plateform stringclasses 1
value | repo_name stringlengths 13 113 | name stringlengths 3 74 | ext stringclasses 1
value | path stringlengths 12 229 | size int64 23 843k | source_encoding stringclasses 9
values | md5 stringlengths 32 32 | text stringlengths 23 843k |
|---|---|---|---|---|---|---|---|---|
github | chaosuo/hctsa-master | TSQ_ForwardFS.m | .m | hctsa-master/PlottingAnalysis/TSQ_ForwardFS.m | 12,912 | utf_8 | cc91b420c4c10c06600eed8219a28ddd | % --------------------------------------------------------------------------
% TSQ_ForwardFS
% --------------------------------------------------------------------------
%
% Performs greedy forward feature selection for a given classification of the
% data. After selecting the features (using specified training indice... |
github | chaosuo/hctsa-master | TSQ_IndividualFeatures.m | .m | hctsa-master/PlottingAnalysis/TSQ_IndividualFeatures.m | 18,825 | utf_8 | 268957d4f593bba976f63f0262ef0f45 | % ------------------------------------------------------------------------------
% TSQ_IndividualFeatures
% ------------------------------------------------------------------------------
%
% Searches for individual features that help distinguish a known classification
% of the time series
%
%---INPUTS
% ClassMeth: the... |
github | chaosuo/hctsa-master | TSQ_plot_distribution.m | .m | hctsa-master/PlottingAnalysis/TSQ_plot_distribution.m | 4,577 | utf_8 | 96317126fe7d7fafc836eb5f4d44178d | % ------------------------------------------------------------------------------
% TSQ_plot_distribution
% ------------------------------------------------------------------------------
%
% Input a vector (FeatureVector) and a labeling of its elements (GroupLabels),
% and this will plot a distribution of it.
%
% Coup... |
github | chaosuo/hctsa-master | TSQ_plot_ColorDendrogram.m | .m | hctsa-master/PlottingAnalysis/TSQ_plot_ColorDendrogram.m | 9,355 | utf_8 | 0bccc11cd1b1bae18874c79e2c3b7cd0 | % ------------------------------------------------------------------------------
% TSQ_plot_ColorDendrogram
% ------------------------------------------------------------------------------
% Creates a dendrogram for distance matrix d with colours at each leaf
%
%---INPUTS:
% d is the (NxN) pairwise distance matrix (e.... |
github | chaosuo/hctsa-master | TSQ_cluster.m | .m | hctsa-master/PlottingAnalysis/TSQ_cluster.m | 7,927 | utf_8 | f45281bea2023be1f3dda06f400ecb37 | % --------------------------------------------------------------------------
% TSQ_cluster
% --------------------------------------------------------------------------
%
% Reads in normalized data from HCTSA_N.mat, clusters the data matrix by
% reordering rows and columns with linkage clustering, and then saves the re... |
github | chaosuo/hctsa-master | TSQ_LabelGroups.m | .m | hctsa-master/PlottingAnalysis/TSQ_LabelGroups.m | 8,502 | utf_8 | 622577da42e5515cf8e05646f036d670 | % --------------------------------------------------------------------------
% TSQ_LabelGroups
% --------------------------------------------------------------------------
%
% You provide a set of keyword options to store a specific grouping of time series.
% Useful when doing a classification task -- can store your c... |
github | chaosuo/hctsa-master | TSQ_plot_timeseries.m | .m | hctsa-master/PlottingAnalysis/TSQ_plot_timeseries.m | 12,001 | utf_8 | 71c6aec082cc0201b8772e863b44d06b | % ------------------------------------------------------------------------------
% TSQ_plot_timeseries
% ------------------------------------------------------------------------------
%
% Plots the time series read from a local file, in a specified format.
%
%---INPUTS:
% whatData, The data to get information from: c... |
github | chaosuo/hctsa-master | TSQ_MHT_pvalue.m | .m | hctsa-master/PlottingAnalysis/TSQ_MHT_pvalue.m | 12,645 | utf_8 | ed7576c92829c09d4993897179c1e176 | % ------------------------------------------------------------------------------
% TSQ_MHT_pvalue
% ------------------------------------------------------------------------------
%
% Returns p-values given a distribution of test stastics and an appropriate null
% distribution. cf. algorithm 18.3, etc. in Hastie's stat... |
github | chaosuo/hctsa-master | TSQ_plot_DataMatrix.m | .m | hctsa-master/PlottingAnalysis/TSQ_plot_DataMatrix.m | 9,498 | utf_8 | 841ed38cb4418debc8dc033dcd17335d | % ------------------------------------------------------------------------------
% TSQ_plot_DataMatrix
% ------------------------------------------------------------------------------
%
% Plot the data matrix.
%
%---INPUTS:
% whatData: specify 'norm' for normalized data in HCTSA_N.mat, 'cl' for clustered
% da... |
github | yusong-shen/CIFAR10_multiclassification-master | checkNumericalGradient.m | .m | CIFAR10_multiclassification-master/checkNumericalGradient.m | 1,982 | utf_8 | 689a352eb2927b0838af5dc508f6374d | function [] = checkNumericalGradient()
% This code can be used to check your numerical gradient implementation
% in computeNumericalGradient.m
% It analytically evaluates the gradient of a very simple function called
% simpleQuadraticFunction (see below) and compares the result with your numerical
% solution. Your num... |
github | yusong-shen/CIFAR10_multiclassification-master | WolfeLineSearch.m | .m | CIFAR10_multiclassification-master/minFunc/WolfeLineSearch.m | 11,023 | utf_8 | e78201dab344cc6fa1a101af3e1e8eec | function [t,f_new,g_new,funEvals,H] = WolfeLineSearch(...
x,t,d,f,g,gtd,c1,c2,LS,maxLS,tolX,debug,doPlot,saveHessianComp,funObj,varargin)
%
% Bracketing Line Search to Satisfy Wolfe Conditions
%
% Inputs:
% x: starting location
% t: initial step size
% d: descent direction
% f: function value at starting lo... |
github | yusong-shen/CIFAR10_multiclassification-master | minFunc_processInputOptions.m | .m | CIFAR10_multiclassification-master/minFunc/minFunc_processInputOptions.m | 3,551 | utf_8 | ea7fbcf303b9cafeca4045921adad934 |
function [verbose,verboseI,debug,doPlot,maxFunEvals,maxIter,tolFun,tolX,method,...
corrections,c1,c2,LS_init,LS,cgSolve,qnUpdate,cgUpdate,initialHessType,...
HessianModify,Fref,useComplex,numDiff,LS_saveHessianComp,...
DerivativeCheck,Damped,HvFunc,bbType,cycle,...
HessianIter,outputFcn,useMex,useNegCu... |
github | suryanshkumar/Structure-from-Motion-master | reconstruct_3d_points.m | .m | Structure-from-Motion-master/src_matlab/mvg_modules/reconstruction_module/reconstruct_3d_points.m | 2,969 | utf_8 | 85926344a0f7f193763c1a166b257130 | % Author: Suryansh Kumar, ETH Zurich
function [R_rel, t_rel, reconstructed_points] = reconstruct_3d_points(...
x_ref, x_nex, K)
% step 1: compute fundamental matrix
% Used Peter Kovesi script for estimation fundamental matrix.
