.151476576924324 Improvement: 0.003076 +[3] (%Time%) 19 iterations Loss: 0.146950766444206 Improvement: 0.004163 +[3] (%Time%) 20 iterations Loss: 0.143808200955391 Improvement: 0.003398 +[3] (%Time%) 21 iterations Loss: 0.141508430242538 Improvement: 0.002574 +[3] (%Time%) 22 iterations Loss: 0.140696823596954 Improvement: 0.001252 +[3] (%Time%) 23 iterations Loss: 0.140071913599968 Improvement: 0.0007818 +[3] 'LBFGS Optimizer' finished in %Time%. +[4] 'Saving model' started. +[4] 'Saving model' finished in %Time%. diff --git a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-rp.txt b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-rp.txt index b6f5c0b23a..3632e10049 100644 --- a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-rp.txt +++ b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/LogisticRegression-Non-Negative-TrainTest-iris-rp.txt @@ -1,4 +1,4 @@ MulticlassLogisticRegression Accuracy(micro-avg) Accuracy(macro-avg) Log-loss Log-loss reduction /l2 /l1 /ot /nt /nn Learner Name Train Dataset Test Dataset Results File Run Time Physical Memory Virtual Memory Command Line Settings -0.98 0.98 0.095534 91.30415 0.1 0.001 0.001 1 + MulticlassLogisticRegression %Data% %Data% %Output% 99 0 0 maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} norm=No dout=%Output% data=%Data% out=%Output% seed=1 /l2:0.1;/l1:0.001;/ot:0.001;/nt:1;/nn:+ +0.98 0.98 0.095534 91.30415 0.1 0.001 0.001 1 + MulticlassLogisticRegression %Data% %Data% %Output% 99 0 0 maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1 nn=+} norm=No dout=%Output% data=%Data% out=%Output% seed=1 xf=Term{col=Label} /l2:0.1;/l1:0.001;/ot:0.001;/nt:1;/nn:+ diff --git a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-out.txt b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-out.txt index 9a7765db33..c4b420c01e 100644 --- a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-out.txt +++ b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-out.txt @@ -1,4 +1,4 @@ -maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} threads=- norm=No dout=%Output% data=%Data% seed=1 +maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} threads=- norm=No dout=%Output% data=%Data% seed=1 xf=Term{col=Label} Not adding a normalizer. Beginning optimization num vars: 15 @@ -53,53 +53,59 @@ Virtual memory usage(MB): %Number% %DateTime% Time elapsed(s): %Number% --- Progress log --- -[1] 'LBFGS data prep' started. -[1] 'LBFGS data prep' finished in %Time%. -[2] 'LBFGS Optimizer' started. -[2] (%Time%) 0 iterations Loss: 1.0986123085022 -[2] (%Time%) 1 iterations Loss: 0.975501239299774 Improvement: 0.1231 -[2] (%Time%) 2 iterations Loss: 0.828468441963196 Improvement: 0.1422 -[2] (%Time%) 3 iterations Loss: 0.49238583445549 Improvement: 0.2899 -[2] (%Time%) 4 iterations Loss: 0.410263001918793 Improvement: 0.1335 -[2] (%Time%) 5 iterations Loss: 0.373202115297318 Improvement: 0.06109 -[2] (%Time%) 6 iterations Loss: 0.326229214668274 Improvement: 0.0505 -[2] (%Time%) 7 iterations Loss: 0.30860298871994 Improvement: 0.02584 -[2] (%Time%) 8 iterations Loss: 0.249911725521088 Improvement: 0.05048 -[2] (%Time%) 9 iterations Loss: 0.197030156850815 Improvement: 0.05228 -[2] (%Time%) 10 iterations Loss: 0.183768630027771 Improvement: 0.02302 -[2] (%Time%) 11 iterations Loss: 0.174268662929535 Improvement: 0.01288 -[2] (%Time%) 12 iterations Loss: 0.1489098072052 Improvement: 0.02224 -[2] (%Time%) 13 iterations Loss: 0.146679118275642 Improvement: 0.007233 -[2] (%Time%) 14 iterations Loss: 0.127629071474075 Improvement: 0.0161 -[2] (%Time%) 15 iterations Loss: 0.127402290701866 Improvement: 0.004194 -[2] (%Time%) 16 iterations Loss: 0.127095967531204 Improvement: 0.001278 -[2] (%Time%) 17 iterations Loss: 0.1268040984869 Improvement: 0.0005385 -[2] 'LBFGS Optimizer' finished in %Time%. -[3] 'LBFGS data prep #2' started. -[3] 'LBFGS data prep #2' finished in %Time%. -[4] 'LBFGS Optimizer #2' started. -[4] (%Time%) 0 iterations Loss: 1.0986123085022 -[4] (%Time%) 1 iterations Loss: 1.03655636310577 Improvement: 0.06206 -[4] (%Time%) 2 iterations Loss: 1.00361847877502 Improvement: 0.03876 -[4] (%Time%) 3 iterations Loss: 0.937079250812531 Improvement: 0.05993 -[4] (%Time%) 4 iterations Loss: 0.819244384765625 Improvement: 0.1035 -[4] (%Time%) 5 iterations Loss: 0.728321373462677 Improvement: 0.09406 -[4] (%Time%) 6 iterations Loss: 0.581992864608765 Improvement: 0.1333 -[4] (%Time%) 7 iterations Loss: 0.440624892711639 Improvement: 0.1393 -[4] (%Time%) 8 iterations Loss: 0.368180394172668 Improvement: 0.08917 -[4] (%Time%) 9 iterations Loss: 