SVM va Naive Bayes

from sklearn.svm import SVC
from sklearn.datasets import load_breast_cancer
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.pipeline import make_pipeline

malumot = load_breast_cancer()
X, y = malumot.data, malumot.target

X_o, X_s, y_o, y_s = train_test_split(
    X, y, test_size=0.3, random_state=42, stratify=y
)

model = make_pipeline(StandardScaler(), SVC(kernel="linear"))
model.fit(X_o, y_o)

print(f"Aniqlik: {model.score(X_s, y_s):.3f}")
print(f"Qo'llab-quvvatlovchi vektorlar: {model.named_steps['svc'].n_support_}")

for yadro in ["linear", "poly", "rbf", "sigmoid"]:
    model = make_pipeline(StandardScaler(), SVC(kernel=yadro))
    model.fit(X_o, y_o)
    print(f"{yadro:8s}: {model.score(X_s, y_s):.3f}")

import numpy as np

print("     C   gamma   sinov")
for c in [0.1, 1, 10, 100]:
    for gamma in ["scale", 0.001, 0.01, 0.1]:
        model = make_pipeline(StandardScaler(), SVC(C=c, gamma=gamma))
        model.fit(X_o, y_o)
        print(f"{c:6.1f}  {str(gamma):>6}  {model.score(X_s, y_s):.3f}")

from sklearn.model_selection import GridSearchCV

tor = {
    "svc__C": [0.1, 1, 10, 100],
    "svc__gamma": ["scale", 0.001, 0.01, 0.1, 1],
}

qidiruv = GridSearchCV(
    make_pipeline(StandardScaler(), SVC()),
    tor, cv=5, n_jobs=-1,
)
qidiruv.fit(X_o, y_o)

print(f"Eng yaxshi: {qidiruv.best_params_}")
print(f"CV natija:  {qidiruv.best_score_:.3f}")
print(f"Sinov:      {qidiruv.score(X_s, y_s):.3f}")

model = make_pipeline(StandardScaler(), SVC(probability=True))
model.fit(X_o, y_o)

print(model.predict_proba(X_s[:3]).round(3))

from sklearn.svm import SVR

model = make_pipeline(StandardScaler(), SVR(kernel="rbf", C=100, epsilon=0.1))
model.fit(X_o, y_o)

from sklearn.naive_bayes import GaussianNB, MultinomialNB, BernoulliNB

model = GaussianNB()
model.fit(X_o, y_o)
print(f"Aniqlik: {model.score(X_s, y_s):.3f}")

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.naive_bayes import MultinomialNB
from sklearn.pipeline import make_pipeline

xatlar = [
    "Bepul chegirma! Hoziroq bosing va sovg'a yutib oling",
    "Ertangi yig'ilish soat 10 da bo'ladi",
    "Siz million so'm yutdingiz! Havolaga o'ting",
    "Hisobot fayli biriktirildi, ko'rib chiqing",
    "Aksiya! Faqat bugun 90 foiz chegirma",
    "Loyiha bo'yicha savollaringiz bormi?",
    "Kredit kartangiz bloklandi, ma'lumot kiriting",
    "Rahmat, hujjatlarni oldim",
]
belgilar = [1, 0, 1, 0, 1, 0, 1, 0]     # 1 - spam

model = make_pipeline(TfidfVectorizer(), MultinomialNB())
model.fit(xatlar, belgilar)

yangi = [
    "Bepul sovg'a yutib oling hoziroq",
    "Yig'ilish vaqti o'zgardi",
]

for matn, bashorat, ehtimol in zip(
    yangi, model.predict(yangi), model.predict_proba(yangi)[:, 1]
):
    natija = "SPAM" if bashorat else "oddiy"
    print(f"{natija:6s} ({ehtimol:.0%})  {matn}")

from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import RandomForestClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.model_selection import cross_val_score
import time

modellar = {
    "Logistik":      make_pipeline(StandardScaler(), LogisticRegression(max_iter=1000)),
    "KNN":           make_pipeline(StandardScaler(), KNeighborsClassifier()),
    "Qaror daraxti": DecisionTreeClassifier(max_depth=5, random_state=42),
    "Random Forest": RandomForestClassifier(n_estimators=100, random_state=42),
    "SVM (RBF)":     make_pipeline(StandardScaler(), SVC()),
    "Naive Bayes":   GaussianNB(),
}

print(f"{'Model':16s} {'CV aniqlik':>12} {'Vaqt (s)':>10}")

for nom, model in modellar.items():
    boshlandi = time.perf_counter()
    ballar = cross_val_score(model, X_o, y_o, cv=5)
    vaqt = time.perf_counter() - boshlandi

    print(f"{nom:16s} {ballar.mean():>10.3f}   {vaqt:>8.2f}")