Files
GeneticAlgo/inc/genetic.h

371 lines
10 KiB
C++

#pragma once
#include <algorithm>
#include <cstdlib>
#include <vector>
#include "sync.h"
#include "rand.h"
using namespace sync;
using namespace std;
namespace genetic {
template <class T> struct Array;
template <class T> struct Stats;
template <class T> struct Strategy;
struct CellTracker;
template <class T> Stats<T> run(Strategy<T>);
template <class T> struct Strategy {
// Number of worker threads that will be evaluating cell fitness
int num_threads;
int batch_size; // Number of cells a worker thread tries to work on in a row
// before accessing/locking the work queue again.
int num_cells; // Size of the population pool
int num_generations; // Number of times (epochs) to run the algorithm
bool test_all; // Sets whether or not every cell's fitness is evaluated every
// generation
float test_chance; // Chance to test any given cell's fitness. Relevant only
// if test_all is false.
bool enable_crossover; // Cells that score well in the evaluation stage
// produce children that replace low-scoring cells
int crossover_parent_num; // Number of unique high-scoring parents in a
// crossover call.
int crossover_parent_stride; // Number of parents to skip over when moving to
// the next set of parents. A stride of 1 would
// produce maximum overlap because the set of
// parents would only change by one every
// crossover.
int crossover_children_num; // Number of children to expect the user to
// produce in the crossover function.
bool enable_mutation; // Cells may be mutated
// before fitness evaluation
float mutation_chance; // Chance for any given cell to be mutated cells during
// the mutation
uint64_t rand_seed;
bool higher_fitness_is_better; // Sets whether or not to consider higher
// fitness values better or worse. Set this to
// false if fitness is an error function.
// User defined functions
T (*make_default_cell)();
void (*mutate)(T &cell_to_modify);
void (*crossover)(const Array<T *> parents, const Array<T *> out_children);
float (*fitness)(const T &cell);
};
template<class T> struct Stats {
std::vector<T> best_cell;
std::vector<float> best_cell_fitness;
TimeSpan setup_time;
TimeSpan run_time;
};
struct CellTracker {
float score;
int cellid;
};
template <class T> struct Array {
T *data;
int len;
T &operator[](int i) { return data[i]; }
};
template <class T> Array<T> make_array(int len) {
return {
.data = (T*)malloc(sizeof(T)*len),
.len = len
};
}
template<class T>
struct MutateJob {
T* cell;
};
template<class T>
struct CrossoverJob {
Array<T*> parents;
Array<T*> children;
};
template<class T>
struct FitnessJob {
T* cell;
CellTracker* track;
};
enum class JobType {
MUTATE,
CROSSOVER,
FITNESS
};
template<class T>
union Job {
MutateJob<T> m;
CrossoverJob<T> c;
FitnessJob<T> f;
};
// Yes. I am aware of variant
// For some reason I like this better
template<class T>
struct TaggedJob {
Job<T> data;
JobType type;
};
template<class T>
struct WorkQueue {
Array<TaggedJob<T>> jobs;
int read_i, write_i, batch_size;
bool done_writing, work_complete, stop; // These catch some edge conditions
Mutex m;
ConditionVar done;
ConditionVar jobs_ready;
};
template<class T>
struct WorkerThreadArgs {
WorkQueue<T> &q;
Strategy<T> &s;
};
template<class T> WorkQueue<T> make_work_queue(int len, int batch_size);
template<class T> bool tryget_job_batch(WorkQueue<T> &q, int len, Array<TaggedJob<T>>* out_batch, bool* out_batch_is_end);
template<class T>
DWORD worker(LPVOID args);
template <class T> Stats<T> run(Strategy<T> strat) {
Stats<T> stats;
// ************* SETUP **************
TimeSpan start_setup = now();
// Create cells
Array<T> cells = make_array<T>(strat.num_cells);
for (int i = 0; i < cells.len; i++) cells[i] = strat.make_default_cell();
// Create cell trackers
Array<CellTracker> trackers = make_array<CellTracker>(strat.num_cells);
for (int i = 0; i < trackers.len; i++) trackers[i] = { .score=0, .cellid=i };
// Create work queue
// Worst case size is every cell mutated, crossed, and evaluated...? Not quite, but 3x should be upper bound
WorkQueue<T> q = make_work_queue<T>(3*strat.num_cells, strat.batch_size);
WorkerThreadArgs<T> args = {q, strat};
// Create worker threads
Thread *threads = (Thread*)malloc(sizeof(Thread*)*strat.num_threads);
for (int i = 0; i < strat.num_threads; i++) {
threads[i] = make_thread(worker<T>, &args);
}
stats.setup_time = now() - start_setup;
// *********** ALGORITHM ************
TimeSpan start_algo = now();
for (int gen = 0; gen < strat.num_generations; gen++) {
// Reset work queue
lock(q.m);
q.read_i = 0;
q.write_i = 0;
q.work_complete = false;
q.done_writing = false;
unlock(q.m);
// 1. mutate
for (int i = 0; i < trackers.len; i++) {
if (abs(norm_rand(strat.rand_seed)) < strat.mutation_chance) {
MutateJob<T> mj = {&cells[trackers[i].cellid]};
TaggedJob<T> job;
job.data.m = mj;
job.type=JobType::MUTATE;
q.jobs[q.write_i++] = job;
}
}
wake_all(q.jobs_ready); // There are available jobs for the worker threads!
