umappp 0.1.2
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/LICENSE.txt +25 -0
- data/README.md +110 -0
- data/ext/umappp/extconf.rb +25 -0
- data/ext/umappp/numo.hpp +867 -0
- data/ext/umappp/umappp.cpp +225 -0
- data/lib/umappp/version.rb +5 -0
- data/lib/umappp.rb +41 -0
- data/vendor/Eigen/Cholesky +45 -0
- data/vendor/Eigen/CholmodSupport +48 -0
- data/vendor/Eigen/Core +384 -0
- data/vendor/Eigen/Dense +7 -0
- data/vendor/Eigen/Eigen +2 -0
- data/vendor/Eigen/Eigenvalues +60 -0
- data/vendor/Eigen/Geometry +59 -0
- data/vendor/Eigen/Householder +29 -0
- data/vendor/Eigen/IterativeLinearSolvers +48 -0
- data/vendor/Eigen/Jacobi +32 -0
- data/vendor/Eigen/KLUSupport +41 -0
- data/vendor/Eigen/LU +47 -0
- data/vendor/Eigen/MetisSupport +35 -0
- data/vendor/Eigen/OrderingMethods +70 -0
- data/vendor/Eigen/PaStiXSupport +49 -0
- data/vendor/Eigen/PardisoSupport +35 -0
- data/vendor/Eigen/QR +50 -0
- data/vendor/Eigen/QtAlignedMalloc +39 -0
- data/vendor/Eigen/SPQRSupport +34 -0
- data/vendor/Eigen/SVD +50 -0
- data/vendor/Eigen/Sparse +34 -0
- data/vendor/Eigen/SparseCholesky +37 -0
- data/vendor/Eigen/SparseCore +69 -0
- data/vendor/Eigen/SparseLU +50 -0
- data/vendor/Eigen/SparseQR +36 -0
- data/vendor/Eigen/StdDeque +27 -0
- data/vendor/Eigen/StdList +26 -0
- data/vendor/Eigen/StdVector +27 -0
- data/vendor/Eigen/SuperLUSupport +64 -0
- data/vendor/Eigen/UmfPackSupport +40 -0
- data/vendor/Eigen/src/Cholesky/LDLT.h +688 -0
- data/vendor/Eigen/src/Cholesky/LLT.h +558 -0
- data/vendor/Eigen/src/Cholesky/LLT_LAPACKE.h +99 -0
- data/vendor/Eigen/src/CholmodSupport/CholmodSupport.h +682 -0
- data/vendor/Eigen/src/Core/ArithmeticSequence.h +413 -0
- data/vendor/Eigen/src/Core/Array.h +417 -0
- data/vendor/Eigen/src/Core/ArrayBase.h +226 -0
- data/vendor/Eigen/src/Core/ArrayWrapper.h +209 -0
- data/vendor/Eigen/src/Core/Assign.h +90 -0
- data/vendor/Eigen/src/Core/AssignEvaluator.h +1010 -0
- data/vendor/Eigen/src/Core/Assign_MKL.h +178 -0
- data/vendor/Eigen/src/Core/BandMatrix.h +353 -0
- data/vendor/Eigen/src/Core/Block.h +448 -0
- data/vendor/Eigen/src/Core/BooleanRedux.h +162 -0
- data/vendor/Eigen/src/Core/CommaInitializer.h +164 -0
- data/vendor/Eigen/src/Core/ConditionEstimator.h +175 -0
- data/vendor/Eigen/src/Core/CoreEvaluators.h +1741 -0
- data/vendor/Eigen/src/Core/CoreIterators.h +132 -0
- data/vendor/Eigen/src/Core/CwiseBinaryOp.h +183 -0
- data/vendor/Eigen/src/Core/CwiseNullaryOp.h +1001 -0
- data/vendor/Eigen/src/Core/CwiseTernaryOp.h +197 -0
- data/vendor/Eigen/src/Core/CwiseUnaryOp.h +103 -0
- data/vendor/Eigen/src/Core/CwiseUnaryView.h +132 -0
- data/vendor/Eigen/src/Core/DenseBase.h +701 -0
- data/vendor/Eigen/src/Core/DenseCoeffsBase.h +685 -0
- data/vendor/Eigen/src/Core/DenseStorage.h +652 -0
- data/vendor/Eigen/src/Core/Diagonal.h +258 -0
- data/vendor/Eigen/src/Core/DiagonalMatrix.h +391 -0
- data/vendor/Eigen/src/Core/DiagonalProduct.h +28 -0
- data/vendor/Eigen/src/Core/Dot.h +318 -0
- data/vendor/Eigen/src/Core/EigenBase.h +160 -0
- data/vendor/Eigen/src/Core/ForceAlignedAccess.h +150 -0
- data/vendor/Eigen/src/Core/Fuzzy.h +155 -0
- data/vendor/Eigen/src/Core/GeneralProduct.h +465 -0
- data/vendor/Eigen/src/Core/GenericPacketMath.h +1040 -0
- data/vendor/Eigen/src/Core/GlobalFunctions.h +194 -0
- data/vendor/Eigen/src/Core/IO.h +258 -0
- data/vendor/Eigen/src/Core/IndexedView.h +237 -0
- data/vendor/Eigen/src/Core/Inverse.h +117 -0
- data/vendor/Eigen/src/Core/Map.h +171 -0
- data/vendor/Eigen/src/Core/MapBase.h +310 -0
- data/vendor/Eigen/src/Core/MathFunctions.h +2057 -0
- data/vendor/Eigen/src/Core/MathFunctionsImpl.h +200 -0
- data/vendor/Eigen/src/Core/Matrix.h +565 -0
- data/vendor/Eigen/src/Core/MatrixBase.h +547 -0
- data/vendor/Eigen/src/Core/NestByValue.h +85 -0
- data/vendor/Eigen/src/Core/NoAlias.h +109 -0
- data/vendor/Eigen/src/Core/NumTraits.h +335 -0
- data/vendor/Eigen/src/Core/PartialReduxEvaluator.h +232 -0
- data/vendor/Eigen/src/Core/PermutationMatrix.h +605 -0
- data/vendor/Eigen/src/Core/PlainObjectBase.h +1128 -0
- data/vendor/Eigen/src/Core/Product.h +191 -0
- data/vendor/Eigen/src/Core/ProductEvaluators.h +1179 -0
- data/vendor/Eigen/src/Core/Random.h +218 -0
- data/vendor/Eigen/src/Core/Redux.h +515 -0
- data/vendor/Eigen/src/Core/Ref.h +381 -0
- data/vendor/Eigen/src/Core/Replicate.h +142 -0
- data/vendor/Eigen/src/Core/Reshaped.h +454 -0
- data/vendor/Eigen/src/Core/ReturnByValue.h +119 -0
- data/vendor/Eigen/src/Core/Reverse.h +217 -0
- data/vendor/Eigen/src/Core/Select.h +164 -0
- data/vendor/Eigen/src/Core/SelfAdjointView.h +365 -0
- data/vendor/Eigen/src/Core/SelfCwiseBinaryOp.h +47 -0
- data/vendor/Eigen/src/Core/Solve.h +188 -0
- data/vendor/Eigen/src/Core/SolveTriangular.h +235 -0
- data/vendor/Eigen/src/Core/SolverBase.h +168 -0
- data/vendor/Eigen/src/Core/StableNorm.h +251 -0
- data/vendor/Eigen/src/Core/StlIterators.h +463 -0
- data/vendor/Eigen/src/Core/Stride.h +116 -0
- data/vendor/Eigen/src/Core/Swap.h +68 -0
- data/vendor/Eigen/src/Core/Transpose.h +464 -0
- data/vendor/Eigen/src/Core/Transpositions.h +386 -0
- data/vendor/Eigen/src/Core/TriangularMatrix.h +1001 -0
- data/vendor/Eigen/src/Core/VectorBlock.h +96 -0
- data/vendor/Eigen/src/Core/VectorwiseOp.h +784 -0
- data/vendor/Eigen/src/Core/Visitor.h +381 -0
- data/vendor/Eigen/src/Core/arch/AVX/Complex.h +372 -0
- data/vendor/Eigen/src/Core/arch/AVX/MathFunctions.h +228 -0
- data/vendor/Eigen/src/Core/arch/AVX/PacketMath.h +1574 -0
- data/vendor/Eigen/src/Core/arch/AVX/TypeCasting.h +115 -0
- data/vendor/Eigen/src/Core/arch/AVX512/Complex.h +422 -0
- data/vendor/Eigen/src/Core/arch/AVX512/MathFunctions.h +362 -0
- data/vendor/Eigen/src/Core/arch/AVX512/PacketMath.h +2303 -0
