import numpy as np
import matplotlib.pyplot as plt
from sklearn import preprocessing
from sklearn.decomposition import PCA

#加载数据
data = np.loadtxt("C:/Users/18578/Desktop/Anaconda/seeds_dataset.txt")
x = data[:,1:7]
print(x)
dataset= preprocessing.scale(x) #对前七列数据进行标准化
print(dataset)

[[14.84    0.871   5.763   3.312   2.221   5.22  ]
 [14.57    0.8811  5.554   3.333   1.018   4.956 ]
 [14.09    0.905   5.291   3.337   2.699   4.825 ]
 ...
 [13.66    0.8883  5.236   3.232   8.315   5.056 ]
 [13.21    0.8521  5.175   2.836   3.598   5.044 ]
 [13.34    0.8684  5.243   2.974   5.637   5.063 ]]
[[ 2.15462437e-01  6.06017918e-05  3.04218199e-01  1.41701823e-01
  -9.86151745e-01 -3.83577423e-01]
 [ 8.22375713e-03  4.28515270e-01 -1.68624664e-01  1.97432229e-01
  -1.78816620e+00 -9.22013487e-01]
 [-3.60200562e-01  1.44238325e+00 -7.63637453e-01  2.08047544e-01
  -6.67479334e-01 -1.18919199e+00]
 ...
 [-6.90247348e-01  7.33948301e-01 -8.88069786e-01 -7.06044846e-02
   3.07658816e+00 -7.18060432e-01]
 [-1.03564515e+00 -8.01701104e-01 -1.02607655e+00 -1.12152071e+00
  -6.81351965e-02 -7.42534799e-01]
 [-9.35863561e-01 -1.10234659e-01 -8.72232943e-01 -7.55292327e-01
   1.29122264e+00 -7.03783718e-01]]

pca = PCA(n_components=2)
pca.fit(dataset)
DRdata=pca.transform(dataset)#实现降维


#对标签进行处理！
labels = data[:,-1]  #获取标签，将标签转化为整数型
labels=labels.astype(int)

Dataset=np.c_[DRdata,labels]  #处理后的数据集
print(Dataset)

[[ 2.88764155e-01  7.81871323e-01  1.00000000e+00]
 [ 2.28909211e-03  1.91385990e+00  1.00000000e+00]
 [-4.09600763e-01  1.91189872e+00  1.00000000e+00]
 [-4.79583667e-01  1.93249919e+00  1.00000000e+00]
 [ 1.02130956e+00  2.06202929e+00  1.00000000e+00]
 [-2.88311012e-01  1.63943107e+00  1.00000000e+00]
 [-1.34128463e-01  4.38095095e-01  1.00000000e+00]
 [-3.42318429e-01  1.39102227e+00  1.00000000e+00]
 [ 1.67256827e+00  2.91527357e-01  1.00000000e+00]
 [ 1.34319530e+00  1.00061187e+00  1.00000000e+00]
 [ 7.22548998e-04 -4.25283839e-01  1.00000000e+00]
 [-4.90069491e-01  1.56030698e+00  1.00000000e+00]
 [-8.74383101e-01  8.83234199e-01  1.00000000e+00]
 [-8.34950965e-01  8.78540305e-01  1.00000000e+00]
 [-9.31696938e-01  9.66476563e-01  1.00000000e+00]
 [-5.29927195e-01  1.06864388e+00  1.00000000e+00]
 [-7.29487114e-01  1.15133653e+00  1.00000000e+00]
 [ 6.24903392e-01  2.16796623e+00  1.00000000e+00]
 [-2.24016069e-01  2.79333127e+00  1.00000000e+00]
 [-1.57559510e+00  3.50866435e-01  1.00000000e+00]
 [-5.07378475e-01  1.61366428e-01  1.00000000e+00]
 [-4.02247651e-01  7.05500357e-01  1.00000000e+00]
 [ 7.93883556e-01  2.31672813e+00  1.00000000e+00]
 [-1.72418579e+00  1.59742351e+00  1.00000000e+00]
 [-2.26391231e-02  1.01490187e+00  1.00000000e+00]
 [ 1.00552365e+00  1.62406979e+00  1.00000000e+00]
 [-1.42339470e+00  5.65376295e-01  1.00000000e+00]
 [-1.52977390e+00  7.59432915e-01  1.00000000e+00]
 [-3.78269777e-01  1.04520557e+00  1.00000000e+00]
 [-9.69946418e-01  1.27495715e-01  1.00000000e+00]
 [-9.39754484e-01  1.58090267e+00  1.00000000e+00]
 [ 3.11377782e-01  2.52588156e-01  1.00000000e+00]
 [-4.10040955e-01 -4.74421401e-01  1.00000000e+00]
 [-5.12220056e-01  1.11984887e+00  1.00000000e+00]
 [ 4.17228852e-01  9.09729855e-01  1.00000000e+00]
 [ 1.08767092e+00  1.29236059e+00  1.00000000e+00]
 [ 9.59573699e-01  2.89272307e-01  1.00000000e+00]
 [ 1.69320697e+00  1.07906010e+00  1.00000000e+00]
 [ 1.68300363e-01  6.07113267e-01  1.00000000e+00]
 [-6.68618860e-01 -4.27861946e-01  1.00000000e+00]
 [-5.86223084e-01  1.22675304e+00  1.00000000e+00]
 [-5.76124791e-01  1.34604942e+00  1.00000000e+00]
 [-1.02018916e+00  2.04110063e+00  1.00000000e+00]
 [ 7.12450098e-01 -4.33823523e-01  1.00000000e+00]
 [ 4.16549378e-01  1.14947983e+00  1.00000000e+00]
 [-6.83890616e-01  1.68762216e+00  1.00000000e+00]
 [ 4.96645671e-01  1.65249243e+00  1.00000000e+00]
 [ 1.80945562e-01  9.83842582e-01  1.00000000e+00]
 [-8.84160847e-02  1.00314282e+00  1.00000000e+00]
 [ 1.43619158e-01  6.84949970e-01  1.00000000e+00]
 [-2.93212780e-01  2.04720621e-01  1.00000000e+00]
 [ 2.72605389e-01 -2.09822804e-01  1.00000000e+00]
 [-3.50976858e-01 -6.58310214e-01  1.00000000e+00]
 [-2.82840811e-01  6.54616964e-01  1.00000000e+00]
 [ 2.12041412e-02  5.54402734e-01  1.00000000e+00]
 [ 2.19881803e-01  6.36903139e-01  1.00000000e+00]
 [-2.79442651e-01  1.12025231e+00  1.00000000e+00]
 [ 2.31238637e-01  2.29996808e+00  1.00000000e+00]
 [ 5.07774221e-01  1.29769317e+00  1.00000000e+00]
 [-2.20274847e+00  1.28029460e+00  1.00000000e+00]
 [-2.47635004e+00  1.35286181e+00  1.00000000e+00]
 [-2.30774566e+00  1.90788288e+00  1.00000000e+00]
 [-1.76619912e+00  1.63905692e+00  1.00000000e+00]
 [-1.12606000e+00  7.88500662e-02  1.00000000e+00]
 [-1.37344854e+00  1.88990250e+00  1.00000000e+00]
 [-1.45953299e+00  1.92200292e+00  1.00000000e+00]
 [-1.43761359e-01  1.36801176e+00  1.00000000e+00]
 [-4.69877077e-01  7.34114023e-01  1.00000000e+00]
 [-1.30587820e-01  1.19570936e+00  1.00000000e+00]
 [-1.63663043e+00 -1.57090387e-01  1.00000000e+00]
 [ 2.01495125e+00 -1.06284460e+00  2.00000000e+00]
 [ 1.31081515e+00 -1.22104712e+00  2.00000000e+00]
 [ 1.57458044e+00 -7.40296908e-01  2.00000000e+00]
 [ 3.21410689e+00  4.75878198e-01  2.00000000e+00]
 [ 1.51855728e+00 -5.08134008e-01  2.00000000e+00]
 [ 1.08218434e+00 -1.21480627e+00  2.00000000e+00]
 [ 1.45455514e+00 -9.77348529e-01  2.00000000e+00]
 [ 3.65374643e+00 -1.55457590e+00  2.00000000e+00]
 [ 2.85508824e+00 -1.73725338e+00  2.00000000e+00]
 [ 1.61969309e+00  4.49133366e-01  2.00000000e+00]
 [ 1.22067265e+00 -1.10794750e+00  2.00000000e+00]
 [ 2.40254121e+00 -6.78343960e-01  2.00000000e+00]
 [ 3.14930478e+00 -1.21732254e+00  2.00000000e+00]
 [ 3.34580570e+00  1.56757898e-01  2.00000000e+00]
 [ 3.13883814e+00 -4.61633748e-01  2.00000000e+00]
 [ 2.68374475e+00  7.12705640e-02  2.00000000e+00]
 [ 2.92399167e+00  8.02152919e-01  2.00000000e+00]
 [ 3.10525430e+00 -1.60055511e+00  2.00000000e+00]
 [ 3.90229808e+00 -1.44413946e+00  2.00000000e+00]
 [ 3.91439739e+00 -9.86981955e-01  2.00000000e+00]
 [ 3.73410896e+00 -3.95058302e-01  2.00000000e+00]
 [ 2.87115376e+00  1.50162347e-01  2.00000000e+00]
 [ 2.76625113e+00 -1.63619557e-01  2.00000000e+00]
 [ 2.43991037e+00 -8.05270579e-01  2.00000000e+00]
 [ 2.59969781e+00 -2.55916301e+00  2.00000000e+00]
 [ 1.62946906e+00 -8.43597755e-01  2.00000000e+00]
 [ 3.12374562e+00 -6.47295970e-01  2.00000000e+00]
 [ 3.04436296e+00 -5.91934987e-01  2.00000000e+00]
 [ 2.39614220e+00 -8.01536018e-01  2.00000000e+00]
 [ 2.72578037e+00  1.07136669e-01  2.00000000e+00]
 [ 1.04061802e+00 -8.07669215e-02  2.00000000e+00]
 [ 2.22834507e+00  9.70938825e-01  2.00000000e+00]
 [ 2.89738388e+00 -3.43553659e-01  2.00000000e+00]
 [ 2.87730602e+00 -8.54519934e-01  2.00000000e+00]
 [ 2.83907436e+00 -4.51804879e-01  2.00000000e+00]
 [ 2.55878020e+00  4.64391279e-01  2.00000000e+00]
 [ 3.10129102e+00  3.41126034e-01  2.00000000e+00]
 [ 1.89887067e+00 -4.31812470e-01  2.00000000e+00]
 [ 3.75715259e+00 -1.12554270e+00  2.00000000e+00]
 [ 2.53250542e+00  6.41239297e-01  2.00000000e+00]
 [ 2.50734257e+00  6.64747844e-01  2.00000000e+00]
 [ 2.71830348e+00 -7.44443783e-01  2.00000000e+00]
 [ 3.10714154e+00  8.63467336e-01  2.00000000e+00]
 [ 2.36902440e+00 -2.21624158e+00  2.00000000e+00]
 [ 3.85791620e+00 -1.28524605e+00  2.00000000e+00]
 [ 3.13317702e+00  3.07730311e-02  2.00000000e+00]
 [ 2.61527511e+00  1.05597945e-01  2.00000000e+00]
 [ 3.06332703e+00 -1.93595682e-01  2.00000000e+00]
 [ 2.81382910e+00 -1.16365861e-02  2.00000000e+00]
 [ 3.56840116e+00 -6.06363329e-01  2.00000000e+00]
 [ 3.25214117e+00 -1.54582207e+00  2.00000000e+00]
 [ 2.05902795e+00 -5.31818002e-01  2.00000000e+00]
 [ 6.61512408e-01 -8.69013563e-01  2.00000000e+00]
 [ 2.55220807e+00  6.40913338e-01  2.00000000e+00]
 [ 5.23756958e-01  1.36681101e+00  2.00000000e+00]
 [ 2.75956906e+00 -2.24681022e-01  2.00000000e+00]
 [ 2.36085010e+00 -1.25724289e+00  2.00000000e+00]
 [ 2.36675277e+00  7.74055116e-01  2.00000000e+00]
 [ 3.41836661e+00 -1.63484916e-01  2.00000000e+00]
 [ 1.58157930e+00 -4.00975347e-01  2.00000000e+00]
 [ 2.60510756e+00  7.36552494e-01  2.00000000e+00]
 [ 2.79623411e+00  2.52898772e-01  2.00000000e+00]
 [ 6.78692622e-01 -8.21999772e-01  2.00000000e+00]
 [ 8.98290745e-01 -8.41914925e-01  2.00000000e+00]
 [ 8.92088242e-01 -7.52036765e-01  2.00000000e+00]
 [ 5.46588027e-01  7.71777409e-01  2.00000000e+00]
 [ 2.01958591e+00 -4.40127851e-01  2.00000000e+00]
 [ 7.69862933e-01 -4.65062109e-01  2.00000000e+00]
 [ 6.81508816e-01 -2.26215406e-01  2.00000000e+00]
 [ 1.39866807e+00 -2.27599970e-01  2.00000000e+00]
