MétaCan
Menu
Back to cohort
Record W4399531343 · doi:10.1145/3644815.3644963

An Exploratory Study of Dataset and Model Management in Open Source Machine Learning Applications

2024· article· en· W4399531343 on OpenAlexafffund
Tajkia Rahman Toma, Cor‐Paul Bezemer

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicAdvanced Data Storage Technologies
Canadian institutionsUniversity of Alberta
FundersUniversity of Alberta
KeywordsComputer scienceTRACE (psycholinguistics)Key (lock)Focus (optics)Open sourceData modelingData scienceData miningPredictive modellingDatabaseMachine learningSoftwareOperating system

Abstract

fetched live from OpenAlex

Datasets and models are two key artifacts in machine learning (ML) applications. Although there exist tools to support dataset and model developers in managing ML artifacts, little is known about how these datasets and models are integrated into ML applications. In this paper, we study how datasets and models in ML applications are managed. In particular, we focus on how these artifacts are stored and versioned alongside the applications. After analyzing 93 repositories, we identified the most common storage location to store datasets and models is the file system, which causes availability issues. Notably, large data and model files, exceeding approximately 60 MB, are stored exclusively in remote storage and downloaded as needed. Most of the datasets and models lack proper integration with the version control system, posing potential trace-ability and reproducibility issues. Additionally, although datasets and models are likely to evolve during the application development, they are rarely updated in application repositories.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.011
metaresearch head score (Gemma)0.069
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesOpen science
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Qualitative · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.998
Threshold uncertainty score0.056

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0110.069
Meta-epidemiology (narrow)0.0000.001
Meta-epidemiology (broad)0.0000.001
Bibliometrics0.0030.005
Science and technology studies0.0020.001
Scholarly communication0.0030.006
Open science0.0020.003
Research integrity0.0010.002
Insufficient payload (model declined to judge)0.0010.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.044
GPT teacher head0.330
Teacher spread0.287 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

Study designQualitative
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations6
Published2024
Admission routes2
Has abstractyes

Explore more

Same topicAdvanced Data Storage TechnologiesFrench-language works237,207