adelorenzo/stable
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

callback function and TDQN model added

Browse Source
This commit is contained in:
MukavaValkku 2022-08-13 20:05:21 +03:00 committed by robcaulk
parent 01232e9a1f
commit cd3fe44424
1 changed files with 116 additions and 27 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

143
freqtrade/freqai/prediction_models/ReinforcementLearning.py
View File

@ -5,14 +5,23 @@ import numpy as np
import numpy.typing as npt import numpy.typing as npt
import pandas as pd import pandas as pd
from pandas import DataFrame from pandas import DataFrame
from stable_baselines.common.callbacks import CallbackList, CheckpointCallback, EvalCallback
from freqtrade.freqai.data_kitchen import FreqaiDataKitchen from freqtrade.freqai.data_kitchen import FreqaiDataKitchen
from freqtrade.freqai.freqai_interface import IFreqaiModel from freqtrade.freqai.freqai_interface import IFreqaiModel
from freqtrade.freqai.prediction_models.RL.RLPrediction_agent import RLPrediction_agent from freqtrade.freqai.prediction_models.RL.RLPrediction_agent import RLPrediction_agent
from freqtrade.freqai.prediction_models.RL.RLPrediction_agent_v2 import TDQN
#from freqtrade.freqai.prediction_models.RL.RLPrediction_env import GymAnytrading #from freqtrade.freqai.prediction_models.RL.RLPrediction_env import GymAnytrading
from freqtrade.freqai.prediction_models.RL.RLPrediction_env import DEnv from freqtrade.freqai.prediction_models.RL.RLPrediction_env import DEnv
from freqtrade.persistence import Trade from freqtrade.persistence import Trade
from stable_baselines3.common.vec_env import SubprocVecEnv
from stable_baselines3.common.monitor import Monitor
import torch as th
from stable_baselines3.common.callbacks import CallbackList, CheckpointCallback, EvalCallback, StopTrainingOnRewardThreshold
from stable_baselines3.common.buffers import ReplayBuffer
from stable_baselines3 import PPO
logger = logging.getLogger(__name__) logger = logging.getLogger(__name__)
@ -74,47 +83,127 @@ class ReinforcementLearningModel(IFreqaiModel):
def fit(self, data_dictionary: Dict[str, Any], pair: str = ''): def fit(self, data_dictionary: Dict[str, Any], pair: str = ''):
train_df = data_dictionary["train_features"] # train_df = data_dictionary["train_features"]
# train_labels = data_dictionary["train_labels"] # # train_labels = data_dictionary["train_labels"]
test_df = data_dictionary["test_features"] # test_df = data_dictionary["test_features"]
# test_labels = data_dictionary["test_labels"] # # test_labels = data_dictionary["test_labels"]
# sep = '/' # # sep = '/'
# coin = pair.split(sep, 1)[0] # # coin = pair.split(sep, 1)[0]
# price = train_df[f"%-{coin}raw_price_{self.config['timeframe']}"] # # price = train_df[f"%-{coin}raw_price_{self.config['timeframe']}"]
# price.reset_index(inplace=True, drop=True) # # price.reset_index(inplace=True, drop=True)
# price = price.to_frame() # # price = price.to_frame()
price = self.dd.historic_data[pair][f"{self.config['timeframe']}"].tail(len(train_df.index)) # 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))
# #train_env = GymAnytrading(train_df, price, self.CONV_WIDTH)
# agent_params = self.freqai_info['model_training_parameters']
# reward_params = self.freqai_info['model_reward_parameters']
model_name = 'ppo' # train_env = DEnv(df=train_df, prices=price, window_size=self.CONV_WIDTH, reward_kwargs=reward_params)
# #eval_env = DEnv(df=test_df, prices=price_test, window_size=self.CONV_WIDTH, reward_kwargs=reward_params)
#env_instance = GymAnytrading(train_df, price, self.CONV_WIDTH) # #env_instance = SubprocVecEnv([DEnv(df=train_df, prices=price, window_size=self.CONV_WIDTH, reward_kwargs=reward_params)])
