Skip to content

D
dlib
Commit 5d259cd7 authored by Davis King's avatar Davis King
Browse files
Options

merged

parents f9f69185 532e2a3e
Hide whitespace changes
Inline Side-by-side
Showing with 42 additions and 29 deletions
dlib/global_optimization/global_function_search.cpp
View file @ 5d259cd7
......@@ -3,7 +3,6 @@
#include "upper_bound_function.h"
#include "../optimization.h"
#include "../timing.h" // TODO, remove
namespace dlib
{
......@@ -205,7 +204,6 @@ namespace dlib
const std::vector<bool>& is_integer_variable
)
{
timing::block oaijsdofijas(1, "pick_next_sample_quad_interp");
DLIB_CASSERT(samples.size() > 0);
// We don't use the QP to optimize integer variables. Instead, we just fix them at
// their best observed value and use the QP to optimize the real variables. So the
......@@ -324,21 +322,16 @@ namespace dlib
max_upper_bound_function pick_next_sample_max_upper_bound_function (
dlib::rand& rnd,
const std::vector<function_evaluation>& samples,
const upper_bound_function& ub,
const matrix<double,0,1>& lower,
const matrix<double,0,1>& upper,
const std::vector<bool>& is_integer_variable,
const double relative_noise_magnitude = 0.001,
const size_t num_random_samples = 5000
const size_t num_random_samples
)
{
timing::block oaijsdofijas(0, "pick_next_sample_max_upper_bound_function");
DLIB_CASSERT(samples.size() > 0);
// TODO, assert everyone has same dims
DLIB_CASSERT(ub.num_points() > 0);
// build the upper bound
upper_bound_function ub(samples, relative_noise_magnitude);
// now do a simple random search to find the maximum upper bound
double best_ub_so_far = -std::numeric_limits<double>::infinity();
......@@ -356,7 +349,7 @@ namespace dlib
}
double max_value = -std::numeric_limits<double>::infinity();
for (auto& v : samples)
for (auto& v : ub.get_points())
max_value = std::max(max_value, v.y);
return max_upper_bound_function(v, best_ub_so_far - max_value, best_ub_so_far);
......@@ -404,18 +397,23 @@ namespace dlib
namespace gopt_impl
{
std::vector<function_evaluation> funct_info::all_function_evals (
upper_bound_function funct_info::build_upper_bound_with_all_function_evals (
) const
{
auto temp = complete_evals;
temp.reserve(temp.size()+incomplete_evals.size());
upper_bound_function tmp(ub);
// we are going to add the incomplete evals into this and assume the
// incomplete evals are going to take y values equal to their nearest
// neighbor complete evals.
for (auto& eval : incomplete_evals)
temp.emplace_back(eval.x, find_nn(complete_evals, eval.x));
{
function_evaluation e;
e.x = eval.x;
e.y = find_nn(ub.get_points(), eval.x);
tmp.add(e);
}
return temp;
return tmp;
}
double funct_info::find_nn (
......@@ -521,7 +519,7 @@ namespace dlib
auto i = std::find(info->incomplete_evals.begin(), info->incomplete_evals.end(), req);
DLIB_CASSERT(i != info->incomplete_evals.end());
info->incomplete_evals.erase(i);
info->complete_evals.emplace_back(req.x,y);
info->ub.add(function_evaluation(req.x,y));
// Now do trust region radius maintenance and keep track of the best objective
......@@ -532,6 +530,8 @@ namespace dlib
// was.
double measured_improvement = y-req.anchor_objective_value;
double rho = measured_improvement/std::abs(req.predicted_improvement);
std::cout << "rho: "<< rho << std::endl;
std::cout << "radius: "<< info->radius << std::endl;
if (rho < 0.25)
info->radius *= 0.5;
else if (rho > 0.75)
......@@ -540,8 +540,9 @@ namespace dlib
if (y > info->best_objective_value)
{
if (length(req.x - info->best_x) > info->radius*1.001)
if (!req.was_trust_region_generated_request && length(req.x - info->best_x) > info->radius*1.001)
{
std::cout << "reset radius because of big move, " << length(req.x - info->best_x) << " radius was " << info->radius << std::endl;
