Function test_cn_ppo_strategy

tests/rl/test_saoe_simple.py:278–306

()

def test_cn_ppo_strategy():
    set_log_with_config(C.logging_config)
    # The data starts with 9:31 and ends with 15:00
    orders = pickle_styled.load_orders(CN_ORDER_DIR, start_time=pd.Timestamp("9:31"), end_time=pd.Timestamp("14:58"))
    assert len(orders) == 40

    state_interp = FullHistoryStateInterpreter(8, 240, 6, PickleProcessedDataProvider(CN_FEATURE_DATA_DIR))
    action_interp = CategoricalActionInterpreter(4)
    network = Recurrent(state_interp.observation_space)
    policy = PPO(network, state_interp.observation_space, action_interp.action_space, 1e-4)
    policy.load_state_dict(torch.load(CN_POLICY_WEIGHTS_DIR / "ppo_recurrent_30min.pth", map_location="cpu"))
    csv_writer = CsvWriter(Path(__file__).parent / ".output")

    backtest(
        partial(SingleAssetOrderExecutionSimple, data_dir=CN_DATA_DIR, ticks_per_step=30),
        state_interp,
        action_interp,
        orders,
        policy,
        [ConsoleWriter(total_episodes=len(orders)), csv_writer],
        concurrency=4,
    )

    metrics = pd.read_csv(Path(__file__).parent / ".output" / "result.csv")
    assert len(metrics) == len(orders)
    assert np.isclose(metrics["ffr"].mean(), 1.0)
    assert np.isclose(metrics["pa"].mean(), -16.21578303474833)
    assert np.isclose(metrics["market_price"].mean(), 58.68277690875527)
    assert np.isclose(metrics["trade_price"].mean(), 58.76063985000002)


def test_ppo_train():