- data/vendor/Eigen/src/Core/arch/AVX512/TypeCasting.h +89 -0
- data/vendor/Eigen/src/Core/arch/AltiVec/Complex.h +417 -0
- data/vendor/Eigen/src/Core/arch/AltiVec/MathFunctions.h +90 -0
- data/vendor/Eigen/src/Core/arch/AltiVec/MatrixProduct.h +2937 -0
- data/vendor/Eigen/src/Core/arch/AltiVec/MatrixProductCommon.h +221 -0
- data/vendor/Eigen/src/Core/arch/AltiVec/MatrixProductMMA.h +629 -0
- data/vendor/Eigen/src/Core/arch/AltiVec/PacketMath.h +2711 -0
- data/vendor/Eigen/src/Core/arch/CUDA/Complex.h +258 -0
- data/vendor/Eigen/src/Core/arch/Default/BFloat16.h +700 -0
- data/vendor/Eigen/src/Core/arch/Default/ConjHelper.h +117 -0
- data/vendor/Eigen/src/Core/arch/Default/GenericPacketMathFunctions.h +1649 -0
- data/vendor/Eigen/src/Core/arch/Default/GenericPacketMathFunctionsFwd.h +110 -0
- data/vendor/Eigen/src/Core/arch/Default/Half.h +942 -0
- data/vendor/Eigen/src/Core/arch/Default/Settings.h +49 -0
- data/vendor/Eigen/src/Core/arch/Default/TypeCasting.h +120 -0
- data/vendor/Eigen/src/Core/arch/GPU/MathFunctions.h +103 -0
- data/vendor/Eigen/src/Core/arch/GPU/PacketMath.h +1685 -0
- data/vendor/Eigen/src/Core/arch/GPU/TypeCasting.h +80 -0
- data/vendor/Eigen/src/Core/arch/HIP/hcc/math_constants.h +23 -0
- data/vendor/Eigen/src/Core/arch/MSA/Complex.h +648 -0
- data/vendor/Eigen/src/Core/arch/MSA/MathFunctions.h +387 -0
- data/vendor/Eigen/src/Core/arch/MSA/PacketMath.h +1233 -0
- data/vendor/Eigen/src/Core/arch/NEON/Complex.h +584 -0
- data/vendor/Eigen/src/Core/arch/NEON/GeneralBlockPanelKernel.h +183 -0
- data/vendor/Eigen/src/Core/arch/NEON/MathFunctions.h +75 -0
- data/vendor/Eigen/src/Core/arch/NEON/PacketMath.h +4587 -0
- data/vendor/Eigen/src/Core/arch/NEON/TypeCasting.h +1419 -0
- data/vendor/Eigen/src/Core/arch/SSE/Complex.h +351 -0
- data/vendor/Eigen/src/Core/arch/SSE/MathFunctions.h +199 -0
- data/vendor/Eigen/src/Core/arch/SSE/PacketMath.h +1505 -0
- data/vendor/Eigen/src/Core/arch/SSE/TypeCasting.h +142 -0
- data/vendor/Eigen/src/Core/arch/SVE/MathFunctions.h +44 -0
- data/vendor/Eigen/src/Core/arch/SVE/PacketMath.h +752 -0
- data/vendor/Eigen/src/Core/arch/SVE/TypeCasting.h +49 -0
- data/vendor/Eigen/src/Core/arch/SYCL/InteropHeaders.h +232 -0
- data/vendor/Eigen/src/Core/arch/SYCL/MathFunctions.h +301 -0
- data/vendor/Eigen/src/Core/arch/SYCL/PacketMath.h +670 -0
- data/vendor/Eigen/src/Core/arch/SYCL/SyclMemoryModel.h +694 -0
- data/vendor/Eigen/src/Core/arch/SYCL/TypeCasting.h +85 -0
- data/vendor/Eigen/src/Core/arch/ZVector/Complex.h +426 -0
- data/vendor/Eigen/src/Core/arch/ZVector/MathFunctions.h +233 -0
- data/vendor/Eigen/src/Core/arch/ZVector/PacketMath.h +1060 -0
- data/vendor/Eigen/src/Core/functors/AssignmentFunctors.h +177 -0
- data/vendor/Eigen/src/Core/functors/BinaryFunctors.h +541 -0
- data/vendor/Eigen/src/Core/functors/NullaryFunctors.h +189 -0
- data/vendor/Eigen/src/Core/functors/StlFunctors.h +166 -0
- data/vendor/Eigen/src/Core/functors/TernaryFunctors.h +25 -0
- data/vendor/Eigen/src/Core/functors/UnaryFunctors.h +1131 -0
- data/vendor/Eigen/src/Core/products/GeneralBlockPanelKernel.h +2645 -0
- data/vendor/Eigen/src/Core/products/GeneralMatrixMatrix.h +517 -0
- data/vendor/Eigen/src/Core/products/GeneralMatrixMatrixTriangular.h +317 -0
- data/vendor/Eigen/src/Core/products/GeneralMatrixMatrixTriangular_BLAS.h +145 -0
- data/vendor/Eigen/src/Core/products/GeneralMatrixMatrix_BLAS.h +124 -0
- data/vendor/Eigen/src/Core/products/GeneralMatrixVector.h +518 -0
- data/vendor/Eigen/src/Core/products/GeneralMatrixVector_BLAS.h +136 -0
- data/vendor/Eigen/src/Core/products/Parallelizer.h +180 -0
- data/vendor/Eigen/src/Core/products/SelfadjointMatrixMatrix.h +544 -0
- data/vendor/Eigen/src/Core/products/SelfadjointMatrixMatrix_BLAS.h +295 -0
- data/vendor/Eigen/src/Core/products/SelfadjointMatrixVector.h +262 -0
- data/vendor/Eigen/src/Core/products/SelfadjointMatrixVector_BLAS.h +118 -0
- data/vendor/Eigen/src/Core/products/SelfadjointProduct.h +133 -0
- data/vendor/Eigen/src/Core/products/SelfadjointRank2Update.h +94 -0
- data/vendor/Eigen/src/Core/products/TriangularMatrixMatrix.h +472 -0
- data/vendor/Eigen/src/Core/products/TriangularMatrixMatrix_BLAS.h +317 -0
- data/vendor/Eigen/src/Core/products/TriangularMatrixVector.h +350 -0
- data/vendor/Eigen/src/Core/products/TriangularMatrixVector_BLAS.h +255 -0
- data/vendor/Eigen/src/Core/products/TriangularSolverMatrix.h +337 -0
- data/vendor/Eigen/src/Core/products/TriangularSolverMatrix_BLAS.h +167 -0
- data/vendor/Eigen/src/Core/products/TriangularSolverVector.h +148 -0
- data/vendor/Eigen/src/Core/util/BlasUtil.h +583 -0
- data/vendor/Eigen/src/Core/util/ConfigureVectorization.h +512 -0
- data/vendor/Eigen/src/Core/util/Constants.h +563 -0
- data/vendor/Eigen/src/Core/util/DisableStupidWarnings.h +106 -0
- data/vendor/Eigen/src/Core/util/ForwardDeclarations.h +322 -0
- data/vendor/Eigen/src/Core/util/IndexedViewHelper.h +186 -0
- data/vendor/Eigen/src/Core/util/IntegralConstant.h +272 -0
- data/vendor/Eigen/src/Core/util/MKL_support.h +137 -0
- data/vendor/Eigen/src/Core/util/Macros.h +1464 -0
- data/vendor/Eigen/src/Core/util/Memory.h +1163 -0
- data/vendor/Eigen/src/Core/util/Meta.h +812 -0
- data/vendor/Eigen/src/Core/util/NonMPL2.h +3 -0
- data/vendor/Eigen/src/Core/util/ReenableStupidWarnings.h +31 -0
- data/vendor/Eigen/src/Core/util/ReshapedHelper.h +51 -0
- data/vendor/Eigen/src/Core/util/StaticAssert.h +221 -0
- data/vendor/Eigen/src/Core/util/SymbolicIndex.h +293 -0
- data/vendor/Eigen/src/Core/util/XprHelper.h +856 -0
- data/vendor/Eigen/src/Eigenvalues/ComplexEigenSolver.h +346 -0
- data/vendor/Eigen/src/Eigenvalues/ComplexSchur.h +462 -0
- data/vendor/Eigen/src/Eigenvalues/ComplexSchur_LAPACKE.h +91 -0
- data/vendor/Eigen/src/Eigenvalues/EigenSolver.h +622 -0
