2019-07-16 04:45:13 +00:00
|
|
|
"""
|
|
|
|
|
IHyperOptLoss interface
|
|
|
|
|
This module defines the interface for the loss-function for hyperopts
|
|
|
|
|
"""
|
|
|
|
|
|
|
|
|
|
from datetime import datetime
|
|
|
|
|
|
|
|
|
|
from pandas import DataFrame
|
|
|
|
|
import numpy as np
|
|
|
|
|
|
2019-07-17 05:14:27 +00:00
|
|
|
from freqtrade.optimize.hyperopt import IHyperOptLoss
|
2019-07-16 04:45:13 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
class SharpeHyperOptLoss(IHyperOptLoss):
|
|
|
|
|
"""
|
|
|
|
|
Defines the a loss function for hyperopt.
|
|
|
|
|
This implementation uses the sharpe ratio calculation.
|
|
|
|
|
"""
|
|
|
|
|
|
|
|
|
|
@staticmethod
|
|
|
|
|
def hyperopt_loss_function(results: DataFrame, trade_count: int,
|
|
|
|
|
min_date: datetime, max_date: datetime,
|
|
|
|
|
*args, **kwargs) -> float:
|
|
|
|
|
"""
|
|
|
|
|
Objective function, returns smaller number for more optimal results
|
|
|
|
|
Using sharpe ratio calculation
|
|
|
|
|
"""
|
|
|
|
|
total_profit = results.profit_percent
|
|
|
|
|
days_period = (max_date - min_date).days
|
|
|
|
|
|
|
|
|
|
# adding slippage of 0.1% per trade
|
|
|
|
|
total_profit = total_profit - 0.0005
|
|
|
|
|
expected_yearly_return = total_profit.sum() / days_period
|
|
|
|
|
|
|
|
|
|
if (np.std(total_profit) != 0.):
|
|
|
|
|
sharp_ratio = expected_yearly_return / np.std(total_profit) * np.sqrt(365)
|
|
|
|
|
else:
|
|
|
|
|
# Define high (negative) sharpe ratio to be clear that this is NOT optimal.
|
|
|
|
|
sharp_ratio = 20.
|
|
|
|
|
|
|
|
|
|
# print(expected_yearly_return, np.std(total_profit), sharp_ratio)
|
|
|
|
|
return -sharp_ratio
|
