408 lignes
32 Kio
MQL5
408 lignes
32 Kio
MQL5
//+------------------------------------------------------------------+
|
|
//| Study.mq5 |
|
|
//| Copyright DNG® |
|
|
//| https://www.mql5.com/ru/users/dng |
|
|
//+------------------------------------------------------------------+
|
|
#property copyright "Copyright DNG®"
|
|
#property link "https://www.mql5.com/ru/users/dng"
|
|
#property version "1.00"
|
|
//+------------------------------------------------------------------+
|
|
//| |
|
|
//+------------------------------------------------------------------+
|
|
#define Study
|
|
#define Critic
|
|
#include "Trajectory.mqh"
|
|
//+------------------------------------------------------------------+
|
|
//| Input parameters |
|
|
//+------------------------------------------------------------------+
|
|
input int Iterations = 100000;
|
|
input int Batch = 60;
|
|
//+------------------------------------------------------------------+
|
|
//| |
|
|
//+------------------------------------------------------------------+
|
|
STrajectory Buffer[];
|
|
CNet Actor;
|
|
CNet Probability;
|
|
//---
|
|
float dError;
|
|
datetime dtStudied;
|
|
//---
|
|
CBufferFloat bState;
|
|
CBufferFloat bTime;
|
|
CBufferFloat bAccount;
|
|
CBufferFloat bGradient;
|
|
CBufferFloat bActions;
|
|
CBufferFloat *Result;
|
|
vector<float> check;
|
|
vector<float> STD_Actor;
|
|
vector<float> STD_Goal;
|
|
//+------------------------------------------------------------------+
|
|
//| Expert initialization function |
|
|
//+------------------------------------------------------------------+
|
|
int OnInit()
|
|
{
|
|
//---
|
|
ResetLastError();
|
|
if(!LoadTotalBase())
|
|
{
|
|
PrintFormat("Error of load study data: %d", GetLastError());
|
|
return INIT_FAILED;
|
|
}
|
|
//--- load models
|
|
float temp;
|
|
CArrayObj *actor = new CArrayObj();
|
|
CArrayObj *probability = new CArrayObj();
|
|
if(!CreateDescriptions(actor, probability))
|
|
{
|
|
delete actor;
|
|
delete probability;
|
|
return INIT_FAILED;
|
|
}
|
|
if(!Actor.Load(FileName + "Act.nnw", temp, temp, temp, dtStudied, true))
|
|
{
|
|
Print("Create new Actor");
|
|
if(!Actor.Create(actor))
|
|
{
|
|
delete actor;
|
|
delete probability;
|
|
return INIT_FAILED;
|
|
}
|
|
}
|
|
if(!Probability.Load(FileName + "Prb.nnw", temp, temp, temp, dtStudied, true))
|
|
{
|
|
Print("Create new Probability model");
|
|
if(!Probability.Create(probability))
|
|
{
|
|
delete actor;
|
|
delete probability;
|
|
return INIT_FAILED;
|
|
}
|
|
}
|
|
delete actor;
|
|
delete probability;
|
|
//---
|
|
Actor.TrainMode(true);
|
|
Probability.TrainMode(true);
|
|
Probability.SetOpenCL(Actor.GetOpenCL());
|
|
//---
|
|
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;
|
|
}
|
|
//---
|
|
Actor.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, NActions), 0) ||
|
|
!bGradient.BufferCreate(Actor.GetOpenCL()))
