/eval — Evaluate a Trained Checkpoint

SkillDev tools

Evaluate a trained checkpoint with visualization

Available today. Use it from your connected AI after setup.

No other account needed.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the /eval — Evaluate a Trained Checkpoint skill

What this skill tells your AI

The instructions your AI receives, as published by rohanpsingh/learninghumanoidwalking in .claude/skills/eval/SKILL.md and read by ahel’s review.

Parse the user's request from $ARGUMENTS and run evaluation.

uv run python run_experiment.py eval --path <PATH> [OPTIONS...]

The user may provide:

A .pt file: Use directly (--path /tmp/.../actor_999.pt)
A run directory: Contains actor*.pt files (--path /tmp/.../26-03-07-00-26-36_cartpole/)
A logdir: Contains timestamped run subdirectories (--logdir /tmp/training_runs)

If no path is given, check /tmp/training_runs for the most recent run.

Use Glob to verify the path exists and resolve it before running.

Flag	Default	Description
`--ep-len`	10	Episode length in seconds
`--seed`	None	Random seed for reproducible eval
`--out-dir`	None	Directory to save videos

Resolve the model path from the user's input. If ambiguous, list available checkpoints and ask.
Show the user which checkpoint will be evaluated (full path).
Run the eval command. This opens an interactive MuJoCo viewer window — it is NOT a background job.
Report the results when done.

Advanced