This is a hobby project for creating classifiers for two problems

• Would a data scientist be eager to change jobs based on knowledge about their current situation (binary classifier)
• Salary prediction based on few variables (multiclass classifier for salary group prediction)

There are two versions of the code: ds_job_change.py and ds_job_change_modular.py. The first one is intended for a walk-through of what has been done and why, as the second one can be used to predict unseen rows from a file. This project also utilizes tools.py to report results and demonstrate examples to be used in explaining the project.

This was the first small project. The dataset was too small for meaningful feature selection, and as blood type was a better predictor for salary than age, it seemed that the dataset was fabricated -> not interesting.