F = fundmatrix([x_ref'; ones(1, length(x_ref))], ...
[x_nex'; ones(1, length(x_nex... |
github | suryanshkumar/Structure-from-Motion-master | triangulate_points.m | .m | Structure-from-Motion-master/src_matlab/mvg_modules/reconstruction_module/triangulate_points.m | 1,007 | utf_8 | d612da404c4a9247e5e76a0ee0295370 | % Author: Suryansh Kumar, ETH Zurich
% Refer Hartley and Zisserman Multiple View Geometry Book
% page 312, Linear Triangulation Method.
% Input:
% x_ref: reference image 2d key-points.
% x_nex: next image corresponding points.
% scale: for homogeneous it is set to 1.
% P_ref: reference image projection matrix.
% P_ne... |
github | suryanshkumar/Structure-from-Motion-master | compute_chiral_estimate.m | .m | Structure-from-Motion-master/src_matlab/mvg_modules/reconstruction_module/compute_chiral_estimate.m | 321 | utf_8 | 1cb85845b55a7a3851d83974dabcfb29 | % Author: Suryansh Kumar, ETH Zurich
function chiral_val = compute_chiral_estimate(P1, P2, X)
X = X'; %for matrix multiplication
p1x = P1*X; %projection to first camera.
p2x = P2*X; %projection to second camera.
chiral_val = (sign(p1x(3,:) ) .* sign (X(4,:))) + ...
(sign(p2x(3,:) ) .* sign (X(4,:)));
end
... |
github | suryanshkumar/Structure-from-Motion-master | compute_possible_poses.m | .m | Structure-from-Motion-master/src_matlab/mvg_modules/pose_module/compute_possible_poses.m | 498 | utf_8 | d1605ad0257d2d3210c572b3fef8247d | % Author: Suryansh Kumar, ETH
% Refer Hartley and Zisserman Multiple View Geometry Book
% page 258-259, Result 9.18-9.19.
function [R1, R2, t1, t2] = compute_possible_poses(E)
W = [0, -1, 0; 1, 0, 0; 0, 0, 1];
[U, ~, V] = svd(E);
% the two possible rotation
R1 = U*W*V';
R2 = U*W'*V';
% the two possible translation
... |
github | suryanshkumar/Structure-from-Motion-master | fundmatrix.m | .m | Structure-from-Motion-master/src_matlab/mvg_modules/pose_module/fundmatrix.m | 4,069 | utf_8 | 632e6f9e26790316764a91117aa2adb9 | % FUNDMATRIX - computes fundamental matrix from 8 or more points
%
% Function computes the fundamental matrix from 8 or more matching points in
% a stereo pair of images. The normalised 8 point algorithm given by
% Hartley and Zisserman p265 is used. To achieve accurate results it is
% recommended that 12 or more poi... |
github | suryanshkumar/Structure-from-Motion-master | normalise2dpts.m | .m | Structure-from-Motion-master/src_matlab/mvg_modules/pose_module/normalise2dpts.m | 2,346 | utf_8 | 95482f59fae27b790b236af0f825ea81 | % NORMALISE2DPTS - normalises 2D homogeneous points
%
% Function translates and normalises a set of 2D homogeneous points
% so that their centroid is at the origin and their mean distance from
% the origin is sqrt(2). This process typically improves the
% conditioning of any equations used to solve homographies, fun... |
github | ExtremeLearningMachines/ELM-MATLAB-and-Online.Sequential.ELM-master | elm_kernel.m | .m | ELM-MATLAB-and-Online.Sequential.ELM-master/elmbase/elm_kernel.m | 7,866 | utf_8 | 7b001834553880add8c9ab5f8d5c0fa0 | function [TrainingTime, TestingTime, TrainingAccuracy, TestingAccuracy,TY] = elm_kernel(TrainingData_File, TestingData_File, Elm_Type, Regularization_coefficient, Kernel_type, Kernel_para)
% Usage: elm(TrainingData_File, TestingData_File, Elm_Type, NumberofHiddenNeurons, ActivationFunction)
% OR: [TrainingTime, ... |
github | AurelienRoy/ardupilot-master | RotToQuat.m | .m | ardupilot-master/libraries/AP_NavEKF/Models/Common/RotToQuat.m | 288 | utf_8 | 9239706354267c8f5f2a29f992c07de9 | % convert froma rotation vector in radians to a quaternion
function quaternion = RotToQuat(rotVec)
vecLength = sqrt(rotVec(1)^2 + rotVec(2)^2 + rotVec(3)^2);
if vecLength < 1e-6
quaternion = [1;0;0;0];
else
quaternion = [cos(0.5*vecLength); rotVec/vecLength*sin(0.5*vecLength)];
end |
github | AurelienRoy/ardupilot-master | NormQuat.m | .m | ardupilot-master/libraries/AP_NavEKF/Models/Common/NormQuat.m | 198 | utf_8 | ed913e87efc9194a2c52b266fced8da7 | % normalise the quaternion
function quaternion = normQuat(quaternion)
quatMag = sqrt(quaternion(1)^2 + quaternion(2)^2 + quaternion(3)^2 + quaternion(4)^2);
quaternion(1:4) = quaternion / quatMag;
|
github | AurelienRoy/ardupilot-master | QuatToEul.m | .m | ardupilot-master/libraries/AP_NavEKF/Models/Common/QuatToEul.m | 436 | utf_8 | c19c9235052d99b8b943a7157e83fc94 | % Convert from a quaternion to a 321 Euler rotation sequence in radians
function Euler = QuatToEul(quat)
Euler = zeros(3,1);
Euler(1) = atan2(2*(quat(3)*quat(4)+quat(1)*quat(2)), quat(1)*quat(1) - quat(2)*quat(2) - quat(3)*quat(3) + quat(4)*quat(4));
Euler(2) = -asin(2*(quat(2)*quat(4)-quat(1)*quat(3)));
Euler(3) =... |
github | mriphysics/circuit_cosimulation-master | optimfun_circ.m | .m | circuit_cosimulation-master/optimfun_circ.m | 1,433 | utf_8 | 173c9586a56e95a55bcb851d0a9197f0 | % Optimisation cost function for calculating optimal capcitor values as
% part of circuit co-simulation framework.
%
% Currently only supports optimisation of capacitor values but can easily
% be modified to optimise all lumped element parameters.
%
% Created by Arian Beqiri, King's College London, November 2014.
