Fine-tune an oh-my-mlip model on your own dataset from a foundation checkpoint. Triggers on requests to fine-tune, continue training, or adapt a model to custom data — "fine-tune MACE on my data", "continue training SevenNet", "train DeePMD with my extxyz frames" — even when "oh-my-mlip" or "finetune" is not named.
Installs into .claude/skills of the current project.
Are you the author of Finetune?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/jinukmoon-finetune)
---
name: finetune
description: Fine-tune an oh-my-mlip model on your own dataset from a foundation checkpoint. Triggers on requests to fine-tune, continue training, or adapt a model to custom data — "fine-tune MACE on my data", "continue training SevenNet", "train DeePMD with my extxyz frames" — even when "oh-my-mlip" or "finetune" is not named.
argument-hint: "<model> --dataset PATH [--out DIR] [--epochs N] [--emit-only] [--slurm]"
---
The hub is `$OH_MY_MLIP_HOME` (default `~/.oh-my-mlip`); if neither exists,
clone it there first (`$OH_MY_MLIP_HOME/AGENTS.md §9.0`). Read the AGENTS.md
and run the scripts of that clone, never the copy inside the plugin directory.
Defer entirely to `AGENTS.md §3C` (fine-tune a model on your own dataset).
Read `AGENTS.md §3C` now. Do not reproduce its content here; follow it verbatim.
Key entry points: `scripts/ft_dataset.py` (canonical dataset converter),
`scripts/ft_run.py <model> --dataset <path>` (resolves `models.json`'s
`finetune` block and materializes the run), `scripts/ft_verify.py <ckpt>
--model <M> --json` (the checkpoint-loads-and-computes oracle).
If the model's env is not yet materialized, run `/oh-my-mlip:setup <model>` first.
Whole-job requests (interview, scoped plan, approval, execution, evidence)
follow the shared recipe `recipes/finetune.md` — read it after the section
above; it says which helpers exist today and which are still planned.
