End to end machine learning workflow using automation tools
Author(s)Nordin, Alexander Friedrich
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science.
MetadataShow full item record
We have developed an open source library named Trane and integrated it with two open source libraries to build an end-to-end machine learning workflow that can facilitate rapid development of machine learning models. The three components of this workflow are Trane, Featuretools and ATM. Trane enumerates tens of prediction problems relevant to any dataset using the meta information about the data. Furthermore, Trane generates training examples required for training machine learning models. Featuretools is an open-source software for automatically generating features from a dataset. Auto Tune Models (ATM), an open source library, performs a high throughput search over modeling options to find the best modeling technique for a problem. We show the capability of these three tools and highlight the open-source development of Trane.
Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2018.Cataloged from PDF version of thesis.Includes bibliographical references (pages 79-80).
DepartmentMassachusetts Institute of Technology. Department of Electrical Engineering and Computer Science.
Massachusetts Institute of Technology
Electrical Engineering and Computer Science.