# source: https://raw.githubusercontent.com/IBMaxin/CodingP1.1/6434372bdc0ae7209dffe25b8e5e88872590cc6c/strategies/MicroSwingATRRSI_v1.py # strategies/Github_IBMaxin_CodingP1_1__MicroSwingATRRSI_v1__20250812_145507.py from __future__ import annotations from typing import Dict, List import numpy as np import pandas as pd from pandas import DataFrame from freqtrade.strategy import IStrategy, IntParameter, DecimalParameter class Github_IBMaxin_CodingP1_1__MicroSwingATRRSI_v1__20250812_145507(IStrategy): """ Mid-horizon (hours, not seconds) micro-swing strategy: - Trend filter: SMA fast > SMA slow - Buys pullbacks toward fast SMA with ATR cushion and RSI relief - Exits on RSI pop or ATR channel touch Hyperoptable + FreqAI-ready (feature + target methods below). """ # --- Core run settings (config will override) --- timeframe = "5m" startup_candle_count = 200 process_only_new_candles = True # Small-cap defaults (config can override these) stake_currency = "USDT" stake_amount = 25 max_open_trades = 3 # Order handling order_types = { "entry": "limit", "exit": "limit", "stoploss": "limit", "stoploss_on_exchange": False, "stoploss_on_exchange_interval": 60, } order_time_in_force = {"entry": "GTC", "exit": "GTC"} # Default ROI/SL — hyperopt will override if you optimize ROI/SL spaces minimal_roi = { "0": 0.12, "30": 0.05, "120": 0.02, "240": 0, } stoploss = -0.12 trailing_stop = False plot_config = { "main_plot": { "sma_fast": {"color": "blue"}, "sma_slow": {"color": "orange"}, "atr_low_band": {"color": "gray"}, "atr_high_band": {"color": "gray"}, }, "subplots": { "RSI": {"rsi": {}}, "ATR%": {"atr_pct": {}}, }, } # -------------------- Hyperopt params -------------------- # Trend & pullback structure sma_fast_len = IntParameter(10, 30, default=20, space="buy") sma_slow_len = IntParameter(40, 100, default=60, space="buy") sma_gap_min = DecimalParameter(0.002, 0.03, decimals=4, default=0.008, space="buy") # Volatility cushion atr_len = IntParameter(10, 30, default=14, space="buy") atr_buy_mult = DecimalParameter(0.4, 1.6, decimals=3, default=0.8, space="buy") atr_exit_mult = DecimalParameter(0.3, 1.2, decimals=3, default=0.6, space="sell") # Momentum filter/exit rsi_len = IntParameter(7, 21, default=14, space="buy") rsi_buy_max = IntParameter(45, 60, default=55, space="buy") rsi_exit = IntParameter(60, 90, default=75, space="sell") # -------------------- Indicator helpers -------------------- @staticmethod def _rsi(series: pd.Series, length: int) -> pd.Series: delta = series.diff() up = delta.clip(lower=0) down = -delta.clip(upper=0) roll_up = up.ewm(alpha=1 / length, adjust=False, min_periods=length).mean() roll_down = down.ewm(alpha=1 / length, adjust=False, min_periods=length).mean() rs = roll_up / (roll_down + 1e-12) return 100 - (100 / (1 + rs)) @staticmethod def _atr(df: DataFrame, length: int) -> pd.Series: prev_close = df["close"].shift(1) tr1 = (df["high"] - df["low"]).abs() tr2 = (df["high"] - prev_close).abs() tr3 = (df["low"] - prev_close).abs() tr = pd.concat([tr1, tr2, tr3], axis=1).max(axis=1) return tr.ewm(alpha=1 / length, adjust=False, min_periods=length).mean() # -------------------- Indicators -------------------- def populate_indicators(self, dataframe: DataFrame, metadata: dict) -> DataFrame: sf = int(self.sma_fast_len.value) ss = int(self.sma_slow_len.value) rl = int(self.rsi_len.value) al = int(self.atr_len.value) dataframe["sma_fast"] = dataframe["close"].rolling(sf).mean() dataframe["sma_slow"] = dataframe["close"].rolling(ss).mean() dataframe["atr"] = self._atr(dataframe, al) dataframe["atr_pct"] = (dataframe["atr"] / dataframe["close"]).fillna(0) dataframe["rsi"] = self._rsi(dataframe["close"], rl) # ATR bands around fast SMA (visual + exits) dataframe["atr_low_band"] = dataframe["sma_fast"] - self.atr_buy_mult.value * dataframe["atr"] dataframe["atr_high_band"] = dataframe["sma_fast"] + self.atr_exit_mult.value * dataframe["atr"] # Gap of fast/slow to ensure trend strength dataframe["sma_gap"] = (dataframe["sma_fast"] - dataframe["sma_slow"]) / dataframe["sma_slow"] dataframe.replace([np.inf, -np.inf], np.nan, inplace=True) return dataframe # -------------------- Entries -------------------- def populate_entry_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame: conditions: List[pd.Series] = [] # Trend: fast > slow & enough separation conditions += [ (dataframe["sma_fast"] > dataframe["sma_slow"]), (dataframe["sma_gap"] > float(self.sma_gap_min.value)), ] # Pullback near/below fast SMA with ATR cushion conditions += [ (dataframe["close"] <= dataframe["atr_low_band"]), (dataframe["rsi"] <= int(self.rsi_buy_max.value)), ] if conditions: dataframe.loc[ np.logical_and.reduce(conditions), ["enter_long", "enter_tag"], ] = (1, "pullback_atr_rsi") return dataframe # -------------------- Exits -------------------- def populate_exit_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame: exit_conditions: List[pd.Series] = [] # Momentum pop or ATR channel top exit_conditions += [ (dataframe["rsi"] >= int(self.rsi_exit.value)) | (dataframe["close"] >= dataframe["atr_high_band"]) ] if exit_conditions: dataframe.loc[ np.logical_and.reduce(exit_conditions), ["exit_long", "exit_tag"], ] = (1, "rsi_or_atr_band") return dataframe # -------------------- FreqAI hooks (optional) -------------------- # These are used only when FreqAI is enabled in config. def feature_engineering_expand_basic(self, dataframe: DataFrame, metadata: dict) -> DataFrame: df = dataframe.copy() # Lightweight, model-friendly features df["ret_1"] = df["close"].pct_change().fillna(0) df["ret_3"] = df["close"].pct_change(3).fillna(0) df["rsi_fe"] = self._rsi(df["close"], int(self.rsi_len.value)).fillna(50) df["atr_pct_fe"] = (self._atr(df, int(self.atr_len.value)) / df["close"]).fillna(0) df["sma_gap_fe"] = ( df["close"].rolling(int(self.sma_fast_len.value)).mean() - df["close"].rolling(int(self.sma_slow_len.value)).mean() ) / df["close"].rolling(int(self.sma_slow_len.value)).mean() df["sma_gap_fe"] = df["sma_gap_fe"].fillna(0) df.replace([np.inf, -np.inf], 0, inplace=True) return df def set_freqai_targets(self, dataframe: DataFrame, metadata: dict) -> DataFrame: """ Create a simple future-return target over ~2 hours (24x5m candles). Matches your config's label_period_candles=24. """ df = dataframe.copy() horizon = 24 df["fai_target_return"] = (df["close"].shift(-horizon) / df["close"] - 1.0).fillna(0) return df