Applied formatter
Project: http://git-wip-us.apache.org/repos/asf/incubator-hivemall/repo Commit: http://git-wip-us.apache.org/repos/asf/incubator-hivemall/commit/c4036695 Tree: http://git-wip-us.apache.org/repos/asf/incubator-hivemall/tree/c4036695 Diff: http://git-wip-us.apache.org/repos/asf/incubator-hivemall/diff/c4036695 Branch: refs/heads/master Commit: c40366959b7b87d9162398d083556aae1ba84065 Parents: ae0d3c4 Author: Makoto Yui <[email protected]> Authored: Fri Apr 27 15:36:44 2018 +0900 Committer: Makoto Yui <[email protected]> Committed: Fri Apr 27 15:36:44 2018 +0900 ---------------------------------------------------------------------- .../java/hivemall/GeneralLearnerBaseUDTF.java | 51 ++--- .../src/main/java/hivemall/LearnerBaseUDTF.java | 19 +- .../java/hivemall/anomaly/ChangeFinderUDF.java | 21 +- core/src/main/java/hivemall/anomaly/SDAR1D.java | 4 +- core/src/main/java/hivemall/anomaly/SDAR2D.java | 8 +- .../anomaly/SingularSpectrumTransform.java | 4 +- .../anomaly/SingularSpectrumTransformUDF.java | 36 ++-- .../hivemall/classifier/AROWClassifierUDTF.java | 10 +- .../hivemall/classifier/AdaGradRDAUDTF.java | 3 +- .../classifier/BinaryOnlineClassifierUDTF.java | 5 +- .../classifier/ConfidenceWeightedUDTF.java | 3 +- .../hivemall/classifier/KPAPredictUDAF.java | 37 ++-- .../KernelExpansionPassiveAggressiveUDTF.java | 3 +- .../classifier/PassiveAggressiveUDTF.java | 10 +- .../classifier/SoftConfideceWeightedUDTF.java | 15 +- .../MulticlassAROWClassifierUDTF.java | 21 +- .../MulticlassConfidenceWeightedUDTF.java | 10 +- .../MulticlassOnlineClassifierUDTF.java | 42 ++-- .../MulticlassPassiveAggressiveUDTF.java | 13 +- .../multiclass/MulticlassPerceptronUDTF.java | 6 +- .../MulticlassSoftConfidenceWeightedUDTF.java | 26 ++- .../LogisticRegressionDataGeneratorUDTF.java | 19 +- .../hivemall/ensemble/ArgminKLDistanceUDAF.java | 3 +- .../main/java/hivemall/ensemble/MaxRowUDAF.java | 6 +- .../ensemble/bagging/WeightVotedAvgUDAF.java | 3 +- .../main/java/hivemall/evaluation/AUCUDAF.java | 74 ++++--- .../evaluation/BinaryResponsesMeasures.java | 4 +- .../java/hivemall/evaluation/F1ScoreUDAF.java | 4 +- .../java/hivemall/evaluation/FMeasureUDAF.java | 60 +++--- .../evaluation/GradedResponsesMeasures.java | 3 +- .../java/hivemall/evaluation/HitRateUDAF.java | 13 +- .../main/java/hivemall/evaluation/MAPUDAF.java | 12 +- .../main/java/hivemall/evaluation/MRRUDAF.java | 9 +- .../main/java/hivemall/evaluation/NDCGUDAF.java | 37 ++-- .../java/hivemall/evaluation/PrecisionUDAF.java | 9 +- .../main/java/hivemall/evaluation/R2UDAF.java | 3 +- .../java/hivemall/evaluation/RecallUDAF.java | 9 +- core/src/main/java/hivemall/fm/Entry.java | 13 +- .../hivemall/fm/FFMStringFeatureMapModel.java | 5 +- .../src/main/java/hivemall/fm/FMArrayModel.java | 4 +- .../java/hivemall/fm/FMHyperParameters.java | 24 +-- .../java/hivemall/fm/FMIntFeatureMapModel.java | 4 +- .../java/hivemall/fm/FMPredictGenericUDAF.java | 18 +- .../hivemall/fm/FactorizationMachineModel.java | 31 +-- .../hivemall/fm/FactorizationMachineUDTF.java | 33 ++-- .../fm/FieldAwareFactorizationMachineModel.java | 36 ++-- .../fm/FieldAwareFactorizationMachineUDTF.java | 14 +- .../main/java/hivemall/ftvec/AddBiasUDF.java | 3 +- .../java/hivemall/ftvec/AddFeatureIndexUDF.java | 3 +- .../java/hivemall/ftvec/ExtractWeightUDF.java | 3 +- .../main/java/hivemall/ftvec/FeatureUDF.java | 10 +- .../ftvec/amplify/RandomAmplifierUDTF.java | 6 +- .../hivemall/ftvec/binning/BuildBinsUDAF.java | 32 +-- .../ftvec/binning/FeatureBinningUDF.java | 21 +- .../ftvec/binning/NumericHistogram.java | 16 +- .../ftvec/conv/ConvertToDenseModelUDAF.java | 3 +- .../ftvec/conv/QuantifyColumnsUDTF.java | 4 +- .../ftvec/hashing/ArrayHashValuesUDF.java | 6 +- .../ftvec/hashing/FeatureHashingUDF.java | 3 +- .../ftvec/pairing/FeaturePairsUDTF.java | 8 +- .../hivemall/ftvec/ranking/BprSamplingUDTF.java | 12 +- .../ftvec/ranking/ItemPairsSamplingUDTF.java | 15 +- .../ftvec/ranking/PopulateNotInUDTF.java | 6 +- .../java/hivemall/ftvec/scaling/RescaleUDF.java | 13 +- .../hivemall/ftvec/selection/ChiSquareUDF.java | 20 +- .../ftvec/selection/SignalNoiseRatioUDAF.java | 70 ++++--- .../ftvec/trans/AddFieldIndicesUDF.java | 3 +- .../hivemall/ftvec/trans/BinarizeLabelUDTF.java | 7 +- .../ftvec/trans/CategoricalFeaturesUDF.java | 31 ++- .../hivemall/ftvec/trans/FFMFeaturesUDF.java | 39 ++-- .../hivemall/ftvec/trans/IndexedFeatures.java | 6 +- .../ftvec/trans/OnehotEncodingUDAF.java | 22 ++- .../ftvec/trans/QuantifiedFeaturesUDTF.java | 13 +- .../ftvec/trans/QuantitativeFeaturesUDF.java | 31 ++- .../ftvec/trans/VectorizeFeaturesUDF.java | 31 ++- .../geospatial/HaversineDistanceUDF.java | 3 +- .../java/hivemall/geospatial/Lat2TileYUDF.java | 3 +- .../java/hivemall/geospatial/Lon2TileXUDF.java | 3 +- .../java/hivemall/geospatial/MapURLUDF.java | 6 +- .../main/java/hivemall/geospatial/TileUDF.java | 3 +- .../java/hivemall/geospatial/TileX2LonUDF.java | 3 +- .../java/hivemall/geospatial/TileY2LatUDF.java | 3 +- .../knn/distance/EuclidDistanceUDF.java | 3 +- .../hivemall/knn/distance/KLDivergenceUDF.java | 3 +- .../main/java/hivemall/knn/lsh/MinHashUDTF.java | 3 +- .../java/hivemall/knn/lsh/MinHashesUDF.java | 3 +- .../java/hivemall/knn/lsh/bBitMinHashUDF.java | 8 +- .../knn/similarity/DIMSUMMapperUDTF.java | 3 +- .../hivemall/math/matrix/AbstractMatrix.java | 6 +- .../java/hivemall/math/matrix/FloatMatrix.java | 3 +- .../builders/ColumnMajorDenseMatrixBuilder.java | 3 +- .../math/matrix/builders/MatrixBuilder.java | 4 +- .../math/matrix/ints/AbstractIntMatrix.java | 6 +- .../hivemall/math/matrix/ints/DoKIntMatrix.java | 12 +- .../hivemall/math/matrix/sparse/CSCMatrix.java | 12 +- .../hivemall/math/matrix/sparse/CSRMatrix.java | 16 +- .../hivemall/math/matrix/sparse/DoKMatrix.java | 7 +- .../matrix/sparse/floats/CSCFloatMatrix.java | 12 +- .../matrix/sparse/floats/CSRFloatMatrix.java | 19 +- .../matrix/sparse/floats/DoKFloatMatrix.java | 7 +- .../random/RandomNumberGeneratorFactory.java | 8 +- .../java/hivemall/mf/BPRMFPredictionUDF.java | 3 +- .../hivemall/mf/BPRMatrixFactorizationUDTF.java | 38 ++-- .../main/java/hivemall/mf/FactorizedModel.java | 8 +- .../main/java/hivemall/mf/MFPredictionUDF.java | 3 +- .../mf/MatrixFactorizationAdaGradUDTF.java | 3 +- .../hivemall/mf/MatrixFactorizationSGDUDTF.java | 3 +- .../mf/OnlineMatrixFactorizationUDTF.java | 43 ++-- .../java/hivemall/mix/MixMessageDecoder.java | 3 +- .../java/hivemall/mix/client/MixClient.java | 4 +- .../hivemall/mix/client/MixRequestRouter.java | 4 +- .../main/java/hivemall/model/DenseModel.java | 3 +- .../main/java/hivemall/model/FeatureValue.java | 3 +- .../main/java/hivemall/model/NewDenseModel.java | 3 +- .../model/NewSpaceEfficientDenseModel.java | 3 +- .../java/hivemall/model/NewSparseModel.java | 3 +- .../model/SpaceEfficientDenseModel.java | 3 +- .../main/java/hivemall/model/SparseModel.java | 3 +- .../model/SynchronizedModelWrapper.java | 3 +- .../optimizer/DenseOptimizerFactory.java | 16 +- .../java/hivemall/optimizer/LossFunctions.java | 4 +- .../main/java/hivemall/optimizer/Optimizer.java | 16 +- .../hivemall/optimizer/OptimizerOptions.java | 3 +- .../java/hivemall/optimizer/Regularization.java | 8 +- .../optimizer/SparseOptimizerFactory.java | 16 +- .../main/java/hivemall/recommend/SlimUDTF.java | 78 ++++---- .../hivemall/regression/AROWRegressionUDTF.java | 12 +- .../java/hivemall/regression/AdaDeltaUDTF.java | 12 +- .../java/hivemall/regression/AdaGradUDTF.java | 3 +- .../java/hivemall/regression/LogressUDTF.java | 3 +- .../PassiveAggressiveRegressionUDTF.java | 15 +- .../java/hivemall/sketch/bloom/BloomAndUDF.java | 3 +- .../hivemall/sketch/bloom/BloomContainsUDF.java | 3 +- .../java/hivemall/sketch/bloom/BloomOrUDF.java | 3 +- .../sketch/hll/ApproxCountDistinctUDAF.java | 6 +- .../smile/classification/DecisionTree.java | 98 +++++----- .../GradientTreeBoostingClassifierUDTF.java | 45 +++-- .../RandomForestClassifierUDTF.java | 30 ++- .../regression/RandomForestRegressionUDTF.java | 39 ++-- .../smile/regression/RegressionTree.java | 77 ++++---- .../smile/tools/RandomForestEnsembleUDAF.java | 33 ++-- .../hivemall/smile/tools/TreeExportUDF.java | 9 +- .../hivemall/smile/tools/TreePredictUDF.java | 14 +- .../hivemall/smile/tools/TreePredictUDFv1.java | 25 ++- .../hivemall/smile/utils/SmileExtUtils.java | 11 +- .../java/hivemall/smile/vm/StackMachine.java | 8 +- .../hivemall/statistics/MovingAverageUDTF.java | 3 +- .../java/hivemall/tools/ConvertLabelUDF.java | 3 +- .../main/java/hivemall/tools/EachTopKUDTF.java | 3 +- .../java/hivemall/tools/GenerateSeriesUDTF.java | 7 +- .../main/java/hivemall/tools/TryCastUDF.java | 5 +- .../hivemall/tools/array/ArrayAppendUDF.java | 6 +- .../tools/array/ArrayAvgGenericUDAF.java | 32 +-- .../hivemall/tools/array/ArrayConcatUDF.java | 8 +- .../hivemall/tools/array/ArrayFlattenUDF.java | 11 +- .../hivemall/tools/array/ArrayIntersectUDF.java | 13 +- .../hivemall/tools/array/ArraySliceUDF.java | 7 +- .../java/hivemall/tools/array/ArraySumUDAF.java | 4 +- .../hivemall/tools/array/ArrayUnionUDF.java | 9 +- .../hivemall/tools/array/CollectAllUDAF.java | 9 +- .../tools/array/ConditionalEmitUDTF.java | 3 +- .../hivemall/tools/array/SelectKBestUDF.java | 13 +- .../tools/array/SortAndUniqArrayUDF.java | 5 +- .../tools/array/SubarrayEndWithUDF.java | 5 +- .../tools/array/SubarrayStartWithUDF.java | 5 +- .../hivemall/tools/bits/BitsCollectUDAF.java | 13 +- .../java/hivemall/tools/bits/BitsORUDF.java | 10 +- .../java/hivemall/tools/bits/ToBitsUDF.java | 6 +- .../java/hivemall/tools/bits/UnBitsUDF.java | 3 +- .../hivemall/tools/compress/DeflateUDF.java | 6 +- .../hivemall/tools/compress/InflateUDF.java | 3 +- .../java/hivemall/tools/json/FromJsonUDF.java | 17 +- .../java/hivemall/tools/json/ToJsonUDF.java | 10 +- .../hivemall/tools/list/UDAFToOrderedList.java | 54 ++--- .../java/hivemall/tools/map/MapTailNUDF.java | 3 +- .../main/java/hivemall/tools/map/UDAFToMap.java | 3 +- .../hivemall/tools/map/UDAFToOrderedMap.java | 30 ++- .../tools/mapred/DistributedCacheLookupUDF.java | 22 +-- .../hivemall/tools/mapred/RowNumberUDF.java | 4 +- .../java/hivemall/tools/math/L2NormUDAF.java | 3 +- .../tools/matrix/TransposeAndDotUDAF.java | 19 +- .../tools/text/NormalizeUnicodeUDF.java | 3 +- .../hivemall/tools/text/SingularizeUDF.java | 4 +- .../java/hivemall/tools/text/SplitWordsUDF.java | 3 +- .../java/hivemall/tools/text/StopwordUDF.java | 16 +- .../java/hivemall/tools/text/WordNgramsUDF.java | 10 +- .../hivemall/tools/vector/VectorAddUDF.java | 10 +- .../hivemall/tools/vector/VectorDotUDF.java | 7 +- .../topicmodel/IncrementalPLSAModel.java | 12 +- .../hivemall/topicmodel/LDAPredictUDAF.java | 48 +++-- .../main/java/hivemall/topicmodel/LDAUDTF.java | 3 +- .../hivemall/topicmodel/OnlineLDAModel.java | 16 +- .../hivemall/topicmodel/PLSAPredictUDAF.java | 45 +++-- .../ProbabilisticTopicModelBaseUDTF.java | 32 ++- .../hivemall/utils/buffer/DynamicByteArray.java | 4 +- .../java/hivemall/utils/buffer/HeapBuffer.java | 8 +- .../hivemall/utils/collections/Fastutil.java | 46 +++-- .../utils/collections/arrays/DoubleArray3D.java | 17 +- .../utils/collections/lists/IntArrayList.java | 4 +- .../maps/Long2DoubleOpenHashTable.java | 16 +- .../maps/Long2FloatOpenHashTable.java | 16 +- .../collections/maps/Long2IntOpenHashTable.java | 23 +-- .../utils/collections/maps/OpenHashTable.java | 16 +- .../utils/geospatial/GeoSpatialUtils.java | 4 +- .../java/hivemall/utils/hadoop/HiveUtils.java | 50 ++--- .../hivemall/utils/hadoop/JsonSerdeUtils.java | 64 +++--- .../utils/io/CompressionStreamFactory.java | 11 +- .../hivemall/utils/io/LimitedInputStream.java | 7 +- .../java/hivemall/utils/lang/ArrayUtils.java | 13 +- .../main/java/hivemall/utils/lang/BitUtils.java | 3 +- .../java/hivemall/utils/lang/HalfFloat.java | 4 +- .../java/hivemall/utils/lang/ObjectUtils.java | 20 +- .../java/hivemall/utils/lang/Preconditions.java | 28 +-- .../hivemall/utils/lang/PrivilegedAccessor.java | 13 +- .../utils/lang/mutable/MutableDouble.java | 8 +- .../utils/lang/mutable/MutableFloat.java | 8 +- .../hivemall/utils/lang/mutable/MutableInt.java | 4 +- .../utils/lang/mutable/MutableLong.java | 4 +- .../main/java/hivemall/utils/math/FastMath.java | 47 ++--- .../java/hivemall/utils/math/MathUtils.java | 17 +- .../java/hivemall/utils/math/MatrixUtils.java | 59 +++--- .../main/java/hivemall/utils/math/Primes.java | 65 +++---- .../java/hivemall/utils/math/StatsUtils.java | 12 +- .../main/java/hivemall/utils/net/NetUtils.java | 7 +- core/src/test/java/hivemall/TestUtils.java | 9 +- .../hivemall/anomaly/ChangeFinder2DTest.java | 10 +- .../classifier/GeneralClassifierUDTFTest.java | 40 ++-- ...ernelExpansionPassiveAggressiveUDTFTest.java | 19 +- .../classifier/PassiveAggressiveUDTFTest.java | 61 +++--- .../hivemall/classifier/PerceptronUDTFTest.java | 36 ++-- .../java/hivemall/evaluation/AUCUDAFTest.java | 59 +++--- .../evaluation/BinaryResponsesMeasuresTest.java | 11 +- .../hivemall/evaluation/FMeasureUDAFTest.java | 71 ++++--- .../evaluation/GradedResponsesMeasuresTest.java | 4 +- .../test/java/hivemall/fm/ArrayModelTest.java | 6 +- .../fm/FactorizationMachineUDTFTest.java | 15 +- core/src/test/java/hivemall/fm/FeatureTest.java | 3 +- .../FieldAwareFactorizationMachineUDTFTest.java | 39 ++-- .../hivemall/fm/IntFeatureMapModelTest.java | 6 +- .../hivemall/fm/StringFeatureMapModelTest.java | 6 +- .../java/hivemall/ftvec/FeatureUDFTest.java | 18 +- .../ftvec/hashing/FeatureHashingUDFTest.java | 13 +- .../pairing/PolynomialFeaturesUDFTest.java | 16 +- .../ftvec/selection/ChiSquareUDFTest.java | 26 +-- .../selection/SignalNoiseRatioUDAFTest.java | 195 +++++++++++-------- .../ftvec/trans/BinarizeLabelUDTFTest.java | 6 +- .../ftvec/trans/QuantifiedFeaturesUDTFTest.java | 7 +- .../ftvec/trans/VectorizeFeaturesUDFTest.java | 7 +- .../geospatial/HaversineDistanceUDFTest.java | 56 +++--- .../hivemall/geospatial/Lat2TileYUDFTest.java | 18 +- .../hivemall/geospatial/Lon2TileXUDFTest.java | 18 +- .../hivemall/geospatial/TileX2LonUDFTest.java | 17 +- .../hivemall/geospatial/TileY2LatUDFTest.java | 17 +- .../knn/distance/EuclidDistanceUDFTest.java | 12 +- .../knn/similarity/CosineSimilarityUDFTest.java | 64 +++--- .../knn/similarity/DIMSUMMapperUDTFTest.java | 31 +-- .../mf/BPRMatrixFactorizationUDTFTest.java | 17 +- .../mf/MatrixFactorizationAdaGradUDTFTest.java | 8 +- .../mf/MatrixFactorizationSGDUDTFTest.java | 44 ++--- .../mix/client/MixRequestRouterTest.java | 4 +- .../java/hivemall/optimizer/OptimizerTest.java | 24 ++- .../java/hivemall/recommend/SlimUDTFTest.java | 17 +- .../hivemall/regression/AdaGradUDTFTest.java | 31 +-- .../regression/GeneralRegressorUDTFTest.java | 43 ++-- .../hivemall/sketch/bloom/BloomAndUDFTest.java | 4 +- .../hivemall/sketch/bloom/BloomOrUDFTest.java | 4 +- .../smile/classification/DecisionTreeTest.java | 10 +- .../RandomForestClassifierUDTFTest.java | 31 +-- .../smile/regression/RegressionTreeTest.java | 16 +- .../smile/tools/TreePredictUDFTest.java | 61 +++--- .../smile/tools/TreePredictUDFv1Test.java | 76 ++++---- .../statistics/MovingAverageUDTFTest.java | 8 +- .../java/hivemall/tools/TryCastUDFTest.java | 14 +- .../tools/array/ArrayAppendUDFTest.java | 29 +-- .../tools/array/ArrayElementAtUDFTest.java | 27 +-- .../tools/array/ArrayFlattenUDFTest.java | 11 +- .../hivemall/tools/array/ArraySliceUDFTest.java | 33 ++-- .../hivemall/tools/array/ArrayUnionUDFTest.java | 29 +-- .../tools/array/ConditionalEmitUDTFTest.java | 23 ++- .../tools/array/FirstElementUDFTest.java | 16 +- .../tools/array/LastElementUDFTest.java | 16 +- .../tools/array/SelectKBestUDFTest.java | 23 ++- .../hivemall/tools/json/FromJsonUDFTest.java | 8 +- .../java/hivemall/tools/json/ToJsonUDFTest.java | 10 +- .../tools/list/UDAFToOrderedListTest.java | 111 ++++++----- .../tools/map/UDAFToOrderedMapTest.java | 42 ++-- .../hivemall/tools/math/L2NormUDAFTest.java | 4 +- .../tools/matrix/TransposeAndDotUDAFTest.java | 19 +- .../hivemall/tools/vector/VectorAddUDFTest.java | 28 +-- .../hivemall/tools/vector/VectorDotUDFTest.java | 27 +-- .../topicmodel/IncrementalPLSAModelTest.java | 24 ++- .../hivemall/topicmodel/LDAPredictUDAFTest.java | 53 +++-- .../java/hivemall/topicmodel/LDAUDTFTest.java | 34 ++-- .../hivemall/topicmodel/OnlineLDAModelTest.java | 29 +-- .../topicmodel/PLSAPredictUDAFTest.java | 53 +++-- .../java/hivemall/topicmodel/PLSAUDTFTest.java | 34 ++-- .../java/hivemall/utils/codec/Base91Test.java | 5 +- .../hivemall/utils/codec/DeflateCodecTest.java | 12 +- .../collections/BoundedPriorityQueueTest.java | 36 ++-- .../utils/geospatial/GeoSpatialUtilsTest.java | 9 +- .../utils/hadoop/JsonSerdeUtilsTest.java | 78 ++++---- .../java/hivemall/utils/lang/HalfFloatTest.java | 24 +-- .../hivemall/utils/math/MatrixUtilsTest.java | 65 ++++--- .../java/hivemall/mix/server/MixServer.java | 11 +- .../hivemall/mix/server/MixServerHandler.java | 3 +- .../hivemall/mix/store/PartialArgminKLD.java | 3 +- .../java/hivemall/mix/store/PartialAverage.java | 3 +- .../java/hivemall/mix/store/SessionStore.java | 7 +- .../mix/server/MixServerHandlerTest.java | 9 +- .../java/hivemall/mix/server/MixServerTest.java | 4 +- .../java/hivemall/test/HivemallTestBase.java | 4 +- .../hivemall/nlp/tokenizer/KuromojiUDF.java | 26 +-- .../java/hivemall/nlp/tokenizer/SmartcnUDF.java | 7 +- .../hivemall/nlp/tokenizer/KuromojiUDFTest.java | 3 +- ...isticRegressionDataGeneratorUDTFWrapper.java | 3 +- .../java/hivemall/ftvec/AddBiasUDFWrapper.java | 3 +- .../ftvec/AddFeatureIndexUDFWrapper.java | 6 +- .../hivemall/ftvec/ExtractWeightUDFWrapper.java | 3 +- .../hivemall/ftvec/SortByFeatureUDFWrapper.java | 9 +- .../scaling/L2NormalizationUDFWrapper.java | 6 +- .../hivemall/knn/lsh/MinHashesUDFWrapper.java | 11 +- .../hivemall/tools/mapred/RowIdUDFWrapper.java | 3 +- .../java/hivemall/docs/FuncsListGenerator.java | 9 +- .../java/hivemall/xgboost/NativeLibLoader.java | 12 +- .../hivemall/xgboost/XGBoostPredictUDTF.java | 8 +- .../main/java/hivemall/xgboost/XGBoostUDTF.java | 14 +- .../java/hivemall/xgboost/XGBoostUtils.java | 3 +- .../XGBoostBinaryClassifierUDTF.java | 3 +- .../XGBoostMulticlassClassifierUDTF.java | 7 +- .../regression/XGBoostRegressionUDTF.java | 3 +- .../tools/XGBoostMulticlassPredictUDTF.java | 3 +- .../xgboost/tools/XGBoostPredictUDTF.java | 3 +- 332 files changed, 3055 