network.putInt(Network.sizeOf(obs));
network.putObservation(obs);
}
protected void onEnvStep() {
Action action = network.getAction();
Reward_observation_terminal rewardObservation = env.env_step(action);
network.clearSendBuffer();
network.putInt(Network.kEnvStep);
network.putInt(Network.sizeOf(rewardObservation));