{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/tensorflow-estimators-managing-simplicity-vs","title":"TensorFlow Estimators: Managing Simplicity vs. Flexibility in High-Level Machine Learning Frameworks","arxiv_id":"1708.02637","date":"2017-08-08","proceeding":null,"authors":["Heng-Tze Cheng","Zakaria Haque","Lichan Hong","Mustafa Ispir","Clemens Mewald","Illia Polosukhin","Georgios Roumpos","D. Sculley","Jamie Smith","David Soergel","Yuan Tang","Philipp Tucker","Martin Wicke","Cassandra Xia","Jianwei Xie"],"abstract":"We present a framework for specifying, training, evaluating, and deploying\nmachine learning models. Our focus is on simplifying cutting edge machine\nlearning for practitioners in order to bring such technologies into production.\nRecognizing the fast evolution of the field of deep learning, we make no\nattempt to capture the design space of all possible model architectures in a\ndomain- specific language (DSL) or similar configuration language. We allow\nusers to write code to define their models, but provide abstractions that guide\ndevelop- ers to write models in ways conducive to productionization. We also\nprovide a unifying Estimator interface, making it possible to write downstream\ninfrastructure (e.g. distributed training, hyperparameter tuning) independent\nof the model implementation. We balance the competing demands for flexibility\nand simplicity by offering APIs at different levels of abstraction, making\ncommon model architectures available out of the box, while providing a library\nof utilities designed to speed up experimentation with model architectures. To\nmake out of the box models flexible and usable across a wide range of problems,\nthese canned Estimators are parameterized not only over traditional\nhyperparameters, but also using feature columns, a declarative specification\ndescribing how to interpret input data. We discuss our experience in using this\nframework in re- search and production environments, and show the impact on\ncode health, maintainability, and development speed.","url_abs":"http://arxiv.org/abs/1708.02637v1","url_pdf":"http://arxiv.org/pdf/1708.02637v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"tensorflow-estimators-managing-simplicity-vs","repo_url":"https://github.com/tensorflow/tensorflow","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}