🧠 NNTicks

NNTicks is a desktop neural-network-based tick predictor built in Python. It’s designed for binary tick trading and supports real-time predictions, neural network training, backtesting, and live visualization — all within a sleek PyQt6 interface.

---

🚧 Project Status

This project is currently in active development.
Features, structure, and UI may change. Contributions and feedback are welcome!

---

🖼️ Screenshots

📈 Live Tick Chart & Prediction Output

📊 Candlestick Prediction Mode

🧪 Neural Network Training Interface

⚙️ Settings Panel

---

🧩 Architecture Diagram (Text-Based)

                          +-----------------+
                          |    config.py    |
                          |-----------------|
                          | API keys,       |
                          | symbol,         |
                          | constants       |
                          +--------+--------+
                                   |
                                   v
+-------------------+      +--------------------+      +---------------------+
|     gui.py        |----->|   data_handler.py  |----->|   prediction.py     |
|-------------------|      |--------------------|      |---------------------|
| Starts interface  |      | Connects to market |      | Predicts tick dir.  |
| & logic           |      | Handles WebSocket  |      | Neural Net training |
+-------------------+      +--------------------+      +---------------------+
                                                             |
                                                             v
                                                      +--------------+
                                                      |  model.pkl   |
                                                      | (saved NN)   |
                                                      +--------------+
                                  GUI
                                   |
                                   v
                          +--------------------+
                          |     PyQt6 GUI      |
                          |--------------------|
                          | Charts, logs,      |
                          | buttons, input     |
                          +--------------------+

---

🔧 Features

• 📡 Real-time WebSocket tick data from Deriv
• 🧠 Neural network prediction with confidence threshold
• 🖥️ Fully interactive PyQt6 GUI
• 📈 Tick and candlestick charting modes
• 🧪 Training mode with XP & level system
• 🕹️ Manual Rise/Fall data entry
• 💾 Save/Load trained models (model.pkl)
• 📤 Export trade logs to PDF
• 🌙 Dark theme with clear, modern visuals

---

📦 Tech Stack

Component	Tech Used
Language	Python
GUI Framework	PyQt6
ML Engine	Custom NumPy-based Neural Net
Data Source	Deriv.com WebSocket API
Charting	Matplotlib (for visual plots)

---

🚀 How to Run

1. Clone the repository:

   git clone https://github.com/Cruelhelp/NNTicks.git
   cd NNTicks

2. Install dependencies:

   pip install -r requirements.txt

3. Launch the app:

   python gui.py

Make sure your API key and WebSocket URL are correctly set in config.py
or enter the details in Settings > WebSocket.

---

📁 File Structure

File	Description
gui.py	Launches the application and GUI logic
data_handler.py	Handles WebSocket connection and data streaming
prediction.py	Neural network logic for prediction and training
config.py	Stores constants, API keys, and symbol info
model.pkl	Auto-saved trained model
screenshots/	Folder for UI preview images
requirements.txt	Python dependencies list
DEVLOG.md	Project development log

---

📓 Development Log

See detailed progress, ongoing features, and upcoming goals in the DEVLOG.md.

---

👤 Author

Ruel McNeil
📧 [email protected]
🔗 GitHub Profile

---

🧾 License

MIT License
Developed by Ruel McNeil | Powered by Grok (xAI)