% Em... |
github | mriphysics/circuit_cosimulation-master | circ_co_sim_optimisation.m | .m | circuit_cosimulation-master/circ_co_sim_optimisation.m | 4,854 | utf_8 | f075f921bf9fdae9057871693f1d3831 | % Circuit Co-Simulation Optimisation function - Top-level script to attempt
% to find optimal capacitor values for a given EM simulation using circuit
% simulation at a given frequency
%
% circ_co_sim_optimisation(S,freqs,nPorts,nLumped,capValsInit,
% portCapInds,lambda,varargin)
%
% ... |
github | antoine-cha/neuroProject-master | preprocessFolder.m | .m | neuroProject-master/matlab/preprocessFolder.m | 1,958 | utf_8 | d0553382adf92ca441a055bf0c3ec6a6 | function preprocessFolder(source, target, vanhateren)
% PREPROCESSFOLDER(source, target)
% Preprocess a folder of images (source)
% Write the new images at target
%
% -----------------------------------
% source : string
% absolute path to the source folder
% target : string
% ... |
github | antoine-cha/neuroProject-master | extractPatches.m | .m | neuroProject-master/matlab/extractPatches.m | 2,931 | utf_8 | 02d631ebb666bb7862efa8cad01e22cb | function extractPatches(source, target)
%EXTRACTPATCHES(source, target)
% extract random patches from images in source folder
% Write the new images at target
%-----------------------------------
%source : string
% absolute path to the source folder
%target : string
% absolute ... |
github | antoine-cha/neuroProject-master | shift.m | .m | neuroProject-master/matlab/shift.m | 438 | utf_8 | 2ac42b171e4ca683ef1d361eaf331870 | % [RES] = shift(MTX, OFFSET)
%
% Circular shift 2D matrix samples by OFFSET (a [Y,X] 2-vector),
% such that RES(POS) = MTX(POS-OFFSET).
function res = shift(mtx, offset)
dims = size(mtx);
offset = mod(-offset,dims);
res = [ mtx(offset(1)+1:dims(1), offset(2)+1:dims(2)), ...
mtx(offset(1)+1:dims(1), 1:o... |
github | antoine-cha/neuroProject-master | learn.m | .m | neuroProject-master/matlab/learn.m | 5,847 | utf_8 | 06ccae064cf81594e24860f6e1a99910 | function learn()
% function learn()
%
% main function
%
% yan karklin. jan 2010.
%% Learning Rates
params.epsy = 0.005; % step size for inference of y's
params.epsb = 0.0000005; % step size for gradient ascent on dL/dB
params.epsw = 0.0000005; % step size for gradient ascent on dL/dw
%%
params.decayw = ... |
github | antoine-cha/neuroProject-master | inferLatents.m | .m | neuroProject-master/matlab/inferLatents.m | 2,120 | utf_8 | 538528047523d35e46b000a160696611 | function Data = inferLatents(Model, Data, params)
% function Data = inferLatents(Model, Data, params)
%
% yan karklin. jan 2010.
N = params.N;
if params.debug, Hist.L = calcL(Model, Data, params); Hist.y = Data.y(1:20);end;
% inference loop to compute \hat{y}
for i = 1:params.yiters,
% start with small steps (... |
github | Korogodin/matlab-glonass-signal-choice-master | resize_array.m | .m | matlab-glonass-signal-choice-master/resize_array.m | 656 | utf_8 | 115f57329e45d8c9413c71b41c3ed6d7 | %/**
% Растяжение массива на N точек
%@param in_a - входной массив
%@param N - размер выходного массива
%*/
function out_a = resize_array(in_a, N)
S = length(in_a);
if N <= S % Размер выходного массива должен быть больше или равен входного
if N == S
out_a = in_a;
else
out_a = NaN;
ret... |
github | JJ-Ho/COSMOSS-master | progressbar.m | .m | COSMOSS-master/MoleculeConstruction/Shared_Func/progressbar.m | 12,126 | utf_8 | 57c81f277bf66bb3875922a2e7b9f221 | function progressbar(varargin)
% Description:
% progressbar() provides an indication of the progress of some task using
% graphics and text. Calling progressbar repeatedly will update the figure and
% automatically estimate the amount of time remaining.
% This implementation of progressbar is intended to be ex... |
github | JJ-Ho/COSMOSS-master | pdbread.m | .m | COSMOSS-master/MoleculeConstruction/Shared_Func/pdbread.m | 53,375 | utf_8 | f03b3b0a7711081c73667ba16a71c57f | function PDB_struct = pdbread(pdbfile,varargin)
%PDBREAD reads a Protein Data Bank (PDB) coordinate entry file.
%
% PDBSTRUCT = PDBREAD(PDBFILE) reads in a PDB coordinate entry file
% PBDFILE and creates a structure PDBSTRUCT containing fields
% corresponding to the PDB records. The coordinate information is stor... |
github | JJ-Ho/COSMOSS-master | Pointer_N.m | .m | COSMOSS-master/SpecFunc/Plotting/Pointer_N.m | 3,115 | utf_8 | bee9235a2f0168c393ab2d9c0f4f1140 | function [] = Pointer_N(S)
% Demonstrate how to display the current location of the mouse in an axes.
% Run the GUI then move the cursor over the axes. The current location of
% the pointer in the axes will be displayed at the top of the plot, in axes
% units.
%
% Suggested exercise: Make this function to take an ax... |
github | JJ-Ho/COSMOSS-master | Pointer_T.m | .m | COSMOSS-master/SpecFunc/Plotting/Pointer_T.m | 933 | utf_8 | faf577c73c090638e1f5dbf3595ccfdb | function [] = Pointer_T(S)
% Put the cursor values at the second line of title
% pPrep variabes
hF = S.hF;
hAx = S.hAx;
% Find old Pointer ui elements and delete if exist
hOld_Number = findobj(hF,'Tag','Pointer_Number');
if ishandle(hOld_Number)
delete(hOld_Number)
end
% Find and replace the original title stri... |
github | JJ-Ho/COSMOSS-master | Cuts.m | .m | COSMOSS-master/AnalysisTools/Cuts.m | 3,791 | utf_8 | 05e88badd7b986d4876a58312694f6b4 | function Cuts(varargin)
%% debug
% hF = figure(1);
% hAx_Plot = 'New';
% Direction = 'X';
%% Select Figure to be cut, even without the hF input
if eq(nargin,0)
hF_All = get(0,'Children');
N_F_All = length(hF_All);
if eq(N_F_All,0)
disp('No figures to be cut...')
return
end
Avaliab... |
github | JJ-Ho/COSMOSS-master | Update_Modes_Figure.m | .m | COSMOSS-master/AnalysisTools/ResponseVisulization/Update_Modes_Figure.m | 7,079 | utf_8 | 7100eee2da1a2b59b3b57f6156413d7c | function Output = Update_Modes_Figure(GUI_Inputs, Structure, SpecData)
%% Inputs parser
GUI_Inputs_C = fieldnames(GUI_Inputs);
GUI_Inputs_C(:,2) = struct2cell(GUI_Inputs);
GUI_Inputs_C = GUI_Inputs_C';
INPUT = inputParser;
INPUT.KeepUnmatched = 1;
% Default values
defaultMode_Type = 'Exciton Modes';
d... |
github | JJ-Ho/COSMOSS-master | arrow3.m | .m | COSMOSS-master/AnalysisTools/ResponseVisulization/arrow3.m | 35,020 | utf_8 | 0efb8c330ac9c3119e766e98cdf11f40 | function hn=arrow3(ax,p1,p2,s,w,h,ip,alpha,beta)
% ARROW3 (R13)
% ARROW3(P1,P2) draws lines from P1 to P2 with directional arrowheads.
% P1 and P2 are either nx2 or nx3 matrices. Each row of P1 is an
% initial point, and each row of P2 is a terminal point.
