数据EDA

# 导入数据
import pandas as pd
import numpy as np

import plotly_express as px
from matplotlib import pyplot as plt
import seaborn as sns

import warnings
warnings.filterwarnings('ignore')

data = pd.read_csv("mushrooms.csv")
data.shape

image.png

data.columns

image.png

# 失值
data.isnull().sum()

image.png

# 有无毒对比
data["class"].value_counts()

image.png

可视化分析

cap = data["cap-color"].value_counts().reset_index()
cap.columns = ["color","number"]
cap

image.png

fig = px.bar(cap,x="color",
            y="number",
            color="number",
            text="number",
            color_continuous_scale="rainbow")

fig.show()

image.png

cap_class = data.groupby(["class","cap-color"]).size().reset_index()
cap_class.columns = ["class","color","number"]
cap_class.head()

image.png

fig = px.bar(cap_class,x="color",
            y="number",
            color="class",
            text="number",
            barmode="group")

fig.show()

image.png

# 菌的气味
odor = data["odor"].value_counts().reset_index()
odor.columns = ["odor","number"]
odor

image.png

fig = px.bar(odor,
            x="odor",
            y="number",
            color="number",
            text="number",
            color_continuous_scale="rainbow")

fig.show()

image.png

odor_class = data.groupby(["class","odor"]).size().reset_index()
odor_class.columns = ["class","odor","number"]
odor_class.head()

image.png

fig = px.bar(odor_class,
            x="odor",
            y="number",
            color="class",
            text="number",
            barmode="group")

fig.show()

image.png

特征工程

# 特征转换
data.head()

image.png

from sklearn.preprocessing import LabelEncoder

labelencoder = LabelEncoder()

for col in data.columns:
    data[col] = labelencoder.fit_transform(data[col])
    
data.head()

image.png

data["stalk-color-above-ring"].unique()

image.png

data.groupby("class").size()

image.png

# 数据分布
data["stalk-color-above-ring"].value_counts()

image.png

ax = sns.boxplot(x='class',
                y='stalk-color-above-ring',
                data=data)
ax = sns.stripplot(x='class',
                  y='stalk-color-above-ring',
                  data=data,
                  jitter=True,
                  edgecolor="gray")
plt.title("Class w.r.t stalkcolor above ring",fontsize=12)

plt.show()

image.png

# 分离特征和标签
x = data.iloc[:,1:23]
y = data.iloc[:,0]

# 数据标准化
from sklearn.preprocessing import StandardScaler

scaler = StandardScaler()
X = scaler.fit_transform(x)
X

image.png

特征相关性

corr = data.corr()
sns.heatmap(corr)

plt.show()

image.png

主成分分析PCA

# PCA过程
from sklearn.decomposition import PCA

pca = PCA()
pca.fit_transform(X)

covariance = pca.get_covariance()

explained_variance = pca.explained_variance_
explained_variance

image.png

with plt.style.context("dark_background"):
    plt.figure(figsize=(6,4))
    
    plt.bar(range(22),
           explained_variance,
           alpha=0.5,
           align="center",
           label="individual explained variance")
    
    plt.ylabel('Explained variance ratio')
    plt.xlabel('Principal components')
    plt.legend(loc="best")
    plt.tight_layout()

image.png

# 2个主成分下的原始数据分布
N = data.values
pca = PCA(n_components=2)
x = pca.fit_transform(N)

plt.figure(figsize=(5,5))
plt.scatter(x[:,0],x[:,1])

plt.show()

image.png

from sklearn.cluster import KMeans

km = KMeans(n_clusters=2,random_state=5)

N = data.values
X_clustered = km.fit_predict(N)

label_color_map = {0:"g",1:"y"}
label_color = [label_color_map[l] for l in X_clustered]

plt.figure(figsize=(5,5))
plt.scatter(x[:,0],x[:,1],c=label_color)

plt.show()

image.png

# 基于17主成分下的建模
pca_modified = PCA(n_components=17)
pca_modified.fit_transform(X)

image.png

from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(X,y,test_size=0.2,random_state=4)

模型1：逻辑回归

from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import cross_val_score
from sklearn import metrics

model_LR = LogisticRegression()
model_LR.fit(X_train,y_train)

image.png

y_prob = model_LR.predict_proba(X_test)[:,1]
y_prob

image.png

y_pred = np.where(y_prob>0.5,1,0)
y_pred

image.png

model_LR.score(X_test,y_pred)

image.png

confusion_matrix = metrics.confusion_matrix(y_test,y_pred)
confusion_matrix

image.png

auc_roc = metrics.roc_auc_score(y_test,y_pred)
auc_roc

image.png

# 真假阳性
from sklearn.metrics import roc_curve,auc

false_positive_rate, true_positive_rate, thresholds = roc_curve(y_test,y_prob)
roc_auc = auc(false_positive_rate,true_positive_rate)
roc_auc

image.png

# ROC曲线
import matplotlib.pyplot as plt

plt.figure(figsize=(10,10))
plt.title("ROC")
plt.plot(false_positive_rate,true_positive_rate,color="red",label="AUC=%0.2f"%roc_auc)

plt.legend(loc="lower right")
plt.plot([0,1],[0,1],linestyle="--")
plt.axis("tight")

plt.ylabel("True Positive Rate")
plt.xlabel("False Positive Rate")

plt.show()

image.png

from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import cross_val_score
from sklearn import metrics

