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
NEU-Gou/kernel-metric-learning-reid-master
icg_roc.m
.m
kernel-metric-learning-reid-master/Assistant Code/KISSME/toolbox/helper/icg_roc.m
1,425
utf_8
11d04e9c4c3db15aa1c3b9b771eff30e
function [tpr,fpr,thresh] = icg_roc(tp,confs) % ICG_ROC computes ROC measures (tpr,fpr) % % Input: % tp - [m x n] matrix of zero-one labels. one row per class. % confs - [m x n] matrix of classifier scores. one row per class. % % Output: % tpr - true positive rate in interval [0,1], [m x n+1] matrix % ...
github
NEU-Gou/kernel-metric-learning-reid-master
make.m
.m
kernel-metric-learning-reid-master/Assistant Code/KISSME/toolbox/lib/liblinear-2.1/matlab/make.m
1,198
utf_8
72532ef957c850421c786167742d0912
% This make.m is for MATLAB and OCTAVE under Windows, Mac, and Unix function make() try % This part is for OCTAVE if(exist('OCTAVE_VERSION', 'builtin')) mex libsvmread.c mex libsvmwrite.c mex -I.. train.c linear_model_matlab.c ../linear.cpp ../tron.cpp ../blas/daxpy.c ../blas/ddot.c ../blas/dnrm2.c ../blas/dsca...
github
NEU-Gou/kernel-metric-learning-reid-master
lbp.m
.m
kernel-metric-learning-reid-master/Assistant Code/LBP/lbp.m
6,280
utf_8
57e3d9b82aa8caaa6045d2ef0e048479
%LBP returns the local binary pattern image or LBP histogram of an image. % J = LBP(I,R,N,MAPPING,MODE) returns either a local binary pattern % coded image or the local binary pattern histogram of an intensity % image I. The LBP codes are computed using N sampling points on a % circle of radius R and using mapping...
github
NEU-Gou/kernel-metric-learning-reid-master
getmapping.m
.m
kernel-metric-learning-reid-master/Assistant Code/LBP/getmapping.m
5,222
utf_8
54332fd445f20849f7554078eddeccef
%GETMAPPING returns a structure containing a mapping table for LBP codes. % MAPPING = GETMAPPING(SAMPLES,MAPPINGTYPE) returns a % structure containing a mapping table for % LBP codes in a neighbourhood of SAMPLES sampling % points. Possible values for MAPPINGTYPE are % 'u2' for uniform LBP % 'ri' fo...
github
viniciusrpb/regiongrowing-master
filtroEDP.m
.m
regiongrowing-master/filtroEDP.m
4,766
UNKNOWN
b3e43d7c519ebff57c8935800b9e6864
%Filtro de Suavizacao e Eliminacao de Ruidos baseado em Equacoes % Diferenciais Parciais, proposto pela prof. Dra. Celia Aparecida % Zorzo Barcelos. %Implementado por Vinicius Ruela Pereira Borges, 08/01/2009 function U = filtroEDP(img,k,dt,T,LAMBDA) % k = fator de suavizacao (quanto menor, maior eh a suavizacao) ...
github
viniciusrpb/regiongrowing-master
cssDescriptor.m
.m
regiongrowing-master/cssDescriptor.m
5,496
utf_8
f46b06ddc6581cb522a38cfec1189043
function [descriptor,K] = cssDescriptor(x,y,sigma,numberOfPoints,thresh,dimensions) nroPontos = length(x); fator2 = 0.15; sigmaAtual(1) = 0.05; niveis = 1; while(sigmaAtual(niveis) < sigma) sigmaAtual(niveis+1) = double(sigmaAtual(niveis) + fator2); niveis = niveis + 1; end...
github
viniciusrpb/regiongrowing-master
regionGrowingProcess.m
.m
regiongrowing-master/regionGrowingProcess.m
3,272
utf_8
6e7fb8a68cfff48e543c1cbf618cef61
% Region growing algorithm based on Gonzalez, 2008: Digital Image Processing % 3. ed. % IMPORTANT: This implementation of region growing follows an iterative % version proposed by Gonzalez et. al, 2008. function bin = regionGrowingProcess(img,x,y,mu1,sigma1,mu2,sigma2,binTemp) % Take the weight and width [hei...
github
viniciusrpb/regiongrowing-master
regionGrowingBasedSegmentation.m
.m
regiongrowing-master/regionGrowingBasedSegmentation.m
8,994
utf_8
d6804b65e2d0e61efbb12c7f492aac1c
% INSTITUTO DE CIENCIAS MATEMATICAS E DE COMPUTACAO % UNIVERSITY OF SAO PAULO, SAO CARLOS % MATLAB CODE % IMAGE SEGMEN % LAST UPDATE: May, 12th 2016 function regionGrowingBasedSegmentation(path,pathSaida) pngPath = sprintf('%s*.png',path); list = dir(pngPath); listSize = length(list); %Flag write...
github
viniciusrpb/regiongrowing-master
curvature.m
.m
regiongrowing-master/curvature.m
1,504
utf_8
cc0ff9432463daedee472728b50699b3
function [kappa,smoothKappa,normKappa] = curvature(x,y) T = 25; dt = 0.1; tam = length(x); smoothKappa(1:tam+2,1:T) = 0; %Calcula angulo entre segmentos kappa(1:tam+2) = 0; xp(2:tam+1) = x; yp(2:tam+1) = y; xp(1) = x(end); yp(1) = y(end); xp(tam+2) = x(1); yp(tam+2)...
github
viniciusrpb/regiongrowing-master
cssSmoothing.m
.m
regiongrowing-master/cssSmoothing.m
557
utf_8
619e8ef9c63093c7f0b7bd3f20a56b4c
% Implementation of the Curvature Scale Space Descriptor % % INPUT: binary image img % OUTPUT: the threshold cut at the CSS Map function cutHere = cssSmoothing(img) % Create background image area = sum(sum(img)); boundaries = Extracao_Contorno(img); x = boundaries(1,:); y = boundaries(2,:); ...
github
TWANG006/G-LS3U-master
fp_unwrapping.m
.m
G-LS3U-master/MatLab/fp_unwrapping.m
1,456
utf_8
a8ab7fc56336415392af2fd6c3f0335d
function p=fp_unwrapping(p,start_x,start_y) %FUNCTION % p=fp_unwrapping(p,start_x,start_y) % %PURPOSE % Line scanning phase unwrapping % %INPUTS % p: wrapped phasethe phase distribution % start_x: starting point for phase unwrapping, x % start_y: starting point for phase unwrapping, y % %OUTPUT...
