Upgrade Existing MT5 CatBoost EA

작업 종료됨

실행 시간 5 일
고객의 피드백
Great work! A highly professional developer.
피고용인의 피드백
Very considerate and give you the best offer.

명시

Upgrade Existing MT5 CatBoost Angle EA → Add Self-Learning Engine + Multi-Symbol + Prop-Firm Risk Layer (with 5-Year High-Quality Tick Backtests)

Project Overview

I already have a fully functional MT5 EA based on a CatBoost Angle Strategy (3-model ensemble).
This project is NOT about rewriting from scratch.

I need an experienced Senior MQL5 Developer to upgrade and extend the EA’s architecture by integrating:

A Self-Training Engine (pattern-score reinforcement learning)

Multi-Symbol independent learning (6 symbols)

A full Prop-Firm Challenge Risk-Control Layer

High-quality (99% modeling) tick-data backtests for 5+ years

Clean, modular OOP architecture

CSV-based model persistence per symbol

All CatBoost logic, angle features, and ensemble prediction must be kept exactly as they are.
Your job is to extend the EA with new modules and integrate them cleanly.

Scope of Work
1. Integrate the Self-Training Engine (Reinforcement-Style, MT5 Internal)

No Python needed.
All learning is inside MQL5.

Includes:

Feature builder
Feature binning → pattern hashing
Pattern table per symbol
Online score update (win/loss smoothing + decay)
MAE/MFE tracking
Model persistence (CSV)
Learning must run in parallel with CatBoost, not replacing it.
Final Decision = CatBoost Prediction + Pattern Score + Risk Layer

2. Multi-Symbol Learning Engine
Must support the following symbols independently:
XAUUSD
EURUSD
GBPUSD
USDJPY
US30
DAX40

Each symbol has its own:
Pattern table
MAE/MFE
Model CSV
Score history
Learning cycle
No cross-contamination allowed.

3. Prop-Firm Challenge Risk Layer
Must include:
Account-Level
Daily loss limit
Max drawdown limit
Equity guard
Account shutdown mode
Execution Level
Spread filter
Slippage filter
News/time blocked windows
Min seconds between trades (anti-HFT)
Max lots per symbol
Max total exposure
Strategy-Level
Swing mode (minimum holding time)
Risk Layer must wrap the trading engine:
CatBoost → Self-Training → RiskEngine.AllowNewTrade → Execution

4. High-Quality Tick-Data Backtests (Mandatory)
Modeling Quality: ≥ 99%

Symbols (≥ 5 years each):
XAUUSD
EURUSD
GBPUSD
USDJPY
US30
DAX40

Timeframes:
M15
H1

Files required per backtest:
HTML report
Equity curve PNG
Monthly returns table
Raw trade history CSV
Tick-data source info
Modeling quality screenshot

5. Architecture Requirements

Must follow a clean module design:
[ Prop-Firm Risk Layer ]
        ↓ allow/deny
[ CatBoost Predictions ]
        ↓
[ Self-Training Engine ]
  ↓ Feature Builder
  ↓ Pattern Hash
  ↓ Pattern Table (per symbol)
  ↓ MAE/MFE Tracking
  ↓ CSV Persistence


No monolithic OnTick code.
Clean classes required.

Deliverables

Updated EA (.mq5 + .ex5)

Self-Training Engine (integrated with existing CatBoost logic)

Multi-Symbol learning engine

Prop-firm-safe risk layer

Model persistence per symbol

All 12 required backtests (6 symbols × 2 TF × 5 years)

