Module 4 Test: Machine Learning
Ye tumhara progress tracker hai. 20 sawal: 5 Easy, 10 Moderate, 5 Hard . Pehle sab sawal khud hal karo (code ko Colab me chalane se pehle apne dimagh se output sochna), phir Answer Key se check karo.
| Score | Matlab |
|---|---|
| 18 se 20 | Zabardast! Module 4 par tumhari grip mazboot hai. |
| 12 se 17 | Acha hai. Jin sawalon me galti hui un chapters ko dobara parho. |
| 12 se kam | Pareshan na ho. Chapters 12 se 20 dobara parho aur code khud chalao, phir test dobara do. |
EASY
Q1. Supervised Learning me training data ke saath kya maujood hota hai?
-
A) Sirf features, koi jawab nahi
-
B) Features ke saath jawab (label / target) bhi
-
C) Sirf images
D) Sirf random numbers
EASY
Q2. Overfitting ka sahi matlab kya hai?
-
A) Model train aur test dono par kharab hai
-
B) Model train par bohat achha magar naye (test) data par kharab hai
-
C) Model bohat tez chalta hai
D) Data me missing values hain
EASY
Q3. K-Means me "K" ka kya matlab hai?
-
A) Padosiyon ki tadaad
-
B) Clusters (groups) ki tadaad
-
C) Features ki tadaad
-
D) Iterations ki tadaad
EASY
Q4. Neeche me se kaunsa metric batata hai ke asli positive cases me se model ne kitne pakde?
- A) Precision
B) Recall
- C) Accuracy
D) Specificity
EASY
Q5. PCA ka bunyadi maqsad kya hai?
-
A) Missing values bharna
-
B) Features ki tadaad kam karna aur zyada se zyada maloomat bachana
-
C) Labels banana
D) Data ko text me badalna
MODERATE
Q6. Is code ka output kya hoga?
import numpy as np
from sklearn.model_selection import train_test_split
X = np.arange(100).reshape(50, 2)
Xtr, Xte = train_test_split(X, test_size=0.2, random_state=0)
print(len(Xtr), len(Xte))
A) 30 20
B) 40 10 C) 45 5 D) 10 40
MODERATE
Q7. Is code ka output kya hoga (precision, recall)?
from sklearn.metrics import precision_score, recall_score
y_true = [1, 1, 1, 0, 0, 0, 0, 0]
y_pred = [1, 1, 0, 1, 1, 0, 0, 0]
print(round(precision_score(y_true, y_pred), 2), round(recall_score(y_true, y_pred), 2))
A) 0.67 0.5 B) 0.5 0.67 C) 0.5 0.5 D) 0.67 0.67
MODERATE
Q8. Decision Tree me do classes barabar barabar (50% - 50%) mix hon to Gini impurity kitni hogi?
print(1 - (0.5 ** 2 + 0.5 ** 2))
A) 0.0 B) 0.25 C) 0.5 D) 1.0
MODERATE
Q9. Is code ka output kya hoga?
import numpy as np
from sklearn.preprocessing import StandardScaler
print(StandardScaler().fit_transform(np.array([[1], [2], [3]])).ravel().round(2))
A) [0. 0.5 1.] B) [-1.22 0. 1.22] C) [-1. 0. 1.] D) [1 2 3]
MODERATE
Q10. Neeche me se kaunsa tareeqa data leakage hai?
A) StandardScaler ko sirf train data par fit karna
B) Poore data par scaler fit karke phir train/test split karna
C) Pipeline ke andar scaler rakhna
D) Cross-validation ke har fold me alag scaler fit hona
MODERATE
Q11. 5-fold cross-validation par cross_val_score kitne scores wapas karta hai?
from sklearn.datasets import load_iris
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import cross_val_score
X, y = load_iris(return_X_y=True)
print(len(cross_val_score(LogisticRegression(max_iter=500), X, y, cv=5)))
A) 1 B) 3 C) 5 D) 150
MODERATE
Q12. Is code ka output kya hoga (K-Means clusters ki sizes)?
import numpy as np
from sklearn.cluster import KMeans
X = [[1, 1], [1.2, 1], [8, 8], [8.2, 8]]
labels = KMeans(n_clusters=2, n_init=10, random_state=0).fit(X).labels_
print(sorted(np.bincount(labels).tolist()))
A) [1, 3] B) [2, 2] C) [4] D) [1, 1, 2]
MODERATE
Q13. SVM me C ko bohat bara karne se kya hota hai?