insertions(+), 2660 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/GeneralLearnerBaseUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/GeneralLearnerBaseUDTF.java b/core/src/main/java/hivemall/GeneralLearnerBaseUDTF.java index 56fe7e0..5c3967b 100644 --- a/core/src/main/java/hivemall/GeneralLearnerBaseUDTF.java +++ b/core/src/main/java/hivemall/GeneralLearnerBaseUDTF.java @@ -166,7 +166,8 @@ public abstract class GeneralLearnerBaseUDTF extends LearnerBaseUDTF { protected Options getOptions() { Options opts = super.getOptions(); opts.addOption("loss", "loss_function", true, getLossOptionDescription()); - opts.addOption("iter", "iterations", true, "The maximum number of iterations [default: 10]"); + opts.addOption("iter", "iterations", true, + "The maximum number of iterations [default: 10]"); // conversion check opts.addOption("disable_cv", "disable_cvtest", false, "Whether to disable convergence check [default: OFF]"); @@ -188,7 +189,8 @@ public abstract class GeneralLearnerBaseUDTF extends LearnerBaseUDTF { if (cl != null) { if (cl.hasOption("loss_function")) { try { - lossFunction = LossFunctions.getLossFunction(cl.getOptionValue("loss_function")); + lossFunction = + LossFunctions.getLossFunction(cl.getOptionValue("loss_function")); } catch (Throwable e) { throw new UDFArgumentException(e.getMessage()); } @@ -312,8 +314,8 @@ public abstract class GeneralLearnerBaseUDTF extends LearnerBaseUDTF { file = File.createTempFile("hivemall_general_learner", ".sgmt"); file.deleteOnExit(); if (!file.canWrite()) { - throw new UDFArgumentException("Cannot write a temporary file: " - + file.getAbsolutePath()); + throw new UDFArgumentException( + "Cannot write a temporary file: " + file.getAbsolutePath()); } logger.info("Record training samples to a file: " + file.getAbsolutePath()); } catch (IOException ioe) { @@ -381,8 +383,8 @@ public abstract class GeneralLearnerBaseUDTF extends LearnerBaseUDTF { feature = Long.valueOf(featureStr); break; default: - throw new IllegalStateException("Unexpected feature type " + featureType - + " for feature: " + featureStr); + throw new IllegalStateException( + "Unexpected feature type " + featureType + " for feature: " + featureStr); } double value = buf.getDouble(); return new FeatureValue(feature, value); @@ -542,8 +544,8 @@ public abstract class GeneralLearnerBaseUDTF extends LearnerBaseUDTF { final long numTrainingExamples = count; final Reporter reporter = getReporter(); - final Counters.Counter iterCounter = (reporter == null) ? null : reporter.getCounter( - "hivemall.GeneralLearnerBase$Counter", "iteration"); + final Counters.Counter iterCounter = (reporter == null) ? null + : reporter.getCounter("hivemall.GeneralLearnerBase$Counter", "iteration"); try { if (dst.getPosition() == 0L) {// run iterations w/o temporary file @@ -578,13 +580,12 @@ public abstract class GeneralLearnerBaseUDTF extends LearnerBaseUDTF { break; } } - logger.info("Performed " - + cvState.getCurrentIteration() - + " iterations of " + logger.info("Performed " + cvState.getCurrentIteration() + " iterations of " + NumberUtils.formatNumber(numTrainingExamples) + " training examples on memory (thus " - + NumberUtils.formatNumber(numTrainingExamples - * cvState.getCurrentIteration()) + " training updates in total) "); + + NumberUtils.formatNumber( + numTrainingExamples * cvState.getCurrentIteration()) + + " training updates in total) "); } else {// read training examples in the temporary file and invoke train for each example // write training examples in buffer to a temporary file if (buf.remaining() > 0) { @@ -593,8 +594,8 @@ public abstract class GeneralLearnerBaseUDTF extends LearnerBaseUDTF { try { dst.flush(); } catch (IOException e) { - throw new HiveException("Failed to flush a file: " - + dst.getFile().getAbsolutePath(), e); + throw new HiveException( + "Failed to flush a file: " + dst.getFile().getAbsolutePath(), e); } if (logger.isInfoEnabled()) { File tmpFile = dst.getFile(); @@ -619,8 +620,8 @@ public abstract class GeneralLearnerBaseUDTF extends LearnerBaseUDTF { try { bytesRead = dst.read(buf); } catch (IOException e) { - throw new HiveException("Failed to read a file: " - + dst.getFile().getAbsolutePath(), e); + throw new HiveException( + "Failed to read a file: " + dst.getFile().getAbsolutePath(), e); } if (bytesRead == 0) { // reached file EOF break; @@ -644,7 +645,8 @@ public abstract class GeneralLearnerBaseUDTF extends LearnerBaseUDTF { } int featureVectorLength = buf.getInt(); - final FeatureValue[] featureVector = new FeatureValue[featureVectorLength]; + final FeatureValue[] featureVector = + new FeatureValue[featureVectorLength]; for (int j = 0; j < featureVectorLength; j++) { featureVector[j] = readFeatureValue(buf, featureType); } @@ -664,13 +666,12 @@ public abstract class GeneralLearnerBaseUDTF extends LearnerBaseUDTF { break; } } - logger.info("Performed " - + cvState.getCurrentIteration() - + " iterations of " + logger.info("Performed " + cvState.getCurrentIteration() + " iterations of " + NumberUtils.formatNumber(numTrainingExamples) + " training examples on a secondary storage (thus " - + NumberUtils.formatNumber(numTrainingExamples - * cvState.getCurrentIteration()) + " training updates in total)"); + + NumberUtils.formatNumber( + numTrainingExamples * cvState.getCurrentIteration()) + + " training updates in total)"); } } catch (Throwable e) { throw new HiveException("Exception caused in the iterative training", e); @@ -679,8 +680,8 @@ public abstract class GeneralLearnerBaseUDTF extends LearnerBaseUDTF { try { dst.close(true); } catch (IOException e) { - throw new HiveException("Failed to close a file: " - + dst.getFile().getAbsolutePath(), e); + throw new HiveException( + "Failed to close a file: " + dst.getFile().getAbsolutePath(), e); } this.inputBuf = null; this.fileIO = null; http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/LearnerBaseUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/LearnerBaseUDTF.java b/core/src/main/java/hivemall/LearnerBaseUDTF.java index 356d739..5407d0e 100644 --- a/core/src/main/java/hivemall/LearnerBaseUDTF.java +++ b/core/src/main/java/hivemall/LearnerBaseUDTF.java @@ -126,18 +126,19 @@ public abstract class LearnerBaseUDTF extends UDTFWithOptions { } disableHalfFloat = cl.hasOption("disable_halffloat"); - miniBatchSize = Primitives.parseInt(cl.getOptionValue("mini_batch_size"), miniBatchSize); + miniBatchSize = + Primitives.parseInt(cl.getOptionValue("mini_batch_size"), miniBatchSize); if (miniBatchSize <= 0) { - throw new UDFArgumentException("mini_batch_size must be greater than 0: " - + miniBatchSize); + throw new UDFArgumentException( + "mini_batch_size must be greater than 0: " + miniBatchSize); } mixConnectInfo = cl.getOptionValue("mix"); mixSessionName = cl.getOptionValue("mix_session"); mixThreshold = Primitives.parseInt(cl.getOptionValue("mix_threshold"), 3); if (mixThreshold > Byte.MAX_VALUE) { - throw new UDFArgumentException("mix_threshold must be in range (0,127]: " - + mixThreshold); + throw new UDFArgumentException( + "mix_threshold must be in range (0,127]: " + mixThreshold); } mixCancel = cl.hasOption("mix_cancel"); ssl = cl.hasOption("ssl"); @@ -181,8 +182,8 @@ public abstract class LearnerBaseUDTF extends UDTFWithOptions { } } else { int initModelSize = getInitialModelSize(); - logger.info("Build a sparse model with initial with " + initModelSize - + " initial dimensions"); + logger.info( + "Build a sparse model with initial with " + initModelSize + " initial dimensions"); model = new SparseModel(initModelSize, useCovar); } if (mixConnectInfo != null) { @@ -212,8 +213,8 @@ public abstract class LearnerBaseUDTF extends UDTFWithOptions { } } else { int initModelSize = getInitialModelSize(); - logger.info("Build a sparse model with initial with " + initModelSize - + " initial dimensions"); + logger.info( + "Build a sparse model with initial with " + initModelSize + " initial dimensions"); model = new NewSparseModel(initModelSize, useCovar); } if (mixConnectInfo != null) { http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/anomaly/ChangeFinderUDF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/anomaly/ChangeFinderUDF.java b/core/src/main/java/hivemall/anomaly/ChangeFinderUDF.java index 47d41e9..5b15edd 100644 --- a/core/src/main/java/hivemall/anomaly/ChangeFinderUDF.java +++ b/core/src/main/java/hivemall/anomaly/ChangeFinderUDF.java @@ -46,8 +46,7 @@ import org.apache.hadoop.hive.serde2.objectinspector.PrimitiveObjectInspector; import org.apache.hadoop.hive.serde2.objectinspector.primitive.PrimitiveObjectInspectorFactory; import org.apache.hadoop.io.BooleanWritable; -@Description( - name = "changefinder", +@Description(name = "changefinder", value = "_FUNC_(double|array<double> x [, const string options])" + " - Returns outlier/change-point scores and decisions using ChangeFinder." + " It will return a tuple <double outlier_score, double changepoint_score [, boolean is_anomaly [, boolean is_changepoint]]") @@ -86,15 +85,9 @@ public final class ChangeFinderUDF extends UDFWithOptions { "Number of past samples to include for calculating outlier score [default: 