mirror of
https://github.com/TREX-CoE/Sherman-Morrison.git
synced 2024-12-26 06:15:08 +01:00
First implementation of Woodbury 2x2 and 3x3 kernels.
This commit is contained in:
parent
288bc83e19
commit
efe96cbeea
@ -56,6 +56,45 @@ void showMatrix(T *matrix, unsigned int M, std::string name) {
|
||||
std::cout << std::endl;
|
||||
}
|
||||
|
||||
template <typename T>
|
||||
void showMatrix2(T *matrix, unsigned int M, unsigned int N, std::string name) {
|
||||
std::cout.precision(17);
|
||||
std::cout << name << " = [" << std::endl;
|
||||
for (unsigned int i = 0; i < M; i++) {
|
||||
std::cout << "[";
|
||||
for (unsigned int j = 0; j < N; j++) {
|
||||
if (matrix[i * N + j] >= 0) {
|
||||
std::cout << " " << matrix[i * N + j] << ",";
|
||||
} else {
|
||||
std::cout << " " << matrix[i * N + j] << ",";
|
||||
}
|
||||
}
|
||||
std::cout << " ]," << std::endl;
|
||||
}
|
||||
std::cout << "]" << std::endl;
|
||||
std::cout << std::endl;
|
||||
}
|
||||
|
||||
|
||||
template <typename T>
|
||||
void showMatrixNS(T *matrix, unsigned int M, unsigned int N, std::string name) {
|
||||
std::cout.precision(17);
|
||||
std::cout << name << " = [" << std::endl;
|
||||
for (unsigned int i = 0; i < M; i++) {
|
||||
std::cout << "[";
|
||||
for (unsigned int j = 0; j < N; j++) {
|
||||
if (matrix[i * N + j] >= 0) {
|
||||
std::cout << " " << matrix[i * N + j] << ",";
|
||||
} else {
|
||||
std::cout << " " << matrix[i * N + j] << ",";
|
||||
}
|
||||
}
|
||||
std::cout << " ]," << std::endl;
|
||||
}
|
||||
std::cout << "]" << std::endl;
|
||||
std::cout << std::endl;
|
||||
}
|
||||
|
||||
template <typename T> T *transpose(T *A, unsigned int M) {
|
||||
T *B = new T[M * M];
|
||||
for (unsigned int i = 0; i < M; i++) {
|
||||
@ -77,6 +116,18 @@ template <typename T> void matMul(T *A, T *B, T *C, unsigned int M) {
|
||||
}
|
||||
}
|
||||
|
||||
template <typename T1, typename T2, typename T3>
|
||||
static inline void matMul2(T1 *A, T2 *B, T3 *C, unsigned int M, unsigned int N, unsigned int P) {
|
||||
for(unsigned int i = 0; i < M; i++) {
|
||||
for(unsigned int j = 0; j < P; j++) {
|
||||
C[i * P + j] = 0;
|
||||
for(unsigned int k = 0; k < N; k++) {
|
||||
C[i * P + j] += A[i * N + k] * B[k * P + j];
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
template <typename T1, typename T2>
|
||||
T1 *outProd(T1 *vec1, T2 *vec2, unsigned int M) {
|
||||
T1 *C = new T1[M * M];
|
||||
|
@ -1,7 +1,7 @@
|
||||
// Woodbury 2x2 kernel
|
||||
void WB2(double *Slater_inv, unsigned int Dim,
|
||||
bool WB2(double *Slater_inv, unsigned int Dim,
|
||||
double *Updates, unsigned int *Updates_index);
|
||||
|
||||
// Woodbury 3x3 kernel
|
||||
void WB3(double *Slater_inv, unsigned int Dim,
|
||||
bool WB3(double *Slater_inv, unsigned int Dim,
|
||||
double *Updates, unsigned int *Updates_index);
|
||||
|
@ -3,6 +3,10 @@ unset THRESHOLD
|
||||
unset MKL
|
||||
ENV=$1
|
||||
|
||||
echo
|
||||
echo "Sherman-Morrison-Woodbury parameters"
|
||||
echo "------------------------------------"
|
||||
|
||||
## Set Sherman-Morrison root dir
|
||||
PWD=$(pwd)
|
||||
SRCDIR=$(dirname $BASH_SOURCE)
|
||||
|
11
src/SMWB.cpp
11
src/SMWB.cpp
@ -5,13 +5,16 @@
|
||||
|
||||
// Sherman-Morrison-Woodbury kernel 1
|
||||
// WB2, WB3, SM2 mixing scheme 1
|
||||
void SMWB1(double *Slater_inv, unsigned int Dim, unsigned int N_updates,
|
||||
double *Updates, unsigned int *Updates_index) {
|
||||
