Technical coursework
Prepare a Jupyter notebook that a marker can follow
A notebook can display convincing output even when the saved cells no longer reproduce it in their visible order. Before submitting, treat the notebook as a document another person must read and run under the course's rules. Make the starting point, dependencies and conclusions clear without assuming the reader watched you develop the work.
Jupyter documents the notebook format and export options, and Microsoft's VS Code guidance covers notebook execution and export in that editor. Your department may provide a different interface or a hosted environment. Use its assessment instructions for the required files and execution conditions, then apply the review process below.
We may earn a commission from retailer links. This guide does not claim first-hand product testing. Our editorial approach
Read the submission contract
Identify whether the course wants the notebook file, a static export, source data, supporting scripts or a separate report. Check file names and any restrictions on external resources. Do not assume that a PDF alone replaces the executable notebook, or that an online link meets a requirement to upload a file.
Make a checklist from those instructions and keep it beside the working folder. If the task uses provided data, record its expected location and permitted redistribution. A marker should not need your personal home-directory path or a login to your private drive to understand how the analysis is meant to run.
Make the reading order coherent
Add concise explanatory text that introduces the question, describes the data and explains the main steps. Place each explanation near the relevant code and output. Avoid leaving an important decision only in a comment hidden inside a long cell. The narrative should show why the analysis takes a particular route.
Remove abandoned experiments from the submission copy where the course allows, or label them clearly if they are part of the required reasoning. Distinguish exploration from the final method. A sequence of unlabeled trials can obscure the work you intend to be assessed, even when each individual calculation is technically valid.
Recheck execution from the beginning
Save a working copy, then use the interface's supported restart and run-all process when appropriate for the assignment. Watch for errors and confirm that cells run in the intended order. This helps reveal hidden dependencies on values created earlier in the session but absent from the saved notebook.
Do not rerun an expensive or externally affecting operation without understanding it. If the notebook accesses a paid service, changes remote data or uses restricted research systems, follow the course's safe testing method. Record any legitimate execution limitation instead of fabricating a successful run or leaving stale output presented as newly verified.
Inspect the outputs as evidence
Check that table headings, chart labels, units and figure captions explain what is shown. Look for clipped output, unreadably long tables or errors hidden below a large plot. Explain the meaning of the result in the surrounding text, including important limitations. A chart is not self-explanatory simply because the software generated it.
Review the notebook for personal information, credentials and material that should not be submitted. Output can expose data even when the code itself looks harmless. Use permitted examples when seeking help, and follow the assessment's instructions for retaining or clearing output. Do not remove evidence that the course explicitly requires without resolving the conflict.
Check the exported deliverable
Create the required export using the supported tool and open it outside the editor. Inspect equations, figures, page breaks and the final conclusion. Export methods can have additional dependencies or format limitations, so test early enough to resolve a missing figure or failed conversion through the university's support route.
Keep the exact submitted files and receipt in the approved location. Include a short setup note if required, with the environment and input files needed for reproduction. A larger laptop is only relevant when a measured resource limitation prevents the required work; many notebook submission problems are solved by clear ordering, complete files and a careful export check.
Your next steps
- Confirm every required submission file.
- Explain the analysis beside the relevant code.
- Check execution order under the course’s conditions.
- Open and inspect the exported deliverable.
Sources and further help
Sources were consulted on 5 September 2026. Provider terms and services can change; check your exact booking, product or university service.
Keep planning
- Prepare Autodesk student access before design classes begin
- Check MATLAB access before choosing engineering course equipment
- Import coursework CSV data without changing identifiers
- Organise a coursework repository before the first commit