7]"); opts.addOption("T2", "y_window", true, "Number of past samples to include for calculating change-point score [default: 7]"); - opts.addOption( - "outlier_threshold", - "x_threshold", - true, + opts.addOption("outlier_threshold", "x_threshold", true, "Score threshold (inclusive) for determining outlier existence [default: -1, do not output decision]"); - opts.addOption( - "changepoint_threshold", - "y_threshold", - true, + opts.addOption("changepoint_threshold", "y_threshold", true, "Score threshold (inclusive) for determining change-point existence [default: -1, do not output decision]"); opts.addOption("loss1", "lossfunc1", true, "Loss function for outlier scoring [default: hellinger, logloss]"); @@ -116,10 +109,10 @@ public final class ChangeFinderUDF extends UDFWithOptions { cl.getOptionValue("outlier_threshold"), _params.outlierThreshold); this._params.changepointThreshold = Primitives.parseDouble( cl.getOptionValue("changepoint_threshold"), _params.changepointThreshold); - this._params.lossFunc1 = LossFunction.resolve(cl.getOptionValue("lossfunc1", - LossFunction.hellinger.name())); - this._params.lossFunc2 = LossFunction.resolve(cl.getOptionValue("lossfunc2", - LossFunction.hellinger.name())); + this._params.lossFunc1 = + LossFunction.resolve(cl.getOptionValue("lossfunc1", LossFunction.hellinger.name())); + this._params.lossFunc2 = + LossFunction.resolve(cl.getOptionValue("lossfunc2", LossFunction.hellinger.name())); Preconditions.checkArgument(_params.k >= 2, "K must be greater than 1: " + _params.k); Preconditions.checkArgument(_params.r1 > 0.d && _params.r1 < 1.d, http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/anomaly/SDAR1D.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/anomaly/SDAR1D.java b/core/src/main/java/hivemall/anomaly/SDAR1D.java index 5b2e76e..ad2c90b 100644 --- a/core/src/main/java/hivemall/anomaly/SDAR1D.java +++ b/core/src/main/java/hivemall/anomaly/SDAR1D.java @@ -65,8 +65,8 @@ public final class SDAR1D { public double update(@Nonnull final double[] x, final int k) { Preconditions.checkArgument(x.length >= 1, "x.length MUST be greater than 1: ", x.length); Preconditions.checkArgument(k >= 0, "k MUST be greater than or equals to 0: ", k); - Preconditions.checkArgument(k < _C.length, "k MUST be less than |C| but ", "k=", k - + ", |C|=", _C.length); + Preconditions.checkArgument(k < _C.length, "k MUST be less than |C| but ", "k=", + k + ", |C|=", _C.length); final double x_t = x[0]; http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/anomaly/SDAR2D.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/anomaly/SDAR2D.java b/core/src/main/java/hivemall/anomaly/SDAR2D.java index 7198067..a6e866c 100644 --- a/core/src/main/java/hivemall/anomaly/SDAR2D.java +++ b/core/src/main/java/hivemall/anomaly/SDAR2D.java @@ -70,8 +70,8 @@ public final class SDAR2D { public RealVector update(@Nonnull final ArrayRealVector[] x, final int k) { Preconditions.checkArgument(x.length >= 1, "x.length MUST be greater than 1: " + x.length); Preconditions.checkArgument(k >= 0, "k MUST be greater than or equals to 0: ", k); - Preconditions.checkArgument(k < _C.length, "k MUST be less than |C| but " + "k=" + k - + ", |C|=" + _C.length); + Preconditions.checkArgument(k < _C.length, + "k MUST be less than |C| but " + "k=" + k + ", |C|=" + _C.length); final ArrayRealVector x_t = x[0]; final int dims = x_t.getDimension(); @@ -145,8 +145,8 @@ public final class SDAR2D { // update model covariance // â := (1-r) â + r (x - \hat{x}) (x - \hat{x})' RealVector xEstimateResidual = x_t.subtract(x_hat); - this._sigma = _sigma.scalarMultiply(1.d - _r).add( - xEstimateResidual.mapMultiply(_r).outerProduct(xEstimateResidual)); + this._sigma = _sigma.scalarMultiply(1.d - _r) + .add(xEstimateResidual.mapMultiply(_r).outerProduct(xEstimateResidual)); return x_hat; } http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/anomaly/SingularSpectrumTransform.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/anomaly/SingularSpectrumTransform.java b/core/src/main/java/hivemall/anomaly/SingularSpectrumTransform.java index 1936da4..1bb585d 100644 --- a/core/src/main/java/hivemall/anomaly/SingularSpectrumTransform.java +++ b/core/src/main/java/hivemall/anomaly/SingularSpectrumTransform.java @@ -154,8 +154,8 @@ final class SingularSpectrumTransform implements SingularSpectrumTransformInterf RealMatrix Q = svdG.getU(); // find the largest singular value for the r principal components - RealMatrix UTQ = UT.getSubMatrix(0, r - 1, 0, window - 1).multiply( - Q.getSubMatrix(0, window - 1, 0, r - 1)); + RealMatrix UTQ = UT.getSubMatrix(0, r - 1, 0, window - 1) + .multiply(Q.getSubMatrix(0, window - 1, 0, r - 1)); SingularValueDecomposition svdUTQ = new SingularValueDecomposition(UTQ); double[] s = svdUTQ.getSingularValues(); http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/anomaly/SingularSpectrumTransformUDF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/anomaly/SingularSpectrumTransformUDF.java b/core/src/main/java/hivemall/anomaly/SingularSpectrumTransformUDF.java index 3369edc..2c40158 100644 --- a/core/src/main/java/hivemall/anomaly/SingularSpectrumTransformUDF.java +++ b/core/src/main/java/hivemall/anomaly/SingularSpectrumTransformUDF.java @@ -49,17 +49,14 @@ import org.apache.hadoop.io.BooleanWritable; * * References: * <ul> - * <li>T. Ide and K. Inoue, - * "Knowledge Discovery from Heterogeneous Dynamic Systems using Change-Point Correlations", SDM'05. - * </li> + * <li>T. Ide and K. Inoue, "Knowledge Discovery from Heterogeneous Dynamic Systems using + * Change-Point Correlations", SDM'05.</li> * <li>T. Ide and K. Tsuda, "Change-point detection using Krylov subspace learning", SDM'07.</li> * </ul> */ -@Description( - name = "sst", - value = "_FUNC_(double|array<double> x [, const string options])" - + " - Returns change-point scores and decisions using Singular Spectrum Transformation (SST)." - + " It will return a tuple <double changepoint_score [, boolean is_changepoint]>") +@Description(name = "sst", value = "_FUNC_(double|array<double> x [, const string options])" + + " - Returns change-point scores and decisions using Singular Spectrum Transformation (SST)." + + " It will return a tuple <double changepoint_score [, boolean is_changepoint]>") @UDFType(deterministic = false, stateful = true) @Since(version = "0.5-rc.1") public final class SingularSpectrumTransformUDF extends UDFWithOptions { @@ -93,16 +90,10 @@ public final class SingularSpectrumTransformUDF extends UDFWithOptions { "Offset of the current windows from the updating sample [default: `-w` = -30]"); opts.addOption("r", "n_component", true, "Number of singular vectors (i.e. principal components) [default: 3]"); - opts.addOption( - "k", - "n_dim", - true, + opts.addOption("k", "n_dim", true, "Number of dimensions for the Krylov subspaces [default: 5 (`2*r` if `r` is even, `2*r-1` otherwise)]"); opts.addOption("score", "scorefunc", true, "Score function [default: svd, ika]"); - opts.addOption( - "th", - "threshold", - true, + opts.addOption("th", "threshold", true, "Score threshold (inclusive) for determining change-point existence [default: -1, do not output decision]"); return opts; } @@ -119,15 +110,15 @@ public final class SingularSpectrumTransformUDF extends UDFWithOptions { this._params.k = Primitives.parseInt(cl.getOptionValue("k"), (_params.r % 2 == 0) ? (2 * _params.r) : (2 * _params.r - 1)); - this._params.scoreFunc = ScoreFunction.resolve(cl.getOptionValue("scorefunc", - ScoreFunction.svd.name())); + this._params.scoreFunc = + ScoreFunction.resolve(cl.getOptionValue("scorefunc", ScoreFunction.svd.name())); if ((_params.w != _params.n || _params.w != _params.m) && _params.scoreFunc == ScoreFunction.ika) { throw new UDFArgumentException("IKA-based efficient SST requires w = n = m"); } - this._params.changepointThreshold = Primitives.parseDouble(cl.getOptionValue("th"), - _params.changepointThreshold); + this._params.changepointThreshold = + Primitives.parseDouble(cl.getOptionValue("th"), _params.changepointThreshold); Preconditions.checkArgument(_params.w >= 2, UDFArgumentException.class, "w must be greater than 1: " + _params.w); @@ -137,8 +128,9 @@ public final class SingularSpectrumTransformUDF extends UDFWithOptions { "k must be greater than 0: " + _params.k); Preconditions.checkArgument(_params.k >= _params.r, UDFArgumentException.class, "k must be equals to or greater than r: k=" + _params.k + ", r" + _params.r); - Preconditions.checkArgument(_params.changepointThreshold > 0.d - && _params.changepointThreshold < 1.d, UDFArgumentException.class, + Preconditions.checkArgument( + _params.changepointThreshold > 0.d && _params.changepointThreshold < 1.d, + UDFArgumentException.class, "changepointThreshold must be in range (0, 1): " + _params.changepointThreshold); return cl; http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/classifier/AROWClassifierUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/classifier/AROWClassifierUDTF.java b/core/src/main/java/hivemall/classifier/AROWClassifierUDTF.java index 959aaaa..30cd909 100644 --- a/core/src/main/java/hivemall/classifier/AROWClassifierUDTF.java +++ b/core/src/main/java/hivemall/classifier/AROWClassifierUDTF.java @@ -41,8 +41,7 @@ import org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector; * In Proc. NIPS, 2009. * </pre> */ -@Description( - name = "train_arow", +@Description(name = "train_arow", value = "_FUNC_(list<string|int|bigint> features, int label [, const string options])" + " - Returns a relation consists of <string|int|bigint feature, float weight, float covar>", extended = "Build a prediction model by Adaptive Regularization of Weight Vectors (AROW) binary classifier") @@ -149,8 +148,7 @@ public class AROWClassifierUDTF extends BinaryOnlineClassifierUDTF { return new WeightValueWithCovar(new_w, new_cov); } - @Description( - name = "train_arowh", + @Description(name = "train_arowh", value = "_FUNC_(list<string|int|bigint> features, int label [, const string options])" + " - Returns a relation consists of <string|int|bigint feature, float weight, float covar>", extended = "Build a prediction model by AROW binary classifier using hinge loss") @@ -176,8 +174,8 @@ public class AROWClassifierUDTF extends BinaryOnlineClassifierUDTF { if (c_str != null) { c = Float.parseFloat(c_str); if (!