- data/vendor/Eigen/src/Eigenvalues/GeneralizedEigenSolver.h +418 -0
- data/vendor/Eigen/src/Eigenvalues/GeneralizedSelfAdjointEigenSolver.h +226 -0
- data/vendor/Eigen/src/Eigenvalues/HessenbergDecomposition.h +374 -0
- data/vendor/Eigen/src/Eigenvalues/MatrixBaseEigenvalues.h +158 -0
- data/vendor/Eigen/src/Eigenvalues/RealQZ.h +657 -0
- data/vendor/Eigen/src/Eigenvalues/RealSchur.h +558 -0
- data/vendor/Eigen/src/Eigenvalues/RealSchur_LAPACKE.h +77 -0
- data/vendor/Eigen/src/Eigenvalues/SelfAdjointEigenSolver.h +904 -0
- data/vendor/Eigen/src/Eigenvalues/SelfAdjointEigenSolver_LAPACKE.h +87 -0
- data/vendor/Eigen/src/Eigenvalues/Tridiagonalization.h +561 -0
- data/vendor/Eigen/src/Geometry/AlignedBox.h +486 -0
- data/vendor/Eigen/src/Geometry/AngleAxis.h +247 -0
- data/vendor/Eigen/src/Geometry/EulerAngles.h +114 -0
- data/vendor/Eigen/src/Geometry/Homogeneous.h +501 -0
- data/vendor/Eigen/src/Geometry/Hyperplane.h +282 -0
- data/vendor/Eigen/src/Geometry/OrthoMethods.h +235 -0
- data/vendor/Eigen/src/Geometry/ParametrizedLine.h +232 -0
- data/vendor/Eigen/src/Geometry/Quaternion.h +870 -0
- data/vendor/Eigen/src/Geometry/Rotation2D.h +199 -0
- data/vendor/Eigen/src/Geometry/RotationBase.h +206 -0
- data/vendor/Eigen/src/Geometry/Scaling.h +188 -0
- data/vendor/Eigen/src/Geometry/Transform.h +1563 -0
- data/vendor/Eigen/src/Geometry/Translation.h +202 -0
- data/vendor/Eigen/src/Geometry/Umeyama.h +166 -0
- data/vendor/Eigen/src/Geometry/arch/Geometry_SIMD.h +168 -0
- data/vendor/Eigen/src/Householder/BlockHouseholder.h +110 -0
- data/vendor/Eigen/src/Householder/Householder.h +176 -0
- data/vendor/Eigen/src/Householder/HouseholderSequence.h +545 -0
- data/vendor/Eigen/src/IterativeLinearSolvers/BasicPreconditioners.h +226 -0
- data/vendor/Eigen/src/IterativeLinearSolvers/BiCGSTAB.h +212 -0
- data/vendor/Eigen/src/IterativeLinearSolvers/ConjugateGradient.h +229 -0
- data/vendor/Eigen/src/IterativeLinearSolvers/IncompleteCholesky.h +394 -0
- data/vendor/Eigen/src/IterativeLinearSolvers/IncompleteLUT.h +453 -0
- data/vendor/Eigen/src/IterativeLinearSolvers/IterativeSolverBase.h +444 -0
- data/vendor/Eigen/src/IterativeLinearSolvers/LeastSquareConjugateGradient.h +198 -0
- data/vendor/Eigen/src/IterativeLinearSolvers/SolveWithGuess.h +117 -0
- data/vendor/Eigen/src/Jacobi/Jacobi.h +483 -0
- data/vendor/Eigen/src/KLUSupport/KLUSupport.h +358 -0
- data/vendor/Eigen/src/LU/Determinant.h +117 -0
- data/vendor/Eigen/src/LU/FullPivLU.h +877 -0
- data/vendor/Eigen/src/LU/InverseImpl.h +432 -0
- data/vendor/Eigen/src/LU/PartialPivLU.h +624 -0
- data/vendor/Eigen/src/LU/PartialPivLU_LAPACKE.h +83 -0
- data/vendor/Eigen/src/LU/arch/InverseSize4.h +351 -0
- data/vendor/Eigen/src/MetisSupport/MetisSupport.h +137 -0
- data/vendor/Eigen/src/OrderingMethods/Amd.h +435 -0
- data/vendor/Eigen/src/OrderingMethods/Eigen_Colamd.h +1863 -0
- data/vendor/Eigen/src/OrderingMethods/Ordering.h +153 -0
- data/vendor/Eigen/src/PaStiXSupport/PaStiXSupport.h +678 -0
- data/vendor/Eigen/src/PardisoSupport/PardisoSupport.h +545 -0
- data/vendor/Eigen/src/QR/ColPivHouseholderQR.h +674 -0
- data/vendor/Eigen/src/QR/ColPivHouseholderQR_LAPACKE.h +97 -0
- data/vendor/Eigen/src/QR/CompleteOrthogonalDecomposition.h +635 -0
- data/vendor/Eigen/src/QR/FullPivHouseholderQR.h +713 -0
- data/vendor/Eigen/src/QR/HouseholderQR.h +434 -0
- data/vendor/Eigen/src/QR/HouseholderQR_LAPACKE.h +68 -0
- data/vendor/Eigen/src/SPQRSupport/SuiteSparseQRSupport.h +335 -0
- data/vendor/Eigen/src/SVD/BDCSVD.h +1366 -0
- data/vendor/Eigen/src/SVD/JacobiSVD.h +812 -0
- data/vendor/Eigen/src/SVD/JacobiSVD_LAPACKE.h +91 -0
- data/vendor/Eigen/src/SVD/SVDBase.h +376 -0
- data/vendor/Eigen/src/SVD/UpperBidiagonalization.h +414 -0
- data/vendor/Eigen/src/SparseCholesky/SimplicialCholesky.h +697 -0
- data/vendor/Eigen/src/SparseCholesky/SimplicialCholesky_impl.h +174 -0
- data/vendor/Eigen/src/SparseCore/AmbiVector.h +378 -0
- data/vendor/Eigen/src/SparseCore/CompressedStorage.h +274 -0
- data/vendor/Eigen/src/SparseCore/ConservativeSparseSparseProduct.h +352 -0
- data/vendor/Eigen/src/SparseCore/MappedSparseMatrix.h +67 -0
- data/vendor/Eigen/src/SparseCore/SparseAssign.h +270 -0
- data/vendor/Eigen/src/SparseCore/SparseBlock.h +571 -0
- data/vendor/Eigen/src/SparseCore/SparseColEtree.h +206 -0
- data/vendor/Eigen/src/SparseCore/SparseCompressedBase.h +370 -0
- data/vendor/Eigen/src/SparseCore/SparseCwiseBinaryOp.h +722 -0
- data/vendor/Eigen/src/SparseCore/SparseCwiseUnaryOp.h +150 -0
- data/vendor/Eigen/src/SparseCore/SparseDenseProduct.h +342 -0
- data/vendor/Eigen/src/SparseCore/SparseDiagonalProduct.h +138 -0
- data/vendor/Eigen/src/SparseCore/SparseDot.h +98 -0
- data/vendor/Eigen/src/SparseCore/SparseFuzzy.h +29 -0
- data/vendor/Eigen/src/SparseCore/SparseMap.h +305 -0
- data/vendor/Eigen/src/SparseCore/SparseMatrix.h +1518 -0
- data/vendor/Eigen/src/SparseCore/SparseMatrixBase.h +398 -0
- data/vendor/Eigen/src/SparseCore/SparsePermutation.h +178 -0
- data/vendor/Eigen/src/SparseCore/SparseProduct.h +181 -0
- data/vendor/Eigen/src/SparseCore/SparseRedux.h +49 -0
- data/vendor/Eigen/src/SparseCore/SparseRef.h +397 -0
- data/vendor/Eigen/src/SparseCore/SparseSelfAdjointView.h +659 -0
- data/vendor/Eigen/src/SparseCore/SparseSolverBase.h +124 -0
- data/vendor/Eigen/src/SparseCore/SparseSparseProductWithPruning.h +198 -0
- data/vendor/Eigen/src/SparseCore/SparseTranspose.h +92 -0
- data/vendor/Eigen/src/SparseCore/SparseTriangularView.h +189 -0
- data/vendor/Eigen/src/SparseCore/SparseUtil.h +186 -0
- data/vendor/Eigen/src/SparseCore/SparseVector.h +478 -0
- data/vendor/Eigen/src/SparseCore/SparseView.h +254 -0
- data/vendor/Eigen/src/SparseCore/TriangularSolver.h +315 -0
