|
Download README.md from tsilva/gradlab-breakout-6127e81d: direct link, hf CLI and curl.
- Browser
- Download file 804 Bytes
-
https://huggingface.co/datasets/tsilva/gradlab-breakout-6127e81d/resolve/3ebbd7b97c5b66cc20ef5fe396814d94e05cd4cf/README.md
- Command line
-
hf download hf://datasets/tsilva/gradlab-breakout-6127e81d@3ebbd7b97c5b66cc20ef5fe396814d94e05cd4cf/README.md
-
curl -L -o README.md https://huggingface.co/datasets/tsilva/gradlab-breakout-6127e81d/resolve/3ebbd7b97c5b66cc20ef5fe396814d94e05cd4cf/README.md
804 Bytes
metadata
configs:
- config_name: episodes
data_files: episodes/*.parquet
Checkpoint monitoring trajectories
Each complete episode row links to immutable transition tables and lossless PNG chunks. The episode index retains Run, training seed, Checkpoint, evaluation, episode conditions, actions, rewards, boundaries and source/runtime provenance. Frames are full unmasked native RGB at the contracted action cadence. Chunk joins preserve the initial and true terminal image without duplicated transitions. Staged chunks without a complete episode index are not dataset episodes.
Monitoring is observational and never Acceptance evidence. Use whole-Run or training-seed-group holdouts to avoid leakage across correlated Checkpoints; split assignment is left to consumers. R2 sources remain canonical.