What is meanshift
資料集的密度為一個隨核密度分佈，能夠在此資料集中找到局部極值，即為一個 kernel density estimation（它不需要預先知道樣本數據的概率密度分佈函數，完全能夠對樣本點的計算），因此將資料分群。
Example
import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
from sklearn.cluster import MeanShift, estimate_bandwidth
from sklearn import datasets
from sklearn import preprocessing
#create datasets
# iris = datasets.load_iris()
# X = iris.data[:, :4]
url = "https://raw.githubusercontent.com/uiuc-cse/data-fa14/gh-pages/data/iris.csv"
data = pd.read_csv(url)
le = preprocessing.LabelEncoder()
data['species'] = le.fit_transform(data.iloc[:,-1])
X = data.iloc[:, 0:4].to_numpy()
y = data.iloc[:,-1].to_numpy()
plt.scatter(X[:, 0], X[:, 1], c="yellow", marker='o', label='see')
plt.show()
# estimate bandwidth
bandwidth = estimate_bandwidth(X, quantile=0.3)
# Mean Shift method
model = MeanShift(bandwidth = bandwidth, bin_seeding = True, max_iter=500)
model.fit(X)
labels = model.fit_predict(X)
centers = model.cluster_centers_
colormap = np.array(['Red','green','blue'])
plt.scatter(data.petal_length, data.petal_width, c=colormap[y], s=50)
plt.title('Classification Actually')
plt.show()
plt.scatter(data.petal_length, data.petal_width, c=colormap[model.labels_], s=50)
plt.title('Classification Prediction')
plt.show()
其中 estimate bandwidth 會影響 Mean Shift。當 bandwidth 越大，則峰就越平滑，越小則相反，這會導致集群分類的結果。