0.287548065185547 Improvement: 0.08277 -[4] (%Time%) 10 iterations Loss: 0.239883854985237 Improvement: 0.05644 -[4] (%Time%) 11 iterations Loss: 0.217700272798538 Improvement: 0.03075 -[4] (%Time%) 12 iterations Loss: 0.206228733062744 Improvement: 0.01629 -[4] (%Time%) 13 iterations Loss: 0.192829161882401 Improvement: 0.01412 -[4] (%Time%) 14 iterations Loss: 0.185032933950424 Improvement: 0.009378 -[4] (%Time%) 15 iterations Loss: 0.181731522083282 Improvement: 0.00482 -[4] (%Time%) 16 iterations Loss: 0.168401405215263 Improvement: 0.0112 -[4] (%Time%) 17 iterations Loss: 0.159209698438644 Improvement: 0.009694 -[4] (%Time%) 18 iterations Loss: 0.150576055049896 Improvement: 0.008899 -[4] (%Time%) 19 iterations Loss: 0.14181961119175 Improvement: 0.008792 -[4] (%Time%) 20 iterations Loss: 0.135607719421387 Improvement: 0.006857 -[4] (%Time%) 21 iterations Loss: 0.134872287511826 Improvement: 0.002266 -[4] (%Time%) 22 iterations Loss: 0.133358553051949 Improvement: 0.001702 -[4] (%Time%) 23 iterations Loss: 0.132842555642128 Improvement: 0.0008124 -[4] 'LBFGS Optimizer #2' finished in %Time%. +[1] 'Building term dictionary' started. +[1] (%Time%) 71 examples Total Terms: 3 +[1] 'Building term dictionary' finished in %Time%. +[2] 'LBFGS data prep' started. +[2] 'LBFGS data prep' finished in %Time%. +[3] 'LBFGS Optimizer' started. +[3] (%Time%) 0 iterations Loss: 1.0986123085022 +[3] (%Time%) 1 iterations Loss: 0.975501239299774 Improvement: 0.1231 +[3] (%Time%) 2 iterations Loss: 0.828468441963196 Improvement: 0.1422 +[3] (%Time%) 3 iterations Loss: 0.49238583445549 Improvement: 0.2899 +[3] (%Time%) 4 iterations Loss: 0.410263001918793 Improvement: 0.1335 +[3] (%Time%) 5 iterations Loss: 0.373202115297318 Improvement: 0.06109 +[3] (%Time%) 6 iterations Loss: 0.326229214668274 Improvement: 0.0505 +[3] (%Time%) 7 iterations Loss: 0.30860298871994 Improvement: 0.02584 +[3] (%Time%) 8 iterations Loss: 0.249911725521088 Improvement: 0.05048 +[3] (%Time%) 9 iterations Loss: 0.197030156850815 Improvement: 0.05228 +[3] (%Time%) 10 iterations Loss: 0.183768630027771 Improvement: 0.02302 +[3] (%Time%) 11 iterations Loss: 0.174268662929535 Improvement: 0.01288 +[3] (%Time%) 12 iterations Loss: 0.1489098072052 Improvement: 0.02224 +[3] (%Time%) 13 iterations Loss: 0.146679118275642 Improvement: 0.007233 +[3] (%Time%) 14 iterations Loss: 0.127629071474075 Improvement: 0.0161 +[3] (%Time%) 15 iterations Loss: 0.127402290701866 Improvement: 0.004194 +[3] (%Time%) 16 iterations Loss: 0.127095967531204 Improvement: 0.001278 +[3] (%Time%) 17 iterations Loss: 0.1268040984869 Improvement: 0.0005385 +[3] 'LBFGS Optimizer' finished in %Time%. +[4] 'Building term dictionary #2' started. +[4] (%Time%) 79 examples Total Terms: 3 +[4] 'Building term dictionary #2' finished in %Time%. +[5] 'LBFGS data prep #2' started. +[5] 'LBFGS data prep #2' finished in %Time%. +[6] 'LBFGS Optimizer #2' started. +[6] (%Time%) 0 iterations Loss: 1.0986123085022 +[6] (%Time%) 1 iterations Loss: 1.03655636310577 Improvement: 0.06206 +[6] (%Time%) 2 iterations Loss: 1.00361847877502 Improvement: 0.03876 +[6] (%Time%) 3 iterations Loss: 0.937079250812531 Improvement: 0.05993 +[6] (%Time%) 4 iterations Loss: 0.819244384765625 Improvement: 0.1035 +[6] (%Time%) 5 iterations Loss: 0.728321373462677 Improvement: 0.09406 +[6] (%Time%) 6 iterations Loss: 0.581992864608765 Improvement: 0.1333 +[6] (%Time%) 7 iterations Loss: 0.440624892711639 Improvement: 0.1393 +[6] (%Time%) 8 iterations Loss: 0.368180394172668 Improvement: 0.08917 +[6] (%Time%) 9 iterations Loss: 0.287548065185547 Improvement: 0.08277 +[6] (%Time%) 10 iterations Loss: 0.239883854985237 Improvement: 0.05644 +[6] (%Time%) 11 iterations Loss: 0.217700272798538 Improvement: 0.03075 +[6] (%Time%) 12 iterations Loss: 0.206228733062744 Improvement: 0.01629 +[6] (%Time%) 13 iterations Loss: 0.192829161882401 Improvement: 0.01412 +[6] (%Time%) 14 iterations Loss: 0.185032933950424 Improvement: 0.009378 +[6] (%Time%) 15 iterations Loss: 0.181731522083282 Improvement: 0.00482 +[6] (%Time%) 16 iterations Loss: 0.168401405215263 Improvement: 0.0112 +[6] (%Time%) 17 iterations Loss: 0.159209698438644 Improvement: 0.009694 +[6] (%Time%) 18 iterations Loss: 0.150576055049896 Improvement: 0.008899 +[6] (%Time%) 19 iterations Loss: 0.14181961119175 Improvement: 0.008792 +[6] (%Time%) 20 iterations Loss: 0.135607719421387 Improvement: 0.006857 +[6] (%Time%) 21 iterations Loss: 0.134872287511826 Improvement: 0.002266 +[6] (%Time%) 22 iterations Loss: 0.133358553051949 Improvement: 0.001702 +[6] (%Time%) 23 iterations