%
% ARROW3(P1,P2,S,W,H,IP,ALPHA,BETA) can be use... |
github | JJ-Ho/COSMOSS-master | distinguishable_colors.m | .m | COSMOSS-master/AnalysisTools/Orientation/distinguishable_colors.m | 5,753 | utf_8 | 57960cf5d13cead2f1e291d1288bccb2 | function colors = distinguishable_colors(n_colors,bg,func)
% DISTINGUISHABLE_COLORS: pick colors that are maximally perceptually distinct
%
% When plotting a set of lines, you may want to distinguish them by color.
% By default, Matlab chooses a small set of colors and cycles among them,
% and so if you have more than ... |
github | eglxiang/icassp15_emotion-master | LassoActiveSet.m | .m | icassp15_emotion-master/slr/lasso/LassoActiveSet.m | 7,567 | utf_8 | 6c65dea7e9626ca06d354197796f76c6 | function [w,wp,iteration] = LassoActiveSet(X, y, t,varargin)
% This function computes the Least Squares parameters
% whose 1-Norm is less than t
%
% Method used:
% Local Linearization and Active Set Method
% Proposed in [Osborne et al., 2000], [Osborne et al., 2000b]
%
% Mode:
% 0: Maintain QR Factorization of X'... |
github | eglxiang/icassp15_emotion-master | LassoProjection.m | .m | icassp15_emotion-master/slr/lasso/LassoProjection.m | 2,954 | utf_8 | ae367ab9088b56a86a9d91ab694dfd7e | function [w,wp,iteration] = LassoSubGradient(X,y,lambda,varargin)
% This function computes the Least Squares parameters
% with a penalty on the L1-norm of the parameters
%
% Method used:
% Two-Metric Projection
[maxIter,verbose,optTol,threshold] = process_options(varargin,'maxIter',10000,'verbose',2,'optTol',1e-5,'z... |
github | eglxiang/icassp15_emotion-master | LassoGrafting.m | .m | icassp15_emotion-master/slr/lasso/LassoGrafting.m | 2,951 | utf_8 | 29edb7545a498b6f5de5ab4d14cf8828 | function [w,wp,it] = LassoGrafting(X, y, lambda,varargin)
% This function computes the Least Squares parameters
% with a penalty on the L1-norm of the parameters
%
% Method used:
% The Grafting method of [Perkins et al., 2003]
% This method uses Matlab's fminunc
[maxIter,verbose,optTol,threshold] = process_options(... |
github | eglxiang/icassp15_emotion-master | LassoNonNegativeSquared.m | .m | icassp15_emotion-master/slr/lasso/LassoNonNegativeSquared.m | 1,689 | utf_8 | c9aef6ac0132e3385a7ba631c594e85a | function [w,fEvals] = LassoUnconstrainedApx(X, y, lambda,varargin)
% This function computes the Least Squares parameters
% with a penalty on the L1-norm of the parameters
%
% Method used:
% Uses a set of non-negative-squared variables to convert the problem
% into an equivalent unconstrained problem (solved with fm... |
github | eglxiang/icassp15_emotion-master | LassoIteratedRidge.m | .m | icassp15_emotion-master/slr/lasso/LassoIteratedRidge.m | 4,584 | utf_8 | 2a1f5505eb9981d45dfd6f12cd9895ff | function [w,wp,i] = LassoIteratedRidge(X, y, lambda,varargin)
% This function computes the Least Squares parameters
% with a penalty on the L1-norm of the parameters
%
% Method used:
% Iterated Ridge Regression using the approximation
% |w| =~ norm(w,2)/norm(w,1)
%
% Mode options:
% 0 - Deal with Numerical Instab... |
github | eglxiang/icassp15_emotion-master | LassoGaussSeidel.m | .m | icassp15_emotion-master/slr/lasso/LassoGaussSeidel.m | 3,855 | utf_8 | 70361adec41dd72dd76a080ec4f00f94 | function [w,wp,iteration] = LassoGaussSeidel(X, y, gamma,varargin)
% This function computes the Least Squares parameters
% with a penalty on the L1-norm of the parameters
%
% Method used:
% The Gauss-Seidel method of [Shevade and Keerthi, 2003]
%
% Mode options:
% 0 - Use bottom-up Method described in paper
% ... |
github | eglxiang/icassp15_emotion-master | LassoConstrained.m | .m | icassp15_emotion-master/slr/lasso/LassoConstrained.m | 4,769 | utf_8 | 886a004cc2edc3ef410533118fa829b8 | function [w,it] = LassoConstrained(X,y,t,varargin)
% This function computes the Least Squares parameters
% with a penalty on the L1-norm of the parameters
%
% Method used:
% Constrained Optimization with linear constraints,
%
% Mode (representation of constraint):
% 0 - min ([X -X]w-y).^2 s.t. w >= 0, sum(w) <= t
%... |
github | eglxiang/icassp15_emotion-master | LassoSubGradient.m | .m | icassp15_emotion-master/slr/lasso/LassoSubGradient.m | 2,798 | utf_8 | beb99efe04e6d38ac92c78a77ff87cd7 | function [w,wp,iteration] = LassoSubGradient(X,y,lambda,varargin)
% This function computes the Least Squares parameters
% with a penalty on the L1-norm of the parameters
%
% Method used:
% Sub-Gradient Descent on non-zero and zero but non-optimal variables
% taking Newton steps on the sub-gradients
%
% Note: This m... |
github | eglxiang/icassp15_emotion-master | LassoUnconstrainedApx.m | .m | icassp15_emotion-master/slr/lasso/LassoUnconstrainedApx.m | 4,514 | utf_8 | 8a6461f2035166b9a878fcd93f213ac2 | function [w,fEvals] = LassoUnconstrainedApx(X, y, lambda,varargin)
% This function computes the Least Squares parameters
% with a penalty on the L1-norm of the parameters
%
% Method used:
% Unconstrained optimization using an approximation to |w|.
% Implements 3 approximations to |w|, and 2 optimization strategies
... |
github | eglxiang/icassp15_emotion-master | process_options.m | .m | icassp15_emotion-master/slr/lasso/sub/process_options.m | 4,394 | utf_8 | 483b50d27e3bdb68fd2903a0cab9df44 | % PROCESS_OPTIONS - Processes options passed to a Matlab function.
% This function provides a simple means of
% parsing attribute-value options. Each option is
% named by a unique string and is given a default
% value.
%
% Usage: [var1, var2, ...... |
github | eglxiang/icassp15_emotion-master | ALM.m | .m | icassp15_emotion-master/src/Recon/ALM.m | 1,722 | utf_8 | bcdfc86e89beda3826c1c5cadf13139f | % modified from Allen Yang's code
% SYM March 13 2013
function [x, nIter] = ALM(A, b)
% Initialize parameters
[m,n] = size(A) ;
tol = 1e-6 ;
tol_apg = 1e-6 ;
rho = 1.25 ;
At = A';
G = At*A ;
opts.disp = 0;
tau = eigs(G,1,'lm',opts);
nIter = 0 ;
mu = 1/tau ;
mubar = 1e10*mu ;
tauInv = 1*mu ;
lambda = ones(m,1) ;
x ... |
github | eglxiang/icassp15_emotion-master | FISTA.m | .m | icassp15_emotion-master/src/Recon/FISTA.m | 5,508 | utf_8 | 2f501b36bfc6b28a0aae8a6a284a59ee | function [x_hat,nIter] = SolveFISTA(A,b, varargin)
% FISTA
% Modified from Alen Yang's code at Berkeley
% b - m x 1 vector of observations/data (required input)
% A - m x n measurement matrix (required input)
%
% tol - tolerance for stopping criterion.
