2026-08-10 03:17:55 +03:00 | | | //+------------------------------------------------------------------+
|
2026-08-13 23:48:47 +03:00 | | | //| StudyOnline.mq5 |
|
2026-08-10 03:17:55 +03:00 | | | //+------------------------------------------------------------------+
|
| | | #property copyright "Copyright DNG®"
|
| | | #property link "https://www.mql5.com/ru/users/dng"
|
| | | #property version "1.00"
|
| | | //+------------------------------------------------------------------+
|
| | | //| Includes |
|
| | | //+------------------------------------------------------------------+
|
| | | #define StudyOnline
|
| | |
|
| | | #include "Trajectory.mqh"
|
| | | #include <Trade\Trade.mqh>
|
| | | #include <Trade\SymbolInfo.mqh>
|
| | | #include <Indicators\Oscilators.mqh>
|
| | | //+------------------------------------------------------------------+
|
| | | //| Input parameters |
|
| | | //+------------------------------------------------------------------+
|
| | | input group "---- Other ----"
|
| | | input int Agent = 1;
|
| | | input group "---- Base Memory ----"
|
| | | input bool LoadBaseMemorySnapshot = true;
|
| | | input string BaseMemorySnapshotPath = "VLADriverBaseMemory.snapshot";
|
| | | input bool PublishMemoryOnOnlineMemorySize = true;
|
| | | input bool PublishMemoryOnEpisodeClosure = true;
|
| | | CNet cActor[2];
|
| | | CNet cCritic[3];
|
| | | CNet cStateEncoder;
|
| | |
|
| | | datetime dtStudied;
|
| | | CBufferFloat bState;
|
| | | CBufferFloat bContext;
|
| | | CBufferFloat bTime;
|
| | | CBufferFloat bGradient;
|
| | | CBufferFloat *bAction;
|
| | | CBufferFloat *Result;
|
| | |
|
| | | double PrevBalance = 0;
|
| | | double PrevEquity = 0;
|
| | | bool bFirstRun = true;
|
| | | bool bFillStack = true;
|
2026-08-13 23:48:47 +03:00 | | | // The library owns records, centroids, snapshots and GPU retrieval.
|
2026-08-10 03:17:55 +03:00 | | | CNeuronRAGMemory RAGMemory;
|
| | | bool BaseMemoryNullBound = true;
|
2026-08-13 23:48:47 +03:00 | | | // These are execution facts only; pending events live in CNeuronRAGMemory.
|
2026-08-10 03:17:55 +03:00 | | | ulong OnlinePendingId = 0;
|
| | | bool OnlinePendingExecuted = false;
|
| | | datetime OnlinePendingStartTime = 0;
|
| | | double OnlinePendingStartBalance = 0;
|
| | | double OnlinePendingStartEquity = 0;
|
| | | double OnlinePendingPeakEquity = 0;
|
| | | double OnlinePendingMaxDrawdown = 0;
|
| | | double OnlinePendingMaxMarginToEquity = 0;
|
| | | bool OnlinePendingHasEquitySample = false;
|
| | | bool OnlinePendingHasRiskSample = false;
|
| | | ulong OnlinePendingExecutionDeal = 0;
|
| | | ulong OnlinePendingPositionId = 0;
|
2026-08-13 23:48:47 +03:00 | | | ulong OnlinePendingPositionIds[];
|
2026-08-10 03:17:55 +03:00 | | | uint OnlinePendingStartPosition = 0;
|
| | | CBufferFloat OnlinePendingAction;
|
| | | CBufferFloat OnlinePendingScenarioEmbedding;
|
| | | bool OnlinePendingPrepared = false;
|
| | | const float RejectedExecutionPenaltyScale = 100.0f;
|
| | | enum ENUM_VLADriverModifyResult
|
| | | {
|
| | | VLADriverModifyNoOp=0,
|
| | | VLADriverModifyAccepted=1,
|
| | | VLADriverModifyRejected=2
|
| | | };
|
| | | void ClearOnlinePendingScenario(void)
|
| | | {
|
| | | OnlinePendingId = 0;
|
| | | OnlinePendingExecuted = false;
|
| | | OnlinePendingStartTime = 0;
|
| | | OnlinePendingStartBalance = 0;
|
| | | OnlinePendingStartEquity = 0;
|
| | | OnlinePendingPeakEquity = 0;
|
| | | OnlinePendingMaxDrawdown = 0;
|
| | | OnlinePendingMaxMarginToEquity = 0;
|
| | | OnlinePendingHasEquitySample = false;
|
| | | OnlinePendingHasRiskSample = false;
|
| | | OnlinePendingExecutionDeal = 0;
|
| | | OnlinePendingPositionId = 0;
|
2026-08-13 23:48:47 +03:00 | | | ArrayResize(OnlinePendingPositionIds,0);
|
2026-08-10 03:17:55 +03:00 | | | OnlinePendingStartPosition = 0;
|
| | | OnlinePendingAction.BufferInit(NActions, 0);
|
| | | OnlinePendingScenarioEmbedding.BufferInit(EmbeddingSize, 0);
|
| | | OnlinePendingPrepared = false;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
2026-08-13 23:48:47 +03:00 | | | //| Capture the observed state before the new market order is sent. |
|
2026-08-10 03:17:55 +03:00 | | | //+------------------------------------------------------------------+
|
| | | bool AnchorOnlinePendingScenario(void)
|
| | | {
|
| | | if(OnlinePendingId != 0 || OnlinePendingPrepared || bAction == NULL)
|
| | | ReturnFalse;
|
| | | CBufferFloat *scenario_embedding = NULL;
|
| | | if(!cStateEncoder.GetLayerOutputDevice(StateScenarioLayer, scenario_embedding))
|
| | | {
|
| | | PrintFormat("%s -> %d pending scenario output failed", __FUNCTION__, __LINE__);
|
| | | ReturnFalse;
|
| | | }
|
| | | if(scenario_embedding.Total() != EmbeddingSize)
|
| | | {
|
| | | PrintFormat("%s -> %d pending scenario embedding size mismatch", __FUNCTION__, __LINE__);
|
| | | ReturnFalse;
|
| | | }
|
| | | float scenario_embedding_data[];
|
| | | uint replaced_values = 0;
|
| | | const bool scenario_read = RAGMemory.ReadScenarioEmbedding(scenario_embedding,
|
| | | scenario_embedding_data, replaced_values);
|
| | | if(replaced_values > 0)
|
| | | PrintFormat("Online RAG scenario anchor replaced %u non-finite values", replaced_values);
|
| | | if(!scenario_read)
|
| | | {
|
| | | PrintFormat("%s -> %d pending scenario anchor read failed", __FUNCTION__, __LINE__);
|
| | | ReturnFalse;
|
| | | }
|
| | | for(int i = 0; i < NActions; i++)
|
| | | if(!OnlinePendingAction.Update(i, bAction[i]))
|
| | | ReturnFalse;
|
| | | if(!OnlinePendingScenarioEmbedding.BufferInit(EmbeddingSize,0))
|
| | | ReturnFalse;
|
| | | if(!OnlinePendingScenarioEmbedding.AssignArray(scenario_embedding_data))
|
| | | ReturnFalse;
|
| | | OnlinePendingStartTime = TimeCurrent();
|
| | | OnlinePendingStartBalance = AccountInfoDouble(ACCOUNT_BALANCE);
|
| | | OnlinePendingStartEquity = AccountInfoDouble(ACCOUNT_EQUITY);
|
| | | OnlinePendingPeakEquity=OnlinePendingStartEquity;
|
| | | OnlinePendingPrepared = true;
|
| | | return true;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
| | | //| Create library pending state only after a market order is known. |
|
| | | //+------------------------------------------------------------------+
|
| | | bool OpenConfirmedOnlinePendingScenario(const ulong deal)
|
| | | {
|
| | | if(!OnlinePendingPrepared || OnlinePendingId!=0 || deal==0 ||
|
| | | !HistoryDealSelect(deal) ||
|
| | | ulong(HistoryDealGetInteger(deal,DEAL_POSITION_ID))==0)
|
| | | {
|
| | | ClearOnlinePendingScenario();
|
| | | ReturnFalse;
|
| | | }
|
| | | float normalized_action[NActions];
|
| | | if(!NormalizeMemoryAction(RAGMemory,GetPointer(OnlinePendingAction),0,
|
| | | OnlinePendingStartEquity,normalized_action))
|
| | | {
|
| | | ClearOnlinePendingScenario();
|
| | | ReturnFalse;
|
| | | }
|
| | | float scenario_embedding_data[EmbeddingSize];
|
| | | for(int i=0;i<EmbeddingSize;i++)
|
| | | scenario_embedding_data[i]=OnlinePendingScenarioEmbedding[i];
|
| | | OnlinePendingId=RAGMemory.OpenPending(scenario_embedding_data,normalized_action);
|
| | | if(OnlinePendingId==0 || !RegisterOnlinePendingExecution(deal))
|
| | | {
|
| | | if(OnlinePendingId!=0)
|
| | | RAGMemory.CancelPending(OnlinePendingId);
|
| | | ClearOnlinePendingScenario();
|
| | | ReturnFalse;
|
| | | }
|
| | | return true;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
| | | //| Apply one uniform Critic-only penalty for rejected execution. |
|
| | | //+------------------------------------------------------------------+
|
