均值漂移算法。
目前为止的代码:
import matplotlib.pyplot as plt
from matplotlib import style
style.use('ggplot')
import numpy as np
X = np.array([[1, 2],
[1.5, 1.8],
[5, 8 ],
[8, 8],
[1, 0.6],
[9,11],
[8,2],
[10,2],
[9,3],])
##plt.scatter(X[:,0], X[:,1], s=150)
##plt.show()
colors = 10*["g","r","c","b","k"]
class Mean_Shift:
def __init__(self, radius=4):
self.radius = radius
def fit(self, data):
centroids = {}
for i in range(len(data)):
centroids[i] = data[i]
while True:
new_centroids = []
for i in centroids:
in_bandwidth = []
centroid = centroids[i]
for featureset in data:
if np.linalg.norm(featureset-centroid) < self.radius:
in_bandwidth.append(featureset)
new_centroid = np.average(in_bandwidth,axis=0)
new_centroids.append(tuple(new_centroid))
uniques = sorted(list(set(new_centroids)))
prev_centroids = dict(centroids)
centroids = {}
for i in range(len(uniques)):
centroids[i] = np.array(uniques[i])
optimized = True
for i in centroids:
if not np.array_equal(centroids[i], prev_centroids[i]):
optimized = False
if not optimized:
break
if optimized:
break
self.centroids = centroids
clf = Mean_Shift()
clf.fit(X)
centroids = clf.centroids
plt.scatter(X[:,0], X[:,1], s=150)
for c in centroids:
plt.scatter(centroids[c][0], centroids[c][1], color='k', marker='*', s=150)
plt.show()
这个代码能够工作,但是我们决定硬编码的半径不好。我们希望做一些更好的事情。首先,我们会修改我们的__init__
方法:
def __init__(self, radius=None, radius_norm_step = 100):
self.radius = radius
self.radius_norm_step = radius_norm_step
所以这里的计划时创建大量的半径,但是逐步处理这个半径,就像带宽一样,或者一些不同长度的半径,我们将其称为步骤。如果特征集靠近半径,它就比远离的点有更大的“权重”。唯一的问题就是,这些步骤应该是什么。现在,开始实现我们的方法:
def fit(self, data):
if self.radius == None:
all_data_centroid = np.average(data, axis=0)
all_data_norm = np.linalg.norm(all_data_centroid)
self.radius = all_data_norm / self.radius_norm_step
centroids = {}
for i in range(len(data)):
centroids[i] = data[i]
这里,如果用户没有硬编码半径,我们就打算寻找所有数据的“中心”。之后,我们会计算数据的模,之后假设每个self.radius
中的半径都是整个数据长度,再除以我们希望的步骤数量。这里,形心的定义和上面的代码相同。现在我们开始while
循环的优化:
weights = [i for i in range(self.radius_norm_step)][::-1]
while True:
new_centroids = []
for i in centroids:
in_bandwidth = []
centroid = centroids[i]
for featureset in data:
#if np.linalg.norm(featureset-centroid) < self.radius:
# in_bandwidth.append(featureset)
distance = np.linalg.norm(featureset-centroid)
if distance == 0:
distance = 0.00000000001
weight_index = int(distance/self.radius)
if weight_index > self.radius_norm_step-1:
weight_index = self.radius_norm_step-1
to_add = (weights[weight_index]**2)*[featureset]
in_bandwidth +=to_add
new_centroid = np.average(in_bandwidth,axis=0)
new_centroids.append(tuple(new_centroid))
uniques = sorted(list(set(new_centroids)))
要注意权重的定义,之后是数据中特征集的改变。
网友评论