void SMWB1(double *Slater_inv, unsigned int Dim, unsigned int N_updates, double *Updates, unsigned int *Updates_index) {
|
||||
std::cerr << "Called Sherman-Morrison-Woodbury kernel 1 with " << N_updates << " updates" << std::endl;
|
||||
WB3(Slater_inv, Dim, Updates, Updates_index);
|
||||
WB2(Slater_inv, Dim, Updates, Updates_index);
|
||||
|
||||
bool ok;
|
||||
ok = WB2(Slater_inv, Dim, Updates, Updates_index);
|
||||
if (!ok) {
|
||||
std::cerr << "Woodbury kernel failed!" << std::endl;
|
||||
SM2(Slater_inv, Dim, N_updates, Updates, Updates_index);
|
||||
}
|
||||
}
|
||||
|
||||
extern "C" {
|
||||
void SMWB1_f(double **linSlater_inv, unsigned int *Dim, unsigned int *N_updates,
|
||||
|
127
src/Woodbury.cpp
127
src/Woodbury.cpp
@ -5,65 +5,126 @@
|
||||
// (S + U * V)^{-1} = S^{-1} - S^{-1} * U * B^{-1} * V * S^{-1}
|
||||
// B := 1 + V * C, 2 x 2
|
||||
// C := S^{-1} * U, dim x 2
|
||||
// D := V * S^{-1}, 2 x dim
|
||||
//
|
||||
// All matrices are stored in row-major order
|
||||
|
||||
#include "Woodbury.hpp"
|
||||
#include "Helpers.hpp"
|
||||
|
||||
// Woodbury 2x2 kernel:
|
||||
void WB2(double *Slater_inv, unsigned int Dim,
|
||||
double *Updates, unsigned int *Updates_index) {
|
||||
bool WB2(double *Slater_inv, unsigned int Dim, double *Updates,
|
||||
unsigned int *Updates_index) {
|
||||
std::cerr << "Called Woodbury 2x2 kernel" << std::endl;
|
||||
|
||||
// Construct V from Updates_index
|
||||
unsigned int *V = new unsigned int[2 * Dim]{0};
|
||||
for (unsigned int i = 0; i < Dim; i++) {
|
||||
if (i == Updates_index[0] - 1) V[i] = 1;
|
||||
if (Dim + i == Updates_index[1] - 1) V[Dim + i] = 1;
|
||||
}
|
||||
unsigned int V[2 * Dim]{0}; // 2 x Dim matrix stored in row-major order
|
||||
V[Updates_index[0] - 1] = 1;
|
||||
V[Dim + Updates_index[1] - 1] = 1;
|
||||
|
||||
// Compute B from U, V and Slater_inv
|
||||
double B[4]{0};
|
||||
double Binv[4]{0};
|
||||
double C[4]{0};
|
||||
double D[2*Dim]{0};
|
||||
// Compute C
|
||||
for (unsigned int i = 0; i < Dim; i++) {
|
||||
for (unsigned int j = 0; j < 2; j++) {
|
||||
for (unsigned int k = 0; k < Dim; k++) {
|
||||
C[i * Dim + j] += Slater_inv[i * Dim + k] * Updates[k * Dim + j];
|
||||
}
|
||||
}
|
||||
}
|
||||
double C[2 * Dim]{0};
|
||||
matMul2(Slater_inv, Updates, C, Dim, Dim, 2);
|
||||
// Compute B
|
||||
for (unsigned int i = 0; i < 2; i++) {
|
||||
for (unsigned int j = 0; j < 2; j++) {
|
||||
for (unsigned int k = 0; k < Dim; k++) {
|
||||
B[i * Dim + j] += (i == j) + V[i * Dim + k] * C[k * Dim + j];
|
||||
}
|
||||
}
|
||||
}
|
||||
double B[4]{0};
|
||||
matMul2(V, C, B, 2, Dim, 2);
|
||||
// Compute 1 + B
|
||||
B[0] += 1;
|
||||
B[3] += 1;
|
||||
|
||||
// Invert B with explicit formula for 2x2 inversion
|
||||
// Invert 1 + B with explicit formula for 2x2 inversion
|
||||
double idet = 1.0 / (B[0] * B[3] - B[1] * B[2]);
|
||||
double Binv[4]{0};
|
||||
Binv[0] = idet * B[3];
|
||||
Binv[1] = -1.0 * idet * B[1];
|
||||
Binv[2] = -1.0 * idet * B[2];
|
||||
Binv[3] = idet * B[0];
|
||||
|
||||
// Compute (S + U * V)^{-1} with Woobury identity
|
||||
// Check if determinant of inverted matrix is not zero
|
||||
double det = B[0] * B[3] - B[1] * B[2];
|
||||
if (std::fabs(det) < threshold()) {
|
||||
std::cerr << "Determinant approached 0!" << std::endl;
|
||||
return false;
|
||||
}
|
||||
|
||||
// Compute (S + U * V)^{-1} with Woobury identity
|
||||
double D[2 * Dim]{0};
|
||||
matMul2(V, Slater_inv, D, 2, Dim, Dim);
|
||||
