bool link_ode(
size_t size ,
size_t repeat ,
CppAD::vector<double> &x ,
CppAD::vector<double> &jacobian
)
{ // -------------------------------------------------------------
// setup
size_t n = size;
assert( x.size() == n );
size_t m = 0;
CppAD::vector<double> f(n);
while(repeat--)
{ // choose next x value
uniform_01(n, x);
// evaluate function
CppAD::ode_evaluate(x, m, f);
}
size_t i;
for(i = 0; i < n; i++)
jacobian[i] = f[i];
return true;
}
void forest::train(const vector<vector<float>>& x, const vector<float>& y, const size_t num_trees, const size_t mtry, const size_t seed)
{
// Initialize
resize(num_trees);
rng.seed(seed);
u01 = [&]()
{
return uniform_01(rng);
};
// Aggregate the statistics over all trees of the random forest
const size_t num_samples = x.size();
const size_t num_variables = x.front().size();
incPurity.resize(num_variables, 0);
incMSE.resize(num_variables, 0);
impSD.resize(num_variables, 0);
vector<float> oobPreds(num_samples, 0);
vector<size_t> oobTimes(num_samples, 0);
for (auto& t : *this)
{
t.train(x, y, mtry, u01, incPurity, incMSE, impSD, oobPreds, oobTimes);
}
// Normalize incPurity, incMSE, impSD
for (size_t i = 0; i < num_variables; ++i)
{
incPurity[i] /= size();
incMSE[i] /= size();
impSD[i] = sqrt((impSD[i] / size() - incMSE[i] * incMSE[i]) / size());
}
// Normalize oobPreds to calculate MSE
mse = 0;
size_t jout = 0;
for (size_t s = 0; s < num_samples; ++s)
{
if (!oobTimes[s]) continue;
const auto p = oobPreds[s] / oobTimes[s];
mse += (p - y[s]) * (p - y[s]);
++jout;
}
mse /= jout;
// Calculate rsq
float yAvg = 0;
for (size_t i = 0; i < num_samples; ++i)
{
yAvg += y[i];
}
yAvg /= num_samples;
float yVar = 0;
for (size_t i = 0; i < num_samples; ++i)
{
yVar += (y[i] - yAvg) * (y[i] - yAvg);
}
rsq = 1 - mse * num_samples / yVar;
}
bool link_ode(
size_t size ,
size_t repeat ,
CppAD::vector<double> &x ,
CppAD::vector<double> &jacobian
)
{
// speed test global option values
if( global_option["atomic"] )
return false;
// optimization options: no conditional skips or compare operators
std::string options="no_compare_op";
// --------------------------------------------------------------------
// setup
assert( x.size() == size );
assert( jacobian.size() == size * size );
typedef CppAD::AD<double> ADScalar;
typedef CppAD::vector<ADScalar> ADVector;
size_t j;
size_t p = 0; // use ode to calculate function values
size_t n = size; // number of independent variables
size_t m = n; // number of dependent variables
ADVector X(n), Y(m); // independent and dependent variables
CppAD::ADFun<double> f; // AD function
// -------------------------------------------------------------
if( ! global_option["onetape"] ) while(repeat--)
{ // choose next x value
uniform_01(n, x);
for(j = 0; j < n; j++)
X[j] = x[j];
// declare the independent variable vector
Independent(X);
// evaluate function
CppAD::ode_evaluate(X, p, Y);
// create function object f : X -> Y
f.Dependent(X, Y);
if( global_option["optimize"] )
f.optimize(options);
// skip comparison operators
f.compare_change_count(0);
jacobian = f.Jacobian(x);
}
else
{ // an x value
uniform_01(n, x);
for(j = 0; j < n; j++)
X[j] = x[j];
// declare the independent variable vector
Independent(X);
// evaluate function
CppAD::ode_evaluate(X, p, Y);
// create function object f : X -> Y
f.Dependent(X, Y);
if( global_option["optimize"] )
f.optimize(options);
// skip comparison operators
f.compare_change_count(0);
while(repeat--)
{ // get next argument value
uniform_01(n, x);
// evaluate jacobian
jacobian = f.Jacobian(x);
}
}
return true;
}