Published inTowards Data ScienceIs your ML model stable? Checking model stability and population drift with PSI and CSIHow to make sure that your Machine Learning model is still performing as trainedMar 28, 20223Mar 28, 20223
Published inTowards Data ScienceTarget-encoding Categorical VariablesOne nice alternative to One-hot encoding your categoriesMar 17, 20223Mar 17, 20223
Published inTowards Data ScienceEvaluating the potential return of a model with Lift, Gain, and Decile AnalysisUse these three tools to understand the usefulness of your machine learning modelsMar 7, 2022Mar 7, 2022
Published inTowards Data ScienceEvaluating classification models with Kolmogorov-Smirnov (KS) testUsing the KS test to evaluate the separation between class distributionsFeb 28, 20224Feb 28, 20224
Published inTowards Data Science5 tips to start a career in dataThe steps that led me to become a Data Scientist in a multinational companyFeb 19, 2022Feb 19, 2022
Published inTowards Data ScienceMulticlass classification evaluation with ROC Curves and ROC AUCAdapting the most used classification evaluation metric to the multiclass classification problem with OvR and OvO strategiesFeb 12, 20224Feb 12, 20224
Published inTowards Data ScienceComparing sample distributions with the Kolmogorov-Smirnov (KS) testHow to compare samples and understand if they come from the same distribution using pythonFeb 7, 20222Feb 7, 20222
Published inTowards Data ScienceInterpreting ROC Curve and ROC AUC for Classification EvaluationHow I wish I was taught ROC Curve when I first learned itJan 31, 20222Jan 31, 20222
Published inTowards Data ScienceBoruta SHAP: an amazing tool for feature selection every data scientist should knowHow we can use Boruta and SHAP to build an amazing feature selection process — with python examplesJan 25, 20224Jan 25, 20224