(c > 0.f)) { - throw new UDFArgumentException("Aggressiveness parameter C must be C > 0: " - + c); + throw new UDFArgumentException( + "Aggressiveness parameter C must be C > 0: " + c); } } } http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/classifier/AdaGradRDAUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/classifier/AdaGradRDAUDTF.java b/core/src/main/java/hivemall/classifier/AdaGradRDAUDTF.java index 6adeaa9..8bb280d 100644 --- a/core/src/main/java/hivemall/classifier/AdaGradRDAUDTF.java +++ b/core/src/main/java/hivemall/classifier/AdaGradRDAUDTF.java @@ -138,7 +138,8 @@ public final class AdaGradRDAUDTF extends BinaryOnlineClassifierUDTF { } else { // x_{t,i} = -sign(u_{t,i}) * \frac{\eta t}{\sqrt{G_{t,ii}}}(|u_{t,i}|/t - \lambda) float weight = -1.f * sign * eta * t * meansOfGradients / (float) Math.sqrt(sum_sqgrad); - IWeightValue new_w = new WeightValueParamsF2(weight, scaled_sum_sqgrad, scaled_sum_grad); + IWeightValue new_w = + new WeightValueParamsF2(weight, scaled_sum_sqgrad, scaled_sum_grad); model.set(x, new_w); } } http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/classifier/BinaryOnlineClassifierUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/classifier/BinaryOnlineClassifierUDTF.java b/core/src/main/java/hivemall/classifier/BinaryOnlineClassifierUDTF.java index 2f4db3a..ffe3186 100644 --- a/core/src/main/java/hivemall/classifier/BinaryOnlineClassifierUDTF.java +++ b/core/src/main/java/hivemall/classifier/BinaryOnlineClassifierUDTF.java @@ -76,9 +76,8 @@ public abstract class BinaryOnlineClassifierUDTF extends LearnerBaseUDTF { @Override public StructObjectInspector initialize(ObjectInspector[] argOIs) throws UDFArgumentException { if (argOIs.length < 2) { - throw new UDFArgumentException( - getClass().getSimpleName() - + " takes 2 arguments: List<Int|BigInt|Text> features, int label [, constant string options]"); + throw new UDFArgumentException(getClass().getSimpleName() + + " takes 2 arguments: List<Int|BigInt|Text> features, int label [, constant string options]"); } PrimitiveObjectInspector featureInputOI = processFeaturesOI(argOIs[0]); this.labelOI = HiveUtils.asIntCompatibleOI(argOIs[1]); http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/classifier/ConfidenceWeightedUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/classifier/ConfidenceWeightedUDTF.java b/core/src/main/java/hivemall/classifier/ConfidenceWeightedUDTF.java index 5c268d2..98c35c5 100644 --- a/core/src/main/java/hivemall/classifier/ConfidenceWeightedUDTF.java +++ b/core/src/main/java/hivemall/classifier/ConfidenceWeightedUDTF.java @@ -43,8 +43,7 @@ import org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector; * * @link http://dl.acm.org/citation.cfm?id=1390190 */ -@Description( - name = "train_cw", +@Description(name = "train_cw", value = "_FUNC_(list<string|int|bigint> features, int label [, const string options])" + " - Returns a relation consists of <string|int|bigint feature, float weight, float covar>", extended = "Build a prediction model by Confidence-Weighted (CW) binary classifier") http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/classifier/KPAPredictUDAF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/classifier/KPAPredictUDAF.java b/core/src/main/java/hivemall/classifier/KPAPredictUDAF.java index 72409d9..9196fe3 100644 --- a/core/src/main/java/hivemall/classifier/KPAPredictUDAF.java +++ b/core/src/main/java/hivemall/classifier/KPAPredictUDAF.java @@ -40,8 +40,7 @@ import org.apache.hadoop.hive.serde2.objectinspector.PrimitiveObjectInspector; import org.apache.hadoop.hive.serde2.objectinspector.primitive.PrimitiveObjectInspectorFactory; import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo; -@Description( - name = "kpa_predict", +@Description(name = "kpa_predict", value = "_FUNC_(@Nonnull double xh, @Nonnull double xk, @Nullable float w0, @Nonnull float w1, @Nonnull float w2, @Nullable float w3)" + " - Returns a prediction value in Double") public final class KPAPredictUDAF extends AbstractGenericUDAFResolver { @@ -54,28 +53,28 @@ public final class KPAPredictUDAF extends AbstractGenericUDAFResolver { + parameters.length); } if (!HiveUtils.isNumberTypeInfo(parameters[0])) { - throw new UDFArgumentTypeException(0, "Number type is expected for xh (1st argument): " - + parameters[0].getTypeName()); + throw new UDFArgumentTypeException(0, + "Number type is expected for xh (1st argument): " + parameters[0].getTypeName()); } if (!HiveUtils.isNumberTypeInfo(parameters[1])) { - throw new UDFArgumentTypeException(1, "Number type is expected for xk (2nd argument): " - + parameters[1].getTypeName()); + throw new UDFArgumentTypeException(1, + "Number type is expected for xk (2nd argument): " + parameters[1].getTypeName()); } if (!HiveUtils.isNumberTypeInfo(parameters[2])) { - throw new UDFArgumentTypeException(2, "Number type is expected for w0 (3rd argument): " - + parameters[2].getTypeName()); + throw new UDFArgumentTypeException(2, + "Number type is expected for w0 (3rd argument): " + parameters[2].getTypeName()); } if (!HiveUtils.isNumberTypeInfo(parameters[3])) { - throw new UDFArgumentTypeException(3, "Number type is expected for w1 (4th argument): " - + parameters[3].getTypeName()); + throw new UDFArgumentTypeException(3, + "Number type is expected for w1 (4th argument): " + parameters[3].getTypeName()); } if (!HiveUtils.isNumberTypeInfo(parameters[4])) { - throw new UDFArgumentTypeException(4, "Number type is expected for w2 (5th argument): " - + parameters[4].getTypeName()); + throw new UDFArgumentTypeException(4, + "Number type is expected for w2 (5th argument): " + parameters[4].getTypeName()); } if (!HiveUtils.isNumberTypeInfo(parameters[5])) { - throw new UDFArgumentTypeException(5, "Number type is expected for w3 (6th argument): " - + parameters[5].getTypeName()); + throw new UDFArgumentTypeException(5, + "Number type is expected for w3 (6th argument): " + parameters[5].getTypeName()); } return new Evaluator(); @@ -126,10 +125,10 @@ public final class KPAPredictUDAF extends AbstractGenericUDAFResolver { final AggrBuffer aggr = (AggrBuffer) agg; - if (parameters[0] /* xh */!= null) { + if (parameters[0] /* xh */ != null) { double xh = HiveUtils.getDouble(parameters[0], xhOI); - if (parameters[1] /* xk */!= null) { - if (parameters[5] /* w3hk */== null) { + if (parameters[1] /* xk */ != null) { + if (parameters[5] /* w3hk */ == null) { return; } // xh, xk, w3hk @@ -137,7 +136,7 @@ public final class KPAPredictUDAF extends AbstractGenericUDAFResolver { double w3hk = HiveUtils.getDouble(parameters[5], w3OI); aggr.addW3(xh, xk, w3hk); } else { - if (parameters[3] /* w1h */== null) { + if (parameters[3] /* w1h */ == null) { return; } // xh, w1h, w2h @@ -146,7 +145,7 @@ public final class KPAPredictUDAF extends AbstractGenericUDAFResolver { double w2h = HiveUtils.getDouble(parameters[4], w2OI); aggr.addW1W2(xh, w1h, w2h); } - } else if (parameters[2] /* w0 */!= null) { + } else if (parameters[2] /* w0 */ != null) { // w0 double w0 = HiveUtils.getDouble(parameters[2], w0OI); aggr.addW0(w0); http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/classifier/KernelExpansionPassiveAggressiveUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/classifier/KernelExpansionPassiveAggressiveUDTF.java b/core/src/main/java/hivemall/classifier/KernelExpansionPassiveAggressiveUDTF.java index 731083d..4ecc028 100644 --- a/core/src/main/java/hivemall/classifier/KernelExpansionPassiveAggressiveUDTF.java +++ b/core/src/main/java/hivemall/classifier/KernelExpansionPassiveAggressiveUDTF.java @@ -116,7 +116,8 @@ public final class KernelExpansionPassiveAggressiveUDTF extends BinaryOnlineClas if (c_str != null) { c = Float.parseFloat(c_str); if (c <= 0.f) { - throw new UDFArgumentException("Aggressiveness parameter C must be C > 0: " + c); + throw new UDFArgumentException( + "Aggressiveness parameter C must be C > 0: " + c); } } algo = cl.getOptionValue("algo", algo); http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/classifier/PassiveAggressiveUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/classifier/PassiveAggressiveUDTF.java b/core/src/main/java/hivemall/classifier/PassiveAggressiveUDTF.java index e4146ce..02a8c6a 100644 --- a/core/src/main/java/hivemall/classifier/PassiveAggressiveUDTF.java +++ b/core/src/main/java/hivemall/classifier/PassiveAggressiveUDTF.java @@ -68,8 +68,7 @@ public class PassiveAggressiveUDTF extends BinaryOnlineClassifierUDTF { return loss / margin.getSquaredNorm(); } - @Description( - name = "train_pa1", + @Description(name = "train_pa1", value = "_FUNC_(list<string|int|bigint> features, int label [, const string options])" + " - Returns a relation consists of <string|int|bigint feature, float weight>", extended = "Build a prediction model by Passive-Aggressive 1 (PA-1) binary classifier") @@ -95,8 +94,8 @@ public class PassiveAggressiveUDTF extends BinaryOnlineClassifierUDTF { if (c_str != null) { c = Float.parseFloat(c_str); if (!