sp
/
tempestpy
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
Browse Source

set sb3 device to auto 

This automatically detects whether a GPU can be used for training.
refactoring
sp 10 months ago
parent
71854bae01
commit
36c04f1b81
1 changed files with 1 additions and 1 deletions
Whitespace
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Unified View
Diff Options
Show Stats Download Patch File Download Diff File
  1. 2
      examples/shields/rl/13_minigridsb.py

2
examples/shields/rl/13_minigridsb.py
View File

@ -35,7 +35,7 @@ def main():
env = MiniGridSbShieldingWrapper(env, shield_handler=shield_handler, create_shield_at_reset=False, mask_actions=args.shielding == ShieldingConfig.Full) env = MiniGridSbShieldingWrapper(env, shield_handler=shield_handler, create_shield_at_reset=False, mask_actions=args.shielding == ShieldingConfig.Full)
env = ActionMasker(env, mask_fn) env = ActionMasker(env, mask_fn)
logDir = create_log_dir(args) logDir = create_log_dir(args)
model = MaskablePPO("CnnPolicy", env, verbose=1, tensorboard_log=logDir)
model = MaskablePPO("CnnPolicy", env, verbose=1, tensorboard_log=logDir, device="auto")
evalCallback = EvalCallback(env, best_model_save_path=logDir, evalCallback = EvalCallback(env, best_model_save_path=logDir,
log_path=logDir, eval_freq=max(500, int(args.steps/30)), log_path=logDir, eval_freq=max(500, int(args.steps/30)),

Write Preview
Loading…
Cancel
Save