332d8ed93fd24838cd0e2ac1e2bad74f72e04784,ch09/02_cartpole_reinforce.py,,,#,42

Before Change

            batch_actions.extend(cur_actions)
            batch_qvals.extend(calc_qvals(cur_rewards))
            cur_states.clear()
            cur_actions.clear()
            cur_rewards.clear()
            batch_episodes += 1

        // handle new rewards

After Change

    batch_episodes = 0
    batch_states, batch_actions, batch_qvals = [], [], []
    cur_rewards = []

    for step_idx, exp in enumerate(exp_source):
        batch_states.append(exp.state)
        batch_actions.append(int(exp.action))

Instances