Loss: 0.132842555642128 Improvement: 0.0008124 +[6] 'LBFGS Optimizer #2' finished in %Time%. diff --git a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-rp.txt b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-rp.txt index a4dc51d9a7..fa88103c95 100644 --- a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-rp.txt +++ b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-CV-iris-rp.txt @@ -1,4 +1,4 @@ MulticlassLogisticRegression Accuracy(micro-avg) Accuracy(macro-avg) Log-loss Log-loss reduction /l2 /l1 /ot /nt Learner Name Train Dataset Test Dataset Results File Run Time Physical Memory Virtual Memory Command Line Settings -0.97397 0.974206 0.088839 91.82503 0.1 0.001 0.001 1 MulticlassLogisticRegression %Data% %Output% 99 0 0 maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} threads=- norm=No dout=%Output% data=%Data% seed=1 /l2:0.1;/l1:0.001;/ot:0.001;/nt:1 +0.97397 0.974206 0.088839 91.82503 0.1 0.001 0.001 1 MulticlassLogisticRegression %Data% %Output% 99 0 0 maml.exe CV tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} threads=- norm=No dout=%Output% data=%Data% seed=1 xf=Term{col=Label} /l2:0.1;/l1:0.001;/ot:0.001;/nt:1 diff --git a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-out.txt b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-out.txt index 0e64ac225a..a48f89b011 100644 --- a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-out.txt +++ b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-out.txt @@ -1,4 +1,4 @@ -maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} norm=No dout=%Output% data=%Data% out=%Output% seed=1 +maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} norm=No dout=%Output% data=%Data% out=%Output% seed=1 xf=Term{col=Label} Not adding a normalizer. Beginning optimization num vars: 15 @@ -33,34 +33,37 @@ Virtual memory usage(MB): %Number% %DateTime% Time elapsed(s): %Number% --- Progress log --- -[1] 'LBFGS data prep' started. -[1] 'LBFGS data prep' finished in %Time%. -[2] 'LBFGS Optimizer' started. -[2] (%Time%) 0 iterations Loss: 1.0986123085022 -[2] (%Time%) 1 iterations Loss: 1.09053671360016 Improvement: 0.008076 -[2] (%Time%) 2 iterations Loss: 0.964357972145081 Improvement: 0.1026 -[2] (%Time%) 3 iterations Loss: 0.874466478824615 Improvement: 0.09291 -[2] (%Time%) 4 iterations Loss: 0.53207802772522 Improvement: 0.2808 -[2] (%Time%) 5 iterations Loss: 0.460592895746231 Improvement: 0.1236 -[2] (%Time%) 6 iterations Loss: 0.381620526313782 Improvement: 0.09013 -[2] (%Time%) 7 iterations Loss: 0.301508545875549 Improvement: 0.08262 -[2] (%Time%) 8 iterations Loss: 0.230116382241249 Improvement: 0.0742 -[2] (%Time%) 9 iterations Loss: 0.170902773737907 Improvement: 0.06296 -[2] (%Time%) 10 iterations Loss: 0.143164187669754 Improvement: 0.03654 -[2] (%Time%) 11 iterations Loss: 0.135387286543846 Improvement: 0.01497 -[2] (%Time%) 12 iterations Loss: 0.133318409323692 Improvement: 0.005294 -[2] (%Time%) 13 iterations Loss: 0.132491216063499 Improvement: 0.001944 -[2] (%Time%) 14 iterations Loss: 0.124604761600494 Improvement: 0.006401 -[2] (%Time%) 15 iterations Loss: 0.120595537126064 Improvement: 0.004607 -[2] (%Time%) 16 iterations Loss: 0.119206272065639 Improvement: 0.002194 -[2] (%Time%) 17 iterations Loss: 0.117203310132027 Improvement: 0.002051 -[2] (%Time%) 18 iterations Loss: 0.116163291037083 Improvement: 0.001293 -[2] (%Time%) 19 iterations Loss: 0.109811097383499 Improvement: 0.005087 -[2] (%Time%) 20 iterations Loss: 0.106156274676323 Improvement: 0.004013 -[2] (%Time%) 21 iterations Loss: 0.104246392846107 Improvement: 0.002436 -[2] (%Time%) 22 iterations Loss: 0.10310410708189 Improvement: 0.001466 -[2] (%Time%) 23 iterations Loss: 0.102218925952911 Improvement: 0.00103 -[2] (%Time%) 24 iterations Loss: 0.101610459387302 Improvement: 0.0007139 -[2] 'LBFGS Optimizer' finished in %Time%. -[3] 'Saving model' started. -[3] 'Saving model' finished in %Time%. +[1] 'Building term dictionary' started. +[1] (%Time%) 150 examples Total Terms: 3 +[1] 'Building term dictionary' finished in %Time%. +[2] 'LBFGS data prep' started. +[2] 'LBFGS data prep' finished in %Time%. +[3] 'LBFGS Optimizer' started. +[3] (%Time%) 0 iterations Loss: 1.0986123085022 +[3] (%Time%) 1 iterations Loss: 1.09053671360016 Improvement: 0.008076 +[3] (%Time%) 2 iterations Loss: 0.964357972145081 Improvement: 0.1026 +[3] (%Time%) 3 iterations Loss: 0.874466478824615 Improvement: 0.09291 +[3] (%Time%) 4 iterations Loss: 0.53207802772522 Improvement: 0.2808 +[3] (%Time%) 5 iterations Loss: 0.460592895746231 Improvement: 0.1236 +[3] (%Time%) 