L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Développement · SKILL.md

training-and-experiments

training-and-experiments

Run and debug H2O LLM Studio training experiments, distributed launchers, backend choices, and experiment artifacts.

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentation de la Skill

GitHub

name: training-and-experiments description: "Run and debug H2O LLM Studio training experiments, distributed launchers, backend choices, and experiment artifacts." disable-model-invocation: true metadata: disco-role: operating license: Apache 2.0

Training And Experiments

Use this sub-skill when the task is to launch, monitor, or debug an H2O LLM Studio experiment after a dataset/config has already been chosen. It owns direct CLI training, GUI-launched training behavior, multi-GPU launch mechanics, backend/performance settings, and the files written to an experiment output directory.

Read These First

  • Training CLI: direct llm_studio/train.py -Y runs, deprecated -C, dynamic --section.field overrides, GUI launch behavior, and multi-GPU commands.
  • Experiment artifacts: output directory contents, status flags, logs, charts, checkpoints, predictions, and how to decide whether a run finished.
  • Backend and performance: CUDA, DeepSpeed, DDP, LoRA/DoRA/RSLoRA, mixed precision, optimizer/scheduler, and memory/runtime trade-offs.
  • Troubleshooting: symptoms, likely causes, recovery checks, and routes to sibling sub-skills for data/model/export issues.

Bundled Helpers

Route Away

  • Dataset columns, schema, YAML construction, config class details, and CSV/Parquet validation belong to ../configuration-and-data/SKILL.md.
  • Model wrapper internals, metrics, losses, prediction semantics, and evaluation internals belong to ../modeling-and-evaluation/SKILL.md.
  • Interactive prompting, model-card/export preflights, and Hugging Face publishing belong to ../export-and-prompt/SKILL.md.
  • App startup, work directories, H2O Wave settings, and server lifecycle belong to ../app-and-ui/SKILL.md.
Skills enregistrées