Managing machine learning models in production
There are many challenges to deploying machine models in production, including managing multiple versions of models, maintaining staging and production models, keeping track of model performance, logging, and scaling. Anand Chitipothu explores the tools, techniques, and system architecture of a cloud platform built to solve these challenges and the new opportunities it opens up.
| Talk Title | Managing machine learning models in production |
| Speakers | Anand Chitipothu (rorodata) |
| Conference | Strata + Hadoop World |
| Conf Tag | Make Data Work |
| Location | Singapore |
| Date | December 6-8, 2016 |
| URL | Talk Page |
| Slides | Talk Slides |
| Video | |
Typically, data scientists build machine learning models and ask IT specialists in their team to deploy these models. With teams becoming smaller and the quest for increased productivity, few data science teams have luxury of specialists at their beck and call. And even with dedicated IT teams, managing models in production is not a trivial task. As the number of models and team size increase, the complexity only grows. So how do you manage multiple versions of a model; version control the datasets used for model building; tag production and staging versions of a model; switch from one version to another seamlessly without any service disruption; or monitor performance of a live model? Anand Chitipothu explores the tools, techniques, and system architecture of a cloud platform built to solve these challenges and the new opportunities it opens up.