% - DEFAULT 1e-7 if omitted or -1.
% maxIter - maxilambdam numb... |
github | phtra2/hctsa-master | TS_LinkOperationsWithMasters.m | .m | hctsa-master/Calculation/TS_LinkOperationsWithMasters.m | 2,721 | utf_8 | 903b2f7a6df2cbd8eb8ec0d5fecf94ad | function [Operations, MasterOperations] = TS_LinkOperationsWithMasters(Operations,MasterOperations)
% TS_LinkOperationsWithMasters Link Operations with MasterOperations using Label field
% ------------------------------------------------------------------------------
% Copyright (C) 2015, Ben D. Fulcher <ben.d.fulc... |
github | phtra2/hctsa-master | NL_DVV.m | .m | hctsa-master/Operations/NL_DVV.m | 7,121 | utf_8 | 9e3d0de2618b6744172ff878b99e3eb5 | function out = NL_DVV(x,m,numDVs,nd,Ntv,numSurr,randomSeed)
% NL_DVV Delay Vector Variance method for real and complex signals.
%
% Uses predictability of the signal in phase space to characterize the
% time series.
%
% This function uses the original code from the DVV toolbox to do the computation
% and produc... |
github | phtra2/hctsa-master | WL_coeffs.m | .m | hctsa-master/Operations/WL_coeffs.m | 3,642 | utf_8 | 542f15edaa6dafa61d62f34d0b0e7ee0 | function out = WL_coeffs(y, wname, level)
% WL_coeffs Wavelet decomposition of the time series.
%
% Performs a wavelet decomposition of the time series using a given wavelet at a
% given level and returns a set of statistics on the coefficients obtained.
%
% Uses Matlab's Wavelet Toolbox.
%
%---INPUTS:
% y, the inp... |
github | phtra2/hctsa-master | CP_l1pwc_sweep_lambda.m | .m | hctsa-master/Operations/CP_l1pwc_sweep_lambda.m | 4,334 | utf_8 | 2c134faf92850dbd090aa22f6e6052a3 | function out = CP_l1pwc_sweep_lambda(y,lambdar)
% CP_l1pwc_sweep_lambda Dependence of step detection on regularization parameter.
%
% Gives information about discrete steps in the signal across a range of
% regularization parameters lambda, using the function l1pwc from Max Little's
% step detection toolkit.
%
% cf... |
github | phtra2/hctsa-master | SB_MotifThree.m | .m | hctsa-master/Operations/SB_MotifThree.m | 7,638 | utf_8 | 7ff80106832fce52ee5929bb5c5640b3 | function out = SB_MotifThree(y,cgHow)
% SB_MotifThree Motifs in a coarse-graining of a time series to a 3-letter alphabet
%
% (As SB_MotifTwo but with a 3-letter alphabet)
%
%---INPUTS:
% y, time series to analyze
% cgHow, the coarse-graining method to use:
% (i) 'quantile': equiprobable alphabet by time-seri... |
github | phtra2/hctsa-master | NL_TISEAN_fnn.m | .m | hctsa-master/Operations/NL_TISEAN_fnn.m | 7,189 | utf_8 | 08eaa9c6ad2c0b7df3d71f299b818538 | function out = NL_TISEAN_fnn(y,tau,maxm,theilerWin,justBest,bestp)
% NL_TISEAN_fnn false nearest neighbors of a time series.
%
%---INPUTS:
% y, the input time series
% tau, the time delay
% maxm, the maximum embedding dimension
% theilerWin, the Theiler window
% justBest, if 1 just outputs a scalar estimate of embe... |
github | phtra2/hctsa-master | NW_VisibilityGraph.m | .m | hctsa-master/Operations/NW_VisibilityGraph.m | 9,952 | utf_8 | 56626516e6cd6807ec6a6f39fdb08546 | function out = NW_VisibilityGraph(y,meth,maxL)
% NW_VisibilityGraph Visibility graph analysis of a time series.
%
% Constructs a visibility graph of the time series and returns various
% statistics on the properties of the resulting network.
%
% cf.: "From time series to complex networks: The visibility graph"
% Lac... |
github | phtra2/hctsa-master | MF_ExpSmoothing.m | .m | hctsa-master/Operations/MF_ExpSmoothing.m | 9,555 | utf_8 | e29ac62c43617b9f1a035c111535fc07 | function out = MF_ExpSmoothing(x,ntrain,alpha)
% MF_ExpSmoothing Exponential smoothing time-series prediction model.
%
% Fits an exponential smoothing model to the time series using a training set to
% fit the optimal smoothing parameter, alpha, and then applies the result to the
% try to predict the rest of the time... |
github | phtra2/hctsa-master | SC_FluctAnal.m | .m | hctsa-master/Operations/SC_FluctAnal.m | 13,692 | utf_8 | 541b573fa14b143ab27831ea5154f962 | function out = SC_FluctAnal(x,q,wtf,tauStep,k,lag,logInc)
% SC_FluctAnal Implements fluctuation analysis by a variety of methods.
%
% Much of our implementation is based on the well-explained discussion of
% scaling methods in:
% "Power spectrum and detrended fluctuation analysis: Application to daily
% temperatures"... |
github | phtra2/hctsa-master | CO_AddNoise.m | .m | hctsa-master/Operations/CO_AddNoise.m | 7,966 | utf_8 | 271e0441e4e2ce65c6792c557408f19c | function out = CO_AddNoise(y,tau,amiMethod,extraParam,randomSeed)
% CO_AddNoise Changes in the automutual information with the addition of noise
%
% Adds Gaussian-distributed noise to the time series with increasing standard
% deviation, eta, across the range eta = 0, 0.1, ..., 2, and measures the
% mutual information... |
github | phtra2/hctsa-master | NL_TSTL_PoincareSection.m | .m | hctsa-master/Operations/NL_TSTL_PoincareSection.m | 8,296 | utf_8 | 402a9b29e3d46a6d807525f48425e3bf | function out = NL_TSTL_PoincareSection(y,ref,embedParams)
% NL_TSTL_PoincareSection Poincare sectino analysis of a time series.
%
% Obtains a Poincare section of the time-delay embedded time series, producing a
% set of vector points projected orthogonal to the tangential vector at the
% specified index using TSTOOL ... |
github | phtra2/hctsa-master | SD_SurrogateTest.m | .m | hctsa-master/Operations/SD_SurrogateTest.m | 12,083 | utf_8 | 8109ed148adc20f5f7355443e1e0b18e | function out = SD_SurrogateTest(x,surrMeth,numSurrs,extrap,theTestStat,randomSeed)
% SD_SurrogateTest Analyzes test statistics obtained from surrogate time series
%
% This function is based on information found in:
% "Surrogate data test for nonlinearity including nonmonotonic transforms"
% D. Kugiumtzis Phys. Rev. E... |
github | phtra2/hctsa-master | SY_Trend.m | .m | hctsa-master/Operations/SY_Trend.m | 2,575 | utf_8 | d46ef4c75208bcc28e29cb9182952a34 | function out = SY_Trend(y)
% SY_Trend Quantifies various measures of trend in a time series.