| | | bool PenalizeRejectedExecution(const double requested_volume)
|
| | | {
|
| | | if(requested_volume<=0)
|
| | | return true;
|
| | | Result.Clear();
|
| | | if(!Result.Add(float(-RejectedExecutionPenaltyScale*requested_volume)) ||
|
| | | !cCritic[0].backProp(Result,GetPointer(cStateEncoder),StateTokenLayer))
|
| | | {
|
| | | PrintFormat("%s -> %d rejected execution Critic update failed",__FUNCTION__,__LINE__);
|
| | | ReturnFalse;
|
| | | }
|
| | | return true;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
2026-08-13 23:48:47 +03:00 | | | //| Classify terminal acknowledgement without treating no-op as |
|
| | | //| data. |
|
2026-08-10 03:17:55 +03:00 | | | //+------------------------------------------------------------------+
|
| | | ENUM_VLADriverModifyResult ClassifyOnlineTradeResult(const bool request_sent)
|
| | | {
|
| | | const uint retcode=Trade.ResultRetcode();
|
| | | if(request_sent && (retcode==TRADE_RETCODE_DONE || retcode==TRADE_RETCODE_DONE_PARTIAL))
|
| | | return VLADriverModifyAccepted;
|
| | | if(retcode==TRADE_RETCODE_NO_CHANGES)
|
| | | return VLADriverModifyNoOp;
|
| | | return VLADriverModifyRejected;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
2026-08-13 23:48:47 +03:00 | | | //| Check whether one position belongs to the current trade plan. |
|
2026-08-10 03:17:55 +03:00 | | | //+------------------------------------------------------------------+
|
2026-08-13 23:48:47 +03:00 | | | bool OnlinePendingHasPosition(const ulong position)
|
2026-08-10 03:17:55 +03:00 | | | {
|
2026-08-13 23:48:47 +03:00 | | | if(position==0)
|
| | | ReturnFalse;
|
| | | const int total=ArraySize(OnlinePendingPositionIds);
|
| | | for(int i=0;i<total;i++)
|
| | | if(OnlinePendingPositionIds[i]==position)
|
| | | return true;
|
| | | return false;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
| | | //| Add a unique position identity to the current trade plan. |
|
| | | //+------------------------------------------------------------------+
|
| | | bool AddOnlinePendingPosition(const ulong position)
|
| | | {
|
| | | if(position==0)
|
| | | ReturnFalse;
|
| | | if(OnlinePendingHasPosition(position))
|
| | | return true;
|
| | | const int total=ArraySize(OnlinePendingPositionIds);
|
| | | if(ArrayResize(OnlinePendingPositionIds,total+1)!=total+1)
|
2026-08-10 03:17:55 +03:00 | | | ReturnFalse;
|
2026-08-13 23:48:47 +03:00 | | | OnlinePendingPositionIds[total]=position;
|
| | | return true;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
| | | //| Undo a just-reserved position when event confirmation failed. |
|
| | | //+------------------------------------------------------------------+
|
| | | bool RemoveOnlinePendingPosition(const ulong position)
|
| | | {
|
| | | const int total=ArraySize(OnlinePendingPositionIds);
|
| | | for(int i=0;i<total;i++)
|
| | | if(OnlinePendingPositionIds[i]==position)
|
| | | {
|
| | | for(int j=i+1;j<total;j++)
|
| | | OnlinePendingPositionIds[j-1]=OnlinePendingPositionIds[j];
|
| | | if(ArrayResize(OnlinePendingPositionIds,total-1)!=total-1)
|
| | | ReturnFalse;
|
| | | return true;
|
| | | }
|
| | | ReturnFalse;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
| | | //| A plan remains open while any of its netting/hedging positions |
|
| | | //| remains open. |
|
| | | //+------------------------------------------------------------------+
|
| | | bool OnlinePendingPositionsAreOpen(void)
|
| | | {
|
| | | if(ArraySize(OnlinePendingPositionIds)<=0)
|
| | | return false;
|
2026-08-10 03:17:55 +03:00 | | | for(int i=0;i<PositionsTotal();i++)
|
| | | if(PositionGetSymbol(i)==Symb.Name() &&
|
2026-08-13 23:48:47 +03:00 | | | OnlinePendingHasPosition(ulong(PositionGetInteger(POSITION_IDENTIFIER))))
|
2026-08-10 03:17:55 +03:00 | | | return true;
|
2026-08-13 23:48:47 +03:00 | | | return false;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
| | | //| Compatibility helper for partial-close classification. |
|
| | | //+------------------------------------------------------------------+
|
| | | bool OnlinePendingPositionIsOpen(void)
|
| | | {
|
| | | return OnlinePendingPositionsAreOpen();
|
2026-08-10 03:17:55 +03:00 | | | }
|
| | | //+------------------------------------------------------------------+
|
| | | //| Execute TP/SL change and classify no-op, accepted and rejected. |
|
| | | //+------------------------------------------------------------------+
|
| | | ENUM_VLADriverModifyResult ModifyOnlinePosition(const ENUM_POSITION_TYPE type,
|
| | | const double sl,const double tp)
|
| | | {
|
| | | const int total=PositionsTotal();
|
| | | const datetime earliest_update=TimeCurrent()-5*PeriodSeconds(TimeFrame);
|
| | | uint result=uint(VLADriverModifyNoOp);
|
| | | for(int i=0;i<total;i++)
|
| | | {
|
| | | if(PositionGetSymbol(i)!=Symb.Name() || PositionGetInteger(POSITION_TYPE)!=type ||
|
| | | PositionGetInteger(POSITION_TIME_UPDATE)>earliest_update)
|
| | | continue;
|
| | | bool modify=false;
|
| | | double position_sl=PositionGetDouble(POSITION_SL);
|
| | | double position_tp=PositionGetDouble(POSITION_TP);
|
| | | if(type==POSITION_TYPE_BUY)
|
| | | {
|
| | | if((sl-position_sl)>=Symb.Point())
|
| | | { position_sl=sl; modify=true; }
|
| | | if(MathAbs(tp-position_tp)>=Symb.Point())
|
| | | { position_tp=tp; modify=true; }
|
| | | }
|
| | | else
|
| | | {
|
| | | if((position_sl-sl)>=Symb.Point())
|
| | | { position_sl=sl; modify=true; }
|
| | | if(MathAbs(tp-position_tp)>=Symb.Point())
|
| | | { position_tp=tp; modify=true; }
|
| | | }
|
| | | if(!modify)
|
| | | continue;
|
| | | const ENUM_VLADriverModifyResult update=ClassifyOnlineTradeResult(
|
| | | Trade.PositionModify(PositionGetInteger(POSITION_TICKET),position_sl,position_tp));
|
| | | result|=uint(update);
|
| | | }
|
| | | return (ENUM_VLADriverModifyResult)result;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
2026-08-13 23:48:47 +03:00 | | | //| Full close is terminal and therefore never creates RAG |
|
| | | //| correction. |
|
2026-08-10 03:17:55 +03:00 | | | //+------------------------------------------------------------------+
|
| | | ENUM_VLADriverModifyResult CloseOnlineByDirection(const ENUM_POSITION_TYPE type)
|
| | | {
|
| | | return ClassifyOnlineTradeResult(CloseByDirection(type));
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
2026-08-13 23:48:47 +03:00 | | | //| Partial close is an action correction only while episode |
|
| | | //| remains. |
|
2026-08-10 03:17:55 +03:00 | | | //+------------------------------------------------------------------+
|
| | | ENUM_VLADriverModifyResult CloseOnlinePartial(const ENUM_POSITION_TYPE type,
|
| | | const double volume)
|
| | | {
|
| | | const ENUM_VLADriverModifyResult result=ClassifyOnlineTradeResult(ClosePartial(type,volume));
|
| | | if(result!=VLADriverModifyAccepted || !OnlinePendingPositionIsOpen())
|
| | | return (result==VLADriverModifyAccepted ? VLADriverModifyNoOp : result);
|
| | | return VLADriverModifyAccepted;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
2026-08-13 23:48:47 +03:00 | | | //| Sample actual account exposure for a confirmed pending |
|
| | | //| execution. |
|
2026-08-10 03:17:55 +03:00 | | | //+------------------------------------------------------------------+
|
| | | void SampleOnlinePendingScenario(void)
|
| | | {
|
| | | if(OnlinePendingId==0 || !OnlinePendingExecuted)
|
| | | return;
|
| | | double equity = AccountInfoDouble(ACCOUNT_EQUITY);
|
| | | if(MathIsValidNumber(equity) && equity > 0)
|
