# 忽略所有警告
import warnings
warnings.filterwarnings("ignore")

# 调用相应的模块
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
pd.set_option('display.max_columns', None)

# 读入数据
data = pd.read_csv("./input/risk_factors_cervical_cancer.csv")

print(data.shape)
data.head()

# 使用np.nan代替？
import numpy as np
data.replace({'?':np.nan},inplace=True)
data.head()

data.info()

# 删除掉缺失严重的两列
data.drop(['STDs: Time since first diagnosis','STDs: Time since last diagnosis'],axis=1,inplace=True)
data.shape

data[data['STDs:HPV'].isnull()].head()

data = data[~data['STDs:HPV'].isnull()]
data.isnull().sum().sort_values(ascending=False).head(15)

data['Number of sexual partners'].fillna(data['Number of sexual partners'].mode().values[0],inplace=True)
data['First sexual intercourse'].fillna(data['First sexual intercourse'].mode().values[0],inplace=True)
data['Num of pregnancies'].fillna(data['Num of pregnancies'].mode().values[0],inplace=True)

data.fillna('0',inplace=True)
print(data.shape)
data.info()

data.iloc[:,1:25] = data.iloc[:,1:25].astype(float).astype(int)
data.head()

data['target'] = data['Hinselmann'] + data['Schiller'] + data['Citology'] + data['Biopsy']
data.drop(['Hinselmann','Schiller','Citology','Biopsy'],axis = 1,inplace = True)

data['target'] = data['target'].apply(lambda x : 1 if x > 0 else 0)

data['target'].value_counts()

data.describe(include = 'all')

data.drop(['STDs:cervical condylomatosis','STDs:AIDS'],axis = 1,inplace = True)

fig,ax = plt.subplots(2,2,figsize = (15,7))

ax1 = plt.subplot(221)
sns.distplot(data['Age'],color = 'coral') 
ax1.set_title('Age Distribution')

ax2 = plt.subplot(222)
sns.distplot(data['First sexual intercourse'],color = 'coral') 
ax2.set_title('First sexual intercourse Distribution')

ax3 = plt.subplot(223)
plt.hist(data['Number of sexual partners'],color = 'slateblue') 
ax3.set_title('Number of sexual partners Distribution')

ax4 = plt.subplot(224)
plt.hist(data['Num of pregnancies'],color = 'slateblue') 
ax4.set_title('Num of pregnancies Distribution')

plt.tight_layout()
plt.show()  

sns.relplot(x='Age', y='First sexual intercourse', hue='target', data = data,aspect=2,height=6)
plt.show()

fig,[ax1,ax2] = plt.subplots(1,2,figsize=(15,5))

sns.countplot(x='Smokes', hue='target', data = data, palette="Set1",ax=ax1)
sns.countplot(x='STDs:HPV', hue='target', data = data, palette="Set1",ax=ax2)

plt.show()

from sklearn.preprocessing import scale
continuous_columns = ['Age','Number of sexual partners','First sexual intercourse','Num of pregnancies','Smokes (years)','Smokes (packs/year)','Hormonal Contraceptives (years)','IUD (years)','STDs (number)','STDs: Number of diagnosis']
data[continuous_columns] = scale(data[continuous_columns])

data.head()

data['target'].value_counts()

from sklearn import model_selection
x = data.drop(['target'],axis=1)
y = data['target']

x_train,x_test,y_train,y_test = model_selection.train_test_split(x, y,test_size=0.2,random_state = 33,stratify=y)

from sklearn.svm import SVC
from sklearn.metrics import accuracy_score,confusion_matrix,classification_report,recall_score

svm = SVC(kernel = 'rbf',class_weight='balanced',random_state = 4)
svm.fit(x_train,y_train)
y_pred = svm.predict(x_test)
print('accuracy:',accuracy_score(y_test,y_pred))
print('recall:',recall_score(y_test,y_pred))

conf_matrix = confusion_matrix(y_test,y_pred)

fig,ax = plt.subplots(figsize=(8,6))
sns.heatmap(conf_matrix,ax=ax,annot=True,annot_kws={'size':15}, fmt='d')
ax.set_ylabel('真实值')
ax.set_xlabel('预测值')
ax.set_title('混淆矩阵热力图')
plt.show()

import xgboost as xgb
from xgboost import XGBClassifier
xgb_model = xgb.XGBClassifier(max_depth = 8,class_weight='balanced',random_state = 4)

xgb_model.fit(x_train, y_train)
y_pred = xgb_model.predict(x_test)
print('accuracy:',accuracy_score(y_test,y_pred))
print('recall:',recall_score(y_test,y_pred))
conf_matrix = confusion_matrix(y_test,y_pred)