 [-1.21034702e+00 -1.21844476e+00  3.00000000e+00]
 [-9.72382014e-01 -1.79426877e+00  3.00000000e+00]
 [-1.14163778e+00 -1.10550338e+00  3.00000000e+00]
 [-1.67771556e+00 -6.21674559e-01  3.00000000e+00]
 [-2.23947735e+00 -1.05746035e+00  3.00000000e+00]
 [-2.70642495e+00 -2.16935210e+00  3.00000000e+00]
 [-2.31559666e+00  2.55031815e-01  3.00000000e+00]
 [-1.66724605e+00  3.18948132e-02  3.00000000e+00]
 [-1.35435096e+00 -1.25967652e-01  3.00000000e+00]
 [-2.87661039e+00 -1.91792628e+00  3.00000000e+00]
 [-1.93759871e+00 -9.54858647e-01  3.00000000e+00]
 [-2.28272305e+00 -2.43841274e+00  3.00000000e+00]
 [-1.78000938e+00 -1.24914720e+00  3.00000000e+00]
 [-2.64475037e+00 -3.46462722e-01  3.00000000e+00]
 [-2.29183135e+00 -5.96543617e-01  3.00000000e+00]
 [-2.68595747e+00 -1.74821031e+00  3.00000000e+00]
 [-2.25155218e+00  5.16837557e-01  3.00000000e+00]
 [-2.32083046e+00 -1.75988165e+00  3.00000000e+00]
 [-2.31073350e+00 -1.64431728e+00  3.00000000e+00]
 [-2.40597742e+00 -1.62803966e+00  3.00000000e+00]
 [-1.25760411e+00 -3.63739993e-01  3.00000000e+00]
 [-1.75004881e+00 -5.58427927e-01  3.00000000e+00]
 [-2.17274349e+00 -8.41583996e-01  3.00000000e+00]
 [-1.53251859e+00 -3.66200324e-01  3.00000000e+00]
 [-2.68711644e+00 -1.04075531e+00  3.00000000e+00]
 [-1.56015053e+00  1.43966387e+00  3.00000000e+00]
 [-1.72135321e+00 -9.08999539e-01  3.00000000e+00]
 [-1.59300643e+00 -4.47653815e-01  3.00000000e+00]
 [-2.63789053e+00 -8.63765768e-01  3.00000000e+00]
 [-2.58858438e+00 -2.25652355e-01  3.00000000e+00]
 [-2.81022135e+00 -2.31070850e+00  3.00000000e+00]
 [-2.58615011e+00 -1.43566306e+00  3.00000000e+00]
 [-2.61153061e+00 -3.81086524e-01  3.00000000e+00]
 [-2.61638025e+00 -1.18891267e+00  3.00000000e+00]
 [-2.90856756e+00 -1.78054690e+00  3.00000000e+00]
 [-2.56412689e+00 -1.73942614e-01  3.00000000e+00]
 [-2.65772062e+00 -1.14987700e+00  3.00000000e+00]
 [-2.98384388e+00 -7.84428634e-01  3.00000000e+00]
 [-2.56095937e+00 -1.06026477e+00  3.00000000e+00]
 [-1.52319446e+00  6.85259249e-01  3.00000000e+00]
 [-2.62935521e+00 -2.34269440e-01  3.00000000e+00]
 [-1.51633357e+00 -2.11876301e-01  3.00000000e+00]
 [-1.81521069e+00 -4.70437028e-01  3.00000000e+00]
 [-2.19921466e+00 -5.60704218e-01  3.00000000e+00]
 [-1.27171610e+00 -1.43117999e+00  3.00000000e+00]
 [-2.40059340e+00 -1.07934425e+00  3.00000000e+00]
 [-2.17946423e+00 -1.63591070e+00  3.00000000e+00]
 [-2.87721231e+00 -4.02189311e-01  3.00000000e+00]
 [-2.69616507e+00 -1.46127591e+00  3.00000000e+00]
 [-2.93461521e+00  1.12251851e-01  3.00000000e+00]
 [-2.87854197e+00 -9.79876707e-01  3.00000000e+00]
 [-2.37653259e+00  1.10706341e-01  3.00000000e+00]
 [-1.62417577e+00  7.19498519e-01  3.00000000e+00]
 [-2.92339679e+00 -1.14543867e+00  3.00000000e+00]
 [-1.75144013e+00  1.50273186e-01  3.00000000e+00]
 [-1.42665713e+00  4.32224438e-01  3.00000000e+00]
 [-1.54305351e+00 -1.13369088e-01  3.00000000e+00]
 [-9.76140476e-01  2.65584524e-01  3.00000000e+00]
 [-1.43610871e+00 -1.12004829e-01  3.00000000e+00]
 [-1.26454006e+00  1.75996062e+00  3.00000000e+00]
 [-1.82158847e+00 -5.80005547e-01  3.00000000e+00]
 [-1.41686337e+00  2.14893176e+00  3.00000000e+00]
 [-2.51329017e+00  5.27298508e-01  3.00000000e+00]
 [-1.70576570e+00 -1.32649466e+00  3.00000000e+00]
 [-1.81440163e+00  9.71683324e-02  3.00000000e+00]
 [-1.76471606e+00  8.75617095e-01  3.00000000e+00]
 [-2.42639769e+00 -1.96172694e-01  3.00000000e+00]
 [-1.29998347e+00 -1.28667583e+00  3.00000000e+00]
 [-2.09559657e+00  1.10920351e-01  3.00000000e+00]
 [-1.75330579e+00 -5.14756105e-01  3.00000000e+00]]