// reset trust region radius since we made a big move. Doing this will
// cause the radius to be reset to the size of the local region.
info->radius = 0;
......@@ -578,7 +579,7 @@ namespace dlib
DLIB_CASSERT(functions_.size() == initial_function_evals.size());
for (size_t i = 0; i < initial_function_evals.size(); ++i)
{
functions[i]->complete_evals = initial_function_evals[i];
functions[i]->ub = upper_bound_function(initial_function_evals[i]);
}
}
......@@ -609,7 +610,7 @@ namespace dlib
for (size_t i = 0; i < functions.size(); ++i)
{
specs.emplace_back(functions[i]->spec);
function_evals.emplace_back(functions[i]->complete_evals);
function_evals.emplace_back(functions[i]->ub.get_points());
}
}
......@@ -645,7 +646,7 @@ namespace dlib
for (auto& info : functions)
{
const long dims = info->spec.lower.size();
if (info->complete_evals.size() < std::max<long>(3,dims))
if (info->ub.num_points() < std::max<long>(3,dims))
{
outstanding_function_eval_request new_req;
new_req.request_id = next_request_id++;
......@@ -663,10 +664,11 @@ namespace dlib
auto info = best_function();
const long dims = info->spec.lower.size();
// if we have enough points to do a trust region step
if (info->complete_evals.size() > dims+1)
if (info->ub.num_points() > dims+1)
{
auto tmp = pick_next_sample_quad_interp(info->complete_evals,
auto tmp = pick_next_sample_quad_interp(info->ub.get_points(),
info->radius, info->spec.lower, info->spec.upper, info->spec.is_integer_variable);
std::cout << "QP predicted improvement: "<< tmp.predicted_improvement << std::endl;
if (tmp.predicted_improvement > qp_eps)
{
do_trust_region_step = false;
......@@ -696,8 +698,8 @@ namespace dlib
for (auto& info : functions)
{
auto tmp = pick_next_sample_max_upper_bound_function(rnd,
info->all_function_evals(), info->spec.lower, info->spec.upper,
info->spec.is_integer_variable, relative_noise_magnitude, num_random_samples);
info->build_upper_bound_with_all_function_evals(), info->spec.lower, info->spec.upper,
info->spec.is_integer_variable, num_random_samples);
if (tmp.predicted_improvement > 0 && tmp.upper_bound > best_upper_bound)
{
best_upper_bound = tmp.upper_bound;
......@@ -757,7 +759,11 @@ namespace dlib
double global_function_search::
get_relative_noise_magnitude (
) const { return relative_noise_magnitude; }
) const
{
return relative_noise_magnitude;
}
void global_function_search::
set_relative_noise_magnitude (
double value
......@@ -765,11 +771,18 @@ namespace dlib
{
DLIB_CASSERT(0 <= value);
relative_noise_magnitude = value;
// recreate all the upper bound functions with the new relative noise magnitude
for (auto& f : functions)
f->ub = upper_bound_function(f->ub.get_points(), relative_noise_magnitude);
}
size_t global_function_search::
get_monte_carlo_upper_bound_sample_num (
) const { return num_random_samples; }
) const
{
return num_random_samples;
}
void global_function_search::
set_monte_carlo_upper_bound_sample_num (
size_t num
......
dlib/global_optimization/global_function_search.h
View file @ 5d259cd7
......@@ -53,7 +53,7 @@ namespace dlib
best_x = zeros_matrix(spec.lower);
}
std::vector<function_evaluation> all_function_evals (
upper_bound_function build_upper_bound_with_all_function_evals (
) const;
static double find_nn (
......@@ -65,7 +65,7 @@ namespace dlib
function_spec spec;
size_t function_idx = 0;
std::shared_ptr<std::mutex> m;
std::vector<function_evaluation> complete_evals;
upper_bound_function ub;
std::vector<outstanding_function_eval_request> incomplete_evals;
matrix<double,0,1> best_x;
double best_objective_value = -std::numeric_limits<double>::infinity();
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment