dhruvbatra

Cool to see that DDPPO training is getting in. Can we see a reward/success vs steps curve?

Interesting. Agreed that the untuned-reward policy indeed looks pretty bad. If you haven't already, consider chatting with @naokiyokoyama about his "gaze" policy.

Great! Glad to see this working. Agreed with the todos and look forward to them. Question -- what is the sensor suite for the spot-pick experiments? Just proprioception or proprioception+arm...

That's odd. If you give us a minimal reproducible example script (with details of the scene), we can look at reproducing it.

Can we get a github repository where we can actually look at code?

CC: @mosra

For MP3D, you can find some trajectories here: https://github.com/vincentcartillier/Semantic-MapNet

Thanks for finding this. Would you mind sending a PR to fix it? And @mpiseno -- perhaps you could review the PR?