用神經網絡模型，預測紅酒質量

用神經網絡模型，預測紅酒質量；
后又用KNN\邏輯回歸\SVM模型試了試，准確率都差不多，神經網絡稍高。原始數據格式如下：

 

 最后輸出如下：

神經網絡模型的預測准確率是： 0.755
KNN模型的預測准確率是：0.7275
LogicRe模型的預測准確率是：0.7325
SVM模型的預測准確率是：0.7425

import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import MinMaxScaler
from sklearn.neural_network import MLPClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC
import seaborn as sns
import matplotlib.pyplot as plt

pd.set_option('expand_frame_repr', False)  # 當列太多時不換行
pd.set_option('display.max_rows', 500)  # 最多顯示數據的行數

file_path = './data/wine_quality.csv'

if __name__ == '__main__':
    data_df = pd.read_csv(file_path)
    all_cols = data_df.columns.tolist()    # 巧妙的取出了所有列名，並轉化為list
    feat_cols = all_cols[:-1]

    # 看看quality值各有多少個
    # sns.countplot(data_df['quality'])    # 這句跟下面一句等價
    # sns.countplot(data=data_df, x='quality')
    # plt.show()

    # 對quality列進行處理，原來若干種分類變為0、1兩種分類
    data_df.loc[data_df['quality'] <= 5,'quality'] = 0
    data_df.loc[data_df['quality'] >= 6,'quality'] = 1
    # sns.countplot(data=data_df, x='quality')
    # plt.show()

    X = data_df[feat_cols]
    y = data_df['quality']

    # 對特征值進行歸一化
    scaler = MinMaxScaler()
    X_process = scaler.fit_transform(X)

    X_train,X_test,y_train,y_test = train_test_split(X_process,y,test_size=0.25,random_state=10)

    # 神經網絡模型；隱藏層也不是越多越好;max_iter設置太小會有警告（沒達到最優），max_iter默認200；random_state設置后每次運行結果一樣。
    mlp_model = MLPClassifier(hidden_layer_sizes=(100,100),max_iter=1000,activation='relu',random_state=17)
    mlp_model.fit(X_train,y_train)
    accuracy = mlp_model.score(X_test,y_test)
    print('神經網絡模型的預測准確率是：',accuracy)

    # KNN\邏輯回歸\SVM模型試試
    model_dict = {
        'KNN': KNeighborsClassifier(n_neighbors=3),
        'LogicRe': LogisticRegression(C=1e3, solver='liblinear', multi_class='auto'),
        'SVM': SVC(C=1e3, gamma='auto')  # C值越小表示越強的正則化，也就是更弱復雜度;C值默認為1.0
    }
    for model_name,model in model_dict.items():
        model.fit(X_train,y_train)
        acc = model.score(X_test,y_test)
        print('{}模型的預測准確率是：{}'.format(model_name,acc))