ML Pipelines and Hyperparameter Tuning

While EDA can be the stepping stone for any Machine learning and analytic activity, what sets apart an optimal model from other sub-optimal ones is the ability itself to swiftly explore, train and contrast multiple models of different class and complexity (regularization). This beginner to intermediate level talk will focus on the regularization, hyper-parameter tuning and how Scikit-Learn pipelines and grid search can be leveraged to effortlessly perform model selection and tuning.

About our speaker:

Atul Saurav is Lead Data Engineer with Genworth Financials. Prior to Genworth, Atul was a consultant on State Farm's Data Warehouse and Big Data Analytics projects. He has also consulted at AT&T and Verizon telecom in past. Atul is pursing his Masters in Decision Analytics at the Virginia Commonwealth University and loves learning new concepts and playing with data in his spare time.