%
%---INPUT:
% y, the input time series.
%
%---OUTPUTS:
% Linearly detrends the time series using detrend, and returns the ratio of
% standard deviations before and after the linear detrending. If a strong linear
% trend is pr... |
github | phtra2/hctsa-master | WL_DetailCoeffs.m | .m | hctsa-master/Operations/WL_DetailCoeffs.m | 5,698 | utf_8 | 38804222795f61e518cc4f98491cf0d4 | function out = WL_DetailCoeffs(y, wname, maxlevel)
% WL_DetailCoeffs Detail coefficients of a wavelet decomposition.
%
% Compares the detail coefficients obtained at each level of the wavelet
% decomposition from 1 to the maximum possible level for the wavelet given the
% length of the input time series (computed usi... |
github | phtra2/hctsa-master | NL_embed_PCA.m | .m | hctsa-master/Operations/NL_embed_PCA.m | 4,933 | utf_8 | 5b412cf7cee832e86e1315879c63d1f0 | function out = NL_embed_PCA(y,tau,m)
% NL_embed_PCA Principal Components analysis of a time series in an embedding space.
%
% Reconstructs the time series as a time-delay embedding, and performs Principal
% Components Analysis on the result using princomp code from
% Matlab's Bioinformatics Toolbox.
%
% This technique... |
github | phtra2/hctsa-master | SP_Summaries.m | .m | hctsa-master/Operations/SP_Summaries.m | 16,882 | utf_8 | 9617af80206ba2404c6ef9d84f0f0c20 | function out = SP_Summaries(y,psdmeth,wmeth,nf,dologabs)
% SP_Summaries Statistics of the power spectrum of a time series
%
% The estimation can be done using a periodogram, using the periodogram code in
% Matlab's Signal Processing Toolbox, or a fast fourier transform, implemented
% using Matlab's fft code.
%
%---INP... |
github | phtra2/hctsa-master | SC_MMA.m | .m | hctsa-master/Operations/SC_MMA.m | 8,770 | utf_8 | 237b3c97b17bf16e6d41c12b2cadc07c | function out = SC_MMA(y,doOverlap,scaleRange,qRange)
% SC_MMA Physionet implementation of multiscale multifractal analysis
%
% Scale-dependent estimates of multifractal scaling in a time series.
% ------------------------------------------------------------------------------
% Modified by Ben Fulcher for use in hcts... |
github | phtra2/hctsa-master | DN_RemovePoints.m | .m | hctsa-master/Operations/DN_RemovePoints.m | 4,669 | utf_8 | 192ac49fd65e8e245d20363947f8b646 | function out = DN_RemovePoints(y,removeHow,p)
% DN_RemovePoints How time-series properties change as points are removed.
%
% A proportion, p, of points are removed from the time series according to some
% rule, and a set of statistics are computed before and after the change.
%
%---INPUTS:
% y, the input time series
... |
github | phtra2/hctsa-master | NL_TSTL_dimensions.m | .m | hctsa-master/Operations/NL_TSTL_dimensions.m | 14,999 | utf_8 | 2d42906471557cfe45d2b0cc23b43030 | function out = NL_TSTL_dimensions(y,nbins,embedParams)
% NL_TSTL_dimensions box counting, information, and correlation dimension of a time series.
%
% Computes the box counting, information, and correlation dimension of a
% time-delay embedded time series using the TSTOOL code 'dimensions'.
% This function contains ext... |
github | phtra2/hctsa-master | SB_MotifTwo.m | .m | hctsa-master/Operations/SB_MotifTwo.m | 6,309 | utf_8 | fe33ad4079fc7d7f67c9f4e54358d043 | function out = SB_MotifTwo(y,binarizeHow)
% SB_MotifTwo Local motifs in a binary symbolization of the time series
%
% Coarse-graining is performed by a given binarization method.
%
%---INPUTS:
% y, the input time series
% binarizeHow, the binary transformation method:
% (i) 'diff': incremental time-series incre... |
github | phtra2/hctsa-master | ST_MomentCorr.m | .m | hctsa-master/Operations/ST_MomentCorr.m | 5,415 | utf_8 | 1756564c1c7a2e9df7eacc247c6c404d | function out = ST_MomentCorr(x,windowLength,wOverlap,mom1,mom2,whatTransform)
% ST_MomentCorr Correlations between simple statistics in local windows of a time series.
%
% Idea to implement by Nick S. Jones.
%
%---INPUTS:
% x, the input time series
%
% windowLength, the sliding window length (can be a fraction to spe... |
github | phtra2/hctsa-master | NL_MS_fnn.m | .m | hctsa-master/Operations/NL_MS_fnn.m | 5,370 | utf_8 | c53c4e877995775ddeefa83943855703 | function out = NL_MS_fnn(y,de,tau,th,kth,justBest,bestp)
% NL_MS_fnn False nearest neighbors of a time series.
%
% Determines the number of false nearest neighbors for the embedded time series
% using Michael Small's false nearest neighbor code, fnn (renamed MS_fnn here).
%
% False nearest neighbors are judged usin... |
github | phtra2/hctsa-master | MF_GP_hyperparameters.m | .m | hctsa-master/Operations/MF_GP_hyperparameters.m | 10,430 | utf_8 | 8fe86cbbe5d3bc0e4b34087a16b422e1 | function out = MF_GP_hyperparameters(y,covFunc,squishorsquash,maxN,resampleHow,randomSeed)
% MF_GP_hyperparameters Gaussian Process time-series model parameters and goodness of fit
%
% Uses GP fitting code from the gpml toolbox, which is available here:
% http://gaussianprocess.org/gpml/code.
%
% The code can accomo... |
github | phtra2/hctsa-master | NL_crptool_fnn.m | .m | hctsa-master/Operations/NL_crptool_fnn.m | 4,953 | utf_8 | e1a79d9e00ccf8d7487ca4f035efbf15 | function out = NL_crptool_fnn(y,maxm,r,taum,th,randomSeed)
% NL_crptool_fnn Analyzes the false-nearest neighbours statistic.
%
%---INPUTS:
% y, the input time series
% maxm, the maximum embedding dimension to consider
% r, the threshold; neighbourhood criterion
% taum, the method of determining the time delay, 'corr... |
github | phtra2/hctsa-master | CO_StickAngles.m | .m | hctsa-master/Operations/CO_StickAngles.m | 10,799 | utf_8 | 0fd74672caa734070ebea06771a37e8b | function out = CO_StickAngles(y)
% CO_StickAngles Analysis of line-of-sight angles between time-series data points.
%
% Line-of-sight angles between time-series points treat each time-series value
% as a stick protruding from an opaque baseline level.
