sklearn 简介
机器学习 Machine Learning
监督学习 supervised learning;
非监督学习 unsupervised learning;
半监督学习 semi-supervised learning;
强化学习 reinforcement learning;
遗传算法 genetic algorithm.
sklearn 安装
Anaconda安装
pip安装
sklearn 一般使用
选择学习方法
通用学习方式
sklearn 强大数据库
sklearn 常用属性与功能
选择学习方法
看图选方法
常用算法可分为四类:分类,回归,聚类,降维。
img1
通用学习方式
要点
创建数据
建立模型-训练-预测
要点:用KNN classifier
,对Iris数据集进行分类
创建数据:
from sklearn import datasetsfrom sklearn.model_selection import train_test_splitfrom sklearn.neighbors import KNeighborsClassifier iris = datasets.load_iris() iris_X = iris.data iris_y = iris.target print(iris_X[:2, :]) print(iris_y)""" [[ 5.1 3.5 1.4 0.2] [ 4.9 3. 1.4 0.2]] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2] """ X_train, X_test, y_train, y_test = train_test_split( iris_X, iris_y, test_size=0.3) print(y_train)""" [2 1 0 1 0 0 1 1 1 1 0 0 1 2 1 1 1 0 2 2 1 1 1 1 0 2 2 0 2 2 2 2 2 0 1 2 2 2 2 2 2 0 1 2 2 1 1 1 0 0 1 2 0 1 0 1 0 1 2 2 0 1 2 2 2 1 1 1 1 2 2 2 1 0 1 1 0 0 0 2 0 1 0 0 1 2 0 2 2 0 0 2 2 2 1 2 0 0 2 1 2 0 0 1 2] """
建立模型-训练-预测:
knn = KNeighborsClassifier() knn.fit(X_train, y_train) print(knn.predict(X_test)) print(y_test)""" [2 0 0 1 2 2 0 0 0 1 2 2 1 1 2 1 2 1 0 0 0 2 1 2 0 0 0 0 1 0 2 0 0 2 1 0 1 0 0 1 0 1 2 0 1] [2 0 0 1 2 1 0 0 0 1 2 2 1 1 2 1 2 1 0 0 0 2 1 2 0 0 0 0 1 0 2 0 0 2 1 0 1 0 0 1 0 1 2 0 1] """
sklearn 强大数据库
要点
导入数据-训练模型
创建虚拟数据-可视化
要点:使用 sklearn 读取数据库和生成虚拟的数据,例如用来训练线性回归模型的数据
img1
img2
sklearn.datasets.make_regression(n_samples=100, n_features=100, n_informative=10, n_targets=1, bias=0.0, effective_rank=None, tail_strength=0.5, noise=0.0, shuffle=True, coef=False, random_state=None)[source]
导入数据-训练模型:
from __future__ import print_functionfrom sklearn import datasetsfrom sklearn.linear_model import LinearRegressionimport matplotlib.pyplot as plt loaded_data = datasets.load_boston() data_X = loaded_data.data data_y = loaded_data.target model = LinearRegression() model.fit(data_X, data_y) print(model.predict(data_X[:4, :])) print(data_y[:4]) “”“ [ 30.00821269 25.0298606 30.5702317 28.60814055] [ 24. 21.6 34.7 33.4] ”“”
创建虚拟数据-可视化:
X, y = datasets.make_regression(n_samples=100, n_features=1, n_targets=1, noise=10) plt.scatter(X, y) plt.show()
img3
X, y = datasets.make_regression(n_samples=100, n_features=1, n_targets=1, noise=50) plt.scatter(X, y) plt.show()
img4
sklearn 常用属性与功能
model.coef_
:输出模型的斜率model.intercept_
:输出模型的截距(与y轴的交点)model.get_params()
:取出之前定义的参数model.score(data_X, data_y)
:对 Model 用 R^2 的方式进行打分
训练和预测:
from sklearn import datasetsfrom sklearn.linear_model import LinearRegression loaded_data = datasets.load_boston() data_X = loaded_data.data data_y = loaded_data.target model = LinearRegression() model.fit(data_X, data_y) print(model.predict(data_X[:4, :]))""" [ 30.00821269 25.0298606 30.5702317 28.60814055] """
参数和分数:
