souti/Freqtrade
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

FrictradeLearning

Browse Source
This commit is contained in:
Jérôme Delacotte
2025-12-20 20:31:09 +01:00
parent 54d028d4f8
commit c26966da45
1 changed files with 17 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

27
FrictradeLearning.py
View File

@@ -73,7 +73,7 @@ RESET = "\033[0m"
class FrictradeLearning(IStrategy): class FrictradeLearning(IStrategy):
startup_candle_count = 200 startup_candle_count = 360
train_model = None train_model = None
model_indicators = [] model_indicators = []
DEFAULT_PARAMS = { DEFAULT_PARAMS = {
@@ -634,6 +634,8 @@ class FrictradeLearning(IStrategy):
informative['rsi'] = talib.RSI(informative['mid'], timeperiod=14) informative['rsi'] = talib.RSI(informative['mid'], timeperiod=14)
self.calculeDerivees(informative, 'rsi', ema_period=12) self.calculeDerivees(informative, 'rsi', ema_period=12)
self.calculateScores(informative, 6)
# informative = self.rsi_trend_probability(informative) # informative = self.rsi_trend_probability(informative)
# probas = self.calculModelInformative(informative) # probas = self.calculModelInformative(informative)
@@ -889,13 +891,7 @@ class FrictradeLearning(IStrategy):
self.calculeDerivees(dataframe, 'sma5', ema_period=3) self.calculeDerivees(dataframe, 'sma5', ema_period=3)
horizon = 180 horizon = 180
dataframe['price_change'] = (dataframe['close'] - dataframe['close'].shift(horizon)) / dataframe['close'].shift(horizon) self.calculateScores(dataframe, horizon)
# dataframe['rsi_delta'] = dataframe['rsi'] - dataframe['rsi'].shift(horizon)
dataframe['price_score'] = (dataframe['price_change'] / 0.05).clip(0, 2)
# dataframe['rsi_score'] = (dataframe['rsi_delta'] / 15).clip(0, 2)
dataframe['heat_score'] = talib.MAX(dataframe['price_score'], timeperiod=horizon) #+ dataframe['rsi_score']
# val = 90000 # val = 90000
# steps = 12 # steps = 12
@@ -957,6 +953,7 @@ class FrictradeLearning(IStrategy):
if self.pairs[pair]['count_of_buys']: if self.pairs[pair]['count_of_buys']:
dca_threshold = self.pairs[pair]['dca_thresholds'][min(self.pairs[pair]['count_of_buys'] - 1, len(self.pairs[pair]['dca_thresholds']) - 1)] dca_threshold = self.pairs[pair]['dca_thresholds'][min(self.pairs[pair]['count_of_buys'] - 1, len(self.pairs[pair]['dca_thresholds']) - 1)]
dataframe[f"next_dca"] = val * (1 - dca_threshold)
print(f"count_of_buys={self.pairs[pair]['count_of_buys']} dca_threshold={dca_threshold} {self.pairs[pair]['dca_thresholds']}") print(f"count_of_buys={self.pairs[pair]['count_of_buys']} dca_threshold={dca_threshold} {self.pairs[pair]['dca_thresholds']}")
print(f"val={val} dca={self.pairs[pair]['dca_thresholds']} ath={self.pairs[pair]['last_ath']} first_price={self.pairs[pair]['first_price']}") print(f"val={val} dca={self.pairs[pair]['dca_thresholds']} ath={self.pairs[pair]['last_ath']} first_price={self.pairs[pair]['first_price']}")
@@ -984,7 +981,7 @@ class FrictradeLearning(IStrategy):
loss_amount += total_stake * dca_previous loss_amount += total_stake * dca_previous
offset = self.dynamic_trailing_offset(pair, total_stake, price=val, ath=ath, count_of_buys=count) offset = self.dynamic_trailing_offset(pair, total_stake, price=val, ath=ath, count_of_buys=count)
if count == self.pairs[pair]['count_of_buys'] - self.pairs[pair]['has_gain']: if count == self.pairs[pair]['count_of_buys'] - self.pairs[pair]['has_gain'] - 1:
print(f"next_buy={round(val * (1 - pct),1)} count={count} pct={round(pct, 4)}") print(f"next_buy={round(val * (1 - pct),1)} count={count} pct={round(pct, 4)}")
dataframe[f"next_buy"] = val * (1 - pct) dataframe[f"next_buy"] = val * (1 - pct)
count += 1 count += 1
@@ -996,6 +993,13 @@ class FrictradeLearning(IStrategy):
return dataframe return dataframe
def calculateScores(self, dataframe, horizon):
dataframe['price_change'] = (dataframe['close'] - dataframe['close'].shift(horizon)) / dataframe['close'].shift(horizon)
# dataframe['rsi_delta'] = dataframe['rsi'] - dataframe['rsi'].shift(horizon)
dataframe['price_score'] = (dataframe['price_change'] / 0.05).clip(0, 2)
# dataframe['rsi_score'] = (dataframe['rsi_delta'] / 15).clip(0, 2)
dataframe['heat_score'] = talib.MAX(dataframe['price_score'], timeperiod=horizon) # + dataframe['rsi_score']
def getOpenTrades(self): def getOpenTrades(self):
# if len(self.trades) == 0: # if len(self.trades) == 0:
self.trades = Trade.get_open_trades() self.trades = Trade.get_open_trades()
@@ -1059,7 +1063,7 @@ class FrictradeLearning(IStrategy):
# & (dataframe['min180'].shift(3) == dataframe['min180']) # & (dataframe['min180'].shift(3) == dataframe['min180'])
, ['enter_long', 'enter_tag'] , ['enter_long', 'enter_tag']
] = (1, f"future") ] = (1, f"future")
dataframe['test'] = np.where(dataframe['enter_long'] == 1, dataframe['close'] * 1.003, np.nan) dataframe['test'] = np.where(dataframe['enter_long'] == 1, dataframe['close'] * 1.01, np.nan)
return dataframe return dataframe
@@ -2334,6 +2338,9 @@ class FrictradeLearning(IStrategy):
and not c.startswith('sma12_deriv1_1h') and not c.startswith('sma12_deriv1_1h')
and not c.startswith('sma12_1h') and not c.startswith('sma12_1h')
and not c.startswith('confidence_index') and not c.startswith('confidence_index')
and not c.startswith('price_change')
and not c.startswith('price_score')
and not c.startswith('heat_score')
] ]
# Étape 3 : remplacer inf et NaN par 0 # Étape 3 : remplacer inf et NaN par 0
dataframe[usable_cols] = dataframe[usable_cols].replace([np.inf, -np.inf], 0).fillna(0) dataframe[usable_cols] = dataframe[usable_cols].replace([np.inf, -np.inf], 0).fillna(0)