Python Programming for Data Science - Part 2



Overview

Data science is a discipline that uses scientific methods, processes and algorithms to extract meaningful information, knowledge and insights from structured and unstructured data.

The aim of this course is to provide insights on intermediate and advanced data science topics, using the Python programming language. The course will explore concepts such as machine learning, deep learning and natural language processing from a practical hands-down point of view. The focus will be on tools and methods rather than diving into the theoretical basis, in order to be appreciated by an audience with a minimal mathematical background.

Experience in using a programming or scripting language is a must. The student should master all the concepts explored in the course Python Programming for Data Science - Part 1.

In order to complete the assignment (and in order to get the full benefit from the course) students will need access to a computer capable of running the open-source software used in the course and access to the Internet. A limited amount of class time will be allocated to working on the class assignment, so students should ensure that they have access to a computer outside of class.

The course will rely on Jupyter Notebooks for interactive Python programming as they are widely used in Data Science.

 

Before attending this course, prospective students will know:

  • All the requirements and topics covered in the "Python Programming for Data Science - Part 1" course, i.e:
    • The fundamentals of linear algebra: what is a matrix and how matrix addition and multiplication are performed.
    • The following fundamental concepts of statistics: mean, median, variance and standard deviation, interquartile range.
    • The fundamentals of algebra: real and complex numbers, exponential and logarithm, and trigonometric functions.
    • How to perform fundamental Python operations such as variable creation, numerical operations on scalar, vectors and matrices, iteration through a collection, manipulation of elements in a collection.
    • how to use NumPy and pandas to import a dataset and extract important statistics from it using techniques such as split-apply-combine (for example, finding the mean, median or max of a quantitative variable for each category in a categorical variable).
    • Given a dataset, how to select the appropriate visualisation graph depending on the information to be conveyed, and use the matplotlib and seaborn library to draw it and add title, captions and figure legends.
    • How to create and add state and behaviour to a class in Python.
    • How to use nltk or spaCy to preprocess a text and convert it to a numerical representation that can be manipulated by information retrieval algorithms (e.g. for sentimental analysis, semantic search or machine learning algorithms).
    • What is, at least conceptually or visually, a derivative and a gradient.

Programme details

Course begins: 23 Apr 2024

Week 1: Introduction to the course. Basic overview of Machine Learning. Linear Regression example.

Week 2: Overview of a data-science preprocessing pipeline. Exploratory Data Analysis

Week 3: Data cleaning and preparation. Supervised Learning: regression.

Half Term - no class on 14th May. 

Week 4: Supervised Learning: classification. 

Week 5: Decision Trees. Ensemble Methods. Hyperparameter Tuning.

Week 6: Dimensionality Reduction and Unsupervised Learning.

Week 7: The Perceptron. Back-propagation. Fully-connected neural networks

Week 8: Deep Learning: the fundamentals.

Week 9: Convolutional Neural Networks (CNNs) for image processing. Recurrent Neural Networks (RNNs) for time series analysis.

Week 10: Attention-based models  (Transformers). Autoencoders. Introduction to ML Interpretability.

The following Python libraries will be used during the course:

  • scikit-learn (weeks 2-7)
  • PyTorch (weeks 7-10)
  • NumPy pandas, matplotlib, seaborn (throughout the course)
  • HuggingFace Transformers (week 10)

Digital Certification

To complete the course and receive a certificate, you will be required to attend and participate in at least 80% of the live sessions on the course and pass your final assignment. Upon successful completion, you will receive a link to download a University of Oxford digital certificate. Information on how to access this digital certificate will be emailed to you after the end of the course. The certificate will show your name, the course title and the dates of the course you attended. You will be able to download your certificate or share it on social media if you choose to do so.

Fees

Description Costs
Course Fee £339.00
Take this course for CATS points £10.00

Funding

If you are in receipt of a UK state benefit, you are a full-time student in the UK or a student on a low income, you may be eligible for a reduction of 50% of tuition fees. Please see the below link for full details:

Concessionary fees for short courses

Tutor

Dr Massi Izzo

Massi is a Departmental Tutor with five years experience teaching Python and Data Science courses for adult learners. He holds a doctorate in Bioengineering and currently works as a Software Engineer at MindFoundry, an Oxford University Company specialised in Artificial Intelligence solutions.

Course aims

1. Explore the landscape con contemporary machine learning (ML) and deep learning.

2. Learn how to use a variety of machine learning algorithms to extract features from the data using Python libraries.

3. Familiarise with the concepts of overfitting and regularisation in ML

4. Gain insights on how to face scaling issues in a "big data" scenario.

Teaching methods

Each week's session will consist of lectures and hands-on programming exercises, class discussions and interactive programming demonstrations by the lecturer.  

Learning outcomes

A the end of the course the students will be able to:

  • choose the right ML task and evaluation metric for a given ML problem and select a set of ML models to be trained;
  • set up a data pre-processing pipeline for data science and machine learning algorithms;
  • use Python machine learning tools (namely scikit-learn and PyTorch) to build up ML models, train and evaluate them on a test set;
  • evaluate whether a model overfits or underfits the data and act accordingly (e.g. opportunely regularise and overfitting model);
  • identify the appropriate and most performant model for a given task and tune appropriately the hyperparameters (parameters that cannot be learned by the model).

Assessment methods

Students will be asked to submit a portfolio of 3 exercises for their coursework assignment. I will give the first exercise on week 5 for early submission, the second and the third on week 8 and 9.

In order to complete the assignment (and in order to get the full benefit from the course) students will need access to a computer capable of running the open source software used in the course and access to the Internet. Only a limited amount of class time will be allocated to working on the assignment, so students should ensure that they have access to a computer outside of class.

Students must submit a completed Declaration of Authorship form at the end of term when submitting your final piece of work. CATS points cannot be awarded without the aforementioned form - Declaration of Authorship form

Application

To earn credit (CATS points) for your course you will need to register and pay an additional £10 fee per course. You can do this by ticking the relevant box at the bottom of the enrolment form or when enrolling online.

Please use the 'Book' or 'Apply' button on this page. Alternatively, please complete an enrolment form (Word) or enrolment form (Pdf).

Level and demands

Students who register for CATS points will receive a Record of CATS points on successful completion of their course assessment.

To earn credit (CATS points) you will need to register and pay an additional £10 fee per course. You can do this by ticking the relevant box at the bottom of the enrolment form or when enrolling online.

Coursework is an integral part of all weekly classes and everyone enrolled will be expected to do coursework in order to benefit fully from the course. Only those who have registered for credit will be awarded CATS points for completing work at the required standard.

Students who do not register for CATS points during the enrolment process can either register for CATS points prior to the start of their course or retrospectively from the January 1st after the current full academic year has been completed. If you are enrolled on the Certificate of Higher Education you need to indicate this on the enrolment form but there is no additional registration fee.

Most of the Department's weekly classes have 10 or 20 CATS points assigned to them. 10 CATS points at FHEQ Level 4 usually consist of ten 2-hour sessions. 20 CATS points at FHEQ Level 4 usually consist of twenty 2-hour sessions. It is expected that, for every 2 hours of tuition you are given, you will engage in eight hours of private study.

Credit Accumulation and Transfer Scheme (CATS)

Terms & conditions for applicants and students

Information on financial support