A simple bayes application
- Download:
git clone https://github.com/zpeterg/bmig_5003_basic_bayes
- Setup environment:
conda env create -f environment.yml
conda activate bayes
python cookie.py
- Assumptions:
- That we never know which bowl the cookie is taken from.
- That the best estimate when the cookie is removed is to subtract the proportional estimate from both bowls.
- I adapted the Likelihood method to work off of the total numbers of cookies and calculate the ratio from them (eg., vanilla-count / (vanilla-count + chocolate-count)).
- I added copied and adapted the Update() method (to UpdateAndRemove()) to make it subtract the cookie ratio from the bowl after it's normalized.