train-smoke

name	train-smoke
description	Run a minimal train/import smoke for ML crates (Rust or Python HF). Use before claiming a training path works.
metadata	{"author":"tz-forge","version":"1.0","domain":"ml"}
allowed-tools	Bash(cargo:*) Bash(uv:*) Bash(python:*) Bash(pytest:*)

Train smoke

When to use

• After scaffolding ml-rust-crate or ml-python-hf
• Before long GPU runs — validate imports and unit tests first
• CI local gate extension for ML kinds

Rust path

./scripts/check.sh --quick   # or cargo test --all-features
cargo test --all-features

Python HF path

uv sync --all-groups 2>/dev/null || uv sync
uv run python -m {{project_snake}}.train_smoke
uv run pytest -q

Rules

1. Prefer CPU / no-download smoke unless user explicitly requests a model pull.
2. Do not commit wandb/, checkpoints/, *.safetensors, large .bin/.pt.
3. Record command + pass/fail in the PR or kickoff note.

Output

Train smoke
===========
kind: ml-rust-crate | ml-python-hf
imports: ✅/❌
tests: ✅/❌
notes: ...