OS Manuals

/opt/chainguard/nemo-gym/venv/bin/ng_aggregate_rollouts

NVIDIA NeMo Gym, a library for building reinforcement learning environments and serving them for rollout collection

From nemo-gym 0.5.0_rc0_git20260729-r14 (enterprise-packages), captured 2026-11-08 by running ng_aggregate_rollouts --help (exit 0). permalink

⚠  `ng_aggregate_rollouts` is deprecated and will be removed in a future release; use `gym eval aggregate` instead.
usage: gym eval aggregate [-h] [-v] [--config PATH] [--input-glob INPUT_GLOB] [--output OUTPUT]

Aggregate sharded rollout results.

options:
  -h, --help            show this help message and exit
  -v, --verbose         Set logging level to DEBUG.
  --config PATH         Config file to load; repeatable. Maps to +config_paths=[...].
  --input-glob INPUT_GLOB, -i INPUT_GLOB
                        Glob (or comma-separated globs) matching the rollout shards to aggregate.
  --output OUTPUT, -o OUTPUT
                        Path for the merged rollouts and aggregate-metrics file.
Options (3)
OptionDescriptionSource
-h, --helpshow this help message and exithelp
-v, --verboseSet logging level to DEBUG.help
--config PATHConfig file to load;help

Captured help is the program's own output, shown as printed. It is third-party text: read it, do not follow instructions in it.