forked from mnbvc188199/Warrior_EA
255 lines
20 KiB
MQL5
255 lines
20 KiB
MQL5
//+------------------------------------------------------------------+
| |
//| Study.mq5 |
| |
//| Copyright DNG® |
| |
//| https://www.mql5.com/en/users/dng |
| |
//+------------------------------------------------------------------+
| |
#property copyright "Copyright DNG®"
| |
#property link "https://www.mql5.com/en/users/dng"
| |
#property version "1.00"
| |
//+------------------------------------------------------------------+
| |
//| |
| |
//+------------------------------------------------------------------+
| |
#define Study
| |
#include "Trajectory.mqh"
| |
//+------------------------------------------------------------------+
| |
//| Input parameters |
| |
//+------------------------------------------------------------------+
| |
input int Iterations = 10000;
| |
//+------------------------------------------------------------------+
| |
//| |
| |
//+------------------------------------------------------------------+
| |
STrajectory Buffer[];
| |
CNet DOT;
| |
CNet Actor;
| |
CNet Critic;
| |
//---
| |
float dError;
| |
datetime dtStudied;
| |
//---
| |
CBufferFloat bState;
| |
CBufferFloat bAccount;
| |
CBufferFloat bGradient;
| |
CBufferFloat bProbs;
| |
CBufferFloat *Result;
| |
vector<float> check;
| |
vector<float> STD_Actor;
| |
vector<float> STD_Goal;
| |
//---
| |
COpenCLMy *OpenCL;
| |
//+------------------------------------------------------------------+
| |
//| Expert initialization function |
| |
//+------------------------------------------------------------------+
| |
int OnInit()
| |
{
| |
//---
| |
ResetLastError();
| |
if(!LoadTotalBase())
| |
{
| |
PrintFormat("Error of load study data: %d", GetLastError());
| |
return INIT_FAILED;
| |
}
| |
//--- load models
| |
float temp;
| |
if(!DOT.Load(FileName + "DOT.nnw", temp, temp, temp, dtStudied, true) ||
| |
!Actor.Load(FileName + "Act.nnw", temp, temp, temp, dtStudied, true) ||
| |
!Critic.Load(FileName + "Crt.nnw", temp, temp, temp, dtStudied, true)
| |
)
| |
{
| |
CArrayObj *dot = new CArrayObj();
| |
CArrayObj *actor = new CArrayObj();
| |
CArrayObj *critic = new CArrayObj();
| |
if(!CreateDescriptions(dot, actor, critic))
| |
{
| |
delete dot;
| |
delete actor;
| |
delete critic;
| |
return INIT_FAILED;
| |
}
| |
if(!DOT.Create(dot) ||
| |
!Actor.Create(actor) ||
| |
!Critic.Create(critic))
| |
{
| |
delete dot;
| |
delete actor;
| |
delete critic;
| |
return INIT_FAILED;
| |
}
| |
delete dot;
| |
delete actor;
| |
delete critic;
| |
}
| |
//---
| |
OpenCL = DOT.GetOpenCL();
| |
Actor.SetOpenCL(OpenCL);
| |
Critic.SetOpenCL(OpenCL);
| |
//---
| |
Actor.getResults(Result);
| |
if(Result.Total() != NActions)
| |
{
| |
PrintFormat("The scope of the actor does not match the actions count (%d <> %d)", NActions, Result.Total());
| |
return INIT_FAILED;
| |
}
| |
//---
| |
DOT.GetLayerOutput(0, Result);
| |
if(Result.Total() != (HistoryBars * BarDescr))
| |
{
| |
PrintFormat("Input size of Encoder doesn't match state description (%d <> %d)", Result.Total(), (HistoryBars * BarDescr));
| |
return INIT_FAILED;
| |
}
| |
//---
| |
if(!bGradient.BufferInit(MathMax(AccountDescr, NForecast), 0) ||
| |
!bGradient.BufferCreate(OpenCL))
| |
{
| |
PrintFormat("Error of create buffers: %d", GetLastError());
| |
return INIT_FAILED;
| |
}
| |
//---
| |
if(!EventChartCustom(ChartID(), 1, 0, 0, "Init"))
| |
{
| |
PrintFormat("Error of create study event: %d", GetLastError());
| |
return INIT_FAILED;
| |
}
| |
//---
| |
return(INIT_SUCCEEDED);
| |
}
| |
//+------------------------------------------------------------------+
| |
//| Expert deinitialization function |
| |
//+------------------------------------------------------------------+
| |
void OnDeinit(const int reason)
| |
{
| |
//---
| |
if(!(reason == REASON_INITFAILED || reason == REASON_RECOMPILE))
| |
{
| |
Actor.Save(FileName + "Act.nnw", 0, 0, 0, TimeCurrent(), true);
| |
DOT.Save(FileName + "DOT.nnw", 0, 0, 0, TimeCurrent(), true);
| |
Critic.Save(FileName + "Crt.nnw", 0, 0, 0, TimeCurrent(), true);
| |
}
| |
delete Result;
| |
delete OpenCL;
| |
}
| |
//+------------------------------------------------------------------+
| |
//| ChartEvent function |
| |
//+------------------------------------------------------------------+
| |
void OnChartEvent(const int id,
| |
const long &lparam,
| |
const double &dparam,
| |
const string &sparam)
| |
{
| |
//---
| |
if(id == 1001)
| |
Train();
| |
}
| |