| | | {
|
| | | if(!OnlinePendingHasEquitySample || equity>OnlinePendingPeakEquity)
|
| | | OnlinePendingPeakEquity=equity;
|
| | | OnlinePendingMaxDrawdown=MathMax(OnlinePendingMaxDrawdown,OnlinePendingPeakEquity-equity);
|
| | | OnlinePendingHasEquitySample=true;
|
| | | double margin = AccountInfoDouble(ACCOUNT_MARGIN);
|
| | | if(MathIsValidNumber(margin) && margin >= 0)
|
| | | {
|
| | | OnlinePendingMaxMarginToEquity=MathMax(OnlinePendingMaxMarginToEquity,margin/equity);
|
| | | OnlinePendingHasRiskSample=true;
|
| | | }
|
| | | }
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
2026-08-13 23:48:47 +03:00 | | | //| Bind only a confirmed market deal to the causal pending anchor. |
|
2026-08-10 03:17:55 +03:00 | | | //+------------------------------------------------------------------+
|
| | | bool RegisterOnlinePendingExecution(const ulong deal)
|
| | | {
|
| | | if(OnlinePendingId==0 || deal==0 || !HistoryDealSelect(deal))
|
| | | {
|
| | | if(OnlinePendingId!=0)
|
| | | RAGMemory.CancelPending(OnlinePendingId);
|
| | | ClearOnlinePendingScenario();
|
| | | ReturnFalse;
|
| | | }
|
| | | ulong position_id = ulong(HistoryDealGetInteger(deal, DEAL_POSITION_ID));
|
| | | if(position_id == 0)
|
| | | {
|
| | | RAGMemory.CancelPending(OnlinePendingId);
|
| | | ClearOnlinePendingScenario();
|
| | | ReturnFalse;
|
| | | }
|
| | | if(!RAGMemory.ConfirmPendingExecution(OnlinePendingId,deal,position_id))
|
| | | {
|
| | | RAGMemory.CancelPending(OnlinePendingId);
|
| | | ClearOnlinePendingScenario();
|
| | | ReturnFalse;
|
| | | }
|
| | | OnlinePendingExecutionDeal=deal;
|
| | | OnlinePendingPositionId=position_id;
|
2026-08-13 23:48:47 +03:00 | | | if(!AddOnlinePendingPosition(position_id))
|
| | | {
|
| | | RAGMemory.CancelPending(OnlinePendingId);
|
| | | ClearOnlinePendingScenario();
|
| | | ReturnFalse;
|
| | | }
|
2026-08-10 03:17:55 +03:00 | | | OnlinePendingExecuted=true;
|
| | | SampleOnlinePendingScenario();
|
| | | return true;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
| | | //| Add only an accepted position correction to the pending episode. |
|
| | | //+------------------------------------------------------------------+
|
2026-08-13 23:48:47 +03:00 | | | bool RegisterOnlinePendingCorrection(const ulong deal=0)
|
2026-08-10 03:17:55 +03:00 | | | {
|
2026-08-13 23:48:47 +03:00 | | | if(OnlinePendingId==0 || !OnlinePendingExecuted || OnlinePendingPositionId==0 || bAction==NULL)
|
| | | ReturnFalse;
|
| | | ulong position=OnlinePendingPositionId;
|
| | | if(deal>0)
|
| | | {
|
| | | if(!HistoryDealSelect(deal))
|
| | | ReturnFalse;
|
| | | position=ulong(HistoryDealGetInteger(deal,DEAL_POSITION_ID));
|
| | | if(position==0 || HistoryDealGetInteger(deal,DEAL_ENTRY)!=DEAL_ENTRY_IN)
|
| | | ReturnFalse;
|
| | | }
|
| | | float normalized_action[NActions];
|
2026-08-10 03:17:55 +03:00 | | | for(int i=0;i<NActions;i++)
|
| | | OnlinePendingAction.Update(i,bAction[i]);
|
| | | if(!NormalizeMemoryAction(RAGMemory,GetPointer(OnlinePendingAction),0,OnlinePendingStartEquity,normalized_action) ||
|
| | | !RAGMemory.CorrectPending(OnlinePendingId,normalized_action))
|
| | | ReturnFalse;
|
2026-08-13 23:48:47 +03:00 | | | if(deal>0)
|
| | | {
|
| | | const bool already_added=OnlinePendingHasPosition(position);
|
| | | if(!already_added && !AddOnlinePendingPosition(position))
|
| | | {
|
| | | RAGMemory.CancelPendingCorrection(OnlinePendingId);
|
| | | ReturnFalse;
|
| | | }
|
| | | if(!RAGMemory.ConfirmPendingCorrection(OnlinePendingId,deal,position))
|
| | | {
|
| | | if(!already_added)
|
| | | RemoveOnlinePendingPosition(position);
|
| | | RAGMemory.CancelPendingCorrection(OnlinePendingId);
|
| | | ReturnFalse;
|
| | | }
|
| | | return true;
|
| | | }
|
2026-08-10 03:17:55 +03:00 | | | if(RAGMemory.ConfirmPendingCorrection(OnlinePendingId,OnlinePendingPositionId))
|
| | | return true;
|
| | | RAGMemory.CancelPendingCorrection(OnlinePendingId);
|
| | | ReturnFalse;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
| | | //| Persist one validated snapshot/manifest publication pair. |
|
| | | //+------------------------------------------------------------------+
|
| | | bool SaveOnlineBaseMemory(void)
|
| | | {
|
| | | if(RAGMemory.ScenarioCount()==0)
|
| | | return true;
|
| | | return RAGMemory.SavePublishedSnapshot(BaseMemorySnapshotPath);
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
| | | //| Rebind both Actor copies before any later forward pass. |
|
| | | //+------------------------------------------------------------------+
|
| | | bool RebindOnlineRAGMemory(void)
|
| | | {
|
| | | const bool actor0_bound=cActor[0].BindRAGMemory(2,GetPointer(RAGMemory));
|
| | | const bool actor1_bound=cActor[1].BindRAGMemory(2,GetPointer(RAGMemory));
|
| | | if(actor0_bound && actor1_bound)
|
| | | return true;
|
| | | PrintFormat("%s -> %d online RAG Actor rebind failed",__FUNCTION__,__LINE__);
|
| | | DetachOnlineRAGMemory();
|
| | | ExpertRemove();
|
| | | ReturnFalse;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
| | | //| Detach both consumers before COW deletes the old device buffers. |
|
| | | //+------------------------------------------------------------------+
|
| | | bool DetachOnlineRAGMemory(void)
|
| | | {
|
| | | const bool actor0_detached=cActor[0].UnbindRAGMemory(2);
|
| | | const bool actor1_detached=cActor[1].UnbindRAGMemory(2);
|
| | | if(actor0_detached && actor1_detached)
|
| | | return true;
|
| | | PrintFormat("%s -> %d online RAG Actor detach failed",__FUNCTION__,__LINE__);
|
| | | ExpertRemove();
|
| | | ReturnFalse;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
2026-08-13 23:48:47 +03:00 | | | //| Build and swap only an immutable Base Memory snapshot. |
|
2026-08-10 03:17:55 +03:00 | | | //+------------------------------------------------------------------+
|
| | | bool PublishOnlineMemory(const bool episode_completed)
|
| | | {
|
| | | const uint pending=RAGMemory.PendingRecordCount();
|
| | | if(pending==0 ||
|
| | | (!(PublishMemoryOnOnlineMemorySize && pending>=OnlineMemorySize) &&
|
| | | !(PublishMemoryOnEpisodeClosure && episode_completed)))
|
| | | return true;
|
| | | if(!DetachOnlineRAGMemory())
|
| | | ReturnFalse;
|
| | | if(!RAGMemory.Publish(episode_completed))
|
| | | {
|
| | | if(!RebindOnlineRAGMemory())
|
| | | ReturnFalse;
|
| | | ReturnFalse;
|
| | | }
|
| | | if(!RebindOnlineRAGMemory())
|
| | | ReturnFalse;
|
| | | BaseMemoryNullBound=false;
|
| | | return SaveOnlineBaseMemory();
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
2026-08-13 23:48:47 +03:00 | | | //| Store the actual, fully closed execution as one completed case. |
|
2026-08-10 03:17:55 +03:00 | | | //+------------------------------------------------------------------+
|
| | | void CollectOnlineCompletedScenario(void)
|
| | | {
|
| | | if(OnlinePendingId==0 || !OnlinePendingExecuted)
|
| | | return;
|
| | | if(OnlinePendingPositionId==0 && OnlinePendingExecutionDeal>0)
|
| | | OnlinePendingPositionId=ulong(HistoryDealGetInteger(OnlinePendingExecutionDeal,DEAL_POSITION_ID));
|
| | | if(OnlinePendingPositionId==0)
|
| | | return;
|
2026-08-13 23:48:47 +03:00 | | | if(ArraySize(OnlinePendingPositionIds)==0)
|
| | | if(!AddOnlinePendingPosition(OnlinePendingPositionId))
|
| | | return;
|
| | | if(OnlinePendingPositionsAreOpen())
|
| | | return;
|
2026-08-10 03:17:55 +03:00 | | | SRAGTerminalOutcome evaluated_terminal;
|
| | | // The simulator is a base/fallback only; live account/deal data below wins.
