[scikit-learn] Rerunning Kmeans with Python
stephen_malcolm at hotmail.com
Mon Oct 5 02:56:30 EDT 2020
I've added some extra code to plot 6 smaller graphs, which should change the initialization of the centroids.
The title of the plots should be the sum of the squared distance of each initialization (which should tell me when they converge).
The smaller graphs appear (unlabelled) however, my data is not being plotted.
Can someone tell me where I'm going wrong?
n_iter = 9
fig, ax = plt.subplots(3, 3, figsize=(16, 16))
ax = np.ravel(ax)
centers = 
for i in range(n_iter):
# Run local implementation of kmeans
random_state=np.random.randint(0, 1000, size=1)
clusters = km.clusters
ax[i].scatter([km.labels == 0, 0], [km.labels == 0, 1],
c='green', label='cluster 1')
ax[i].scatter([km.labels == 1, 0], [km.labels == 1, 1],
c='blue', label='cluster 2')
ax[i].scatter(clusters[:, 0], clusters[:, 1],
c='r', marker='*', s=300, label='centroid')
From: Stephen Malcolm <stephen_malcolm at hotmail.com>
Sent: 04 October 2020 21:14
To: scikit-learn at python.org <scikit-learn at python.org>
Subject: Rerunning Kmeans with Python
I've written some code to run Kmeans on a data set (please see below).
And I've plotted the results, with my two clusters/ centroids.
However, I've to re-run Kmeans several times and pull up different plots (showing the different centroid positions).
Can someone point me in the right direction how to write this extra code to perform this task?
Then I've to conclude if Kmeans is stable. I believe this is the lowest sum of squared errors?
Thanking you in advance.
#pandas used to read dataset and return the data
#numpy and matplotlib to represent and visualize the data
#sklearn to implement kmeans algorithm
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler
#import the data
data = pd.read_csv('file.csv')
V1_V2 = np.column_stack ((V1, V2))
km_res = KMeans (n_clusters= 2).fit(V1_V2)
y_kmeans = km_res.predict(V1_V2)
plt.scatter(V1, V2, c=y_kmeans, cmap='viridis', s = 50, alpha = 0.5)
plt.title('Visualization of raw data');
clusters = km_res.cluster_centers_
plt.scatter(clusters[:,0], clusters[:,1], c='blue', s=150)
Get Outlook for iOS<https://aka.ms/o0ukef>
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the scikit-learn