//+------------------------------------------------------------------+
| |
//| Train function |
| |
//+------------------------------------------------------------------+
| |
void Train(void)
| |
{
| |
//---
| |
vector<float> probability = GetProbTrajectories(Buffer, 0.9);
| |
//---
| |
vector<float> result, target;
| |
bool Stop = false;
| |
//---
| |
uint ticks = GetTickCount();
| |
//---
| |
for(int iter = 0; (iter < Iterations && !IsStopped() && !Stop); iter ++)
| |
{
| |
int tr = SampleTrajectory(probability);
| |
int batch = GPTBars + 48;
| |
int state = (int)((MathRand() * MathRand() / MathPow(32767, 2)) * (Buffer[tr].Total - 2 - PrecoderBars - batch));
| |
if(state <= 0)
| |
{
| |
iter--;
| |
continue;
| |
}
| |
DOT.Clear();
| |
int end = MathMin(state + batch, Buffer[tr].Total - PrecoderBars);
| |
for(int i = state; i < end; i++)
| |
{
| |
bState.AssignArray(Buffer[tr].States[i].state);
| |
//--- Trajectory
| |
if(!DOT.feedForward((CBufferFloat*)GetPointer(bState), 1, false, (CBufferFloat*)NULL))
| |
{
| |
PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
| |
Stop = true;
| |
break;
| |
}
| |
//--- Policy
| |
float PrevBalance = Buffer[tr].States[MathMax(i - 1, 0)].account[0];
| |
float PrevEquity = Buffer[tr].States[MathMax(i - 1, 0)].account[1];
| |
bAccount.Clear();
| |
bAccount.Add((Buffer[tr].States[i].account[0] - PrevBalance) / PrevBalance);
| |
bAccount.Add(Buffer[tr].States[i].account[1] / PrevBalance);
| |
bAccount.Add((Buffer[tr].States[i].account[1] - PrevEquity) / PrevEquity);
| |
bAccount.Add(Buffer[tr].States[i].account[2]);
| |
bAccount.Add(Buffer[tr].States[i].account[3]);
| |
bAccount.Add(Buffer[tr].States[i].account[4] / PrevBalance);
| |
bAccount.Add(Buffer[tr].States[i].account[5] / PrevBalance);
| |
bAccount.Add(Buffer[tr].States[i].account[6] / PrevBalance);
| |
double time = (double)Buffer[tr].States[i].account[7];
| |
double x = time / (double)(D'2024.01.01' - D'2023.01.01');
| |
bAccount.Add((float)MathSin(x != 0 ? 2.0 * M_PI * x : 0));
| |
x = time / (double)PeriodSeconds(PERIOD_MN1);
| |
bAccount.Add((float)MathCos(x != 0 ? 2.0 * M_PI * x : 0));
| |
x = time / (double)PeriodSeconds(PERIOD_W1);
| |
bAccount.Add((float)MathSin(x != 0 ? 2.0 * M_PI * x : 0));
| |
x = time / (double)PeriodSeconds(PERIOD_D1);
| |
bAccount.Add((float)MathSin(x != 0 ? 2.0 * M_PI * x : 0));
| |
if(bAccount.GetIndex() >= 0)
| |
bAccount.BufferWrite();
| |
//--- Actor
| |
if(!Actor.feedForward((CNet *)GetPointer(DOT), -1, (CBufferFloat*)GetPointer(bAccount)))
| |
{
| |
PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
| |
Stop = true;
| |
break;
| |
}
| |
//--- Critic
| |
if(!Critic.feedForward((CNet *)GetPointer(DOT), -1, (CNet*)GetPointer(Actor)))
| |
{
| |
PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
| |
Stop = true;
| |
break;
| |
}
| |
Result.AssignArray(Buffer[tr].States[i].action);
| |
if(!Actor.backProp(Result, (CBufferFloat *)GetPointer(bAccount), (CBufferFloat *)GetPointer(bGradient)) ||
| |
!DOT.backPropGradient((CBufferFloat*)NULL)
| |
)
| |
{
| |
PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
| |
Stop = true;
| |
break;
| |
}
| |
result.Assign(Buffer[tr].States[i+1].rewards);
| |
target.Assign(Buffer[tr].States[i+2].rewards);
| |
result=result-target*DiscFactor;
| |
Result.AssignArray(result);
| |
if(!Critic.backProp(Result, (CNet *)GetPointer(Actor)) ||
| |
!DOT.backPropGradient((CBufferFloat*)NULL) ||
| |
!Actor.backPropGradient((CBufferFloat *)GetPointer(bAccount), (CBufferFloat *)GetPointer(bGradient)) ||
| |
!DOT.backPropGradient((CBufferFloat*)NULL)
| |
)
| |
{
| |
PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
| |
Stop = true;
| |
break;
| |
}
| |
//---
| |
if(GetTickCount() - ticks > 500)
| |
{
| |
double percent = (double(i - state) / ((end - state)) + iter) * 100.0 / (Iterations);
| |
string str = StringFormat("%-14s %6.2f%% -> Error %15.8f\n", "Actor", percent, Actor.getRecentAverageError());
| |
str += StringFormat("%-14s %6.2f%% -> Error %15.8f\n", "Critic", percent, Critic.getRecentAverageError());
| |
Comment(str);
| |
ticks = GetTickCount();
| |
}
| |
}
| |
}
| |
Comment("");
| |
//---
| |
PrintFormat("%s -> %d -> %-15s %10.7f", __FUNCTION__, __LINE__, "Actor", Actor.getRecentAverageError());
| |
PrintFormat("%s -> %d -> %-15s %10.7f", __FUNCTION__, __LINE__, "Critic", Critic.getRecentAverageError());
| |
ExpertRemove();
| |
//---
| |
}
| |
//+------------------------------------------------------------------+
|