Fast Machine Learning Hyperparameters Model Optimization using Hyperopt
Details
Meeting Agenda:
6:00 – 6:40 pm: Pizza, water and networking.
6:40 – 6:45 pm: Welcome message by Ernest Bonat, Ph.D.
6:45 – 8:15 pm: Presentation and open discussions. “Fast Machine Learning Hyperparameters Model Optimization using Hyperopt” by Ernest Bonat, Ph.D., Senior Data Scientist.
Abstract:
Hyperopt (Distributed Asynchronous Hyperparameter Optimization in Python) - is a Python library for optimizing over awkward search spaces with real-valued, discrete, and conditional dimensions. Currently two algorithms are implemented in hyperopt: 1.Random Search and 2.Tree of Parzen Estimators (TPE). We’ll be covering the second one TPE in details. Hyperopt Installation and Regression/Classification real examples will be provided. After you understand and learn this tool, you’ll never use GridSearchCV/RandomizedSearchCV from scikit-learn framework anymore.