print(model.coef_) print(model.intercept_)""" [ -1.07170557e-01 4.63952195e-02 2.08602395e-02 2.68856140e+00 -1.77957587e+01 3.80475246e+00 7.51061703e-04 -1.47575880e+00 3.05655038e-01 -1.23293463e-02 -9.53463555e-01 9.39251272e-03 -5.25466633e-01] 36.4911032804 """print(model.get_params())""" {'copy_X': True, 'normalize': False, 'n_jobs': 1, 'fit_intercept': True} """print(model.score(data_X, data_y)) # R^2 coefficient of determination""" 0.740607742865 """
sklearn 高级使用
正规化 Normalization
检查神经网络(Evaluation)
交叉验证 1 Cross-validation
交叉验证 2 Cross-validation
交叉验证 3 Cross-validation
保存模型
正规化 Normalization
数据标准化
数据标准化对机器学习成效的影响
数据标准化:
from sklearn import preprocessing #标准化数据模块import numpy as np#建立Arraya = np.array([[10, 2.7, 3.6], [-100, 5, -2], [120, 20, 40]], dtype=np.float64)#将normalized后的a打印出print(preprocessing.scale(a))# [[ 0. -0.85170713 -0.55138018]# [-1.22474487 -0.55187146 -0.852133 ]# [ 1.22474487 1.40357859 1.40351318]]
数据标准化对机器学习成效的影响:
# 标准化数据模块from sklearn import preprocessing import numpy as np# 将资料分割成train与test的模块from sklearn.model_selection import train_test_split# 生成适合做classification资料的模块from sklearn.datasets.samples_generator import make_classification # Support Vector Machine中的Support Vector Classifierfrom sklearn.svm import SVC # 可视化数据的模块import matplotlib.pyplot as plt #生成具有2种属性的300笔数据X, y = make_classification( n_samples=300, n_features=2, n_redundant=0, n_informative=2, random_state=22, n_clusters_per_class=1, scale=100)#可视化数据plt.scatter(X[:, 0], X[:, 1], c=y) plt.show()
img1
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3) clf = SVC() clf.fit(X_train, y_train)print(clf.score(X_test, y_test))# 0.477777777778
img1
X = preprocessing.scale(X) X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3) clf = SVC() clf.fit(X_train, y_train)print(clf.score(X_test, y_test))# 0.9
检查神经网络(Evaluation)
Training and Test data
误差曲线
准确度曲线
正规化
交叉验证
Training and Test data:70% training,30% testing
img1
误差曲线:
img2
准确度曲线:
img3
正规化:
img4
交叉验证:
img5
交叉验证 1 Cross-validation
Model 基础验证法
Model 交叉验证法(Cross Validation)
以准确率(accuracy)判断
以平均方差(Mean squared error)
Model 基础验证法:
from sklearn.datasets import load_iris # iris数据集from sklearn.model_selection import train_test_split # 分割数据模块from sklearn.neighbors import KNeighborsClassifier # K最近邻(kNN,k-NearestNeighbor)分类算法#加载iris数据集iris = load_iris() X = iris.data y = iris.target#分割数据并X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=4)#建立模型knn = KNeighborsClassifier()#训练模型knn.fit(X_train, y_train)#将准确率打印出print(knn.score(X_test, y_test))# 0.973684210526
Model 交叉验证法(Cross Validation):
from sklearn.cross_validation import cross_val_score # K折交叉验证模块#使用K折交叉验证模块scores = cross_val_score(knn, X, y, cv=5, scoring='accuracy')#将5次的预测准确率打印出print(scores)# [ 0.96666667 1. 0.93333333 0.96666667 1. ]#将5次的预测准确平均率打印出print(scores.mean())# 0.973333333333
以准确率(accuracy)判断
import matplotlib.pyplot as plt #可视化模块#建立测试参数集k_range = range(1, 31) k_scores = []#藉由迭代的方式来计算不同参数对模型的影响,并返回交叉验证后的平均准确率for k in k_range: knn = KNeighborsClassifier(n_neighbors=k) scores = cross_val_score(knn, X, y, cv=10, scoring='accuracy') k_scores.append(scores.mean())#可视化数据plt.plot(k_range, k_scores) plt.xlabel('Value of K for KNN') plt.ylabel('Cross-Validated Accuracy') plt.show()
