# 以下都是例行公事，直接拷贝即可
import pandas as pd
import numpy as np
# 导入matplotlib.pyplot绘图库，其中plt.plot()是最常用的绘图函数之一
import matplotlib.pyplot as plt
import seaborn as sns
sns.set() # 默认用seaborn的绘图样式
plt.rcParams["font.sans-serif"]=["SimHei"] #设置字体。如果不设置，中文会乱码。这里采用黑体SimHei，也可以考虑其他字体，如宋体SimSun等
plt.rcParams["axes.unicode_minus"]=False #该语句解决图像中的“-”负号的乱码问题
# 绘图使用'svg'后端：svg是矢量格式，可以任意缩放均保持清晰，各种屏幕的显示良好。
%config InlineBackend.figure_formats = ['svg']23 机器学习分类算法
23.1 无监督学习
23.1.1 k-means
23.2
np.random.seed(1000)
np.set_printoptions(suppress=True, precision=4) from sklearn.datasets import make_blobs
X, y = make_blobs(n_samples=250, centers=4,
random_state=500, cluster_std=1.25)
plt.figure(figsize=(8, 5))
plt.scatter(X[:, 0], X[:, 1], s=20); 
from sklearn.cluster import KMeansmodel = KMeans(n_clusters=4, random_state=0)
model.fit(X)
y_kmeans = model.predict(X)plt.figure(figsize=(8, 5))
plt.scatter(X[:, 0], X[:, 1], c=y_kmeans, cmap='coolwarm'); 
23.3 有监督学习
from sklearn.datasets import make_classification
n_samples = 100
X, y = make_classification(n_samples=n_samples, n_features=2,
n_informative=2, n_redundant=0,
n_repeated=0, random_state=250)
plt.figure(figsize=(8, 5))
plt.scatter(x=X[:, 0], y=X[:, 1], c=y, cmap='coolwarm'); 
23.3.1 高斯朴素贝叶斯
from sklearn.naive_bayes import GaussianNB
from sklearn.metrics import accuracy_score model = GaussianNB()
model.fit(X, y) GaussianNB()model.predict_proba(X).round(4)[:5]array([[0.0041, 0.9959],
[0.8534, 0.1466],
[0.9947, 0.0053],
[0.0182, 0.9818],
[0.5156, 0.4844]])pred = model.predict(X)
print(pred[:5])
accuracy_score(y, pred)[1 0 0 1 0]0.87Xc = X[y == pred]
Xf = X[y != pred]
plt.figure(figsize=(8, 5))
plt.scatter(x=Xc[:, 0], y=Xc[:, 1], c=y[y == pred],
marker='o', cmap='coolwarm')
plt.scatter(x=Xf[:, 0], y=Xf[:, 1], c=y[y != pred],
marker='x', cmap='coolwarm');
23.3.2 逻辑回归
from sklearn.linear_model import LogisticRegression
model = LogisticRegression(C=1, solver='lbfgs') model.fit(X,y)
pred = model.predict(X)
accuracy_score(y, pred) 0.9Xc = X[y == pred]
Xf = X[y != pred]
plt.figure(figsize=(8, 5))
plt.scatter(x=Xc[:, 0], y=Xc[:, 1], c=y[y == pred],
marker='o', cmap='coolwarm')
plt.scatter(x=Xf[:, 0], y=Xf[:, 1], c=y[y != pred],
marker='x', cmap='coolwarm');
23.3.3 决策树
from sklearn.tree import DecisionTreeClassifier
model = DecisionTreeClassifier(max_depth=4)model.fit(X, y)
pred = model.predict(X)
accuracy_score(y, pred) 0.97Xc = X[y == pred]
Xf = X[y != pred]
plt.figure(figsize=(8, 5))
plt.scatter(x=Xc[:, 0], y=Xc[:, 1], c=y[y == pred],
marker='o', cmap='coolwarm')
plt.scatter(x=Xf[:, 0], y=Xf[:, 1], c=y[y != pred],
marker='x', cmap='coolwarm');
23.3.4 深度神经网络
from sklearn.neural_network import MLPClassifier
model = MLPClassifier(solver='lbfgs', alpha=1e-5,
hidden_layer_sizes=2 * [75], random_state=10)%time model.fit(X, y) CPU times: user 1.49 s, sys: 38.3 ms, total: 1.52 s
Wall time: 195 ms/home/lee/anaconda3/lib/python3.9/site-packages/sklearn/neural_network/_multilayer_perceptron.py:549: ConvergenceWarning: lbfgs failed to converge (status=1):
STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.
Increase the number of iterations (max_iter) or scale the data as shown in:
https://scikit-learn.org/stable/modules/preprocessing.html
self.n_iter_ = _check_optimize_result("lbfgs", opt_res, self.max_iter)MLPClassifier(alpha=1e-05, hidden_layer_sizes=[75, 75], random_state=10,
solver='lbfgs')pred = model.predict(X)
accuracy_score(y, pred) 1.023.3.5 支持向量机
from sklearn.svm import SVC
from sklearn.model_selection import train_test_splittrain_x, test_x, train_y, test_y = train_test_split(X, y, test_size=0.33, random_state=0)
model = SVC(C=1, kernel='linear') model.fit(train_x, train_y)SVC(C=1, kernel='linear')pred_train = model.predict(train_x) accuracy_score(train_y, pred_train)0.9402985074626866