{"id":"https://openalex.org/W6893122931","doi":"https://doi.org/10.5281/zenodo.14128996","title":"Replication Package for 'How do Machine Learning Models Change?'","display_name":"Replication Package for 'How do Machine Learning Models Change?'","publication_year":2025,"publication_date":"2025-09-05","ids":{"openalex":"https://openalex.org/W6893122931","doi":"https://doi.org/10.5281/zenodo.14128996"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.14128996","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.14128996","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Dataset"},"type":"dataset","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.5281/zenodo.14128996","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Casta\u00f1o Fern\u00e1ndez, Joel","orcid":"https://orcid.org/0009-0005-6385-9255"},"institutions":[{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Casta\u00f1o Fern\u00e1ndez, Joel","raw_affiliation_strings":["Universitat Polit\u00e8cnica de Catalunya"],"raw_orcid":"https://orcid.org/0009-0005-6385-9255","affiliations":[{"raw_affiliation_string":"Universitat Polit\u00e8cnica de Catalunya","institution_ids":["https://openalex.org/I9617848"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Caba\u00f1as, Rafael","orcid":"https://orcid.org/0000-0002-5034-582X"},"institutions":[{"id":"https://openalex.org/I52354020","display_name":"University of Almer\u00eda","ror":"https://ror.org/003d3xx08","country_code":"ES","type":"education","lineage":["https://openalex.org/I52354020"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Caba\u00f1as, Rafael","raw_affiliation_strings":["University of Almer\u00eda"],"raw_orcid":"https://orcid.org/0000-0002-5034-582X","affiliations":[{"raw_affiliation_string":"University of Almer\u00eda","institution_ids":["https://openalex.org/I52354020"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Salmer\u00f3n, Antonio","orcid":"https://orcid.org/0000-0003-4982-8725"},"institutions":[{"id":"https://openalex.org/I52354020","display_name":"University of Almer\u00eda","ror":"https://ror.org/003d3xx08","country_code":"ES","type":"education","lineage":["https://openalex.org/I52354020"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Salmer\u00f3n, Antonio","raw_affiliation_strings":["University of Almer\u00eda"],"raw_orcid":"https://orcid.org/0000-0003-4982-8725","affiliations":[{"raw_affiliation_string":"University of Almer\u00eda","institution_ids":["https://openalex.org/I52354020"]}]},{"author_position":"middle","author":{"id":null,"display_name":"David, Lo","orcid":"https://orcid.org/0000-0002-4367-7201"},"institutions":[{"id":"https://openalex.org/I79891267","display_name":"Singapore Management University","ror":"https://ror.org/050qmg959","country_code":"SG","type":"education","lineage":["https://openalex.org/I79891267"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"David, Lo","raw_affiliation_strings":["Singapore Management University"],"raw_orcid":"https://orcid.org/0000-0002-4367-7201","affiliations":[{"raw_affiliation_string":"Singapore Management University","institution_ids":["https://openalex.org/I79891267"]}]},{"author_position":"last","author":{"id":null,"display_name":"Mart\u00ednez-Fern\u00e1ndez, Silverio","orcid":"https://orcid.org/0000-0001-9928-133X"},"institutions":[{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Mart\u00ednez-Fern\u00e1ndez, Silverio","raw_affiliation_strings":["Universitat Polit\u00e8cnica de Catalunya"],"raw_orcid":"https://orcid.org/0000-0001-9928-133X","affiliations":[{"raw_affiliation_string":"Universitat