mirror of
https://github.com/UpsilonNumworks/Upsilon.git
synced 2026-01-18 16:27:34 +01:00
80 lines
2.4 KiB
C++
80 lines
2.4 KiB
C++
#include "power_model.h"
|
|
#include "../store.h"
|
|
#include <math.h>
|
|
#include <assert.h>
|
|
#include <poincare/code_point_layout.h>
|
|
#include <poincare/horizontal_layout.h>
|
|
#include <poincare/vertical_offset_layout.h>
|
|
|
|
using namespace Poincare;
|
|
|
|
namespace Regression {
|
|
|
|
Layout PowerModel::layout() {
|
|
if (m_layout.isUninitialized()) {
|
|
m_layout = HorizontalLayout::Builder({
|
|
CodePointLayout::Builder('a', k_layoutFont),
|
|
CodePointLayout::Builder(UCodePointMiddleDot, k_layoutFont),
|
|
CodePointLayout::Builder('X', k_layoutFont),
|
|
VerticalOffsetLayout::Builder(
|
|
CodePointLayout::Builder('b', k_layoutFont),
|
|
VerticalOffsetLayoutNode::Position::Superscript
|
|
),
|
|
});
|
|
}
|
|
return m_layout;
|
|
}
|
|
|
|
double PowerModel::evaluate(double * modelCoefficients, double x) const {
|
|
double a = modelCoefficients[0];
|
|
double b = modelCoefficients[1];
|
|
return a*pow(x,b);
|
|
}
|
|
|
|
double PowerModel::levelSet(double * modelCoefficients, double xMin, double step, double xMax, double y, Poincare::Context * context) {
|
|
double a = modelCoefficients[0];
|
|
double b = modelCoefficients[1];
|
|
if (a == 0 || b == 0|| y/a <= 0) {
|
|
return NAN;
|
|
}
|
|
return exp(log(y/a)/b);
|
|
}
|
|
|
|
double PowerModel::partialDerivate(double * modelCoefficients, int derivateCoefficientIndex, double x) const {
|
|
double a = modelCoefficients[0];
|
|
double b = modelCoefficients[1];
|
|
if (derivateCoefficientIndex == 0) {
|
|
// Derivate with respect to a: pow(x,b)
|
|
return pow(x,b);
|
|
}
|
|
assert(derivateCoefficientIndex == 1);
|
|
assert(x >= 0);
|
|
/* We assume all xi are positive.
|
|
* For x = 0, a*pow(x,b) = 0, the partial derivate with respect to b is 0
|
|
* For x > 0, a*pow(x,b) = a*exp(b*ln(x)), the partial derivate with respect
|
|
* to b is ln(x)*a*pow(x,b) */
|
|
return x == 0.0 ? 0.0 : log(x) * a * pow(x, b);
|
|
}
|
|
|
|
void PowerModel::fit(Store * store, int series, double * modelCoefficients, Poincare::Context * context) {
|
|
/* Y1 = aX1^b => ln(Y1) = ln(a) + b*ln(X1)*/
|
|
modelCoefficients[0] = exp(store->yIntercept(series, true));
|
|
modelCoefficients[1] = store->slope(series, true);
|
|
}
|
|
|
|
bool PowerModel::dataSuitableForFit(Store * store, int series) const {
|
|
if (!Model::dataSuitableForFit(store, series)) {
|
|
return false;
|
|
}
|
|
int numberOfPairs = store->numberOfPairsOfSeries(series);
|
|
for (int j = 0; j < numberOfPairs; j++) {
|
|
if (store->get(series, 0, j) < 0) {
|
|
return false;
|
|
}
|
|
}
|
|
return true;
|
|
}
|
|
|
|
|
|
}
|