module moplayground.learning.inference

---

function load_mo_policy

load_mo_policy(
    config,
    tradeoff: numpy.ndarray,
    network_factory=<function make_moppo_networks at 0x797cd2801b20>,
    deterministic: bool = True
)

---

function load_moppo_network

load_moppo_network(
    config,
    network_factory=<function make_moppo_networks at 0x797cd2801b20>,
    path=None
) → tuple[moplayground.moppo.factory.MOPPONetworks, dict]

Loads the MOPPO object

---

function load_hypernetwork

load_hypernetwork(
    config,
    network_factory=<function make_moppo_networks at 0x797cd2801b20>,
    path=None
)

Loads policy inference function from PPO checkpoint.

---

function rollout_policy

rollout_policy(
    env,
    config,
    tradeoff=None,
    T=10.0,
    camera='track',
    width=1080,
    height=720
)

---

function get_num_objectives

get_num_objectives(config)