% Statistics are returned on the raw time series, where sticks pr... |
github | phtra2/hctsa-master | TS_subset.m | .m | hctsa-master/PeripheryFunctions/TS_subset.m | 6,074 | utf_8 | 9133009ebbef7e83e85f9a07bdd2e45f | function [TS_DataMat,TimeSeries,Operations] = TS_subset(whatData,ts_ids_keep,op_ids_keep,doSave,outputFileName)
% TS_subset Save a given subset of an hctsa dataset, based on time series and operation IDs
%
%---INPUTS:
% whatData, the source of the hctsa dataset (default, 'HCTSA_N.mat', cf. TS_LoadData)
% ts_ids_keep, ... |
github | phtra2/hctsa-master | BF_NormalizeMatrix.m | .m | hctsa-master/PeripheryFunctions/BF_NormalizeMatrix.m | 7,750 | utf_8 | a765f70a44bc426ea584d85aa51c851d | function dataMatrixNorm = BF_NormalizeMatrix(dataMatrix,normMethod,itrain)
% BF_NormalizeMatrix Normalizes all columns of an input matrix.
%
%---INPUTS:
% dataMatrix, the input data matrix
% normMethod, the normalization method to use (see body of the code for options)
% itrain, learn the normalization parameters ju... |
github | phtra2/hctsa-master | BF_pdist.m | .m | hctsa-master/PeripheryFunctions/BF_pdist.m | 7,653 | utf_8 | 22939b37587fd5068d115809f9a3eacd | function R = BF_pdist(dataMatrix,distMetric,toVector,opts,beSilent,minPropGood)
% BF_pdist Pairwise distances between rows of a data matrix.
%
% Same as pdist but then goes through and fills in NaNs with indiviually
% calculated values using an overlapping range of good values.
% -------------------------------------... |
github | phtra2/hctsa-master | TS_local_clear_remove.m | .m | hctsa-master/PeripheryFunctions/TS_local_clear_remove.m | 6,542 | utf_8 | 8218628ff36c294ef22f941fbd08068f | function TS_local_clear_remove(tsOrOps,idRange,doRemove,whatData)
% TS_local_clear_remove Clear or remove data from an hctsa dataset
%
% 'Clear' means clearing any calculations performed about a given time series
% or operation, but keeping it in the dataset.
% 'Remove' means removing the time series or operation f... |
github | phtra2/hctsa-master | BF_AnnotatePoints.m | .m | hctsa-master/PeripheryFunctions/BF_AnnotatePoints.m | 8,749 | utf_8 | 095f78aa318b91e42db07023abe26347 | function BF_AnnotatePoints(xy,TimeSeries,annotateParams)
% BF_AnnotatePoints Annotates time series/metadata to a plot
%
%---INPUTS:
% xy, a vector (or cell) of x-y co-ordinates of points on the plot
% TimeSeries, a structure array of time series making up the plot
% annotateParams, structure of custom plotting para... |
github | phtra2/hctsa-master | BF_MutualInformation.m | .m | hctsa-master/PeripheryFunctions/BF_MutualInformation.m | 4,573 | utf_8 | 57928a4e3581df155882e6638ed89482 | function mi = BF_MutualInformation(v1,v2,r1,r2,numBins)
% BF_MutualInformation Mutual information between two data vectors using bin counting.
%
% Mutual information computed using a histogram-based, bin-counting method.
%
%---INPUTS:
% v1, the first input vector
% v2, the second input vector
% r1, the bin-partitionin... |
github | phtra2/hctsa-master | ML_l1pwcar1.m | .m | hctsa-master/Toolboxes/Max_Little/steps_bumps_toolkit/ML_l1pwcar1.m | 2,540 | utf_8 | 9d54dd6018fa164663f21d16a1757904 | % Performs discrete correlated total variation denoising (CTVD) using a
% primal-dual interior-point solver. It minimizes the following discrete
% functional:
%
% E=(1/2)||y_0-ay_1-x||_2^2+lambda*||Dx||_1,
%
% over the variable x, given the input signal y, according to each
% value of the regularization parameter lamb... |
github | phtra2/hctsa-master | ML_ckfilter.m | .m | hctsa-master/Toolboxes/Max_Little/steps_bumps_toolkit/ML_ckfilter.m | 2,305 | utf_8 | ce08bfcf3e205227e46cdc08b1bd0124 | % Implements the Chung-Kennedy sliding window nonlinear step filter. This
% filter is similar to a centred moving average filter of length K, but the
% centre sample in the window is replaced by a weighted sum of forward and
% backward moving average filters. The weights are inversely proportional
% to the one-step-ahe... |
github | phtra2/hctsa-master | ML_l1pwclmax.m | .m | hctsa-master/Toolboxes/Max_Little/steps_bumps_toolkit/ML_l1pwclmax.m | 1,072 | utf_8 | e392aeba5a721f8fdc65204f75643b3c | % Calculate the value of lambda so that if lambda >= lambdamax, the TVD
% functional solved by l1pwc is minimized by the trivial constant
% solution x = mean(y). This can then be used to determine a useful range
% of values of lambda, for example.
%
% Usage:
% lambdamax = l1pwclmax(y)
%
% Input arguments:
% - y ... |
github | phtra2/hctsa-master | ML_l1pwc.m | .m | hctsa-master/Toolboxes/Max_Little/steps_bumps_toolkit/ML_l1pwc.m | 6,386 | utf_8 | 35a2f72b14c9e9dc8e8f6be4307fd245 | % Performs discrete total variation denoising (TVD) using a primal-dual
% interior-point solver. It minimizes the following discrete functional:
%
% E=(1/2)||y-x||_2^2+lambda*||Dx||_1,
%
% over the variable x, given the input signal y, according to each
% value of the regularization parameter lambda > 0. D is the firs... |
github | phtra2/hctsa-master | ML_kvsteps.m | .m | hctsa-master/Toolboxes/Max_Little/steps_bumps_toolkit/ML_kvsteps.m | 1,280 | utf_8 | affbbbbf7715764bb93eca5d0d45c377 | % Implements the Kalafut-Visscher step detection method, using the MEX
% wrapper of the C version.
%
% Usage:
% [y, steps] = ML_kvsteps(x)
%
% Inputs
% x - Input signal
%
% Outputs
% y - Estimated piecewise constant approximation to the input signal
% steps - Vector of estimated step-change points i... |
github | phtra2/hctsa-master | ML_fastdfa.m | .m | hctsa-master/Toolboxes/Max_Little/fastdfa/ML_fastdfa.m | 1,237 | utf_8 | f34bf3cc4bf90912bc4a07fc585024e6 | % Performs fast detrended fluctuation analysis on a nonstationary input signal to
% obtain an estimate for the scaling exponent.
%
% Useage:
% [alpha, intervals, flucts] = fastdfa(x)
% [alpha, intervals, flucts] = fastdfa(x, intervals)
% Inputs
% x - input signal: must be a column vector
% Optional inputs
%... |
github | phtra2/hctsa-master | likBeta.m | .m | hctsa-master/Toolboxes/gpml/lik/likBeta.m | 4,830 | utf_8 | 8e503690924874d07a77dc48bc238db1 | function [varargout] = likBeta(link, hyp, y, mu, s2, inf, i)
% likBeta - Beta likelihood function for interval data y from [0,1].
