<Discuss>How does BIC score calculate?

[xy-whu]

Subject of the issue

I just want to know how the BIC score really calculates. So i tried to use pgmpy to calculate the BIC score with an example dataset from a lecture. However, the results returned by pgmpy does not equal to the value provided in the lecture.

Your environment

• pgmpy 0.1.24
• Python 3.9.1
• window 10

Steps to reproduce

import pandas as pd
import numpy as np
from pgmpy.models import BayesianNetwork
from pgmpy.estimators import BicScore
data={"M":[0,1,0,1,1,1,0,1,1,0,1,1],"C":[1,0,1,0,0,0,1,0,0,0,0,1]} 
data = pd.DataFrame(data)
ret=BicScore(data).score(BayesianNetwork([['M','C']]))
print(ret)

Expected behaviour

According to the lecture, the BicScore value is -23.98
https://www.youtube.com/watch?v=QGRsmRgN0P8&t=236s

Actual behaviour

the BicScore value computed by pgmpy is -16.62903186274648