# Create a virtual environment (recommended)
python -m venv trading_env
source trading_env/bin/activate # Linux/Mac
trading_env\Scripts\activate   # Windows

# Install core libraries
pip install pandas numpy scikit-learn
pip install MetaTrader5 ccxt
pip install ta-lib yfinance
pip install tensorflow keras
pip install matplotlib seaborn
pip install backtrader

import MetaTrader5 as mt5
import pandas as pd
from datetime import datetime

# Initialize MT5 connection
if not mt5.initialize():
  print("MT5 initialization failed")
  mt5.shutdown()

# Fetch 5 years of H4 data for EURUSD
rates = mt5.copy_rates_range(
  "EURUSD",
  mt5.TIMEFRAME_H4,
  datetime(2020, 1, 1),
  datetime.now()
)

# Convert to DataFrame and clean
df = pd.DataFrame(rates)
df['time'] = pd.to_datetime(df['time'], unit='s')
df.set_index('time', inplace=True)
df.dropna(inplace=True)

print(df.head())
print(f"Total rows: {len(df)}")

mt5.shutdown()

import pandas as pd
import numpy as np
from ta.momentum import RSIIndicator
from ta.trend import MACD, EMAIndicator
from ta.volatility import BollingerBands
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import classification_report

def add_features(df):
  # RSI (momentum)
  df['rsi'] = RSIIndicator(df['close'], window=14).rsi()

  # MACD (trend)
  macd = MACD(df['close'])
  df['macd_diff'] = macd.macd_diff()

  # Bollinger Bands (volatility)
  bb = BollingerBands(df['close'])
  df['bb_pct'] = bb.bollinger_pband()

  # EMA cross (trend direction)
  df['ema_20'] = EMAIndicator(df['close'], 20).ema_indicator()
  df['ema_50'] = EMAIndicator(df['close'], 50).ema_indicator()
  df['ema_cross'] = (df['ema_20'] > df['ema_50']).astype(int)

  # Target: 1 if next candle closes higher, 0 if lower
  df['target'] = (df['close'].shift(-1) > df['close']).astype(int)
  return df.dropna()

# Prepare data
df = add_features(df)
features = ['rsi', 'macd_diff', 'bb_pct', 'ema_cross']
X = df[features]
y = df['target']

# Train / test split (time-based, not random)
split = int(len(df) * 0.8)
X_train, X_test = X[:split], X[split:]
y_train, y_test = y[:split], y[split:]

# Train Random Forest model
model = RandomForestClassifier(
  n_estimators=200,
  max_depth=5,
  random_state=42
)
model.fit(X_train, y_train)

print(classification_report(y_test, model.predict(X_test)))

import numpy as np

def backtest(df, model, features,
      spread_pips=1.5, risk_per_trade=0.01):
  signals = model.predict(df[features])
  returns = []
  equity = [10000] # Start with $10,000
  spread_cost = spread_pips * 0.0001

  for i in range(len(df) - 1):
    price_change = df['close'].iloc[i+1] - df['close'].iloc[i]

    if signals[i] == 1: # Long signal
      pnl = price_change - spread_cost
    elif signals[i] == 0: # Short signal
      pnl = -price_change - spread_cost
    else:
      pnl = 0

    returns.append(pnl)
    equity.append(equity[-1] + pnl * equity[-1] * 10000)

  returns = np.array(returns)

  # Performance metrics
  sharpe = (np.mean(returns) / np.std(returns)) * np.sqrt(252)
  max_dd = max_drawdown(equity)
  win_rate = np.sum(returns > 0) / len(returns)

  print(f"Sharpe Ratio: {sharpe:.2f}")
  print(f"Max Drawdown: {max_dd:.1%}")
  print(f"Win Rate:   {win_rate:.1%}")
  print(f"Final Equity: ${equity[-1]:,.0f}")

  return equity

import MetaTrader5 as mt5
import time

# Risk parameters NEVER skip these
MAX_RISK_PER_TRADE = 0.01  # 1% of account per trade
MAX_DAILY_DRAWDOWN = 0.03  # Stop trading if -3% today
MAX_OPEN_TRADES  = 3    # Maximum concurrent positions
STOP_LOSS_PIPS  = 30   # Hard stop on every trade

def get_lot_size(account_balance, stop_loss_pips):
  "Calculate position size using 1% risk rule"
  risk_amount = account_balance * MAX_RISK_PER_TRADE
  pip_value = 10 # USD per pip for 1 standard lot
  return round(risk_amount / (stop_loss_pips * pip_value), 2)

def place_order(symbol, direction, lot_size, sl_pips):
  price = mt5.symbol_info_tick(symbol).ask
  sl = price - sl_pips * 0.0001 if direction == 'buy' else price + sl_pips * 0.0001

  request = {
    "action": mt5.TRADE_ACTION_DEAL,
    "symbol": symbol,
    "volume": lot_size,
    "type": mt5.ORDER_TYPE_BUY if direction == 'buy' else mt5.ORDER_TYPE_SELL,
    "price": price,
    "sl": round(sl, 5),
    "deviation": 20,
    "magic": 20260527,
    "comment": "AI_BOT",
    "type_time": mt5.ORDER_TIME_GTC,
    "type_filling": mt5.ORDER_FILLING_IOC,
  }
  return mt5.order_send(request)