% The expression for the likelihood is
% likBeta(f) = 1/Z * y^(mu*phi-1) * (1-y)^((1-mu)*phi-1) with
% mean=mu and variance=mu*(1-mu)/(1+phi) where mu = g(f) is the Beta intensity,
% f ... |
github | phtra2/hctsa-master | likT.m | .m | hctsa-master/Toolboxes/gpml/lik/likT.m | 4,776 | utf_8 | 6463e0fed8f6484854dd3dd212db5202 | function [varargout] = likT(hyp, y, mu, s2, inf, i)
% likT - Student's t likelihood function for regression.
% The expression for the likelihood is
% likT(t) = Z * ( 1 + (t-y)^2/(nu*sn^2) ).^(-(nu+1)/2),
% where Z = gamma((nu+1)/2) / (gamma(nu/2)*sqrt(nu*pi)*sn)
% and y is the mean (for nu>1) and nu*sn^2/(nu-2) is ... |
github | phtra2/hctsa-master | likLaplace.m | .m | hctsa-master/Toolboxes/gpml/lik/likLaplace.m | 6,922 | iso_8859_13 | 9673b9c57508bdbfd0dc917f10944f80 | function [varargout] = likLaplace(hyp, y, mu, s2, inf, i)
% likLaplace - Laplacian likelihood function for regression.
% The expression for the likelihood is
% likLaplace(t) = exp(-|t-y|/b)/(2*b) with b = sn/sqrt(2),
% where y is the mean and sn^2 is the variance.
%
% The hyperparameters are:
%
% hyp = [ log(sn) ... |
github | phtra2/hctsa-master | likGaussWarp.m | .m | hctsa-master/Toolboxes/gpml/lik/likGaussWarp.m | 9,118 | utf_8 | baca6bc6eb9f081dff2f85d7a4eb8318 | function [varargout] = likGaussWarp(warp, hyp, y, mu, varargin)
% likGaussWarp - Warped Gaussian likelihood for regression.
% The expression for the likelihood is
% likGaussWarp( y | t ) = likGauss( g(y) | t ) * g'(y),
% where likGauss is the Gaussian likelihood and g is the warping function.
%
% The hyperparamete... |
github | phtra2/hctsa-master | likWeibull.m | .m | hctsa-master/Toolboxes/gpml/lik/likWeibull.m | 4,548 | utf_8 | 5134b34b56b016f15d716469fb93c583 | function [varargout] = likWeibull(link, hyp, y, mu, s2, inf, i)
% likWeibull - Weibull likelihood function for strictly positive data y. The
% expression for the likelihood is
% likWeibull(f) = g1*ka/mu * (g1*y/mu)^(ka-1) * exp(-(g1*y/mu)^ka) with
% gj = gamma(1+j/ka), mean=mu and variance=mu^2*(g2/g1^2-1) where mu... |
github | phtra2/hctsa-master | likGamma.m | .m | hctsa-master/Toolboxes/gpml/lik/likGamma.m | 4,573 | utf_8 | 30195b20deb79baed3429087b58977a8 | function [varargout] = likGamma(link, hyp, y, mu, s2, inf, i)
% likGamma - Gamma likelihood function for strictly positive data y. The
% expression for the likelihood is
% likGamma(f) = al^al*y^(al-1)/gamma(al) * exp(-y*al/mu) / mu^al with
% mean=mu and variance=mu^2/al where mu = g(f) is the Gamma intensity, f is... |
github | phtra2/hctsa-master | likInvGauss.m | .m | hctsa-master/Toolboxes/gpml/lik/likInvGauss.m | 4,679 | utf_8 | 1bffc204bfdee3ee427008906bce81ad | function [varargout] = likInvGauss(link, hyp, y, mu, s2, inf, i)
% likInvGauss - Inverse Gaussian likelihood function for strictly positive data
% y. The expression for the likelihood is
% likInvGauss(f) = sqrt(lam/(2*pi*y^3))*exp(-lam*(mu-y)^2/(2*mu^2*y)) with
% mean=mu and variance=mu^3/lam where mu = g(f) is th... |
github | phtra2/hctsa-master | likPoisson.m | .m | hctsa-master/Toolboxes/gpml/lik/likPoisson.m | 4,178 | utf_8 | 9bdb4f7a4905445839d4697149efc827 | function [varargout] = likPoisson(link, hyp, y, mu, s2, inf, i)
% likPoisson - Poisson likelihood function for count data y. The expression for
% the likelihood is
% likPoisson(f) = mu^y * exp(-mu) / y! with mean=variance=mu
% where mu = g(f) is the Poisson intensity, f is a
% Gaussian process, y is the non-negativ... |
github | phtra2/hctsa-master | likLogistic.m | .m | hctsa-master/Toolboxes/gpml/lik/likLogistic.m | 6,137 | utf_8 | 0227c40f8798f8f47d1f32e9dfd6e946 | function [varargout] = likLogistic(hyp, y, mu, s2, inf, i)
% likLogistic - logistic function for binary classification or logit regression.
% The expression for the likelihood is
% likLogistic(t) = 1./(1+exp(-t)).
%
% Several modes are provided, for computing likelihoods, derivatives and moments
% respectively, see... |
github | phtra2/hctsa-master | likSech2.m | .m | hctsa-master/Toolboxes/gpml/lik/likSech2.m | 8,514 | utf_8 | 25a639e43b4bcdc60d8fd113ded18611 | function [varargout] = likSech2(hyp, y, mu, s2, inf, i)
% likSech2 - sech-square likelihood function for regression. Often, the sech-
% square distribution is also referred to as the logistic distribution not to be
% confused with the logistic function for classification. The expression for the
% likelihood is
% li... |
github | phtra2/hctsa-master | likGumbel.m | .m | hctsa-master/Toolboxes/gpml/lik/likGumbel.m | 3,976 | utf_8 | e181712e58f8360c4d43c5c354d8431a | function [varargout] = likGumbel(sign, hyp, y, mu, s2, inf, i)
% likGumbel - Gumbel likelihood function for extremal value regression.
% The expression for the likelihood is
% likGumbel(t) = exp(-z-exp(-z))/be, z = ga+s*(y-t)/be, be = sn*sqrt(6)/pi
% where s={+1,-1} is a sign switching between left and right skewed... |
github | phtra2/hctsa-master | priorSmoothBox1.m | .m | hctsa-master/Toolboxes/gpml/prior/priorSmoothBox1.m | 1,617 | utf_8 | df60218e999e45adf5f4204501f3c42f | function [lp,dlp] = priorSmoothBox1(a,b,eta,x)
% Univariate smoothed box prior distribution with linear decay in the log domain
% and infinite support over the whole real axis.
% Compute log-likelihood and its derivative or draw a random sample.
% The prior distribution is parameterized as:
%
% p(x) = sigmoid(... |
github | phtra2/hctsa-master | logphi.m | .m | hctsa-master/Toolboxes/gpml/util/logphi.m | 2,261 | utf_8 | 69fbcfc9d9913da15644d5f0a0368d5f | % Safe computation of logphi(z) = log(normcdf(z)) and its derivatives
% dlogphi(z) = normpdf(x)/normcdf(x).
% The function is based on index 5725 in Hart et al. and gsl_sf_log_erfc_e.
%
% Copyright (c) by Carl Edward Rasmussen and Hannes Nickisch, 2013-11-13.
function [lp,dlp,d2lp,d3lp] = logphi(z)
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.