notebooks/README.md

Training on Colab Pro+ (start a run in ~3 cells)

Fine-tunes the ScrubData planner (Qwen3-4B-Instruct-2507) with QLoRA, exports a GGUF, and pushes to the Hub. ~1–2h on an A100; Pro+ background execution finishes it unattended.

One-time prep (on this machine)

uv run training/build_dataset.py --n 2000 --out data/train.jsonl   # verified SFT data
huggingface-cli login                                              # HF token with WRITE
uv run scripts/push_dataset.py --repo <you>/scrubdata-sft          # data -> HF dataset

In Colab

1. New notebook → Runtime → Change runtime type → A100 GPU + High-RAM. (Fallback L4; T4 works too — the script auto-switches to 4-bit there.)
2. Add HF_TOKEN in Colab Secrets (🔑), value = your HF write token.
3. Run these cells:

Cell 1 — deps + repo

!pip install -q unsloth
!git clone https://github.com/<you>/hackaton-small.git || true   # or upload notebooks/train_qlora.py

Cell 2 — token

import os
from google.colab import userdata
os.environ["HF_TOKEN"] = userdata.get("HF_TOKEN")

Cell 3 — train + export + push

!python hackaton-small/notebooks/train_qlora.py \
    --data-repo <you>/scrubdata-sft \
    --out-repo  <you>/scrubdata-qwen3-4b \
    --epochs 2

Artifacts land at https://huggingface.co/<you>/scrubdata-qwen3-4b (adapter) and ...-gguf (llama.cpp). The Colab VM is ephemeral — the script pushes to the Hub so a finished run isn't lost.

After training — measure it

Pull the GGUF locally and plug it into the eval harness as a planner (same shape as scrubdata/model_planner.py), then:

uv run eval/run_model.py --model <local-gguf-or-ollama-id>   # vs heuristic/oracle
uv run eval/run_real.py                                       # real OOD slice

Compare against the goalposts in eval/README.md (recovery ≥0.95, canon_f1 ≥0.85).

Can Claude start this for me?

Not headlessly — Google removed external Colab runtime control and tunneling violates ToS. Options: (a) run the 3 cells yourself (Pro+ background execution carries it), or (b) connect colab-mcp with a Colab tab open and Claude Code can drive the cells live in your browser (co-pilot, not unattended). HF Jobs is the headless alternative if you ever want a no-browser run.