github
TWANG006/G-LS3U-master
fp_wft2f.m
.m
G-LS3U-master/MatLab/fp_wft2f.m
7,113
utf_8
cf0a15743672a522b91924255c402081
function z=fp_wft2f(type,f,sigmax,wxl,wxi,wxh,sigmay,wyl,wyi,wyh,thr) %FUNCTION % z=fp_wft2f(type,f,sigmax,wxl,wxi,wxh,sigmay,wyl,wyi,wyh,thr) % %PURPOSE % 2-D WFT [Fourier version]: Fourier transform is used to compute % convolutions. % %INPUT % type: 'wff' or 'wfr' % f: 2D input signal % sigmax: s...
github
TWANG006/G-LS3U-master
fexpand.m
.m
G-LS3U-master/MatLab/fexpand.m
368
utf_8
f4a53900abc7172106df6bc3bca744e7
%INFO % Last update: 28/07/2011, 05/10/2012, 12/12/2012 % Contact: mkmqian@ntu.edu.sg (Dr Qian Kemao) % Copyright reserved. %expand f to [m n] %this function can be realized by padarray, but is slower function f=fexpand(f,mm,nn) %size f [m n]=size(f); %store f f0=f; %generate a larger matrix with size [mm nn] f=...
github
TWANG006/G-LS3U-master
fp_aia.m
.m
G-LS3U-master/MatLab/fp_aia.m
2,791
utf_8
8aa8e74c2df306eca3e82f276c92712e
function [phi delta iter err] = fp_aia(f, delta, max_iter, max_err) %FUNCTION % [phi delta iter err]=aia(f, delta, max_iter, max_err) % %PURPOSE % Wang and Han's AIA for phase-shifting with arbitrary phase-shifts. % %INPUT % f: 3-D matrix of size m*n*l (image size m*n, frame number l) % delta: ini...
github
facundoq/courses-master
submit.m
.m
courses-master/mlclassng/mlclass-ex4-005/submit.m
17,129
utf_8
bf5c5c1ceefb6cad621d632888b78dd1
function submit(partId, webSubmit) %SUBMIT Submit your code and output to the ml-class servers % SUBMIT() will connect to the ml-class server and submit your solution fprintf('==\n== [ml-class] Submitting Solutions | Programming Exercise %s\n==\n', ... homework_id()); if ~exist('partId', 'var') || isem...
github
facundoq/courses-master
submitWeb.m
.m
courses-master/mlclassng/mlclass-ex4-005/submitWeb.m
827
utf_8
bfb2fa08cac9d8d797e3071d3fdd7ca1
% submitWeb Creates files from your code and output for web submission. % % If the submit function does not work for you, use the web-submission mechanism. % Call this function to produce a file for the part you wish to submit. Then, % submit the file to the class servers using the "Web Submission" button on ...
github
facundoq/courses-master
submit.m
.m
courses-master/mlclassng/mlclass-ex5-005/submit.m
17,211
utf_8
13a9995decf628307987cfb3364dd6a1
function submit(partId, webSubmit) %SUBMIT Submit your code and output to the ml-class servers % SUBMIT() will connect to the ml-class server and submit your solution fprintf('==\n== [ml-class] Submitting Solutions | Programming Exercise %s\n==\n', ... homework_id()); if ~exist('partId', 'var') || isem...
github
facundoq/courses-master
submitWeb.m
.m
courses-master/mlclassng/mlclass-ex5-005/submitWeb.m
827
utf_8
bfb2fa08cac9d8d797e3071d3fdd7ca1
% submitWeb Creates files from your code and output for web submission. % % If the submit function does not work for you, use the web-submission mechanism. % Call this function to produce a file for the part you wish to submit. Then, % submit the file to the class servers using the "Web Submission" button on ...
github
facundoq/courses-master
submit.m
.m
courses-master/mlclassng/mlclass-ex8-005/submit.m
17,515
utf_8
f1320acecad5e41355ce270b4195c3db
function submit(partId, webSubmit) %SUBMIT Submit your code and output to the ml-class servers % SUBMIT() will connect to the ml-class server and submit your solution fprintf('==\n== [ml-class] Submitting Solutions | Programming Exercise %s\n==\n', ... homework_id()); if ~exist('partId', 'var') || isem...
github
facundoq/courses-master
submitWeb.m
.m
courses-master/mlclassng/mlclass-ex8-005/submitWeb.m
827
utf_8
bfb2fa08cac9d8d797e3071d3fdd7ca1
% submitWeb Creates files from your code and output for web submission. % % If the submit function does not work for you, use the web-submission mechanism. % Call this function to produce a file for the part you wish to submit. Then, % submit the file to the class servers using the "Web Submission" button on ...
github
facundoq/courses-master
submit.m
.m
courses-master/mlclassng/mlclass-ex1-005/submit.m
17,317
utf_8
c92fa7713d3c5747ef93f067dddcc54d
function submit(partId, webSubmit) %SUBMIT Submit your code and output to the ml-class servers % SUBMIT() will connect to the ml-class server and submit your solution fprintf('==\n== [ml-class] Submitting Solutions | Programming Exercise %s\n==\n', ... homework_id()); if ~exist('partId', 'var') || isem...
github
facundoq/courses-master
submitWeb.m
.m
courses-master/mlclassng/mlclass-ex1-005/submitWeb.m
827
utf_8
bfb2fa08cac9d8d797e3071d3fdd7ca1
% submitWeb Creates files from your code and output for web submission. % % If the submit function does not work for you, use the web-submission mechanism. % Call this function to produce a file for the part you wish to submit. Then, % submit the file to the class servers using the "Web Submission" button on ...
github
facundoq/courses-master
submit.m
.m
courses-master/mlclassng/mlclass-ex2-005/mlclass-ex2/submit.m
17,086
utf_8
6f26069c12237f7a430ed09fc0c1b344
function submit(partId, webSubmit) %SUBMIT Submit your code and output to the ml-class servers % SUBMIT() will connect to the ml-class server and submit your solution fprintf('==\n== [ml-class] Submitting Solutions | Programming Exercise %s\n==\n', ... homework_id()); if ~exist('partId', 'var') || isem...
