stable/freqtrade/freqai/RL/ReinforcementLearnerCustomAgent.py

262 lines
9.3 KiB
Python

import logging
import torch as th
from typing import Any, Dict, List, Optional, Tuple, Type, Union
from freqtrade.freqai.RL.BaseReinforcementLearningModel import BaseReinforcementLearningModel
from stable_baselines3 import DQN
from stable_baselines3.common.buffers import ReplayBuffer
from freqtrade.freqai.data_kitchen import FreqaiDataKitchen
from pathlib import Path
from stable_baselines3.dqn.policies import (CnnPolicy, DQNPolicy, MlpPolicy,
QNetwork)
from torch import nn
import gym
from stable_baselines3.common.torch_layers import (BaseFeaturesExtractor,
FlattenExtractor)
from stable_baselines3.common.type_aliases import GymEnv, Schedule
from stable_baselines3.common.policies import BasePolicy
logger = logging.getLogger(__name__)
class ReinforcementLearnerCustomAgent(BaseReinforcementLearningModel):
"""
User can customize agent by defining the class and using it directly.
Here the example is "TDQN"
Warning!
This is an advanced example of how a user may create and use a highly
customized model class (which can inherit from existing classes,
similar to how the example below inherits from DQN).
This file is for example purposes only, and should not be run.
"""
def fit_rl(self, data_dictionary: Dict[str, Any], dk: FreqaiDataKitchen):
train_df = data_dictionary["train_features"]
total_timesteps = self.freqai_info["rl_config"]["train_cycles"] * len(train_df)
policy_kwargs = dict(activation_fn=th.nn.ReLU,
net_arch=[256, 256, 128])
# TDQN is a custom agent defined below
model = TDQN(self.policy_type, self.train_env,
tensorboard_log=str(Path(dk.data_path / "tensorboard")),
policy_kwargs=policy_kwargs,
**self.freqai_info['model_training_parameters']
)
model.learn(
total_timesteps=int(total_timesteps),
callback=self.eval_callback
)
if Path(dk.data_path / "best_model.zip").is_file():
logger.info('Callback found a best model.')
best_model = self.MODELCLASS.load(dk.data_path / "best_model")
return best_model
logger.info('Couldnt find best model, using final model instead.')
return model
# User creates their custom agent and networks as shown below
def create_mlp_(
input_dim: int,
output_dim: int,
net_arch: List[int],
activation_fn: Type[nn.Module] = nn.ReLU,
squash_output: bool = False,
) -> List[nn.Module]:
dropout = 0.2
if len(net_arch) > 0:
number_of_neural = net_arch[0]
modules = [
nn.Linear(input_dim, number_of_neural),
nn.BatchNorm1d(number_of_neural),
nn.LeakyReLU(),
nn.Dropout(dropout),
nn.Linear(number_of_neural, number_of_neural),
nn.BatchNorm1d(number_of_neural),
nn.LeakyReLU(),
nn.Dropout(dropout),
nn.Linear(number_of_neural, number_of_neural),
nn.BatchNorm1d(number_of_neural),
nn.LeakyReLU(),
nn.Dropout(dropout),
nn.Linear(number_of_neural, number_of_neural),
nn.BatchNorm1d(number_of_neural),
nn.LeakyReLU(),
nn.Dropout(dropout),
nn.Linear(number_of_neural, output_dim)
]
return modules
class TDQNetwork(QNetwork):
def __init__(self,
observation_space: gym.spaces.Space,
action_space: gym.spaces.Space,
features_extractor: nn.Module,
features_dim: int,
net_arch: Optional[List[int]] = None,
activation_fn: Type[nn.Module] = nn.ReLU,
normalize_images: bool = True
):
super().__init__(
observation_space=observation_space,
action_space=action_space,
features_extractor=features_extractor,
features_dim=features_dim,
net_arch=net_arch,
activation_fn=activation_fn,
normalize_images=normalize_images
)
action_dim = self.action_space.n
q_net = create_mlp_(self.features_dim, action_dim, self.net_arch, self.activation_fn)
self.q_net = nn.Sequential(*q_net).apply(self.init_weights)
def init_weights(self, m):
if type(m) == nn.Linear:
th.nn.init.kaiming_uniform_(m.weight)
class TDQNPolicy(DQNPolicy):
def __init__(
self,
observation_space: gym.spaces.Space,
