freqtrade_origin/freqtrade/optimize/hyperopt.py
2019-10-27 10:56:38 +01:00

435 lines
17 KiB
Python

# pragma pylint: disable=too-many-instance-attributes, pointless-string-statement
"""
This module contains the hyperopt logic
"""
import logging
import sys
from collections import OrderedDict
from operator import itemgetter
from pathlib import Path
from pprint import pprint
from typing import Any, Dict, List, Optional
import rapidjson
from colorama import init as colorama_init
from colorama import Fore, Style
from joblib import Parallel, delayed, dump, load, wrap_non_picklable_objects, cpu_count
from pandas import DataFrame
from skopt import Optimizer
from skopt.space import Dimension
from freqtrade.data.history import get_timeframe, trim_dataframe
from freqtrade.misc import round_dict
from freqtrade.optimize.backtesting import Backtesting
# Import IHyperOpt and IHyperOptLoss to allow unpickling classes from these modules
from freqtrade.optimize.hyperopt_interface import IHyperOpt # noqa: F4
from freqtrade.optimize.hyperopt_loss_interface import IHyperOptLoss # noqa: F4
from freqtrade.resolvers.hyperopt_resolver import HyperOptResolver, HyperOptLossResolver
logger = logging.getLogger(__name__)
INITIAL_POINTS = 30
# Keep no more than 2*SKOPT_MODELS_MAX_NUM models
# in the skopt models list
SKOPT_MODELS_MAX_NUM = 10
MAX_LOSS = 100000 # just a big enough number to be bad result in loss optimization
class Hyperopt:
"""
Hyperopt class, this class contains all the logic to run a hyperopt simulation
To run a backtest:
hyperopt = Hyperopt(config)
hyperopt.start()
"""
def __init__(self, config: Dict[str, Any]) -> None:
self.config = config
self.custom_hyperopt = HyperOptResolver(self.config).hyperopt
self.backtesting = Backtesting(self.config)
self.custom_hyperoptloss = HyperOptLossResolver(self.config).hyperoptloss
self.calculate_loss = self.custom_hyperoptloss.hyperopt_loss_function
self.trials_file = (self.config['user_data_dir'] /
'hyperopt_results' / 'hyperopt_results.pickle')
self.tickerdata_pickle = (self.config['user_data_dir'] /
'hyperopt_results' / 'hyperopt_tickerdata.pkl')
self.total_epochs = config.get('epochs', 0)
self.current_best_loss = 100
if not self.config.get('hyperopt_continue'):
self.clean_hyperopt()
else:
logger.info("Continuing on previous hyperopt results.")
# Previous evaluations
self.trials: List = []
# Populate functions here (hasattr is slow so should not be run during "regular" operations)
if hasattr(self.custom_hyperopt, 'populate_indicators'):
self.backtesting.strategy.advise_indicators = \
self.custom_hyperopt.populate_indicators # type: ignore
if hasattr(self.custom_hyperopt, 'populate_buy_trend'):
self.backtesting.strategy.advise_buy = \
self.custom_hyperopt.populate_buy_trend # type: ignore
if hasattr(self.custom_hyperopt, 'populate_sell_trend'):
self.backtesting.strategy.advise_sell = \
self.custom_hyperopt.populate_sell_trend # type: ignore
# Use max_open_trades for hyperopt as well, except --disable-max-market-positions is set
if self.config.get('use_max_market_positions', True):
self.max_open_trades = self.config['max_open_trades']
else:
logger.debug('Ignoring max_open_trades (--disable-max-market-positions was used) ...')
self.max_open_trades = 0
self.position_stacking = self.config.get('position_stacking', False)
if self.has_space('sell'):
# Make sure use_sell_signal is enabled
if 'ask_strategy' not in self.config:
self.config['ask_strategy'] = {}
self.config['ask_strategy']['use_sell_signal'] = True
@staticmethod
def get_lock_filename(config) -> str:
return str(config['user_data_dir'] / 'hyperopt.lock')
def clean_hyperopt(self):
"""
Remove hyperopt pickle files to restart hyperopt.
"""
for f in [self.tickerdata_pickle, self.trials_file]:
p = Path(f)
if p.is_file():
logger.info(f"Removing `{p}`.")
p.unlink()
def get_args(self, params):
dimensions = self.dimensions
# Ensure the number of dimensions match
# the number of parameters in the list x.
if len(params) != len(dimensions):
raise ValueError('Mismatch in number of search-space dimensions. '
f'len(dimensions)=={len(dimensions)} and len(x)=={len(params)}')