img1
以平均方差(Mean squared error):
import matplotlib.pyplot as plt k_range = range(1, 31) k_scores = []for k in k_range: knn = KNeighborsClassifier(n_neighbors=k) loss = -cross_val_score(knn, X, y, cv=10, scoring='mean_squared_error') k_scores.append(loss.mean()) plt.plot(k_range, k_scores) plt.xlabel('Value of K for KNN') plt.ylabel('Cross-Validated MSE') plt.show()
img2
交叉验证 2 Cross-validation
sklearn.learning_curve
: 检视过拟合
Learning curve 检视过拟合:
from sklearn.learning_curve import learning_curve #学习曲线模块from sklearn.datasets import load_digits #digits数据集from sklearn.svm import SVC #Support Vector Classifierimport matplotlib.pyplot as plt #可视化模块import numpy as np digits = load_digits() X = digits.data y = digits.target train_sizes, train_loss, test_loss = learning_curve( SVC(gamma=0.001), X, y, cv=10, scoring='mean_squared_error', train_sizes=[0.1, 0.25, 0.5, 0.75, 1])#平均每一轮所得到的平均方差(共5轮,分别为样本10%、25%、50%、75%、100%)train_loss_mean = -np.mean(train_loss, axis=1) test_loss_mean = -np.mean(test_loss, axis=1) plt.plot(train_sizes, train_loss_mean, 'o-', color="r", label="Training") plt.plot(train_sizes, test_loss_mean, 'o-', color="g", label="Cross-validation") plt.xlabel("Training examples") plt.ylabel("Loss") plt.legend(loc="best") plt.show()
img1
交叉验证 3 Cross-validation
sklearn.validation_curve
:检视过拟合
validation_curve 检视过拟合:
from sklearn.learning_curve import validation_curve #validation_curve模块from sklearn.datasets import load_digits from sklearn.svm import SVC import matplotlib.pyplot as plt import numpy as np#digits数据集digits = load_digits() X = digits.data y = digits.target#建立参数测试集param_range = np.logspace(-6, -2.3, 5)#使用validation_curve快速找出参数对模型的影响train_loss, test_loss = validation_curve( SVC(), X, y, param_name='gamma', param_range=param_range, cv=10, scoring='mean_squared_error')#平均每一轮的平均方差train_loss_mean = -np.mean(train_loss, axis=1) test_loss_mean = -np.mean(test_loss, axis=1)#可视化图形plt.plot(param_range, train_loss_mean, 'o-', color="r", label="Training") plt.plot(param_range, test_loss_mean, 'o-', color="g", label="Cross-validation") plt.xlabel("gamma") plt.ylabel("Loss") plt.legend(loc="best") plt.show()
img1
保存模型
使用 pickle 保存
使用 joblib 保存
使用 pickle 保存:
from sklearn import svmfrom sklearn import datasets clf = svm.SVC() iris = datasets.load_iris() X, y = iris.data, iris.target clf.fit(X,y)import pickle #pickle模块#保存Model(注:save文件夹要预先建立,否则会报错)with open('save/clf.pickle', 'wb') as f: pickle.dump(clf, f)#读取Modelwith open('save/clf.pickle', 'rb') as f: clf2 = pickle.load(f) #测试读取后的Model print(clf2.predict(X[0:1]))# [0]
使用 joblib 保存:
from sklearn.externals import joblib #jbolib模块#保存Model(注:save文件夹要预先建立,否则会报错)joblib.dump(clf, 'save/clf.pkl')#读取Modelclf3 = joblib.load('save/clf.pkl')#测试读取后的Modelprint(clf3.predict(X[0:1]))# [0]
作者:CrazyWolf_081c
链接:https://www.jianshu.com/p/74f059916e00
共同学习,写下你的评论
评论加载中...
作者其他优质文章