The tool includes over 20,000 reinforcement learning experiments.
This is Constellation, our lightweight aggregate experiment visualizer. Open source, hackable. Single-file C implementation. Ships with 20,000+ RL experiments worth of data from PufferLib 4.
"how many experiments have you run" yes
Users praise Constellation's single-file C visualizer for 20k+ RL experiments as an impressive feat of simple efficient shipping.
@jsuarez How do you recommend someone starting with PufferLib 4 with a new env creation and then the training? Good with python but not used C in the last 10 years. Any good tutorial (blog and/or video) to take me through the journey?
It's first 2 weeks of your first systems course level C. Check the website. Copy one of the tutorial envs as a template and implement something that is only a couple hundred lines. I had also done exclusively Python for almost 10 years when I started on Puffer's low level. Now I prefer C.
This is Constellation, our lightweight aggregate experiment visualizer. Open source, hackable. Single-file C implementation. Ships with 20,000+ RL experiments worth of data from PufferLib 4.
"how many experiments have you run" yes
@jsuarez Are you guys watching that? Automatic delivery
@jsuarez twenty thousand rl experiments and a single-file c visualizer. meanwhile most ai labs need a slack channel to draft a meeting. shipping is a skill most teams forgot they had.
@jsuarez the fact that its a single file C implementation is wild
@jsuarez Single-file C is a flex. 20k experiments is the part that makes me want to actually open it.
The tool includes over 20,000 reinforcement learning experiments.
This is Constellation, our lightweight aggregate experiment visualizer. Open source, hackable. Single-file C implementation. Ships with 20,000+ RL experiments worth of data from PufferLib 4.
"how many experiments have you run" yes