pythonがsvmを実現して、トビの花を分類します(トビの尾の花のデータがあります).
7719 ワード
トビの花の分類トビの花データリンク:http://bj.bcebos.com/v1/ai-studio-online/93e8a07d6624465c943f60a0b4ec5fd959d44b5e5453410a8b2452ed3720c32f?responseContentDisposition=attachment%3B%20filename%3Diris.data&authorization=bce-auth-v 1%2 F 0 ef 6765 c 1 e 4918 bc 0 d 3 c 3 c 3 c 6 d 1%2 F 2010 8-12-12 T 14%3 A 57%3 A 54 Z%2 F-1%2 F 2 cbe 86672 d 2 d 2 d 44278 cc 3 f 7678930700 c 5 aeffa 85569803 fd 6 e 7 d 625 b 43ca 2方法一
方法2
import numpy as np
from sklearn import model_selection as mo
from sklearn import svm
import matplotlib.pyplot as plt
from matplotlib import colors
import matplotlib as mpl
def iris_type(s):
# ， string int
it = {b'Iris-setosa': 0, b'Iris-versicolor': 1, b'Iris-virginica': 2}
return it[s]
data = np.loadtxt(r'D:\PycharmProjects\untitled\ \iris.data', dtype=float, delimiter=',', converters={4:iris_type})
'''
def loadtxt(fname, dtype=float, comments='#', delimiter=None,
converters=None, skiprows=0, usecols=None, unpack=False,
ndmin=0, encoding='bytes', max_rows=None):
'''
x, y = np.split(data, (4, ), axis=1)
x_train, x_test, y_train, y_test = mo.train_test_split(x, y, random_state=1, test_size=0.3)
'''
train_data：
train_target：
test_size： ， 0-1 ， ；
random_state： 。
： ， ， 。 1， 。 0 ， 。
， ：
， ； ， 。
'''
clf = svm.SVC(C=0.5, kernel='linear', decision_function_shape='ovr')
clf.fit(x_train, y_train, sample_weight=None)
print(x_train.shape)
#print(x_train)
#print(y_train)
#print(x_test)
#print(y_test)
acc = clf.predict(x_train) == y_train.flat
print('Accuracy:%f' % (np.mean(acc)))
x1 = x[:, :2]
x_train, x_test, y_train, y_test = mo.train_test_split(x1, y,random_state=1, test_size=0.3)
clf.fit(x_train, y_train, sample_weight=None)
x1_min, x1_max = x1[:, 0].min(), x1[:, 0].max()
x2_min, x2_max = x1[:, 1].min(), x1[:, 1].max()
x1, x2 = np.mgrid[x1_min:x1_max:200j, x2_min:x2_max:200j]
g_test = np.stack((x1.flat, x2.flat), axis=1)
print(g_test.shape)
g_map = clf.predict(g_test).reshape(x1.shape)
y = clf.predict(x_test)
cm_light = colors.ListedColormap(['#A0FFA0', '#FFA0A0', '#A0A0FF'])
cm_dack = colors.ListedColormap(['r', 'g', 'b'])
plt.pcolormesh(x1, x2, g_map, cmap=cm_light)
plt.scatter(x_test[:, 0], x_test[:, 1],c=np.squeeze(y.flat), s=50, cmap=cm_dack)
plt.plot()
plt.grid()
plt.show()
方法2import numpy as np
from matplotlib import colors
from sklearn import svm
from sklearn.svm import SVC
from sklearn import model_selection
import matplotlib.pyplot as plt
import matplotlib as mpl
def load_data():
#
data = np.loadtxt(r'D:\PycharmProjects\untitled\ \iris.data', dtype=float, delimiter=',', converters={4: iris_type})
return data
def iris_type(s):
# ， string int
it = {b'Iris-setosa': 0, b'Iris-versicolor': 1, b'Iris-virginica': 2}
return it[s]
#
def classifier():
clf = svm.SVC(C=0.5, #
kernel='linear', # kenrel="rbf":
decision_function_shape='ovr') #
