不同分类器对数据的处理

"""
    基于鸢尾花的不同分类器的效果比对:
        step1：
            准备数据；
            提取数据的特征向量X,Y
            将Y数据采用LabelEncoder转化为数值型数据;
        step2:
            将提取的特征向量X,Y进行拆分(训练集与测试集)
        step3:
            构建不同分类器并设置参数，例如：KNN,RF,SVM,deng；
        step4:
            循环遍历model列表：
                设置时间戳；
                每循环一个模型，进行模型训练，模型验证或测试打分并将结果进行记录；
        step5：plot(画图)
"""
import time
import warnings
import sys

import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import matplotlib.pylab as mpl
from sklearn.preprocessing import LabelEncoder       ###数据清洗
from sklearn.model_selection import train_test_split
from sklearn.svm import SVC, SVR
from sklearn.neighbors import KNeighborsClassifier   ###KNN
from sklearn.ensemble import RandomForestClassifier  ###随机森林
from sklearn.linear_model import LogisticRegression  ###逻辑回归器


# 溢出警告设置为0
warnings.filterwarnings('ignore')

# 防止中文乱码
mpl.rcParams['font.sans-serif'] = [u'simHei']
mpl.rcParams['axes.unicode_minus'] = 'False'

# 数据路径
path = 'iris.data'
names = ['A', 'B', 'C', 'D', 'cls']
# 创建读取数据对象
data = pd.read_csv(filepath_or_buffer=path, header=None, names=names)
print(data)
# print(data['cls'].values)
# sys.exit()
# 读取特征向量X, Y
X = data.iloc[:, :2]
# Y = data.iloc[:, :-1]
Y = data['cls']

# 将字符型数据Y转成数值型  数据清洗或调用LabelEncoder
class_label = {'Iris-setosa': 0, 'Iris-versicolor': 1, 'Iris-virginica': 2}
# 对目标属性做一个类别的转换，将字符串的数据转换为从0开始的int值
### 方式一: LabelEncoder方法
# label_encoder = LabelEncoder()
# Y = label_encoder.fit_transform(Y)
# print(Y)
### 方式二：采用map一一印射关系+匿名函数
data['cls'] = list(map(lambda cls: class_label[cls], data['cls'].values))
print(data['cls'].values)

# 数据拆分训练集和测试集
x_train, x_test, y_train, y_test = train_test_split(X, Y, test_size=0.2, random_state=10)

# 构建不同分类器
svc = SVC(C=0.2, kernel='rbf', decision_function_shape='ovr')
knn = KNeighborsClassifier(n_neighbors=5)
log = LogisticRegression()
rand = RandomForestClassifier(n_estimators=150, max_depth=3)
models = np.array([svc, knn, log, rand])

# 创建新列表记录数据
T = []
TRAIN_SORCE = []
TEST_SORCE = []

# 遍历不同分类器进行训练、打分和时间花销，并记录方便后续画图
for i in models:
    N = time.clock()
    i.fit(x_train, y_train)
    M = time.clock()
    T.append(M-N)
    TRAIN_SORCE.append(i.score(x_train, y_train))
    TEST_SORCE.append(i.score(x_test, y_test))

# 画图
plt.figure(num=1)
plt.plot(['svc01', 'knn02', 'log03', 'rand04'], TRAIN_SORCE, 'r-', linewidth='3', label='TRAIN_SORCE')
plt.plot(['svc01', 'knn02', 'log03', 'rand04'], TEST_SORCE, 'b-o', linewidth='3', label='TEST_SORCE')
# 设置Y轴尺度
plt.ylim(0.5, 1.2)

plt.figure(num=2)
plt.plot(['svc01', 'knn02', 'log03', 'rand04'], T, 'g-o', linewidth='3', label='time')
plt.show()


E:\myprogram\anaconda\envs\python3.6\python.exe E:/XXX/L-SVM/_differ-classifiies.py
       A    B    C    D             cls
0    5.1  3.5  1.4  0.2     Iris-setosa
1    4.9  3.0  1.4  0.2     Iris-setosa
2    4.7  3.2  1.3  0.2     Iris-setosa
3    4.6  3.1  1.5  0.2     Iris-setosa
4    5.0  3.6  1.4  0.2     Iris-setosa
..   ...  ...  ...  ...             ...
145  6.7  3.0  5.2  2.3  Iris-virginica
146  6.3  2.5  5.0  1.9  Iris-virginica
147  6.5  3.0  5.2  2.0  Iris-virginica
148  6.2  3.4  5.4  2.3  Iris-virginica
149  5.9  3.0  5.1  1.8  Iris-virginica

[150 rows x 5 columns]
[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]

Process finished with exit code 0