Data Management, Numerical Methods and Analysis

Learning Outcomes

Learning Outcomes

Having successfully completed this module you will be able to:

• Use machine learning and artificial intelligence approaches to analyse data.
• Choose appropriate data reduction processes to analyse large or complex data sets
• Understand the limits of data and the effects of sampling on the accuracy and reliability of outputs.
• Understand data standards
• Use best practise in the management of data including the use of repositories and the preparation of appropriate documentation

Arrays, loops, vector addition and dot products, matrix multiplication

Matrix diagonalization, Jacobi method

Discrete Fourier Transform

Discretisation. Numerical derivatives and error.

Numerical integration, trapezium rule, Simpson’s rule and higher order quadratures

Optimisation using gradients (steepest descents, conjugate gradients) and 2nd derivatives (Newton Raphson)

Matrix functions. Calculation of matrix exponentials, matrix inverse and other matrix functions.

Discretisation of Newton’s law of motion, molecular dynamics concepts, Verlet integrator algorithm for MD

Data standards and how to deal with sparse or missing data

Data handling and management

Approaches to data reduction

Error analysis and regression

Sampling and bootstrapping

A range of computational practicals involving setting up operating systems on virtual machines, version control systems, development of models based on your chemical knowledge and write code to test these against literature results, running large scale computers, schedulers and queueing

Teaching and Learning Methods will focus on directed computational practical work that lead you through the various aspects of scientific and data coding, mainly in Python. These will be supported by lectures and more open workshops where you will get to test your new skills on previously unseen problems.

Summative

This is how we’ll formally assess what you have learned in this module.