Polit\u00e8cnica de Catalunya","institution_ids":["https://openalex.org/I9617848"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":true,"primary_topic":null,"topics":[],"keywords":[{"id":"https://openalex.org/keywords/commit","display_name":"Commit","score":0.7922999858856201},{"id":"https://openalex.org/keywords/metadata","display_name":"Metadata","score":0.534600019454956},{"id":"https://openalex.org/keywords/replication","display_name":"Replication (statistics)","score":0.504800021648407},{"id":"https://openalex.org/keywords/preprocessor","display_name":"Preprocessor","score":0.4708000123500824},{"id":"https://openalex.org/keywords/categorization","display_name":"Categorization","score":0.4296000003814697},{"id":"https://openalex.org/keywords/documentation","display_name":"Documentation","score":0.3991999924182892},{"id":"https://openalex.org/keywords/face","display_name":"Face (sociological concept)","score":0.38749998807907104}],"concepts":[{"id":"https://openalex.org/C153180980","wikidata":"https://www.wikidata.org/wiki/Q19776675","display_name":"Commit","level":2,"score":0.7922999858856201},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7807000279426575},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.6664999723434448},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6539000272750854},{"id":"https://openalex.org/C93518851","wikidata":"https://www.wikidata.org/wiki/Q180160","display_name":"Metadata","level":2,"score":0.534600019454956},{"id":"https://openalex.org/C12590798","wikidata":"https://www.wikidata.org/wiki/Q3933199","display_name":"Replication (statistics)","level":2,"score":0.504800021648407},{"id":"https://openalex.org/C34736171","wikidata":"https://www.wikidata.org/wiki/Q918333","display_name":"Preprocessor","level":2,"score":0.4708000123500824},{"id":"https://openalex.org/C94124525","wikidata":"https://www.wikidata.org/wiki/Q912550","display_name":"Categorization","level":2,"score":0.4296000003814697},{"id":"https://openalex.org/C56666940","wikidata":"https://www.wikidata.org/wiki/Q788790","display_name":"Documentation","level":2,"score":0.3991999924182892},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.38749998807907104},{"id":"https://openalex.org/C10551718","wikidata":"https://www.wikidata.org/wiki/Q5227332","display_name":"Data pre-processing","level":2,"score":0.3675000071525574},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.3544999957084656},{"id":"https://openalex.org/C52001869","wikidata":"https://www.wikidata.org/wiki/Q812530","display_name":"Naive Bayes classifier","level":3,"score":0.34779998660087585},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.33399999141693115},{"id":"https://openalex.org/C133462117","wikidata":"https://www.wikidata.org/wiki/Q4929239","display_name":"Data collection","level":2,"score":0.30149999260902405},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.2915000021457672},{"id":"https://openalex.org/C55439883","wikidata":"https://www.wikidata.org/wiki/Q360812","display_name":"Correctness","level":2,"score":0.29019999504089355},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.2865000069141388},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.26179999113082886},{"id":"https://openalex.org/C195807954","wikidata":"https://www.wikidata.org/wiki/Q1662562","display_name":"Information