案例-红酒品质分类

---

1. 数据集介绍

数据集共包含 11 个特征，共计 3269 条数据. 我们通过训练模型来预测红酒的品质, 品质共有 6 个各类别，分别使用数字: 0、1、2、3、4、5 来表示。

2. 案例实现

2.1 导入需要的库文件

import joblib
import numpy as np
import xgboost as xgb
import pandas as pd
import numpy as np
from collections import Counter
from sklearn.model_selection import train_test_split
from sklearn.metrics import classification_report
from sklearn.model_selection import StratifiedKFold

2.2 数据基本处理

def test01():

    # 1. 加载训练数据
    data = pd.read_csv('data/红酒品质分类.csv')
    x = data.iloc[:, :-1]
    y = data.iloc[:, -1] - 3

    # 2. 数据集分割
    x_train, x_valid, y_train, y_valid = train_test_split(x, y, test_size=0.2, stratify=y, random_state=22)

    # 3. 存储数据
    pd.concat([x_train, y_train], axis=1).to_csv('data/红酒品质分类-train.csv')
    pd.concat([x_valid, y_valid], axis=1).to_csv('data/红酒品质分类-valid.csv')

2.3 模型基本训练

def test02():

    # 1. 加载训练数据
    train_data = pd.read_csv('data/红酒品质分类-train.csv')
    valid_data = pd.read_csv('data/红酒品质分类-valid.csv')

    # 训练集
    x_train = train_data.iloc[:, :-1]
    y_train = train_data.iloc[:, -1]

    # 测试集
    x_valid = valid_data.iloc[:, :-1]
    y_valid = valid_data.iloc[:, -1]

    # 2. XGBoost模型训练
    estimator = xgb.XGBClassifier(n_estimators=100,
                                  objective='multi:softmax',
                                  eval_metric='merror',
                                  eta=0.1,
                                  use_label_encoder=False,
                                  random_state=22)
    estimator.fit(x_train, y_train)

    # 3. 模型评估
    y_pred = estimator.predict(x_valid)
    print(classification_report(y_true=y_valid, y_pred=y_pred))

    # 4. 模型保存
    joblib.dump(estimator, 'model/xgboost.pth')

2.4 模型参数调优

# 样本不均衡问题处理
from sklearn.utils import class_weight
classes_weights = class_weight.compute_sample_weight(class_weight='balanced',y=y_train)
# 训练的时候，指定样本的权重
estimator.fit(x_train, y_train,sample_weight = classes_weights)
y_pred = estimator.predict(x_valid)
print(classification_report(y_true=y_valid, y_pred=y_pred))

# 交叉验证，网格搜索
train_data = pd.read_csv('data/红酒品质分类-train.csv')
valid_data = pd.read_csv('data/红酒品质分类-valid.csv')

# 训练集
x_train = train_data.iloc[:, :-1]
y_train = train_data.iloc[:, -1]

# 测试集
x_valid = valid_data.iloc[:, :-1]
y_valid = valid_data.iloc[:, -1]

spliter = StratifiedKFold(n_splits=5, shuffle=True)
# 2. 定义超参数
param_grid = {'max_depth': np.arange(3, 5, 1),
              'n_estimators': np.arange(50, 150, 50),
              'eta': np.arange(0.1, 1, 0.3)}
estimator = xgb.XGBClassifier(n_estimators=100,
                              objective='multi:softmax',
                              eval_metric='merror',
                              eta=0.1,
                              use_label_encoder=False,
                              random_state=22)
cv = GridSearchCV(estimator,param_grid,cv=spliter)
y_pred = cv.predict(x_valid)
print(classification_report(y_true=y_valid, y_pred=y_pred))