You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
So, I am now training things on tier 3, and although I think I probably won't get astonishing results before the deadline, I am quite happy to see my drones trying without using the imo-cheat noisy positional information :)
During my work with my approach on this project (end-to-end Deep Reinforcement Learning), I missed a few API functions. It would definitely be nice to have these in the future:
A way to directly specify the (reduced) size of the images from the drones without touching the json file (because I assume the competition setting will not let us modify this file ?)
A way to synchronously get the disqualification information
A way to synchronously get the gate-passing information (with gate id). Actually there is already a way to do this and I use it, but it is a bit geometrically intense.
A way to really control time (basically changing simContinueForTime() for something that is simulated-time based instead of cpu-time based should do the trick)
The text was updated successfully, but these errors were encountered:
Hello everyone,
So, I am now training things on tier 3, and although I think I probably won't get astonishing results before the deadline, I am quite happy to see my drones trying without using the imo-cheat noisy positional information :)
During my work with my approach on this project (end-to-end Deep Reinforcement Learning), I missed a few API functions. It would definitely be nice to have these in the future:
The text was updated successfully, but these errors were encountered: