ybgdgh / L3MVN

Leveraging Large Language Models for Visual Target Navigation
https://sites.google.com/view/l3mvn
62 stars 13 forks source link

Question about "-n" and "--num_eval_episodes" #21

Open Nieysh opened 3 months ago

Nieysh commented 3 months ago

Hi! Thank you for your great work! I have some questions about evaluating episodes:

Nieysh commented 3 months ago

@ybgdgh The L3MVN paper says using the standard HM3D splits of 20 validation scenes, when evaluating, I notice each thread will be assigned to several scenes from the val spilts, how does each thread handle the scene and episodes data? Thanks!