Saturday 4:00 PM–4:45 PM in Room 1
Deploying Machine Learning using sklearn pipelines
Kevin Goetsch
- Audience level:
- Intermediate
Description
Sklearn pipeline objects provide an framework that simplifies the lifecycle of data science models. This talk will cover the how and why of encoding feature engineering, estimators, and model ensembles in a single deployable object.
Abstract
Sklearn pipelines have advantages during both the development and deployment lifecycle of data science models. This talk will focus on what Sklearn pipelines are, how you benefit from using them in your workflow, and an example of a technical implementation (with some practical tips).
Some the advantages include:
- Reusable feature engineering
- Maintable and readable code
- Creating objects which are deployable to production environments
- Ease of changing and redeploying quickly and cleanly
Sponsors
