Idanwo ti o sopọ mọ itọsọna · Alabọde Ipele

K-Nitosi Awọn aladugbo adanwo

Covers neighbor voting, feature scaling, the cost of prediction, metric choice, and practical tradeoffs when storing training examples.

Jẹmọ itọsọna awọn ọnaK Awọn aladugbo to sunmọ
Ibeere 1 ti 8

Bawo ni KNN ṣe lẹtọ aaye data tuntun kan?

Awotẹlẹ gbogbo ibeere8 ibeere

Awọn idahun ati awọn alaye han bi o ṣe n mu ibeere naa.

  1. Bawo ni KNN ṣe lẹtọ aaye data tuntun kan?
    • Nipa gbigbe ibo to poju ti awọn aladugbo K rẹ ti o sunmọ
    • Nipa ikẹkọ nẹtiwọọki nkankikan lori rẹ
    • Nipa ṣiṣe iṣiro laini ipadasẹhin iwuwo
    • Nipa fifun kilasi laileto
  2. Why is KNN often called a “lazy learner”?
    • It never stores training data
    • It stores training examples and can defer much model-specific work until prediction, though indexes may be built during fitting
    • It always trains a deep neural network first
    • It does no work during either fitting or prediction
  3. Kilode ti igbelowọn ẹya ṣe pataki fun KNN?
    • O mu ki ikẹkọ yarayara
    • O dinku nọmba awọn aladugbo ti o nilo
    • Nitorinaa awọn ẹya pẹlu awọn sakani nla ko jẹ gaba lori ijinna naa
    • O yipada K laifọwọyi
  4. Kini yoo ṣẹlẹ ti o ba yan K kekere kan, bii K=1?
    • Awọn awoṣe di pupọ dan
    • Awọn awoṣe foju awọn data
    • Awọn asọtẹlẹ di eyiti ko ṣee ṣe
    • Awọn awoṣe di kókó si ariwo ati ki o le overfit
  5. In a brute-force KNN implementation, why can prediction be slow on a large dataset?
    • It may compute distances from a query to many or all stored examples
    • It must update all model weights for every query
    • It reruns the entire training dataset through a neural network
    • It cannot use an index or approximate search
  6. Imọ-ẹrọ igbalode wo ni o gbẹkẹle imọran aladugbo ti o sunmọ julọ?
    • Aworan funmorawon pẹlu JPEG
    • Wiwa Vector ati imupadabọ-imudara iran
    • ID nọmba iran
    • Awọn agbekalẹ lẹja
  7. Why should a feature scaler be fit only on the training split before KNN evaluation?
    • To make the distance metric ignore every feature
    • To guarantee all classes have equal counts
    • To avoid using information from validation or test examples when setting the scale
    • To remove the need to choose K
  8. How can KNN make a prediction for a regression task?
    • Choose the most common class label only
    • Fit a new deep network for each query
    • Ignore the stored examples and use a fixed intercept
    • Aggregate the target values of the K nearest training examples, often by averaging