Everything, as files

Downloads

You do not need any of this to follow the course: the lectures run in your browser and the notebooks open in Colab with the data loading by itself. This page is for working offline, on your own machine, which is what you will do at the exam.

All the notebooks and data, in one file

The ZIP includes a local readiness notebook. See the setup guide for instructions and troubleshooting.

Download just the readiness notebook and put it beside data/ in your unzipped course folder.

Every exercise notebook, every case notebook and every data file, arranged so they work as soon as you unzip them, with the list of package versions the course was checked with. Always the current version.

Unzip it, then in VS Code use File → Open Folder and pick the unzipped folder. Opening the folder rather than a single file is what makes data/prices.csv resolve. See the setup guide.

Where each data file comes from, and the terms it comes with, is in the table on the Resources page.

Session 1 · Beginning Python for financial data

FileWhat it is
session_01_exercises.ipynb55 exercises, hints and solutions folded in download
session_01_case.ipynbThe case, Part 1 download
aapl_2024_closes.csvApple's 2024 closing prices download
ko_2024_closes.csvCoca-Cola's 2024 closing prices download
The lectureRuns in your browser, nothing to install open
The lecture, as a PDFEvery slide, with the result of every code cell; for revising offline download

Session 2 · Functions, loops and dictionaries

FileWhat it is
session_02_exercises.ipynb64 exercises download
session_02_case.ipynbThe case, Part 2 download
Five single-stock 2024 files: Apple, Coca-Cola, Nvidia, Johnson & Johnson, JPMorgan AAPL · KO · NVDA · JNJ · JPM
The lectureRuns in your browser, nothing to install open
The lecture, as a PDFEvery slide, with the result of every code cell; for revising offline download

Session 3 · Packages: NumPy, pandas and matplotlib

FileWhat it is
session_03_exercises.ipynb76 exercises download
session_03_case.ipynbThe case, Part 3 download
prices.csvAll eleven instruments, 2015 to 2024, 27,676 rows download
aapl_2024_closes.csvUsed by the array exercises download
The lectureRuns in your browser, nothing to install open
The lecture, as a PDFEvery slide, with the result of every code cell; for revising offline download

Session 4 · Foundations of machine learning

FileWhat it is
session_04_exercises.ipynb73 exercises download
session_04_case.ipynbThe case, Part 4 download
prices.csvAll eleven instruments, 2015 to 2024, 27,676 rows download
The lectureRuns in your browser, nothing to install open
The lecture, as a PDFEvery slide, with the result of every code cell; for revising offline download

Session 5 · Model selection and cross-validation

FileWhat it is
session_05_exercises.ipynb63 exercises download
session_05_case.ipynbThe case, Part 5 download
prices.csvAll eleven instruments, 2015 to 2024, 27,676 rows download
The lectureRuns in your browser, nothing to install open
The lecture, as a PDFEvery slide, with the result of every code cell; for revising offline download

Session 6 · Penalised regression

FileWhat it is
session_06_exercises.ipynb58 exercises download
session_06_case.ipynbThe case, Part 6 download
prices.csvAll eleven instruments, 2015 to 2024, 27,676 rows download
market_features.csvThe lecture's nineteen-column table for the index, 2,376 rows download
The lectureRuns in your browser, nothing to install open
The lecture, as a PDFEvery slide, with the result of every code cell; for revising offline download

Session 7 · Pricing a house in Aarhus

FileWhat it is
aarhus_houses.csv7,621 ordinary house sales in the Aarhus postcodes, January 2021 to September 2024, thirteen columns; no addresses, locations to about 100 metres download
The lectureRuns in your browser, nothing to install; no exercises or case this session open
The lecture, as a PDFEvery slide, with the result of every code cell; for revising offline download
The mapEvery sale in the table on a live map open

Session 8 · Classification with logistic regression

FileWhat it is
session_08_exercises.ipynb64 exercises download
session_08_case.ipynbThe case, Part 8 download
prices.csvAll eleven instruments, 2015 to 2024, 27,676 rows download
market_features.csvThe 19-column table for the index, 2,376 rows; the deck and the exercises make their label from it in one line download
The lectureRuns in your browser, nothing to install open
The lecture, as a PDFEvery slide, with the result of every code cell; for revising offline download

