算法python实现_常用机器学习算法的python实现

﹏ヽ暗。殇╰゛Y 2022-10-28 12:29 158阅读 0赞

KNN

参数指定版

from sklearn.neighbors import KNeighborsClassifier

from sklearn.model\_selection import train\_test\_split

import pandas as pd

\#read data

data = pd.read\_csv('D:/column\_2C\_weka.csv')

\#n\_neighbors: K. In this example it is 3. it means that Look at the 3 closest labeled data points

knn = KNeighborsClassifier(n\_neighbors = 3)

\#x: features y: target variables

x,y = data.loc\[:,data.columns != 'class'\], data.loc\[:,'class'\]

\#split our data train and test sets

x\_train,x\_test,y\_train,y\_test = train\_test\_split(x,y,test\_size = 0.3,random\_state = 1)

knn.fit(x\_train,y\_train)   \#train the data

prediction = knn.predict(x\_test) \#predicts the data

print('Prediction: \{\}'.format(prediction))

print('With KNN (K=3) accuracy is: ',knn.score(x\_test,y\_test)) \# accuracy

调参版,其他算法的调参均可参照于此

from sklearn.model\_selection import GridSearchCV

grid = \{'n\_neighbors': np.arange(1,50)\}

knn = KNeighborsClassifier()

knn\_cv = GridSearchCV(knn, grid, cv=3) \# GridSearchCV

knn\_cv.fit(x,y)\# Fit

\# Print hyperparameter

print("Tuned hyperparameter k: \{\}".format(knn\_cv.best\_params\_))

print("Best score: \{\}".format(knn\_cv.best\_score\_))

数据切分

将数据集划分为训练集与测试集

\# train/test split

from sklearn.model\_selection import train\_test\_split

x,y = data.loc\[:,data.columns != 'class'\], data.loc\[:,'class'\]

\#test\_size:percentage of test size, train\_size = 1 - test\_size

\#random\_state: sets a seed. If this seed is same number,

\#train\_test\_split() produce exact same split at each time

x\_train,x\_test,y\_train,y\_test = train\_test\_split(x,y,test\_size = 0.3,random\_state = 1)

knn = KNeighborsClassifier(n\_neighbors = 3)

\#train with train\_set

knn.fit(x\_train,y\_train)

\#predict with test\_set

prediction = knn.predict(x\_test)

print('With KNN (K=3) accuracy is: ',knn.score(x\_test,y\_test)) \# accuracy

LR-线性回归

from sklearn.metrics import roc\_curve

from sklearn.metrics import confusion\_matrix, classification\_report

\# LinearRegression

from sklearn.linear\_model import LinearRegression

reg = LinearRegression()

\# Predict space

predict\_space = np.linspace(min(x), max(x)).reshape(-1,1)

\# Fit

reg.fit(x,y)

\# Predict

predicted = reg.predict(predict\_space)

\# R^2

print('R^2 score: ',reg.score(x, y))

y\_pred\_prob = reg.predict\_proba(x\_test)\[:,1\]

fpr, tpr, thresholds = roc\_curve(y\_test, y\_pred\_prob)

\# Plot ROC curve

plt.plot(\[0, 1\], \[0, 1\], 'k--')

plt.plot(fpr, tpr)

plt.xlabel('False Positive Rate')

plt.ylabel('True Positive Rate')

plt.title('ROC')

plt.show()

K fold CROSS VALIDATION-k fold CV-K折交叉验证

参数指定版

from sklearn.model\_selection import cross\_val\_score

reg = LinearRegression()

k = 5

cv\_result = cross\_val\_score(reg,x,y,cv=k) \# uses R^2 as score

print('CV Scores: ',cv\_result)

print('CV scores average: ',np.sum(cv\_result)/k)

调参版

\# grid search cross validation with 2 hyperparameter

\# 1. hyperparameter is C:logistic regression regularization parameter

\# 2. penalty l1 or l2

\# Hyperparameter grid

param\_grid = \{'C': np.logspace(-3, 3, 7), 'penalty': \['l1', 'l2'\]\}

x\_train, x\_test, y\_train, y\_test = train\_test\_split(x,y,

test\_size = 0.3,random\_state = 12)

logreg = LogisticRegression()

logreg\_cv = GridSearchCV(logreg,param\_grid,cv=3)

logreg\_cv.fit(x\_train,y\_train)

\# Print the optimal parameters and best score

print("Tuned hyperparameters : \{\}".format(logreg\_cv.best\_params\_))

print("Best Accuracy: \{\}".format(logreg\_cv.best\_score\_))

CROSS VALIDATION-正则化回归

损失函数正则化，分别采用L2和L1正则化

\# Ridge-L2

from sklearn.linear\_model import Ridge

x\_train,x\_test,y\_train,y\_test = train\_test\_split(x,y,random\_state = 2, test\_size = 0.3)

\# alpha-正则化项的系数

ridge = Ridge(alpha = 0.1, normalize = True)

ridge.fit(x\_train,y\_train)

ridge\_predict = ridge.predict(x\_test)

print('Ridge score: ',ridge.score(x\_test,y\_test))

\# Lasso-L1

from sklearn.linear\_model import Lasso

x = np.array(data1.loc\[:,\['pelvic\_incidence','pelvic\_tilt numeric','lumbar\_lordosis\_angle','pelvic\_radius'\]\])

x\_train,x\_test,y\_train,y\_test = train\_test\_split(x,y,random\_state = 3, test\_size = 0.3)

lasso = Lasso(alpha = 0.1, normalize = True)

lasso.fit(x\_train,y\_train)

ridge\_predict = lasso.predict(x\_test)

print('Lasso score: ',lasso.score(x\_test,y\_test))

print('Lasso coefficients: ',lasso.coef\_)

Random forest

\# Confusion matrix with random forest

from sklearn.metrics import classification\_report, confusion\_matrix

from sklearn.ensemble import RandomForestClassifier

x,y = data.loc\[:,data.columns != 'class'\], data.loc\[:,'class'\]

x\_train,x\_test,y\_train,y\_test = train\_test\_split(x,y,test\_size = 0.3,

random\_state = 1)

rf = RandomForestClassifier(random\_state = 4)

rf.fit(x\_train,y\_train)

y\_pred = rf.predict(x\_test)

cm = confusion\_matrix(y\_test,y\_pred)

print('Confusion matrix: \\n',cm)

print('Classification report: \\n',classification\_report(y\_test,y\_pred))

算法python实现_常用机器学习算法的python实现

发表评论取消回复

还没有评论，来说两句吧...

相关阅读

相关机器学习聚类算法 python实现

相关 Python机器学习算法如何实现

相关机器学习_感知器算法学习(Python实现)

相关 k近邻算法python_k近邻算法python实现 -- 《机器学习实战》

相关算法python实现_常用机器学习算法的python实现

相关 Python实现常用排序算法

相关 python3实现机器学习--KNN算法详解

相关 python机器学习实现K-近邻算法（KNN)

相关机器学习中线性回归算法python实现

相关常用排序算法的python实现