# source: https://raw.githubusercontent.com/4tie/fortiesr/d5cc40760e3cbaff381058a0ea8733d2be469578/backend/engine/strategy_generator.py """ Strategy Generator Engine Pure business logic for generating trading strategies No dependencies on FastAPI, file I/O, or external services """ from typing import List, Dict, Any, Optional from dataclasses import dataclass from enum import Enum class StrategyStyle(Enum): SCALPING = "scalping" INTRADAY = "intraday" SWING = "swing" POSITION = "position" class RiskProfile(Enum): CONSERVATIVE = "conservative" BALANCED = "balanced" AGGRESSIVE = "aggressive" @dataclass class StrategyTemplate: """Template for strategy generation""" name: str style: StrategyStyle indicators: List[str] timeframe: str description: str @dataclass class GeneratedStrategy: """Generated trading strategy""" id: str name: str code: str style: StrategyStyle timeframe: str indicators: List[str] parameters: Dict[str, Any] description: str class StrategyGeneratorEngine: """ Pure business logic for generating trading strategies Can be tested independently without FastAPI or file I/O """ # Predefined strategy templates TEMPLATES = { StrategyStyle.SCALPING: [ StrategyTemplate( name="RSI Scalper", style=StrategyStyle.SCALPING, indicators=["RSI", "EMA", "ATR"], timeframe="1m", description="Fast RSI-based scalping with EMA trend filter" ), StrategyTemplate( name="MACD Scalper", style=StrategyStyle.SCALPING, indicators=["MACD", "EMA", "Volume"], timeframe="5m", description="MACD momentum scalping with volume confirmation" ), StrategyTemplate( name="STOCH Scalper", style=StrategyStyle.SCALPING, indicators=["STOCH", "ATR", "EMA"], timeframe="5m", description="Stochastic oscillator scalping with ATR volatility filter" ), ], StrategyStyle.INTRADAY: [ StrategyTemplate( name="EMA Crossover", style=StrategyStyle.INTRADAY, indicators=["EMA", "RSI", "ATR"], timeframe="15m", description="Classic EMA crossover with RSI filter" ), StrategyTemplate( name="Bollinger Breakout", style=StrategyStyle.INTRADAY, indicators=["Bollinger Bands", "RSI", "Volume"], timeframe="30m", description="Bollinger band breakout with RSI confirmation" ), StrategyTemplate( name="CCI Reversal", style=StrategyStyle.INTRADAY, indicators=["CCI", "EMA", "ATR"], timeframe="15m", description="Commodity Channel Index reversal with EMA trend" ), ], StrategyStyle.SWING: [ StrategyTemplate( name="Swing RSI", style=StrategyStyle.SWING, indicators=["RSI", "EMA", "MACD"], timeframe="1h", description="Swing trading with RSI and EMA trend" ), StrategyTemplate( name="MACD Swing", style=StrategyStyle.SWING, indicators=["MACD", "EMA", "ATR"], timeframe="4h", description="MACD-based swing trading with ATR stops" ), StrategyTemplate( name="MFI Swing", style=StrategyStyle.SWING, indicators=["MFI", "EMA", "SAR"], timeframe="1h", description="Money Flow Index swing with Parabolic SAR" ), ], StrategyStyle.POSITION: [ StrategyTemplate( name="Position Trend", style=StrategyStyle.POSITION, indicators=["EMA", "MACD", "RSI"], timeframe="1d", description="Long-term trend following with multiple indicators" ), StrategyTemplate( name="KAMA Trend", style=StrategyStyle.POSITION, indicators=["KAMA", "ADX", "ATR"], timeframe="1d", description="Kaufman Adaptive Moving Average with ADX trend strength" ), ], } def __init__(self): self.templates = self.TEMPLATES def generate_strategies( self, style: StrategyStyle, risk_profile: RiskProfile, count: int = 10, timeframe: Optional[str] = None ) -> List[GeneratedStrategy]: """ Generate trading strategies based on style and risk profile Args: style: Trading style (scalping, intraday, swing, position) risk_profile: Risk tolerance (conservative, balanced, aggressive) count: Number of strategies to generate timeframe: Specific timeframe (auto-select if None) Returns: List of generated strategies """ templates = self.templates.get(style, []) if not templates: return [] strategies = [] for i in range(count): # Cycle