Upload 3 files

patterns.py

@@ -0,0 +1,173 @@
+import numpy as np
+import pandas as pd
+
+def identify_patterns(df):
+    """Identify candlestick patterns in the data using basic calculations"""
+    patterns = pd.DataFrame(index=df.index)
+
+    # Calculate basic candlestick properties
+    body = df['Close'] - df['Open']
+    body_abs = abs(body)
+    upper_shadow = df['High'] - df[['Open', 'Close']].max(axis=1)
+    lower_shadow = df[['Open', 'Close']].min(axis=1) - df['Low']
+
+    # 1. Hammer Pattern
+    patterns['HAMMER'] = np.where(
+        (lower_shadow > 2 * body_abs) &  # Long lower shadow
+        (upper_shadow <= 0.1 * body_abs) &  # Minimal upper shadow
+        (body > 0),  # Bullish close
+        1, 0
+    )
+
+    # 2. Inverted Hammer Pattern
+    patterns['INVERTED_HAMMER'] = np.where(
+        (upper_shadow > 2 * body_abs) &  # Long upper shadow
+        (lower_shadow <= 0.1 * body_abs) &  # Minimal lower shadow
+        (body > 0),  # Bullish close
+        1, 0
+    )
+
+    # 3. Piercing Line Pattern
+    patterns['PIERCING_LINE'] = np.where(
+        (body.shift(1) < 0) &  # Previous candle bearish
+        (body > 0) &  # Current candle bullish
+        (df['Open'] < df['Close'].shift(1)) &  # Opens below previous close
+        (df['Close'] > (df['Open'].shift(1) + df['Close'].shift(1)) / 2),  # Closes above midpoint
+        1, 0
+    )
+
+    # 4. Bullish Engulfing Pattern
+    patterns['BULLISH_ENGULFING'] = np.where(
+        (body.shift(1) < 0) &  # Previous candle bearish
+        (body > 0) &  # Current candle bullish
+        (df['Open'] < df['Close'].shift(1)) &  # Opens below previous close
+        (df['Close'] > df['Open'].shift(1)),  # Closes above previous open
+        1, 0
+    )
+
+    # 5. Morning Star Pattern
+    patterns['MORNING_STAR'] = np.where(
+        (body.shift(2) < 0) &  # First candle bearish
+        (abs(body.shift(1)) < abs(body.shift(2)) * 0.3) &  # Second candle small
+        (body > 0) &  # Third candle bullish
+        (df['Close'] > df['Close'].shift(2) * 0.5),  # Closes above midpoint
+        1, 0
+    )
+
+    # 6. Three White Soldiers
+    patterns['THREE_WHITE_SOLDIERS'] = np.where(
+        (body > 0) &  # Current candle bullish
+        (body.shift(1) > 0) &  # Previous candle bullish
+        (body.shift(2) > 0) &  # Two candles ago bullish
+        (df['Close'] > df['Close'].shift(1)) &  # Each closes higher
+        (df['Close'].shift(1) > df['Close'].shift(2)),
+        1, 0
+    )
+
+    # 7. Bullish Harami
+    patterns['BULLISH_HARAMI'] = np.where(
+        (body.shift(1) < 0) &  # Previous candle bearish
+        (body > 0) &  # Current candle bullish
+        (df['Open'] > df['Close'].shift(1)) &  # Opens inside previous body
+        (df['Close'] < df['Open'].shift(1)),  # Closes inside previous body
+        1, 0
+    )
+
+    # 8. Hanging Man
+    patterns['HANGING_MAN'] = np.where(
+        (lower_shadow > 2 * body_abs) &  # Long lower shadow
+        (upper_shadow <= 0.1 * body_abs) &  # Minimal upper shadow
+        (body < 0),  # Bearish close
+        1, 0
+    )
+
+    # 9. Dark Cloud Cover
+    patterns['DARK_CLOUD_COVER'] = np.where(
+        (body.shift(1) > 0) &  # Previous candle bullish
