def learn()

in phasic_policy_gradient/ppg.py [0:0]

• 13 lines of code
• 1 McCabe index (conditional complexity)

def learn(
    *,
    model,
    venv,
    ppo_hps,
    aux_lr,
    aux_mbsize,
    n_aux_epochs=6,
    n_pi=32,
    kl_ewma_decay=None,
    interacts_total=float("inf"),
    name2coef=None,
    comm=None,