[Refactor] Update logging backend to use wandb

Author: Ibinarriaga8

sota-implementations/cql/cql_offline.py

@@ -191,10 +191,6 @@ def update(data, policy_eval_start, iteration):
             metrics_to_log.update(timeit.todict(prefix="time"))
             metrics_to_log["time/speed"] = pbar.format_dict["rate"]
 
-            if i % evaluation_interval == 0:
-                print(
-                    f"Step {i}: loss={loss.item():.4f}, eval_reward={eval_reward:.4f}"
-                )
             log_metrics(logger, metrics_to_log, i)
 
     pbar.close()

sota-implementations/cql/offline_config.yaml

@@ -9,7 +9,7 @@ env:
 
 # logger
 logger:
-  backend: csv
+  backend: wandb
   project_name: torchrl_example_cql
   group_name: null
   exp_name: cql_${replay_buffer.dataset}