-
Notifications
You must be signed in to change notification settings - Fork 41
/
main.cpp
92 lines (74 loc) · 2.51 KB
/
main.cpp
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
#include <iostream>
#include <Eigen/Eigen>
#include "alglib/src/ap.h"
#include "alglib/src/specialfunctions.h"
#include "vbfa.h"
#include "sim.h"
using namespace Eigen;
using namespace std;
using namespace PEER;
//using Cwise;
using alglib::randomreal;
using alglib::psi;
using alglib::gammafunction;
void playing()
{
// insert code here...
PMatrix m(2,2);
PVector v1,v2;
v1 = PVector::Constant(3, 2.1);
v2 = PVector::Constant(3, 1.3);
v1 = v2;
v2(1) = 3;
std::cout << PMatrix::Ones(3,4)*1.2 << std::endl;
std::cout << v1 << std::endl << v2 + PVector::Ones(3)*2.0 << std::endl;
m(0,0) = 3;
m(1,0) = 2.5;
m(0,1) = -1;
m(1,1) = m(1,0) + m(0,1);
std::cout << m << std::endl << m.rowwise().sum() << std::endl << m.colwise().sum() << std::endl;
std::cout << randomreal() << std::endl << psi(10) << std::endl << gammafunction(10) << std::endl;
}
/** Trying out update equations etc */
void play_matrix(){
int N = 3;
int P = 4;
int K = 2;
PMatrix alpha = PMatrix::Zero(K,K);//randn(K,1);
alpha.diagonal() = randn(K,1);
cout << alpha << endl;
PMatrix XE2s = randn(K,K);
PMatrix X = randn(N,K);
PMatrix eps = randn(P,1);
PMatrix E1 = randn(P,K);
PMatrix E2S = randn(K,K);
PMatrix pheno = randn(N,P);
PMatrix b = 0.1 + 0.5*E2S.diagonal().array();
PMatrix a = (10.0 + 0.5*P)*(PMatrix::Ones(K, 1).array());
// W update
for(int i=0; i<P; ++i){
PMatrix prec = alpha + XE2s*eps(i);
PMatrix cov = prec.inverse();
// cout << alpha << endl << XE2s << endl << XE2s*eps(i) << endl << prec << endl << cov << endl;
cout << pheno.col(i).rows() << pheno.col(i).cols() << endl << X.rows() << X.cols() << endl;
cout << eps(i)*cov*X.transpose()*pheno.col(i); // self.E1[d,:] = S.dot(dcov[:,:],Eps[d]*S.dot(_S.E1.T,net.dataNode.E1[ :,d]))
E1.row(i) = eps(i)*cov*X.transpose()*pheno.col(i); // self.E1[d,:] = S.dot(dcov[:,:],Eps[d]*S.dot(_S.E1.T,net.dataNode.E1[ :,d]))
// E1(i) = eps(i)*cov*X*pheno(i); // self.E1[d,:] = S.dot(dcov[:,:],Eps[d]*S.dot(_S.E1.T,net.dataNode.E1[ :,d]))
PMatrix outer = E1.row(i).transpose()*E1.row(i);
E2S += (cov + outer); // E2 = dcov + outer(self.E1[d], self.E1[d])
}
}
int main (int argc, char * const argv[]) {
//play_matrix();
//1. simulate small dataset
PMatrix Y = simulate_expression(100,20,5,0.01).expr;
PMatrix Yvar = 0.1*PMatrix::Ones(10,100);
//2. create object
//test with nonexisting variance:
Yvar = PMatrix();
cVBFA vb(Y,Yvar, PMatrix(), 8);
vb.setAdd_mean(false);
vb.setNmax_iterations(50);
vb.update();
cout << vb.Alpha.E1 << endl << vb.X.E1.col(0);
}