SUSS

The python package that powers interactive study labguide. Work closely with kaggle's learntools.

Get started

NOTE: github package don't work with python packages

git clone https://github.com/suss-vli/suss.git
cd suss
python3 -m pip install --upgrade build
python3 -m build

if you are developing labguide

you can just run the following command to upgrade this package

pip3 install ../suss --upgrade

Installing

pip3 install <path-to-suss>

Use pygount

python3 -m venv venv; source venv/bin/activate
pygount --format=summary --out=pygount_17july.txt . 

Usage

from suss.ict133.lab1 import *

Notes

Constrains

x.get_source_code / x.create_object_from_source_code - When we call either of the two functions, it will generate the an error ____ is not callable when any .check() is uncommented in the notebook.

• The solution is comment out all the .check() code.

✅ Everyone should read this first! - Updates to suss check

For ICT133 and ICT162, you will see a difference where we use the FunctionProblem and directly access the student's written function for the answer. and the format for our code is slightly different where instead of writing our customised check in check, it is under check_testbook: https://github.com/suss-vli/suss/blob/main/src/suss/ict133/lab1_questions/q1.py#L28

From ICT233 onwards, we begin using get_source_code .

and we write our checks within the check function. get_source_code allows us to get the exact codes written by students to check. With every get_source_code, a test_for_none_xxx must accompany it. This is to navigate the issue of difference in the cell number to get_source_code from due to missing cells, additional cells.

There's also a more structured format from ict233 onwards where there is:

produce_expected() _var _expected _test_cases check()

The standardised format is meant to help us to eventually use AI to generate new labs or checks for us and the instructors.