Ci bi mu melGis bi ci topp
Recall, Precision ak Elusion Testing ci Xool Mbind
Xarala
GUIDE bu am solo
Precision ak fàttaliku
Precision ak recall ñaari metric yuy mottali lañu ngir jàngat classifier yi, rawatina sudee class yi tolloowul.
Déglul tegtal biiIA-nettali · ~2 min
Ci xët wii2 simili jàng
Résumé
Ñoom ñépp dañuy wane li njubte bu leer di nëbb — ñaata yoon la li ñuy wax lu baax ci model bi di dëggu, ak ñaata yoon lay jàpp lu baax dëgg.
Plongeur bu xóot
Sudee ab model dafay màndargaal mbir yi ni lu baax, ñaari laaj ñoo am solo. Precision dafay laaj: ci lépp luñu màndargaal, ñaata moo nekkoon lu baax dëgg? Dafay méngoo ak positive dëgg yi nga xaaj ko ak positive yiñ séentu yépp, di daan alarm yu baaxul yi. Rappel (sensitivite) dafay laaj: ci mbir yu baax yi am ci biti, ñaata lañu jàpp? Dafay méngoo ak positive dëgg yi nga xaaj ko ak positive dëgg yépp, di yar ñi ñàkk. Yii dañuy faral di njaay: wàññi dogal bi dafay gëna jàpp lu baax (gëna gëna fàttaliku) waaye dafay gëna fësal junk (gëna néew njub), ak luko moy. Li nga wara njëkka tànn mingi aju ci njëg yi - filtre spam bi dafay bëgg njub (bul mbalit mail dëgg), fekk ekraŋ cancer bi dafay bëgg fàttaliku (bul ñàkka am tumër). Njortu F1, seen moyenne harmonique, ñoom ñaar ñu ngi boole ci benn nimero.
Doxal ci àdduna dëgg
Seggal spam yi dañu leen di defar ngir ñu mëna jaar yoon, suko defee ñu baña yónnee imeel yu baax yi ci dosiye spam bi ci anam wu jaarul yoon.
Test depistaasoŋ medikaal yi dañu njëkka fàttali nit ñi ngir moytu ñu réer malaad yi am feebar bi dëgg, nangu ñu gëna bari njuumte ngir topp.
Sistemu seetlu ak xalaat yi dañuy rapoor precision@k (ñaata ci resultaa yi gëna mag ñoo am njariñ) ngir natt kalite rang yi.
Gis njuuj njaaj dafay ekilibre njub ak fàttaliku jaaraleko ci poñ F1, ndax alarm yu baaxul ak njuuj njaaj yépp dañu seer lool.
Ëlëgu njub ak fàttaliku
Bi IA duggee ci domen yu am solo - saytu pajum, ëmbiit li, njuuj njaaj - ekip yi dañuy gëna xamle njub ak fàttaliku (ak seeni courbe) du njub kese, ak tune thresholds ngir méngoo ak njëg yi am ci àdduna bi ak ay jafe-jafe yoon. Grupp bu nekk ci njub/ñuy fàttaliku audit yi ñu ngi nekk standard ngir gis njuumte yu wuute yi ci demografik yi. Xaarandil metrics yu gëna am solo ci njëg, probabilite yuñ kalibre, ak jumtukaay yuy may boroom alal yi ñu tànnee poñ yi ñuy jëfandikoo ci anam wu weccoo xalaat ludul nangu 0.5 threshold biñ jagleel.
Teg ko ci jëf
Lu tax mu am solo
- dogal yu gëna leer. Daf lay jàppale nga tàqale kàddu yu leer ci wàllu xarala ak làkku fësal njaay.
- Njëgg ak budget. Mën nga laaj laaj yu gëna baax ci samp gi balaa ngay dugal xaalis wala sa jotu liggéey.
- Ekip ak def liggéey. Ekip yi bokk xam-xam ñoo gëna mëna jël yenn dogal ci wàllu produit, politik ak jàng.
Moytul
- Ekip yu bari mën nañu jëfandikoo benn baat ci anam wu wuute, kon teela leeral yaatuwaayam.
- Benchmark yi mën nañu nuru lu am doole waaye performance yi ci àdduna bi duñu tolloo.
- Bëgg kalite done ak palaŋu jàngat dafay faral di jur njariñ yu yomba dagg.
Jéego yi ci topp
- Tàmbaleel ci joxe leeral ci làkk wu leer ci njariñ li nga soxla.
- Tannal benn metric bu baax ak benn anam bu baaxul balaa ngay saytu.
- Doxal ab pilote bu ndaw ak ay done yu representatif, du ab demo bu leer.
- Bindal fi Precision ak Recall di jàppale ak fi pexe yu gëna yomba gëna baax.
Weyal di banneexu
Bulletin bu amul fayda
Jàngal xibaar AI bu bes
Ñatti jaar-jaari IA yuñ firndeel bis bu nekk ci ayu-bis bi, ñu bind leen ci Àngle bu leer. Gratis ba fàww, amul yëgle.
Benn email bis bu nekk ci ayu-bis bi. Benn klik nga dindi abonemaa bi. Dunu musa jaay wala séddoo sa adres.
Nattal sa bopp
Take the Precision and Recall quiz
Feedback ci saasi ci bépp tontu, ak sertifikaa buñ mëna séddoo ak ID buñ mëna saytu soo jàllee ci njàng mi.
Jàppale njàngum IA te doo fay. AI Understanding mbootaay la buy def te Yàlla tax — amul yëgle, amul paywall, musul am. Defal ab sarax
Laaj yi ñuy faral di laaj
Luy njub ak fàttaliku?
Precision ak recall ñaari metric yuy mottali lañu ngir jàngat classifier yi, rawatina sudee class yi tolloowul. Ñoom ñépp dañuy wane li njubte bu leer di nëbb — ñaata yoon la li ñuy wax lu baax ci model bi di dëggu, ak ñaata yoon lay jàpp lu baax dëgg.
Precision mooy tontu ban laaj?
Precision = TP / (TP + FP) dafay natt njubteg xam-xam bu baax bi ci model bi — ni darapo bu baax mëna wóolu.
On a fixed evaluation set, lowering a positive-class decision threshold usually produces which tradeoff?
Lowering the threshold expands or preserves the predicted-positive set. It may catch additional true positives and additional false positives; changes depend on which items cross the threshold.
Why can precision and recall reveal a weakness hidden by high accuracy on a dataset dominated by negatives?
Accuracy can be dominated by correct negative predictions. Precision and recall examine positive flags and missed positives using TP, FP, and FN.
Six of the top ten search results are relevant. What is precision@10 for that result list?
Precision@10 measures the relevant fraction among the top ten results: 6/10 = 60%.
Weyal di jàng
Gid yu jëm ci loolu
Tann nañu yeneen njiit ngir topic bii