Fix typo in README and import error (#18)

README.md

@@ -82,16 +82,16 @@ OSRL uses the `WandbLogger` in [FSRL](https://github.com/liuzuxin/FSRL) and [Pyr
 For example, to train the `bcql` method, simply run by overriding the default parameters:
 
 ```shell
-python examples/train/train_bcql.py --task OfflineCarCirvle-v0 --param1 args1 ...
+python examples/train/train_bcql.py --task OfflineCarCircle-v0 --param1 args1 ...
 ```
 By default, the config file and the logs during training will be written to `logs\` folder and the training plots can be viewed online using Wandb.
 
 You can also launch a sequence of experiments or in parallel via the [EasyRunner](https://github.com/liuzuxin/easy-runner) package, see `examples/train_all_tasks.py` for details.
 
 ### Evaluation
 To evaluate a trained agent, for example, a BCQ agent, simply run
-```
-python example/eval/eval_bcql.py --path path_to_model --eval_episodes 20
+```shell
+python examples/eval/eval_bcql.py --path path_to_model --eval_episodes 20
 ```
 It will load config file from `path_to_model/config.yaml` and model file from `path_to_model/checkpoints/model.pt`, run 20 episodes, and print the average normalized reward and cost. The pretrained checkpoints for all datasets are available [here](https://drive.google.com/drive/folders/1lZmw2NVNR4YGUdrkih9o3rTMDrWCI_jw?usp=sharing) for reference.
 

examples/eval/eval_bc.py

@@ -2,7 +2,6 @@
 from typing import Any, DefaultDict, Dict, List, Optional, Tuple
 
 import dsrl
-import gymnasium as gym  # noqa
 import numpy as np
 import pyrallis
 import torch
@@ -33,6 +32,9 @@ def eval(args: EvalConfig):
 
     if "Metadrive" in cfg["task"]:
         import gym
+    else:
+        import gymnasium as gym  # noqa
+
     env = gym.make(cfg["task"])
     env.set_target_cost(cfg["cost_limit"])
 

examples/eval/eval_bcql.py

@@ -2,7 +2,6 @@
 from typing import Any, DefaultDict, Dict, List, Optional, Tuple
 
 import dsrl
-import gymnasium as gym  # noqa
 import numpy as np
 import pyrallis
 import torch
@@ -33,6 +32,9 @@ def eval(args: EvalConfig):
 
     if "Metadrive" in cfg["task"]:
         import gym
+    else:
+        import gymnasium as gym  # noqa
+
     env = wrap_env(
         env=gym.make(cfg["task"]),
         reward_scale=cfg["reward_scale"],

examples/eval/eval_bearl.py

@@ -2,7 +2,6 @@
 from typing import Any, DefaultDict, Dict, List, Optional, Tuple
 
 import dsrl
-import gymnasium as gym  # noqa
 import numpy as np
 import pyrallis
 import torch
@@ -33,6 +32,9 @@ def eval(args: EvalConfig):
 
     if "Metadrive" in cfg["task"]:
         import gym
+    else:
+        import gymnasium as gym  # noqa
+
     env = wrap_env(
         env=gym.make(cfg["task"]),
         reward_scale=cfg["reward_scale"],

examples/eval/eval_cdt.py

@@ -2,7 +2,6 @@
 from typing import Any, DefaultDict, Dict, List, Optional, Tuple
 
 import dsrl
-import gymnasium as gym  # noqa
 import numpy as np
 import pyrallis
 import torch
@@ -35,6 +34,9 @@ def eval(args: EvalConfig):
 
     if "Metadrive" in cfg["task"]:
         import gym
+    else:
+        import gymnasium as gym  # noqa
+
     env = wrap_env(
         env=gym.make(cfg["task"]),
         reward_scale=cfg["reward_scale"],

examples/eval/eval_coptidice.py

@@ -2,7 +2,6 @@
 from typing import Any, DefaultDict, Dict, List, Optional, Tuple
 
 import dsrl
-import gymnasium as gym  # noqa
 import numpy as np
 import pyrallis
 import torch
@@ -33,6 +32,9 @@ def eval(args: EvalConfig):
 
     if "Metadrive" in cfg["task"]:
         import gym
+    else:
+        import gymnasium as gym  # noqa
+
     env = wrap_env(
         env=gym.make(cfg["task"]),
         reward_scale=cfg["reward_scale"],

examples/eval/eval_cpq.py

@@ -2,7 +2,6 @@
 from typing import Any, DefaultDict, Dict, List, Optional, Tuple
 
 import dsrl
-import gymnasium as gym  # noqa
 import numpy as np
 import pyrallis
 import torch
@@ -33,6 +32,9 @@ def eval(args: EvalConfig):
 
     if "Metadrive" in cfg["task"]:
         import gym
+    else:
+        import gymnasium as gym  # noqa
+
     env = wrap_env(
         env=gym.make(cfg["task"]),
         reward_scale=cfg["reward_scale"],