# #train_env.reset()
# #eval_env.reset()
# # model
# #policy_kwargs = dict(net_arch=[512, 512, 512])
# policy_kwargs = dict(activation_fn=th.nn.Tanh,
# net_arch=[256, 256, 256])
# agent = RLPrediction_agent(train_env)
# #eval_agent = RLPrediction_agent(eval_env)
# # PPO
# model_name = 'ppo'
# model = agent.get_model(model_name, model_kwargs=agent_params, policy_kwargs=policy_kwargs)
# trained_model = agent.train_model(model=model,
# tb_log_name=model_name,
# model_kwargs=agent_params,
# train_df=train_df,
# test_df=test_df,
# price=price,
# price_test=price_test,
# window_size=self.CONV_WIDTH)
# # best_model = eval_agent.train_model(model=model,
# # tb_log_name=model_name,
# # model_kwargs=agent_params,
# # eval=eval_env)
# # TDQN
# # model_name = 'TDQN'
# # model = TDQN('TMultiInputPolicy', train_env, policy_kwargs=policy_kwargs, tensorboard_log='./tensorboard_log/',
# # learning_rate=agent_params["learning_rate"], gamma=0.9,
# # target_update_interval=5000, buffer_size=50000,
# # exploration_initial_eps=1, exploration_final_eps=0.1,
# # replay_buffer_class=ReplayBuffer
# # )
# # trained_model = agent.train_model(model=model,
# # tb_log_name=model_name,
# # model_kwargs=agent_params)
# #model.learn(
# # total_timesteps=5000,
# # callback=callback
# # )
agent_params = self.freqai_info['model_training_parameters'] agent_params = self.freqai_info['model_training_parameters']
reward_params = self.freqai_info['model_reward_parameters'] reward_params = self.freqai_info['model_reward_parameters']
train_df = data_dictionary["train_features"]
test_df = data_dictionary["test_features"]
env_instance = DEnv(df=train_df, prices=price, window_size=self.CONV_WIDTH, reward_kwargs=reward_params) # price data for model training and evaluation
agent = RLPrediction_agent(env_instance) 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))
# checkpoint_callback = CheckpointCallback(save_freq=1000, save_path='./logs/') # environments
# eval_callback = EvalCallback(test_df, best_model_save_path='./models/', train_env = DEnv(df=train_df, prices=price, window_size=self.CONV_WIDTH, reward_kwargs=reward_params)
# log_path='./logs/', eval_freq=10000, eval = DEnv(df=test_df, prices=price_test, window_size=self.CONV_WIDTH, reward_kwargs=reward_params)
# deterministic=True, render=False) eval_env = Monitor(eval, ".")
eval_env.reset()
# #Create the callback list # this should be in config - TODO
# callback = CallbackList([checkpoint_callback, eval_callback]) agent_type = 'tdqn'
model = agent.get_model(model_name, model_kwargs=agent_params) path = self.dk.data_path
trained_model = agent.train_model(model=model, eval_callback = EvalCallback(eval_env, best_model_save_path=f"{path}/",
tb_log_name=model_name, log_path=f"{path}/{agent_type}/logs/", eval_freq=10000,
model_kwargs=agent_params) deterministic=True, render=False)
#eval_callback=callback)
# model arch
policy_kwargs = dict(activation_fn=th.nn.Tanh,
net_arch=[512, 512, 512])
if agent_type == 'tdqn':
model = TDQN('TMultiInputPolicy', train_env, policy_kwargs=policy_kwargs, tensorboard_log=f"{path}/{agent_type}/tensorboard/",
learning_rate=0.00025, gamma=0.9,
target_update_interval=5000, buffer_size=50000,
exploration_initial_eps=1, exploration_final_eps=0.1,
replay_buffer_class=ReplayBuffer
)
elif agent_type == 'ppo':
model = PPO('MultiInputPolicy', train_env, policy_kwargs=policy_kwargs, tensorboard_log=f"{path}/{agent_type}/tensorboard/",
learning_rate=0.00025, gamma=0.9
)
model.learn(
total_timesteps=agent_params["total_timesteps"],
callback=eval_callback
)
print('Training finished!') print('Training finished!')
return trained_model return model
def get_state_info(self, pair): def get_state_info(self, pair):
open_trades = Trade.get_trades(trade_filter=Trade.is_open.is_(True)) open_trades = Trade.get_trades(trade_filter=Trade.is_open.is_(True))