- data/vendor/Eigen/src/SparseLU/SparseLU.h +923 -0
- data/vendor/Eigen/src/SparseLU/SparseLUImpl.h +66 -0
- data/vendor/Eigen/src/SparseLU/SparseLU_Memory.h +226 -0
- data/vendor/Eigen/src/SparseLU/SparseLU_Structs.h +110 -0
- data/vendor/Eigen/src/SparseLU/SparseLU_SupernodalMatrix.h +375 -0
- data/vendor/Eigen/src/SparseLU/SparseLU_Utils.h +80 -0
- data/vendor/Eigen/src/SparseLU/SparseLU_column_bmod.h +181 -0
- data/vendor/Eigen/src/SparseLU/SparseLU_column_dfs.h +179 -0
- data/vendor/Eigen/src/SparseLU/SparseLU_copy_to_ucol.h +107 -0
- data/vendor/Eigen/src/SparseLU/SparseLU_gemm_kernel.h +280 -0
- data/vendor/Eigen/src/SparseLU/SparseLU_heap_relax_snode.h +126 -0
- data/vendor/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +130 -0
- data/vendor/Eigen/src/SparseLU/SparseLU_panel_bmod.h +223 -0
- data/vendor/Eigen/src/SparseLU/SparseLU_panel_dfs.h +258 -0
- data/vendor/Eigen/src/SparseLU/SparseLU_pivotL.h +137 -0
- data/vendor/Eigen/src/SparseLU/SparseLU_pruneL.h +136 -0
- data/vendor/Eigen/src/SparseLU/SparseLU_relax_snode.h +83 -0
- data/vendor/Eigen/src/SparseQR/SparseQR.h +758 -0
- data/vendor/Eigen/src/StlSupport/StdDeque.h +116 -0
- data/vendor/Eigen/src/StlSupport/StdList.h +106 -0
- data/vendor/Eigen/src/StlSupport/StdVector.h +131 -0
- data/vendor/Eigen/src/StlSupport/details.h +84 -0
- data/vendor/Eigen/src/SuperLUSupport/SuperLUSupport.h +1025 -0
- data/vendor/Eigen/src/UmfPackSupport/UmfPackSupport.h +642 -0
- data/vendor/Eigen/src/misc/Image.h +82 -0
- data/vendor/Eigen/src/misc/Kernel.h +79 -0
- data/vendor/Eigen/src/misc/RealSvd2x2.h +55 -0
- data/vendor/Eigen/src/misc/blas.h +440 -0
- data/vendor/Eigen/src/misc/lapack.h +152 -0
- data/vendor/Eigen/src/misc/lapacke.h +16292 -0
- data/vendor/Eigen/src/misc/lapacke_mangling.h +17 -0
- data/vendor/Eigen/src/plugins/ArrayCwiseBinaryOps.h +358 -0
- data/vendor/Eigen/src/plugins/ArrayCwiseUnaryOps.h +696 -0
- data/vendor/Eigen/src/plugins/BlockMethods.h +1442 -0
- data/vendor/Eigen/src/plugins/CommonCwiseBinaryOps.h +115 -0
- data/vendor/Eigen/src/plugins/CommonCwiseUnaryOps.h +177 -0
- data/vendor/Eigen/src/plugins/IndexedViewMethods.h +262 -0
- data/vendor/Eigen/src/plugins/MatrixCwiseBinaryOps.h +152 -0
- data/vendor/Eigen/src/plugins/MatrixCwiseUnaryOps.h +95 -0
- data/vendor/Eigen/src/plugins/ReshapedMethods.h +149 -0
- data/vendor/aarand/aarand.hpp +114 -0
- data/vendor/annoy/annoylib.h +1495 -0
- data/vendor/annoy/kissrandom.h +120 -0
- data/vendor/annoy/mman.h +242 -0
- data/vendor/hnswlib/bruteforce.h +152 -0
- data/vendor/hnswlib/hnswalg.h +1192 -0
- data/vendor/hnswlib/hnswlib.h +108 -0
- data/vendor/hnswlib/space_ip.h +282 -0
- data/vendor/hnswlib/space_l2.h +281 -0
- data/vendor/hnswlib/visited_list_pool.h +79 -0
- data/vendor/irlba/irlba.hpp +575 -0
- data/vendor/irlba/lanczos.hpp +212 -0
- data/vendor/irlba/parallel.hpp +474 -0
- data/vendor/irlba/utils.hpp +224 -0
- data/vendor/irlba/wrappers.hpp +228 -0
- data/vendor/kmeans/Base.hpp +75 -0
- data/vendor/kmeans/Details.hpp +79 -0
- data/vendor/kmeans/HartiganWong.hpp +492 -0
- data/vendor/kmeans/InitializeKmeansPP.hpp +144 -0
- data/vendor/kmeans/InitializeNone.hpp +44 -0
- data/vendor/kmeans/InitializePCAPartition.hpp +309 -0
- data/vendor/kmeans/InitializeRandom.hpp +91 -0
- data/vendor/kmeans/Kmeans.hpp +161 -0
- data/vendor/kmeans/Lloyd.hpp +134 -0
- data/vendor/kmeans/MiniBatch.hpp +269 -0
- data/vendor/kmeans/QuickSearch.hpp +179 -0
- data/vendor/kmeans/compute_centroids.hpp +32 -0
- data/vendor/kmeans/compute_wcss.hpp +27 -0
- data/vendor/kmeans/is_edge_case.hpp +42 -0
- data/vendor/kmeans/random.hpp +55 -0
- data/vendor/knncolle/Annoy/Annoy.hpp +193 -0
- data/vendor/knncolle/BruteForce/BruteForce.hpp +120 -0
- data/vendor/knncolle/Hnsw/Hnsw.hpp +225 -0
- data/vendor/knncolle/Kmknn/Kmknn.hpp +286 -0
- data/vendor/knncolle/VpTree/VpTree.hpp +256 -0
- data/vendor/knncolle/knncolle.hpp +34 -0
- data/vendor/knncolle/utils/Base.hpp +100 -0
- data/vendor/knncolle/utils/NeighborQueue.hpp +94 -0
- data/vendor/knncolle/utils/distances.hpp +98 -0
- data/vendor/knncolle/utils/find_nearest_neighbors.hpp +112 -0
- data/vendor/powerit/PowerIterations.hpp +157 -0
- data/vendor/umappp/NeighborList.hpp +37 -0
- data/vendor/umappp/Umap.hpp +662 -0
- data/vendor/umappp/combine_neighbor_sets.hpp +95 -0
- data/vendor/umappp/find_ab.hpp +157 -0
- data/vendor/umappp/neighbor_similarities.hpp +136 -0
- data/vendor/umappp/optimize_layout.hpp +285 -0
- data/vendor/umappp/spectral_init.hpp +181 -0
- data/vendor/umappp/umappp.hpp +13 -0
- metadata +465 -0
@@ -0,0 +1,662 @@
|
|
1
|
+
#ifndef UMAPPP_UMAP_HPP
|
2
|
+
#define UMAPPP_UMAP_HPP
|
3
|
+
|
4
|
+
#include "NeighborList.hpp"
|
5
|
+
#include "combine_neighbor_sets.hpp"
|
6
|
+
#include "find_ab.hpp"
|
7
|
+
#include "neighbor_similarities.hpp"
|
8
|
+
#include "optimize_layout.hpp"
|
9
|
+
#include "spectral_init.hpp"
|
10
|
+
|
11
|
+
#ifndef UMAPPP_CUSTOM_NEIGHBORS
|
12
|
+
#include "knncolle/knncolle.hpp"
|
13
|
+
#endif
|
14
|
+
|
15
|
+
#include <random>
|
16
|
+
#include <cstdint>
|
17
|
+
|
18
|
+
/**
|
19
|
+
* @file Umap.hpp
|
20
|
+
*
|
21
|
+
* @brief Run the UMAP algorithm.
|
22
|
+
*/
|
23
|
+
|
24
|
+
namespace umappp {
|
25
|
+
|
26
|
+
/**
|
27
|
+
* How should the initial coordinates of the embedding be obtained?
|
28
|
+
*
|
29
|
+
* - `SPECTRAL`: attempts initialization based on spectral decomposition of the graph Laplacian.
|
30
|
+
* If that fails, we fall back to random draws from a normal distribution.
|
31
|
+
* - `SPECTRAL_ONLY`: attempts spectral initialization as before,
|
32
|
+
* but if that fails, we use the existing values in the supplied embedding array.
|
33
|
+
* - `RANDOM`: fills the embedding with random draws from a normal distribution.
