diff --git a/README.md b/README.md index 1b8928b..88b6b45 100644 --- a/README.md +++ b/README.md @@ -24,7 +24,7 @@ and go to `localhost:6006` in your browser. If you're running headlessly, try po The console output is also available in a form: ``` -| train | E: 1 | S: 1000 | D: 0.8 s | R: 0.0000 | BR: 0.0000 | ALOSS: 0.0000 | CLOSS: 0.0000 | RLOSS: 0.0000 +| train | E: 1 | S: 1000 | D: 0.8 s | R: 0.0000 | BR: 0.0000 | A_LOSS: 0.0000 | CR_LOSS: 0.0000 | CU_LOSS: 0.0000 ``` a training entry decodes as: ``` diff --git a/train.py b/train.py index a3178b0..3f27ed3 100644 --- a/train.py +++ b/train.py @@ -106,8 +106,8 @@ def evaluate(env, agent, video, num_episodes, L, step, args): mean_ep_reward = np.mean(all_ep_rewards) best_ep_reward = np.max(all_ep_rewards) - L.log('eval/' + prefix + 'mean_episode_reward', mean_ep_reward, step) - L.log('eval/' + prefix + 'best_episode_reward', best_ep_reward, step) + L.log('eval/' + prefix + '_mean_episode_reward', mean_ep_reward, step) + L.log('eval/' + prefix + '_best_episode_reward', best_ep_reward, step) run_eval_loop(sample_stochastically=True) run_eval_loop(sample_stochastically=False)