LR_model = LogisticRegression()
tuned_parameters = {"C":[0.001,0.01,0.1,1,10,100,1000],
                   "penalty":['l1','l2']}

from sklearn.model_selection import GridSearchCV

LR = GridSearchCV(LR_model,tuned_parameters,cv=10)
LR.fit(X_train,y_train)

print(LR.best_params_)

image.png

y_prob = LR.predict_proba(X_test)[:,1]
y_pred = np.where(y_prob>0.5,1,0)
LR.score(X_test,y_pred)

image.png

confusion_matrix = metrics.confusion_matrix(y_test,y_pred)
confusion_matrix

image.png

auc_roc = metrics.classification_report(y_test,y_pred)
print(auc_roc)

image.png

auc_roc = metrics.roc_auc_score(y_test,y_pred)
auc_roc

image.png

# ROC曲线情况
from sklearn.metrics import roc_curve,auc

false_positive_rate,true_positive_rate,thresholds = roc_curve(y_test,y_prob)

import matplotlib.pyplot as plt

plt.figure(figsize=(10,10))
plt.title("ROC")
plt.plot(false_positive_rate,true_positive_rate,color="red",label="AUC = %0.2f"%roc_auc)

plt.legend(loc="lower right")
plt.plot([0,1],[0,1],linestyle="--")
plt.axis("tight")
plt.ylabel("True Positive Rate")
plt.xlabel("False Positive Rate")

plt.show()

image.png

模型2：高斯朴素贝叶斯

from sklearn.naive_bayes import GaussianNB

model_naive = GaussianNB()
model_naive.fit(X_train,y_train)

y_prob = model_naive.predict_proba(X_test)[:,1]
y_pred = np.where(y_prob>0.5,1,0)
model_naive.score(X_test,y_pred)

image.png

print(f"Number of mislabeled points from {X_test.shape[0]} points:{(y_test!=y_pred).sum()}")

image.png

# 交叉验证
scores = cross_val_score(model_naive,X,y,cv=10,scoring="accuracy")
scores

image.png

scores.mean()

image.png

# 混淆矩阵和AUC
confusion_matrix = metrics.confusion_matrix(y_test,y_pred)
confusion_matrix

image.png

auc_roc = metrics.classification_report(y_test,y_pred)
print(auc_roc)

image.png

# 真假阳性
from sklearn.metrics import roc_curve,auc

false_positive_rate,true_positive_rate,thresholds = roc_curve(y_test,y_prob)
roc_auc = auc(false_positive_rate,true_positive_rate)
roc_auc

image.png

# ROC曲线
import matplotlib.pyplot as plt

plt.figure(figsize=(10,10))
plt.title("ROC")
plt.plot(false_positive_rate,true_positive_rate,color="red",label="AUC = %0.2f"%roc_auc)

plt.legend(loc="lower right")
plt.plot([0,1],[0,1],linestyle='--')
plt.axis("tight")
plt.xlabel("False Positive Rate")
plt.ylabel("True Positive Rate")

plt.show()

image.png

模型3：支持向量机SVM

# 建模过程
from sklearn.svm import SVC

svm_model = SVC()
tuned_parameters = {
    'C':[1,10,100,500,1000],
    'kernel':['linear','rbf'],
    'C':[1,10,100,500,1000],
    'gamma':[1,0.1,0.01,0.001,0.0001],
    'kernel':['rbf']
}

# 随机网络搜索-RandomizedSearchCV
from sklearn.model_selection import RandomizedSearchCV

model_svm = RandomizedSearchCV(svm_model,
                              tuned_parameters,
                              cv=10,
                              scoring="accuracy",
                              n_iter=20)
model_svm.fit(X_train,y_train)

image.png

print(model_svm.best_score_)

image.png

model_svm.best_params_

image.png

y_pred = model_svm.predict(X_test)
metrics.accuracy_score(y_pred,y_test)

image.png

# 混淆矩阵
metrics.confusion_matrix(y_test,y_pred)

image.png

print(metrics.classification_report(y_test,y_pred))

image.png

metrics.roc_auc_score(y_test,y_pred)

image.png

# ROC曲线
from sklearn.metrics import roc_curve,auc

false_positive_rate,true_positive_rate,thresholds = roc_curve(y_test,y_pred)
roc_auc = auc(false_positive_rate,true_positive_rate)

import matplotlib.pyplot as plt

plt.figure(figsize=(10,10))
plt.title("ROC")
plt.plot(false_positive_rate,true_positive_rate,color="red",label="AUC = %0.2f"%roc_auc)

plt.legend(loc="lower right")
plt.plot([0,1],[0,1],linestyle="--")
plt.axis("tight")
plt.ylabel("True Positive Rate")
plt.xlabel("False Positive Rate")

plt.show()

image.png

模型4：随机森林

# 建模拟合
from sklearn.ensemble import RandomForestClassifier

model_RR = RandomForestClassifier()
model_RR.fit(X_train,y_train)

