3bb0fd78af4f9dbddf7c01dde71844a283099650,ch06/01_dqn_pong.py,,,#,71

Before Change

        return self.fc(conv_out)


if __name__ == "__main__":
    env = BufferWrapper(ImageWrapper(gym.make("Pong-v4")), n_steps=4)
    net = DQN(env.observation_space.shape, env.action_space.n)
    print(net)
    out = net(Variable(torch.FloatTensor([env.reset()])))
    print(out.size())
    pass

After Change

        return self.fc(conv_out)


Experience = collections.namedtuple("Experience", field_names=["state", "action", "reward", "done", "new_state"])


class ExperienceBuffer:
    def __init__(self, capacity):

Instances