# kmeans

class MyKmeans:
    def __init__(self, k, n=20):
        self.k = k
        self.n = n
    def fit(self, x, centers=None):
        # 第一步，随机选择 K 个点, 或者指定
        if centers is None:
            idx = np.random.randint(low=0, high=len(x), size=self.k)
            centers = x[idx]
        #print(centers)
        inters = 0
        while inters < self.n:
            #print(inters)
            #print(centers)
            points_set = {key: [] for key in range(self.k)}
            # 第二步，遍历所有点 P，将 P 放入最近的聚类中心的集合中
            for p in x:
                nearest_index = np.argmin(np.sum((centers - p) ** 2, axis=1) ** 0.5)
                points_set[nearest_index].append(p)
            # 第三步，遍历每一个点集，计算新的聚类中心
            for i_k in range(self.k):
                centers[i_k] = sum(points_set[i_k])/len(points_set[i_k])
                
            inters += 1
        return points_set, centers

#将点进行分类
label=Dataset[np.where(labels==1)]
label2=Dataset[np.where(labels==2)]
label3=Dataset[np.where(labels==3)]


for k in range(1,6):
    mykmean = MyKmeans(k,1000)
    points_set, centers = mykmean.fit(DRdata)
    print(centers)
    
#     x = Dataset[:,0]  #取x轴，y轴
#     y = Dataset[:,1]
    
    
    plt.scatter(label[:,0], label[:,1], color='green')
    plt.scatter(label2[:,0], label2[:,1], color='red')
    plt.scatter(label3[:,0], label3[:,1], color='blue')
    for ix, p in enumerate(centers):
        plt.scatter(p[0], p[1], color='C{}'.format(ix), marker='^', edgecolor='black', s=256)
    plt.show()