(c > 0.f)) { - throw new UDFArgumentException("Aggressiveness parameter C must be C > 0: " - + c); + throw new UDFArgumentException( + "Aggressiveness parameter C must be C > 0: " + c); } } } @@ -114,8 +113,7 @@ public class PassiveAggressiveUDTF extends BinaryOnlineClassifierUDTF { } - @Description( - name = "train_pa2", + @Description(name = "train_pa2", value = "_FUNC_(list<string|int|bigint> features, int label [, const string options])" + " - Returns a relation consists of <string|int|bigint feature, float weight>", extended = "Build a prediction model by Passive-Aggressive 2 (PA-2) binary classifier") http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/classifier/SoftConfideceWeightedUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/classifier/SoftConfideceWeightedUDTF.java b/core/src/main/java/hivemall/classifier/SoftConfideceWeightedUDTF.java index e95d3ba..449b213 100644 --- a/core/src/main/java/hivemall/classifier/SoftConfideceWeightedUDTF.java +++ b/core/src/main/java/hivemall/classifier/SoftConfideceWeightedUDTF.java @@ -102,7 +102,8 @@ public abstract class SoftConfideceWeightedUDTF extends BinaryOnlineClassifierUD if (c_str != null) { c = Float.parseFloat(c_str); if (!(c > 0.f)) { - throw new UDFArgumentException("Aggressiveness parameter C must be C > 0: " + c); + throw new UDFArgumentException( + "Aggressiveness parameter C must be C > 0: " + c); } } } @@ -145,8 +146,7 @@ public abstract class SoftConfideceWeightedUDTF extends BinaryOnlineClassifierUD protected abstract float getBeta(PredictionResult margin, float alpha); - @Description( - name = "train_scw", + @Description(name = "train_scw", value = "_FUNC_(list<string|int|bigint> features, int label [, const string options])" + " - Returns a relation consists of <string|int|bigint feature, float weight, float covar>", extended = "Build a prediction model by Soft Confidence-Weighted (SCW-1) binary classifier") @@ -172,10 +172,8 @@ public abstract class SoftConfideceWeightedUDTF extends BinaryOnlineClassifierUD float m = margin.getScore(); float var = margin.getVariance(); - float alpha_numer = -m - * psi - + (float) Math.sqrt((m * m * squared_phi * squared_phi / 4.f) - + (var * squared_phi * zeta)); + float alpha_numer = -m * psi + (float) Math.sqrt( + (m * m * squared_phi * squared_phi / 4.f) + (var * squared_phi * zeta)); float alpha_denom = var * zeta; if (alpha_denom == 0.f) { return 0.f; @@ -210,8 +208,7 @@ public abstract class SoftConfideceWeightedUDTF extends BinaryOnlineClassifierUD } - @Description( - name = "train_scw2", + @Description(name = "train_scw2", value = "_FUNC_(list<string|int|bigint> features, int label [, const string options])" + " - Returns a relation consists of <string|int|bigint feature, float weight, float covar>", extended = "Build a prediction model by Soft Confidence-Weighted 2 (SCW-2) binary classifier") http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/classifier/multiclass/MulticlassAROWClassifierUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/classifier/multiclass/MulticlassAROWClassifierUDTF.java b/core/src/main/java/hivemall/classifier/multiclass/MulticlassAROWClassifierUDTF.java index 1387803..c3fce21 100644 --- a/core/src/main/java/hivemall/classifier/multiclass/MulticlassAROWClassifierUDTF.java +++ b/core/src/main/java/hivemall/classifier/multiclass/MulticlassAROWClassifierUDTF.java @@ -41,8 +41,7 @@ import org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector; * In Proc. NIPS, 2009. * </pre> */ -@Description( - name = "train_multiclass_arow", +@Description(name = "train_multiclass_arow", value = "_FUNC_(list<string|int|bigint> features, {int|string} label [, const string options])" + " - Returns a relation consists of <{int|string} label, {string|int|bigint} feature, float weight, float covar>", extended = "Build a prediction model by Adaptive Regularization of Weight Vectors (AROW) multiclass classifier") @@ -116,8 +115,8 @@ public class MulticlassAROWClassifierUDTF extends MulticlassOnlineClassifierUDTF final Object missed_label, final float alpha, final float beta) { assert (actual_label != null); if (actual_label.equals(missed_label)) { - throw new IllegalArgumentException("Actual label equals to missed label: " - + actual_label); + throw new IllegalArgumentException( + "Actual label equals to missed label: " + actual_label); } PredictionModel model2add = label2model.get(actual_label); @@ -142,13 +141,14 @@ public class MulticlassAROWClassifierUDTF extends MulticlassOnlineClassifierUDTF final float v = f.getValueAsFloat(); IWeightValue old_correctclass_w = model2add.get(k); - IWeightValue new_correctclass_w = getNewWeight(old_correctclass_w, v, alpha, beta, true); + IWeightValue new_correctclass_w = + getNewWeight(old_correctclass_w, v, alpha, beta, true); model2add.set(k, new_correctclass_w); if (model2sub != null) { IWeightValue old_wrongclass_w = model2sub.get(k); - IWeightValue new_wrongclass_w = getNewWeight(old_wrongclass_w, v, alpha, beta, - false); + IWeightValue new_wrongclass_w = + getNewWeight(old_wrongclass_w, v, alpha, beta, false); model2sub.set(k, new_wrongclass_w); } } @@ -173,8 +173,7 @@ public class MulticlassAROWClassifierUDTF extends MulticlassOnlineClassifierUDTF return new WeightValueWithCovar(new_w, new_cov); } - @Description( - name = "train_multiclass_arowh", + @Description(name = "train_multiclass_arowh", value = "_FUNC_(list<string|int|bigint> features, int|string label [, const string options])" + " - Returns a relation consists of <int|string label, string|int|bigint feature, float weight, float covar>", extended = "Build a prediction model by Adaptive Regularization of Weight Vectors (AROW) multiclass classifier using hinge loss") @@ -200,8 +199,8 @@ public class MulticlassAROWClassifierUDTF extends MulticlassOnlineClassifierUDTF if (c_str != null) { c = Float.parseFloat(c_str); if (!(c > 0.f)) { - throw new UDFArgumentException("Aggressiveness parameter C must be C > 0: " - + c); + throw new UDFArgumentException( + "Aggressiveness parameter C must be C > 0: " + c); } } } http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/classifier/multiclass/MulticlassConfidenceWeightedUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/classifier/multiclass/MulticlassConfidenceWeightedUDTF.java b/core/src/main/java/hivemall/classifier/multiclass/MulticlassConfidenceWeightedUDTF.java index 033c9b8..298ca3b 100644 --- a/core/src/main/java/hivemall/classifier/multiclass/MulticlassConfidenceWeightedUDTF.java +++ b/core/src/main/java/hivemall/classifier/multiclass/MulticlassConfidenceWeightedUDTF.java @@ -47,8 +47,7 @@ import org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector; * @link http://dl.acm.org/citation.cfm?id=1390190 * @link http://dl.acm.org/citation.cfm?id=1699577 */ -@Description( - name = "train_multiclass_cw", +@Description(name = "train_multiclass_cw", value = "_FUNC_(list<string|int|bigint> features, {int|string} label [, const string options])" + " - Returns a relation consists of <{int|string} label, {string|int|bigint} feature, float weight, float covar>", extended = "Build a prediction model by Confidence-Weighted (CW) multiclass classifier") @@ -137,8 +136,8 @@ public class MulticlassConfidenceWeightedUDTF extends MulticlassOnlineClassifier Object missed_label) { assert (actual_label != null); if (actual_label.equals(missed_label)) { - throw new IllegalArgumentException("Actual label equals to missed label: " - + actual_label); + throw new IllegalArgumentException( + "Actual label equals to missed label: " + actual_label); } PredictionModel model2add = label2model.get(actual_label); @@ -168,7 +167,8 @@ public class MulticlassConfidenceWeightedUDTF extends MulticlassOnlineClassifier if (model2sub != null) { IWeightValue old_wrongclass_w = model2sub.get(k); - IWeightValue new_wrongclass_w = getNewWeight(old_wrongclass_w, v, alpha, phi, false); + IWeightValue new_wrongclass_w = + getNewWeight(old_wrongclass_w, v, alpha, phi, false); model2sub.set(k, new_wrongclass_w); } } http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/classifier/multiclass/MulticlassOnlineClassifierUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/classifier/multiclass/MulticlassOnlineClassifierUDTF.java