github
facundoq/courses-master
submitWeb.m
.m
courses-master/mlclassng/mlclass-ex2-005/mlclass-ex2/submitWeb.m
827
utf_8
bfb2fa08cac9d8d797e3071d3fdd7ca1
% submitWeb Creates files from your code and output for web submission. % % If the submit function does not work for you, use the web-submission mechanism. % Call this function to produce a file for the part you wish to submit. Then, % submit the file to the class servers using the "Web Submission" button on ...
github
facundoq/courses-master
submit.m
.m
courses-master/mlclassng/mlclass-ex6-005/submit.m
16,836
utf_8
bbf9b999a1dae2f9a208e9edbfb6981a
function submit(partId, webSubmit) %SUBMIT Submit your code and output to the ml-class servers % SUBMIT() will connect to the ml-class server and submit your solution fprintf('==\n== [ml-class] Submitting Solutions | Programming Exercise %s\n==\n', ... homework_id()); if ~exist('partId', 'var') || isem...
github
facundoq/courses-master
porterStemmer.m
.m
courses-master/mlclassng/mlclass-ex6-005/porterStemmer.m
9,902
utf_8
7ed5acd925808fde342fc72bd62ebc4d
function stem = porterStemmer(inString) % Applies the Porter Stemming algorithm as presented in the following % paper: % Porter, 1980, An algorithm for suffix stripping, Program, Vol. 14, % no. 3, pp 130-137 % Original code modeled after the C version provided at: % http://www.tartarus.org/~martin/PorterStemmer/c.tx...
github
facundoq/courses-master
submitWeb.m
.m
courses-master/mlclassng/mlclass-ex6-005/submitWeb.m
827
utf_8
bfb2fa08cac9d8d797e3071d3fdd7ca1
% submitWeb Creates files from your code and output for web submission. % % If the submit function does not work for you, use the web-submission mechanism. % Call this function to produce a file for the part you wish to submit. Then, % submit the file to the class servers using the "Web Submission" button on ...
github
facundoq/courses-master
submit.m
.m
courses-master/mlclassng/mlclass-ex3-005/mlclass-ex3/submit.m
17,041
utf_8
3222eb8403790dbbf400354e1e71d0de
function submit(partId, webSubmit) %SUBMIT Submit your code and output to the ml-class servers % SUBMIT() will connect to the ml-class server and submit your solution fprintf('==\n== [ml-class] Submitting Solutions | Programming Exercise %s\n==\n', ... homework_id()); if ~exist('partId', 'var') || isem...
github
facundoq/courses-master
submitWeb.m
.m
courses-master/mlclassng/mlclass-ex3-005/mlclass-ex3/submitWeb.m
827
utf_8
bfb2fa08cac9d8d797e3071d3fdd7ca1
% submitWeb Creates files from your code and output for web submission. % % If the submit function does not work for you, use the web-submission mechanism. % Call this function to produce a file for the part you wish to submit. Then, % submit the file to the class servers using the "Web Submission" button on ...
github
facundoq/courses-master
submit.m
.m
courses-master/mlclassng/mlclass-ex7-005/submit.m
16,958
utf_8
44af178969606b83916b59695dd1a568
function submit(partId, webSubmit) %SUBMIT Submit your code and output to the ml-class servers % SUBMIT() will connect to the ml-class server and submit your solution fprintf('==\n== [ml-class] Submitting Solutions | Programming Exercise %s\n==\n', ... homework_id()); if ~exist('partId', 'var') || isem...
github
facundoq/courses-master
submitWeb.m
.m
courses-master/mlclassng/mlclass-ex7-005/submitWeb.m
827
utf_8
bfb2fa08cac9d8d797e3071d3fdd7ca1
% submitWeb Creates files from your code and output for web submission. % % If the submit function does not work for you, use the web-submission mechanism. % Call this function to produce a file for the part you wish to submit. Then, % submit the file to the class servers using the "Web Submission" button on ...
github
facundoq/courses-master
ComputeExactMarginalsBP.m
.m
courses-master/pgm/p4/ComputeExactMarginalsBP.m
2,013
utf_8
356d51f3bd67254e5a4ff1a82fc31d4f
%COMPUTEEXACTMARGINALSBP Runs exact inference and returns the marginals %over all the variables (if isMax == 0) or the max-marginals (if isMax == 1). % % M = COMPUTEEXACTMARGINALSBP(F, E, isMax) takes a list of factors F, % evidence E, and a flag isMax, runs exact inference and returns the % final marginals for ...
github
facundoq/courses-master
FactorSum.m
.m
courses-master/pgm/p4/FactorSum.m
2,340
utf_8
d2e83cf81bd5e92fe61deaa53e6f4643
% FactorSum Computes the sum of two factors. % C = FactorProduct(A,B) computes the sum between two factors, A and B, % where each factor is defined over a set of variables with given dimension. % The factor data structure has the following fields: % .var Vector of variables in the factor, e.g. [1 2 3] % ...
github
facundoq/courses-master
EliminateVar.m
.m
courses-master/pgm/p4/EliminateVar.m
1,346
utf_8
d107c19a06837def63e9460f88420c58
% Function used in production of clique trees % % Copyright (C) Daphne Koller, Stanford University, 2012 function [newF C E] = EliminateVar(F, C, E, Z) useFactors = []; scope = []; for i=1:length(F) if any(F(i).var == Z) useFactors = [useFactors i]; scope = union(scope, F(i).var); end end % ...
github
facundoq/courses-master
IndexToAssignment.m
.m
courses-master/pgm/p4/IndexToAssignment.m
641
utf_8
4500d5b7500f21f11c89b80922de3c42
% IndexToAssignment Convert index to variable assignment. % % A = IndexToAssignment(I, D) converts an index, I, into the .val vector % into an assignment over variables with cardinality D. If I is a vector, % then the function produces a matrix of assignments, one assignment % per row. % % See also Assignme...
github
facundoq/courses-master
CliqueTreeCalibrate.m
.m
courses-master/pgm/p4/CliqueTreeCalibrate.m
3,610
utf_8
2c54be8808db19a03421528e2b0e6412
%CLIQUETREECALIBRATE Performs sum-product or max-product algorithm for %clique tree calibration. % P = CLIQUETREECALIBRATE(P, isMax) calibrates a given clique tree, P % according to the value of isMax flag. If isMax is 1, it uses max-sum % message passing, otherwise uses sum-product. This function % returns...
