Support for online guidance with vbd_online (#431)

* Delete obsolete vbd functions

* Format

* Remove all instances of use_vbd and alike -> unified guidance mode

* Create dataclass struct for VBD online predictionsions

* Delete world_time_steps as it is no longer used

* Leaving a todo

* Make sure to demean the VBD predicted trajectories

* Data analysis and minor changes

* Guidance data analysis notebook

* Script to process guidance data

* Fixes

* Fixes

* Update nb

* Fixes and make sure to always wrap the yaws.

* Fixes and make sure to always wrap the yaws.

* Example data to work with

Author: daphne-cornelisse

.gitignore

@@ -29,7 +29,7 @@ data/processed/training/*
 data/processed/testing/*
 data/processed/pop_play/*
 data/processed/hand_designed/*
-analyze/figures/*
+figures/
 data/other/*
 wosac/
 

baselines/ppo/config/ppo_base_puffer.yaml

@@ -42,11 +42,7 @@ environment: # Overrides default environment configs (see pygpudrive/env/config.
   obs_radius: 50.0 # Visibility radius of the agents
   action_space_steer_disc: 13
   action_space_accel_disc: 7
-  # Versatile Behavior Diffusion (VBD): This will slow down training
-  use_vbd: false
   init_steps: 0
-  vbd_trajectory_weight: 0.1 # Importance of distance to the vbd trajectories in the reward function
-  vbd_in_obs: false
 
 wandb:
   entity: ""

baselines/ppo/config/ppo_guided_autonomy.yaml

@@ -75,7 +75,7 @@ train:
 
   # # # PPO # # #
   torch_deterministic: false
-  total_timesteps: 2_000_000_000
+  total_timesteps: 4_000_000_000
   batch_size: 131072
   minibatch_size: 8192
   learning_rate: 3e-4
@@ -105,7 +105,7 @@ train:
     num_parameters: 0 # Total trainable parameters, to be filled at runtime
 
   # # # Checkpointing # # #
-  checkpoint_interval: 50 # Save policy every k iterations
+  checkpoint_interval: 500 # Save policy every k iterations
   checkpoint_path: "./runs"
 
   # # # Rendering # # #

baselines/ppo/config/ppo_population.yaml

@@ -42,11 +42,6 @@ environment: # Overrides default environment configs (see pygpudrive/env/config.
   action_space_steer_disc: 13
   action_space_accel_disc: 7
   init_steps: 0 # Warmup steps
-  # Versatile Behavior Diffusion (VBD): This will slow down training
-  use_vbd: false
-  vbd_model_path: "gpudrive/integrations/vbd/weights/epoch=18.ckpt"
-  vbd_trajectory_weight: 0.1 # Importance of distance to the vbd trajectories in the reward function
-  vbd_in_obs: false
 
 wandb:
   entity: ""

baselines/ppo/ppo_guided_autonomy.py

@@ -125,10 +125,7 @@ def run(
     obs_radius: Annotated[Optional[float], typer.Option(help="The radius for the observation")] = None,
     collision_behavior: Annotated[Optional[str], typer.Option(help="The collision behavior; 'ignore' or 'remove'")] = None,
     remove_non_vehicles: Annotated[Optional[int], typer.Option(help="Remove non-vehicles from the scene; 0 or 1")] = None,
-    use_vbd: Annotated[Optional[bool], typer.Option(help="Use VBD model for trajectory predictions")] = False,
     vbd_model_path: Annotated[Optional[str], typer.Option(help="Path to VBD model checkpoint")] = None,
-    vbd_trajectory_weight: Annotated[Optional[float], typer.Option(help="Weight for VBD trajectory deviation penalty")] = 0.1,
-    vbd_in_obs: Annotated[Optional[bool], typer.Option(help="Include VBD predictions in the observation")] = False,
     init_steps: Annotated[Optional[int], typer.Option(help="Environment warmup steps")] = 0,
 
     # Train options
@@ -185,10 +182,7 @@ def run(
         "remove_non_vehicles": None
         if remove_non_vehicles is None
         else bool(remove_non_vehicles),
-        "use_vbd": use_vbd,
         "vbd_model_path": vbd_model_path,
-        "vbd_trajectory_weight": vbd_trajectory_weight,
-        "vbd_in_obs": vbd_in_obs,
         "init_steps": init_steps,
     }
     config.environment.update(

baselines/ppo/ppo_pufferlib.py

@@ -172,9 +172,6 @@ def run(
     obs_radius: Annotated[Optional[float], typer.Option(help="The radius for the observation")] = None,
     collision_behavior: Annotated[Optional[str], typer.Option(help="The collision behavior; 'ignore' or 'remove'")] = None,
     remove_non_vehicles: Annotated[Optional[int], typer.Option(help="Remove non-vehicles from the scene; 0 or 1")] = None,
-    use_vbd: Annotated[Optional[bool], typer.Option(help="Use VBD model for trajectory predictions")] = False,
-    vbd_trajectory_weight: Annotated[Optional[float], typer.Option(help="Weight for VBD trajectory deviation penalty")] = 0.1,
-    vbd_in_obs: Annotated[Optional[bool], typer.Option(help="Include VBD predictions in the observation")] = False,
     init_steps: Annotated[Optional[int], typer.Option(help="Environment warmup steps")] = 0,
     # Train options
     seed: Annotated[Optional[int], typer.Option(help="The seed for training")] = None,
@@ -227,9 +224,6 @@ def run(
         "remove_non_vehicles": None
         if remove_non_vehicles is None
         else bool(remove_non_vehicles),
-        "use_vbd": use_vbd,
-        "vbd_trajectory_weight": vbd_trajectory_weight,
-        "vbd_in_obs": vbd_in_obs,
         "init_steps": init_steps,
     }
     config.environment.update(