Merge branch 'nil-custom-obs-wrapper' into refactor

experiments/flatland_random_sparse_small/local_conflict_obs_fc_net/ppo.yaml

+flatland-random-sparse-small-local-conflict-fc-ppo:
+    run: PPO
+    env: flatland_random_sparse_small
+    stop:
+        timesteps_total: 10000000  # 1e7
+    checkpoint_freq: 10
+    checkpoint_at_end: True
+    keep_checkpoints_num: 5
+    checkpoint_score_attr: episode_reward_mean
+    config:
+        clip_rewards: True
+        clip_param: 0.1
+        vf_clip_param: 500.0
+        entropy_coeff: 0.01
+        # effective batch_size: train_batch_size * num_agents_in_each_environment [5, 10]
+        # see https://github.com/ray-project/ray/issues/4628
+        train_batch_size: 1000  # 5000
+        rollout_fragment_length: 50  # 100
+        sgd_minibatch_size: 100  # 500
+        num_sgd_iter: 10
+        num_workers: 1
+        num_envs_per_worker: 1
+        batch_mode: truncate_episodes
+        observation_filter: NoFilter
+        vf_share_layers: True
+        vf_loss_coeff: 0.5
+        num_gpus: 0
+
+        env_config:
+            min_seed: 1002
+            max_seed: 213783
+            min_test_seed: 0
+            max_test_seed: 100
+            # After how many episodes the level should be regenerated:
+            reset_env_freq: 1
+            observation: localConflict
+            observation_config:
+                max_depth: 2
+                shortest_path_max_depth: 30
+                n_local: 5
+            regenerate_rail_on_reset: True
+            regenerate_schedule_on_reset: True
+            render: False
+
+        model:
+            fcnet_activation: relu
+            fcnet_hiddens: [256, 256]
+            vf_share_layers: True  # False