MCPcopy Index your code
hub / github.com/rlcode/reinforcement-learning / parse_args

Function parse_args

3-atari/env.py:71–83  ·  view source on GitHub ↗
()

Source from the content-addressed store, hash-verified

69
70
71def parse_args():
72 p = argparse.ArgumentParser()
73 p.add_argument("--env", choices=list(ENV_IDS), default="breakout",
74 help="which Atari game to train on")
75 p.add_argument("--render", action="store_true",
76 help="open a window during training (much slower)")
77 p.add_argument("--test", action="store_true",
78 help="load the saved checkpoint and just play (no learning)")
79 p.add_argument("--device", choices=["auto", "cpu", "cuda", "mps"], default="auto",
80 help="override the auto-selected torch device")
81 p.add_argument("--wandb", action="store_true",
82 help="log metrics to Weights & Biases")
83 return p.parse_args()
84
85
86def make_env(args):

Callers 2

2-ppo.pyFile · 0.90
1-dqn.pyFile · 0.90

Calls

no outgoing calls

Tested by

no test coverage detected