Your data is bigger than your spreadsheet, and your analysis needs to be repeatable.

Most researchers reach a point where the tool stops fitting the question - the dataset is too large, the test you need is not in the menu, or a reviewer asks you to re-run everything with one variable changed. Python solves all three, and this course teaches it from the beginning, with research data rather than toy examples.

You will install Miniconda, Python and Jupyter Notebook and write your first program, then cover data types, control flow, functions and modules. From there into the work itself: file handling, directories, and importing and cleaning datasets with Pandas.

The analysis

• Scientific computing with NumPy and SciPy

• Descriptive statistics, correlations, ANOVA and t-tests

• Zonal statistics on spatial data

• Three sections of plotting, from first graphs to specialised techniques

Then the part most Python courses for researchers leave out

• Setting up GPU acceleration for machine learning

• Detecting and counting plants with computer vision

• Building a convolutional neural network in PyTorch for image classification

• Training a machine learning model for crop health analysis

Three full case studies

• Leaf Area Index and Land Surface Temperature analysis

• India's air quality data analysed with machine learning, across four lectures

• Climate data analysis, across four lectures

Before you enrol

No programming experience is needed. The setup lectures are recorded on Windows, though the code runs on macOS and Linux as well, and every tool used is free. Fourteen quizzes run through the course so you can check your understanding as you go.

Taught by Dr. Azad Rasul, Assistant Professor of Remote Sensing, with a PhD in Geography, over 60 peer-reviewed publications, and more than 150,000 students enrolled across his Udemy courses.

Enrol now and start doing your analysis in code that you - and your reviewers - can run again.