|
| | | EvaluateTerminalOutcome(RAGMemory,GetPointer(OnlinePendingAction),OnlinePendingStartBalance,
|
| | | OnlinePendingStartPosition,evaluated_terminal);
|
| | | double reward = 0;
|
| | | double costs = 0;
|
| | | bool has_deals = false;
|
| | | if(HistorySelect(OnlinePendingStartTime, TimeCurrent()))
|
| | | for(int i = 0; i < HistoryDealsTotal(); i++)
|
| | | {
|
| | | ulong deal = HistoryDealGetTicket(i);
|
2026-08-13 23:48:47 +03:00 | | | if(deal==0 || !OnlinePendingHasPosition(ulong(HistoryDealGetInteger(deal,DEAL_POSITION_ID))))
|
2026-08-10 03:17:55 +03:00 | | | continue;
|
| | | double profit = HistoryDealGetDouble(deal, DEAL_PROFIT);
|
| | | double swap = HistoryDealGetDouble(deal, DEAL_SWAP);
|
| | | double commission = HistoryDealGetDouble(deal, DEAL_COMMISSION);
|
| | | reward += profit + swap + commission;
|
| | | costs += -MathMin(swap, 0.0) - MathMin(commission, 0.0);
|
| | | has_deals = true;
|
| | | }
|
| | | bool has_actual_reward = has_deals;
|
| | | if(!has_actual_reward && OnlinePendingStartBalance>0)
|
| | | {
|
| | | double end_balance = AccountInfoDouble(ACCOUNT_BALANCE);
|
| | | if(MathIsValidNumber(end_balance))
|
| | | {
|
| | | reward=end_balance-OnlinePendingStartBalance;
|
| | | has_actual_reward = true;
|
| | | }
|
| | | }
|
| | | bool partial_metrics=(!has_deals || !OnlinePendingHasEquitySample ||
|
| | | !OnlinePendingHasRiskSample || OnlinePendingStartTime<=0);
|
| | | const int period_seconds = PeriodSeconds(TimeFrame);
|
| | | SRAGTerminalOutcome terminal=evaluated_terminal;
|
| | | if(has_actual_reward && period_seconds > 0)
|
| | | {
|
| | | const double risk_budget=OnlinePendingStartEquity*MemoryRiskBudgetFraction;
|
| | | RAGMemory.NormalizeTerminalOutcome(reward,OnlinePendingMaxDrawdown,costs,
|
| | | OnlinePendingMaxMarginToEquity*risk_budget,
|
| | | double(TimeCurrent()-OnlinePendingStartTime),
|
| | | OnlinePendingStartEquity,risk_budget,
|
| | | period_seconds*MathMax(NForecast,1),terminal);
|
| | | }
|
| | | else
|
| | | partial_metrics = true;
|
| | | if(partial_metrics)
|
| | | Print("Online RAG record has partial terminal metrics");
|
| | | if(!RAGMemory.ClosePending(OnlinePendingId,terminal))
|
| | | {
|
| | | PrintFormat("%s -> %d online RAG close failed", __FUNCTION__, __LINE__);
|
| | | return;
|
| | | }
|
| | | ClearOnlinePendingScenario();
|
| | | if(!PublishOnlineMemory(true))
|
| | | PrintFormat("%s -> %d online memory publication failed", __FUNCTION__, __LINE__);
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
| | | //| Expert initialization function |
|
| | | //+------------------------------------------------------------------+
|
| | | int OnInit()
|
| | | {
|
| | | if(!MQLInfoInteger(MQL_TESTER))
|
| | | return INIT_FAILED;
|
| | | BaseMemoryNullBound = true;
|
| | | ClearOnlinePendingScenario();
|
| | | if(!Symb.Name(_Symbol))
|
| | | return INIT_FAILED;
|
| | | Symb.Refresh();
|
| | | if(!RSI.Create(Symb.Name(), TimeFrame, RSIPeriod, RSIPrice))
|
| | | return INIT_FAILED;
|
| | | if(!CCI.Create(Symb.Name(), TimeFrame, CCIPeriod, CCIPrice))
|
| | | return INIT_FAILED;
|
| | | if(!ATR.Create(Symb.Name(), TimeFrame, ATRPeriod))
|
| | | return INIT_FAILED;
|
| | | if(!MACD.Create(Symb.Name(), TimeFrame, FastPeriod, SlowPeriod, SignalPeriod, MACDPrice))
|
| | | return INIT_FAILED;
|
| | | if(!RSI.BufferResize(StackSize + HistoryBars) || !CCI.BufferResize(StackSize + HistoryBars) ||
|
| | | !ATR.BufferResize(StackSize + HistoryBars) || !MACD.BufferResize(StackSize + HistoryBars))
|
| | | {
|
| | | PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
| | | return INIT_FAILED;
|
| | | }
|
| | | if(!Trade.SetTypeFillingBySymbol(Symb.Name()))
|
| | | return INIT_FAILED;
|
2026-08-13 23:48:47 +03:00 | | | // Load Models.
|
2026-08-10 03:17:55 +03:00 | | | float temp;
|
| | | CArrayObj *actor = new CArrayObj();
|
| | | CArrayObj *critic = new CArrayObj();
|
| | | if(!CreateDescriptions(actor, critic))
|
| | | {
|
| | | delete actor;
|
| | | delete critic;
|
| | | return INIT_FAILED;
|
| | | }
|
| | | if(!cStateEncoder.Load(FileName + "StEnc.nnw", temp, temp, temp, dtStudied, true))
|
| | | {
|
| | | PrintFormat("Error of load StateEncoder: %d", GetLastError());
|
| | | delete actor;
|
| | | delete critic;
|
| | | return INIT_FAILED;
|
| | | }
|
| | | if(!cActor[0].Load(FileName + "Act.nnw", temp, temp, temp, dtStudied, true) ||
|
| | | !cActor[1].Load(FileName + "Act.nnw", temp, temp, temp, dtStudied, true))
|
| | | {
|
| | | Print("Create new Actor");
|
| | | if(!cActor[0].Create(actor) ||
|
| | | !cActor[1].Create(actor))
|
| | | {
|
| | | delete actor;
|
| | | delete critic;
|
| | | return INIT_FAILED;
|
| | | }
|
| | | cActor[1].SetOpenCL(cActor[0].GetOpenCL());
|
| | | cActor[1].WeightsUpdate(GetPointer(cActor[0]), 1);
|
| | | }
|
| | | bool result = true;
|
| | | for(uint i = 0; (i < cCritic.Size() && result); i++)
|
| | | if(!cCritic[i].Load(FileName + CriticCheckpointFile, temp, temp, temp, dtStudied, true))
|
| | | result = false;
|
| | | if(!result)
|
| | | {
|
| | | Print("Create new Critic model");
|
| | | for(uint i = 0; i < cCritic.Size(); i++)
|
| | | if(!cCritic[i].Create(critic))
|
| | | {
|
| | | DeleteObj(actor);
|
| | | DeleteObj(critic);
|
| | | return INIT_FAILED;
|
| | | }
|
| | | for(uint i = 1; i < cCritic.Size(); i++)
|
| | | {
|
| | | cCritic[i].SetOpenCL(cCritic[0].GetOpenCL());
|
| | | if(!cCritic[i].WeightsUpdate(GetPointer(cCritic[0]), 1))
|
| | | {
|
| | | DeleteObj(actor);
|
| | | DeleteObj(critic);
|
| | | return INIT_FAILED;
|
| | | }
|
| | | }
|
| | | }
|
| | | DeleteObj(actor);
|
| | | DeleteObj(critic);
|
| | | for(int i = 0; i < 2; i++)
|
| | | {
|
| | | cActor[i].TrainMode(i == 0);
|
| | | cCritic[i].TrainMode(i == 0);
|
| | | cCritic[i].SetOpenCL(cActor[0].GetOpenCL());
|
| | | cActor[i].Clear();
|
