Ikibazo gihuza ibibazo · Hagati Urwego

Impuzandengo y'Ibisobanuro na PR-AUC Ikibazo

Witoze neza-kwibutsa ibicuruzwa biva mu mahanga, impuzandengo-yuzuye yo kubara n'ingaruka ziganje mugihe usuzumye icyitegererezo cyerekana ibyiza bidasanzwe.

Inzira zijyanye no kuyoboraImpuzandengo ya Precision na Pr Auc
Ikibazo 1 Bya 8

Ni igipimo ki precision ipima ku murongo w'isesengura watoranijwe?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Ni igipimo ki precision ipima ku murongo w'isesengura watoranijwe?
    • The fraction of all positive cases that were selected
    • The fraction of selected cases that are positive
    • The fraction of all negative cases that were excluded
    • The average score assigned to every unselected case
  2. Mu gutondeka hatarimo amanota angana, ibya mbere n'icya gatatu ni byo byiza gusa. Ni iyihe precision y'ubusanzwe ikurikira?
    • 0.5
    • 0.667
    • 1.0
    • About 0.833
  3. Kuki ubuso bwa trapezoidal PR-AUC bushobora gutandukana na precision y'ubusanzwe?
    • They use different interpolation and summation rules
    • Average precision averages precision equally across every threshold
    • Trapezoidal PR-AUC weights only thresholds where a positive case enters
    • Average precision always evaluates exactly one decision threshold
  4. Ni ibihe byinjiza bikomeza amakuru y'urutonde akeneye precision y'ubusanzwe?
    • Only the final hard decisions and their total count
    • Observed labels and one shared threshold, without per-case scores
    • Observed labels and model prediction scores
    • Only a confusion matrix at one threshold
  5. Dataseti ebyiri zifite ubwinshi butandukanye bw'ibintu byiza. Ni iki kigomba kujyana n'igereranya ry'amanota yabo ya precision-recall?
    • Only overall accuracy on each dataset
    • The positive fractions and evaluation sampling designs
    • Only whether the two models use the same threshold
    • Only whether the predicted scores occupy the same numeric range
  6. What happens to recall as a score threshold is lowered and more cases are included?
    • It must strictly increase at every step
    • It can decrease when a negative case is added
    • It must become identical to precision
    • It cannot decrease
  7. Why must tied prediction scores be handled carefully when calculating average precision?
    • Equal-score cases cross a threshold together
    • Changing the order of tied rows always leaves the no-ties rank shortcut unchanged
    • A threshold can select tied positives before tied negatives
    • Tied examples should each receive a different threshold based on their true labels
  8. A model has high average precision. Which deployment question still needs a separate check?
    • Whether the model achieved a useful summary across thresholds
    • Whether the reported summary used precision and recall
    • Whether precision and workload are acceptable at the chosen operating threshold
    • Whether recall increases were used as weights in the reported AP calculation