Пост import numpy as np

import numpy as np import pandas as pd import joblib from sklearn.model_selection import train_test_split from sklearn.compose import ColumnTransformer from sklearn.preprocessing import OneHotEncoder, StandardScaler from sklearn.pipeline import Pipeline from sklearn.linear_model import LinearRegression from sklearn.metrics import ( mean_absolute_error, mean_squared_error, r2_score, ) # -------------------------------------------------- # 1. Create a reproducible random number generator # -------------------------------------------------- rng = np.random.default_rng(42) n_samples = 300 brand_names = ["A", "B", "C", "D"] # -------------------------------------------------- # 2. Generate synthetic input features # -------------------------------------------------- mileage = rng.integers( low=5_000, high=220_001, size=n_samples, ) age = rng.integers( low=0, high=16, size=n_samples, ) engine_size = rng.choice( [1.3, 1.6, 2.0, 2.5], size=n_samples, ) brand = rng.choice( brand_names, size=n_samples, p=[0.30, 0.30, 0.25, 0.15], ) automatic = rng.choice( ["yes", "no"], size=n_samples, p=[0.65, 0.35], ) # -------------------------------------------------- # 3. Generate the synthetic target # -------------------------------------------------- brand_effect = { "A": 3_500, "B": 1_500, "C": 0, "D": -2_000, } brand_bonus = np.array( [brand_effect[current_brand] for current_brand in brand] ) automatic_bonus = np.where( automatic == "yes", 1_200, 0, ) noise = rng.normal( loc=0, scale=1_800, size=n_samples, ) price = ( 42_000 - 0.09 * mileage - 950 * age + 1_800 * engine_size + brand_bonus + automatic_bonus + noise ) price = np.maximum(price, 2_500) price = price.round(0) # -------------------------------------------------- # 4. Create the DataFrame # -------------------------------------------------- cars = pd.DataFrame( { "mileage": mileage, "age": age, "engine_size": engine_size, "brand": brand, "automatic": automatic, "price": price, } ) # -------------------------------------------------- # 5. Separate features and target # -------------------------------------------------- feature_columns = [ "mileage", "age", "engine_size", "brand", "automatic", ] X = cars[feature_columns].copy() y = cars["price"].copy() # -------------------------------------------------- # 6. Split into training and test sets # -------------------------------------------------- X_train, X_test, y_train, y_test = train_test_split( X, y, test_size=0.20, random_state=42, ) # -------------------------------------------------- # 7. Define feature groups # -------------------------------------------------- numeric_features = [ "mileage", "age", "engine_size", ] categorical_features = [ "brand", "automatic", ] # -------------------------------------------------- # 8. Create preprocessing pipelines # -------------------------------------------------- numeric_pipeline = Pipeline( steps=[ ( "scaler", StandardScaler(), ) ] ) categorical_pipeline = Pipeline( steps=[ ( "onehot", OneHotEncoder( drop="first", handle_unknown="ignore", sparse_output=False, ), ) ] ) # -------------------------------------------------- # 9. Apply transformations to selected columns # -------------------------------------------------- preprocessor = ColumnTransformer( transformers=[ ( "num", numeric_pipeline, numeric_features, ), ( "cat", categorical_pipeline, categorical_features, ), ], remainder="drop", verbose_feature_names_out=False, )