diff --git a/irlc/lectures/lec02/lecture_02_optimal_inventory.py b/irlc/lectures/lec02/lecture_02_optimal_inventory.py new file mode 100644 index 0000000000000000000000000000000000000000..48eb5e1320a987ac14f60aab9cf427c080709cbd --- /dev/null +++ b/irlc/lectures/lec02/lecture_02_optimal_inventory.py @@ -0,0 +1,15 @@ +# This file may not be shared/redistributed without permission. Please read copyright notice in the git repo. If this file contains other copyright notices disregard this text. +from irlc import Agent, interactive +from irlc.lectures.lec01.viz_inventory_environment import VizInventoryEnvironment + +if __name__ == "__main__": + env = VizInventoryEnvironment(render_mode='human') + from irlc.ex02.inventory import InventoryDPModel + from irlc.ex02.dp_agent import DynamicalProgrammingAgent + + from irlc import train, interactive + agent = DynamicalProgrammingAgent(env, model=InventoryDPModel()) + + env, agent = interactive(env, agent) + n = 400 + stats, _ = train(env, agent, max_steps=n, num_episodes=1000, return_trajectory=False, verbose=False)