+        (body < 0) &  # Current candle bearish
+        (df['Open'] > df['High'].shift(1)) &  # Opens above previous high
+        (df['Close'] < (df['Open'].shift(1) + df['Close'].shift(1)) / 2),  # Closes below midpoint
+        1, 0
+    )
+
+    # 10. Bearish Engulfing
+    patterns['BEARISH_ENGULFING'] = np.where(
+        (body.shift(1) > 0) &  # Previous candle bullish
+        (body < 0) &  # Current candle bearish
+        (df['Open'] > df['Close'].shift(1)) &  # Opens above previous close
+        (df['Close'] < df['Open'].shift(1)),  # Closes below previous open
+        1, 0
+    )
+
+    # 11. Evening Star
+    patterns['EVENING_STAR'] = np.where(
+        (body.shift(2) > 0) &  # First candle bullish
+        (abs(body.shift(1)) < abs(body.shift(2)) * 0.3) &  # Second candle small
+        (body < 0) &  # Third candle bearish
+        (df['Close'] < df['Close'].shift(2) * 0.5),  # Closes below midpoint
+        1, 0
+    )
+
+    # 12. Three Black Crows
+    patterns['THREE_BLACK_CROWS'] = np.where(
+        (body < 0) &  # Current candle bearish
+        (body.shift(1) < 0) &  # Previous candle bearish
+        (body.shift(2) < 0) &  # Two candles ago bearish
+        (df['Close'] < df['Close'].shift(1)) &  # Each closes lower
+        (df['Close'].shift(1) < df['Close'].shift(2)),
+        1, 0
+    )
+
+    # 13. Shooting Star
+    patterns['SHOOTING_STAR'] = np.where(
+        (upper_shadow > 2 * body_abs) &  # Long upper shadow
+        (lower_shadow <= 0.1 * body_abs) &  # Minimal lower shadow
+        (body < 0),  # Bearish close
+        1, 0
+    )
+
+    # 14. Doji Patterns
+    patterns['DOJI'] = np.where(
+        abs(body) <= 0.1 * (df['High'] - df['Low']),  # Very small body
+        1, 0
+    )
+
+    # 15. Dragonfly Doji
+    patterns['DRAGONFLY_DOJI'] = np.where(
+        (abs(body) <= 0.1 * (df['High'] - df['Low'])) &  # Doji body
+        (upper_shadow <= 0.1 * (df['High'] - df['Low'])) &  # Minimal upper shadow
+        (lower_shadow >= 0.7 * (df['High'] - df['Low'])),  # Long lower shadow
+        1, 0
+    )
+
+    # 16. Gravestone Doji
+    patterns['GRAVESTONE_DOJI'] = np.where(
+        (abs(body) <= 0.1 * (df['High'] - df['Low'])) &  # Doji body
+        (lower_shadow <= 0.1 * (df['High'] - df['Low'])) &  # Minimal lower shadow
+        (upper_shadow >= 0.7 * (df['High'] - df['Low'])),  # Long upper shadow
+        1, 0
+    )
+
+    return patterns
+
+def calculate_technical_indicators(df):
+    """Calculate technical indicators for analysis"""
+    # RSI
+    delta = df['Close'].diff()
+    gain = (delta.where(delta > 0, 0)).rolling(window=14).mean()
+    loss = (-delta.where(delta < 0, 0)).rolling(window=14).mean()
+    rs = gain / loss
+    df['RSI'] = 100 - (100 / (1 + rs))
+
+    # MACD
+    exp1 = df['Close'].ewm(span=12, adjust=False).mean()
+    exp2 = df['Close'].ewm(span=26, adjust=False).mean()
+    df['MACD'] = exp1 - exp2
+    df['MACD_Signal'] = df['MACD'].ewm(span=9, adjust=False).mean()
+    df['MACD_Hist'] = df['MACD'] - df['MACD_Signal']
+
+    # Moving Averages
+    df['SMA_20'] = df['Close'].rolling(window=20).mean()
+    df['SMA_50'] = df['Close'].rolling(window=50).mean()
+
+    return df