github
facundoq/courses-master
FactorMarginalization.m
.m
courses-master/pgm/p4/FactorMarginalization.m
1,691
utf_8
1a6c278432109b94853fa27ff43fe933
% FactorMarginalization Sums given variables out of a factor. % B = FactorMarginalization(A,V) computes the factor with the variables % in V summed out. The factor data structure has the following fields: % .var Vector of variables in the factor, e.g. [1 2 3] % .card Vector of cardinalities corresp...
github
facundoq/courses-master
DecodedMarginalsToChars.m
.m
courses-master/pgm/p4/DecodedMarginalsToChars.m
218
utf_8
02261a4f2b86ec598a6e1c7e98525853
% Copyright (C) Daphne Koller, Stanford University, 2012 function DecodedMarginalsToChars(decodedMarginals) chars = 'abcdefghijklmnopqrstuvwxyz'; fprintf('%c', chars(decodedMarginals)); fprintf('\n'); end
github
facundoq/courses-master
MaxDecoding.m
.m
courses-master/pgm/p4/MaxDecoding.m
933
utf_8
5a3a81860c3e4efd46b5b7b9cb67d6f0
%MAXDECODING Finds the best assignment for each variable from the marginals M %passed in. Returns A such that A(i) returns the index of the best %instantiation for variable i. % % For instance: Let's say we have two variables 1 and 2. % Marginals for 1 = [0.1, 0.3, 0.6] % Marginals for 2 = [0.92, 0.08] % A(1) ...
github
facundoq/courses-master
FactorsSum.m
.m
courses-master/pgm/p4/FactorsSum.m
707
utf_8
a03c697932279d1b1c7e0b84e8f119f8
% FactorSum Computes the sum of two factors. % C = FactorProduct(A,B) computes the sum between two factors, A and B, % where each factor is defined over a set of variables with given dimension. % The factor data structure has the following fields: % .var Vector of variables in the factor, e.g. [1 2 3] % ...
github
facundoq/courses-master
FactorProduct.m
.m
courses-master/pgm/p4/FactorProduct.m
2,357
utf_8
06035e60b1296257aaf43c3e4e8aa15c
% FactorProduct Computes the product of two factors. % C = FactorProduct(A,B) computes the product between two factors, A and B, % where each factor is defined over a set of variables with given dimension. % The factor data structure has the following fields: % .var Vector of variables in the factor, e.g...
github
facundoq/courses-master
submit.m
.m
courses-master/pgm/p4/submit.m
28,930
utf_8
3e767e0b1938f984f7b29e22e066321d
function submit(partId, webSubmit) %SUBMIT Submit your code and output to the pgm-class servers % SUBMIT() will connect to the pgm-class server and submit your solution % There is no penalty for submitting, so go ahead and try this! % % If this function does not work for you, use the web-submission mechanism. % ...
github
facundoq/courses-master
StandardizeFactors.m
.m
courses-master/pgm/p4/StandardizeFactors.m
621
utf_8
f622aaecac8d0b7e4493aacd074eda6c
% Function that sorts the variables in F and returns an equivalent % factor G. Used only to standardize output for grading purposes. function G = StandardizeFactors(F) G = struct('var', [], 'card', [], 'val', []); for i = 1:length(F) G(i) = StandardizeFactor(F(i)); end function G = StandardizeFactor(F); G = stru...
github
facundoq/courses-master
ComputeInitialPotentials.m
.m
courses-master/pgm/p4/ComputeInitialPotentials.m
2,044
utf_8
a99695a5590d08313c5386fec70909a4
%COMPUTEINITIALPOTENTIALS Sets up the cliques in the clique tree that is %passed in as a parameter. % % P = COMPUTEINITIALPOTENTIALS(C) Takes the clique tree skeleton C which is a % struct with three fields: % - nodes: cell array representing the cliques in the tree. % - edges: represents the adjacency matrix o...
github
facundoq/courses-master
GetValueOfAssignment.m
.m
courses-master/pgm/p4/GetValueOfAssignment.m
838
utf_8
44eefd1a7a3ae4efe72c80711a4ac67d
% GetValueOfAssignment Gets the value of a variable assignment in a factor. % % v = GetValueOfAssignment(F, A) returns the value of a variable assignment, % A, in factor F. The order of the variables in A are assumed to be the % same as the order in F.var. % % v = GetValueOfAssignment(F, A, VO) gets the value o...
github
facundoq/courses-master
GetNextCliques.m
.m
courses-master/pgm/p4/GetNextCliques.m
2,759
utf_8
7e4effa84213f2cc1c89be9b6030a174
%GETNEXTCLIQUES Find a pair of cliques ready for message passing % [i, j] = GETNEXTCLIQUES(P, messages) finds ready cliques in a given % clique tree, P, and a matrix of current messages. Returns indices i and j % such that clique i is ready to transmit a message to clique j. % % We are doing clique tree message...
github
facundoq/courses-master
FactorMaxMarginalization.m
.m
courses-master/pgm/p4/FactorMaxMarginalization.m
1,968
utf_8
e662d931faa6c7201e6216627575204b
% FactorMaxMarginalization Max-marginalizes a factor % by taking the max over a given set variables. % % B = FactorMaxMarginalization(A,V) computes the factor with the variables % in V maxed out. The factor data structure has the following fields: % .var Vector of variables in the factor, e.g. [1 2 3] % ...
github
facundoq/courses-master
ComputeMarginal.m
.m
courses-master/pgm/p4/ComputeMarginal.m
1,227
utf_8
67b61710048f402adfa7ea23875b4872
%ComputeMarginal Computes the marginal over a set of given variables % M = ComputeMarginal(V, F, E) computes the marginal over variables V % in the distribution induced by the set of factors F, given evidence E % % M is a factor containing the marginal over variables V % V is a vector containing the variables i...
github
facundoq/courses-master
AssignmentToIndex.m
.m
courses-master/pgm/p4/AssignmentToIndex.m
622
utf_8
3da521179d034588af2bfe11ed13f8f7
% AssignmentToIndex Convert assignment to index. % % I = AssignmentToIndex(A, D) converts an assignment, A, over variables % with cardinality D to an index into the .val vector for a factor. % If A is a matrix then the function converts each row of A to an index. % % See also IndexToAssignment.m % % Copyright ...