b/core/src/main/java/hivemall/classifier/multiclass/MulticlassOnlineClassifierUDTF.java index 08a040b..e547d4e 100644 --- a/core/src/main/java/hivemall/classifier/multiclass/MulticlassOnlineClassifierUDTF.java +++ b/core/src/main/java/hivemall/classifier/multiclass/MulticlassOnlineClassifierUDTF.java @@ -88,17 +88,16 @@ public abstract class MulticlassOnlineClassifierUDTF extends LearnerBaseUDTF { @Override public StructObjectInspector initialize(ObjectInspector[] argOIs) throws UDFArgumentException { if (argOIs.length < 2) { - throw new UDFArgumentException( - getClass().getSimpleName() - + " takes 2 arguments: List<Int|BigInt|Text> features, {Int|BitInt|Text} label [, constant text options]"); + throw new UDFArgumentException(getClass().getSimpleName() + + " takes 2 arguments: List<Int|BigInt|Text> features, {Int|BitInt|Text} label [, constant text options]"); } PrimitiveObjectInspector featureInputOI = processFeaturesOI(argOIs[0]); this.labelInputOI = HiveUtils.asPrimitiveObjectInspector(argOIs[1]); String labelTypeName = labelInputOI.getTypeName(); if (!STRING_TYPE_NAME.equals(labelTypeName) && !INT_TYPE_NAME.equals(labelTypeName) && !BIGINT_TYPE_NAME.equals(labelTypeName)) { - throw new UDFArgumentTypeException(0, "label must be a type [Int|BigInt|Text]: " - + labelTypeName); + throw new UDFArgumentTypeException(0, + "label must be a type [Int|BigInt|Text]: " + labelTypeName); } processOptions(argOIs); @@ -343,8 +342,8 @@ public abstract class MulticlassOnlineClassifierUDTF extends LearnerBaseUDTF { Object missed_label) { assert (actual_label != null); if (actual_label.equals(missed_label)) { - throw new IllegalArgumentException("Actual label equals to missed label: " - + actual_label); + throw new IllegalArgumentException( + "Actual label equals to missed label: " + actual_label); } PredictionModel model2add = label2model.get(actual_label); @@ -496,9 +495,12 @@ public abstract class MulticlassOnlineClassifierUDTF extends LearnerBaseUDTF { StructField c1ref = lineOI.getStructFieldRef("c1"); StructField c2ref = lineOI.getStructFieldRef("c2"); StructField c3ref = lineOI.getStructFieldRef("c3"); - PrimitiveObjectInspector c1refOI = (PrimitiveObjectInspector) c1ref.getFieldObjectInspector(); - PrimitiveObjectInspector c2refOI = (PrimitiveObjectInspector) c2ref.getFieldObjectInspector(); - FloatObjectInspector c3refOI = (FloatObjectInspector) c3ref.getFieldObjectInspector(); + PrimitiveObjectInspector c1refOI = + (PrimitiveObjectInspector) c1ref.getFieldObjectInspector(); + PrimitiveObjectInspector c2refOI = + (PrimitiveObjectInspector) c2ref.getFieldObjectInspector(); + FloatObjectInspector c3refOI = + (FloatObjectInspector) c3ref.getFieldObjectInspector(); BufferedReader reader = null; try { @@ -548,17 +550,21 @@ public abstract class MulticlassOnlineClassifierUDTF extends LearnerBaseUDTF { covarOI); } } else { - LazySimpleSerDe serde = HiveUtils.getLineSerde(labelOI, featureOI, weightOI, - covarOI); + LazySimpleSerDe serde = + HiveUtils.getLineSerde(labelOI, featureOI, weightOI, covarOI); StructObjectInspector lineOI = (StructObjectInspector) serde.getObjectInspector(); StructField c1ref = lineOI.getStructFieldRef("c1"); StructField c2ref = lineOI.getStructFieldRef("c2"); StructField c3ref = lineOI.getStructFieldRef("c3"); StructField c4ref = lineOI.getStructFieldRef("c4"); - PrimitiveObjectInspector c1refOI = (PrimitiveObjectInspector) c1ref.getFieldObjectInspector(); - PrimitiveObjectInspector c2refOI = (PrimitiveObjectInspector) c2ref.getFieldObjectInspector(); - FloatObjectInspector c3refOI = (FloatObjectInspector) c3ref.getFieldObjectInspector(); - FloatObjectInspector c4refOI = (FloatObjectInspector) c4ref.getFieldObjectInspector(); + PrimitiveObjectInspector c1refOI = + (PrimitiveObjectInspector) c1ref.getFieldObjectInspector(); + PrimitiveObjectInspector c2refOI = + (PrimitiveObjectInspector) c2ref.getFieldObjectInspector(); + FloatObjectInspector c3refOI = + (FloatObjectInspector) c3ref.getFieldObjectInspector(); + FloatObjectInspector c4refOI = + (FloatObjectInspector) c4ref.getFieldObjectInspector(); BufferedReader reader = null; try { @@ -584,8 +590,8 @@ public abstract class MulticlassOnlineClassifierUDTF extends LearnerBaseUDTF { } Object k = c2refOI.getPrimitiveWritableObject(c2refOI.copyObject(f1)); float v = c3refOI.get(f2); - float cov = (f3 == null) ? WeightValueWithCovar.DEFAULT_COVAR - : c4refOI.get(f3); + float cov = + (f3 == null) ? WeightValueWithCovar.DEFAULT_COVAR : c4refOI.get(f3); model.set(k, new WeightValueWithCovar(v, cov, false)); } } finally { http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/classifier/multiclass/MulticlassPassiveAggressiveUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/classifier/multiclass/MulticlassPassiveAggressiveUDTF.java b/core/src/main/java/hivemall/classifier/multiclass/MulticlassPassiveAggressiveUDTF.java index d665c59..3a70f72 100644 --- a/core/src/main/java/hivemall/classifier/multiclass/MulticlassPassiveAggressiveUDTF.java +++ b/core/src/main/java/hivemall/classifier/multiclass/MulticlassPassiveAggressiveUDTF.java @@ -29,8 +29,7 @@ import org.apache.hadoop.hive.ql.exec.UDFArgumentException; import org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector; import org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector; -@Description( - name = "train_multiclass_pa", +@Description(name = "train_multiclass_pa", value = "_FUNC_(list<string|int|bigint> features, {int|string} label [, const string options])" + " - Returns a relation consists of <{int|string} label, {string|int|bigint} feature, float weight>", extended = "Build a prediction model by Passive-Aggressive (PA) multiclass classifier") @@ -71,8 +70,7 @@ public class MulticlassPassiveAggressiveUDTF extends MulticlassOnlineClassifierU return loss / (2.f * sqnorm); } - @Description( - name = "train_multiclass_pa1", + @Description(name = "train_multiclass_pa1", value = "_FUNC_(list<string|int|bigint> features, {int|string} label [, const string options])" + " - Returns a relation consists of <{int|string} label, {string|int|bigint} feature, float weight>", extended = "Build a prediction model by Passive-Aggressive 1 (PA-1) multiclass classifier") @@ -91,8 +89,8 @@ public class MulticlassPassiveAggressiveUDTF extends MulticlassOnlineClassifierU if (c_str != null) { c = Float.parseFloat(c_str); if (!(c > 0.f)) { - throw new UDFArgumentException("Aggressiveness parameter C must be C > 0: " - + c); + throw new UDFArgumentException( + "Aggressiveness parameter C must be C > 0: " + c); } } } @@ -109,8 +107,7 @@ public class MulticlassPassiveAggressiveUDTF extends MulticlassOnlineClassifierU } - @Description( - name = "train_multiclass_pa2", + @Description(name = "train_multiclass_pa2", value = "_FUNC_(list<string|int|bigint> features, {int|string} label [, const string options])" + " - Returns a relation consists of <{int|string} label, {string|int|bigint} feature, float weight>", extended = "Build a prediction model by Passive-Aggressive 2 (PA-2) multiclass classifier") http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/classifier/multiclass/MulticlassPerceptronUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/classifier/multiclass/MulticlassPerceptronUDTF.java b/core/src/main/java/hivemall/classifier/multiclass/MulticlassPerceptronUDTF.java index 2430389..539eb0f 100644 --- a/core/src/main/java/hivemall/classifier/multiclass/MulticlassPerceptronUDTF.java +++ b/core/src/main/java/hivemall/classifier/multiclass/MulticlassPerceptronUDTF.java @@ -28,8 +28,7 @@ import org.apache.hadoop.hive.ql.exec.UDFArgumentException; import org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector; import org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector; -@Description( - name = "train_multiclass_perceptron", +@Description(name = "train_multiclass_perceptron", value = "_FUNC_(list<string|int|bigint> features, {int|string} label [, const string options])" + " - Returns a relation consists of <{int|string} label, {string|int|bigint} feature, float weight>", extended = "Build a prediction model by Perceptron multiclass classifier") @@ -47,7 +46,8 @@ public final class MulticlassPerceptronUDTF extends MulticlassOnlineClassifierUD } @Override - protected void train(@Nonnull final FeatureValue[] features, @Nonnull final Object actual_label) { + protected void train(@Nonnull final FeatureValue[] features, + @Nonnull final Object actual_label) { PredictionResult predicted = classify(features); Object predicted_label = predicted.getLabel(); http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/classifier/multiclass/MulticlassSoftConfidenceWeightedUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/classifier/multiclass/MulticlassSoftConfidenceWeightedUDTF.java b/core/src/main/java/hivemall/classifier/multiclass/MulticlassSoftConfidenceWeightedUDTF.java index b2e7a45..f702342 100644 --- a/core/src/main/java/hivemall/classifier/multiclass/MulticlassSoftConfidenceWeightedUDTF.java +++ b/core/src/main/java/hivemall/classifier/multiclass/MulticlassSoftConfidenceWeightedUDTF.java @@ -103,7 +103,8 @@ public abstract class MulticlassSoftConfidenceWeightedUDTF extends MulticlassOnl if (c_str != null) { c = Float.parseFloat(c_str); if (!