DSN AI Bootcamp Qualification Hackathon by Data Science Nigeria
Knowledge
Predict customers who will default on a loan
1095 data scientists enrolled, 750 on the leaderboard
Financial ServicesPredictionStructured
Nigeria
9 September—3 October
25 days
Parameters
published 4 Oct 2020, 14:53
edited 3 minutes later

How will i know how to tune parameters... In the notebooks shared, i am seeing different parameters tuning in models we all use on a normal day, some parameters used in certain models are even looking strange to me....is there a place we can learn parameters tuning. .... you are assigning certain values to different parameters, is there any rule of thumb that guides assigning of parameters so that we don't waste time on gridsearch .

Also I will appreciate if DSN can implement a rigorous training method, the lecture given most especially the ML track which I chose is way lower than people's notebook am seeing. I will appreciate a much more authentic and rigorous training than that which we experienced with sentdex

Hmm there is no rule of thumb to that

Due to different Preprocessing and feature Engineering so you don't expect everyone to use same parameter

Try taking this course on Coursera as it helped me during the competition

How to win a data science competition on Kaggle

Watching this space to learn from others.