|
34
|
+
* - `NONE`: uses the existing values in the supplied embedding array.
|
35
|
+
*/
|
36
|
+
enum InitMethod { SPECTRAL, SPECTRAL_ONLY, RANDOM, NONE };
|
37
|
+
|
38
|
+
/**
|
39
|
+
* @cond
|
40
|
+
*/
|
41
|
+
int choose_num_epochs(int num_epochs, size_t size) {
|
42
|
+
if (num_epochs < 0) {
|
43
|
+
// Choosing the number of epochs. We use a simple formula to decrease
|
44
|
+
// the number of epochs with increasing size, with the aim being that
|
45
|
+
// the 'extra work' beyond the minimal 200 epochs should be the same
|
46
|
+
// regardless of the numbe of observations. Given one calculation per
|
47
|
+
// observation per epoch, this amounts to 300 * 10000 calculations at
|
48
|
+
// the lower bound, so we simply choose a number of epochs that
|
49
|
+
// equalizes the number of calculations for any number of observations.
|
50
|
+
if (num_epochs < 0) {
|
51
|
+
constexpr int limit = 10000, minimal = 200, maximal = 300;
|
52
|
+
if (size <= limit) {
|
53
|
+
num_epochs = minimal + maximal;
|
54
|
+
} else {
|
55
|
+
num_epochs = minimal + static_cast<int>(std::ceil(maximal * limit / static_cast<double>(size)));
|
56
|
+
}
|
57
|
+
}
|
58
|
+
}
|
59
|
+
return num_epochs;
|
60
|
+
}
|
61
|
+
/**
|
62
|
+
* @endcond
|
63
|
+
*/
|
64
|
+
|
65
|
+
/**
|
66
|
+
* @brief Wrapper class to run UMAP.
|
67
|
+
*
|
68
|
+
* The Uniform Manifold Approximation and Projection (UMAP) algorithm is an efficient dimensionality reduction method based on nearest neighbors.
|
69
|
+
* The general idea is to find a low-dimensional embedding that preserves the neighborhood of each observation from the original space;
|
70
|
+
* this is achieved by applying attractive forces between each observation and its neighbors while repelling to all other cells.
|
71
|
+
* Further theoretical details can be found in the [original UMAP documentation](https://umap-learn.readthedocs.io/en/latest/how_umap_works.html),
|
72
|
+
* though this particular implementation is derived from the C++ code in the [**uwot** R package](https://github.com/jlmelville/uwot).
|
73
|
+
*
|
74
|
+
* @tparam Float Floating-point type.
|
75
|
+
* Defaults to `double` to be conservative, but most applications can make do with `float` for some extra speed.
|
76
|
+
*
|
77
|
+
* @see
|
78
|
+
* McInnes L, Healy J and Melville J (2020).
|
79
|
+
* UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction.
|
80
|
+
* _arXiv_, https://arxiv.org/abs/1802.03426
|
81
|
+
*/
|
82
|
+
template<typename Float = double>
|
83
|
+
class Umap {
|
84
|
+
public:
|
85
|
+
/**
|
86
|
+
* @brief Default values for all UMAP parameters.
|
87
|
+
*/
|
88
|
+
struct Defaults {
|
89
|
+
/**
|
90
|
+
* See `set_local_connectivity()`.
|
91
|
+
*/
|
92
|
+
static constexpr Float local_connectivity = 1.0;
|
93
|
+
|
94
|
+
/**
|
95
|
+
* See `set_bandwidth()`.
|
96
|
+
*/
|
97
|
+
static constexpr Float bandwidth = 1;
|
98
|
+
|
99
|
+
/**
|
100
|
+
* See `set_mix_ratio()`.
|
101
|
+
*/
|
102
|
+
static constexpr Float mix_ratio = 1;
|
103
|
+
|
104
|
+
/**
|
105
|
+
* See `set_spread()`.
|
106
|
+
*/
|
107
|
+
static constexpr Float spread = 1;
|
108
|
+
|
109
|
+
/**
|
110
|
+
* See `set_min_dist()`.
|
111
|
+
*/
|
112
|
+
static constexpr Float min_dist = 0.01;
|
113
|
+
|
114
|
+
/**
|
115
|
+
* See `set_a()`.
|
116
|
+
*/
|
117
|
+
static constexpr Float a = 0;
|
118
|
+
|
119
|
+
/**
|
120
|
+
* See `set_b()`.
|
121
|
+
*/
|
122
|
+
static constexpr Float b = 0;
|
123
|
+
|
124
|
+
/**
|
125
|
+
* See `set_repulsion_strength()`.
|
126
|
+
*/
|
127
|
+
static constexpr Float repulsion_strength = 1;
|
128
|
+
|
129
|
+
/**
|
130
|
+
* See `set_initialize()`.
|
131
|
+
*/
|
132
|
+
static constexpr InitMethod initialize = SPECTRAL;
|
133
|
+
|
134
|
+
/**
|
135
|
+
* See `set_num_epochs()`.
|
136
|
+
*/
|
137
|
+
static constexpr int num_epochs = -1;
|
138
|
+
|
139
|
+
/**
|
140
|
+
* See `set_learning_rate()`.
|
141
|
+
*/
|
142
|
+
static constexpr Float learning_rate = 1;
|
143
|
+
|
144
|
+
/**
|
145
|
+
* See `set_negative_sample_rate()`.
|
146
|
+
*/
|
147
|
+
static constexpr Float negative_sample_rate = 5;
|
148
|
+
|
149
|
+
/**
|
150
|
+
* See `set_num_neighbors()`.
|
151
|
+
*/
|
152
|
+
static constexpr int num_neighbors = 15;
|
153
|
+
|
154
|
+
/**
|
155
|
+
* See `set_seed()`.
|
156
|
+
*/
|
157
|
+
static constexpr uint64_t seed = 1234567890;
|
158
|
+
|
159
|
+
/**
|
160
|
+
* See `set_batch()`.
|
161
|
+
*/
|
162
|
+
static constexpr bool batch = false;
|
163
|
+
|
164
|
+
/**
|
165
|
+
* See `set_num_threads()`.
|
166
|
+
*/
|
167
|
+
static constexpr int num_threads = 1;
|
168
|
+
};
|
169
|
+
|
170
|
+
private:
|
171
|
+
InitMethod init = Defaults::initialize;
|
172
|
+
int num_neighbors = Defaults::num_neighbors;
|
173
|
+
Float local_connectivity = Defaults::local_connectivity;
|
174
|
+
Float bandwidth = Defaults::bandwidth;
|
175
|
+
Float mix_ratio = Defaults::mix_ratio;
|
176
|
+
Float spread = Defaults::spread;
|
177
|
+
Float min_dist = Defaults::min_dist;
|
178
|
+
int num_epochs = Defaults::num_epochs;
|
179
|
+
Float negative_sample_rate = Defaults::negative_sample_rate;
|
180
|
+
uint64_t seed = Defaults::seed;
|
181
|
+
|
182
|
+
struct RuntimeParameters {
|
183
|
+
Float a = Defaults::a;
|
184
|
+
Float b = Defaults::b;
|
185
|
+
Float repulsion_strength = Defaults::repulsion_strength;
|
186
|
+
Float learning_rate = Defaults::learning_rate;
|
187
|
+
bool batch = Defaults::batch;
|
188
|
+
int nthreads = Defaults::num_threads;
|
189
|
+
};
|
190
|
+
|
191
|
+
RuntimeParameters rparams;
|
192
|
+
|
193
|
+
public:
|
194
|
+
/**
|
195
|
+
* @param i How to initialize the embedding, see `InitMethod` for more details.
|
196
|
+
* Some choices may use the existing coordinates provided to `run()` via the `embedding` argument.
|
197
|
+
*
|
198
|
+
* @return A reference to this `Umap` object.
|
199
|
+
*/
|
200
|
+
Umap& set_initialize(InitMethod i = Defaults::initialize) {
|
201
|
+
init = i;
|
202
|
+
return *this;
|
203
|
+
}
|
204
|
+
|
205
|
+
/**
|
206
|
+
* @param n Number of neighbors to use to define the fuzzy sets.
|
207
|
+
* Larger values improve connectivity and favor preservation of global structure, at the cost of increased computational work.
|
208
|
+
* This argument is only used in certain `run()` and `initialize()` methods that perform identification of the nearest neighbors.
|
209
|
+
*
|
210
|
+
* @return A reference to this `Umap` object.