return clf
def train(clf, x_train, y_train):
# x_train：
# y_train：
#
clf.fit(x_train, y_train.ravel(),sample_weight=None) # flnumpy.ravelatten
def show_accuracy(a, b, tip):
acc = a.ravel() == b.ravel()
print(a)
print(b)
print(acc)
print('%s Accuracy:%.3f' % (tip, np.mean(acc)))
def print_accuracy(clf, x_train, y_train, x_test, y_test):
#print(x_train)
show_accuracy(clf.predict(x_train), y_train, 'traing data')
show_accuracy(clf.predict(x_test), y_test, 'testing data')
#print(x_train)
#print(y_train.ravel())
#print(clf.predict(x_train))
def draw(clf, x): # ， ：
'''
print(x.shape)
(150, 2)
'''
iris_feature = 'sepal length', 'sepal width', 'petal lenght', 'petal width'
x1_min, x1_max = x[:, 0].min(), x[:, 0].max() # 0
x2_min, x2_max = x[:, 1].min(), x[:, 1].max() # 1
x1, x2 = np.mgrid[x1_min:x1_max:200j, x2_min:x2_max:200j] #
grid_test = np.stack((x1.flat, x2.flat), axis=1) #
'''
print(grid_test.shape)
(40000, 2)
'''
#print('grid_test:
', grid_test)
z = clf.decision_function(grid_test)
#print('the distance to decision plane:
', z)
grid_hat = clf.predict(grid_test) # 【0,0.。。。2,2,2】
'''
print(grid_hat.shape)
(40000,)
'''
#print('grid_hat:
', grid_hat)
grid_hat = grid_hat.reshape(x1.shape) # reshape grid_hat x1
# 3*3 e， e.shape() 3*3, 3 3
cm_light = mpl.colors.ListedColormap(['#A0FFA0', '#FFA0A0', '#A0A0FF'])
'''
x3 = np.mgrid[x1_min:x1_max:200j, x2_min:x2_max:200j] #
print(x3.shape)
(2, 200, 200)
print(x1.shape)
print(x2.shape)
print(grid_hat.shape)
(200, 200)
(200, 200)
x1+x2=x3
(200, 200)
'''
cm_dark = mpl.colors.ListedColormap(['g', 'b', 'r'])
print(grid_hat.shape)
plt.pcolormesh(x1, x2, grid_hat, cmap=cm_light) # pcolormesh(x,y,z,cmap)
# x1，x2，grid_hat，cmap=cm_light 。
plt.scatter(x[:, 0], x[:, 1], c=np.squeeze(y), edgecolor='k', s=50, cmap=cm_dark) #
plt.scatter(x_test[:, 0], x_test[:, 1], s=120, facecolor='none', zorder=10) #
plt.xlabel(iris_feature[0], fontsize=20)
plt.ylabel(iris_feature[1], fontsize=20)
plt.xlim(x1_min, x1_max)
plt.ylim(x2_min, x2_max)
plt.title('svm in iris data classification', fontsize=30)
plt.grid()
plt.show()
# ：
data = load_data()
x, y = np.split(data, (4,), axis=1) # x ，y ，x ，y
# data=(150,5),x=(150,4),y=(150,1)
# x_train,x_test,y_train,y_test = ， ， ，
x_train, x_test, y_train, y_test = model_selection.train_test_split(x, y, random_state=1,
test_size=0.3) # 70%30%
clf = classifier() # svm
train(clf, x_train, y_train) #
print_accuracy(clf, x_train, y_train, x_test, y_test)
# （ ）
data = load_data()
#print(np.shape(data))
x,y = np.split(data,(4,),axis=1) # x ，y ，x ，y
#print(np.shape(x))
#print(np.shape(y))
x=x[:,:2] # ， ，
#print(np.shape(x))
x_train,x_test,y_train,y_test=model_selection.train_test_split(x,y,random_state=1,test_size=0.3)
clf = classifier()
train(clf,x_train,y_train)
print_accuracy(clf,x_train,y_train,x_test,y_test)
draw(clf,x)
