Kiyayya daga ƙididdiga na koɗin koɗin jiya yana nufin auna yadda suke aiki sosai — ta amfani da ƙididdiga da suke dace (daidaita, dabarar saita, dabarar karbawar, da sauransu) akan bayanan gwaje waɗanda ƙididdiga ba su gani ba. Kiyayya da ta dace ta muhimmu ne don sanin ko ƙididdiga na gaske yana aiki kuma ta tabbatannu.
Kiyayya akan bayanan da ba a gani ba
→ evaluate on a TEST set the model did NOT train on → measures GENERALIZATION (real performance)
→ training accuracy alone is misleading (a model can memorize training data)
→ train/validation/test split; cross-validation → reliable performance estimates
Ƙididdiga da aka sani
CLASSIFICATION:
ACCURACY → % correct (but misleading for IMBALANCED data — e.g. 99% 'not fraud')
PRECISION → of predicted positives, how many are actually positive (avoid false positives)
RECALL → of actual positives, how many were found (avoid false negatives/missing cases)
F1 → balance of precision and recall
CONFUSION MATRIX → true/false positives/negatives breakdown
REGRESSION:
MAE, MSE/RMSE → average prediction error (how far off predictions are)
→ choose metrics that fit the problem (accuracy isn't always right)