fig,ax = plt.subplots(figsize=(8,6))
sns.heatmap(conf_matrix,ax=ax,annot=True,annot_kws={'size':15}, fmt='d')
ax.set_ylabel('真实值')
ax.set_xlabel('预测值')
ax.set_title('混淆矩阵热力图')
plt.show()

from sklearn.svm import OneClassSVM

clf = OneClassSVM(nu=0.35, kernel="rbf")
y_pred = clf.fit(x).predict(x)

y_pred = np.where(y_pred==1, 0, y_pred) 
y_pred = np.where(y_pred==-1, 1, y_pred) 
print('accuracy:',accuracy_score(y,y_pred))
print('recall:',recall_score(y,y_pred))

conf_matrix = confusion_matrix(y_pred,y)

fig,ax = plt.subplots(figsize=(8,6))
sns.heatmap(conf_matrix,ax=ax,annot=True,annot_kws={'size':15}, fmt='d')
ax.set_ylabel('真实值')
ax.set_xlabel('预测值')
ax.set_title('混淆矩阵热力图')
plt.show()

from sklearn.ensemble import IsolationForest
clf = IsolationForest(n_estimators=120,max_features=6)
y_pred  = clf.fit(x).predict(x)

y_pred = np.where(y_pred==1, 0, y_pred) 
y_pred = np.where(y_pred==-1, 1, y_pred) 
print('accuracy:',accuracy_score(y,y_pred))
print('recall:',recall_score(y,y_pred))

conf_matrix = confusion_matrix(y_pred,y)

fig,ax = plt.subplots(figsize=(8,6))
sns.heatmap(conf_matrix,ax=ax,annot=True,annot_kws={'size':15}, fmt='d')
ax.set_ylabel('真实值')
ax.set_xlabel('预测值')
ax.set_title('混淆矩阵热力图')
plt.show()

from imblearn.over_sampling import SMOTE
x_train_resampled, y_train_resampled = SMOTE(random_state=4).fit_sample(x_train, y_train)

svm = SVC(kernel = 'linear',class_weight='balanced',random_state = 4)
svm.fit(x_train_resampled, y_train_resampled)
y_pred = svm.predict(x_test)
print('svm:','\n','accuracy:',accuracy_score(y_test,y_pred))
print('recall:',recall_score(y_test,y_pred))
conf_matrix = confusion_matrix(y_test,y_pred)

fig,ax = plt.subplots(figsize=(8,6))
sns.heatmap(conf_matrix,ax=ax,annot=True,annot_kws={'size':15}, fmt='d')
ax.set_ylabel('真实值')
ax.set_xlabel('预测值')
ax.set_title('混淆矩阵热力图')
plt.show()

xgb_model = xgb.XGBClassifier(random_state=42,max_depth = 8 )
xgb_model.fit(x_train_resampled, y_train_resampled)
y_pred = xgb_model.predict(np.array(x_test))

print('xgboost:','\n','accuracy:',accuracy_score(y_test,y_pred))
print('recall:',recall_score(y_test,y_pred))

conf_matrix = confusion_matrix(y_test,y_pred)

fig,ax = plt.subplots(figsize=(8,6))
sns.heatmap(conf_matrix,ax=ax,annot=True,annot_kws={'size':15}, fmt='d')
ax.set_ylabel('真实值')
ax.set_xlabel('预测值')
ax.set_title('混淆矩阵热力图')
plt.show()

from imblearn.over_sampling import ADASYN
x_train_resampled, y_train_resampled = ADASYN(random_state=4).fit_sample(x_train, y_train)

svm = SVC(kernel = 'linear',class_weight='balanced',random_state = 4)
svm.fit(x_train_resampled, y_train_resampled)
y_pred = svm.predict(x_test)
print('svm:','\n','accuracy:',accuracy_score(y_test,y_pred))
print('recall:',recall_score(y_test,y_pred))

conf_matrix = confusion_matrix(y_test,y_pred)

fig,ax = plt.subplots(figsize=(8,6))
sns.heatmap(conf_matrix,ax=ax,annot=True,annot_kws={'size':15}, fmt='d')
ax.set_ylabel('真实值')
ax.set_xlabel('预测值')
ax.set_title('混淆矩阵热力图')
plt.show()

xgb_model = xgb.XGBClassifier(random_state=42,max_depth = 8 )
xgb_model.fit(x_train_resampled, y_train_resampled)
y_pred = xgb_model.predict(np.array(x_test))
print('xgboost:','\n','accuracy:',accuracy_score(y_test,y_pred))
print('recall:',recall_score(y_test,y_pred))

conf_matrix = confusion_matrix(y_test,y_pred)

fig,ax = plt.subplots(figsize=(8,6))
sns.heatmap(conf_matrix,ax=ax,annot=True,annot_kws={'size':15}, fmt='d')
ax.set_ylabel('真实值')
ax.set_xlabel('预测值')
ax.set_title('混淆矩阵热力图')
plt.show()