Documentation

Clean, extensible OOP architecture

응답함

1
개발자 1
등급
(22)
프로젝트
28
7%
중재
9
33% / 33%
기한 초과
1
4%
작업중
2
개발자 2
등급
(4)
프로젝트
8
0%
중재
3
33% / 67%
기한 초과
4
50%
무료
3
개발자 3
등급
(3)
프로젝트
1
100%
중재
3
0% / 100%
기한 초과
0
무료
4
개발자 4
등급
(177)
프로젝트
189
47%
중재
3
33% / 33%
기한 초과
1
1%
로드됨
5
개발자 5
등급
(4)
프로젝트
3
33%
중재
2
0% / 100%
기한 초과
0
무료
6
개발자 6
등급
(52)
프로젝트
64
39%
중재
15
27% / 60%
기한 초과
1
2%
작업중
7
개발자 7
등급
프로젝트
0
0%
중재
0
기한 초과
0
무료
8
개발자 8
등급
프로젝트
0
0%
중재
1
0% / 100%
기한 초과
0
무료
9
개발자 9
등급
(33)
프로젝트
34
41%
중재
0
기한 초과
3
9%
작업중
10
개발자 10
등급
(1)
프로젝트
1
100%
중재
0
기한 초과
0
무료
11
개발자 11
등급
(1)
프로젝트
2
0%
중재
1
0% / 100%
기한 초과
1
50%
무료
게재됨: 4 코드
12
개발자 12
등급
(29)
프로젝트
33
27%
중재
20
10% / 50%
기한 초과
11
33%
무료
13
개발자 13
등급
(271)
프로젝트
553
50%
중재
57
40% / 37%
기한 초과
227
41%
작업중
14
개발자 14
등급
프로젝트
0
0%
중재
0
기한 초과
0
무료
15
개발자 15
등급
프로젝트
2
0%
중재
4
25% / 50%
기한 초과
1
50%
무료
16
개발자 16
등급
(24)
프로젝트
32
38%
중재
4
50% / 25%
기한 초과
5
16%
작업중
17
개발자 17
등급
(2)
프로젝트
3
0%
중재
1
100% / 0%
기한 초과
0
무료
18
개발자 18
등급
(14)
프로젝트
14
36%
중재
1
0% / 100%
기한 초과
1
7%
작업중
19
개발자 19
등급
(1)
프로젝트
1
0%
중재
0
기한 초과
0
무료
20
개발자 20
등급
프로젝트
1
100%
중재
0
기한 초과
0
무료
게재됨: 3 코드
21
개발자 21
등급
프로젝트
0
0%
중재
0
기한 초과
0
무료
비슷한 주문
UnixFirmanMln 30 - 200 USD
I am an MQL5 Developer specializing in Forex trading automation, Expert Advisors, custom indicators, and trading strategy optimization. I provide services for developing, debugging, modifying, and improving trading robots and indicators. My technical skills include MQL5, C++, Linux, algorithmic trading logic, debugging, and performance optimization. I pay close attention to trading rules, execution logic, risk
it starts with a directional Buy or Sell entry, then uses grid/martingale for recovery if needed. X also has updated risk controls and filters. Only Experts Allowed please. No Newbies also Drop your Quotes
I DO NOT need any programming or strategy development. I already have a working NinjaTrader 8 automated strategy based on a 3/5 EMA crossover. I need you to run my existing strategy through NinjaTrader Strategy Analyzer/Optimizer, test the existing adjustable parameters, and find robust settings with the best profit factor and lowest reasonable drawdown. I will provide the existing NinjaScript ZIP. I do not want the
I am looking to acquire an EXISTING and PROVEN MetaTrader 5 Expert Advisor. I am NOT looking for someone to develop a random new strategy from scratch. The objective is to find a robust EA with an existing track record, verify its performance and robustness, and purchase the MQL5 source code together with clearly defined commercial rights. MAIN REQUIREMENTS The EA should meet most or all of the following
Looking to acquire an existing, proven EA. Not looking for new strategy development. REQUIREMENTS — non-negotiable: Minimum 12 months verified LIVE history (Myfxbook or MQL5 Signals, real money — not demo, not backtest) Hard stop loss on EVERY trade, set at order placement NO martingale, NO grid, NO averaging into losers, NO lot multiplication after a loss Max historical drawdown under 30% Minimum 5 trades per
Platform: TradingView Programming Language: Pine Script v6 Type: Custom Indicator Project Name: Scalping Reaction Zones + Valid Order Blocks MAIN GOAL I need a custom TradingView indicator for scalping. The indicator must detect: 1. Reaction / Explosion Zones 2. Valid Order Blocks 3. Combined Reaction Zone + Order Block zones The indicator should NOT generate: Buy signals Sell signals TP SL Entry signals Trading
Hello, I trade on HFM MT5. Need SCALPING EA for scalping, budget $30. SYMBOLS: XAUUSD (GOLD), US30, NAS100, GER40, EURUSD TIMEFRAME: M1 and M5 PLATFORM: MT5, must work on HFM VPS STRATEGY: - EMA 9 crosses EMA 21 - RSI 14: BUY only if RSI >45, SELL only if RSI <55 - Bollinger Bands: Only trade when price returns inside BB after breakout - 1 trade at a time per symbol - SL: 150 points Gold / 100 points indices (must be
Project Description I am looking for an experienced MQL5/MT5 Expert Advisor developer to develop an automated trading EA for XAUUSD on the M3 timeframe , running on an Exness account . The EA will automate a manual strategy based on SNR/GAP zones , using two setup types: Price Rejection Price Correction The EA should identify valid BUY/SELL setups around predefined SNR/GAP areas, apply configurable RSI, SMA, EMA and
I need an MT5 Expert Advisor for XAUUSD (Gold) running on an Exness account. STRATEGY LOGIC The exact entry and exit rules will be provided in writing before development starts. Please quote for a single rule-based strategy with clearly separated logic, so the rules can be adjusted later without rewriting the whole EA. RISK MANAGEMENT (this is the priority of the project) - Risk per trade as a percentage of account
Modification of Trade_Panel_7 7.05 Objective . I have this utility “Trade_Panel_7 7.05” with which I open and close positions. Please refer to attached “Trade Panel Modification.jpg”. It shows an input parameter “Volume Step”. This parameter has the initial value of 0.02 as shown. This feature needs to be amended. Present Status . Value of lot size is increased by this “Volume Step”. I have built a whole series of

프로젝트 정보

예산
550 - 700 USD
기한
 10 일