A) Margin chora aur model saada (underfit) B) Margin tang aur model train data ko ratta lagane ki taraf (overfit) C) Model tez ho jata hai D) Kuch nahi badalta
MODERATE
Q14. Sigmoid function ka z = 0 par output kya hota hai?
import numpy as np
print(1 / (1 + np.exp(0)))
A) 0.0 B) 0.5 C) 1.0 D) -1.0
MODERATE
Q15. Neeche me se kis algorithm ko features ki scaling ki zaroorat nahi hoti?
A) KNN B) SVM (rbf) C) PCA D) Random Forest
HARD
Q16. Do features bilkul correlated hain (y = 2x). PCA ka pehla component kitna phailao (variance) pakadta hai?
import numpy as np
from sklearn.decomposition import PCA
X = np.array([[1, 2], [2, 4], [3, 6], [4, 8]])
print(round(PCA(n_components=1).fit(X).explained_variance_ratio_[0], 2))
A) 0.50 B) 0.75 C) 1.0 D) 0.0
HARD
Q17. Ridge regression me alpha = 10 lagane se slope ka kya hota hai? (asli slope 2.0)
import numpy as np
from sklearn.linear_model import Ridge
X = np.array([[1], [2], [3], [4]])
y = [2, 4, 6, 8]
print(round(Ridge(alpha=10).fit(X, y).coef_[0], 2))
A) 2.0 B) 0.67 C) 10.0 D) 0.0
HARD
Q18. Probabilities [0.2, 0.45, 0.6, 0.9] par threshold 0.5 aur 0.4 se predictions kya hongi?
import numpy as np
probs = np.array([0.2, 0.45, 0.6, 0.9])
print((probs >= 0.5).astype(int).tolist(), (probs >= 0.4).astype(int).tolist())
A) [0, 0, 1, 1] [0, 1, 1, 1] B) [0, 1, 1, 1] [0, 0, 1, 1] C) [0, 0, 1, 1] [0, 0, 1, 1] D) [1, 1, 1, 1] [1, 1, 1, 1]
HARD
Q19. Ek dataset me 95% negative aur 5% positive hain. Model hamesha "negative" kehta hai. Accuracy aur recall kya hongi?
from sklearn.metrics import accuracy_score, recall_score
y_true = [0] * 95 + [1] * 5
y_pred = [0] * 100
print(accuracy_score(y_true, y_pred), recall_score(y_true, y_pred))
A) 0.95 0.95 B) 0.95 0.0 C) 0.5 0.0 D) 1.0 1.0
HARD
Q20. KNN me K = 1 aur K = 4 par train accuracy (chhote dataset par) kya hogi?
