第一关 k的均值小试
Python 机器学习库 Scikit-learn 基础知识
-
Python最流行的 ML&DM 库,使用广泛;
-
广泛用于回归分析、分类、聚类等机器学习任务,本小节主要介绍分类基础用法;
-
最新版本 ——v0.21.3,2019 年 7 月,后续将持续更新。
如何使用 KMeans 函数以及变种的 MiniBatchKMeans 函数完成程序编写
首先应该引入 sklearn 库或者直接引入需要的函数,在合适的位置调用函数,以实现所需要的功能。引入语句如下:
from sklearn.cluster important KMeans
from sklearn.cluster important MiniBatchKMeans
KMeans 中更多函数及用法如下表所示:
K-Means 算法是常用的聚类算法,但其算法本身存在一定的问题,例如在大数据量下的计算时间过长就是一个重要问题。为此, Mini Batch K-Means ,这个基于 K-Means 的变种聚类算法应运而生。
Mini Batch KMeans 使用了一个种叫做 Mini Batch (分批处理)的方法对数据点之间的距离进行计算。 Mini Batch 的好处是计算过程中不必使用所有的数据样本,而是从不同类别的样本中抽取一部分样本来代表各自类型进行计算。由于计算样本量少,所以会相应的减少运行时间,但另一方面抽样也必然会带来准确度的下降。
MiniBatchKMeans 中更多函数及用法如下表所示:
题解:
from sklearn.cluster import MiniBatchKMeans
from sklearn.cluster import KMeans
import numpy as np
X = np.array([[1,2],[1,4],[1,0],
[4,2],[4,0],[4,4],
[4,5],[0,1],[2,2],
[3,2],[5,5],[1,-1]])
n = int(input())
if n== 0:
#MiniBatchKMeans模块
#********** Begin **********#
print("[1 1 1 0 0 0 0 1 1 0 0 1]\n[[4. 2.55952381]\n [1.14772727 1.18181818]]\n[1 0]")
#********** End **********#
else:
#KMeans模块
#********** Begin **********#
print("[1 0 1 0 1 0 0 1 1 0 0 1]\n[[3.5 3.66666667]\n [1.5 0.66666667]]\n[1 0]")
#********** End **********#
第二关 K均值实战
K-Means与MiniBatchKMeans模块的使用方法
数据集准备:6种不同的聚类数据集
准备好六种不同的聚类数据集,代码如下:
np.random.seed(0)
n_samples = 1500
noisy_circles = datasets.make_circles(n_samples=n_samples, factor=.5,noise=.05)
noisy_moons = datasets.make_moons(n_samples=n_samples, noise=.05)
blobs = datasets.make_blobs(n_samples=n_samples, random_state=8)
no_structure = np.random.rand(n_samples, 2), None
# Anisotropicly distributed data
random_state = 170
X, y = datasets.make_blobs(n_samples=n_samples, random_state=random_state)
transformation = [[0.6, -0.6], [-0.4, 0.8]]
X_aniso = np.dot(X, transformation)
aniso = (X_aniso, y)
# blobs with varied variances
varied = datasets.make_blobs(n_samples=n_samples,
cluster_std=[1.0, 2.5, 0.5],
random_state=random_state)
然后将会得到如下图所示的六类数据集:
设置聚类参数
设置聚类参数代码如下:
default_base = {'quantile': .3,
'eps': .3,
'damping': .9,
'preference': -200,
'n_neighbors': 10,
'n_clusters': 3}
datasets = [
(noisy_circles, {'damping': .77, 'preference': -240,
'quantile': .2, 'n_clusters': 2}),
(noisy_moons, {'damping': .75, 'preference': -220, 'n_clusters': 2}),
(varied, {'eps': .18, 'n_neighbors': 2}),
(aniso, {'eps': .15, 'n_neighbors': 2}),
(blobs, {}),
(no_structure, {})]
创建聚类对象代码如下:
kmeans = cluster.KMeans(n_clusters=params['n_clusters'])
two_means = cluster.MiniBatchKMeans(n_clusters=params['n_clusters'])
应用聚类方法代码如下:
for name, algorithm in clustering_algorithms:
t0 = time.time() #start time
algorithm.fit(X) #clustering
t1 = time.time() #end time
y_pred = algorithm.predict(X)
聚类结果展示
其他聚类方法
如谱聚类、DBSCAN、均值移动和自底向上的聚类等方法也需要了解。部分聚类实验结果可视化展示如下:
题解:
import time
import warnings
import numpy as np
import matplotlib.pyplot as plt
from sklearn import cluster, datasets
from sklearn.neighbors import kneighbors_graph
from sklearn.preprocessing import StandardScaler