action_space: gym.spaces.Space,
lr_schedule: Schedule,
net_arch: Optional[List[int]] = None,
activation_fn: Type[nn.Module] = nn.ReLU,
features_extractor_class: Type[BaseFeaturesExtractor] = FlattenExtractor,
features_extractor_kwargs: Optional[Dict[str, Any]] = None,
normalize_images: bool = True,
optimizer_class: Type[th.optim.Optimizer] = th.optim.Adam,
optimizer_kwargs: Optional[Dict[str, Any]] = None,
):
super().__init__(
observation_space=observation_space,
action_space=action_space,
lr_schedule=lr_schedule,
net_arch=net_arch,
activation_fn=activation_fn,
features_extractor_class=features_extractor_class,
features_extractor_kwargs=features_extractor_kwargs,
normalize_images=normalize_images,
optimizer_class=optimizer_class,
optimizer_kwargs=optimizer_kwargs
)
@staticmethod
def init_weights(module: nn.Module, gain: float = 1) -> None:
"""
Orthogonal initialization (used in PPO and A2C)
"""
if isinstance(module, (nn.Linear, nn.Conv2d)):
nn.init.kaiming_uniform_(module.weight)
if module.bias is not None:
module.bias.data.fill_(0.0)
def make_q_net(self) -> TDQNetwork:
# Make sure we always have separate networks for features extractors etc
net_args = self._update_features_extractor(self.net_args, features_extractor=None)
return TDQNetwork(**net_args).to(self.device)
class TMultiInputPolicy(TDQNPolicy):
def __init__(
self,
observation_space: gym.spaces.Space,
action_space: gym.spaces.Space,
lr_schedule: Schedule,
net_arch: Optional[List[int]] = None,
activation_fn: Type[nn.Module] = nn.ReLU,
features_extractor_class: Type[BaseFeaturesExtractor] = FlattenExtractor,
features_extractor_kwargs: Optional[Dict[str, Any]] = None,
normalize_images: bool = True,
optimizer_class: Type[th.optim.Optimizer] = th.optim.Adam,
optimizer_kwargs: Optional[Dict[str, Any]] = None,
):
super().__init__(
observation_space,
action_space,
lr_schedule,
net_arch,
activation_fn,
features_extractor_class,
features_extractor_kwargs,
normalize_images,
optimizer_class,
optimizer_kwargs,
)
class TDQN(DQN):
policy_aliases: Dict[str, Type[BasePolicy]] = {
"MlpPolicy": MlpPolicy,
"CnnPolicy": CnnPolicy,
"TMultiInputPolicy": TMultiInputPolicy,
}
def __init__(
self,
policy: Union[str, Type[TDQNPolicy]],
env: Union[GymEnv, str],
learning_rate: Union[float, Schedule] = 1e-4,
buffer_size: int = 1000000, # 1e6
learning_starts: int = 50000,
batch_size: int = 32,
tau: float = 1.0,
gamma: float = 0.99,
train_freq: Union[int, Tuple[int, str]] = 4,
gradient_steps: int = 1,
replay_buffer_class: Optional[ReplayBuffer] = None,
replay_buffer_kwargs: Optional[Dict[str, Any]] = None,
optimize_memory_usage: bool = False,
target_update_interval: int = 10000,
exploration_fraction: float = 0.1,
exploration_initial_eps: float = 1.0,
exploration_final_eps: float = 0.05,
max_grad_norm: float = 10,
tensorboard_log: Optional[str] = None,
create_eval_env: bool = False,
policy_kwargs: Optional[Dict[str, Any]] = None,
verbose: int = 1,
seed: Optional[int] = None,
device: Union[th.device, str] = "auto",
_init_setup_model: bool = True,
):
super().__init__(
policy=policy,
env=env,
learning_rate=learning_rate,
buffer_size=buffer_size,
learning_starts=learning_starts,
batch_size=batch_size,
tau=tau,
gamma=gamma,
train_freq=train_freq,
gradient_steps=gradient_steps,
replay_buffer_class=replay_buffer_class, # No action noise
replay_buffer_kwargs=replay_buffer_kwargs,
optimize_memory_usage=optimize_memory_usage,
target_update_interval=target_update_interval,
exploration_fraction=exploration_fraction,
exploration_initial_eps=exploration_initial_eps,
exploration_final_eps=exploration_final_eps,
max_grad_norm=max_grad_norm,
tensorboard_log=tensorboard_log,
create_eval_env=create_eval_env,
policy_kwargs=policy_kwargs,
verbose=verbose,
seed=seed,
device=device,
_init_setup_model=_init_setup_model
)