MarketEnv module
- class MarketEnv.MarketEnvironment[source]
Bases:
EnvA custom environment that simulates a market with two sellers and four buyers.
Each seller sets a price (the action), and each buyer has a certain willingness to pay. The state of the environment consists of the two prices set by the sellers and the current willingness to pay of each buyer.
The episode ends when all buyers have bought the products.
- reset()[source]
Reset the state of the environment to an initial state.
Returns
- array_like
The initial state of the environment.
- step(action)[source]
Execute one time step within the environment.
Parameters
- actionarray_like
The prices set by the sellers.
Returns
- array_like
The new state of the environment after taking the action.
- float
The reward achieved by the action.
- bool
A flag denoting whether the episode has ended.
- dict
An empty dictionary for additional info (unused in this environment).