Forex Trading Bot with Transformer and EMA

This project is a forex trading bot that connects to the IQ Option API, uses a Transformer model to predict market movements, and incorporates an Exponential Moving Average (EMA) strategy to make informed trading decisions. The bot operates only when the predicted probability, combined with the EMA trend, exceeds a 70% confidence threshold.

Features

Real-time connection to IQ Option: The bot fetches real-time candle data (OHLC) from IQ Option and processes it.
Transformer model for prediction: A trained Transformer model predicts whether the market will rise (call) or fall (put).
EMA (Exponential Moving Average): The bot uses a 20-period EMA to adjust its decision-making process.
Confidence threshold: The bot only operates when the probability of a successful trade exceeds 70%.

Requirements

Python 3.7+
An IQ Option account (credentials stored securely in a .env file)
Required Python libraries (see installation below)

Installation

Clone the repository:

git clone https://github.com/Seem1019/forexia.git
cd forex-trading-bot

Set up a virtual environment (optional but recommended):

python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

Install dependencies:
```
pip install -r requirements.txt
```
Set up environment variables:
- Create a .env file in the root directory and add your IQ Option credentials:
```
IQ_OPTION_EMAIL=your-email@example.com
IQ_OPTION_PASSWORD=your-password
```
Download the trained Transformer model:
- Ensure the trained model (transformer_model.pth) is available in the models/ directory.

How to Use

To start the bot, execute the mainia.py file:

python mainia.py

The bot will:

Connect to the IQ Option API.
Fetch real-time candle data for the specified currency pair (e.g., EUR/USD OTC).
Use the Transformer model to predict whether to make a call (buy) or put (sell) based on the current market.
Adjust the prediction using the EMA to determine the market trend.
Only place a trade if the adjusted prediction probability is greater than 70%.

How It Works

Transformer Model

The bot uses a Transformer-based model trained on historical candle data to predict market movements.
It processes 20 consecutive candles (OHLC data) and outputs a prediction for whether the price will rise (call) or fall (put).

EMA (Exponential Moving Average)

The bot calculates a 20-period EMA using the closing prices of the most recent candles.
If the current price is above the EMA, it indicates an uptrend, and a bias of +0.2 is added to the model’s prediction.
If the current price is below the EMA, it indicates a downtrend, and a bias of -0.2 is applied.

Confidence Threshold

After combining the model’s prediction with the EMA bias, the bot operates only if the final probability is greater than 70%.
If the probability is lower, no trade is executed, and the bot waits for new data.

Customization

Modify Trading Pair

You can change the trading pair (e.g., EUR/USD, USD/JPY) directly in the mainia.py file:
```
ciclo_de_operaciones_transformer(iq_handler, "EURUSD-OTC", model)
```

Adjust the EMA Period

To change the EMA period from the default of 20 to another value, modify the calcular_ema function in mainia.py:
```
ema_actual = calcular_ema(close_prices, timeperiod=20)
```

Set a Different Confidence Threshold

You can adjust the confidence threshold for placing trades by changing this value in the hacer_inferencia_transformer function:
```
if probabilidad >= 0.7:
    # Proceed with trade
```

Example Output

Upon running the bot, you should see output like the following:

Connected to IQ Option successfully.
EURUSD-OTC is available for trading.
Fetching 25 candles for EURUSD-OTC...
Logits: 0.62, Probability: 0.72
Uptrend (price above EMA). Adjusted probability: 0.92
Action: Call

Disclaimer

This project is for educational purposes only. Trading involves substantial risk and is not suitable for every investor. You are solely responsible for any trades or decisions made based on the results of this bot.