úÎ!<Ó9=5      !"#$%&'()*+,-./01234None-QSTV]95emd>Treats every item in a "plateu" as a local minimum or maximum.5None -FSTVi G6emd :https://en.wikipedia.org/wiki/Tridiagonal_matrix_algorithm Will return 78 if the matrix is not invertible. This will happen if: +The first item in the main diagonal is zeroœThere is any i such that b_{i + 1} = a_i * c_i. That is, an item in the main diagonal is equal to the product of the off-diagonal elements a row above itAnother mystery condition!6emda: Bottom diagonal of Memdb: Main diagonal of Memdc: Upper diagonal of Memdyemdx such that M x = y6(c) Justin Le 2018BSD3 justin@jle.im experimental non-portableNone "#&'-/STVi¥ emd1D Cubic spline8emda9emdb:emdy_{i-1};emdy_i<emd x_i - x_{i-1}emdEnd condition for splineemd!Sample a spline at a given point.emdaBuild a cubic spline based on control points using given end conditions (not-a-knot, or natural) 2https://en.wikipedia.org/wiki/Spline_interpolationemd(x, y)(c) Justin Le 2018BSD3 justin@jle.im experimental non-portableNone"#&',-FSTV])£ emdVThe result of a sifting operation. Each sift either yields a residual, or a new IMF. emdnumber of iterations emdAn   v n a5 is a Hilbert-Huang transform of a time series with n items of type a stored in a vector v.emd#Stop conditions for sifting processemdStop using standard SD methodemd'Stop after a fixed number of iterationsemdone or the otheremdboth conditions metemdOptions for EMD composition.emdstop condition for siftingemd#end conditions for envelope splinesemdif =<, use time series endpoints as part of min and max envelopesemdDefault emdDefault >emd= if stopemdhEMD decomposition (Hilbert-Huang Transform) of a given time series with a given sifting stop condition.$Takes a sized vector to ensure that: The resulting  D contains IMFs that are all the same length as the input vector,We provide a vector of size of at least one.emdn, but tracing results to stdout as IMFs are found. Useful for debugging to see how long each step is taking.emd$ with a callback for each found IMF. emdFIterated sifting process, used to produce either an IMF or a residual.?emd Single sift!emd@Returns cubic splines of local minimums and maximums. Returns 7J if there are not enough local minimum or maximums to create the splines.@emd7Build a splined vector against a map of control points.  !  !(c) Justin Le 2018BSD3 justin@jle.im experimental non-portableNone &'-STV]i8i)emdVThe result of a sifting operation. Each sift either yields a residual, or a new IMF.+emdnumber of iterations,emdAn , v aC is a Hilbert-Huang transform of a time series with items of type a stored in a vector v.0emdhEMD decomposition (Hilbert-Huang Transform) of a given time series with a given sifting stop condition.Returns 7 if given an empty vector.See E for a type-safe version with guaruntees on the output vector sizes.1emd0n, but tracing results to stdout as IMFs are found. Useful for debugging to see how long each step is taking.Returns 7 if given an empty vector.2emd0$ with a callback for each found IMF.Returns 7 if given an empty vector.3emdFIterated sifting process, used to produce either an IMF or a residual.Returns 7 if given an empty vector.)*+,-./0123012,-./3)*+A      !"#$%&'()*+,-#$%-./0123456789:;<=>"emd-0.1.1.0-EM5YMZeOmXAAJtQRPcHziONumeric.EMD.Internal.Spline Numeric.EMDNumeric.EMD.UnsizedNumeric.EMD.Internal.Extrema Numeric.EMD.Internal.TridiagonalemdSpline SplineEnd SENotAKnot SENatural sampleSpline makeSpline$fShowSplineEnd $fEqSplineEnd$fOrdSplineEnd$fShowSplineCoef SiftResult SRResidualSRIMFEMDemdIMFs emdResidual SiftConditionSCStdDevSCTimesSCOrSCAndEMDOptsEOeoSiftCondition eoSplineEndeoClampEnvelope defaultEO defaultSCemdTraceemd'sift envelopes$fShowSiftCondition$fEqSiftCondition$fOrdSiftCondition $fShowEMDOpts $fEqEMDOpts $fOrdEMDOpts $fShowEMDextremasolveTridiagonalbaseGHC.BaseNothing_scAlpha_scBeta _scGamma0 _scGamma1_scDeltaghc-prim GHC.TypesTrue testConditionsift' splineAgainst