|
|
{
|
|
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);
|
|
Probability.Save(FileName + "Prb.nnw", 0, 0, 0, TimeCurrent(), true);
|
|
}
|
|
delete Result;
|
|
}
|
|
//+------------------------------------------------------------------+
|
|
//| 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 = vector<float>::Full(Buffer.Size(), 1.0f / Buffer.Size());
|
|
//---
|
|
vector<float> result, target, state;
|
|
matrix<float> fstate = matrix<float>::Zeros(1, NForecast * BarDescr);
|
|
bool Stop = false;
|
|
//---
|
|
uint ticks = GetTickCount();
|
|
//---
|
|
for(int iter = 0; (iter < Iterations && !IsStopped() && !Stop); iter += Batch)
|
|
{
|
|
int tr = SampleTrajectory(probability);
|
|
int start = (int)((MathRand() * MathRand() / MathPow(32767, 2)) * (Buffer[tr].Total - 2 - NForecast - Batch));
|
|
if(start <= 0)
|
|
{
|
|
iter -= Batch;
|
|
continue;
|
|
}
|
|
if(!Actor.Clear())
|
|
{
|
|
PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
|
Stop = true;
|
|
break;
|
|
}
|
|
result = vector<float>::Zeros(NActions);
|
|
for(int i = start; i < MathMin(Buffer[tr].Total, start + Batch); i++)
|
|
{
|
|
if(!state.Assign(Buffer[tr].States[i].state) ||
|
|
MathAbs(state).Sum() == 0 ||
|
|
!bState.AssignArray(state))
|
|
{
|
|
iter -= Batch + start - i;
|
|
break;
|
|
}
|
|
//---
|
|
bTime.Clear();
|
|
double time = (double)Buffer[tr].States[i].account[7];
|
|
double x = time / (double)(D'2024.01.01' - D'2023.01.01');
|
|
bTime.Add((float)MathSin(x != 0 ? 2.0 * M_PI * x : 0));
|
|
x = time / (double)PeriodSeconds(PERIOD_MN1);
|
|
bTime.Add((float)MathCos(x != 0 ? 2.0 * M_PI * x : 0));
|
|
x = time / (double)PeriodSeconds(PERIOD_W1);
|
|
bTime.Add((float)MathSin(x != 0 ? 2.0 * M_PI * x : 0));
|
|
x = time / (double)PeriodSeconds(PERIOD_D1);
|
|
bTime.Add((float)MathSin(x != 0 ? 2.0 * M_PI * x : 0));
|
|
if(bTime.GetIndex() >= 0)
|
|
bTime.BufferWrite();
|
|
//--- Account
|
|
float PrevBalance = Buffer[tr].States[MathMax(i - 1, 0)].account[0];
|
|
float PrevEquity = Buffer[tr].States[MathMax(i - 1, 0)].account[1];
|
|
float profit = float(bState[0] / _Point * (result[0] - result[3]));
|
|
bAccount.Clear();
|
|
bAccount.Add(1);
|
|
bAccount.Add((PrevEquity + profit) / PrevEquity);
|
|
bAccount.Add(profit / PrevEquity);
|
|
bAccount.Add(MathMax(result[0] - result[3], 0));
|
|
bAccount.Add(MathMax(result[3] - result[0], 0));
|
|
bAccount.Add((bAccount[3] > 0 ? profit / PrevEquity : 0));
|
|
bAccount.Add((bAccount[4] > 0 ? profit / PrevEquity : 0));
|
|
bAccount.Add(0);
|
|
bAccount.AddArray(GetPointer(bTime));
|
|
if(bAccount.GetIndex() >= 0)
|
|
bAccount.BufferWrite();
|
|
//--- Feed Forward
|
|
if(!Actor.feedForward((CBufferFloat*)GetPointer(bState), 1, false, GetPointer(bAccount)))
|
|
{
|
|
PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
|
Stop = true;
|
|
break;
|
|
}
|
|
