| Battery performance datasets are typically non-normal and multicollinear. Extrapolating such datasets for model predictions needs attention to such characteristics. This study explores the impact of data normality in building machine learning models. In this work, tree-based regression models and multiple linear regressions models are each built from a non-normal dataset with multicollinearity and compared. Several techniques are necessary, such as data transformation, to achieve a good multiple linear regression model with this dataset; the most useful techniques are discussed. With these techniques, the best multiple linear regression model achieved anR2 = 81.23% and exhibited no multicollinearity effect for the dataset used in this study. Tree-basedmodels perform better on this dataset, as they are non-parametric, capable of handling complexrelationships among variables and not affected by multicollinearity. I show that bagging, in the use of Random Forests, reduces overfitting. Our best tree-based model achieved accuracy ofR2 = 97.73%. This study explains why tree-based regressions promise as a machine learning model for non-normally distributed, multicollinear data. This work applies machine learning tools toachieve the early life prediction of li-ion battery life. The prediction accuracy of different machine learning algorithms are compared in the battery database. Among various algorithms, the random forest (RF) method exhibits the highest accuracy of 97.73% to predict the battery cycle life using early cycle discharge capacity. The best model predicts battery cycle life with 4.05% test error when battery reaches 97% of nominal capacity and 9.69% test error when battery reaches 99% of nominal capacity.Keywords: Machine Learning, Random Forest, Battery Diagnostic, Lifetime prediction, Lithium-ion battery, Data driven method, Decision Tree, Data Transformation, Stepwise Regression |