// 2. crossover
if (strat.enable_crossover) {
int npar = strat.crossover_parent_num;
int nchild = strat.crossover_children_num;
int parent_end = npar;
int child_begin = trackers.len-nchild;
while (parent_end <= child_begin) {
// TODO: Variable size arrays please. This is rediculous.
Array<T*> parents = make_array<T*>(npar);
Array<T*> children = make_array<T*>(nchild);
// Get pointers to all the parent cells
for (int i = parent_end-npar; i < parent_end; i++) {
parents[i - (parent_end-npar)] = &cells[trackers[i].cellid];
}
// Get pointers to all the child cells (these will be overwritten)
for (int i = child_begin; i < child_begin+nchild; i++) {
children[i-child_begin] = &cells[trackers[i].cellid];
}
CrossoverJob<T> cj = {parents, children};
TaggedJob<T> job;
job.data.c=cj;
job.type=JobType::CROSSOVER;
q.jobs[q.write_i++] = job;
parent_end += strat.crossover_parent_stride;
child_begin -= nchild;
}
wake_all(q.jobs_ready); // There are available jobs for the worker threads!
}
// 3. evaluate
if (strat.test_all) {
for (int i = 0; i < trackers.len; i++) {
FitnessJob<T> fj = {&cells[trackers[i].cellid], &trackers[i]};
TaggedJob<T> job;
job.data.f=fj;
job.type=JobType::FITNESS;
if (i == trackers.len-1) lock(q.m);
q.jobs[q.write_i++] = job;
if (i == trackers.len-1) { q.done_writing = true; unlock(q.m); }
}
} else {
lock(q.m);
for (int i = 0; i < trackers.len; i++) {
if (abs(norm_rand(strat.rand_seed)) < strat.test_chance) {
FitnessJob<T> fj = {&cells[trackers[i].cellid], &trackers[i]};
TaggedJob<T> job;
job.data.f=fj;
job.type=JobType::FITNESS;
q.jobs[q.write_i++] = job;
}
}
q.done_writing = true;
unlock(q.m);
}
wake_all(q.jobs_ready);
// Wait until the work is finished
lock(q.m);
if (!q.work_complete)
wait(q.done, q.m, infinite_ts);
unlock(q.m);
// 4. sort
std::sort(&trackers[0], &trackers[trackers.len-1], [strat](CellTracker &a, CellTracker &b){ return strat.higher_fitness_is_better ? a.score > b.score : a.score < b.score; });
printf("Gen: %d, Best Score: %f\n", gen, trackers[0].score);
stats.best_cell.push_back(cells[trackers[0].cellid]);
stats.best_cell_fitness.push_back(trackers[0].score);
}
q.stop = true;
wake_all(q.jobs_ready);
// TODO: join all threads
// TODO: There's some data freeing that should really be done here
stats.run_time = now() - start_algo;
return stats;
}
template<class T> WorkQueue<T> make_work_queue(int len, int batch_size) {
return {
.jobs=make_array<TaggedJob<T>>(len),
.read_i=0,
.write_i=0,
.batch_size=batch_size,
.done_writing=false,
.work_complete=false,
.m=make_mutex(),
.done=make_condition_var(),
.jobs_ready=make_condition_var()
};
}
template<class T> bool tryget_job_batch(WorkQueue<T> &q, Array<TaggedJob<T>>* out_batch, bool* out_batch_is_end) {
lock(q.m);
if (q.stop) {
unlock(q.m);
return false;
}
// Keep waiting till jobs are available
while (q.read_i >= q.write_i) {
wait(q.jobs_ready, q.m, infinite_ts);
if (q.stop) {
unlock(q.m);
return false;
}
}
// Yay! Let's grab some jobs to do
// If the batch we're about to grab moves read_i to write_i and the producer
// is done writing, we should let our callee know it's handling this gen's last
// batch know that way it sets work_complete and signals done.
*out_batch_is_end = q.done_writing && q.read_i + q.batch_size >= q.write_i;
out_batch->data = &q.jobs[q.read_i];
out_batch->len = min(q.batch_size, q.write_i - q.read_i);
q.read_i += q.batch_size;
unlock(q.m);
return true;
}
template<class T>
void work_batch(Array<TaggedJob<T>> batch, Strategy<T> &s) {
for (int i = 0; i < batch.len; i++) {
switch (batch[i].type) {
case JobType::MUTATE: {
MutateJob<T> mj = batch[i].data.m;
s.mutate(*mj.cell);
} break;
case JobType::CROSSOVER: {
CrossoverJob<T> cj = batch[i].data.c;
s.crossover(cj.parents, cj.children);
} break;
case JobType::FITNESS: {
FitnessJob<T> fj = batch[i].data.f;
fj.track->score = s.fitness(*fj.cell);
} break;
default: {
assert(false);
}
}
}
}
template<class T>
DWORD worker(LPVOID args) {
WorkerThreadArgs<T>* wa = static_cast<WorkerThreadArgs<T>*>(args);
WorkQueue<T> &q = wa->q;
Strategy<T> &s = wa->s;
// These are written by tryget_job_batch
bool batch_is_end;
Array<TaggedJob<T>> batch;
while (tryget_job_batch(q, &batch, &batch_is_end)) {
work_batch(batch, s);
if (batch_is_end) {
lock(q.m);
q.work_complete = true;
wake_one(q.done);
unlock(q.m);
}
}
return NULL;
}
} // namespace genetic