if(!Probability.feedForward(GetPointer(Actor), LatentLayer, (CBufferFloat*)NULL))
|
|
{
|
|
PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
|
Stop = true;
|
|
break;
|
|
}
|
|
//--- Look for target
|
|
target = vector<float>::Zeros(NActions);
|
|
bActions.AssignArray(target);
|
|
if(!state.Assign(Buffer[tr].States[i + NForecast].state) ||
|
|
!state.Resize(NForecast * BarDescr) ||
|
|
MathAbs(state).Sum() == 0)
|
|
{
|
|
iter -= Batch + start - i;
|
|
break;
|
|
}
|
|
if(!fstate.Resize(1, NForecast * BarDescr) ||
|
|
!fstate.Row(state, 0) ||
|
|
!fstate.Reshape(NForecast, BarDescr))
|
|
{
|
|
iter -= Batch + start - i;
|
|
break;
|
|
}
|
|
for(int j = 0; j < NForecast / 2; j++)
|
|
{
|
|
if(!fstate.SwapRows(j, NForecast - j - 1))
|
|
{
|
|
PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
|
Stop = true;
|
|
break;
|
|
}
|
|
}
|
|
target = fstate.Col(0).CumSum();
|
|
if(result[0] > result[3])
|
|
{
|
|
float tp = 0;
|
|
float sl = 0;
|
|
float cur_sl = float(-(result[2] > 0 ? result[2] : 1) * MaxSL * Point());
|
|
int pos = 0;
|
|
for(int j = 0; j < NForecast; j++)
|
|
{
|
|
tp = MathMax(tp, target[j] + fstate[j, 1] - fstate[j, 0]);
|
|
pos = j;
|
|
if(cur_sl >= target[j] + fstate[j, 2] - fstate[j, 0])
|
|
break;
|
|
sl = MathMin(sl, target[j] + fstate[j, 2] - fstate[j, 0]);
|
|
}
|
|
if(tp > 0)
|
|
{
|
|
sl = (float)MathMax(MathMin(MathAbs(sl) / (MaxSL * Point()), 1), 0.01);
|
|
tp = float(MathMin(tp / (MaxTP * Point()), 1));
|
|
result[0] = MathMax(result[0] - result[3], 0.011f);
|
|
result[5] = result[1] = tp;
|
|
result[4] = result[2] = sl;
|
|
result[3] = 0;
|
|
bActions.AssignArray(result);
|
|
}
|
|
}
|
|
else
|
|
{
|
|
if(result[0] < result[3])
|
|
{
|
|
float tp = 0;
|
|
float sl = 0;
|
|
float cur_sl = float((result[5] > 0 ? result[5] : 1) * MaxSL * Point());
|
|
int pos = 0;
|
|
for(int j = 0; j < NForecast; j++)
|
|
{
|
|
tp = MathMin(tp, target[j] + fstate[j, 2] - fstate[j, 0]);
|
|
pos = j;
|
|
if(cur_sl <= target[j] + fstate[j, 1] - fstate[j, 0])
|
|
break;
|
|
sl = MathMax(sl, target[j] + fstate[j, 1] - fstate[j, 0]);
|
|
}
|
|
if(tp < 0)
|
|
{
|
|
sl = (float)MathMax(MathMin(MathAbs(sl) / (MaxSL * Point()), 1), 0.01);
|
|
tp = float(MathMin(-tp / (MaxTP * Point()), 1));
|
|
result[3] = MathMax(result[3] - result[0], 0.011f);
|
|
result[2] = result[4] = tp;
|
|
result[1] = result[5] = sl;
|
|
result[0] = 0;
|
|
bActions.AssignArray(result);
|
|
}
|
|
}
|
|
else
|
|
{
|
|
ulong argmin = target.ArgMin();
|
|
ulong argmax = target.ArgMax();
|
|
float max_sl = float(MaxSL * Point());
|
|
while(argmax > 0 && argmin > 0)
|
|
{
|
|
if(argmax < argmin && target[argmax] / 2 > MathAbs(target[argmin]) && MathAbs(target[argmin]) < max_sl)
|
|
break;
|
|
if(argmax > argmin && target[argmax] < MathAbs(target[argmin] / 2) && target[argmax] < max_sl)
|
|
break;
|
|
target.Resize(MathMin(argmax, argmin));
|
|