extraction","level":2,"score":0.2563000023365021}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.5281/zenodo.14128996","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.14128996","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Dataset"}],"best_oa_location":{"id":"doi:10.5281/zenodo.14128996","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.14128996","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Dataset"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Replication":[0],"Package:":[1],"How":[2,764],"Do":[3,16],"Machine":[4,17],"Learning":[5,18],"Models":[6,19],"Change?":[7],"Overview":[8],"This":[9,152,737,767],"replication":[10,461],"package":[11,153,462],"accompanies":[12],"the":[13,45,116,135,156,164,180,185,263,278,317,324,390,424,456,486,539,547,574,580,585,605,635,639,646,658,670,681,690,699,707,716,725,741,758,801,807,841,850,856,862,867,878,898,948],"paper":[14],"\"How":[15],"Change?\".":[20],"In":[21],"this":[22,327,829,944],"study,":[23],"we":[24,245,329],"conducted":[25],"a":[26,66,100,236,247,331,348,364,375,385,416,434,449,774,818],"large-scale":[27],"analysis":[28,540,549,581,759],"of":[29,41,89,103,111,120,139,239,265,326,352,387,409,419,437,550,614,657,757],"over":[30,94,202,344,595],"680,000":[31],"commits":[32,96,288,444,598],"from":[33,39,179,201,493,625,762,924],"100,000":[34],"models":[35,43,59,205,910,916],"and":[36,71,106,118,137,146,160,172,221,293,476,500,515,522,527,531,601,634,660,772,812,882,887,893],"2,251":[37,615],"releases":[38,617,923],"202":[40,925],"these":[42],"on":[44,97,374,403],"Hugging":[46,181,186,494],"Face":[47,182,187],"(HF)":[48],"platform.":[49],"Our":[50,81],"goal":[51],"was":[52,193,257,341,357,401,431],"to":[53,126,162,196,260,289,309,472,608,752,765,814,848,914],"understand":[54],"how":[55,143],"machine":[56],"learning":[57],"(ML)":[58],"evolve":[60],"by":[61,359,368,872],"classifying":[62],"commit":[63,121,219,270,307,501,588],"types":[64,105,122],"using":[65,78,123,184,316,347,448],"detailed":[67,101,237,454],"ML":[68,313],"change":[69,104,148,314,552],"taxonomy":[70,315],"analyzing":[72,142],"temporal":[73,128],"patterns":[74],"in":[75,167,242,455,579,806,817,843,855,869],"their":[76,107],"activities":[77],"Bayesian":[79,124],"networks.":[80],"research":[82,543],"addresses":[83],"three":[84],"main":[85,586],"aspects:":[86],"1.":[87,335,559,770],"Categorization":[88],"Commit":[90,112,232,274,302,652],"Changes:":[91,233],"We":[92,114,133,176,215,276,304,441,826],"classified":[93,305,443,587,597,616],"960,000":[95,596],"HF,":[98],"providing":[99],"breakdown":[102],"distribution.":[108],"2.":[109,394,565,790],"Analysis":[110,555,561,567,802],"Sequences:":[113],"examined":[115],"sequence":[117],"dependencies":[119],"networks":[125],"identify":[127,290],"patterns.":[129,553],"3.":[130,571],"Release":[131,213],"Analysis**:":[132],"investigated":[134],"distribution":[136],"evolution":[138],"release":[140,222,505,533],"types,":[141],"model":[144,217,499,626],"attributes":[145],"metadata":[147,477,624,731],"across":[149],"successive":[150],"releases.":[151,609],"provides":[154],"all":[155,482],"necessary":[157],"code,":[158,474],"data,":[159,475],"documentation":[161],"reproduce":[163,833],"results":[165],"presented":[166],"our":[168,227,834,954,960],"paper.":[169,457,961],"Data":[170,174,272],"Collection":[171,175,821],"Preprocessing":[173,273,823],"collected":[177,216],"data":[178,191,491,512,524,636,880],"platform":[183],"Hub":[188],"API.":[189],"The":[190,339,355,380,398,412,460,467,591,611,621,664,803],"extraction":[192,492],"performed":[194],"up":[195],"May":[197],"2025,":[198],"capturing":[199],"details":[200,656],"1":[203],"million":[204],"available":[206],"at":[207],"that":[208,876],"time.":[209],"-":[210,230,831,859,903,919,927],"Model":[211],"&":[212,906],"Information:":[214],"metadata,":[218],"histories,":[220],"information":[223],"(Git":[224],"tags)":[225],"for":[226,268,280,300,478,485,490,511,525,537,541,556,562,568,599,618,628,638,669,680,689,698,706,715,724,897,911,917],"sampled":[228],"models.":[229],"Detailed":[231],"To":[234,322,832,860],"get":[235],"list":[238,264],"files":[240,267,284,627,633,749],"modified":[241],"each":[243,253,306,542],"commit,":[244],"implemented":[246,330],"direct":[248],"Git":[249],"processing":[250],"approach.":[251],"For":[252,904,920,928],"model,":[254,452],"its":[255,674],"repository":[256],"temporarily":[258],"cloned":[259],"programmatically":[261],"extract":[262],"changed":[266],"every":[269],"SHA.":[271],"Diffs":[275],"computed":[277],"differences":[279],"key":[281,575],"JSON":[282],"configuration":[283,632],"(e.g.,":[285],"`config.json`)":[286],"between":[287],"added,":[291],"deleted,":[292],"updated":[294],"keys,":[295],"which":[296,429],"served":[297],"as":[298,453,465],"input":[299],"classification.":[301,516],"Classification":[303],"according":[308],"Bhatia":[310],"et":[311],"al.'s":[312],"Gemini":[318],"2.5":[319],"Flash":[320],"LLM.":[321,899],"ensure":[323],"reliability":[325],"process,":[328],"rigorous":[332],"two-phase":[333],"validation:":[334],"Prompt":[336],"Refinement":[337],"(Training):":[338],"prompt":[340,383,400,667],"iteratively":[342],"refined":[343,382],"6":[345],"cycles":[346],"curated":[349],"training":[350,391,683,691,700,717],"set":[351],"143":[353],"commits.":[354,411],"process":[356],"guided":[358],"comparing":[360],"LLM":[361,671],"classifications":[362,414],"against":[363,389,423],"gold":[365,392,427,647],"standard":[366,648,679,705],"created":[367],"two":[369],"human":[370],"annotators":[371],"(Human-Human":[372],"IRR":[373,436,692,701,718,727],"subset:":[376],"\ud835\udf05":[377,438],"=":[378,439],"0.7798).":[379],"final":[381,576,592,612],"achieved":[384,415],"Kappa":[386,418],"0.9068":[388],"standard.":[393],"Final":[395],"Validation":[396],"(Testing):":[397],"validated":[399,432,666],"tested":[402],"an":[404],"independent,":[405],"statistically":[406],"significant":[407,890],"sample":[408],"384":[410],"LLM's":[413],"Cohen's":[417],"0.8568":[420],"when":[421],"compared":[422],"test":[425,709,726],"set's":[426],"standard,":[428],"itself":[430],"with":[433,594,840,866,943],"human-human":[435],"0.8150.":[440],"also":[442],"into":[445],"Swanson's":[446],"categories":[447],"fine-tuned":[450],"DistilBERT":[451],"Folder":[458],"Structure":[459],"is":[463,885],"organized":[464],"follows.":[466],"structure":[468],"has":[469],"been":[470],"designed":[471],"separate":[473],"clarity.":[479],"`code/`:":[480],"Contains":[481,546,573,584,604,631],"Jupyter":[483,804],"notebooks":[484,805,842,868],"study.":[487],"`Collection/`:":[488],"Scripts":[489,510],"Face.":[495],"`HFExtraction.ipynb`:":[496],"Collects":[497,504],"primary":[498],"information.":[502,508],"`HFReleasesExtraction.ipynb`:":[503],"(tag)":[506],"specific":[507,755],"`Preprocessing/`:":[509],"cleaning,":[513],"processing,":[514],"`HFCommitsPreprocessing.ipynb`:":[517],"Processes":[518,530],"commits,":[519],"computes":[520],"diffs,":[521],"prepares":[523],"classification":[526,672,883],"analysis.":[528],"`HFReleasesPreprocessing.ipynb`:":[529],"classifies":[532],"data.":[534,649],"`Analysis/`:":[535],"Notebooks":[536],"reproducing":[538],"question.":[544],"`HFFileChanges.ipynb`:":[545],"preliminary":[548],"file":[551,732],"`RQ1_Analysis.ipynb`:":[554],"Research":[557,563,569],"Question":[558,564,570],"`RQ2_Analysis.ipynb`:":[560],"`RQ3_Analysis.ipynb`:":[566],"`datasets/`:":[572],"datasets":[577,606,853],"used":[578,637,668,733],"notebooks.":[582],"`commits_datasets/`:":[583],"dataset.":[589],"`HFCommitsClassification_final.csv`:":[590],"dataset":[593,613],"RQ1":[600,905],"RQ2.":[602],"`releases_datasets/`:":[603],"related":[607],"`HFReleasesClassification.csv`:":[610],"RQ3.":[619],"`model_metadata.csv`:":[620],"extracted":[622],"internal":[623],"RQ3.4.":[629],"`metadata/`:":[630],"validation":[640,661],"process.":[641],"`validation_data/`:":[642],"A":[643],"sub-folder":[644],"containing":[645,655],"`Agreement":[650],"TOSEM":[651],"Changes.xlsx`:":[653],"Excel":[654],"classication":[659],"processes.":[662],"`prompt_refinement.txt`:":[663],"final,":[665,851],"along":[673],"previous":[675],"iterations.":[676],"`training_set_ground_truth.json`:":[677],"Gold":[678,704],"143-commit":[682],"set.":[684,710],"`training_set_first_classification.json`:":[685],"First":[686,712],"annotator's":[687,696,713,722],"labels":[688,697,714,723],"subset.":[693,702,719,728],"`training_set_second_classification.json`:":[694],"Second":[695,721],"`test_set_ground_truth.json`:":[703],"384-commit":[708],"`test_first_classification.json`:":[711],"`test_set_second_classification.json`:":[720],"`tags_metadata.yaml`:":[729],"Auxiliary":[730],"during":[734],"preprocessing.":[735],"`README.md`:":[736],"file.":[738],"`requirements.txt`:":[739],"Lists":[740],"required":[742,792],"Python":[743],"packages.":[744,793],"*Note:":[745],"Other":[746],"intermediate":[747],"CSV":[748],"are":[750,810,846],"provided":[751,854],"facilitate":[753],"re-running":[754],"parts":[756],"without":[760],"starting":[761],"scratch.*":[763],"Use":[766],"Package":[768],"Setup":[769],"Create":[771],"activate":[773],"virtual":[775],"environment":[776],"(recommended).":[777],"```bash":[778,794],"python":[779],"-m":[780],"venv":[781,782],"source":[783],"venv/bin/activate":[784],"#":[785],"On":[786],"Windows:":[787],"venv\\Scripts\\activate":[788],"```":[789,799],"Install":[791],"pip":[795],"install":[796],"-r":[797],"requirements.txt":[798],"Running":[800],"`code/`":[808],"directory":[809],"numbered":[811],"named":[813],"be":[815],"run":[816],"logical":[819],"sequence:":[820],"->":[822,824],"Analysis.":[825],"recommend":[827],"following":[828],"order.":[830],"findings":[835],"directly,":[836],"you":[837,936,952],"can":[838],"start":[839,865],"`code/Analysis/`.":[844],"They":[845],"configured":[847],"load":[849],"processed":[852],"`datasets/`":[857],"folder.":[858],"re-run":[861],"entire":[863],"pipeline,":[864],"`code/Collection/`,":[870],"followed":[871],"`code/Preprocessing/`.":[873],"Please":[874],"note":[875],"running":[877],"full":[879],"collection":[881],"pipeline":[884],"time-consuming":[886],"may":[888],"require":[889],"computational":[891],"resources":[892],"appropriate":[894],"API":[895],"keys":[896],"Key":[900],"Datasets":[901],"Provided":[902],"RQ2:":[907],"`datasets/commits_datasets/HFCommitsClassification_final.csv`":[908],"(100,000":[909],"RQ1;":[912],"filtered":[913],"14,343":[915],"RQ2).":[918],"RQ3.1-3.3:`datasets/releases_datasets/HFReleasesClassification.csv`":[921],"(2,251":[922],"models).":[926,933],"RQ3.4:":[929],"`datasets/releases_datasets/model_metadata.csv`":[930],"(from":[931],"28":[932],"Contact":[934],"If":[935,951],"have":[937],"any":[938],"questions":[939],"or":[940],"encounter":[941],"issues":[942],"package,":[945],"please":[946,957],"contact":[947],"corresponding":[949],"author.":[950],"find":[953],"work":[955],"useful,":[956],"consider":[958],"citing":[959]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2025-10-10T00:00:00"}