| | | cCritic[i].Clear();
|
| | | }
|
| | | cActor[1].SetOpenCL(cActor[0].GetOpenCL());
|
| | | for(uint i = 2; i < cCritic.Size(); i++)
|
| | | {
|
| | | cCritic[i].TrainMode(false);
|
| | | cCritic[i].SetOpenCL(cActor[0].GetOpenCL());
|
| | | cCritic[i].Clear();
|
| | | }
|
| | | cStateEncoder.TrainMode(false);
|
| | | cStateEncoder.SetOpenCL(cActor[0].GetOpenCL());
|
| | | cStateEncoder.Clear();
|
| | | CLayerDescription memory_description;
|
| | | if(!CreateRAGMemoryDescription(memory_description) ||
|
| | | !RAGMemory.Init(0,0,cActor[0].GetOpenCL(),memory_description) ||
|
| | | !RAGMemory.SetOnlineMemorySize(OnlineMemorySize))
|
| | | return INIT_FAILED;
|
| | | if(LoadBaseMemorySnapshot && RAGMemory.LoadPublishedSnapshot(BaseMemorySnapshotPath))
|
| | | BaseMemoryNullBound=false;
|
| | | else
|
| | | if(!RAGMemory.BindNullInference())
|
| | | return INIT_FAILED;
|
| | | if(!RebindOnlineRAGMemory())
|
| | | return INIT_FAILED;
|
| | | if(!bGradient.BufferInit(EmbeddingSize, 0.0f) ||
|
| | | !bGradient.BufferCreate(cActor[0].GetOpenCL()))
|
| | | return INIT_FAILED;
|
| | | cActor[0].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;
|
| | | }
|
| | | cActor[0].GetLayerOutput(0, Result);
|
| | | if(Result.Total() != AccountDescr)
|
| | | {
|
| | | PrintFormat("Input size of Actor doesn't match context description (%d <> %d)", Result.Total(), AccountDescr);
|
| | | return INIT_FAILED;
|
| | | }
|
| | | cStateEncoder.GetLayerOutput(0, Result);
|
| | | if(Result.Total() != (HistoryBars * BarDescr))
|
| | | {
|
| | | PrintFormat("Input size of StateEncoder doesn't match market state description (%d <> %d)", Result.Total(), (HistoryBars * BarDescr));
|
| | | return INIT_FAILED;
|
| | | }
|
| | | cStateEncoder.GetLayerOutput(StateTokenLayer, Result);
|
| | | if(Result.Total() != (BarDescr * EmbeddingSize))
|
| | | {
|
| | | PrintFormat("StateEncoder RankTCM layer doesn't match Critic context (%d <> %d)", Result.Total(), (BarDescr * EmbeddingSize));
|
| | | return INIT_FAILED;
|
| | | }
|
| | | cStateEncoder.GetLayerOutput(StateScenarioLayer, Result);
|
| | | if(Result.Total() != EmbeddingSize)
|
| | | {
|
| | | PrintFormat("StateEncoder token layer doesn't match pooled scenario embedding (%d <> %d)", Result.Total(), EmbeddingSize);
|
| | | return INIT_FAILED;
|
| | | }
|
| | | PrevBalance = AccountInfoDouble(ACCOUNT_BALANCE);
|
| | | PrevEquity = AccountInfoDouble(ACCOUNT_EQUITY);
|
| | | bFirstRun = true;
|
| | | return(INIT_SUCCEEDED);
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
| | | //| Expert deinitialization function |
|
| | | //+------------------------------------------------------------------+
|
| | | void OnDeinit(const int reason)
|
| | | {
|
| | | CollectOnlineCompletedScenario();
|
| | | if(OnlinePendingId!=0 && !OnlinePendingExecuted)
|
| | | RAGMemory.CancelPending(OnlinePendingId);
|
| | | if(reason!=REASON_INITFAILED && !SaveOnlineBaseMemory())
|
| | | Print("Error of save Base Memory snapshot");
|
| | | if(!(reason == REASON_RECOMPILE || reason == REASON_INITFAILED))
|
| | | {
|
| | | cActor[0].Save(FileName + "Act.nnw", 0, 0, 0, TimeCurrent(), true);
|
| | | cCritic[0].Save(FileName + CriticCheckpointFile, 0, 0, 0, TimeCurrent(), true);
|
| | | }
|
| | | DeleteObj(Result);
|
| | | DeleteObj(bAction);
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
| | | //| Expert tick function |
|
| | | //+------------------------------------------------------------------+
|
| | | void OnTick()
|
| | | {
|
| | | if(bFillStack)
|
| | | {
|
| | | if(!FillStack())
|
| | | {
|
| | | PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
| | | return;
|
| | | }
|
| | | bFillStack = false;
|
| | | }
|
| | | // Capture the live equity/margin observation on every tick while pending.
|
| | | SampleOnlinePendingScenario();
|
| | | if(!IsNewBar())
|
| | | return;
|
| | | int bars = CopyRates(Symb.Name(), TimeFrame, iTime(Symb.Name(), TimeFrame, 1), HistoryBars, Rates);
|
| | | if(bars < 0)
|
| | | {
|
| | | PrintFormat("%s -> %d CopyRates failed (%d)", __FUNCTION__, __LINE__, GetLastError());
|
| | | return;
|
| | | }
|
| | | if(!ArraySetAsSeries(Rates, true))
|
| | | {
|
| | | PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
| | | return;
|
| | | }
|
| | | RSI.Refresh();
|
| | | CCI.Refresh();
|
| | | ATR.Refresh();
|
| | | MACD.Refresh();
|
| | | Symb.Refresh();
|
| | | Symb.RefreshRates();
|
| | | bTime.Clear();
|
| | | bTime.Reserve(HistoryBars);
|
| | | if(!CreateBuffers(0, GetPointer(bState), GetPointer(bTime), (CBufferFloat*)NULL))
|
| | | {
|
| | | PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
| | | return;
|
| | | }
|
| | | double buy_value = 0, sell_value = 0, buy_profit = 0, sell_profit = 0;
|
| | | bool account_has_position = false;
|
| | | double position_discount = 0;
|
| | | double multiplier = 1.0 / (60.0 * 60.0 * 10.0);
|
| | | int total = PositionsTotal();
|
| | | datetime current = TimeCurrent();
|
| | | for(int i = 0; i < total; i++)
|
| | | {
|
| | | string position_symbol=PositionGetSymbol(i);
|
| | | if(position_symbol==NULL)
|
| | | continue;
|
| | | account_has_position=true;
|
| | | if(position_symbol != Symb.Name())
|
| | | continue;
|
| | | double profit = PositionGetDouble(POSITION_PROFIT);
|
| | | switch((int)PositionGetInteger(POSITION_TYPE))
|
| | | {
|
| | | case POSITION_TYPE_BUY:
|
| | | buy_value += PositionGetDouble(POSITION_VOLUME);
|
| | | buy_profit += profit;
|
| | | break;
|
| | | case POSITION_TYPE_SELL:
|
| | | sell_value += PositionGetDouble(POSITION_VOLUME);
|
| | | sell_profit += profit;
|
| | | break;
|
| | | }
|
| | | position_discount += (current - PositionGetInteger(POSITION_TIME)) * multiplier * MathAbs(profit);
|
| | | }
|
2026-08-13 23:48:47 +03:00 | | | // A bound position disappears only after its complete episode closes.
