freqtrade_origin/freqtrade/optimize/hyperopt_loss_sharpe.py

47 lines
1.4 KiB
Python
Raw Normal View History

2019-07-16 04:45:13 +00:00
"""
2019-07-23 15:51:24 +00:00
SharpeHyperOptLoss
2019-07-16 04:45:13 +00:00
2019-07-23 15:51:24 +00:00
This module defines the alternative HyperOptLoss class which can be used for
Hyperoptimization.
"""
2019-07-16 04:45:13 +00:00
from datetime import datetime
import numpy as np
2020-09-28 17:39:41 +00:00
from pandas import DataFrame
2019-07-16 04:45:13 +00:00
from freqtrade.optimize.hyperopt import IHyperOptLoss
2019-07-16 04:45:13 +00:00
class SharpeHyperOptLoss(IHyperOptLoss):
"""
2019-07-23 15:51:24 +00:00
Defines the loss function for hyperopt.
This implementation uses the Sharpe Ratio calculation.
2019-07-16 04:45:13 +00:00
"""
@staticmethod
def hyperopt_loss_function(results: DataFrame, trade_count: int,
min_date: datetime, max_date: datetime,
*args, **kwargs) -> float:
"""
2019-07-23 15:51:24 +00:00
Objective function, returns smaller number for more optimal results.
Uses Sharpe Ratio calculation.
2019-07-16 04:45:13 +00:00
"""
total_profit = results["profit_ratio"]
2019-07-16 04:45:13 +00:00
days_period = (max_date - min_date).days
# adding slippage of 0.1% per trade
total_profit = total_profit - 0.0005
expected_returns_mean = total_profit.sum() / days_period
up_stdev = np.std(total_profit)
2019-07-16 04:45:13 +00:00
if up_stdev != 0:
sharp_ratio = expected_returns_mean / up_stdev * np.sqrt(365)
2019-07-16 04:45:13 +00:00
else:
# Define high (negative) sharpe ratio to be clear that this is NOT optimal.
sharp_ratio = -20.
2019-07-16 04:45:13 +00:00
# print(expected_returns_mean, up_stdev, sharp_ratio)
2019-07-16 04:45:13 +00:00
return -sharp_ratio