1. 概述
官网：scikit-learn：Machine Learning in Python
开源（Open source），可移植（reusable），基于NumPy、SciPy、matplotlib，用于数据挖掘（data mining）和数据分析（data analysis）的库。
目录：
（1）Classification 分类
Identifying to which category an object belongs to.
应用：虚假评价检测（Spam detection），图像识别（Image recognition）
算法：SVM，最近邻法（nearest neighbors），随机森林（random forest）
（2）Regression 回归
Predicting a continuous-valued attribute associated with an object.
应用：药物反应（Drug response），股票价格（Stock prices）
算法：支持向量回归（SVR），岭回归（ridge regression），lasso回归（Lasso）
（3）Clustering 聚类
Automatic grouping of similar objects into sets.
应用：顾客区分（Customer segmentation），分组实验结果（Grouping experiment outcomes）
算法：k均值（k-means），谱聚类（spectral clustering），均值偏移（mean-shift）
（4）Dimensionality reduction 降维
Reducing the number of random variables to consider.
应用：可视化（Visualization），增加效率（Increased efficiency）
算法：PCA，特征选择（feature selection），非负矩阵分解（non-negative matrix factorization）
（5）Model selection 模型选择
Comparing, validating and choosing parameters and models.
目标：通过改变参数提高准确度
模块：grid search 网格搜索，cross validation 交叉验证，metrics
（6）Preprocessing 预处理
Feature extraction and normalization. 特征提取及归一化
应用：Transforming input data such as text for use with machine learning algorithms.
模块：preprocessing 预处理，feature extortion 特征提取
2. 安装
conda install scikit-learn
or
pip install -U scikit-learn
注：本人使用conda安装tf成功，但是安装sklearn失败。遂使用pip安装sklearn，成功。
附1. conda更新或删除软件
conda update scikit-learn
conda remove scikit-learn
附2. 更新 pip
You are using pip version 9.0.1, however version 19.2.3 is available.
You should consider upgrading via the 'pip install --upgrade pip' command.
参考：Installing scikit-learn
3. 快速开始示例
（1）导入简单数据
from sklearn import datasetsiris = datasets.load_iris()digits = datasets.load_digits()print(digits.data)
[[ 0. 0. 5. ... 0. 0. 0.]
[ 0. 0. 0. ... 10. 0. 0.]
[ 0. 0. 0. ... 16. 9. 0.]
...
[ 0. 0. 1. ... 6. 0. 0.]
[ 0. 0. 2. ... 12. 0. 0.]
[ 0. 0. 10. ... 12. 1. 0.]]
array([0, 1, 2, ..., 8, 9, 8])
array([[ 0., 0., 5., 13., 9., 1., 0., 0.],
[ 0., 0., 13., 15., 10., 15., 5., 0.],
[ 0., 3., 15., 2., 0., 11., 8., 0.],
[ 0., 4., 12., 0., 0., 8., 8., 0.],
[ 0., 5., 8., 0., 0., 9., 8., 0.],
[ 0., 4., 11., 0., 1., 12., 7., 0.],
[ 0., 2., 14., 5., 10., 12., 0., 0.],
[ 0., 0., 6., 13., 10., 0., 0., 0.]])
（2）学习和预测
from sklearn import svmclf = svm.SVC(gamma=0.001, C=100.)clf.fit(digits.data[:-1], digits.target[:-1])
SVC(C=100.0, cache_size=200, class_weight=None, coef0=0.0,
decision_function_shape='ovr', degree=3, gamma=0.001, kernel='rbf',
max_iter=-1, probability=False, random_state=None, shrinking=True,
tol=0.001, verbose=False)
clf.predict(digits.data[-1:])
array([8])
（3）模型持久性
from sklearn import svmfrom sklearn import datasetsclf = svm.SVC(gamma='scale')iris = datasets.load_iris()X, y = iris.data, iris.targetclf.fit(X, y)
SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,
decision_function_shape='ovr', degree=3, gamma='scale', kernel='rbf',
max_iter=-1, probability=False, random_state=None, shrinking=True,
tol=0.001, verbose=False)
import pickle
s = pickle.dumps(clf)
clf2 = pickle.loads(s)
clf2.predict(X[0:1])
array([0])
0
from joblib import dump, load dump(clf, 'filename.joblib')
['filename.joblib']
clf = load('filename.joblib')
（4）Conventions
a. 类型转换 Type casting
import numpy as npfrom sklearn import random_projectionrng = np.random.RandomState(0)X = rng.rand(10, 2000)X = np.array(X, dtype='float32')X.dtype
dtype('float32')
transformer = random_projection.GaussianRandomProjection() X_new = transformer.fit_transform(X) X_new.dtype
dtype('float64')
以及
from sklearn import datasetsfrom sklearn.svm import SVCiris = datasets.load_iris()clf = SVC(gamma='scale')clf.fit(iris.data, iris.target)
SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,
decision_function_shape='ovr', degree=3, gamma='scale', kernel='rbf',
max_iter=-1, probability=False, random_state=None, shrinking=True,
tol=0.001, verbose=False)
list(clf.predict(iris.data[:3]))
[0, 0, 0]
clf.fit(iris.data, iris.target_names[iris.target])
SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,
decision_function_shape='ovr', degree=3, gamma='scale', kernel='rbf',
max_iter=-1, probability=False, random_state=None, shrinking=True,
tol=0.001, verbose=False)
list(clf.predict(iris.data[:3]))
['setosa', 'setosa', 'setosa']
b. 重拟合及更新参数
import numpy as np
from sklearn.datasets import load_iris
from sklearn.svm import SVC
X, y = load_iris(return_X_y=True)
clf = SVC()
clf.set_params(kernel='linear').fit(X, y)
SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,
decision_function_shape='ovr', degree=3, gamma='auto_deprecated',
kernel='linear', max_iter=-1, probability=False, random_state=None,
shrinking=True, tol=0.001, verbose=False)
array([0, 0, 0, 0, 0])
clf.set_params(kernel='rbf', gamma='scale').fit(X, y)
SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,
decision_function_shape='ovr', degree=3, gamma='scale', kernel='rbf',
max_iter=-1, probability=False, random_state=None, shrinking=True,
tol=0.001, verbose=False)
array([0, 0, 0, 0, 0])
c. 多类别 vs 多标签 拟合分类
from sklearn.svm import SVCfrom sklearn.multiclass import OneVsRestClassifierfrom sklearn.preprocessing import LabelBinarizerX = [[1, 2], [2, 4], [4, 5], [3, 2], [3, 1]]y = [0, 0, 1, 1, 2]classif = OneVsRestClassifier(estimator=SVC(gamma='scale', random_state=0))classif.fit(X, y).predict(X)
array([0, 0, 1, 1, 2])
y = LabelBinarizer().fit_transform(y)
classif.fit(X, y).predict(X)
array([[1, 0, 0],
[1, 0, 0],
[0, 1, 0],
[0, 0, 0],
[0, 0, 0]])
from sklearn.preprocessing import MultiLabelBinarizer
y = [[0, 1], [0, 2], [1, 3], [0, 2, 3], [2, 4]]
y = MultiLabelBinarizer().fit_transform(y)
classif.fit(X, y).predict(X)
array([[1, 1, 0, 0, 0],
[1, 0, 1, 0, 0],
[0, 1, 0, 1, 0],
[1, 0, 1, 0, 0],
[1, 0, 1, 0, 0]])
参考：An introduction to machine learning with scikit-learn
|