image.png

# 预测得分
y_prob = model_RR.predict_proba(X_test)[:,1]
y_prob

image.png

y_pred = np.where(y_prob>0.5,1,0)
model_RR.score(X_test,y_pred)

image.png

# 混淆矩阵
metrics.confusion_matrix(y_test,y_pred)

image.png

print(metrics.classification_report(y_test,y_pred))

image.png

metrics.roc_auc_score(y_test,y_pred)

image.png

# ROC曲线
from sklearn.metrics import roc_curve,auc

false_positive_rate,true_positive_rate,thresholds = roc_curve(y_test,y_prob)
roc_auc = auc(false_positive_rate,true_positive_rate)
roc_auc

import matplotlib.pyplot as plt

plt.figure(figsize=(10,10))
plt.title("ROC")
plt.plot(false_positive_rate,true_positive_rate,color="red",label="AUC = %0.2f"%roc_auc)

plt.legend(loc="lower right")
plt.plot([0,1],[0,1],linestyle="--")
plt.axis("tight")
plt.ylabel("True Positive Rate")
plt.xlabel("False Positive Rate")

plt.show()

image.png

模型5：决策树(CART)

# 建模
from sklearn.tree import DecisionTreeClassifier

model_tree = DecisionTreeClassifier()
model_tree.fit(X_train,y_train)
y_prob = model_tree.predict_proba(X_test)[:,1]
y_pred = np.where(y_prob>0.5,1,0)
model_tree.score(X_test,y_pred)

image.png

# 混淆矩阵
metrics.confusion_matrix(y_test,y_pred)

image.png

print(metrics.classification_report(y_test,y_pred))

image.png

metrics.roc_auc_score(y_test,y_pred)

image.png

# ROC曲线
from sklearn.metrics import roc_curve,auc

false_positive_rate,true_positive_rate,thresholds = roc_curve(y_test,y_prob)
roc_auc = auc(false_positive_rate,true_positive_rate)
roc_auc

import matplotlib.pyplot as plt

plt.figure(figsize=(10,10))
plt.title("ROC")
plt.plot(false_positive_rate,true_positive_rate,color="red",label="AUC = %0.2f"%roc_auc)

plt.legend(loc="lower right")
plt.plot([0,1],[0,1],linestyle="--")
plt.axis("tight")
plt.xlabel("False Positive Rate")
plt.ylabel("True Positive Rate")

plt.show()

image.png

模型6：神经网络ANN

# 建模
from sklearn.neural_network import MLPClassifier

mlp = MLPClassifier()
mlp.fit(X_train,y_train)

image.png

y_prob = mlp.predict_proba(X_test)[:,1]
y_pred = np.where(y_prob>0.5,1,0)
mlp.score(X_test,y_pred)

image.png

# 混淆矩阵
metrics.confusion_matrix(y_test,y_pred)

image.png

print(metrics.classification_report(y_test,y_pred))

image.png

metrics.roc_auc_score(y_test,y_pred)

image.png

# ROC曲线
from sklearn.metrics import roc_curve,auc

false_positive_rate,true_positive_rate,thresholds = roc_curve(y_test,y_prob)
roc_auc = auc(false_positive_rate,true_positive_rate)
roc_auc

import matplotlib.pyplot as plt

plt.figure(figsize=(10,10))
plt.title("ROC")
plt.plot(false_positive_rate,true_positive_rate,color="red",label="AUC = %0.2f"%roc_auc)

plt.legend(loc="lower right")
plt.plot([0,1],[0,1],linestyle="--")
plt.axis("tight")
plt.ylabel("True Positive Rate")
plt.xlabel("False Positive Rate")

plt.show()

image.png

# 网格搜索
from sklearn.neural_network import MLPClassifier

mlp_model = MLPClassifier()

tuned_parameters = {'hidden_layer_sizes':range(1,200,10),
                   'activation':['tanh','logistic','relu'],
                   'alpha':[0.0001,0.001,0.01,0.1,1,10],
                   'max_iter':range(50,200,50)}
model_mlp = RandomizedSearchCV(mlp_model,tuned_parameters,cv=10,scoring='accuracy',n_iter=5,n_jobs=-1,random_state=5)
model_mlp.fit(X_train,y_train)

image.png

# 模型属性
model_mlp.best_score_

image.png

model_mlp.best_params_

image.png

model_svm.cv_results_

image.png

# ROC曲线
from sklearn.metrics import roc_curve,auc

false_positive_rate,true_positive_rate,thresholds = roc_curve(y_test,y_prob)
roc_auc = auc(false_positive_rate,true_positive_rate)
roc_auc

import matplotlib.pyplot as plt

plt.figure(figsize=(10,10))
plt.title("ROC")
plt.plot(false_positive_rate,true_positive_rate,color="red",label="AUC = %0.2f"%roc_auc)

plt.legend(loc="lower right")
plt.plot([0,1],[0,1],linestyle="--")
plt.axis("tight")
plt.xlabel("False Positive Rate")
plt.ylabel("True Positive Rate")

plt.show()

image.png