ONNX.Price.Prediction/ONNX.Price.Prediction.Test.mq5

<EFBFBD><EFBFBD>//+------------------------------------------------------------------+
//|                                   ONNX.Price.Prediction.Test.mq5 |
//|                                  Copyright 2023, MetaQuotes Ltd. |
//|                                             https://www.mql5.com |
//+------------------------------------------------------------------+
#property copyright   "Copyright 2023, MetaQuotes Ltd."
#property link        "https://www.mql5.com"
#property version     "1.00"
#property description "Evaluation of the quality of the next Close price prediction by Python\\model.onnx.\n"
"Start in strategy tester on EURUSD,H1, open prices, from 2023.01.01 to 2023.02.01"

#resource "Python/model.onnx" as uchar ExtModel[]

#define SAMPLE_SIZE 10  // bars count in sample as defined in Python\PricePredictionTraining.py
// X, y = collect_dataset(df, history_size=10)

long     ExtHandle=INVALID_HANDLE;
double   ExtPredicted=0;
datetime ExtNextBar=0;
long     ExtTests=0;
long     ExtRightDirection=0;
double   ExtSumAbsoluteError=0.0;
//+------------------------------------------------------------------+
//| Expert initialization function                                   |
//+------------------------------------------------------------------+
int OnInit()
  {
   /*
    if(Symbol()!="EURUSD" || Period()!=PERIOD_H1)
      {
       Print("model must work with EURUSD,H1");
       return(INIT_FAILED);
      }
      */
//--- create a model from static buffer
   ExtHandle=OnnxCreateFromBuffer(ExtModel,ONNX_DEFAULT);
   if(ExtHandle==INVALID_HANDLE)
     {
      Print("OnnxCreateFromBuffer error ",GetLastError());
      return(INIT_FAILED);
     }
//--- since not all sizes defined in the input tensor we must set them explicitly
//--- first index - batch size, second index - series size, third index - number of series (OHLC)
   const long input_shape[] = {1,SAMPLE_SIZE,4};
   if(!OnnxSetInputShape(ExtHandle,0,input_shape))
     {
      Print("OnnxSetInputShape error ",GetLastError());
      return(INIT_FAILED);
     }
//--- since not all sizes defined in the output tensor we must set them explicitly
//--- first index - batch size, must match the batch size of the input tensor
//--- second index - number of predicted prices (we only predict Close)
   const long output_shape[] = {1,1};
   if(!OnnxSetOutputShape(ExtHandle,0,output_shape))
     {
      Print("OnnxSetOutputShape error ",GetLastError());
      return(INIT_FAILED);
     }
//---
   return(INIT_SUCCEEDED);
  }
//+------------------------------------------------------------------+
//| Expert deinitialization function                                 |
//+------------------------------------------------------------------+
void OnDeinit(const int reason)
  {
   if(ExtHandle!=INVALID_HANDLE)
     {
      OnnxRelease(ExtHandle);
      ExtHandle=INVALID_HANDLE;
     }
  }
//+------------------------------------------------------------------+
//| Expert tick function                                             |
//+------------------------------------------------------------------+
void OnTick()
  {
//--- check new bar
   if(TimeCurrent()<ExtNextBar)
      return;
   ExtTests++;
//--- set next bar time
   ExtNextBar=TimeCurrent();
   ExtNextBar-=ExtNextBar%PeriodSeconds();
   ExtNextBar+=PeriodSeconds();
//--- check predicted price
   CheckPredicted();
//--- predict next price
   PredictPrice();
  }
//+------------------------------------------------------------------+
//| Check predicted price                                            |
//+------------------------------------------------------------------+
void CheckPredicted(void)
  {
   if(ExtPredicted!=0.0)
     {
      static double closes[2];
      static datetime times[2];
      if(CopyClose(Symbol(),Period(),1,2,closes)==2 && CopyTime(Symbol(),Period(),1,2,times)==2)
        {
         ExtSumAbsoluteError+=MathAbs(ExtPredicted-closes[1]);
         double delta_predict=ExtPredicted-closes[0];
         double delta_actual=closes[1]-closes[0];
         if((delta_predict>0 && delta_actual>0) || (delta_predict<0 && delta_actual<0))
            ExtRightDirection++;
         Print("----------------------------------");
         Print("times[0]: ", times[0], "; closes[0]: ", closes[0], "; times[1]: ", times[1], "; closes[1]: ", closes[1]);
         Print("ExtPredicted: ", ExtPredicted, "; ExtSumAbsoluteError (ExtPredicted-closes[1]): ", ExtSumAbsoluteError, "; delta_predict (ExtPredicted-closes[0]): ", delta_predict, "; delta_actual (closes[1]-closes[0]): ", delta_actual, "; ExtRightDirection: ", ExtRightDirection, "; ExtTests: ", ExtTests);
        }
     }
  }
//+------------------------------------------------------------------+
//| Predict next price                                               |
//+------------------------------------------------------------------+
void PredictPrice(void)
  {
   static matrixf input_data(SAMPLE_SIZE,4);   // matrix for prepared input data
   static vectorf output_data(1);              // vector to get result
   static matrix  mm(SAMPLE_SIZE,4);           // matrix of horizontal vectors Mean
   static matrix  ms(SAMPLE_SIZE,4);           // matrix of horizontal vectors Std
   static matrix  x_norm(SAMPLE_SIZE,4);       // matrix for prices normalize
//--- prepare input data
   matrix rates;
//--- request last bars
   if(!rates.CopyRates(Symbol(),PERIOD_D1,COPY_RATES_OHLC,1,SAMPLE_SIZE))
     {
      ExtPredicted=0.0;
      return;
     }
//--- get series Mean
   vector m=rates.Mean(1);
//--- get series Std
   vector s=rates.Std(1);
//--- prepare matrices for prices normalization
   for(int i=0; i<SAMPLE_SIZE; i++)
     {
      mm.Row(m,i);
      ms.Row(s,i);
     }
//--- the input of the model must be a set of vertical OHLC vectors
   x_norm=rates.Transpose();
//--- normalize prices
   x_norm-=mm;
   x_norm/=ms;
//--- run the inference
   input_data.Assign(x_norm);
   if(!OnnxRun(ExtHandle,ONNX_NO_CONVERSION,input_data,output_data))
     {
      ExtPredicted=0.0;
      return;
     }
//--- denormalize the price from the output value
   ExtPredicted=output_data[0]*s[3]+m[3];
  }
//+------------------------------------------------------------------+
//| Tester function                                                  |
//+------------------------------------------------------------------+
double OnTester()
  {
   double mae=ExtSumAbsoluteError/ExtTests;
   Print("mae = ",mae);
   double right_directions=(ExtRightDirection*100.0)/ExtTests;
   PrintFormat("right_directions = %.2f%%",right_directions);
//---
   return(right_directions);
  }
//+------------------------------------------------------------------+