Daily Notebooks

One notebook per day — tutorial + Corpus Lab in a single file

Each day has one Colab notebook that you work through top to bottom and submit at the end of the day. Every notebook has two parts:

Part A · Tutorial — the guided, run-along section you do together in class.
Part B · Corpus Lab — the independent hands-on practice for that day.

You open each notebook directly in Colab (Tohoku Google account, no setup), Run all, then File → Download → Download .ipynb and submit that one file.

The pipeline is the same all week

load gold → format prompt → call model → evaluate → inspect errors. Only the task and the prompt change. The datasets behind these notebooks are catalogued under Datasets.

Which backend each day uses

Day 1 — Colab’s built-in Gemini (colab.ai), keyless. Your first live call; notice it varies.
Day 2 — no model call. You evaluate frozen predictions so the metrics numbers hold still.
Day 3 onward — you run the model yourself via the Gemini API (temperature=0 + a fixed seed, so results are reproducible for the autograded labs). One-time setup: get a free key.

The notebooks

Day	Notebook	Tutorial (Part A)	Corpus Lab (Part B)
1	`day1_python_and_first_llm.ipynb`	Python basics + your first LLM call	Python practice exercises (self-checked)
2	`day2_gold_standards_and_evaluation.ipynb`	The pipeline on CEFR-SP (P/R/F1 + confusion matrix)	Code the metrics from scratch, checked vs. scikit-learn
3	`day3_prompt_design.ipynb`	Zero-shot → few-shot → chain-of-thought on CEFR	Your own prompt-iteration study (coming soon)
4	`day4_pipeline_and_sampling.ipynb`	Sample a balanced gold subset from a pool	QC & adjudicate your gold set (coming soon)
5	`day5_project_finalization.ipynb`	Assemble the pipeline end-to-end (coming soon)	Draft the one-page report (coming soon)

Deeper notes

The prose walkthroughs below expand on the in-notebook narration:

Day 2 — Annotation, gold standards & evaluation
Day 3 — Replicating Kim & Lu with open data — the discourse-move mini-project track.

When you are ready to run your own study, move on to the mini-project starter tracks.

--- title: "Daily Notebooks" subtitle: "One notebook per day — tutorial + Corpus Lab in a single file" toc: true --- Each day has **one Colab notebook** that you work through top to bottom and submit at the end of the day. Every notebook has two parts: - **Part A · Tutorial** — the guided, run-along section you do together in class. - **Part B · Corpus Lab** — the independent hands-on practice for that day. You open each notebook directly in Colab (Tohoku Google account, no setup), **Run all**, then **File → Download → Download `.ipynb`** and submit that one file. ::: {.callout-note} ## The pipeline is the same all week **load gold → format prompt → call model → evaluate → inspect errors.** Only the *task* and the *prompt* change. The datasets behind these notebooks are catalogued under [Datasets](../datasets/index.md). ::: ::: {.callout-tip} ## Which backend each day uses - **Day 1** — Colab's built-in Gemini (`colab.ai`), keyless. Your first live call; notice it varies. - **Day 2** — *no model call.* You evaluate **frozen** predictions so the metrics numbers hold still. - **Day 3 onward** — you run the model yourself via the **Gemini API** (`temperature=0` + a fixed seed, so results are reproducible for the autograded labs). One-time setup: [get a free key](../tools/gemini-api-key.md). ::: ## The notebooks | Day | Notebook | Tutorial (Part A) | Corpus Lab (Part B) | |---|---|---|---| | 1 | [`day1_python_and_first_llm.ipynb`](./day1_python_and_first_llm.ipynb) | Python basics + your first LLM call | Python practice exercises (self-checked) | | 2 | [`day2_gold_standards_and_evaluation.ipynb`](./day2_gold_standards_and_evaluation.ipynb) | The pipeline on CEFR-SP (P/R/F1 + confusion matrix) | Code the metrics from scratch, checked vs. scikit-learn | | 3 | [`day3_prompt_design.ipynb`](./day3_prompt_design.ipynb) | Zero-shot → few-shot → chain-of-thought on CEFR | Your own prompt-iteration study *(coming soon)* | | 4 | [`day4_pipeline_and_sampling.ipynb`](./day4_pipeline_and_sampling.ipynb) | Sample a balanced gold subset from a pool | QC & adjudicate your gold set *(coming soon)* | | 5 | [`day5_project_finalization.ipynb`](./day5_project_finalization.ipynb) | Assemble the pipeline end-to-end *(coming soon)* | Draft the one-page report *(coming soon)* | ## Deeper notes The prose walkthroughs below expand on the in-notebook narration: - [Day 2 — Annotation, gold standards & evaluation](./tutorial-day2-annotation-eval.md) - [Day 3 — Replicating Kim & Lu with open data](./tutorial-day3-move-replication.md) — the discourse-move mini-project track. When you are ready to run your own study, move on to the [mini-project starter tracks](../datasets/mini-project-tracks.md).