(c > 0.f)) { - throw new UDFArgumentException("Aggressiveness parameter C must be C > 0: " + c); + throw new UDFArgumentException( + "Aggressiveness parameter C must be C > 0: " + c); } } } @@ -146,8 +147,7 @@ public abstract class MulticlassSoftConfidenceWeightedUDTF extends MulticlassOnl protected abstract float getBeta(Margin margin, float alpha); - @Description( - name = "train_multiclass_scw", + @Description(name = "train_multiclass_scw", value = "_FUNC_(list<string|int|bigint> features, {int|string} label [, const string options])" + " - Returns a relation consists of <{int|string} label, {string|int|bigint} feature, float weight, float covar>", extended = "Build a prediction model by Soft Confidence-Weighted (SCW-1) multiclass classifier") @@ -173,10 +173,8 @@ public abstract class MulticlassSoftConfidenceWeightedUDTF extends MulticlassOnl float m = margin.get(); float var = margin.getVariance(); - float alpha_numer = -m - * psi - + (float) Math.sqrt((m * m * squared_phi * squared_phi / 4.f) - + (var * squared_phi * zeta)); + float alpha_numer = -m * psi + (float) Math.sqrt( + (m * m * squared_phi * squared_phi / 4.f) + (var * squared_phi * zeta)); float alpha_denom = var * zeta; if (alpha_denom == 0.f) { return 0.f; @@ -211,8 +209,7 @@ public abstract class MulticlassSoftConfidenceWeightedUDTF extends MulticlassOnl } - @Description( - name = "train_multiclass_scw2", + @Description(name = "train_multiclass_scw2", value = "_FUNC_(list<string|int|bigint> features, {int|string} label [, const string options])" + " - Returns a relation consists of <{int|string} label, {string|int|bigint} feature, float weight, float covar>", extended = "Build a prediction model by Soft Confidence-Weighted 2 (SCW-2) multiclass classifier") @@ -250,8 +247,8 @@ public abstract class MulticlassSoftConfidenceWeightedUDTF extends MulticlassOnl final Object missed_label, final float alpha, final float beta) { assert (actual_label != null); if (actual_label.equals(missed_label)) { - throw new IllegalArgumentException("Actual label equals to missed label: " - + actual_label); + throw new IllegalArgumentException( + "Actual label equals to missed label: " + actual_label); } PredictionModel model2add = label2model.get(actual_label); @@ -276,13 +273,14 @@ public abstract class MulticlassSoftConfidenceWeightedUDTF extends MulticlassOnl final float v = f.getValueAsFloat(); IWeightValue old_correctclass_w = model2add.get(k); - IWeightValue new_correctclass_w = getNewWeight(old_correctclass_w, v, alpha, beta, true); + IWeightValue new_correctclass_w = + getNewWeight(old_correctclass_w, v, alpha, beta, true); model2add.set(k, new_correctclass_w); if (model2sub != null) { IWeightValue old_wrongclass_w = model2sub.get(k); - IWeightValue new_wrongclass_w = getNewWeight(old_wrongclass_w, v, alpha, beta, - false); + IWeightValue new_wrongclass_w = + getNewWeight(old_wrongclass_w, v, alpha, beta, false); model2sub.set(k, new_wrongclass_w); } } http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/dataset/LogisticRegressionDataGeneratorUDTF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/dataset/LogisticRegressionDataGeneratorUDTF.java b/core/src/main/java/hivemall/dataset/LogisticRegressionDataGeneratorUDTF.java index d07089a..5b87183 100644 --- a/core/src/main/java/hivemall/dataset/LogisticRegressionDataGeneratorUDTF.java +++ b/core/src/main/java/hivemall/dataset/LogisticRegressionDataGeneratorUDTF.java @@ -40,8 +40,7 @@ import org.apache.hadoop.hive.serde2.objectinspector.ObjectInspectorFactory; import org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector; import org.apache.hadoop.hive.serde2.objectinspector.primitive.PrimitiveObjectInspectorFactory; -@Description( - name = "lr_datagen", +@Description(name = "lr_datagen", value = "_FUNC_(options string) - Generates a logistic regression dataset", extended = "WITH dual AS (SELECT 1) SELECT lr_datagen('-n_examples 1k -n_features 10') FROM dual;") public final class LogisticRegressionDataGeneratorUDTF extends UDTFWithOptions { @@ -79,9 +78,7 @@ public final class LogisticRegressionDataGeneratorUDTF extends UDTFWithOptions { opts.addOption("p1", "prob_one", true, " Probability in [0, 1.0) that a label is 1 [DEFAULT: 0.6]"); opts.addOption("seed", true, "The seed value for random number generator [DEFAULT: 43L]"); - opts.addOption( - "dense", - false, + opts.addOption("dense", false, "Make a dense dataset or not. If not specified, a sparse dataset is generated.\n" + "For sparse, n_dims should be much larger than n_features. When disabled, n_features must be equals to n_dims "); opts.addOption("sort", false, "Sort features if specified (used only for sparse dataset)"); @@ -114,9 +111,9 @@ public final class LogisticRegressionDataGeneratorUDTF extends UDTFWithOptions { } if (dense) { if (n_features != n_dimensions) { - throw new UDFArgumentException("n_features '" + n_features - + "' must be equals to n_dimensions '" + n_dimensions - + "' when making a dense dataset"); + throw new UDFArgumentException( + "n_features '" + n_features + "' must be equals to n_dimensions '" + + n_dimensions + "' when making a dense dataset"); } } @@ -135,9 +132,11 @@ public final class LogisticRegressionDataGeneratorUDTF extends UDTFWithOptions { fieldOIs.add(PrimitiveObjectInspectorFactory.javaFloatObjectInspector); fieldNames.add("features"); if (dense) { - fieldOIs.add(ObjectInspectorFactory.getStandardListObjectInspector(PrimitiveObjectInspectorFactory.javaFloatObjectInspector)); + fieldOIs.add(ObjectInspectorFactory.getStandardListObjectInspector( + PrimitiveObjectInspectorFactory.javaFloatObjectInspector)); } else { - fieldOIs.add(ObjectInspectorFactory.getStandardListObjectInspector(PrimitiveObjectInspectorFactory.javaStringObjectInspector)); + fieldOIs.add(ObjectInspectorFactory.getStandardListObjectInspector( + PrimitiveObjectInspectorFactory.javaStringObjectInspector)); } return ObjectInspectorFactory.getStandardStructObjectInspector(fieldNames, fieldOIs); } http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/ensemble/ArgminKLDistanceUDAF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/ensemble/ArgminKLDistanceUDAF.java b/core/src/main/java/hivemall/ensemble/ArgminKLDistanceUDAF.java index d6cd7d4..136ca0d 100644 --- a/core/src/main/java/hivemall/ensemble/ArgminKLDistanceUDAF.java +++ b/core/src/main/java/hivemall/ensemble/ArgminKLDistanceUDAF.java @@ -24,8 +24,7 @@ import org.apache.hadoop.hive.ql.exec.UDAFEvaluator; import org.apache.hadoop.io.FloatWritable; @SuppressWarnings("deprecation") -@Description( - name = "argmin_kld", +@Description(name = "argmin_kld", value = "_FUNC_(float mean, float covar) - Returns mean or covar that minimize a KL-distance among distributions", extended = "The returned value is (1.0 / (sum(1.0 / covar))) * (sum(mean / covar)") public final class ArgminKLDistanceUDAF extends UDAF { http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/ensemble/MaxRowUDAF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/ensemble/MaxRowUDAF.java b/core/src/main/java/hivemall/ensemble/MaxRowUDAF.java index 29d1563..c4ee658 100644 --- a/core/src/main/java/hivemall/ensemble/MaxRowUDAF.java +++ b/core/src/main/java/hivemall/ensemble/MaxRowUDAF.java @@ -38,14 +38,14 @@ import org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector; import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo; import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; -@Description( - name = "maxrow", +@Description(name = "maxrow", value = "_FUNC_(ANY compare, ...) - Returns a row that has maximum value in the 1st argument") public final class MaxRowUDAF extends AbstractGenericUDAFResolver { @Override public GenericUDAFEvaluator getEvaluator(TypeInfo[] parameters) throws SemanticException { - ObjectInspector oi = TypeInfoUtils.getStandardJavaObjectInspectorFromTypeInfo(parameters[0]); + ObjectInspector oi = + TypeInfoUtils.getStandardJavaObjectInspectorFromTypeInfo(parameters[0]); if (!ObjectInspectorUtils.compareSupported(oi)) { throw new UDFArgumentTypeException(0, "Cannot support comparison of map<> type or complex type containing map<>."); http://git-wip-us.apache.org/repos/asf/incubator-hivemall/blob/c4036695/core/src/main/java/hivemall/ensemble/bagging/WeightVotedAvgUDAF.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/hivemall/ensemble/bagging/WeightVotedAvgUDAF.java b/core/src/main/java/hivemall/ensemble/bagging/WeightVotedAvgUDAF.java index 4e7ea1b..2679116 100644 --- a/core/src/main/java/hivemall/ensemble/bagging/WeightVotedAvgUDAF.java +++ b/core/src/main/java/hivemall/ensemble/bagging/WeightVotedAvgUDAF.java @@ -26,8 +26,7 @@ import org.apache.hadoop.hive.ql.exec.UDAFEvaluator; import org.apache.hadoop.hive.serde2.io.DoubleWritable; @SuppressWarnings("deprecation") -@Description( - name = "weight_voted_avg", +@Description(name = "weight_voted_avg", value = "_FUNC_(expr) - Returns an averaged value by considering sum of positive/negative weights") public final class WeightVotedAvgUDAF extends UDAF {
