擅长:python、mysql、java
<p>因为您已经在尝试scikit学习:<code>sklearn.cluster.KMeans</code>应该比<code>Ward</code>更适合缩放,并且支持多核机器上的并行拟合。<a href="http://scikit-learn.org/stable/modules/generated/sklearn.cluster.MiniBatchKMeans.html#sklearn.cluster.MiniBatchKMeans" rel="noreferrer">^{<cd3>}</a>更好,但不会为您随机重新启动。</p>
<pre><code>>>> from sklearn.cluster import MiniBatchKMeans
>>> X = np.random.randn(50000, 7)
>>> %timeit MiniBatchKMeans(30).fit(X)
1 loops, best of 3: 114 ms per loop
</code></pre>