|
211
|
+
*/
|
212
|
+
Umap& set_num_neighbors(Float n = Defaults::num_neighbors) {
|
213
|
+
num_neighbors = n;
|
214
|
+
return *this;
|
215
|
+
}
|
216
|
+
|
217
|
+
/**
|
218
|
+
* @param l The number of nearest neighbors that are assumed to be always connected, with maximum membership confidence.
|
219
|
+
* Larger values increase the connectivity of the embedding and reduce the focus on local structure.
|
220
|
+
*
|
221
|
+
* @return A reference to this `Umap` object.
|
222
|
+
*/
|
223
|
+
Umap& set_local_connectivity(Float l = Defaults::local_connectivity) {
|
224
|
+
local_connectivity = l;
|
225
|
+
return *this;
|
226
|
+
}
|
227
|
+
|
228
|
+
/**
|
229
|
+
* @param b Effective bandwidth of the kernel when converting the distance to a neighbor into a fuzzy set membership confidence.
|
230
|
+
* Larger values reduce the decay in confidence with respect to distance, increasing connectivity and favoring global structure.
|
231
|
+
*
|
232
|
+
* @return A reference to this `Umap` object.
|
233
|
+
*/
|
234
|
+
Umap& set_bandwidth(Float b = Defaults::bandwidth) {
|
235
|
+
bandwidth = b;
|
236
|
+
return *this;
|
237
|
+
}
|
238
|
+
|
239
|
+
/**
|
240
|
+
* @param m Mixing ratio to use when combining fuzzy sets.
|
241
|
+
* This symmetrizes the sets by ensuring that the confidence of $A$ belonging to $B$'s set is the same as the confidence of $B$ belonging to $A$'s set.
|
242
|
+
* A mixing ratio of 1 will take the union of confidences, a ratio of 0 will take the intersection, and intermediate values will interpolate between them.
|
243
|
+
* Larger values (up to 1) favor connectivity and more global structure.
|
244
|
+
*
|
245
|
+
* @return A reference to this `Umap` object.
|
246
|
+
*/
|
247
|
+
Umap& set_mix_ratio(Float m = Defaults::mix_ratio) {
|
248
|
+
mix_ratio = m;
|
249
|
+
return *this;
|
250
|
+
}
|
251
|
+
|
252
|
+
/**
|
253
|
+
* @param s Scale of the coordinates of the final low-dimensional embedding.
|
254
|
+
*
|
255
|
+
* @return A reference to this `Umap` object.
|
256
|
+
*/
|
257
|
+
Umap& set_spread(Float s = Defaults::spread) {
|
258
|
+
spread = s;
|
259
|
+
return *this;
|
260
|
+
}
|
261
|
+
|
262
|
+
/**
|
263
|
+
* @param m Minimum distance between observations in the final low-dimensional embedding.
|
264
|
+
* Smaller values will increase local clustering while larger values favors a more even distribution.
|
265
|
+
* This is interpreted relative to the spread of points in `set_spread()`.
|
266
|
+
*
|
267
|
+
* @return A reference to this `Umap` object.
|
268
|
+
*/
|
269
|
+
Umap& set_min_dist(Float m = Defaults::min_dist) {
|
270
|
+
min_dist = m;
|
271
|
+
return *this;
|
272
|
+
}
|
273
|
+
|
274
|
+
/**
|
275
|
+
* @param n Number of epochs for the gradient descent, i.e., optimization iterations.
|
276
|
+
*
|
277
|
+
* Larger values improve accuracy at the cost of computational work.
|
278
|
+
* If the requested number of epochs is negative, a value is automatically chosen based on the size of the dataset:
|
279
|
+
*
|
280
|
+
* - For datasets with no more than 10000 observations, the number of epochs is set to 500.
|
281
|
+
* - For larger datasets, the number of epochs decreases from 500 according to the number of cells beyond 10000, to a lower limit of 200.
|
282
|
+
*
|
283
|
+
* This choice aims to reduce computational work for very large datasets.
|
284
|
+
*
|
285
|
+
* @return A reference to this `Umap` object.
|
286
|
+
*/
|
287
|
+
Umap& set_num_epochs(int n = Defaults::num_epochs) {
|
288
|
+
num_epochs = n;
|
289
|
+
return *this;
|
290
|
+
}
|
291
|
+
|
292
|
+
/**
|
293
|
+
* @param n Rate of sampling negative observations to compute repulsive forces.
|
294
|
+
* This is interpreted with respect to the number of neighbors with attractive forces, i.e., for each attractive interaction, `n` negative samples are taken for repulsive interactions.
|
295
|
+
* Smaller values can improve the speed of convergence but at the cost of stability.
|
296
|
+
*
|
297
|
+
* @return A reference to this `Umap` object.
|
298
|
+
*/
|
299
|
+
Umap& set_negative_sample_rate(Float n = Defaults::negative_sample_rate) {
|
300
|
+
negative_sample_rate = n;
|
301
|
+
return *this;
|
302
|
+
}
|
303
|
+
|
304
|
+
/**
|
305
|
+
* @param s Seed to use for the Mersenne Twister when sampling negative observations.
|
306
|
+
*
|
307
|
+
* @return A reference to this `Umap` object.
|
308
|
+
*/
|
309
|
+
Umap& set_seed(uint64_t s = Defaults::seed) {
|
310
|
+
seed = s;
|
311
|
+
return *this;
|
312
|
+
}
|
313
|
+
|
314
|
+
/**
|
315
|
+
* @param a Positive value for the $a$ parameter for the fuzzy set membership strength calculations.
|
316
|
+
* Larger values yield a sharper decay in membership strength with increasing distance between observations.
|
317
|
+
*
|
318
|
+
* If this or `set_b()` is set to zero, a suitable value for this parameter is automatically determined from the values provided to `set_spread()` and `set_min_dist()`.
|
319
|
+
*
|
320
|
+
* @return A reference to this `Umap` object.
|
321
|
+
*/
|
322
|
+
Umap& set_a(Float a = Defaults::a) {
|
323
|
+
rparams.a = a;
|
324
|
+
return *this;
|
325
|
+
}
|
326
|
+
|
327
|
+
/**
|
328
|
+
* @param b Value in $(0, 1)$ for the $b$ parameter for the fuzzy set membership strength calculations.
|
329
|
+
* Larger values yield an earlier decay in membership strength with increasing distance between observations.
|
330
|
+
*
|
331
|
+
* If this or `set_a()` is set to zero, a suitable value for this parameter is automatically determined from the values provided to `set_spread()` and `set_min_dist()`.
|
332
|
+
*
|
333
|
+
* @return A reference to this `Umap` object.
|
334
|
+
*/
|
335
|
+
Umap& set_b(Float b = Defaults::b) {
|
336
|
+
rparams.b = b;
|
337
|
+
return *this;
|
338
|
+
}
|
339
|
+
|
340
|
+
/**
|
341
|
+
* @param r Modifier for the repulsive force.
|
342
|
+
* Larger values increase repulsion and favor local structure.
|
343
|
+
*
|
344
|
+
* @return A reference to this `Umap` object.
|
345
|
+
*/
|
346
|
+
Umap& set_repulsion_strength(Float r = Defaults::repulsion_strength) {
|
347
|
+
rparams.repulsion_strength = r;
|
348
|
+
return *this;
|
349
|
+
}
|
350
|
+
|
351
|
+
/**
|
352
|
+
* @param l Initial learning rate used in the gradient descent.
|
353
|
+
* Larger values can improve the speed of convergence but at the cost of stability.
|
354
|
+
*
|
355
|
+
* @return A reference to this `Umap` object.
|
356
|
+
*/
|
357
|
+
Umap& set_learning_rate(Float l = Defaults::learning_rate) {
|
358
|
+
rparams.learning_rate = l;
|
359
|
+
return *this;
|
360
|
+
}
|
361
|
+
|
362
|
+
/**
|
363
|
+
* @param b Whether to optimize in batch mode.
|
364
|
+
* Batch mode is required for effective parallelization via OpenMP but may reduce the stability of the gradient descent.
|
365
|
+
*
|
366
|
+
* Batch mode involves computing forces for all observations and applying them simultaneously.
|
367
|
+
* This is in contrast to the default where the location of observation is updated before the forces are computed for the next observation.
|
368
|
+
* As each observation's forces are computed independently, batch mode is more amenable to parallelization;
|
369
|
+
* however, this comes at the cost of stability as the force calculations for later observations are not aware of updates to the positions of earlier observations.
|
370
|
+
*
|
371
|
+
* @return A reference to this `Umap` object.
|
372
|
+
*/
|
373
|
+
Umap& set_batch(bool b = Defaults::batch) {
|
374
|
+
rparams.batch = b;
|
375
|
+
return *this;
|
376
|
+
}
|
377
|
+
|
378
|
+
/**
|
379
|
+
* @param n Number of threads to use.