|
2026-08-10 03:17:55 +03:00 | | | CollectOnlineCompletedScenario();
|
| | | vector<float> account = vector<float>::Zeros(AccountDescr);
|
| | | account[0] = float(AccountInfoDouble(ACCOUNT_BALANCE) / EtalonBalance);
|
| | | account[1] = float((AccountInfoDouble(ACCOUNT_BALANCE) - PrevBalance) / MathMax(PrevBalance, 1.0));
|
| | | account[2] = float(AccountInfoDouble(ACCOUNT_EQUITY) / MathMax(PrevBalance, 1.0));
|
| | | account[3] = float((AccountInfoDouble(ACCOUNT_EQUITY) - PrevEquity) / MathMax(PrevEquity, 1.0));
|
| | | account[4] = (float)buy_value;
|
| | | account[5] = (float)sell_value;
|
| | | account[6] = float(buy_profit / MathMax(PrevBalance, 1.0));
|
| | | account[7] = float(sell_profit / MathMax(PrevBalance, 1.0));
|
| | | account[8] = float(position_discount / MathMax(PrevBalance, 1.0));
|
| | | double time = (double)Rates[0].time;
|
| | | double x = time / (double)(D'2024.01.01' - D'2023.01.01');
|
| | | account[9] = (float)MathSin(x != 0 ? 2.0 * M_PI * x : 0);
|
| | | x = time / (double)PeriodSeconds(PERIOD_MN1);
|
| | | account[10] = (float)MathCos(x != 0 ? 2.0 * M_PI * x : 0);
|
| | | x = time / (double)PeriodSeconds(PERIOD_W1);
|
| | | account[11] = (float)MathSin(x != 0 ? 2.0 * M_PI * x : 0);
|
| | | x = time / (double)PeriodSeconds(PERIOD_D1);
|
| | | account[12] = (float)MathSin(x != 0 ? 2.0 * M_PI * x : 0);
|
| | | if(!bContext.AssignArray(account))
|
| | | {
|
| | | PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
| | | return;
|
| | | }
|
| | | if(!cStateEncoder.feedForward((CBufferFloat*)GetPointer(bState), 1, false, (CBufferFloat*)NULL))
|
| | | {
|
| | | PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
| | | return;
|
| | | }
|
| | | if(!RAGMemory.Retrieve(GetPointer(cStateEncoder),StateScenarioLayer))
|
| | | {
|
| | | PrintFormat("%s -> %d RAG retrieval failed", __FUNCTION__, __LINE__);
|
| | | return;
|
| | | }
|
| | | if(!bFirstRun)
|
| | | {
|
2026-08-13 23:48:47 +03:00 | | | // Target Nets.
|
2026-08-10 03:17:55 +03:00 | | | if(!cActor[1].feedForward((CBufferFloat*)GetPointer(bContext), 1, false,
|
| | | GetPointer(cStateEncoder),StateScenarioLayer)
|
| | | || !cCritic[1].feedForward(GetPointer(cActor[1]), -1, GetPointer(cStateEncoder), StateTokenLayer)
|
| | | || !cCritic[2].feedForward(GetPointer(cActor[1]), -1, GetPointer(cStateEncoder), StateTokenLayer)
|
| | | )
|
| | | {
|
| | | PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
| | | bFirstRun = true;
|
| | | return;
|
| | | }
|
2026-08-13 23:48:47 +03:00 | | | // Critic.
|
2026-08-10 03:17:55 +03:00 | | | cCritic[1].getResults(Result);
|
| | | float forward = Result[0];
|
| | | cCritic[2].getResults(Result);
|
| | | forward = MathMin(forward, Result[0]);
|
| | | float reward = float(forward * DiscFactor +
|
| | | (PrevEquity - AccountInfoDouble(ACCOUNT_EQUITY) +
|
| | | PrevBalance - AccountInfoDouble(ACCOUNT_BALANCE)) * PrevBalance / EtalonBalance);
|
| | | if(MathMax(buy_value, sell_value) < Symb.LotsMin())
|
| | | {
|
| | | double marg = 0;
|
| | | if(!OrderCalcMargin(ORDER_TYPE_BUY, Symb.Name(), 1, Symb.Ask(), marg))
|
| | | marg = 200;
|
| | | double point_cost = Symb.TickValue() / Symb.TickSize();
|
| | | double loss = MathAbs(bState[0]) *
|
| | | point_cost * PrevBalance / (10 * marg);
|
| | | reward -= float(loss * PrevBalance / EtalonBalance);
|
| | | }
|
| | | Result.Clear();
|
| | | if(!Result.Add(reward)
|
| | | || !cCritic[0].backProp(Result, GetPointer(cStateEncoder), StateTokenLayer))
|
| | | {
|
| | | PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
| | | return;
|
| | | }
|
| | | if((PrevEquity - AccountInfoDouble(ACCOUNT_EQUITY) + PrevBalance - AccountInfoDouble(ACCOUNT_BALANCE)) <= 0 &&
|
| | | MathAbs(bState[0]) > Symb.StopsLevel()*Symb.Point())
|
| | | {
|
| | | double point_cost = Symb.TickValue() / Symb.TickSize();
|
| | | if(point_cost > 0)
|
| | | {
|
| | | vector<float> oracul = vector<float>::Zeros(NActions);
|
| | | double marg = 0;
|
| | | if(!OrderCalcMargin(ORDER_TYPE_BUY, Symb.Name(), 1, Symb.Ask(), marg))
|
| | | marg = 200;
|
| | | double lot = MathMax(PrevBalance / (10 * marg), Symb.LotsMin());
|
| | | if(bState[0] > 0)
|
| | | {
|
| | | lot = MathMin(lot, PrevBalance / ((-100 * bState[2 * HistoryBars] + Symb.Spread() * Symb.Point()) * point_cost));
|
| | | oracul[0] = (float)MathMin(lot, 1);
|
| | | oracul[1] = (float)MathMin(3 * bState[6] / (Symb.Point() * MaxTP), 1);
|
| | | oracul[2] = (float)MathMin(bState[6] / (Symb.Point() * MaxSL), 1);
|
| | | }
|
| | | else
|
| | | {
|
| | | lot = MathMin(lot, PrevBalance / ((100 * bState[HistoryBars] + Symb.Spread() * Symb.Point()) * point_cost));
|
| | | oracul[3] = (float)MathMin(lot, 1);
|
| | | oracul[4] = (float)MathMin(3 * bState[6] / (Symb.Point() * MaxTP), 1);
|
| | | oracul[5] = (float)MathMin(bState[6] / (Symb.Point() * MaxSL), 1);
|
| | | }
|
| | | reward = float(MathAbs(bState[0]) * point_cost * lot);
|
| | | reward = float(forward * DiscFactor + reward * PrevBalance / EtalonBalance);
|
| | | if(!Result.AssignArray(oracul) ||
|
| | | !cCritic[0].feedForward(Result, 1, false, (CNet*)GetPointer(cStateEncoder), StateTokenLayer))
|
| | | {
|
| | | PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
| | | bFirstRun = true;
|
| | | return;
|
| | | }
|
| | | if(!Result.Update(0, reward)
|
| | | || !cCritic[0].backProp(Result, GetPointer(cStateEncoder), StateTokenLayer))
|
| | | {
|
| | | PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
| | | bFirstRun = true;
|
| | | return;
|
| | | }
|
| | | }
|
| | | }
|
| | | if((MathRand() % ActorUpdate) == 0)
|
| | | {
|
| | | cCritic[0].TrainMode(false);
|
| | | if(!cCritic[0].feedForward(GetPointer(cActor[0]), -1, GetPointer(cStateEncoder), StateTokenLayer))
|
| | | {
|
| | | PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
| | | return;
|
| | | }
|
| | | cCritic[0].getResults(Result);
|
| | | CBufferFloat *actor_market_gradient=NULL;
|
| | | if(!cStateEncoder.GetLayerOutputDevice(StateScenarioLayer,actor_market_gradient))
|
| | | {
|
| | | PrintFormat("%s -> %d actor market gradient input failed", __FUNCTION__, __LINE__);
|
| | | return;
|
| | | }
|
| | | if(!Result.Update(0, float(MathMax(Result[0], 0) + (PrevBalance * 0.01 + PrevBalance) / EtalonBalance)) ||
|
| | | !cCritic[0].backProp(Result, GetPointer(cStateEncoder), StateTokenLayer) ||
|
| | | !cActor[0].backPropGradient(actor_market_gradient, GetPointer(bGradient), -1, true))
|
| | | {
|
| | | PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
| | | return;
|
| | | }
|
| | | cCritic[0].TrainMode(true);
|
| | | }
|
| | | if(PrevBalance < 20)
|
| | | ExpertRemove();
|
| | | }
|
2026-08-13 23:48:47 +03:00 | | | // New State.
|
2026-08-10 03:17:55 +03:00 | | | if(!cActor[0].feedForward((CBufferFloat*)GetPointer(bContext), 1, false,
|
| | | GetPointer(cStateEncoder),StateScenarioLayer)
|
| | | || !cCritic[0].feedForward((CNet*)GetPointer(cActor[0]), -1, (CNet*)GetPointer(cStateEncoder), StateTokenLayer)
|
| | | )
|
| | | {
|
| | | PrintFormat("%s -> %d", __FUNCTION__, __LINE__);
|
| | | return;
|
| | | }
|
| | | PrevBalance = AccountInfoDouble(ACCOUNT_BALANCE);
|
| | | PrevEquity = AccountInfoDouble(ACCOUNT_EQUITY);
|
| | | cActor[0].getResults(bAction);
|
| | | if(bAction.Total() < NActions)
|
| | | bAction.BufferInit(NActions, 0);
|
| | | double min_lot = Symb.LotsMin();
|
| | | double step_lot = Symb.LotsStep();
|
| | | double stops = (MathMax(Symb.StopsLevel(), 1) + Symb.Spread()) * Symb.Point();
|
| | | bool online_execution_attempted = false;
|
| | | if(bAction[0] >= bAction[3])
|
| | | {
|
| | | bAction.Update(0, (bAction[0] - bAction[3]));
|
| | | bAction.Update(3, 0);
|
| | | }
|
| | | else
|
| | | {
|
| | | bAction.Update(3, (bAction[3] - bAction[0]));
|
| | | bAction.Update(0, 0);
|
| | | }
|
2026-08-13 23:48:47 +03:00 | | | // Buy Control.
