PPO model evaluation with docker about habitat-challenge HOT 4 CLOSED

Looks like the resolution is different than expected. We ran last year's challenge with 256x256, but now it's 640x480 (99712 / 25088 == (640x480)/(256x256)).

from habitat-challenge.

As @erikwijmans mentioned make sure you used latest challenge config: challenge_pointnav2020.local.rgbd.yaml

from habitat-challenge.

@mathfac @erikwijmans thanks for answers!
As I understand I trained ppo model with last year's challenge configuration. So, now the question is how to train ppo model (from habitat-api/habitat_baselines) with gibson dataset for pointnav task with 2020 challenge configuration ?

Should I use configurations from challenge_pointnav2020.local.rgbd.yaml instead of pointav_rgbd.yaml?

from habitat-challenge.

@AdventureO, correct. Make sure you're using latest version.

from habitat-challenge.