|
380
|
+
*
|
381
|
+
* @return A reference to this `Umap` object.
|
382
|
+
*
|
383
|
+
* This setting affects nearest neighbor detection (if an existing list of neighbors is not supplied in `initialize()` or `run()`) and spectral initialization.
|
384
|
+
* If `set_batch()` is `true`, multiple threads will also be used during layout optimization.
|
385
|
+
*
|
386
|
+
* The `UMAPPP_CUSTOM_PARALLEL` macro can be set to a function that specifies a custom parallelization scheme.
|
387
|
+
* This function should be a template that accept three arguments:
|
388
|
+
*
|
389
|
+
* - `njobs`, an integer specifying the number of jobs.
|
390
|
+
* - `fun`, a lambda that accepts two arguments, `start` and `end`.
|
391
|
+
* - `nthreads`, an integer specifying the number of threads to use.
|
392
|
+
*
|
393
|
+
* The function should split `[0, njobs)` into any number of contiguous, non-overlapping intervals, and call `fun` on each interval, possibly in different threads.
|
394
|
+
* The details of the splitting and evaluation are left to the discretion of the developer defining the macro.
|
395
|
+
* The function should only return once all evaluations of `fun` are complete.
|
396
|
+
*
|
397
|
+
* If `UMAPPP_CUSTOM_PARALLEL` is set, the `IRLBA_CUSTOM_PARALLEL` macro is also set if it is not already defined.
|
398
|
+
* This ensures that any custom parallelization scheme is propagated to all of **umappp**'s dependencies.
|
399
|
+
* If **irlba** is used outside of **umappp**, some care is required to ensure that the macros are consistently defined throughout the client library/application;
|
400
|
+
* otherwise, developers may observe ODR compilation errors.
|
401
|
+
*/
|
402
|
+
Umap& set_num_threads(int n = Defaults::num_threads) {
|
403
|
+
rparams.nthreads = n;
|
404
|
+
return *this;
|
405
|
+
}
|
406
|
+
|
407
|
+
public:
|
408
|
+
/**
|
409
|
+
* @brief Status of the UMAP optimization iterations.
|
410
|
+
*/
|
411
|
+
struct Status {
|
412
|
+
/**
|
413
|
+
* @cond
|
414
|
+
*/
|
415
|
+
Status(EpochData<Float> e, uint64_t seed, RuntimeParameters p) : epochs(std::move(e)), engine(seed), rparams(std::move(p)) {}
|
416
|
+
|
417
|
+
EpochData<Float> epochs;
|
418
|
+
std::mt19937_64 engine;
|
419
|
+
RuntimeParameters rparams;
|
420
|
+
/**
|
421
|
+
* @endcond
|
422
|
+
*/
|
423
|
+
|
424
|
+
/**
|
425
|
+
* @return Current epoch.
|
426
|
+
*/
|
427
|
+
int epoch() const {
|
428
|
+
return epochs.current_epoch;
|
429
|
+
}
|
430
|
+
|
431
|
+
/**
|
432
|
+
* @return Total number of epochs.
|
433
|
+
* This is equal to the value set by `set_num_epochs()` when the `Status` object is created.
|
434
|
+
*/
|
435
|
+
int num_epochs() const {
|
436
|
+
return epochs.total_epochs;
|
437
|
+
}
|
438
|
+
|
439
|
+
/**
|
440
|
+
* @return The number of observations in the dataset.
|
441
|
+
*/
|
442
|
+
size_t nobs() const {
|
443
|
+
return epochs.head.size();
|
444
|
+
}
|
445
|
+
|
446
|
+
/**
|
447
|
+
* @param ndim Number of dimensions of the embedding.
|
448
|
+
* @param[in, out] embedding Two-dimensional array where rows are dimensions (`ndim`) and columns are observations.
|
449
|
+
* This contains the initial coordinates and is updated to store the final embedding.
|
450
|
+
* @param epoch_limit Number of epochs to run to.
|
451
|
+
* The actual number of epochs performed is equal to the difference between `epoch_limit` and the current number of epochs in `epoch()`.
|
452
|
+
* `epoch_limit` should be not less than `epoch()` and no greater than the maximum number of epochs specified in `Umap::set_num_epochs()`.
|
453
|
+
* If zero, defaults to the maximum number of epochs.
|
454
|
+
*
|
455
|
+
* @return The status of the algorithm and the coordinates in `embedding` are updated to the specified number of epochs.
|
456
|
+
*/
|
457
|
+
void run(int ndim, Float* embedding, int epoch_limit = 0) {
|
458
|
+
if (!rparams.batch) {
|
459
|
+
optimize_layout(
|
460
|
+
ndim,
|
461
|
+
embedding,
|
462
|
+
epochs,
|
463
|
+
rparams.a,
|
464
|
+
rparams.b,
|
465
|
+
rparams.repulsion_strength,
|
466
|
+
rparams.learning_rate,
|
467
|
+
engine,
|
468
|
+
epoch_limit
|
469
|
+
);
|
470
|
+
} else {
|
471
|
+
optimize_layout_batched(
|
472
|
+
ndim,
|
473
|
+
embedding,
|
474
|
+
epochs,
|
475
|
+
rparams.a,
|
476
|
+
rparams.b,
|
477
|
+
rparams.repulsion_strength,
|
478
|
+
rparams.learning_rate,
|
479
|
+
[&]() -> auto { return engine(); },
|
480
|
+
[](decltype(engine()) s) -> auto { return std::mt19937_64(s); },
|
481
|
+
epoch_limit,
|
482
|
+
rparams.nthreads
|
483
|
+
);
|
484
|
+
}
|
485
|
+
return;
|
486
|
+
}
|
487
|
+
};
|
488
|
+
|
489
|
+
/**
|
490
|
+
* @param x Indices and distances to the nearest neighbors for each observation.
|
491
|
+
* Note the expectations in the `NeighborList` documentation.
|
492
|
+
* @param ndim Number of dimensions of the embedding.
|
493
|
+
* @param[in, out] embedding Two-dimensional array to store the embedding,
|
494
|
+
* where rows are dimensions (`ndim`) and columns are observations (`x.size()`).
|
495
|
+
*
|
496
|
+
* @return A `Status` object containing the initial state of the UMAP algorithm, to be used in `run()`.
|
497
|
+
* If `set_initialize()` is `NONE` or if spectral initialization fails with `SPECTRAL_ONLY`, `embedding` should contain the initial coordinates and will not be altered;
|
498
|
+
* otherwise, it is filled with initial coordinates.
|
499
|
+
*/
|
500
|
+
Status initialize(NeighborList<Float> x, int ndim, Float* embedding) const {
|
501
|
+
neighbor_similarities(x, local_connectivity, bandwidth);
|
502
|
+
combine_neighbor_sets(x, mix_ratio);
|
503
|
+
|
504
|
+
// Choosing the manner of initialization.
|
505
|
+
if (init == SPECTRAL || init == SPECTRAL_ONLY) {
|
506
|
+
bool attempt = spectral_init(x, ndim, embedding, rparams.nthreads);
|
507
|
+
if (!attempt && init == SPECTRAL) {
|
508
|
+
random_init(x.size(), ndim, embedding);
|
509
|
+
}
|
510
|
+
} else if (init == RANDOM) {
|
511
|
+
random_init(x.size(), ndim, embedding);
|
512
|
+
}
|
513
|
+
|
514
|
+
// Finding a good a/b pair.
|
515
|
+
auto pcopy = rparams;
|
516
|
+
if (pcopy.a <= 0 || pcopy.b <= 0) {
|
517
|
+
auto found = find_ab(spread, min_dist);
|
518
|
+
pcopy.a = found.first;
|
519
|
+
pcopy.b = found.second;
|
520
|
+
}
|
521
|
+
|
522
|
+
int num_epochs_to_do = choose_num_epochs(num_epochs, x.size());
|
523
|
+
|
524
|
+
return Status(
|
525
|
+
similarities_to_epochs(x, num_epochs_to_do, negative_sample_rate),
|
526
|
+
seed,
|
527
|
+
std::move(pcopy)
|
528
|
+
);
|
529
|
+
}
|
530
|
+
|
531
|
+
public:
|
532
|
+
/**
|
533
|
+
* @tparam Algorithm `knncolle::Base` subclass implementing a nearest neighbor search algorithm.
|
534
|
+
*
|
535
|
+
* @param searcher Pointer to a `knncolle::Base` subclass with a `find_nearest_neighbors()` method.
