Add Assertion on the dataset while creating MatchingData

[abhishek-ch]

Add few data quality check inside MatchingData.
ex: If the population column will struggle with boolean value like 0 and 1, it must catch them early

[sprivite]

Can you please show me the code that is causing trouble?

[abhishek-ch]

This code raised issue for me

match = matcher.get_best_match()
m_data = m.copy().get_population(0)

Assuming I have 0 and 1 in the population column

[sprivite]

I cannot reproduce:

from pybalance.utils.balance_calculators import *
from pybalance.utils import MatchingData
from pybalance.sim import load_paper_dataset

m =load_paper_dataset()
data = m.data
data.loc[data.population == 'pool', 'population'] = 0
data.loc[data.population == 'target', 'population'] = 1
m = MatchingData(data)
m.copy().get_population(0)

[sprivite]

Can you please give the steps to reproduce?

[abhishek-ch]

large_confounding_adjustment_dataset.csv
Here is the sample dataset

[sprivite]

What matcher are you using?

[sprivite]

Can you please paste the full code along with the error?