Introduction to Data Analytics and Machine Learning with Python

Ideal for established Python programmers, this online course is an introduction to machine learning and data analytics leveraging the most widely used Python libraries, developed and maintained by big companies like Google, Facebook and Twitter.

As both data analytics and machine learning fields are vast and fast expanding, we will focus our efforts on grasping the foundations. This could enable you to get a junior position as a data analyst and/or machine learning engineer.

Libraries that will be taught in this course:

• Jupyter Notebook
• NumPy
• SciPy
• matplotlib
• pandas
• Scikit-learn

Put your foundation in Python programming to work making sense of big data.

The course will  be taught using Python 3, the most recent iteration of the Python language.

Who is it for?

Understand the key principles of analysis and machine learning, perfect for experienced Python programmers. This course is designed to open the vast world of data analytics and machine learning to those without prior experience of that field, using Python. Delegates must have Python already. Knowledge of mathematical concepts such as described at http://www.gcseguide.co.uk/mathsgcseguide.htm is

Timetable

The Introduction to Data Analytics and Machine Learning with Python short course is offered in various formats to fit your availability:

1. Weekly evening classes – taught one evening a week for 10 consecutive weeks, allowing you to continue in full-time employment
2. Saturday classes – taught over five consecutive Saturdays, designed for weekend learners
3. Summer School – taught intensively over one week in the summer, ideal for learning quickly or for staff training

Benefits

• Delivered by industry professionals
• Taught in small group sizes
• Awarded a City, University of London certificate on completion

• Jupyter notebook: a quick tour of the data engineers' IDE of choice.
• Introduction to numpy: N-dimensional arrays, broadcasting functions, linear algebra abstractions and random number generators.
• Exploratory data analysis with pandas: manipulating data: loading, storing, cleaning, transforming, merging, reshaping.
• Visualising and plotting with matplotlib: generate plots, histograms, power spectra, bar charts, errorcharts, scatterplots. Visualize and understand different types of data.
• Introduction to scipy with statistics, is mainly focused at providing a quick introduction to the scipy.stats package. We will be looking at distributions, fitting distributions and random numbers.
• Introduction to machine learning concepts with scikit-learn, training and evaluating learning algorithms. We will be looking at: decision trees, perceptrons, support vector machines, and neural networks.
• Scikit-learn delving deeper: using data validation and cross-validation. Also some other methods to improve the accuracy of your learning algorithms

Information about the libraries taught in this course

• Jupyter Notebook: a web application that allows you to create and share documents that contain live code, equations, visualizations and explanatory text. Uses include: data cleaning and transformation, numerical simulation, statistical modeling, machine learning and much more
• NumPy: the fundamental package for scientific computing with Python, which contains useful things like: a powerful N-dimensional array object; sophisticated (broadcasting) functions; useful linear algebra, Fourier transform, and random number capabilities. We will also be using it as an efficient multi-dimensional container of generic data.
• SciPy: provides many user-friendly and efficient numerical routines such as routines for numerical integration and optimization. This library builds on top of NumPy and makes heavy use of all the features that we will be learning in NumPy.
• matplotlib: a plotting library which produces publication quality figures and can also be used to do image manipulation. You can generate plots, histograms, power spectra, bar charts, error charts, scatter plots, with just a few lines of code.
• pandas: is an easy to use data structuring and data analysis library which we will be using. It has advanced data manipulation capabilities and can use data objects in the same way we use databases. It can also import and export data from a vast number of formats.
• Scikit-learn: built on top of NumPy, SciPy, and matplotlib this is one of the most widely used machine learning libraries in industry and research. It covers a truly impressive number of machine learning techniques and methods, some of which include: classification, regression, clustering, dimensionality reduction, model selection, data pre-processing, etc.

You will be awarded an official City, University of London certificate if you attend at least 70 per cent of the classes. The course is not formally accredited.

Assessment

Informal assessment through optional weekly assignments, which will build into a final project that will solve a real world problem using real world data, applying state of the art techniques taught during the course.

As this is an introductory data analytics course you are not expected to have any data analytics or machine learning experience, but please note the following:

- Applicants must have successfully completed the Introduction to Programming with Python or have Python to a similar standard, as a baseline.

(NB Candidates with strong programming skills in other languages - C++, Java etc. - may be able to convert as they go but should consult the syllabus noted above for guidance.)

- Knowledge of mathematical concepts such as those presented here is essential.

English requirements

Applicants must have fluent written and spoken English.