|
536
|
+
* @param ndim Number of dimensions of the embedding.
|
537
|
+
* @param[out] embedding Two-dimensional array to store the embedding,
|
538
|
+
* where rows are dimensions (`ndim`) and columns are observations (`searcher->nobs()`).
|
539
|
+
*
|
540
|
+
* @return A `Status` object containing the initial state of the UMAP algorithm, to be used in `run()`.
|
541
|
+
* If `set_initialize()` is true, `embedding` is filled with initial coordinates derived from the fuzzy set graph;
|
542
|
+
* otherwise it is ignored.
|
543
|
+
*/
|
544
|
+
template<class Algorithm>
|
545
|
+
Status initialize(const Algorithm* searcher, int ndim, Float* embedding) {
|
546
|
+
const size_t N = searcher->nobs();
|
547
|
+
NeighborList<Float> output(N);
|
548
|
+
|
549
|
+
#ifndef UMAPPP_CUSTOM_PARALLEL
|
550
|
+
#pragma omp parallel for num_threads(rparams.nthreads)
|
551
|
+
for (size_t i = 0; i < N; ++i) {
|
552
|
+
#else
|
553
|
+
UMAPPP_CUSTOM_PARALLEL(N, [&](size_t first, size_t last) -> void {
|
554
|
+
for (size_t i = first; i < last; ++i) {
|
555
|
+
#endif
|
556
|
+
|
557
|
+
output[i] = searcher->find_nearest_neighbors(i, num_neighbors);
|
558
|
+
|
559
|
+
#ifndef UMAPPP_CUSTOM_PARALLEL
|
560
|
+
}
|
561
|
+
#else
|
562
|
+
}
|
563
|
+
}, rparams.nthreads);
|
564
|
+
#endif
|
565
|
+
|
566
|
+
return initialize(std::move(output), ndim, embedding);
|
567
|
+
}
|
568
|
+
|
569
|
+
#ifndef UMAPPP_CUSTOM_NEIGHBORS
|
570
|
+
/**
|
571
|
+
* @tparam Input Floating point type for the input data.
|
572
|
+
*
|
573
|
+
* @param ndim_in Number of dimensions.
|
574
|
+
* @param nobs Number of observations.
|
575
|
+
* @param[in] input Pointer to a 2D array containing the input high-dimensional data, with number of rows and columns equal to `ndim_in` and `nobs`, respectively.
|
576
|
+
* The array is treated as column-major where each row corresponds to a dimension and each column corresponds to an observation.
|
577
|
+
* @param ndim_out Number of dimensions of the embedding.
|
578
|
+
* @param[out] embedding Two-dimensional array to store the embedding,
|
579
|
+
* where rows are dimensions (`ndim`) and columns are observations (`searcher->nobs()`).
|
580
|
+
*
|
581
|
+
* @return A `Status` object containing various pre-computed structures required for the epochs in `run()`.
|
582
|
+
* If `set_initialize()` is `NONE` or if spectral initialization fails with `SPECTRAL_ONLY`, `embedding` should contain the initial coordinates and will not be altered;
|
583
|
+
* otherwise, it is filled with initial coordinates.
|
584
|
+
*
|
585
|
+
* This differs from the other `initialize()` methods in that it will internally compute the nearest neighbors for each observation.
|
586
|
+
* It will use vantage point trees for the search - see the other `initialize()` methods to specify a custom search algorithm.
|
587
|
+
*/
|
588
|
+
template<typename Input = Float>
|
589
|
+
Status initialize(int ndim_in, size_t nobs, const Input* input, int ndim_out, Float* embedding) {
|
590
|
+
knncolle::VpTreeEuclidean<> searcher(ndim_in, nobs, input);
|
591
|
+
return initialize(&searcher, ndim_out, embedding);
|
592
|
+
}
|
593
|
+
#endif
|
594
|
+
|
595
|
+
public:
|
596
|
+
/**
|
597
|
+
* @tparam Algorithm `knncolle::Base` subclass implementing a nearest neighbor search algorithm.
|
598
|
+
*
|
599
|
+
* @param searcher Pointer to a `knncolle::Base` subclass with a `find_nearest_neighbors()` method.
|
600
|
+
* @param ndim Number of dimensions of the embedding.
|
601
|
+
* @param[in, out] embedding Two-dimensional array where rows are dimensions (`ndim`) and columns are observations (`searcher->nobs()`).
|
602
|
+
* This is filled with the final embedding on output.
|
603
|
+
* If `set_initialize()` is false, this is assumed to contain the initial coordinates on input.
|
604
|
+
* @param epoch_limit Number of epochs to run - see `Status::run()`.
|
605
|
+
*
|
606
|
+
* @return The status of the algorithm is returned after running up to `epoch_limit`; this can be used for further iterations by invoking `Status::run()`.
|
607
|
+
* `embedding` is updated with the embedding at the specified epoch limit.
|
608
|
+
*/
|
609
|
+
template<class Algorithm>
|
610
|
+
Status run(const Algorithm* searcher, int ndim, Float* embedding, int epoch_limit = 0) {
|
611
|
+
auto status = initialize(searcher, ndim, embedding);
|
612
|
+
status.run(ndim, embedding, epoch_limit);
|
613
|
+
return status;
|
614
|
+
}
|
615
|
+
|
616
|
+
/**
|
617
|
+
* @param x Indices and distances to the nearest neighbors for each observation.
|
618
|
+
* Note the expectations in the `NeighborList` documentation.
|
619
|
+
* @param ndim Number of dimensions of the embedding.
|
620
|
+
* @param[in, out] embedding Two-dimensional array where rows are dimensions (`ndim`) and columns are observations.
|
621
|
+
* This is filled with the final embedding on output.
|
622
|
+
* If `set_initialize()` is `NONE` or if spectral initialization fails with `SPECTRAL_ONLY`, `embedding` is assumed to contain the initial coordinates on input.
|
623
|
+
* @param epoch_limit Number of epochs to run - see `Status::run()`.
|
624
|
+
*
|
625
|
+
* @return The status of the algorithm is returned after running up to `epoch_limit`; this can be used for further iterations by invoking `Status::run()`.
|
626
|
+
* `embedding` is updated with the embedding at the specified epoch limit.
|
627
|
+
*/
|
628
|
+
Status run(NeighborList<Float> x, int ndim, Float* embedding, int epoch_limit = 0) const {
|
629
|
+
auto status = initialize(std::move(x), ndim, embedding);
|
630
|
+
status.run(ndim, embedding, epoch_limit);
|
631
|
+
return status;
|
632
|
+
}
|
633
|
+
|
634
|
+
#ifndef UMAPPP_CUSTOM_NEIGHBORS
|
635
|
+
/**
|
636
|
+
* @tparam Input Floating point type for the input data.
|
637
|
+
*
|
638
|
+
* @param ndim_in Number of dimensions.
|
639
|
+
* @param nobs Number of observations.
|
640
|
+
* @param[in] input Pointer to a 2D array containing the input high-dimensional data, with number of rows and columns equal to `ndim_in` and `nobs`, respectively.
|
641
|
+
* The array is treated as column-major where each row corresponds to a dimension and each column corresponds to an observation.
|
642
|
+
* @param ndim_out Number of dimensions of the embedding.
|
643
|
+
* @param[in, out] embedding Two-dimensional array where rows are dimensions (`ndim`) and columns are observations (`searcher->nobs()`).
|
644
|
+
* This is filled with the final embedding on output.
|
645
|
+
* If `set_initialize()` is `NONE` or if spectral initialization fails with `SPECTRAL_ONLY`, `embedding` is assumed to contain the initial coordinates on input.
|
646
|
+
* @param epoch_limit Number of epochs to run - see `Status::run()`.
|
647
|
+
*
|
648
|
+
* @return The status of the algorithm is returned after running up to `epoch_limit`; this can be used for further iterations by invoking `Status::run()`.
|
649
|
+
* `embedding` is updated with the embedding at the specified epoch limit.
|
650
|
+
*/
|
651
|
+
template<typename Input = Float>
|
652
|
+
Status run(int ndim_in, size_t nobs, const Input* input, int ndim_out, Float* embedding, int epoch_limit = 0) {
|
653
|
+
auto status = initialize(ndim_in, nobs, input, ndim_out, embedding);
|
654
|
+
status.run(ndim_out, embedding, epoch_limit);
|
655
|
+
return status;
|
656
|
+
}
|
657
|
+
#endif
|
658
|
+
};
|
659
|
+
|
660
|
+
}
|
661
|
+
|
662
|
+
#endif
|