|
2026-08-10 03:17:55 +03:00 | | | if(bAction[0] < min_lot ||
|
| | | (bAction[1] * MaxTP * Symb.Point()) <= 2 * stops ||
|
| | | (bAction[2] * MaxSL * Symb.Point()) <= stops
|
| | | )
|
| | | {
|
| | | if(buy_value > 0)
|
| | | {
|
| | | const ENUM_VLADriverModifyResult buy_close=CloseOnlineByDirection(POSITION_TYPE_BUY);
|
| | | if((buy_close&VLADriverModifyRejected)!=0 && !PenalizeRejectedExecution(buy_value))
|
| | | return;
|
| | | }
|
| | | }
|
| | | else
|
| | | {
|
| | | double buy_lot = min_lot + MathRound((double)(bAction[0] - min_lot) / step_lot) * step_lot;
|
| | | double buy_tp = NormalizeDouble(Symb.Ask() + bAction[1] * MaxTP * Symb.Point(), Symb.Digits());
|
| | | double buy_sl = NormalizeDouble(Symb.Ask() - bAction[2] * MaxSL * Symb.Point(), Symb.Digits());
|
2026-08-13 23:48:47 +03:00 | | | bool buy_correction=false;
|
| | | ulong buy_correction_deal=0;
|
2026-08-10 03:17:55 +03:00 | | | ENUM_VLADriverModifyResult buy_modify=VLADriverModifyNoOp;
|
| | | if(buy_value>0)
|
| | | buy_modify=ModifyOnlinePosition(POSITION_TYPE_BUY,buy_sl,buy_tp);
|
| | | if((buy_modify&VLADriverModifyRejected)!=0 && !PenalizeRejectedExecution(buy_lot))
|
| | | return;
|
| | | if((buy_modify&VLADriverModifyAccepted)!=0)
|
| | | buy_correction=true;
|
| | | if(buy_value != buy_lot)
|
| | | {
|
| | | if((buy_value - buy_lot) >= min_lot)
|
| | | {
|
| | | const ENUM_VLADriverModifyResult buy_partial=CloseOnlinePartial(POSITION_TYPE_BUY,buy_value-buy_lot);
|
| | | if((buy_partial&VLADriverModifyRejected)!=0 && !PenalizeRejectedExecution(buy_value-buy_lot))
|
| | | return;
|
| | | if((buy_partial&VLADriverModifyAccepted)!=0)
|
| | | buy_correction=true;
|
| | | }
|
2026-08-13 23:48:47 +03:00 | | | else
|
| | | if((buy_lot-buy_value)>=min_lot)
|
| | | {
|
| | | // A same-direction volume increase is a correction of the current episode.
|
| | | if(buy_value>0 && sell_value<=0)
|
2026-08-10 03:17:55 +03:00 | | | {
|
2026-08-13 23:48:47 +03:00 | | | const ENUM_VLADriverModifyResult buy_increase=ClassifyOnlineTradeResult(
|
| | | Trade.Buy(buy_lot-buy_value,Symb.Name(),Symb.Ask(),buy_sl,buy_tp));
|
| | | if((buy_increase&VLADriverModifyRejected)!=0 &&
|
| | | !PenalizeRejectedExecution(buy_lot-buy_value))
|
2026-08-10 03:17:55 +03:00 | | | return;
|
2026-08-13 23:48:47 +03:00 | | | if((buy_increase&VLADriverModifyAccepted)!=0)
|
| | | {
|
| | | buy_correction_deal=Trade.ResultDeal();
|
| | | if(buy_correction_deal>0)
|
| | | buy_correction=true;
|
| | | else
|
| | | Print("Accepted BUY volume correction has no deal and is not added to RAG memory");
|
2026-08-10 03:17:55 +03:00 | | | }
|
| | | }
|
| | | else
|
2026-08-13 23:48:47 +03:00 | | | if(buy_value<=0 && sell_value<=0 && !account_has_position && OnlinePendingId==0 &&
|
| | | !online_execution_attempted && AnchorOnlinePendingScenario())
|
| | | {
|
| | | online_execution_attempted = true;
|
| | | if(Trade.Buy(buy_lot-buy_value,Symb.Name(),Symb.Ask(),buy_sl,buy_tp))
|
| | | {
|
| | | if(!OpenConfirmedOnlinePendingScenario(Trade.ResultDeal()))
|
| | | {
|
| | | if(!PenalizeRejectedExecution(buy_lot-buy_value))
|
| | | return;
|
| | | ExpertRemove();
|
| | | return;
|
| | | }
|
| | | }
|
| | | else
|
| | | {
|
| | | ClearOnlinePendingScenario();
|
| | | if(!PenalizeRejectedExecution(buy_lot-buy_value))
|
| | | return;
|
| | | }
|
| | | }
|
2026-08-10 03:17:55 +03:00 | | | }
|
2026-08-13 23:48:47 +03:00 | | | if(buy_correction && OnlinePendingId!=0 &&
|
| | | !RegisterOnlinePendingCorrection(buy_correction_deal))
|
2026-08-10 03:17:55 +03:00 | | | {
|
| | | PrintFormat("%s -> %d accepted BUY RAG correction registration failed",__FUNCTION__,__LINE__);
|
| | | ExpertRemove();
|
| | | return;
|
| | | }
|
| | | }
|
| | | }
|
2026-08-13 23:48:47 +03:00 | | | // Sell Control.
|
2026-08-10 03:17:55 +03:00 | | | if(bAction[3] < min_lot ||
|
| | | (bAction[4] * MaxTP * Symb.Point()) <= 2 * stops ||
|
| | | (bAction[5] * MaxSL * Symb.Point()) <= stops
|
| | | )
|
| | | {
|
| | | if(sell_value > 0)
|
| | | {
|
| | | const ENUM_VLADriverModifyResult sell_close=CloseOnlineByDirection(POSITION_TYPE_SELL);
|
| | | if((sell_close&VLADriverModifyRejected)!=0 && !PenalizeRejectedExecution(sell_value))
|
| | | return;
|
| | | }
|
| | | }
|
| | | else
|
| | | {
|
| | | double sell_lot = min_lot + MathRound((double)(bAction[3] - min_lot) / step_lot) * step_lot;
|
| | | double sell_tp = NormalizeDouble(Symb.Bid() - bAction[4] * MaxTP * Symb.Point(), Symb.Digits());
|
| | | double sell_sl = NormalizeDouble(Symb.Bid() + bAction[5] * MaxSL * Symb.Point(), Symb.Digits());
|
2026-08-13 23:48:47 +03:00 | | | bool sell_correction=false;
|
| | | ulong sell_correction_deal=0;
|
2026-08-10 03:17:55 +03:00 | | | ENUM_VLADriverModifyResult sell_modify=VLADriverModifyNoOp;
|
| | | if(sell_value>0)
|
| | | sell_modify=ModifyOnlinePosition(POSITION_TYPE_SELL,sell_sl,sell_tp);
|
| | | if((sell_modify&VLADriverModifyRejected)!=0 && !PenalizeRejectedExecution(sell_lot))
|
| | | return;
|
| | | if((sell_modify&VLADriverModifyAccepted)!=0)
|
| | | sell_correction=true;
|
| | | if(sell_value != sell_lot)
|
| | | {
|
| | | if((sell_value - sell_lot) >= min_lot)
|
| | | {
|
| | | const ENUM_VLADriverModifyResult sell_partial=CloseOnlinePartial(POSITION_TYPE_SELL,sell_value-sell_lot);
|
| | | if((sell_partial&VLADriverModifyRejected)!=0 && !PenalizeRejectedExecution(sell_value-sell_lot))
|
| | | return;
|
| | | if((sell_partial&VLADriverModifyAccepted)!=0)
|
| | | sell_correction=true;
|
| | | }
|
2026-08-13 23:48:47 +03:00 | | | else
|
| | | if((sell_lot-sell_value)>=min_lot)
|
| | | {
|
| | | // A same-direction volume increase is a correction of the current episode.