github
facundoq/courses-master
CreateCliqueTree.m
.m
courses-master/pgm/p4/CreateCliqueTree.m
2,238
utf_8
0b309fb32a6aee40615f6221d60a710e
%CREATECLIQUETREE Takes in a list of factors F, Evidence and returns a %clique tree after calling ComputeInitialPotentials at the end. % % P = CREATECLIQUETREE(F, Evidence) Takes a list of factors and creates a clique % tree. The value of the cliques should be initialized to % the initial potential. % It ret...
github
facundoq/courses-master
submitWeb.m
.m
courses-master/pgm/p4/submitWeb.m
581
utf_8
40868cac2a1f7de8fbdc5855c255d44d
% submitWeb Creates files from your code and output for web submission. % % If the submit function does not work for you, use the web-submission mechanism. % Call this function to produce a file for the part you wish to submit. Then, % submit the file to the class servers using the "Web Submission" button on ...
github
facundoq/courses-master
ComputeJointDistribution.m
.m
courses-master/pgm/p4/ComputeJointDistribution.m
1,270
utf_8
f5233de99da1f71ac79340c59698670d
%ComputeJointDistribution Computes the joint distribution defined by a set % of given factors % % Joint = ComputeJointDistribution(F) computes the joint distribution % defined by a set of given factors % % Joint is a factor that encapsulates the joint distribution given by F % F is a vector of factors (s...
github
facundoq/courses-master
eq_factors.m
.m
courses-master/pgm/p4/eq_factors.m
372
utf_8
0425650440c2e53ea5fb0979fb30200d
function eq=eq_factors(F1,F2) assert(length(F1)==length(F2),'Both factor lists must have the same number of elements'); eq=true; for i=1:length(F1) if ~eq_factor(F1(i),F2(i)) eq=false; break; end end end function eq=eq_factor(f1,f2) eq= eq_eps(f1.var,f2.var) && e...
github
facundoq/courses-master
comparedata.m
.m
courses-master/pgm/p4/comparedata.m
12,980
utf_8
a7242d9a91fcd2959131f5fce367f085
function retval=comparedata(data1, data2, context, Params) % function retval=comparedata(data1, data2, context, Params) % compares to see if data1 and data2 are roughly recursively equal. "Rough" here is defined by % the Params. Note that matlabs ISEQUAL function is a test for exact equality. comparedata % ...
github
facundoq/courses-master
ObserveEvidence.m
.m
courses-master/pgm/p4/ObserveEvidence.m
2,193
utf_8
4af4a267a25aaff99190410099accb86
% ObserveEvidence Modify a vector of factors given some evidence. % F = ObserveEvidence(F, E) sets all entries in the vector of factors, F, % that are not consistent with the evidence, E, to zero. F is a vector of % factors, each a data structure with the following fields: % .var Vector of variables in the...
github
facundoq/courses-master
SetValueOfAssignment.m
.m
courses-master/pgm/p4/SetValueOfAssignment.m
1,181
utf_8
6a83e76d1be90e51b6e7eba649f9f209
% SetValueOfAssignment Sets the value of a variable assignment in a factor. % % F = SetValueOfAssignment(F, A, v) sets the value of a variable assignment, % A, in factor F to v. The order of the variables in A are assumed to be the % same as the order in F.var. % % F = SetValueOfAssignment(F, A, v, VO) sets the...
github
facundoq/courses-master
set_inclusion.m
.m
courses-master/pgm/p4/set_inclusion.m
127
utf_8
911e20355be2c91e5c65cb8350c2612a
%set_inclusion(a,b) % true if all elements of vector a are in vector b function r=set_inclusion(a,b) r=all(ismember(a,b)); end
github
facundoq/courses-master
EM_HMM.m
.m
courses-master/pgm/p9/EM_HMM.m
6,667
utf_8
94b98a6c15a38822e34e8dbad97b37e6
% File: EM_HMM.m % % Copyright (C) Daphne Koller, Stanford Univerity, 2012 function [P, loglikelihood, ClassProb, PairProb] = EM_HMM(actionData, poseData, G, InitialClassProb, InitialPairProb, maxIter) % INPUTS % actionData: structure holding the actions as described in the PA % poseData: N x 10 x 3 matrix, where N i...
github
facundoq/courses-master
ShowPose.m
.m
courses-master/pgm/p9/ShowPose.m
1,415
utf_8
d88d1852aab394ef850174ee2185f4bd
% File: ShowPose.m % % Copyright (C) Daphne Koller, Stanford Univerity, 2012 % visualize a configuration the body parts function img = ShowPose( pose ) % pose 10 x 3. 10 : body parts, % 3 : y, x, alpha pose(:,1) = pose(:,1) + 100; pose(:,2) = pose(:,2) + 150; pose = reshape(pose, [10 3]); part_leng...
github
facundoq/courses-master
lognormpdf.m
.m
courses-master/pgm/p9/lognormpdf.m
556
utf_8
93179aac759911fb170081a38b7025a8
% File: lognormpdf.m % % Copyright (C) Daphne Koller, Stanford Univerity, 2012 function [log_prob] = lognormpdf(x,mu,sigma) % LOGNORMPDF Natural logarithm of the normal probability density function (pdf) % Y = lognormpdf(X,MU,SIGMA) returns the log of the pdf of the normal % distribution parameterized by mean MU and ...
github
facundoq/courses-master
IndexToAssignment.m
.m
courses-master/pgm/p9/IndexToAssignment.m
587
utf_8
16baab12b20308058dd13e7565a96e60
% IndexToAssignment Convert index to variable assignment. % % A = IndexToAssignment(I, D) converts an index, I, into the .val vector % into an assignment over variables with cardinality D. If I is a vector, % then the function produces a matrix of assignments, one assignment % per row. % % See also Assignme...
github
facundoq/courses-master
CliqueTreeCalibrate.m
.m
courses-master/pgm/p9/CliqueTreeCalibrate.m
3,196
utf_8
b2d6a4c7d27c772e31acc30dbf52518b
%CLIQUETREECALIBRATE Performs sum-product or max-product algorithm for %clique tree calibration. % P = CLIQUETREECALIBRATE(P, isMax) calibrates a given clique tree, P % according to the value of isMax flag. If isMax is 1, it uses max-product % message passing, otherwise uses sum-product. This function % ret...