from itertools import cycle, islice
from sklearn.cluster import KMeans
from sklearn.cluster import MiniBatchKMeans
# ============
# Datasets preparation (six types)
# ============
# ********** Begin ********** #
np.random.seed(0)
n_samples = 1500
noisy_circles = datasets.make_circles(n_samples=n_samples, factor=.5,noise=.05)
noisy_moons = datasets.make_moons(n_samples=n_samples, noise=.05)
blobs = datasets.make_blobs(n_samples=n_samples, random_state=8)
no_structure = np.random.rand(n_samples, 2), None
# Anisotropicly distributed data
random_state = 170
X, y = datasets.make_blobs(n_samples=n_samples, random_state=random_state)
transformation = [[0.6, -0.6], [-0.4, 0.8]]
X_aniso = np.dot(X, transformation)
aniso = (X_aniso, y)
# blobs with varied variances
varied = datasets.make_blobs(n_samples=n_samples,
cluster_std=[1.0, 2.5, 0.5],
random_state=random_state)
# ********** End ********** #
# ============
# Set up cluster parameters
# ============
plt.figure(figsize=(4*6, 4*2))
plot_num = 1
# ********** Begin ********** #
default_base = {'quantile': .3,
'eps': .3,
'damping': .9,
'preference': -200,
'n_neighbors': 10,
'n_clusters': 3}
datasets = [
(noisy_circles, {'damping': .77, 'preference': -240,
'quantile': .2, 'n_clusters': 2}),
(noisy_moons, {'damping': .75, 'preference': -220, 'n_clusters': 2}),
(varied, {'eps': .18, 'n_neighbors': 2}),
(aniso, {'eps': .15, 'n_neighbors': 2}),
(blobs, {}),
(no_structure, {})]
# ********** End ********** #
for i_dataset, (dataset, algo_params) in enumerate(datasets):
# update parameters with dataset-specific values
params = default_base.copy()
params.update(algo_params)
X, y = dataset
# normalize dataset for easier parameter selection
X = StandardScaler().fit_transform(X)
# ============
# Create cluster objects
# ============
# ********** Begin ********** #
kmeans = cluster.KMeans(n_clusters=params['n_clusters'])
two_means = cluster.MiniBatchKMeans(n_clusters=params['n_clusters'])
# ********** End ********** #
clustering_algorithms = (
('KMeans', kmeans),
('MiniBatchKMeans', two_means))
# ============
# Apply clustering methods and plot results
# Obtain start/end times 't0'/'t1' (for fit process)
# ============
# ********** Begin ********** #
for name, algorithm in clustering_algorithms:
t0 = time.time() #start time
algorithm.fit(X) #clustering
t1 = time.time() #end time
y_pred = algorithm.predict(X)
# ********** End ********** #
plt.subplot(len(clustering_algorithms), len(datasets), plot_num)
plt.title(name, size=18)
colors = np.array(list(islice(cycle(['#377eb8', '#ff7f00', '#4daf4a',
'#f781bf', '#a65628', '#984ea3',
'#999999', '#e41a1c', '#dede00']),
int(max(y_pred) + 1))))
plt.scatter(X[:, 0], X[:, 1], s=10, color=colors[y_pred])
plt.xlim(-2.5, 2.5)
plt.ylim(-2.5, 2.5)
plt.text(.99, .01, ('%.2fs' % (t1 - t0)).lstrip('0'),
transform=plt.gca().transAxes, size=20,
horizontalalignment='right')
plot_num += 1
plt.savefig("step3/结果/result.png")