import numpy as np
from sklearn.neighbors import KNeighborsClassifier
X = np.array([[0], [1], [2], [3]])
y = [0, 1, 0, 1]
print(KNeighborsClassifier(n_neighbors=1).fit(X, y).score(X, y),
KNeighborsClassifier(n_neighbors=4).fit(X, y).score(X, y))
A) 1.0 1.0 B) 0.5 0.5 C) 1.0 0.5 D) 0.5 1.0
MODULE 4 TEST
Module 4 Answer Key (Jawabat aur wazahat)
| # | Level | Jawab | Wazahat |
|---|---|---|---|
| 1 | Easy | B) Features ke saath jawab (label / target) bhi | Supervised learning me har example ka sahi jawab (label) maujood hota hai, jis se model input se jawab tak ka rishta seekhta hai. |
| 2 | Easy | B) Model train par bohat achha magar naye (test) data par kharab hai | Overfitting me model training data ko ratta maar leta hai, isliye train score bohat zyada aur test score kam hota hai. |
| 3 | Easy | B) Clusters (groups) ki tadaad | K-Means me K wo clusters ki tadaad hai jo hum pehle se batate hain. (KNN me K padosiyon ki tadaad hota hai.) |
| 4 | Easy | B) Recall | Recall = TP / (TP + FN): jitne asli positive the un me se kitne pakde. Precision batata hai jo positive kaha wo kitne sahi the. |
| 5 | Easy | B) Features ki tadaad kam karna aur zyada se zyada maloomat bachana | PCA nayi axes (principal components) banata hai jo sab se zyada phailao pakadti hain, aur kam components rakh kar dimensions ghata deta hai. |
| 6 | Moderate | B) 40 10 | 50 rows me se 20% (10 rows) test me gaye, baqi 40 train me. |
| 7 | Moderate | B) 0.5 0.67 | TP = 2, FP = 2, FN = 1. Precision = 2 / (2 + 2) = 0.5. Recall = 2 / (2 + 1) = 0.67. |
| 8 | Moderate | C) 0.5 | Gini = 1 - (0.5<sup>2</sup>+ 0.5<sup>2</sup>) = 0.5, jo do classes ke liye sab se zyada (sab se mixed) impurity hai. |
| 9 | Moderate | B) [-1.22 0. 1.22] | Mean = 2, std (population) = 0.816. Z-score: (1-2)/0.816 = -1.22, 0, +1.22. |
| 10 | Moderate | B) Poore data par scaler fit karke phir train/test split karna | Poore data par fit karne se test data ki maloomat (mean, std) scaler me chali jati hai. Sahi tareeqa: split pehle, fit sirf train par. |
| 11 | Moderate | C) 5 | 5 folds me 5 alag train/test hote hain, isliye 5 scores milte hain. |
| 12 | Moderate | B) [2, 2] | Do points (1, 1) ke qareeb aur do (8, 8) ke qareeb hain, isliye K = 2 par har cluster me 2 points aate hain. |
| 13 | Moderate | B) Margin tang aur model train data ko ratta lagane ki taraf (overfit) | Bara C ghalti ko sakht saza deta hai, isliye margin tang ho jata hai aur model har training point sahi karne ki koshish me overfit ho sakta hai. |
| 14 | Moderate | B) 0.5 | sigmoid(0) = 1 / (1 + 1) = 0.5. Isliye logistic regression me decision boundary par probability 0.5 hoti hai. |
| 15 | Moderate | D) Random Forest | Trees aur forests sawal (threshold) par chalte hain, doori par nahi, isliye scaling ki zaroorat nahi. KNN, SVM aur PCA ko hoti hai. |
| 16 | Hard | C) 1.0 | Saray points ek seedhi lakeer par hain, to ek hi axis (PC1) poora phailao pakad leti hai: 100% (1.0). |
| 17 | Hard | B) 0.67 | Ridge coefficient ko chhota kar deta hai: slope = Sxy / (Sxx + alpha) = 10 / (5 + 10) = 0.67. Alpha barhane se slope aur chhota hota hai. |
| 18 | Hard | A) [0, 0, 1, 1] [0, 1, 1, 1] | Threshold kam karne par (0.4) 0.45 wala point bhi positive ho jata hai: recall barhta hai, magar jhoote alarm bhi. |
| 19 | Hard | B) 0.95 0.0 | Accuracy 95% hai magar ek bhi positive nahi pakda gaya (recall = 0). Ye Accuracy Paradox hai. |
| 20 | Hard | C) 1.0 0.5 | K = 1 par har point apna hi sab se qareebi padosi hai (100% ratta). K = 4 par saray points ka vote 2 - 2 barabar, to ek hi class milti hai aur sirf aadhe sahi (0.5). |
Tumhara score: ______ / 20