scikit-learn: أول نموذج

1. import            ← استورد الخوارزمية
2. prepare X/y       ← جهّز البيانات
3. split             ← قسّم train/test
4. instantiate       ← أنشئ كائن النموذج
5. fit               ← درّب النموذج
6. predict           ← تنبّؤ
7. evaluate          ← قيّم الأداء

from sklearn.linear_model import LinearRegression
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier

import numpy as np

# X: 5 صفوف، 3 ميزات (مساحة، غرف، عمر)
X = np.array([
    [120, 3, 10],
    [200, 4,  5],
    [ 85, 2, 20],
    [300, 5,  2],
    [150, 3, 15],
])

# y: 5 أسعار
y = np.array([300, 550, 230, 800, 400])  # آلاف

import pandas as pd

df = pd.DataFrame({
    'area': [120, 200, 85, 300, 150],
    'rooms': [3, 4, 2, 5, 3],
    'price': [300, 550, 230, 800, 400]
})

X = df[['area', 'rooms']]    # DataFrame
y = df['price']              # Series

from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(
    X, y,
    test_size=0.2,
    random_state=42,      # للتكرار
    stratify=y            # للتصنيف فقط — يحافظ على نسبة الفئات
)

model = LinearRegression()
# أو
model = LogisticRegression(max_iter=1000)
# أو
model = RandomForestClassifier(n_estimators=100, max_depth=5, random_state=42)

model.fit(X_train, y_train)

ما يحدث داخل fit (LinearRegression):
- يحسب w (المعاملات) و b (التقاطع) عبر Normal Equation.
- يخزّنها في model.coef_ و model.intercept_.

ما يحدث داخل fit (RandomForestClassifier):
- يبني 100 شجرة قرار.
- كل شجرة تدرّب على bootstrap sample.
- يخزّن كل الأشجار.

y_pred = model.predict(X_test)

# LogisticRegression / RandomForestClassifier
y_proba = model.predict_proba(X_test)
# [[0.92, 0.08], [0.15, 0.85], ...] — احتمالات كل فئة

from sklearn.metrics import accuracy_score, classification_report

accuracy = accuracy_score(y_test, y_pred)
print(f"Accuracy: {accuracy:.3f}")
print(classification_report(y_test, y_pred))

from sklearn.metrics import mean_absolute_error, r2_score

mae = mean_absolute_error(y_test, y_pred)
r2 = r2_score(y_test, y_pred)
print(f"MAE: {mae:.2f}")
print(f"R²: {r2:.3f}")

# Classification: returns accuracy
print(model.score(X_test, y_test))   # 0.92

# Regression: returns R²
print(model.score(X_test, y_test))   # 0.85

from sklearn.linear_model import LinearRegression
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_absolute_error, r2_score
import numpy as np

# 1. تحضير
X = np.array([[50], [70], [85], [100], [120], [140], [160], [200]])
y = np.array([150, 200, 230, 270, 320, 370, 400, 480])

# 2. تقسيم
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.25, random_state=42
)

# 3. تدريب
model = LinearRegression()
model.fit(X_train, y_train)

# 4. تنبّؤ
y_pred = model.predict(X_test)

# 5. تقييم
mae = mean_absolute_error(y_test, y_pred)
r2 = r2_score(y_test, y_pred)
print(f"MAE: {mae:.2f}")
print(f"R²: {r2:.3f}")

from sklearn.ensemble import RandomForestClassifier
from sklearn.datasets import make_classification
from sklearn.model_selection import train_test_split
from sklearn.metrics import classification_report

# 1. بيانات تجريبية
X, y = make_classification(n_samples=300, n_features=8, random_state=42)

# 2. تقسيم
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.25, random_state=42, stratify=y
)

# 3. تدريب
model = RandomForestClassifier(n_estimators=100, max_depth=5, random_state=42)
model.fit(X_train, y_train)

# 4. تنبّؤ
y_pred = model.predict(X_test)

# 5. تقييم
print(classification_report(y_test, y_pred))