tensortrade.environments.trading_environment module¶
-
class
tensortrade.environments.trading_environment.
TradingEnvironment
(action_strategy, reward_strategy, exchange, **kwargs)[source]¶ Bases:
gym.core.Env
A trading environment made for use with Gym-compatible reinforcement learning algorithms.
-
__init__
(action_strategy, reward_strategy, exchange, **kwargs)[source]¶ - Parameters
action_strategy (
ActionStrategy
) – The strategy for transforming an action into a TradeDefinition at each timestep.reward_strategy (
RewardStrategy
) – The strategy for determining the reward at each timestep.exchange (
AssetExchange
) – The AssetExchange that will be used to feed data from and execute trades within.kwargs (optional) – Additional arguments for tuning the environment, logging, etc.
-
reset
()[source]¶ Resets the state of the environment and returns an initial observation.
- Return type
pandas.DataFrame
- Returns
observation – the initial observation.
-
step
(action)[source]¶ Run one timestep within the environment based on the specified action.
- Parameters
action – The trade action provided by the agent for this timestep.
- Return type
- Returns
observation (pandas.DataFrame) – Provided by the environment’s exchange, often OHLCV or tick trade history data points. reward (float): An amount corresponding to the advantage gained by the agent in this timestep. done (bool): If True, the environment is complete and should be restarted. info (dict): Any auxiliary, diagnostic, or debugging information to output.
-