Graph Matrix JSP Environment Example

from jsp_instance_utils.instances import ft06, ft06_makespan
from graph_matrix_jsp_env.disjunctive_jsp_env import DisjunctiveGraphJspEnv
import numpy as np

def main():
    env = DisjunctiveGraphJspEnv(
        jsp_instance=ft06,
        c_lb=ft06_makespan,
        reward_function="makespan"
    )

    done = False
    info = {}
    while not done:
        # get valid action mask. sample expects it to be a numpy array of type int8
        mask = np.array(env.valid_action_mask()).astype(np.int8)
        action = env.action_space.sample(mask=mask)
        state, reward, terminated, truncated, info = env.step(action)

        done = terminated or truncated

    # console rendering
    env.render()
    print(f"makespan: {env.get_makespan()}")

if __name__ == '__main__':
    main()

         ╔═══════════════════════════════════════════════════════╗
Job 0    ║    ███  ██████████           ██████                   ║ Machine 0   █
Job 1    ║   ██████   ████    ███████     ███████     ███████████║ Machine 1   █
Job 2    ║█████████████                        ████████ ██████   ║ Machine 2   █
Job 3    ║                ████     ██████████     █████████████  ║ Machine 3   █
Job 4    ║     ███████ █████████████                             ║ Machine 4   █
Job 5    ║███                ██  ██████████████████              ║ Machine 5   █
         ╚╦════╤════╤════╤════╤════╦════╤════╤════╤════╤════╦════╝
          0.0                      32.7                     65.5
makespan: 71