import pandas as pd ;import matplotlib.

pyplot as plt ;from

sklearn.model_selection import train_test_split; from
sklearn.preprocessing import PolynomialFeatures; import matplotlib.pyplot as plt ;from sklearn.datasets
from sklearn.linear_model import LinearRegression; mpg = import fetch_olivetti_faces ;from sklearn.model_selection
pd.read_csv(r"C:\Users\Downloads\auto-mpg.csv"); import train_test_split, cross_val_score ;
mpg['horsepower'] = pd.to_numeric(mpg['horsepower'], from sklearn.naive_bayes import GaussianNB ;from
errors='coerce'); sklearn.metrics import accuracy_score,
mpg.dropna(subset=['horsepower','mpg'], inplace=True); Xtr_m, classification_report, confusion_matrix ;X, y =
Xte_m, ytr_m, yte_m = train_test_split(mpg[['horsepower']], fetch_olivetti_faces(shuffle=True, random_state=42,
mpg['mpg'], test_size=0.2, random_state=42) return_X_y=True) ; Xtr, Xte, ytr, yte = train_test_split(X, y,
poly = PolynomialFeatures(2); Xtr_poly, Xte_poly = test_size=0.3, random_state=42) ;m = GaussianNB().fit(Xtr,
poly.fit_transform(Xtr_m), poly.transform(Xte_m) ypred_mpg = ytr); pred = m.predict(Xte);
LinearRegression().fit(Xtr_poly, ytr_m).predict(Xte_poly) ; print(f"Acc: {accuracy_score(yte,
housing = pred)*100:.2f}%");print("\nReport:\n",
pd.read_csv(r"C:\Users\Downloads\BostonHousing.csv"); Xtr_h, classification_report(yte, pred,
Xte_h, ytr_h, yte_h = train_test_split(housing[['rm']], zero_division=1));print("Confusion:\n",
housing['medv'], test_size=0.2, random_state=42) confusion_matrix(yte, pred));
ypred_h = LinearRegression().fit(Xtr_h, ytr_h).predict(Xte_h); print(f"\nCV Acc: {cross_val_score(m, X, y,
fig, axs = plt.subplots(1, 2, figsize=(12,5)); axs[0].scatter(Xte_m, cv=5).mean()*100:.2f}%");fig, axs = plt.subplots(3, 5,
yte_m, alpha=0.5, label='Actual'); axs[0].scatter(Xte_m, figsize=(12, 8)) ;for ax, img, t, p in zip(axs.ravel(), Xte, yte,
ypred_mpg, c='r', alpha=0.5, label='Predicted') ; pred): ;
axs[0].set(title='Auto MPG: Polynomial Regression', ax.imshow(img.reshape(64, 64), cmap='gray') ;
xlabel='Horsepower', ylabel='MPG'); axs[0].legend(); ax.set_title(f"Test:{t} Pred:{p}", fontsize=9) ; ax.axis('off');
axs[1].scatter(Xte_h, yte_h, alpha=0.5, label='Actual'); plt.tight_layout(); plt.show();
axs[1].plot(Xte_h, ypred_h, c='r', label='Linear Fit');
axs[1].set(title='Boston Housing: Linear Regression', xlabel='Avg
Rooms (rm)', ylabel='Median Value (medv)'); axs[1].legend() ;
plt.tight_layout(); plt.show()
 import pandas as pd; import matplotlib.pyplot as plt; import
seaborn as sns;from sklearn.datasets import
import numpy as np, matplotlib.pyplot as pltfrom load_breast_cancer; from sklearn.cluster import KMeans;
sklearn.neighbors import from sklearn.preprocessing import StandardScaler; from
KNeighborsClassifiernp.random.seed(42)X = sklearn.decomposition import PCA; from sklearn.metrics
np.random.rand(100,1)y = (X[:50,0]>0.5).astype(int)ks = import confusion_matrix, classification_report;
[1,2,3,4,5,20,30]plt.figure(figsize=(15,10))for i,k in X, y = load_breast_cancer(return_X_y=True); X_pca =
enumerate(ks,1): p = PCA(2).fit_transform(StandardScaler().fit_transform(X));clus
KNeighborsClassifier(k).fit(X[:50],y).predict(X[50:]) ters = KMeans(2, random_state=42).fit_predict(X_pca) ;
print(f'k={k}:', p) plt.subplot(3,3,i) plt.scatter(X[:50], print("Confusion Matrix:\n", confusion_matrix(y, clusters)) ;
np.zeros(50), c=y, cmap='coolwarm', edgecolors='k') print("\nClassification Report:\n", classification_report(y,
plt.scatter(X[50:], np.ones(50)*0.1, c=p, clusters)) ; df = pd.DataFrame(X_pca, columns=['PC1',
cmap='coolwarm', marker='x') plt.title(k); plt.yticks([]); 'PC2']);df['Cluster'], df['Label'] = clusters, y ; def plot(title,
plt.xlabel('x'); plt.grid() plt.tight_layout(); plt.show() hue, palette, centroids=False): ; plt.figure(figsize=(7,5));
sns.scatterplot(data=df, x='PC1', y='PC2', hue=hue,
palette=palette, s=80, edgecolor='k', alpha=0.7); if
centroids:;
plt.scatter(*KMeans(2).fit(X_pca).cluster_centers_.T,
c='red', s=200, marker='X', label='Centroids'); plt.title(title);
plt.legend(); plt.tight_layout(); plt.show(); plot("K-Means
Clustering", 'Cluster', 'Set1'); plot("True Labels", 'Label',
'coolwarm');
plot("K-Means with Centroids", 'Cluster', 'Set1',
centroids=True);