basic_profit_max

Classes

BasicProfitMax	Base class for reward functions in trading environments.
BasicRealizedProfitMax	Base class for reward functions in trading environments.
SharpeRatio	Base class for reward functions in trading environments.
SortinoRatio	Base class for reward functions in trading environments.
CalmarRatio	Base class for reward functions in trading environments.

Module Contents

class basic_profit_max.BasicProfitMax(cfg, initial_state)

Bases: trading.src.alg.environments.reward_functions.base_reward_function.RewardFunction

Base class for reward functions in trading environments. @TODO pass in the profit calculated from the position step

Parameters:

• cfg (trading.cli.alg.config.RewardConfig)

• initial_state (numpy.ndarray)

initial_net

previous_net

__repr__()

Return type:

str

reset()

Reset the reward function state.

compute_reward(pf, df, realized_profit)

Compute the reward for a given action and state transition.

Parameters:

• pf (trading.src.portfolio.portfolio.Portfolio)

• df (pandas.DataFrame)

• realized_profit (float)

Return type:

float

class basic_profit_max.BasicRealizedProfitMax(cfg, initial_state)

Bases: trading.src.alg.environments.reward_functions.base_reward_function.RewardFunction

Base class for reward functions in trading environments. @TODO pass in the profit calculated from the position step

Parameters:

• cfg (trading.cli.alg.config.RewardConfig)

• initial_state (numpy.ndarray)

compute_reward(pf, df, realized_profit)

Compute the reward for a given action and state transition.

Parameters:

• pf (trading.src.portfolio.portfolio.Portfolio)

• df (pandas.DataFrame)

• realized_profit (float)

Return type:

float

class basic_profit_max.SharpeRatio(cfg, initial_state)

Bases: trading.src.alg.environments.reward_functions.base_reward_function.RewardFunction

Base class for reward functions in trading environments. @TODO pass in the profit calculated from the position step

Parameters:

• cfg (trading.cli.alg.config.RewardConfig)

• initial_state (pandas.DataFrame)

initial_state

risk_free_rate

compute_reward(pf, df, realized_profit)

Compute the reward for a given action and state transition.

Parameters:

• pf (trading.src.portfolio.portfolio.Portfolio)

• df (pandas.DataFrame)

• realized_profit (float)

Return type:

float

class basic_profit_max.SortinoRatio(cfg, initial_state)

Bases: trading.src.alg.environments.reward_functions.base_reward_function.RewardFunction

Base class for reward functions in trading environments. @TODO pass in the profit calculated from the position step

Parameters:

• cfg (trading.cli.alg.config.RewardConfig)

• initial_state (pandas.DataFrame)

initial_state

risk_free_rate

compute_reward(pf, df, realized_profit)

Compute the reward for a given action and state transition.

Parameters:

• pf (trading.src.portfolio.portfolio.Portfolio)

• df (pandas.DataFrame)

• realized_profit (float)

Return type:

float

class basic_profit_max.CalmarRatio(cfg, initial_state)

Bases: trading.src.alg.environments.reward_functions.base_reward_function.RewardFunction

Base class for reward functions in trading environments. @TODO pass in the profit calculated from the position step

Parameters:

• cfg (trading.cli.alg.config.RewardConfig)

• initial_state (pandas.DataFrame)

initial_state

risk_free_rate

compute_reward(pf, df, realized_profit)

Compute the reward for a given action and state transition.

Parameters:

• pf (trading.src.portfolio.portfolio.Portfolio)

• df (pandas.DataFrame)

• realized_profit (float)

Return type:

float