ba37ff3652f24e8416cf4eb13e9dd66976f843e8,examples/tf/ppo_memorize_digits.py,,ppo_memorize_digits,#Any#Any#Any#,21

Before Change

                                      padding="VALID",
                                      hidden_sizes=(256, ))  // yapf: disable

        baseline = GaussianCNNBaseline(
            env_spec=env.spec,
            regressor_args=dict(filters=(
                                            (32, (5, 5)),
                                            (64, (3, 3)),
                                            (64, (2, 2)),
                                        ),
                                strides=(4, 2, 1),
                                padding="VALID",
                                hidden_sizes=(256, ),
                                use_trust_region=True))  // yapf: disable

        algo = PPO(env_spec=env.spec,
                   policy=policy,

After Change

                                      padding="VALID",
                                      hidden_sizes=(256, ))  // yapf: disable

        baseline = GaussianCNNBaseline(
            env_spec=env.spec,
            filters=(
                (32, (5, 5)),
                (64, (3, 3)),
                (64, (2, 2)),
            ),
            strides=(4, 2, 1),
            padding="VALID",
            hidden_sizes=(256, ),
            use_trust_region=True)  // yapf: disable

        algo = PPO(env_spec=env.spec,
                   policy=policy,

Instances