//+------------------------------------------------------------------+
//|                                                     Learning.mq5 |
//|                                                           Eugene |
//|                                             https://www.mql5.com |
//+------------------------------------------------------------------+
#property copyright "Eugene"
#property link      "https://www.mql5.com"
#property version   "1.00"
#include <CategoricalHMM\CategoricalHMM.mqh>
#include <Graphics\Graphic.mqh>
//+------------------------------------------------------------------+
//| Script program start function                                    |
//+------------------------------------------------------------------+
void OnStart()
  {
   int n_states = 2;
   int n_emissions = 6;
   int n_samples = 300;

//--- 1. Создаем модель (seed=99 для повторяемости) ---
   CCategoricalHMM model(n_states, n_emissions, 99);

// --- Задаем параметры модели ---
// --- 2. Устанавливаем матрицу переходов (Tr) ---
   matrix tr = {{0.95, 0.05},
        {0.1,  0.9}
     };

// --- 3.Устанавливаем матрицу эмиссий (E) ---
   matrix e = {{1.0/6, 1.0/6, 1.0/6, 1.0/6, 1.0/6, 1.0/6},
        {0.1,   0.1,   0.1,   0.1,   0.1,   0.5}
     };

// --- 4. Генерируем последовательность наблюдений и скрытых состояний ---
   vector y,z;
   model.Sample(n_samples, tr, e, y, z);

// --- 5. Создаем новую модель и обучаем на сгенерированых данных ---
   CCategoricalHMM hmm(n_states, n_emissions);
// --- В качестве начальных параметров зададим: ---
   matrix initialTR = {{0.7, 0.3},
        {0.3, 0.7}
     };
   matrix initialE  = {{0.15, 0.05, 0.1, 0.1, 0.4, 0.2},
        {0.05, 0.15, 0.1, 0.1, 0.3, 0.3}
     };

   hmm.m_TR = initialTR;
   hmm.m_E = initialE;

   double tol = 1e-6;
   if(hmm.Fit(y,100, tol, true))
     {
      Print("\n--- Результаты ---");
      PrintFormat("Final LogLikelihood: %.6f", hmm.m_logliks[hmm.m_logliks.Size() - 1]);
      Print("Estimated TR:\n", hmm.m_TR);
      Print("True TR:\n", tr);
      Print("Estimated E:\n", hmm.m_E);
      Print("True E:\n", e);
     }

// --- БЛОК ВИЗУАЛИЗАЦИИ ---
   int plotLimit = (int)hmm.m_logliks.Size();
   double x[], y_[];
   ArrayResize(x, plotLimit);
   ArrayResize(y_, plotLimit);
   for(int i = 0; i < plotLimit; i++)
     {
      x[i] = i;
      y_[i] = hmm.m_logliks[i];
     }
   ChartSetInteger(0,CHART_SHOW,false);
   CGraphic graphic;
   ulong width = ChartGetInteger(0,CHART_WIDTH_IN_PIXELS);
   ulong height = ChartGetInteger(0,CHART_HEIGHT_IN_PIXELS);
   graphic.Create(0,"HMM_States_Plot",0,0,0,int(width),int(height));
   CCurve *curve = graphic.CurveAdd(x, y_, ColorToARGB(clrDodgerBlue, 255), CURVE_LINES,"LL");
   curve.LinesStyle(STYLE_SOLID);
   curve.LinesWidth(2);
   graphic.BackgroundMainColor(ColorToARGB(clrBlack,255));
   graphic.BackgroundMainSize(24);
   graphic.BackgroundMain("EM Training Progress");
   graphic.XAxis().Name("Iteration");
   graphic.XAxis().NameSize(18);
   graphic.YAxis().Name("Log Likelihood");
   graphic.YAxis().NameSize(18);
   graphic.CurvePlotAll();
   graphic.Update();
   Sleep(5*1000);
   ChartSetInteger(0,CHART_SHOW,true);
   graphic.Destroy();
   ChartRedraw(0);
  }
//+------------------------------------------------------------------+