argmin = target.ArgMin();
|
|
argmax = target.ArgMax();
|
|
}
|
|
if(argmin == 0 || (argmax < argmin && argmax > 0))
|
|
{
|
|
float tp = 0;
|
|
float sl = 0;
|
|
float cur_sl = - float(MaxSL * Point());
|
|
ulong pos = 0;
|
|
for(ulong j = 0; j < argmax; j++)
|
|
{
|
|
tp = MathMax(tp, target[j] + fstate[j, 1] - fstate[j, 0]);
|
|
pos = j;
|
|
if(cur_sl >= target[j] + fstate[j, 2] - fstate[j, 0])
|
|
break;
|
|
sl = MathMin(sl, target[j] + fstate[j, 2] - fstate[j, 0]);
|
|
}
|
|
if(tp > 0)
|
|
{
|
|
sl = (float)MathMax(MathMin(MathAbs(sl) / (MaxSL * Point()), 1), 0.01);
|
|
tp = (float)MathMin(tp / (MaxTP * Point()), 1);
|
|
result[0] = float(MathMax(Buffer[tr].States[i].account[0] / 100 * 0.01, 0.011));
|
|
result[5] = result[1] = tp;
|
|
result[4] = result[2] = sl;
|
|
result[3] = 0;
|
|
bActions.AssignArray(result);
|
|
}
|
|
}
|
|
else
|
|
{
|
|
if(argmax == 0 || argmax > argmin)
|
|
{
|
|
float tp = 0;
|
|
float sl = 0;
|
|
float cur_sl = float(MaxSL * Point());
|
|
ulong pos = 0;
|
|
for(ulong j = 0; j < argmin; j++)
|
|
{
|
|
tp = MathMin(tp, target[j] + fstate[j, 2] - fstate[j, 0]);
|
|
pos = j;
|
|
if(cur_sl <= target[j] + fstate[j, 1] - fstate[j, 0])
|
|
break;
|
|
sl = MathMax(sl, target[j] + fstate[j, 1] - fstate[j, 0]);
|
|
}
|
|
if(tp < 0)
|
|
{
|
|
sl = (float)MathMax(MathMin(MathAbs(sl) / (MaxSL * Point()), 1), 0.01);
|
|
tp = (float)MathMin(-tp / (MaxTP * Point()), 1);
|
|
result[3] = float(MathMax(Buffer[tr].States[i].account[0] / 100 * 0.01, 0.011));
|
|
result[2] = result[4] = tp;
|
|
result[1] = result[5] = sl;
|
|
result[0] = 0;
|
|
bActions.AssignArray(result);
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
//--- Actor Policy
|
|
if(!Actor.backProp(GetPointer(bActions), (CBufferFloat*)GetPointer(bAccount), GetPointer(bGradient)))
|
|
{
|
|
PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
|
Stop = true;
|
|
break;
|
|
}
|
|
target = vector<float>::Zeros(NActions / 3);
|
|
for(int a = 0; a < NActions; a += 3)
|
|
target[a / 3] = float(result[a] > 0);
|
|
if(!Result.AssignArray(target) ||
|
|
!Probability.backProp(Result, GetPointer(Actor), LatentLayer) ||
|
|
!Actor.backPropGradient((CBufferFloat*)NULL, (CBufferFloat*)NULL, LatentLayer))
|
|
{
|
|
PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
|
Stop = true;
|
|
break;
|
|
}
|
|
//---
|
|
if(GetTickCount() - ticks > 500)
|
|
{
|
|
double percent = double(iter + i - start) * 100.0 / (Iterations);
|
|
string str = StringFormat("%-14s %6.2f%% -> Error %15.8f\n", "Actor", percent, Actor.getRecentAverageError());
|
|
str += StringFormat("%-13s %6.2f%% -> Error %15.8f\n", "Probability", percent, Probability.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__, "Probability", Probability.getRecentAverageError());
|
|
ExpertRemove();
|
|
//---
|
|
}
|
|
//+------------------------------------------------------------------+
|