github
facundoq/courses-master
CreateCliqueTreeHMM.m
.m
courses-master/pgm/p9/CreateCliqueTreeHMM.m
2,423
utf_8
89c8f4dc2a761a58cf501fad96aa5b9a
% CreateCliqueTreeHMM Takes in a list of factors F and returns a clique % tree. Should only be called when F meets the following conditions: % % 1) Factors are over 1 or 2 variables % 2) All 2-variable factors are over variables (i,i+1) % 3) All variables have the same cardinality % % Roughly, these conditions mean tha...
github
facundoq/courses-master
FitG.m
.m
courses-master/pgm/p9/FitG.m
352
utf_8
1c8b5ca906775262ec10c160aa613b83
% File: FitG.m % % Copyright (C) Daphne Koller, Stanford Univerity, 2012 function [mu sigma] = FitG(X, W) % X: (N x 1): N examples (1 dimensional) % W: (N x 1): Weights over examples (W(i) is the weight for X(i)) % Fit N(mu, sigma^2) to the empirical distribution mu = 0; sigma = 1; mu = W'*X/sum(W); v = W'*(X.*X)/...
github
facundoq/courses-master
FactorMarginalization.m
.m
courses-master/pgm/p9/FactorMarginalization.m
821
utf_8
9481d98b85585038c22e4faa9583992c
% FactorMarginalization Sums given variables out of a factor in log space. % B = FactorMarginalization(A,V) computes the factor with the variables % in V summed out. The factor data structure has the following fields: % .var Vector of variables in the factor, e.g. [1 2 3] % .card Vector of cardin...
github
facundoq/courses-master
RecognizeActions.m
.m
courses-master/pgm/p9/RecognizeActions.m
3,320
utf_8
397eca61a70e4305fe6aea2c683e9a87
% File: RecognizeActions.m % % Copyright (C) Daphne Koller, Stanford Univerity, 2012 function [accuracy, predicted_labels] = RecognizeActions(datasetTrain, datasetTest, G, maxIter) % INPUTS % datasetTrain: dataset for training models, see PA for details % datasetTest: dataset for testing models, see PA for detail...
github
facundoq/courses-master
estimate_initial_probabilities.m
.m
courses-master/pgm/p9/estimate_initial_probabilities.m
1,383
utf_8
b597e18e197e28fee0bc8407b90e5618
function dataset=estimate_initial_probabilities(dataset,k_poses) for i=1:length(dataset) action_samples=dataset(i); [action_samples.InitialClassProb, action_samples.InitialPairProb]=estimate_initial_probabilities_class(action_samples,k_poses); dataset(i)=action_samples; end end function [class_prob,pair_...
github
facundoq/courses-master
ComputeExactMarginalsHMM.m
.m
courses-master/pgm/p9/ComputeExactMarginalsHMM.m
1,613
utf_8
d2dad5ff1748ef025220bd22df9876ad
%COMPUTEEXACTMARGINALSHMM Runs exact inference and returns the marginals %over all the variables and the calibrated clique tree. % M = COMPUTEEXACTMARGINALSHMM(F) Takes a list of factors F, % and runs exact inference and returns the calibrated clique tree (unnormalized) and % final marginals (normalized) for the...
github
facundoq/courses-master
submit.m
.m
courses-master/pgm/p9/submit.m
22,612
utf_8
3c90403ef5f9b0b52e55bbf559b97d22
function submit(partId, webSubmit) %SUBMIT Submit your code and output to the pgm-class servers % SUBMIT() will connect to the pgm-class server and submit your solution % There is no penalty for submitting, so go ahead and try this! % % If this function does not work for you, use the web-submission mechanism. % ...
github
facundoq/courses-master
ComputeClassLikelihoodSample.m
.m
courses-master/pgm/p9/ComputeClassLikelihoodSample.m
2,485
utf_8
a4d345390cfad113fd8a5395f8a8f243
function class_l = ComputeClassLikelihoodSample(P, G, sample) % You should compute the log likelihood of data as in eq. (12) and (13) % in the PA description % Hint: Use lognormpdf instead of log(normpdf) to prevent underflow. % You may use log(sum(exp(logProb))) to do addition in the original % space, sum...
github
facundoq/courses-master
VisualizeDataset.m
.m
courses-master/pgm/p9/VisualizeDataset.m
243
utf_8
ce8d4d0fc87937b06c2daa5552ff0330
% File: VisualizeDataset.m % % Copyright (C) Daphne Koller, Stanford Univerity, 2012 function VisualizeDataset(Dataset) figure for i=1:size(Dataset,1) img = ShowPose(reshape(Dataset(i,:,:), [10 3])); imshow(img); pause(0.3); end
github
facundoq/courses-master
AssignmentToIndex.m
.m
courses-master/pgm/p9/AssignmentToIndex.m
609
utf_8
3c17a18df90fc3a49aeebcb3e418e23f
% AssignmentToIndex Convert assignment to index. % % I = AssignmentToIndex(A, D) converts an assignment, A, over variables % with cardinality D to an index into the .val vector for a factor. % If A is a matrix then the function converts each row of A to an index. % % See also IndexToAssignment.m % % Copyright ...
github
facundoq/courses-master
submitWeb.m
.m
courses-master/pgm/p9/submitWeb.m
580
utf_8
5f4510147426716d140b1e22e95d36d7
% submitWeb Creates files from your code and output for web submission. % % If the submit function does not work for you, use the web-submission mechanism. % Call this function to produce a file for the part you wish to submit. Then, % submit the file to the class servers using the "Web Submission" button on ...
github
facundoq/courses-master
FitLG.m
.m
courses-master/pgm/p9/FitLG.m
1,648
utf_8
fb35b5bf14764213e24bf53eeba348ea
% File: FitLG.m % % Copyright (C) Daphne Koller, Stanford Univerity, 2012 function [Beta sigma] = FitLG(X, U, W) % Estimate parameters of the linear Gaussian model: % X|U ~ N(Beta(1)*U(1) + ... + Beta(K)*U(K) + Beta(K+1), sigma^2); % Note that Matlab index from 1, we can't write Beta(0). So Beta(K+1) is % essentiall...
github
facundoq/courses-master
PrintFactor.m
.m
courses-master/pgm/p9/PrintFactor.m
526
utf_8
7ab925539a614f01ebbef8c4c001d3c2
% Copyright (C) Daphne Koller, Stanford University, 2012 function [] = PrintFactor(F) % Pretty print the factor F. % The first row lists the variables and subsequent rows are % the joint assignment and their associated factor value in % the last column. for i=1:length(F.var) fprintf(1, '%d\t', F.var(i...