double tmp[2 * Dim]{0};
|
||||
matMul2(Binv, D, tmp, 2, 2, Dim);
|
||||
double tmp2[Dim * Dim]{0};
|
||||
matMul2(C, tmp, tmp2, Dim, 2, Dim);
|
||||
for (unsigned int i = 0; i < Dim * Dim; i++) {
|
||||
Slater_inv[i] -= tmp2[i];
|
||||
}
|
||||
return true;
|
||||
}
|
||||
|
||||
// Woodbury 3x3 kernel
|
||||
void WB3(double *Slater_inv, unsigned int Dim,
|
||||
double *Updates, unsigned int *Updates_index) {
|
||||
bool WB3(double *Slater_inv, unsigned int Dim, double *Updates,
|
||||
unsigned int *Updates_index) {
|
||||
std::cerr << "Called Woodbury 3x3 kernel" << std::endl;
|
||||
|
||||
// Construct V from Updates_index
|
||||
unsigned int V[3 * Dim]{0}; // 2 x Dim matrix stored in row-major order
|
||||
V[Updates_index[0] - 1] = 1;
|
||||
V[Dim + Updates_index[1] - 1] = 1;
|
||||
V[2 * Dim + Updates_index[2] - 1] = 1;
|
||||
|
||||
// Compute B from U, V and Slater_inv
|
||||
// Compute C
|
||||
double C[3 * Dim]{0};
|
||||
matMul2(Slater_inv, Updates, C, Dim, Dim, 3);
|
||||
// Compute B
|
||||
double B[9]{0};
|
||||
matMul2(V, C, B, 3, Dim, 3);
|
||||
// Compute 1 + B
|
||||
B[0] += 1;
|
||||
B[4] += 1;
|
||||
B[8] += 1;
|
||||
|
||||
// Invert B with explicit formula for 3x3 inversion
|
||||
double Binv[9];
|
||||
Binv[0] = B[4] * B[8] - B[5] * B[7];
|
||||
Binv[3] = B[5] * B[6] - B[3] * B[8];
|
||||
Binv[6] = B[3] * B[7] - B[4] * B[6];
|
||||
|
||||
Binv[1] = B[2] * B[7] - B[1] * B[8];
|
||||
Binv[4] = B[0] * B[8] - B[2] * B[6];
|
||||
Binv[7] = B[1] * B[6] - B[0] * B[7];
|
||||
|
||||
Binv[2] = B[1] * B[5] - B[2] * B[4];
|
||||
Binv[5] = B[2] * B[3] - B[0] * B[5];
|
||||
Binv[8] = B[0] * B[4] - B[1] * B[3];
|
||||
|
||||
// Check if determinant of inverted matrix is not zero
|
||||
// If so, exigt and return false.
|
||||
double det;
|
||||
det = B[0] * (B[4] * B[8] - B[5] * B[7]) -
|
||||
B[1] * (B[3] * B[8] - B[5] * B[6]) + B[2] * (B[3] * B[7] - B[4] * B[6]);
|
||||
if (std::fabs(det) < threshold()) {
|
||||
std::cerr << "Determinant approached 0!" << std::endl;
|
||||
return false;
|
||||
}
|
||||
|
||||
// Compute (S + U * V)^{-1} with Woobury identity
|
||||
|
||||
double D[3 * Dim]{0};
|
||||
matMul2(V, Slater_inv, D, 3, Dim, Dim);
|
||||
double tmp[3 * Dim]{0};
|
||||
matMul2(Binv, D, tmp, 3, 3, Dim);
|
||||
double tmp2[Dim * Dim]{0};
|
||||
matMul2(C, tmp, tmp2, Dim, 3, Dim);
|
||||
for (unsigned int i = 0; i < Dim * Dim; i++) {
|
||||
Slater_inv[i] -= tmp2[i];
|
||||
}
|
||||
return true;
|
||||
}
|
||||
|
||||
extern "C" {
|
||||
bool WB2_f(double **linSlater_inv, unsigned int *Dim, double **linUpdates,
|
||||
unsigned int **Updates_index) {
|
||||
WB2(*linSlater_inv, *Dim, *linUpdates, *Updates_index);
|
||||
}
|
||||
bool WB3_f(double **linSlater_inv, unsigned int *Dim, double **linUpdates,
|
||||
unsigned int **Updates_index) {
|
||||
WB3(*linSlater_inv, *Dim, *linUpdates, *Updates_index);
|
||||
}
|
||||
}
|
||||
|
10
todo.txt
10
todo.txt
@ -1,10 +0,0 @@
|
||||
-=={TODO}==-
|
||||
|
||||
* Looking at the eigenvalues of the intermediate matrix after a problematic update is
|
||||
* plotting the ratio of the moduli of the maximum and the minimum eigenvalues of
|
||||
|
||||
* Use valgrind to find the problem with MaponiA3S
|
||||
* Contact Claudia if she would be interested in sharing datasets from Champ to test with SM
|
||||
|
||||
* Make a representative selection of update cycles for presentatoin puroses
|
||||
|
Loading…
Reference in New Issue
Block a user