# Create a dict where the keys are the names of the dimensions
# and the values are taken from the list of parameters x.
arg_dict = {dim.name: value for dim, value in zip(dimensions, params)}
return arg_dict
def save_trials(self) -> None:
"""
Save hyperopt trials to file
"""
if self.trials:
logger.info("Saving %d evaluations to '%s'", len(self.trials), self.trials_file)
dump(self.trials, self.trials_file)
def read_trials(self) -> List:
"""
Read hyperopt trials file
"""
logger.info("Reading Trials from '%s'", self.trials_file)
trials = load(self.trials_file)
self.trials_file.unlink()
return trials
def log_trials_result(self) -> None:
"""
Display Best hyperopt result
"""
results = sorted(self.trials, key=itemgetter('loss'))
best_result = results[0]
params = best_result['params']
log_str = self.format_results_logstring(best_result)
print(f"\nBest result:\n\n{log_str}\n")
if self.config.get('print_json'):
result_dict: Dict = {}
if self.has_space('buy') or self.has_space('sell'):
result_dict['params'] = {}
if self.has_space('buy'):
result_dict['params'].update({p.name: params.get(p.name)
for p in self.hyperopt_space('buy')})
if self.has_space('sell'):
result_dict['params'].update({p.name: params.get(p.name)
for p in self.hyperopt_space('sell')})
if self.has_space('roi'):
# Convert keys in min_roi dict to strings because
# rapidjson cannot dump dicts with integer keys...
# OrderedDict is used to keep the numeric order of the items
# in the dict.
result_dict['minimal_roi'] = OrderedDict(
(str(k), v) for k, v in self.custom_hyperopt.generate_roi_table(params).items()
)
if self.has_space('stoploss'):
result_dict['stoploss'] = params.get('stoploss')
print(rapidjson.dumps(result_dict, default=str, number_mode=rapidjson.NM_NATIVE))
else:
if self.has_space('buy'):
print('Buy hyperspace params:')
pprint({p.name: params.get(p.name) for p in self.hyperopt_space('buy')},
indent=4)
if self.has_space('sell'):
print('Sell hyperspace params:')
pprint({p.name: params.get(p.name) for p in self.hyperopt_space('sell')},
indent=4)
if self.has_space('roi'):
print("ROI table:")
# Round printed values to 5 digits after the decimal point
pprint(round_dict(self.custom_hyperopt.generate_roi_table(params), 5), indent=4)
if self.has_space('stoploss'):
# Also round to 5 digits after the decimal point
print(f"Stoploss: {round(params.get('stoploss'), 5)}")
def log_results(self, results) -> None:
"""
Log results if it is better than any previous evaluation
"""
print_all = self.config.get('print_all', False)
is_best_loss = results['loss'] < self.current_best_loss
if print_all or is_best_loss:
if is_best_loss:
self.current_best_loss = results['loss']
log_str = self.format_results_logstring(results)
# Colorize output
if self.config.get('print_colorized', False):
if results['total_profit'] > 0:
log_str = Fore.GREEN + log_str
if print_all and is_best_loss:
log_str = Style.BRIGHT + log_str
if print_all:
print(log_str)
else:
print('\n' + log_str)
else:
print('.', end='')
sys.stdout.flush()
def format_results_logstring(self, results) -> str:
# Output human-friendly index here (starting from 1)
current = results['current_epoch'] + 1
total = self.total_epochs
res = results['results_explanation']
loss = results['loss']
log_str = f'{current:5d}/{total}: {res} Objective: {loss:.5f}'
log_str = f'*{log_str}' if results['is_initial_point'] else f' {log_str}'
return log_str
def has_space(self, space: str) -> bool:
"""
Tell if a space value is contained in the configuration
"""
return any(s in self.config['spaces'] for s in [space, 'all'])
def hyperopt_space(self, space: Optional[str] = None) -> List[Dimension]:
"""
Return the dimensions in the hyperoptimization space.
:param space: Defines hyperspace to return dimensions for.
If None, then the self.has_space() will be used to return dimensions
for all hyperspaces used.
"""
spaces: List[Dimension] = []
if space == 'buy' or (space is None and self.has_space('buy')):
logger.debug("Hyperopt has 'buy' space")
spaces += self.custom_hyperopt.indicator_space()
if space == 'sell' or (space is None and self.has_space('sell')):
logger.debug("Hyperopt has 'sell' space")
spaces += self.custom_hyperopt.sell_indicator_space()
if space == 'roi' or (space is None and self.has_space('roi')):
logger.debug("Hyperopt has 'roi' space")
spaces += self.custom_hyperopt.roi_space()
if space == 'stoploss' or (space is None and self.has_space('stoploss')):
logger.debug("Hyperopt has 'stoploss' space")
spaces += self.custom_hyperopt.stoploss_space()
return spaces
def generate_optimizer(self, _params: Dict, iteration=None) -> Dict:
"""
Used Optimize function. Called once per epoch to optimize whatever is configured.
Keep this function as optimized as possible!
