Update dataset optim to equality in ppo

Author: AntoinePrv

irl/algo/policy_methods.py

@@ -325,7 +325,7 @@ def terminate_trajectory_and_data_collection(engine):
 
     @agent.on(Events.EPOCH_COMPLETED)
     def optimize(engine):
-        if len(engine.state.trajectories) > dataset_size:
+        if len(engine.state.trajectories) >= dataset_size:
             engine.terminate()
             sample_elem = engine.state.trajectories[0]
             dataloader = DataLoader(