github
facundoq/courses-master
EM_cluster.m
.m
courses-master/pgm/p9/EM_cluster.m
3,444
utf_8
2a65249f9f25c300a24cfe746ca6f417
% File: EM_cluster.m % % Copyright (C) Daphne Koller, Stanford Univerity, 2012 function [P, loglikelihood, soft_labels] = EM_cluster(poseData, G, InitialClassProb, maxIter) % INPUTS % poseData: N x 10 x 3 matrix, where N is number of poses; % poseData(i,:,:) yields the 10x3 matrix for pose i. % G: graph parameteriz...
github
facundoq/courses-master
CompareData.m
.m
courses-master/pgm/p9/CompareData.m
12,989
utf_8
c611d62554c31e24c84df3ec9447f6ce
function retval = CompareData(data1, data2, context, Params) % function retval = comparedata(data1, data2, context, Params) % compares to see if data1 and data2 are roughly recursively equal. "Rough" here is defined by % the Params. Note that matlabs ISEQUAL function is a test for exact equality. comparedata % ...
github
facundoq/courses-master
expand_dataset.m
.m
courses-master/pgm/p9/expand_dataset.m
2,139
utf_8
0d6505d19cd7a2361aef5b353deb1717
function datasetTrain=expand_dataset(datasetTrain,expand_factor,perturbation_factor) expand_factor=expand_factor-1; if expand_factor<=0 return end for i=1:length(datasetTrain) class_samples=datasetTrain(i); [new_actions,new_poses]=perturb_class_actions(class_samples.actionData,class_samples.poseData,expand...
github
facundoq/courses-master
logsumexp.m
.m
courses-master/pgm/p9/logsumexp.m
341
utf_8
4a201a5a36500a989a523c436ac4b69e
% File: logsumexp.m % % Copyright (C) Daphne Koller, Stanford Univerity, 2012 function out = logsumexp(A) % LOGSUMEXP % Computes log( sum( exp( ) ) ) of each row in A in a way that avoids underflow. % If A is an N x M matrix, then out is a N x 1 vector. pi_max = max(A, [], 2); out = pi_max + log(sum(exp(bsxfun(@minu...
github
facundoq/courses-master
LearnCPDsGivenGraph.m
.m
courses-master/pgm/p9/LearnCPDsGivenGraph.m
2,802
utf_8
eb48720161e07326645907476762e2bf
function [P] = LearnCPDsGivenGraph(dataset, G, soft_labels) % % Inputs: % dataset: N x 10 x 3, N poses represented by 10 parts in (y, x, alpha) % G: graph parameterization as explained in PA description % labels: N x 2 true class labels for the examples. labels(i,j)=1 if the % the ith example belongs to class ...
github
facundoq/courses-master
ComputeExactMarginalsBP.m
.m
courses-master/pgm/p7/ComputeExactMarginalsBP.m
2,261
utf_8
7c623bc99a6cabc2bfb5379ab75c4e06
%COMPUTEEXACTMARGINALSBP Runs exact inference and returns the marginals %over all the variables. % M = COMPUTEEXACTMARGINALSBP(F,E, isMax) Takes a list of factors F, % evidence E and a flag isMax and run exact inference and returns the % final marginals for the variables in the network. If isMax is 1, then % ...
github
facundoq/courses-master
FactorSum.m
.m
courses-master/pgm/p7/FactorSum.m
2,208
utf_8
b32940673c307cf1e23efcd1dc3cc9fe
% FactorProduct Computes the product of two factors. % C = FactorSum(A,B) computes the sum of two factors, A and B, % where each factor is defined over a set of variables with given dimension. % The factor data structure has the following fields: % .var Vector of variables in the factor, e.g. [1 2 3] % ...
github
facundoq/courses-master
LRSearchLambdaSGD.m
.m
courses-master/pgm/p7/LRSearchLambdaSGD.m
1,256
utf_8
8e8df067cee7994d967166097b9e4aeb
% function allAcc = LRSearchLambdaSGD(Xtrain, Ytrain, Xvalidation, Yvalidation, lambdas) % For each value of lambda provided, fit parameters to the training data and return % the accuracy in the validation data in the corresponding entry of allAcc. % For instance, allAcc(i) = accuracy in the validation set using lambda...
github
facundoq/courses-master
LRPredict.m
.m
courses-master/pgm/p7/LRPredict.m
514
utf_8
0cd38f90588722af4558c1227da54620
% pred = LRPredict(X, theta) uses the LR classifier encoded by theta % to predict the labels on data X. % % Inputs: % X data. (numInstances x numFeatures matrix) % theta LR parameters. (numFeatures x 1 vector) % % Outputs: % pred predicted labels for X. (numInstances...
github
facundoq/courses-master
EliminateVar.m
.m
courses-master/pgm/p7/EliminateVar.m
1,344
utf_8
5eee22473a7397f75d71157c426f2ea7
% Function used in production of clique trees % % Copyright (C) Daphne Koller, Stanford Univerity, 2012 function [newF C E] = EliminateVar(F, C, E, Z) useFactors = []; scope = []; for i=1:length(F) if any(F(i).var == Z) useFactors = [useFactors i]; scope = union(scope, F(i).var); end end % u...
github
facundoq/courses-master
StochasticGradientDescent.m
.m
courses-master/pgm/p7/StochasticGradientDescent.m
1,649
utf_8
680d4fdda11ed5ed5db6772f70e8ecd3
% function thetaOpt = StochasticGradientDescent (gradFunc, theta0, maxiter) % runs gradient descent until convergence, returning the optimal parameters thetaOpt. % % Inputs: % gradFunc function handle to a function [cost, grad] = gradFunc(theta, i) % that computes the LR cost / objective function and...
github
facundoq/courses-master
IndexToAssignment.m
.m
courses-master/pgm/p7/IndexToAssignment.m
587
utf_8
16baab12b20308058dd13e7565a96e60
% IndexToAssignment Convert index to variable assignment. % % A = IndexToAssignment(I, D) converts an index, I, into the .val vector % into an assignment over variables with cardinality D. If I is a vector, % then the function produces a matrix of assignments, one assignment % per row. % % See also Assignme...