through templates template = templates[i % len(templates)] # Generate strategy strategy = self._generate_from_template( template=template, risk_profile=risk_profile, index=i, timeframe=timeframe ) strategies.append(strategy) return strategies def _generate_from_template( self, template: StrategyTemplate, risk_profile: RiskProfile, index: int, timeframe: Optional[str] = None ) -> GeneratedStrategy: """Generate a single strategy from template""" # Determine timeframe if timeframe: strategy_timeframe = timeframe else: strategy_timeframe = template.timeframe # Generate parameters based on risk profile parameters = self._generate_parameters( template.indicators, risk_profile, strategy_timeframe ) # Generate code code = self._generate_code( template.name, template.indicators, parameters, strategy_timeframe ) return GeneratedStrategy( id=f"strat_{index}_{template.style.value}", name=f"{template.name} #{index + 1}", code=code, style=template.style, timeframe=strategy_timeframe, indicators=template.indicators, parameters=parameters, description=template.description ) def _generate_parameters( self, indicators: List[str], risk_profile: RiskProfile, timeframe: str ) -> Dict[str, Any]: """Generate strategy parameters based on risk profile""" base_params = {} # Risk-based parameter adjustment if risk_profile == RiskProfile.CONSERVATIVE: risk_multiplier = 0.8 stop_loss = 0.02 take_profit = 0.04 elif risk_profile == RiskProfile.BALANCED: risk_multiplier = 1.0 stop_loss = 0.03 take_profit = 0.06 else: # AGGRESSIVE risk_multiplier = 1.2 stop_loss = 0.05 take_profit = 0.10 # Indicator-specific parameters for indicator in indicators: if indicator == "RSI": base_params["rsi_period"] = int(14 * risk_multiplier) base_params["rsi_overbought"] = 70 base_params["rsi_oversold"] = 30 elif indicator == "EMA": base_params["ema_fast"] = int(9 * risk_multiplier) base_params["ema_slow"] = int(21 * risk_multiplier) elif indicator == "MACD": base_params["macd_fast"] = int(12 * risk_multiplier) base_params["macd_slow"] = int(26 * risk_multiplier) base_params["macd_signal"] = int(9 * risk_multiplier) elif indicator == "ATR": base_params["atr_period"] = int(14 * risk_multiplier) base_params["atr_multiplier"] = 2.0 elif indicator == "Bollinger Bands": base_params["bb_period"] = int(20 * risk_multiplier) base_params["bb_std"] = 2.0 elif indicator == "STOCH": base_params["stoch_fastk_period"] = int(14 * risk_multiplier) base_params["stoch_slowk_period"] = int(3 * risk_multiplier) base_params["stoch_slowd_period"] = int(3 * risk_multiplier) elif indicator == "CCI": base_params["cci_period"] = int(20 * risk_multiplier) elif indicator == "MFI": base_params["mfi_period"] = int(14 * risk_multiplier) elif indicator == "SAR": base_params["sar_acceleration"] = 0.02 base_params["sar_maximum"] = 0.2 elif indicator == "ADX": base_params["adx_period"] = int(14 * risk_multiplier) elif indicator == "KAMA": base_params["kama_period"] = int(10 * risk_multiplier) elif indicator == "SMA": base_params["sma_period"] = int(20 * risk_multiplier) elif indicator == "WMA": base_params["wma_period"] = int(20 * risk_multiplier) elif indicator == "DEMA": base_params["dema_period"] = int(20 * risk_multiplier) elif indicator == "TEMA": base_params["tema_period"] = int(20 * risk_multiplier) elif indicator == "WILLR": base_params["willr_period"] = int(14 * risk_multiplier) elif indicator == "NATR": base_params["natr_period"] = int(14 * risk_multiplier) elif indicator == "ROC": base_params["roc_period"] = int(10 * risk_multiplier) # Risk management parameters base_params["stop_loss"] = stop_loss base_params["take_profit"] = take_profit base_params["risk_per_trade"] = 0.01 if risk_profile == RiskProfile.CONSERVATIVE else 0.02 return base_params def _generate_code( self, name: str, indicators: List[str], parameters: Dict[str, Any], timeframe: str ) -> str: """Generate strategy code (simplified template)""" code = f""" # {name} # Timeframe: {timeframe} # Indicators: {', '.join(indicators)} from freqtrade.strategy import IStrategy from pandas import DataFrame import talib.abstract as ta class {name.replace(' ', Github_4tie_fortiesr__strategy_generator__20260709_213639(IStrategy): timeframe = '{timeframe}' stoploss = {parameters.get('stop_loss', 0.03)} def populate_indicators(self, dataframe: DataFrame, metadata: dict) -> DataFrame: """ # Add indicator code for indicator in indicators: if indicator == "RSI": code += f""" dataframe['rsi'] = ta.RSI(dataframe, timeperiod={parameters.get('rsi_period', 14)}) """ elif indicator == "EMA": code += f""" dataframe['ema_fast'] = ta.EMA(dataframe, timeperiod={parameters.get('ema_fast', 9)}) dataframe['ema_slow'] = ta.EMA(dataframe, timeperiod={parameters.get('ema_slow', 21)}) """ elif indicator == "MACD": code += f""" macd = ta.MACD(dataframe, fastperiod={parameters.get('macd_fast', 12)}, slowperiod={parameters.get('macd_slow', 26)}, signalperiod={parameters.get('macd_signal', 9)}) dataframe['macd'] = macd['macd'] dataframe['macd_signal'] = macd['macdsignal'] """ elif indicator == "ATR": code += f""" dataframe['atr'] = ta.ATR(dataframe, timeperiod={parameters.get('atr_period', 14)}) """ elif indicator == "STOCH": code += f""" stoch = ta.STOCH(dataframe, fastk_period={parameters.get('stoch_fastk_period', 14)}, slowk_period={parameters.get('stoch_slowk_period', 3)}, slowd_period={parameters.get('stoch_slowd_period', 3)}) dataframe['stoch_slowk'] = stoch['slowk'] dataframe['stoch_slowd'] = stoch['slowd'] """ elif indicator == "CCI": code += f""" dataframe['cci'] = ta.CCI(dataframe, timeperiod={parameters.get('cci_period', 20)}) """ elif indicator == "MFI": code += f""" dataframe['mfi'] = ta.MFI(dataframe, timeperiod={parameters.get('mfi_period', 14)}) """ elif indicator == "SAR": code += f""" dataframe['sar'] = ta.SAR(dataframe, acceleration={parameters.get('sar_acceleration', 0.02)}, maximum={parameters.get('sar_maximum', 0.2)}) """ elif indicator == "ADX": code += f""" dataframe['adx'] = ta.ADX(dataframe, timeperiod={parameters.get('adx_period', 14)}) """ elif indicator == "KAMA": code += f""" dataframe['kama'] = ta.KAMA(dataframe, timeperiod={parameters.get('kama_period', 10)}) """ elif indicator == "SMA": code += f""" dataframe['sma'] = ta.SMA(dataframe, timeperiod={parameters.get('sma_period', 20)}) """ elif indicator == "WMA": code += f""" dataframe['wma'] = ta.WMA(dataframe, timeperiod={parameters.get('wma_period', 20)}) """ elif indicator == "DEMA": code += f""" dataframe['dema'] = ta.DEMA(dataframe, timeperiod={parameters.get('dema_period', 20)}) """ elif indicator == "TEMA": code += f""" dataframe['tema'] = ta.TEMA(dataframe, timeperiod={parameters.get('tema_period', 20)}) """ elif indicator == "WILLR": code += f""" dataframe['willr'] = ta.WILLR(dataframe, timeperiod={parameters.get('willr_period', 14)}) """ elif indicator == "NATR": code += f""" dataframe['natr'] = ta.NATR(dataframe, timeperiod={parameters.get('natr_period', 14)}) """ elif indicator == "ROC": code += f""" dataframe['roc'] = ta.ROC(dataframe, timeperiod={parameters.get('roc_period', 10)}) """ elif indicator == "OBV": code += """ dataframe['obv'] = ta.OBV(dataframe) """ elif indicator == "AD": code += """ dataframe['ad'] = ta.AD(dataframe) """ elif indicator == "Volume": code += """ dataframe['volume'] = dataframe['volume'] """ code += """ return dataframe def populate_buy_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame: dataframe.loc[ ( # Add your buy conditions here ), 'buy' ] = 1 return dataframe def populate_sell_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame: dataframe.loc[ ( # Add your sell conditions here ), 'sell' ] = 1 return dataframe """ return code def get_available_styles(self) -> List[StrategyStyle]: """Get available trading styles""" return list(StrategyStyle) def get_available_risk_profiles(self) -> List[RiskProfile]: """Get available risk profiles""" return list(RiskProfile)