-
Notifications
You must be signed in to change notification settings - Fork 256
Env - Artillery3D #325
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: 3.0
Are you sure you want to change the base?
Env - Artillery3D #325
Conversation
EndDrawing(); | ||
} | ||
|
||
void init(Artillery3D* env) { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Put env logic above render logic
get_random_start(env); | ||
} | ||
|
||
void allocate(Artillery3D* env) { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Latest template envs etc have this in the .c, since that's the only place you need it. Can just inline obs reward etc. allocs there
int action = env->actions[0]; | ||
env->tick += 1; | ||
step_frame(env, action); | ||
compute_observations(env); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Pretty solid env. Like plat. Few small things could be simplified here and there. Overall gestalt is good. Didn't catch any major messups on RL side. Having targets be position dependent is going to make learning a bit harder vs. sorting them. Not huge. Only other thing is possible the env is ticking too fast. Like atari with no frameskip. Nothing wrong with that if you want the benchmark to be a bit artificially harder. But you can solve the same problem easier on lower tick rate
3D Version of Artillery meant to simulate real world scale anti drone defense.