github
facundoq/courses-master
LRCostSGD.m
.m
courses-master/pgm/p7/LRCostSGD.m
1,496
utf_8
d971795e0a6b27c9bc5e105958f368f5
% [cost, grad] = LRCostSGD(X, y, theta, lambda, i) calculates the LR cost / objective % function with respect to data instance (i mod n), given the LR classifier parameterized % by theta and where n = the number of data instances. Also returns the gradient of % the cost function with respect to data instance (i mod ...
github
facundoq/courses-master
CliqueTreeCalibrate.m
.m
courses-master/pgm/p7/CliqueTreeCalibrate.m
5,314
utf_8
08c84518c832001d732bd798e1518e67
%CLIQUETREECALIBRATE Performs sum-product or max-product algorithm for %clique tree calibration. % [P] = CLIQUETREECALIBRATE(P, isMax) calibrates a given clique tree, P % according to the value of isMax flag. If isMax is 1, it uses max-product % message passing, otherwise uses sum-product. This function % r...
github
facundoq/courses-master
GenerateAllFeatures.m
.m
courses-master/pgm/p7/GenerateAllFeatures.m
2,513
utf_8
d5e82ae09bde41bb49215839c7ce6c1b
% This function is called by InstanceNegLogLikelihood. % Its input/output is specified there. % If you're interested in the implementation details of CRFs, % feel free to read through this code! % For the purposes of this assignment, though, you don't % have to understand how this code works. % % Copyright (C) Daphne K...
github
facundoq/courses-master
FactorMarginalization.m
.m
courses-master/pgm/p7/FactorMarginalization.m
1,551
utf_8
470be9bc98bbb2fb7e2b10913db1da6d
% FactorMarginalization Sums given variables out of a factor. % B = FactorMarginalization(A,V) computes the factor with the variables % in V summed out. The factor data structure has the following fields: % .var Vector of variables in the factor, e.g. [1 2 3] % .card Vector of cardinalities corresp...
github
facundoq/courses-master
LRTrainSGD.m
.m
courses-master/pgm/p7/LRTrainSGD.m
1,353
utf_8
d3ef910e60f9f60de975d70858e94283
% thetaOpt = LRTrainSGD(X, y, lambda) trains a logistic regression % classifier using stochastic gradient descent. It returns the optimal theta values. % % Inputs: % X data. (numInstances x numFeatures matrix) % X(:,1) is all ones, i.e., it encodes the intercept/bias term. %...
github
facundoq/courses-master
MaxDecoding.m
.m
courses-master/pgm/p7/MaxDecoding.m
659
utf_8
9168d0f003f618eb31f0006d6c83970d
%MAXDECODING Finds the best assignment for each variable from the marginals %passed in. Returns A such that A(i) returns the index of the best %instantiation for variable i. % % For instance: Let's say we have two variables 1 and 2. % Marginals for 1 = [0.1, 0.3, 0.6] % Marginals for 2 = [0.92, 0.08] % A(1) = ...
github
facundoq/courses-master
LRAccuracy.m
.m
courses-master/pgm/p7/LRAccuracy.m
658
utf_8
bf10b837298e6f94c490327c020e201e
% function acc = LRAccuracy(GroundTruth, Predictions) compares the % vector of predictions with the vector of ground truth values, % and returns the accuracy (fraction of predictions that are correct). % % Input: % GroundTruth (numInstances x 1 vector) % Predictions (numInstances x 1 vector) % % Output: % err...
github
facundoq/courses-master
FactorProduct.m
.m
courses-master/pgm/p7/FactorProduct.m
2,160
utf_8
7f87add7a3c8bfac085a23c8cb2ba03c
% FactorProduct Computes the product of two factors. % C = FactorProduct(A,B) computes the product between two factors, A and B, % where each factor is defined over a set of variables with given dimension. % The factor data structure has the following fields: % .var Vector of variables in the factor, e.g...
github
facundoq/courses-master
submit.m
.m
courses-master/pgm/p7/submit.m
20,967
utf_8
c1248b5f208f3aacb05dd313678cdb8b
function submit(partId, webSubmit) %SUBMIT Submit your code and output to the pgm-class servers % SUBMIT() will connect to the pgm-class server and submit your solution % There is no penalty for submitting, so go ahead and try this! % % If this function does not work for you, use the web-submission mechanism. % ...
github
facundoq/courses-master
ComputeInitialPotentials.m
.m
courses-master/pgm/p7/ComputeInitialPotentials.m
3,604
utf_8
737d72093cf516b5431db5e4878b5891
%COMPUTEINITIALPOTENTIALS Sets up the cliques in the clique tree that is %passed in as a parameter. % P = COMPUTEINITIALPOTENTIALS(C) Takes the clique tree C which is a % struct with three fields: % - nodes: represents the cliques in the tree. % - edges: represents the adjacency matrix of the tree. % - facto...
github
facundoq/courses-master
GetValueOfAssignment.m
.m
courses-master/pgm/p7/GetValueOfAssignment.m
834
utf_8
0a65aef1740618807fb9e95821268dab
%GETVALUEOFASSIGNMENT Gets the value of a variable assignment in a factor. % % v = GETVALUEOFASSIGNMENT(F, A) returns the value of a variable assignment, % A, in factor F. The order of the variables in A are assumed to be the % same as the order in F.var. % % v = GETVALUEOFASSIGNMENT(F, A, VO) gets the va...
github
facundoq/courses-master
GetNextCliques.m
.m
courses-master/pgm/p7/GetNextCliques.m
1,742
utf_8
ba635097f61dfeba09e3885c9ee936b4
%GETNEXTCLIQUES Find a pair of cliques ready for message passing % [i, j] = GETNEXTCLIQUES(P, messages) finds ready cliques in a given % clique tree, P, and current messages. Returns indices i and j % such that clique i is ready to transmit a message to clique j. % If no such cliques exist, returns i = j = 0 % ...
github
facundoq/courses-master
FactorMaxMarginalization.m
.m
courses-master/pgm/p7/FactorMaxMarginalization.m
1,694
utf_8
eaa4b78b713845fc6b57c262e4764292
% FactorMaxMarginalization Takes the max of given variables when marginalizing out of a factor. % B = FactorMarginalization(A,V) computes the factor with the variables % in V summed out. The factor data structure has the following fields: % .var Vector of variables in the factor, e.g. [1 2 3] % .card...