Session 9 · Classification in practice

FileWhat it is
session_09_exercises.ipynb68 exercises download
session_09_case.ipynbThe case, Part 9 download
credit.csv30,000 credit card holders at a Taiwanese bank in 2005, seven columns and whether each defaulted the month after; from the UCI Machine Learning Repository, CC BY 4.0 download
prices.csvAll eleven instruments, 2015 to 2024, 27,676 rows download
market_features.csvThe 19-column table for the index, 2,376 rows download
The lectureRuns in your browser, nothing to install open
The lecture, as a PDFEvery slide, with the result of every code cell; for revising offline download

Session 10 · Electricity prices in West Denmark

FileWhat it is
power.csv26,133 hours of day-ahead prices for West Denmark (DK1), 2022 to 2024, with the day-ahead wind and sun forecasts; open data from Energinet's Energi Data Service download
The lectureRuns in your browser, nothing to install; no exercises or case this session open
The lecture, as a PDFEvery slide, with the result of every code cell; for revising offline download

Session 11 · Trees and forests

FileWhat it is
session_11_exercises.ipynb62 exercises download
session_11_case.ipynbThe case, Part 11 download
market_features.csvThe 19-column table for the index, 2,376 rows download
credit.csv30,000 credit card holders at a Taiwanese bank in 2005, seven columns and whether each defaulted the month after; from the UCI Machine Learning Repository, CC BY 4.0 download
prices.csvAll eleven instruments, 2015 to 2024, 27,676 rows download
The lectureRuns in your browser, nothing to install open
The lecture, as a PDFEvery slide, with the result of every code cell; for revising offline download

Session 12 · Boosting and reading a model

FileWhat it is
session_12_exercises.ipynb59 exercises download
session_12_case.ipynbThe case, Part 12, the last part download
market_features.csvThe 19-column table for the index, 2,376 rows download
credit.csv30,000 credit card holders at a Taiwanese bank in 2005, seven columns and whether each defaulted the month after; from the UCI Machine Learning Repository, CC BY 4.0 download
prices.csvAll eleven instruments, 2015 to 2024, 27,676 rows download
The lectureRuns in your browser, nothing to install open
The lecture, as a PDFEvery slide, with the result of every code cell; for revising offline download

Session 13 · The two cases with trees

FileWhat it is
aarhus_houses.csvThe house sales of Session 7: 7,621 ordinary house sales in the Aarhus postcodes, January 2021 to September 2024 download
power.csvThe hours of Session 10: 26,133 hours of day-ahead prices for West Denmark (DK1), 2022 to 2024, with the wind and sun forecasts download
The lectureRuns in your browser, nothing to install; no exercises or case this session open
The lecture, as a PDFEvery slide, with the result of every code cell; for revising offline download

Running the notebooks locally

  1. Follow the setup guide once: Python 3.14, VS Code, the Python extension, and the course's packages, installed from a notebook cell with %pip install -r https://theill95.github.io/mlfin-2026/requirements.txt.
  2. Put the notebook and a data/ folder in the same place, exactly as in the ZIP.
  3. In VS Code, File → Open Folder and choose that folder. Not the file: the folder.
  4. Open the notebook, pick your Python when asked, and run the first cell.
If you skip the data files entirely, the notebooks still work: when they cannot find a local data/ folder they download what they need from this repository. You only need the CSVs if you want to work with no internet at all.

A note on the lectures

The interactive lectures are not in the ZIP. Their code cells run Python inside the browser, and that only works when the page is served over the web rather than opened from a file on your disk. It is a browser restriction, not a course one. So the slides are best used online, from the sessions page, where they always work and are always current.

For offline reading, every lecture is also a PDF with the result of every code cell: each session's table above has it, and the Exam info page has all thirteen in one place. Everything you actually write code in, the exercises and the case, is in the ZIP and works completely offline.