Run any Skill in Manus with one click

run-experiment

Stars2

Forks0

UpdatedJune 10, 2026 at 00:37

Plans and launches an hmfast experiment (training, inference, or scan) on the cluster with safe resource limits. Use when the user wants to run something that costs nontrivial compute, especially on GPU.

Installation

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

Run Skill in Manus

Source

licongxu

licongxu/JXPaint

View GitHub Repository View Creator Repositories

Download

Run Skill in Manus

Related occupationsSOC

Based on SOC occupation classification

Network and Computer Systems AdministratorsComputer and Mathematical Occupations·SOC 15-1244

SKILL.md

readonly

name	run-experiment
description	Plans and launches an hmfast experiment (training, inference, or scan) on the cluster with safe resource limits. Use when the user wants to run something that costs nontrivial compute, especially on GPU.

Run experiment

Follow .claude/rules/cluster-usage.md and, for autonomous sessions, .claude/rules/autonomous-operation.md.

Pre-flight

Identify the device. GPU partition? Which one? Quick nvidia-smi check on the target node.
Estimate memory. Sum array sizes in floats; multiply by 8 bytes (x64 default). Confirm it fits with headroom for JAX's default 75% preallocation. Show the arithmetic.
Estimate runtime. Run a tiny version (small N, few steps) and extrapolate. If the extrapolation is unclear, ask the user before launching the full run.
Pick a seed and config. Save both into the output directory.

Launch

Always via SLURM if estimated runtime > 10 min.
Job script must set --time, --mem, --gres=gpu:1 (when GPU is needed), and --output=logs/%x-%j.out.

Activate the env inside the job script:

source /scratch/scratch-lxu/venv/cmbagent_env/bin/activate

Outputs go under /scratch/scratch-lxu/... (working) or /rds/rds-lxu/... (persistent), never inside the repo.

During and after

Stream the log: tail -f logs/<jobname>-<jobid>.out.
On finish, verify saved arrays are finite and have the expected shapes.
Write a run.json manifest with: git hash, full config, seed, MaxRSS, wall time, output paths.

Stop conditions

If during the run you see OOM, repeated NaNs, or runtime exceeding the estimate by more than 2x, stop and report before resubmitting. Do not blindly retry with the same parameters.

More from this repository

same repository

code-review

licongxu/JXPaint

Reviews proposed changes to hmfast for correctness, JAX compatibility, API layering, test coverage, and hardware-cost surprises. Use when the user asks for a review, a sanity check on a diff, or before merging a feature branch.

2026-06-102

cosmology-emulators

licongxu/JXPaint

Guides changes to hmfast Cosmology, emulator loading, and cosmological parameter sets. Use when editing cosmology.py, emulator_load.py, download.py, or adding new emulator_set keys.

2026-06-102

debug-hpc-job

licongxu/JXPaint

Diagnoses a failed or stuck SLURM/HPC job for hmfast (OOM, timeout, NaNs, CUDA errors, missing outputs). Use when the user asks "why did my job fail" or shows a SLURM log.

2026-06-102

halo-model-workflow

licongxu/JXPaint

Guides edits to hmfast's differentiable JAX halo model (HaloModel, mass function, bias, concentration, profiles). Use when changing src/hmfast/halos/, power-spectrum methods, or halo-model consistency flags.

2026-06-102

tracers-and-profiles

licongxu/JXPaint

Guides adding or modifying hmfast tracers and halo profiles. Use when editing src/hmfast/tracers/ or src/hmfast/halos/profiles/, or integrating new observables with HaloModel.

2026-06-102

write-paper-section

licongxu/JXPaint

Drafts or revises a section of an astrophysics or cosmology manuscript based on hmfast results. Use when the user asks for a draft, an edit pass, or a publishability review of a section, including its figures.

2026-06-102

name	run-experiment
description	Plans and launches an hmfast experiment (training, inference, or scan) on the cluster with safe resource limits. Use when the user wants to run something that costs nontrivial compute, especially on GPU.

Run experiment

Follow .claude/rules/cluster-usage.md and, for autonomous sessions, .claude/rules/autonomous-operation.md.

Pre-flight

Identify the device. GPU partition? Which one? Quick nvidia-smi check on the target node.
Estimate memory. Sum array sizes in floats; multiply by 8 bytes (x64 default). Confirm it fits with headroom for JAX's default 75% preallocation. Show the arithmetic.
Estimate runtime. Run a tiny version (small N, few steps) and extrapolate. If the extrapolation is unclear, ask the user before launching the full run.
Pick a seed and config. Save both into the output directory.

Launch

Always via SLURM if estimated runtime > 10 min.
Job script must set --time, --mem, --gres=gpu:1 (when GPU is needed), and --output=logs/%x-%j.out.

Activate the env inside the job script:

source /scratch/scratch-lxu/venv/cmbagent_env/bin/activate

Outputs go under /scratch/scratch-lxu/... (working) or /rds/rds-lxu/... (persistent), never inside the repo.

During and after

Stream the log: tail -f logs/<jobname>-<jobid>.out.
On finish, verify saved arrays are finite and have the expected shapes.
Write a run.json manifest with: git hash, full config, seed, MaxRSS, wall time, output paths.

Stop conditions

If during the run you see OOM, repeated NaNs, or runtime exceeding the estimate by more than 2x, stop and report before resubmitting. Do not blindly retry with the same parameters.