predictions.py

@@ -0,0 +1,33 @@
+import numpy as np
+from sklearn.ensemble import RandomForestClassifier
+from sklearn.preprocessing import StandardScaler
+
+def prepare_features(df):
+    """Prepare features for prediction model"""
+    features = df[['RSI', 'MACD', 'MACD_Signal', 'MACD_Hist']].copy()
+    features['SMA_Ratio'] = df['SMA_20'] / df['SMA_50']
+    features['Price_Change'] = df['Close'].pct_change()
+    features['Volatility'] = df['Close'].rolling(window=20).std()
+    
+    return features.dropna()
+
+def predict_movement(df, lookback_period=30):
+    """Predict price movement for next 30 minutes"""
+    features = prepare_features(df)
+    if len(features) < lookback_period:
+        return None, None
+    
+    # Prepare training data
+    X = features[:-1].values
+    y = (df['Close'].shift(-1) > df['Close'])[:-1].values
+    
+    # Train model
+    model = RandomForestClassifier(n_estimators=100, random_state=42)
+    model.fit(X, y)
+    
+    # Make prediction
+    latest_features = features.iloc[-1:].values
+    prediction = model.predict(latest_features)[0]
+    probability = model.predict_proba(latest_features)[0]
+    
+    return prediction, probability

trading.py

@@ -0,0 +1,71 @@
+import yfinance as yf
+import pandas as pd
+from datetime import datetime, timedelta
+import pytz
+
+def is_market_open():
+    """Check if US market is currently open"""
+    now = datetime.now(pytz.timezone('America/New_York'))
+    # Market hours are 9:30 AM - 4:00 PM Eastern Time, Monday to Friday
+    is_weekday = now.weekday() < 5
+    is_market_hours = 9.5 <= now.hour + (now.minute / 60) <= 16
+
+    # Pre-market (4:00 AM - 9:30 AM) and After-hours (4:00 PM - 8:00 PM)
+    is_extended_hours = (4 <= now.hour + (now.minute / 60) <= 20)
+
+    return is_weekday and (is_market_hours or is_extended_hours)
+
+def fetch_market_data(symbol, period='1d', interval='15m'):
+    """Fetch market data from Yahoo Finance"""
+    try:
+        # Check if we have a valid symbol
+        if not symbol:
+            raise Exception("Please enter a valid trading symbol")
+
+        # Always fetch 1 day of data to ensure we have enough history
+        ticker = yf.Ticker(symbol)
+        df = ticker.history(period='1d', interval='15m', prepost=True)
+
+        if df.empty:
+            raise Exception(f"No data available for {symbol}. Please verify the symbol is correct.")
+
+        # Get the market status
+        market_status = "Market Open" if is_market_open() else "Market Closed"
+
+        # Trim data based on selected timeframe
+        now = datetime.now(pytz.timezone('America/New_York'))
+        if period.endswith('m'):
+            minutes = int(period[:-1])
+            cutoff_time = now - timedelta(minutes=minutes)
+        else:
+            hours = int(period[:-1])
+            cutoff_time = now - timedelta(hours=hours)
+
+        df = df[df.index >= cutoff_time]
+
+        if df.empty:
+            raise Exception(f"No recent data available. {market_status}.")
+
+        return df
+
+    except Exception as e:
+        if "symbol may be delisted" in str(e).lower():
+            raise Exception(f"Symbol {symbol} not found. Please verify the symbol is correct.")
+        raise Exception(f"Data fetch error: {str(e)}")
+
+def calculate_performance_metrics(predictions, actual):
+    """Calculate prediction performance metrics"""
+    if len(predictions) == 0 or len(actual) == 0:
+        return {
+            'accuracy': 0,
+            'success_rate': 0,
+            'total_predictions': 0
+        }
+
+    correct_predictions = sum(p == a for p, a in zip(predictions, actual))
+
+    return {
+        'accuracy': correct_predictions / len(predictions),
+        'success_rate': correct_predictions / len(predictions) * 100,
+        'total_predictions': len(predictions)
+    }