<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<p>There was a bug in 0.18 that was fixed here
<a class="moz-txt-link-freetext" href="https://github.com/scikit-learn/scikit-learn/pull/9105">https://github.com/scikit-learn/scikit-learn/pull/9105</a></p>
<p>The results from 0.20 should be correct.<br>
</p>
<p>It looks like you're still using Python 2, please be aware that <b>scikit-learn
will drop support for python 2 in the next release</b>!</p>
<p>Nicolas<br>
</p>
<div class="moz-cite-prefix">On 2/15/19 7:52 PM, Huan Tran wrote:<br>
</div>
<blockquote type="cite"
cite="mid:CAK9T6Gh2BHjyeHvRTZi+6W7hucr=6FfThX8aoZtncfkE9Gk4Vg@mail.gmail.com">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div>Dear community,</div>
<div><br>
</div>
<div>I did a very small pca analysis on a 3D data to
print out the explained_variance. I found that with
scikit-learn 0.18.1 AND 0.20.2, the results are
significantly different. In particular, for 0.18.1 I
got<br>
</div>
<div>+3.875925353581E+00 +3.270175297443E+00
+2.207814537475E+00</div>
<div><br>
</div>
<div>and with 0.20.2, I got</div>
<div>+4.651110424297E+00 +3.924210356932E+00
+2.649377444970E+00<br>
</div>
<div><br>
</div>
<div>Could anyone has a hint on what is going on? FYI,
my data and code are enclosed. Many thanks. <br>
</div>
<div><br>
</div>
<div>Huan<br>
</div>
<div> </div>
<div>My data is</div>
<div><br>
</div>
<div> -3.117642E+00, 1.453819E+00, -7.952874E-02<br>
3.081224E+00, 1.453819E+00, -7.952874E-02<br>
1.376932E-01, -2.491454E+00, -1.908521E-01<br>
9.578602E-02, 3.632759E+00, -1.908521E-01<br>
-1.238644E-01, 5.396424E-02, -3.147031E+00<br>
6.335262E-01, 1.393937E+00, 2.500474E+00<br>
</div>
<div><br>
</div>
<div>and my code is <br>
</div>
<div><br>
</div>
<div>import pandas as pd<br>
import numpy as np<br>
from sklearn import decomposition<br>
<br>
df = pd.read_csv('data', delimiter=',', header=None)<br>
data = np.array(df)<br>
<br>
X = data[:,:]<br>
data_size = X.shape[0]<br>
feature_dim = X.shape[1]<br>
<br>
print X<br>
<br>
pca = decomposition.PCA(n_components=feature_dim)<br>
X_transformed = pca.fit_transform(X)<br>
print "%+4.12E %+4.12E %+4.12E"
%(pca.explained_variance_[0],
pca.explained_variance_[1],
pca.explained_variance_[2])<br>
<br>
</div>
<div><br>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<pre class="moz-quote-pre" wrap="">_______________________________________________
scikit-learn mailing list
<a class="moz-txt-link-abbreviated" href="mailto:scikit-learn@python.org">scikit-learn@python.org</a>
<a class="moz-txt-link-freetext" href="https://mail.python.org/mailman/listinfo/scikit-learn">https://mail.python.org/mailman/listinfo/scikit-learn</a>
</pre>
</blockquote>
</body>
</html>