forked from nique_372/AiDataTaskRuner
5.4 KiB
5.4 KiB
Automate data generation and AI model training through a simple, easy-to-use GUI.
Features
- Data generation | Main tab — choose timeframe, symbol, date range and add tasks to the queue.
- Data generation | Tasks tab — view and manage all pending, running and completed tasks.
- Settings tab — choose the interface language.
- Training tab | Run — Choose the py trainer settings and run it
- Training tab | Results — View the results of the py training in real time
- Feature editor — Modify\Compile (csv)\Save\Load\etc the features that your AI model will use through a simple editor for the FGBLC DSL
- Utils tab — Encrypt/Decrypt/Clean folders Useful if you want to distribute the onnx files, etc., generated by the panel
- Workflows tab — Run workflows without touching the panel Useful for the mass automation of data generation/training of ML models
- AI tab — configure the LLM model and control the panel via Telegram in natural language Or use MCP.
Currently supported via Telegram and MCP:
- Query task status and count
- Add, save and load tasks
- Execute and clean all tasks
- Load and save panel configuration
- Query panel folders and settings
- Query MT5 logs, manage files, check system time
- Results from panel operations are delivered asynchronously directly to the chat.
- A subtab has now also been added to allow the use of MCP with the panel.
Repository Structure
AiDataTaskRuner/
├── Backend/ # The backend of the panel is where the code that the tester executes for data generation is located.
├── FileInstaler/ # This is where the panel's file installer is located... such as ini files, etc.
├── DllFilseSelctDialog/ # C++ code for the DLL (this allows the use of a custom file select dialog without sandbox limitations)
├── UI/ # Frontend code for the panel (tabs and main window)
├── Images/ # Repository banner
└── Utils/ # Utility scripts for working with .set
Requirements
- For final users:
- ICTLibrary (https://www.thebotplace.com/bot/ict-fastlibrary)
- FastNL (https://www.thebotplace.com/bot/fast-newslibrary)
- McpServerByLeo (https://www.thebotplace.com/bot/mcpserverbyleo)
- All files instaled from releases
- Py server from (and all py requeriments): https://forge.mql5.io/nique_372/AiDataTaskRunerMcp
- For devs:
- Check dependencies.json
Installation of repo code
cd "C:\Users\YOUR USER\AppData\Roaming\MetaQuotes\Terminal\YOUR ID\MQL5\Shared Projects"
tsndep install "https://forge.mql5.io/nique_372/AiDataTaskRuner.git"
- For use tsndep command requerid tsndep pacakage (avaible in pypi).. This command automatically downloads all dependencies and installs all requirements from the repositories.
- If any part of the system is private, then it will fail... contact me so I can give you access (if it's a product, you can buy it; if you have any questions, don't hesitate to contact me).
Quick Start for final users
1. Install
Install the 6 files located in the releases section (most recent ones preferred).
- Place all EAs (5) .ex5 files in your MQL5\Experts\ folder
- Place the DLL DllFileSelctDialog.dll in your MQL5\Libraries\ folder
- Place the executable file py_trainer.exe in a folder relative to your terminal path or a custom location.
2. Check the wiki's quick start
Visit the wiki: https://forge.mql5.io/nique_372/AiDataTaskRuner/wiki/QuickStartPanelUse
Documentation
Full documentation including tab-by-tab guides, parameter reference and configuration examples is available in the Wiki.
Disclaimer
Trading involves substantial risk of loss.
- This software is a technical tool, not financial advice
- Past performance does not guarantee future results
- You are solely responsible for your trading decisions
- Always test thoroughly before deploying with real capital
- Apply appropriate risk management at all times
The authors assume no liability for trading losses, system failures, or any damages arising from the use of this software.
License
Copyright © 2026 Nique-Leo.