Building a Reproducible Machine Learning Pipeline

Reproducibility of modeling is a problem that exists for any machine learning practitioner, whether in industry or academia. The consequences of an irreproducible model can include significant financial costs, lost time, and even loss of personal reputation (if results prove unable to be replicated)...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Sugimura, Peter, Hartl, Florian
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Sugimura, Peter
Hartl, Florian
description Reproducibility of modeling is a problem that exists for any machine learning practitioner, whether in industry or academia. The consequences of an irreproducible model can include significant financial costs, lost time, and even loss of personal reputation (if results prove unable to be replicated). This paper will first discuss the problems we have encountered while building a variety of machine learning models, and subsequently describe the framework we built to tackle the problem of model reproducibility. The framework is comprised of four main components (data, feature, scoring, and evaluation layers), which are themselves comprised of well defined transformations. This enables us to not only exactly replicate a model, but also to reuse the transformations across different models. As a result, the platform has dramatically increased the speed of both offline and online experimentation while also ensuring model reproducibility.
doi_str_mv 10.48550/arxiv.1810.04570
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1810_04570</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1810_04570</sourcerecordid><originalsourceid>FETCH-LOGICAL-a670-20814eda97239979f3a70815b83ca0a7aa19854166bfbf15560c14164b4ffbb33</originalsourceid><addsrcrecordid>eNotjk0KwjAYRLNxIdUDuDIXqCYmaZKlin9QUcR9-dImGqi1RBS9vW11NcwbGB5CI0omXAlBphDe_jWhqgGEC0n6iC6evix8dcGAT7YO9-KZe1NavIf86iuLUwuhavejr23ZkAHqOSgfdvjPCJ3Xq_NyG6eHzW45T2NIJIlnRFFuC9ByxrSW2jGQDRJGsRwISACqleA0SYwzjgqRkJw2lRvunDGMRWj8u-2cszr4G4RP1rpnnTv7AgojPT8</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Building a Reproducible Machine Learning Pipeline</title><source>arXiv.org</source><creator>Sugimura, Peter ; Hartl, Florian</creator><creatorcontrib>Sugimura, Peter ; Hartl, Florian</creatorcontrib><description>Reproducibility of modeling is a problem that exists for any machine learning practitioner, whether in industry or academia. The consequences of an irreproducible model can include significant financial costs, lost time, and even loss of personal reputation (if results prove unable to be replicated). This paper will first discuss the problems we have encountered while building a variety of machine learning models, and subsequently describe the framework we built to tackle the problem of model reproducibility. The framework is comprised of four main components (data, feature, scoring, and evaluation layers), which are themselves comprised of well defined transformations. This enables us to not only exactly replicate a model, but also to reuse the transformations across different models. As a result, the platform has dramatically increased the speed of both offline and online experimentation while also ensuring model reproducibility.</description><identifier>DOI: 10.48550/arxiv.1810.04570</identifier><language>eng</language><subject>Computer Science - Learning ; Computer Science - Software Engineering ; Statistics - Machine Learning</subject><creationdate>2018-10</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/1810.04570$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1810.04570$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Sugimura, Peter</creatorcontrib><creatorcontrib>Hartl, Florian</creatorcontrib><title>Building a Reproducible Machine Learning Pipeline</title><description>Reproducibility of modeling is a problem that exists for any machine learning practitioner, whether in industry or academia. The consequences of an irreproducible model can include significant financial costs, lost time, and even loss of personal reputation (if results prove unable to be replicated). This paper will first discuss the problems we have encountered while building a variety of machine learning models, and subsequently describe the framework we built to tackle the problem of model reproducibility. The framework is comprised of four main components (data, feature, scoring, and evaluation layers), which are themselves comprised of well defined transformations. This enables us to not only exactly replicate a model, but also to reuse the transformations across different models. As a result, the platform has dramatically increased the speed of both offline and online experimentation while also ensuring model reproducibility.</description><subject>Computer Science - Learning</subject><subject>Computer Science - Software Engineering</subject><subject>Statistics - Machine Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2018</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotjk0KwjAYRLNxIdUDuDIXqCYmaZKlin9QUcR9-dImGqi1RBS9vW11NcwbGB5CI0omXAlBphDe_jWhqgGEC0n6iC6evix8dcGAT7YO9-KZe1NavIf86iuLUwuhavejr23ZkAHqOSgfdvjPCJ3Xq_NyG6eHzW45T2NIJIlnRFFuC9ByxrSW2jGQDRJGsRwISACqleA0SYwzjgqRkJw2lRvunDGMRWj8u-2cszr4G4RP1rpnnTv7AgojPT8</recordid><startdate>20181009</startdate><enddate>20181009</enddate><creator>Sugimura, Peter</creator><creator>Hartl, Florian</creator><scope>AKY</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20181009</creationdate><title>Building a Reproducible Machine Learning Pipeline</title><author>Sugimura, Peter ; Hartl, Florian</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a670-20814eda97239979f3a70815b83ca0a7aa19854166bfbf15560c14164b4ffbb33</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2018</creationdate><topic>Computer Science - Learning</topic><topic>Computer Science - Software Engineering</topic><topic>Statistics - Machine Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Sugimura, Peter</creatorcontrib><creatorcontrib>Hartl, Florian</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Sugimura, Peter</au><au>Hartl, Florian</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Building a Reproducible Machine Learning Pipeline</atitle><date>2018-10-09</date><risdate>2018</risdate><abstract>Reproducibility of modeling is a problem that exists for any machine learning practitioner, whether in industry or academia. The consequences of an irreproducible model can include significant financial costs, lost time, and even loss of personal reputation (if results prove unable to be replicated). This paper will first discuss the problems we have encountered while building a variety of machine learning models, and subsequently describe the framework we built to tackle the problem of model reproducibility. The framework is comprised of four main components (data, feature, scoring, and evaluation layers), which are themselves comprised of well defined transformations. This enables us to not only exactly replicate a model, but also to reuse the transformations across different models. As a result, the platform has dramatically increased the speed of both offline and online experimentation while also ensuring model reproducibility.</abstract><doi>10.48550/arxiv.1810.04570</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.1810.04570
ispartof
issn
language eng
recordid cdi_arxiv_primary_1810_04570
source arXiv.org
subjects Computer Science - Learning
Computer Science - Software Engineering
Statistics - Machine Learning
title Building a Reproducible Machine Learning Pipeline
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-05T09%3A45%3A59IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Building%20a%20Reproducible%20Machine%20Learning%20Pipeline&rft.au=Sugimura,%20Peter&rft.date=2018-10-09&rft_id=info:doi/10.48550/arxiv.1810.04570&rft_dat=%3Carxiv_GOX%3E1810_04570%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true