import pandas as pd
import matplotlib.pyplot as plt
import matplotlib as mpl
import numpy as np
from imblearn.over_sampling import SMOTE

df  = pd.read_csv('cuisines.csv')
df.head()

df.cuisine.value_counts().plot.barh() #根据不同国家对数据集进行划分

thai_df = df[(df.cuisine == "thai")]#提取泰国美食
japanese_df = df[(df.cuisine == "japanese")]#提取日本美食
chinese_df = df[(df.cuisine == "chinese")]#提取中国美食
indian_df = df[(df.cuisine == "indian")]#提取印度美食
korean_df = df[(df.cuisine == "korean")]#提取韩国美食
print(f'thai df: {thai_df.shape}')#输出数据结构
print(f'japanese df: {japanese_df.shape}')#输出数据结构
print(f'chinese df: {chinese_df.shape}')#输出数据结构
print(f'indian df: {indian_df.shape}')#输出数据结构
print(f'korean df: {korean_df.shape}')#输出数据结构

def create_ingredient_df(df):
    ingredient_df = df.T.drop(['cuisine','Unnamed: 0']).sum(axis=1).to_frame('value')#从原始数据中删除无效列并统计axis=1的总和赋值给value列
    ingredient_df = ingredient_df[(ingredient_df.T != 0).any()]#提取所有非0值的样本
    ingredient_df = ingredient_df.sort_values(by='value', ascending=False,
    inplace=False)#按照数值的大小进行排序
    return ingredient_df#返回处理并排序后的结果

feature_df= df.drop(['cuisine','Unnamed: 0','rice','garlic','ginger'], axis=1)#删去最常见的这几列以平衡不同国家之间的混淆
labels_df = df.cuisine 

oversample = SMOTE()
transformed_feature_df, transformed_label_df = oversample.fit_resample(feature_df, labels_df)
print(f'new label count: {transformed_label_df.value_counts()}')
print(f'old label count: {df.cuisine.value_counts()}')

transformed_df.to_csv("cleaned_cuisines.csv")#保存至新文件方便下一次直接调用

from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split, cross_val_score
from sklearn.metrics import accuracy_score,precision_score,confusion_matrix,classification_report, precision_recall_curve
from sklearn.svm import SVC
import numpy as np
import pandas as pd
cuisines_df = pd.read_csv("cleaned_cuisines.csv")
print(cuisines_df.info())
cuisines_df.head()

cuisines_label_df = cuisines_df['cuisine'] #以cuisine列为预测标签
cuisines_feature_df = cuisines_df.drop(['Unnamed: 0', 'cuisine'], axis=1) #删去无用的列，取食材列为可训练特征
X_train, X_test, Y_train, Y_test = train_test_split(cuisines_feature_df, cuisines_label_df, test_size=0.3)#以7：3比例划分训练集与测试集

lr = LogisticRegression(multi_class='ovr',solver='liblinear')
model = lr.fit(X_train, np.ravel(Y_train))

accuracy = model.score(X_test, Y_test)
print ("Accuracy is {}".format(accuracy))
Y_pred = model.predict(X_test)
print(classification_report(Y_test,Y_pred))

print(f'ingredients: {X_test.iloc[50][X_test.iloc[50]!=0].keys()}')
print(f'cuisine: {Y_test.iloc[50]}')
test= X_test.iloc[50].values.reshape(-1, 1).T
proba = model.predict_proba(test)
classes = model.classes_
resultdf = pd.DataFrame(data=proba, columns=classes)
topPrediction = resultdf.T.sort_values(by=[0], ascending = [False])
topPrediction.head()

from sklearn.neighbors import KNeighborsClassifier
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC
from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier
from sklearn.model_selection import train_test_split, cross_val_score
from sklearn.metrics import accuracy_score,precision_score,confusion_matrix,classification_report, precision_recall_curve
import numpy as np

C = 10 #正则化参数设置为10
# Create different classifiers.
classifiers = {
    'Linear SVC': SVC(kernel='linear', C=C, probability=True,random_state=0)
}#创建分类器列表

n_classifiers = len(classifiers)

for index, (name, classifier) in enumerate(classifiers.items()):
    classifier.fit(X_train, np.ravel(y_train))

    y_pred = classifier.predict(X_test)
    accuracy = accuracy_score(y_test, y_pred)
    print("Accuracy (train) for %s: %0.1f%% " % (name, accuracy * 100))
    print(classification_report(y_test,y_pred))

'KNN classifier': KNeighborsClassifier(C),

'SVC': SVC(),

​       fantastic！精度上升到了82.4%。

'RFST': RandomForestClassifier(n_estimators=100), 
'ADA': AdaBoostClassifier(n_estimators=100)

​        很不戳！RFST随机森林的精度达到了83.4%，AdaBoost似乎不太好，不过这不重要。

import pickle
model_filename = 'cuisines.pkl'
pickle.dump(model, open(model_filename,'wb'))
model = pickle.load(open('cuisines.pkl','rb'))
print(model.predict([range(1,381)]))

​

<!DOCTYPE html>
<html>
  <head>
    <meta charset="UTF-8">
    <title>亚洲美食预测器</title>
    <link rel="stylesheet" href="{
     { url_for('static', filename='css/styles.css') }}">
  </head>

  <body>
    <div class="grid">

      <div class="box">

        <p>根据拥有的食材来预测UFO出现的国家！</p>
        <p>包含的食材填1，不包含的填0</p>
        <form action="{
     { url_for('predict')}}" method="post">
          <input type="number" name="cayenne" placeholder="cayenne" required="required" min="0" max="60" />
          <input type="number" name="onion" placeholder="onion" required="required" />
          <input type="number" name="pear" placeholder="pear" required="required" />
		  <input type="number" name="radish" placeholder="radish" required="required" />
		  <input type="number" name="sake" placeholder="sake" required="required" />
		  <input type="number" name="scallion" placeholder="scallion" required="required" />
		  <input type="number" name="sesame_oil" placeholder="sesame_oil" required="required" />
		  <input type="number" name="vegetable" placeholder="vegetable" required="required" />
		  <input type="number" name="vegetable_oil" placeholder="vegetable_oil" required="required" />
          <button type="submit" class="btn">预测</button>
        </form>

        <p>{
     { prediction_text }}</p>

      </div>

    </div>

  </body>
</html>

import numpy as np
from flask import Flask, request, render_template
import pickle

app = Flask(__name__)#初始化APP

model = pickle.load(open("cuisines.pkl", "rb"))#加载模型


@app.route("/")#装饰器
def home():
    return render_template("index.html")#先引入index.html，同时根据后面传入的参数，对html进行修改渲染。


@app.route("/predict", methods=["POST"])
def predict():

    int_features = [int(x) for x in request.form.values()]#存储用户输入的参数
    int_features = int_features+list(range(1,372)) #凑数
    final_features = [np.array(int_features)]#将用户输入的值转化为一个数组
    prediction = model.predict(final_features)#输入模型进行预测
    output = prediction[0]#将预测值传入output


    return render_template(
        "index.html", prediction_text="Likely country: {}".format(output)#将预测值返回到Web界面，使我们看到
    )


if __name__ == "__main__":
    app.run(debug=True)#调试模式下运行文件，实时反应结果。仅限测试使用，生产模式下不要使用

​

​​

​

​