6 iterations Loss: 0.381620526313782 Improvement: 0.09013 +[3] (%Time%) 7 iterations Loss: 0.301508545875549 Improvement: 0.08262 +[3] (%Time%) 8 iterations Loss: 0.230116382241249 Improvement: 0.0742 +[3] (%Time%) 9 iterations Loss: 0.170902773737907 Improvement: 0.06296 +[3] (%Time%) 10 iterations Loss: 0.143164187669754 Improvement: 0.03654 +[3] (%Time%) 11 iterations Loss: 0.135387286543846 Improvement: 0.01497 +[3] (%Time%) 12 iterations Loss: 0.133318409323692 Improvement: 0.005294 +[3] (%Time%) 13 iterations Loss: 0.132491216063499 Improvement: 0.001944 +[3] (%Time%) 14 iterations Loss: 0.124604761600494 Improvement: 0.006401 +[3] (%Time%) 15 iterations Loss: 0.120595537126064 Improvement: 0.004607 +[3] (%Time%) 16 iterations Loss: 0.119206272065639 Improvement: 0.002194 +[3] (%Time%) 17 iterations Loss: 0.117203310132027 Improvement: 0.002051 +[3] (%Time%) 18 iterations Loss: 0.116163291037083 Improvement: 0.001293 +[3] (%Time%) 19 iterations Loss: 0.109811097383499 Improvement: 0.005087 +[3] (%Time%) 20 iterations Loss: 0.106156274676323 Improvement: 0.004013 +[3] (%Time%) 21 iterations Loss: 0.104246392846107 Improvement: 0.002436 +[3] (%Time%) 22 iterations Loss: 0.10310410708189 Improvement: 0.001466 +[3] (%Time%) 23 iterations Loss: 0.102218925952911 Improvement: 0.00103 +[3] (%Time%) 24 iterations Loss: 0.101610459387302 Improvement: 0.0007139 +[3] 'LBFGS Optimizer' finished in %Time%. +[4] 'Saving model' started. +[4] 'Saving model' finished in %Time%. diff --git a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-rp.txt b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-rp.txt index 02b121c96d..817c1490f6 100644 --- a/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-rp.txt +++ b/test/BaselineOutput/SingleRelease/MulticlassLogisticRegression/MulticlassLogisticRegression-TrainTest-iris-rp.txt @@ -1,4 +1,4 @@ MulticlassLogisticRegression Accuracy(micro-avg) Accuracy(macro-avg) Log-loss Log-loss reduction /l2 /l1 /ot /nt Learner Name Train Dataset Test Dataset Results File Run Time Physical Memory Virtual Memory Command Line Settings -0.98 0.98 0.072218 93.42639 0.1 0.001 0.001 1 MulticlassLogisticRegression %Data% %Data% %Output% 99 0 0 maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} norm=No dout=%Output% data=%Data% out=%Output% seed=1 /l2:0.1;/l1:0.001;/ot:0.001;/nt:1 +0.98 0.98 0.072218 93.42639 0.1 0.001 0.001 1 MulticlassLogisticRegression %Data% %Data% %Output% 99 0 0 maml.exe TrainTest test=%Data% tr=MulticlassLogisticRegression{l1=0.001 l2=0.1 ot=1e-3 nt=1} norm=No dout=%Output% data=%Data% out=%Output% seed=1 xf=Term{col=Label} /l2:0.1;/l1:0.001;/ot:0.001;/nt:1 diff --git a/test/Microsoft.ML.Benchmarks.Tests/BenchmarksTest.cs b/test/Microsoft.ML.Benchmarks.Tests/BenchmarksTest.cs index 9dc6854e8b..ccf1898dce 100644 --- a/test/Microsoft.ML.Benchmarks.Tests/BenchmarksTest.cs +++ b/test/Microsoft.ML.Benchmarks.Tests/BenchmarksTest.cs @@ -3,10 +3,8 @@ // See the LICENSE file in the project root for more information. using System; -using System.Collections.Generic; using System.Linq; using System.Reflection; -using BenchmarkDotNet.Attributes; using BenchmarkDotNet.Configs; using BenchmarkDotNet.Jobs; using BenchmarkDotNet.Loggers; diff --git a/test/Microsoft.ML.Benchmarks/PredictionEngineBench.cs b/test/Microsoft.ML.Benchmarks/PredictionEngineBench.cs index 0453fc07dd..49a07fc085 100644 --- a/test/Microsoft.ML.Benchmarks/PredictionEngineBench.cs +++ b/test/Microsoft.ML.Benchmarks/PredictionEngineBench.cs @@ -1,4 +1,4 @@ -// Licensed to the .NET Foundation under one or more agreements. +// Licensed to the .NET Foundation under one or more agreements. // The .NET Foundation licenses this file to you under the MIT license. // See the LICENSE file in the project root for more information. @@ -57,8 +57,9 @@ public void SetupIrisPipeline() IDataView data = loader.Load(_irisDataPath); var pipeline = new ColumnConcatenatingEstimator(env, "Features", new[] { "SepalLength", "SepalWidth", "PetalLength", "PetalWidth" }) + .Append(env.Transforms.Conversion.MapValueToKey("Label")) .Append(env.MulticlassClassification.Trainers.StochasticDualCoordinateAscent( - new SdcaMultiClassTrainer.Options {NumberOfThreads = 1, ConvergenceTolerance = 1e-2f, })); + new SdcaMultiClassTrainer.Options { NumberOfThreads = 1, ConvergenceTolerance = 1e-2f, })); var model = pipeline.Fit(data); @@ -93,7 +94,7 @@ public void SetupSentimentPipeline() var pipeline = mlContext.Transforms.Text.FeaturizeText("Features", "SentimentText") .Append(mlContext.BinaryClassification.Trainers.StochasticDualCoordinateAscentNonCalibrated( - new