kkdsdesilva / diabetes-ucidata

Analyzing clinical care data, this repo aims to predict hospital readmission 🏥 within 30 days using ML models: Logistic Regression, Decision Tree 🌳, Random Forest 🌲, XGBoost 🚀, and Neural Networks 🧠. It targets diabetic inpatient encounters, focusing on outcomes and treatments.
1 stars 0 forks source link

Sampling #14

Open kkdsdesilva opened 7 months ago

kkdsdesilva commented 7 months ago

If we label the target variable as 'No':0, '>30':0 and '<30':1, it becomes really imbalanced.

So we might have to sample the data so that there is no imbalance, to get better performance.