add multiprocessing variant of ReinforcementLearningPPO

This commit is contained in:
robcaulk 2022-08-15 13:46:12 +02:00
parent cf0731095f
commit acf3484e88
2 changed files with 118 additions and 2 deletions

View File

@ -35,10 +35,12 @@ class Base3ActionRLEnv(gym.Env):
metadata = {'render.modes': ['human']} metadata = {'render.modes': ['human']}
def __init__(self, df, prices, reward_kwargs, window_size=10, starting_point=True, ): def __init__(self, df, prices, reward_kwargs, window_size=10, starting_point=True,
id: str = 'baseenv-1', seed: int = 1):
assert df.ndim == 2 assert df.ndim == 2
self.seed() self.id = id
self.seed(seed)
self.df = df self.df = df
self.signal_features = self.df self.signal_features = self.df
self.prices = prices self.prices = prices

View File

@ -0,0 +1,114 @@
import logging
from typing import Any, Dict # , Tuple
import numpy as np
# import numpy.typing as npt
# import pandas as pd
import torch as th
# from pandas import DataFrame
from typing import Callable
from stable_baselines3 import PPO
from stable_baselines3.common.callbacks import EvalCallback
from stable_baselines3.common.vec_env import SubprocVecEnv
from stable_baselines3.common.utils import set_random_seed
from freqtrade.freqai.RL.Base3ActionRLEnv import Base3ActionRLEnv, Actions, Positions
from freqtrade.freqai.RL.BaseReinforcementLearningModel import BaseReinforcementLearningModel
import gym
logger = logging.getLogger(__name__)
def make_env(env_id: str, rank: int, seed: int, train_df, price,
reward_params, window_size) -> Callable:
"""
Utility function for multiprocessed env.
:param env_id: (str) the environment ID
:param num_env: (int) the number of environment you wish to have in subprocesses
:param seed: (int) the inital seed for RNG
:param rank: (int) index of the subprocess
:return: (Callable)
"""
def _init() -> gym.Env:
env = MyRLEnv(df=train_df, prices=price, window_size=window_size,
reward_kwargs=reward_params, id=env_id, seed=seed + rank)
return env
set_random_seed(seed)
return _init
class ReinforcementLearningPPO_multiproc(BaseReinforcementLearningModel):
"""
User created Reinforcement Learning Model prediction model.
"""
def fit(self, data_dictionary: Dict[str, Any], pair: str = ''):
agent_params = self.freqai_info['model_training_parameters']
reward_params = self.freqai_info['model_reward_parameters']
train_df = data_dictionary["train_features"]
test_df = data_dictionary["test_features"]
eval_freq = agent_params.get("eval_cycles", 4) * len(test_df)
total_timesteps = agent_params["train_cycles"] * len(train_df)
# price data for model training and evaluation
price = self.dd.historic_data[pair][f"{self.config['timeframe']}"].tail(len(train_df.index))
price_test = self.dd.historic_data[pair][f"{self.config['timeframe']}"].tail(
len(test_df.index))
env_id = "CartPole-v1"
num_cpu = 4
train_env = SubprocVecEnv([make_env(env_id, i, 1, train_df, price, reward_params,
self.CONV_WIDTH) for i in range(num_cpu)])
eval_env = SubprocVecEnv([make_env(env_id, i, 1, test_df, price_test, reward_params,
self.CONV_WIDTH) for i in range(num_cpu)])
path = self.dk.data_path
eval_callback = EvalCallback(eval_env, best_model_save_path=f"{path}/",
log_path=f"{path}/ppo/logs/", eval_freq=int(eval_freq),
deterministic=True, render=False)
# model arch
policy_kwargs = dict(activation_fn=th.nn.ReLU,
net_arch=[256, 256, 128])
model = PPO('MlpPolicy', train_env, policy_kwargs=policy_kwargs,
tensorboard_log=f"{path}/ppo/tensorboard/", learning_rate=0.00025, gamma=0.9, verbose=1
)
model.learn(
total_timesteps=int(total_timesteps),
callback=eval_callback
)
print('Training finished!')
return model
class MyRLEnv(Base3ActionRLEnv):
"""
User can override any function in BaseRLEnv and gym.Env
"""
def calculate_reward(self, action):
if self._last_trade_tick is None:
return 0.
# close long
if (action == Actions.Short.value or
action == Actions.Neutral.value) and self._position == Positions.Long:
last_trade_price = self.add_buy_fee(self.prices.iloc[self._last_trade_tick].open)
current_price = self.add_sell_fee(self.prices.iloc[self._current_tick].open)
return float(np.log(current_price) - np.log(last_trade_price))
# close short
if (action == Actions.Long.value or
action == Actions.Neutral.value) and self._position == Positions.Short:
last_trade_price = self.add_sell_fee(self.prices.iloc[self._last_trade_tick].open)
current_price = self.add_buy_fee(self.prices.iloc[self._current_tick].open)
return float(np.log(last_trade_price) - np.log(current_price))
return 0.