|
| | | if(sell_value>0 && buy_value<=0)
|
2026-08-10 03:17:55 +03:00 | | | {
|
2026-08-13 23:48:47 +03:00 | | | const ENUM_VLADriverModifyResult sell_increase=ClassifyOnlineTradeResult(
|
| | | Trade.Sell(sell_lot-sell_value,Symb.Name(),Symb.Bid(),sell_sl,sell_tp));
|
| | | if((sell_increase&VLADriverModifyRejected)!=0 &&
|
| | | !PenalizeRejectedExecution(sell_lot-sell_value))
|
2026-08-10 03:17:55 +03:00 | | | return;
|
2026-08-13 23:48:47 +03:00 | | | if((sell_increase&VLADriverModifyAccepted)!=0)
|
| | | {
|
| | | sell_correction_deal=Trade.ResultDeal();
|
| | | if(sell_correction_deal>0)
|
| | | sell_correction=true;
|
| | | else
|
| | | Print("Accepted SELL volume correction has no deal and is not added to RAG memory");
|
2026-08-10 03:17:55 +03:00 | | | }
|
| | | }
|
| | | else
|
2026-08-13 23:48:47 +03:00 | | | if(buy_value<=0 && sell_value<=0 && !account_has_position && OnlinePendingId==0 &&
|
| | | !online_execution_attempted && AnchorOnlinePendingScenario())
|
| | | {
|
| | | online_execution_attempted = true;
|
| | | if(Trade.Sell(sell_lot-sell_value,Symb.Name(),Symb.Bid(),sell_sl,sell_tp))
|
| | | {
|
| | | if(!OpenConfirmedOnlinePendingScenario(Trade.ResultDeal()))
|
| | | {
|
| | | if(!PenalizeRejectedExecution(sell_lot-sell_value))
|
| | | return;
|
| | | ExpertRemove();
|
| | | return;
|
| | | }
|
| | | }
|
| | | else
|
| | | {
|
| | | ClearOnlinePendingScenario();
|
| | | if(!PenalizeRejectedExecution(sell_lot-sell_value))
|
| | | return;
|
| | | }
|
| | | }
|
2026-08-10 03:17:55 +03:00 | | | }
|
2026-08-13 23:48:47 +03:00 | | | if(sell_correction && OnlinePendingId!=0 &&
|
| | | !RegisterOnlinePendingCorrection(sell_correction_deal))
|
2026-08-10 03:17:55 +03:00 | | | {
|
| | | PrintFormat("%s -> %d accepted SELL RAG correction registration failed",__FUNCTION__,__LINE__);
|
| | | ExpertRemove();
|
| | | return;
|
| | | }
|
| | | }
|
| | | }
|
| | | if((OnlinePendingId!=0 && !OnlinePendingExecuted) ||
|
| | | (OnlinePendingPrepared && OnlinePendingId==0))
|
| | | {
|
| | | if(OnlinePendingId!=0)
|
| | | RAGMemory.CancelPending(OnlinePendingId);
|
| | | ClearOnlinePendingScenario();
|
| | | }
|
| | | bFirstRun = false;
|
| | | if((int(Rates[0].time / PeriodSeconds(TimeFrame)) % TargetUpdate) == 0)
|
| | | {
|
| | | if(MathRand() / 32767.0 > 0.5)
|
| | | cCritic[1].WeightsUpdate(GetPointer(cCritic[0]), tau);
|
| | | else
|
| | | cCritic[2].WeightsUpdate(GetPointer(cCritic[0]), tau);
|
| | | cActor[1].WeightsUpdate(GetPointer(cActor[0]), tau);
|
| | | }
|
| | | }
|
| | | bool FillStack(void)
|
| | | {
|
| | | int start = StackSize + HistoryBars;
|
| | | int end = 0;
|
| | | int bars = CopyRates(Symb.Name(), TimeFrame, 2, start, Rates);
|
| | | if(bars < 0)
|
| | | ReturnFalse;
|
| | | if(!RSI.BufferResize(bars) || !CCI.BufferResize(bars) ||
|
| | | !ATR.BufferResize(bars) || !MACD.BufferResize(bars))
|
| | | ReturnFalse;
|
| | | if(RSI.BarsCalculated() < bars ||
|
| | | CCI.BarsCalculated() < bars ||
|
| | | ATR.BarsCalculated() < bars ||
|
| | | MACD.BarsCalculated() < bars)
|
| | | ReturnFalse;
|
| | | RSI.Refresh();
|
| | | CCI.Refresh();
|
| | | ATR.Refresh();
|
| | | MACD.Refresh();
|
| | | if(!ArraySetAsSeries(Rates, true))
|
| | | ReturnFalse;
|
| | | bars -= end + HistoryBars;
|
| | | if(bars < 0)
|
| | | ReturnFalse;
|
| | | vector<float> result, target, neg_target;
|
| | | uint ticks = GetTickCount();
|
| | | uint tester_print = 0;
|
| | | for(uint i = 0; i < cActor.Size(); i++)
|
| | | {
|
| | | if(!cActor[i].Clear())
|
| | | ReturnFalse;
|
| | | cActor[i].TrainMode(false);
|
| | | }
|
| | | for(uint i = 0; i < cCritic.Size(); i++)
|
| | | {
|
| | | if(!cCritic[i].Clear())
|
| | | ReturnFalse;
|
| | | cCritic[i].TrainMode(false);
|
| | | }
|
| | | if(!cStateEncoder.Clear())
|
| | | ReturnFalse;
|
| | | cStateEncoder.TrainMode(false);
|
| | | for(int posit = start - HistoryBars - 1; (posit >= end && !IsStopped()); posit--)
|
| | | {
|
| | | if(!CreateBuffers(posit, GetPointer(bState), GetPointer(bTime), NULL))
|
| | | ReturnFalse;
|
| | | vector<float> account = vector<float>::Zeros(AccountDescr);
|
| | | account[0] = float(AccountInfoDouble(ACCOUNT_BALANCE) / EtalonBalance);
|
| | | account[2] = 1;
|
| | | double time = (double)bTime[0];
|
| | | double x = time / (double)(D'2024.01.01' - D'2023.01.01');
|
| | | account[9] = (float)MathSin(x != 0 ? 2.0 * M_PI * x : 0);
|
| | | x = time / (double)PeriodSeconds(PERIOD_MN1);
|
| | | account[10] = (float)MathCos(x != 0 ? 2.0 * M_PI * x : 0);
|
| | | x = time / (double)PeriodSeconds(PERIOD_W1);
|
| | | account[11] = (float)MathSin(x != 0 ? 2.0 * M_PI * x : 0);
|
| | | x = time / (double)PeriodSeconds(PERIOD_D1);
|
| | | account[12] = (float)MathSin(x != 0 ? 2.0 * M_PI * x : 0);
|
| | | if(!bContext.AssignArray(account))
|
| | | ReturnFalse;
|
| | | if(!cStateEncoder.feedForward((CBufferFloat*)GetPointer(bState), 1, false, (CBufferFloat*)NULL))
|
| | | ReturnFalse;
|
| | | if(!RAGMemory.Retrieve(GetPointer(cStateEncoder),StateScenarioLayer))
|
| | | {
|
| | | PrintFormat("%s -> %d RAG retrieval failed", __FUNCTION__, __LINE__);
|
| | | ReturnFalse;
|
| | | }
|
2026-08-13 23:48:47 +03:00 | | | // Feed Forward.
|
2026-08-10 03:17:55 +03:00 | | | for(uint i = 0; (i < cActor.Size() && !IsStopped()); i++)
|
| | | if(!cActor[i].feedForward((CBufferFloat*)GetPointer(bContext), 1, false,
|
| | | GetPointer(cStateEncoder),StateScenarioLayer))
|
| | | ReturnFalse;
|
| | | for(uint i = 0; (i < cCritic.Size() && !IsStopped()); i++)
|
| | | if(!cCritic[i].feedForward(GetPointer(cActor[int(i > 0)]), -1,
|
| | | GetPointer(cStateEncoder), StateTokenLayer))
|
| | | ReturnFalse;
|
| | | if(GetTickCount() - ticks > 500)
|
| | | {
|
| | | double percent = (1.0 - double(posit - end) / (start - end - HistoryBars - NForecast)) * 100.0;
|
| | | string str = StringFormat("%-12s %6.2f%%", "Fill stack", percent);
|
| | | if(BaseMemoryNullBound)
|
| | | str += "\nRAG Null Memory";
|
| | | Comment(str);
|
| | | ticks = GetTickCount();
|
| | | if(MQLInfoInteger(MQL_TESTER) && percent >= tester_print)
|
| | | {
|
| | | Print(str);
|
| | | tester_print += 10;
|
| | | }
|
| | | }
|
| | | }
|
| | | cActor[0].TrainMode(true);
|
| | | cCritic[0].TrainMode(true);
|
| | | Comment("");
|
| | | if(MQLInfoInteger(MQL_TESTER))
|
| | | Print("Fill stack Done");
|
| | | return true;
|
| | | }
|
| | | //+------------------------------------------------------------------+
|
2026-08-13 23:48:47 +03:00 | | |
|