SdcaNonCalibratedBinaryTrainer.Options {NumberOfThreads = 1, ConvergenceTolerance = 1e-2f, })); + new SdcaNonCalibratedBinaryTrainer.Options { NumberOfThreads = 1, ConvergenceTolerance = 1e-2f, })); var model = pipeline.Fit(data); diff --git a/test/Microsoft.ML.Benchmarks/StochasticDualCoordinateAscentClassifierBench.cs b/test/Microsoft.ML.Benchmarks/StochasticDualCoordinateAscentClassifierBench.cs index 890e3bea46..d8929e926d 100644 --- a/test/Microsoft.ML.Benchmarks/StochasticDualCoordinateAscentClassifierBench.cs +++ b/test/Microsoft.ML.Benchmarks/StochasticDualCoordinateAscentClassifierBench.cs @@ -1,4 +1,4 @@ -// Licensed to the .NET Foundation under one or more agreements. +// Licensed to the .NET Foundation under one or more agreements. // The .NET Foundation licenses this file to you under the MIT license. // See the LICENSE file in the project root for more information. @@ -71,6 +71,7 @@ private TransformerChain { "Text" }, - new TextFeaturizingEstimator.Options { + data = new TextFeaturizingEstimator(Env, "Features", new List { "Text" }, + new TextFeaturizingEstimator.Options + { UseStopRemover = true, }).Fit(data).Transform(data); } @@ -2015,13 +2016,23 @@ public void EntryPointLinearSVM() [Fact] public void EntryPointBinaryEnsemble() { - TestEntryPointRoutine("iris.txt", "Trainers.EnsembleBinaryClassifier"); + TestEntryPointRoutine("iris.txt", "Trainers.EnsembleBinaryClassifier", xfNames: + new[] { "Transforms.ColumnTypeConverter" }, + xfArgs: + new[] { + @"'Column': [{'Name': 'Label','Source': 'Label','Type': 'BL'}]" + }); } [Fact] public void EntryPointClassificationEnsemble() { - TestEntryPointRoutine("iris.txt", "Trainers.EnsembleClassification"); + TestEntryPointRoutine("iris.txt", "Trainers.EnsembleClassification", xfNames: + new[] { "Transforms.TextToKeyConverter" }, + xfArgs: + new[] { + @"'Column': [{'Name': 'Label','Source': 'Label'}]" + }); } [Fact] @@ -2352,10 +2363,29 @@ internal void TestEntryPointPipelineRoutine(string dataFile, string schema, stri cmd.Run(); } - internal void TestEntryPointRoutine(string dataFile, string trainerName, string loader = null, string trainerArgs = null) + internal void TestEntryPointRoutine(string dataFile, string trainerName, string loader = null, string trainerArgs = null, string[] xfNames = null, string[] xfArgs = null) { var dataPath = GetDataPath(dataFile); var outputPath = DeleteOutputPath("model.zip"); + string xfTemplate = + @"'Name': '{0}', + 'Inputs': {{ + 'Data': '$data{1}', + {2}, + }}, + 'Outputs': {{ + 'OutputData': '$data{3}' + }}"; + var transforms = ""; + + for (int i = 0; i < Utils.Size(xfNames); i++) + { + transforms = + $@"{transforms} + {{ + {string.Format(xfTemplate, xfNames[i], i + 1, xfArgs[i], i + 2)} + }},"; + } string inputGraph = string.Format(@" {{ 'Nodes': [ @@ -2369,10 +2399,11 @@ internal void TestEntryPointRoutine(string dataFile, string trainerName, string 'Data': '$data1' }} }}, + {5} {{ 'Name': '{2}', 'Inputs': {{ - 'TrainingData': '$data1' + 'TrainingData': '$data{6}' {4} }}, 'Outputs': {{ @@ -2387,9 +2418,11 @@ internal void TestEntryPointRoutine(string dataFile, string trainerName, string 'model' : '{1}' }} }}", EscapePath(dataPath), EscapePath(outputPath), trainerName, - string.IsNullOrWhiteSpace(loader) ? "" : string.Format(",'CustomSchema': 'sparse+ {0}'", loader), - string.IsNullOrWhiteSpace(trainerArgs) ? "" : trainerArgs - ); + string.IsNullOrWhiteSpace(loader) ? "" : string.Format(",'CustomSchema': 'sparse+ {0}'", loader), + string.IsNullOrWhiteSpace(trainerArgs) ? "" : trainerArgs, + transforms, + xfNames != null ? xfNames.Length + 1 : 1 + ); var jsonPath = DeleteOutputPath("graph.json"); File.WriteAllLines(jsonPath, new[] { inputGraph }); @@ -2952,8 +2985,8 @@ public void EntryPointChainedTrainTestMacros() var metrics = runner.GetOutput("OverallMetrics"); - Action validateAuc = (metricsIdv) => - { + Action validateAuc = (metricsIdv) => + { Assert.NotNull(metricsIdv); using (var cursor = metricsIdv.GetRowCursorForAllColumns()) { @@ -3143,8 +3176,8 @@ public void EntryPointChainedCrossValMacros() var metrics = runner.GetOutput("OverallMetrics"); - Action aucValidate = (metricsIdv) => - { + Action aucValidate = (metricsIdv) => + { Assert.NotNull(metricsIdv); using (var cursor = metrics.GetRowCursorForAllColumns()) { diff --git a/test/Microsoft.ML.Functional.Tests/Evaluation.cs b/test/Microsoft.ML.Functional.Tests/Evaluation.cs index b10dec8c05..841b2c1aa0 100644 --- a/test/Microsoft.ML.Functional.Tests/Evaluation.cs +++ b/test/Microsoft.ML.Functional.Tests/Evaluation.cs @@ -149,6 +149,7 @@ public void TrainAndEvaluateMulticlassClassification() // Create a training pipeline. var pipeline = mlContext.Transforms.Concatenate("Features", Iris.Features) + .Append(mlContext.Transforms.Conversion.MapValueToKey("Label")) .AppendCacheCheckpoint(mlContext) .Append(mlContext.MulticlassClassification.Trainers.StochasticDualCoordinateAscent( new SdcaMultiClassTrainer.Options { NumberOfThreads = 1})); diff --git a/test/Microsoft.ML.TestFramework/Datasets.cs b/test/Microsoft.ML.TestFramework/Datasets.cs index 23cb24dc62..a0aa1d7dac 100644 --- a/test/Microsoft.ML.TestFramework/Datasets.cs +++ b/test/Microsoft.ML.TestFramework/Datasets.cs @@ -458,7 +458,8 @@ public static class TestDatasets trainFilename = @"iris.txt", testFilename = @"iris.txt", fileHasHeader = true, - fileSeparator = '\t' + fileSeparator = '\t', + mamlExtraSettings = new[] { "xf=Term{col=Label}" } }; public static TestDataset irisMissing = new TestDataset() diff --git a/test/Microsoft.ML.Tests/FeatureContributionTests.cs b/test/Microsoft.ML.Tests/FeatureContributionTests.cs index fc6469f809..b59cbb1d0a 100644 --- a/test/Microsoft.ML.Tests/FeatureContributionTests.cs +++ b/test/Microsoft.ML.Tests/FeatureContributionTests.cs @@ -109,7 +109,7 @@ public void TestLightGbmRanking() { TestFeatureContribution(ML.Ranking.Trainers.LightGbm(), GetSparseDataset(TaskType.Ranking, 100), "LightGbmRanking"); } - + // Tests for binary classification trainers that implement IFeatureContributionMapper interface. [Fact] public void TestAveragePerceptronBinary() @@ -158,7 +158,7 @@ public void TestSDCABinary() public void TestSGDBinary() { TestFeatureContribution(ML.BinaryClassification.Trainers.StochasticGradientDescent( - new SgdBinaryTrainer.Options { NumberOfThreads = 1}), + new SgdBinaryTrainer.Options { NumberOfThreads = 1 }), GetSparseDataset(TaskType.BinaryClassification, 100), "SGDBinary"); } @@ -181,7 +181,7 @@ private void TestFeatureContribution( int precision = 6) { // Train the model. - var model = trainer.Fit(data); + var model = trainer.Fit(data); // Extract the predictor, check that it supports feature contribution. var predictor = model.Model as ICalculateFeatureContribution; @@ -274,8 +274,13 @@ private IDataView GetSparseDataset(TaskType task = TaskType.Regression, int numb var pipeline = ML.Transforms.Concatenate("Features", "X1", "X2VBuffer", "X3Important") .Append(ML.Transforms.Normalize("Features")); - // Create a keytype for Ranking - if (task == TaskType.Ranking) + if (task == TaskType.BinaryClassification) + return pipeline.Append(ML.Transforms.Conversion.ConvertType("Label", outputKind: DataKind.Boolean)) + .Fit(srcDV).Transform(srcDV); + else if (task == TaskType.MulticlassClassification) + return pipeline.Append(ML.Transforms.Conversion.MapValueToKey("Label")) + .Fit(srcDV).Transform(srcDV); + else if (task == TaskType.Ranking) return pipeline.Append(ML.Transforms.Conversion.MapValueToKey("GroupId")) .Fit(srcDV).Transform(srcDV); diff --git a/test/Microsoft.ML.Tests/PermutationFeatureImportanceTests.cs b/test/Microsoft.ML.Tests/PermutationFeatureImportanceTests.cs index 92ff2fbddf..51dedb1fbd 100644 --- a/test/Microsoft.ML.Tests/PermutationFeatureImportanceTests.cs +++ b/test/Microsoft.ML.Tests/PermutationFeatureImportanceTests.cs @@ -88,7 +88,7 @@ public void TestPfiRegressionStandardDeviationAndErrorOnDenseFeatures() Assert.Equal(3, MinDeltaIndex(pfi, m => m.RSquared.StandardDeviation)); Assert.Equal(1, MaxDeltaIndex(pfi, m => m.RSquared.StandardDeviation)); - + // Stardard Error will scale with the magnitude of the measure (as it's SD/sqrt(N)) Assert.Equal(3, MinDeltaIndex(pfi, m => m.MeanAbsoluteError.StandardError)); Assert.Equal(1, MaxDeltaIndex(pfi, m => m.MeanAbsoluteError.StandardError)); @@ -173,7 +173,7 @@ public void TestPfiBinaryClassificationOnDenseFeatures() Assert.Equal(1, MinDeltaIndex(pfi, m => m.PositiveRecall.Mean)); Assert.Equal(3, MaxDeltaIndex(pfi, m => m.NegativePrecision.Mean)); Assert.Equal(1, MinDeltaIndex(pfi, m => m.NegativePrecision.Mean)); - Assert.Equal(0, MaxDeltaIndex(pfi, m => m.NegativeRecall.Mean)); + Assert.Equal(3, MaxDeltaIndex(pfi, m => m.NegativeRecall.Mean)); Assert.Equal(1, MinDeltaIndex(pfi, m => m.NegativeRecall.Mean)); Assert.Equal(3, MaxDeltaIndex(pfi, m => m.F1Score.Mean)); Assert.Equal(1, MinDeltaIndex(pfi, m => m.F1Score.Mean)); @@ -403,7 +403,7 @@ private IDataView GetDenseDataset(TaskType task = TaskType.Regression) } // If binary classification, modify the labels - if (task == TaskType.BinaryClassification || + if (task == TaskType.BinaryClassification || task == TaskType.MulticlassClassification) GetBinaryClassificationLabels(yArray); else if (task == TaskType.Ranking) @@ -422,9 +422,13 @@ private IDataView GetDenseDataset(TaskType task = TaskType.Regression) var pipeline = ML.Transforms.Concatenate("Features", "X1", "X2Important", "X3", "X4Rand") .Append(ML.Transforms.Normalize("Features")); - - // Create a keytype for Ranking - if (task == TaskType.Ranking) + if (task == TaskType.BinaryClassification) + return pipeline.Append(ML.Transforms.Conversion.ConvertType("Label", outputKind: DataKind.Boolean)) + .Fit(srcDV).Transform(srcDV); + else if (task == TaskType.MulticlassClassification) + return pipeline.Append(ML.Transforms.Conversion.MapValueToKey("Label")) + .Fit(srcDV).Transform(srcDV); + else if (task == TaskType.Ranking) return pipeline.Append(ML.Transforms.Conversion.MapValueToKey("GroupId")) .Fit(srcDV).Transform(srcDV); @@ -498,9 +502,17 @@ private IDataView GetSparseDataset(TaskType task = TaskType.Regression) var pipeline = ML.Transforms.Concatenate("Features", "X1", "X2VBuffer", "X3Important") .Append(ML.Transforms.Normalize("Features")); - - // Create a keytype for Ranking - if (task == TaskType.Ranking) + if (task == TaskType.BinaryClassification) + { + return pipeline.Append(ML.Transforms.Conversion.ConvertType("Label", outputKind: DataKind.Boolean)) + .Fit(srcDV).Transform(srcDV); + } + else if (task == TaskType.MulticlassClassification) + { + return pipeline.Append(ML.Transforms.Conversion.MapValueToKey("Label")) + .Fit(srcDV).Transform(srcDV); + } + else if (task == TaskType.Ranking) return pipeline.Append(ML.Transforms.Conversion.MapValueToKey("GroupId")) .Fit(srcDV).Transform(srcDV); diff --git a/test/Microsoft.ML.Tests/Scenarios/Api/TestApi.cs b/test/Microsoft.ML.Tests/Scenarios/Api/TestApi.cs index 574db0f901..df292e43c0 100644 --- a/test/Microsoft.ML.Tests/Scenarios/Api/TestApi.cs +++ b/test/Microsoft.ML.Tests/Scenarios/Api/TestApi.cs @@ -174,10 +174,10 @@ public void TrainAveragedPerceptronWithCache() var dataFile = GetDataPath("breast-cancer.txt"); var loader = TextLoader.Create(mlContext, new TextLoader.Options(), new MultiFileSource(dataFile)); var globalCounter = 0; - var xf = LambdaTransform.CreateFilter(mlContext, loader, + IDataView xf = LambdaTransform.CreateFilter(mlContext, loader, (i, s) => true, s => { globalCounter++; }); - + xf = mlContext.Transforms.Conversion.ConvertType("Label", outputKind: DataKind.Boolean).Fit(xf).Transform(xf); // The baseline result of this was generated with everything cached in memory. As auto-cache is removed, // an explicit step of caching is required to make this test ok. var cached = mlContext.Data.Cache(xf); @@ -295,7 +295,6 @@ private List ReadBreastCancerExamples() public void TestTrainTestSplit() { var mlContext = new MLContext(0); - var dataPath = GetDataPath("adult.tiny.with-schema.txt"); // Create the reader: define the data columns and where to find them in the text file. var input = mlContext.Data.LoadFromTextFile(dataPath, new[] { diff --git a/test/Microsoft.ML.Tests/Scenarios/IrisPlantClassificationTests.cs b/test/Microsoft.ML.Tests/Scenarios/IrisPlantClassificationTests.cs index 06f1a166ef..c8cf115396 100644 --- a/test/Microsoft.ML.Tests/Scenarios/IrisPlantClassificationTests.cs +++ b/test/Microsoft.ML.Tests/Scenarios/IrisPlantClassificationTests.cs @@ -30,6 +30,7 @@ public void TrainAndPredictIrisModelTest() var pipe = mlContext.Transforms.Concatenate("Features", "SepalLength", "SepalWidth", "PetalLength", "PetalWidth") .Append(mlContext.Transforms.Normalize("Features")) + .Append(mlContext.Transforms.Conversion.MapValueToKey("Label")) .AppendCacheCheckpoint(mlContext) .Append(mlContext.MulticlassClassification.Trainers.StochasticDualCoordinateAscent( new SdcaMultiClassTrainer.Options { NumberOfThreads = 1 })); diff --git a/test/Microsoft.ML.Tests/Scenarios/OvaTest.cs b/test/Microsoft.ML.Tests/Scenarios/OvaTest.cs index c6e47bb8eb..3b4de327bc 100644 --- a/test/Microsoft.ML.Tests/Scenarios/OvaTest.cs +++ b/test/Microsoft.ML.Tests/Scenarios/OvaTest.cs @@ -28,8 +28,9 @@ public void OvaLogisticRegression() } }); - // Data - var data = reader.Load(GetDataPath(dataPath)); + var textData = reader.Load(GetDataPath(dataPath)); + var data = mlContext.Data.Cache(mlContext.Transforms.Conversion.MapValueToKey("Label") + .Fit(textData).Transform(textData)); // Pipeline var logReg = mlContext.BinaryClassification.Trainers.LogisticRegression(); @@ -61,11 +62,14 @@ public void OvaAveragedPerceptron() }); // Data - var data = mlContext.Data.Cache(reader.Load(GetDataPath(dataPath))); + var textData = reader.Load(GetDataPath(dataPath)); + var