"""
params = self.get_args(_params)
if self.has_space('roi'):
self.backtesting.strategy.minimal_roi = \
self.custom_hyperopt.generate_roi_table(params)
if self.has_space('buy'):
self.backtesting.strategy.advise_buy = \
self.custom_hyperopt.buy_strategy_generator(params)
if self.has_space('sell'):
self.backtesting.strategy.advise_sell = \
self.custom_hyperopt.sell_strategy_generator(params)
if self.has_space('stoploss'):
self.backtesting.strategy.stoploss = params['stoploss']
processed = load(self.tickerdata_pickle)
min_date, max_date = get_timeframe(processed)
results = self.backtesting.backtest(
{
'stake_amount': self.config['stake_amount'],
'processed': processed,
'max_open_trades': self.max_open_trades,
'position_stacking': self.position_stacking,
'start_date': min_date,
'end_date': max_date,
}
)
results_explanation = self.format_results(results)
trade_count = len(results.index)
total_profit = results.profit_abs.sum()
# If this evaluation contains too short amount of trades to be
# interesting -- consider it as 'bad' (assigned max. loss value)
# in order to cast this hyperspace point away from optimization
# path. We do not want to optimize 'hodl' strategies.
if trade_count < self.config['hyperopt_min_trades']:
return {
'loss': MAX_LOSS,
'params': params,
'results_explanation': results_explanation,
'total_profit': total_profit,
}
loss = self.calculate_loss(results=results, trade_count=trade_count,
min_date=min_date.datetime, max_date=max_date.datetime)
return {
'loss': loss,
'params': params,
'results_explanation': results_explanation,
'total_profit': total_profit,
}
def format_results(self, results: DataFrame) -> str:
"""
Return the formatted results explanation in a string
"""
trades = len(results.index)
avg_profit = results.profit_percent.mean() * 100.0
total_profit = results.profit_abs.sum()
stake_cur = self.config['stake_currency']
profit = results.profit_percent.sum() * 100.0
duration = results.trade_duration.mean()
return (f'{trades:6d} trades. Avg profit {avg_profit: 5.2f}%. '
f'Total profit {total_profit: 11.8f} {stake_cur} '
f'({profit: 7.2f}Σ%). Avg duration {duration:5.1f} mins.')
def get_optimizer(self, dimensions, cpu_count) -> Optimizer:
return Optimizer(
dimensions,
base_estimator="ET",
acq_optimizer="auto",
n_initial_points=INITIAL_POINTS,
acq_optimizer_kwargs={'n_jobs': cpu_count},
random_state=self.config.get('hyperopt_random_state', None)
)
def fix_optimizer_models_list(self):
"""
WORKAROUND: Since skopt is not actively supported, this resolves problems with skopt
memory usage, see also: https://github.com/scikit-optimize/scikit-optimize/pull/746
This may cease working when skopt updates if implementation of this intrinsic
part changes.
"""
n = len(self.opt.models) - SKOPT_MODELS_MAX_NUM
# Keep no more than 2*SKOPT_MODELS_MAX_NUM models in the skopt models list,
# remove the old ones. These are actually of no use, the current model
# from the estimator is the only one used in the skopt optimizer.
# Freqtrade code also does not inspect details of the models.
if n >= SKOPT_MODELS_MAX_NUM:
logger.debug(f"Fixing skopt models list, removing {n} old items...")
del self.opt.models[0:n]
def run_optimizer_parallel(self, parallel, asked, i) -> List:
return parallel(delayed(
wrap_non_picklable_objects(self.generate_optimizer))(v, i) for v in asked)
def load_previous_results(self):
""" read trials file if we have one """
if self.trials_file.is_file() and self.trials_file.stat().st_size > 0:
self.trials = self.read_trials()
logger.info(
'Loaded %d previous evaluations from disk.',
len(self.trials)
)
def start(self) -> None:
data, timerange = self.backtesting.load_bt_data()
preprocessed = self.backtesting.strategy.tickerdata_to_dataframe(data)
# Trim startup period from analyzed dataframe
for pair, df in preprocessed.items():
preprocessed[pair] = trim_dataframe(df, timerange)
min_date, max_date = get_timeframe(data)
logger.info(
'Hyperopting with data from %s up to %s (%s days)..',
min_date.isoformat(), max_date.isoformat(), (max_date - min_date).days
)
dump(preprocessed, self.tickerdata_pickle)
# We don't need exchange instance anymore while running hyperopt
self.backtesting.exchange = None # type: ignore
self.load_previous_results()
cpus = cpu_count()
logger.info(f"Found {cpus} CPU cores. Let's make them scream!")
config_jobs = self.config.get('hyperopt_jobs', -1)
logger.info(f'Number of parallel jobs set as: {config_jobs}')
self.dimensions = self.hyperopt_space()
self.opt = self.get_optimizer(self.dimensions, config_jobs)
if self.config.get('print_colorized', False):
colorama_init(autoreset=True)
try:
with Parallel(n_jobs=config_jobs) as parallel:
jobs = parallel._effective_n_jobs()
logger.info(f'Effective number of parallel workers used: {jobs}')
EVALS = max(self.total_epochs // jobs, 1)
for i in range(EVALS):
asked = self.opt.ask(n_points=jobs)
f_val = self.run_optimizer_parallel(parallel, asked, i)
self.opt.tell(asked, [v['loss'] for v in f_val])
self.fix_optimizer_models_list()
for j in range(jobs):
current = i * jobs + j
val = f_val[j]
val['current_epoch'] = current
val['is_initial_point'] = current < INITIAL_POINTS
self.log_results(val)
self.trials.append(val)
logger.debug(f"Optimizer epoch evaluated: {val}")
except KeyboardInterrupt:
print('User interrupted..')
self.save_trials()
self.log_trials_result()