data = mlContext.Data.Cache(mlContext.Transforms.Conversion.MapValueToKey("Label") + .Fit(textData).Transform(textData)); // Pipeline var ap = mlContext.BinaryClassification.Trainers.AveragedPerceptron( new AveragedPerceptronTrainer.Options { Shuffle = true }); + var pipeline = mlContext.MulticlassClassification.Trainers.OneVersusAll(ap, useProbabilities: false); var model = pipeline.Fit(data); @@ -73,7 +77,7 @@ public void OvaAveragedPerceptron() // Metrics var metrics = mlContext.MulticlassClassification.Evaluate(predictions); - Assert.True(metrics.MicroAccuracy > 0.71); + Assert.True(metrics.MicroAccuracy > 0.66); } [Fact] @@ -94,7 +98,9 @@ public void OvaFastTree() }); // Data - var data = reader.Load(GetDataPath(dataPath)); + var textData = reader.Load(GetDataPath(dataPath)); + var data = mlContext.Data.Cache(mlContext.Transforms.Conversion.MapValueToKey("Label") + .Fit(textData).Transform(textData)); // Pipeline var pipeline = mlContext.MulticlassClassification.Trainers.OneVersusAll( @@ -125,9 +131,10 @@ public void OvaLinearSvm() new TextLoader.Column("Features", DataKind.Single, new [] { new TextLoader.Range(1, 4) }), } }); - // Data - var data = mlContext.Data.Cache(reader.Load(GetDataPath(dataPath))); + var textData = reader.Load(GetDataPath(dataPath)); + var data = mlContext.Data.Cache(mlContext.Transforms.Conversion.MapValueToKey("Label") + .Fit(textData).Transform(textData)); // Pipeline var pipeline = mlContext.MulticlassClassification.Trainers.OneVersusAll( diff --git a/test/Microsoft.ML.Tests/ScenariosWithDirectInstantiation/IrisPlantClassificationTests.cs b/test/Microsoft.ML.Tests/ScenariosWithDirectInstantiation/IrisPlantClassificationTests.cs index 04c3770233..3390b677ca 100644 --- a/test/Microsoft.ML.Tests/ScenariosWithDirectInstantiation/IrisPlantClassificationTests.cs +++ b/test/Microsoft.ML.Tests/ScenariosWithDirectInstantiation/IrisPlantClassificationTests.cs @@ -28,6 +28,7 @@ public void TrainAndPredictIrisModelUsingDirectInstantiationTest() var pipe = mlContext.Transforms.Concatenate("Features", "SepalLength", "SepalWidth", "PetalLength", "PetalWidth") .Append(mlContext.Transforms.Normalize("Features")) + .Append(mlContext.Transforms.Conversion.MapValueToKey("Label")) .AppendCacheCheckpoint(mlContext) .Append(mlContext.MulticlassClassification.Trainers.StochasticDualCoordinateAscent( new SdcaMultiClassTrainer.Options { NumberOfThreads = 1 })); diff --git a/test/Microsoft.ML.Tests/TrainerEstimators/SdcaTests.cs b/test/Microsoft.ML.Tests/TrainerEstimators/SdcaTests.cs index e96ef170ad..10976d9b5d 100644 --- a/test/Microsoft.ML.Tests/TrainerEstimators/SdcaTests.cs +++ b/test/Microsoft.ML.Tests/TrainerEstimators/SdcaTests.cs @@ -20,21 +20,25 @@ public void SdcaWorkout() var data = TextLoaderStatic.CreateLoader(Env, ctx => (Label: ctx.LoadFloat(0), Features: ctx.LoadFloat(1, 10))) .Load(dataPath).Cache(); + var binaryData = ML.Transforms.Conversion.ConvertType("Label", outputKind: DataKind.Boolean) + .Fit(data.AsDynamic).Transform(data.AsDynamic); + var binaryTrainer = ML.BinaryClassification.Trainers.StochasticDualCoordinateAscent( - new SdcaBinaryTrainer.Options { ConvergenceTolerance = 1e-2f }); - TestEstimatorCore(binaryTrainer, data.AsDynamic); + new SdcaBinaryTrainer.Options { ConvergenceTolerance = 1e-2f, MaximumNumberOfIterations = 10 }); + TestEstimatorCore(binaryTrainer, binaryData); var nonCalibratedBinaryTrainer = ML.BinaryClassification.Trainers.StochasticDualCoordinateAscentNonCalibrated( - new SdcaNonCalibratedBinaryTrainer.Options { ConvergenceTolerance = 1e-2f }); - TestEstimatorCore(nonCalibratedBinaryTrainer, data.AsDynamic); + new SdcaNonCalibratedBinaryTrainer.Options { ConvergenceTolerance = 1e-2f, MaximumNumberOfIterations = 10 }); + TestEstimatorCore(nonCalibratedBinaryTrainer, binaryData); var regressionTrainer = ML.Regression.Trainers.StochasticDualCoordinateAscent( - new SdcaRegressionTrainer.Options { ConvergenceTolerance = 1e-2f }); + new SdcaRegressionTrainer.Options { ConvergenceTolerance = 1e-2f, MaximumNumberOfIterations = 10 }); TestEstimatorCore(regressionTrainer, data.AsDynamic); + var mcData = ML.Transforms.Conversion.MapValueToKey("Label").Fit(data.AsDynamic).Transform(data.AsDynamic); var mcTrainer = ML.MulticlassClassification.Trainers.StochasticDualCoordinateAscent( - new SdcaMultiClassTrainer.Options { ConvergenceTolerance = 1e-2f }); - TestEstimatorCore(mcTrainer, data.AsDynamic); + new SdcaMultiClassTrainer.Options { ConvergenceTolerance = 1e-2f, MaximumNumberOfIterations = 10 }); + TestEstimatorCore(mcTrainer, mcData); Done(); }